Patch set 1 of 3 - The kdb / kgdb merge
This is 3rd and hopefully final kgdb/kdb series originally posted in
December 2009. The next version will hopefully be a pull request to
Linus, depending on any further feedback from the community.
Thank you very much to everyone who provided feedback on the prior
patch sets, as well as to Martin Hicks for some code contributions
which have been folded into the series.
This patch set is stable, and there are several people in the
community making use of the patches.
You can preview the documetation at:
http://kernel.org/pub/linux/kernel/people/jwessel/kdb/
In total there are presently 43 patches, but they have been broken up
into three separate series.
1) 28 patches - The kdb core
2) 7 patches - The debug core kms hooks
3) 8 patches - Early debugging capability with kdb & kgdb
Differences in the V3 patch set:
* The ARM hooks were dropped and implemented another way by RMK
* X86 hw breakpoint regression patches were merged in 2.6.33-rc7
* All the NR_CPUS arrays were factored out of the kdb shell
* The kdb init was divided into an early and late section
to allow extended commands definitions to use kernel memory
allocator
* The KMS debugger API function names were updated for consistency
* Several minor bugs fixed in the kdb shell with the pager
* The early debug series was added to kgdb-next
* Allows early hardware breakpoints and trap on X86
* Allows kgdb and kdb over the EHCI USB Debug Port
* Allows kdb debugging in conjunction with earlyprintk=vga
Differences in the V2 patch set:
* All the code is now in linux-next via kgdb-next
* 6 different build breakages were repaired around different .config
options not compiling
* kms / kdb / kgdb was built tested as kernel modules
* The kdb core was split into 2 patches, 1 for the core, and one for all
the files touched outside the kdb core
* kdb_seq_printf added to eliminate the duplicate meminfo commands in procfs
* kdb breakpoint code refactored to remove unused globals and struct members
* Fixed the possibility to loop infinitely in the kdb md command
* Cleaned up the kdb kernel/modules.c patch
* Removed the unused kdb report helper function in kernel/sched.c
* HW instruction and data access breakpoints implementation completed
* The kdb keyboard code no longer modifies include/linux/keyboard.h
* The kernel_probe_read and kernel_probe_write were cleaned up for blackfin
* The blackfin arch specific kgdb implementation was cleaned up and merged
* The debug core now turns off tracing while the kernel debugger is active
* New in the series is a command for kdb called ftdump, which allows
dumping the ftrace buffer
* kdb will automatically set the correct number of rows on the display
via the information from the vt_console when using kms
* There is preliminary documentation for kdb and kms
* The drm hacks patch is still a hack, but cleaned up and using macros
The acks were added for the blackfin, sparc and sh arch specific code by the
respective maintainers.
-- From v1 patch set --
Back in May 2009 an initial kdb prototype was posted to lkml as an RFC
to see if there was sufficient interest to merge kdb and kgdb together.
There was enough interest to continue on with the project, and it is
now in a state where further work can occur to review, and hopefully
merge the code into a future kernel.
If this is the first time you have heard about kgdb or kdb, you might
consider taking a look at the presentation slides from LPC 2009.
http://kgdb.wiki.kernel.org/index.php/Main_Page#Linux_Plumbers_Conference_presentation_2009
The code included in this series does not include any of the USB code
referenced in the presentation.
Several changes have occurred since the original prototype was posted.
All of the architectures using the kgdb core today have had the
functions implemented to make use of the kdb shell. This list
includes x86, arm, ppc, mips, sh, sparc, and blackfin. The proposed
atomic kernel modesetting project now has an initial implementation
for the intel i915 driver, and it should be possible to add atomic
mode setting hooks for any other video driver which makes use of
builtin kernel mode setting. All the dead code and functions that
were not previously implemented in the RFC have been removed or
implemented.
The directory of the kernel debugger changed as well. All the kernel
debugger pieces, the debug core, gdbstub and kdb now live under
kernel/debug.
While the patch set may be slightly large, it is broken down into
logical, incremental functionality. Only the first 12 patches are
needed to see kdb in action on a serial port which has a kgdboc driver
for example.
It is important to understand that this work was directly derived from
the original kdb, and the intent is to deprecate the out of tree kdb
and move all its functionality to this code base. This version of kdb
is completely wired into the debug core and the kgdboc polled I/O
model using the same API used by the gdbstub (which people call kgdb).
The kdb front end in this patch series works a little differently than
if you were to take the original kdb patch set from:
ftp://oss.sgi.com/projects/kdb/download/v4.4/
In the kernel .config you should enable the following options:
CONFIG_KGDB=y
CONFIG_KGDB_SERIAL_CONSOLE=y
CONFIG_KGDB_LOW_LEVEL_TRAP=y
CONFIG_KGDB_KDB=y
CONFIG_KDB_KEYBOARD=y
To use kdb with a serial port you would use the kgdb/kgdboc way of
doing things. You would used a kernel command line with:
console=ttyS0,115200 kgdboc=ttyS0 kgdbwait
That will get you access to kdb just after the console has been
registered. If you want to use the keyboard, you could use the
following (NOTE it is kbd and not kdb, kbd is short for keyboard):
console=tty0 kgdboc=kbd kgdbwait
You can also use the keyboard and or serial console:
console=ttyS0,115200 console=tty0 kgdboc=kbd,ttyS0
In terms of breaking into the debugger after the system is up, you
must use the sysrq-g sequence. That means you could run:
echo g > /proc/sysrq-trigger
Or you can use the SysRq key on your key board. On a typical laptop
you might have to do the following:
press and hold ALT -- You will be holding this the whole time
press and hold FN
press and release the key with the SysRq label
release FN
press and release g
release ALT
Once you are in kdb you can run help to see a limited list of
commands.
You can also still connect gdb or re-enter kdb without leaving the
exception state, if you are using a serial port. To get out of kgdb
mode you can type "$3#33", or to get into kgdb mode from kdb, you can
type "kgdb".
>From gdb you can issue commands to the kdb front end, via gdb's
monitor command. For instance you could issue "monitor lsmod".
Allowing the gdb monitor extension was certainly another motivation
behind the prototype.
Included among the kdb patches are some kgdb specific fixes for the
next kernel merge window, and if kdb were to not make the cut, these
patches will be split out of the series. Anything too ugly for a
possible merge is also a candidate to get removed from the series.
Thanks,
Jason.
You can find the development branch for this here:
http://git.kernel.org/?p=linux/kernel/git/jwessel/linux-2.6-kgdb.git;a=shortlog;h=kdb-merge
---
Jason Wessel (28):
Move kernel/kgdb.c to kernel/debug/debug_core.c
Separate the gdbstub from the debug core
kgdb: eliminate kgdb_wait(), all cpus enter the same way
kgdb,sparc: Add in kgdb_arch_set_pc for sparc
kgdb,sh: update superh kgdb exception handling
kgdb,blackfin: Add in kgdb_arch_set_pc for blackfin
kdb: core for kgdb back end (1 of 2)
kdb: core for kgdb back end (2 of 2)
kgdb: core changes to support kdb
kgdb,8250,pl011: Return immediately from console poll
sh,sh-sci: Use NO_POLL_CHAR in the SCIF polled console code
sparc,sunzilog: Add console polling support for sunzilog serial driver
kgdb: gdb "monitor" -> kdb passthrough
kgdboc,keyboard: Keyboard driver for kdb with kgdb
kgdb,docs: Update the kgdb docs to include kdb
kgdb: remove post_primary_code references
x86,kgdb: Add low level debug hook
powerpc,kgdb: Introduce low level trap catching
mips,kgdb: kdb low level trap catch and stack trace
kgdb: Add the ability to schedule a breakpoint via a tasklet
kgdboc,kdb: Allow kdb to work on a non open console port
printk,kdb: capture printk() when in kdb shell
keyboard, input: Add hook to input to allow low level event clear
debug_core,kdb: Allow the debug core to process a recursive debug entry
MAINTAINERS: update kgdb, kdb, and debug_core info
kdb,panic,debug_core: Allow the debug core to receive a panic before smp_send_stop()
kgdbts,sh: Add in breakpoint pc offset for superh
debug_core: Turn off tracing while in the debugger
Documentation/DocBook/kgdb.tmpl | 640 ++++++--
Documentation/kernel-parameters.txt | 15 +-
MAINTAINERS | 6 +-
arch/arm/include/asm/kmap_types.h | 1 +
arch/arm/kernel/kgdb.c | 5 +
arch/blackfin/kernel/kgdb.c | 5 +
arch/mips/include/asm/kgdb.h | 2 +
arch/mips/kernel/kgdb.c | 27 +-
arch/mips/kernel/traps.c | 13 +
arch/powerpc/include/asm/kmap_types.h | 1 +
arch/powerpc/kernel/kgdb.c | 12 +-
arch/powerpc/kernel/traps.c | 7 +
arch/sh/kernel/kgdb.c | 14 +-
arch/sparc/kernel/kgdb_32.c | 6 +
arch/sparc/kernel/kgdb_64.c | 6 +
arch/x86/include/asm/kgdb.h | 3 +
arch/x86/kernel/kgdb.c | 56 +-
arch/x86/kernel/traps.c | 6 +
drivers/char/Makefile | 1 +
drivers/char/kdb_keyboard.c | 204 +++
drivers/char/kdb_keyboard.h | 143 ++
drivers/char/keyboard.c | 29 +-
drivers/input/input.c | 15 +
drivers/misc/kgdbts.c | 6 +
drivers/serial/8250.c | 4 +-
drivers/serial/amba-pl011.c | 6 +-
drivers/serial/kgdboc.c | 88 +-
drivers/serial/sh-sci.c | 6 +-
drivers/serial/sunzilog.c | 50 +
fs/proc/internal.h | 4 +-
fs/proc/meminfo.c | 15 +-
fs/proc/mmu.c | 8 +-
include/asm-generic/kmap_types.h | 3 +-
include/linux/input.h | 10 +
include/linux/kdb.h | 117 ++
include/linux/kgdb.h | 41 +-
include/linux/serial_core.h | 1 +
include/linux/swap.h | 2 +
init/main.c | 2 +
kernel/Makefile | 2 +-
kernel/debug/Makefile | 7 +
kernel/debug/debug_core.c | 966 +++++++++++
kernel/debug/debug_core.h | 81 +
kernel/debug/gdbstub.c | 1022 ++++++++++++
kernel/debug/kdb/.gitignore | 1 +
kernel/debug/kdb/Makefile | 24 +
kernel/debug/kdb/kdb_bp.c | 564 +++++++
kernel/debug/kdb/kdb_bt.c | 210 +++
kernel/debug/kdb/kdb_cmds | 35 +
kernel/debug/kdb/kdb_debugger.c | 169 ++
kernel/debug/kdb/kdb_io.c | 826 ++++++++++
kernel/debug/kdb/kdb_main.c | 2855 +++++++++++++++++++++++++++++++++
kernel/debug/kdb/kdb_private.h | 312 ++++
kernel/debug/kdb/kdb_support.c | 991 ++++++++++++
kernel/kallsyms.c | 21 +
kernel/kgdb.c | 1763 --------------------
kernel/module.c | 4 +
kernel/panic.c | 7 +-
kernel/printk.c | 25 +
kernel/sched.c | 7 +-
kernel/signal.c | 40 +
lib/Kconfig.kgdb | 24 +-
mm/swapfile.c | 10 +-
63 files changed, 9528 insertions(+), 2018 deletions(-)
create mode 100644 drivers/char/kdb_keyboard.c
create mode 100644 drivers/char/kdb_keyboard.h
create mode 100644 include/linux/kdb.h
create mode 100644 kernel/debug/Makefile
create mode 100644 kernel/debug/debug_core.c
create mode 100644 kernel/debug/debug_core.h
create mode 100644 kernel/debug/gdbstub.c
create mode 100644 kernel/debug/kdb/.gitignore
create mode 100644 kernel/debug/kdb/Makefile
create mode 100644 kernel/debug/kdb/kdb_bp.c
create mode 100644 kernel/debug/kdb/kdb_bt.c
create mode 100644 kernel/debug/kdb/kdb_cmds
create mode 100644 kernel/debug/kdb/kdb_debugger.c
create mode 100644 kernel/debug/kdb/kdb_io.c
create mode 100644 kernel/debug/kdb/kdb_main.c
create mode 100644 kernel/debug/kdb/kdb_private.h
create mode 100644 kernel/debug/kdb/kdb_support.c
delete mode 100644 kernel/kgdb.c
The new debug core api requires all architectures that use to debug
core to implement a function to set the program counter.
Signed-off-by: Jason Wessel <[email protected]>
Acked-by: David S. Miller <[email protected]>
---
arch/sparc/kernel/kgdb_32.c | 6 ++++++
arch/sparc/kernel/kgdb_64.c | 6 ++++++
2 files changed, 12 insertions(+), 0 deletions(-)
diff --git a/arch/sparc/kernel/kgdb_32.c b/arch/sparc/kernel/kgdb_32.c
index 04df4ed..539243b 100644
--- a/arch/sparc/kernel/kgdb_32.c
+++ b/arch/sparc/kernel/kgdb_32.c
@@ -158,6 +158,12 @@ void kgdb_arch_exit(void)
{
}
+void kgdb_arch_set_pc(struct pt_regs *regs, unsigned long ip)
+{
+ regs->pc = ip;
+ regs->npc = regs->pc + 4;
+}
+
struct kgdb_arch arch_kgdb_ops = {
/* Breakpoint instruction: ta 0x7d */
.gdb_bpt_instr = { 0x91, 0xd0, 0x20, 0x7d },
diff --git a/arch/sparc/kernel/kgdb_64.c b/arch/sparc/kernel/kgdb_64.c
index f5a0fd4..9bf387b 100644
--- a/arch/sparc/kernel/kgdb_64.c
+++ b/arch/sparc/kernel/kgdb_64.c
@@ -180,6 +180,12 @@ void kgdb_arch_exit(void)
{
}
+void kgdb_arch_set_pc(struct pt_regs *regs, unsigned long ip)
+{
+ regs->tpc = ip;
+ regs->tnpc = regs->tpc + 4;
+}
+
struct kgdb_arch arch_kgdb_ops = {
/* Breakpoint instruction: ta 0x72 */
.gdb_bpt_instr = { 0x91, 0xd0, 0x20, 0x72 },
--
1.6.4.rc1
Remove all the references to the kgdb_post_primary_code. This
function serves no useful purpose because you can obtain the same
information from the "struct kgdb_state *ks" from with in the
debugger, if for some reason you want the data.
Also remove the unintentional duplicate assignment for ks->ex_vector.
Signed-off-by: Jason Wessel <[email protected]>
---
arch/x86/kernel/kgdb.c | 29 -----------------------------
include/linux/kgdb.h | 14 --------------
kernel/debug/debug_core.c | 8 --------
3 files changed, 0 insertions(+), 51 deletions(-)
diff --git a/arch/x86/kernel/kgdb.c b/arch/x86/kernel/kgdb.c
index c7a20d7..c38c8fb 100644
--- a/arch/x86/kernel/kgdb.c
+++ b/arch/x86/kernel/kgdb.c
@@ -47,20 +47,8 @@
#include <asm/debugreg.h>
#include <asm/apicdef.h>
#include <asm/system.h>
-
#include <asm/apic.h>
-/*
- * Put the error code here just in case the user cares:
- */
-static int gdb_x86errcode;
-
-/*
- * Likewise, the vector number here (since GDB only gets the signal
- * number through the usual means, and that's not very specific):
- */
-static int gdb_x86vector = -1;
-
/**
* pt_regs_to_gdb_regs - Convert ptrace regs to GDB regs
* @gdb_regs: A pointer to hold the registers in the order GDB wants.
@@ -399,23 +387,6 @@ void kgdb_disable_hw_debug(struct pt_regs *regs)
}
}
-/**
- * kgdb_post_primary_code - Save error vector/code numbers.
- * @regs: Original pt_regs.
- * @e_vector: Original error vector.
- * @err_code: Original error code.
- *
- * This is needed on architectures which support SMP and KGDB.
- * This function is called after all the slave cpus have been put
- * to a know spin state and the primary CPU has control over KGDB.
- */
-void kgdb_post_primary_code(struct pt_regs *regs, int e_vector, int err_code)
-{
- /* primary processor is completely in the debugger */
- gdb_x86vector = e_vector;
- gdb_x86errcode = err_code;
-}
-
#ifdef CONFIG_SMP
/**
* kgdb_roundup_cpus - Get other CPUs into a holding pattern
diff --git a/include/linux/kgdb.h b/include/linux/kgdb.h
index 407edb1..406f6f9 100644
--- a/include/linux/kgdb.h
+++ b/include/linux/kgdb.h
@@ -36,20 +36,6 @@ struct pt_regs;
extern int kgdb_skipexception(int exception, struct pt_regs *regs);
/**
- * kgdb_post_primary_code - (optional) Save error vector/code numbers.
- * @regs: Original pt_regs.
- * @e_vector: Original error vector.
- * @err_code: Original error code.
- *
- * This is usually needed on architectures which support SMP and
- * KGDB. This function is called after all the secondary cpus have
- * been put to a know spin state and the primary CPU has control over
- * KGDB.
- */
-extern void kgdb_post_primary_code(struct pt_regs *regs, int e_vector,
- int err_code);
-
-/**
* kgdb_disable_hw_debug - (optional) Disable hardware debugging hook
* @regs: Current &struct pt_regs.
*
diff --git a/kernel/debug/debug_core.c b/kernel/debug/debug_core.c
index dbf81aa..9cfc71d 100644
--- a/kernel/debug/debug_core.c
+++ b/kernel/debug/debug_core.c
@@ -203,12 +203,6 @@ int __weak kgdb_skipexception(int exception, struct pt_regs *regs)
return 0;
}
-void __weak
-kgdb_post_primary_code(struct pt_regs *regs, int e_vector, int err_code)
-{
- return;
-}
-
/**
* kgdb_disable_hw_debug - Disable hardware debugging while we in kgdb.
* @regs: Current &struct pt_regs.
@@ -586,7 +580,6 @@ return_normal:
* At this point the primary processor is completely
* in the debugger and all secondary CPUs are quiescent
*/
- kgdb_post_primary_code(ks->linux_regs, ks->ex_vector, ks->err_code);
dbg_deactivate_sw_breakpoints();
kgdb_single_step = 0;
kgdb_contthread = current;
@@ -670,7 +663,6 @@ kgdb_handle_exception(int evector, int signo, int ecode, struct pt_regs *regs)
ks->cpu = raw_smp_processor_id();
ks->ex_vector = evector;
ks->signo = signo;
- ks->ex_vector = evector;
ks->err_code = ecode;
ks->kgdb_usethreadid = 0;
ks->linux_regs = regs;
--
1.6.4.rc1
The kgdb test suite mimics the behavior of gdb. For the sh
architecture the pc must be decremented by 2 for software breakpoint.
Signed-off-by: Jason Wessel <[email protected]>
Acked-by: Paul Mundt <[email protected]>
---
drivers/misc/kgdbts.c | 6 ++++++
1 files changed, 6 insertions(+), 0 deletions(-)
diff --git a/drivers/misc/kgdbts.c b/drivers/misc/kgdbts.c
index fcb6ec1..7245023 100644
--- a/drivers/misc/kgdbts.c
+++ b/drivers/misc/kgdbts.c
@@ -295,6 +295,10 @@ static int check_and_rewind_pc(char *put_str, char *arg)
/* On x86 a breakpoint stop requires it to be decremented */
if (addr + 1 == kgdbts_regs.ip)
offset = -1;
+#elif defined(CONFIG_SUPERH)
+ /* On SUPERH a breakpoint stop requires it to be decremented */
+ if (addr + 2 == kgdbts_regs.pc)
+ offset = -2;
#endif
if (strcmp(arg, "silent") &&
instruction_pointer(&kgdbts_regs) + offset != addr) {
@@ -305,6 +309,8 @@ static int check_and_rewind_pc(char *put_str, char *arg)
#ifdef CONFIG_X86
/* On x86 adjust the instruction pointer if needed */
kgdbts_regs.ip += offset;
+#elif defined(CONFIG_SUPERH)
+ kgdbts_regs.pc += offset;
#endif
return 0;
}
--
1.6.4.rc1
The kernel debugger should turn off kernel tracing any time the
debugger is active and restore it on resume.
CC: Steven Rostedt <[email protected]>
Signed-off-by: Jason Wessel <[email protected]>
---
kernel/debug/debug_core.c | 8 ++++++++
1 files changed, 8 insertions(+), 0 deletions(-)
diff --git a/kernel/debug/debug_core.c b/kernel/debug/debug_core.c
index 580fc3b..f7ebb7f 100644
--- a/kernel/debug/debug_core.c
+++ b/kernel/debug/debug_core.c
@@ -474,6 +474,7 @@ static int kgdb_cpu_enter(struct kgdb_state *ks, struct pt_regs *regs)
int sstep_tries = 100;
int error;
int i, cpu;
+ int trace_on = 0;
acquirelock:
/*
* Interrupts will be restored by the 'trap return' code, except when
@@ -518,6 +519,8 @@ return_normal:
*/
if (arch_kgdb_ops.correct_hw_break)
arch_kgdb_ops.correct_hw_break();
+ if (trace_on)
+ tracing_on();
atomic_set(&cpu_in_kgdb[cpu], 0);
touch_softlockup_watchdog_sync();
clocksource_touch_watchdog();
@@ -592,6 +595,9 @@ return_normal:
kgdb_single_step = 0;
kgdb_contthread = current;
exception_level = 0;
+ trace_on = tracing_is_on();
+ if (trace_on)
+ tracing_off();
while (1) {
cpu_master_loop:
@@ -645,6 +651,8 @@ kgdb_restore:
else
kgdb_sstep_pid = 0;
}
+ if (trace_on)
+ tracing_on();
/* Free kgdb_active */
atomic_set(&kgdb_active, -1);
touch_softlockup_watchdog_sync();
--
1.6.4.rc1
Certain calls from the kdb shell will call out to printk(), and any of
these calls should get vectored back to the kdb_printf() so that the
kdb pager and processing can be used, as well as to properly channel
I/O to the polled I/O devices.
CC: Randy Dunlap <[email protected]>
CC: Andrew Morton <[email protected]>
Signed-off-by: Jason Wessel <[email protected]>
---
include/linux/kdb.h | 3 +++
kernel/debug/kdb/kdb_bt.c | 2 ++
kernel/debug/kdb/kdb_io.c | 24 ++++++++++++++++++++----
kernel/debug/kdb/kdb_main.c | 4 ++++
kernel/printk.c | 9 +++++++++
5 files changed, 38 insertions(+), 4 deletions(-)
diff --git a/include/linux/kdb.h b/include/linux/kdb.h
index d72fa39..ccb2b3e 100644
--- a/include/linux/kdb.h
+++ b/include/linux/kdb.h
@@ -78,6 +78,9 @@ typedef enum {
KDB_REASON_SSTEP, /* Single Step trap. - regs valid */
} kdb_reason_t;
+extern int kdb_trap_printk;
+extern int vkdb_printf(const char *fmt, va_list args)
+ __attribute__ ((format (printf, 1, 0)));
extern int kdb_printf(const char *, ...)
__attribute__ ((format (printf, 1, 2)));
typedef int (*kdb_printf_t)(const char *, ...)
diff --git a/kernel/debug/kdb/kdb_bt.c b/kernel/debug/kdb/kdb_bt.c
index 483fa4e..2f62fe8 100644
--- a/kernel/debug/kdb/kdb_bt.c
+++ b/kernel/debug/kdb/kdb_bt.c
@@ -23,6 +23,7 @@ static void kdb_show_stack(struct task_struct *p, void *addr)
{
int old_lvl = console_loglevel;
console_loglevel = 15;
+ kdb_trap_printk++;
kdb_set_current_task(p);
if (addr) {
show_stack((struct task_struct *)p, addr);
@@ -36,6 +37,7 @@ static void kdb_show_stack(struct task_struct *p, void *addr)
show_stack(p, NULL);
}
console_loglevel = old_lvl;
+ kdb_trap_printk--;
}
/*
diff --git a/kernel/debug/kdb/kdb_io.c b/kernel/debug/kdb/kdb_io.c
index 58be7e9..c9b7f4f 100644
--- a/kernel/debug/kdb/kdb_io.c
+++ b/kernel/debug/kdb/kdb_io.c
@@ -29,6 +29,7 @@
#define CMD_BUFLEN 256
char kdb_prompt_str[CMD_BUFLEN];
+int kdb_trap_printk;
static void kgdb_transition_check(char *buffer)
{
@@ -533,12 +534,12 @@ static int kdb_search_string(char *searched, char *searchfor)
return 0;
}
-int kdb_printf(const char *fmt, ...)
+int vkdb_printf(const char *fmt, va_list ap)
{
- va_list ap;
int diag;
int linecount;
int logging, saved_loglevel = 0;
+ int saved_trap_printk;
int got_printf_lock = 0;
int retlen = 0;
int fnd, len;
@@ -549,6 +550,9 @@ int kdb_printf(const char *fmt, ...)
unsigned long uninitialized_var(flags);
preempt_disable();
+ saved_trap_printk = kdb_trap_printk;
+ kdb_trap_printk = 0;
+
/* Serialize kdb_printf if multiple cpus try to write at once.
* But if any cpu goes recursive in kdb, just print the output,
* even if it is interleaved with any other text.
@@ -575,9 +579,7 @@ int kdb_printf(const char *fmt, ...)
next_avail = kdb_buffer;
size_avail = sizeof(kdb_buffer);
}
- va_start(ap, fmt);
vsnprintf(next_avail, size_avail, fmt, ap);
- va_end(ap);
/*
* If kdb_parse() found that the command was cmd xxx | grep yyy
@@ -805,6 +807,20 @@ kdb_print_out:
} else {
__release(kdb_printf_lock);
}
+ kdb_trap_printk = saved_trap_printk;
preempt_enable();
return retlen;
}
+
+int kdb_printf(const char *fmt, ...)
+{
+ va_list ap;
+ int r;
+
+ va_start(ap, fmt);
+ r = vkdb_printf(fmt, ap);
+ va_end(ap);
+
+ return r;
+}
+
diff --git a/kernel/debug/kdb/kdb_main.c b/kernel/debug/kdb/kdb_main.c
index a5b82cd..945132f 100644
--- a/kernel/debug/kdb/kdb_main.c
+++ b/kernel/debug/kdb/kdb_main.c
@@ -1056,7 +1056,9 @@ static void kdb_dumpregs(struct pt_regs *regs)
{
int old_lvl = console_loglevel;
console_loglevel = 15;
+ kdb_trap_printk++;
show_regs(regs);
+ kdb_trap_printk--;
kdb_printf("\n");
console_loglevel = old_lvl;
}
@@ -1823,7 +1825,9 @@ static int kdb_sr(int argc, const char **argv)
__sysrq_enabled = 1;
}
+ kdb_trap_printk++;
handle_sysrq(*argv[1], NULL);
+ kdb_trap_printk--;
return 0;
}
diff --git a/kernel/printk.c b/kernel/printk.c
index d86c91a..be9bd70 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -33,6 +33,7 @@
#include <linux/bootmem.h>
#include <linux/syscalls.h>
#include <linux/kexec.h>
+#include <linux/kdb.h>
#include <linux/ratelimit.h>
#include <linux/kmsg_dump.h>
@@ -609,6 +610,14 @@ asmlinkage int printk(const char *fmt, ...)
va_list args;
int r;
+#ifdef CONFIG_KGDB_KDB
+ if (unlikely(kdb_trap_printk)) {
+ va_start(args, fmt);
+ r = vkdb_printf(fmt, args);
+ va_end(args);
+ return r;
+ }
+#endif
va_start(args, fmt);
r = vprintk(fmt, args);
va_end(args);
--
1.6.4.rc1
It is highly desirable to trap into kdb on panic, and in order
to do so the notifier must be called before smp_send_stop(),
else the debug_core will not be able to correctly trap
all the CPUs.
CC: Ingo Molnar <[email protected]>
CC: Andrew Morton <[email protected]>
Signed-off-by: Jason Wessel <[email protected]>
---
kernel/debug/debug_core.c | 18 ++++++++++++++++++
kernel/panic.c | 7 ++++---
2 files changed, 22 insertions(+), 3 deletions(-)
diff --git a/kernel/debug/debug_core.c b/kernel/debug/debug_core.c
index bfdba97..580fc3b 100644
--- a/kernel/debug/debug_core.c
+++ b/kernel/debug/debug_core.c
@@ -754,11 +754,27 @@ static struct sysrq_key_op sysrq_dbg_op = {
};
#endif
+static int kgdb_panic_event(struct notifier_block *self,
+ unsigned long val,
+ void *data)
+{
+ if (dbg_kdb_mode)
+ kdb_printf("PANIC: %s\n", (char *)data);
+ kgdb_breakpoint();
+ return NOTIFY_DONE;
+}
+
+static struct notifier_block kgdb_panic_event_nb = {
+ .notifier_call = kgdb_panic_event,
+};
+
static void kgdb_register_callbacks(void)
{
if (!kgdb_io_module_registered) {
kgdb_io_module_registered = 1;
kgdb_arch_init();
+ atomic_notifier_chain_register(&panic_notifier_list,
+ &kgdb_panic_event_nb);
#ifdef CONFIG_MAGIC_SYSRQ
register_sysrq_key('g', &sysrq_dbg_op);
#endif
@@ -778,6 +794,8 @@ static void kgdb_unregister_callbacks(void)
*/
if (kgdb_io_module_registered) {
kgdb_io_module_registered = 0;
+ atomic_notifier_chain_unregister(&panic_notifier_list,
+ &kgdb_panic_event_nb);
kgdb_arch_exit();
#ifdef CONFIG_MAGIC_SYSRQ
unregister_sysrq_key('g', &sysrq_dbg_op);
diff --git a/kernel/panic.c b/kernel/panic.c
index c787333..428d15b 100644
--- a/kernel/panic.c
+++ b/kernel/panic.c
@@ -66,10 +66,13 @@ NORET_TYPE void panic(const char * fmt, ...)
*/
preempt_disable();
- bust_spinlocks(1);
va_start(args, fmt);
vsnprintf(buf, sizeof(buf), fmt, args);
va_end(args);
+
+ atomic_notifier_call_chain(&panic_notifier_list, 0, buf);
+
+ bust_spinlocks(1);
printk(KERN_EMERG "Kernel panic - not syncing: %s\n",buf);
#ifdef CONFIG_DEBUG_BUGVERBOSE
dump_stack();
@@ -91,8 +94,6 @@ NORET_TYPE void panic(const char * fmt, ...)
*/
smp_send_stop();
- atomic_notifier_call_chain(&panic_notifier_list, 0, buf);
-
bust_spinlocks(0);
if (!panic_blink)
--
1.6.4.rc1
When using a keyboard with kdb, on resuming the system there needs to
be a hook to allow for the keyboard state to get reset.
This is mainly because there is no way to force the end user to hold
down the original keys that were pressed prior to entering kdb.
CC: Dmitry Torokhov <[email protected]>
Signed-off-by: Jason Wessel <[email protected]>
---
drivers/char/keyboard.c | 29 ++++++++++++++++++++++++-----
drivers/input/input.c | 15 +++++++++++++++
drivers/serial/kgdboc.c | 13 +++++++++++++
include/linux/input.h | 10 ++++++++++
4 files changed, 62 insertions(+), 5 deletions(-)
diff --git a/drivers/char/keyboard.c b/drivers/char/keyboard.c
index f706b1d..d61145a 100644
--- a/drivers/char/keyboard.c
+++ b/drivers/char/keyboard.c
@@ -1195,6 +1195,11 @@ static void kbd_keycode(unsigned int keycode, int down, int hw_raw)
if (keycode < BTN_MISC && printk_ratelimit())
printk(KERN_WARNING "keyboard.c: can't emulate rawmode for keycode %d\n", keycode);
+ if (down)
+ set_bit(keycode, key_down);
+ else
+ clear_bit(keycode, key_down);
+
#ifdef CONFIG_MAGIC_SYSRQ /* Handle the SysRq Hack */
if (keycode == KEY_SYSRQ && (sysrq_down || (down == 1 && sysrq_alt))) {
if (!sysrq_down) {
@@ -1237,11 +1242,6 @@ static void kbd_keycode(unsigned int keycode, int down, int hw_raw)
raw_mode = 1;
}
- if (down)
- set_bit(keycode, key_down);
- else
- clear_bit(keycode, key_down);
-
if (rep &&
(!vc_kbd_mode(kbd, VC_REPEAT) ||
(tty && !L_ECHO(tty) && tty_chars_in_buffer(tty)))) {
@@ -1410,6 +1410,22 @@ static const struct input_device_id kbd_ids[] = {
MODULE_DEVICE_TABLE(input, kbd_ids);
+#ifdef CONFIG_KDB_KEYBOARD
+void kbd_clear_keys(void)
+{
+ int i, j, k;
+
+ for (i = 0; i < ARRAY_SIZE(key_down); i++) {
+ k = i * BITS_PER_LONG;
+ for (j = 0; j < BITS_PER_LONG; j++, k++) {
+ if (test_bit(k, key_down)) {
+ kbd_keycode(k, 0, 0);
+ }
+ }
+ }
+}
+#endif
+
static struct input_handler kbd_handler = {
.event = kbd_event,
.connect = kbd_connect,
@@ -1417,6 +1433,9 @@ static struct input_handler kbd_handler = {
.start = kbd_start,
.name = "kbd",
.id_table = kbd_ids,
+#ifdef CONFIG_KDB_KEYBOARD
+ .dbg_clear_keys = kbd_clear_keys,
+#endif
};
int __init kbd_init(void)
diff --git a/drivers/input/input.c b/drivers/input/input.c
index 86cb2d2..7473300 100644
--- a/drivers/input/input.c
+++ b/drivers/input/input.c
@@ -1692,6 +1692,21 @@ int input_register_handler(struct input_handler *handler)
}
EXPORT_SYMBOL(input_register_handler);
+#ifdef CONFIG_KDB_KEYBOARD
+/* input_db_clear_keys - Clear any keyboards if they have a call back,
+ * after returning from the kernel debugger
+ */
+void input_dbg_clear_keys(void)
+{
+ struct input_handler *handler;
+
+ list_for_each_entry(handler, &input_handler_list, node)
+ if (handler->dbg_clear_keys)
+ handler->dbg_clear_keys();
+}
+EXPORT_SYMBOL_GPL(input_dbg_clear_keys);
+#endif
+
/**
* input_unregister_handler - unregisters an input handler
* @handler: handler to be unregistered
diff --git a/drivers/serial/kgdboc.c b/drivers/serial/kgdboc.c
index b765ab4..201cdf5 100644
--- a/drivers/serial/kgdboc.c
+++ b/drivers/serial/kgdboc.c
@@ -17,6 +17,7 @@
#include <linux/kdb.h>
#include <linux/tty.h>
#include <linux/console.h>
+#include <linux/input.h>
#define MAX_CONFIG_LEN 40
@@ -35,12 +36,16 @@ static struct tty_driver *kgdb_tty_driver;
static int kgdb_tty_line;
#ifdef CONFIG_KDB_KEYBOARD
+static int kgdboc_use_kbd; /* 1 if we use a keyboard */
+
static int kgdboc_register_kbd(char **cptr)
{
+ kgdboc_use_kbd = 0;
if (strncmp(*cptr, "kbd", 3) == 0) {
if (kdb_poll_idx < KDB_POLL_FUNC_MAX) {
kdb_poll_funcs[kdb_poll_idx] = kdb_get_kbd_char;
kdb_poll_idx++;
+ kgdboc_use_kbd = 1;
if (cptr[0][3] == ',')
*cptr += 4;
else
@@ -63,9 +68,16 @@ static void kgdboc_unregister_kbd(void)
}
}
}
+
+static inline void kgdboc_clear_kbd(void)
+{
+ if (kgdboc_use_kbd)
+ input_dbg_clear_keys(); /* Release all pressed keys */
+}
#else /* ! CONFIG_KDB_KEYBOARD */
#define kgdboc_register_kbd(x) 0
#define kgdboc_unregister_kbd()
+#define kgdboc_clear_kbd()
#endif /* ! CONFIG_KDB_KEYBOARD */
static int kgdboc_option_setup(char *opt)
@@ -213,6 +225,7 @@ static void kgdboc_post_exp_handler(void)
/* decrement the module count when the debugger detaches */
if (!kgdb_connected)
module_put(THIS_MODULE);
+ kgdboc_clear_kbd();
}
static struct kgdb_io kgdboc_io_ops = {
diff --git a/include/linux/input.h b/include/linux/input.h
index 735ceaf..c311871 100644
--- a/include/linux/input.h
+++ b/include/linux/input.h
@@ -1231,6 +1231,9 @@ struct input_handler {
int (*connect)(struct input_handler *handler, struct input_dev *dev, const struct input_device_id *id);
void (*disconnect)(struct input_handle *handle);
void (*start)(struct input_handle *handle);
+#ifdef CONFIG_KDB_KEYBOARD
+ void (*dbg_clear_keys)(void);
+#endif
const struct file_operations *fops;
int minor;
@@ -1316,6 +1319,13 @@ int input_flush_device(struct input_handle* handle, struct file* file);
void input_event(struct input_dev *dev, unsigned int type, unsigned int code, int value);
void input_inject_event(struct input_handle *handle, unsigned int type, unsigned int code, int value);
+#ifdef CONFIG_KDB_KEYBOARD
+void input_dbg_clear_keys(void);
+#else
+static inline void input_dbg_clear_keys(void)
+{}
+#endif
+
static inline void input_report_key(struct input_dev *dev, unsigned int code, int value)
{
input_event(dev, EV_KEY, code, !!value);
--
1.6.4.rc1
Update the kgdb docs to reflect the new directory structure and API.
Merge in the kdb shell information.
Signed-off-by: Jason Wessel <[email protected]>
---
Documentation/DocBook/kgdb.tmpl | 640 ++++++++++++++++++++++++++---------
Documentation/kernel-parameters.txt | 7 +-
include/linux/kgdb.h | 12 +-
lib/Kconfig.kgdb | 2 +-
4 files changed, 495 insertions(+), 166 deletions(-)
diff --git a/Documentation/DocBook/kgdb.tmpl b/Documentation/DocBook/kgdb.tmpl
index 5cff41a..cc81879 100644
--- a/Documentation/DocBook/kgdb.tmpl
+++ b/Documentation/DocBook/kgdb.tmpl
@@ -4,7 +4,7 @@
<book id="kgdbOnLinux">
<bookinfo>
- <title>Using kgdb and the kgdb Internals</title>
+ <title>Using kgdb, kdb and the kernel debugger internals</title>
<authorgroup>
<author>
@@ -17,33 +17,8 @@
</affiliation>
</author>
</authorgroup>
-
- <authorgroup>
- <author>
- <firstname>Tom</firstname>
- <surname>Rini</surname>
- <affiliation>
- <address>
- <email>[email protected]</email>
- </address>
- </affiliation>
- </author>
- </authorgroup>
-
- <authorgroup>
- <author>
- <firstname>Amit S.</firstname>
- <surname>Kale</surname>
- <affiliation>
- <address>
- <email>[email protected]</email>
- </address>
- </affiliation>
- </author>
- </authorgroup>
-
<copyright>
- <year>2008</year>
+ <year>2008,2010</year>
<holder>Wind River Systems, Inc.</holder>
</copyright>
<copyright>
@@ -69,41 +44,76 @@
<chapter id="Introduction">
<title>Introduction</title>
<para>
- kgdb is a source level debugger for linux kernel. It is used along
- with gdb to debug a linux kernel. The expectation is that gdb can
- be used to "break in" to the kernel to inspect memory, variables
- and look through call stack information similar to what an
- application developer would use gdb for. It is possible to place
- breakpoints in kernel code and perform some limited execution
- stepping.
+ The kernel has two different debugger front ends (kdb and kgdb)
+ which interface to the debug core. It is possible to use either
+ of the debugger front ends and dynamically transition between them
+ if you configure the kernel properly at compile and runtime.
+ </para>
+ <para>
+ Kdb is simplistic shell style interface which you can use on a
+ system console with a keyboard or serial console. You can use it
+ to inspect memory, registers, process lists, dmesg, and even set
+ breakpoints to stop in a certain location. Kdb is not a source
+ level debugger, although you can set breakpoints and execute some
+ basic kernel run control. Kdb is mainly aimed at doing some
+ analysis to aid in development or diagnosing kernel problems. You
+ can access some symbols by name in kernel builtins or in kernel
+ modules if the code was built
+ with <symbol>CONFIG_KALLSYMS</symbol>.
</para>
<para>
- Two machines are required for using kgdb. One of these machines is a
- development machine and the other is a test machine. The kernel
- to be debugged runs on the test machine. The development machine
- runs an instance of gdb against the vmlinux file which contains
- the symbols (not boot image such as bzImage, zImage, uImage...).
- In gdb the developer specifies the connection parameters and
- connects to kgdb. The type of connection a developer makes with
- gdb depends on the availability of kgdb I/O modules compiled as
- builtin's or kernel modules in the test machine's kernel.
+ Kgdb was intended to be used as a source level debugger for the
+ linux kernel. It is used along with gdb to debug a linux kernel.
+ The expectation is that gdb can be used to "break in" to the
+ kernel to inspect memory, variables and look through call stack
+ information similar to the way an application developer would use
+ gdb to debug an application. It is possible to place breakpoints
+ in kernel code and perform some limited execution stepping.
+ </para>
+ <para>
+ Two machines are required for using kgdb. One of these machines is
+ a development machine and the other is the target machine. The
+ kernel to be debugged runs on the target machine. The development
+ machine runs an instance of gdb against the vmlinux file which
+ contains the symbols (not boot image such as bzImage, zImage,
+ uImage...). In gdb the developer specifies the connection
+ parameters and connects to kgdb. The type of connection a
+ developer makes with gdb depends on the availability of kgdb I/O
+ modules compiled as builtins or kernel modules in the test
+ machine's kernel.
</para>
</chapter>
<chapter id="CompilingAKernel">
- <title>Compiling a kernel</title>
+ <title>Compiling a kernel</title>
+ <para>
+ <itemizedlist>
+ <listitem><para>In order to enable compilation of kdb, you must first enable kgdb.</para></listitem>
+ <listitem><para>The kgdb test compile options are described in the kgdb test suite chapter.</para></listitem>
+ </itemizedlist>
+ </para>
+ <sect1 id="CompileKGDB">
+ <title>Kernel config options for kgdb</title>
<para>
To enable <symbol>CONFIG_KGDB</symbol> you should first turn on
"Prompt for development and/or incomplete code/drivers"
(CONFIG_EXPERIMENTAL) in "General setup", then under the
- "Kernel debugging" select "KGDB: kernel debugging with remote gdb".
+ "Kernel debugging" select "KGDB: kernel debugger".
+ </para>
+ <para>
+ While it is not a hard requirement that you have symbols in your
+ vmlinux file, gdb tends not to be very useful without the symbolic
+ data, so you will want to turn
+ on <symbol>CONFIG_DEBUG_INFO</symbol> which is called "Compile the
+ kernel with debug info" in the menu config.
</para>
<para>
It is advised, but not required that you turn on the
- CONFIG_FRAME_POINTER kernel option. This option inserts code to
- into the compiled executable which saves the frame information in
- registers or on the stack at different points which will allow a
- debugger such as gdb to more accurately construct stack back traces
- while debugging the kernel.
+ <symbol>CONFIG_FRAME_POINTER</symbol> kernel option which is called "Compile the
+ kernel with frame pointers" in the menu config. This option
+ inserts code to into the compiled executable which saves the frame
+ information in registers or on the stack at different points which
+ will allow a debugger such as gdb to more accurately construct
+ stack back traces while debugging the kernel.
</para>
<para>
If the architecture that you are using supports the kernel option
@@ -116,38 +126,146 @@
this option.
</para>
<para>
- Next you should choose one of more I/O drivers to interconnect debugging
- host and debugged target. Early boot debugging requires a KGDB
- I/O driver that supports early debugging and the driver must be
- built into the kernel directly. Kgdb I/O driver configuration
- takes place via kernel or module parameters, see following
- chapter.
+ Next you should choose one of more I/O drivers to interconnect
+ debugging host and debugged target. Early boot debugging requires
+ a KGDB I/O driver that supports early debugging and the driver
+ must be built into the kernel directly. Kgdb I/O driver
+ configuration takes place via kernel or module parameters which
+ you can learn more about in the in the section that describes the
+ kernel parameter kgdboc.
</para>
- <para>
- The kgdb test compile options are described in the kgdb test suite chapter.
+ <para>Here is an example set of .config to enable kgdb:
+ <itemizedlist>
+ <listitem><para># CONFIG_DEBUG_RODATA is not set</para></listitem>
+ <listitem><para>CONFIG_FRAME_POINTER=y</para></listitem>
+ <listitem><para>CONFIG_KGDB=y</para></listitem>
+ <listitem><para>CONFIG_KGDB_SERIAL_CONSOLE=y</para></listitem>
+ </itemizedlist>
</para>
-
+ </sect1>
+ <sect1 id="CompileKDB">
+ <title>Kernel config options for kdb</title>
+ <para>Kdb is quite a bit more complex than the simple gdbstub
+ sitting on top of the kernel's debug core. Kdb must implement an
+ shell, and also adds some helper functions in other parts of the
+ kernel, responsible for printing out interesting data such as what
+ you would see if you ran "lsmod", or "ps". In order to build kdb
+ into the kernel you follow the same steps as you would for kgdb.
+ </para>
+ <para>The main config option for kdb
+ is <symbol>CONFIG_KGDB_KDB</symbol> which is called "KGDB_KDB:
+ include kdb frontend for kgdb" in the menu config. In theory you
+ would have already also selected an I/O driver such as the
+ CONFIG_KGDB_SERIAL_CONSOLE interface if you plan on using kdb on a
+ serial port, when you were configuring kgdb.
+ </para>
+ <para>If you want to use a ps/2 style keyboard with kdb you would
+ select CONFIG_KDB_KEYBOARD which is called "KGDB_KDB: keyboard as
+ input device" in the menu config. This option really does do
+ anything for the gdb interface to kgdb it only works with kdb,
+ because it is not really meaningful to manually type in gdb
+ remote protocol commands.
+ </para>
+ <para>Here is an example set of .config to enable kdb:
+ <itemizedlist>
+ <listitem><para># CONFIG_DEBUG_RODATA is not set</para></listitem>
+ <listitem><para>CONFIG_FRAME_POINTER=y</para></listitem>
+ <listitem><para>CONFIG_KGDB=y</para></listitem>
+ <listitem><para>CONFIG_KGDB_SERIAL_CONSOLE=y</para></listitem>
+ <listitem><para>CONFIG_KGDB_KDB=y</para></listitem>
+ <listitem><para>CONFIG_KDB_KEYBOARD=y</para></listitem>
+ </itemizedlist>
+ </para>
+ </sect1>
</chapter>
- <chapter id="EnableKGDB">
- <title>Enable kgdb for debugging</title>
- <para>
- In order to use kgdb you must activate it by passing configuration
- information to one of the kgdb I/O drivers. If you do not pass any
- configuration information kgdb will not do anything at all. Kgdb
- will only actively hook up to the kernel trap hooks if a kgdb I/O
- driver is loaded and configured. If you unconfigure a kgdb I/O
- driver, kgdb will unregister all the kernel hook points.
+ <chapter id="kgdbKernelArgs">
+ <title>Kernel Debugger Boot Arguments</title>
+ <para>This section will describe the various runtime kernel
+ parameters that affect the configuration of the kernel debugger.
+ The following chapter will cover using kdb and kgdb as well as
+ provide some examples of the configuration parameters.</para>
+ <sect1 id="kgdboc">
+ <title>Kernel parameter: kgdboc</title>
+ <para>The kgdboc driver was originally an abbreviation meant to
+ stand for "kgdb over console". Today it is the primary mechanism
+ to configure how to communicate from gdb to kgdb as well as the
+ devices you want to use to interact with the kdb shell.
</para>
- <para>
- All drivers can be reconfigured at run time, if
- <symbol>CONFIG_SYSFS</symbol> and <symbol>CONFIG_MODULES</symbol>
- are enabled, by echo'ing a new config string to
- <constant>/sys/module/<driver>/parameter/<option></constant>.
- The driver can be unconfigured by passing an empty string. You cannot
- change the configuration while the debugger is attached. Make sure
- to detach the debugger with the <constant>detach</constant> command
- prior to trying unconfigure a kgdb I/O driver.
+ <para>For kgdb/gdb, kgdboc is designed to work with a single serial
+ port. It was meant to cover the circumstance where you wanted to
+ use a serial console as your primary console as well as using it to
+ perform kernel debugging. It is also possible to use kgdb on a
+ serial port which is not designated as a system console. Kgdboc
+ may be configured as a kernel built-in or a kernel module. You can
+ only make use of <constant>kgdbwait</constant> and early debugging
+ if you build kgdboc into the kernel as a builtins.
+ </para>
+ <sect2 id="kgdbocArgs">
+ <title>kgdboc arguments</title>
+ <para>Usage: <constant>kgdboc=[kbd][[,]serial_device][,baud]</constant></para>
+ <para>You can configure kgdboc to use the keyboard, and or a serial device
+ depending on if you are using kdb and or kgdb, in one of the
+ following scenarios.
+ <orderedlist>
+ <listitem><para>kdb and kgdb over only a serial port</para>
+ <para><constant>kgdboc=<serial_device>[,baud]</constant></para>
+ <para>Example: <constant>kgdboc=ttyS0,115200</constant></para>
+ </listitem>
+ <listitem><para>kdb and kgdb with keyboard and a serial port</para>
+ <para><constant>kgdboc=kbd,<serial_device>[,baud]</constant></para>
+ <para>Example: <constant>kgdboc=kbd,ttyS0,115200</constant></para>
+ </listitem>
+ <listitem><para>kdb with a keyboard</para>
+ <para><constant>kgdboc=kbd</constant></para>
+ </listitem>
+ </orderedlist>
+ </para>
+ <para>You can configure kgdboc via sysfs or a module or kernel boot line
+ parameter depending on if you build with CONFIG_KGDBOC as a module
+ or built-in.
+ <orderedlist>
+ <listitem><para>From the module load or build-in</para>
+ <para>As a kernel argument: <constant>kgdboc=<tty-device>,[baud]</constant></para>
+ <para>As a kernel module: <constant>modprobe kgdboc kgdboc=<tty-device>,[baud]</constant></para>
+ <para>Here are two examples, the first for an x86 target using the
+ first serial port and the second for the ARM Versatile AB using the
+ second serial port.
+ <orderedlist>
+ <listitem><para><constant>kgdboc=ttyS0,115200</constant></para></listitem>
+ <listitem><para><constant>kgdboc=ttyAMA1,115200</constant></para></listitem>
+ </orderedlist>
+ </para>
+ </listitem>
+ <listitem><para>From sysfs</para>
+ <para><constant>echo ttyS0 > /sys/module/kgdboc/parameters/kgdboc</constant></para>
+ <para>NOTE: You do not need to specify the baud if you are
+ configuring the console on tty which is already configured or
+ open.</para>
+ </listitem>
+ </orderedlist>
+ </para>
+ <para>NOTE: Kgdboc does not support interrupting the target via the
+ gdb remote protocol. You must manually send a sysrq-g unless you
+ have a proxy that splits console output to a terminal program.
+ A console proxy has a separate tcp port for the debugger and a separate
+ tcp port for the "human" console. The proxy can take care of sending
+ the sysrq-g for you.
</para>
+ <para>When using kgdboc with no debugger proxy, you can end up
+ connecting the debugger for one of two entry points. If an
+ exception occurs after you have loaded kgdboc a message should
+ print on the console stating it is waiting for the debugger. In
+ case you disconnect your terminal program and then connect the
+ debugger in its place. If you want to interrupt the target system
+ and forcibly enter a debug session you have to issue a Sysrq
+ sequence and then type the letter <constant>g</constant>. Then
+ you disconnect the terminal session and connect gdb. Your options
+ if you don't like this are to hack gdb to send the sysrq-g for you
+ as well as on the initial connect, or to use a debugger proxy that
+ allows an unmodified gdb to do the debugging.
+ </para>
+ </sect2>
+ </sect1>
<sect1 id="kgdbwait">
<title>Kernel parameter: kgdbwait</title>
<para>
@@ -166,99 +284,193 @@
kgdb I/O driver as a kernel module kgdbwait will not do anything.
</para>
</sect1>
- <sect1 id="kgdboc">
- <title>Kernel parameter: kgdboc</title>
- <para>
- The kgdboc driver was originally an abbreviation meant to stand for
- "kgdb over console". Kgdboc is designed to work with a single
- serial port. It was meant to cover the circumstance
- where you wanted to use a serial console as your primary console as
- well as using it to perform kernel debugging. Of course you can
- also use kgdboc without assigning a console to the same port.
+ <sect1 id="kgdbcon">
+ <title>Kernel parameter: kgdbcon</title>
+ <para> The kgdbcon feature allows you to see printk() messages
+ inside gdb while gdb is connected to the kernel. Kdb does not make
+ use of the kgdbcon feature.
+ </para>
+ <para>Kgdb supports using the gdb serial protocol to send console
+ messages to the debugger when the debugger is connected and running.
+ There are two ways to activate this feature.
+ <orderedlist>
+ <listitem><para>Activate with the kernel command line option:</para>
+ <para><constant>kgdbcon</constant></para>
+ </listitem>
+ <listitem><para>Use sysfs before configuring an io driver</para>
+ <para>
+ <constant>echo 1 > /sys/module/kgdb/parameters/kgdb_use_con</constant>
+ </para>
+ <para>
+ NOTE: If you do this after you configure the kgdb I/O driver, the
+ setting will not take effect until the next point the I/O is
+ reconfigured.
+ </para>
+ </listitem>
+ </orderedlist>
+ <para>IMPORTANT NOTE: You cannot use kgdboc + kgdbcon on a tty that is an
+ active system console IE: console=ttyS0,115200 kgdboc=ttyS0 kgdbcon
+ </para>
+ <para>It is possible to use this option with kgdboc on a tty that is not a system console.
+ </para>
</para>
- <sect2 id="UsingKgdboc">
- <title>Using kgdboc</title>
- <para>
- You can configure kgdboc via sysfs or a module or kernel boot line
- parameter depending on if you build with CONFIG_KGDBOC as a module
- or built-in.
- <orderedlist>
- <listitem><para>From the module load or build-in</para>
- <para><constant>kgdboc=<tty-device>,[baud]</constant></para>
+ </sect1>
+ </chapter>
+ <chapter id="usingKDB">
+ <title>Using kdb</title>
<para>
- The example here would be if your console port was typically ttyS0, you would use something like <constant>kgdboc=ttyS0,115200</constant> or on the ARM Versatile AB you would likely use <constant>kgdboc=ttyAMA0,115200</constant>
+ </para>
+ <sect1 id="quickKDBserial">
+ <title>Quick start for kdb on a serial port</title>
+ <para>This is a quick example of how to use kdb</para>
+ <para><orderedlist>
+ <listitem><para>Boot kernel with arguments:
+ <itemizedlist>
+ <listitem><para><constant>console=ttyS0,115200 kgdboc=ttyS0,115200</constant></para></listitem>
+ </itemizedlist></para>
+ <para>OR</para>
+ <para>Configure kgdboc after the kernel booted assuming your are using a serial port console:
+ <itemizedlist>
+ <listitem><para><constant>echo ttyS0 > /sys/module/kgdboc/parameters/kgdboc</constant></para></listitem>
+ </itemizedlist>
</para>
</listitem>
- <listitem><para>From sysfs</para>
- <para><constant>echo ttyS0 > /sys/module/kgdboc/parameters/kgdboc</constant></para>
+ <listitem><para>Enter the kernel debugger manually or by waiting for an opps or fault. There are several ways you can enter the kernel debugger manually, all involve using the sysrq-g, which means you have to enabled CONFIG_MAGIC_SYSRQ=y in your kernel config</para>
+ <itemizedlist>
+ <listitem><para>When logged in as root or with a super user session you can run:</para>
+ <para><constant>echo g > /proc/sysrq-trigger</constant></para></listitem>
+ <listitem><para>Example using minicom 2.2</para>
+ <para>Press: <constant>Control-a</constant></para>
+ <para>Press: <constant>f</constant></para>
+ <para>Press: <constant>g</constant></para>
</listitem>
- </orderedlist>
- </para>
- <para>
- NOTE: Kgdboc does not support interrupting the target via the
- gdb remote protocol. You must manually send a sysrq-g unless you
- have a proxy that splits console output to a terminal problem and
- has a separate port for the debugger to connect to that sends the
- sysrq-g for you.
+ <listitem><para>When you have telneted to a terminal server that supports sending a remote break</para>
+ <para>Press: <constant>Control-]</constant></para>
+ <para>Type in:<constant>send break</constant></para>
+ <para>Press: <constant>Enter</constant></para>
+ <para>Press: <constant>g</constant></para>
+ </listitem>
+ </itemizedlist>
+ </listitem>
+ <listitem><para>From the kdb prompt you can run the "help" command to see a complete list of the commands that are available.</para>
+ <para>Some useful commands in kdb include:
+ <itemizedlist>
+ <listitem><para>lsmod -- Shows where kernel modules are loaded</para></listitem>
+ <listitem><para>ps -- Displays only the active processes</para></listitem>
+ <listitem><para>ps A -- Shows all the processes</para></listitem>
+ <listitem><para>summary -- Shows kernel version info and memory usage</para></listitem>
+ <listitem><para>bt -- Get a backtrace of the current process using dump_stack()</para></listitem>
+ <listitem><para>dmesg -- View the kernel syslog buffer</para></listitem>
+ <listitem><para>go -- Continue the system</para></listitem>
+ </itemizedlist>
</para>
- <para>When using kgdboc with no debugger proxy, you can end up
- connecting the debugger for one of two entry points. If an
- exception occurs after you have loaded kgdboc a message should print
- on the console stating it is waiting for the debugger. In case you
- disconnect your terminal program and then connect the debugger in
- its place. If you want to interrupt the target system and forcibly
- enter a debug session you have to issue a Sysrq sequence and then
- type the letter <constant>g</constant>. Then you disconnect the
- terminal session and connect gdb. Your options if you don't like
- this are to hack gdb to send the sysrq-g for you as well as on the
- initial connect, or to use a debugger proxy that allows an
- unmodified gdb to do the debugging.
+ </listitem>
+ <listitem>
+ <para>When you are done using kdb you need to consider rebooting the system or using the "go" command to resuming normal kernel execution. If you have paused the kernel for a lengthy period of time, applications that rely on timely networking or anything to do with real wall clock time could get adversely affected, so you should take this into consideration when using the kernel debugger.</para>
+ </listitem>
+ </orderedlist></para>
+ </sect1>
+ <sect1 id="quickKDBkeyboard">
+ <title>Quick start for kdb using a keyboard connected console</title>
+ <para>This is a quick example of how to use kdb with a keyboard</para>
+ <para><orderedlist>
+ <listitem><para>Boot kernel with arguments:
+ <itemizedlist>
+ <listitem><para><constant>kgdboc=kbd</constant></para></listitem>
+ </itemizedlist></para>
+ <para>OR</para>
+ <para>Configure kgdboc after the kernel booted:
+ <itemizedlist>
+ <listitem><para><constant>echo kbd > /sys/module/kgdboc/parameters/kgdboc</constant></para></listitem>
+ </itemizedlist>
</para>
- </sect2>
+ </listitem>
+ <listitem><para>Enter the kernel debugger manually or by waiting for an opps or fault. There are several ways you can enter the kernel debugger manually, all involve using the sysrq-g, which means you have to enabled CONFIG_MAGIC_SYSRQ=y in your kernel config</para>
+ <itemizedlist>
+ <listitem><para>When logged in as root or with a super user session you can run:</para>
+ <para><constant>echo g > /proc/sysrq-trigger</constant></para></listitem>
+ <listitem><para>Example using a laptop keyboard</para>
+ <para>Press and hold down: <constant>Alt</constant></para>
+ <para>Press and hold down: <constant>Fn</constant></para>
+ <para>Press and release the key with the label: <constant>SysRq</constant></para>
+ <para>Release: <constant>Fn</constant></para>
+ <para>Press and release: <constant>g</constant></para>
+ <para>Release: <constant>Alt</constant></para>
+ </listitem>
+ <listitem><para>Example using a PS/2 101 keyboard</para>
+ <para>Press and hold down: <constant>Alt</constant></para>
+ <para>Press and release the key with the label: <constant>SysRq</constant></para>
+ <para>Press and release: <constant>g</constant></para>
+ <para>Release: <constant>Alt</constant></para>
+ </listitem>
+ </itemizedlist>
+ </listitem>
+ <listitem>
+ <para>Now type in a kdb command such as "help", "dmesg", "bt" or "go" to continue kernel execution.</para>
+ </listitem>
+ </orderedlist></para>
</sect1>
- <sect1 id="kgdbcon">
- <title>Kernel parameter: kgdbcon</title>
- <para>
- Kgdb supports using the gdb serial protocol to send console messages
- to the debugger when the debugger is connected and running. There
- are two ways to activate this feature.
+ </chapter>
+ <chapter id="EnableKGDB">
+ <title>Using kgdb / gdb</title>
+ <para>In order to use kgdb you must activate it by passing
+ configuration information to one of the kgdb I/O drivers. If you
+ do not pass any configuration information kgdb will not do anything
+ at all. Kgdb will only actively hook up to the kernel trap hooks
+ if a kgdb I/O driver is loaded and configured. If you unconfigure
+ a kgdb I/O driver, kgdb will unregister all the kernel hook points.
+ </para>
+ <para> All kgdb I/O drivers can be reconfigured at run time, if
+ <symbol>CONFIG_SYSFS</symbol> and <symbol>CONFIG_MODULES</symbol>
+ are enabled, by echo'ing a new config string to
+ <constant>/sys/module/<driver>/parameter/<option></constant>.
+ The driver can be unconfigured by passing an empty string. You cannot
+ change the configuration while the debugger is attached. Make sure
+ to detach the debugger with the <constant>detach</constant> command
+ prior to trying unconfigure a kgdb I/O driver.
+ </para>
+ <sect1 id="ConnectingGDB">
+ <title>Connecting with gdb to a serial port</title>
<orderedlist>
- <listitem><para>Activate with the kernel command line option:</para>
- <para><constant>kgdbcon</constant></para>
+ <listitem><para>Configure kgdboc</para>
+ <para>Boot kernel with arguments:
+ <itemizedlist>
+ <listitem><para><constant>kgdboc=ttyS0,115200</constant></para></listitem>
+ </itemizedlist></para>
+ <para>OR</para>
+ <para>Configure kgdboc after the kernel booted:
+ <itemizedlist>
+ <listitem><para><constant>echo ttyS0 > /sys/module/kgdboc/parameters/kgdboc</constant></para></listitem>
+ </itemizedlist></para>
</listitem>
- <listitem><para>Use sysfs before configuring an io driver</para>
- <para>
- <constant>echo 1 > /sys/module/kgdb/parameters/kgdb_use_con</constant>
- </para>
- <para>
- NOTE: If you do this after you configure the kgdb I/O driver, the
- setting will not take effect until the next point the I/O is
- reconfigured.
- </para>
+ <listitem>
+ <para>Stop kernel execution (break into the debugger)</para>
+ <para>In order to connect to gdb to the via kgdboc the kernel must
+ first be stopped. There are several ways to stop the kernel which
+ include using kgdbwait as a boot argument, via a sysrq-g, or running
+ the kernel until it takes an an exception where it waits for the
+ debugger to attach.
+ <itemizedlist>
+ <listitem><para>When logged in as root or with a super user session you can run:</para>
+ <para><constant>echo g > /proc/sysrq-trigger</constant></para></listitem>
+ <listitem><para>Example using minicom 2.2</para>
+ <para>Press: <constant>Control-a</constant></para>
+ <para>Press: <constant>f</constant></para>
+ <para>Press: <constant>g</constant></para>
</listitem>
- </orderedlist>
- </para>
- <para>
- IMPORTANT NOTE: Using this option with kgdb over the console
- (kgdboc) is not supported.
+ <listitem><para>When you have telneted to a terminal server that supports sending a remote break</para>
+ <para>Press: <constant>Control-]</constant></para>
+ <para>Type in:<constant>send break</constant></para>
+ <para>Press: <constant>Enter</constant></para>
+ <para>Press: <constant>g</constant></para>
+ </listitem>
+ </itemizedlist>
</para>
- </sect1>
- </chapter>
- <chapter id="ConnectingGDB">
- <title>Connecting gdb</title>
- <para>
- If you are using kgdboc, you need to have used kgdbwait as a boot
- argument, issued a sysrq-g, or the system you are going to debug
- has already taken an exception and is waiting for the debugger to
- attach before you can connect gdb.
- </para>
- <para>
- If you are not using different kgdb I/O driver other than kgdboc,
- you should be able to connect and the target will automatically
- respond.
- </para>
+ </listitem>
+ <listitem>
+ <para>Connect from from gdb</para>
<para>
- Example (using a serial port):
+ Example (using a directly connected port):
</para>
<programlisting>
% gdb ./vmlinux
@@ -283,6 +495,82 @@
communications. You do this prior to issuing the <constant>target
remote</constant> command by typing in: <constant>set debug remote 1</constant>
</para>
+ </listitem>
+ </orderedlist>
+ <para>Remember if you continue in gdb, and need "break in" again,
+ you need to issue an other sysrq. It is easy to create an simple
+ entry point by putting a breakpoint at <constant>sys_sync</constant>
+ and then you can run "sync" from a shell or script to break into the
+ debugger.</para>
+ </sect1>
+ </chapter>
+ <chapter id="switchKdbKgdb">
+ <title>kgdb and kdb interoperability</title>
+ <para>It is possible to transition between kdb and kgdb dynamically.
+ The debug core will remember which you used the last time and
+ automatically start in the same mode.</para>
+ <sect1>
+ <title>Switching between kdb and kgdb</title>
+ <sect2>
+ <title>Switching from kgdb to kdb</title>
+ <para>
+ There are two ways to switch from kgdb to kdb, you can use gdb to
+ issue a maintenance packet, or you can blindly type the command $3#33.
+ When ever kernel debugger stops in kgdb mode it will print the
+ message <constant>KGDB or $3#33 for KDB</constant>. It is important
+ to note, that you have to type the sequence correctly in one pass.
+ You cannot type a backspace or delete because kgdb will interpret
+ that as part of the debug stream.
+ <orderedlist>
+ <listitem><para>Change from kgdb to kdb by blindly typing:</para>
+ <para><constant>$3#33</constant></para></listitem>
+ <listitem><para>Change from kgdb to kdb with gdb</para>
+ <para><constant>maintenance packet 3</constant></para>
+ <para>NOTE: Now you must kill gdb</para></listitem>
+ </orderedlist>
+ </para>
+ </sect2>
+ <sect2>
+ <title>Change from kdb to kgdb</title>
+ <para>There are two ways you can change from kdb to kgdb. You can
+ manually enter kgdb mode by issuing the kgdb command from the kdb
+ shell prompt, or you can connect gdb while the kdb shell prompt is
+ active. The kdb shell looks for the typical first commands that gdb
+ would issue with the gdb remote protocol and if it sees one of those
+ commands it automatically changes into kgdb mode.</para>
+ <orderedlist>
+ <listitem><para>From kdb issue the command:</para>
+ <para><constant>kgdb</constant></para>
+ <para>Now disconnect your terminal program and connect gdb in its place</para></listitem>
+ <listitem><para>At the kdb prompt, disconnect the terminal program and connect gdb in its place.</para></listitem>
+ </orderedlist>
+ </sect2>
+ </sect1>
+ <sect1>
+ <title>Running kdb commands from gdb</title>
+ <para>It is possible to run a limited set of kdb commands from gdb,
+ using the gdb monitor command. You don't want to execute any of the
+ run control or breakpoint operations, because it can disrupt the
+ state of the kernel debugger. You should be using gdb for
+ breakpoints and run control operations if you have gdb connected.
+ The more useful commands to run are things like lsmod, dmesg, ps or
+ possibly some of the memory information. To see all the kdb
+ commands you can run <constant>monitor help</constant>.</para>
+ <para>Example:
+ <informalexample><programlisting>
+(gdb) monitor ps
+1 idle process (state I) and
+27 sleeping system daemon (state M) processes suppressed,
+use 'ps A' to see all.
+Task Addr Pid Parent [*] cpu State Thread Command
+
+0xc78291d0 1 0 0 0 S 0xc7829404 init
+0xc7954150 942 1 0 0 S 0xc7954384 dropbear
+0xc78789c0 944 1 0 0 S 0xc7878bf4 sh
+(gdb)
+ </programlisting></informalexample>
+ </para>
+ </sect1>
</chapter>
<chapter id="KGDBTestSuite">
<title>kgdb Test Suite</title>
@@ -309,22 +597,22 @@
</para>
</chapter>
<chapter id="CommonBackEndReq">
- <title>KGDB Internals</title>
+ <title>Kernel Debugger Internals</title>
<sect1 id="kgdbArchitecture">
<title>Architecture Specifics</title>
<para>
- Kgdb is organized into three basic components:
+ The kernel debugger is organized into a number of components:
<orderedlist>
- <listitem><para>kgdb core</para>
+ <listitem><para>The debug core</para>
<para>
- The kgdb core is found in kernel/kgdb.c. It contains:
+ The debug core is found in kernel/debugger/debug_core.c. It contains:
<itemizedlist>
- <listitem><para>All the logic to implement the gdb serial protocol</para></listitem>
<listitem><para>A generic OS exception handler which includes sync'ing the processors into a stopped state on an multi cpu system.</para></listitem>
<listitem><para>The API to talk to the kgdb I/O drivers</para></listitem>
<listitem><para>The API to make calls to the arch specific kgdb implementation</para></listitem>
<listitem><para>The logic to perform safe memory reads and writes to memory while using the debugger</para></listitem>
<listitem><para>A full implementation for software breakpoints unless overridden by the arch</para></listitem>
+ <listitem><para>The API to invoke either the kdb or kgdb frontend to the debug core.</para></listitem>
</itemizedlist>
</para>
</listitem>
@@ -347,9 +635,31 @@
</itemizedlist>
</para>
</listitem>
+ <listitem><para>gdbstub frontend (aka kgdb)</para>
+ <para>The gdbstub is located in kernel/debug/gdbstub.c. It contains:</para>
+ <itemizedlist>
+ <listitem><para>All the logic to implement the gdb serial protocol</para></listitem>
+ </itemizedlist>
+ </listitem>
+ <listitem><para>kdb frontend</para>
+ <para>The kdb debugger shell is broken down into a number of
+ components. The kdb core is located in kernel/debug/kdb. There
+ are number of helper functions in some of the other kernel
+ components to make it possible for kdb to examine report
+ information about the kernel without taking locks that could
+ cause a kernel deadlock. The kdb core contains implements the following functionality.</para>
+ <itemizedlist>
+ <listitem><para>A simple shell</para></listitem>
+ <listitem><para>The kdb core command set</para></listitem>
+ <listitem><para>A registration API to register additional kdb shell commands.</para>
+ <para>A good example of a self contained kdb module is the "ftdump" command for dumping the ftrace buffer. See: kernel/trace/trace_kdb.c</para></listitem>
+ <listitem><para>The implementation for kdb_printf() which emit messages directly to I/O drivers, bypassing the kernel log.</para></listitem>
+ <listitem><para>SW / HW breakpoint management for the kdb shell</para></listitem>
+ </itemizedlist>
+ </listitem>
<listitem><para>kgdb I/O driver</para>
<para>
- Each kgdb I/O driver has to provide an implemenation for the following:
+ Each kgdb I/O driver has to provide an implementation for the following:
<itemizedlist>
<listitem><para>configuration via builtin or module</para></listitem>
<listitem><para>dynamic configuration and kgdb hook registration calls</para></listitem>
@@ -453,6 +763,10 @@
<itemizedlist>
<listitem><para>Jason Wessel<email>[email protected]</email></para></listitem>
</itemizedlist>
+ In Jan 2010 this document was updated to include kdb.
+ <itemizedlist>
+ <listitem><para>Jason Wessel<email>[email protected]</email></para></listitem>
+ </itemizedlist>
</para>
</chapter>
</book>
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index f3be1e6..af8b8e8 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -57,6 +57,7 @@ parameter is applicable:
ISAPNP ISA PnP code is enabled.
ISDN Appropriate ISDN support is enabled.
JOY Appropriate joystick support is enabled.
+ KGDB Kernel debugger support is enabled.
KVM Kernel Virtual Machine support is enabled.
LIBATA Libata driver is enabled
LP Printer support is enabled.
@@ -1102,13 +1103,17 @@ and is between 256 and 4096 characters. It is defined in the file
use the HighMem zone if it exists, and the Normal
zone if it does not.
- kgdboc= [HW] kgdb over consoles.
+ kgdboc= [KGDB,HW] kgdb over consoles.
Requires a tty driver that supports console polling,
or a supported polling keyboard driver (non-usb).
Serial only format: <serial_device>[,baud]
keyboard only format: kbd
keyboard and serial format: kbd,<serial_device>[,baud]
+ kgdbwait [KGDB] Stop kernel execution and enter the
+ kernel debugger at the earliest opportunity if
+ kgdb was configured on as a boot argument.
+
kmac= [MIPS] korina ethernet MAC address.
Configure the RouterBoard 532 series on-chip
Ethernet adapter MAC address.
diff --git a/include/linux/kgdb.h b/include/linux/kgdb.h
index 5b37df0..407edb1 100644
--- a/include/linux/kgdb.h
+++ b/include/linux/kgdb.h
@@ -204,6 +204,17 @@ kgdb_arch_handle_exception(int vector, int signo, int err_code,
*/
extern void kgdb_roundup_cpus(unsigned long flags);
+/**
+ * kgdb_arch_set_pc - Generic call back to the program counter
+ * @regs: Current &struct pt_regs.
