The of-tree subsystem does not currently preserve the IBM vTPM 1.2 and
vTPM 2.0 measurement logs across a kexec on PowerVM and PowerKVM. This
series fixes this for the kexec_file_load() syscall using the flattened
device tree (fdt) to carry the TPM measurement log's buffer across kexec.
Stefan
v5:
- Rebased on 1 more patch that would otherwise create merge conflicts
v4:
- Rebased on 2 patches that would otherwise create merge conflicts;
posting these patches in this series with several tags removed so
krobot can test the series already
- Changes to individual patches documented in patch descripitons
v3:
- Moved TPM Open Firmware related function to drivers/char/tpm/eventlog/tpm_of.c
v2:
- rearranged patches
- fixed compilation issues for x86
Jonathan McDowell (1):
x86/kexec: Carry forward IMA measurement log on kexec
Palmer Dabbelt (1):
drivers: of: kexec ima: Support 32-bit platforms
Stefan Berger (3):
tpm: of: Make of-tree specific function commonly available
of: kexec: Refactor IMA buffer related functions to make them reusable
tpm/kexec: Duplicate TPM measurement log in of-tree for kexec
Vaibhav Jain (1):
of: check previous kernel's ima-kexec-buffer against memory bounds
arch/x86/Kconfig | 1 +
arch/x86/include/uapi/asm/bootparam.h | 9 +
arch/x86/kernel/e820.c | 6 +-
arch/x86/kernel/kexec-bzimage64.c | 42 +++-
arch/x86/kernel/setup.c | 63 +++++
drivers/char/tpm/eventlog/of.c | 31 +--
drivers/of/kexec.c | 341 ++++++++++++++++++++++----
include/linux/ima.h | 5 +
include/linux/kexec.h | 6 +
include/linux/of.h | 10 +-
include/linux/tpm.h | 27 ++
kernel/kexec_file.c | 6 +
security/integrity/ima/ima_kexec.c | 2 +-
13 files changed, 467 insertions(+), 82 deletions(-)
base-commit: 03c765b0e3b4cb5063276b086c76f7a612856a9a
--
2.35.1
Simplify tpm_read_log_of() by moving reusable parts of the code into
an inline function that makes it commonly available so it can be
used also for kexec support. Call the new of_tpm_get_sml_parameters()
function from the TPM Open Firmware driver.
Signed-off-by: Stefan Berger <[email protected]>
Cc: Jarkko Sakkinen <[email protected]>
Cc: Jason Gunthorpe <[email protected]>
Cc: Rob Herring <[email protected]>
Cc: Frank Rowand <[email protected]>
---
v4:
- converted to inline function
---
drivers/char/tpm/eventlog/of.c | 31 +++++--------------------------
include/linux/tpm.h | 27 +++++++++++++++++++++++++++
2 files changed, 32 insertions(+), 26 deletions(-)
diff --git a/drivers/char/tpm/eventlog/of.c b/drivers/char/tpm/eventlog/of.c
index a9ce66d09a75..f9462d19632e 100644
--- a/drivers/char/tpm/eventlog/of.c
+++ b/drivers/char/tpm/eventlog/of.c
@@ -12,6 +12,7 @@
#include <linux/slab.h>
#include <linux/of.h>
+#include <linux/tpm.h>
#include <linux/tpm_eventlog.h>
#include "../tpm.h"
@@ -20,11 +21,10 @@
int tpm_read_log_of(struct tpm_chip *chip)
{
struct device_node *np;
- const u32 *sizep;
- const u64 *basep;
struct tpm_bios_log *log;
u32 size;
u64 base;
+ int ret;
log = &chip->log;
if (chip->dev.parent && chip->dev.parent->of_node)
@@ -35,30 +35,9 @@ int tpm_read_log_of(struct tpm_chip *chip)
if (of_property_read_bool(np, "powered-while-suspended"))
chip->flags |= TPM_CHIP_FLAG_ALWAYS_POWERED;
- sizep = of_get_property(np, "linux,sml-size", NULL);
- basep = of_get_property(np, "linux,sml-base", NULL);
- if (sizep == NULL && basep == NULL)
- return -ENODEV;
- if (sizep == NULL || basep == NULL)
- return -EIO;
-
- /*
- * For both vtpm/tpm, firmware has log addr and log size in big
- * endian format. But in case of vtpm, there is a method called
- * sml-handover which is run during kernel init even before
- * device tree is setup. This sml-handover function takes care
- * of endianness and writes to sml-base and sml-size in little
- * endian format. For this reason, vtpm doesn't need conversion
- * but physical tpm needs the conversion.
- */
- if (of_property_match_string(np, "compatible", "IBM,vtpm") < 0 &&
- of_property_match_string(np, "compatible", "IBM,vtpm20") < 0) {
- size = be32_to_cpup((__force __be32 *)sizep);
- base = be64_to_cpup((__force __be64 *)basep);
- } else {
- size = *sizep;
- base = *basep;
- }
+ ret = of_tpm_get_sml_parameters(np, &base, &size);
+ if (ret < 0)
+ return ret;
if (size == 0) {
dev_warn(&chip->dev, "%s: Event log area empty\n", __func__);
diff --git a/include/linux/tpm.h b/include/linux/tpm.h
index dfeb25a0362d..b3dff255bc58 100644
--- a/include/linux/tpm.h
+++ b/include/linux/tpm.h
@@ -460,4 +460,31 @@ static inline struct tpm_chip *tpm_default_chip(void)
return NULL;
}
#endif
+
+#ifdef CONFIG_OF
+static inline int of_tpm_get_sml_parameters(struct device_node *np,
+ u64 *base, u32 *size)
+{
+ const u32 *sizep;
+ const u64 *basep;
+
+ sizep = of_get_property(np, "linux,sml-size", NULL);
+ basep = of_get_property(np, "linux,sml-base", NULL);
+ if (sizep == NULL && basep == NULL)
+ return -ENODEV;
+ if (sizep == NULL || basep == NULL)
+ return -EIO;
+
+ if (of_property_match_string(np, "compatible", "IBM,vtpm") < 0 &&
+ of_property_match_string(np, "compatible", "IBM,vtpm20") < 0) {
+ *size = be32_to_cpup((__force __be32 *)sizep);
+ *base = be64_to_cpup((__force __be64 *)basep);
+ } else {
+ *size = *sizep;
+ *base = *basep;
+ }
+ return 0;
+}
+#endif
+
#endif
--
2.35.1
The memory area of the TPM measurement log is currently not properly
duplicated for carrying it across kexec when an Open Firmware
Devicetree is used. Therefore, the contents of the log get corrupted.