+ * @pc: The new value for the program counter
+ *
+ * This function handles updating the program counter and requires an
+ * architecture specific implementation.
+ */
+extern void kgdb_arch_set_pc(struct pt_regs *regs, unsigned long pc);
+
+
/* Optional functions. */
extern int kgdb_validate_break_address(unsigned long addr);
extern int kgdb_arch_set_breakpoint(unsigned long addr, char *saved_instr);
@@ -264,7 +275,6 @@ extern struct kgdb_arch arch_kgdb_ops;
extern unsigned long __weak kgdb_arch_pc(int exception, struct pt_regs *regs);
-extern void kgdb_arch_set_pc(struct pt_regs *regs, unsigned long pc);
extern int kgdb_register_io_module(struct kgdb_io *local_kgdb_io_ops);
extern void kgdb_unregister_io_module(struct kgdb_io *local_kgdb_io_ops);
extern struct kgdb_io *dbg_io_ops;
diff --git a/lib/Kconfig.kgdb b/lib/Kconfig.kgdb
index 0fb633c..c80de7d 100644
--- a/lib/Kconfig.kgdb
+++ b/lib/Kconfig.kgdb
@@ -3,7 +3,7 @@ config HAVE_ARCH_KGDB
bool
menuconfig KGDB
- bool "KGDB: kernel debugging with remote gdb"
+ bool "KGDB: kernel debugger"
depends on HAVE_ARCH_KGDB
depends on DEBUG_KERNEL && EXPERIMENTAL
help
--
1.6.4.rc1
The only way the debugger can handle a trap in inside rcu_lock,
notify_die, or atomic_notifier_call_chain without a recursive fault is
to have a low level "first opportunity handler" low level
program_check_exception() handler.
The other change here is to make sure that kgdb_handle_exception() is
called with correct parameters when catching an oops, because kdb
needs to know if the entry was an oops, single step, or breakpoint
exception.
CC: Benjamin Herrenschmidt <[email protected]>
CC: Paul Mackerras <[email protected]>
Signed-off-by: Jason Wessel <[email protected]>
---
arch/powerpc/kernel/kgdb.c | 7 +++++--
arch/powerpc/kernel/traps.c | 7 +++++++
lib/Kconfig.kgdb | 2 +-
3 files changed, 13 insertions(+), 3 deletions(-)
diff --git a/arch/powerpc/kernel/kgdb.c b/arch/powerpc/kernel/kgdb.c
index 7aafcc5..c06fe55 100644
--- a/arch/powerpc/kernel/kgdb.c
+++ b/arch/powerpc/kernel/kgdb.c
@@ -20,6 +20,7 @@
#include <linux/smp.h>
#include <linux/signal.h>
#include <linux/ptrace.h>
+#include <linux/kdebug.h>
#include <asm/current.h>
#include <asm/processor.h>
#include <asm/machdep.h>
@@ -115,7 +116,9 @@ void kgdb_roundup_cpus(unsigned long flags)
/* KGDB functions to use existing PowerPC64 hooks. */
static int kgdb_debugger(struct pt_regs *regs)
{
- return kgdb_handle_exception(0, computeSignal(TRAP(regs)), 0, regs);
+ if (kgdb_handle_exception(1, computeSignal(TRAP(regs)), DIE_OOPS, regs))
+ return 0;
+ return 1;
}
static int kgdb_handle_breakpoint(struct pt_regs *regs)
@@ -123,7 +126,7 @@ static int kgdb_handle_breakpoint(struct pt_regs *regs)
if (user_mode(regs))
return 0;
- if (kgdb_handle_exception(0, SIGTRAP, 0, regs) != 0)
+ if (kgdb_handle_exception(1, SIGTRAP, 0, regs) != 0)
return 0;
if (*(u32 *) (regs->nip) == *(u32 *) (&arch_kgdb_ops.gdb_bpt_instr))
diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c
index d069ff8..379104a 100644
--- a/arch/powerpc/kernel/traps.c
+++ b/arch/powerpc/kernel/traps.c
@@ -809,12 +809,19 @@ void __kprobes program_check_exception(struct pt_regs *regs)
return;
}
if (reason & REASON_TRAP) {
+
+#ifdef CONFIG_KGDB_LOW_LEVEL_TRAP
+ if (debugger_bpt(regs))
+ return;
+#endif /* CONFIG_KGDB_LOW_LEVEL_TRAP */
/* trap exception */
if (notify_die(DIE_BPT, "breakpoint", regs, 5, 5, SIGTRAP)
== NOTIFY_STOP)
return;
+#ifndef CONFIG_KGDB_LOW_LEVEL_TRAP
if (debugger_bpt(regs))
return;
+#endif /* ! CONFIG_KGDB_LOW_LEVEL_TRAP */
if (!(regs->msr & MSR_PR) && /* not user-mode */
report_bug(regs->nip, regs) == BUG_TRAP_TYPE_WARN) {
diff --git a/lib/Kconfig.kgdb b/lib/Kconfig.kgdb
index 0a0e049..2b1601b 100644
--- a/lib/Kconfig.kgdb
+++ b/lib/Kconfig.kgdb
@@ -59,7 +59,7 @@ config KGDB_TESTS_BOOT_STRING
config KGDB_LOW_LEVEL_TRAP
bool "KGDB: Allow debugging with traps in notifiers"
- depends on X86
+ depends on X86 || PPC
default n
help
This will add an extra call back to kgdb for the breakpoint
--
1.6.4.rc1
Some kgdb I/O modules require the ability to create a breakpoint
tasklet, such as kgdboc and external modules such as kgdboe. The
breakpoint tasklet is used as an asynchronous entry point into the
debugger which will have a different function scope than the current
execution path where it might not be safe to have an inline
breakpoint. This is true of some of the kgdb I/O drivers which share
code with kgdb and rest of the kernel users.
Signed-off-by: Jason Wessel <[email protected]>
---
include/linux/kgdb.h | 1 +
kernel/debug/debug_core.c | 26 ++++++++++++++++++++++++++
2 files changed, 27 insertions(+), 0 deletions(-)
diff --git a/include/linux/kgdb.h b/include/linux/kgdb.h
index 19d1b29..ee007ea 100644
--- a/include/linux/kgdb.h
+++ b/include/linux/kgdb.h
@@ -271,6 +271,7 @@ extern int kgdb_mem2hex(char *mem, char *buf, int count);
extern int kgdb_hex2mem(char *buf, char *mem, int count);
extern int kgdb_isremovedbreak(unsigned long addr);
+extern void kgdb_schedule_breakpoint(void);
extern int
kgdb_handle_exception(int ex_vector, int signo, int err_code,
diff --git a/kernel/debug/debug_core.c b/kernel/debug/debug_core.c
index 8db8e40..ef46af3 100644
--- a/kernel/debug/debug_core.c
+++ b/kernel/debug/debug_core.c
@@ -114,6 +114,7 @@ EXPORT_SYMBOL_GPL(kgdb_active);
*/
static atomic_t passive_cpu_wait[NR_CPUS];
static atomic_t cpu_in_kgdb[NR_CPUS];
+static atomic_t kgdb_break_tasklet_var;
atomic_t kgdb_setting_breakpoint;
struct task_struct *kgdb_usethread;
@@ -781,6 +782,31 @@ static void kgdb_unregister_callbacks(void)
}
}
+/*
+ * There are times a tasklet needs to be used vs a compiled in
+ * break point so as to cause an exception outside a kgdb I/O module,
+ * such as is the case with kgdboe, where calling a breakpoint in the
+ * I/O driver itself would be fatal.
+ */
+static void kgdb_tasklet_bpt(unsigned long ing)
+{
+ kgdb_breakpoint();
+ atomic_set(&kgdb_break_tasklet_var, 0);
+}
+
+static DECLARE_TASKLET(kgdb_tasklet_breakpoint, kgdb_tasklet_bpt, 0);
+
+void kgdb_schedule_breakpoint(void)
+{
+ if (atomic_read(&kgdb_break_tasklet_var) ||
+ atomic_read(&kgdb_active) != -1 ||
+ atomic_read(&kgdb_setting_breakpoint))
+ return;
+ atomic_inc(&kgdb_break_tasklet_var);
+ tasklet_schedule(&kgdb_tasklet_breakpoint);
+}
+EXPORT_SYMBOL_GPL(kgdb_schedule_breakpoint);
+
static void kgdb_initial_breakpoint(void)
{
kgdb_break_asap = 0;
--
1.6.4.rc1
This allows kdb to debug a crash with in the kms code with a
single level recursive re-entry.
Signed-off-by: Jason Wessel <[email protected]>
---
kernel/debug/debug_core.c | 7 +++++++
1 files changed, 7 insertions(+), 0 deletions(-)
diff --git a/kernel/debug/debug_core.c b/kernel/debug/debug_core.c
index ef46af3..bfdba97 100644
--- a/kernel/debug/debug_core.c
+++ b/kernel/debug/debug_core.c
@@ -445,6 +445,10 @@ static int kgdb_reenter_check(struct kgdb_state *ks)
}
printk(KERN_CRIT "KGDB: re-enter exception: ALL breakpoints killed\n");
+#ifdef CONFIG_KGDB_KDB
+ /* Allow kdb to debug itself one level */
+ return 0;
+#endif
dump_stack();
panic("Recursive entry to debugger");
@@ -489,6 +493,9 @@ acquirelock:
smp_wmb();
atomic_set(&cpu_in_kgdb[cpu], 1);
+ if (exception_level == 1)
+ goto cpu_master_loop;
+
/*
* CPU will loop if it is a slave or request to become a kgdb
* master cpu and acquire the kgdb_active lock:
--
1.6.4.rc1
If kdb is open on a serial port that is not actually a console make
sure to call the poll routines to emit and receive characters.
Signed-off-by: Jason Wessel <[email protected]>
---
drivers/serial/kgdboc.c | 14 ++++++++++++++
include/linux/kgdb.h | 3 +++
kernel/debug/kdb/kdb_io.c | 16 ++++++++++++++++
3 files changed, 33 insertions(+), 0 deletions(-)
diff --git a/drivers/serial/kgdboc.c b/drivers/serial/kgdboc.c
index ecef6e1..b765ab4 100644
--- a/drivers/serial/kgdboc.c
+++ b/drivers/serial/kgdboc.c
@@ -16,6 +16,7 @@
#include <linux/kgdb.h>
#include <linux/kdb.h>
#include <linux/tty.h>
+#include <linux/console.h>
#define MAX_CONFIG_LEN 40
@@ -93,12 +94,14 @@ static int configure_kgdboc(void)
int tty_line = 0;
int err;
char *cptr = config;
+ struct console *cons;
err = kgdboc_option_setup(config);
if (err || !strlen(config) || isspace(config[0]))
goto noconfig;
err = -ENODEV;
+ kgdboc_io_ops.is_console = 0;
kgdb_tty_driver = NULL;
if (kgdboc_register_kbd(&cptr))
@@ -108,6 +111,17 @@ static int configure_kgdboc(void)
if (!p)
goto noconfig;
+ cons = console_drivers;
+ while (cons) {
+ int idx;
+ if (cons->device && cons->device(cons, &idx) == p &&
+ idx == tty_line) {
+ kgdboc_io_ops.is_console = 1;
+ break;
+ }
+ cons = cons->next;
+ }
+
kgdb_tty_driver = p;
kgdb_tty_line = tty_line;
diff --git a/include/linux/kgdb.h b/include/linux/kgdb.h
index ee007ea..6c784ab 100644
--- a/include/linux/kgdb.h
+++ b/include/linux/kgdb.h
@@ -247,6 +247,8 @@ struct kgdb_arch {
* the I/O driver.
* @post_exception: Pointer to a function that will do any cleanup work
* for the I/O driver.
+ * @is_console: 1 if the end device is a console 0 if the I/O device is
+ * not a console
*/
struct kgdb_io {
const char *name;
@@ -256,6 +258,7 @@ struct kgdb_io {
int (*init) (void);
void (*pre_exception) (void);
void (*post_exception) (void);
+ int is_console;
};
extern struct kgdb_arch arch_kgdb_ops;
diff --git a/kernel/debug/kdb/kdb_io.c b/kernel/debug/kdb/kdb_io.c
index 8339b29..58be7e9 100644
--- a/kernel/debug/kdb/kdb_io.c
+++ b/kernel/debug/kdb/kdb_io.c
@@ -673,6 +673,14 @@ kdb_printit:
if (!dbg_kdb_mode && kgdb_connected) {
gdbstub_msg_write(kdb_buffer, retlen);
} else {
+ if (!dbg_io_ops->is_console) {
+ len = strlen(kdb_buffer);
+ cp = kdb_buffer;
+ while (len--) {
+ dbg_io_ops->write_char(*cp);
+ cp++;
+ }
+ }
while (c) {
c->write(c, kdb_buffer, retlen);
touch_nmi_watchdog();
@@ -719,6 +727,14 @@ kdb_printit:
kdb_input_flush();
c = console_drivers;
+ if (!dbg_io_ops->is_console) {
+ len = strlen(moreprompt);
+ cp = moreprompt;
+ while (len--) {
+ dbg_io_ops->write_char(*cp);
+ cp++;
+ }
+ }
while (c) {
c->write(c, moreprompt, strlen(moreprompt));
touch_nmi_watchdog();
--
1.6.4.rc1
The only way the debugger can handle a trap in inside rcu_lock,
notify_die, or atomic_notifier_call_chain without a recursive fault is
to have a low level "first opportunity handler" do_trap_or_bp() handler.
Generally this will be something the vast majority of folks will not
need, but for those who need it, it is added as a kernel .config
option called KGDB_LOW_LEVEL_TRAP.
Also added was a die notification for oops such that kdb can catch an
oops for analysis.
There appeared to be no obvious way to pass the struct pt_regs from
the original exception back to the stack back tracer, so a special
case was added to show_stack() for when kdb is active because you
generally desire to generally look at the back trace of the original
exception.
CC: Ralf Baechle <[email protected]>
Signed-off-by: Jason Wessel <[email protected]>
---
arch/mips/include/asm/kgdb.h | 2 ++
arch/mips/kernel/kgdb.c | 22 +++++++++++++++++++++-
arch/mips/kernel/traps.c | 13 +++++++++++++
include/linux/kdb.h | 3 +++
kernel/debug/kdb/kdb_private.h | 1 -
lib/Kconfig.kgdb | 2 +-
6 files changed, 40 insertions(+), 3 deletions(-)
diff --git a/arch/mips/include/asm/kgdb.h b/arch/mips/include/asm/kgdb.h
index 48223b0..19002d6 100644
--- a/arch/mips/include/asm/kgdb.h
+++ b/arch/mips/include/asm/kgdb.h
@@ -38,6 +38,8 @@ extern int kgdb_early_setup;
extern void *saved_vectors[32];
extern void handle_exception(struct pt_regs *regs);
extern void breakinst(void);
+extern int kgdb_ll_trap(int cmd, const char *str,
+ struct pt_regs *regs, long err, int trap, int sig);
#endif /* __KERNEL__ */
diff --git a/arch/mips/kernel/kgdb.c b/arch/mips/kernel/kgdb.c
index 6ed4c83..9b78ff6 100644
--- a/arch/mips/kernel/kgdb.c
+++ b/arch/mips/kernel/kgdb.c
@@ -203,7 +203,7 @@ static int kgdb_mips_notify(struct notifier_block *self, unsigned long cmd,
if (atomic_read(&kgdb_active) != -1)
kgdb_nmicallback(smp_processor_id(), regs);
- if (kgdb_handle_exception(trap, compute_signal(trap), 0, regs))
+ if (kgdb_handle_exception(trap, compute_signal(trap), cmd, regs))
return NOTIFY_DONE;
if (atomic_read(&kgdb_setting_breakpoint))
@@ -217,6 +217,26 @@ static int kgdb_mips_notify(struct notifier_block *self, unsigned long cmd,
return NOTIFY_STOP;
}
+#ifdef CONFIG_KGDB_LOW_LEVEL_TRAP
+int kgdb_ll_trap(int cmd, const char *str,
+ struct pt_regs *regs, long err, int trap, int sig)
+{
+ struct die_args args = {
+ .regs = regs,
+ .str = str,
+ .err = err,
+ .trapnr = trap,
+ .signr = sig,
+
+ };
+
+ if (!kgdb_io_module_registered)
+ return NOTIFY_DONE;
+
+ return kgdb_mips_notify(NULL, cmd, &args);
+}
+#endif /* CONFIG_KGDB_LOW_LEVEL_TRAP */
+
static struct notifier_block kgdb_notifier = {
.notifier_call = kgdb_mips_notify,
};
diff --git a/arch/mips/kernel/traps.c b/arch/mips/kernel/traps.c
index 31b204b..241b0d1 100644
--- a/arch/mips/kernel/traps.c
+++ b/arch/mips/kernel/traps.c
@@ -26,6 +26,7 @@
#include <linux/kgdb.h>
#include <linux/kdebug.h>
#include <linux/notifier.h>
+#include <linux/kdb.h>
#include <asm/bootinfo.h>
#include <asm/branch.h>
@@ -184,6 +185,11 @@ void show_stack(struct task_struct *task, unsigned long *sp)
regs.regs[29] = task->thread.reg29;
regs.regs[31] = 0;
regs.cp0_epc = task->thread.reg31;
+#ifdef CONFIG_KGDB_KDB
+ } else if (atomic_read(&kgdb_active) != -1 &&
+ kdb_current_regs) {
+ memcpy(®s, kdb_current_regs, sizeof(regs));
+#endif /* CONFIG_KGDB_KDB */
} else {
prepare_frametrace(®s);
}
@@ -358,6 +364,8 @@ void __noreturn die(const char * str, const struct pt_regs * regs)
unsigned long dvpret = dvpe();
#endif /* CONFIG_MIPS_MT_SMTC */
+ notify_die(DIE_OOPS, str, (struct pt_regs *)regs, SIGSEGV, 0, 0);
+
console_verbose();
spin_lock_irq(&die_lock);
bust_spinlocks(1);
@@ -698,6 +706,11 @@ static void do_trap_or_bp(struct pt_regs *regs, unsigned int code,
siginfo_t info;
char b[40];
+#ifdef CONFIG_KGDB_LOW_LEVEL_TRAP
+ if (kgdb_ll_trap(DIE_TRAP, str, regs, code, 0, 0) == NOTIFY_STOP)
+ return;
+#endif /* CONFIG_KGDB_LOW_LEVEL_TRAP */
+
if (notify_die(DIE_TRAP, str, regs, code, 0, 0) == NOTIFY_STOP)
return;
diff --git a/include/linux/kdb.h b/include/linux/kdb.h
index f44aa5f..d72fa39 100644
--- a/include/linux/kdb.h
+++ b/include/linux/kdb.h
@@ -99,6 +99,9 @@ int kdb_process_cpu(const struct task_struct *p)
return cpu;
}
+/* kdb access to register set for stack dumping */
+extern struct pt_regs *kdb_current_regs;
+
#else /* ! CONFIG_KGDB_KDB */
#define kdb_printf(...)
#define kdb_init(x)
diff --git a/kernel/debug/kdb/kdb_private.h b/kernel/debug/kdb/kdb_private.h
index 42af29b..546066e 100644
--- a/kernel/debug/kdb/kdb_private.h
+++ b/kernel/debug/kdb/kdb_private.h
@@ -288,7 +288,6 @@ extern void debug_kusage(void);
extern void kdb_set_current_task(struct task_struct *);
extern struct task_struct *kdb_current_task;
-extern struct pt_regs *kdb_current_regs;
#ifdef CONFIG_MODULES
extern struct list_head *kdb_modules;
#endif /* CONFIG_MODULES */
diff --git a/lib/Kconfig.kgdb b/lib/Kconfig.kgdb
index 2b1601b..14b61f4 100644
--- a/lib/Kconfig.kgdb
+++ b/lib/Kconfig.kgdb
@@ -59,7 +59,7 @@ config KGDB_TESTS_BOOT_STRING
config KGDB_LOW_LEVEL_TRAP
bool "KGDB: Allow debugging with traps in notifiers"
- depends on X86 || PPC
+ depends on X86 || PPC || MIPS
default n
help
This will add an extra call back to kgdb for the breakpoint
--
1.6.4.rc1
This patch adds in the kdb PS/2 keyboard driver. This was mostly a
direct port from the original kdb where I cleaned up the code against
checkpatch.pl and added the glue to stitch it into kgdb.
This patch also enables early kdb debug via kgdbwait and the keyboard.
All the access to configure kdb using either a serial console or the
keyboard is done via kgdboc.
If you want to use only the keyboard and want to break in early you
would add to your kernel command arguments:
kgdboc=kbd kgdbwait
If you wanted serial and or the keyboard access you could use:
kgdboc=kbd,ttyS0
You can also configure kgdboc as a kernel module or at run time with
the sysfs where you can activate and deactivate kgdb.
Turn it on:
echo kbd,ttyS0 > /sys/module/kgdboc/parameters/kgdboc
Turn it off:
echo "" > /sys/module/kgdboc/parameters/kgdboc
CC: Dmitry Torokhov <[email protected]>
Signed-off-by: Jason Wessel <[email protected]>
---
Documentation/kernel-parameters.txt | 8 +-
drivers/char/Makefile | 1 +
drivers/char/kdb_keyboard.c | 204 +++++++++++++++++++++++++++++++++++
drivers/char/kdb_keyboard.h | 143 ++++++++++++++++++++++++
drivers/serial/kgdboc.c | 61 +++++++++--
lib/Kconfig.kgdb | 7 ++
6 files changed, 414 insertions(+), 10 deletions(-)
create mode 100644 drivers/char/kdb_keyboard.c
create mode 100644 drivers/char/kdb_keyboard.h
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 736d456..f3be1e6 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1103,9 +1103,11 @@ and is between 256 and 4096 characters. It is defined in the file
zone if it does not.
kgdboc= [HW] kgdb over consoles.
- Requires a tty driver that supports console polling.
- (only serial supported for now)
- Format: <serial_device>[,baud]
+ Requires a tty driver that supports console polling,
+ or a supported polling keyboard driver (non-usb).
+ Serial only format: <serial_device>[,baud]
+ keyboard only format: kbd
+ keyboard and serial format: kbd,<serial_device>[,baud]
kmac= [MIPS] korina ethernet MAC address.
Configure the RouterBoard 532 series on-chip
diff --git a/drivers/char/Makefile b/drivers/char/Makefile
index f957edf..dbc579e 100644
--- a/drivers/char/Makefile
+++ b/drivers/char/Makefile
@@ -13,6 +13,7 @@ obj-$(CONFIG_LEGACY_PTYS) += pty.o
obj-$(CONFIG_UNIX98_PTYS) += pty.o
obj-y += misc.o
obj-$(CONFIG_VT) += vt_ioctl.o vc_screen.o selection.o keyboard.o
+obj-$(CONFIG_KDB_KEYBOARD) += kdb_keyboard.o
obj-$(CONFIG_BFIN_JTAG_COMM) += bfin_jtag_comm.o
obj-$(CONFIG_CONSOLE_TRANSLATIONS) += consolemap.o consolemap_deftbl.o
obj-$(CONFIG_HW_CONSOLE) += vt.o defkeymap.o
diff --git a/drivers/char/kdb_keyboard.c b/drivers/char/kdb_keyboard.c
new file mode 100644
index 0000000..95aa102
--- /dev/null
+++ b/drivers/char/kdb_keyboard.c
@@ -0,0 +1,204 @@
+/*
+ * Kernel Debugger Architecture Dependent Console I/O handler
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.
+ *
+ * Copyright (c) 1999-2006 Silicon Graphics, Inc. All Rights Reserved.
+ * Copyright (c) 2009 Wind River Systems, Inc. All Rights Reserved.
+ */
+
+#include <linux/kdb.h>
+#include <linux/keyboard.h>
+#include <linux/ctype.h>
+#include <linux/module.h>
+#include <linux/io.h>
+#include "kdb_keyboard.h"
+
+
+static int kbd_exists;
+
+/*
+ * Check if the keyboard controller has a keypress for us.
+ * Some parts (Enter Release, LED change) are still blocking polled here,
+ * but hopefully they are all short.
+ */
+int kdb_get_kbd_char(void)
+{
+ int scancode, scanstatus;
+ static int shift_lock; /* CAPS LOCK state (0-off, 1-on) */
+ static int shift_key; /* Shift next keypress */
+ static int ctrl_key;
+ u_short keychar;
+
+ if (KDB_FLAG(NO_I8042) || KDB_FLAG(NO_VT_CONSOLE) ||
+ (inb(KBD_STATUS_REG) == 0xff && inb(KBD_DATA_REG) == 0xff)) {
+ kbd_exists = 0;
+ return -1;
+ }
+ kbd_exists = 1;
+
+ if ((inb(KBD_STATUS_REG) & KBD_STAT_OBF) == 0)
+ return -1;
+
+ /*
+ * Fetch the scancode
+ */
+ scancode = inb(KBD_DATA_REG);
+ scanstatus = inb(KBD_STATUS_REG);
+
+ /*
+ * Ignore mouse events.
+ */
+ if (scanstatus & KBD_STAT_MOUSE_OBF)
+ return -1;
+
+ /*
+ * Ignore release, trigger on make
+ * (except for shift keys, where we want to
+ * keep the shift state so long as the key is
+ * held down).
+ */
+
+ if (((scancode&0x7f) == 0x2a) || ((scancode&0x7f) == 0x36)) {
+ /*
+ * Next key may use shift table
+ */
+ if ((scancode & 0x80) == 0)
+ shift_key = 1;
+ else
+ shift_key = 0;
+ return -1;
+ }
+
+ if ((scancode&0x7f) == 0x1d) {
+ /*
+ * Left ctrl key
+ */
+ if ((scancode & 0x80) == 0)
+ ctrl_key = 1;
+ else
+ ctrl_key = 0;
+ return -1;
+ }
+
+ if ((scancode & 0x80) != 0)
+ return -1;
+
+ scancode &= 0x7f;
+
+ /*
+ * Translate scancode
+ */
+
+ if (scancode == 0x3a) {
+ /*
+ * Toggle caps lock
+ */
+ shift_lock ^= 1;
+
+#ifdef KDB_BLINK_LED
+ kdb_toggleled(0x4);
+#endif
+ return -1;
+ }
+
+ if (scancode == 0x0e) {
+ /*
+ * Backspace
+ */
+ return 8;
+ }
+
+ /* Special Key */
+ switch (scancode) {
+ case 0xF: /* Tab */
+ return 9;
+ case 0x53: /* Del */
+ return 4;
+ case 0x47: /* Home */
+ return 1;
+ case 0x4F: /* End */
+ return 5;
+ case 0x4B: /* Left */
+ return 2;
+ case 0x48: /* Up */
+ return 16;
+ case 0x50: /* Down */
+ return 14;
+ case 0x4D: /* Right */
+ return 6;
+ }
+
+ if (scancode == 0xe0)
+ return -1;
+
+ /*
+ * For Japanese 86/106 keyboards
+ * See comment in drivers/char/pc_keyb.c.
+ * - Masahiro Adegawa
+ */
+ if (scancode == 0x73)
+ scancode = 0x59;
+ else if (scancode == 0x7d)
+ scancode = 0x7c;
+
+ if (!shift_lock && !shift_key && !ctrl_key) {
+ keychar = plain_map[scancode];
+ } else if ((shift_lock || shift_key) && key_maps[1]) {
+ keychar = key_maps[1][scancode];
+ } else if (ctrl_key && key_maps[4]) {
+ keychar = key_maps[4][scancode];
+ } else {
+ keychar = 0x0020;
+ kdb_printf("Unknown state/scancode (%d)\n", scancode);
+ }
+ keychar &= 0x0fff;
+ if (keychar == '\t')
+ keychar = ' ';
+ switch (KTYP(keychar)) {
+ case KT_LETTER:
+ case KT_LATIN:
+ if (isprint(keychar))
+ break; /* printable characters */
+ /* drop through */
+ case KT_SPEC:
+ if (keychar == K_ENTER)
+ break;
+ /* drop through */
+ default:
+ return -1; /* ignore unprintables */
+ }
+
+ if ((scancode & 0x7f) == 0x1c) {
+ /*
+ * enter key. All done. Absorb the release scancode.
+ */
+ while ((inb(KBD_STATUS_REG) & KBD_STAT_OBF) == 0)
+ ;
+
+ /*
+ * Fetch the scancode
+ */
+ scancode = inb(KBD_DATA_REG);
+ scanstatus = inb(KBD_STATUS_REG);
+
+ while (scanstatus & KBD_STAT_MOUSE_OBF) {
+ scancode = inb(KBD_DATA_REG);
+ scanstatus = inb(KBD_STATUS_REG);
+ }
+
+ if (scancode != 0x9c) {
+ /*
+ * Wasn't an enter-release, why not?
+ */
+ kdb_printf("kdb: expected enter got 0x%x status 0x%x\n",
+ scancode, scanstatus);
+ }
+
+ return 13;
+ }
+
+ return keychar & 0xff;
+}
+EXPORT_SYMBOL_GPL(kdb_get_kbd_char);
diff --git a/drivers/char/kdb_keyboard.h b/drivers/char/kdb_keyboard.h
new file mode 100644
index 0000000..5541668
--- /dev/null
+++ b/drivers/char/kdb_keyboard.h
@@ -0,0 +1,143 @@
+/*
+ * include/linux/pc_keyb.h
+ *
+ * PC Keyboard And Keyboard Controller
+ *
+ * (c) 1997 Martin Mares <[email protected]>
+ */
+
+/*
+ * Configuration Switches
+ */
+
+#undef KBD_REPORT_ERR /* Report keyboard errors */
+#define KBD_REPORT_UNKN /* Report unknown scan codes */
+#define KBD_REPORT_TIMEOUTS /* Report keyboard timeouts */
+#undef KBD_IS_FOCUS_9000 /* We have the brain-damaged
+ * FOCUS-9000 keyboard */
+#undef INITIALIZE_MOUSE /* Define if your PS/2 mouse
+ * needs initialization. */
+
+#define KBD_INIT_TIMEOUT 1000 /* Timeout in ms for initializing the
+ * keyboard */
+#define KBC_TIMEOUT 250 /* Timeout in ms for sending
+ * to keyboard controller */
+#define KBD_TIMEOUT 1000 /* Timeout in ms for keyboard
+ * command acknowledge */
+
+/*
+ * Internal variables of the driver
+ */
+
+extern unsigned char pckbd_read_mask;
+extern unsigned char aux_device_present;
+
+/*
+ * Keyboard Controller Registers on normal PCs.
+ */
+
+#define KBD_STATUS_REG 0x64 /* Status register (R) */
+#define KBD_CNTL_REG 0x64 /* Controller command register (W) */
+#define KBD_DATA_REG 0x60 /* Keyboard data register (R/W) */
+
+/*
+ * Keyboard Controller Commands
+ */
+
+#define KBD_CCMD_READ_MODE 0x20 /* Read mode bits */
+#define KBD_CCMD_WRITE_MODE 0x60 /* Write mode bits */
+#define KBD_CCMD_GET_VERSION 0xA1 /* Get controller version */
+#define KBD_CCMD_MOUSE_DISABLE 0xA7 /* Disable mouse interface */
+#define KBD_CCMD_MOUSE_ENABLE 0xA8 /* Enable mouse interface */
+#define KBD_CCMD_TEST_MOUSE 0xA9 /* Mouse interface test */
+#define KBD_CCMD_SELF_TEST 0xAA /* Controller self test */
+#define KBD_CCMD_KBD_TEST 0xAB /* Keyboard interface test */
+#define KBD_CCMD_KBD_DISABLE 0xAD /* Keyboard interface disable */
+#define KBD_CCMD_KBD_ENABLE 0xAE /* Keyboard interface enable */
+#define KBD_CCMD_WRITE_AUX_OBUF 0xD3 /* Write to output buffer as if
+ initiated by the auxiliary device */
+#define KBD_CCMD_WRITE_MOUSE 0xD4 /* Write the following byte to
+ * the mouse */
+
+/*
+ * Keyboard Commands
+ */
+
+#define KBD_CMD_SET_LEDS 0xED /* Set keyboard leds */
+#define KBD_CMD_SET_RATE 0xF3 /* Set typematic rate */
+#define KBD_CMD_ENABLE 0xF4 /* Enable scanning */
+#define KBD_CMD_DISABLE 0xF5 /* Disable scanning */
+#define KBD_CMD_RESET 0xFF /* Reset */
+
+/*
+ * Keyboard Replies
+ */
+
+#define KBD_REPLY_POR 0xAA /* Power on reset */
+#define KBD_REPLY_ACK 0xFA /* Command ACK */
+#define KBD_REPLY_RESEND 0xFE /* Command NACK, send the cmd again */
+
+/*
+ * Status Register Bits
+ */
+
+#define KBD_STAT_OBF 0x01 /* Keyboard output buffer full */
+#define KBD_STAT_IBF 0x02 /* Keyboard input buffer full */
+#define KBD_STAT_SELFTEST 0x04 /* Self test successful */
+#define KBD_STAT_CMD 0x08 /* Last write was a command
+ * write (0=data) */
+#define KBD_STAT_UNLOCKED 0x10 /* Zero if keyboard locked */
+#define KBD_STAT_MOUSE_OBF 0x20 /* Mouse output buffer full */
+#define KBD_STAT_GTO 0x40 /* General receive/xmit timeout */
+#define KBD_STAT_PERR 0x80 /* Parity error */
+
+#define AUX_STAT_OBF (KBD_STAT_OBF | KBD_STAT_MOUSE_OBF)
+
+/*
+ * Controller Mode Register Bits
+ */
+
+#define KBD_MODE_KBD_INT 0x01 /* Keyboard data generate IRQ1 */
+#define KBD_MODE_MOUSE_INT 0x02 /* Mouse data generate IRQ12 */
+#define KBD_MODE_SYS 0x04 /* The system flag (?) */
+#define KBD_MODE_NO_KEYLOCK 0x08 /* The keylock doesn't affect
+ * the keyboard if set */
+#define KBD_MODE_DISABLE_KBD 0x10 /* Disable keyboard interface */
+#define KBD_MODE_DISABLE_MOUSE 0x20 /* Disable mouse interface */
+#define KBD_MODE_KCC 0x40 /* Scan code conversion to PC format */
+#define KBD_MODE_RFU 0x80
+
+/*
+ * Mouse Commands
+ */
+
+#define AUX_SET_RES 0xE8 /* Set resolution */
+#define AUX_SET_SCALE11 0xE6 /* Set 1:1 scaling */
+#define AUX_SET_SCALE21 0xE7 /* Set 2:1 scaling */
+#define AUX_GET_SCALE 0xE9 /* Get scaling factor */
+#define AUX_SET_STREAM 0xEA /* Set stream mode */
+#define AUX_SET_SAMPLE 0xF3 /* Set sample rate */
+#define AUX_ENABLE_DEV 0xF4 /* Enable aux device */
+#define AUX_DISABLE_DEV 0xF5 /* Disable aux device */
+#define AUX_RESET 0xFF /* Reset aux device */
+#define AUX_ACK 0xFA /* Command byte ACK. */
+
+#define AUX_BUF_SIZE 2048 /* This might be better divisible by
+ three to make overruns stay in sync
+ but then the read function would need
+ a lock etc - ick */
+
+struct aux_queue {
+ unsigned long head;
+ unsigned long tail;
+ wait_queue_head_t proc_list;
+ struct fasync_struct *fasync;
+ unsigned char buf[AUX_BUF_SIZE];
+};
+
+
+/* How to access the keyboard macros on this platform. */
+#define kbd_read_input() inb(KBD_DATA_REG)
+#define kbd_read_status() inb(KBD_STATUS_REG)
+#define kbd_write_output(val) outb(val, KBD_DATA_REG)
+#define kbd_write_command(val) outb(val, KBD_CNTL_REG)
diff --git a/drivers/serial/kgdboc.c b/drivers/serial/kgdboc.c
index eadc1ab..ecef6e1 100644
--- a/drivers/serial/kgdboc.c
+++ b/drivers/serial/kgdboc.c
@@ -14,6 +14,7 @@
#include <linux/kernel.h>
#include <linux/ctype.h>
#include <linux/kgdb.h>
+#include <linux/kdb.h>
#include <linux/tty.h>
#define MAX_CONFIG_LEN 40
@@ -32,6 +33,40 @@ static struct kparam_string kps = {
static struct tty_driver *kgdb_tty_driver;
static int kgdb_tty_line;
+#ifdef CONFIG_KDB_KEYBOARD
+static int kgdboc_register_kbd(char **cptr)
+{
+ if (strncmp(*cptr, "kbd", 3) == 0) {
+ if (kdb_poll_idx < KDB_POLL_FUNC_MAX) {
+ kdb_poll_funcs[kdb_poll_idx] = kdb_get_kbd_char;
+ kdb_poll_idx++;
+ if (cptr[0][3] == ',')
+ *cptr += 4;
+ else
+ return 1;
+ }
+ }
+ return 0;
+}
+
+static void kgdboc_unregister_kbd(void)
+{
+ int i;
+
+ for (i = 0; i < kdb_poll_idx; i++) {
+ if (kdb_poll_funcs[i] == kdb_get_kbd_char) {
+ kdb_poll_idx--;
+ kdb_poll_funcs[i] = kdb_poll_funcs[kdb_poll_idx];
+ kdb_poll_funcs[kdb_poll_idx] = NULL;
+ i--;
+ }
+ }
+}
+#else /* ! CONFIG_KDB_KEYBOARD */
+#define kgdboc_register_kbd(x) 0
+#define kgdboc_unregister_kbd()
+#endif /* ! CONFIG_KDB_KEYBOARD */
+
static int kgdboc_option_setup(char *opt)
{
if (strlen(opt) > MAX_CONFIG_LEN) {
@@ -45,25 +80,38 @@ static int kgdboc_option_setup(char *opt)
__setup("kgdboc=", kgdboc_option_setup);
+static void cleanup_kgdboc(void)
+{
+ kgdboc_unregister_kbd();
+ if (configured == 1)
+ kgdb_unregister_io_module(&kgdboc_io_ops);
+}
+
static int configure_kgdboc(void)
{
struct tty_driver *p;
int tty_line = 0;
int err;
+ char *cptr = config;
err = kgdboc_option_setup(config);
if (err || !strlen(config) || isspace(config[0]))
goto noconfig;
err = -ENODEV;
+ kgdb_tty_driver = NULL;
+
+ if (kgdboc_register_kbd(&cptr))
+ goto do_register;
- p = tty_find_polling_driver(config, &tty_line);
+ p = tty_find_polling_driver(cptr, &tty_line);
if (!p)
goto noconfig;
kgdb_tty_driver = p;
kgdb_tty_line = tty_line;
+do_register:
err = kgdb_register_io_module(&kgdboc_io_ops);
if (err)
goto noconfig;
@@ -75,6 +123,7 @@ static int configure_kgdboc(void)
noconfig:
config[0] = 0;
configured = 0;
+ cleanup_kgdboc();
return err;
}
@@ -88,20 +137,18 @@ static int __init init_kgdboc(void)
return configure_kgdboc();
}
-static void cleanup_kgdboc(void)
-{
- if (configured == 1)
- kgdb_unregister_io_module(&kgdboc_io_ops);
-}
-
static int kgdboc_get_char(void)
{
+ if (!kgdb_tty_driver)
+ return -1;
return kgdb_tty_driver->ops->poll_get_char(kgdb_tty_driver,
kgdb_tty_line);
}
static void kgdboc_put_char(u8 chr)
{
+ if (!kgdb_tty_driver)
+ return;
kgdb_tty_driver->ops->poll_put_char(kgdb_tty_driver,
kgdb_tty_line, chr);
}
diff --git a/lib/Kconfig.kgdb b/lib/Kconfig.kgdb
index 02b46de..0fb633c 100644
--- a/lib/Kconfig.kgdb
+++ b/lib/Kconfig.kgdb
@@ -63,4 +63,11 @@ config KGDB_KDB
help
KDB frontend for kernel
+config KDB_KEYBOARD
+ bool "KGDB_KDB: keyboard as input device"
+ depends on VT && KGDB_KDB
+ default y
+ help
+ KDB can use a PS/2 type keyboard for an input device
+
endif # KGDB
--
1.6.4.rc1
Update the maintained files sections.
Signed-off-by: Jason Wessel <[email protected]>
---
MAINTAINERS | 6 ++++--
1 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/MAINTAINERS b/MAINTAINERS
index 412eff6..b7b192a 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -3172,15 +3172,17 @@ S: Maintained
F: include/linux/kexec.h
F: kernel/kexec.c
-KGDB
+KGDB / KDB / debug_core
M: Jason Wessel <[email protected]>
L: [email protected]
S: Maintained
F: Documentation/DocBook/kgdb.tmpl
+F: drivers/char/kdb_keyboard.*
F: drivers/misc/kgdbts.c
F: drivers/serial/kgdboc.c
F: include/linux/kgdb.h
-F: kernel/kgdb.c
+F: include/linux/kdb.h
+F: kernel/debug/
KMEMCHECK
M: Vegard Nossum <[email protected]>
--
1.6.4.rc1
One of the driving forces behind integrating another front end (kdb)
to the debug core is to allow front end commands to be accessible via
gdb's monitor command. It is true that you could write gdb macros to
get certain data, but you may want to just use gdb to access the
commands that are available in the kdb front end.
This patch implements the Rcmd gdb stub packet. In gdb you access
this with the "monitor" command. For instance you could type "monitor
help", "monitor lsmod" or "monitor ps A" etc...
There is no error checking or command restrictions on what you can and
cannot access at this point. Doing something like trying to set
breakpoints with the monitor command is going to cause nothing but
problems. Perhaps in the future only the commands that are actually
known to work with the gdb monitor command will be available.
Signed-off-by: Jason Wessel <[email protected]>
---
kernel/debug/debug_core.c | 2 +-
kernel/debug/debug_core.h | 2 ++
kernel/debug/gdbstub.c | 22 ++++++++++++++++++++++
kernel/debug/kdb/kdb_io.c | 13 +++++++++----
kernel/debug/kdb/kdb_private.h | 1 -
5 files changed, 34 insertions(+), 6 deletions(-)
diff --git a/kernel/debug/debug_core.c b/kernel/debug/debug_core.c
index 5bd2670..dbf81aa 100644
--- a/kernel/debug/debug_core.c
+++ b/kernel/debug/debug_core.c
@@ -82,7 +82,7 @@ static int kgdb_use_con;
int dbg_switch_cpu;
/* Use kdb or gdbserver mode */
-static int dbg_kdb_mode = 1;
+int dbg_kdb_mode = 1;
static int __init opt_kgdb_con(char *str)
{
diff --git a/kernel/debug/debug_core.h b/kernel/debug/debug_core.h
index a7051fa..888cc4e 100644
--- a/kernel/debug/debug_core.h
+++ b/kernel/debug/debug_core.h
@@ -66,9 +66,11 @@ extern void gdbstub_msg_write(const char *s, int len);
/* gdbstub functions used for kdb <-> gdbstub transition */
extern int gdbstub_state(struct kgdb_state *ks, char *cmd);
+extern int dbg_kdb_mode;
#ifdef CONFIG_KGDB_KDB
extern int kdb_stub(struct kgdb_state *ks);
+extern int kdb_parse(const char *cmdstr);
#else /* ! CONFIG_KGDB_KDB */
static inline int kdb_stub(struct kgdb_state *ks)
{
diff --git a/kernel/debug/gdbstub.c b/kernel/debug/gdbstub.c
index 6965b5c..cee2c0e 100644
--- a/kernel/debug/gdbstub.c
+++ b/kernel/debug/gdbstub.c
@@ -201,6 +201,9 @@ void gdbstub_msg_write(const char *s, int len)
int wcount;
int i;
+ if (len == 0)
+ len = strlen(s);
+
/* 'O'utput */
gdbmsgbuf[0] = 'O';
@@ -690,6 +693,25 @@ static void gdb_cmd_query(struct kgdb_state *ks)
kgdb_mem2hex(tmpstr, remcom_out_buffer, strlen(tmpstr));
}
break;
+#ifdef CONFIG_KGDB_KDB
+ case 'R':
+ if (strncmp(remcom_in_buffer, "qRcmd,", 6) == 0) {
+ int len = strlen(remcom_in_buffer + 6);
+
+ if ((len % 2) != 0) {
+ strcpy(remcom_out_buffer, "E01");
+ break;
+ }
+ kgdb_hex2mem(remcom_in_buffer + 6,
+ remcom_out_buffer, len);
+ len = len / 2;
+ remcom_out_buffer[len++] = 0;
+
+ kdb_parse(remcom_out_buffer);
+ strcpy(remcom_out_buffer, "OK");
+ }
+ break;
+#endif
}
}
diff --git a/kernel/debug/kdb/kdb_io.c b/kernel/debug/kdb/kdb_io.c
index 9e3cec7..8339b29 100644
--- a/kernel/debug/kdb/kdb_io.c
+++ b/kernel/debug/kdb/kdb_io.c
@@ -21,6 +21,7 @@
#include <linux/smp.h>
#include <linux/nmi.h>
#include <linux/delay.h>
+#include <linux/kgdb.h>
#include <linux/kdb.h>
#include <linux/kallsyms.h>
#include "kdb_private.h"
@@ -669,10 +670,14 @@ kdb_printit:
* Write to all consoles.
*/
retlen = strlen(kdb_buffer);
- while (c) {
- c->write(c, kdb_buffer, retlen);
- touch_nmi_watchdog();
- c = c->next;
+ if (!dbg_kdb_mode && kgdb_connected) {
+ gdbstub_msg_write(kdb_buffer, retlen);
+ } else {
+ while (c) {
+ c->write(c, kdb_buffer, retlen);
+ touch_nmi_watchdog();
+ c = c->next;
+ }
}
if (logging) {
saved_loglevel = console_loglevel;
diff --git a/kernel/debug/kdb/kdb_private.h b/kernel/debug/kdb/kdb_private.h
index 5183b69..42af29b 100644
--- a/kernel/debug/kdb/kdb_private.h
+++ b/kernel/debug/kdb/kdb_private.h
@@ -254,7 +254,6 @@ extern unsigned long kdb_task_state(const struct task_struct *p,
unsigned long mask);
extern void kdb_ps_suppressed(void);
extern void kdb_ps1(const struct task_struct *p);
-extern int kdb_parse(const char *cmdstr);
extern void kdb_print_nameval(const char *name, unsigned long val);
extern void kdb_send_sig_info(struct task_struct *p, struct siginfo *info);
extern void kdb_meminfo_proc_show(void);
--
1.6.4.rc1
This patch contains the hooks and instrumentation into kernel which
live outside the kernel/debug directory, which the kdb core
will call to run commands like lsmod, dmesg, bt etc...
CC: [email protected]
CC: [email protected]
Signed-off-by: Jason Wessel <[email protected]>
---
arch/arm/include/asm/kmap_types.h | 1 +
arch/powerpc/include/asm/kmap_types.h | 1 +
fs/proc/internal.h | 4 +-
fs/proc/meminfo.c | 15 ++++++++++--
fs/proc/mmu.c | 8 ++++--
include/asm-generic/kmap_types.h | 3 +-
include/linux/swap.h | 2 +
init/main.c | 2 +
kernel/kallsyms.c | 21 +++++++++++++++++
kernel/module.c | 4 +++
kernel/printk.c | 16 +++++++++++++
kernel/sched.c | 7 ++++-
kernel/signal.c | 40 +++++++++++++++++++++++++++++++++
mm/swapfile.c | 10 ++++++-
14 files changed, 121 insertions(+), 13 deletions(-)
diff --git a/arch/arm/include/asm/kmap_types.h b/arch/arm/include/asm/kmap_types.h
index c019949..3a9fb57 100644
--- a/arch/arm/include/asm/kmap_types.h
+++ b/arch/arm/include/asm/kmap_types.h
@@ -19,6 +19,7 @@ enum km_type {
KM_SOFTIRQ0,
KM_SOFTIRQ1,
KM_L2_CACHE,
+ KM_KDB,
KM_TYPE_NR
};
diff --git a/arch/powerpc/include/asm/kmap_types.h b/arch/powerpc/include/asm/kmap_types.h
index 9163695..bca8fdc 100644
--- a/arch/powerpc/include/asm/kmap_types.h
+++ b/arch/powerpc/include/asm/kmap_types.h
@@ -26,6 +26,7 @@ enum km_type {
KM_SOFTIRQ1,
KM_PPC_SYNC_PAGE,
KM_PPC_SYNC_ICACHE,
+ KM_KDB,
KM_TYPE_NR
};
diff --git a/fs/proc/internal.h b/fs/proc/internal.h
index 1f24a3e..36d55e1 100644
--- a/fs/proc/internal.h
+++ b/fs/proc/internal.h
@@ -32,11 +32,11 @@ extern struct mm_struct *mm_for_maps(struct task_struct *);
#ifdef CONFIG_MMU
#define VMALLOC_TOTAL (VMALLOC_END - VMALLOC_START)
-extern void get_vmalloc_info(struct vmalloc_info *vmi);
+extern void get_vmalloc_info(struct vmalloc_info *vmi, int lock);
#else
#define VMALLOC_TOTAL 0UL
-#define get_vmalloc_info(vmi) \
+#define get_vmalloc_info(vmi, lock) \
do { \
(vmi)->used = 0; \
(vmi)->largest_chunk = 0; \
diff --git a/fs/proc/meminfo.c b/fs/proc/meminfo.c
index a65239c..b967cf2 100644
--- a/fs/proc/meminfo.c
+++ b/fs/proc/meminfo.c
@@ -10,6 +10,7 @@
#include <linux/seq_file.h>
#include <linux/swap.h>
#include <linux/vmstat.h>
+#include <linux/kdb.h>
#include <asm/atomic.h>
#include <asm/page.h>
#include <asm/pgtable.h>
@@ -19,7 +20,7 @@ void __attribute__((weak)) arch_report_meminfo(struct seq_file *m)
{
}
-static int meminfo_proc_show(struct seq_file *m, void *v)
+int _meminfo_proc_show(struct seq_file *m, void *v, int lock)
{
struct sysinfo i;
unsigned long committed;
@@ -34,7 +35,10 @@ static int meminfo_proc_show(struct seq_file *m, void *v)
*/
#define K(x) ((x) << (PAGE_SHIFT - 10))
si_meminfo(&i);
- si_swapinfo(&i);
+ if (lock)
+ si_swapinfo(&i);
+ else
+ __si_swapinfo(&i);
committed = percpu_counter_read_positive(&vm_committed_as);
allowed = ((totalram_pages - hugetlb_total_pages())
* sysctl_overcommit_ratio / 100) + total_swap_pages;
@@ -44,7 +48,7 @@ static int meminfo_proc_show(struct seq_file *m, void *v)
if (cached < 0)
cached = 0;
- get_vmalloc_info(&vmi);
+ get_vmalloc_info(&vmi, lock);
for (lru = LRU_BASE; lru < NR_LRU_LISTS; lru++)
pages[lru] = global_page_state(NR_LRU_BASE + lru);
@@ -161,6 +165,11 @@ static int meminfo_proc_show(struct seq_file *m, void *v)
#undef K
}
+static int meminfo_proc_show(struct seq_file *m, void *v)
+{
+ return _meminfo_proc_show(m, v, 1);
+}
+
static int meminfo_proc_open(struct inode *inode, struct file *file)
{
return single_open(file, meminfo_proc_show, NULL);
diff --git a/fs/proc/mmu.c b/fs/proc/mmu.c
index 8ae221d..10a0f8b 100644
--- a/fs/proc/mmu.c
+++ b/fs/proc/mmu.c
@@ -14,7 +14,7 @@
#include <asm/pgtable.h>
#include "internal.h"
-void get_vmalloc_info(struct vmalloc_info *vmi)
+void get_vmalloc_info(struct vmalloc_info *vmi, int lock)
{
struct vm_struct *vma;
unsigned long free_area_size;
@@ -30,7 +30,8 @@ void get_vmalloc_info(struct vmalloc_info *vmi)
prev_end = VMALLOC_START;
- read_lock(&vmlist_lock);
+ if (lock)
+ read_lock(&vmlist_lock);
for (vma = vmlist; vma; vma = vma->next) {
unsigned long addr = (unsigned long) vma->addr;
@@ -55,6 +56,7 @@ void get_vmalloc_info(struct vmalloc_info *vmi)
if (VMALLOC_END - prev_end > vmi->largest_chunk)
vmi->largest_chunk = VMALLOC_END - prev_end;
- read_unlock(&vmlist_lock);
+ if (lock)
+ read_unlock(&vmlist_lock);
}
}
diff --git a/include/asm-generic/kmap_types.h b/include/asm-generic/kmap_types.h
index e5f234a..97e807c 100644
--- a/include/asm-generic/kmap_types.h
+++ b/include/asm-generic/kmap_types.h
@@ -28,7 +28,8 @@ KMAP_D(15) KM_UML_USERCOPY,
KMAP_D(16) KM_IRQ_PTE,
KMAP_D(17) KM_NMI,
KMAP_D(18) KM_NMI_PTE,
-KMAP_D(19) KM_TYPE_NR
+KMAP_D(19) KM_KDB,
+KMAP_D(20) KM_TYPE_NR
};
#undef KMAP_D
diff --git a/include/linux/swap.h b/include/linux/swap.h
index a2602a8..c326282 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -312,6 +312,7 @@ extern struct page *swapin_readahead(swp_entry_t, gfp_t,
/* linux/mm/swapfile.c */
extern long nr_swap_pages;
extern long total_swap_pages;
+extern void __si_swapinfo(struct sysinfo *);
extern void si_swapinfo(struct sysinfo *);
extern swp_entry_t get_swap_page(void);
extern swp_entry_t get_swap_page_of_type(int);
@@ -377,6 +378,7 @@ static inline void mem_cgroup_uncharge_swap(swp_entry_t ent)
#define si_swapinfo(val) \
do { (val)->freeswap = (val)->totalswap = 0; } while (0)
+#define __si_swapinfo(val) si_swapinfo(val)
/* only sparc can not include linux/pagemap.h in this file
* so leave page_cache_release and release_pages undeclared... */
#define free_page_and_swap_cache(page) \
diff --git a/init/main.c b/init/main.c
index 4cb47a1..9d415a1 100644
--- a/init/main.c
+++ b/init/main.c
@@ -63,6 +63,7 @@
#include <linux/sched.h>
#include <linux/signal.h>
#include <linux/idr.h>
+#include <linux/kdb.h>
#include <linux/ftrace.h>
#include <linux/async.h>
#include <linux/kmemcheck.h>
@@ -659,6 +660,7 @@ asmlinkage void __init start_kernel(void)
key_init();
radix_tree_init();
security_init();
+ kdb_init(KDB_INIT_FULL);
vfs_caches_init(totalram_pages);
signals_init();
/* rootfs populating might need page-writeback */
diff --git a/kernel/kallsyms.c b/kernel/kallsyms.c
index 8e5288a..dc08f8b 100644
--- a/kernel/kallsyms.c
+++ b/kernel/kallsyms.c
@@ -16,6 +16,7 @@
#include <linux/init.h>
#include <linux/seq_file.h>
#include <linux/fs.h>
+#include <linux/kdb.h>
#include <linux/err.h>
#include <linux/proc_fs.h>
#include <linux/sched.h> /* for cond_resched */
@@ -515,6 +516,26 @@ static int kallsyms_open(struct inode *inode, struct file *file)
return ret;
}
+#ifdef CONFIG_KGDB_KDB
+const char *kdb_walk_kallsyms(loff_t *pos)
+{
+ static struct kallsym_iter kdb_walk_kallsyms_iter;
+ if (*pos == 0) {
+ memset(&kdb_walk_kallsyms_iter, 0,
+ sizeof(kdb_walk_kallsyms_iter));
+ reset_iter(&kdb_walk_kallsyms_iter, 0);
+ }
+ while (1) {
+ if (!update_iter(&kdb_walk_kallsyms_iter, *pos))
+ return NULL;
+ ++*pos;
+ /* Some debugging symbols have no name. Ignore them. */
+ if (kdb_walk_kallsyms_iter.name[0])
+ return kdb_walk_kallsyms_iter.name;
+ }
+}
+#endif /* CONFIG_KGDB_KDB */
+
static const struct file_operations kallsyms_operations = {
.open = kallsyms_open,
.read = seq_read,
diff --git a/kernel/module.c b/kernel/module.c
index f82386b..e59aca1 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -79,6 +79,10 @@ EXPORT_TRACEPOINT_SYMBOL(module_get);
DEFINE_MUTEX(module_mutex);
EXPORT_SYMBOL_GPL(module_mutex);
static LIST_HEAD(modules);
+#ifdef CONFIG_KGDB_KDB
+struct list_head *kdb_modules = &modules; /* kdb needs the list of modules */
+#endif /* CONFIG_KGDB_KDB */
+
/* Block module loading/unloading? */
int modules_disabled = 0;
diff --git a/kernel/printk.c b/kernel/printk.c
index 1751c45..d86c91a 100644
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -420,6 +420,22 @@ SYSCALL_DEFINE3(syslog, int, type, char __user *, buf, int, len)
return do_syslog(type, buf, len);
}
+#ifdef CONFIG_KGDB_KDB
+/* kdb dmesg command needs access to the syslog buffer. do_syslog()
+ * uses locks so it cannot be used during debugging. Just tell kdb
+ * where the start and end of the physical and logical logs are. This
+ * is equivalent to do_syslog(3).
+ */
+void kdb_syslog_data(char *syslog_data[4])
+{
+ syslog_data[0] = log_buf;
+ syslog_data[1] = log_buf + log_buf_len;
+ syslog_data[2] = log_buf + log_end -
+ (logged_chars < log_buf_len ? logged_chars : log_buf_len);
+ syslog_data[3] = log_buf + log_end;
+}
+#endif /* CONFIG_KGDB_KDB */
+
/*
* Call the console drivers on a range of log_buf
*/
diff --git a/kernel/sched.c b/kernel/sched.c
index 3a8fb30..92ab8b8 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -9802,9 +9802,9 @@ void normalize_rt_tasks(void)
#endif /* CONFIG_MAGIC_SYSRQ */
-#ifdef CONFIG_IA64
+#if defined(CONFIG_IA64) || defined(CONFIG_KGDB_KDB)
/*
- * These functions are only useful for the IA64 MCA handling.
+ * These functions are only useful for the IA64 MCA handling, or kdb.
*
* They can only be called when the whole system has been
* stopped - every CPU needs to be quiescent, and no scheduling
@@ -9824,6 +9824,9 @@ struct task_struct *curr_task(int cpu)
return cpu_curr(cpu);
}
+#endif /* defined(CONFIG_IA64) || defined(CONFIG_KGDB_KDB) */
+
+#ifdef CONFIG_IA64
/**
* set_curr_task - set the current task for a given cpu.
* @cpu: the processor in question.
diff --git a/kernel/signal.c b/kernel/signal.c
index 934ae5e..96ecbf8 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -2718,3 +2718,43 @@ void __init signals_init(void)
{
sigqueue_cachep = KMEM_CACHE(sigqueue, SLAB_PANIC);
}
+
+#ifdef CONFIG_KGDB_KDB
+#include <linux/kdb.h>
+/*
+ * kdb_send_sig_info - Allows kdb to send signals without exposing
+ * signal internals. This function checks if the required locks are
+ * available before calling the main signal code, to avoid kdb
+ * deadlocks.
+ */
+void
+kdb_send_sig_info(struct task_struct *t, struct siginfo *info)
+{
+ static struct task_struct *kdb_prev_t;
+ int sig, new_t;
+ if (!spin_trylock(&t->sighand->siglock)) {
+ kdb_printf("Can't do kill command now.\n"
+ "The sigmask lock is held somewhere else in "
+ "kernel, try again later\n");
+ return;
+ }
+ spin_unlock(&t->sighand->siglock);
+ new_t = kdb_prev_t != t;
+ kdb_prev_t = t;
+ if (t->state != TASK_RUNNING && new_t) {
+ kdb_printf("Process is not RUNNING, sending a signal from "
+ "kdb risks deadlock\n"
+ "on the run queue locks. "
+ "The signal has _not_ been sent.\n"
+ "Reissue the kill command if you want to risk "
+ "the deadlock.\n");
+ return;
+ }
+ sig = info->si_signo;
+ if (send_sig_info(sig, info, t))
+ kdb_printf("Fail to deliver Signal %d to process %d.\n",
+ sig, t->pid);
+ else
+ kdb_printf("Signal %d is sent to process %d.\n", sig, t->pid);
+}
+#endif /* CONFIG_KGDB_KDB */
diff --git a/mm/swapfile.c b/mm/swapfile.c
index 6c0585b..dc2039e 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -13,6 +13,7 @@
#include <linux/swap.h>
#include <linux/vmalloc.h>
#include <linux/pagemap.h>
+#include <linux/kdb.h>
#include <linux/namei.h>
#include <linux/shm.h>
#include <linux/blkdev.h>
@@ -2056,12 +2057,11 @@ out:
return error;
}
-void si_swapinfo(struct sysinfo *val)
+void __si_swapinfo(struct sysinfo *val)
{
unsigned int type;
unsigned long nr_to_be_unused = 0;
- spin_lock(&swap_lock);
for (type = 0; type < nr_swapfiles; type++) {
struct swap_info_struct *si = swap_info[type];
@@ -2070,6 +2070,12 @@ void si_swapinfo(struct sysinfo *val)
}
val->freeswap = nr_swap_pages + nr_to_be_unused;
val->totalswap = total_swap_pages + nr_to_be_unused;
+}
+
+void si_swapinfo(struct sysinfo *val)
+{
+ spin_lock(&swap_lock);
+ __si_swapinfo(val);
spin_unlock(&swap_lock);
}
--
1.6.4.rc1
This patch contains only the kdb core. Because the change set was
large, it was split. The next patch in the series includes the
instrumentation into the core kernel which are mainly helper functions
for kdb.
This work is directly derived from kdb v4.4 found at:
ftp://oss.sgi.com/projects/kdb/download/v4.4/
The kdb internals have been re-organized to make them mostly platform
independent and to connect everything to the debug core which is used by
gdbstub (which has long been known as kgdb).
The original version of kdb was 58,000 lines worth of changes to
support x86. From that implementation only the kdb shell, and basic
commands for memory access, runcontrol, lsmod, and dmesg where carried
forward.
This is a generic implementation which aims to cover all the current
architectures using the kgdb core: ppc, arm, x86, mips, sparc, sh and
blackfin. More archictectures can be added by implementing the
architecture specific kgdb functions.
[[email protected]: Compile fix with hugepages enabled]
[[email protected]: Clean breakpoint code renaming kdba_ -> kdb_]
[[email protected]: fix new line after printing registers]
[[email protected]: Remove the concept of global vs. local breakpoints]
[[email protected]: Rework kdb_si_swapinfo to use more generic name]
[[email protected]: fix the information dump macros, remove 'arch' from the names]
CC: [email protected]
CC: [email protected]
Signed-off-by: Jason Wessel <[email protected]>
---
include/linux/kdb.h | 110 ++
kernel/debug/Makefile | 4 +-
kernel/debug/kdb/.gitignore | 1 +
kernel/debug/kdb/Makefile | 24 +
kernel/debug/kdb/kdb_bp.c | 564 ++++++++
kernel/debug/kdb/kdb_bt.c | 208 +++
kernel/debug/kdb/kdb_cmds | 35 +
kernel/debug/kdb/kdb_debugger.c | 159 +++
kernel/debug/kdb/kdb_io.c | 789 +++++++++++
kernel/debug/kdb/kdb_main.c | 2851 +++++++++++++++++++++++++++++++++++++++
kernel/debug/kdb/kdb_private.h | 314 +++++
kernel/debug/kdb/kdb_support.c | 991 ++++++++++++++
12 files changed, 6049 insertions(+), 1 deletions(-)
create mode 100644 include/linux/kdb.h
create mode 100644 kernel/debug/kdb/.gitignore
create mode 100644 kernel/debug/kdb/Makefile
create mode 100644 kernel/debug/kdb/kdb_bp.c
create mode 100644 kernel/debug/kdb/kdb_bt.c
create mode 100644 kernel/debug/kdb/kdb_cmds
create mode 100644 kernel/debug/kdb/kdb_debugger.c
create mode 100644 kernel/debug/kdb/kdb_io.c
create mode 100644 kernel/debug/kdb/kdb_main.c
create mode 100644 kernel/debug/kdb/kdb_private.h
create mode 100644 kernel/debug/kdb/kdb_support.c
diff --git a/include/linux/kdb.h b/include/linux/kdb.h
new file mode 100644
index 0000000..b3ac39a
--- /dev/null
+++ b/include/linux/kdb.h
@@ -0,0 +1,110 @@
+#ifndef _KDB_H
+#define _KDB_H
+
+/*
+ * Kernel Debugger Architecture Independent Global Headers
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (c) 2000-2007 Silicon Graphics, Inc. All Rights Reserved.
+ * Copyright (C) 2000 Stephane Eranian <[email protected]>
+ * Copyright (C) 2009 Jason Wessel <[email protected]>
+ */
+
+#ifdef CONFIG_KGDB_KDB
+#include <linux/init.h>
+#include <linux/sched.h>
+#include <asm/atomic.h>
+
+#define KDB_POLL_FUNC_MAX 5
+
+/*
+ * kdb_initial_cpu is initialized to -1, and is set to the cpu
+ * number whenever the kernel debugger is entered.
+ */
+extern int kdb_initial_cpu;
+extern atomic_t kdb_event;
+
+/*
+ * kdb_diemsg
+ *
+ * Contains a pointer to the last string supplied to the
+ * kernel 'die' panic function.
+ */
+extern const char *kdb_diemsg;
+
+#define KDB_FLAG_EARLYKDB (1 << 0) /* set from boot parameter kdb=early */
+#define KDB_FLAG_CATASTROPHIC (1 << 1) /* A catastrophic event has occurred */
+#define KDB_FLAG_CMD_INTERRUPT (1 << 2) /* Previous command was interrupted */
+#define KDB_FLAG_NOIPI (1 << 3) /* Do not send IPIs */
+#define KDB_FLAG_ONLY_DO_DUMP (1 << 4) /* Only do a dump, used when
+ * kdb is off */
+#define KDB_FLAG_NO_CONSOLE (1 << 5) /* No console is available,
+ * kdb is disabled */
+#define KDB_FLAG_NO_VT_CONSOLE (1 << 6) /* No VT console is available, do
+ * not use keyboard */
+#define KDB_FLAG_NO_I8042 (1 << 7) /* No i8042 chip is available, do
+ * not use keyboard */
+
+extern int kdb_flags; /* Global flags, see kdb_state for per cpu state */
+
+extern void kdb_save_flags(void);
+extern void kdb_restore_flags(void);
+
+#define KDB_FLAG(flag) (kdb_flags & KDB_FLAG_##flag)
+#define KDB_FLAG_SET(flag) ((void)(kdb_flags |= KDB_FLAG_##flag))
+#define KDB_FLAG_CLEAR(flag) ((void)(kdb_flags &= ~KDB_FLAG_##flag))
+
+/*
+ * External entry point for the kernel debugger. The pt_regs
+ * at the time of entry are supplied along with the reason for
+ * entry to the kernel debugger.
+ */
+
+typedef enum {
+ KDB_REASON_ENTER = 1, /* KDB_ENTER() trap/fault - regs valid */
+ KDB_REASON_ENTER_SLAVE, /* KDB_ENTER_SLAVE() trap/fault - regs valid */
+ KDB_REASON_BREAK, /* Breakpoint inst. - regs valid */
+ KDB_REASON_DEBUG, /* Debug Fault - regs valid */
+ KDB_REASON_OOPS, /* Kernel Oops - regs valid */
+ KDB_REASON_SWITCH, /* CPU switch - regs valid*/
+ KDB_REASON_KEYBOARD, /* Keyboard entry - regs valid */
+ KDB_REASON_NMI, /* Non-maskable interrupt; regs valid */
+ KDB_REASON_RECURSE, /* Recursive entry to kdb;
+ * regs probably valid */
+ KDB_REASON_SSTEP, /* Single Step trap. - regs valid */
+} kdb_reason_t;
+
+extern int kdb_printf(const char *, ...)
+ __attribute__ ((format (printf, 1, 2)));
+typedef int (*kdb_printf_t)(const char *, ...)
+ __attribute__ ((format (printf, 1, 2)));
+
+extern void kdb_init(int level);
+
+/* Access to kdb specific polling devices */
+typedef int (*get_char_func)(void);
+extern get_char_func kdb_poll_funcs[];
+extern int kdb_get_kbd_char(void);
+
+static inline
+int kdb_process_cpu(const struct task_struct *p)
+{
+ unsigned int cpu = task_thread_info(p)->cpu;
+ if (cpu > num_possible_cpus())
+ cpu = 0;
+ return cpu;
+}
+
+#else /* ! CONFIG_KGDB_KDB */
+#define kdb_printf(...)
+#define kdb_init(x)
+#endif /* CONFIG_KGDB_KDB */
+enum {
+ KDB_NOT_INITIALIZED,
+ KDB_INIT_EARLY,
+ KDB_INIT_FULL,
+};
+#endif /* !_KDB_H */
diff --git a/kernel/debug/Makefile b/kernel/debug/Makefile
index 3657c64..c72de00 100644
--- a/kernel/debug/Makefile
+++ b/kernel/debug/Makefile
@@ -2,4 +2,6 @@
# Makefile for the linux kernel debugger
#
-obj-$(CONFIG_KGDB) += debug_core.o gdbstub.o
\ No newline at end of file
+obj-$(CONFIG_KGDB) += debug_core.o gdbstub.o
+obj-$(CONFIG_KGDB_KDB) += kdb/
+
diff --git a/kernel/debug/kdb/.gitignore b/kernel/debug/kdb/.gitignore
new file mode 100644
index 0000000..396d12e
--- /dev/null
+++ b/kernel/debug/kdb/.gitignore
@@ -0,0 +1 @@
+gen-kdb_cmds.c
diff --git a/kernel/debug/kdb/Makefile b/kernel/debug/kdb/Makefile
new file mode 100644
index 0000000..d1e925e
--- /dev/null
+++ b/kernel/debug/kdb/Makefile
@@ -0,0 +1,24 @@
+# This file is subject to the terms and conditions of the GNU General Public
+# License. See the file "COPYING" in the main directory of this archive
+# for more details.
+#
+# Copyright (c) 1999-2004 Silicon Graphics, Inc. All Rights Reserved.
+# Copyright (c) 2009 Wind River Systems, Inc. All Rights Reserved.
+#
+
+CCVERSION := $(shell $(CC) -v 2>&1 | sed -ne '$$p')
+obj-y := kdb_io.o kdb_main.o kdb_support.o kdb_bt.o gen-kdb_cmds.o kdb_bp.o kdb_debugger.o
+
+clean-files := gen-kdb_cmds.c
+
+quiet_cmd_gen-kdb = GENKDB $@
+ cmd_gen-kdb = $(AWK) 'BEGIN {print "\#include <linux/stddef.h>"; print "\#include <linux/init.h>"} \
+ /^\#/{next} \
+ /^[ \t]*$$/{next} \
+ {gsub(/"/, "\\\"", $$0); \
+ print "static __initdata char kdb_cmd" cmds++ "[] = \"" $$0 "\\n\";"} \
+ END {print "extern char *kdb_cmds[]; char __initdata *kdb_cmds[] = {"; for (i = 0; i < cmds; ++i) {print " kdb_cmd" i ","}; print(" NULL\n};");}' \
+ $(filter-out %/Makefile,$^) > $@#
+
+$(obj)/gen-kdb_cmds.c: $(src)/kdb_cmds $(src)/Makefile
+ $(call cmd,gen-kdb)
diff --git a/kernel/debug/kdb/kdb_bp.c b/kernel/debug/kdb/kdb_bp.c
new file mode 100644
index 0000000..75bd9b3
--- /dev/null
+++ b/kernel/debug/kdb/kdb_bp.c
@@ -0,0 +1,564 @@
+/*
+ * Kernel Debugger Architecture Independent Breakpoint Handler
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (c) 1999-2004 Silicon Graphics, Inc. All Rights Reserved.
+ * Copyright (c) 2009 Wind River Systems, Inc. All Rights Reserved.
+ */
+
+#include <linux/string.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/kdb.h>
+#include <linux/kgdb.h>
+#include <linux/smp.h>
+#include <linux/sched.h>
+#include <linux/interrupt.h>
+#include "kdb_private.h"
+
+/*
+ * Table of kdb_breakpoints
+ */
+kdb_bp_t kdb_breakpoints[KDB_MAXBPT];
+
+static void kdb_setsinglestep(struct pt_regs *regs)
+{
+ KDB_STATE_SET(DOING_SS);
+}
+
+static char *kdb_rwtypes[] = {
+ "Instruction(i)",
+ "Instruction(Register)",
+ "Data Write",
+ "I/O",
+ "Data Access"
+};
+
+static char *kdb_bptype(kdb_bp_t *bp)
+{
+ if (bp->bp_type < 0 || bp->bp_type > 4)
+ return "";
+
+ return kdb_rwtypes[bp->bp_type];
+}
+
+static int kdb_parsebp(int argc, const char **argv, int *nextargp, kdb_bp_t *bp)
+{
+ int nextarg = *nextargp;
+ int diag;
+
+ bp->bph_length = 1;
+ if ((argc + 1) != nextarg) {
+ if (strnicmp(argv[nextarg], "datar", sizeof("datar")) == 0)
+ bp->bp_type = BP_ACCESS_WATCHPOINT;
+ else if (strnicmp(argv[nextarg], "dataw", sizeof("dataw")) == 0)
+ bp->bp_type = BP_WRITE_WATCHPOINT;
+ else if (strnicmp(argv[nextarg], "inst", sizeof("inst")) == 0)
+ bp->bp_type = BP_HARDWARE_BREAKPOINT;
+ else
+ return KDB_ARGCOUNT;
+
+ bp->bph_length = 1;
+
+ nextarg++;
+
+ if ((argc + 1) != nextarg) {
+ unsigned long len;
+
+ diag = kdbgetularg((char *)argv[nextarg],
+ &len);
+ if (diag)
+ return diag;
+
+
+ if (len > 8)
+ return KDB_BADLENGTH;
+
+ bp->bph_length = len;
+ nextarg++;
+ }
+
+ if ((argc + 1) != nextarg)
+ return KDB_ARGCOUNT;
+ }
+
+ *nextargp = nextarg;
+ return 0;
+}
+
+static int _kdb_bp_remove(kdb_bp_t *bp)
+{
+ int ret = 1;
+ if (!bp->bp_installed)
+ return ret;
+ if (!bp->bp_type)
+ ret = dbg_remove_sw_break(bp->bp_addr);
+ else
+ ret = arch_kgdb_ops.remove_hw_breakpoint(bp->bp_addr,
+ bp->bph_length,
+ bp->bp_type);
+ if (ret == 0)
+ bp->bp_installed = 0;
+ return ret;
+}
+
+static void kdb_handle_bp(struct pt_regs *regs, kdb_bp_t *bp)
+{
+ if (KDB_DEBUG(BP))
+ kdb_printf("regs->ip = 0x%lx\n", instruction_pointer(regs));
+
+ /*
+ * Setup single step
+ */
+ kdb_setsinglestep(regs);
+
+ /*
+ * Reset delay attribute
+ */
+ bp->bp_delay = 0;
+ bp->bp_delayed = 1;
+}
+
+static int _kdb_bp_install(struct pt_regs *regs, kdb_bp_t *bp)
+{
+ int ret;
+ /*
+ * Install the breakpoint, if it is not already installed.
+ */
+
+ if (KDB_DEBUG(BP))
+ kdb_printf("%s: bp_installed %d\n",
+ __func__, bp->bp_installed);
+ if (!KDB_STATE(SSBPT))
+ bp->bp_delay = 0;
+ if (bp->bp_installed)
+ return 1;
+ if (bp->bp_delay || (bp->bp_delayed && KDB_STATE(DOING_SS))) {
+ if (KDB_DEBUG(BP))
+ kdb_printf("%s: delayed bp\n", __func__);
+ kdb_handle_bp(regs, bp);
+ return 0;
+ }
+ if (!bp->bp_type)
+ ret = dbg_set_sw_break(bp->bp_addr);
+ else
+ ret = arch_kgdb_ops.set_hw_breakpoint(bp->bp_addr,
+ bp->bph_length,
+ bp->bp_type);
+ if (ret == 0) {
+ bp->bp_installed = 1;
+ } else {
+ kdb_printf("%s: failed to set breakpoint at 0x%lx\n",
+ __func__, bp->bp_addr);
+ return 1;
+ }
+ return 0;
+}
+
+/*
+ * kdb_bp_install
+ *
+ * Install kdb_breakpoints prior to returning from the
+ * kernel debugger. This allows the kdb_breakpoints to be set
+ * upon functions that are used internally by kdb, such as
+ * printk(). This function is only called once per kdb session.
+ */
+void kdb_bp_install(struct pt_regs *regs)
+{
+ int i;
+
+ for (i = 0; i < KDB_MAXBPT; i++) {
+ kdb_bp_t *bp = &kdb_breakpoints[i];
+
+ if (KDB_DEBUG(BP)) {
+ kdb_printf("%s: bp %d bp_enabled %d\n",
+ __func__, i, bp->bp_enabled);
+ }
+ if (bp->bp_enabled)
+ _kdb_bp_install(regs, bp);
+ }
+}
+
+/*
+ * kdb_bp_remove
+ *
+ * Remove kdb_breakpoints upon entry to the kernel debugger.
+ *
+ * Parameters:
+ * None.
+ * Outputs:
+ * None.
+ * Returns:
+ * None.
+ * Locking:
+ * None.
+ * Remarks:
+ */
+void kdb_bp_remove(void)
+{
+ int i;
+
+ for (i = KDB_MAXBPT - 1; i >= 0; i--) {
+ kdb_bp_t *bp = &kdb_breakpoints[i];
+
+ if (KDB_DEBUG(BP)) {
+ kdb_printf("%s: bp %d bp_enabled %d\n",
+ __func__, i, bp->bp_enabled);
+ }
+ if (bp->bp_enabled)
+ _kdb_bp_remove(bp);
+ }
+}
+
+
+/*
+ * kdb_printbp
+ *
+ * Internal function to format and print a breakpoint entry.
+ *
+ * Parameters:
+ * None.
+ * Outputs:
+ * None.
+ * Returns:
+ * None.
+ * Locking:
+ * None.
+ * Remarks:
+ */
+
+static void kdb_printbp(kdb_bp_t *bp, int i)
+{
+ kdb_printf("%s ", kdb_bptype(bp));
+ kdb_printf("BP #%d at ", i);
+ kdb_symbol_print(bp->bp_addr, NULL, KDB_SP_DEFAULT);
+
+ if (bp->bp_enabled)
+ kdb_printf("\n is enabled");
+ else
+ kdb_printf("\n is disabled");
+
+ kdb_printf("\taddr at %016lx, hardtype=%d installed=%d\n",
+ bp->bp_addr, bp->bp_type, bp->bp_installed);
+
+ kdb_printf("\n");
+}
+
+/*
+ * kdb_bp
+ *
+ * Handle the bp commands.
+ *
+ * [bp|bph] <addr-expression> [DATAR|DATAW]
+ *
+ * Parameters:
+ * argc Count of arguments in argv
+ * argv Space delimited command line arguments
+ * Outputs:
+ * None.
+ * Returns:
+ * Zero for success, a kdb diagnostic if failure.
+ * Locking:
+ * None.
+ * Remarks:
+ *
+ * bp Set breakpoint on all cpus. Only use hardware assist if need.
+ * bph Set breakpoint on all cpus. Force hardware register
+ */
+
+static int kdb_bp(int argc, const char **argv)
+{
+ int i, bpno;
+ kdb_bp_t *bp, *bp_check;
+ int diag;
+ int free;
+ char *symname = NULL;
+ long offset = 0ul;
+ int nextarg;
+ kdb_bp_t template = {0};
+
+ if (argc == 0) {
+ /*
+ * Display breakpoint table
+ */
+ for (bpno = 0, bp = kdb_breakpoints; bpno < KDB_MAXBPT;
+ bpno++, bp++) {
+ if (bp->bp_free)
+ continue;
+ kdb_printbp(bp, bpno);
+ }
+
+ return 0;
+ }
+
+ nextarg = 1;
+ diag = kdbgetaddrarg(argc, argv, &nextarg, &template.bp_addr,
+ &offset, &symname);
+ if (diag)
+ return diag;
+ if (!template.bp_addr)
+ return KDB_BADINT;
+
+ /*
+ * Find an empty bp structure to allocate
+ */
+ free = KDB_MAXBPT;
+ for (bpno = 0, bp = kdb_breakpoints; bpno < KDB_MAXBPT; bpno++, bp++) {
+ if (bp->bp_free)
+ break;
+ }
+
+ if (bpno == KDB_MAXBPT)
+ return KDB_TOOMANYBPT;
+
+ if (strcmp(argv[0], "bph") == 0) {
+ template.bp_type = BP_HARDWARE_BREAKPOINT;
+ diag = kdb_parsebp(argc, argv, &nextarg, &template);
+ if (diag)
+ return diag;
+ } else {
+ template.bp_type = BP_BREAKPOINT;
+ }
+
+ /*
+ * Check for clashing breakpoints.
+ *
+ * Note, in this design we can't have hardware breakpoints
+ * enabled for both read and write on the same address.
+ */
+ for (i = 0, bp_check = kdb_breakpoints; i < KDB_MAXBPT;
+ i++, bp_check++) {
+ if (!bp_check->bp_free &&
+ bp_check->bp_addr == template.bp_addr) {
+ kdb_printf("You already have a breakpoint at "
+ kdb_bfd_vma_fmt0 "\n", template.bp_addr);
+ return KDB_DUPBPT;
+ }
+ }
+
+ template.bp_enabled = 1;
+
+ /*
+ * Actually allocate the breakpoint found earlier
+ */
+ *bp = template;
+ bp->bp_free = 0;
+
+ kdb_printbp(bp, bpno);
+
+ return 0;
+}
+
+/*
+ * kdb_bc
+ *
+ * Handles the 'bc', 'be', and 'bd' commands
+ *
+ * [bd|bc|be] <breakpoint-number>
+ * [bd|bc|be] *
+ *
+ * Parameters:
+ * argc Count of arguments in argv
+ * argv Space delimited command line arguments
+ * Outputs:
+ * None.
+ * Returns:
+ * Zero for success, a kdb diagnostic for failure
+ * Locking:
+ * None.
+ * Remarks:
+ */
+static int kdb_bc(int argc, const char **argv)
+{
+ unsigned long addr;
+ kdb_bp_t *bp = NULL;
+ int lowbp = KDB_MAXBPT;
+ int highbp = 0;
+ int done = 0;
+ int i;
+ int diag = 0;
+
+ int cmd; /* KDBCMD_B? */
+#define KDBCMD_BC 0
+#define KDBCMD_BE 1
+#define KDBCMD_BD 2
+
+ if (strcmp(argv[0], "be") == 0)
+ cmd = KDBCMD_BE;
+ else if (strcmp(argv[0], "bd") == 0)
+ cmd = KDBCMD_BD;
+ else
+ cmd = KDBCMD_BC;
+
+ if (argc != 1)
+ return KDB_ARGCOUNT;
+
+ if (strcmp(argv[1], "*") == 0) {
+ lowbp = 0;
+ highbp = KDB_MAXBPT;
+ } else {
+ diag = kdbgetularg(argv[1], &addr);
+ if (diag)
+ return diag;
+
+ /*
+ * For addresses less than the maximum breakpoint number,
+ * assume that the breakpoint number is desired.
+ */
+ if (addr < KDB_MAXBPT) {
+ bp = &kdb_breakpoints[addr];
+ lowbp = highbp = addr;
+ highbp++;
+ } else {
+ for (i = 0, bp = kdb_breakpoints; i < KDB_MAXBPT;
+ i++, bp++) {
+ if (bp->bp_addr == addr) {
+ lowbp = highbp = i;
+ highbp++;
+ break;
+ }
+ }
+ }
+ }
+
+ /*
+ * Now operate on the set of breakpoints matching the input
+ * criteria (either '*' for all, or an individual breakpoint).
+ */
+ for (bp = &kdb_breakpoints[lowbp], i = lowbp;
+ i < highbp;
+ i++, bp++) {
+ if (bp->bp_free)
+ continue;
+
+ done++;
+
+ switch (cmd) {
+ case KDBCMD_BC:
+ bp->bp_enabled = 0;
+
+ kdb_printf("Breakpoint %d at "
+ kdb_bfd_vma_fmt " cleared\n",
+ i, bp->bp_addr);
+
+ bp->bp_addr = 0;
+ bp->bp_free = 1;
+
+ break;
+ case KDBCMD_BE:
+ bp->bp_enabled = 1;
+
+ kdb_printf("Breakpoint %d at "
+ kdb_bfd_vma_fmt " enabled",
+ i, bp->bp_addr);
+
+ kdb_printf("\n");
+ break;
+ case KDBCMD_BD:
+ if (!bp->bp_enabled)
+ break;
+
+ bp->bp_enabled = 0;
+
+ kdb_printf("Breakpoint %d at "
+ kdb_bfd_vma_fmt " disabled\n",
+ i, bp->bp_addr);
+
+ break;
+ }
+ if (bp->bp_delay && (cmd == KDBCMD_BC || cmd == KDBCMD_BD)) {
+ bp->bp_delay = 0;
+ KDB_STATE_CLEAR(SSBPT);
+ }
+ }
+
+ return (!done) ? KDB_BPTNOTFOUND : 0;
+}
+
+/*
+ * kdb_ss
+ *
+ * Process the 'ss' (Single Step) and 'ssb' (Single Step to Branch)
+ * commands.
+ *
+ * ss
+ * ssb
+ *
+ * Parameters:
+ * argc Argument count
+ * argv Argument vector
+ * Outputs:
+ * None.
+ * Returns:
+ * KDB_CMD_SS[B] for success, a kdb error if failure.
+ * Locking:
+ * None.
+ * Remarks:
+ *
+ * Set the arch specific option to trigger a debug trap after the next
+ * instruction.
+ *
+ * For 'ssb', set the trace flag in the debug trap handler
+ * after printing the current insn and return directly without
+ * invoking the kdb command processor, until a branch instruction
+ * is encountered.
+ */
+
+static int kdb_ss(int argc, const char **argv)
+{
+ int ssb = 0;
+
+ ssb = (strcmp(argv[0], "ssb") == 0);
+ if (argc != 0)
+ return KDB_ARGCOUNT;
+ /*
+ * Set trace flag and go.
+ */
+ KDB_STATE_SET(DOING_SS);
+ if (ssb) {
+ KDB_STATE_SET(DOING_SSB);
+ return KDB_CMD_SSB;
+ }
+ return KDB_CMD_SS;
+}
+
+/* Initialize the breakpoint table and register breakpoint commands. */
+
+void __init kdb_initbptab(void)
+{
+ int i;
+ kdb_bp_t *bp;
+
+ /*
+ * First time initialization.
+ */
+ memset(&kdb_breakpoints, '\0', sizeof(kdb_breakpoints));
+
+ for (i = 0, bp = kdb_breakpoints; i < KDB_MAXBPT; i++, bp++)
+ bp->bp_free = 1;
+
+ kdb_register_repeat("bp", kdb_bp, "[<vaddr>]",
+ "Set/Display breakpoints", 0, KDB_REPEAT_NO_ARGS);
+ kdb_register_repeat("bl", kdb_bp, "[<vaddr>]",
+ "Display breakpoints", 0, KDB_REPEAT_NO_ARGS);
+ if (arch_kgdb_ops.flags & KGDB_HW_BREAKPOINT)
+ kdb_register_repeat("bph", kdb_bp, "[<vaddr>]",
+ "[datar [length]|dataw [length]] Set hw brk", 0, KDB_REPEAT_NO_ARGS);
+ kdb_register_repeat("bc", kdb_bc, "<bpnum>",
+ "Clear Breakpoint", 0, KDB_REPEAT_NONE);
+ kdb_register_repeat("be", kdb_bc, "<bpnum>",
+ "Enable Breakpoint", 0, KDB_REPEAT_NONE);
+ kdb_register_repeat("bd", kdb_bc, "<bpnum>",
+ "Disable Breakpoint", 0, KDB_REPEAT_NONE);
+
+ kdb_register_repeat("ss", kdb_ss, "",
+ "Single Step", 1, KDB_REPEAT_NO_ARGS);
+ kdb_register_repeat("ssb", kdb_ss, "",
+ "Single step to branch/call", 0, KDB_REPEAT_NO_ARGS);
+ /*
+ * Architecture dependent initialization.
+ */
+}
diff --git a/kernel/debug/kdb/kdb_bt.c b/kernel/debug/kdb/kdb_bt.c
new file mode 100644
index 0000000..483fa4e
--- /dev/null
+++ b/kernel/debug/kdb/kdb_bt.c
@@ -0,0 +1,208 @@
+/*
+ * Kernel Debugger Architecture Independent Stack Traceback
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (c) 1999-2004 Silicon Graphics, Inc. All Rights Reserved.
+ * Copyright (c) 2009 Wind River Systems, Inc. All Rights Reserved.
+ */
+
+#include <linux/ctype.h>
+#include <linux/string.h>
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/kdb.h>
+#include <linux/nmi.h>
+#include <asm/system.h>
+#include "kdb_private.h"
+
+
+static void kdb_show_stack(struct task_struct *p, void *addr)
+{
+ int old_lvl = console_loglevel;
+ console_loglevel = 15;
+ kdb_set_current_task(p);
+ if (addr) {
+ show_stack((struct task_struct *)p, addr);
+ } else if (kdb_current_regs) {
+#ifdef CONFIG_X86
+ show_stack(p, &kdb_current_regs->sp);
+#else
+ show_stack(p, NULL);
+#endif
+ } else {
+ show_stack(p, NULL);
+ }
+ console_loglevel = old_lvl;
+}
+
+/*
+ * kdb_bt
+ *
+ * This function implements the 'bt' command. Print a stack
+ * traceback.
+ *
+ * bt [<address-expression>] (addr-exp is for alternate stacks)
+ * btp <pid> Kernel stack for <pid>
+ * btt <address-expression> Kernel stack for task structure at
+ * <address-expression>
+ * bta [DRSTCZEUIMA] All useful processes, optionally
+ * filtered by state
+ * btc [<cpu>] The current process on one cpu,
+ * default is all cpus
+ *
+ * bt <address-expression> refers to a address on the stack, that location
+ * is assumed to contain a return address.
+ *
+ * btt <address-expression> refers to the address of a struct task.
+ *
+ * Inputs:
+ * argc argument count
+ * argv argument vector
+ * Outputs:
+ * None.
+ * Returns:
+ * zero for success, a kdb diagnostic if error
+ * Locking:
+ * none.
+ * Remarks:
+ * Backtrack works best when the code uses frame pointers. But even
+ * without frame pointers we should get a reasonable trace.
+ *
+ * mds comes in handy when examining the stack to do a manual traceback or
+ * to get a starting point for bt <address-expression>.
+ */
+
+static int
+kdb_bt1(struct task_struct *p, unsigned long mask,
+ int argcount, int btaprompt)
+{
+ char buffer[2];
+ if (kdb_getarea(buffer[0], (unsigned long)p) ||
+ kdb_getarea(buffer[0], (unsigned long)(p+1)-1))
+ return KDB_BADADDR;
+ if (!kdb_task_state(p, mask))
+ return 0;
+ kdb_printf("Stack traceback for pid %d\n", p->pid);
+ kdb_ps1(p);
+ kdb_show_stack(p, NULL);
+ if (btaprompt) {
+ kdb_getstr(buffer, sizeof(buffer),
+ "Enter <q> to end, <cr> to continue:");
+ if (buffer[0] == 'q') {
+ kdb_printf("\n");
+ return 1;
+ }
+ }
+ touch_nmi_watchdog();
+ return 0;
+}
+
+int
+kdb_bt(int argc, const char **argv)
+{
+ int diag;
+ int argcount = 5;
+ int btaprompt = 1;
+ int nextarg;
+ unsigned long addr;
+ long offset;
+
+ kdbgetintenv("BTARGS", &argcount); /* Arguments to print */
+ kdbgetintenv("BTAPROMPT", &btaprompt); /* Prompt after each
+ * proc in bta */
+
+ if (strcmp(argv[0], "bta") == 0) {
+ struct task_struct *g, *p;
+ unsigned long cpu;
+ unsigned long mask = kdb_task_state_string(argc ? argv[1] :
+ NULL);
+ if (argc == 0)
+ kdb_ps_suppressed();
+ /* Run the active tasks first */
+ for_each_online_cpu(cpu) {
+ p = kdb_curr_task(cpu);
+ if (kdb_bt1(p, mask, argcount, btaprompt))
+ return 0;
+ }
+ /* Now the inactive tasks */
+ kdb_do_each_thread(g, p) {
+ if (task_curr(p))
+ continue;
+ if (kdb_bt1(p, mask, argcount, btaprompt))
+ return 0;
+ } kdb_while_each_thread(g, p);
+ } else if (strcmp(argv[0], "btp") == 0) {
+ struct task_struct *p;
+ unsigned long pid;
+ if (argc != 1)
+ return KDB_ARGCOUNT;
+ diag = kdbgetularg((char *)argv[1], &pid);
+ if (diag)
+ return diag;
+ p = find_task_by_pid_ns(pid, &init_pid_ns);
+ if (p) {
+ kdb_set_current_task(p);
+ return kdb_bt1(p, ~0UL, argcount, 0);
+ }
+ kdb_printf("No process with pid == %ld found\n", pid);
+ return 0;
+ } else if (strcmp(argv[0], "btt") == 0) {
+ if (argc != 1)
+ return KDB_ARGCOUNT;
+ diag = kdbgetularg((char *)argv[1], &addr);
+ if (diag)
+ return diag;
+ kdb_set_current_task((struct task_struct *)addr);
+ return kdb_bt1((struct task_struct *)addr, ~0UL, argcount, 0);
+ } else if (strcmp(argv[0], "btc") == 0) {
+ unsigned long cpu = ~0;
+ struct task_struct *save_current_task = kdb_current_task;
+ char buf[80];
+ if (argc > 1)
+ return KDB_ARGCOUNT;
+ if (argc == 1) {
+ diag = kdbgetularg((char *)argv[1], &cpu);
+ if (diag)
+ return diag;
+ }
+ /* Recursive use of kdb_parse, do not use argv after
+ * this point */
+ argv = NULL;
+ if (cpu != ~0) {
+ if (cpu >= num_possible_cpus() || !cpu_online(cpu)) {
+ kdb_printf("no process for cpu %ld\n", cpu);
+ return 0;
+ }
+ sprintf(buf, "btt 0x%p\n", KDB_TSK(cpu));
+ kdb_parse(buf);
+ return 0;
+ }
+ kdb_printf("btc: cpu status: ");
+ kdb_parse("cpu\n");
+ for_each_online_cpu(cpu) {
+ sprintf(buf, "btt 0x%p\n", KDB_TSK(cpu));
+ kdb_parse(buf);
+ touch_nmi_watchdog();
+ }
+ kdb_set_current_task(save_current_task);
+ return 0;
+ } else {
+ if (argc) {
+ nextarg = 1;
+ diag = kdbgetaddrarg(argc, argv, &nextarg, &addr,
+ &offset, NULL);
+ if (diag)
+ return diag;
+ kdb_show_stack(kdb_current_task, (void *)addr);
+ return 0;
+ } else {
+ return kdb_bt1(kdb_current_task, ~0UL, argcount, 0);
+ }
+ }
+
+ /* NOTREACHED */
+ return 0;
+}
diff --git a/kernel/debug/kdb/kdb_cmds b/kernel/debug/kdb/kdb_cmds
new file mode 100644
index 0000000..56c88e4
--- /dev/null
+++ b/kernel/debug/kdb/kdb_cmds
@@ -0,0 +1,35 @@
+# Initial commands for kdb, alter to suit your needs.
+# These commands are executed in kdb_init() context, no SMP, no
+# processes. Commands that require process data (including stack or
+# registers) are not reliable this early. set and bp commands should
+# be safe. Global breakpoint commands affect each cpu as it is booted.
+
+# Standard debugging information for first level support, just type archkdb
+# or archkdbcpu or archkdbshort at the kdb prompt.
+
+defcmd dumpcommon "" "Common kdb debugging"
+ set BTAPROMPT 0
+ set LINES 10000
+ -summary
+ -cpu
+ -ps
+ -dmesg 600
+ -bt
+endefcmd
+
+defcmd dumpall "" "First line debugging"
+ set BTSYMARG 1
+ set BTARGS 9
+ pid R
+ -dumpcommon
+ -bta
+endefcmd
+
+defcmd dumpcpu "" "Same as dumpall but only tasks on cpus"
+ set BTSYMARG 1
+ set BTARGS 9
+ pid R
+ -dumpcommon
+ -btc
+endefcmd
+
diff --git a/kernel/debug/kdb/kdb_debugger.c b/kernel/debug/kdb/kdb_debugger.c
new file mode 100644
index 0000000..f024c0c
--- /dev/null
+++ b/kernel/debug/kdb/kdb_debugger.c
@@ -0,0 +1,159 @@
+/*
+ * Created by: Jason Wessel <[email protected]>
+ *
+ * Copyright (c) 2009 Wind River Systems, Inc. All Rights Reserved.
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#include <linux/kgdb.h>
+#include <linux/kdb.h>
+#include <linux/kdebug.h>
+#include "kdb_private.h"
+#include "../debug_core.h"
+
+/*
+ * KDB interface to KGDB internals
+ */
+get_char_func kdb_poll_funcs[] = {
+ dbg_io_get_char,
+ NULL,
+};
+
+int kdb_stub(struct kgdb_state *ks)
+{
+ int error = 0;
+ kdb_bp_t *bp;
+ unsigned long addr = kgdb_arch_pc(ks->ex_vector, ks->linux_regs);
+ kdb_reason_t reason = KDB_REASON_OOPS;
+ kdb_dbtrap_t db_result = KDB_DB_NOBPT;
+ int i;
+
+ if (KDB_STATE(REENTRY)) {
+ reason = KDB_REASON_SWITCH;
+ KDB_STATE_CLEAR(REENTRY);
+ addr = instruction_pointer(ks->linux_regs);
+ }
+ ks->pass_exception = 0;
+ if (atomic_read(&kgdb_setting_breakpoint))
+ reason = KDB_REASON_KEYBOARD;
+
+ for (i = 0, bp = kdb_breakpoints; i < KDB_MAXBPT; i++, bp++) {
+ if ((bp->bp_enabled) && (bp->bp_addr == addr)) {
+ reason = KDB_REASON_BREAK;
+ db_result = KDB_DB_BPT;
+ if (addr != instruction_pointer(ks->linux_regs))
+ kgdb_arch_set_pc(ks->linux_regs, addr);
+ break;
+ }
+ }
+ if (reason == KDB_REASON_BREAK || reason == KDB_REASON_SWITCH) {
+ for (i = 0, bp = kdb_breakpoints; i < KDB_MAXBPT; i++, bp++) {
+ if (bp->bp_free)
+ continue;
+ if (bp->bp_addr == addr) {
+ bp->bp_delay = 1;
+ bp->bp_delayed = 1;
+ /*
+ * SSBPT is set when the kernel debugger must single step a
+ * task in order to re-establish an instruction breakpoint
+ * which uses the instruction replacement mechanism. It is
+ * cleared by any action that removes the need to single-step
+ * the breakpoint.
+ */
+ reason = KDB_REASON_BREAK;
+ db_result = KDB_DB_BPT;
+ KDB_STATE_SET(SSBPT);
+ break;
+ }
+ }
+ }
+
+ if (reason != KDB_REASON_BREAK && ks->ex_vector == 0 &&
+ ks->signo == SIGTRAP) {
+ reason = KDB_REASON_SSTEP;
+ db_result = KDB_DB_BPT;
+ }
+ /* Set initial kdb state variables */
+ KDB_STATE_CLEAR(KGDB_TRANS);
+ kdb_initial_cpu = ks->cpu;
+ kdb_current_task = kgdb_info[ks->cpu].task;
+ kdb_current_regs = kgdb_info[ks->cpu].debuggerinfo;
+ /* Remove any breakpoints as needed by kdb and clear single step */
+ kdb_bp_remove();
+ KDB_STATE_CLEAR(DOING_SS);
+ KDB_STATE_CLEAR(DOING_SSB);
+ /* zero out any offline cpu data */
+ for_each_present_cpu(i) {
+ if (!cpu_online(i)) {
+ kgdb_info[i].debuggerinfo = NULL;
+ kgdb_info[i].task = NULL;
+ }
+ }
+ if (ks->err_code == DIE_OOPS || reason == KDB_REASON_OOPS) {
+ ks->pass_exception = 1;
+ KDB_FLAG_SET(CATASTROPHIC);
+ }
+ kdb_initial_cpu = ks->cpu;
+ if (KDB_STATE(SSBPT) && reason == KDB_REASON_SSTEP) {
+ KDB_STATE_CLEAR(SSBPT);
+ KDB_STATE_CLEAR(DOING_SS);
+ } else {
+ /* Start kdb main loop */
+ error = kdb_main_loop(KDB_REASON_ENTER, reason,
+ ks->err_code, db_result, ks->linux_regs);
+ }
+ /*
+ * Upon exit from the kdb main loop setup break points and restart
+ * the system based on the requested continue state
+ */
+ kdb_initial_cpu = -1;
+ kdb_current_task = NULL;
+ kdb_current_regs = NULL;
+ kdbnearsym_cleanup();
+ if (error == KDB_CMD_KGDB) {
+ if (KDB_STATE(DOING_KGDB) || KDB_STATE(DOING_KGDB2)) {
+ /*
+ * This inteface glue which allows kdb to transition in into
+ * the gdb stub. In order to do this the '?' or '' gdb serial
+ * packet response is processed here. And then control is
+ * passed to the gdbstub.
+ */
+ if (KDB_STATE(DOING_KGDB))
+ gdbstub_state(ks, "?");
+ else
+ gdbstub_state(ks, "");
+ KDB_STATE_CLEAR(DOING_KGDB);
+ KDB_STATE_CLEAR(DOING_KGDB2);
+ }
+ return DBG_PASS_EVENT;
+ }
+ kdb_bp_install(ks->linux_regs);
+ dbg_activate_sw_breakpoints();
+ /* Set the exit state to a single step or a continue */
+ if (KDB_STATE(DOING_SS))
+ gdbstub_state(ks, "s");
+ else
+ gdbstub_state(ks, "c");
+
+ KDB_FLAG_CLEAR(CATASTROPHIC);
+
+ /* Invoke arch specific exception handling prior to system resume */
+ kgdb_info[ks->cpu].ret_state = gdbstub_state(ks, "e");
+ if (ks->pass_exception)
+ kgdb_info[ks->cpu].ret_state = 1;
+ if (error == KDB_CMD_CPU) {
+ KDB_STATE_SET(REENTRY);
+ /*
+ * Force clear the single step bit because kdb emulates this
+ * differently vs the gdbstub
+ */
+ kgdb_single_step = 0;
+ dbg_deactivate_sw_breakpoints();
+ return DBG_SWITCH_CPU_EVENT;
+ }
+ return kgdb_info[ks->cpu].ret_state;
+}
+
diff --git a/kernel/debug/kdb/kdb_io.c b/kernel/debug/kdb/kdb_io.c
new file mode 100644
index 0000000..9e3cec7
--- /dev/null
+++ b/kernel/debug/kdb/kdb_io.c
@@ -0,0 +1,789 @@
+/*
+ * Kernel Debugger Architecture Independent Console I/O handler
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (c) 1999-2006 Silicon Graphics, Inc. All Rights Reserved.
+ * Copyright (c) 2009 Wind River Systems, Inc. All Rights Reserved.
+ */
+
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/ctype.h>
+#include <linux/kernel.h>
+#include <linux/init.h>
+#include <linux/kdev_t.h>
+#include <linux/console.h>
+#include <linux/string.h>
+#include <linux/sched.h>
+#include <linux/smp.h>
+#include <linux/nmi.h>
+#include <linux/delay.h>
+#include <linux/kdb.h>
+#include <linux/kallsyms.h>
+#include "kdb_private.h"
+
+#define CMD_BUFLEN 256
+char kdb_prompt_str[CMD_BUFLEN];
+
+
+static void kgdb_transition_check(char *buffer)
+{
+ int slen = strlen(buffer);
+ if (strncmp(buffer, "$?#3f", slen) != 0 &&
+ strncmp(buffer, "$qSupported#37", slen) != 0 &&
+ strncmp(buffer, "+$qSupported#37", slen) != 0) {
+ KDB_STATE_SET(KGDB_TRANS);
+ kdb_printf("%s", buffer);
+ }
+}
+
+static int kdb_read_get_key(char *buffer, size_t bufsize)
+{
+#define ESCAPE_UDELAY 1000
+#define ESCAPE_DELAY (2*1000000/ESCAPE_UDELAY) /* 2 seconds worth of udelays */
+ char escape_data[5]; /* longest vt100 escape sequence is 4 bytes */
+ char *ped = escape_data;
+ int escape_delay = 0;
+ get_char_func *f, *f_escape = NULL;
+ int key;
+
+ for (f = &kdb_poll_funcs[0]; ; ++f) {
+ if (*f == NULL) {
+ /* Reset NMI watchdog once per poll loop */
+ touch_nmi_watchdog();
+ f = &kdb_poll_funcs[0];
+ }
+ if (escape_delay == 2) {
+ *ped = '\0';
+ ped = escape_data;
+ --escape_delay;
+ }
+ if (escape_delay == 1) {
+ key = *ped++;
+ if (!*ped)
+ --escape_delay;
+ break;
+ }
+ key = (*f)();
+ if (key == -1) {
+ if (escape_delay) {
+ udelay(ESCAPE_UDELAY);
+ --escape_delay;
+ }
+ continue;
+ }
+ if (bufsize <= 2) {
+ if (key == '\r')
+ key = '\n';
+ *buffer++ = key;
+ *buffer = '\0';
+ return -1;
+ }
+ if (escape_delay == 0 && key == '\e') {
+ escape_delay = ESCAPE_DELAY;
+ ped = escape_data;
+ f_escape = f;
+ }
+ if (escape_delay) {
+ *ped++ = key;
+ if (f_escape != f) {
+ escape_delay = 2;
+ continue;
+ }
+ if (ped - escape_data == 1) {
+ /* \e */
+ continue;
+ } else if (ped - escape_data == 2) {
+ /* \e<something> */
+ if (key != '[')
+ escape_delay = 2;
+ continue;
+ } else if (ped - escape_data == 3) {
+ /* \e[<something> */
+ int mapkey = 0;
+ switch (key) {
+ case 'A': /* \e[A, up arrow */
+ mapkey = 16;
+ break;
+ case 'B': /* \e[B, down arrow */
+ mapkey = 14;
+ break;
+ case 'C': /* \e[C, right arrow */
+ mapkey = 6;
+ break;
+ case 'D': /* \e[D, left arrow */
+ mapkey = 2;
+ break;
+ case '1': /* dropthrough */
+ case '3': /* dropthrough */
+ /* \e[<1,3,4>], may be home, del, end */
+ case '4':
+ mapkey = -1;
+ break;
+ }
+ if (mapkey != -1) {
+ if (mapkey > 0) {
+ escape_data[0] = mapkey;
+ escape_data[1] = '\0';
+ }
+ escape_delay = 2;
+ }
+ continue;
+ } else if (ped - escape_data == 4) {
+ /* \e[<1,3,4><something> */
+ int mapkey = 0;
+ if (key == '~') {
+ switch (escape_data[2]) {
+ case '1': /* \e[1~, home */
+ mapkey = 1;
+ break;
+ case '3': /* \e[3~, del */
+ mapkey = 4;
+ break;
+ case '4': /* \e[4~, end */
+ mapkey = 5;
+ break;
+ }
+ }
+ if (mapkey > 0) {
+ escape_data[0] = mapkey;
+ escape_data[1] = '\0';
+ }
+ escape_delay = 2;
+ continue;
+ }
+ }
+ break; /* A key to process */
+ }
+ return key;
+}
+
+/*
+ * kdb_read
+ *
+ * This function reads a string of characters, terminated by
+ * a newline, or by reaching the end of the supplied buffer,
+ * from the current kernel debugger console device.
+ * Parameters:
+ * buffer - Address of character buffer to receive input characters.
+ * bufsize - size, in bytes, of the character buffer
+ * Returns:
+ * Returns a pointer to the buffer containing the received
+ * character string. This string will be terminated by a
+ * newline character.
+ * Locking:
+ * No locks are required to be held upon entry to this
+ * function. It is not reentrant - it relies on the fact
+ * that while kdb is running on only one "master debug" cpu.
+ * Remarks:
+ *
+ * The buffer size must be >= 2. A buffer size of 2 means that the caller only
+ * wants a single key.
+ *
+ * An escape key could be the start of a vt100 control sequence such as \e[D
+ * (left arrow) or it could be a character in its own right. The standard
+ * method for detecting the difference is to wait for 2 seconds to see if there
+ * are any other characters. kdb is complicated by the lack of a timer service
+ * (interrupts are off), by multiple input sources and by the need to sometimes
+ * return after just one key. Escape sequence processing has to be done as
+ * states in the polling loop.
+ */
+
+static char *kdb_read(char *buffer, size_t bufsize)
+{
+ char *cp = buffer;
+ char *bufend = buffer+bufsize-2; /* Reserve space for newline
+ * and null byte */
+ char *lastchar;
+ char *p_tmp;
+ char tmp;
+ static char tmpbuffer[CMD_BUFLEN];
+ int len = strlen(buffer);
+ int len_tmp;
+ int tab = 0;
+ int count;
+ int i;
+ int diag, dtab_count;
+ int key;
+
+
+ diag = kdbgetintenv("DTABCOUNT", &dtab_count);
+ if (diag)
+ dtab_count = 30;
+
+ if (len > 0) {
+ cp += len;
+ if (*(buffer+len-1) == '\n')
+ cp--;
+ }
+
+ lastchar = cp;
+ *cp = '\0';
+ kdb_printf("%s", buffer);
+poll_again:
+ key = kdb_read_get_key(buffer, bufsize);
+ if (key == -1)
+ return buffer;
+ if (key != 9)
+ tab = 0;
+ switch (key) {
+ case 8: /* backspace */
+ if (cp > buffer) {
+ if (cp < lastchar) {
+ memcpy(tmpbuffer, cp, lastchar - cp);
+ memcpy(cp-1, tmpbuffer, lastchar - cp);
+ }
+ *(--lastchar) = '\0';
+ --cp;
+ kdb_printf("\b%s \r", cp);
+ tmp = *cp;
+ *cp = '\0';
+ kdb_printf(kdb_prompt_str);
+ kdb_printf("%s", buffer);
+ *cp = tmp;
+ }
+ break;
+ case 13: /* enter */
+ *lastchar++ = '\n';
+ *lastchar++ = '\0';
+ kdb_printf("\n");
+ return buffer;
+ case 4: /* Del */
+ if (cp < lastchar) {
+ memcpy(tmpbuffer, cp+1, lastchar - cp - 1);
+ memcpy(cp, tmpbuffer, lastchar - cp - 1);
+ *(--lastchar) = '\0';
+ kdb_printf("%s \r", cp);
+ tmp = *cp;
+ *cp = '\0';
+ kdb_printf(kdb_prompt_str);
+ kdb_printf("%s", buffer);
+ *cp = tmp;
+ }
+ break;
+ case 1: /* Home */
+ if (cp > buffer) {
+ kdb_printf("\r");
+ kdb_printf(kdb_prompt_str);
+ cp = buffer;
+ }
+ break;
+ case 5: /* End */
+ if (cp < lastchar) {
+ kdb_printf("%s", cp);
+ cp = lastchar;
+ }
+ break;
+ case 2: /* Left */
+ if (cp > buffer) {
+ kdb_printf("\b");
+ --cp;
+ }
+ break;
+ case 14: /* Down */
+ memset(tmpbuffer, ' ',
+ strlen(kdb_prompt_str) + (lastchar-buffer));
+ *(tmpbuffer+strlen(kdb_prompt_str) +
+ (lastchar-buffer)) = '\0';
+ kdb_printf("\r%s\r", tmpbuffer);
+ *lastchar = (char)key;
+ *(lastchar+1) = '\0';
+ return lastchar;
+ case 6: /* Right */
+ if (cp < lastchar) {
+ kdb_printf("%c", *cp);
+ ++cp;
+ }
+ break;
+ case 16: /* Up */
+ memset(tmpbuffer, ' ',
+ strlen(kdb_prompt_str) + (lastchar-buffer));
+ *(tmpbuffer+strlen(kdb_prompt_str) +
+ (lastchar-buffer)) = '\0';
+ kdb_printf("\r%s\r", tmpbuffer);
+ *lastchar = (char)key;
+ *(lastchar+1) = '\0';
+ return lastchar;
+ case 9: /* Tab */
+ if (tab < 2)
+ ++tab;
+ p_tmp = buffer;
+ while (*p_tmp == ' ')
+ p_tmp++;
+ if (p_tmp > cp)
+ break;
+ memcpy(tmpbuffer, p_tmp, cp-p_tmp);
+ *(tmpbuffer + (cp-p_tmp)) = '\0';
+ p_tmp = strrchr(tmpbuffer, ' ');
+ if (p_tmp)
+ ++p_tmp;
+ else
+ p_tmp = tmpbuffer;
+ len = strlen(p_tmp);
+ count = kallsyms_symbol_complete(p_tmp,
+ sizeof(tmpbuffer) -
+ (p_tmp - tmpbuffer));
+ if (tab == 2 && count > 0) {
+ kdb_printf("\n%d symbols are found.", count);
+ if (count > dtab_count) {
+ count = dtab_count;
+ kdb_printf(" But only first %d symbols will"
+ " be printed.\nYou can change the"
+ " environment variable DTABCOUNT.",
+ count);
+ }
+ kdb_printf("\n");
+ for (i = 0; i < count; i++) {
+ if (kallsyms_symbol_next(p_tmp, i) < 0)
+ break;
+ kdb_printf("%s ", p_tmp);
+ *(p_tmp + len) = '\0';
+ }
+ if (i >= dtab_count)
+ kdb_printf("...");
+ kdb_printf("\n");
+ kdb_printf(kdb_prompt_str);
+ kdb_printf("%s", buffer);
+ } else if (tab != 2 && count > 0) {
+ len_tmp = strlen(p_tmp);
+ strncpy(p_tmp+len_tmp, cp, lastchar-cp+1);
+ len_tmp = strlen(p_tmp);
+ strncpy(cp, p_tmp+len, len_tmp-len + 1);
+ len = len_tmp - len;
+ kdb_printf("%s", cp);
+ cp += len;
+ lastchar += len;
+ }
+ kdb_nextline = 1; /* reset output line number */
+ break;
+ default:
+ if (key >= 32 && lastchar < bufend) {
+ if (cp < lastchar) {
+ memcpy(tmpbuffer, cp, lastchar - cp);
+ memcpy(cp+1, tmpbuffer, lastchar - cp);
+ *++lastchar = '\0';
+ *cp = key;
+ kdb_printf("%s\r", cp);
+ ++cp;
+ tmp = *cp;
+ *cp = '\0';
+ kdb_printf(kdb_prompt_str);
+ kdb_printf("%s", buffer);
+ *cp = tmp;
+ } else {
+ *++lastchar = '\0';
+ *cp++ = key;
+ /* The kgdb transition check will hide
+ * printed characters if we think that
+ * kgdb is connecting, until the check
+ * fails */
+ if (!KDB_STATE(KGDB_TRANS))
+ kgdb_transition_check(buffer);
+ else
+ kdb_printf("%c", key);
+ }
+ /* Special escape to kgdb */
+ if (lastchar - buffer >= 5 &&
+ strcmp(lastchar - 5, "$?#3f") == 0) {
+ strcpy(buffer, "kgdb");
+ KDB_STATE_SET(DOING_KGDB);
+ return buffer;
+ }
+ if (lastchar - buffer >= 14 &&
+ strcmp(lastchar - 14, "$qSupported#37") == 0) {
+ strcpy(buffer, "kgdb");
+ KDB_STATE_SET(DOING_KGDB2);
+ return buffer;
+ }
+ }
+ break;
+ }
+ goto poll_again;
+}
+
+/*
+ * kdb_getstr
+ *
+ * Print the prompt string and read a command from the
+ * input device.
+ *
+ * Parameters:
+ * buffer Address of buffer to receive command
+ * bufsize Size of buffer in bytes
+ * prompt Pointer to string to use as prompt string
+ * Returns:
+ * Pointer to command buffer.
+ * Locking:
+ * None.
+ * Remarks:
+ * For SMP kernels, the processor number will be
+ * substituted for %d, %x or %o in the prompt.
+ */
+
+char *kdb_getstr(char *buffer, size_t bufsize, char *prompt)
+{
+ if (prompt && kdb_prompt_str != prompt)
+ strncpy(kdb_prompt_str, prompt, CMD_BUFLEN);
+ kdb_printf(kdb_prompt_str);
+ kdb_nextline = 1; /* Prompt and input resets line number */
+ return kdb_read(buffer, bufsize);
+}
+
+/*
+ * kdb_input_flush
+ *
+ * Get rid of any buffered console input.
+ *
+ * Parameters:
+ * none
+ * Returns:
+ * nothing
+ * Locking:
+ * none
+ * Remarks:
+ * Call this function whenever you want to flush input. If there is any
+ * outstanding input, it ignores all characters until there has been no
+ * data for approximately 1ms.
+ */
+
+static void kdb_input_flush(void)
+{
+ get_char_func *f;
+ int res;
+ int flush_delay = 1;
+ while (flush_delay) {
+ flush_delay--;
+empty:
+ touch_nmi_watchdog();
+ for (f = &kdb_poll_funcs[0]; *f; ++f) {
+ res = (*f)();
+ if (res != -1) {
+ flush_delay = 1;
+ goto empty;
+ }
+ }
+ if (flush_delay)
+ mdelay(1);
+ }
+}
+
+/*
+ * kdb_printf
+ *
+ * Print a string to the output device(s).
+ *
+ * Parameters:
+ * printf-like format and optional args.
+ * Returns:
+ * 0
+ * Locking:
+ * None.
+ * Remarks:
+ * use 'kdbcons->write()' to avoid polluting 'log_buf' with
+ * kdb output.
+ *
+ * If the user is doing a cmd args | grep srch
+ * then kdb_grepping_flag is set.
+ * In that case we need to accumulate full lines (ending in \n) before
+ * searching for the pattern.
+ */
+
+static char kdb_buffer[256]; /* A bit too big to go on stack */
+static char *next_avail = kdb_buffer;
+static int size_avail;
+static int suspend_grep;
+
+/*
+ * search arg1 to see if it contains arg2
+ * (kdmain.c provides flags for ^pat and pat$)
+ *
+ * return 1 for found, 0 for not found
+ */
+static int kdb_search_string(char *searched, char *searchfor)
+{
+ char firstchar, *cp;
+ int len1, len2;
+
+ /* not counting the newline at the end of "searched" */
+ len1 = strlen(searched)-1;
+ len2 = strlen(searchfor);
+ if (len1 < len2)
+ return 0;
+ if (kdb_grep_leading && kdb_grep_trailing && len1 != len2)
+ return 0;
+ if (kdb_grep_leading) {
+ if (!strncmp(searched, searchfor, len2))
+ return 1;
+ } else if (kdb_grep_trailing) {
+ if (!strncmp(searched+len1-len2, searchfor, len2))
+ return 1;
+ } else {
+ firstchar = *searchfor;
+ cp = searched;
+ while ((cp = strchr(cp, firstchar))) {
+ if (!strncmp(cp, searchfor, len2))
+ return 1;
+ cp++;
+ }
+ }
+ return 0;
+}
+
+int kdb_printf(const char *fmt, ...)
+{
+ va_list ap;
+ int diag;
+ int linecount;
+ int logging, saved_loglevel = 0;
+ int got_printf_lock = 0;
+ int retlen = 0;
+ int fnd, len;
+ char *cp, *cp2, *cphold = NULL, replaced_byte = ' ';
+ char *moreprompt = "more> ";
+ struct console *c = console_drivers;
+ static DEFINE_SPINLOCK(kdb_printf_lock);
+ unsigned long uninitialized_var(flags);
+
+ preempt_disable();
+ /* Serialize kdb_printf if multiple cpus try to write at once.
+ * But if any cpu goes recursive in kdb, just print the output,
+ * even if it is interleaved with any other text.
+ */
+ if (!KDB_STATE(PRINTF_LOCK)) {
+ KDB_STATE_SET(PRINTF_LOCK);
+ spin_lock_irqsave(&kdb_printf_lock, flags);
+ got_printf_lock = 1;
+ atomic_inc(&kdb_event);
+ } else {
+ __acquire(kdb_printf_lock);
+ }
+
+ diag = kdbgetintenv("LINES", &linecount);
+ if (diag || linecount <= 1)
+ linecount = 24;
+
+ diag = kdbgetintenv("LOGGING", &logging);
+ if (diag)
+ logging = 0;
+
+ if (!kdb_grepping_flag || suspend_grep) {
+ /* normally, every vsnprintf starts a new buffer */
+ next_avail = kdb_buffer;
+ size_avail = sizeof(kdb_buffer);
+ }
+ va_start(ap, fmt);
+ vsnprintf(next_avail, size_avail, fmt, ap);
+ va_end(ap);
+
+ /*
+ * If kdb_parse() found that the command was cmd xxx | grep yyy
+ * then kdb_grepping_flag is set, and kdb_grep_string contains yyy
+ *
+ * Accumulate the print data up to a newline before searching it.
+ * (vsnprintf does null-terminate the string that it generates)
+ */
+
+ /* skip the search if prints are temporarily unconditional */
+ if (!suspend_grep && kdb_grepping_flag) {
+ cp = strchr(kdb_buffer, '\n');
+ if (!cp) {
+ /*
+ * Special cases that don't end with newlines
+ * but should be written without one:
+ * The "[nn]kdb> " prompt should
+ * appear at the front of the buffer.
+ *
+ * The "[nn]more " prompt should also be
+ * (MOREPROMPT -> moreprompt)
+ * written * but we print that ourselves,
+ * we set the suspend_grep flag to make
+ * it unconditional.
+ *
+ */
+ if (next_avail == kdb_buffer) {
+ /*
+ * these should occur after a newline,
+ * so they will be at the front of the
+ * buffer
+ */
+ cp2 = kdb_buffer;
+ len = strlen(kdb_prompt_str);
+ if (!strncmp(cp2, kdb_prompt_str, len)) {
+ /*
+ * We're about to start a new
+ * command, so we can go back
+ * to normal mode.
+ */
+ kdb_grepping_flag = 0;
+ goto kdb_printit;
+ }
+ }
+ /* no newline; don't search/write the buffer
+ until one is there */
+ len = strlen(kdb_buffer);
+ next_avail = kdb_buffer + len;
+ size_avail = sizeof(kdb_buffer) - len;
+ goto kdb_print_out;
+ }
+
+ /*
+ * The newline is present; print through it or discard
+ * it, depending on the results of the search.
+ */
+ cp++; /* to byte after the newline */
+ replaced_byte = *cp; /* remember what/where it was */
+ cphold = cp;
+ *cp = '\0'; /* end the string for our search */
+
+ /*
+ * We now have a newline at the end of the string
+ * Only continue with this output if it contains the
+ * search string.
+ */
+ fnd = kdb_search_string(kdb_buffer, kdb_grep_string);
+ if (!fnd) {
+ /*
+ * At this point the complete line at the start
+ * of kdb_buffer can be discarded, as it does
+ * not contain what the user is looking for.
+ * Shift the buffer left.
+ */
+ *cphold = replaced_byte;
+ strcpy(kdb_buffer, cphold);
+ len = strlen(kdb_buffer);
+ next_avail = kdb_buffer + len;
+ size_avail = sizeof(kdb_buffer) - len;
+ goto kdb_print_out;
+ }
+ /*
+ * at this point the string is a full line and
+ * should be printed, up to the null.
+ */
+ }
+kdb_printit:
+
+ /*
+ * Write to all consoles.
+ */
+ retlen = strlen(kdb_buffer);
+ while (c) {
+ c->write(c, kdb_buffer, retlen);
+ touch_nmi_watchdog();
+ c = c->next;
+ }
+ if (logging) {
+ saved_loglevel = console_loglevel;
+ console_loglevel = 0;
+ printk(KERN_INFO "%s", kdb_buffer);
+ }
+
+ if (KDB_STATE(PAGER) && strchr(kdb_buffer, '\n'))
+ kdb_nextline++;
+
+ /* check for having reached the LINES number of printed lines */
+ if (kdb_nextline == linecount) {
+ char buf1[16] = "";
+#if defined(CONFIG_SMP)
+ char buf2[32];
+#endif
+
+ /* Watch out for recursion here. Any routine that calls
+ * kdb_printf will come back through here. And kdb_read
+ * uses kdb_printf to echo on serial consoles ...
+ */
+ kdb_nextline = 1; /* In case of recursion */
+
+ /*
+ * Pause until cr.
+ */
+ moreprompt = kdbgetenv("MOREPROMPT");
+ if (moreprompt == NULL)
+ moreprompt = "more> ";
+
+#if defined(CONFIG_SMP)
+ if (strchr(moreprompt, '%')) {
+ sprintf(buf2, moreprompt, get_cpu());
+ put_cpu();
+ moreprompt = buf2;
+ }
+#endif
+
+ kdb_input_flush();
+ c = console_drivers;
+
+ while (c) {
+ c->write(c, moreprompt, strlen(moreprompt));
+ touch_nmi_watchdog();
+ c = c->next;
+ }
+
+ if (logging)
+ printk("%s", moreprompt);
+
+ kdb_read(buf1, 2); /* '2' indicates to return
+ * immediately after getting one key. */
+ kdb_nextline = 1; /* Really set output line 1 */
+
+ /* empty and reset the buffer: */
+ kdb_buffer[0] = '\0';
+ next_avail = kdb_buffer;
+ size_avail = sizeof(kdb_buffer);
+ if ((buf1[0] == 'q') || (buf1[0] == 'Q')) {
+ /* user hit q or Q */
+ KDB_FLAG_SET(CMD_INTERRUPT); /* command interrupted */
+ KDB_STATE_CLEAR(PAGER);
+ /* end of command output; back to normal mode */
+ kdb_grepping_flag = 0;
+ kdb_printf("\n");
+ } else if (buf1[0] == ' ') {
+ kdb_printf("\n");
+ suspend_grep = 1; /* for this recursion */
+ } else if (buf1[0] == '\n') {
+ kdb_nextline = linecount - 1;
+ kdb_printf("\r");
+ suspend_grep = 1; /* for this recursion */
+ } else if (buf1[0] && buf1[0] != '\n') {
+ /* user hit something other than enter */
+ suspend_grep = 1; /* for this recursion */
+ kdb_printf("\nOnly 'q' or 'Q' are processed at more "
+ "prompt, input ignored\n");
+ } else if (kdb_grepping_flag) {
+ /* user hit enter */
+ suspend_grep = 1; /* for this recursion */
+ kdb_printf("\n");
+ }
+ kdb_input_flush();
+ }
+
+ /*
+ * For grep searches, shift the printed string left.
+ * replaced_byte contains the character that was overwritten with
+ * the terminating null, and cphold points to the null.
+ * Then adjust the notion of available space in the buffer.
+ */
+ if (kdb_grepping_flag && !suspend_grep) {
+ *cphold = replaced_byte;
+ strcpy(kdb_buffer, cphold);
+ len = strlen(kdb_buffer);
+ next_avail = kdb_buffer + len;
+ size_avail = sizeof(kdb_buffer) - len;
+ }
+
+kdb_print_out:
+ suspend_grep = 0; /* end of what may have been a recursive call */
+ if (logging)
+ console_loglevel = saved_loglevel;
+ if (KDB_STATE(PRINTF_LOCK) && got_printf_lock) {
+ got_printf_lock = 0;
+ spin_unlock_irqrestore(&kdb_printf_lock, flags);
+ KDB_STATE_CLEAR(PRINTF_LOCK);
+ atomic_dec(&kdb_event);
+ } else {
+ __release(kdb_printf_lock);
+ }
+ preempt_enable();
+ return retlen;
+}
diff --git a/kernel/debug/kdb/kdb_main.c b/kernel/debug/kdb/kdb_main.c
new file mode 100644
index 0000000..a5b82cd
--- /dev/null
+++ b/kernel/debug/kdb/kdb_main.c
@@ -0,0 +1,2851 @@
+/*
+ * Kernel Debugger Architecture Independent Main Code
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1999-2004 Silicon Graphics, Inc. All Rights Reserved.
+ * Copyright (C) 2000 Stephane Eranian <[email protected]>
+ * Xscale (R) modifications copyright (C) 2003 Intel Corporation.
+ * Copyright (c) 2009 Wind River Systems, Inc. All Rights Reserved.
+ */
+
+#include <linux/ctype.h>
+#include <linux/string.h>
+#include <linux/kernel.h>
+#include <linux/reboot.h>
+#include <linux/sched.h>
+#include <linux/sysrq.h>
+#include <linux/smp.h>
+#include <linux/utsname.h>
+#include <linux/vmalloc.h>
+#include <linux/module.h>
+#include <linux/mm.h>
+#include <linux/init.h>
+#include <linux/kallsyms.h>
+#include <linux/kgdb.h>
+#include <linux/kdb.h>
+#include <linux/notifier.h>
+#include <linux/interrupt.h>
+#include <linux/delay.h>
+#include <linux/nmi.h>
+#include <linux/time.h>
+#include <linux/ptrace.h>
+#include <linux/sysctl.h>
+#include <linux/cpu.h>
+#include <linux/kdebug.h>
+#include <linux/proc_fs.h>
+#include <linux/uaccess.h>
+#include <linux/swap.h>
+#include "kdb_private.h"
+
+#define GREP_LEN 256
+char kdb_grep_string[GREP_LEN];
+int kdb_grepping_flag;
+EXPORT_SYMBOL(kdb_grepping_flag);
+int kdb_grep_leading;
+int kdb_grep_trailing;
+
+/*
+ * Kernel debugger state flags
+ */
+int kdb_flags;
+atomic_t kdb_event;
+
+/*
+ * kdb_lock protects updates to kdb_initial_cpu. Used to
+ * single thread processors through the kernel debugger.
+ */
+int kdb_initial_cpu = -1; /* cpu number that owns kdb */
+int kdb_nextline = 1;
+int kdb_state; /* General KDB state */
+
+struct task_struct *kdb_current_task;
+EXPORT_SYMBOL(kdb_current_task);
+struct pt_regs *kdb_current_regs;
+
+const char *kdb_diemsg;
+static int kdb_go_count;
+#ifdef CONFIG_KDB_CONTINUE_CATASTROPHIC
+static unsigned int kdb_continue_catastrophic =
+ CONFIG_KDB_CONTINUE_CATASTROPHIC;
+#else
+static unsigned int kdb_continue_catastrophic;
+#endif
+
+/* kdb_commands describes the available commands. */
+static kdbtab_t *kdb_commands;
+#define KDB_BASE_CMD_MAX 50
+static int kdb_max_commands = KDB_BASE_CMD_MAX;
+static kdbtab_t kdb_base_commands[50];
+#define for_each_kdbcmd(cmd, num) \
+ for ((cmd) = kdb_base_commands, (num) = 0; \
+ num < kdb_max_commands; \
+ num == KDB_BASE_CMD_MAX ? cmd = kdb_commands : cmd++, num++)
+
+typedef struct _kdbmsg {
+ int km_diag; /* kdb diagnostic */
+ char *km_msg; /* Corresponding message text */
+} kdbmsg_t;
+
+#define KDBMSG(msgnum, text) \
+ { KDB_##msgnum, text }
+
+static kdbmsg_t kdbmsgs[] = {
+ KDBMSG(NOTFOUND, "Command Not Found"),
+ KDBMSG(ARGCOUNT, "Improper argument count, see usage."),
+ KDBMSG(BADWIDTH, "Illegal value for BYTESPERWORD use 1, 2, 4 or 8, "
+ "8 is only allowed on 64 bit systems"),
+ KDBMSG(BADRADIX, "Illegal value for RADIX use 8, 10 or 16"),
+ KDBMSG(NOTENV, "Cannot find environment variable"),
+ KDBMSG(NOENVVALUE, "Environment variable should have value"),
+ KDBMSG(NOTIMP, "Command not implemented"),
+ KDBMSG(ENVFULL, "Environment full"),
+ KDBMSG(ENVBUFFULL, "Environment buffer full"),
+ KDBMSG(TOOMANYBPT, "Too many breakpoints defined"),
+#ifdef CONFIG_CPU_XSCALE
+ KDBMSG(TOOMANYDBREGS, "More breakpoints than ibcr registers defined"),
+#else
+ KDBMSG(TOOMANYDBREGS, "More breakpoints than db registers defined"),
+#endif
+ KDBMSG(DUPBPT, "Duplicate breakpoint address"),
+ KDBMSG(BPTNOTFOUND, "Breakpoint not found"),
+ KDBMSG(BADMODE, "Invalid IDMODE"),
+ KDBMSG(BADINT, "Illegal numeric value"),
+ KDBMSG(INVADDRFMT, "Invalid symbolic address format"),
+ KDBMSG(BADREG, "Invalid register name"),
+ KDBMSG(BADCPUNUM, "Invalid cpu number"),
+ KDBMSG(BADLENGTH, "Invalid length field"),
+ KDBMSG(NOBP, "No Breakpoint exists"),
+ KDBMSG(BADADDR, "Invalid address"),
+};
+#undef KDBMSG
+
+static const int __nkdb_err = sizeof(kdbmsgs) / sizeof(kdbmsg_t);
+
+
+/*
+ * Initial environment. This is all kept static and local to
+ * this file. We don't want to rely on the memory allocation
+ * mechanisms in the kernel, so we use a very limited allocate-only
+ * heap for new and altered environment variables. The entire
+ * environment is limited to a fixed number of entries (add more
+ * to __env[] if required) and a fixed amount of heap (add more to
+ * KDB_ENVBUFSIZE if required).
+ */
+
+static char *__env[] = {
+#if defined(CONFIG_SMP)
+ "PROMPT=[%d]kdb> ",
+ "MOREPROMPT=[%d]more> ",
+#else
+ "PROMPT=kdb> ",
+ "MOREPROMPT=more> ",
+#endif
+ "RADIX=16",
+ "MDCOUNT=8", /* lines of md output */
+ "BTARGS=9", /* 9 possible args in bt */
+ KDB_PLATFORM_ENV,
+ "DTABCOUNT=30",
+ "NOSECT=1",
+ (char *)0,
+ (char *)0,
+ (char *)0,
+ (char *)0,
+ (char *)0,
+ (char *)0,
+ (char *)0,
+ (char *)0,
+ (char *)0,
+ (char *)0,
+ (char *)0,
+ (char *)0,
+ (char *)0,
+ (char *)0,
+ (char *)0,
+ (char *)0,
+ (char *)0,
+ (char *)0,
+ (char *)0,
+ (char *)0,
+ (char *)0,
+ (char *)0,
+ (char *)0,
+};
+
+static const int __nenv = (sizeof(__env) / sizeof(char *));
+
+struct task_struct *kdb_curr_task(int cpu)
+{
+ struct task_struct *p = curr_task(cpu);
+#ifdef _TIF_MCA_INIT
+ if ((task_thread_info(p)->flags & _TIF_MCA_INIT) && KDB_TSK(cpu))
+ p = krp->p;
+#endif
+ return p;
+}
+
+/*
+ * kdbgetenv - This function will return the character string value of
+ * an environment variable.
+ * Parameters:
+ * match A character string representing an environment variable.
+ * Returns:
+ * NULL No environment variable matches 'match'
+ * char* Pointer to string value of environment variable.
+ */
+char *kdbgetenv(const char *match)
+{
+ char **ep = __env;
+ int matchlen = strlen(match);
+ int i;
+
+ for (i = 0; i < __nenv; i++) {
+ char *e = *ep++;
+
+ if (!e)
+ continue;
+
+ if ((strncmp(match, e, matchlen) == 0)
+ && ((e[matchlen] == '\0')
+ || (e[matchlen] == '='))) {
+ char *cp = strchr(e, '=');
+ return cp ? ++cp : "";
+ }
+ }
+ return NULL;
+}
+
+/*
+ * kdballocenv - This function is used to allocate bytes for
+ * environment entries.
+ * Parameters:
+ * match A character string representing a numeric value
+ * Outputs:
+ * *value the unsigned long representation of the env variable 'match'
+ * Returns:
+ * Zero on success, a kdb diagnostic on failure.
+ * Remarks:
+ * We use a static environment buffer (envbuffer) to hold the values
+ * of dynamically generated environment variables (see kdb_set). Buffer
+ * space once allocated is never free'd, so over time, the amount of space
+ * (currently 512 bytes) will be exhausted if env variables are changed
+ * frequently.
+ */
+static char *kdballocenv(size_t bytes)
+{
+#define KDB_ENVBUFSIZE 512
+ static char envbuffer[KDB_ENVBUFSIZE];
+ static int envbufsize;
+ char *ep = NULL;
+
+ if ((KDB_ENVBUFSIZE - envbufsize) >= bytes) {
+ ep = &envbuffer[envbufsize];
+ envbufsize += bytes;
+ }
+ return ep;
+}
+
+/*
+ * kdbgetulenv - This function will return the value of an unsigned
+ * long-valued environment variable.
+ * Parameters:
+ * match A character string representing a numeric value
+ * Outputs:
+ * *value the unsigned long represntation of the env variable 'match'
+ * Returns:
+ * Zero on success, a kdb diagnostic on failure.
+ */
+static int kdbgetulenv(const char *match, unsigned long *value)
+{
+ char *ep;
+
+ ep = kdbgetenv(match);
+ if (!ep)
+ return KDB_NOTENV;
+ if (strlen(ep) == 0)
+ return KDB_NOENVVALUE;
+
+ *value = simple_strtoul(ep, NULL, 0);
+
+ return 0;
+}
+
+/*
+ * kdbgetintenv - This function will return the value of an
+ * integer-valued environment variable.
+ * Parameters:
+ * match A character string representing an integer-valued env variable
+ * Outputs:
+ * *value the integer representation of the environment variable 'match'
+ * Returns:
+ * Zero on success, a kdb diagnostic on failure.
+ */
+int kdbgetintenv(const char *match, int *value)
+{
+ unsigned long val;
+ int diag;
+
+ diag = kdbgetulenv(match, &val);
+ if (!diag)
+ *value = (int) val;
+ return diag;
+}
+
+/*
+ * kdbgetularg - This function will convert a numeric string into an
+ * unsigned long value.
+ * Parameters:
+ * arg A character string representing a numeric value
+ * Outputs:
+ * *value the unsigned long represntation of arg.
+ * Returns:
+ * Zero on success, a kdb diagnostic on failure.
+ */
+int kdbgetularg(const char *arg, unsigned long *value)
+{
+ char *endp;
+ unsigned long val;
+
+ val = simple_strtoul(arg, &endp, 0);
+
+ if (endp == arg) {
+ /*
+ * Try base 16, for us folks too lazy to type the
+ * leading 0x...
+ */
+ val = simple_strtoul(arg, &endp, 16);
+ if (endp == arg)
+ return KDB_BADINT;
+ }
+
+ *value = val;
+
+ return 0;
+}
+
+/*
+ * kdb_set - This function implements the 'set' command. Alter an
+ * existing environment variable or create a new one.
+ */
+int kdb_set(int argc, const char **argv)
+{
+ int i;
+ char *ep;
+ size_t varlen, vallen;
+
+ /*
+ * we can be invoked two ways:
+ * set var=value argv[1]="var", argv[2]="value"
+ * set var = value argv[1]="var", argv[2]="=", argv[3]="value"
+ * - if the latter, shift 'em down.
+ */
+ if (argc == 3) {
+ argv[2] = argv[3];
+ argc--;
+ }
+
+ if (argc != 2)
+ return KDB_ARGCOUNT;
+
+ /*
+ * Check for internal variables
+ */
+ if (strcmp(argv[1], "KDBDEBUG") == 0) {
+ unsigned int debugflags;
+ char *cp;
+
+ debugflags = simple_strtoul(argv[2], &cp, 0);
+ if (cp == argv[2] || debugflags & ~KDB_DEBUG_FLAG_MASK) {
+ kdb_printf("kdb: illegal debug flags '%s'\n",
+ argv[2]);
+ return 0;
+ }
+ kdb_flags = (kdb_flags &
+ ~(KDB_DEBUG_FLAG_MASK << KDB_DEBUG_FLAG_SHIFT))
+ | (debugflags << KDB_DEBUG_FLAG_SHIFT);
+
+ return 0;
+ }
+
+ /*
+ * Tokenizer squashed the '=' sign. argv[1] is variable
+ * name, argv[2] = value.
+ */
+ varlen = strlen(argv[1]);
+ vallen = strlen(argv[2]);
+ ep = kdballocenv(varlen + vallen + 2);
+ if (ep == (char *)0)
+ return KDB_ENVBUFFULL;
+
+ sprintf(ep, "%s=%s", argv[1], argv[2]);
+
+ ep[varlen+vallen+1] = '\0';
+
+ for (i = 0; i < __nenv; i++) {
+ if (__env[i]
+ && ((strncmp(__env[i], argv[1], varlen) == 0)
+ && ((__env[i][varlen] == '\0')
+ || (__env[i][varlen] == '=')))) {
+ __env[i] = ep;
+ return 0;
+ }
+ }
+
+ /*
+ * Wasn't existing variable. Fit into slot.
+ */
+ for (i = 0; i < __nenv-1; i++) {
+ if (__env[i] == (char *)0) {
+ __env[i] = ep;
+ return 0;
+ }
+ }
+
+ return KDB_ENVFULL;
+}
+
+static int kdb_check_regs(void)
+{
+ if (!kdb_current_regs) {
+ kdb_printf("No current kdb registers."
+ " You may need to select another task\n");
+ return KDB_BADREG;
+ }
+ return 0;
+}
+
+/*
+ * kdbgetaddrarg - This function is responsible for parsing an
+ * address-expression and returning the value of the expression,
+ * symbol name, and offset to the caller.
+ *
+ * The argument may consist of a numeric value (decimal or
+ * hexidecimal), a symbol name, a register name (preceeded by the
+ * percent sign), an environment variable with a numeric value
+ * (preceeded by a dollar sign) or a simple arithmetic expression
+ * consisting of a symbol name, +/-, and a numeric constant value
+ * (offset).
+ * Parameters:
+ * argc - count of arguments in argv
+ * argv - argument vector
+ * *nextarg - index to next unparsed argument in argv[]
+ * regs - Register state at time of KDB entry
+ * Outputs:
+ * *value - receives the value of the address-expression
+ * *offset - receives the offset specified, if any
+ * *name - receives the symbol name, if any
+ * *nextarg - index to next unparsed argument in argv[]
+ * Returns:
+ * zero is returned on success, a kdb diagnostic code is
+ * returned on error.
+ */
+int kdbgetaddrarg(int argc, const char **argv, int *nextarg,
+ unsigned long *value, long *offset,
+ char **name)
+{
+ unsigned long addr;
+ unsigned long off = 0;
+ int positive;
+ int diag;
+ int found = 0;
+ char *symname;
+ char symbol = '\0';
+ char *cp;
+ kdb_symtab_t symtab;
+
+ /*
+ * Process arguments which follow the following syntax:
+ *
+ * symbol | numeric-address [+/- numeric-offset]
+ * %register
+ * $environment-variable
+ */
+
+ if (*nextarg > argc)
+ return KDB_ARGCOUNT;
+
+ symname = (char *)argv[*nextarg];
+
+ /*
+ * If there is no whitespace between the symbol
+ * or address and the '+' or '-' symbols, we
+ * remember the character and replace it with a
+ * null so the symbol/value can be properly parsed
+ */
+ cp = strpbrk(symname, "+-");
+ if (cp != NULL) {
+ symbol = *cp;
+ *cp++ = '\0';
+ }
+
+ if (symname[0] == '$') {
+ diag = kdbgetulenv(&symname[1], &addr);
+ if (diag)
+ return diag;
+ } else if (symname[0] == '%') {
+ diag = kdb_check_regs();
+ if (diag)
+ return diag;
+ /* Implement register values with % at a later time as it is
+ * arch optional.
+ */
+ return KDB_NOTIMP;
+ } else {
+ found = kdbgetsymval(symname, &symtab);
+ if (found) {
+ addr = symtab.sym_start;
+ } else {
+ diag = kdbgetularg(argv[*nextarg], &addr);
+ if (diag)
+ return diag;
+ }
+ }
+
+ if (!found)
+ found = kdbnearsym(addr, &symtab);
+
+ (*nextarg)++;
+
+ if (name)
+ *name = symname;
+ if (value)
+ *value = addr;
+ if (offset && name && *name)
+ *offset = addr - symtab.sym_start;
+
+ if ((*nextarg > argc)
+ && (symbol == '\0'))
+ return 0;
+
+ /*
+ * check for +/- and offset
+ */
+
+ if (symbol == '\0') {
+ if ((argv[*nextarg][0] != '+')
+ && (argv[*nextarg][0] != '-')) {
+ /*
+ * Not our argument. Return.
+ */
+ return 0;
+ } else {
+ positive = (argv[*nextarg][0] == '+');
+ (*nextarg)++;
+ }
+ } else
+ positive = (symbol == '+');
+
+ /*
+ * Now there must be an offset!
+ */
+ if ((*nextarg > argc)
+ && (symbol == '\0')) {
+ return KDB_INVADDRFMT;
+ }
+
+ if (!symbol) {
+ cp = (char *)argv[*nextarg];
+ (*nextarg)++;
+ }
+
+ diag = kdbgetularg(cp, &off);
+ if (diag)
+ return diag;
+
+ if (!positive)
+ off = -off;
+
+ if (offset)
+ *offset += off;
+
+ if (value)
+ *value += off;
+
+ return 0;
+}
+
+static void kdb_cmderror(int diag)
+{
+ int i;
+
+ if (diag >= 0) {
+ kdb_printf("no error detected (diagnostic is %d)\n", diag);
+ return;
+ }
+
+ for (i = 0; i < __nkdb_err; i++) {
+ if (kdbmsgs[i].km_diag == diag) {
+ kdb_printf("diag: %d: %s\n", diag, kdbmsgs[i].km_msg);
+ return;
+ }
+ }
+
+ kdb_printf("Unknown diag %d\n", -diag);
+}
+
+/*
+ * kdb_defcmd, kdb_defcmd2 - This function implements the 'defcmd'
+ * command which defines one command as a set of other commands,
+ * terminated by endefcmd. kdb_defcmd processes the initial
+ * 'defcmd' command, kdb_defcmd2 is invoked from kdb_parse for
+ * the following commands until 'endefcmd'.