Fix this for the kexec_file_load() syscall by allocating a buffer and
copying the contents of the existing log into it. The new buffer is
preserved across the kexec and a pointer to it is available when the new
kernel is started. To achieve this, store the allocated buffer's address
in the flattened device tree (fdt) under the name linux,tpm-kexec-buffer
and search for this entry early in the kernel startup before the TPM
subsystem starts up. Adjust the pointer in the of-tree stored under
linux,sml-base to point to this buffer holding the preserved log. The TPM
driver can then read the base address from this entry when making the log
available. Invalidate the log by removing 'linux,sml-base' from the
devicetree if anything goes wrong with updating the buffer.
Use subsys_initcall() to call the function to restore the buffer even if
the TPM subsystem or driver are not used. This allows the buffer to be
carried across the next kexec without involvement of the TPM subsystem
and ensures a valid buffer pointed to by the of-tree.
Use the subsys_initcall(), rather than an ealier initcall, since
page_is_ram() in get_kexec_buffer() only starts working at this stage.
Signed-off-by: Stefan Berger <[email protected]>
Cc: Rob Herring <[email protected]>
Cc: Frank Rowand <[email protected]>
Cc: Eric Biederman <[email protected]>
---
v4:
- Added #include <linux/vmalloc.h> due to parisc
- Use phys_addr_t for physical address rather than void *
- Remove linux,sml-base if the buffer cannot be updated after a kexec
- Added __init to functions where possible
---
drivers/of/kexec.c | 216 +++++++++++++++++++++++++++++++++++++++++-
include/linux/kexec.h | 6 ++
include/linux/of.h | 8 +-
kernel/kexec_file.c | 6 ++
4 files changed, 233 insertions(+), 3 deletions(-)
diff --git a/drivers/of/kexec.c b/drivers/of/kexec.c
index 404a86bb3978..6d07e2120296 100644
--- a/drivers/of/kexec.c
+++ b/drivers/of/kexec.c
@@ -19,6 +19,8 @@
#include <linux/random.h>
#include <linux/slab.h>
#include <linux/types.h>
+#include <linux/tpm.h>
+#include <linux/vmalloc.h>
#define RNG_SEED_SIZE 128
@@ -116,7 +118,6 @@ static int do_get_kexec_buffer(const void *prop, int len, unsigned long *addr,
return 0;
}
-#ifdef CONFIG_HAVE_IMA_KEXEC
static int get_kexec_buffer(const char *name, unsigned long *addr, size_t *size)
{
int ret, len;
@@ -150,6 +151,7 @@ static int get_kexec_buffer(const char *name, unsigned long *addr, size_t *size)
return 0;
}
+#ifdef CONFIG_HAVE_IMA_KEXEC
/**
* ima_get_kexec_buffer - get IMA buffer from the previous kernel
* @addr: On successful return, set to point to the buffer contents.
@@ -238,7 +240,6 @@ static void remove_ima_buffer(void *fdt, int chosen_node)
remove_buffer(fdt, chosen_node, "linux,ima-kexec-buffer");
}
-#ifdef CONFIG_IMA_KEXEC
static int setup_buffer(void *fdt, int chosen_node, const char *name,
phys_addr_t addr, size_t size)
{
@@ -262,6 +263,7 @@ static int setup_buffer(void *fdt, int chosen_node, const char *name,
}
+#ifdef CONFIG_IMA_KEXEC
/**
* setup_ima_buffer - add IMA buffer information to the fdt
* @image: kexec image being loaded.
@@ -284,6 +286,213 @@ static inline int setup_ima_buffer(const struct kimage *image, void *fdt,
}
#endif /* CONFIG_IMA_KEXEC */
+/**
+ * tpm_get_kexec_buffer - get TPM log buffer from the previous kernel
+ * @phyaddr: On successful return, set to physical address of buffer
+ * @size: On successful return, set to the buffer size.
+ *
+ * Return: 0 on success, negative errno on error.
+ */
+static int __init tpm_get_kexec_buffer(phys_addr_t *phyaddr, size_t *size)
+{
+ unsigned long tmp_addr;
+ size_t tmp_size;
+ int ret;
+
+ ret = get_kexec_buffer("linux,tpm-kexec-buffer", &tmp_addr, &tmp_size);
+ if (ret)
+ return ret;
+
+ *phyaddr = (phys_addr_t)tmp_addr;
+ *size = tmp_size;
+
+ return 0;
+}
+
+/**
+ * tpm_of_remove_kexec_buffer - remove the linux,tpm-kexec-buffer node
+ */
+static int __init tpm_of_remove_kexec_buffer(void)
+{
+ struct property *prop;
+
+ prop = of_find_property(of_chosen, "linux,tpm-kexec-buffer", NULL);
+ if (!prop)
+ return -ENOENT;
+
+ return of_remove_property(of_chosen, prop);
+}
+
+/**
+ * remove_tpm_buffer - remove the TPM log buffer property and reservation from @fdt
+ *
+ * @fdt: Flattened Device Tree to update
+ * @chosen_node: Offset to the chosen node in the device tree
+ *
+ * The TPM log measurement buffer is of no use to a subsequent kernel, so we always
+ * remove it from the device tree.
+ */
+static void remove_tpm_buffer(void *fdt, int chosen_node)
+{
+ if (!IS_ENABLED(CONFIG_PPC64))
+ return;
+
+ remove_buffer(fdt, chosen_node, "linux,tpm-kexec-buffer");
+}
+
+/**
+ * setup_tpm_buffer - add TPM measurement log buffer information to the fdt
+ * @image: kexec image being loaded.
+ * @fdt: Flattened device tree for the next kernel.
+ * @chosen_node: Offset to the chosen node.
+ *
+ * Return: 0 on success, or negative errno on error.
+ */
+static int setup_tpm_buffer(const struct kimage *image, void *fdt,
+ int chosen_node)
+{
+ if (!IS_ENABLED(CONFIG_PPC64))
+ return 0;
+
+ return setup_buffer(fdt, chosen_node, "linux,tpm-kexec-buffer",
+ image->tpm_buffer_addr, image->tpm_buffer_size);
+}
+
+void tpm_add_kexec_buffer(struct kimage *image)
+{
+ struct kexec_buf kbuf = { .image = image, .buf_align = 1,
+ .buf_min = 0, .buf_max = ULONG_MAX,
+ .top_down = true };
+ struct device_node *np;
+ void *buffer;
+ u32 size;
+ u64 base;
+ int ret;
+
+ if (!IS_ENABLED(CONFIG_PPC64))
+ return;
+
+ np = of_find_node_by_name(NULL, "vtpm");
+ if (!np)
+ return;
+
+ if (of_tpm_get_sml_parameters(np, &base, &size) < 0)
+ return;
+
+ buffer = vmalloc(size);
+ if (!buffer)
+ return;
+ memcpy(buffer, __va(base), size);
+
+ kbuf.buffer = buffer;
+ kbuf.bufsz = size;
+ kbuf.memsz = size;
+ ret = kexec_add_buffer(&kbuf);
+ if (ret) {
+ pr_err("Error passing over kexec TPM measurement log buffer: %d\n",
+ ret);
+ return;
+ }
+
+ image->tpm_buffer = buffer;
+ image->tpm_buffer_addr = kbuf.mem;
+ image->tpm_buffer_size = size;
+}
+
+/**
+ * tpm_post_kexec - Make stored TPM log buffer available in of-tree
+ */
+static int __init tpm_post_kexec(void)
+{
+ struct property *newprop, *p;
+ struct device_node *np;
+ phys_addr_t phyaddr;
+ u32 oflogsize;
+ size_t size;
+ u64 unused;
+ int ret;
+
+ if (!IS_ENABLED(CONFIG_PPC64))
+ return 0;
+
+ np = of_find_node_by_name(NULL, "vtpm");
+ if (!np)
+ return 0;
+
+ if (!of_get_property(of_chosen, "linux,tpm-kexec-buffer", NULL)) {
+ /*
+ * linux,tpm-kexec-buffer may be missing on initial boot
+ * or if previous kernel didn't pass a buffer.
+ */
+ if (of_get_property(of_chosen, "linux,booted-from-kexec", NULL)) {
+ /* no buffer but kexec'd: remove 'linux,sml-base' */
+ ret = -EINVAL;
+ goto err_remove_sml_base;
+ }
+ return 0;
+ }
+
+ /*
+ * If any one of the following steps fails we remove linux,sml-base
+ * to invalidate the TPM log.
+ */
+ ret = tpm_get_kexec_buffer(&phyaddr, &size);
+ if (ret)
+ goto err_remove_kexec_buffer;
+
+ /* logsize must not have changed */
+ ret = of_tpm_get_sml_parameters(np, &unused, &oflogsize);
+ if (ret < 0)
+ goto err_free_memblock;
+ ret = -EINVAL;
+ if (oflogsize != size)
+ goto err_free_memblock;
+
+ /* replace linux,sml-base with new physical address of buffer */
+ ret = -ENOMEM;
+ newprop = kzalloc(sizeof(*newprop), GFP_KERNEL);
+ if (!newprop)
+ goto err_free_memblock;
+
+ newprop->name = kstrdup("linux,sml-base", GFP_KERNEL);
+ newprop->length = sizeof(phyaddr);
+ newprop->value = kmalloc(sizeof(phyaddr), GFP_KERNEL);
+ if (!newprop->name || !newprop->value)
+ goto err_free_newprop_struct;
+
+ if (of_property_match_string(np, "compatible", "IBM,vtpm") < 0 &&
+ of_property_match_string(np, "compatible", "IBM,vtpm20") < 0) {
+ ret = -ENODEV;
+ goto err_free_newprop_struct;
+ } else {
+ *(phys_addr_t *)newprop->value = phyaddr;
+ }
+
+ ret = of_update_property(np, newprop);
+ if (ret) {
+ pr_err("Could not update linux,sml-base with new address");
+ goto err_free_newprop_struct;
+ }
+
+ return 0;
+
+err_free_newprop_struct:
+ kfree(newprop->value);
+ kfree(newprop->name);
+ kfree(newprop);
+err_free_memblock:
+ memblock_phys_free((phys_addr_t)phyaddr, size);
+err_remove_kexec_buffer:
+ tpm_of_remove_kexec_buffer();
+err_remove_sml_base:
+ p = of_find_property(np, "linux,sml-base", NULL);
+ if (p)
+ of_remove_property(np, p);
+
+ return ret;
+}
+subsys_initcall(tpm_post_kexec);
+
/*
* of_kexec_alloc_and_setup_fdt - Alloc and setup a new Flattened Device Tree
*
@@ -482,6 +691,9 @@ void *of_kexec_alloc_and_setup_fdt(const struct kimage *image,
remove_ima_buffer(fdt, chosen_node);
ret = setup_ima_buffer(image, fdt, fdt_path_offset(fdt, "/chosen"));