+ * Inputs:
+ * argc argument count
+ * argv argument vector
+ * Returns:
+ * zero for success, a kdb diagnostic if error
+ */
+struct defcmd_set {
+ int count;
+ int usable;
+ char *name;
+ char *usage;
+ char *help;
+ char **command;
+};
+static struct defcmd_set *defcmd_set;
+static int defcmd_set_count;
+static int defcmd_in_progress;
+
+/* Forward references */
+static int kdb_exec_defcmd(int argc, const char **argv);
+
+static int kdb_defcmd2(const char *cmdstr, const char *argv0)
+{
+ struct defcmd_set *s = defcmd_set + defcmd_set_count - 1;
+ char **save_command = s->command;
+ if (strcmp(argv0, "endefcmd") == 0) {
+ defcmd_in_progress = 0;
+ if (!s->count)
+ s->usable = 0;
+ if (s->usable)
+ kdb_register(s->name, kdb_exec_defcmd,
+ s->usage, s->help, 0);
+ return 0;
+ }
+ if (!s->usable)
+ return KDB_NOTIMP;
+ s->command = kmalloc((s->count + 1) * sizeof(*(s->command)), GFP_KDB);
+ if (!s->command) {
+ kdb_printf("Could not allocate new kdb_defcmd table for %s\n",
+ cmdstr);
+ s->usable = 0;
+ return KDB_NOTIMP;
+ }
+ memcpy(s->command, save_command, s->count * sizeof(*(s->command)));
+ s->command[s->count++] = kdb_strdup(cmdstr, GFP_KDB);
+ kfree(save_command);
+ return 0;
+}
+
+static int kdb_defcmd(int argc, const char **argv)
+{
+ struct defcmd_set *save_defcmd_set = defcmd_set, *s;
+ if (defcmd_in_progress) {
+ kdb_printf("kdb: nested defcmd detected, assuming missing "
+ "endefcmd\n");
+ kdb_defcmd2("endefcmd", "endefcmd");
+ }
+ if (argc == 0) {
+ int i;
+ for (s = defcmd_set; s < defcmd_set + defcmd_set_count; ++s) {
+ kdb_printf("defcmd %s \"%s\" \"%s\"\n", s->name,
+ s->usage, s->help);
+ for (i = 0; i < s->count; ++i)
+ kdb_printf("%s", s->command[i]);
+ kdb_printf("endefcmd\n");
+ }
+ return 0;
+ }
+ if (argc != 3)
+ return KDB_ARGCOUNT;
+ defcmd_set = kmalloc((defcmd_set_count + 1) * sizeof(*defcmd_set),
+ GFP_KDB);
+ if (!defcmd_set) {
+ kdb_printf("Could not allocate new defcmd_set entry for %s\n",
+ argv[1]);
+ defcmd_set = save_defcmd_set;
+ return KDB_NOTIMP;
+ }
+ memcpy(defcmd_set, save_defcmd_set,
+ defcmd_set_count * sizeof(*defcmd_set));
+ kfree(save_defcmd_set);
+ s = defcmd_set + defcmd_set_count;
+ memset(s, 0, sizeof(*s));
+ s->usable = 1;
+ s->name = kdb_strdup(argv[1], GFP_KDB);
+ s->usage = kdb_strdup(argv[2], GFP_KDB);
+ s->help = kdb_strdup(argv[3], GFP_KDB);
+ if (s->usage[0] == '"') {
+ strcpy(s->usage, s->usage+1);
+ s->usage[strlen(s->usage)-1] = '\0';
+ }
+ if (s->help[0] == '"') {
+ strcpy(s->help, s->help+1);
+ s->help[strlen(s->help)-1] = '\0';
+ }
+ ++defcmd_set_count;
+ defcmd_in_progress = 1;
+ return 0;
+}
+
+/*
+ * kdb_exec_defcmd - Execute the set of commands associated with this
+ * defcmd name.
+ * Inputs:
+ * argc argument count
+ * argv argument vector
+ * Returns:
+ * zero for success, a kdb diagnostic if error
+ */
+static int kdb_exec_defcmd(int argc, const char **argv)
+{
+ int i, ret;
+ struct defcmd_set *s;
+ if (argc != 0)
+ return KDB_ARGCOUNT;
+ for (s = defcmd_set, i = 0; i < defcmd_set_count; ++i, ++s) {
+ if (strcmp(s->name, argv[0]) == 0)
+ break;
+ }
+ if (i == defcmd_set_count) {
+ kdb_printf("kdb_exec_defcmd: could not find commands for %s\n",
+ argv[0]);
+ return KDB_NOTIMP;
+ }
+ for (i = 0; i < s->count; ++i) {
+ /* Recursive use of kdb_parse, do not use argv after
+ * this point */
+ argv = NULL;
+ kdb_printf("[%s]kdb> %s\n", s->name, s->command[i]);
+ ret = kdb_parse(s->command[i]);
+ if (ret)
+ return ret;
+ }
+ return 0;
+}
+
+/* Command history */
+#define KDB_CMD_HISTORY_COUNT 32
+#define CMD_BUFLEN 200 /* kdb_printf: max printline
+ * size == 256 */
+static unsigned int cmd_head, cmd_tail;
+static unsigned int cmdptr;
+static char cmd_hist[KDB_CMD_HISTORY_COUNT][CMD_BUFLEN];
+static char cmd_cur[CMD_BUFLEN];
+
+/*
+ * The "str" argument may point to something like | grep xyz
+ */
+static void parse_grep(const char *str)
+{
+ int len;
+ char *cp = (char *)str, *cp2;
+
+ /* sanity check: we should have been called with the \ first */
+ if (*cp != '|')
+ return;
+ cp++;
+ while (isspace(*cp))
+ cp++;
+ if (strncmp(cp, "grep ", 5)) {
+ kdb_printf("invalid 'pipe', see grephelp\n");
+ return;
+ }
+ cp += 5;
+ while (isspace(*cp))
+ cp++;
+ cp2 = strchr(cp, '\n');
+ if (cp2)
+ *cp2 = '\0'; /* remove the trailing newline */
+ len = strlen(cp);
+ if (len == 0) {
+ kdb_printf("invalid 'pipe', see grephelp\n");
+ return;
+ }
+ /* now cp points to a nonzero length search string */
+ if (*cp == '"') {
+ /* allow it be "x y z" by removing the "'s - there must
+ be two of them */
+ cp++;
+ cp2 = strchr(cp, '"');
+ if (!cp2) {
+ kdb_printf("invalid quoted string, see grephelp\n");
+ return;
+ }
+ *cp2 = '\0'; /* end the string where the 2nd " was */
+ }
+ kdb_grep_leading = 0;
+ if (*cp == '^') {
+ kdb_grep_leading = 1;
+ cp++;
+ }
+ len = strlen(cp);
+ kdb_grep_trailing = 0;
+ if (*(cp+len-1) == '$') {
+ kdb_grep_trailing = 1;
+ *(cp+len-1) = '\0';
+ }
+ len = strlen(cp);
+ if (!len)
+ return;
+ if (len >= GREP_LEN) {
+ kdb_printf("search string too long\n");
+ return;
+ }
+ strcpy(kdb_grep_string, cp);
+ kdb_grepping_flag++;
+ return;
+}
+
+/*
+ * kdb_parse - Parse the command line, search the command table for a
+ * matching command and invoke the command function. This
+ * function may be called recursively, if it is, the second call
+ * will overwrite argv and cbuf. It is the caller's
+ * responsibility to save their argv if they recursively call
+ * kdb_parse().
+ * Parameters:
+ * cmdstr The input command line to be parsed.
+ * regs The registers at the time kdb was entered.
+ * Returns:
+ * Zero for success, a kdb diagnostic if failure.
+ * Remarks:
+ * Limited to 20 tokens.
+ *
+ * Real rudimentary tokenization. Basically only whitespace
+ * is considered a token delimeter (but special consideration
+ * is taken of the '=' sign as used by the 'set' command).
+ *
+ * The algorithm used to tokenize the input string relies on
+ * there being at least one whitespace (or otherwise useless)
+ * character between tokens as the character immediately following
+ * the token is altered in-place to a null-byte to terminate the
+ * token string.
+ */
+
+#define MAXARGC 20
+
+int kdb_parse(const char *cmdstr)
+{
+ static char *argv[MAXARGC];
+ static int argc;
+ static char cbuf[CMD_BUFLEN+2];
+ char *cp;
+ char *cpp, quoted;
+ kdbtab_t *tp;
+ int i, escaped, ignore_errors = 0, check_grep;
+
+ /*
+ * First tokenize the command string.
+ */
+ cp = (char *)cmdstr;
+ kdb_grepping_flag = check_grep = 0;
+
+ if (KDB_FLAG(CMD_INTERRUPT)) {
+ /* Previous command was interrupted, newline must not
+ * repeat the command */
+ KDB_FLAG_CLEAR(CMD_INTERRUPT);
+ KDB_STATE_SET(PAGER);
+ argc = 0; /* no repeat */
+ }
+
+ if (*cp != '\n' && *cp != '\0') {
+ argc = 0;
+ cpp = cbuf;
+ while (*cp) {
+ /* skip whitespace */
+ while (isspace(*cp))
+ cp++;
+ if ((*cp == '\0') || (*cp == '\n') ||
+ (*cp == '#' && !defcmd_in_progress))
+ break;
+ /* special case: check for | grep pattern */
+ if (*cp == '|') {
+ check_grep++;
+ break;
+ }
+ if (cpp >= cbuf + CMD_BUFLEN) {
+ kdb_printf("kdb_parse: command buffer "
+ "overflow, command ignored\n%s\n",
+ cmdstr);
+ return KDB_NOTFOUND;
+ }
+ if (argc >= MAXARGC - 1) {
+ kdb_printf("kdb_parse: too many arguments, "
+ "command ignored\n%s\n", cmdstr);
+ return KDB_NOTFOUND;
+ }
+ argv[argc++] = cpp;
+ escaped = 0;
+ quoted = '\0';
+ /* Copy to next unquoted and unescaped
+ * whitespace or '=' */
+ while (*cp && *cp != '\n' &&
+ (escaped || quoted || !isspace(*cp))) {
+ if (cpp >= cbuf + CMD_BUFLEN)
+ break;
+ if (escaped) {
+ escaped = 0;
+ *cpp++ = *cp++;
+ continue;
+ }
+ if (*cp == '\\') {
+ escaped = 1;
+ ++cp;
+ continue;
+ }
+ if (*cp == quoted)
+ quoted = '\0';
+ else if (*cp == '\'' || *cp == '"')
+ quoted = *cp;
+ *cpp = *cp++;
+ if (*cpp == '=' && !quoted)
+ break;
+ ++cpp;
+ }
+ *cpp++ = '\0'; /* Squash a ws or '=' character */
+ }
+ }
+ if (!argc)
+ return 0;
+ if (check_grep)
+ parse_grep(cp);
+ if (defcmd_in_progress) {
+ int result = kdb_defcmd2(cmdstr, argv[0]);
+ if (!defcmd_in_progress) {
+ argc = 0; /* avoid repeat on endefcmd */
+ *(argv[0]) = '\0';
+ }
+ return result;
+ }
+ if (argv[0][0] == '-' && argv[0][1] &&
+ (argv[0][1] < '0' || argv[0][1] > '9')) {
+ ignore_errors = 1;
+ ++argv[0];
+ }
+
+ for_each_kdbcmd(tp, i) {
+ if (tp->cmd_name) {
+ /*
+ * If this command is allowed to be abbreviated,
+ * check to see if this is it.
+ */
+
+ if (tp->cmd_minlen
+ && (strlen(argv[0]) <= tp->cmd_minlen)) {
+ if (strncmp(argv[0],
+ tp->cmd_name,
+ tp->cmd_minlen) == 0) {
+ break;
+ }
+ }
+
+ if (strcmp(argv[0], tp->cmd_name) == 0)
+ break;
+ }
+ }
+
+ /*
+ * If we don't find a command by this name, see if the first
+ * few characters of this match any of the known commands.
+ * e.g., md1c20 should match md.
+ */
+ if (i == kdb_max_commands) {
+ for_each_kdbcmd(tp, i) {
+ if (tp->cmd_name) {
+ if (strncmp(argv[0],
+ tp->cmd_name,
+ strlen(tp->cmd_name)) == 0) {
+ break;
+ }
+ }
+ }
+ }
+
+ if (i < kdb_max_commands) {
+ int result;
+ KDB_STATE_SET(CMD);
+ result = (*tp->cmd_func)(argc-1, (const char **)argv);
+ if (result && ignore_errors && result > KDB_CMD_GO)
+ result = 0;
+ KDB_STATE_CLEAR(CMD);
+ switch (tp->cmd_repeat) {
+ case KDB_REPEAT_NONE:
+ argc = 0;
+ if (argv[0])
+ *(argv[0]) = '\0';
+ break;
+ case KDB_REPEAT_NO_ARGS:
+ argc = 1;
+ if (argv[1])
+ *(argv[1]) = '\0';
+ break;
+ case KDB_REPEAT_WITH_ARGS:
+ break;
+ }
+ return result;
+ }
+
+ /*
+ * If the input with which we were presented does not
+ * map to an existing command, attempt to parse it as an
+ * address argument and display the result. Useful for
+ * obtaining the address of a variable, or the nearest symbol
+ * to an address contained in a register.
+ */
+ {
+ unsigned long value;
+ char *name = NULL;
+ long offset;
+ int nextarg = 0;
+
+ if (kdbgetaddrarg(0, (const char **)argv, &nextarg,
+ &value, &offset, &name)) {
+ return KDB_NOTFOUND;
+ }
+
+ kdb_printf("%s = ", argv[0]);
+ kdb_symbol_print(value, NULL, KDB_SP_DEFAULT);
+ kdb_printf("\n");
+ return 0;
+ }
+}
+
+
+static int handle_ctrl_cmd(char *cmd)
+{
+#define CTRL_P 16
+#define CTRL_N 14
+
+ /* initial situation */
+ if (cmd_head == cmd_tail)
+ return 0;
+ switch (*cmd) {
+ case CTRL_P:
+ if (cmdptr != cmd_tail)
+ cmdptr = (cmdptr-1) % KDB_CMD_HISTORY_COUNT;
+ strncpy(cmd_cur, cmd_hist[cmdptr], CMD_BUFLEN);
+ return 1;
+ case CTRL_N:
+ if (cmdptr != cmd_head)
+ cmdptr = (cmdptr+1) % KDB_CMD_HISTORY_COUNT;
+ strncpy(cmd_cur, cmd_hist[cmdptr], CMD_BUFLEN);
+ return 1;
+ }
+ return 0;
+}
+
+/*
+ * kdb_reboot - This function implements the 'reboot' command. Reboot
+ * the system immediately, or loop for ever on failure.
+ */
+static int kdb_reboot(int argc, const char **argv)
+{
+ emergency_restart();
+ kdb_printf("Hmm, kdb_reboot did not reboot, spinning here\n");
+ while (1)
+ cpu_relax();
+ /* NOTREACHED */
+ return 0;
+}
+
+static void kdb_dumpregs(struct pt_regs *regs)
+{
+ int old_lvl = console_loglevel;
+ console_loglevel = 15;
+ show_regs(regs);
+ kdb_printf("\n");
+ console_loglevel = old_lvl;
+}
+
+void kdb_set_current_task(struct task_struct *p)
+{
+ kdb_current_task = p;
+
+ if (kdb_task_has_cpu(p)) {
+ kdb_current_regs = KDB_TSKREGS(kdb_process_cpu(p));
+ return;
+ }
+ kdb_current_regs = NULL;
+}
+
+/*
+ * kdb_local - The main code for kdb. This routine is invoked on a
+ * specific processor, it is not global. The main kdb() routine
+ * ensures that only one processor at a time is in this routine.
+ * This code is called with the real reason code on the first
+ * entry to a kdb session, thereafter it is called with reason
+ * SWITCH, even if the user goes back to the original cpu.
+ * Inputs:
+ * reason The reason KDB was invoked
+ * error The hardware-defined error code
+ * regs The exception frame at time of fault/breakpoint.
+ * db_result Result code from the break or debug point.
+ * Returns:
+ * 0 KDB was invoked for an event which it wasn't responsible
+ * 1 KDB handled the event for which it was invoked.
+ * KDB_CMD_GO User typed 'go'.
+ * KDB_CMD_CPU User switched to another cpu.
+ * KDB_CMD_SS Single step.
+ * KDB_CMD_SSB Single step until branch.
+ */
+static int kdb_local(kdb_reason_t reason, int error, struct pt_regs *regs,
+ kdb_dbtrap_t db_result)
+{
+ char *cmdbuf;
+ int diag;
+ struct task_struct *kdb_current =
+ kdb_curr_task(raw_smp_processor_id());
+
+ KDB_DEBUG_STATE("kdb_local 1", reason);
+ kdb_go_count = 0;
+ if (reason == KDB_REASON_DEBUG) {
+ /* special case below */
+ } else {
+ kdb_printf("\nEntering kdb (current=0x%p, pid %d) ",
+ kdb_current, kdb_current->pid);
+#if defined(CONFIG_SMP)
+ kdb_printf("on processor %d ", raw_smp_processor_id());
+#endif
+ }
+
+ switch (reason) {
+ case KDB_REASON_DEBUG:
+ {
+ /*
+ * If re-entering kdb after a single step
+ * command, don't print the message.
+ */
+ switch (db_result) {
+ case KDB_DB_BPT:
+ kdb_printf("\nEntering kdb (0x%p, pid %d) ",
+ kdb_current, kdb_current->pid);
+#if defined(CONFIG_SMP)
+ kdb_printf("on processor %d ", raw_smp_processor_id());
+#endif
+ kdb_printf("due to Debug @ " kdb_machreg_fmt "\n",
+ instruction_pointer(regs));
+ break;
+ case KDB_DB_SSB:
+ /*
+ * In the midst of ssb command. Just return.
+ */
+ KDB_DEBUG_STATE("kdb_local 3", reason);
+ return KDB_CMD_SSB; /* Continue with SSB command */
+
+ break;
+ case KDB_DB_SS:
+ break;
+ case KDB_DB_SSBPT:
+ KDB_DEBUG_STATE("kdb_local 4", reason);
+ return 1; /* kdba_db_trap did the work */
+ default:
+ kdb_printf("kdb: Bad result from kdba_db_trap: %d\n",
+ db_result);
+ break;
+ }
+
+ }
+ break;
+ case KDB_REASON_ENTER:
+ if (KDB_STATE(KEYBOARD))
+ kdb_printf("due to Keyboard Entry\n");
+ else
+ kdb_printf("due to KDB_ENTER()\n");
+ break;
+ case KDB_REASON_KEYBOARD:
+ KDB_STATE_SET(KEYBOARD);
+ kdb_printf("due to Keyboard Entry\n");
+ break;
+ case KDB_REASON_ENTER_SLAVE:
+ /* drop through, slaves only get released via cpu switch */
+ case KDB_REASON_SWITCH:
+ kdb_printf("due to cpu switch\n");
+ break;
+ case KDB_REASON_OOPS:
+ kdb_printf("Oops: %s\n", kdb_diemsg);
+ kdb_printf("due to oops @ " kdb_machreg_fmt "\n",
+ instruction_pointer(regs));
+ kdb_dumpregs(regs);
+ break;
+ case KDB_REASON_NMI:
+ kdb_printf("due to NonMaskable Interrupt @ "
+ kdb_machreg_fmt "\n",
+ instruction_pointer(regs));
+ kdb_dumpregs(regs);
+ break;
+ case KDB_REASON_SSTEP:
+ case KDB_REASON_BREAK:
+ kdb_printf("due to %s @ " kdb_machreg_fmt "\n",
+ reason == KDB_REASON_BREAK ?
+ "Breakpoint" : "SS trap", instruction_pointer(regs));
+ /*
+ * Determine if this breakpoint is one that we
+ * are interested in.
+ */
+ if (db_result != KDB_DB_BPT) {
+ kdb_printf("kdb: error return from kdba_bp_trap: %d\n",
+ db_result);
+ KDB_DEBUG_STATE("kdb_local 6", reason);
+ return 0; /* Not for us, dismiss it */
+ }
+ break;
+ case KDB_REASON_RECURSE:
+ kdb_printf("due to Recursion @ " kdb_machreg_fmt "\n",
+ instruction_pointer(regs));
+ break;
+ default:
+ kdb_printf("kdb: unexpected reason code: %d\n", reason);
+ KDB_DEBUG_STATE("kdb_local 8", reason);
+ return 0; /* Not for us, dismiss it */
+ }
+
+ while (1) {
+ /*
+ * Initialize pager context.
+ */
+ kdb_nextline = 1;
+ KDB_STATE_CLEAR(SUPPRESS);
+
+ cmdbuf = cmd_cur;
+ *cmdbuf = '\0';
+ *(cmd_hist[cmd_head]) = '\0';
+
+ if (KDB_FLAG(ONLY_DO_DUMP)) {
+ /* kdb is off but a catastrophic error requires a dump.
+ * Take the dump and reboot.
+ * Turn on logging so the kdb output appears in the log
+ * buffer in the dump.
+ */
+ const char *setargs[] = { "set", "LOGGING", "1" };
+ kdb_set(2, setargs);
+ kdb_reboot(0, NULL);
+ /*NOTREACHED*/
+ }
+
+do_full_getstr:
+#if defined(CONFIG_SMP)
+ snprintf(kdb_prompt_str, CMD_BUFLEN, kdbgetenv("PROMPT"),
+ raw_smp_processor_id());
+#else
+ snprintf(kdb_prompt_str, CMD_BUFLEN, kdbgetenv("PROMPT"));
+#endif
+ if (defcmd_in_progress)
+ strncat(kdb_prompt_str, "[defcmd]", CMD_BUFLEN);
+
+ /*
+ * Fetch command from keyboard
+ */
+ cmdbuf = kdb_getstr(cmdbuf, CMD_BUFLEN, kdb_prompt_str);
+ if (*cmdbuf != '\n') {
+ if (*cmdbuf < 32) {
+ if (cmdptr == cmd_head) {
+ strncpy(cmd_hist[cmd_head], cmd_cur,
+ CMD_BUFLEN);
+ *(cmd_hist[cmd_head] +
+ strlen(cmd_hist[cmd_head])-1) = '\0';
+ }
+ if (!handle_ctrl_cmd(cmdbuf))
+ *(cmd_cur+strlen(cmd_cur)-1) = '\0';
+ cmdbuf = cmd_cur;
+ goto do_full_getstr;
+ } else {
+ strncpy(cmd_hist[cmd_head], cmd_cur,
+ CMD_BUFLEN);
+ }
+
+ cmd_head = (cmd_head+1) % KDB_CMD_HISTORY_COUNT;
+ if (cmd_head == cmd_tail)
+ cmd_tail = (cmd_tail+1) % KDB_CMD_HISTORY_COUNT;
+ }
+
+ cmdptr = cmd_head;
+ diag = kdb_parse(cmdbuf);
+ if (diag == KDB_NOTFOUND) {
+ kdb_printf("Unknown kdb command: '%s'\n", cmdbuf);
+ diag = 0;
+ }
+ if (diag == KDB_CMD_GO
+ || diag == KDB_CMD_CPU
+ || diag == KDB_CMD_SS
+ || diag == KDB_CMD_SSB
+ || diag == KDB_CMD_KGDB)
+ break;
+
+ if (diag)
+ kdb_cmderror(diag);
+ }
+ KDB_DEBUG_STATE("kdb_local 9", diag);
+ return diag;
+}
+
+
+/*
+ * kdb_print_state - Print the state data for the current processor
+ * for debugging.
+ * Inputs:
+ * text Identifies the debug point
+ * value Any integer value to be printed, e.g. reason code.
+ */
+void kdb_print_state(const char *text, int value)
+{
+ kdb_printf("state: %s cpu %d value %d initial %d state %x\n",
+ text, raw_smp_processor_id(), value, kdb_initial_cpu,
+ kdb_state);
+}
+
+/*
+ * kdb_main_loop - After initial setup and assignment of the
+ * controlling cpu, all cpus are in this loop. One cpu is in
+ * control and will issue the kdb prompt, the others will spin
+ * until 'go' or cpu switch.
+ *
+ * To get a consistent view of the kernel stacks for all
+ * processes, this routine is invoked from the main kdb code via
+ * an architecture specific routine. kdba_main_loop is
+ * responsible for making the kernel stacks consistent for all
+ * processes, there should be no difference between a blocked
+ * process and a running process as far as kdb is concerned.
+ * Inputs:
+ * reason The reason KDB was invoked
+ * error The hardware-defined error code
+ * reason2 kdb's current reason code.
+ * Initially error but can change
+ * acording to kdb state.
+ * db_result Result code from break or debug point.
+ * regs The exception frame at time of fault/breakpoint.
+ * should always be valid.
+ * Returns:
+ * 0 KDB was invoked for an event which it wasn't responsible
+ * 1 KDB handled the event for which it was invoked.
+ */
+int kdb_main_loop(kdb_reason_t reason, kdb_reason_t reason2, int error,
+ kdb_dbtrap_t db_result, struct pt_regs *regs)
+{
+ int result = 1;
+ /* Stay in kdb() until 'go', 'ss[b]' or an error */
+ while (1) {
+ /*
+ * All processors except the one that is in control
+ * will spin here.
+ */
+ KDB_DEBUG_STATE("kdb_main_loop 1", reason);
+ while (KDB_STATE(HOLD_CPU)) {
+ /* state KDB is turned off by kdb_cpu to see if the
+ * other cpus are still live, each cpu in this loop
+ * turns it back on.
+ */
+ if (!KDB_STATE(KDB))
+ KDB_STATE_SET(KDB);
+ }
+
+ KDB_STATE_CLEAR(SUPPRESS);
+ KDB_DEBUG_STATE("kdb_main_loop 2", reason);
+ if (KDB_STATE(LEAVING))
+ break; /* Another cpu said 'go' */
+ /* Still using kdb, this processor is in control */
+ result = kdb_local(reason2, error, regs, db_result);
+ KDB_DEBUG_STATE("kdb_main_loop 3", result);
+
+ if (result == KDB_CMD_CPU)
+ break;
+
+ if (result == KDB_CMD_SS) {
+ KDB_STATE_SET(DOING_SS);
+ break;
+ }
+
+ if (result == KDB_CMD_SSB) {
+ KDB_STATE_SET(DOING_SS);
+ KDB_STATE_SET(DOING_SSB);
+ break;
+ }
+
+ if (result == KDB_CMD_KGDB) {
+ if (!(KDB_STATE(DOING_KGDB) || KDB_STATE(DOING_KGDB2)))
+ kdb_printf("Entering please attach debugger "
+ "or use $D#44+ or $3#33\n");
+ break;
+ }
+ if (result && result != 1 && result != KDB_CMD_GO)
+ kdb_printf("\nUnexpected kdb_local return code %d\n",
+ result);
+ KDB_DEBUG_STATE("kdb_main_loop 4", reason);
+ break;
+ }
+ if (KDB_STATE(DOING_SS))
+ KDB_STATE_CLEAR(SSBPT);
+
+ return result;
+}
+
+/*
+ * kdb_mdr - This function implements the guts of the 'mdr', memory
+ * read command.
+ * mdr <addr arg>,<byte count>
+ * Inputs:
+ * addr Start address
+ * count Number of bytes
+ * Returns:
+ * Always 0. Any errors are detected and printed by kdb_getarea.
+ */
+static int kdb_mdr(unsigned long addr, unsigned int count)
+{
+ unsigned char c;
+ while (count--) {
+ if (kdb_getarea(c, addr))
+ return 0;
+ kdb_printf("%02x", c);
+ addr++;
+ }
+ kdb_printf("\n");
+ return 0;
+}
+
+/*
+ * kdb_md - This function implements the 'md', 'md1', 'md2', 'md4',
+ * 'md8' 'mdr' and 'mds' commands.
+ *
+ * md|mds [<addr arg> [<line count> [<radix>]]]
+ * mdWcN [<addr arg> [<line count> [<radix>]]]
+ * where W = is the width (1, 2, 4 or 8) and N is the count.
+ * for eg., md1c20 reads 20 bytes, 1 at a time.
+ * mdr <addr arg>,<byte count>
+ */
+static void kdb_md_line(const char *fmtstr, unsigned long addr,
+ int symbolic, int nosect, int bytesperword,
+ int num, int repeat, int phys)
+{
+ /* print just one line of data */
+ kdb_symtab_t symtab;
+ char cbuf[32];
+ char *c = cbuf;
+ int i;
+ unsigned long word;
+
+ memset(cbuf, '\0', sizeof(cbuf));
+ if (phys)
+ kdb_printf("phys " kdb_machreg_fmt0 " ", addr);
+ else
+ kdb_printf(kdb_machreg_fmt0 " ", addr);
+
+ for (i = 0; i < num && repeat--; i++) {
+ if (phys) {
+ if (kdb_getphysword(&word, addr, bytesperword))
+ break;
+ } else if (kdb_getword(&word, addr, bytesperword))
+ break;
+ kdb_printf(fmtstr, word);
+ if (symbolic)
+ kdbnearsym(word, &symtab);
+ else
+ memset(&symtab, 0, sizeof(symtab));
+ if (symtab.sym_name) {
+ kdb_symbol_print(word, &symtab, 0);
+ if (!nosect) {
+ kdb_printf("\n");
+ kdb_printf(" %s %s "
+ kdb_machreg_fmt " "
+ kdb_machreg_fmt " "
+ kdb_machreg_fmt, symtab.mod_name,
+ symtab.sec_name, symtab.sec_start,
+ symtab.sym_start, symtab.sym_end);
+ }
+ addr += bytesperword;
+ } else {
+ union {
+ u64 word;
+ unsigned char c[8];
+ } wc;
+ unsigned char *cp;
+#ifdef __BIG_ENDIAN
+ cp = wc.c + 8 - bytesperword;
+#else
+ cp = wc.c;
+#endif
+ wc.word = word;
+#define printable_char(c) \
+ ({unsigned char __c = c; isascii(__c) && isprint(__c) ? __c : '.'; })
+ switch (bytesperword) {
+ case 8:
+ *c++ = printable_char(*cp++);
+ *c++ = printable_char(*cp++);
+ *c++ = printable_char(*cp++);
+ *c++ = printable_char(*cp++);
+ addr += 4;
+ case 4:
+ *c++ = printable_char(*cp++);
+ *c++ = printable_char(*cp++);
+ addr += 2;
+ case 2:
+ *c++ = printable_char(*cp++);
+ addr++;
+ case 1:
+ *c++ = printable_char(*cp++);
+ addr++;
+ break;
+ }
+#undef printable_char
+ }
+ }
+ kdb_printf("%*s %s\n", (int)((num-i)*(2*bytesperword + 1)+1),
+ " ", cbuf);
+}
+
+static int kdb_md(int argc, const char **argv)
+{
+ static unsigned long last_addr;
+ static int last_radix, last_bytesperword, last_repeat;
+ int radix = 16, mdcount = 8, bytesperword = KDB_WORD_SIZE, repeat;
+ int nosect = 0;
+ char fmtchar, fmtstr[64];
+ unsigned long addr;
+ unsigned long word;
+ long offset = 0;
+ int symbolic = 0;
+ int valid = 0;
+ int phys = 0;
+
+ kdbgetintenv("MDCOUNT", &mdcount);
+ kdbgetintenv("RADIX", &radix);
+ kdbgetintenv("BYTESPERWORD", &bytesperword);
+
+ /* Assume 'md <addr>' and start with environment values */
+ repeat = mdcount * 16 / bytesperword;
+
+ if (strcmp(argv[0], "mdr") == 0) {
+ if (argc != 2)
+ return KDB_ARGCOUNT;
+ valid = 1;
+ } else if (isdigit(argv[0][2])) {
+ bytesperword = (int)(argv[0][2] - '0');
+ if (bytesperword == 0) {
+ bytesperword = last_bytesperword;
+ if (bytesperword == 0)
+ bytesperword = 4;
+ }
+ last_bytesperword = bytesperword;
+ repeat = mdcount * 16 / bytesperword;
+ if (!argv[0][3])
+ valid = 1;
+ else if (argv[0][3] == 'c' && argv[0][4]) {
+ char *p;
+ repeat = simple_strtoul(argv[0] + 4, &p, 10);
+ mdcount = ((repeat * bytesperword) + 15) / 16;
+ valid = !*p;
+ }
+ last_repeat = repeat;
+ } else if (strcmp(argv[0], "md") == 0)
+ valid = 1;
+ else if (strcmp(argv[0], "mds") == 0)
+ valid = 1;
+ else if (strcmp(argv[0], "mdp") == 0) {
+ phys = valid = 1;
+ }
+ if (!valid)
+ return KDB_NOTFOUND;
+
+ if (argc == 0) {
+ if (last_addr == 0)
+ return KDB_ARGCOUNT;
+ addr = last_addr;
+ radix = last_radix;
+ bytesperword = last_bytesperword;
+ repeat = last_repeat;
+ mdcount = ((repeat * bytesperword) + 15) / 16;
+ }
+
+ if (argc) {
+ unsigned long val;
+ int diag, nextarg = 1;
+ diag = kdbgetaddrarg(argc, argv, &nextarg, &addr,
+ &offset, NULL);
+ if (diag)
+ return diag;
+ if (argc > nextarg+2)
+ return KDB_ARGCOUNT;
+
+ if (argc >= nextarg) {
+ diag = kdbgetularg(argv[nextarg], &val);
+ if (!diag) {
+ mdcount = (int) val;
+ repeat = mdcount * 16 / bytesperword;
+ }
+ }
+ if (argc >= nextarg+1) {
+ diag = kdbgetularg(argv[nextarg+1], &val);
+ if (!diag)
+ radix = (int) val;
+ }
+ }
+
+ if (strcmp(argv[0], "mdr") == 0)
+ return kdb_mdr(addr, mdcount);
+
+ switch (radix) {
+ case 10:
+ fmtchar = 'd';
+ break;
+ case 16:
+ fmtchar = 'x';
+ break;
+ case 8:
+ fmtchar = 'o';
+ break;
+ default:
+ return KDB_BADRADIX;
+ }
+
+ last_radix = radix;
+
+ if (bytesperword > KDB_WORD_SIZE)
+ return KDB_BADWIDTH;
+
+ switch (bytesperword) {
+ case 8:
+ sprintf(fmtstr, "%%16.16l%c ", fmtchar);
+ break;
+ case 4:
+ sprintf(fmtstr, "%%8.8l%c ", fmtchar);
+ break;
+ case 2:
+ sprintf(fmtstr, "%%4.4l%c ", fmtchar);
+ break;
+ case 1:
+ sprintf(fmtstr, "%%2.2l%c ", fmtchar);
+ break;
+ default:
+ return KDB_BADWIDTH;
+ }
+
+ last_repeat = repeat;
+ last_bytesperword = bytesperword;
+
+ if (strcmp(argv[0], "mds") == 0) {
+ symbolic = 1;
+ /* Do not save these changes as last_*, they are temporary mds
+ * overrides.
+ */
+ bytesperword = KDB_WORD_SIZE;
+ repeat = mdcount;
+ kdbgetintenv("NOSECT", &nosect);
+ }
+
+ /* Round address down modulo BYTESPERWORD */
+
+ addr &= ~(bytesperword-1);
+
+ while (repeat > 0) {
+ unsigned long a;
+ int n, z, num = (symbolic ? 1 : (16 / bytesperword));
+
+ if (KDB_FLAG(CMD_INTERRUPT))
+ return 0;
+ for (a = addr, z = 0; z < repeat; a += bytesperword, ++z) {
+ if (phys) {
+ if (kdb_getphysword(&word, a, bytesperword)
+ || word)
+ break;
+ } else if (kdb_getword(&word, a, bytesperword) || word)
+ break;
+ }
+ n = min(num, repeat);
+ kdb_md_line(fmtstr, addr, symbolic, nosect, bytesperword,
+ num, repeat, phys);
+ addr += bytesperword * n;
+ repeat -= n;
+ z = (z + num - 1) / num;
+ if (z > 2) {
+ int s = num * (z-2);
+ kdb_printf(kdb_machreg_fmt0 "-" kdb_machreg_fmt0
+ " zero suppressed\n",
+ addr, addr + bytesperword * s - 1);
+ addr += bytesperword * s;
+ repeat -= s;
+ }
+ }
+ last_addr = addr;
+
+ return 0;
+}
+
+/*
+ * kdb_mm - This function implements the 'mm' command.
+ * mm address-expression new-value
+ * Remarks:
+ * mm works on machine words, mmW works on bytes.
+ */
+static int kdb_mm(int argc, const char **argv)
+{
+ int diag;
+ unsigned long addr;
+ long offset = 0;
+ unsigned long contents;
+ int nextarg;
+ int width;
+
+ if (argv[0][2] && !isdigit(argv[0][2]))
+ return KDB_NOTFOUND;
+
+ if (argc < 2)
+ return KDB_ARGCOUNT;
+
+ nextarg = 1;
+ diag = kdbgetaddrarg(argc, argv, &nextarg, &addr, &offset, NULL);
+ if (diag)
+ return diag;
+
+ if (nextarg > argc)
+ return KDB_ARGCOUNT;
+ diag = kdbgetaddrarg(argc, argv, &nextarg, &contents, NULL, NULL);
+ if (diag)
+ return diag;
+
+ if (nextarg != argc + 1)
+ return KDB_ARGCOUNT;
+
+ width = argv[0][2] ? (argv[0][2] - '0') : (KDB_WORD_SIZE);
+ diag = kdb_putword(addr, contents, width);
+ if (diag)
+ return diag;
+
+ kdb_printf(kdb_machreg_fmt " = " kdb_machreg_fmt "\n", addr, contents);
+
+ return 0;
+}
+
+/*
+ * kdb_go - This function implements the 'go' command.
+ * go [address-expression]
+ */
+static int kdb_go(int argc, const char **argv)
+{
+ unsigned long addr;
+ int diag;
+ int nextarg;
+ long offset;
+
+ if (argc == 1) {
+ if (raw_smp_processor_id() != kdb_initial_cpu) {
+ kdb_printf("go <address> must be issued from the "
+ "initial cpu, do cpu %d first\n",
+ kdb_initial_cpu);
+ return KDB_ARGCOUNT;
+ }
+ nextarg = 1;
+ diag = kdbgetaddrarg(argc, argv, &nextarg,
+ &addr, &offset, NULL);
+ if (diag)
+ return diag;
+ } else if (argc) {
+ return KDB_ARGCOUNT;
+ }
+
+ diag = KDB_CMD_GO;
+ if (KDB_FLAG(CATASTROPHIC)) {
+ kdb_printf("Catastrophic error detected\n");
+ kdb_printf("kdb_continue_catastrophic=%d, ",
+ kdb_continue_catastrophic);
+ if (kdb_continue_catastrophic == 0 && kdb_go_count++ == 0) {
+ kdb_printf("type go a second time if you really want "
+ "to continue\n");
+ return 0;
+ }
+ if (kdb_continue_catastrophic == 2) {
+ kdb_printf("forcing reboot\n");
+ kdb_reboot(0, NULL);
+ }
+ kdb_printf("attempting to continue\n");
+ }
+ return diag;
+}
+
+/*
+ * kdb_rd - This function implements the 'rd' command.
+ */
+static int kdb_rd(int argc, const char **argv)
+{
+ int diag = kdb_check_regs();
+ if (diag)
+ return diag;
+
+ kdb_dumpregs(kdb_current_regs);
+ return 0;
+}
+
+/*
+ * kdb_rm - This function implements the 'rm' (register modify) command.
+ * rm register-name new-contents
+ * Remarks:
+ * Currently doesn't allow modification of control or
+ * debug registers.
+ */
+static int kdb_rm(int argc, const char **argv)
+{
+ int diag;
+ int ind = 0;
+ unsigned long contents;
+
+ if (argc != 2)
+ return KDB_ARGCOUNT;
+ /*
+ * Allow presence or absence of leading '%' symbol.
+ */
+ if (argv[1][0] == '%')
+ ind = 1;
+
+ diag = kdbgetularg(argv[2], &contents);
+ if (diag)
+ return diag;
+
+ diag = kdb_check_regs();
+ if (diag)
+ return diag;
+ kdb_printf("ERROR: Register set currently not implemented\n");
+ return 0;
+}
+
+#if defined(CONFIG_MAGIC_SYSRQ)
+/*
+ * kdb_sr - This function implements the 'sr' (SYSRQ key) command
+ * which interfaces to the soi-disant MAGIC SYSRQ functionality.
+ * sr <magic-sysrq-code>
+ */
+static int kdb_sr(int argc, const char **argv)
+{
+ if (argc != 1)
+ return KDB_ARGCOUNT;
+ if (!__sysrq_enabled) {
+ kdb_printf("Auto activating sysrq\n");
+ __sysrq_enabled = 1;
+ }
+
+ handle_sysrq(*argv[1], NULL);
+
+ return 0;
+}
+#endif /* CONFIG_MAGIC_SYSRQ */
+
+/*
+ * kdb_ef - This function implements the 'regs' (display exception
+ * frame) command. This command takes an address and expects to
+ * find an exception frame at that address, formats and prints
+ * it.
+ * regs address-expression
+ * Remarks:
+ * Not done yet.
+ */
+static int kdb_ef(int argc, const char **argv)
+{
+ int diag;
+ unsigned long addr;
+ long offset;
+ int nextarg;
+
+ if (argc != 1)
+ return KDB_ARGCOUNT;
+
+ nextarg = 1;
+ diag = kdbgetaddrarg(argc, argv, &nextarg, &addr, &offset, NULL);
+ if (diag)
+ return diag;
+ show_regs((struct pt_regs *)addr);
+ return 0;
+}
+
+#if defined(CONFIG_MODULES)
+/* modules using other modules */
+struct module_use {
+ struct list_head list;
+ struct module *module_which_uses;
+};
+
+/*
+ * kdb_lsmod - This function implements the 'lsmod' command. Lists
+ * currently loaded kernel modules.
+ * Mostly taken from userland lsmod.
+ */
+static int kdb_lsmod(int argc, const char **argv)
+{
+ struct module *mod;
+
+ if (argc != 0)
+ return KDB_ARGCOUNT;
+
+ kdb_printf("Module Size modstruct Used by\n");
+ list_for_each_entry(mod, kdb_modules, list) {
+
+ kdb_printf("%-20s%8u 0x%p ", mod->name,
+ mod->core_size, (void *)mod);
+#ifdef CONFIG_MODULE_UNLOAD
+ kdb_printf("%4d ", module_refcount(mod));
+#endif
+ if (mod->state == MODULE_STATE_GOING)
+ kdb_printf(" (Unloading)");
+ else if (mod->state == MODULE_STATE_COMING)
+ kdb_printf(" (Loading)");
+ else
+ kdb_printf(" (Live)");
+
+#ifdef CONFIG_MODULE_UNLOAD
+ {
+ struct module_use *use;
+ kdb_printf(" [ ");
+ list_for_each_entry(use, &mod->modules_which_use_me,
+ list)
+ kdb_printf("%s ", use->module_which_uses->name);
+ kdb_printf("]\n");
+ }
+#endif
+ }
+
+ return 0;
+}
+
+#endif /* CONFIG_MODULES */
+
+/*
+ * kdb_env - This function implements the 'env' command. Display the
+ * current environment variables.
+ */
+
+static int kdb_env(int argc, const char **argv)
+{
+ int i;
+
+ for (i = 0; i < __nenv; i++) {
+ if (__env[i])
+ kdb_printf("%s\n", __env[i]);
+ }
+
+ if (KDB_DEBUG(MASK))
+ kdb_printf("KDBFLAGS=0x%x\n", kdb_flags);
+
+ return 0;
+}
+
+#ifdef CONFIG_PRINTK
+/*
+ * kdb_dmesg - This function implements the 'dmesg' command to display
+ * the contents of the syslog buffer.
+ * dmesg [lines] [adjust]
+ */
+static int kdb_dmesg(int argc, const char **argv)
+{
+ char *syslog_data[4], *start, *end, c = '\0', *p;
+ int diag, logging, logsize, lines = 0, adjust = 0, n;
+
+ if (argc > 2)
+ return KDB_ARGCOUNT;
+ if (argc) {
+ char *cp;
+ lines = simple_strtol(argv[1], &cp, 0);
+ if (*cp)
+ lines = 0;
+ if (argc > 1) {
+ adjust = simple_strtoul(argv[2], &cp, 0);
+ if (*cp || adjust < 0)
+ adjust = 0;
+ }
+ }
+
+ /* disable LOGGING if set */
+ diag = kdbgetintenv("LOGGING", &logging);
+ if (!diag && logging) {
+ const char *setargs[] = { "set", "LOGGING", "0" };
+ kdb_set(2, setargs);
+ }
+
+ /* syslog_data[0,1] physical start, end+1. syslog_data[2,3]
+ * logical start, end+1. */
+ kdb_syslog_data(syslog_data);
+ if (syslog_data[2] == syslog_data[3])
+ return 0;
+ logsize = syslog_data[1] - syslog_data[0];
+ start = syslog_data[2];
+ end = syslog_data[3];
+#define KDB_WRAP(p) (((p - syslog_data[0]) % logsize) + syslog_data[0])
+ for (n = 0, p = start; p < end; ++p) {
+ c = *KDB_WRAP(p);
+ if (c == '\n')
+ ++n;
+ }
+ if (c != '\n')
+ ++n;
+ if (lines < 0) {
+ if (adjust >= n)
+ kdb_printf("buffer only contains %d lines, nothing "
+ "printed\n", n);
+ else if (adjust - lines >= n)
+ kdb_printf("buffer only contains %d lines, last %d "
+ "lines printed\n", n, n - adjust);
+ if (adjust) {
+ for (; start < end && adjust; ++start) {
+ if (*KDB_WRAP(start) == '\n')
+ --adjust;
+ }
+ if (start < end)
+ ++start;
+ }
+ for (p = start; p < end && lines; ++p) {
+ if (*KDB_WRAP(p) == '\n')
+ ++lines;
+ }
+ end = p;
+ } else if (lines > 0) {
+ int skip = n - (adjust + lines);
+ if (adjust >= n) {
+ kdb_printf("buffer only contains %d lines, "
+ "nothing printed\n", n);
+ skip = n;
+ } else if (skip < 0) {
+ lines += skip;
+ skip = 0;
+ kdb_printf("buffer only contains %d lines, first "
+ "%d lines printed\n", n, lines);
+ }
+ for (; start < end && skip; ++start) {
+ if (*KDB_WRAP(start) == '\n')
+ --skip;
+ }
+ for (p = start; p < end && lines; ++p) {
+ if (*KDB_WRAP(p) == '\n')
+ --lines;
+ }
+ end = p;
+ }
+ /* Do a line at a time (max 200 chars) to reduce protocol overhead */
+ c = '\n';
+ while (start != end) {
+ char buf[201];
+ p = buf;
+ if (KDB_FLAG(CMD_INTERRUPT))
+ return 0;
+ while (start < end && (c = *KDB_WRAP(start)) &&
+ (p - buf) < sizeof(buf)-1) {
+ ++start;
+ *p++ = c;
+ if (c == '\n')
+ break;
+ }
+ *p = '\0';
+ kdb_printf("%s", buf);
+ }
+ if (c != '\n')
+ kdb_printf("\n");
+
+ return 0;
+}
+#endif /* CONFIG_PRINTK */
+/*
+ * kdb_cpu - This function implements the 'cpu' command.
+ * cpu [<cpunum>]
+ * Returns:
+ * KDB_CMD_CPU for success, a kdb diagnostic if error
+ */
+static void kdb_cpu_status(void)
+{
+ int i, start_cpu, first_print = 1;
+ char state, prev_state = '?';
+
+ kdb_printf("Currently on cpu %d\n", raw_smp_processor_id());
+ kdb_printf("Available cpus: ");
+ for (start_cpu = -1, i = 0; i < NR_CPUS; i++) {
+ if (!cpu_online(i)) {
+ state = 'F'; /* cpu is offline */
+ } else {
+ state = ' '; /* cpu is responding to kdb */
+ if (kdb_task_state_char(KDB_TSK(i)) == 'I')
+ state = 'I'; /* idle task */
+ }
+ if (state != prev_state) {
+ if (prev_state != '?') {
+ if (!first_print)
+ kdb_printf(", ");
+ first_print = 0;
+ kdb_printf("%d", start_cpu);
+ if (start_cpu < i-1)
+ kdb_printf("-%d", i-1);
+ if (prev_state != ' ')
+ kdb_printf("(%c)", prev_state);
+ }
+ prev_state = state;
+ start_cpu = i;
+ }
+ }
+ /* print the trailing cpus, ignoring them if they are all offline */
+ if (prev_state != 'F') {
+ if (!first_print)
+ kdb_printf(", ");
+ kdb_printf("%d", start_cpu);
+ if (start_cpu < i-1)
+ kdb_printf("-%d", i-1);
+ if (prev_state != ' ')
+ kdb_printf("(%c)", prev_state);
+ }
+ kdb_printf("\n");
+}
+
+static int kdb_cpu(int argc, const char **argv)
+{
+ unsigned long cpunum;
+ int diag;
+
+ if (argc == 0) {
+ kdb_cpu_status();
+ return 0;
+ }
+
+ if (argc != 1)
+ return KDB_ARGCOUNT;
+
+ diag = kdbgetularg(argv[1], &cpunum);
+ if (diag)
+ return diag;
+
+ /*
+ * Validate cpunum
+ */
+ if ((cpunum > NR_CPUS) || !cpu_online(cpunum))
+ return KDB_BADCPUNUM;
+
+ dbg_switch_cpu = cpunum;
+
+ /*
+ * Switch to other cpu
+ */
+ return KDB_CMD_CPU;
+}
+
+/* The user may not realize that ps/bta with no parameters does not print idle
+ * or sleeping system daemon processes, so tell them how many were suppressed.
+ */
+void kdb_ps_suppressed(void)
+{
+ int idle = 0, daemon = 0;
+ unsigned long mask_I = kdb_task_state_string("I"),
+ mask_M = kdb_task_state_string("M");
+ unsigned long cpu;
+ const struct task_struct *p, *g;
+ for_each_online_cpu(cpu) {
+ p = kdb_curr_task(cpu);
+ if (kdb_task_state(p, mask_I))
+ ++idle;
+ }
+ kdb_do_each_thread(g, p) {
+ if (kdb_task_state(p, mask_M))
+ ++daemon;
+ } kdb_while_each_thread(g, p);
+ if (idle || daemon) {
+ if (idle)
+ kdb_printf("%d idle process%s (state I)%s\n",
+ idle, idle == 1 ? "" : "es",
+ daemon ? " and " : "");
+ if (daemon)
+ kdb_printf("%d sleeping system daemon (state M) "
+ "process%s", daemon,
+ daemon == 1 ? "" : "es");
+ kdb_printf(" suppressed,\nuse 'ps A' to see all.\n");
+ }
+}
+
+/*
+ * kdb_ps - This function implements the 'ps' command which shows a
+ * list of the active processes.
+ * ps [DRSTCZEUIMA] All processes, optionally filtered by state
+ */
+void kdb_ps1(const struct task_struct *p)
+{
+ int cpu;
+ unsigned long tmp;
+
+ if (!p || probe_kernel_read(&tmp, (char *)p, sizeof(unsigned long)))
+ return;
+
+ cpu = kdb_process_cpu(p);
+ kdb_printf("0x%p %8d %8d %d %4d %c 0x%p %c%s\n",
+ (void *)p, p->pid, p->parent->pid,
+ kdb_task_has_cpu(p), kdb_process_cpu(p),
+ kdb_task_state_char(p),
+ (void *)(&p->thread),
+ p == kdb_curr_task(raw_smp_processor_id()) ? '*' : ' ',
+ p->comm);
+ if (kdb_task_has_cpu(p)) {
+ if (!KDB_TSK(cpu)) {
+ kdb_printf(" Error: no saved data for this cpu\n");
+ } else {
+ if (KDB_TSK(cpu) != p)
+ kdb_printf(" Error: does not match running "
+ "process table (0x%p)\n", KDB_TSK(cpu));
+ }
+ }
+}
+
+static int kdb_ps(int argc, const char **argv)
+{
+ struct task_struct *g, *p;
+ unsigned long mask, cpu;
+
+ if (argc == 0)
+ kdb_ps_suppressed();
+ kdb_printf("%-*s Pid Parent [*] cpu State %-*s Command\n",
+ (int)(2*sizeof(void *))+2, "Task Addr",
+ (int)(2*sizeof(void *))+2, "Thread");
+ mask = kdb_task_state_string(argc ? argv[1] : NULL);
+ /* Run the active tasks first */
+ for_each_online_cpu(cpu) {
+ if (KDB_FLAG(CMD_INTERRUPT))
+ return 0;
+ p = kdb_curr_task(cpu);
+ if (kdb_task_state(p, mask))
+ kdb_ps1(p);
+ }
+ kdb_printf("\n");
+ /* Now the real tasks */
+ kdb_do_each_thread(g, p) {
+ if (KDB_FLAG(CMD_INTERRUPT))
+ return 0;
+ if (kdb_task_state(p, mask))
+ kdb_ps1(p);
+ } kdb_while_each_thread(g, p);
+
+ return 0;
+}
+
+/*
+ * kdb_pid - This function implements the 'pid' command which switches
+ * the currently active process.
+ * pid [<pid> | R]
+ */
+static int kdb_pid(int argc, const char **argv)
+{
+ struct task_struct *p;
+ unsigned long val;
+ int diag;
+
+ if (argc > 1)
+ return KDB_ARGCOUNT;
+
+ if (argc) {
+ if (strcmp(argv[1], "R") == 0) {
+ p = KDB_TSK(kdb_initial_cpu);
+ } else {
+ diag = kdbgetularg(argv[1], &val);
+ if (diag)
+ return KDB_BADINT;
+
+ p = find_task_by_pid_ns((pid_t)val, &init_pid_ns);
+ if (!p) {
+ kdb_printf("No task with pid=%d\n", (pid_t)val);
+ return 0;
+ }
+ }
+ kdb_set_current_task(p);
+ }
+ kdb_printf("KDB current process is %s(pid=%d)\n",
+ kdb_current_task->comm,
+ kdb_current_task->pid);
+
+ return 0;
+}
+
+/*
+ * kdb_ll - This function implements the 'll' command which follows a
+ * linked list and executes an arbitrary command for each
+ * element.
+ */
+static int kdb_ll(int argc, const char **argv)
+{
+ int diag;
+ unsigned long addr;
+ long offset = 0;
+ unsigned long va;
+ unsigned long linkoffset;
+ int nextarg;
+ const char *command;
+
+ if (argc != 3)
+ return KDB_ARGCOUNT;
+
+ nextarg = 1;
+ diag = kdbgetaddrarg(argc, argv, &nextarg, &addr, &offset, NULL);
+ if (diag)
+ return diag;
+
+ diag = kdbgetularg(argv[2], &linkoffset);
+ if (diag)
+ return diag;
+
+ /*
+ * Using the starting address as
+ * the first element in the list, and assuming that
+ * the list ends with a null pointer.
+ */
+
+ va = addr;
+ command = kdb_strdup(argv[3], GFP_KDB);
+ if (!command) {
+ kdb_printf("%s: cannot duplicate command\n", __func__);
+ return 0;
+ }
+ /* Recursive use of kdb_parse, do not use argv after this point */
+ argv = NULL;
+
+ while (va) {
+ char buf[80];
+
+ sprintf(buf, "%s " kdb_machreg_fmt "\n", command, va);
+ diag = kdb_parse(buf);
+ if (diag)
+ return diag;
+
+ addr = va + linkoffset;
+ if (kdb_getword(&va, addr, sizeof(va)))
+ return 0;
+ }
+ kfree(command);
+
+ return 0;
+}
+
+static int kdb_kgdb(int argc, const char **argv)
+{
+ return KDB_CMD_KGDB;
+}
+
+/*
+ * kdb_help - This function implements the 'help' and '?' commands.
+ */
+static int kdb_help(int argc, const char **argv)
+{
+ kdbtab_t *kt;
+ int i;
+
+ kdb_printf("%-15.15s %-20.20s %s\n", "Command", "Usage", "Description");
+ kdb_printf("-----------------------------"
+ "-----------------------------\n");
+ for_each_kdbcmd(kt, i) {
+ if (kt->cmd_name)
+ kdb_printf("%-15.15s %-20.20s %s\n", kt->cmd_name,
+ kt->cmd_usage, kt->cmd_help);
+ if (KDB_FLAG(CMD_INTERRUPT))
+ return 0;
+ }
+ return 0;
+}
+
+/*
+ * kdb_kill - This function implements the 'kill' commands.
+ */
+static int kdb_kill(int argc, const char **argv)
+{
+ long sig, pid;
+ char *endp;
+ struct task_struct *p;
+ struct siginfo info;
+
+ if (argc != 2)
+ return KDB_ARGCOUNT;
+
+ sig = simple_strtol(argv[1], &endp, 0);
+ if (*endp)
+ return KDB_BADINT;
+ if (sig >= 0) {
+ kdb_printf("Invalid signal parameter.<-signal>\n");
+ return 0;
+ }
+ sig = -sig;
+
+ pid = simple_strtol(argv[2], &endp, 0);
+ if (*endp)
+ return KDB_BADINT;
+ if (pid <= 0) {
+ kdb_printf("Process ID must be large than 0.\n");
+ return 0;
+ }
+
+ /* Find the process. */
+ p = find_task_by_pid_ns(pid, &init_pid_ns);
+ if (!p) {
+ kdb_printf("The specified process isn't found.\n");
+ return 0;
+ }
+ p = p->group_leader;
+ info.si_signo = sig;
+ info.si_errno = 0;
+ info.si_code = SI_USER;
+ info.si_pid = pid; /* same capabilities as process being signalled */
+ info.si_uid = 0; /* kdb has root authority */
+ kdb_send_sig_info(p, &info);
+ return 0;
+}
+
+struct kdb_tm {
+ int tm_sec; /* seconds */
+ int tm_min; /* minutes */
+ int tm_hour; /* hours */
+ int tm_mday; /* day of the month */
+ int tm_mon; /* month */
+ int tm_year; /* year */
+};
+
+static void kdb_gmtime(struct timespec *tv, struct kdb_tm *tm)
+{
+ /* This will work from 1970-2099, 2100 is not a leap year */
+ static int mon_day[] = { 31, 29, 31, 30, 31, 30, 31,
+ 31, 30, 31, 30, 31 };
+ memset(tm, 0, sizeof(*tm));
+ tm->tm_sec = tv->tv_sec % (24 * 60 * 60);
+ tm->tm_mday = tv->tv_sec / (24 * 60 * 60) +
+ (2 * 365 + 1); /* shift base from 1970 to 1968 */
+ tm->tm_min = tm->tm_sec / 60 % 60;
+ tm->tm_hour = tm->tm_sec / 60 / 60;
+ tm->tm_sec = tm->tm_sec % 60;
+ tm->tm_year = 68 + 4*(tm->tm_mday / (4*365+1));
+ tm->tm_mday %= (4*365+1);
+ mon_day[1] = 29;
+ while (tm->tm_mday >= mon_day[tm->tm_mon]) {
+ tm->tm_mday -= mon_day[tm->tm_mon];
+ if (++tm->tm_mon == 12) {
+ tm->tm_mon = 0;
+ ++tm->tm_year;
+ mon_day[1] = 28;
+ }
+ }
+ ++tm->tm_mday;
+}
+
+/*
+ * Most of this code has been lifted from kernel/timer.c::sys_sysinfo().
+ * I cannot call that code directly from kdb, it has an unconditional
+ * cli()/sti() and calls routines that take locks which can stop the debugger.
+ */
+static void kdb_sysinfo(struct sysinfo *val)
+{
+ struct timespec uptime;
+ do_posix_clock_monotonic_gettime(&uptime);
+ memset(val, 0, sizeof(*val));
+ val->uptime = uptime.tv_sec;
+ val->loads[0] = avenrun[0];
+ val->loads[1] = avenrun[1];
+ val->loads[2] = avenrun[2];
+ val->procs = nr_threads-1;
+ si_meminfo(val);
+ __si_swapinfo(val);
+
+ return;
+}
+
+/*
+ * kdb_summary - This function implements the 'summary' command.
+ */
+static int kdb_summary(int argc, const char **argv)
+{
+ struct kdb_tm tm;
+ struct sysinfo val;
+
+ if (argc)
+ return KDB_ARGCOUNT;
+
+ kdb_printf("sysname %s\n", init_uts_ns.name.sysname);
+ kdb_printf("release %s\n", init_uts_ns.name.release);
+ kdb_printf("version %s\n", init_uts_ns.name.version);
+ kdb_printf("machine %s\n", init_uts_ns.name.machine);
+ kdb_printf("nodename %s\n", init_uts_ns.name.nodename);
+ kdb_printf("domainname %s\n", init_uts_ns.name.domainname);
+ kdb_printf("ccversion %s\n", __stringify(CCVERSION));
+
+ kdb_gmtime(&xtime, &tm);
+ kdb_printf("date %04d-%02d-%02d %02d:%02d:%02d "
+ "tz_minuteswest %d\n",
+ 1900+tm.tm_year, tm.tm_mon+1, tm.tm_mday,
+ tm.tm_hour, tm.tm_min, tm.tm_sec,
+ sys_tz.tz_minuteswest);
+
+ kdb_sysinfo(&val);
+ kdb_printf("uptime ");
+ if (val.uptime > (24*60*60)) {
+ int days = val.uptime / (24*60*60);
+ val.uptime %= (24*60*60);
+ kdb_printf("%d day%s ", days, days == 1 ? "" : "s");
+ }
+ kdb_printf("%02ld:%02ld\n", val.uptime/(60*60), (val.uptime/60)%60);
+
+ /* lifted from fs/proc/proc_misc.c::loadavg_read_proc() */
+
+#define LOAD_INT(x) ((x) >> FSHIFT)
+#define LOAD_FRAC(x) LOAD_INT(((x) & (FIXED_1-1)) * 100)
+ kdb_printf("load avg %ld.%02ld %ld.%02ld %ld.%02ld\n",
+ LOAD_INT(val.loads[0]), LOAD_FRAC(val.loads[0]),
+ LOAD_INT(val.loads[1]), LOAD_FRAC(val.loads[1]),
+ LOAD_INT(val.loads[2]), LOAD_FRAC(val.loads[2]));
+ kdb_printf("\n");
+#undef LOAD_INT
+#undef LOAD_FRAC
+
+ kdb_seq_file_reset();
+ _meminfo_proc_show(&kdb_seq_file, NULL, 0); /* in fs/proc/meminfo.c */
+ kdb_seq_file_print();
+
+ return 0;
+}
+
+/*
+ * kdb_per_cpu - This function implements the 'per_cpu' command.
+ */
+static int kdb_per_cpu(int argc, const char **argv)
+{
+ char buf[256], fmtstr[64];
+ kdb_symtab_t symtab;
+ cpumask_t suppress = CPU_MASK_NONE;
+ int cpu, diag;
+ unsigned long addr, val, bytesperword = 0, whichcpu = ~0UL;
+
+ if (argc < 1 || argc > 3)
+ return KDB_ARGCOUNT;
+
+ snprintf(buf, sizeof(buf), "per_cpu__%s", argv[1]);
+ if (!kdbgetsymval(buf, &symtab)) {
+ kdb_printf("%s is not a per_cpu variable\n", argv[1]);
+ return KDB_BADADDR;
+ }
+ if (argc >= 2) {
+ diag = kdbgetularg(argv[2], &bytesperword);
+ if (diag)
+ return diag;
+ }
+ if (!bytesperword)
+ bytesperword = KDB_WORD_SIZE;
+ else if (bytesperword > KDB_WORD_SIZE)
+ return KDB_BADWIDTH;
+ sprintf(fmtstr, "%%0%dlx ", (int)(2*bytesperword));
+ if (argc >= 3) {
+ diag = kdbgetularg(argv[3], &whichcpu);
+ if (diag)
+ return diag;
+ if (!cpu_online(whichcpu)) {
+ kdb_printf("cpu %ld is not online\n", whichcpu);
+ return KDB_BADCPUNUM;
+ }
+ }
+
+ /* Most architectures use __per_cpu_offset[cpu], some use
+ * __per_cpu_offset(cpu), smp has no __per_cpu_offset.
+ */
+#ifdef __per_cpu_offset
+#define KDB_PCU(cpu) __per_cpu_offset(cpu)
+#else
+#ifdef CONFIG_SMP
+#define KDB_PCU(cpu) __per_cpu_offset[cpu]
+#else
+#define KDB_PCU(cpu) 0
+#endif
+#endif
+
+ for_each_online_cpu(cpu) {
+ if (whichcpu != ~0UL && whichcpu != cpu)
+ continue;
+ addr = symtab.sym_start + KDB_PCU(cpu);
+ diag = kdb_getword(&val, addr, bytesperword);
+ if (diag) {
+ kdb_printf("%5d " kdb_bfd_vma_fmt0 " - unable to "
+ "read, diag=%d\n", cpu, addr, diag);
+ continue;
+ }
+#ifdef CONFIG_SMP
+ if (!val) {
+ cpu_set(cpu, suppress);
+ continue;
+ }
+#endif /* CONFIG_SMP */
+ kdb_printf("%5d ", cpu);
+ kdb_md_line(fmtstr, addr,
+ bytesperword == KDB_WORD_SIZE,
+ 1, bytesperword, 1, 1, 0);
+ }
+ if (cpus_weight(suppress) == 0)
+ return 0;
+ kdb_printf("Zero suppressed cpu(s):");
+ for (cpu = first_cpu(suppress); cpu < num_possible_cpus();
+ cpu = next_cpu(cpu, suppress)) {
+ kdb_printf(" %d", cpu);
+ if (cpu == num_possible_cpus() - 1 ||
+ next_cpu(cpu, suppress) != cpu + 1)
+ continue;
+ while (cpu < num_possible_cpus() &&
+ next_cpu(cpu, suppress) == cpu + 1)
+ ++cpu;
+ kdb_printf("-%d", cpu);
+ }
+ kdb_printf("\n");
+
+#undef KDB_PCU
+
+ return 0;
+}
+
+/*
+ * display help for the use of cmd | grep pattern
+ */
+static int kdb_grep_help(int argc, const char **argv)
+{
+ kdb_printf("Usage of cmd args | grep pattern:\n");
+ kdb_printf(" Any command's output may be filtered through an ");
+ kdb_printf("emulated 'pipe'.\n");
+ kdb_printf(" 'grep' is just a key word.\n");
+ kdb_printf(" The pattern may include a very limited set of "
+ "metacharacters:\n");
+ kdb_printf(" pattern or ^pattern or pattern$ or ^pattern$\n");
+ kdb_printf(" And if there are spaces in the pattern, you may "
+ "quote it:\n");
+ kdb_printf(" \"pat tern\" or \"^pat tern\" or \"pat tern$\""
+ " or \"^pat tern$\"\n");
+ return 0;
+}
+
+/*
+ * kdb_register_repeat - This function is used to register a kernel
+ * debugger command.
+ * Inputs:
+ * cmd Command name
+ * func Function to execute the command
+ * usage A simple usage string showing arguments
+ * help A simple help string describing command
+ * repeat Does the command auto repeat on enter?
+ * Returns:
+ * zero for success, one if a duplicate command.
+ */
+#define kdb_command_extend 50 /* arbitrary */
+int kdb_register_repeat(char *cmd,
+ kdb_func_t func,
+ char *usage,
+ char *help,
+ short minlen,
+ kdb_repeat_t repeat)
+{
+ int i;
+ kdbtab_t *kp;
+
+ /*
+ * Brute force method to determine duplicates
+ */
+ for_each_kdbcmd(kp, i) {
+ if (kp->cmd_name && (strcmp(kp->cmd_name, cmd) == 0)) {
+ kdb_printf("Duplicate kdb command registered: "
+ "%s, func %p help %s\n", cmd, func, help);
+ return 1;
+ }
+ }
+
+ /*
+ * Insert command into first available location in table
+ */
+ for_each_kdbcmd(kp, i) {
+ if (kp->cmd_name == NULL)
+ break;
+ }
+
+ if (i >= kdb_max_commands) {
+ kdbtab_t *new = kmalloc((kdb_max_commands - KDB_BASE_CMD_MAX +
+ kdb_command_extend) * sizeof(*new), GFP_KDB);
+ if (!new) {
+ kdb_printf("Could not allocate new kdb_command "
+ "table\n");
+ return 1;
+ }
+ if (kdb_commands) {
+ memcpy(new, kdb_commands,
+ kdb_max_commands * sizeof(*new));
+ kfree(kdb_commands);
+ }
+ memset(new + kdb_max_commands, 0,
+ kdb_command_extend * sizeof(*new));
+ kdb_commands = new;
+ kp = kdb_commands + kdb_max_commands;
+ kdb_max_commands += kdb_command_extend;
+ }
+
+ kp->cmd_name = cmd;
+ kp->cmd_func = func;
+ kp->cmd_usage = usage;
+ kp->cmd_help = help;
+ kp->cmd_flags = 0;
+ kp->cmd_minlen = minlen;
+ kp->cmd_repeat = repeat;
+
+ return 0;
+}
+
+/*
+ * kdb_register - Compatibility register function for commands that do
+ * not need to specify a repeat state. Equivalent to
+ * kdb_register_repeat with KDB_REPEAT_NONE.
+ * Inputs:
+ * cmd Command name
+ * func Function to execute the command
+ * usage A simple usage string showing arguments
+ * help A simple help string describing command
+ * Returns:
+ * zero for success, one if a duplicate command.
+ */
+int kdb_register(char *cmd,
+ kdb_func_t func,
+ char *usage,
+ char *help,
+ short minlen)
+{
+ return kdb_register_repeat(cmd, func, usage, help, minlen,
+ KDB_REPEAT_NONE);
+}
+
+/*
+ * kdb_unregister - This function is used to unregister a kernel
+ * debugger command. It is generally called when a module which
+ * implements kdb commands is unloaded.
+ * Inputs:
+ * cmd Command name
+ * Returns:
+ * zero for success, one command not registered.
+ */
+int kdb_unregister(char *cmd)
+{
+ int i;
+ kdbtab_t *kp;
+
+ /*
+ * find the command.