+ remove_tpm_buffer(fdt, chosen_node);
+ ret = setup_tpm_buffer(image, fdt, fdt_path_offset(fdt, "/chosen"));
+
out:
if (ret) {
kvfree(fdt);
diff --git a/include/linux/kexec.h b/include/linux/kexec.h
index ce6536f1d269..f0940396c3d4 100644
--- a/include/linux/kexec.h
+++ b/include/linux/kexec.h
@@ -349,6 +349,12 @@ struct kimage {
void *elf_headers;
unsigned long elf_headers_sz;
unsigned long elf_load_addr;
+
+ /* Virtual address of TPM log buffer for kexec syscall */
+ void *tpm_buffer;
+
+ phys_addr_t tpm_buffer_addr;
+ size_t tpm_buffer_size;
};
/* kexec interface functions */
diff --git a/include/linux/of.h b/include/linux/of.h
index 20a4e7cb7afe..d6c776b22f47 100644
--- a/include/linux/of.h
+++ b/include/linux/of.h
@@ -100,6 +100,8 @@ struct of_reconfig_data {
struct property *old_prop;
};
+struct kimage;
+
/* initialize a node */
extern struct kobj_type of_node_ktype;
extern const struct fwnode_operations of_fwnode_ops;
@@ -436,11 +438,11 @@ int of_map_id(struct device_node *np, u32 id,
phys_addr_t of_dma_get_max_cpu_address(struct device_node *np);
-struct kimage;
void *of_kexec_alloc_and_setup_fdt(const struct kimage *image,
unsigned long initrd_load_addr,
unsigned long initrd_len,
const char *cmdline, size_t extra_fdt_size);
+void tpm_add_kexec_buffer(struct kimage *image);
#else /* CONFIG_OF */
static inline void of_core_init(void)
@@ -842,6 +844,10 @@ static inline phys_addr_t of_dma_get_max_cpu_address(struct device_node *np)
return PHYS_ADDR_MAX;
}
+static inline void tpm_add_kexec_buffer(struct kimage *image)
+{
+}
+
#define of_match_ptr(_ptr) NULL
#define of_match_node(_matches, _node) NULL
#endif /* CONFIG_OF */
diff --git a/kernel/kexec_file.c b/kernel/kexec_file.c
index 145321a5e798..d2e0751852dc 100644
--- a/kernel/kexec_file.c
+++ b/kernel/kexec_file.c
@@ -27,6 +27,7 @@
#include <linux/kernel_read_file.h>
#include <linux/syscalls.h>
#include <linux/vmalloc.h>
+#include <linux/of.h>
#include "kexec_internal.h"
static int kexec_calculate_store_digests(struct kimage *image);
@@ -137,6 +138,9 @@ void kimage_file_post_load_cleanup(struct kimage *image)
image->ima_buffer = NULL;
#endif /* CONFIG_IMA_KEXEC */
+ vfree(image->tpm_buffer);
+ image->tpm_buffer = NULL;
+
/* See if architecture has anything to cleanup post load */
arch_kimage_file_post_load_cleanup(image);
@@ -243,6 +247,8 @@ kimage_file_prepare_segments(struct kimage *image, int kernel_fd, int initrd_fd,
/* IMA needs to pass the measurement list to the next kernel. */
ima_add_kexec_buffer(image);
+ /* Pass the TPM measurement log to next kernel */
+ tpm_add_kexec_buffer(image);
/* Call arch image load handlers */
ldata = arch_kexec_kernel_image_load(image);
--
2.35.1
From: Vaibhav Jain <[email protected]>
Presently ima_get_kexec_buffer() doesn't check if the previous kernel's
ima-kexec-buffer lies outside the addressable memory range. This can result
in a kernel panic if the new kernel is booted with 'mem=X' arg and the
ima-kexec-buffer was allocated beyond that range by the previous kernel.
The panic is usually of the form below:
$ sudo kexec --initrd initrd vmlinux --append='mem=16G'
<snip>
BUG: Unable to handle kernel data access on read at 0xc000c01fff7f0000
Faulting instruction address: 0xc000000000837974
Oops: Kernel access of bad area, sig: 11 [#1]
<snip>
NIP [c000000000837974] ima_restore_measurement_list+0x94/0x6c0
LR [c00000000083b55c] ima_load_kexec_buffer+0xac/0x160
Call Trace:
[c00000000371fa80] [c00000000083b55c] ima_load_kexec_buffer+0xac/0x160
[c00000000371fb00] [c0000000020512c4] ima_init+0x80/0x108
[c00000000371fb70] [c0000000020514dc] init_ima+0x4c/0x120
[c00000000371fbf0] [c000000000012240] do_one_initcall+0x60/0x2c0
[c00000000371fcc0] [c000000002004ad0] kernel_init_freeable+0x344/0x3ec
[c00000000371fda0] [c0000000000128a4] kernel_init+0x34/0x1b0
[c00000000371fe10] [c00000000000ce64] ret_from_kernel_thread+0x5c/0x64
Instruction dump:
f92100b8 f92100c0 90e10090 910100a0 4182050c 282a0017 3bc00000 40810330
7c0802a6 fb610198 7c9b2378 f80101d0 <a1240000> 2c090001 40820614 e9240010
---[ end trace 0000000000000000 ]---
Fix this issue by checking returned PFN range of previous kernel's
ima-kexec-buffer with page_is_ram() to ensure correct memory bounds.