+ */
+ for (i = 0, kp = kdb_commands; i < kdb_max_commands; i++, kp++) {
+ if (kp->cmd_name && (strcmp(kp->cmd_name, cmd) == 0)) {
+ kp->cmd_name = NULL;
+ return 0;
+ }
+ }
+
+ /* Couldn't find it. */
+ return 1;
+}
+
+/* Initialize the kdb command table. */
+static void __init kdb_inittab(void)
+{
+ int i;
+ kdbtab_t *kp;
+
+ for_each_kdbcmd(kp, i)
+ kp->cmd_name = NULL;
+
+ kdb_register_repeat("md", kdb_md, "<vaddr>",
+ "Display Memory Contents, also mdWcN, e.g. md8c1", 1,
+ KDB_REPEAT_NO_ARGS);
+ kdb_register_repeat("mdr", kdb_md, "<vaddr> <bytes>",
+ "Display Raw Memory", 0, KDB_REPEAT_NO_ARGS);
+ kdb_register_repeat("mdp", kdb_md, "<paddr> <bytes>",
+ "Display Physical Memory", 0, KDB_REPEAT_NO_ARGS);
+ kdb_register_repeat("mds", kdb_md, "<vaddr>",
+ "Display Memory Symbolically", 0, KDB_REPEAT_NO_ARGS);
+ kdb_register_repeat("mm", kdb_mm, "<vaddr> <contents>",
+ "Modify Memory Contents", 0, KDB_REPEAT_NO_ARGS);
+ kdb_register_repeat("go", kdb_go, "[<vaddr>]",
+ "Continue Execution", 1, KDB_REPEAT_NONE);
+ kdb_register_repeat("rd", kdb_rd, "",
+ "Display Registers", 0, KDB_REPEAT_NONE);
+ kdb_register_repeat("rm", kdb_rm, "<reg> <contents>",
+ "Modify Registers", 0, KDB_REPEAT_NONE);
+ kdb_register_repeat("ef", kdb_ef, "<vaddr>",
+ "Display exception frame", 0, KDB_REPEAT_NONE);
+ kdb_register_repeat("bt", kdb_bt, "[<vaddr>]",
+ "Stack traceback", 1, KDB_REPEAT_NONE);
+ kdb_register_repeat("btp", kdb_bt, "<pid>",
+ "Display stack for process <pid>", 0, KDB_REPEAT_NONE);
+ kdb_register_repeat("bta", kdb_bt, "[DRSTCZEUIMA]",
+ "Display stack all processes", 0, KDB_REPEAT_NONE);
+ kdb_register_repeat("btc", kdb_bt, "",
+ "Backtrace current process on each cpu", 0, KDB_REPEAT_NONE);
+ kdb_register_repeat("btt", kdb_bt, "<vaddr>",
+ "Backtrace process given its struct task address", 0,
+ KDB_REPEAT_NONE);
+ kdb_register_repeat("ll", kdb_ll, "<first-element> <linkoffset> <cmd>",
+ "Execute cmd for each element in linked list", 0, KDB_REPEAT_NONE);
+ kdb_register_repeat("env", kdb_env, "",
+ "Show environment variables", 0, KDB_REPEAT_NONE);
+ kdb_register_repeat("set", kdb_set, "",
+ "Set environment variables", 0, KDB_REPEAT_NONE);
+ kdb_register_repeat("help", kdb_help, "",
+ "Display Help Message", 1, KDB_REPEAT_NONE);
+ kdb_register_repeat("?", kdb_help, "",
+ "Display Help Message", 0, KDB_REPEAT_NONE);
+ kdb_register_repeat("cpu", kdb_cpu, "<cpunum>",
+ "Switch to new cpu", 0, KDB_REPEAT_NONE);
+ kdb_register_repeat("kgdb", kdb_kgdb, "",
+ "Enter kgdb mode", 0, KDB_REPEAT_NONE);
+ kdb_register_repeat("ps", kdb_ps, "[<flags>|A]",
+ "Display active task list", 0, KDB_REPEAT_NONE);
+ kdb_register_repeat("pid", kdb_pid, "<pidnum>",
+ "Switch to another task", 0, KDB_REPEAT_NONE);
+ kdb_register_repeat("reboot", kdb_reboot, "",
+ "Reboot the machine immediately", 0, KDB_REPEAT_NONE);
+#if defined(CONFIG_MODULES)
+ kdb_register_repeat("lsmod", kdb_lsmod, "",
+ "List loaded kernel modules", 0, KDB_REPEAT_NONE);
+#endif
+#if defined(CONFIG_MAGIC_SYSRQ)
+ kdb_register_repeat("sr", kdb_sr, "<key>",
+ "Magic SysRq key", 0, KDB_REPEAT_NONE);
+#endif
+#if defined(CONFIG_PRINTK)
+ kdb_register_repeat("dmesg", kdb_dmesg, "[lines]",
+ "Display syslog buffer", 0, KDB_REPEAT_NONE);
+#endif
+ kdb_register_repeat("defcmd", kdb_defcmd, "name \"usage\" \"help\"",
+ "Define a set of commands, down to endefcmd", 0, KDB_REPEAT_NONE);
+ kdb_register_repeat("kill", kdb_kill, "<-signal> <pid>",
+ "Send a signal to a process", 0, KDB_REPEAT_NONE);
+ kdb_register_repeat("summary", kdb_summary, "",
+ "Summarize the system", 4, KDB_REPEAT_NONE);
+ kdb_register_repeat("per_cpu", kdb_per_cpu, "",
+ "Display per_cpu variables", 3, KDB_REPEAT_NONE);
+ kdb_register_repeat("grephelp", kdb_grep_help, "",
+ "Display help on | grep", 0, KDB_REPEAT_NONE);
+}
+
+/* Execute any commands defined in kdb_cmds. */
+static void __init kdb_cmd_init(void)
+{
+ int i, diag;
+ for (i = 0; kdb_cmds[i]; ++i) {
+ diag = kdb_parse(kdb_cmds[i]);
+ if (diag)
+ kdb_printf("kdb command %s failed, kdb diag %d\n",
+ kdb_cmds[i], diag);
+ }
+ if (defcmd_in_progress) {
+ kdb_printf("Incomplete 'defcmd' set, forcing endefcmd\n");
+ kdb_parse("endefcmd");
+ }
+}
+
+/* Intialize kdb_printf, breakpoint tables and kdb state */
+void __init kdb_init(int lvl)
+{
+ static int kdb_init_lvl = KDB_NOT_INITIALIZED;
+ int i;
+
+ if (kdb_init_lvl == KDB_INIT_FULL || lvl <= kdb_init_lvl)
+ return;
+ for (i = kdb_init_lvl; i < lvl; i++) {
+ switch (i) {
+ case KDB_NOT_INITIALIZED:
+ kdb_inittab(); /* Initialize Command Table */
+ kdb_initbptab(); /* Initialize Breakpoints */
+ break;
+ case KDB_INIT_EARLY:
+ kdb_cmd_init(); /* Build kdb_cmds tables */
+ break;
+ }
+ }
+ kdb_init_lvl = lvl;
+}
diff --git a/kernel/debug/kdb/kdb_private.h b/kernel/debug/kdb/kdb_private.h
new file mode 100644
index 0000000..5183b69
--- /dev/null
+++ b/kernel/debug/kdb/kdb_private.h
@@ -0,0 +1,314 @@
+#ifndef _KDBPRIVATE_H
+#define _KDBPRIVATE_H
+
+/*
+ * Kernel Debugger Architecture Independent Private Headers
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (c) 2000-2004 Silicon Graphics, Inc. All Rights Reserved.
+ * Copyright (c) 2009 Wind River Systems, Inc. All Rights Reserved.
+ */
+
+#include <linux/kgdb.h>
+#include "../debug_core.h"
+
+/* Kernel Debugger Error codes. Must not overlap with command codes. */
+#define KDB_NOTFOUND (-1)
+#define KDB_ARGCOUNT (-2)
+#define KDB_BADWIDTH (-3)
+#define KDB_BADRADIX (-4)
+#define KDB_NOTENV (-5)
+#define KDB_NOENVVALUE (-6)
+#define KDB_NOTIMP (-7)
+#define KDB_ENVFULL (-8)
+#define KDB_ENVBUFFULL (-9)
+#define KDB_TOOMANYBPT (-10)
+#define KDB_TOOMANYDBREGS (-11)
+#define KDB_DUPBPT (-12)
+#define KDB_BPTNOTFOUND (-13)
+#define KDB_BADMODE (-14)
+#define KDB_BADINT (-15)
+#define KDB_INVADDRFMT (-16)
+#define KDB_BADREG (-17)
+#define KDB_BADCPUNUM (-18)
+#define KDB_BADLENGTH (-19)
+#define KDB_NOBP (-20)
+#define KDB_BADADDR (-21)
+
+/* Kernel Debugger Command codes. Must not overlap with error codes. */
+#define KDB_CMD_GO (-1001)
+#define KDB_CMD_CPU (-1002)
+#define KDB_CMD_SS (-1003)
+#define KDB_CMD_SSB (-1004)
+#define KDB_CMD_KGDB (-1005)
+#define KDB_CMD_KGDB2 (-1006)
+
+/* Internal debug flags */
+#define KDB_DEBUG_FLAG_BP 0x0002 /* Breakpoint subsystem debug */
+#define KDB_DEBUG_FLAG_BB_SUMM 0x0004 /* Basic block analysis, summary only */
+#define KDB_DEBUG_FLAG_AR 0x0008 /* Activation record, generic */
+#define KDB_DEBUG_FLAG_ARA 0x0010 /* Activation record, arch specific */
+#define KDB_DEBUG_FLAG_BB 0x0020 /* All basic block analysis */
+#define KDB_DEBUG_FLAG_STATE 0x0040 /* State flags */
+#define KDB_DEBUG_FLAG_MASK 0xffff /* All debug flags */
+#define KDB_DEBUG_FLAG_SHIFT 16 /* Shift factor for dbflags */
+
+#define KDB_DEBUG(flag) (kdb_flags & \
+ (KDB_DEBUG_FLAG_##flag << KDB_DEBUG_FLAG_SHIFT))
+#define KDB_DEBUG_STATE(text, value) if (KDB_DEBUG(STATE)) \
+ kdb_print_state(text, value)
+
+#if BITS_PER_LONG == 32
+
+#define KDB_PLATFORM_ENV "BYTESPERWORD=4"
+
+#define kdb_machreg_fmt "0x%lx"
+#define kdb_machreg_fmt0 "0x%08lx"
+#define kdb_bfd_vma_fmt "0x%lx"
+#define kdb_bfd_vma_fmt0 "0x%08lx"
+#define kdb_elfw_addr_fmt "0x%x"
+#define kdb_elfw_addr_fmt0 "0x%08x"
+#define kdb_f_count_fmt "%d"
+
+#elif BITS_PER_LONG == 64
+
+#define KDB_PLATFORM_ENV "BYTESPERWORD=8"
+
+#define kdb_machreg_fmt "0x%lx"
+#define kdb_machreg_fmt0 "0x%016lx"
+#define kdb_bfd_vma_fmt "0x%lx"
+#define kdb_bfd_vma_fmt0 "0x%016lx"
+#define kdb_elfw_addr_fmt "0x%x"
+#define kdb_elfw_addr_fmt0 "0x%016x"
+#define kdb_f_count_fmt "%ld"
+
+#endif
+
+/*
+ * KDB_MAXBPT describes the total number of breakpoints
+ * supported by this architecure.
+ */
+#define KDB_MAXBPT 16
+
+/* Maximum number of arguments to a function */
+#define KDB_MAXARGS 16
+
+typedef enum {
+ KDB_REPEAT_NONE = 0, /* Do not repeat this command */
+ KDB_REPEAT_NO_ARGS, /* Repeat the command without arguments */
+ KDB_REPEAT_WITH_ARGS, /* Repeat the command including its arguments */
+} kdb_repeat_t;
+
+typedef int (*kdb_func_t)(int, const char **);
+
+/* Symbol table format returned by kallsyms. */
+typedef struct __ksymtab {
+ unsigned long value; /* Address of symbol */
+ const char *mod_name; /* Module containing symbol or
+ * "kernel" */
+ unsigned long mod_start;
+ unsigned long mod_end;
+ const char *sec_name; /* Section containing symbol */
+ unsigned long sec_start;
+ unsigned long sec_end;
+ const char *sym_name; /* Full symbol name, including
+ * any version */
+ unsigned long sym_start;
+ unsigned long sym_end;
+ } kdb_symtab_t;
+extern int kallsyms_symbol_next(char *prefix_name, int flag);
+extern int kallsyms_symbol_complete(char *prefix_name, int max_len);
+
+/* Exported Symbols for kernel loadable modules to use. */
+extern int kdb_register(char *, kdb_func_t, char *, char *, short);
+extern int kdb_register_repeat(char *, kdb_func_t, char *, char *,
+ short, kdb_repeat_t);
+extern int kdb_unregister(char *);
+
+extern int kdb_getarea_size(void *, unsigned long, size_t);
+extern int kdb_putarea_size(unsigned long, void *, size_t);
+
+/*
+ * Like get_user and put_user, kdb_getarea and kdb_putarea take variable
+ * names, not pointers. The underlying *_size functions take pointers.
+ */
+#define kdb_getarea(x, addr) kdb_getarea_size(&(x), addr, sizeof((x)))
+#define kdb_putarea(addr, x) kdb_putarea_size(addr, &(x), sizeof((x)))
+
+extern int kdb_getphysword(unsigned long *word,
+ unsigned long addr, size_t size);
+extern int kdb_getword(unsigned long *, unsigned long, size_t);
+extern int kdb_putword(unsigned long, unsigned long, size_t);
+
+extern int kdbgetularg(const char *, unsigned long *);
+extern int kdb_set(int, const char **);
+extern char *kdbgetenv(const char *);
+extern int kdbgetintenv(const char *, int *);
+extern int kdbgetaddrarg(int, const char **, int*, unsigned long *,
+ long *, char **);
+extern int kdbgetsymval(const char *, kdb_symtab_t *);
+extern int kdbnearsym(unsigned long, kdb_symtab_t *);
+extern void kdbnearsym_cleanup(void);
+extern char *kdb_strdup(const char *str, gfp_t type);
+extern void kdb_symbol_print(unsigned long, const kdb_symtab_t *, unsigned int);
+
+/* Routine for debugging the debugger state. */
+extern void kdb_print_state(const char *, int);
+
+extern int kdb_state;
+#define KDB_STATE_KDB 0x00000001 /* Cpu is inside kdb */
+#define KDB_STATE_LEAVING 0x00000002 /* Cpu is leaving kdb */
+#define KDB_STATE_CMD 0x00000004 /* Running a kdb command */
+#define KDB_STATE_KDB_CONTROL 0x00000008 /* This cpu is under
+ * kdb control */
+#define KDB_STATE_HOLD_CPU 0x00000010 /* Hold this cpu inside kdb */
+#define KDB_STATE_DOING_SS 0x00000020 /* Doing ss command */
+#define KDB_STATE_DOING_SSB 0x00000040 /* Doing ssb command,
+ * DOING_SS is also set */
+#define KDB_STATE_SSBPT 0x00000080 /* Install breakpoint
+ * after one ss, independent of
+ * DOING_SS */
+#define KDB_STATE_REENTRY 0x00000100 /* Valid re-entry into kdb */
+#define KDB_STATE_SUPPRESS 0x00000200 /* Suppress error messages */
+#define KDB_STATE_PAGER 0x00000400 /* pager is available */
+#define KDB_STATE_GO_SWITCH 0x00000800 /* go is switching
+ * back to initial cpu */
+#define KDB_STATE_PRINTF_LOCK 0x00001000 /* Holds kdb_printf lock */
+#define KDB_STATE_WAIT_IPI 0x00002000 /* Waiting for kdb_ipi() NMI */
+#define KDB_STATE_RECURSE 0x00004000 /* Recursive entry to kdb */
+#define KDB_STATE_IP_ADJUSTED 0x00008000 /* Restart IP has been
+ * adjusted */
+#define KDB_STATE_GO1 0x00010000 /* go only releases one cpu */
+#define KDB_STATE_KEYBOARD 0x00020000 /* kdb entered via
+ * keyboard on this cpu */
+#define KDB_STATE_KEXEC 0x00040000 /* kexec issued */
+#define KDB_STATE_DOING_KGDB 0x00080000 /* kgdb enter now issued */
+#define KDB_STATE_DOING_KGDB2 0x00100000 /* kgdb enter now issued */
+#define KDB_STATE_KGDB_TRANS 0x00200000 /* Transition to kgdb */
+#define KDB_STATE_ARCH 0xff000000 /* Reserved for arch
+ * specific use */
+
+#define KDB_STATE(flag) (kdb_state & KDB_STATE_##flag)
+#define KDB_STATE_SET(flag) ((void)(kdb_state |= KDB_STATE_##flag))
+#define KDB_STATE_CLEAR(flag) ((void)(kdb_state &= ~KDB_STATE_##flag))
+
+extern int kdb_nextline; /* Current number of lines displayed */
+
+typedef struct _kdb_bp {
+ unsigned long bp_addr; /* Address breakpoint is present at */
+ unsigned int bp_free:1; /* This entry is available */
+ unsigned int bp_enabled:1; /* Breakpoint is active in register */
+ unsigned int bp_type:4; /* Uses hardware register */
+ unsigned int bp_installed:1; /* Breakpoint is installed */
+ unsigned int bp_delay:1; /* Do delayed bp handling */
+ unsigned int bp_delayed:1; /* Delayed breakpoint */
+ unsigned int bph_length; /* HW break length */
+} kdb_bp_t;
+
+#ifdef CONFIG_KGDB_KDB
+extern kdb_bp_t kdb_breakpoints[/* KDB_MAXBPT */];
+
+/* The KDB shell command table */
+typedef struct _kdbtab {
+ char *cmd_name; /* Command name */
+ kdb_func_t cmd_func; /* Function to execute command */
+ char *cmd_usage; /* Usage String for this command */
+ char *cmd_help; /* Help message for this command */
+ short cmd_flags; /* Parsing flags */
+ short cmd_minlen; /* Minimum legal # command
+ * chars required */
+ kdb_repeat_t cmd_repeat; /* Does command auto repeat on enter? */
+} kdbtab_t;
+
+extern int kdb_bt(int, const char **); /* KDB display back trace */
+
+/* KDB breakpoint management functions */
+extern void kdb_initbptab(void);
+extern void kdb_bp_install(struct pt_regs *);
+extern void kdb_bp_remove(void);
+
+typedef enum {
+ KDB_DB_BPT, /* Breakpoint */
+ KDB_DB_SS, /* Single-step trap */
+ KDB_DB_SSB, /* Single step to branch */
+ KDB_DB_SSBPT, /* Single step over breakpoint */
+ KDB_DB_NOBPT /* Spurious breakpoint */
+} kdb_dbtrap_t;
+
+extern int kdb_main_loop(kdb_reason_t, kdb_reason_t,
+ int, kdb_dbtrap_t, struct pt_regs *);
+
+/* Miscellaneous functions and data areas */
+extern int kdb_grepping_flag;
+extern char kdb_grep_string[];
+extern int kdb_grep_leading;
+extern int kdb_grep_trailing;
+extern char *kdb_cmds[];
+extern void kdb_syslog_data(char *syslog_data[]);
+extern unsigned long kdb_task_state_string(const char *);
+extern char kdb_task_state_char (const struct task_struct *);
+extern unsigned long kdb_task_state(const struct task_struct *p,
+ unsigned long mask);
+extern void kdb_ps_suppressed(void);
+extern void kdb_ps1(const struct task_struct *p);
+extern int kdb_parse(const char *cmdstr);
+extern void kdb_print_nameval(const char *name, unsigned long val);
+extern void kdb_send_sig_info(struct task_struct *p, struct siginfo *info);
+extern void kdb_meminfo_proc_show(void);
+extern const char *kdb_walk_kallsyms(loff_t *pos);
+extern char *kdb_getstr(char *, size_t, char *);
+
+/* Defines for kdb_symbol_print */
+#define KDB_SP_SPACEB 0x0001 /* Space before string */
+#define KDB_SP_SPACEA 0x0002 /* Space after string */
+#define KDB_SP_PAREN 0x0004 /* Parenthesis around string */
+#define KDB_SP_VALUE 0x0008 /* Print the value of the address */
+#define KDB_SP_SYMSIZE 0x0010 /* Print the size of the symbol */
+#define KDB_SP_NEWLINE 0x0020 /* Newline after string */
+#define KDB_SP_DEFAULT (KDB_SP_VALUE|KDB_SP_PAREN)
+
+#define KDB_TSK(cpu) kgdb_info[cpu].task
+#define KDB_TSKREGS(cpu) kgdb_info[cpu].debuggerinfo
+
+extern struct task_struct *kdb_curr_task(int);
+
+#define kdb_task_has_cpu(p) (task_curr(p))
+
+/* Simplify coexistence with NPTL */
+#define kdb_do_each_thread(g, p) do_each_thread(g, p)
+#define kdb_while_each_thread(g, p) while_each_thread(g, p)
+
+#define GFP_KDB (in_interrupt() ? GFP_ATOMIC : GFP_KERNEL)
+
+extern void *debug_kmalloc(size_t size, gfp_t flags);
+extern void debug_kfree(void *);
+extern void debug_kusage(void);
+
+extern void kdb_set_current_task(struct task_struct *);
+extern struct task_struct *kdb_current_task;
+extern struct pt_regs *kdb_current_regs;
+#ifdef CONFIG_MODULES
+extern struct list_head *kdb_modules;
+#endif /* CONFIG_MODULES */
+
+extern char kdb_prompt_str[];
+extern struct seq_file kdb_seq_file;
+extern void kdb_seq_file_reset(void);
+extern void kdb_seq_file_print(void);
+/* Access to un-exported kernel internals */
+#ifdef CONFIG_PROC_FS
+extern int _meminfo_proc_show(struct seq_file *m, void *v, int lock);
+#else
+static inline int _meminfo_proc_show(struct seq_file *m, void *v, int lock)
+{
+ return 0;
+}
+#endif
+
+#define KDB_WORD_SIZE ((int)sizeof(unsigned long))
+
+#endif /* CONFIG_KGDB_KDB */
+#endif /* !_KDBPRIVATE_H */
diff --git a/kernel/debug/kdb/kdb_support.c b/kernel/debug/kdb/kdb_support.c
new file mode 100644
index 0000000..6a37b09
--- /dev/null
+++ b/kernel/debug/kdb/kdb_support.c
@@ -0,0 +1,991 @@
+/*
+ * Kernel Debugger Architecture Independent Support Functions
+ *
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License. See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (c) 1999-2004 Silicon Graphics, Inc. All Rights Reserved.
+ * Copyright (c) 2009 Wind River Systems, Inc. All Rights Reserved.
+ * 03/02/13 added new 2.5 kallsyms <[email protected]>
+ */
+
+#include <stdarg.h>
+#include <linux/types.h>
+#include <linux/sched.h>
+#include <linux/mm.h>
+#include <linux/kallsyms.h>
+#include <linux/stddef.h>
+#include <linux/vmalloc.h>
+#include <linux/ptrace.h>
+#include <linux/module.h>
+#include <linux/highmem.h>
+#include <linux/hardirq.h>
+#include <linux/delay.h>
+#include <linux/uaccess.h>
+#include <linux/seq_file.h>
+#include <linux/kdb.h>
+#include "kdb_private.h"
+
+#define KDB_SEQ_BUF_SIZE 2048
+static char kdb_seq_buf[KDB_SEQ_BUF_SIZE];
+
+static void *kdb_seq_next(struct seq_file *seq, void *v, loff_t *pos)
+{
+ (*pos)++;
+ if (*pos >= KDB_SEQ_BUF_SIZE)
+ return NULL;
+ return pos;
+}
+
+static void *kdb_seq_start(struct seq_file *seq, loff_t *pos)
+{
+ return (*pos < KDB_SEQ_BUF_SIZE) ? pos : NULL;
+}
+
+static void kdb_seq_stop(struct seq_file *seq, void *v)
+{
+ seq->count = 0;
+}
+
+static int kdb_seq_show(struct seq_file *seq, void *v)
+{
+ return 0;
+}
+
+static const struct seq_operations kdb_seq_ops = {
+ .start = kdb_seq_start,
+ .next = kdb_seq_next,
+ .stop = kdb_seq_stop,
+ .show = kdb_seq_show,
+};
+
+struct seq_file kdb_seq_file = {
+ .buf = kdb_seq_buf,
+ .size = KDB_SEQ_BUF_SIZE,
+ .op = &kdb_seq_ops
+};
+
+void kdb_seq_file_reset(void)
+{
+ kdb_seq_file.count = 0;
+}
+
+void kdb_seq_file_print(void)
+{
+ char *ptr = kdb_seq_file.buf;
+ char *tmp;
+
+ if (kdb_seq_file.count < kdb_seq_file.size)
+ kdb_seq_file.buf[kdb_seq_file.count] = '\0';
+ else
+ kdb_seq_file.buf[kdb_seq_file.size - 1] = '\0';
+ while (ptr < &kdb_seq_file.buf[kdb_seq_file.count]) {
+ if (KDB_FLAG(CMD_INTERRUPT))
+ return;
+ tmp = strchr(ptr, '\n');
+ if (tmp)
+ *tmp = '\0';
+ kdb_printf("%s\n", ptr);
+ ptr += strlen(ptr) + 1;
+ }
+}
+
+/*
+ * kdbgetsymval - Return the address of the given symbol.
+ *
+ * Parameters:
+ * symname Character string containing symbol name
+ * symtab Structure to receive results
+ * Returns:
+ * 0 Symbol not found, symtab zero filled
+ * 1 Symbol mapped to module/symbol/section, data in symtab
+ */
+int kdbgetsymval(const char *symname, kdb_symtab_t *symtab)
+{
+ if (KDB_DEBUG(AR))
+ kdb_printf("kdbgetsymval: symname=%s, symtab=%p\n", symname,
+ symtab);
+ memset(symtab, 0, sizeof(*symtab));
+ symtab->sym_start = kallsyms_lookup_name(symname);
+ if (symtab->sym_start) {
+ if (KDB_DEBUG(AR))
+ kdb_printf("kdbgetsymval: returns 1, "
+ "symtab->sym_start=0x%lx\n",
+ symtab->sym_start);
+ return 1;
+ }
+ if (KDB_DEBUG(AR))
+ kdb_printf("kdbgetsymval: returns 0\n");
+ return 0;
+}
+EXPORT_SYMBOL(kdbgetsymval);
+
+static char *kdb_name_table[100]; /* arbitrary size */
+
+/*
+ * kdbnearsym - Return the name of the symbol with the nearest address
+ * less than 'addr'.
+ *
+ * Parameters:
+ * addr Address to check for symbol near
+ * symtab Structure to receive results
+ * Returns:
+ * 0 No sections contain this address, symtab zero filled
+ * 1 Address mapped to module/symbol/section, data in symtab
+ * Remarks:
+ * 2.6 kallsyms has a "feature" where it unpacks the name into a
+ * string. If that string is reused before the caller expects it
+ * then the caller sees its string change without warning. To
+ * avoid cluttering up the main kdb code with lots of kdb_strdup,
+ * tests and kfree calls, kdbnearsym maintains an LRU list of the
+ * last few unique strings. The list is sized large enough to
+ * hold active strings, no kdb caller of kdbnearsym makes more
+ * than ~20 later calls before using a saved value.
+ */
+int kdbnearsym(unsigned long addr, kdb_symtab_t *symtab)
+{
+ int ret = 0;
+ unsigned long symbolsize;
+ unsigned long offset;
+#define knt1_size 128 /* must be >= kallsyms table size */
+ char *knt1 = NULL;
+
+ if (KDB_DEBUG(AR))
+ kdb_printf("kdbnearsym: addr=0x%lx, symtab=%p\n", addr, symtab);
+ memset(symtab, 0, sizeof(*symtab));
+
+ if (addr < 4096)
+ goto out;
+ knt1 = debug_kmalloc(knt1_size, GFP_ATOMIC);
+ if (!knt1) {
+ kdb_printf("kdbnearsym: addr=0x%lx cannot kmalloc knt1\n",
+ addr);
+ goto out;
+ }
+ symtab->sym_name = kallsyms_lookup(addr, &symbolsize , &offset,
+ (char **)(&symtab->mod_name), knt1);
+ if (offset > 8*1024*1024) {
+ symtab->sym_name = NULL;
+ addr = offset = symbolsize = 0;
+ }
+ symtab->sym_start = addr - offset;
+ symtab->sym_end = symtab->sym_start + symbolsize;
+ ret = symtab->sym_name != NULL && *(symtab->sym_name) != '\0';
+
+ if (ret) {
+ int i;
+ /* Another 2.6 kallsyms "feature". Sometimes the sym_name is
+ * set but the buffer passed into kallsyms_lookup is not used,
+ * so it contains garbage. The caller has to work out which
+ * buffer needs to be saved.
+ *
+ * What was Rusty smoking when he wrote that code?
+ */
+ if (symtab->sym_name != knt1) {
+ strncpy(knt1, symtab->sym_name, knt1_size);
+ knt1[knt1_size-1] = '\0';
+ }
+ for (i = 0; i < ARRAY_SIZE(kdb_name_table); ++i) {
+ if (kdb_name_table[i] &&
+ strcmp(kdb_name_table[i], knt1) == 0)
+ break;
+ }
+ if (i >= ARRAY_SIZE(kdb_name_table)) {
+ debug_kfree(kdb_name_table[0]);
+ memcpy(kdb_name_table, kdb_name_table+1,
+ sizeof(kdb_name_table[0]) *
+ (ARRAY_SIZE(kdb_name_table)-1));
+ } else {
+ debug_kfree(knt1);
+ knt1 = kdb_name_table[i];
+ memcpy(kdb_name_table+i, kdb_name_table+i+1,
+ sizeof(kdb_name_table[0]) *
+ (ARRAY_SIZE(kdb_name_table)-i-1));
+ }
+ i = ARRAY_SIZE(kdb_name_table) - 1;
+ kdb_name_table[i] = knt1;
+ symtab->sym_name = kdb_name_table[i];
+ knt1 = NULL;
+ }
+
+ if (symtab->mod_name == NULL)
+ symtab->mod_name = "kernel";
+ if (KDB_DEBUG(AR))
+ kdb_printf("kdbnearsym: returns %d symtab->sym_start=0x%lx, "
+ "symtab->mod_name=%p, symtab->sym_name=%p (%s)\n", ret,
+ symtab->sym_start, symtab->mod_name, symtab->sym_name,
+ symtab->sym_name);
+
+out:
+ debug_kfree(knt1);
+ return ret;
+}
+
+void kdbnearsym_cleanup(void)
+{
+ int i;
+ for (i = 0; i < ARRAY_SIZE(kdb_name_table); ++i) {
+ if (kdb_name_table[i]) {
+ debug_kfree(kdb_name_table[i]);
+ kdb_name_table[i] = NULL;
+ }
+ }
+}
+
+static char ks_namebuf[KSYM_NAME_LEN+1], ks_namebuf_prev[KSYM_NAME_LEN+1];
+
+/*
+ * kallsyms_symbol_complete
+ *
+ * Parameters:
+ * prefix_name prefix of a symbol name to lookup
+ * max_len maximum length that can be returned
+ * Returns:
+ * Number of symbols which match the given prefix.
+ * Notes:
+ * prefix_name is changed to contain the longest unique prefix that
+ * starts with this prefix (tab completion).
+ */
+int kallsyms_symbol_complete(char *prefix_name, int max_len)
+{
+ loff_t pos = 0;
+ int prefix_len = strlen(prefix_name), prev_len = 0;
+ int i, number = 0;
+ const char *name;
+
+ while ((name = kdb_walk_kallsyms(&pos))) {
+ if (strncmp(name, prefix_name, prefix_len) == 0) {
+ strcpy(ks_namebuf, name);
+ /* Work out the longest name that matches the prefix */
+ if (++number == 1) {
+ prev_len = min_t(int, max_len-1,
+ strlen(ks_namebuf));
+ memcpy(ks_namebuf_prev, ks_namebuf, prev_len);
+ ks_namebuf_prev[prev_len] = '\0';
+ continue;
+ }
+ for (i = 0; i < prev_len; i++) {
+ if (ks_namebuf[i] != ks_namebuf_prev[i]) {
+ prev_len = i;
+ ks_namebuf_prev[i] = '\0';
+ break;
+ }
+ }
+ }
+ }
+ if (prev_len > prefix_len)
+ memcpy(prefix_name, ks_namebuf_prev, prev_len+1);
+ return number;
+}
+
+/*
+ * kallsyms_symbol_next
+ *
+ * Parameters:
+ * prefix_name prefix of a symbol name to lookup
+ * flag 0 means search from the head, 1 means continue search.
+ * Returns:
+ * 1 if a symbol matches the given prefix.
+ * 0 if no string found
+ */
+int kallsyms_symbol_next(char *prefix_name, int flag)
+{
+ int prefix_len = strlen(prefix_name);
+ static loff_t pos;
+ const char *name;
+
+ if (!flag)
+ pos = 0;
+
+ while ((name = kdb_walk_kallsyms(&pos))) {
+ if (strncmp(name, prefix_name, prefix_len) == 0) {
+ strncpy(prefix_name, name, strlen(name)+1);
+ return 1;
+ }
+ }
+ return 0;
+}
+
+/*
+ * kdb_symbol_print - Standard method for printing a symbol name and offset.
+ * Inputs:
+ * addr Address to be printed.
+ * symtab Address of symbol data, if NULL this routine does its
+ * own lookup.
+ * punc Punctuation for string, bit field.
+ * Remarks:
+ * The string and its punctuation is only printed if the address
+ * is inside the kernel, except that the value is always printed
+ * when requested.
+ */
+void kdb_symbol_print(unsigned long addr, const kdb_symtab_t *symtab_p,
+ unsigned int punc)
+{
+ kdb_symtab_t symtab, *symtab_p2;
+ if (symtab_p) {
+ symtab_p2 = (kdb_symtab_t *)symtab_p;
+ } else {
+ symtab_p2 = &symtab;
+ kdbnearsym(addr, symtab_p2);
+ }
+ if (!(symtab_p2->sym_name || (punc & KDB_SP_VALUE)))
+ return;
+ if (punc & KDB_SP_SPACEB)
+ kdb_printf(" ");
+ if (punc & KDB_SP_VALUE)
+ kdb_printf(kdb_machreg_fmt0, addr);
+ if (symtab_p2->sym_name) {
+ if (punc & KDB_SP_VALUE)
+ kdb_printf(" ");
+ if (punc & KDB_SP_PAREN)
+ kdb_printf("(");
+ if (strcmp(symtab_p2->mod_name, "kernel"))
+ kdb_printf("[%s]", symtab_p2->mod_name);
+ kdb_printf("%s", symtab_p2->sym_name);
+ if (addr != symtab_p2->sym_start)
+ kdb_printf("+0x%lx", addr - symtab_p2->sym_start);
+ if (punc & KDB_SP_SYMSIZE)
+ kdb_printf("/0x%lx",
+ symtab_p2->sym_end - symtab_p2->sym_start);
+ if (punc & KDB_SP_PAREN)
+ kdb_printf(")");
+ }
+ if (punc & KDB_SP_SPACEA)
+ kdb_printf(" ");
+ if (punc & KDB_SP_NEWLINE)
+ kdb_printf("\n");
+}
+
+/*
+ * kdb_strdup - kdb equivalent of strdup, for disasm code.
+ * Inputs:
+ * str The string to duplicate.
+ * type Flags to kmalloc for the new string.
+ * Returns:
+ * Address of the new string, NULL if storage could not be allocated.
+ * Remarks:
+ * This is not in lib/string.c because it uses kmalloc which is not
+ * available when string.o is used in boot loaders.
+ */
+char *kdb_strdup(const char *str, gfp_t type)
+{
+ int n = strlen(str)+1;
+ char *s = kmalloc(n, type);
+ if (!s)
+ return NULL;
+ return strcpy(s, str);
+}
+
+/*
+ * kdb_getarea_size - Read an area of data. The kdb equivalent of
+ * copy_from_user, with kdb messages for invalid addresses.
+ * Inputs:
+ * res Pointer to the area to receive the result.
+ * addr Address of the area to copy.
+ * size Size of the area.
+ * Returns:
+ * 0 for success, < 0 for error.
+ */
+int kdb_getarea_size(void *res, unsigned long addr, size_t size)
+{
+ int ret = probe_kernel_read((char *)res, (char *)addr, size);
+ if (ret) {
+ if (!KDB_STATE(SUPPRESS)) {
+ kdb_printf("kdb_getarea: Bad address 0x%lx\n", addr);
+ KDB_STATE_SET(SUPPRESS);
+ }
+ ret = KDB_BADADDR;
+ } else {
+ KDB_STATE_CLEAR(SUPPRESS);
+ }
+ return ret;
+}
+
+/*
+ * kdb_putarea_size - Write an area of data. The kdb equivalent of
+ * copy_to_user, with kdb messages for invalid addresses.
+ * Inputs:
+ * addr Address of the area to write to.
+ * res Pointer to the area holding the data.
+ * size Size of the area.
+ * Returns:
+ * 0 for success, < 0 for error.
+ */
+int kdb_putarea_size(unsigned long addr, void *res, size_t size)
+{
+ int ret = probe_kernel_read((char *)addr, (char *)res, size);
+ if (ret) {
+ if (!KDB_STATE(SUPPRESS)) {
+ kdb_printf("kdb_putarea: Bad address 0x%lx\n", addr);
+ KDB_STATE_SET(SUPPRESS);
+ }
+ ret = KDB_BADADDR;
+ } else {
+ KDB_STATE_CLEAR(SUPPRESS);
+ }
+ return ret;
+}
+
+/*
+ * kdb_getphys - Read data from a physical address. Validate the
+ * address is in range, use kmap_atomic() to get data
+ * similar to kdb_getarea() - but for phys addresses
+ * Inputs:
+ * res Pointer to the word to receive the result
+ * addr Physical address of the area to copy
+ * size Size of the area
+ * Returns:
+ * 0 for success, < 0 for error.
+ */
+static int kdb_getphys(void *res, unsigned long addr, size_t size)
+{
+ unsigned long pfn;
+ void *vaddr;
+ struct page *page;
+
+ pfn = (addr >> PAGE_SHIFT);
+ if (!pfn_valid(pfn))
+ return 1;
+ page = pfn_to_page(pfn);
+ vaddr = kmap_atomic(page, KM_KDB);
+ memcpy(res, vaddr + (addr & (PAGE_SIZE - 1)), size);
+ kunmap_atomic(vaddr, KM_KDB);
+
+ return 0;
+}
+
+/*
+ * kdb_getphysword
+ * Inputs:
+ * word Pointer to the word to receive the result.
+ * addr Address of the area to copy.
+ * size Size of the area.
+ * Returns:
+ * 0 for success, < 0 for error.
+ */
+int kdb_getphysword(unsigned long *word, unsigned long addr, size_t size)
+{
+ int diag;
+ __u8 w1;
+ __u16 w2;
+ __u32 w4;
+ __u64 w8;
+ *word = 0; /* Default value if addr or size is invalid */
+
+ switch (size) {
+ case 1:
+ diag = kdb_getphys(&w1, addr, sizeof(w1));
+ if (!diag)
+ *word = w1;
+ break;
+ case 2:
+ diag = kdb_getphys(&w2, addr, sizeof(w2));
+ if (!diag)
+ *word = w2;
+ break;
+ case 4:
+ diag = kdb_getphys(&w4, addr, sizeof(w4));
+ if (!diag)
+ *word = w4;
+ break;
+ case 8:
+ if (size <= sizeof(*word)) {
+ diag = kdb_getphys(&w8, addr, sizeof(w8));
+ if (!diag)
+ *word = w8;
+ break;
+ }
+ /* drop through */
+ default:
+ diag = KDB_BADWIDTH;
+ kdb_printf("kdb_getphysword: bad width %ld\n", (long) size);
+ }
+ return diag;
+}
+
+/*
+ * kdb_getword - Read a binary value. Unlike kdb_getarea, this treats
+ * data as numbers.
+ * Inputs:
+ * word Pointer to the word to receive the result.
+ * addr Address of the area to copy.
+ * size Size of the area.
+ * Returns:
+ * 0 for success, < 0 for error.
+ */
+int kdb_getword(unsigned long *word, unsigned long addr, size_t size)
+{
+ int diag;
+ __u8 w1;
+ __u16 w2;
+ __u32 w4;
+ __u64 w8;
+ *word = 0; /* Default value if addr or size is invalid */
+ switch (size) {
+ case 1:
+ diag = kdb_getarea(w1, addr);
+ if (!diag)
+ *word = w1;
+ break;
+ case 2:
+ diag = kdb_getarea(w2, addr);
+ if (!diag)
+ *word = w2;
+ break;
+ case 4:
+ diag = kdb_getarea(w4, addr);
+ if (!diag)
+ *word = w4;
+ break;
+ case 8:
+ if (size <= sizeof(*word)) {
+ diag = kdb_getarea(w8, addr);
+ if (!diag)
+ *word = w8;
+ break;
+ }
+ /* drop through */
+ default:
+ diag = KDB_BADWIDTH;
+ kdb_printf("kdb_getword: bad width %ld\n", (long) size);
+ }
+ return diag;
+}
+
+/*
+ * kdb_putword - Write a binary value. Unlike kdb_putarea, this
+ * treats data as numbers.
+ * Inputs:
+ * addr Address of the area to write to..
+ * word The value to set.
+ * size Size of the area.
+ * Returns:
+ * 0 for success, < 0 for error.
+ */
+int kdb_putword(unsigned long addr, unsigned long word, size_t size)
+{
+ int diag;
+ __u8 w1;
+ __u16 w2;
+ __u32 w4;
+ __u64 w8;
+ switch (size) {
+ case 1:
+ w1 = word;
+ diag = kdb_putarea(addr, w1);
+ break;
+ case 2:
+ w2 = word;
+ diag = kdb_putarea(addr, w2);
+ break;
+ case 4:
+ w4 = word;
+ diag = kdb_putarea(addr, w4);
+ break;
+ case 8:
+ if (size <= sizeof(word)) {
+ w8 = word;
+ diag = kdb_putarea(addr, w8);
+ break;
+ }
+ /* drop through */
+ default:
+ diag = KDB_BADWIDTH;
+ kdb_printf("kdb_putword: bad width %ld\n", (long) size);
+ }
+ return diag;
+}
+
+/*
+ * kdb_task_state_string - Convert a string containing any of the
+ * letters DRSTCZEUIMA to a mask for the process state field and
+ * return the value. If no argument is supplied, return the mask
+ * that corresponds to environment variable PS, DRSTCZEU by
+ * default.
+ * Inputs:
+ * s String to convert
+ * Returns:
+ * Mask for process state.
+ * Notes:
+ * The mask folds data from several sources into a single long value, so
+ * be carefull not to overlap the bits. TASK_* bits are in the LSB,
+ * special cases like UNRUNNABLE are in the MSB. As of 2.6.10-rc1 there
+ * is no overlap between TASK_* and EXIT_* but that may not always be
+ * true, so EXIT_* bits are shifted left 16 bits before being stored in
+ * the mask.
+ */
+
+/* unrunnable is < 0 */
+#define UNRUNNABLE (1UL << (8*sizeof(unsigned long) - 1))
+#define RUNNING (1UL << (8*sizeof(unsigned long) - 2))
+#define IDLE (1UL << (8*sizeof(unsigned long) - 3))
+#define DAEMON (1UL << (8*sizeof(unsigned long) - 4))
+
+unsigned long kdb_task_state_string(const char *s)
+{
+ long res = 0;
+ if (!s) {
+ s = kdbgetenv("PS");
+ if (!s)
+ s = "DRSTCZEU"; /* default value for ps */
+ }
+ while (*s) {
+ switch (*s) {
+ case 'D':
+ res |= TASK_UNINTERRUPTIBLE;
+ break;
+ case 'R':
+ res |= RUNNING;
+ break;
+ case 'S':
+ res |= TASK_INTERRUPTIBLE;
+ break;
+ case 'T':
+ res |= TASK_STOPPED;
+ break;
+ case 'C':
+ res |= TASK_TRACED;
+ break;
+ case 'Z':
+ res |= EXIT_ZOMBIE << 16;
+ break;
+ case 'E':
+ res |= EXIT_DEAD << 16;
+ break;
+ case 'U':
+ res |= UNRUNNABLE;
+ break;
+ case 'I':
+ res |= IDLE;
+ break;
+ case 'M':
+ res |= DAEMON;
+ break;
+ case 'A':
+ res = ~0UL;
+ break;
+ default:
+ kdb_printf("%s: unknown flag '%c' ignored\n",
+ __func__, *s);
+ break;
+ }
+ ++s;
+ }
+ return res;
+}
+
+/*
+ * kdb_task_state_char - Return the character that represents the task state.
+ * Inputs:
+ * p struct task for the process
+ * Returns:
+ * One character to represent the task state.
+ */
+char kdb_task_state_char (const struct task_struct *p)
+{
+ int cpu;
+ char state;
+ unsigned long tmp;
+
+ if (!p || probe_kernel_read(&tmp, (char *)p, sizeof(unsigned long)))
+ return 'E';
+
+ cpu = kdb_process_cpu(p);
+ state = (p->state == 0) ? 'R' :
+ (p->state < 0) ? 'U' :
+ (p->state & TASK_UNINTERRUPTIBLE) ? 'D' :
+ (p->state & TASK_STOPPED) ? 'T' :
+ (p->state & TASK_TRACED) ? 'C' :
+ (p->exit_state & EXIT_ZOMBIE) ? 'Z' :
+ (p->exit_state & EXIT_DEAD) ? 'E' :
+ (p->state & TASK_INTERRUPTIBLE) ? 'S' : '?';
+ if (p->pid == 0) {
+ /* Idle task. Is it really idle, apart from the kdb
+ * interrupt? */
+ if (!kdb_task_has_cpu(p) || kgdb_info[cpu].irq_depth == 1) {
+ if (cpu != kdb_initial_cpu)
+ state = 'I'; /* idle task */
+ }
+ } else if (!p->mm && state == 'S') {
+ state = 'M'; /* sleeping system daemon */
+ }
+ return state;
+}
+
+/*
+ * kdb_task_state - Return true if a process has the desired state
+ * given by the mask.
+ * Inputs:
+ * p struct task for the process
+ * mask mask from kdb_task_state_string to select processes
+ * Returns:
+ * True if the process matches at least one criteria defined by the mask.
+ */
+unsigned long kdb_task_state(const struct task_struct *p, unsigned long mask)
+{
+ char state[] = { kdb_task_state_char(p), '\0' };
+ return (mask & kdb_task_state_string(state)) != 0;
+}
+
+/*
+ * kdb_print_nameval - Print a name and its value, converting the
+ * value to a symbol lookup if possible.
+ * Inputs:
+ * name field name to print
+ * val value of field
+ */
+void kdb_print_nameval(const char *name, unsigned long val)
+{
+ kdb_symtab_t symtab;
+ kdb_printf(" %-11.11s ", name);
+ if (kdbnearsym(val, &symtab))
+ kdb_symbol_print(val, &symtab,
+ KDB_SP_VALUE|KDB_SP_SYMSIZE|KDB_SP_NEWLINE);
+ else
+ kdb_printf("0x%lx\n", val);
+}
+
+/* Last ditch allocator for debugging, so we can still debug even when
+ * the GFP_ATOMIC pool has been exhausted. The algorithms are tuned
+ * for space usage, not for speed. One smallish memory pool, the free
+ * chain is always in ascending address order to allow coalescing,
+ * allocations are done in brute force best fit.
+ */
+
+struct debug_alloc_header {
+ u32 next; /* offset of next header from start of pool */
+ u32 size;
+ void *caller;
+};
+
+/* The memory returned by this allocator must be aligned, which means
+ * so must the header size. Do not assume that sizeof(struct
+ * debug_alloc_header) is a multiple of the alignment, explicitly
+ * calculate the overhead of this header, including the alignment.
+ * The rest of this code must not use sizeof() on any header or
+ * pointer to a header.
+ */
+#define dah_align 8
+#define dah_overhead ALIGN(sizeof(struct debug_alloc_header), dah_align)
+
+static u64 debug_alloc_pool_aligned[256*1024/dah_align]; /* 256K pool */
+static char *debug_alloc_pool = (char *)debug_alloc_pool_aligned;
+static u32 dah_first, dah_first_call = 1, dah_used, dah_used_max;
+
+/* Locking is awkward. The debug code is called from all contexts,
+ * including non maskable interrupts. A normal spinlock is not safe
+ * in NMI context. Try to get the debug allocator lock, if it cannot
+ * be obtained after a second then give up. If the lock could not be
+ * previously obtained on this cpu then only try once.
+ *
+ * sparse has no annotation for "this function _sometimes_ acquires a
+ * lock", so fudge the acquire/release notation.
+ */
+static DEFINE_SPINLOCK(dap_lock);
+static int get_dap_lock(void)
+ __acquires(dap_lock)
+{
+ static int dap_locked = -1;
+ int count;
+ if (dap_locked == smp_processor_id())
+ count = 1;
+ else
+ count = 1000;
+ while (1) {
+ if (spin_trylock(&dap_lock)) {
+ dap_locked = -1;
+ return 1;
+ }
+ if (!count--)
+ break;
+ udelay(1000);
+ }
+ dap_locked = smp_processor_id();
+ __acquire(dap_lock);
+ return 0;
+}
+
+void *debug_kmalloc(size_t size, gfp_t flags)
+{
+ unsigned int rem, h_offset;
+ struct debug_alloc_header *best, *bestprev, *prev, *h;
+ void *p = NULL;
+ if (!get_dap_lock()) {
+ __release(dap_lock); /* we never actually got it */
+ return NULL;
+ }
+ h = (struct debug_alloc_header *)(debug_alloc_pool + dah_first);
+ if (dah_first_call) {
+ h->size = sizeof(debug_alloc_pool_aligned) - dah_overhead;
+ dah_first_call = 0;
+ }
+ size = ALIGN(size, dah_align);
+ prev = best = bestprev = NULL;
+ while (1) {
+ if (h->size >= size && (!best || h->size < best->size)) {
+ best = h;
+ bestprev = prev;
+ if (h->size == size)
+ break;
+ }
+ if (!h->next)
+ break;
+ prev = h;
+ h = (struct debug_alloc_header *)(debug_alloc_pool + h->next);
+ }
+ if (!best)
+ goto out;
+ rem = best->size - size;
+ /* The pool must always contain at least one header */
+ if (best->next == 0 && bestprev == NULL && rem < dah_overhead)
+ goto out;
+ if (rem >= dah_overhead) {
+ best->size = size;
+ h_offset = ((char *)best - debug_alloc_pool) +
+ dah_overhead + best->size;
+ h = (struct debug_alloc_header *)(debug_alloc_pool + h_offset);
+ h->size = rem - dah_overhead;
+ h->next = best->next;
+ } else
+ h_offset = best->next;
+ best->caller = __builtin_return_address(0);
+ dah_used += best->size;
+ dah_used_max = max(dah_used, dah_used_max);
+ if (bestprev)
+ bestprev->next = h_offset;
+ else
+ dah_first = h_offset;
+ p = (char *)best + dah_overhead;
+ memset(p, POISON_INUSE, best->size - 1);
+ *((char *)p + best->size - 1) = POISON_END;
+out:
+ spin_unlock(&dap_lock);
+ return p;
+}
+
+void debug_kfree(void *p)
+{
+ struct debug_alloc_header *h;
+ unsigned int h_offset;
+ if (!p)
+ return;
+ if ((char *)p < debug_alloc_pool ||
+ (char *)p >= debug_alloc_pool + sizeof(debug_alloc_pool_aligned)) {
+ kfree(p);
+ return;
+ }
+ if (!get_dap_lock()) {
+ __release(dap_lock); /* we never actually got it */
+ return; /* memory leak, cannot be helped */
+ }
+ h = (struct debug_alloc_header *)((char *)p - dah_overhead);
+ memset(p, POISON_FREE, h->size - 1);
+ *((char *)p + h->size - 1) = POISON_END;
+ h->caller = NULL;
+ dah_used -= h->size;
+ h_offset = (char *)h - debug_alloc_pool;
+ if (h_offset < dah_first) {
+ h->next = dah_first;
+ dah_first = h_offset;
+ } else {
+ struct debug_alloc_header *prev;
+ unsigned int prev_offset;
+ prev = (struct debug_alloc_header *)(debug_alloc_pool +
+ dah_first);
+ while (1) {
+ if (!prev->next || prev->next > h_offset)
+ break;
+ prev = (struct debug_alloc_header *)
+ (debug_alloc_pool + prev->next);
+ }
+ prev_offset = (char *)prev - debug_alloc_pool;
+ if (prev_offset + dah_overhead + prev->size == h_offset) {
+ prev->size += dah_overhead + h->size;
+ memset(h, POISON_FREE, dah_overhead - 1);
+ *((char *)h + dah_overhead - 1) = POISON_END;
+ h = prev;
+ h_offset = prev_offset;
+ } else {
+ h->next = prev->next;
+ prev->next = h_offset;
+ }
+ }
+ if (h_offset + dah_overhead + h->size == h->next) {
+ struct debug_alloc_header *next;
+ next = (struct debug_alloc_header *)
+ (debug_alloc_pool + h->next);
+ h->size += dah_overhead + next->size;
+ h->next = next->next;
+ memset(next, POISON_FREE, dah_overhead - 1);
+ *((char *)next + dah_overhead - 1) = POISON_END;
+ }
+ spin_unlock(&dap_lock);
+}
+
+void debug_kusage(void)
+{
+ struct debug_alloc_header *h_free, *h_used;
+#ifdef CONFIG_IA64
+ /* FIXME: using dah for ia64 unwind always results in a memory leak.
+ * Fix that memory leak first, then set debug_kusage_one_time = 1 for
+ * all architectures.
+ */
+ static int debug_kusage_one_time;
+#else
+ static int debug_kusage_one_time = 1;
+#endif
+ if (!get_dap_lock()) {
+ __release(dap_lock); /* we never actually got it */
+ return;
+ }
+ h_free = (struct debug_alloc_header *)(debug_alloc_pool + dah_first);
+ if (dah_first == 0 &&
+ (h_free->size == sizeof(debug_alloc_pool_aligned) - dah_overhead ||
+ dah_first_call))
+ goto out;
+ if (!debug_kusage_one_time)
+ goto out;
+ debug_kusage_one_time = 0;
+ kdb_printf("%s: debug_kmalloc memory leak dah_first %d\n",
+ __func__, dah_first);
+ if (dah_first) {
+ h_used = (struct debug_alloc_header *)debug_alloc_pool;
+ kdb_printf("%s: h_used %p size %d\n", __func__, h_used,
+ h_used->size);
+ }
+ do {
+ h_used = (struct debug_alloc_header *)
+ ((char *)h_free + dah_overhead + h_free->size);
+ kdb_printf("%s: h_used %p size %d caller %p\n",
+ __func__, h_used, h_used->size, h_used->caller);
+ h_free = (struct debug_alloc_header *)
+ (debug_alloc_pool + h_free->next);
+ } while (h_free->next);
+ h_used = (struct debug_alloc_header *)
+ ((char *)h_free + dah_overhead + h_free->size);
+ if ((char *)h_used - debug_alloc_pool !=
+ sizeof(debug_alloc_pool_aligned))
+ kdb_printf("%s: h_used %p size %d caller %p\n",
+ __func__, h_used, h_used->size, h_used->caller);
+out:
+ spin_unlock(&dap_lock);
+}
+
+/* Maintain a small stack of kdb_flags to allow recursion without disturbing
+ * the global kdb state.
+ */
+
+static int kdb_flags_stack[4], kdb_flags_index;
+
+void kdb_save_flags(void)
+{
+ BUG_ON(kdb_flags_index >= ARRAY_SIZE(kdb_flags_stack));
+ kdb_flags_stack[kdb_flags_index++] = kdb_flags;
+}
+
+void kdb_restore_flags(void)
+{
+ BUG_ON(kdb_flags_index <= 0);
+ kdb_flags = kdb_flags_stack[--kdb_flags_index];
+}
--
1.6.4.rc1
The only way the debugger can handle a trap in inside rcu_lock,
notify_die, or atomic_notifier_call_chain without a triple fault is
to have a low level "first opportunity handler" in the int3 exception
handler.
Generally this will be something the vast majority of folks will not
need, but for those who need it, it is added as a kernel .config
option called KGDB_LOW_LEVEL_TRAP.
CC: Ingo Molnar <[email protected]>
Signed-off-by: Jason Wessel <[email protected]>
---
arch/x86/include/asm/kgdb.h | 3 +++
arch/x86/kernel/kgdb.c | 22 +++++++++++++++++++++-
arch/x86/kernel/traps.c | 6 ++++++
include/linux/kgdb.h | 1 +
kernel/debug/debug_core.c | 2 +-
lib/Kconfig.kgdb | 9 +++++++++
6 files changed, 41 insertions(+), 2 deletions(-)
diff --git a/arch/x86/include/asm/kgdb.h b/arch/x86/include/asm/kgdb.h
index e6c6c80..006da36 100644
--- a/arch/x86/include/asm/kgdb.h
+++ b/arch/x86/include/asm/kgdb.h
@@ -76,4 +76,7 @@ static inline void arch_kgdb_breakpoint(void)
#define BREAK_INSTR_SIZE 1
#define CACHE_FLUSH_IS_SAFE 1
+extern int kgdb_ll_trap(int cmd, const char *str,
+ struct pt_regs *regs, long err, int trap, int sig);
+
#endif /* _ASM_X86_KGDB_H */
diff --git a/arch/x86/kernel/kgdb.c b/arch/x86/kernel/kgdb.c
index c38c8fb..3b762bc 100644
--- a/arch/x86/kernel/kgdb.c
+++ b/arch/x86/kernel/kgdb.c
@@ -538,7 +538,7 @@ static int __kgdb_notify(struct die_args *args, unsigned long cmd)
return NOTIFY_DONE;
}
- if (kgdb_handle_exception(args->trapnr, args->signr, args->err, regs))
+ if (kgdb_handle_exception(args->trapnr, args->signr, cmd, regs))
return NOTIFY_DONE;
/* Must touch watchdog before return to normal operation */
@@ -546,6 +546,26 @@ static int __kgdb_notify(struct die_args *args, unsigned long cmd)
return NOTIFY_STOP;
}
+#ifdef CONFIG_KGDB_LOW_LEVEL_TRAP
+int kgdb_ll_trap(int cmd, const char *str,
+ struct pt_regs *regs, long err, int trap, int sig)
+{
+ struct die_args args = {
+ .regs = regs,
+ .str = str,
+ .err = err,
+ .trapnr = trap,
+ .signr = sig,
+
+ };
+
+ if (!kgdb_io_module_registered)
+ return NOTIFY_DONE;
+
+ return __kgdb_notify(&args, cmd);
+}
+#endif /* CONFIG_KGDB_LOW_LEVEL_TRAP */
+
static int
kgdb_notify(struct notifier_block *self, unsigned long cmd, void *ptr)
{
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index 3339917..6ae2122 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -15,6 +15,7 @@
#include <linux/kprobes.h>
#include <linux/uaccess.h>
#include <linux/kdebug.h>
+#include <linux/kgdb.h>
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/ptrace.h>
@@ -460,6 +461,11 @@ void restart_nmi(void)
/* May run on IST stack. */
dotraplinkage void __kprobes do_int3(struct pt_regs *regs, long error_code)
{
+#ifdef CONFIG_KGDB_LOW_LEVEL_TRAP
+ if (kgdb_ll_trap(DIE_INT3, "int3", regs, error_code, 3, SIGTRAP)
+ == NOTIFY_STOP)
+ return;
+#endif /* CONFIG_KGDB_LOW_LEVEL_TRAP */
#ifdef CONFIG_KPROBES
if (notify_die(DIE_INT3, "int3", regs, error_code, 3, SIGTRAP)
== NOTIFY_STOP)
diff --git a/include/linux/kgdb.h b/include/linux/kgdb.h
index 406f6f9..19d1b29 100644
--- a/include/linux/kgdb.h
+++ b/include/linux/kgdb.h
@@ -60,6 +60,7 @@ struct uart_port;
void kgdb_breakpoint(void);
extern int kgdb_connected;
+extern int kgdb_io_module_registered;
extern atomic_t kgdb_setting_breakpoint;
extern atomic_t kgdb_cpu_doing_single_step;
diff --git a/kernel/debug/debug_core.c b/kernel/debug/debug_core.c
index 9cfc71d..8db8e40 100644
--- a/kernel/debug/debug_core.c
+++ b/kernel/debug/debug_core.c
@@ -66,7 +66,7 @@ int kgdb_connected;
EXPORT_SYMBOL_GPL(kgdb_connected);
/* All the KGDB handlers are installed */
-static int kgdb_io_module_registered;
+int kgdb_io_module_registered;
/* Guard for recursive entry */
static int exception_level;
diff --git a/lib/Kconfig.kgdb b/lib/Kconfig.kgdb
index c80de7d..0a0e049 100644
--- a/lib/Kconfig.kgdb
+++ b/lib/Kconfig.kgdb
@@ -57,6 +57,15 @@ config KGDB_TESTS_BOOT_STRING
information about other strings you could use beyond the
default of V1F100.
+config KGDB_LOW_LEVEL_TRAP
+ bool "KGDB: Allow debugging with traps in notifiers"
+ depends on X86
+ default n
+ help
+ This will add an extra call back to kgdb for the breakpoint
+ exception handler on which will will allow kgdb to step
+ through a notify handler.
+
config KGDB_KDB
bool "KGDB_KDB: include kdb frontend for kgdb"
default n
--
1.6.4.rc1
The new debug core api requires all architectures that use to debug
core to implement a function to set the program counter.
CC: Mike Frysinger <[email protected]>
Signed-off-by: Jason Wessel <[email protected]>
Acked-by: Sonic Zhang <[email protected]>
---
arch/blackfin/kernel/kgdb.c | 5 +++++
1 files changed, 5 insertions(+), 0 deletions(-)
diff --git a/arch/blackfin/kernel/kgdb.c b/arch/blackfin/kernel/kgdb.c
index 34c7c3e..8cb6204 100644
--- a/arch/blackfin/kernel/kgdb.c
+++ b/arch/blackfin/kernel/kgdb.c
@@ -439,6 +439,11 @@ int kgdb_validate_break_address(unsigned long addr)
return -EFAULT;
}
+void kgdb_arch_set_pc(struct pt_regs *regs, unsigned long ip)
+{
+ regs->retx = ip;
+}
+
int kgdb_arch_init(void)
{
kgdb_single_step = 0;
--
1.6.4.rc1
Implement kgdb_arch_pc() which adjusts the pc if it needs to be
adjusted after a software breakpoint trap.
Implement kgdb_arch_set_pc() which is a new required function in the
kgdb debug core.
When processing a single step return zero in the error exception field
so that the debug core can distinguish between a single step trap and
a breakpoint trap generically.
Signed-off-by: Jason Wessel <[email protected]>
Acked-by: Paul Mundt <[email protected]>
---
arch/sh/kernel/kgdb.c | 14 +++++++++++++-
1 files changed, 13 insertions(+), 1 deletions(-)
diff --git a/arch/sh/kernel/kgdb.c b/arch/sh/kernel/kgdb.c
index 3e532d0..1ec03c9 100644
--- a/arch/sh/kernel/kgdb.c
+++ b/arch/sh/kernel/kgdb.c
@@ -237,6 +237,18 @@ int kgdb_arch_handle_exception(int e_vector, int signo, int err_code,
return -1;
}
+unsigned long kgdb_arch_pc(int exception, struct pt_regs *regs)
+{
+ if (exception == 60)
+ return instruction_pointer(regs) - 2;
+ return instruction_pointer(regs);
+}
+
+void kgdb_arch_set_pc(struct pt_regs *regs, unsigned long ip)
+{
+ regs->pc = ip;
+}
+
/*
* The primary entry points for the kgdb debug trap table entries.
*/
@@ -247,7 +259,7 @@ BUILD_TRAP_HANDLER(singlestep)
local_irq_save(flags);
regs->pc -= instruction_size(__raw_readw(regs->pc - 4));
- kgdb_handle_exception(vec >> 2, SIGTRAP, 0, regs);
+ kgdb_handle_exception(0, SIGTRAP, 0, regs);
local_irq_restore(flags);
}
--
1.6.4.rc1
Split the former kernel/kgdb.c into debug_core.c which contains the
kernel debugger exception logic and to the gdbstub.c which contains
the logic for allowing gdb to talk to the debug core.
This also created a private include file called debug_core.h which
contains all the definitions to glue the debug_core to any other
debugger connections.
CC: Ingo Molnar <[email protected]>
Signed-off-by: Jason Wessel <[email protected]>
---
include/linux/kgdb.h | 1 +
kernel/debug/Makefile | 2 +-
kernel/debug/debug_core.c | 992 ++-------------------------------------------
kernel/debug/debug_core.h | 48 +++
kernel/debug/gdbstub.c | 939 ++++++++++++++++++++++++++++++++++++++++++
5 files changed, 1028 insertions(+), 954 deletions(-)
create mode 100644 kernel/debug/debug_core.h
create mode 100644 kernel/debug/gdbstub.c
diff --git a/include/linux/kgdb.h b/include/linux/kgdb.h
index 19ec41a..4830142 100644
--- a/include/linux/kgdb.h
+++ b/include/linux/kgdb.h
@@ -264,6 +264,7 @@ extern unsigned long __weak kgdb_arch_pc(int exception, struct pt_regs *regs);
extern int kgdb_register_io_module(struct kgdb_io *local_kgdb_io_ops);
extern void kgdb_unregister_io_module(struct kgdb_io *local_kgdb_io_ops);
+extern struct kgdb_io *dbg_io_ops;
extern int kgdb_hex2long(char **ptr, unsigned long *long_val);
extern int kgdb_mem2hex(char *mem, char *buf, int count);
diff --git a/kernel/debug/Makefile b/kernel/debug/Makefile
index 5287fc8..3657c64 100644
--- a/kernel/debug/Makefile
+++ b/kernel/debug/Makefile
@@ -2,4 +2,4 @@
# Makefile for the linux kernel debugger
#
-obj-$(CONFIG_KGDB) += debug_core.o
\ No newline at end of file
+obj-$(CONFIG_KGDB) += debug_core.o gdbstub.o
\ No newline at end of file
diff --git a/kernel/debug/debug_core.c b/kernel/debug/debug_core.c
index 761fdd2..ad3caaa 100644
--- a/kernel/debug/debug_core.c
+++ b/kernel/debug/debug_core.c
@@ -1,5 +1,5 @@
/*
- * KGDB stub.
+ * Kernel Debug Core
*
* Maintainer: Jason Wessel <[email protected]>
*
@@ -9,7 +9,7 @@
* Copyright (C) 2004 Pavel Machek <[email protected]>
* Copyright (C) 2004-2006 Tom Rini <[email protected]>
* Copyright (C) 2004-2006 LinSysSoft Technologies Pvt. Ltd.
- * Copyright (C) 2005-2008 Wind River Systems, Inc.
+ * Copyright (C) 2005-2009 Wind River Systems, Inc.
* Copyright (C) 2007 MontaVista Software, Inc.
* Copyright (C) 2008 Red Hat, Inc., Ingo Molnar <[email protected]>
*
@@ -37,7 +37,6 @@
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/ptrace.h>
-#include <linux/reboot.h>
#include <linux/string.h>
#include <linux/delay.h>
#include <linux/sched.h>
@@ -52,27 +51,12 @@
#include <asm/byteorder.h>
#include <asm/atomic.h>
#include <asm/system.h>
-#include <asm/unaligned.h>
-static int kgdb_break_asap;
+#include "debug_core.h"
-#define KGDB_MAX_THREAD_QUERY 17
-struct kgdb_state {
- int ex_vector;
- int signo;
- int err_code;
- int cpu;
- int pass_exception;
- unsigned long thr_query;
- unsigned long threadid;
- long kgdb_usethreadid;
- struct pt_regs *linux_regs;
-};
+static int kgdb_break_asap;
-static struct debuggerinfo_struct {
- void *debuggerinfo;
- struct task_struct *task;
-} kgdb_info[NR_CPUS];
+struct debuggerinfo_struct kgdb_info[NR_CPUS];
/**
* kgdb_connected - Is a host GDB connected to us?
@@ -86,7 +70,7 @@ static int kgdb_io_module_registered;
/* Guard for recursive entry */
static int exception_level;
-static struct kgdb_io *kgdb_io_ops;
+struct kgdb_io *dbg_io_ops;
static DEFINE_SPINLOCK(kgdb_registration_lock);
/* kgdb console driver is loaded */
@@ -129,16 +113,7 @@ struct task_struct *kgdb_usethread;
struct task_struct *kgdb_contthread;
int kgdb_single_step;
-pid_t kgdb_sstep_pid;
-
-/* Our I/O buffers. */
-static char remcom_in_buffer[BUFMAX];
-static char remcom_out_buffer[BUFMAX];
-
-/* Storage for the registers, in GDB format. */
-static unsigned long gdb_regs[(NUMREGBYTES +
- sizeof(unsigned long) - 1) /
- sizeof(unsigned long)];
+static pid_t kgdb_sstep_pid;
/* to keep track of the CPU which is doing the single stepping*/
atomic_t kgdb_cpu_doing_single_step = ATOMIC_INIT(-1);
@@ -240,329 +215,6 @@ void __weak kgdb_disable_hw_debug(struct pt_regs *regs)
}
/*
- * GDB remote protocol parser:
- */
-
-static int hex(char ch)
-{
- if ((ch >= 'a') && (ch <= 'f'))
- return ch - 'a' + 10;
- if ((ch >= '0') && (ch <= '9'))
- return ch - '0';
- if ((ch >= 'A') && (ch <= 'F'))
- return ch - 'A' + 10;
- return -1;
-}
-
-/* scan for the sequence $<data>#<checksum> */
-static void get_packet(char *buffer)
-{
- unsigned char checksum;
- unsigned char xmitcsum;
- int count;
- char ch;
-
- do {
- /*
- * Spin and wait around for the start character, ignore all
- * other characters:
- */
- while ((ch = (kgdb_io_ops->read_char())) != '$')
- /* nothing */;
-
- kgdb_connected = 1;
- checksum = 0;
- xmitcsum = -1;
-
- count = 0;
-
- /*
- * now, read until a # or end of buffer is found:
- */
- while (count < (BUFMAX - 1)) {
- ch = kgdb_io_ops->read_char();
- if (ch == '#')
- break;
- checksum = checksum + ch;
- buffer[count] = ch;
- count = count + 1;
- }
- buffer[count] = 0;
-
- if (ch == '#') {
- xmitcsum = hex(kgdb_io_ops->read_char()) << 4;
- xmitcsum += hex(kgdb_io_ops->read_char());
-
- if (checksum != xmitcsum)
- /* failed checksum */
- kgdb_io_ops->write_char('-');
- else
- /* successful transfer */
- kgdb_io_ops->write_char('+');
- if (kgdb_io_ops->flush)
- kgdb_io_ops->flush();
- }
- } while (checksum != xmitcsum);
-}
-
-/*
- * Send the packet in buffer.
- * Check for gdb connection if asked for.
- */
-static void put_packet(char *buffer)
-{
- unsigned char checksum;
- int count;
- char ch;
-
- /*
- * $<packet info>#<checksum>.
- */
- while (1) {
- kgdb_io_ops->write_char('$');
- checksum = 0;
- count = 0;
-
- while ((ch = buffer[count])) {
- kgdb_io_ops->write_char(ch);
- checksum += ch;
- count++;
- }
-
- kgdb_io_ops->write_char('#');
- kgdb_io_ops->write_char(hex_asc_hi(checksum));
- kgdb_io_ops->write_char(hex_asc_lo(checksum));
- if (kgdb_io_ops->flush)
- kgdb_io_ops->flush();
-
- /* Now see what we get in reply. */
- ch = kgdb_io_ops->read_char();
-
- if (ch == 3)
- ch = kgdb_io_ops->read_char();
-
- /* If we get an ACK, we are done. */
- if (ch == '+')
- return;
-
- /*
- * If we get the start of another packet, this means
- * that GDB is attempting to reconnect. We will NAK
- * the packet being sent, and stop trying to send this
- * packet.
- */
- if (ch == '$') {
- kgdb_io_ops->write_char('-');
- if (kgdb_io_ops->flush)
- kgdb_io_ops->flush();
- return;
- }
- }
-}
-
-/*
- * Convert the memory pointed to by mem into hex, placing result in buf.
- * Return a pointer to the last char put in buf (null). May return an error.
- */
-int kgdb_mem2hex(char *mem, char *buf, int count)
-{
- char *tmp;
- int err;
-
- /*
- * We use the upper half of buf as an intermediate buffer for the
- * raw memory copy. Hex conversion will work against this one.
- */
- tmp = buf + count;
-
- err = probe_kernel_read(tmp, mem, count);
- if (!err) {
- while (count > 0) {
- buf = pack_hex_byte(buf, *tmp);
- tmp++;
- count--;
- }
-
- *buf = 0;
- }
-
- return err;
-}
-
-/*
- * Copy the binary array pointed to by buf into mem. Fix $, #, and
- * 0x7d escaped with 0x7d. Return a pointer to the character after
- * the last byte written.
- */
-static int kgdb_ebin2mem(char *buf, char *mem, int count)
-{
- int err = 0;
- char c;
-
- while (count-- > 0) {
- c = *buf++;
- if (c == 0x7d)
- c = *buf++ ^ 0x20;
-
- err = probe_kernel_write(mem, &c, 1);
- if (err)
- break;
-
- mem++;
- }
-
- return err;
-}
-
-/*
- * Convert the hex array pointed to by buf into binary to be placed in mem.
- * Return a pointer to the character AFTER the last byte written.
- * May return an error.
- */
-int kgdb_hex2mem(char *buf, char *mem, int count)
-{
- char *tmp_raw;
- char *tmp_hex;
-
- /*
- * We use the upper half of buf as an intermediate buffer for the
- * raw memory that is converted from hex.
- */
- tmp_raw = buf + count * 2;
-
- tmp_hex = tmp_raw - 1;
- while (tmp_hex >= buf) {
- tmp_raw--;
- *tmp_raw = hex(*tmp_hex--);
- *tmp_raw |= hex(*tmp_hex--) << 4;
- }
-
- return probe_kernel_write(mem, tmp_raw, count);
-}
-
-/*
- * While we find nice hex chars, build a long_val.
- * Return number of chars processed.
- */
-int kgdb_hex2long(char **ptr, unsigned long *long_val)
-{
- int hex_val;
- int num = 0;
- int negate = 0;
-
- *long_val = 0;
-
- if (**ptr == '-') {
- negate = 1;
- (*ptr)++;
- }
- while (**ptr) {
- hex_val = hex(**ptr);
- if (hex_val < 0)
- break;
-
- *long_val = (*long_val << 4) | hex_val;
- num++;
- (*ptr)++;
- }
-
- if (negate)
- *long_val = -*long_val;
-
- return num;
-}
-
-/* Write memory due to an 'M' or 'X' packet. */
-static int write_mem_msg(int binary)
-{
- char *ptr = &remcom_in_buffer[1];
- unsigned long addr;
- unsigned long length;
- int err;
-
- if (kgdb_hex2long(&ptr, &addr) > 0 && *(ptr++) == ',' &&
- kgdb_hex2long(&ptr, &length) > 0 && *(ptr++) == ':') {
- if (binary)
- err = kgdb_ebin2mem(ptr, (char *)addr, length);
- else
- err = kgdb_hex2mem(ptr, (char *)addr, length);
- if (err)
- return err;
- if (CACHE_FLUSH_IS_SAFE)
- flush_icache_range(addr, addr + length);
- return 0;
- }
-
- return -EINVAL;
-}
-
-static void error_packet(char *pkt, int error)
-{
- error = -error;
- pkt[0] = 'E';
- pkt[1] = hex_asc[(error / 10)];
- pkt[2] = hex_asc[(error % 10)];
- pkt[3] = '\0';
-}
-
-/*
- * Thread ID accessors. We represent a flat TID space to GDB, where
- * the per CPU idle threads (which under Linux all have PID 0) are
- * remapped to negative TIDs.