Fixes: 467d27824920 ("powerpc: ima: get the kexec buffer passed by the previous kernel")
Cc: Frank Rowand <[email protected]>
Cc: Prakhar Srivastava <[email protected]>
Cc: Lakshmi Ramasubramanian <[email protected]>
Cc: Thiago Jung Bauermann <[email protected]>
Cc: Rob Herring <[email protected]>
Cc: Ritesh Harjani <[email protected]>
Cc: Robin Murphy <[email protected]>
Signed-off-by: Vaibhav Jain <[email protected]>
Signed-off-by: Rob Herring <[email protected]>
Link: https://lore.kernel.org/r/[email protected]
---
drivers/of/kexec.c | 17 +++++++++++++++++
1 file changed, 17 insertions(+)
diff --git a/drivers/of/kexec.c b/drivers/of/kexec.c
index 8d374cc552be..91b04b04eec4 100644
--- a/drivers/of/kexec.c
+++ b/drivers/of/kexec.c
@@ -126,6 +126,7 @@ int ima_get_kexec_buffer(void **addr, size_t *size)
{
int ret, len;
unsigned long tmp_addr;
+ unsigned long start_pfn, end_pfn;
size_t tmp_size;
const void *prop;
@@ -140,6 +141,22 @@ int ima_get_kexec_buffer(void **addr, size_t *size)
if (ret)
return ret;
+ /* Do some sanity on the returned size for the ima-kexec buffer */
+ if (!tmp_size)
+ return -ENOENT;
+
+ /*
+ * Calculate the PFNs for the buffer and ensure
+ * they are with in addressable memory.
+ */
+ start_pfn = PHYS_PFN(tmp_addr);
+ end_pfn = PHYS_PFN(tmp_addr + tmp_size - 1);
+ if (!page_is_ram(start_pfn) || !page_is_ram(end_pfn)) {
+ pr_warn("IMA buffer at 0x%lx, size = 0x%zx beyond memory\n",
+ tmp_addr, tmp_size);
+ return -EINVAL;
+ }
+
*addr = __va(tmp_addr);
*size = tmp_size;
--
2.35.1
Refactor IMA buffer related functions to make them reusable for carrying
TPM logs across kexec.
Signed-off-by: Stefan Berger <[email protected]>
Cc: Rob Herring <[email protected]>
Cc: Frank Rowand <[email protected]>
Cc: Mimi Zohar <[email protected]>
---
v5:
- Rebased on Jonathan McDowell's commit "b69a2afd5afc x86/kexec: Carry
forward IMA measurement log on kexec"
v4:
- Move debug output into setup_buffer()
---
drivers/of/kexec.c | 125 ++++++++++++++++++++++++++-------------------
1 file changed, 73 insertions(+), 52 deletions(-)
diff --git a/drivers/of/kexec.c b/drivers/of/kexec.c
index 548dd5b1b5c1..404a86bb3978 100644
--- a/drivers/of/kexec.c
+++ b/drivers/of/kexec.c
@@ -117,45 +117,56 @@ static int do_get_kexec_buffer(const void *prop, int len, unsigned long *addr,
}
#ifdef CONFIG_HAVE_IMA_KEXEC
-/**
- * ima_get_kexec_buffer - get IMA buffer from the previous kernel
- * @addr: On successful return, set to point to the buffer contents.
- * @size: On successful return, set to the buffer size.
- *
- * Return: 0 on success, negative errno on error.
- */
-int __init ima_get_kexec_buffer(void **addr, size_t *size)
+static int get_kexec_buffer(const char *name, unsigned long *addr, size_t *size)
{
int ret, len;
- unsigned long tmp_addr;
unsigned long start_pfn, end_pfn;
- size_t tmp_size;
const void *prop;
- prop = of_get_property(of_chosen, "linux,ima-kexec-buffer", &len);
+ prop = of_get_property(of_chosen, name, &len);
if (!prop)
return -ENOENT;
- ret = do_get_kexec_buffer(prop, len, &tmp_addr, &tmp_size);
+ ret = do_get_kexec_buffer(prop, len, addr, size);
if (ret)
return ret;
- /* Do some sanity on the returned size for the ima-kexec buffer */
- if (!tmp_size)
+ /* Do some sanity on the returned size for the kexec buffer */
+ if (!*size)
return -ENOENT;
/*
* Calculate the PFNs for the buffer and ensure
* they are with in addressable memory.
*/
- start_pfn = PHYS_PFN(tmp_addr);
- end_pfn = PHYS_PFN(tmp_addr + tmp_size - 1);
+ start_pfn = PHYS_PFN(*addr);
+ end_pfn = PHYS_PFN(*addr + *size - 1);
if (!page_is_ram(start_pfn) || !page_is_ram(end_pfn)) {
- pr_warn("IMA buffer at 0x%lx, size = 0x%zx beyond memory\n",
- tmp_addr, tmp_size);
+ pr_warn("%s buffer at 0x%lx, size = 0x%zx beyond memory\n",
+ name, *addr, *size);
return -EINVAL;
}
+ return 0;
+}
+
+/**
+ * ima_get_kexec_buffer - get IMA buffer from the previous kernel
+ * @addr: On successful return, set to point to the buffer contents.
+ * @size: On successful return, set to the buffer size.
+ *
+ * Return: 0 on success, negative errno on error.
+ */
+int __init ima_get_kexec_buffer(void **addr, size_t *size)
+{
+ int ret;
+ unsigned long tmp_addr;
+ size_t tmp_size;
+
+ ret = get_kexec_buffer("linux,ima-kexec-buffer", &tmp_addr, &tmp_size);
+ if (ret)
+ return ret;
+
*addr = __va(tmp_addr);
*size = tmp_size;
@@ -188,72 +199,82 @@ int __init ima_free_kexec_buffer(void)
}
#endif
-/**
- * remove_ima_buffer - remove the IMA buffer property and reservation from @fdt
- *
- * @fdt: Flattened Device Tree to update
- * @chosen_node: Offset to the chosen node in the device tree
- *
- * The IMA measurement buffer is of no use to a subsequent kernel, so we always
- * remove it from the device tree.
- */
-static void remove_ima_buffer(void *fdt, int chosen_node)
+static int remove_buffer(void *fdt, int chosen_node, const char *name)
{
int ret, len;
unsigned long addr;
size_t size;
const void *prop;
- if (!IS_ENABLED(CONFIG_HAVE_IMA_KEXEC))
- return;
-
- prop = fdt_getprop(fdt, chosen_node, "linux,ima-kexec-buffer", &len);
+ prop = fdt_getprop(fdt, chosen_node, name, &len);
if (!prop)
- return;
+ return -ENOENT;
ret = do_get_kexec_buffer(prop, len, &addr, &size);
- fdt_delprop(fdt, chosen_node, "linux,ima-kexec-buffer");
+ fdt_delprop(fdt, chosen_node, name);
if (ret)
- return;
+ return ret;
ret = fdt_find_and_del_mem_rsv(fdt, addr, size);
if (!ret)
- pr_debug("Removed old IMA buffer reservation.\n");
+ pr_debug("Remove old %s buffer reserveration", name);
+ return ret;
}
-#ifdef CONFIG_IMA_KEXEC
/**
- * setup_ima_buffer - add IMA buffer information to the fdt
- * @image: kexec image being loaded.
- * @fdt: Flattened device tree for the next kernel.
- * @chosen_node: Offset to the chosen node.
+ * remove_ima_buffer - remove the IMA buffer property and reservation from @fdt
*
- * Return: 0 on success, or negative errno on error.
+ * @fdt: Flattened Device Tree to update
+ * @chosen_node: Offset to the chosen node in the device tree
+ *
+ * The IMA measurement buffer is of no use to a subsequent kernel, so we always
+ * remove it from the device tree.
*/
-static int setup_ima_buffer(const struct kimage *image, void *fdt,
- int chosen_node)
+static void remove_ima_buffer(void *fdt, int chosen_node)
+{
+ if (!IS_ENABLED(CONFIG_HAVE_IMA_KEXEC))
+ return;
+
+ remove_buffer(fdt, chosen_node, "linux,ima-kexec-buffer");
+}
+
+#ifdef CONFIG_IMA_KEXEC
+static int setup_buffer(void *fdt, int chosen_node, const char *name,
+ phys_addr_t addr, size_t size)
{
int ret;
- if (!image->ima_buffer_size)
+ if (!size)
return 0;
ret = fdt_appendprop_addrrange(fdt, 0, chosen_node,
- "linux,ima-kexec-buffer",
- image->ima_buffer_addr,
- image->ima_buffer_size);
+ name, addr, size);
if (ret < 0)
return -EINVAL;
- ret = fdt_add_mem_rsv(fdt, image->ima_buffer_addr,
- image->ima_buffer_size);
+ ret = fdt_add_mem_rsv(fdt, addr, size);
if (ret)
return -EINVAL;
- pr_debug("IMA buffer at 0x%pa, size = 0x%zx\n",
- &image->ima_buffer_addr, image->ima_buffer_size);
+ pr_debug("%s at 0x%pa, size = 0x%zx\n", name, &addr, size);
return 0;
+
+}
+
+/**
+ * setup_ima_buffer - add IMA buffer information to the fdt
+ * @image: kexec image being loaded.
+ * @fdt: Flattened device tree for the next kernel.
+ * @chosen_node: Offset to the chosen node.
+ *
+ * Return: 0 on success, or negative errno on error.
+ */
+static int setup_ima_buffer(const struct kimage *image, void *fdt,
+ int chosen_node)
+{
+ return setup_buffer(fdt, chosen_node, "linux,ima-kexec-buffer",
+ image->ima_buffer_addr, image->ima_buffer_size);
}
#else /* CONFIG_IMA_KEXEC */
static inline int setup_ima_buffer(const struct kimage *image, void *fdt,
--
2.35.1
From: Jonathan McDowell <[email protected]>
On kexec file load, the Integrity Measurement Architecture (IMA)
subsystem may verify the IMA signature of the kernel and initramfs, and
measure it. The command line parameters passed to the kernel in the
kexec call may also be measured by IMA.
A remote attestation service can verify a TPM quote based on the TPM
event log, the IMA measurement list and the TPM PCR data. This can
be achieved only if the IMA measurement log is carried over from the
current kernel to the next kernel across the kexec call.
PowerPC and ARM64 both achieve this using device tree with a
"linux,ima-kexec-buffer" node. x86 platforms generally don't make use of
device tree, so use the setup_data mechanism to pass the IMA buffer to
the new kernel.