- */
-
-#define BUF_THREAD_ID_SIZE 16
-
-static char *pack_threadid(char *pkt, unsigned char *id)
-{
- char *limit;
-
- limit = pkt + BUF_THREAD_ID_SIZE;
- while (pkt < limit)
- pkt = pack_hex_byte(pkt, *id++);
-
- return pkt;
-}
-
-static void int_to_threadref(unsigned char *id, int value)
-{
- unsigned char *scan;
- int i = 4;
-
- scan = (unsigned char *)id;
- while (i--)
- *scan++ = 0;
- put_unaligned_be32(value, scan);
-}
-
-static struct task_struct *getthread(struct pt_regs *regs, int tid)
-{
- /*
- * Non-positive TIDs are remapped to the cpu shadow information
- */
- if (tid == 0 || tid == -1)
- tid = -atomic_read(&kgdb_active) - 2;
- if (tid < -1 && tid > -NR_CPUS - 2) {
- if (kgdb_info[-tid - 2].task)
- return kgdb_info[-tid - 2].task;
- else
- return idle_task(-tid - 2);
- }
- if (tid <= 0) {
- printk(KERN_ERR "KGDB: Internal thread select error\n");
- dump_stack();
- return NULL;
- }
-
- /*
- * find_task_by_pid_ns() does not take the tasklist lock anymore
- * but is nicely RCU locked - hence is a pretty resilient
- * thing to use:
- */
- return find_task_by_pid_ns(tid, &init_pid_ns);
-}
-
-/*
* CPU debug state control:
*/
@@ -625,7 +277,7 @@ static void kgdb_flush_swbreak_addr(unsigned long addr)
/*
* SW breakpoint management:
*/
-static int kgdb_activate_sw_breakpoints(void)
+int dbg_activate_sw_breakpoints(void)
{
unsigned long addr;
int error;
@@ -651,7 +303,7 @@ static int kgdb_activate_sw_breakpoints(void)
return ret;
}
-static int kgdb_set_sw_break(unsigned long addr)
+int dbg_set_sw_break(unsigned long addr)
{
int err = kgdb_validate_break_address(addr);
int breakno = -1;
@@ -716,7 +368,7 @@ static int kgdb_deactivate_sw_breakpoints(void)
return ret;
}
-static int kgdb_remove_sw_break(unsigned long addr)
+int dbg_remove_sw_break(unsigned long addr)
{
int i;
@@ -742,7 +394,7 @@ int kgdb_isremovedbreak(unsigned long addr)
return 0;
}
-static int remove_all_break(void)
+int dbg_remove_all_break(void)
{
unsigned long addr;
int error;
@@ -770,53 +422,6 @@ setundefined:
}
/*
- * Remap normal tasks to their real PID,
- * CPU shadow threads are mapped to -CPU - 2
- */
-static inline int shadow_pid(int realpid)
-{
- if (realpid)
- return realpid;
-
- return -raw_smp_processor_id() - 2;
-}
-
-static char gdbmsgbuf[BUFMAX + 1];
-
-static void kgdb_msg_write(const char *s, int len)
-{
- char *bufptr;
- int wcount;
- int i;
-
- /* 'O'utput */
- gdbmsgbuf[0] = 'O';
-
- /* Fill and send buffers... */
- while (len > 0) {
- bufptr = gdbmsgbuf + 1;
-
- /* Calculate how many this time */
- if ((len << 1) > (BUFMAX - 2))
- wcount = (BUFMAX - 2) >> 1;
- else
- wcount = len;
-
- /* Pack in hex chars */
- for (i = 0; i < wcount; i++)
- bufptr = pack_hex_byte(bufptr, s[i]);
- *bufptr = '\0';
-
- /* Move up */
- s += wcount;
- len -= wcount;
-
- /* Write packet */
- put_packet(gdbmsgbuf);
- }
-}
-
-/*
* Return true if there is a valid kgdb I/O module. Also if no
* debugger is attached a message can be printed to the console about
* waiting for the debugger to attach.
@@ -827,7 +432,7 @@ static void kgdb_msg_write(const char *s, int len)
*/
static int kgdb_io_ready(int print_wait)
{
- if (!kgdb_io_ops)
+ if (!dbg_io_ops)
return 0;
if (kgdb_connected)
return 1;
@@ -838,525 +443,6 @@ static int kgdb_io_ready(int print_wait)
return 1;
}
-/*
- * All the functions that start with gdb_cmd are the various
- * operations to implement the handlers for the gdbserial protocol
- * where KGDB is communicating with an external debugger
- */
-
-/* Handle the '?' status packets */
-static void gdb_cmd_status(struct kgdb_state *ks)
-{
- /*
- * We know that this packet is only sent
- * during initial connect. So to be safe,
- * we clear out our breakpoints now in case
- * GDB is reconnecting.
- */
- remove_all_break();
-
- remcom_out_buffer[0] = 'S';
- pack_hex_byte(&remcom_out_buffer[1], ks->signo);
-}
-
-/* Handle the 'g' get registers request */
-static void gdb_cmd_getregs(struct kgdb_state *ks)
-{
- struct task_struct *thread;
- void *local_debuggerinfo;
- int i;
-
- thread = kgdb_usethread;
- if (!thread) {
- thread = kgdb_info[ks->cpu].task;
- local_debuggerinfo = kgdb_info[ks->cpu].debuggerinfo;
- } else {
- local_debuggerinfo = NULL;
- for_each_online_cpu(i) {
- /*
- * Try to find the task on some other
- * or possibly this node if we do not
- * find the matching task then we try
- * to approximate the results.
- */
- if (thread == kgdb_info[i].task)
- local_debuggerinfo = kgdb_info[i].debuggerinfo;
- }
- }
-
- /*
- * All threads that don't have debuggerinfo should be
- * in schedule() sleeping, since all other CPUs
- * are in kgdb_wait, and thus have debuggerinfo.
- */
- if (local_debuggerinfo) {
- pt_regs_to_gdb_regs(gdb_regs, local_debuggerinfo);
- } else {
- /*
- * Pull stuff saved during switch_to; nothing
- * else is accessible (or even particularly
- * relevant).
- *
- * This should be enough for a stack trace.
- */
- sleeping_thread_to_gdb_regs(gdb_regs, thread);
- }
- kgdb_mem2hex((char *)gdb_regs, remcom_out_buffer, NUMREGBYTES);
-}
-
-/* Handle the 'G' set registers request */
-static void gdb_cmd_setregs(struct kgdb_state *ks)
-{
- kgdb_hex2mem(&remcom_in_buffer[1], (char *)gdb_regs, NUMREGBYTES);
-
- if (kgdb_usethread && kgdb_usethread != current) {
- error_packet(remcom_out_buffer, -EINVAL);
- } else {
- gdb_regs_to_pt_regs(gdb_regs, ks->linux_regs);
- strcpy(remcom_out_buffer, "OK");
- }
-}
-
-/* Handle the 'm' memory read bytes */
-static void gdb_cmd_memread(struct kgdb_state *ks)
-{
- char *ptr = &remcom_in_buffer[1];
- unsigned long length;
- unsigned long addr;
- int err;
-
- if (kgdb_hex2long(&ptr, &addr) > 0 && *ptr++ == ',' &&
- kgdb_hex2long(&ptr, &length) > 0) {
- err = kgdb_mem2hex((char *)addr, remcom_out_buffer, length);
- if (err)
- error_packet(remcom_out_buffer, err);
- } else {
- error_packet(remcom_out_buffer, -EINVAL);
- }
-}
-
-/* Handle the 'M' memory write bytes */
-static void gdb_cmd_memwrite(struct kgdb_state *ks)
-{
- int err = write_mem_msg(0);
-
- if (err)
- error_packet(remcom_out_buffer, err);
- else
- strcpy(remcom_out_buffer, "OK");
-}
-
-/* Handle the 'X' memory binary write bytes */
-static void gdb_cmd_binwrite(struct kgdb_state *ks)
-{
- int err = write_mem_msg(1);
-
- if (err)
- error_packet(remcom_out_buffer, err);
- else
- strcpy(remcom_out_buffer, "OK");
-}
-
-/* Handle the 'D' or 'k', detach or kill packets */
-static void gdb_cmd_detachkill(struct kgdb_state *ks)
-{
- int error;
-
- /* The detach case */
- if (remcom_in_buffer[0] == 'D') {
- error = remove_all_break();
- if (error < 0) {
- error_packet(remcom_out_buffer, error);
- } else {
- strcpy(remcom_out_buffer, "OK");
- kgdb_connected = 0;
- }
- put_packet(remcom_out_buffer);
- } else {
- /*
- * Assume the kill case, with no exit code checking,
- * trying to force detach the debugger:
- */
- remove_all_break();
- kgdb_connected = 0;
- }
-}
-
-/* Handle the 'R' reboot packets */
-static int gdb_cmd_reboot(struct kgdb_state *ks)
-{
- /* For now, only honor R0 */
- if (strcmp(remcom_in_buffer, "R0") == 0) {
- printk(KERN_CRIT "Executing emergency reboot\n");
- strcpy(remcom_out_buffer, "OK");
- put_packet(remcom_out_buffer);
-
- /*
- * Execution should not return from
- * machine_emergency_restart()
- */
- machine_emergency_restart();
- kgdb_connected = 0;
-
- return 1;
- }
- return 0;
-}
-
-/* Handle the 'q' query packets */
-static void gdb_cmd_query(struct kgdb_state *ks)
-{
- struct task_struct *g;
- struct task_struct *p;
- unsigned char thref[8];
- char *ptr;
- int i;
- int cpu;
- int finished = 0;
-
- switch (remcom_in_buffer[1]) {
- case 's':
- case 'f':
- if (memcmp(remcom_in_buffer + 2, "ThreadInfo", 10)) {
- error_packet(remcom_out_buffer, -EINVAL);
- break;
- }
-
- i = 0;
- remcom_out_buffer[0] = 'm';
- ptr = remcom_out_buffer + 1;
- if (remcom_in_buffer[1] == 'f') {
- /* Each cpu is a shadow thread */
- for_each_online_cpu(cpu) {
- ks->thr_query = 0;
- int_to_threadref(thref, -cpu - 2);
- pack_threadid(ptr, thref);
- ptr += BUF_THREAD_ID_SIZE;
- *(ptr++) = ',';
- i++;
- }
- }
-
- do_each_thread(g, p) {
- if (i >= ks->thr_query && !finished) {
- int_to_threadref(thref, p->pid);
- pack_threadid(ptr, thref);
- ptr += BUF_THREAD_ID_SIZE;
- *(ptr++) = ',';
- ks->thr_query++;
- if (ks->thr_query % KGDB_MAX_THREAD_QUERY == 0)
- finished = 1;
- }
- i++;
- } while_each_thread(g, p);
-
- *(--ptr) = '\0';
- break;
-
- case 'C':
- /* Current thread id */
- strcpy(remcom_out_buffer, "QC");
- ks->threadid = shadow_pid(current->pid);
- int_to_threadref(thref, ks->threadid);
- pack_threadid(remcom_out_buffer + 2, thref);
- break;
- case 'T':
- if (memcmp(remcom_in_buffer + 1, "ThreadExtraInfo,", 16)) {
- error_packet(remcom_out_buffer, -EINVAL);
- break;
- }
- ks->threadid = 0;
- ptr = remcom_in_buffer + 17;
- kgdb_hex2long(&ptr, &ks->threadid);
- if (!getthread(ks->linux_regs, ks->threadid)) {
- error_packet(remcom_out_buffer, -EINVAL);
- break;
- }
- if ((int)ks->threadid > 0) {
- kgdb_mem2hex(getthread(ks->linux_regs,
- ks->threadid)->comm,
- remcom_out_buffer, 16);
- } else {
- static char tmpstr[23 + BUF_THREAD_ID_SIZE];
-
- sprintf(tmpstr, "shadowCPU%d",
- (int)(-ks->threadid - 2));
- kgdb_mem2hex(tmpstr, remcom_out_buffer, strlen(tmpstr));
- }
- break;
- }
-}
-
-/* Handle the 'H' task query packets */
-static void gdb_cmd_task(struct kgdb_state *ks)
-{
- struct task_struct *thread;
- char *ptr;
-
- switch (remcom_in_buffer[1]) {
- case 'g':
- ptr = &remcom_in_buffer[2];
- kgdb_hex2long(&ptr, &ks->threadid);
- thread = getthread(ks->linux_regs, ks->threadid);
- if (!thread && ks->threadid > 0) {
- error_packet(remcom_out_buffer, -EINVAL);
- break;
- }
- kgdb_usethread = thread;
- ks->kgdb_usethreadid = ks->threadid;
- strcpy(remcom_out_buffer, "OK");
- break;
- case 'c':
- ptr = &remcom_in_buffer[2];
- kgdb_hex2long(&ptr, &ks->threadid);
- if (!ks->threadid) {
- kgdb_contthread = NULL;
- } else {
- thread = getthread(ks->linux_regs, ks->threadid);
- if (!thread && ks->threadid > 0) {
- error_packet(remcom_out_buffer, -EINVAL);
- break;
- }
- kgdb_contthread = thread;
- }
- strcpy(remcom_out_buffer, "OK");
- break;
- }
-}
-
-/* Handle the 'T' thread query packets */
-static void gdb_cmd_thread(struct kgdb_state *ks)
-{
- char *ptr = &remcom_in_buffer[1];
- struct task_struct *thread;
-
- kgdb_hex2long(&ptr, &ks->threadid);
- thread = getthread(ks->linux_regs, ks->threadid);
- if (thread)
- strcpy(remcom_out_buffer, "OK");
- else
- error_packet(remcom_out_buffer, -EINVAL);
-}
-
-/* Handle the 'z' or 'Z' breakpoint remove or set packets */
-static void gdb_cmd_break(struct kgdb_state *ks)
-{
- /*
- * Since GDB-5.3, it's been drafted that '0' is a software
- * breakpoint, '1' is a hardware breakpoint, so let's do that.
- */
- char *bpt_type = &remcom_in_buffer[1];
- char *ptr = &remcom_in_buffer[2];
- unsigned long addr;
- unsigned long length;
- int error = 0;
-
- if (arch_kgdb_ops.set_hw_breakpoint && *bpt_type >= '1') {
- /* Unsupported */
- if (*bpt_type > '4')
- return;
- } else {
- if (*bpt_type != '0' && *bpt_type != '1')
- /* Unsupported. */
- return;
- }
-
- /*
- * Test if this is a hardware breakpoint, and
- * if we support it:
- */
- if (*bpt_type == '1' && !(arch_kgdb_ops.flags & KGDB_HW_BREAKPOINT))
- /* Unsupported. */
- return;
-
- if (*(ptr++) != ',') {
- error_packet(remcom_out_buffer, -EINVAL);
- return;
- }
- if (!kgdb_hex2long(&ptr, &addr)) {
- error_packet(remcom_out_buffer, -EINVAL);
- return;
- }
- if (*(ptr++) != ',' ||
- !kgdb_hex2long(&ptr, &length)) {
- error_packet(remcom_out_buffer, -EINVAL);
- return;
- }
-
- if (remcom_in_buffer[0] == 'Z' && *bpt_type == '0')
- error = kgdb_set_sw_break(addr);
- else if (remcom_in_buffer[0] == 'z' && *bpt_type == '0')
- error = kgdb_remove_sw_break(addr);
- else if (remcom_in_buffer[0] == 'Z')
- error = arch_kgdb_ops.set_hw_breakpoint(addr,
- (int)length, *bpt_type - '0');
- else if (remcom_in_buffer[0] == 'z')
- error = arch_kgdb_ops.remove_hw_breakpoint(addr,
- (int) length, *bpt_type - '0');
-
- if (error == 0)
- strcpy(remcom_out_buffer, "OK");
- else
- error_packet(remcom_out_buffer, error);
-}
-
-/* Handle the 'C' signal / exception passing packets */
-static int gdb_cmd_exception_pass(struct kgdb_state *ks)
-{
- /* C09 == pass exception
- * C15 == detach kgdb, pass exception
- */
- if (remcom_in_buffer[1] == '0' && remcom_in_buffer[2] == '9') {
-
- ks->pass_exception = 1;
- remcom_in_buffer[0] = 'c';
-
- } else if (remcom_in_buffer[1] == '1' && remcom_in_buffer[2] == '5') {
-
- ks->pass_exception = 1;
- remcom_in_buffer[0] = 'D';
- remove_all_break();
- kgdb_connected = 0;
- return 1;
-
- } else {
- kgdb_msg_write("KGDB only knows signal 9 (pass)"
- " and 15 (pass and disconnect)\n"
- "Executing a continue without signal passing\n", 0);
- remcom_in_buffer[0] = 'c';
- }
-
- /* Indicate fall through */
- return -1;
-}
-
-/*
- * This function performs all gdbserial command procesing
- */
-static int gdb_serial_stub(struct kgdb_state *ks)
-{
- int error = 0;
- int tmp;
-
- /* Clear the out buffer. */
- memset(remcom_out_buffer, 0, sizeof(remcom_out_buffer));
-
- if (kgdb_connected) {
- unsigned char thref[8];
- char *ptr;
-
- /* Reply to host that an exception has occurred */
- ptr = remcom_out_buffer;
- *ptr++ = 'T';
- ptr = pack_hex_byte(ptr, ks->signo);
- ptr += strlen(strcpy(ptr, "thread:"));
- int_to_threadref(thref, shadow_pid(current->pid));
- ptr = pack_threadid(ptr, thref);
- *ptr++ = ';';
- put_packet(remcom_out_buffer);
- }
-
- kgdb_usethread = kgdb_info[ks->cpu].task;
- ks->kgdb_usethreadid = shadow_pid(kgdb_info[ks->cpu].task->pid);
- ks->pass_exception = 0;
-
- while (1) {
- error = 0;
-
- /* Clear the out buffer. */
- memset(remcom_out_buffer, 0, sizeof(remcom_out_buffer));
-
- get_packet(remcom_in_buffer);
-
- switch (remcom_in_buffer[0]) {
- case '?': /* gdbserial status */
- gdb_cmd_status(ks);
- break;
- case 'g': /* return the value of the CPU registers */
- gdb_cmd_getregs(ks);
- break;
- case 'G': /* set the value of the CPU registers - return OK */
- gdb_cmd_setregs(ks);
- break;
- case 'm': /* mAA..AA,LLLL Read LLLL bytes at address AA..AA */
- gdb_cmd_memread(ks);
- break;
- case 'M': /* MAA..AA,LLLL: Write LLLL bytes at address AA..AA */
- gdb_cmd_memwrite(ks);
- break;
- case 'X': /* XAA..AA,LLLL: Write LLLL bytes at address AA..AA */
- gdb_cmd_binwrite(ks);
- break;
- /* kill or detach. KGDB should treat this like a
- * continue.
- */
- case 'D': /* Debugger detach */
- case 'k': /* Debugger detach via kill */
- gdb_cmd_detachkill(ks);
- goto default_handle;
- case 'R': /* Reboot */
- if (gdb_cmd_reboot(ks))
- goto default_handle;
- break;
- case 'q': /* query command */
- gdb_cmd_query(ks);
- break;
- case 'H': /* task related */
- gdb_cmd_task(ks);
- break;
- case 'T': /* Query thread status */
- gdb_cmd_thread(ks);
- break;
- case 'z': /* Break point remove */
- case 'Z': /* Break point set */
- gdb_cmd_break(ks);
- break;
- case 'C': /* Exception passing */
- tmp = gdb_cmd_exception_pass(ks);
- if (tmp > 0)
- goto default_handle;
- if (tmp == 0)
- break;
- /* Fall through on tmp < 0 */
- case 'c': /* Continue packet */
- case 's': /* Single step packet */
- if (kgdb_contthread && kgdb_contthread != current) {
- /* Can't switch threads in kgdb */
- error_packet(remcom_out_buffer, -EINVAL);
- break;
- }
- kgdb_activate_sw_breakpoints();
- /* Fall through to default processing */
- default:
-default_handle:
- error = kgdb_arch_handle_exception(ks->ex_vector,
- ks->signo,
- ks->err_code,
- remcom_in_buffer,
- remcom_out_buffer,
- ks->linux_regs);
- /*
- * Leave cmd processing on error, detach,
- * kill, continue, or single step.
- */
- if (error >= 0 || remcom_in_buffer[0] == 'D' ||
- remcom_in_buffer[0] == 'k') {
- error = 0;
- goto kgdb_exit;
- }
-
- }
-
- /* reply to the request */
- put_packet(remcom_out_buffer);
- }
-
-kgdb_exit:
- if (ks->pass_exception)
- error = 1;
- return error;
-}
-
static int kgdb_reenter_check(struct kgdb_state *ks)
{
unsigned long addr;
@@ -1375,17 +461,17 @@ static int kgdb_reenter_check(struct kgdb_state *ks)
* user because the user planted a breakpoint in a place that
* KGDB needs in order to function.
*/
- if (kgdb_remove_sw_break(addr) == 0) {
+ if (dbg_remove_sw_break(addr) == 0) {
exception_level = 0;
kgdb_skipexception(ks->ex_vector, ks->linux_regs);
- kgdb_activate_sw_breakpoints();
+ dbg_activate_sw_breakpoints();
printk(KERN_CRIT "KGDB: re-enter error: breakpoint removed %lx\n",
addr);
WARN_ON_ONCE(1);
return 1;
}
- remove_all_break();
+ dbg_remove_all_break();
kgdb_skipexception(ks->ex_vector, ks->linux_regs);
if (exception_level > 1) {
@@ -1462,7 +548,7 @@ acquirelock:
if (!kgdb_io_ready(1)) {
error = 1;
- goto kgdb_restore; /* No I/O connection, so resume the system */
+ goto kgdb_restore; /* No I/O connection, resume the system */
}
/*
@@ -1472,8 +558,8 @@ acquirelock:
goto kgdb_restore;
/* Call the I/O driver's pre_exception routine */
- if (kgdb_io_ops->pre_exception)
- kgdb_io_ops->pre_exception();
+ if (dbg_io_ops->pre_exception)
+ dbg_io_ops->pre_exception();
kgdb_info[ks->cpu].debuggerinfo = ks->linux_regs;
kgdb_info[ks->cpu].task = current;
@@ -1523,8 +609,8 @@ acquirelock:
error = gdb_serial_stub(ks);
/* Call the I/O driver's post_exception routine */
- if (kgdb_io_ops->post_exception)
- kgdb_io_ops->post_exception();
+ if (dbg_io_ops->post_exception)
+ dbg_io_ops->post_exception();
kgdb_info[ks->cpu].debuggerinfo = NULL;
kgdb_info[ks->cpu].task = NULL;
@@ -1584,7 +670,7 @@ static void kgdb_console_write(struct console *co, const char *s,
return;
local_irq_save(flags);
- kgdb_msg_write(s, count);
+ gdbstub_msg_write(s, count);
local_irq_restore(flags);
}
@@ -1596,9 +682,9 @@ static struct console kgdbcons = {
};
#ifdef CONFIG_MAGIC_SYSRQ
-static void sysrq_handle_gdb(int key, struct tty_struct *tty)
+static void sysrq_handle_dbg(int key, struct tty_struct *tty)
{
- if (!kgdb_io_ops) {
+ if (!dbg_io_ops) {
printk(KERN_CRIT "ERROR: No KGDB I/O module available\n");
return;
}
@@ -1608,8 +694,8 @@ static void sysrq_handle_gdb(int key, struct tty_struct *tty)
kgdb_breakpoint();
}
-static struct sysrq_key_op sysrq_gdb_op = {
- .handler = sysrq_handle_gdb,
+static struct sysrq_key_op sysrq_dbg_op = {
+ .handler = sysrq_handle_dbg,
.help_msg = "debug(G)",
.action_msg = "DEBUG",
};
@@ -1621,7 +707,7 @@ static void kgdb_register_callbacks(void)
kgdb_io_module_registered = 1;
kgdb_arch_init();
#ifdef CONFIG_MAGIC_SYSRQ
- register_sysrq_key('g', &sysrq_gdb_op);
+ register_sysrq_key('g', &sysrq_dbg_op);
#endif
if (kgdb_use_con && !kgdb_con_registered) {
register_console(&kgdbcons);
@@ -1641,7 +727,7 @@ static void kgdb_unregister_callbacks(void)
kgdb_io_module_registered = 0;
kgdb_arch_exit();
#ifdef CONFIG_MAGIC_SYSRQ
- unregister_sysrq_key('g', &sysrq_gdb_op);
+ unregister_sysrq_key('g', &sysrq_dbg_op);
#endif
if (kgdb_con_registered) {
unregister_console(&kgdbcons);
@@ -1660,17 +746,17 @@ static void kgdb_initial_breakpoint(void)
/**
* kgdb_register_io_module - register KGDB IO module
- * @new_kgdb_io_ops: the io ops vector
+ * @new_dbg_io_ops: the io ops vector
*
* Register it with the KGDB core.
*/
-int kgdb_register_io_module(struct kgdb_io *new_kgdb_io_ops)
+int kgdb_register_io_module(struct kgdb_io *new_dbg_io_ops)
{
int err;
spin_lock(&kgdb_registration_lock);
- if (kgdb_io_ops) {
+ if (dbg_io_ops) {
spin_unlock(&kgdb_registration_lock);
printk(KERN_ERR "kgdb: Another I/O driver is already "
@@ -1678,20 +764,20 @@ int kgdb_register_io_module(struct kgdb_io *new_kgdb_io_ops)
return -EBUSY;
}
- if (new_kgdb_io_ops->init) {
- err = new_kgdb_io_ops->init();
+ if (new_dbg_io_ops->init) {
+ err = new_dbg_io_ops->init();
if (err) {
spin_unlock(&kgdb_registration_lock);
return err;
}
}
- kgdb_io_ops = new_kgdb_io_ops;
+ dbg_io_ops = new_dbg_io_ops;
spin_unlock(&kgdb_registration_lock);
printk(KERN_INFO "kgdb: Registered I/O driver %s.\n",
- new_kgdb_io_ops->name);
+ new_dbg_io_ops->name);
/* Arm KGDB now. */
kgdb_register_callbacks();
@@ -1705,11 +791,11 @@ EXPORT_SYMBOL_GPL(kgdb_register_io_module);
/**
* kkgdb_unregister_io_module - unregister KGDB IO module
- * @old_kgdb_io_ops: the io ops vector
+ * @old_dbg_io_ops: the io ops vector
*
* Unregister it with the KGDB core.
*/
-void kgdb_unregister_io_module(struct kgdb_io *old_kgdb_io_ops)
+void kgdb_unregister_io_module(struct kgdb_io *old_dbg_io_ops)
{
BUG_ON(kgdb_connected);
@@ -1721,14 +807,14 @@ void kgdb_unregister_io_module(struct kgdb_io *old_kgdb_io_ops)
spin_lock(&kgdb_registration_lock);
- WARN_ON_ONCE(kgdb_io_ops != old_kgdb_io_ops);
- kgdb_io_ops = NULL;
+ WARN_ON_ONCE(dbg_io_ops != old_dbg_io_ops);
+ dbg_io_ops = NULL;
spin_unlock(&kgdb_registration_lock);
printk(KERN_INFO
"kgdb: Unregistered I/O driver %s, debugger disabled.\n",
- old_kgdb_io_ops->name);
+ old_dbg_io_ops->name);
}
EXPORT_SYMBOL_GPL(kgdb_unregister_io_module);
diff --git a/kernel/debug/debug_core.h b/kernel/debug/debug_core.h
new file mode 100644
index 0000000..4b9d512
--- /dev/null
+++ b/kernel/debug/debug_core.h
@@ -0,0 +1,48 @@
+/*
+ * Created by: Jason Wessel <[email protected]>
+ *
+ * Copyright (c) 2009 Wind River Systems, Inc. All Rights Reserved.
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ */
+
+#ifndef _DEBUG_CORE_H_
+#define _DEBUG_CORE_H_
+/*
+ * These are the private implementation headers between the kernel
+ * debugger core and the debugger front end code.
+ */
+
+/* kernel debug core data structures */
+struct kgdb_state {
+ int ex_vector;
+ int signo;
+ int err_code;
+ int cpu;
+ int pass_exception;
+ unsigned long thr_query;
+ unsigned long threadid;
+ long kgdb_usethreadid;
+ struct pt_regs *linux_regs;
+};
+
+struct debuggerinfo_struct {
+ void *debuggerinfo;
+ struct task_struct *task;
+};
+
+extern struct debuggerinfo_struct kgdb_info[];
+
+/* kernel debug core break point routines */
+extern int dbg_remove_all_break(void);
+extern int dbg_set_sw_break(unsigned long addr);
+extern int dbg_remove_sw_break(unsigned long addr);
+extern int dbg_activate_sw_breakpoints(void);
+
+/* gdbstub interface functions */
+extern int gdb_serial_stub(struct kgdb_state *ks);
+extern void gdbstub_msg_write(const char *s, int len);
+
+#endif /* _DEBUG_CORE_H_ */
diff --git a/kernel/debug/gdbstub.c b/kernel/debug/gdbstub.c
new file mode 100644
index 0000000..ab29bce
--- /dev/null
+++ b/kernel/debug/gdbstub.c
@@ -0,0 +1,939 @@
+/*
+ * Kernel Debug Core
+ *
+ * Maintainer: Jason Wessel <[email protected]>
+ *
+ * Copyright (C) 2000-2001 VERITAS Software Corporation.
+ * Copyright (C) 2002-2004 Timesys Corporation
+ * Copyright (C) 2003-2004 Amit S. Kale <[email protected]>
+ * Copyright (C) 2004 Pavel Machek <[email protected]>
+ * Copyright (C) 2004-2006 Tom Rini <[email protected]>
+ * Copyright (C) 2004-2006 LinSysSoft Technologies Pvt. Ltd.
+ * Copyright (C) 2005-2009 Wind River Systems, Inc.
+ * Copyright (C) 2007 MontaVista Software, Inc.
+ * Copyright (C) 2008 Red Hat, Inc., Ingo Molnar <[email protected]>
+ *
+ * Contributors at various stages not listed above:
+ * Jason Wessel ( [email protected] )
+ * George Anzinger <[email protected]>
+ * Anurekh Saxena ([email protected])
+ * Lake Stevens Instrument Division (Glenn Engel)
+ * Jim Kingdon, Cygnus Support.
+ *
+ * Original KGDB stub: David Grothe <[email protected]>,
+ * Tigran Aivazian <[email protected]>
+ *
+ * This file is licensed under the terms of the GNU General Public License
+ * version 2. This program is licensed "as is" without any warranty of any
+ * kind, whether express or implied.
+ */
+
+#include <linux/kernel.h>
+#include <linux/kgdb.h>
+#include <linux/reboot.h>
+#include <linux/uaccess.h>
+#include <asm/cacheflush.h>
+#include <asm/unaligned.h>
+#include "debug_core.h"
+
+#define KGDB_MAX_THREAD_QUERY 17
+
+/* Our I/O buffers. */
+static char remcom_in_buffer[BUFMAX];
+static char remcom_out_buffer[BUFMAX];
+
+/* Storage for the registers, in GDB format. */
+static unsigned long gdb_regs[(NUMREGBYTES +
+ sizeof(unsigned long) - 1) /
+ sizeof(unsigned long)];
+
+/*
+ * GDB remote protocol parser:
+ */
+
+static int hex(char ch)
+{
+ if ((ch >= 'a') && (ch <= 'f'))
+ return ch - 'a' + 10;
+ if ((ch >= '0') && (ch <= '9'))
+ return ch - '0';
+ if ((ch >= 'A') && (ch <= 'F'))
+ return ch - 'A' + 10;
+ return -1;
+}
+
+/* scan for the sequence $<data>#<checksum> */
+static void get_packet(char *buffer)
+{
+ unsigned char checksum;
+ unsigned char xmitcsum;
+ int count;
+ char ch;
+
+ do {
+ /*
+ * Spin and wait around for the start character, ignore all
+ * other characters:
+ */
+ while ((ch = (dbg_io_ops->read_char())) != '$')
+ /* nothing */;
+
+ kgdb_connected = 1;
+ checksum = 0;
+ xmitcsum = -1;
+
+ count = 0;
+
+ /*
+ * now, read until a # or end of buffer is found:
+ */
+ while (count < (BUFMAX - 1)) {
+ ch = dbg_io_ops->read_char();
+ if (ch == '#')
+ break;
+ checksum = checksum + ch;
+ buffer[count] = ch;
+ count = count + 1;
+ }
+ buffer[count] = 0;
+
+ if (ch == '#') {
+ xmitcsum = hex(dbg_io_ops->read_char()) << 4;
+ xmitcsum += hex(dbg_io_ops->read_char());
+
+ if (checksum != xmitcsum)
+ /* failed checksum */
+ dbg_io_ops->write_char('-');
+ else
+ /* successful transfer */
+ dbg_io_ops->write_char('+');
+ if (dbg_io_ops->flush)
+ dbg_io_ops->flush();
+ }
+ } while (checksum != xmitcsum);
+}
+
+/*
+ * Send the packet in buffer.
+ * Check for gdb connection if asked for.
+ */
+static void put_packet(char *buffer)
+{
+ unsigned char checksum;
+ int count;
+ char ch;
+
+ /*
+ * $<packet info>#<checksum>.
+ */
+ while (1) {
+ dbg_io_ops->write_char('$');
+ checksum = 0;
+ count = 0;
+
+ while ((ch = buffer[count])) {
+ dbg_io_ops->write_char(ch);
+ checksum += ch;
+ count++;
+ }
+
+ dbg_io_ops->write_char('#');
+ dbg_io_ops->write_char(hex_asc_hi(checksum));
+ dbg_io_ops->write_char(hex_asc_lo(checksum));
+ if (dbg_io_ops->flush)
+ dbg_io_ops->flush();
+
+ /* Now see what we get in reply. */
+ ch = dbg_io_ops->read_char();
+
+ if (ch == 3)
+ ch = dbg_io_ops->read_char();
+
+ /* If we get an ACK, we are done. */
+ if (ch == '+')
+ return;
+
+ /*
+ * If we get the start of another packet, this means
+ * that GDB is attempting to reconnect. We will NAK
+ * the packet being sent, and stop trying to send this
+ * packet.
+ */
+ if (ch == '$') {
+ dbg_io_ops->write_char('-');
+ if (dbg_io_ops->flush)
+ dbg_io_ops->flush();
+ return;
+ }
+ }
+}
+
+static char gdbmsgbuf[BUFMAX + 1];
+
+void gdbstub_msg_write(const char *s, int len)
+{
+ char *bufptr;
+ int wcount;
+ int i;
+
+ /* 'O'utput */
+ gdbmsgbuf[0] = 'O';
+
+ /* Fill and send buffers... */
+ while (len > 0) {
+ bufptr = gdbmsgbuf + 1;
+
+ /* Calculate how many this time */
+ if ((len << 1) > (BUFMAX - 2))
+ wcount = (BUFMAX - 2) >> 1;
+ else
+ wcount = len;
+
+ /* Pack in hex chars */
+ for (i = 0; i < wcount; i++)
+ bufptr = pack_hex_byte(bufptr, s[i]);
+ *bufptr = '\0';
+
+ /* Move up */
+ s += wcount;
+ len -= wcount;
+
+ /* Write packet */
+ put_packet(gdbmsgbuf);
+ }
+}
+
+/*
+ * Convert the memory pointed to by mem into hex, placing result in
+ * buf. Return a pointer to the last char put in buf (null). May
+ * return an error.
+ */
+int kgdb_mem2hex(char *mem, char *buf, int count)
+{
+ char *tmp;
+ int err;
+
+ /*
+ * We use the upper half of buf as an intermediate buffer for the
+ * raw memory copy. Hex conversion will work against this one.
+ */
+ tmp = buf + count;
+
+ err = probe_kernel_read(tmp, mem, count);
+ if (!err) {
+ while (count > 0) {
+ buf = pack_hex_byte(buf, *tmp);
+ tmp++;
+ count--;
+ }
+
+ *buf = 0;
+ }
+
+ return err;
+}
+
+/*
+ * Convert the hex array pointed to by buf into binary to be placed in
+ * mem. Return a pointer to the character AFTER the last byte
+ * written. May return an error.
+ */
+int kgdb_hex2mem(char *buf, char *mem, int count)
+{
+ char *tmp_raw;
+ char *tmp_hex;
+
+ /*
+ * We use the upper half of buf as an intermediate buffer for the
+ * raw memory that is converted from hex.
+ */
+ tmp_raw = buf + count * 2;
+
+ tmp_hex = tmp_raw - 1;
+ while (tmp_hex >= buf) {
+ tmp_raw--;
+ *tmp_raw = hex(*tmp_hex--);
+ *tmp_raw |= hex(*tmp_hex--) << 4;
+ }
+
+ return probe_kernel_write(mem, tmp_raw, count);
+}
+
+/*
+ * While we find nice hex chars, build a long_val.
+ * Return number of chars processed.
+ */
+int kgdb_hex2long(char **ptr, unsigned long *long_val)
+{
+ int hex_val;
+ int num = 0;
+ int negate = 0;
+
+ *long_val = 0;
+
+ if (**ptr == '-') {
+ negate = 1;
+ (*ptr)++;
+ }
+ while (**ptr) {
+ hex_val = hex(**ptr);
+ if (hex_val < 0)
+ break;
+
+ *long_val = (*long_val << 4) | hex_val;
+ num++;
+ (*ptr)++;
+ }
+
+ if (negate)
+ *long_val = -*long_val;
+
+ return num;
+}
+
+/*
+ * Copy the binary array pointed to by buf into mem. Fix $, #, and
+ * 0x7d escaped with 0x7d. Return a pointer to the character after
+ * the last byte written.
+ */
+static int kgdb_ebin2mem(char *buf, char *mem, int count)
+{
+ int err = 0;
+ char c;
+
+ while (count-- > 0) {
+ c = *buf++;
+ if (c == 0x7d)
+ c = *buf++ ^ 0x20;
+
+ err = probe_kernel_write(mem, &c, 1);
+ if (err)
+ break;
+
+ mem++;
+ }
+
+ return err;
+}
+
+/* Write memory due to an 'M' or 'X' packet. */
+static int write_mem_msg(int binary)
+{
+ char *ptr = &remcom_in_buffer[1];
+ unsigned long addr;
+ unsigned long length;
+ int err;
+
+ if (kgdb_hex2long(&ptr, &addr) > 0 && *(ptr++) == ',' &&
+ kgdb_hex2long(&ptr, &length) > 0 && *(ptr++) == ':') {
+ if (binary)
+ err = kgdb_ebin2mem(ptr, (char *)addr, length);
+ else
+ err = kgdb_hex2mem(ptr, (char *)addr, length);
+ if (err)
+ return err;
+ if (CACHE_FLUSH_IS_SAFE)
+ flush_icache_range(addr, addr + length);
+ return 0;
+ }
+
+ return -EINVAL;
+}
+
+static void error_packet(char *pkt, int error)
+{
+ error = -error;
+ pkt[0] = 'E';
+ pkt[1] = hex_asc[(error / 10)];
+ pkt[2] = hex_asc[(error % 10)];
+ pkt[3] = '\0';
+}
+
+/*
+ * Thread ID accessors. We represent a flat TID space to GDB, where
+ * the per CPU idle threads (which under Linux all have PID 0) are
+ * remapped to negative TIDs.
+ */
+
+#define BUF_THREAD_ID_SIZE 16
+
+static char *pack_threadid(char *pkt, unsigned char *id)
+{
+ char *limit;
+
+ limit = pkt + BUF_THREAD_ID_SIZE;
+ while (pkt < limit)
+ pkt = pack_hex_byte(pkt, *id++);
+
+ return pkt;
+}
+
+static void int_to_threadref(unsigned char *id, int value)
+{
+ unsigned char *scan;
+ int i = 4;
+
+ scan = (unsigned char *)id;
+ while (i--)
+ *scan++ = 0;
+ put_unaligned_be32(value, scan);
+}
+
+static struct task_struct *getthread(struct pt_regs *regs, int tid)
+{
+ /*
+ * Non-positive TIDs are remapped to the cpu shadow information
+ */
+ if (tid == 0 || tid == -1)
+ tid = -atomic_read(&kgdb_active) - 2;
+ if (tid < -1 && tid > -NR_CPUS - 2) {
+ if (kgdb_info[-tid - 2].task)
+ return kgdb_info[-tid - 2].task;
+ else
+ return idle_task(-tid - 2);
+ }
+ if (tid <= 0) {
+ printk(KERN_ERR "KGDB: Internal thread select error\n");
+ dump_stack();
+ return NULL;
+ }
+
+ /*
+ * find_task_by_pid_ns() does not take the tasklist lock anymore
+ * but is nicely RCU locked - hence is a pretty resilient
+ * thing to use:
+ */
+ return find_task_by_pid_ns(tid, &init_pid_ns);
+}
+
+
+/*
+ * Remap normal tasks to their real PID,
+ * CPU shadow threads are mapped to -CPU - 2
+ */
+static inline int shadow_pid(int realpid)
+{
+ if (realpid)
+ return realpid;
+
+ return -raw_smp_processor_id() - 2;
+}
+
+/*
+ * All the functions that start with gdb_cmd are the various
+ * operations to implement the handlers for the gdbserial protocol
+ * where KGDB is communicating with an external debugger
+ */
+
+/* Handle the '?' status packets */
+static void gdb_cmd_status(struct kgdb_state *ks)
+{
+ /*
+ * We know that this packet is only sent
+ * during initial connect. So to be safe,
+ * we clear out our breakpoints now in case
+ * GDB is reconnecting.
+ */
+ dbg_remove_all_break();
+
+ remcom_out_buffer[0] = 'S';
+ pack_hex_byte(&remcom_out_buffer[1], ks->signo);
+}
+
+/* Handle the 'g' get registers request */
+static void gdb_cmd_getregs(struct kgdb_state *ks)
+{
+ struct task_struct *thread;
+ void *local_debuggerinfo;
+ int i;
+
+ thread = kgdb_usethread;
+ if (!thread) {
+ thread = kgdb_info[ks->cpu].task;
+ local_debuggerinfo = kgdb_info[ks->cpu].debuggerinfo;
+ } else {
+ local_debuggerinfo = NULL;
+ for_each_online_cpu(i) {
+ /*
+ * Try to find the task on some other
+ * or possibly this node if we do not
+ * find the matching task then we try
+ * to approximate the results.
+ */
+ if (thread == kgdb_info[i].task)
+ local_debuggerinfo = kgdb_info[i].debuggerinfo;
+ }
+ }
+
+ /*
+ * All threads that don't have debuggerinfo should be
+ * in schedule() sleeping, since all other CPUs
+ * are in kgdb_wait, and thus have debuggerinfo.
+ */
+ if (local_debuggerinfo) {
+ pt_regs_to_gdb_regs(gdb_regs, local_debuggerinfo);
+ } else {
+ /*
+ * Pull stuff saved during switch_to; nothing
+ * else is accessible (or even particularly
+ * relevant).
+ *
+ * This should be enough for a stack trace.
+ */
+ sleeping_thread_to_gdb_regs(gdb_regs, thread);
+ }
+ kgdb_mem2hex((char *)gdb_regs, remcom_out_buffer, NUMREGBYTES);
+}
+
+/* Handle the 'G' set registers request */
+static void gdb_cmd_setregs(struct kgdb_state *ks)
+{
+ kgdb_hex2mem(&remcom_in_buffer[1], (char *)gdb_regs, NUMREGBYTES);
+
+ if (kgdb_usethread && kgdb_usethread != current) {
+ error_packet(remcom_out_buffer, -EINVAL);
+ } else {
+ gdb_regs_to_pt_regs(gdb_regs, ks->linux_regs);
+ strcpy(remcom_out_buffer, "OK");
+ }
+}
+
+/* Handle the 'm' memory read bytes */
+static void gdb_cmd_memread(struct kgdb_state *ks)
+{
+ char *ptr = &remcom_in_buffer[1];
+ unsigned long length;
+ unsigned long addr;
+ int err;
+
+ if (kgdb_hex2long(&ptr, &addr) > 0 && *ptr++ == ',' &&
+ kgdb_hex2long(&ptr, &length) > 0) {
+ err = kgdb_mem2hex((char *)addr, remcom_out_buffer, length);
+ if (err)
+ error_packet(remcom_out_buffer, err);
+ } else {
+ error_packet(remcom_out_buffer, -EINVAL);
+ }
+}
+
+/* Handle the 'M' memory write bytes */
+static void gdb_cmd_memwrite(struct kgdb_state *ks)
+{
+ int err = write_mem_msg(0);
+
+ if (err)
+ error_packet(remcom_out_buffer, err);
+ else
+ strcpy(remcom_out_buffer, "OK");
+}
+
+/* Handle the 'X' memory binary write bytes */
+static void gdb_cmd_binwrite(struct kgdb_state *ks)
+{
+ int err = write_mem_msg(1);
+
+ if (err)
+ error_packet(remcom_out_buffer, err);
+ else
+ strcpy(remcom_out_buffer, "OK");
+}
+
+/* Handle the 'D' or 'k', detach or kill packets */
+static void gdb_cmd_detachkill(struct kgdb_state *ks)
+{
+ int error;
+
+ /* The detach case */
+ if (remcom_in_buffer[0] == 'D') {
+ error = dbg_remove_all_break();
+ if (error < 0) {
+ error_packet(remcom_out_buffer, error);
+ } else {
+ strcpy(remcom_out_buffer, "OK");
+ kgdb_connected = 0;
+ }
+ put_packet(remcom_out_buffer);
+ } else {
+ /*
+ * Assume the kill case, with no exit code checking,
+ * trying to force detach the debugger:
+ */
+ dbg_remove_all_break();
+ kgdb_connected = 0;
+ }
+}
+
+/* Handle the 'R' reboot packets */
+static int gdb_cmd_reboot(struct kgdb_state *ks)
+{
+ /* For now, only honor R0 */
+ if (strcmp(remcom_in_buffer, "R0") == 0) {
+ printk(KERN_CRIT "Executing emergency reboot\n");
+ strcpy(remcom_out_buffer, "OK");
+ put_packet(remcom_out_buffer);
+
+ /*
+ * Execution should not return from
+ * machine_emergency_restart()
+ */
+ machine_emergency_restart();
+ kgdb_connected = 0;
+
+ return 1;
+ }
+ return 0;
+}
+
+/* Handle the 'q' query packets */
+static void gdb_cmd_query(struct kgdb_state *ks)
+{
+ struct task_struct *g;
+ struct task_struct *p;
+ unsigned char thref[8];
+ char *ptr;
+ int i;
+ int cpu;
+ int finished = 0;
+
+ switch (remcom_in_buffer[1]) {
+ case 's':
+ case 'f':
+ if (memcmp(remcom_in_buffer + 2, "ThreadInfo", 10)) {
+ error_packet(remcom_out_buffer, -EINVAL);
+ break;
+ }
+
+ i = 0;
+ remcom_out_buffer[0] = 'm';
+ ptr = remcom_out_buffer + 1;
+ if (remcom_in_buffer[1] == 'f') {
+ /* Each cpu is a shadow thread */
+ for_each_online_cpu(cpu) {
+ ks->thr_query = 0;
+ int_to_threadref(thref, -cpu - 2);
+ pack_threadid(ptr, thref);
+ ptr += BUF_THREAD_ID_SIZE;
+ *(ptr++) = ',';
+ i++;
+ }
+ }
+
+ do_each_thread(g, p) {
+ if (i >= ks->thr_query && !finished) {
+ int_to_threadref(thref, p->pid);
+ pack_threadid(ptr, thref);
+ ptr += BUF_THREAD_ID_SIZE;
+ *(ptr++) = ',';
+ ks->thr_query++;
+ if (ks->thr_query % KGDB_MAX_THREAD_QUERY == 0)
+ finished = 1;
+ }
+ i++;
+ } while_each_thread(g, p);
+
+ *(--ptr) = '\0';
+ break;
+
+ case 'C':
+ /* Current thread id */
+ strcpy(remcom_out_buffer, "QC");
+ ks->threadid = shadow_pid(current->pid);
+ int_to_threadref(thref, ks->threadid);
+ pack_threadid(remcom_out_buffer + 2, thref);
+ break;
+ case 'T':
+ if (memcmp(remcom_in_buffer + 1, "ThreadExtraInfo,", 16)) {
+ error_packet(remcom_out_buffer, -EINVAL);
+ break;
+ }
+ ks->threadid = 0;
+ ptr = remcom_in_buffer + 17;
+ kgdb_hex2long(&ptr, &ks->threadid);
+ if (!getthread(ks->linux_regs, ks->threadid)) {
+ error_packet(remcom_out_buffer, -EINVAL);
+ break;
+ }
+ if ((int)ks->threadid > 0) {
+ kgdb_mem2hex(getthread(ks->linux_regs,
+ ks->threadid)->comm,
+ remcom_out_buffer, 16);
+ } else {
+ static char tmpstr[23 + BUF_THREAD_ID_SIZE];
+
+ sprintf(tmpstr, "shadowCPU%d",
+ (int)(-ks->threadid - 2));
+ kgdb_mem2hex(tmpstr, remcom_out_buffer, strlen(tmpstr));
+ }
+ break;
+ }
+}
+
+/* Handle the 'H' task query packets */
+static void gdb_cmd_task(struct kgdb_state *ks)
+{
+ struct task_struct *thread;
+ char *ptr;
+
+ switch (remcom_in_buffer[1]) {
+ case 'g':
+ ptr = &remcom_in_buffer[2];
+ kgdb_hex2long(&ptr, &ks->threadid);
+ thread = getthread(ks->linux_regs, ks->threadid);
+ if (!thread && ks->threadid > 0) {
+ error_packet(remcom_out_buffer, -EINVAL);
+ break;
+ }
+ kgdb_usethread = thread;
+ ks->kgdb_usethreadid = ks->threadid;
+ strcpy(remcom_out_buffer, "OK");
+ break;
+ case 'c':
+ ptr = &remcom_in_buffer[2];
+ kgdb_hex2long(&ptr, &ks->threadid);
+ if (!ks->threadid) {
+ kgdb_contthread = NULL;
+ } else {
+ thread = getthread(ks->linux_regs, ks->threadid);
+ if (!thread && ks->threadid > 0) {
+ error_packet(remcom_out_buffer, -EINVAL);
+ break;
+ }
+ kgdb_contthread = thread;
+ }
+ strcpy(remcom_out_buffer, "OK");
+ break;
+ }
+}
+
+/* Handle the 'T' thread query packets */
+static void gdb_cmd_thread(struct kgdb_state *ks)
+{
+ char *ptr = &remcom_in_buffer[1];
+ struct task_struct *thread;
+
+ kgdb_hex2long(&ptr, &ks->threadid);
+ thread = getthread(ks->linux_regs, ks->threadid);
+ if (thread)
+ strcpy(remcom_out_buffer, "OK");
+ else
+ error_packet(remcom_out_buffer, -EINVAL);
+}
+
+/* Handle the 'z' or 'Z' breakpoint remove or set packets */
+static void gdb_cmd_break(struct kgdb_state *ks)
+{
+ /*
+ * Since GDB-5.3, it's been drafted that '0' is a software
+ * breakpoint, '1' is a hardware breakpoint, so let's do that.
+ */
+ char *bpt_type = &remcom_in_buffer[1];
+ char *ptr = &remcom_in_buffer[2];
+ unsigned long addr;
+ unsigned long length;
+ int error = 0;
+
+ if (arch_kgdb_ops.set_hw_breakpoint && *bpt_type >= '1') {
+ /* Unsupported */
+ if (*bpt_type > '4')
+ return;
+ } else {
+ if (*bpt_type != '0' && *bpt_type != '1')
+ /* Unsupported. */
+ return;
+ }
+
+ /*
+ * Test if this is a hardware breakpoint, and
+ * if we support it:
+ */
+ if (*bpt_type == '1' && !(arch_kgdb_ops.flags & KGDB_HW_BREAKPOINT))
+ /* Unsupported. */
+ return;
+
+ if (*(ptr++) != ',') {
+ error_packet(remcom_out_buffer, -EINVAL);
+ return;
+ }
+ if (!kgdb_hex2long(&ptr, &addr)) {
+ error_packet(remcom_out_buffer, -EINVAL);
+ return;
+ }
+ if (*(ptr++) != ',' ||
+ !kgdb_hex2long(&ptr, &length)) {
+ error_packet(remcom_out_buffer, -EINVAL);
+ return;
+ }
+
+ if (remcom_in_buffer[0] == 'Z' && *bpt_type == '0')
+ error = dbg_set_sw_break(addr);
+ else if (remcom_in_buffer[0] == 'z' && *bpt_type == '0')
+ error = dbg_remove_sw_break(addr);
+ else if (remcom_in_buffer[0] == 'Z')
+ error = arch_kgdb_ops.set_hw_breakpoint(addr,
+ (int)length, *bpt_type - '0');
+ else if (remcom_in_buffer[0] == 'z')
+ error = arch_kgdb_ops.remove_hw_breakpoint(addr,
+ (int) length, *bpt_type - '0');
+
+ if (error == 0)
+ strcpy(remcom_out_buffer, "OK");
+ else
+ error_packet(remcom_out_buffer, error);
+}
+
+/* Handle the 'C' signal / exception passing packets */
+static int gdb_cmd_exception_pass(struct kgdb_state *ks)
+{
+ /* C09 == pass exception
+ * C15 == detach kgdb, pass exception
+ */
+ if (remcom_in_buffer[1] == '0' && remcom_in_buffer[2] == '9') {
+
+ ks->pass_exception = 1;
+ remcom_in_buffer[0] = 'c';
+
+ } else if (remcom_in_buffer[1] == '1' && remcom_in_buffer[2] == '5') {
+
+ ks->pass_exception = 1;
+ remcom_in_buffer[0] = 'D';
+ dbg_remove_all_break();
+ kgdb_connected = 0;
+ return 1;
+
+ } else {
+ gdbstub_msg_write("KGDB only knows signal 9 (pass)"
+ " and 15 (pass and disconnect)\n"
+ "Executing a continue without signal passing\n", 0);
+ remcom_in_buffer[0] = 'c';
+ }
+
+ /* Indicate fall through */
+ return -1;
+}
+
+/*
+ * This function performs all gdbserial command procesing
+ */
+int gdb_serial_stub(struct kgdb_state *ks)
+{
+ int error = 0;
+ int tmp;
+
+ /* Clear the out buffer. */
+ memset(remcom_out_buffer, 0, sizeof(remcom_out_buffer));
+
+ if (kgdb_connected) {
+ unsigned char thref[8];
+ char *ptr;
+
+ /* Reply to host that an exception has occurred */
+ ptr = remcom_out_buffer;
+ *ptr++ = 'T';
+ ptr = pack_hex_byte(ptr, ks->signo);
+ ptr += strlen(strcpy(ptr, "thread:"));
+ int_to_threadref(thref, shadow_pid(current->pid));
+ ptr = pack_threadid(ptr, thref);
+ *ptr++ = ';';
+ put_packet(remcom_out_buffer);
+ }
+
+ kgdb_usethread = kgdb_info[ks->cpu].task;
+ ks->kgdb_usethreadid = shadow_pid(kgdb_info[ks->cpu].task->pid);
+ ks->pass_exception = 0;
+
+ while (1) {
+ error = 0;
+
+ /* Clear the out buffer. */
+ memset(remcom_out_buffer, 0, sizeof(remcom_out_buffer));
+
+ get_packet(remcom_in_buffer);
+
+ switch (remcom_in_buffer[0]) {
+ case '?': /* gdbserial status */
+ gdb_cmd_status(ks);
+ break;
+ case 'g': /* return the value of the CPU registers */
+ gdb_cmd_getregs(ks);
+ break;
+ case 'G': /* set the value of the CPU registers - return OK */
+ gdb_cmd_setregs(ks);
+ break;
+ case 'm': /* mAA..AA,LLLL Read LLLL bytes at address AA..AA */
+ gdb_cmd_memread(ks);
+ break;
+ case 'M': /* MAA..AA,LLLL: Write LLLL bytes at address AA..AA */
+ gdb_cmd_memwrite(ks);
+ break;
+ case 'X': /* XAA..AA,LLLL: Write LLLL bytes at address AA..AA */
+ gdb_cmd_binwrite(ks);
+ break;
+ /* kill or detach. KGDB should treat this like a
+ * continue.
+ */
+ case 'D': /* Debugger detach */
+ case 'k': /* Debugger detach via kill */
+ gdb_cmd_detachkill(ks);
+ goto default_handle;
+ case 'R': /* Reboot */
+ if (gdb_cmd_reboot(ks))
+ goto default_handle;
+ break;
+ case 'q': /* query command */
+ gdb_cmd_query(ks);
+ break;
+ case 'H': /* task related */
+ gdb_cmd_task(ks);
+ break;
+ case 'T': /* Query thread status */
+ gdb_cmd_thread(ks);
+ break;
+ case 'z': /* Break point remove */
+ case 'Z': /* Break point set */
+ gdb_cmd_break(ks);
+ break;
+ case 'C': /* Exception passing */
+ tmp = gdb_cmd_exception_pass(ks);
+ if (tmp > 0)
+ goto default_handle;
+ if (tmp == 0)
+ break;
+ /* Fall through on tmp < 0 */
+ case 'c': /* Continue packet */
+ case 's': /* Single step packet */
+ if (kgdb_contthread && kgdb_contthread != current) {
+ /* Can't switch threads in kgdb */
+ error_packet(remcom_out_buffer, -EINVAL);
+ break;
+ }
+ dbg_activate_sw_breakpoints();
+ /* Fall through to default processing */
+ default:
+default_handle:
+ error = kgdb_arch_handle_exception(ks->ex_vector,
+ ks->signo,
+ ks->err_code,
+ remcom_in_buffer,
+ remcom_out_buffer,
+ ks->linux_regs);
+ /*
+ * Leave cmd processing on error, detach,
+ * kill, continue, or single step.
+ */
+ if (error >= 0 || remcom_in_buffer[0] == 'D' ||
+ remcom_in_buffer[0] == 'k') {
+ error = 0;
+ goto kgdb_exit;
+ }
+
+ }
+
+ /* reply to the request */
+ put_packet(remcom_out_buffer);
+ }
+
+kgdb_exit:
+ if (ks->pass_exception)
+ error = 1;
+ return error;
+}
--
1.6.4.rc1
These are the minimum changes to the kgdb core in order to enable an
API to connect a new front end (kdb) to the debug core.
This patch introduces the dbg_kdb_mode variable controls where the
user level I/O is routed. It will be routed to the gdbstub (kgdb) or
to the kdb front end which is a simple shell available over the kgdboc
connection.
You can switch back and forth between kdb or the gdb stub mode of
operation dynamically. From gdb stub mode you can blindly type
"$3#33", or from the kdb mode you can enter "kgdb" to switch to the
gdb stub.
The logic in the debug core depends on kdb to look for the typical gdb
connection sequences and return immediately with KGDB_PASS_EVENT if a
gdb serial command sequence is detected. That should allow a
reasonably seamless transition between kdb -> gdb without leaving the
kernel exception state. The two gdb serial queries that kdb is
responsible for detecting are the "?" and "qSupported" packets.
CC: Ingo Molnar <[email protected]>
Signed-off-by: Jason Wessel <[email protected]>
---
arch/arm/kernel/kgdb.c | 5 ++
arch/mips/kernel/kgdb.c | 5 ++
arch/powerpc/kernel/kgdb.c | 5 ++
arch/x86/kernel/kgdb.c | 5 ++
include/linux/kgdb.h | 11 ++++-
kernel/debug/debug_core.c | 104 +++++++++++++++++++++++++++++++++++++-------
kernel/debug/debug_core.h | 24 ++++++++++
kernel/debug/gdbstub.c | 36 +++++++++++++++
lib/Kconfig.kgdb | 6 +++
9 files changed, 183 insertions(+), 18 deletions(-)
diff --git a/arch/arm/kernel/kgdb.c b/arch/arm/kernel/kgdb.c
index ba8ccfe..bb17e9b 100644
--- a/arch/arm/kernel/kgdb.c
+++ b/arch/arm/kernel/kgdb.c
@@ -97,6 +97,11 @@ sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *task)
gdb_regs[_CPSR] = thread_regs->ARM_cpsr;
}
+void kgdb_arch_set_pc(struct pt_regs *regs, unsigned long pc)
+{
+ regs->ARM_pc = pc;
+}
+
static int compiled_break;
int kgdb_arch_handle_exception(int exception_vector, int signo,
diff --git a/arch/mips/kernel/kgdb.c b/arch/mips/kernel/kgdb.c
index 50c9bb8..6ed4c83 100644
--- a/arch/mips/kernel/kgdb.c
+++ b/arch/mips/kernel/kgdb.c
@@ -180,6 +180,11 @@ void sleeping_thread_to_gdb_regs(unsigned long *gdb_regs, struct task_struct *p)
*(ptr++) = regs->cp0_epc;
}
+void kgdb_arch_set_pc(struct pt_regs *regs, unsigned long pc)
+{
+ regs->cp0_epc = pc;
+}
+
/*
* Calls linux_debug_hook before the kernel dies. If KGDB is enabled,
* then try to fall into the debugger
diff --git a/arch/powerpc/kernel/kgdb.c b/arch/powerpc/kernel/kgdb.c
index b6bd1ea..7aafcc5 100644
--- a/arch/powerpc/kernel/kgdb.c
+++ b/arch/powerpc/kernel/kgdb.c
@@ -309,6 +309,11 @@ void gdb_regs_to_pt_regs(unsigned long *gdb_regs, struct pt_regs *regs)
(unsigned long)(((void *)gdb_regs) + NUMREGBYTES));
}
+void kgdb_arch_set_pc(struct pt_regs *regs, unsigned long pc)
+{
+ regs->nip = pc;
+}
+
/*
* This function does PowerPC specific procesing for interfacing to gdb.
*/
diff --git a/arch/x86/kernel/kgdb.c b/arch/x86/kernel/kgdb.c
index bfba601..c7a20d7 100644
--- a/arch/x86/kernel/kgdb.c
+++ b/arch/x86/kernel/kgdb.c
@@ -690,6 +690,11 @@ unsigned long kgdb_arch_pc(int exception, struct pt_regs *regs)
return instruction_pointer(regs);
}
+void kgdb_arch_set_pc(struct pt_regs *regs, unsigned long ip)
+{
+ regs->ip = ip;
+}
+
struct kgdb_arch arch_kgdb_ops = {
/* Breakpoint instruction: */
.gdb_bpt_instr = { 0xcc },
diff --git a/include/linux/kgdb.h b/include/linux/kgdb.h
index 4830142..5b37df0 100644
--- a/include/linux/kgdb.h
+++ b/include/linux/kgdb.h
@@ -16,10 +16,12 @@
#include <linux/serial_8250.h>
#include <linux/linkage.h>
#include <linux/init.h>
-
#include <asm/atomic.h>
+#ifdef CONFIG_HAVE_ARCH_KGDB
#include <asm/kgdb.h>
+#endif
+#ifdef CONFIG_KGDB
struct pt_regs;
/**
@@ -262,6 +264,7 @@ extern struct kgdb_arch arch_kgdb_ops;
extern unsigned long __weak kgdb_arch_pc(int exception, struct pt_regs *regs);
+extern void kgdb_arch_set_pc(struct pt_regs *regs, unsigned long pc);
extern int kgdb_register_io_module(struct kgdb_io *local_kgdb_io_ops);
extern void kgdb_unregister_io_module(struct kgdb_io *local_kgdb_io_ops);
extern struct kgdb_io *dbg_io_ops;
@@ -279,5 +282,9 @@ extern int kgdb_nmicallback(int cpu, void *regs);
extern int kgdb_single_step;
extern atomic_t kgdb_active;
-
+#define in_dbg_master() \
+ (raw_smp_processor_id() == atomic_read(&kgdb_active))
+#else /* ! CONFIG_KGDB */
+#define in_dbg_master() (0)
+#endif /* ! CONFIG_KGDB */
#endif /* _KGDB_H_ */
diff --git a/kernel/debug/debug_core.c b/kernel/debug/debug_core.c
index 91286e8..2e01595 100644
--- a/kernel/debug/debug_core.c
+++ b/kernel/debug/debug_core.c
@@ -43,6 +43,7 @@
#include <linux/sysrq.h>
#include <linux/init.h>
#include <linux/kgdb.h>
+#include <linux/kdb.h>
#include <linux/pid.h>
#include <linux/smp.h>
#include <linux/mm.h>
@@ -77,6 +78,11 @@ static DEFINE_SPINLOCK(kgdb_registration_lock);
static int kgdb_con_registered;
/* determine if kgdb console output should be used */
static int kgdb_use_con;
+/* Next cpu to become the master debug core */
+int dbg_switch_cpu;
+
+/* Use kdb or gdbserver mode */
+static int dbg_kdb_mode = 1;
static int __init opt_kgdb_con(char *str)
{
@@ -100,6 +106,7 @@ static struct kgdb_bkpt kgdb_break[KGDB_MAX_BREAKPOINTS] = {
* The CPU# of the active CPU, or -1 if none:
*/
atomic_t kgdb_active = ATOMIC_INIT(-1);
+EXPORT_SYMBOL_GPL(kgdb_active);
/*
* We use NR_CPUs not PERCPU, in case kgdb is used to debug early
@@ -301,7 +308,7 @@ int dbg_set_sw_break(unsigned long addr)
return 0;
}
-static int kgdb_deactivate_sw_breakpoints(void)
+int dbg_deactivate_sw_breakpoints(void)
{
unsigned long addr;
int error;
@@ -395,8 +402,14 @@ static int kgdb_io_ready(int print_wait)
return 1;
if (atomic_read(&kgdb_setting_breakpoint))
return 1;
- if (print_wait)
+ if (print_wait) {
+#ifdef CONFIG_KGDB_KDB
+ if (!dbg_kdb_mode)
+ printk(KERN_CRIT "KGDB: waiting... or $3#33 for KDB\n");
+#else
printk(KERN_CRIT "KGDB: Waiting for remote debugger\n");
+#endif
+ }
return 1;
}
@@ -410,7 +423,7 @@ static int kgdb_reenter_check(struct kgdb_state *ks)
/* Panic on recursive debugger calls: */
exception_level++;
addr = kgdb_arch_pc(ks->ex_vector, ks->linux_regs);
- kgdb_deactivate_sw_breakpoints();
+ dbg_deactivate_sw_breakpoints();
/*
* If the break point removed ok at the place exception
@@ -443,11 +456,24 @@ static int kgdb_reenter_check(struct kgdb_state *ks)
return 1;
}
+static void dbg_cpu_switch(int cpu, int next_cpu)
+{
+ /* Mark the cpu we are switching away from as a slave when it
+ * holds the kgdb_active token. This must be done so that the
+ * that all the cpus wait in for the debug core will not enter
+ * again as the master. */
+ if (cpu == atomic_read(&kgdb_active)) {
+ kgdb_info[cpu].exception_state |= DCPU_IS_SLAVE;
+ kgdb_info[cpu].exception_state &= ~DCPU_WANT_MASTER;
+ }
+ kgdb_info[next_cpu].exception_state |= DCPU_NEXT_MASTER;
+}
+
static int kgdb_cpu_enter(struct kgdb_state *ks, struct pt_regs *regs)
{
unsigned long flags;
int sstep_tries = 100;
- int error = 0;
+ int error;
int i, cpu;
acquirelock:
/*
@@ -459,6 +485,8 @@ acquirelock:
cpu = ks->cpu;
kgdb_info[cpu].debuggerinfo = regs;
kgdb_info[cpu].task = current;
+ kgdb_info[cpu].ret_state = 0;
+ kgdb_info[cpu].irq_depth = hardirq_count() >> HARDIRQ_SHIFT;
/*
* Make sure the above info reaches the primary CPU before
* our cpu_in_kgdb[] flag setting does:
@@ -471,7 +499,11 @@ acquirelock:
* master cpu and acquire the kgdb_active lock:
*/
while (1) {
- if (kgdb_info[cpu].exception_state & DCPU_WANT_MASTER) {
+cpu_loop:
+ if (kgdb_info[cpu].exception_state & DCPU_NEXT_MASTER) {
+ kgdb_info[cpu].exception_state &= ~DCPU_NEXT_MASTER;
+ goto cpu_master_loop;
+ } else if (kgdb_info[cpu].exception_state & DCPU_WANT_MASTER) {
if (atomic_cmpxchg(&kgdb_active, -1, cpu) == cpu)
break;
} else if (kgdb_info[cpu].exception_state & DCPU_IS_SLAVE) {
@@ -511,7 +543,7 @@ return_normal:
}
if (!kgdb_io_ready(1)) {
- error = 1;
+ kgdb_info[cpu].ret_state = 1;
goto kgdb_restore; /* No I/O connection, resume the system */
}
@@ -555,13 +587,31 @@ return_normal:
* in the debugger and all secondary CPUs are quiescent
*/
kgdb_post_primary_code(ks->linux_regs, ks->ex_vector, ks->err_code);
- kgdb_deactivate_sw_breakpoints();
+ dbg_deactivate_sw_breakpoints();
kgdb_single_step = 0;
kgdb_contthread = current;
exception_level = 0;
- /* Talk to debugger with gdbserial protocol */
- error = gdb_serial_stub(ks);
+ while (1) {
+cpu_master_loop:
+ if (dbg_kdb_mode) {
+ kgdb_connected = 1;
+ error = kdb_stub(ks);
+ } else {
+ error = gdb_serial_stub(ks);
+ }
+
+ if (error == DBG_PASS_EVENT) {
+ dbg_kdb_mode = !dbg_kdb_mode;
+ kgdb_connected = 0;
+ } else if (error == DBG_SWITCH_CPU_EVENT) {
+ dbg_cpu_switch(cpu, dbg_switch_cpu);
+ goto cpu_loop;
+ } else {
+ kgdb_info[cpu].ret_state = error;
+ break;
+ }
+ }
/* Call the I/O driver's post_exception routine */
if (dbg_io_ops->post_exception)
@@ -573,11 +623,15 @@ return_normal:
for (i = NR_CPUS-1; i >= 0; i--)
atomic_set(&passive_cpu_wait[i], 0);
/*
- * Wait till all the CPUs have quit
- * from the debugger.
+ * Wait till all the CPUs have quit from the debugger,
+ * but allow a CPU that hit an exception and is
+ * waiting to become the master to remain in the debug
+ * core.
*/
for_each_online_cpu(i) {
- while (atomic_read(&cpu_in_kgdb[i]))
+ while (atomic_read(&cpu_in_kgdb[i]) &&
+ !(kgdb_info[i].exception_state &
+ DCPU_WANT_MASTER))
cpu_relax();
}
}
@@ -596,7 +650,7 @@ kgdb_restore:
clocksource_touch_watchdog();
local_irq_restore(flags);
- return error;
+ return kgdb_info[cpu].ret_state;
}
/*
@@ -625,7 +679,8 @@ kgdb_handle_exception(int evector, int signo, int ecode, struct pt_regs *regs)
return 0; /* Ouch, double exception ! */
kgdb_info[ks->cpu].exception_state |= DCPU_WANT_MASTER;
ret = kgdb_cpu_enter(ks, regs);
- kgdb_info[ks->cpu].exception_state &= ~DCPU_WANT_MASTER;
+ kgdb_info[ks->cpu].exception_state &= ~(DCPU_WANT_MASTER |
+ DCPU_IS_SLAVE);
return ret;
}
@@ -658,7 +713,7 @@ static void kgdb_console_write(struct console *co, const char *s,
/* If we're debugging, or KGDB has not connected, don't try
* and print. */
- if (!kgdb_connected || atomic_read(&kgdb_active) != -1)
+ if (!kgdb_connected || atomic_read(&kgdb_active) != -1 || dbg_kdb_mode)
return;
local_irq_save(flags);
@@ -680,8 +735,14 @@ static void sysrq_handle_dbg(int key, struct tty_struct *tty)
printk(KERN_CRIT "ERROR: No KGDB I/O module available\n");
return;
}
- if (!kgdb_connected)
+ if (!kgdb_connected) {
+#ifdef CONFIG_KGDB_KDB
+ if (!dbg_kdb_mode)
+ printk(KERN_CRIT "KGDB or $3#33 for KDB\n");
+#else
printk(KERN_CRIT "Entering KGDB\n");
+#endif
+ }
kgdb_breakpoint();
}
@@ -810,6 +871,16 @@ void kgdb_unregister_io_module(struct kgdb_io *old_dbg_io_ops)
}
EXPORT_SYMBOL_GPL(kgdb_unregister_io_module);
+int dbg_io_get_char(void)
+{
+ int ret = dbg_io_ops->read_char();
+ if (!dbg_kdb_mode)
+ return ret;
+ if (ret == 127)
+ return 8;
+ return ret;
+}
+
/**
* kgdb_breakpoint - generate breakpoint exception
*
@@ -832,6 +903,7 @@ static int __init opt_kgdb_wait(char *str)
{
kgdb_break_asap = 1;
+ kdb_init(KDB_INIT_EARLY);
if (kgdb_io_module_registered)
kgdb_initial_breakpoint();
diff --git a/kernel/debug/debug_core.h b/kernel/debug/debug_core.h
index 9825c5e..a7051fa 100644
--- a/kernel/debug/debug_core.h
+++ b/kernel/debug/debug_core.h
@@ -38,6 +38,8 @@ struct debuggerinfo_struct {
void *debuggerinfo;
struct task_struct *task;
int exception_state;
+ int ret_state;
+ int irq_depth;
};
extern struct debuggerinfo_struct kgdb_info[];
@@ -47,9 +49,31 @@ extern int dbg_remove_all_break(void);
extern int dbg_set_sw_break(unsigned long addr);
extern int dbg_remove_sw_break(unsigned long addr);
extern int dbg_activate_sw_breakpoints(void);
+extern int dbg_deactivate_sw_breakpoints(void);
+
+/* polled character access to i/o module */
+extern int dbg_io_get_char(void);
+
+/* stub return value for switching between the gdbstub and kdb */
+#define DBG_PASS_EVENT -12345
+/* Switch from one cpu to another */
+#define DBG_SWITCH_CPU_EVENT -123456
+extern int dbg_switch_cpu;
/* gdbstub interface functions */
extern int gdb_serial_stub(struct kgdb_state *ks);
extern void gdbstub_msg_write(const char *s, int len);
+/* gdbstub functions used for kdb <-> gdbstub transition */
+extern int gdbstub_state(struct kgdb_state *ks, char *cmd);
+
+#ifdef CONFIG_KGDB_KDB
+extern int kdb_stub(struct kgdb_state *ks);
+#else /* ! CONFIG_KGDB_KDB */
+static inline int kdb_stub(struct kgdb_state *ks)
+{
+ return DBG_PASS_EVENT;
+}
+#endif /* CONFIG_KGDB_KDB */
+
#endif /* _DEBUG_CORE_H_ */
diff --git a/kernel/debug/gdbstub.c b/kernel/debug/gdbstub.c
index ab29bce..195ffbd 100644
--- a/kernel/debug/gdbstub.c
+++ b/kernel/debug/gdbstub.c
@@ -892,6 +892,13 @@ int gdb_serial_stub(struct kgdb_state *ks)
case 'Z': /* Break point set */
gdb_cmd_break(ks);
break;
+#ifdef CONFIG_KGDB_KDB
+ case '3': /* Escape into back into kdb */
+ if (remcom_in_buffer[1] == '\0') {
+ gdb_cmd_detachkill(ks);
+ return DBG_PASS_EVENT;
+ }
+#endif
case 'C': /* Exception passing */
tmp = gdb_cmd_exception_pass(ks);
if (tmp > 0)
@@ -937,3 +944,32 @@ kgdb_exit:
error = 1;
return error;
}
+
+int gdbstub_state(struct kgdb_state *ks, char *cmd)
+{
+ int error;
+
+ switch (cmd[0]) {
+ case 'e':
+ error = kgdb_arch_handle_exception(ks->ex_vector,
+ ks->signo,
+ ks->err_code,
+ remcom_in_buffer,
+ remcom_out_buffer,
+ ks->linux_regs);
+ return error;
+ case 's':
+ case 'c':
+ strcpy(remcom_in_buffer, cmd);
+ return 0;
+ case '?':
+ gdb_cmd_status(ks);
+ break;
+ case '\0':
+ strcpy(remcom_out_buffer, "");
+ break;
+ }
+ dbg_io_ops->write_char('+');
+ put_packet(remcom_out_buffer);
+ return 0;
+}
diff --git a/lib/Kconfig.kgdb b/lib/Kconfig.kgdb
index 9b5d1d7..02b46de 100644
--- a/lib/Kconfig.kgdb
+++ b/lib/Kconfig.kgdb
@@ -57,4 +57,10 @@ config KGDB_TESTS_BOOT_STRING
information about other strings you could use beyond the
default of V1F100.
+config KGDB_KDB
+ bool "KGDB_KDB: include kdb frontend for kgdb"
+ default n
+ help
+ KDB frontend for kernel
+
endif # KGDB
--
1.6.4.rc1
The design of kdb required that every device that can provide input to
kdb have a polling routine that exits immediately if there is no
character available.
This is required in order to get the page scrolling mechanism working,
it is also a reasonable requirement for future kgdb I/O drivers
because that allows for the possibility of multiple input channels.
NO_POLL_CHAR will be the return code to the polling routine when ever
there is no character available. There are several other console
polling drivers which can be modified, but for the prototype only the
8250 and pl011 driver have been changed to make use of this.
CC: [email protected]
Signed-off-by: Jason Wessel <[email protected]>
---
drivers/serial/8250.c | 4 ++--
drivers/serial/amba-pl011.c | 6 +++---
include/linux/kdb.h | 1 +
include/linux/serial_core.h | 1 +
kernel/debug/debug_core.c | 2 ++
kernel/debug/gdbstub.c | 37 +++++++++++++++++++++++++++++++------
kernel/debug/kdb/kdb_debugger.c | 10 ++++++++++
7 files changed, 50 insertions(+), 11 deletions(-)
diff --git a/drivers/serial/8250.c b/drivers/serial/8250.c
index c3e37c8..a488c60 100644
--- a/drivers/serial/8250.c
+++ b/drivers/serial/8250.c
@@ -1895,8 +1895,8 @@ static int serial8250_get_poll_char(struct uart_port *port)
struct uart_8250_port *up = (struct uart_8250_port *)port;
unsigned char lsr = serial_inp(up, UART_LSR);
- while (!(lsr & UART_LSR_DR))
- lsr = serial_inp(up, UART_LSR);
+ if (!(lsr & UART_LSR_DR))
+ return NO_POLL_CHAR;
return serial_inp(up, UART_RX);
}
diff --git a/drivers/serial/amba-pl011.c b/drivers/serial/amba-pl011.c
index ef7adc8..b411115 100644
--- a/drivers/serial/amba-pl011.c
+++ b/drivers/serial/amba-pl011.c
@@ -335,9 +335,9 @@ static int pl010_get_poll_char(struct uart_port *port)
struct uart_amba_port *uap = (struct uart_amba_port *)port;
unsigned int status;
- do {
- status = readw(uap->port.membase + UART01x_FR);
- } while (status & UART01x_FR_RXFE);
+ status = readw(uap->port.membase + UART01x_FR);
+ if (status & UART01x_FR_RXFE)
+ return NO_POLL_CHAR;
return readw(uap->port.membase + UART01x_DR);
}
diff --git a/include/linux/kdb.h b/include/linux/kdb.h
index b3ac39a..f44aa5f 100644
--- a/include/linux/kdb.h
+++ b/include/linux/kdb.h
@@ -19,6 +19,7 @@
#include <asm/atomic.h>
#define KDB_POLL_FUNC_MAX 5
+extern int kdb_poll_idx;
/*
* kdb_initial_cpu is initialized to -1, and is set to the cpu
diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h
index 8c3dd36..d40db83 100644
--- a/include/linux/serial_core.h
+++ b/include/linux/serial_core.h
@@ -246,6 +246,7 @@ struct uart_ops {
#endif
};
+#define NO_POLL_CHAR 0x00ff0000
#define UART_CONFIG_TYPE (1 << 0)
#define UART_CONFIG_IRQ (1 << 1)
diff --git a/kernel/debug/debug_core.c b/kernel/debug/debug_core.c
index 2e01595..5bd2670 100644
--- a/kernel/debug/debug_core.c
+++ b/kernel/debug/debug_core.c
@@ -874,6 +874,8 @@ EXPORT_SYMBOL_GPL(kgdb_unregister_io_module);
int dbg_io_get_char(void)
{
int ret = dbg_io_ops->read_char();
+ if (ret == NO_POLL_CHAR)
+ return -1;
if (!dbg_kdb_mode)
return ret;
if (ret == 127)
diff --git a/kernel/debug/gdbstub.c b/kernel/debug/gdbstub.c
index 195ffbd..6965b5c 100644
--- a/kernel/debug/gdbstub.c
+++ b/kernel/debug/gdbstub.c
@@ -30,6 +30,7 @@
#include <linux/kernel.h>
#include <linux/kgdb.h>
+#include <linux/kdb.h>
#include <linux/reboot.h>
#include <linux/uaccess.h>
#include <asm/cacheflush.h>
@@ -62,6 +63,30 @@ static int hex(char ch)
return -1;
}
+#ifdef CONFIG_KGDB_KDB
+static int gdbstub_read_wait(void)
+{
+ int ret = -1;
+ int i;
+
+ /* poll any additional I/O interfaces that are defined */
+ while (ret < 0)
+ for (i = 0; kdb_poll_funcs[i] != NULL; i++) {
+ ret = kdb_poll_funcs[i]();
+ if (ret > 0)
+ break;
+ }
+ return ret;
+}
+#else
+static int gdbstub_read_wait(void)
+{
+ int ret = dbg_io_ops->read_char();
+ while (ret == NO_POLL_CHAR)
+ ret = dbg_io_ops->read_char();
+ return ret;
+}
+#endif
/* scan for the sequence $<data>#<checksum> */
static void get_packet(char *buffer)
{
@@ -75,7 +100,7 @@ static void get_packet(char *buffer)
* Spin and wait around for the start character, ignore all
* other characters:
*/
- while ((ch = (dbg_io_ops->read_char())) != '$')
+ while ((ch = (gdbstub_read_wait())) != '$')
/* nothing */;
kgdb_connected = 1;
@@ -88,7 +113,7 @@ static void get_packet(char *buffer)
* now, read until a # or end of buffer is found:
*/
while (count < (BUFMAX - 1)) {
- ch = dbg_io_ops->read_char();
+ ch = gdbstub_read_wait();
if (ch == '#')
break;
checksum = checksum + ch;
@@ -98,8 +123,8 @@ static void get_packet(char *buffer)
buffer[count] = 0;
if (ch == '#') {
- xmitcsum = hex(dbg_io_ops->read_char()) << 4;
- xmitcsum += hex(dbg_io_ops->read_char());
+ xmitcsum = hex(gdbstub_read_wait()) << 4;
+ xmitcsum += hex(gdbstub_read_wait());
if (checksum != xmitcsum)
/* failed checksum */
@@ -144,10 +169,10 @@ static void put_packet(char *buffer)
dbg_io_ops->flush();
/* Now see what we get in reply. */
- ch = dbg_io_ops->read_char();
+ ch = gdbstub_read_wait();
if (ch == 3)
- ch = dbg_io_ops->read_char();
+ ch = gdbstub_read_wait();
/* If we get an ACK, we are done. */
if (ch == '+')
diff --git a/kernel/debug/kdb/kdb_debugger.c b/kernel/debug/kdb/kdb_debugger.c
index f024c0c..bf6e827 100644
--- a/kernel/debug/kdb/kdb_debugger.c
+++ b/kernel/debug/kdb/kdb_debugger.c
@@ -20,7 +20,15 @@
get_char_func kdb_poll_funcs[] = {
dbg_io_get_char,
NULL,
+ NULL,
+ NULL,
+ NULL,
+ NULL,
};
+EXPORT_SYMBOL_GPL(kdb_poll_funcs);
+
+int kdb_poll_idx = 1;
+EXPORT_SYMBOL_GPL(kdb_poll_idx);
int kdb_stub(struct kgdb_state *ks)
{
@@ -85,6 +93,7 @@ int kdb_stub(struct kgdb_state *ks)
kdb_bp_remove();
KDB_STATE_CLEAR(DOING_SS);
KDB_STATE_CLEAR(DOING_SSB);
+ KDB_STATE_SET(PAGER);
/* zero out any offline cpu data */
for_each_present_cpu(i) {
if (!cpu_online(i)) {
@@ -112,6 +121,7 @@ int kdb_stub(struct kgdb_state *ks)
kdb_initial_cpu = -1;
kdb_current_task = NULL;
kdb_current_regs = NULL;
+ KDB_STATE_CLEAR(PAGER);
kdbnearsym_cleanup();
if (error == KDB_CMD_KGDB) {
if (KDB_STATE(DOING_KGDB) || KDB_STATE(DOING_KGDB2)) {
--
1.6.4.rc1
Allow kgdboc to work on sparc hardware with the Zilog serial chips.
Signed-off-by: Jason Wessel <[email protected]>
Acked-by: David S. Miller <[email protected]>
---
drivers/serial/sunzilog.c | 50 +++++++++++++++++++++++++++++++++++++++++++++
1 files changed, 50 insertions(+), 0 deletions(-)
diff --git a/drivers/serial/sunzilog.c b/drivers/serial/sunzilog.c
index 2c7a66a..978b3ce 100644
--- a/drivers/serial/sunzilog.c
+++ b/drivers/serial/sunzilog.c
@@ -102,6 +102,8 @@ struct uart_sunzilog_port {
#endif
};
+static void sunzilog_putchar(struct uart_port *port, int ch);
+
#define ZILOG_CHANNEL_FROM_PORT(PORT) ((struct zilog_channel __iomem *)((PORT)->membase))
#define UART_ZILOG(PORT) ((struct uart_sunzilog_port *)(PORT))
@@ -996,6 +998,50 @@ static int sunzilog_verify_port(struct uart_port *port, struct serial_struct *se
return -EINVAL;
}
+#ifdef CONFIG_CONSOLE_POLL
+static int sunzilog_get_poll_char(struct uart_port *port)
+{
+ unsigned char ch, r1;
+ struct uart_sunzilog_port *up = (struct uart_sunzilog_port *) port;
+ struct zilog_channel __iomem *channel
+ = ZILOG_CHANNEL_FROM_PORT(&up->port);
+
+
+ r1 = read_zsreg(channel, R1);
+ if (r1 & (PAR_ERR | Rx_OVR | CRC_ERR)) {
+ writeb(ERR_RES, &channel->control);
+ ZSDELAY();
+ ZS_WSYNC(channel);
+ }
+
+ ch = readb(&channel->control);
+ ZSDELAY();
+
+ /* This funny hack depends upon BRK_ABRT not interfering
+ * with the other bits we care about in R1.
+ */
+ if (ch & BRK_ABRT)
+ r1 |= BRK_ABRT;
+
+ if (!(ch & Rx_CH_AV))
+ return NO_POLL_CHAR;
+
+ ch = readb(&channel->data);
+ ZSDELAY();
+
+ ch &= up->parity_mask;
+ return ch;
+}
+
+static void sunzilog_put_poll_char(struct uart_port *port,
+ unsigned char ch)
+{
+ struct uart_sunzilog_port *up = (struct uart_sunzilog_port *)port;
+
+ sunzilog_putchar(&up->port, ch);
+}
+#endif /* CONFIG_CONSOLE_POLL */
+
static struct uart_ops sunzilog_pops = {
.tx_empty = sunzilog_tx_empty,
.set_mctrl = sunzilog_set_mctrl,
@@ -1013,6 +1059,10 @@ static struct uart_ops sunzilog_pops = {
.request_port = sunzilog_request_port,
.config_port = sunzilog_config_port,
.verify_port = sunzilog_verify_port,
+#ifdef CONFIG_CONSOLE_POLL
+ .poll_get_char = sunzilog_get_poll_char,
+ .poll_put_char = sunzilog_put_poll_char,
+#endif
};
static int uart_chip_count;
--
1.6.4.rc1
This is a kgdb architectural change to have all the cpus (master or
slave) enter the same function.
A cpu that hits an exception (wants to be the master cpu) will call
kgdb_handle_exception() from the trap handler and then invoke a
kgdb_roundup_cpu() to synchronize the other cpus and bring them into
the kgdb_handle_exception() as well.
A slave cpu will enter kgdb_handle_exception() from the
kgdb_nmicallback() and set the exception state to note that the
processor is a slave.
Previously the salve cpu would have called kgdb_wait(). This change
allows the debug core to change cpus without resuming the system in
order to inspect arch specific cpu information.
Signed-off-by: Jason Wessel <[email protected]>
---
kernel/debug/debug_core.c | 158 +++++++++++++++++++++-----------------------
kernel/debug/debug_core.h | 7 ++
2 files changed, 82 insertions(+), 83 deletions(-)
diff --git a/kernel/debug/debug_core.c b/kernel/debug/debug_core.c
index ad3caaa..91286e8 100644
--- a/kernel/debug/debug_core.c
+++ b/kernel/debug/debug_core.c
@@ -215,49 +215,6 @@ void __weak kgdb_disable_hw_debug(struct pt_regs *regs)
}
/*
- * CPU debug state control:
- */
-
-#ifdef CONFIG_SMP
-static void kgdb_wait(struct pt_regs *regs)
-{
- unsigned long flags;
- int cpu;
-
- local_irq_save(flags);
- cpu = raw_smp_processor_id();
- kgdb_info[cpu].debuggerinfo = regs;
- kgdb_info[cpu].task = current;
- /*
- * Make sure the above info reaches the primary CPU before
- * our cpu_in_kgdb[] flag setting does:
- */
- smp_wmb();
- atomic_set(&cpu_in_kgdb[cpu], 1);
-
- /* Disable any cpu specific hw breakpoints */
- kgdb_disable_hw_debug(regs);
-
- /* Wait till primary CPU is done with debugging */
- while (atomic_read(&passive_cpu_wait[cpu]))
- cpu_relax();
-
- kgdb_info[cpu].debuggerinfo = NULL;
- kgdb_info[cpu].task = NULL;
-
- /* fix up hardware debug registers on local cpu */
- if (arch_kgdb_ops.correct_hw_break)
- arch_kgdb_ops.correct_hw_break();
-
- /* Signal the primary CPU that we are done: */
- atomic_set(&cpu_in_kgdb[cpu], 0);
- touch_softlockup_watchdog_sync();
- clocksource_touch_watchdog();
- local_irq_restore(flags);
-}
-#endif
-
-/*
* Some architectures need cache flushes when we set/clear a
* breakpoint:
*/
@@ -486,34 +443,12 @@ static int kgdb_reenter_check(struct kgdb_state *ks)
return 1;
}
-/*
- * kgdb_handle_exception() - main entry point from a kernel exception
- *
- * Locking hierarchy:
- * interface locks, if any (begin_session)
- * kgdb lock (kgdb_active)
- */
-int
-kgdb_handle_exception(int evector, int signo, int ecode, struct pt_regs *regs)
+static int kgdb_cpu_enter(struct kgdb_state *ks, struct pt_regs *regs)
{
- struct kgdb_state kgdb_var;
- struct kgdb_state *ks = &kgdb_var;
unsigned long flags;
int sstep_tries = 100;
int error = 0;
int i, cpu;
-
- ks->cpu = raw_smp_processor_id();
- ks->ex_vector = evector;
- ks->signo = signo;
- ks->ex_vector = evector;
- ks->err_code = ecode;
- ks->kgdb_usethreadid = 0;
- ks->linux_regs = regs;
-
- if (kgdb_reenter_check(ks))
- return 0; /* Ouch, double exception ! */
-
acquirelock:
/*
* Interrupts will be restored by the 'trap return' code, except when
@@ -521,13 +456,42 @@ acquirelock:
*/
local_irq_save(flags);
- cpu = raw_smp_processor_id();
+ cpu = ks->cpu;
+ kgdb_info[cpu].debuggerinfo = regs;
+ kgdb_info[cpu].task = current;
+ /*
+ * Make sure the above info reaches the primary CPU before
+ * our cpu_in_kgdb[] flag setting does:
+ */
+ smp_wmb();
+ atomic_set(&cpu_in_kgdb[cpu], 1);
/*
- * Acquire the kgdb_active lock:
+ * CPU will loop if it is a slave or request to become a kgdb
+ * master cpu and acquire the kgdb_active lock:
*/
- while (atomic_cmpxchg(&kgdb_active, -1, cpu) != -1)
+ while (1) {
+ if (kgdb_info[cpu].exception_state & DCPU_WANT_MASTER) {
+ if (atomic_cmpxchg(&kgdb_active, -1, cpu) == cpu)
+ break;
+ } else if (kgdb_info[cpu].exception_state & DCPU_IS_SLAVE) {
+ if (!atomic_read(&passive_cpu_wait[cpu]))
+ goto return_normal;
+ } else {
+return_normal:
+ /* Return to normal operation by executing any
+ * hw breakpoint fixup.
+ */
+ if (arch_kgdb_ops.correct_hw_break)
+ arch_kgdb_ops.correct_hw_break();
+ atomic_set(&cpu_in_kgdb[cpu], 0);
+ touch_softlockup_watchdog_sync();
+ clocksource_touch_watchdog();
+ local_irq_restore(flags);
+ return 0;
+ }
cpu_relax();
+ }
/*
* For single stepping, try to only enter on the processor
@@ -561,9 +525,6 @@ acquirelock:
if (dbg_io_ops->pre_exception)
dbg_io_ops->pre_exception();
- kgdb_info[ks->cpu].debuggerinfo = ks->linux_regs;
- kgdb_info[ks->cpu].task = current;
-
kgdb_disable_hw_debug(ks->linux_regs);
/*
@@ -575,12 +536,6 @@ acquirelock:
atomic_set(&passive_cpu_wait[i], 1);
}
- /*
- * spin_lock code is good enough as a barrier so we don't
- * need one here:
- */
- atomic_set(&cpu_in_kgdb[ks->cpu], 1);
-
#ifdef CONFIG_SMP
/* Signal the other CPUs to enter kgdb_wait() */
if ((!kgdb_single_step) && kgdb_do_roundup)
@@ -612,8 +567,6 @@ acquirelock:
if (dbg_io_ops->post_exception)
dbg_io_ops->post_exception();
- kgdb_info[ks->cpu].debuggerinfo = NULL;
- kgdb_info[ks->cpu].task = NULL;
atomic_set(&cpu_in_kgdb[ks->cpu], 0);
if (!kgdb_single_step) {
@@ -646,13 +599,52 @@ kgdb_restore:
return error;
}
+/*
+ * kgdb_handle_exception() - main entry point from a kernel exception
+ *
+ * Locking hierarchy:
+ * interface locks, if any (begin_session)
+ * kgdb lock (kgdb_active)
+ */
+int
+kgdb_handle_exception(int evector, int signo, int ecode, struct pt_regs *regs)
+{
+ struct kgdb_state kgdb_var;
+ struct kgdb_state *ks = &kgdb_var;
+ int ret;
+
+ ks->cpu = raw_smp_processor_id();
+ ks->ex_vector = evector;
+ ks->signo = signo;
+ ks->ex_vector = evector;
+ ks->err_code = ecode;
+ ks->kgdb_usethreadid = 0;
+ ks->linux_regs = regs;
+
+ if (kgdb_reenter_check(ks))
+ return 0; /* Ouch, double exception ! */
+ kgdb_info[ks->cpu].exception_state |= DCPU_WANT_MASTER;
+ ret = kgdb_cpu_enter(ks, regs);
+ kgdb_info[ks->cpu].exception_state &= ~DCPU_WANT_MASTER;
+ return ret;
+}
+
int kgdb_nmicallback(int cpu, void *regs)
{
#ifdef CONFIG_SMP
+ struct kgdb_state kgdb_var;
+ struct kgdb_state *ks = &kgdb_var;
+
+ memset(ks, 0, sizeof(struct kgdb_state));
+ ks->cpu = cpu;
+ ks->linux_regs = regs;
+
if (!atomic_read(&cpu_in_kgdb[cpu]) &&
- atomic_read(&kgdb_active) != cpu &&
- atomic_read(&cpu_in_kgdb[atomic_read(&kgdb_active)])) {
- kgdb_wait((struct pt_regs *)regs);
+ atomic_read(&kgdb_active) != -1 &&
+ atomic_read(&kgdb_active) != cpu) {
+ kgdb_info[cpu].exception_state |= DCPU_IS_SLAVE;
+ kgdb_cpu_enter(ks, regs);
+ kgdb_info[cpu].exception_state &= ~DCPU_IS_SLAVE;
return 0;
}
#endif
diff --git a/kernel/debug/debug_core.h b/kernel/debug/debug_core.h
index 4b9d512..9825c5e 100644
--- a/kernel/debug/debug_core.h
+++ b/kernel/debug/debug_core.h
@@ -28,9 +28,16 @@ struct kgdb_state {
struct pt_regs *linux_regs;
};
+/* Exception state values */
+#define DCPU_WANT_MASTER 0x1 /* Waiting to become a master kgdb cpu */
+#define DCPU_NEXT_MASTER 0x2 /* Transition from one master cpu to another */
+#define DCPU_IS_SLAVE 0x4 /* Slave cpu enter exception */
+#define DCPU_SSTEP 0x8 /* CPU is single stepping */
+
struct debuggerinfo_struct {
void *debuggerinfo;
struct task_struct *task;
+ int exception_state;
};
extern struct debuggerinfo_struct kgdb_info[];
--
1.6.4.rc1
The sci_poll_get_char() needs to return immediately if there is no
input from the chip to process, and must return a value of
NO_POLL_CHAR.
This allows kgdboc to process multiple polled devices while kgdb is
active.
Signed-off-by: Jason Wessel <[email protected]>
Acked-by: Paul Mundt <[email protected]>
---
drivers/serial/sh-sci.c | 6 +++++-
1 files changed, 5 insertions(+), 1 deletions(-)
diff --git a/drivers/serial/sh-sci.c b/drivers/serial/sh-sci.c
index 42f3333..2cf3b6d 100644
--- a/drivers/serial/sh-sci.c
+++ b/drivers/serial/sh-sci.c
@@ -126,7 +126,11 @@ static int sci_poll_get_char(struct uart_port *port)
handle_error(port);
continue;
}
- } while (!(status & SCxSR_RDxF(port)));
+ break;
+ } while (1);
+
+ if (!(status & SCxSR_RDxF(port)))
+ return NO_POLL_CHAR;
c = sci_in(port, SCxRDR);
--
1.6.4.rc1
Move kgdb.c in preparation for to separate the gdbstub from the debug
core and exception handling.n
CC: Ingo Molnar <[email protected]>
Signed-off-by: Jason Wessel <[email protected]>
---
kernel/Makefile | 2 +-
kernel/debug/Makefile | 5 +++++
kernel/{kgdb.c => debug/debug_core.c} | 0
3 files changed, 6 insertions(+), 1 deletions(-)
create mode 100644 kernel/debug/Makefile
rename kernel/{kgdb.c => debug/debug_core.c} (100%)
diff --git a/kernel/Makefile b/kernel/Makefile
index 864ff75..a18b039 100644
--- a/kernel/Makefile
+++ b/kernel/Makefile
@@ -74,7 +74,7 @@ obj-$(CONFIG_AUDITSYSCALL) += auditsc.o
obj-$(CONFIG_GCOV_KERNEL) += gcov/
obj-$(CONFIG_AUDIT_TREE) += audit_tree.o
obj-$(CONFIG_KPROBES) += kprobes.o
-obj-$(CONFIG_KGDB) += kgdb.o
+obj-$(CONFIG_KGDB) += debug/
obj-$(CONFIG_DETECT_SOFTLOCKUP) += softlockup.o
obj-$(CONFIG_DETECT_HUNG_TASK) += hung_task.o
obj-$(CONFIG_GENERIC_HARDIRQS) += irq/
diff --git a/kernel/debug/Makefile b/kernel/debug/Makefile
new file mode 100644
index 0000000..5287fc8
--- /dev/null
+++ b/kernel/debug/Makefile
@@ -0,0 +1,5 @@
+#
+# Makefile for the linux kernel debugger
+#
+
+obj-$(CONFIG_KGDB) += debug_core.o
\ No newline at end of file
diff --git a/kernel/kgdb.c b/kernel/debug/debug_core.c
similarity index 100%
rename from kernel/kgdb.c
rename to kernel/debug/debug_core.c
--
1.6.4.rc1
On Fri, 12 Feb 2010 16:35:37 -0600 Jason Wessel <[email protected]> wrote:
> @@ -609,6 +610,14 @@ asmlinkage int printk(const char *fmt, ...)
> va_list args;
> int r;
>
> +#ifdef CONFIG_KGDB_KDB
> + if (unlikely(kdb_trap_printk)) {
> + va_start(args, fmt);
> + r = vkdb_printf(fmt, args);
> + va_end(args);
> + return r;
> + }
> +#endif
> va_start(args, fmt);
> r = vprintk(fmt, args);
> va_end(args);
this?
--- a/kernel/printk.c~a
+++ a/kernel/printk.c
@@ -594,7 +594,11 @@ asmlinkage int printk(const char *fmt, .
int r;
va_start(args, fmt);
+#ifdef CONFIG_KGDB_KDB
+ r = vkdb_printf(fmt, args);
+#else
r = vprintk(fmt, args);
+#endif
va_end(args);
return r;
_
On Fri, 12 Feb 2010 16:35:41 -0600 Jason Wessel <[email protected]> wrote:
> It is highly desirable to trap into kdb on panic,
s/kdb/kgdb/
But is it necessary to do it in-kernel? Why not just put a
breakpoint at panic(), add that to your .gdbinit?
> diff --git a/kernel/panic.c b/kernel/panic.c
> index c787333..428d15b 100644
> --- a/kernel/panic.c
> +++ b/kernel/panic.c
> @@ -66,10 +66,13 @@ NORET_TYPE void panic(const char * fmt, ...)
> */
> preempt_disable();
>
> - bust_spinlocks(1);
> va_start(args, fmt);
> vsnprintf(buf, sizeof(buf), fmt, args);
> va_end(args);
> +
> + atomic_notifier_call_chain(&panic_notifier_list, 0, buf);
> +
> + bust_spinlocks(1);
> printk(KERN_EMERG "Kernel panic - not syncing: %s\n",buf);
> #ifdef CONFIG_DEBUG_BUGVERBOSE
> dump_stack();
> @@ -91,8 +94,6 @@ NORET_TYPE void panic(const char * fmt, ...)
> */
> smp_send_stop();
>
> - atomic_notifier_call_chain(&panic_notifier_list, 0, buf);
> -
> bust_spinlocks(0);
>
> if (!panic_blink)
So the notifier call now happens before all the printks and the kexec
and kmsg_dump handling. What effect does this have upon the code which
implements kexec and kmsg_dump?
Andrew Morton wrote:
> On Fri, 12 Feb 2010 16:35:37 -0600 Jason Wessel <[email protected]> wrote:
>
>
>> @@ -609,6 +610,14 @@ asmlinkage int printk(const char *fmt, ...)
>> va_list args;
>> int r;
>>
>> +#ifdef CONFIG_KGDB_KDB
>> + if (unlikely(kdb_trap_printk)) {
>> + va_start(args, fmt);
>> + r = vkdb_printf(fmt, args);
>> + va_end(args);
>> + return r;
>> + }
>> +#endif
>> va_start(args, fmt);
>> r = vprintk(fmt, args);
>> va_end(args);
>>
>
> this?
>
> --- a/kernel/printk.c~a
> +++ a/kernel/printk.c
> @@ -594,7 +594,11 @@ asmlinkage int printk(const char *fmt, .
> int r;
>
> va_start(args, fmt);
> +#ifdef CONFIG_KGDB_KDB
> + r = vkdb_printf(fmt, args);
> +#else
> r = vprintk(fmt, args);
> +#endif
> va_end(args);
>
> return r;
> _
>
If you feel that would be cleaner, I can make that change, but I would
also have to call vprintk() from vkdb_printf when ever kdb_trap_printk
is not set. Do I understand your recommendation correctly?
The only time the kdb_trap_printk is ever set comes when we are inside
the kernel debug shell, and do something like "bt". It serves the
purpose of letting folks run the stack dumper and a few other
functions. We want to trap printk's in this instance because we don't
want them in the log them and also for the case when your debug
connection is not actually a system log console, or if you run the
command via gdb (example is "monitor bt").
Thanks,
Jason.
Andrew Morton wrote:
> On Fri, 12 Feb 2010 16:35:41 -0600 Jason Wessel <[email protected]> wrote:
>
>
>> It is highly desirable to trap into kdb on panic,
>>
>
> s/kdb/kgdb/
>
> But is it necessary to do it in-kernel? Why not just put a
> breakpoint at panic(), add that to your .gdbinit?
>
>
If you don't mind consuming a breakpoint that is ok. Certainly there is
also the possibility of making this configurable as well.
For now I will omit this patch from any kind of future pull request
while we hash out what makes sense.
>> diff --git a/kernel/panic.c b/kernel/panic.c
>> index c787333..428d15b 100644
>> --- a/kernel/panic.c
>> +++ b/kernel/panic.c
>> @@ -66,10 +66,13 @@ NORET_TYPE void panic(const char * fmt, ...)
>> */
>> preempt_disable();
>>
>> - bust_spinlocks(1);
>> va_start(args, fmt);
>> vsnprintf(buf, sizeof(buf), fmt, args);
>> va_end(args);
>> +
>> + atomic_notifier_call_chain(&panic_notifier_list, 0, buf);
>> +
>> + bust_spinlocks(1);
>> printk(KERN_EMERG "Kernel panic - not syncing: %s\n",buf);
>> #ifdef CONFIG_DEBUG_BUGVERBOSE
>> dump_stack();
>> @@ -91,8 +94,6 @@ NORET_TYPE void panic(const char * fmt, ...)
>> */
>> smp_send_stop();
>>
>> - atomic_notifier_call_chain(&panic_notifier_list, 0, buf);
>> -
>> bust_spinlocks(0);
>>
>> if (!panic_blink)
>>
>
> So the notifier call now happens before all the printks and the kexec
> and kmsg_dump handling. What effect does this have upon the code which
> implements kexec and kmsg_dump?
>
>
I certainly don't want to break kexec or alter any behavior, does that
mean kgdb / kdb should hook the kexec for notification?
I think ideally it is a end user's preference as to if they want in via
kexec or the kernel debugger. Calling the smp_send_stop() prior to the
notifier was a death sentence for the kernel debugger.
Perhaps I can move the notifier before smp_send_stop()?
Jason.
On Fri, Feb 12, 2010 at 17:35, Jason Wessel wrote:
> The new debug core api requires all architectures that use to debug
> core to implement a function to set the program counter.
>
> CC: Mike Frysinger <[email protected]>
sorry, thought i Acked-by this already ...
-mike
On Fri, 12 Feb 2010 17:20:22 -0600 Jason Wessel <[email protected]> wrote:
> Andrew Morton wrote:
> > On Fri, 12 Feb 2010 16:35:37 -0600 Jason Wessel <[email protected]> wrote:
> >
> >
> >> @@ -609,6 +610,14 @@ asmlinkage int printk(const char *fmt, ...)
> >> va_list args;
> >> int r;
> >>
> >> +#ifdef CONFIG_KGDB_KDB
> >> + if (unlikely(kdb_trap_printk)) {
> >> + va_start(args, fmt);
> >> + r = vkdb_printf(fmt, args);
> >> + va_end(args);
> >> + return r;
> >> + }
> >> +#endif
> >> va_start(args, fmt);
> >> r = vprintk(fmt, args);
> >> va_end(args);
> >>
> >
> > this?
> >
> > --- a/kernel/printk.c~a
> > +++ a/kernel/printk.c
> > @@ -594,7 +594,11 @@ asmlinkage int printk(const char *fmt, .
> > int r;
> >
> > va_start(args, fmt);
> > +#ifdef CONFIG_KGDB_KDB
> > + r = vkdb_printf(fmt, args);
> > +#else
> > r = vprintk(fmt, args);
> > +#endif
> > va_end(args);
> >
> > return r;
> > _
> >
>
> If you feel that would be cleaner, I can make that change, but I would
> also have to call vprintk() from vkdb_printf when ever kdb_trap_printk
> is not set. Do I understand your recommendation correctly?
>
err, no, I must have been in a drunken stupor. Was trying to find
a way to avoid the code duplication.
On Fri, 12 Feb 2010 17:43:39 -0600 Jason Wessel <[email protected]> wrote:
> >> printk(KERN_EMERG "Kernel panic - not syncing: %s\n",buf);
> >> #ifdef CONFIG_DEBUG_BUGVERBOSE
> >> dump_stack();
> >> @@ -91,8 +94,6 @@ NORET_TYPE void panic(const char * fmt, ...)
> >> */
> >> smp_send_stop();
> >>
> >> - atomic_notifier_call_chain(&panic_notifier_list, 0, buf);
> >> -
> >> bust_spinlocks(0);
> >>
> >> if (!panic_blink)
> >>
> >
> > So the notifier call now happens before all the printks and the kexec
> > and kmsg_dump handling. What effect does this have upon the code which
> > implements kexec and kmsg_dump?
> >
> >
>
> I certainly don't want to break kexec or alter any behavior, does that
> mean kgdb / kdb should hook the kexec for notification?
>
> I think ideally it is a end user's preference as to if they want in via
> kexec or the kernel debugger. Calling the smp_send_stop() prior to the
> notifier was a death sentence for the kernel debugger.
>
> Perhaps I can move the notifier before smp_send_stop()?
Well. My question can be simplified to "does this break existing code"?
On 02/12/10 14:35, Jason Wessel wrote:
> Update the kgdb docs to reflect the new directory structure and API.
>
> Merge in the kdb shell information.
>
> Signed-off-by: Jason Wessel <[email protected]>
> ---
> diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
> index f3be1e6..af8b8e8 100644
> --- a/Documentation/kernel-parameters.txt
> +++ b/Documentation/kernel-parameters.txt
> @@ -1102,13 +1103,17 @@ and is between 256 and 4096 characters. It is defined in the file
> use the HighMem zone if it exists, and the Normal
> zone if it does not.
>
> - kgdboc= [HW] kgdb over consoles.
> + kgdboc= [KGDB,HW] kgdb over consoles.
> Requires a tty driver that supports console polling,
> or a supported polling keyboard driver (non-usb).
> Serial only format: <serial_device>[,baud]
> keyboard only format: kbd
> keyboard and serial format: kbd,<serial_device>[,baud]
>
> + kgdbwait [KGDB] Stop kernel execution and enter the
> + kernel debugger at the earliest opportunity if
> + kgdb was configured on as a boot argument.
What does "kgdb was configured on as a boot argument" mean?
> +
> kmac= [MIPS] korina ethernet MAC address.
> Configure the RouterBoard 532 series on-chip
> Ethernet adapter MAC address.
--
~Randy
On 02/12/10 14:35, Jason Wessel wrote:
> Update the kgdb docs to reflect the new directory structure and API.
>
> Merge in the kdb shell information.
See comments/corrections inline below.
> Signed-off-by: Jason Wessel <[email protected]>
> ---
> Documentation/DocBook/kgdb.tmpl | 640 ++++++++++++++++++++++++++---------
> Documentation/kernel-parameters.txt | 7 +-
> include/linux/kgdb.h | 12 +-
> lib/Kconfig.kgdb | 2 +-
> 4 files changed, 495 insertions(+), 166 deletions(-)
>
> diff --git a/Documentation/DocBook/kgdb.tmpl b/Documentation/DocBook/kgdb.tmpl
> index 5cff41a..cc81879 100644
> --- a/Documentation/DocBook/kgdb.tmpl
> +++ b/Documentation/DocBook/kgdb.tmpl
> @@ -4,7 +4,7 @@
>
> <book id="kgdbOnLinux">
> <bookinfo>
> - <title>Using kgdb and the kgdb Internals</title>
> + <title>Using kgdb, kdb and the kernel debugger internals</title>
>
> <authorgroup>
> <author>
> @@ -17,33 +17,8 @@
> </affiliation>
> </author>
> </authorgroup>
> -
> - <authorgroup>
> - <author>
> - <firstname>Tom</firstname>
> - <surname>Rini</surname>
> - <affiliation>
> - <address>
> - <email>[email protected]</email>
> - </address>
> - </affiliation>
> - </author>
> - </authorgroup>
> -
> - <authorgroup>
> - <author>
> - <firstname>Amit S.</firstname>
> - <surname>Kale</surname>
> - <affiliation>
> - <address>
> - <email>[email protected]</email>
> - </address>
> - </affiliation>
> - </author>
> - </authorgroup>
> -
> <copyright>
> - <year>2008</year>
> + <year>2008,2010</year>
> <holder>Wind River Systems, Inc.</holder>
> </copyright>
> <copyright>
> @@ -69,41 +44,76 @@
> <chapter id="Introduction">
> <title>Introduction</title>
> <para>
> - kgdb is a source level debugger for linux kernel. It is used along
> - with gdb to debug a linux kernel. The expectation is that gdb can
> - be used to "break in" to the kernel to inspect memory, variables
> - and look through call stack information similar to what an
> - application developer would use gdb for. It is possible to place
> - breakpoints in kernel code and perform some limited execution
> - stepping.
> + The kernel has two different debugger front ends (kdb and kgdb)
> + which interface to the debug core. It is possible to use either
> + of the debugger front ends and dynamically transition between them
> + if you configure the kernel properly at compile and runtime.
> + </para>
> + <para>
> + Kdb is simplistic shell style interface which you can use on a
shell-style
> + system console with a keyboard or serial console. You can use it
> + to inspect memory, registers, process lists, dmesg, and even set
> + breakpoints to stop in a certain location. Kdb is not a source
> + level debugger, although you can set breakpoints and execute some
> + basic kernel run control. Kdb is mainly aimed at doing some
> + analysis to aid in development or diagnosing kernel problems. You
> + can access some symbols by name in kernel builtins or in kernel
> + modules if the code was built
> + with <symbol>CONFIG_KALLSYMS</symbol>.
> </para>
> <para>
> - Two machines are required for using kgdb. One of these machines is a
> - development machine and the other is a test machine. The kernel
> - to be debugged runs on the test machine. The development machine
> - runs an instance of gdb against the vmlinux file which contains
> - the symbols (not boot image such as bzImage, zImage, uImage...).
> - In gdb the developer specifies the connection parameters and
> - connects to kgdb. The type of connection a developer makes with
> - gdb depends on the availability of kgdb I/O modules compiled as
> - builtin's or kernel modules in the test machine's kernel.
> + Kgdb was intended to be used as a source level debugger for the
is (?)
> + linux kernel. It is used along with gdb to debug a linux kernel.
Linux Linux
> + The expectation is that gdb can be used to "break in" to the
> + kernel to inspect memory, variables and look through call stack
> + information similar to the way an application developer would use
> + gdb to debug an application. It is possible to place breakpoints
> + in kernel code and perform some limited execution stepping.
> + </para>
> + <para>
> + Two machines are required for using kgdb. One of these machines is
> + a development machine and the other is the target machine. The
> + kernel to be debugged runs on the target machine. The development
> + machine runs an instance of gdb against the vmlinux file which
> + contains the symbols (not boot image such as bzImage, zImage,
> + uImage...). In gdb the developer specifies the connection
> + parameters and connects to kgdb. The type of connection a
> + developer makes with gdb depends on the availability of kgdb I/O
> + modules compiled as builtins or kernel modules in the test
> + machine's kernel.
> </para>
> </chapter>
> <chapter id="CompilingAKernel">
> - <title>Compiling a kernel</title>
> + <title>Compiling a kernel</title>
> + <para>
> + <itemizedlist>
> + <listitem><para>In order to enable compilation of kdb, you must first enable kgdb.</para></listitem>
> + <listitem><para>The kgdb test compile options are described in the kgdb test suite chapter.</para></listitem>
> + </itemizedlist>
> + </para>
> + <sect1 id="CompileKGDB">
> + <title>Kernel config options for kgdb</title>
> <para>
> To enable <symbol>CONFIG_KGDB</symbol> you should first turn on
> "Prompt for development and/or incomplete code/drivers"
> (CONFIG_EXPERIMENTAL) in "General setup", then under the
> - "Kernel debugging" select "KGDB: kernel debugging with remote gdb".
> + "Kernel debugging" select "KGDB: kernel debugger".
> + </para>
> + <para>
> + While it is not a hard requirement that you have symbols in your
> + vmlinux file, gdb tends not to be very useful without the symbolic
> + data, so you will want to turn
> + on <symbol>CONFIG_DEBUG_INFO</symbol> which is called "Compile the
> + kernel with debug info" in the menu config.
s/menu config/config menu/ throughout.
> </para>
> <para>
> It is advised, but not required that you turn on the
> - CONFIG_FRAME_POINTER kernel option. This option inserts code to
> - into the compiled executable which saves the frame information in
> - registers or on the stack at different points which will allow a
> - debugger such as gdb to more accurately construct stack back traces
> - while debugging the kernel.
> + <symbol>CONFIG_FRAME_POINTER</symbol> kernel option which is called "Compile the
> + kernel with frame pointers" in the menu config. This option
> + inserts code to into the compiled executable which saves the frame
> + information in registers or on the stack at different points which
> + will allow a debugger such as gdb to more accurately construct
allows a debugger
> + stack back traces while debugging the kernel.
> </para>
> <para>
> If the architecture that you are using supports the kernel option
> @@ -116,38 +126,146 @@
> this option.
> </para>
> <para>
> - Next you should choose one of more I/O drivers to interconnect debugging
> - host and debugged target. Early boot debugging requires a KGDB
> - I/O driver that supports early debugging and the driver must be
> - built into the kernel directly. Kgdb I/O driver configuration
> - takes place via kernel or module parameters, see following
> - chapter.
> + Next you should choose one of more I/O drivers to interconnect
> + debugging host and debugged target. Early boot debugging requires
> + a KGDB I/O driver that supports early debugging and the driver
> + must be built into the kernel directly. Kgdb I/O driver
> + configuration takes place via kernel or module parameters which
> + you can learn more about in the in the section that describes the
> + kernel parameter kgdboc.
parameter "kgdboc".
> </para>
> - <para>
> - The kgdb test compile options are described in the kgdb test suite chapter.
> + <para>Here is an example set of .config to enable kgdb:
of .config symbols to enable or disable for kgdb:
> + <itemizedlist>
> + <listitem><para># CONFIG_DEBUG_RODATA is not set</para></listitem>
> + <listitem><para>CONFIG_FRAME_POINTER=y</para></listitem>
> + <listitem><para>CONFIG_KGDB=y</para></listitem>
> + <listitem><para>CONFIG_KGDB_SERIAL_CONSOLE=y</para></listitem>
> + </itemizedlist>
> </para>
> -
> + </sect1>
> + <sect1 id="CompileKDB">
> + <title>Kernel config options for kdb</title>
> + <para>Kdb is quite a bit more complex than the simple gdbstub
> + sitting on top of the kernel's debug core. Kdb must implement an
implement a
> + shell, and also adds some helper functions in other parts of the
> + kernel, responsible for printing out interesting data such as what
> + you would see if you ran "lsmod", or "ps". In order to build kdb
> + into the kernel you follow the same steps as you would for kgdb.
> + </para>
> + <para>The main config option for kdb
> + is <symbol>CONFIG_KGDB_KDB</symbol> which is called "KGDB_KDB:
> + include kdb frontend for kgdb" in the menu config. In theory you
> + would have already also selected an I/O driver such as the
> + CONFIG_KGDB_SERIAL_CONSOLE interface if you plan on using kdb on a
> + serial port, when you were configuring kgdb.
> + </para>
> + <para>If you want to use a ps/2 style keyboard with kdb you would
PS/2-style keyboard with kdb, you would
> + select CONFIG_KDB_KEYBOARD which is called "KGDB_KDB: keyboard as
> + input device" in the menu config. This option really does do
does not do (??)
> + anything for the gdb interface to kgdb it only works with kdb,
to kgdb; it only
> + because it is not really meaningful to manually type in gdb
> + remote protocol commands.
> + </para>
> + <para>Here is an example set of .config to enable kdb:
set of .config symbols to enable/disable for kdb:
> + <itemizedlist>
> + <listitem><para># CONFIG_DEBUG_RODATA is not set</para></listitem>
> + <listitem><para>CONFIG_FRAME_POINTER=y</para></listitem>
> + <listitem><para>CONFIG_KGDB=y</para></listitem>
> + <listitem><para>CONFIG_KGDB_SERIAL_CONSOLE=y</para></listitem>
> + <listitem><para>CONFIG_KGDB_KDB=y</para></listitem>
> + <listitem><para>CONFIG_KDB_KEYBOARD=y</para></listitem>
> + </itemizedlist>
> + </para>
> + </sect1>
> </chapter>
> - <chapter id="EnableKGDB">
> - <title>Enable kgdb for debugging</title>
> - <para>
> - In order to use kgdb you must activate it by passing configuration
> - information to one of the kgdb I/O drivers. If you do not pass any
> - configuration information kgdb will not do anything at all. Kgdb
> - will only actively hook up to the kernel trap hooks if a kgdb I/O
> - driver is loaded and configured. If you unconfigure a kgdb I/O
> - driver, kgdb will unregister all the kernel hook points.
> + <chapter id="kgdbKernelArgs">
> + <title>Kernel Debugger Boot Arguments</title>
> + <para>This section will describe the various runtime kernel
section describes
> + parameters that affect the configuration of the kernel debugger.
> + The following chapter will cover using kdb and kgdb as well as
chapter covers
> + provide some examples of the configuration parameters.</para>
provides
> + <sect1 id="kgdboc">
> + <title>Kernel parameter: kgdboc</title>
> + <para>The kgdboc driver was originally an abbreviation meant to
> + stand for "kgdb over console". Today it is the primary mechanism
> + to configure how to communicate from gdb to kgdb as well as the
> + devices you want to use to interact with the kdb shell.
> </para>
> - <para>
> - All drivers can be reconfigured at run time, if
> - <symbol>CONFIG_SYSFS</symbol> and <symbol>CONFIG_MODULES</symbol>
> - are enabled, by echo'ing a new config string to
> - <constant>/sys/module/<driver>/parameter/<option></constant>.
> - The driver can be unconfigured by passing an empty string. You cannot
> - change the configuration while the debugger is attached. Make sure
> - to detach the debugger with the <constant>detach</constant> command
> - prior to trying unconfigure a kgdb I/O driver.
> + <para>For kgdb/gdb, kgdboc is designed to work with a single serial
> + port. It was meant to cover the circumstance where you wanted to
is meant you want to
> + use a serial console as your primary console as well as using it to
> + perform kernel debugging. It is also possible to use kgdb on a
> + serial port which is not designated as a system console. Kgdboc
> + may be configured as a kernel built-in or a kernel module. You can
kernel loadable module.
> + only make use of <constant>kgdbwait</constant> and early debugging
> + if you build kgdboc into the kernel as a builtins.
as a builtin.
> + </para>
> + <sect2 id="kgdbocArgs">
> + <title>kgdboc arguments</title>
> + <para>Usage: <constant>kgdboc=[kbd][[,]serial_device][,baud]</constant></para>
> + <para>You can configure kgdboc to use the keyboard, and or a serial device
> + depending on if you are using kdb and or kgdb, in one of the
> + following scenarios.
> + <orderedlist>
> + <listitem><para>kdb and kgdb over only a serial port</para>
> + <para><constant>kgdboc=<serial_device>[,baud]</constant></para>
> + <para>Example: <constant>kgdboc=ttyS0,115200</constant></para>
> + </listitem>
> + <listitem><para>kdb and kgdb with keyboard and a serial port</para>
> + <para><constant>kgdboc=kbd,<serial_device>[,baud]</constant></para>
> + <para>Example: <constant>kgdboc=kbd,ttyS0,115200</constant></para>
> + </listitem>
> + <listitem><para>kdb with a keyboard</para>
> + <para><constant>kgdboc=kbd</constant></para>
> + </listitem>
> + </orderedlist>
> + </para>
> + <para>You can configure kgdboc via sysfs or a module or kernel boot line
> + parameter depending on if you build with CONFIG_KGDBOC as a module
as a loadable module
> + or built-in.
> + <orderedlist>
> + <listitem><para>From the module load or build-in</para>
built-in (?)
> + <para>As a kernel argument: <constant>kgdboc=<tty-device>,[baud]</constant></para>
> + <para>As a kernel module: <constant>modprobe kgdboc kgdboc=<tty-device>,[baud]</constant></para>
As a kernel loadable module:
> + <para>Here are two examples, the first for an x86 target using the
> + first serial port and the second for the ARM Versatile AB using the
> + second serial port.
> + <orderedlist>
> + <listitem><para><constant>kgdboc=ttyS0,115200</constant></para></listitem>
> + <listitem><para><constant>kgdboc=ttyAMA1,115200</constant></para></listitem>
> + </orderedlist>
> + </para>
> + </listitem>
> + <listitem><para>From sysfs</para>
> + <para><constant>echo ttyS0 > /sys/module/kgdboc/parameters/kgdboc</constant></para>
> + <para>NOTE: You do not need to specify the baud if you are
> + configuring the console on tty which is already configured or
> + open.</para>
> + </listitem>
> + </orderedlist>
> + </para>
> + <para>NOTE: Kgdboc does not support interrupting the target via the
> + gdb remote protocol. You must manually send a sysrq-g unless you
> + have a proxy that splits console output to a terminal program.
> + A console proxy has a separate tcp port for the debugger and a separate
TCP
> + tcp port for the "human" console. The proxy can take care of sending
TCP
> + the sysrq-g for you.
> </para>
> + <para>When using kgdboc with no debugger proxy, you can end up
> + connecting the debugger for one of two entry points. If an
at (?)
> + exception occurs after you have loaded kgdboc a message should
kgdboc,
> + print on the console stating it is waiting for the debugger. In
> + case you disconnect your terminal program and then connect the
this case,
> + debugger in its place. If you want to interrupt the target system
> + and forcibly enter a debug session you have to issue a Sysrq
> + sequence and then type the letter <constant>g</constant>. Then
> + you disconnect the terminal session and connect gdb. Your options
> + if you don't like this are to hack gdb to send the sysrq-g for you
> + as well as on the initial connect, or to use a debugger proxy that
> + allows an unmodified gdb to do the debugging.
> + </para>
> + </sect2>
> + </sect1>
> <sect1 id="kgdbwait">
> <title>Kernel parameter: kgdbwait</title>
> <para>
> @@ -166,99 +284,193 @@
> kgdb I/O driver as a kernel module kgdbwait will not do anything.
> </para>
> </sect1>
> - <sect1 id="kgdboc">
> - <title>Kernel parameter: kgdboc</title>
> - <para>
> - The kgdboc driver was originally an abbreviation meant to stand for
> - "kgdb over console". Kgdboc is designed to work with a single
> - serial port. It was meant to cover the circumstance
> - where you wanted to use a serial console as your primary console as
> - well as using it to perform kernel debugging. Of course you can
> - also use kgdboc without assigning a console to the same port.
> + <sect1 id="kgdbcon">
> + <title>Kernel parameter: kgdbcon</title>
> + <para> The kgdbcon feature allows you to see printk() messages
> + inside gdb while gdb is connected to the kernel. Kdb does not make
> + use of the kgdbcon feature.
> + </para>
> + <para>Kgdb supports using the gdb serial protocol to send console
> + messages to the debugger when the debugger is connected and running.
> + There are two ways to activate this feature.
> + <orderedlist>
> + <listitem><para>Activate with the kernel command line option:</para>
> + <para><constant>kgdbcon</constant></para>
> + </listitem>
> + <listitem><para>Use sysfs before configuring an io driver</para>
I/O
> + <para>
> + <constant>echo 1 > /sys/module/kgdb/parameters/kgdb_use_con</constant>
> + </para>
> + <para>
> + NOTE: If you do this after you configure the kgdb I/O driver, the
> + setting will not take effect until the next point the I/O is
> + reconfigured.
> + </para>
> + </listitem>
> + </orderedlist>
> + <para>IMPORTANT NOTE: You cannot use kgdboc + kgdbcon on a tty that is an
> + active system console IE: console=ttyS0,115200 kgdboc=ttyS0 kgdbcon
console;
and then EG: or e.g.: would be better than IE: there.
> + </para>
> + <para>It is possible to use this option with kgdboc on a tty that is not a system console.
> + </para>
> </para>
> - <sect2 id="UsingKgdboc">
> - <title>Using kgdboc</title>
> - <para>
> - You can configure kgdboc via sysfs or a module or kernel boot line
> - parameter depending on if you build with CONFIG_KGDBOC as a module
> - or built-in.
> - <orderedlist>
> - <listitem><para>From the module load or build-in</para>
> - <para><constant>kgdboc=<tty-device>,[baud]</constant></para>
> + </sect1>
> + </chapter>
> + <chapter id="usingKDB">
> + <title>Using kdb</title>
> <para>
> - The example here would be if your console port was typically ttyS0, you would use something like <constant>kgdboc=ttyS0,115200</constant> or on the ARM Versatile AB you would likely use <constant>kgdboc=ttyAMA0,115200</constant>
> + </para>
> + <sect1 id="quickKDBserial">
> + <title>Quick start for kdb on a serial port</title>
> + <para>This is a quick example of how to use kdb</para>
End sentence with .
> + <para><orderedlist>
> + <listitem><para>Boot kernel with arguments:
> + <itemizedlist>
> + <listitem><para><constant>console=ttyS0,115200 kgdboc=ttyS0,115200</constant></para></listitem>
> + </itemizedlist></para>
> + <para>OR</para>
> + <para>Configure kgdboc after the kernel booted assuming your are using a serial port console:
booted; assuming you are
> + <itemizedlist>
> + <listitem><para><constant>echo ttyS0 > /sys/module/kgdboc/parameters/kgdboc</constant></para></listitem>
> + </itemizedlist>
> </para>
> </listitem>
> - <listitem><para>From sysfs</para>
> - <para><constant>echo ttyS0 > /sys/module/kgdboc/parameters/kgdboc</constant></para>
> + <listitem><para>Enter the kernel debugger manually or by waiting for an opps or fault. There are several ways you can enter the kernel debugger manually, all involve using the sysrq-g, which means you have to enabled CONFIG_MAGIC_SYSRQ=y in your kernel config</para>
s/opps/oops/
s/manually,/manually;/
s/have to enabled/must have enabled/
End last sentence with .
> + <itemizedlist>
> + <listitem><para>When logged in as root or with a super user session you can run:</para>
> + <para><constant>echo g > /proc/sysrq-trigger</constant></para></listitem>
> + <listitem><para>Example using minicom 2.2</para>
> + <para>Press: <constant>Control-a</constant></para>
> + <para>Press: <constant>f</constant></para>
> + <para>Press: <constant>g</constant></para>
> </listitem>
> - </orderedlist>
> - </para>
> - <para>
> - NOTE: Kgdboc does not support interrupting the target via the
> - gdb remote protocol. You must manually send a sysrq-g unless you
> - have a proxy that splits console output to a terminal problem and
> - has a separate port for the debugger to connect to that sends the
> - sysrq-g for you.
> + <listitem><para>When you have telneted to a terminal server that supports sending a remote break</para>
> + <para>Press: <constant>Control-]</constant></para>
> + <para>Type in:<constant>send break</constant></para>
> + <para>Press: <constant>Enter</constant></para>
> + <para>Press: <constant>g</constant></para>
> + </listitem>
> + </itemizedlist>
> + </listitem>
> + <listitem><para>From the kdb prompt you can run the "help" command to see a complete list of the commands that are available.</para>
> + <para>Some useful commands in kdb include:
> + <itemizedlist>
> + <listitem><para>lsmod -- Shows where kernel modules are loaded</para></listitem>
> + <listitem><para>ps -- Displays only the active processes</para></listitem>
> + <listitem><para>ps A -- Shows all the processes</para></listitem>
> + <listitem><para>summary -- Shows kernel version info and memory usage</para></listitem>
> + <listitem><para>bt -- Get a backtrace of the current process using dump_stack()</para></listitem>
> + <listitem><para>dmesg -- View the kernel syslog buffer</para></listitem>
> + <listitem><para>go -- Continue the system</para></listitem>
> + </itemizedlist>
> </para>
> - <para>When using kgdboc with no debugger proxy, you can end up
> - connecting the debugger for one of two entry points. If an
> - exception occurs after you have loaded kgdboc a message should print
> - on the console stating it is waiting for the debugger. In case you
> - disconnect your terminal program and then connect the debugger in
> - its place. If you want to interrupt the target system and forcibly
> - enter a debug session you have to issue a Sysrq sequence and then
> - type the letter <constant>g</constant>. Then you disconnect the
> - terminal session and connect gdb. Your options if you don't like
> - this are to hack gdb to send the sysrq-g for you as well as on the
> - initial connect, or to use a debugger proxy that allows an
> - unmodified gdb to do the debugging.
> + </listitem>
> + <listitem>
> + <para>When you are done using kdb you need to consider rebooting the system or using the "go" command to resuming normal kernel execution. If you have paused the kernel for a lengthy period of time, applications that rely on timely networking or anything to do with real wall clock time could get adversely affected, so you should take this into consideration when using the kernel debugger.</para>
s/could get/could be/
> + </listitem>
> + </orderedlist></para>
> + </sect1>
> + <sect1 id="quickKDBkeyboard">
> + <title>Quick start for kdb using a keyboard connected console</title>
> + <para>This is a quick example of how to use kdb with a keyboard</para>
End sentence with .
> + <para><orderedlist>
> + <listitem><para>Boot kernel with arguments:
> + <itemizedlist>
> + <listitem><para><constant>kgdboc=kbd</constant></para></listitem>
> + </itemizedlist></para>
> + <para>OR</para>
> + <para>Configure kgdboc after the kernel booted:
> + <itemizedlist>
> + <listitem><para><constant>echo kbd > /sys/module/kgdboc/parameters/kgdboc</constant></para></listitem>
> + </itemizedlist>
> </para>
> - </sect2>
> + </listitem>
> + <listitem><para>Enter the kernel debugger manually or by waiting for an opps or fault. There are several ways you can enter the kernel debugger manually, all involve using the sysrq-g, which means you have to enabled CONFIG_MAGIC_SYSRQ=y in your kernel config</para>
s/opps/oops/
s/manually,/manually;/
s/have to enabled/must have enabled/
End last sentence with .
> + <itemizedlist>
> + <listitem><para>When logged in as root or with a super user session you can run:</para>
> + <para><constant>echo g > /proc/sysrq-trigger</constant></para></listitem>
> + <listitem><para>Example using a laptop keyboard</para>
> + <para>Press and hold down: <constant>Alt</constant></para>
> + <para>Press and hold down: <constant>Fn</constant></para>
> + <para>Press and release the key with the label: <constant>SysRq</constant></para>
> + <para>Release: <constant>Fn</constant></para>
> + <para>Press and release: <constant>g</constant></para>
> + <para>Release: <constant>Alt</constant></para>
> + </listitem>
> + <listitem><para>Example using a PS/2 101 keyboard</para>
101-key keyboard
> + <para>Press and hold down: <constant>Alt</constant></para>
> + <para>Press and release the key with the label: <constant>SysRq</constant></para>
> + <para>Press and release: <constant>g</constant></para>
> + <para>Release: <constant>Alt</constant></para>
> + </listitem>
> + </itemizedlist>
> + </listitem>
> + <listitem>
> + <para>Now type in a kdb command such as "help", "dmesg", "bt" or "go" to continue kernel execution.</para>
> + </listitem>
> + </orderedlist></para>
> </sect1>
> - <sect1 id="kgdbcon">
> - <title>Kernel parameter: kgdbcon</title>
> - <para>
> - Kgdb supports using the gdb serial protocol to send console messages
> - to the debugger when the debugger is connected and running. There
> - are two ways to activate this feature.
> + </chapter>
> + <chapter id="EnableKGDB">
> + <title>Using kgdb / gdb</title>
> + <para>In order to use kgdb you must activate it by passing
> + configuration information to one of the kgdb I/O drivers. If you
> + do not pass any configuration information kgdb will not do anything
> + at all. Kgdb will only actively hook up to the kernel trap hooks
> + if a kgdb I/O driver is loaded and configured. If you unconfigure
> + a kgdb I/O driver, kgdb will unregister all the kernel hook points.
> + </para>
> + <para> All kgdb I/O drivers can be reconfigured at run time, if
> + <symbol>CONFIG_SYSFS</symbol> and <symbol>CONFIG_MODULES</symbol>
> + are enabled, by echo'ing a new config string to
> + <constant>/sys/module/<driver>/parameter/<option></constant>.
> + The driver can be unconfigured by passing an empty string. You cannot
> + change the configuration while the debugger is attached. Make sure
> + to detach the debugger with the <constant>detach</constant> command
> + prior to trying unconfigure a kgdb I/O driver.
prior to trying to unconfigure
> + </para>
> + <sect1 id="ConnectingGDB">
> + <title>Connecting with gdb to a serial port</title>
> <orderedlist>
> - <listitem><para>Activate with the kernel command line option:</para>
> - <para><constant>kgdbcon</constant></para>
> + <listitem><para>Configure kgdboc</para>
> + <para>Boot kernel with arguments:
> + <itemizedlist>
> + <listitem><para><constant>kgdboc=ttyS0,115200</constant></para></listitem>
> + </itemizedlist></para>
> + <para>OR</para>
> + <para>Configure kgdboc after the kernel booted:
> + <itemizedlist>
> + <listitem><para><constant>echo ttyS0 > /sys/module/kgdboc/parameters/kgdboc</constant></para></listitem>
> + </itemizedlist></para>
> </listitem>
> - <listitem><para>Use sysfs before configuring an io driver</para>
> - <para>
> - <constant>echo 1 > /sys/module/kgdb/parameters/kgdb_use_con</constant>
> - </para>
> - <para>
> - NOTE: If you do this after you configure the kgdb I/O driver, the
> - setting will not take effect until the next point the I/O is
> - reconfigured.
> - </para>
> + <listitem>
> + <para>Stop kernel execution (break into the debugger)</para>
> + <para>In order to connect to gdb to the via kgdboc the kernel must
In order to connect to gdb via kgdboc, the kernel must
> + first be stopped. There are several ways to stop the kernel which
> + include using kgdbwait as a boot argument, via a sysrq-g, or running
> + the kernel until it takes an an exception where it waits for the
takes an exception
> + debugger to attach.
> + <itemizedlist>
> + <listitem><para>When logged in as root or with a super user session you can run:</para>
> + <para><constant>echo g > /proc/sysrq-trigger</constant></para></listitem>
> + <listitem><para>Example using minicom 2.2</para>
> + <para>Press: <constant>Control-a</constant></para>
> + <para>Press: <constant>f</constant></para>
> + <para>Press: <constant>g</constant></para>
> </listitem>
> - </orderedlist>
> - </para>
> - <para>
> - IMPORTANT NOTE: Using this option with kgdb over the console
> - (kgdboc) is not supported.
> + <listitem><para>When you have telneted to a terminal server that supports sending a remote break</para>
> + <para>Press: <constant>Control-]</constant></para>
> + <para>Type in:<constant>send break</constant></para>
> + <para>Press: <constant>Enter</constant></para>
> + <para>Press: <constant>g</constant></para>
> + </listitem>
> + </itemizedlist>
> </para>
> - </sect1>
> - </chapter>
> - <chapter id="ConnectingGDB">
> - <title>Connecting gdb</title>
> - <para>
> - If you are using kgdboc, you need to have used kgdbwait as a boot
> - argument, issued a sysrq-g, or the system you are going to debug
> - has already taken an exception and is waiting for the debugger to
> - attach before you can connect gdb.
> - </para>
> - <para>
> - If you are not using different kgdb I/O driver other than kgdboc,
> - you should be able to connect and the target will automatically
> - respond.
> - </para>
> + </listitem>
> + <listitem>
> + <para>Connect from from gdb</para>
> <para>
> - Example (using a serial port):
> + Example (using a directly connected port):
> </para>
> <programlisting>
> % gdb ./vmlinux
> @@ -283,6 +495,82 @@
> communications. You do this prior to issuing the <constant>target
> remote</constant> command by typing in: <constant>set debug remote 1</constant>
> </para>
> + </listitem>
> + </orderedlist>
> + <para>Remember if you continue in gdb, and need "break in" again,
need to
> + you need to issue an other sysrq. It is easy to create an simple
another sysrq-g. It is easy to create a simple
> + entry point by putting a breakpoint at <constant>sys_sync</constant>
> + and then you can run "sync" from a shell or script to break into the
> + debugger.</para>
> + </sect1>
> + </chapter>
> + <chapter id="switchKdbKgdb">
> + <title>kgdb and kdb interoperability</title>
> + <para>It is possible to transition between kdb and kgdb dynamically.
> + The debug core will remember which you used the last time and
> + automatically start in the same mode.</para>
> + <sect1>
> + <title>Switching between kdb and kgdb</title>
> + <sect2>
> + <title>Switching from kgdb to kdb</title>
> + <para>
> + There are two ways to switch from kgdb to kdb, you can use gdb to
to kdb:
> + issue a maintenance packet, or you can blindly type the command $3#33.
> + When ever kernel debugger stops in kgdb mode it will print the
Whenever
> + message <constant>KGDB or $3#33 for KDB</constant>. It is important
> + to note, that you have to type the sequence correctly in one pass.
to note that
> + You cannot type a backspace or delete because kgdb will interpret
> + that as part of the debug stream.
> + <orderedlist>
> + <listitem><para>Change from kgdb to kdb by blindly typing:</para>
> + <para><constant>$3#33</constant></para></listitem>
> + <listitem><para>Change from kgdb to kdb with gdb</para>
> + <para><constant>maintenance packet 3</constant></para>
> + <para>NOTE: Now you must kill gdb</para></listitem>
tell how to kill?
> + </orderedlist>
> + </para>
> + </sect2>
> + <sect2>
> + <title>Change from kdb to kgdb</title>
> + <para>There are two ways you can change from kdb to kgdb. You can
> + manually enter kgdb mode by issuing the kgdb command from the kdb
> + shell prompt, or you can connect gdb while the kdb shell prompt is
> + active. The kdb shell looks for the typical first commands that gdb
> + would issue with the gdb remote protocol and if it sees one of those
> + commands it automatically changes into kgdb mode.</para>
> + <orderedlist>
> + <listitem><para>From kdb issue the command:</para>
> + <para><constant>kgdb</constant></para>
> + <para>Now disconnect your terminal program and connect gdb in its place</para></listitem>
> + <listitem><para>At the kdb prompt, disconnect the terminal program and connect gdb in its place.</para></listitem>
> + </orderedlist>
> + </sect2>
> + </sect1>
> + <sect1>
> + <title>Running kdb commands from gdb</title>
> + <para>It is possible to run a limited set of kdb commands from gdb,
> + using the gdb monitor command. You don't want to execute any of the
> + run control or breakpoint operations, because it can disrupt the
> + state of the kernel debugger. You should be using gdb for
> + breakpoints and run control operations if you have gdb connected.
> + The more useful commands to run are things like lsmod, dmesg, ps or
> + possibly some of the memory information. To see all the kdb
information commands.
> + commands you can run <constant>monitor help</constant>.</para>
> + <para>Example:
> + <informalexample><programlisting>
> +(gdb) monitor ps
> +1 idle process (state I) and
> +27 sleeping system daemon (state M) processes suppressed,
> +use 'ps A' to see all.
> +Task Addr Pid Parent [*] cpu State Thread Command
> +
> +0xc78291d0 1 0 0 0 S 0xc7829404 init
> +0xc7954150 942 1 0 0 S 0xc7954384 dropbear
> +0xc78789c0 944 1 0 0 S 0xc7878bf4 sh
> +(gdb)
> + </programlisting></informalexample>
> + </para>
> + </sect1>
> </chapter>
> <chapter id="KGDBTestSuite">
> <title>kgdb Test Suite</title>
> @@ -309,22 +597,22 @@
> </para>
> </chapter>
> <chapter id="CommonBackEndReq">
> - <title>KGDB Internals</title>
> + <title>Kernel Debugger Internals</title>
> <sect1 id="kgdbArchitecture">
> <title>Architecture Specifics</title>
> <para>
> - Kgdb is organized into three basic components:
> + The kernel debugger is organized into a number of components:
> <orderedlist>
> - <listitem><para>kgdb core</para>
> + <listitem><para>The debug core</para>
> <para>
> - The kgdb core is found in kernel/kgdb.c. It contains:
> + The debug core is found in kernel/debugger/debug_core.c. It contains:
> <itemizedlist>
> - <listitem><para>All the logic to implement the gdb serial protocol</para></listitem>
> <listitem><para>A generic OS exception handler which includes sync'ing the processors into a stopped state on an multi cpu system.</para></listitem>
multi-CPU
> <listitem><para>The API to talk to the kgdb I/O drivers</para></listitem>
> <listitem><para>The API to make calls to the arch specific kgdb implementation</para></listitem>
arch-specific
> <listitem><para>The logic to perform safe memory reads and writes to memory while using the debugger</para></listitem>
> <listitem><para>A full implementation for software breakpoints unless overridden by the arch</para></listitem>
> + <listitem><para>The API to invoke either the kdb or kgdb frontend to the debug core.</para></listitem>
> </itemizedlist>
> </para>
> </listitem>
> @@ -347,9 +635,31 @@
> </itemizedlist>
> </para>
> </listitem>
> + <listitem><para>gdbstub frontend (aka kgdb)</para>
> + <para>The gdbstub is located in kernel/debug/gdbstub.c. It contains:</para>
> + <itemizedlist>
> + <listitem><para>All the logic to implement the gdb serial protocol</para></listitem>
> + </itemizedlist>
> + </listitem>
> + <listitem><para>kdb frontend</para>
> + <para>The kdb debugger shell is broken down into a number of
> + components. The kdb core is located in kernel/debug/kdb. There
> + are number of helper functions in some of the other kernel
are a number
> + components to make it possible for kdb to examine report
examine and report
> + information about the kernel without taking locks that could
> + cause a kernel deadlock. The kdb core contains implements the following functionality.</para>
> + <itemizedlist>
> + <listitem><para>A simple shell</para></listitem>
> + <listitem><para>The kdb core command set</para></listitem>
> + <listitem><para>A registration API to register additional kdb shell commands.</para>
> + <para>A good example of a self contained kdb module is the "ftdump" command for dumping the ftrace buffer. See: kernel/trace/trace_kdb.c</para></listitem>
self-contained
> + <listitem><para>The implementation for kdb_printf() which emit messages directly to I/O drivers, bypassing the kernel log.</para></listitem>
emits
> + <listitem><para>SW / HW breakpoint management for the kdb shell</para></listitem>
> + </itemizedlist>
> + </listitem>
> <listitem><para>kgdb I/O driver</para>
> <para>
> - Each kgdb I/O driver has to provide an implemenation for the following:
> + Each kgdb I/O driver has to provide an implementation for the following:
> <itemizedlist>
> <listitem><para>configuration via builtin or module</para></listitem>
> <listitem><para>dynamic configuration and kgdb hook registration calls</para></listitem>
> @@ -453,6 +763,10 @@
> <itemizedlist>
> <listitem><para>Jason Wessel<email>[email protected]</email></para></listitem>
> </itemizedlist>
> + In Jan 2010 this document was updated to include kdb.
> + <itemizedlist>
> + <listitem><para>Jason Wessel<email>[email protected]</email></para></listitem>
> + </itemizedlist>
> </para>
> </chapter>
> </book>
--
~Randy
On Fri, 2010-02-12 at 16:35 -0600, Jason Wessel wrote:
> @@ -115,7 +116,9 @@ void kgdb_roundup_cpus(unsigned long flags)
> /* KGDB functions to use existing PowerPC64 hooks. */
> static int kgdb_debugger(struct pt_regs *regs)
> {
> - return kgdb_handle_exception(0, computeSignal(TRAP(regs)), 0, regs);
> + if (kgdb_handle_exception(1, computeSignal(TRAP(regs)), DIE_OOPS, regs))
> + return 0;
> + return 1;
> }
I'm no fan of logic inversions like that but I suppose you are trying to
fit into existing hooks. However, I'd rather then do:
return !kgdb...