Signed-off-by: Jonathan McDowell <[email protected]>
Signed-off-by: Borislav Petkov <[email protected]>
Reviewed-by: Mimi Zohar <[email protected]> # IMA function definitions
Link: https://lore.kernel.org/r/YmKyvlF3my1yWTvK@noodles-fedora-PC23Y6EG
---
arch/x86/Kconfig | 1 +
arch/x86/include/uapi/asm/bootparam.h | 9 ++++
arch/x86/kernel/e820.c | 6 +--
arch/x86/kernel/kexec-bzimage64.c | 42 +++++++++++++++++-
arch/x86/kernel/setup.c | 63 +++++++++++++++++++++++++++
drivers/of/kexec.c | 13 +++---
include/linux/ima.h | 5 +++
include/linux/of.h | 2 -
security/integrity/ima/ima_kexec.c | 2 +-
9 files changed, 127 insertions(+), 16 deletions(-)
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index be0b95e51df6..670e0edc074f 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -2033,6 +2033,7 @@ config KEXEC_FILE
bool "kexec file based system call"
select KEXEC_CORE
select BUILD_BIN2C
+ select HAVE_IMA_KEXEC if IMA
depends on X86_64
depends on CRYPTO=y
depends on CRYPTO_SHA256=y
diff --git a/arch/x86/include/uapi/asm/bootparam.h b/arch/x86/include/uapi/asm/bootparam.h
index bea5cdcdf532..ca0796ac4403 100644
--- a/arch/x86/include/uapi/asm/bootparam.h
+++ b/arch/x86/include/uapi/asm/bootparam.h
@@ -11,6 +11,7 @@
#define SETUP_APPLE_PROPERTIES 5
#define SETUP_JAILHOUSE 6
#define SETUP_CC_BLOB 7
+#define SETUP_IMA 8
#define SETUP_INDIRECT (1<<31)
@@ -172,6 +173,14 @@ struct jailhouse_setup_data {
} __attribute__((packed)) v2;
} __attribute__((packed));
+/*
+ * IMA buffer setup data information from the previous kernel during kexec
+ */
+struct ima_setup_data {
+ __u64 addr;
+ __u64 size;
+} __attribute__((packed));
+
/* The so-called "zeropage" */
struct boot_params {
struct screen_info screen_info; /* 0x000 */
diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c
index f267205f2d5a..9dac24680ff8 100644
--- a/arch/x86/kernel/e820.c
+++ b/arch/x86/kernel/e820.c
@@ -1017,10 +1017,10 @@ void __init e820__reserve_setup_data(void)
e820__range_update(pa_data, sizeof(*data)+data->len, E820_TYPE_RAM, E820_TYPE_RESERVED_KERN);
/*
- * SETUP_EFI is supplied by kexec and does not need to be
- * reserved.
+ * SETUP_EFI and SETUP_IMA are supplied by kexec and do not need
+ * to be reserved.
*/
- if (data->type != SETUP_EFI)
+ if (data->type != SETUP_EFI && data->type != SETUP_IMA)
e820__range_update_kexec(pa_data,
sizeof(*data) + data->len,
E820_TYPE_RAM, E820_TYPE_RESERVED_KERN);
diff --git a/arch/x86/kernel/kexec-bzimage64.c b/arch/x86/kernel/kexec-bzimage64.c
index 170d0fd68b1f..c63974e94272 100644
--- a/arch/x86/kernel/kexec-bzimage64.c
+++ b/arch/x86/kernel/kexec-bzimage64.c
@@ -186,11 +186,38 @@ setup_efi_state(struct boot_params *params, unsigned long params_load_addr,
}
#endif /* CONFIG_EFI */
+static void
+setup_ima_state(const struct kimage *image, struct boot_params *params,
+ unsigned long params_load_addr,
+ unsigned int ima_setup_data_offset)
+{
+#ifdef CONFIG_IMA_KEXEC
+ struct setup_data *sd = (void *)params + ima_setup_data_offset;
+ unsigned long setup_data_phys;
+ struct ima_setup_data *ima;
+
+ if (!image->ima_buffer_size)
+ return;
+
+ sd->type = SETUP_IMA;
+ sd->len = sizeof(*ima);
+
+ ima = (void *)sd + sizeof(struct setup_data);
+ ima->addr = image->ima_buffer_addr;
+ ima->size = image->ima_buffer_size;
+
+ /* Add setup data */
+ setup_data_phys = params_load_addr + ima_setup_data_offset;
+ sd->next = params->hdr.setup_data;
+ params->hdr.setup_data = setup_data_phys;
+#endif /* CONFIG_IMA_KEXEC */
+}
+
static int
setup_boot_parameters(struct kimage *image, struct boot_params *params,
unsigned long params_load_addr,
unsigned int efi_map_offset, unsigned int efi_map_sz,
- unsigned int efi_setup_data_offset)
+ unsigned int setup_data_offset)
{
unsigned int nr_e820_entries;
unsigned long long mem_k, start, end;
@@ -245,8 +272,15 @@ setup_boot_parameters(struct kimage *image, struct boot_params *params,
#ifdef CONFIG_EFI
/* Setup EFI state */
setup_efi_state(params, params_load_addr, efi_map_offset, efi_map_sz,
- efi_setup_data_offset);
+ setup_data_offset);
+ setup_data_offset += sizeof(struct setup_data) +
+ sizeof(struct efi_setup_data);
#endif
+
+ /* Setup IMA log buffer state */
+ setup_ima_state(image, params, params_load_addr,
+ setup_data_offset);
+
/* Setup EDD info */
memcpy(params->eddbuf, boot_params.eddbuf,
EDDMAXNR * sizeof(struct edd_info));
@@ -403,6 +437,10 @@ static void *bzImage64_load(struct kimage *image, char *kernel,
sizeof(struct setup_data) +
sizeof(struct efi_setup_data);
+ if (IS_ENABLED(CONFIG_IMA_KEXEC))
+ kbuf.bufsz += sizeof(struct setup_data) +
+ sizeof(struct ima_setup_data);
+
params = kzalloc(kbuf.bufsz, GFP_KERNEL);
if (!params)
return ERR_PTR(-ENOMEM);
diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c
index bd6c6fd373ae..53f863f28b4c 100644
--- a/arch/x86/kernel/setup.c
+++ b/arch/x86/kernel/setup.c
@@ -11,6 +11,7 @@