> static int kgdb_handle_breakpoint(struct pt_regs *regs)
> @@ -123,7 +126,7 @@ static int kgdb_handle_breakpoint(struct pt_regs *regs)
> if (user_mode(regs))
> return 0;
>
> - if (kgdb_handle_exception(0, SIGTRAP, 0, regs) != 0)
> + if (kgdb_handle_exception(1, SIGTRAP, 0, regs) != 0)
> return 0;
>
> if (*(u32 *) (regs->nip) == *(u32 *) (&arch_kgdb_ops.gdb_bpt_instr))
> diff --git a/arch/powerpc/kernel/traps.c b/arch/powerpc/kernel/traps.c
> index d069ff8..379104a 100644
> --- a/arch/powerpc/kernel/traps.c
> +++ b/arch/powerpc/kernel/traps.c
> @@ -809,12 +809,19 @@ void __kprobes program_check_exception(struct pt_regs *regs)
> return;
> }
> if (reason & REASON_TRAP) {
> +
> +#ifdef CONFIG_KGDB_LOW_LEVEL_TRAP
> + if (debugger_bpt(regs))
> + return;
> +#endif /* CONFIG_KGDB_LOW_LEVEL_TRAP */
> /* trap exception */
> if (notify_die(DIE_BPT, "breakpoint", regs, 5, 5, SIGTRAP)
> == NOTIFY_STOP)
> return;
> +#ifndef CONFIG_KGDB_LOW_LEVEL_TRAP
> if (debugger_bpt(regs))
> return;
> +#endif /* ! CONFIG_KGDB_LOW_LEVEL_TRAP */
>
> if (!(regs->msr & MSR_PR) && /* not user-mode */
> report_bug(regs->nip, regs) == BUG_TRAP_TYPE_WARN) {
> diff --git a/lib/Kconfig.kgdb b/lib/Kconfig.kgdb
> index 0a0e049..2b1601b 100644
> --- a/lib/Kconfig.kgdb
> +++ b/lib/Kconfig.kgdb
> @@ -59,7 +59,7 @@ config KGDB_TESTS_BOOT_STRING
>
> config KGDB_LOW_LEVEL_TRAP
> bool "KGDB: Allow debugging with traps in notifiers"
> - depends on X86
> + depends on X86 || PPC
> default n
> help
> This will add an extra call back to kgdb for the breakpoint
No firm objection, but it -is- a bit ugly... Should we just
unconditionally move the debugger_bpt() early on with a comment about
why we do so ? Is there any drawback you can think of ?
Cheers,
Ben.
Andrew Morton wrote:
> On Fri, 12 Feb 2010 17:20:22 -0600 Jason Wessel <[email protected]> wrote:
>
>
>> If you feel that would be cleaner, I can make that change, but I would
>> also have to call vprintk() from vkdb_printf when ever kdb_trap_printk
>> is not set. Do I understand your recommendation correctly?
>>
>>
>
> Was trying to find
> a way to avoid the code duplication.
>
>
>
You did make me think about the interface a bit further. If you were
going to ack one of the versions of this patch, which would you prefer?
The previous version, this new version or something else?
The difference here is that we use a replaced function call vs using the
if statement. There are more lines source in the kdb, but the interface
api is clean.
Thanks,
Jason.
Benjamin Herrenschmidt wrote:
> On Fri, 2010-02-12 at 16:35 -0600, Jason Wessel wrote:
>
>
>> @@ -115,7 +116,9 @@ void kgdb_roundup_cpus(unsigned long flags)
>> /* KGDB functions to use existing PowerPC64 hooks. */
>> static int kgdb_debugger(struct pt_regs *regs)
>> {
>> - return kgdb_handle_exception(0, computeSignal(TRAP(regs)), 0, regs);
>> + if (kgdb_handle_exception(1, computeSignal(TRAP(regs)), DIE_OOPS, regs))
>> + return 0;
>> + return 1;
>> }
>>
>
> I'm no fan of logic inversions like that but I suppose you are trying to
> fit into existing hooks. However, I'd rather then do:
>
> return !kgdb...
>
I agree and I made the change to return !kgdb...
>> +++ b/arch/powerpc/kernel/traps.c
>> @@ -809,12 +809,19 @@ void __kprobes program_check_exception(struct pt_regs *regs)
>> return;
>> }
>> if (reason & REASON_TRAP) {
>> +
>> +#ifdef CONFIG_KGDB_LOW_LEVEL_TRAP
>> + if (debugger_bpt(regs))
>> + return;
>> +#endif /* CONFIG_KGDB_LOW_LEVEL_TRAP */
>> /* trap exception */
>> if (notify_die(DIE_BPT, "breakpoint", regs, 5, 5, SIGTRAP)
>> == NOTIFY_STOP)
>> return;
>> +#ifndef CONFIG_KGDB_LOW_LEVEL_TRAP
>> if (debugger_bpt(regs))
>> return;
>> +#endif /* ! CONFIG_KGDB_LOW_LEVEL_TRAP */
>>
>> if (!(regs->msr & MSR_PR) && /* not user-mode */
>> report_bug(regs->nip, regs) == BUG_TRAP_TYPE_WARN) {
>>
>>
> No firm objection, but it -is- a bit ugly... Should we just
> unconditionally move the debugger_bpt() early on with a comment about
> why we do so ? Is there any drawback you can think of ?
>
>
I took a peek at xmon, and it suffers from the same problem where you
can place a breakpoint in any part of rcu_lock, notify_die, or
atomic_notifier_call_chain and meet with recursive faults. I also
checked that xmon appears to correctly return so as to continue if the
exception was not intended for xmon.
The reason I had not just moved the code block previously is that I was
not looking to break anything such as xmon, which is the the only other
user of this function.
I'll add your ack, if you agree with the new version of the patch.
Thanks,
Jason.
On Tue, 2010-02-16 at 13:33 -0600, Jason Wessel wrote:
> I took a peek at xmon, and it suffers from the same problem where you
> can place a breakpoint in any part of rcu_lock, notify_die, or
> atomic_notifier_call_chain and meet with recursive faults. I also
> checked that xmon appears to correctly return so as to continue if the
> exception was not intended for xmon.
>
> The reason I had not just moved the code block previously is that I was
> not looking to break anything such as xmon, which is the the only other
> user of this function.
>
> I'll add your ack, if you agree with the new version of the patch.
Almost, just stick a blank line here:
+ /* Debugger is first in line to stop recursive faults in
+ * rcu_lock, notify_die, or atomic_notifier_call_chain */
+ if (debugger_bpt(regs))
+ return;
---->
/* trap exception */
if (notify_die(DIE_BPT, "breakpoint", regs, 5, 5, SIGTRAP)
== NOTIFY_STOP)
return;
Please :-)
Acked-by: Benjamin Herrenschmidt <[email protected]>
Cheers,
Ben.
Benjamin Herrenschmidt wrote:
> Almost, just stick a blank line here:
>
> + /* Debugger is first in line to stop recursive faults in
> + * rcu_lock, notify_die, or atomic_notifier_call_chain */
> + if (debugger_bpt(regs))
> + return;
> ---->
> /* trap exception */
> if (notify_die(DIE_BPT, "breakpoint", regs, 5, 5, SIGTRAP)
> == NOTIFY_STOP)
> return;
>
> Please :-)
>
>
Done and pushed to kgdb-next.
Thanks,
Jason.
Randy Dunlap wrote:
> On 02/12/10 14:35, Jason Wessel wrote:
>> + kgdbwait [KGDB] Stop kernel execution and enter the
>> + kernel debugger at the earliest opportunity if
>> + kgdb was configured on as a boot argument.
>
> What does "kgdb was configured on as a boot argument" mean?
>
Changed to:
--
kgdbwait [KGDB] Stop kernel execution and enter the
kernel debugger at the earliest opportunity.
--
We'll assume someone actually configured the kernel debugger with
ekgdboc, kgdboc, or kgdbdbgp. The previous text was attempting to
indicate that kgdbwait only does something when you also configure a
kgdb I/O driver.
Where it stops varies on how early the I/O driver and architecture
trap processing is available.
Jason.
Randy Dunlap wrote:
> On 02/12/10 14:35, Jason Wessel wrote:
>> Update the kgdb docs to reflect the new directory structure and API.
>>
>> Merge in the kdb shell information.
>
> See comments/corrections inline below.
>
Thank very much for the comments. Would you like a credit in the
manual?
I merged in all your edits.
Here is the interdiff in case you were curious.
Thanks,
Jason.
DocBook/kgdb.tmpl | 208 ++++++++++++++++++++++++++++----------------------
kernel-parameters.txt | 3
2 files changed, 118 insertions(+), 93 deletions(-)
diff -u b/Documentation/DocBook/kgdb.tmpl b/Documentation/DocBook/kgdb.tmpl
--- b/Documentation/DocBook/kgdb.tmpl
+++ b/Documentation/DocBook/kgdb.tmpl
@@ -50,20 +50,20 @@
if you configure the kernel properly at compile and runtime.
</para>
<para>
- Kdb is simplistic shell style interface which you can use on a
+ Kdb is simplistic shell-style interface which you can use on a
system console with a keyboard or serial console. You can use it
to inspect memory, registers, process lists, dmesg, and even set
breakpoints to stop in a certain location. Kdb is not a source
level debugger, although you can set breakpoints and execute some
basic kernel run control. Kdb is mainly aimed at doing some
analysis to aid in development or diagnosing kernel problems. You
- can access some symbols by name in kernel builtins or in kernel
+ can access some symbols by name in kernel built-ins or in kernel
modules if the code was built
with <symbol>CONFIG_KALLSYMS</symbol>.
</para>
<para>
- Kgdb was intended to be used as a source level debugger for the
- linux kernel. It is used along with gdb to debug a linux kernel.
+ Kgdb is intended to be used as a source level debugger for the
+ Linux kernel. It is used along with gdb to debug a Linux kernel.
The expectation is that gdb can be used to "break in" to the
kernel to inspect memory, variables and look through call stack
information similar to the way an application developer would use
@@ -79,7 +79,7 @@
uImage...). In gdb the developer specifies the connection
parameters and connects to kgdb. The type of connection a
developer makes with gdb depends on the availability of kgdb I/O
- modules compiled as builtins or kernel modules in the test
+ modules compiled as built-ins or loadable kernel modules in the test
machine's kernel.
</para>
</chapter>
@@ -104,15 +104,15 @@
vmlinux file, gdb tends not to be very useful without the symbolic
data, so you will want to turn
on <symbol>CONFIG_DEBUG_INFO</symbol> which is called "Compile the
- kernel with debug info" in the menu config.
+ kernel with debug info" in the config menu.
</para>
<para>
It is advised, but not required that you turn on the
<symbol>CONFIG_FRAME_POINTER</symbol> kernel option which is called "Compile the
- kernel with frame pointers" in the menu config. This option
+ kernel with frame pointers" in the config menu. This option
inserts code to into the compiled executable which saves the frame
information in registers or on the stack at different points which
- will allow a debugger such as gdb to more accurately construct
+ allows a debugger such as gdb to more accurately construct
stack back traces while debugging the kernel.
</para>
<para>
@@ -132,9 +132,10 @@
must be built into the kernel directly. Kgdb I/O driver
configuration takes place via kernel or module parameters which
you can learn more about in the in the section that describes the
- kernel parameter kgdboc.
+ parameter "kgdboc".
</para>
- <para>Here is an example set of .config to enable kgdb:
+ <para>Here is an example set of .config symbols to enable or
+ disable for kgdb:
<itemizedlist>
<listitem><para># CONFIG_DEBUG_RODATA is not set</para></listitem>
<listitem><para>CONFIG_FRAME_POINTER=y</para></listitem>
@@ -146,7 +147,7 @@
<sect1 id="CompileKDB">
<title>Kernel config options for kdb</title>
<para>Kdb is quite a bit more complex than the simple gdbstub
- sitting on top of the kernel's debug core. Kdb must implement an
+ sitting on top of the kernel's debug core. Kdb must implement a
shell, and also adds some helper functions in other parts of the
kernel, responsible for printing out interesting data such as what
you would see if you ran "lsmod", or "ps". In order to build kdb
@@ -154,19 +155,18 @@
</para>
<para>The main config option for kdb
is <symbol>CONFIG_KGDB_KDB</symbol> which is called "KGDB_KDB:
- include kdb frontend for kgdb" in the menu config. In theory you
+ include kdb frontend for kgdb" in the config menu. In theory you
would have already also selected an I/O driver such as the
CONFIG_KGDB_SERIAL_CONSOLE interface if you plan on using kdb on a
serial port, when you were configuring kgdb.
</para>
- <para>If you want to use a ps/2 style keyboard with kdb you would
+ <para>If you want to use a PS/2-style keyboard with kdb, you would
select CONFIG_KDB_KEYBOARD which is called "KGDB_KDB: keyboard as
- input device" in the menu config. This option really does do
- anything for the gdb interface to kgdb it only works with kdb,
- because it is not really meaningful to manually type in gdb
- remote protocol commands.
+ input device" in the config menu. The CONFIG_KDB_KEYBOARD option
+ is not used for anything in the gdb interface to kgdb. The
+ CONFIG_KDB_KEYBOARD option only works with kdb.
</para>
- <para>Here is an example set of .config to enable kdb:
+ <para>Here is an example set of .config symbols to enable/disable kdb:
<itemizedlist>
<listitem><para># CONFIG_DEBUG_RODATA is not set</para></listitem>
<listitem><para>CONFIG_FRAME_POINTER=y</para></listitem>
@@ -180,10 +180,10 @@
</chapter>
<chapter id="kgdbKernelArgs">
<title>Kernel Debugger Boot Arguments</title>
- <para>This section will describe the various runtime kernel
+ <para>This section describes the various runtime kernel
parameters that affect the configuration of the kernel debugger.
- The following chapter will cover using kdb and kgdb as well as
- provide some examples of the configuration parameters.</para>
+ The following chapter covers using kdb and kgdb as well as
+ provides some examples of the configuration parameters.</para>
<sect1 id="kgdboc">
<title>Kernel parameter: kgdboc</title>
<para>The kgdboc driver was originally an abbreviation meant to
@@ -192,17 +192,54 @@
devices you want to use to interact with the kdb shell.
</para>
<para>For kgdb/gdb, kgdboc is designed to work with a single serial
- port. It was meant to cover the circumstance where you wanted to
+ port. It is intended to cover the circumstance where you want to
use a serial console as your primary console as well as using it to
perform kernel debugging. It is also possible to use kgdb on a
serial port which is not designated as a system console. Kgdboc
- may be configured as a kernel built-in or a kernel module. You can
- only make use of <constant>kgdbwait</constant> and early debugging
- if you build kgdboc into the kernel as a builtins.
+ may be configured as a kernel built-in or a kernel loadable module.
+ You can only make use of <constant>kgdbwait</constant> and early
+ debugging if you build kgdboc into the kernel as a built-in.
</para>
<sect2 id="kgdbocArgs">
<title>kgdboc arguments</title>
<para>Usage: <constant>kgdboc=[kbd][[,]serial_device][,baud]</constant></para>
+ <sect3 id="kgdbocArgs1">
+ <title>Using loadable module or built-in</title>
+ <para>
+ <orderedlist>
+ <listitem><para>As a kernel built-in:</para>
+ <para>Use the kernel boot argument: <constant>kgdboc=<tty-device>,[baud]</constant></para></listitem>
+ <listitem>
+ <para>As a kernel loadable module:</para>
+ <para>Use the command: <constant>modprobe kgdboc kgdboc=<tty-device>,[baud]</constant></para>
+ <para>Here are two examples of how you might formate the kgdboc
+ string. The first is for an x86 target using the first serial port.
+ The second example is for the ARM Versatile AB using the second
+ serial port.
+ <orderedlist>
+ <listitem><para><constant>kgdboc=ttyS0,115200</constant></para></listitem>
+ <listitem><para><constant>kgdboc=ttyAMA1,115200</constant></para></listitem>
+ </orderedlist>
+ </para>
+ </listitem>
+ </orderedlist></para>
+ </sect3>
+ <sect3 id="kgdbocArgs2">
+ <title>Configure kgdboc at runtime with sysfs</title>
+ <para>At run time you can enable or disable kgdboc by echoing a
+ parameters into the sysfs. Here are two examples:</para>
+ <orderedlist>
+ <listitem><para>Enable kgdboc on ttyS0</para>
+ <para><constant>echo ttyS0 > /sys/module/kgdboc/parameters/kgdboc</constant></para></listitem>
+ <listitem><para>Disable kgdboc</para>
+ <para><constant>echo "" > /sys/module/kgdboc/parameters/kgdboc</constant></para></listitem>
+ </orderedlist>
+ <para>NOTE: You do not need to specify the baud if you are
+ configuring the console on tty which is already configured or
+ open.</para>
+ </sect3>
+ <sect3 id="kgdbocArgs3">
+ <title>More examples</title>
<para>You can configure kgdboc to use the keyboard, and or a serial device
depending on if you are using kdb and or kgdb, in one of the
following scenarios.
@@ -220,42 +257,19 @@
</listitem>
</orderedlist>
</para>
- <para>You can configure kgdboc via sysfs or a module or kernel boot line
- parameter depending on if you build with CONFIG_KGDBOC as a module
- or built-in.
- <orderedlist>
- <listitem><para>From the module load or build-in</para>
- <para>As a kernel argument: <constant>kgdboc=<tty-device>,[baud]</constant></para>
- <para>As a kernel module: <constant>modprobe kgdboc kgdboc=<tty-device>,[baud]</constant></para>
- <para>Here are two examples, the first for an x86 target using the
- first serial port and the second for the ARM Versatile AB using the
- second serial port.
- <orderedlist>
- <listitem><para><constant>kgdboc=ttyS0,115200</constant></para></listitem>
- <listitem><para><constant>kgdboc=ttyAMA1,115200</constant></para></listitem>
- </orderedlist>
- </para>
- </listitem>
- <listitem><para>From sysfs</para>
- <para><constant>echo ttyS0 > /sys/module/kgdboc/parameters/kgdboc</constant></para>
- <para>NOTE: You do not need to specify the baud if you are
- configuring the console on tty which is already configured or
- open.</para>
- </listitem>
- </orderedlist>
- </para>
+ </sect3>
<para>NOTE: Kgdboc does not support interrupting the target via the
gdb remote protocol. You must manually send a sysrq-g unless you
have a proxy that splits console output to a terminal program.
- A console proxy has a separate tcp port for the debugger and a separate
- tcp port for the "human" console. The proxy can take care of sending
+ A console proxy has a separate TCP port for the debugger and a separate
+ TCP port for the "human" console. The proxy can take care of sending
the sysrq-g for you.
</para>
<para>When using kgdboc with no debugger proxy, you can end up
- connecting the debugger for one of two entry points. If an
- exception occurs after you have loaded kgdboc a message should
+ connecting the debugger at one of two entry points. If an
+ exception occurs after you have loaded kgdboc, a message should
print on the console stating it is waiting for the debugger. In
- case you disconnect your terminal program and then connect the
+ this case you disconnect your terminal program and then connect the
debugger in its place. If you want to interrupt the target system
and forcibly enter a debug session you have to issue a Sysrq
sequence and then type the letter <constant>g</constant>. Then
@@ -280,8 +294,9 @@
</para>
<para>
The kernel will stop and wait as early as the I/O driver and
- architecture will allow when you use this option. If you build the
- kgdb I/O driver as a kernel module kgdbwait will not do anything.
+ architecture allows when you use this option. If you build the
+ kgdb I/O driver as a loadable kernel module kgdbwait will not do
+ anything.
</para>
</sect1>
<sect1 id="kgdbcon">
@@ -297,7 +312,7 @@
<listitem><para>Activate with the kernel command line option:</para>
<para><constant>kgdbcon</constant></para>
</listitem>
- <listitem><para>Use sysfs before configuring an io driver</para>
+ <listitem><para>Use sysfs before configuring an I/O driver</para>
<para>
<constant>echo 1 > /sys/module/kgdb/parameters/kgdb_use_con</constant>
</para>
@@ -309,7 +324,7 @@
</listitem>
</orderedlist>
<para>IMPORTANT NOTE: You cannot use kgdboc + kgdbcon on a tty that is an
- active system console IE: console=ttyS0,115200 kgdboc=ttyS0 kgdbcon
+ active system console. An example incorrect usage is <constant>console=ttyS0,115200 kgdboc=ttyS0 kgdbcon</constant>
</para>
<para>It is possible to use this option with kgdboc on a tty that is not a system console.
</para>
@@ -322,20 +337,20 @@
</para>
<sect1 id="quickKDBserial">
<title>Quick start for kdb on a serial port</title>
- <para>This is a quick example of how to use kdb</para>
+ <para>This is a quick example of how to use kdb.</para>
<para><orderedlist>
<listitem><para>Boot kernel with arguments:
<itemizedlist>
<listitem><para><constant>console=ttyS0,115200 kgdboc=ttyS0,115200</constant></para></listitem>
</itemizedlist></para>
<para>OR</para>
- <para>Configure kgdboc after the kernel booted assuming your are using a serial port console:
+ <para>Configure kgdboc after the kernel booted; assuming you are using a serial port console:
<itemizedlist>
<listitem><para><constant>echo ttyS0 > /sys/module/kgdboc/parameters/kgdboc</constant></para></listitem>
</itemizedlist>
</para>
</listitem>
- <listitem><para>Enter the kernel debugger manually or by waiting for an opps or fault. There are several ways you can enter the kernel debugger manually, all involve using the sysrq-g, which means you have to enabled CONFIG_MAGIC_SYSRQ=y in your kernel config</para>
+ <listitem><para>Enter the kernel debugger manually or by waiting for an oops or fault. There are several ways you can enter the kernel debugger manually; all involve using the sysrq-g, which means you must have enabled CONFIG_MAGIC_SYSRQ=y in your kernel config.</para>
<itemizedlist>
<listitem><para>When logged in as root or with a super user session you can run:</para>
<para><constant>echo g > /proc/sysrq-trigger</constant></para></listitem>
@@ -366,13 +381,19 @@
</para>
</listitem>
<listitem>
- <para>When you are done using kdb you need to consider rebooting the system or using the "go" command to resuming normal kernel execution. If you have paused the kernel for a lengthy period of time, applications that rely on timely networking or anything to do with real wall clock time could get adversely affected, so you should take this into consideration when using the kernel debugger.</para>
+ <para>When you are done using kdb you need to consider rebooting the
+ system or using the "go" command to resuming normal kernel
+ execution. If you have paused the kernel for a lengthy period of
+ time, applications that rely on timely networking or anything to do
+ with real wall clock time could be adversely affected, so you
+ should take this into consideration when using the kernel
+ debugger.</para>
</listitem>
</orderedlist></para>
</sect1>
<sect1 id="quickKDBkeyboard">
<title>Quick start for kdb using a keyboard connected console</title>
- <para>This is a quick example of how to use kdb with a keyboard</para>
+ <para>This is a quick example of how to use kdb with a keyboard.</para>
<para><orderedlist>
<listitem><para>Boot kernel with arguments:
<itemizedlist>
@@ -385,7 +406,7 @@
</itemizedlist>
</para>
</listitem>
- <listitem><para>Enter the kernel debugger manually or by waiting for an opps or fault. There are several ways you can enter the kernel debugger manually, all involve using the sysrq-g, which means you have to enabled CONFIG_MAGIC_SYSRQ=y in your kernel config</para>
+ <listitem><para>Enter the kernel debugger manually or by waiting for an oops or fault. There are several ways you can enter the kernel debugger manually; all involve using the sysrq-g, which means you must have enabled CONFIG_MAGIC_SYSRQ=y in your kernel config.</para>
<itemizedlist>
<listitem><para>When logged in as root or with a super user session you can run:</para>
<para><constant>echo g > /proc/sysrq-trigger</constant></para></listitem>
@@ -397,7 +418,7 @@
<para>Press and release: <constant>g</constant></para>
<para>Release: <constant>Alt</constant></para>
</listitem>
- <listitem><para>Example using a PS/2 101 keyboard</para>
+ <listitem><para>Example using a PS/2 101-key keyboard</para>
<para>Press and hold down: <constant>Alt</constant></para>
<para>Press and release the key with the label: <constant>SysRq</constant></para>
<para>Press and release: <constant>g</constant></para>
@@ -427,7 +448,7 @@
The driver can be unconfigured by passing an empty string. You cannot
change the configuration while the debugger is attached. Make sure
to detach the debugger with the <constant>detach</constant> command
- prior to trying unconfigure a kgdb I/O driver.
+ prior to trying to unconfigure a kgdb I/O driver.
</para>
<sect1 id="ConnectingGDB">
<title>Connecting with gdb to a serial port</title>
@@ -445,10 +466,10 @@
</listitem>
<listitem>
<para>Stop kernel execution (break into the debugger)</para>
- <para>In order to connect to gdb to the via kgdboc the kernel must
+ <para>In order to connect to gdb via kgdboc, the kernel must
first be stopped. There are several ways to stop the kernel which
include using kgdbwait as a boot argument, via a sysrq-g, or running
- the kernel until it takes an an exception where it waits for the
+ the kernel until it takes an exception where it waits for the
debugger to attach.
<itemizedlist>
<listitem><para>When logged in as root or with a super user session you can run:</para>
@@ -478,7 +499,7 @@
(gdb) target remote /dev/ttyS0
</programlisting>
<para>
- Example (kgdb to a terminal server on tcp port 2012):
+ Example (kgdb to a terminal server on TCP port 2012):
</para>
<programlisting>
% gdb ./vmlinux
@@ -497,8 +518,8 @@
</para>
</listitem>
</orderedlist>
- <para>Remember if you continue in gdb, and need "break in" again,
- you need to issue an other sysrq. It is easy to create an simple
+ <para>Remember if you continue in gdb, and need to "break in" again,
+ you need to issue an other sysrq-g. It is easy to create a simple
entry point by putting a breakpoint at <constant>sys_sync</constant>
and then you can run "sync" from a shell or script to break into the
debugger.</para>
@@ -514,11 +535,11 @@
<sect2>
<title>Switching from kgdb to kdb</title>
<para>
- There are two ways to switch from kgdb to kdb, you can use gdb to
+ There are two ways to switch from kgdb to kdb: you can use gdb to
issue a maintenance packet, or you can blindly type the command $3#33.
- When ever kernel debugger stops in kgdb mode it will print the
+ Whenever kernel debugger stops in kgdb mode it will print the
message <constant>KGDB or $3#33 for KDB</constant>. It is important
- to note, that you have to type the sequence correctly in one pass.
+ to note that you have to type the sequence correctly in one pass.
You cannot type a backspace or delete because kgdb will interpret
that as part of the debug stream.
<orderedlist>
@@ -526,7 +547,8 @@
<para><constant>$3#33</constant></para></listitem>
<listitem><para>Change from kgdb to kdb with gdb</para>
<para><constant>maintenance packet 3</constant></para>
- <para>NOTE: Now you must kill gdb</para></listitem>
+ <para>NOTE: Now you must kill gdb. Typically you press control-z and
+ issue the command: kill -9 %</para></listitem>
</orderedlist>
</para>
</sect2>
@@ -554,7 +576,7 @@
state of the kernel debugger. You should be using gdb for
breakpoints and run control operations if you have gdb connected.
The more useful commands to run are things like lsmod, dmesg, ps or
- possibly some of the memory information. To see all the kdb
+ possibly some of the memory information commands. To see all the kdb
commands you can run <constant>monitor help</constant>.</para>
<para>Example:
<informalexample><programlisting>
@@ -607,24 +629,26 @@
<para>
The debug core is found in kernel/debugger/debug_core.c. It contains:
<itemizedlist>
- <listitem><para>A generic OS exception handler which includes sync'ing the processors into a stopped state on an multi cpu system.</para></listitem>
+ <listitem><para>A generic OS exception handler which includes
+ sync'ing the processors into a stopped state on an multi-CPU
+ system.</para></listitem>
<listitem><para>The API to talk to the kgdb I/O drivers</para></listitem>
- <listitem><para>The API to make calls to the arch specific kgdb implementation</para></listitem>
+ <listitem><para>The API to make calls to the arch-specific kgdb implementation</para></listitem>
<listitem><para>The logic to perform safe memory reads and writes to memory while using the debugger</para></listitem>
<listitem><para>A full implementation for software breakpoints unless overridden by the arch</para></listitem>
<listitem><para>The API to invoke either the kdb or kgdb frontend to the debug core.</para></listitem>
</itemizedlist>
</para>
</listitem>
- <listitem><para>kgdb arch specific implementation</para>
+ <listitem><para>kgdb arch-specific implementation</para>
<para>
This implementation is generally found in arch/*/kernel/kgdb.c.
As an example, arch/x86/kernel/kgdb.c contains the specifics to
implement HW breakpoint as well as the initialization to
dynamically register and unregister for the trap handlers on
- this architecture. The arch specific portion implements:
+ this architecture. The arch-specific portion implements:
<itemizedlist>
- <listitem><para>contains an arch specific trap catcher which
+ <listitem><para>contains an arch-specific trap catcher which
invokes kgdb_handle_exception() to start kgdb about doing its
work</para></listitem>
<listitem><para>translation to and from gdb specific packet format to pt_regs</para></listitem>
@@ -644,16 +668,18 @@
<listitem><para>kdb frontend</para>
<para>The kdb debugger shell is broken down into a number of
components. The kdb core is located in kernel/debug/kdb. There
- are number of helper functions in some of the other kernel
- components to make it possible for kdb to examine report
+ are a number of helper functions in some of the other kernel
+ components to make it possible for kdb to examine and report
information about the kernel without taking locks that could
cause a kernel deadlock. The kdb core contains implements the following functionality.</para>
<itemizedlist>
<listitem><para>A simple shell</para></listitem>
<listitem><para>The kdb core command set</para></listitem>
<listitem><para>A registration API to register additional kdb shell commands.</para>
- <para>A good example of a self contained kdb module is the "ftdump" command for dumping the ftrace buffer. See: kernel/trace/trace_kdb.c</para></listitem>
- <listitem><para>The implementation for kdb_printf() which emit messages directly to I/O drivers, bypassing the kernel log.</para></listitem>
+ <para>A good example of a self-contained kdb module is the "ftdump" command for dumping the ftrace buffer. See: kernel/trace/trace_kdb.c</para></listitem>
+ <listitem><para>The implementation for kdb_printf() which
+ emits messages directly to I/O drivers, bypassing the kernel
+ log.</para></listitem>
<listitem><para>SW / HW breakpoint management for the kdb shell</para></listitem>
</itemizedlist>
</listitem>
@@ -661,7 +687,7 @@
<para>
Each kgdb I/O driver has to provide an implementation for the following:
<itemizedlist>
- <listitem><para>configuration via builtin or module</para></listitem>
+ <listitem><para>configuration via built-in or module</para></listitem>
<listitem><para>dynamic configuration and kgdb hook registration calls</para></listitem>
<listitem><para>read and write character interface</para></listitem>
<listitem><para>A cleanup handler for unconfiguring from the kgdb core</para></listitem>
@@ -726,15 +752,15 @@
underlying low level to the hardware driver having "polling hooks"
which the to which the tty driver is attached. In the initial
implementation of kgdboc it the serial_core was changed to expose a
- low level uart hook for doing polled mode reading and writing of a
+ low level UART hook for doing polled mode reading and writing of a
single character while in an atomic context. When kgdb makes an I/O
request to the debugger, kgdboc invokes a call back in the serial
- core which in turn uses the call back in the uart driver. It is
- certainly possible to extend kgdboc to work with non-uart based
+ core which in turn uses the call back in the UART driver. It is
+ certainly possible to extend kgdboc to work with non-UART based
consoles in the future.
</para>
<para>
- When using kgdboc with a uart, the uart driver must implement two callbacks in the <constant>struct uart_ops</constant>. Example from drivers/8250.c:<programlisting>
+ When using kgdboc with a UART, the UART driver must implement two callbacks in the <constant>struct uart_ops</constant>. Example from drivers/8250.c:<programlisting>
#ifdef CONFIG_CONSOLE_POLL
.poll_get_char = serial8250_get_poll_char,
.poll_put_char = serial8250_put_poll_char,
@@ -744,7 +770,7 @@
<constant>#ifdef CONFIG_CONSOLE_POLL</constant>, as shown above.
Keep in mind that polling hooks have to be implemented in such a way
that they can be called from an atomic context and have to restore
- the state of the uart chip on return such that the system can return
+ the state of the UART chip on return such that the system can return
to normal when the debugger detaches. You need to be very careful
with any kind of lock you consider, because failing here is most
going to mean pressing the reset button.
diff -u b/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
--- b/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1111,8 +1111,7 @@
keyboard and serial format: kbd,<serial_device>[,baud]
kgdbwait [KGDB] Stop kernel execution and enter the
- kernel debugger at the earliest opportunity if
- kgdb was configured on as a boot argument.
+ kernel debugger at the earliest opportunity.
kmac= [MIPS] korina ethernet MAC address.
Configure the RouterBoard 532 series on-chip
On 02/17/10 13:00, Jason Wessel wrote:
> Randy Dunlap wrote:
>> On 02/12/10 14:35, Jason Wessel wrote:
>>> Update the kgdb docs to reflect the new directory structure and API.
>>>
>>> Merge in the kdb shell information.
>>
>> See comments/corrections inline below.
>>
>
> Thank very much for the comments. Would you like a credit in the
> manual?
not needed.
> I merged in all your edits.
>
> Here is the interdiff in case you were curious.
>
> Thanks,
> Jason.
>
> DocBook/kgdb.tmpl | 208 ++++++++++++++++++++++++++++----------------------
> kernel-parameters.txt | 3
> 2 files changed, 118 insertions(+), 93 deletions(-)
thanks,
--
~Randy
Andrew Morton <[email protected]> writes:
> On Fri, 12 Feb 2010 17:43:39 -0600 Jason Wessel <[email protected]> wrote:
>
>> >> printk(KERN_EMERG "Kernel panic - not syncing: %s\n",buf);
>> >> #ifdef CONFIG_DEBUG_BUGVERBOSE
>> >> dump_stack();
>> >> @@ -91,8 +94,6 @@ NORET_TYPE void panic(const char * fmt, ...)
>> >> */
>> >> smp_send_stop();
>> >>
>> >> - atomic_notifier_call_chain(&panic_notifier_list, 0, buf);
>> >> -
>> >> bust_spinlocks(0);
>> >>
>> >> if (!panic_blink)
>> >>
>> >
>> > So the notifier call now happens before all the printks and the kexec
>> > and kmsg_dump handling. What effect does this have upon the code which
>> > implements kexec and kmsg_dump?
>> >
>> >
>>
>> I certainly don't want to break kexec or alter any behavior, does that
>> mean kgdb / kdb should hook the kexec for notification?
>>
>> I think ideally it is a end user's preference as to if they want in via
>> kexec or the kernel debugger. Calling the smp_send_stop() prior to the
>> notifier was a death sentence for the kernel debugger.
>>
>> Perhaps I can move the notifier before smp_send_stop()?
>
> Well. My question can be simplified to "does this break existing code"?
Yes. Removing the bust_spinlocks(1) and moving the panic notification up
will measurably decrease the reliability of crash dump capture. I expect
it will even result in dead_locks where we fail to print the panic message
for people who are trying to debug via the console.
Eric
Jason Wessel <[email protected]> writes:
> This patch contains the hooks and instrumentation into kernel which
> live outside the kernel/debug directory, which the kdb core
> will call to run commands like lsmod, dmesg, bt etc...
You know this dropping the locks from vmalloc_info and swap_info
is down right ugly, and I don't believe it is safe. That code
was not designed to run while the write_lock is held.
How does kdb build if NOMMU is set?
Similarly with si_swapinfo. What makes it safe to run the
code without it's locks. Safe as in no null pointer deferences
or other nasties.
I am not impressed with the mess you make of meminfo_proc_show.
That looks like a maintenance hazard if I ever saw one. An innocent
change to get some additional info (that happens to take a lock)
will cause kdb to deadlock.
The weird magic include of kdb.h in fs/proc/meminfo.c is also ugly.
Eric
> diff --git a/fs/proc/internal.h b/fs/proc/internal.h
> index 1f24a3e..36d55e1 100644
> --- a/fs/proc/internal.h
> +++ b/fs/proc/internal.h
> @@ -32,11 +32,11 @@ extern struct mm_struct *mm_for_maps(struct task_struct *);
>
> #ifdef CONFIG_MMU
> #define VMALLOC_TOTAL (VMALLOC_END - VMALLOC_START)
> -extern void get_vmalloc_info(struct vmalloc_info *vmi);
> +extern void get_vmalloc_info(struct vmalloc_info *vmi, int lock);
> #else
>
> #define VMALLOC_TOTAL 0UL
> -#define get_vmalloc_info(vmi) \
> +#define get_vmalloc_info(vmi, lock) \
> do { \
> (vmi)->used = 0; \
> (vmi)->largest_chunk = 0; \
> diff --git a/fs/proc/meminfo.c b/fs/proc/meminfo.c
> index a65239c..b967cf2 100644
> --- a/fs/proc/meminfo.c
> +++ b/fs/proc/meminfo.c
> @@ -10,6 +10,7 @@
> #include <linux/seq_file.h>
> #include <linux/swap.h>
> #include <linux/vmstat.h>
> +#include <linux/kdb.h>
> #include <asm/atomic.h>
> #include <asm/page.h>
> #include <asm/pgtable.h>
> @@ -19,7 +20,7 @@ void __attribute__((weak)) arch_report_meminfo(struct seq_file *m)
> {
> }
>
> -static int meminfo_proc_show(struct seq_file *m, void *v)
> +int _meminfo_proc_show(struct seq_file *m, void *v, int lock)
> {
> struct sysinfo i;
> unsigned long committed;
> @@ -34,7 +35,10 @@ static int meminfo_proc_show(struct seq_file *m, void *v)
> */
> #define K(x) ((x) << (PAGE_SHIFT - 10))
> si_meminfo(&i);
> - si_swapinfo(&i);
> + if (lock)
> + si_swapinfo(&i);
> + else
> + __si_swapinfo(&i);
> committed = percpu_counter_read_positive(&vm_committed_as);
> allowed = ((totalram_pages - hugetlb_total_pages())
> * sysctl_overcommit_ratio / 100) + total_swap_pages;
> @@ -44,7 +48,7 @@ static int meminfo_proc_show(struct seq_file *m, void *v)
> if (cached < 0)
> cached = 0;
>
> - get_vmalloc_info(&vmi);
> + get_vmalloc_info(&vmi, lock);
>
> for (lru = LRU_BASE; lru < NR_LRU_LISTS; lru++)
> pages[lru] = global_page_state(NR_LRU_BASE + lru);
> @@ -161,6 +165,11 @@ static int meminfo_proc_show(struct seq_file *m, void *v)
> #undef K
> }
>
> +static int meminfo_proc_show(struct seq_file *m, void *v)
> +{
> + return _meminfo_proc_show(m, v, 1);
> +}
> +
> static int meminfo_proc_open(struct inode *inode, struct file *file)
> {
> return single_open(file, meminfo_proc_show, NULL);
> diff --git a/fs/proc/mmu.c b/fs/proc/mmu.c
> index 8ae221d..10a0f8b 100644
> --- a/fs/proc/mmu.c
> +++ b/fs/proc/mmu.c
> @@ -14,7 +14,7 @@
> #include <asm/pgtable.h>
> #include "internal.h"
>
> -void get_vmalloc_info(struct vmalloc_info *vmi)
> +void get_vmalloc_info(struct vmalloc_info *vmi, int lock)
> {
> struct vm_struct *vma;
> unsigned long free_area_size;
> @@ -30,7 +30,8 @@ void get_vmalloc_info(struct vmalloc_info *vmi)
>
> prev_end = VMALLOC_START;
>
> - read_lock(&vmlist_lock);
> + if (lock)
> + read_lock(&vmlist_lock);
>
> for (vma = vmlist; vma; vma = vma->next) {
> unsigned long addr = (unsigned long) vma->addr;
> @@ -55,6 +56,7 @@ void get_vmalloc_info(struct vmalloc_info *vmi)
> if (VMALLOC_END - prev_end > vmi->largest_chunk)
> vmi->largest_chunk = VMALLOC_END - prev_end;
>
> - read_unlock(&vmlist_lock);
> + if (lock)
> + read_unlock(&vmlist_lock);
> }
> }
> diff --git a/include/asm-generic/kmap_types.h b/include/asm-generic/kmap_types.h
> index e5f234a..97e807c 100644
> --- a/include/asm-generic/kmap_types.h
> +++ b/include/asm-generic/kmap_types.h
> @@ -28,7 +28,8 @@ KMAP_D(15) KM_UML_USERCOPY,
> KMAP_D(16) KM_IRQ_PTE,
> KMAP_D(17) KM_NMI,
> KMAP_D(18) KM_NMI_PTE,
> -KMAP_D(19) KM_TYPE_NR
> +KMAP_D(19) KM_KDB,
> +KMAP_D(20) KM_TYPE_NR
> };
>
> #undef KMAP_D
> diff --git a/include/linux/swap.h b/include/linux/swap.h
> index a2602a8..c326282 100644
> --- a/include/linux/swap.h
> +++ b/include/linux/swap.h
> @@ -312,6 +312,7 @@ extern struct page *swapin_readahead(swp_entry_t, gfp_t,
> /* linux/mm/swapfile.c */
> extern long nr_swap_pages;
> extern long total_swap_pages;
> +extern void __si_swapinfo(struct sysinfo *);
> extern void si_swapinfo(struct sysinfo *);
> extern swp_entry_t get_swap_page(void);
> extern swp_entry_t get_swap_page_of_type(int);
> @@ -377,6 +378,7 @@ static inline void mem_cgroup_uncharge_swap(swp_entry_t ent)
>
> #define si_swapinfo(val) \
> do { (val)->freeswap = (val)->totalswap = 0; } while (0)
> +#define __si_swapinfo(val) si_swapinfo(val)
> /* only sparc can not include linux/pagemap.h in this file
> * so leave page_cache_release and release_pages undeclared... */
> #define free_page_and_swap_cache(page) \
> diff --git a/init/main.c b/init/main.c
> index 4cb47a1..9d415a1 100644
> --- a/init/main.c
> +++ b/init/main.c
> @@ -63,6 +63,7 @@
> #include <linux/sched.h>
> #include <linux/signal.h>
> #include <linux/idr.h>
> +#include <linux/kdb.h>
> #include <linux/ftrace.h>
> #include <linux/async.h>
> #include <linux/kmemcheck.h>
> @@ -659,6 +660,7 @@ asmlinkage void __init start_kernel(void)
> key_init();
> radix_tree_init();
> security_init();
> + kdb_init(KDB_INIT_FULL);
> vfs_caches_init(totalram_pages);
> signals_init();
> /* rootfs populating might need page-writeback */
> diff --git a/kernel/kallsyms.c b/kernel/kallsyms.c
> index 8e5288a..dc08f8b 100644
> --- a/kernel/kallsyms.c
> +++ b/kernel/kallsyms.c
> @@ -16,6 +16,7 @@
> #include <linux/init.h>
> #include <linux/seq_file.h>
> #include <linux/fs.h>
> +#include <linux/kdb.h>
> #include <linux/err.h>
> #include <linux/proc_fs.h>
> #include <linux/sched.h> /* for cond_resched */
> @@ -515,6 +516,26 @@ static int kallsyms_open(struct inode *inode, struct file *file)
> return ret;
> }
>
> +#ifdef CONFIG_KGDB_KDB
> +const char *kdb_walk_kallsyms(loff_t *pos)
> +{
> + static struct kallsym_iter kdb_walk_kallsyms_iter;
> + if (*pos == 0) {
> + memset(&kdb_walk_kallsyms_iter, 0,
> + sizeof(kdb_walk_kallsyms_iter));
> + reset_iter(&kdb_walk_kallsyms_iter, 0);
> + }
> + while (1) {
> + if (!update_iter(&kdb_walk_kallsyms_iter, *pos))
> + return NULL;
> + ++*pos;
> + /* Some debugging symbols have no name. Ignore them. */
> + if (kdb_walk_kallsyms_iter.name[0])
> + return kdb_walk_kallsyms_iter.name;
> + }
> +}
> +#endif /* CONFIG_KGDB_KDB */
> +
> static const struct file_operations kallsyms_operations = {
> .open = kallsyms_open,
> .read = seq_read,
> diff --git a/kernel/module.c b/kernel/module.c
> index f82386b..e59aca1 100644
> --- a/kernel/module.c
> +++ b/kernel/module.c
> @@ -79,6 +79,10 @@ EXPORT_TRACEPOINT_SYMBOL(module_get);
> DEFINE_MUTEX(module_mutex);
> EXPORT_SYMBOL_GPL(module_mutex);
> static LIST_HEAD(modules);
> +#ifdef CONFIG_KGDB_KDB
> +struct list_head *kdb_modules = &modules; /* kdb needs the list of modules */
> +#endif /* CONFIG_KGDB_KDB */
> +
>
> /* Block module loading/unloading? */
> int modules_disabled = 0;
> diff --git a/kernel/printk.c b/kernel/printk.c
> index 1751c45..d86c91a 100644
> --- a/kernel/printk.c
> +++ b/kernel/printk.c
> @@ -420,6 +420,22 @@ SYSCALL_DEFINE3(syslog, int, type, char __user *, buf, int, len)
> return do_syslog(type, buf, len);
> }
>
> +#ifdef CONFIG_KGDB_KDB
> +/* kdb dmesg command needs access to the syslog buffer. do_syslog()
> + * uses locks so it cannot be used during debugging. Just tell kdb
> + * where the start and end of the physical and logical logs are. This
> + * is equivalent to do_syslog(3).
> + */
> +void kdb_syslog_data(char *syslog_data[4])
> +{
> + syslog_data[0] = log_buf;
> + syslog_data[1] = log_buf + log_buf_len;
> + syslog_data[2] = log_buf + log_end -
> + (logged_chars < log_buf_len ? logged_chars : log_buf_len);
> + syslog_data[3] = log_buf + log_end;
> +}
> +#endif /* CONFIG_KGDB_KDB */
> +
> /*
> * Call the console drivers on a range of log_buf
> */
> diff --git a/kernel/sched.c b/kernel/sched.c
> index 3a8fb30..92ab8b8 100644
> --- a/kernel/sched.c
> +++ b/kernel/sched.c
> @@ -9802,9 +9802,9 @@ void normalize_rt_tasks(void)
>
> #endif /* CONFIG_MAGIC_SYSRQ */
>
> -#ifdef CONFIG_IA64
> +#if defined(CONFIG_IA64) || defined(CONFIG_KGDB_KDB)
> /*
> - * These functions are only useful for the IA64 MCA handling.
> + * These functions are only useful for the IA64 MCA handling, or kdb.
> *
> * They can only be called when the whole system has been
> * stopped - every CPU needs to be quiescent, and no scheduling
> @@ -9824,6 +9824,9 @@ struct task_struct *curr_task(int cpu)
> return cpu_curr(cpu);
> }
>
> +#endif /* defined(CONFIG_IA64) || defined(CONFIG_KGDB_KDB) */
> +
> +#ifdef CONFIG_IA64
> /**
> * set_curr_task - set the current task for a given cpu.
> * @cpu: the processor in question.
> diff --git a/kernel/signal.c b/kernel/signal.c
> index 934ae5e..96ecbf8 100644
> --- a/kernel/signal.c
> +++ b/kernel/signal.c
> @@ -2718,3 +2718,43 @@ void __init signals_init(void)
> {
> sigqueue_cachep = KMEM_CACHE(sigqueue, SLAB_PANIC);
> }
> +
> +#ifdef CONFIG_KGDB_KDB
> +#include <linux/kdb.h>
> +/*
> + * kdb_send_sig_info - Allows kdb to send signals without exposing
> + * signal internals. This function checks if the required locks are
> + * available before calling the main signal code, to avoid kdb
> + * deadlocks.
> + */
> +void
> +kdb_send_sig_info(struct task_struct *t, struct siginfo *info)
> +{
> + static struct task_struct *kdb_prev_t;
> + int sig, new_t;
> + if (!spin_trylock(&t->sighand->siglock)) {
> + kdb_printf("Can't do kill command now.\n"
> + "The sigmask lock is held somewhere else in "
> + "kernel, try again later\n");
> + return;
> + }
> + spin_unlock(&t->sighand->siglock);
> + new_t = kdb_prev_t != t;
> + kdb_prev_t = t;
> + if (t->state != TASK_RUNNING && new_t) {
> + kdb_printf("Process is not RUNNING, sending a signal from "
> + "kdb risks deadlock\n"
> + "on the run queue locks. "
> + "The signal has _not_ been sent.\n"
> + "Reissue the kill command if you want to risk "
> + "the deadlock.\n");
> + return;
> + }
> + sig = info->si_signo;
> + if (send_sig_info(sig, info, t))
> + kdb_printf("Fail to deliver Signal %d to process %d.\n",
> + sig, t->pid);
> + else
> + kdb_printf("Signal %d is sent to process %d.\n", sig, t->pid);
> +}
> +#endif /* CONFIG_KGDB_KDB */
> diff --git a/mm/swapfile.c b/mm/swapfile.c
> index 6c0585b..dc2039e 100644
> --- a/mm/swapfile.c
> +++ b/mm/swapfile.c
> @@ -13,6 +13,7 @@
> #include <linux/swap.h>
> #include <linux/vmalloc.h>
> #include <linux/pagemap.h>
> +#include <linux/kdb.h>
> #include <linux/namei.h>
> #include <linux/shm.h>
> #include <linux/blkdev.h>
> @@ -2056,12 +2057,11 @@ out:
> return error;
> }
>
> -void si_swapinfo(struct sysinfo *val)
> +void __si_swapinfo(struct sysinfo *val)
> {
> unsigned int type;
> unsigned long nr_to_be_unused = 0;
>
> - spin_lock(&swap_lock);
> for (type = 0; type < nr_swapfiles; type++) {
> struct swap_info_struct *si = swap_info[type];
>
> @@ -2070,6 +2070,12 @@ void si_swapinfo(struct sysinfo *val)
> }
> val->freeswap = nr_swap_pages + nr_to_be_unused;
> val->totalswap = total_swap_pages + nr_to_be_unused;
> +}
> +
> +void si_swapinfo(struct sysinfo *val)
> +{
> + spin_lock(&swap_lock);
> + __si_swapinfo(val);
> spin_unlock(&swap_lock);
> }
>
> --
> 1.6.4.rc1
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
Eric W. Biederman wrote:
> Jason Wessel <[email protected]> writes:
>
>
>> This patch contains the hooks and instrumentation into kernel which
>> live outside the kernel/debug directory, which the kdb core
>> will call to run commands like lsmod, dmesg, bt etc...
>>
>
> You know this dropping the locks from vmalloc_info and swap_info
> is down right ugly, and I don't believe it is safe. That code
> was not designed to run while the write_lock is held.
>
Perhaps we can find some middle ground. I don't mind simply not
allowing the information to be queried from kdb if the locks are not
available.
Which is less ugly you, making the swap_lock global or adding a function
to query it?
> How does kdb build if NOMMU is set?
>
I did not see a kernel config option called NOMMU.
Can you point me to a kernel config that uses this? Or is it the case
that it is something like "# CONFIG_MMU is not set"?
Perhaps we just set kdb to work only when CONFIG_MMU is enabled? It is
hard to say it works or not without a kernel configuration, file system
and hardware to test it with.
> Similarly with si_swapinfo. What makes it safe to run the
> code without it's locks. Safe as in no null pointer deferences
> or other nasties.
>
>
It looks to me like the original kdb took the approach of calling the
setjmp() longjmp() and if there was any kind of fault, it long jumped
back to the original context. Obviously that doesn't solve any kind of
problem with a list loop.
As I mentioned earlier, I think we should check for the lock and if it
isn't available you don't get to execute the function, else a separate
function has to be written that does the same thing safely. I do not
see any value in making a separate function to safely print the
information at this time. If you want to probe around, consider using a
gdb macro.
> I am not impressed with the mess you make of meminfo_proc_show.
> That looks like a maintenance hazard if I ever saw one. An innocent
> change to get some additional info (that happens to take a lock)
> will cause kdb to deadlock.
>
>
This code goes away if we expose a way to check for the lock.
Jason.
Jason Wessel <[email protected]> writes:
> Eric W. Biederman wrote:
>> Jason Wessel <[email protected]> writes:
>>
>>
>>> This patch contains the hooks and instrumentation into kernel which
>>> live outside the kernel/debug directory, which the kdb core
>>> will call to run commands like lsmod, dmesg, bt etc...
>>>
>>
>> You know this dropping the locks from vmalloc_info and swap_info
>> is down right ugly, and I don't believe it is safe. That code
>> was not designed to run while the write_lock is held.
>>
>
> Perhaps we can find some middle ground. I don't mind simply not
> allowing the information to be queried from kdb if the locks are not
> available.
>
> Which is less ugly you, making the swap_lock global or adding a function
> to query it?
My recommendation would be to simply drop the swap_info and meminfo
information for now, and have kdb do what is good at.
Skimming through the history of the discussion it appears Christoph
Hellwig asked you to do something about these bits on the first
review.
>From a maintenance point of view anything where you have to know which
locks a function will take is fragile. It is entirely too easy to create
a change that is fine in it's normal context but breaks kdb.
Alt-sysrq already allows capturing that kind of information, without any
thorny maintenance issues. By contrast kdb appears to be a much larger
and inferior tool.
Eric
Eric W. Biederman wrote:
> Jason Wessel <[email protected]> writes:
>
>
>> Eric W. Biederman wrote:
>>
>>> Jason Wessel <[email protected]> writes:
>>>
>>>
>>>
>>>> This patch contains the hooks and instrumentation into kernel which
>>>> live outside the kernel/debug directory, which the kdb core
>>>> will call to run commands like lsmod, dmesg, bt etc...
>>>>
>>>>
>>> You know this dropping the locks from vmalloc_info and swap_info
>>> is down right ugly, and I don't believe it is safe. That code
>>> was not designed to run while the write_lock is held.
>>>
>>>
>> Perhaps we can find some middle ground. I don't mind simply not
>> allowing the information to be queried from kdb if the locks are not
>> available.
>>
>> Which is less ugly you, making the swap_lock global or adding a function
>> to query it?
>>
>
> My recommendation would be to simply drop the swap_info and meminfo
> information for now, and have kdb do what is good at.
>
I have no objection to dropping this for now.
> Skimming through the history of the discussion it appears Christoph
> Hellwig asked you to do something about these bits on the first
> review.
>
>
If you are referring to the statement from the RFC patch:
"Patch 1 still containes a lot random junk. E.g. all those meminfo and
whatever hooks aren't required for a very basic kernel debugger"
Most of that was restructured in that code base vs what you have taken a
look at and we are down to a handful of instrumentation points.
> From a maintenance point of view anything where you have to know which
> locks a function will take is fragile. It is entirely too easy to create
> a change that is fine in it's normal context but breaks kdb.
>
>
Sure, but that is life for kdb in the current state. You either make
safe query macros or create something to call helper functions to
inspect code. Either way it is maintenance.
> Alt-sysrq already allows capturing that kind of information, without any
> thorny maintenance issues.
Alt-sysrq cannot be called from all the same contexts that you can
invoke kdb. The maintenance issue you speak of is avoided because of
the points you can actually invoke a sysrq. Certainly you are not going
to be able to call Alt-sysrq, in an nmi as well as a number of other
contexts.
> By contrast kdb appears to be a much larger
> and inferior tool.
>
>
I am not certain as to what comparison you are trying to draw here. If
you are only using kdb to obtain meminfo and swapinfo, sure I buy your
argument. If you are using kdb to modify, inspect memory and execute
back traces at certain places then I think this is not an apples to
apples comparison.
I'll post a new version of the patch with the hooks you asked about
stripped out. The commands will also get removed from the kdb command
shell.
Thanks,
Jason.
On Thu, Feb 18, 2010 at 09:04:36AM -0600, Jason Wessel wrote:
> Eric W. Biederman wrote:
> > Jason Wessel <[email protected]> writes:
> >
> >
> >> This patch contains the hooks and instrumentation into kernel which
> >> live outside the kernel/debug directory, which the kdb core
> >> will call to run commands like lsmod, dmesg, bt etc...
> >>
> >
> > You know this dropping the locks from vmalloc_info and swap_info
> > is down right ugly, and I don't believe it is safe. That code
> > was not designed to run while the write_lock is held.
> >
>
> Perhaps we can find some middle ground. I don't mind simply not
> allowing the information to be queried from kdb if the locks are not
> available.
IIRC the original KDB would stop all the cpus when entered,
thus locking to avoid concurrent access to data
was not necessary when displaying kernel data structures. However,
KDB user and developers were assumed to be aware that when KDB was
entered the system context was in an indeterminate state particularly
with respect to linked lists and other non-tabular data structures.
KDB code that displayed data structures which were kept in a non-table
data structure (linked list, tree, etc.) was be required to both
validate each pointer it tries to follow as well as ensure that it
detects loops (either by terminating the list traversal after a certain
number of elements or by allowing the KDB user to terminate the traversal
with e.g. 'q').
>
> It looks to me like the original kdb took the approach of calling the
> setjmp() longjmp() and if there was any kind of fault, it long jumped
> back to the original context. Obviously that doesn't solve any kind of
> problem with a list loop.
Yes. The list loop was expected to be handled either by the display
code terminating after some number of traversal step or by the KDB user
terminating the command via the keyboard (e.g. 'q' at a more-type prompt).
If the new KDB framework allows other cpus to continue to run while kdb
data structure display commands are running, then much more care must
be taken in the display command code to avoid inconsistent data causing
loops or #PF.
scott
Scott Lurndal wrote:
>
> IIRC the original KDB would stop all the cpus when entered,
> thus locking to avoid concurrent access to data
> was not necessary when displaying kernel data structures.
Just because the system is "effectively frozen" does not mean you can
safely walk a structure or call something that takes a lock.
One of the key points Eric and others have made is that they do not want
some of this helper code in the kernel core, nor do they want alternate
lock semantics while the kernel debugger is active. You can achieve the
same sort of interrogation with a gdb helper macro. At some point kdb
could be extended to have the same sort of functionality if someone
finds they just cannot live without it.
> However,
> KDB user and developers were assumed to be aware that when KDB was
> entered the system context was in an indeterminate state particularly
> with respect to linked lists and other non-tabular data structures.
>
> KDB code that displayed data structures which were kept in a non-table
> data structure (linked list, tree, etc.) was be required to both
> validate each pointer it tries to follow as well as ensure that it
> detects loops (either by terminating the list traversal after a certain
> number of elements or by allowing the KDB user to terminate the traversal
> with e.g. 'q').
>
>
>> It looks to me like the original kdb took the approach of calling the
>> setjmp() longjmp() and if there was any kind of fault, it long jumped
>> back to the original context. Obviously that doesn't solve any kind of
>> problem with a list loop.
>>
>
> Yes. The list loop was expected to be handled either by the display
> code terminating after some number of traversal step or by the KDB user
> terminating the command via the keyboard (e.g. 'q' at a more-type prompt).
>
>
The new kdb has a pager as well as abort operations, but it does not
make use of setjmp() longjmp() to handle faults while executing other
helper print code.
> If the new KDB framework allows other cpus to continue to run while kdb
> data structure display commands are running, then much more care must
> be taken in the display command code to avoid inconsistent data causing
> loops or #PF.
>
>
In the new kdb. The system is fully stopped by the kernel debug core.
There is the concept of the master CPU (the one running the debug shell)
and the slave CPUs which are all the other cores. All the slaves spin
in a control loop, and you may switch cpus with the kdb cpu command
without exiting the debug context. The master can "trade places" with
an online slave cpu.
Jason.
Jason Wessel wrote:
> I'll post a new version of the patch with the hooks you asked about
> stripped out. The commands will also get removed from the kdb command
> shell.
Thanks for your comments Eric.
The summary of the clean up is as follows because some of the kdb core
had to be touched in order to properly remove this.
fs/proc/internal.h | 4 +-
fs/proc/meminfo.c | 15 +--------
fs/proc/mmu.c | 8 +----
include/linux/swap.h | 2 -
kernel/debug/kdb/kdb_main.c | 13 +++-----
kernel/debug/kdb/kdb_private.h | 12 -------
kernel/debug/kdb/kdb_support.c | 65 -----------------------------------------
mm/swapfile.c | 10 +-----
8 files changed, 15 insertions(+), 114 deletions(-)
Below is the new version of the basic kernel core kdb modifications.
If you have further comments, please let me know. I would also be
happy to add an ack if you approve of the new patch.
Thanks,
Jason.
---
From: Jason Wessel <[email protected]>
Subject: [PATCH] kdb: core for kgdb back end (2 of 2)
This patch contains the hooks and instrumentation into kernel which
live outside the kernel/debug directory, which the kdb core
will call to run commands like lsmod, dmesg, bt etc...
CC: [email protected]
CC: [email protected]
Signed-off-by: Jason Wessel <[email protected]>
---
arch/arm/include/asm/kmap_types.h | 1
arch/powerpc/include/asm/kmap_types.h | 1
include/asm-generic/kmap_types.h | 3 +-
init/main.c | 2 +
kernel/kallsyms.c | 21 +++++++++++++++++
kernel/module.c | 4 +++
kernel/printk.c | 16 +++++++++++++
kernel/sched.c | 7 ++++-
kernel/signal.c | 40 ++++++++++++++++++++++++++++++++++
9 files changed, 92 insertions(+), 3 deletions(-)
--- a/arch/arm/include/asm/kmap_types.h
+++ b/arch/arm/include/asm/kmap_types.h
@@ -19,6 +19,7 @@ enum km_type {
KM_SOFTIRQ0,
KM_SOFTIRQ1,
KM_L2_CACHE,
+ KM_KDB,
KM_TYPE_NR
};
--- a/arch/powerpc/include/asm/kmap_types.h
+++ b/arch/powerpc/include/asm/kmap_types.h
@@ -26,6 +26,7 @@ enum km_type {
KM_SOFTIRQ1,
KM_PPC_SYNC_PAGE,
KM_PPC_SYNC_ICACHE,
+ KM_KDB,
KM_TYPE_NR
};
--- a/include/asm-generic/kmap_types.h
+++ b/include/asm-generic/kmap_types.h
@@ -28,7 +28,8 @@ KMAP_D(15) KM_UML_USERCOPY,
KMAP_D(16) KM_IRQ_PTE,
KMAP_D(17) KM_NMI,
KMAP_D(18) KM_NMI_PTE,
-KMAP_D(19) KM_TYPE_NR
+KMAP_D(19) KM_KDB,
+KMAP_D(20) KM_TYPE_NR
};
#undef KMAP_D
--- a/init/main.c
+++ b/init/main.c
@@ -63,6 +63,7 @@
#include <linux/sched.h>
#include <linux/signal.h>
#include <linux/idr.h>
+#include <linux/kdb.h>
#include <linux/ftrace.h>
#include <linux/async.h>
#include <linux/kmemcheck.h>
@@ -659,6 +660,7 @@ asmlinkage void __init start_kernel(void
key_init();
radix_tree_init();
security_init();
+ kdb_init(KDB_INIT_FULL);
vfs_caches_init(totalram_pages);
signals_init();
/* rootfs populating might need page-writeback */
--- a/kernel/kallsyms.c
+++ b/kernel/kallsyms.c
@@ -16,6 +16,7 @@
#include <linux/init.h>
#include <linux/seq_file.h>
#include <linux/fs.h>
+#include <linux/kdb.h>
#include <linux/err.h>
#include <linux/proc_fs.h>
#include <linux/sched.h> /* for cond_resched */
@@ -515,6 +516,26 @@ static int kallsyms_open(struct inode *i
return ret;
}
+#ifdef CONFIG_KGDB_KDB
+const char *kdb_walk_kallsyms(loff_t *pos)
+{
+ static struct kallsym_iter kdb_walk_kallsyms_iter;
+ if (*pos == 0) {
+ memset(&kdb_walk_kallsyms_iter, 0,
+ sizeof(kdb_walk_kallsyms_iter));
+ reset_iter(&kdb_walk_kallsyms_iter, 0);
+ }
+ while (1) {
+ if (!update_iter(&kdb_walk_kallsyms_iter, *pos))
+ return NULL;
+ ++*pos;
+ /* Some debugging symbols have no name. Ignore them. */
+ if (kdb_walk_kallsyms_iter.name[0])
+ return kdb_walk_kallsyms_iter.name;
+ }
+}
+#endif /* CONFIG_KGDB_KDB */
+
static const struct file_operations kallsyms_operations = {
.open = kallsyms_open,
.read = seq_read,
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -79,6 +79,10 @@ EXPORT_TRACEPOINT_SYMBOL(module_get);
DEFINE_MUTEX(module_mutex);
EXPORT_SYMBOL_GPL(module_mutex);
static LIST_HEAD(modules);
+#ifdef CONFIG_KGDB_KDB
+struct list_head *kdb_modules = &modules; /* kdb needs the list of modules */
+#endif /* CONFIG_KGDB_KDB */
+
/* Block module loading/unloading? */
int modules_disabled = 0;
--- a/kernel/printk.c
+++ b/kernel/printk.c
@@ -420,6 +420,22 @@ SYSCALL_DEFINE3(syslog, int, type, char
return do_syslog(type, buf, len);
}
+#ifdef CONFIG_KGDB_KDB
+/* kdb dmesg command needs access to the syslog buffer. do_syslog()
+ * uses locks so it cannot be used during debugging. Just tell kdb
+ * where the start and end of the physical and logical logs are. This
+ * is equivalent to do_syslog(3).
+ */
+void kdb_syslog_data(char *syslog_data[4])
+{
+ syslog_data[0] = log_buf;
+ syslog_data[1] = log_buf + log_buf_len;
+ syslog_data[2] = log_buf + log_end -
+ (logged_chars < log_buf_len ? logged_chars : log_buf_len);
+ syslog_data[3] = log_buf + log_end;
+}
+#endif /* CONFIG_KGDB_KDB */
+
/*
* Call the console drivers on a range of log_buf
*/
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -9802,9 +9802,9 @@ void normalize_rt_tasks(void)
#endif /* CONFIG_MAGIC_SYSRQ */
-#ifdef CONFIG_IA64
+#if defined(CONFIG_IA64) || defined(CONFIG_KGDB_KDB)
/*
- * These functions are only useful for the IA64 MCA handling.
+ * These functions are only useful for the IA64 MCA handling, or kdb.
*
* They can only be called when the whole system has been
* stopped - every CPU needs to be quiescent, and no scheduling
@@ -9824,6 +9824,9 @@ struct task_struct *curr_task(int cpu)
return cpu_curr(cpu);
}
+#endif /* defined(CONFIG_IA64) || defined(CONFIG_KGDB_KDB) */
+
+#ifdef CONFIG_IA64
/**
* set_curr_task - set the current task for a given cpu.
* @cpu: the processor in question.
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -2718,3 +2718,43 @@ void __init signals_init(void)
{
sigqueue_cachep = KMEM_CACHE(sigqueue, SLAB_PANIC);
}
+
+#ifdef CONFIG_KGDB_KDB
+#include <linux/kdb.h>
+/*
+ * kdb_send_sig_info - Allows kdb to send signals without exposing
+ * signal internals. This function checks if the required locks are
+ * available before calling the main signal code, to avoid kdb
+ * deadlocks.
+ */
+void
+kdb_send_sig_info(struct task_struct *t, struct siginfo *info)
+{
+ static struct task_struct *kdb_prev_t;
+ int sig, new_t;
+ if (!spin_trylock(&t->sighand->siglock)) {
+ kdb_printf("Can't do kill command now.\n"
+ "The sigmask lock is held somewhere else in "
+ "kernel, try again later\n");
+ return;
+ }
+ spin_unlock(&t->sighand->siglock);
+ new_t = kdb_prev_t != t;
+ kdb_prev_t = t;
+ if (t->state != TASK_RUNNING && new_t) {
+ kdb_printf("Process is not RUNNING, sending a signal from "
+ "kdb risks deadlock\n"
+ "on the run queue locks. "
+ "The signal has _not_ been sent.\n"
+ "Reissue the kill command if you want to risk "
+ "the deadlock.\n");
+ return;
+ }
+ sig = info->si_signo;
+ if (send_sig_info(sig, info, t))
+ kdb_printf("Fail to deliver Signal %d to process %d.\n",
+ sig, t->pid);
+ else
+ kdb_printf("Signal %d is sent to process %d.\n", sig, t->pid);
+}
+#endif /* CONFIG_KGDB_KDB */
Eric W. Biederman wrote:
> Andrew Morton <[email protected]> writes:
>
>> On Fri, 12 Feb 2010 17:43:39 -0600 Jason Wessel <[email protected]> wrote:
>>
>>>>> printk(KERN_EMERG "Kernel panic - not syncing: %s\n",buf);
>>>>> #ifdef CONFIG_DEBUG_BUGVERBOSE
>>>>> dump_stack();
>>>>> @@ -91,8 +94,6 @@ NORET_TYPE void panic(const char * fmt, ...)
>>>>> */
>>>>> smp_send_stop();
>>>>>
>>>>> - atomic_notifier_call_chain(&panic_notifier_list, 0, buf);
>>>>> -
>>>>> bust_spinlocks(0);
>>>>>
>>>>> if (!panic_blink)
>>>>>
>>>> So the notifier call now happens before all the printks and the kexec
>>>> and kmsg_dump handling. What effect does this have upon the code which
>>>> implements kexec and kmsg_dump?
>>>>
>>>>
>>> I certainly don't want to break kexec or alter any behavior, does that
>>> mean kgdb / kdb should hook the kexec for notification?
>>>
>>> I think ideally it is a end user's preference as to if they want in via
>>> kexec or the kernel debugger. Calling the smp_send_stop() prior to the
>>> notifier was a death sentence for the kernel debugger.
>>>
>>> Perhaps I can move the notifier before smp_send_stop()?
>> Well. My question can be simplified to "does this break existing code"?
>
> Yes. Removing the bust_spinlocks(1) and moving the panic notification up
The bust_spinlocks(1) was not removed but moved around, but the point
is moot. Yes, it confirmed beyond the shadow of a doubt to break
existing code.
The revised patch does not touch anything in kernel/panic.c. Instead
we make a best effort to get the kernel debugger first in line to the
notifier. It turns out one of the other notifiers was the culprit in
the hanging the system during the panic tests.
If anyone further objects, I will simply drop the kdb panic
notification registration entirely.
Thanks,
Jason.
---
From: Jason Wessel <[email protected]>
Subject: [PATCH] kdb,debug_core: Allow the debug core to receive a panic notification
It is highly desirable to trap into kdb on panic. The debug core will
attempt to register as the first in line for the panic notifier.
CC: Ingo Molnar <[email protected]>
CC: Andrew Morton <[email protected]>
CC: Eric W. Biederman <[email protected]>
Signed-off-by: Jason Wessel <[email protected]>
---
kernel/debug/debug_core.c | 19 +++++++++++++++++++
1 file changed, 19 insertions(+)
--- a/kernel/debug/debug_core.c
+++ b/kernel/debug/debug_core.c
@@ -754,11 +754,28 @@ static struct sysrq_key_op sysrq_dbg_op
};
#endif
+static int kgdb_panic_event(struct notifier_block *self,
+ unsigned long val,
+ void *data)
+{
+ if (dbg_kdb_mode)
+ kdb_printf("PANIC: %s\n", (char *)data);
+ kgdb_breakpoint();
+ return NOTIFY_DONE;
+}
+
+static struct notifier_block kgdb_panic_event_nb = {
+ .notifier_call = kgdb_panic_event,
+ .priority = INT_MAX,
+};
+
static void kgdb_register_callbacks(void)
{
if (!kgdb_io_module_registered) {
kgdb_io_module_registered = 1;
kgdb_arch_init();
+ atomic_notifier_chain_register(&panic_notifier_list,
+ &kgdb_panic_event_nb);
#ifdef CONFIG_MAGIC_SYSRQ
register_sysrq_key('g', &sysrq_dbg_op);
#endif
@@ -778,6 +795,8 @@ static void kgdb_unregister_callbacks(vo
*/
if (kgdb_io_module_registered) {
kgdb_io_module_registered = 0;
+ atomic_notifier_chain_unregister(&panic_notifier_list,
+ &kgdb_panic_event_nb);
kgdb_arch_exit();
#ifdef CONFIG_MAGIC_SYSRQ
unregister_sysrq_key('g', &sysrq_dbg_op);
On Fri, Feb 12, 2010 at 04:35:34PM -0600, Jason Wessel wrote:
> The only way the debugger can handle a trap in inside rcu_lock,
> notify_die, or atomic_notifier_call_chain without a recursive fault is
> to have a low level "first opportunity handler" do_trap_or_bp() handler.
>
> Generally this will be something the vast majority of folks will not
> need, but for those who need it, it is added as a kernel .config
> option called KGDB_LOW_LEVEL_TRAP.
>
> Also added was a die notification for oops such that kdb can catch an
> oops for analysis.
>
> There appeared to be no obvious way to pass the struct pt_regs from
> the original exception back to the stack back tracer, so a special
> case was added to show_stack() for when kdb is active because you
> generally desire to generally look at the back trace of the original
> exception.
>
> CC: Ralf Baechle <[email protected]>
> Signed-off-by: Jason Wessel <[email protected]>
Acked-by: Ralf Baechle <[email protected]>
Ralf