#include <linux/dma-map-ops.h>
#include <linux/dmi.h>
#include <linux/efi.h>
+#include <linux/ima.h>
#include <linux/init_ohci1394_dma.h>
#include <linux/initrd.h>
#include <linux/iscsi_ibft.h>
@@ -140,6 +141,11 @@ __visible unsigned long mmu_cr4_features __ro_after_init;
__visible unsigned long mmu_cr4_features __ro_after_init = X86_CR4_PAE;
#endif
+#ifdef CONFIG_IMA
+static phys_addr_t ima_kexec_buffer_phys;
+static size_t ima_kexec_buffer_size;
+#endif
+
/* Boot loader ID and version as integers, for the benefit of proc_dointvec */
int bootloader_type, bootloader_version;
@@ -330,6 +336,60 @@ static void __init reserve_initrd(void)
}
#endif /* CONFIG_BLK_DEV_INITRD */
+static void __init add_early_ima_buffer(u64 phys_addr)
+{
+#ifdef CONFIG_IMA
+ struct ima_setup_data *data;
+
+ data = early_memremap(phys_addr + sizeof(struct setup_data), sizeof(*data));
+ if (!data) {
+ pr_warn("setup: failed to memremap ima_setup_data entry\n");
+ return;
+ }
+
+ if (data->size) {
+ memblock_reserve(data->addr, data->size);
+ ima_kexec_buffer_phys = data->addr;
+ ima_kexec_buffer_size = data->size;
+ }
+
+ early_memunmap(data, sizeof(*data));
+#else
+ pr_warn("Passed IMA kexec data, but CONFIG_IMA not set. Ignoring.\n");
+#endif
+}
+
+#if defined(CONFIG_HAVE_IMA_KEXEC) && !defined(CONFIG_OF_FLATTREE)
+int __init ima_free_kexec_buffer(void)
+{
+ int rc;
+
+ if (!ima_kexec_buffer_size)
+ return -ENOENT;
+
+ rc = memblock_phys_free(ima_kexec_buffer_phys,
+ ima_kexec_buffer_size);
+ if (rc)
+ return rc;
+
+ ima_kexec_buffer_phys = 0;
+ ima_kexec_buffer_size = 0;
+
+ return 0;
+}
+
+int __init ima_get_kexec_buffer(void **addr, size_t *size)
+{
+ if (!ima_kexec_buffer_size)
+ return -ENOENT;
+
+ *addr = __va(ima_kexec_buffer_phys);
+ *size = ima_kexec_buffer_size;
+
+ return 0;
+}
+#endif
+
static void __init parse_setup_data(void)
{
struct setup_data *data;
@@ -355,6 +415,9 @@ static void __init parse_setup_data(void)
case SETUP_EFI:
parse_efi_setup(pa_data, data_len);
break;
+ case SETUP_IMA:
+ add_early_ima_buffer(pa_data);
+ break;
default:
break;
}
diff --git a/drivers/of/kexec.c b/drivers/of/kexec.c
index c4f9b6655a2e..548dd5b1b5c1 100644
--- a/drivers/of/kexec.c
+++ b/drivers/of/kexec.c
@@ -9,6 +9,7 @@
* Copyright (C) 2016 IBM Corporation
*/
+#include <linux/ima.h>
#include <linux/kernel.h>
#include <linux/kexec.h>
#include <linux/memblock.h>
@@ -115,6 +116,7 @@ static int do_get_kexec_buffer(const void *prop, int len, unsigned long *addr,
return 0;
}
+#ifdef CONFIG_HAVE_IMA_KEXEC
/**
* ima_get_kexec_buffer - get IMA buffer from the previous kernel
* @addr: On successful return, set to point to the buffer contents.
@@ -122,7 +124,7 @@ static int do_get_kexec_buffer(const void *prop, int len, unsigned long *addr,
*
* Return: 0 on success, negative errno on error.
*/
-int ima_get_kexec_buffer(void **addr, size_t *size)
+int __init ima_get_kexec_buffer(void **addr, size_t *size)
{
int ret, len;
unsigned long tmp_addr;
@@ -130,9 +132,6 @@ int ima_get_kexec_buffer(void **addr, size_t *size)
size_t tmp_size;
const void *prop;
- if (!IS_ENABLED(CONFIG_HAVE_IMA_KEXEC))
- return -ENOTSUPP;
-
prop = of_get_property(of_chosen, "linux,ima-kexec-buffer", &len);
if (!prop)
return -ENOENT;
@@ -166,16 +165,13 @@ int ima_get_kexec_buffer(void **addr, size_t *size)
/**
* ima_free_kexec_buffer - free memory used by the IMA buffer
*/
-int ima_free_kexec_buffer(void)
+int __init ima_free_kexec_buffer(void)
{
int ret;
unsigned long addr;
size_t size;
struct property *prop;
- if (!IS_ENABLED(CONFIG_HAVE_IMA_KEXEC))
- return -ENOTSUPP;
-
prop = of_find_property(of_chosen, "linux,ima-kexec-buffer", NULL);
if (!prop)
return -ENOENT;
@@ -190,6 +186,7 @@ int ima_free_kexec_buffer(void)
return memblock_phys_free(addr, size);
}
+#endif
/**
* remove_ima_buffer - remove the IMA buffer property and reservation from @fdt
diff --git a/include/linux/ima.h b/include/linux/ima.h
index 426b1744215e..81708ca0ebc7 100644
--- a/include/linux/ima.h
+++ b/include/linux/ima.h
@@ -140,6 +140,11 @@ static inline int ima_measure_critical_data(const char *event_label,
#endif /* CONFIG_IMA */
+#ifdef CONFIG_HAVE_IMA_KEXEC
+int __init ima_free_kexec_buffer(void);
+int __init ima_get_kexec_buffer(void **addr, size_t *size);
+#endif
+
#ifdef CONFIG_IMA_SECURE_AND_OR_TRUSTED_BOOT
extern bool arch_ima_get_secureboot(void);
extern const char * const *arch_get_ima_policy(void);
diff --git a/include/linux/of.h b/include/linux/of.h
index f0a5d6b10c5a..20a4e7cb7afe 100644
--- a/include/linux/of.h
+++ b/include/linux/of.h
@@ -441,8 +441,6 @@ void *of_kexec_alloc_and_setup_fdt(const struct kimage *image,
unsigned long initrd_load_addr,
unsigned long initrd_len,
const char *cmdline, size_t extra_fdt_size);
-int ima_get_kexec_buffer(void **addr, size_t *size);
-int ima_free_kexec_buffer(void);
#else /* CONFIG_OF */
static inline void of_core_init(void)
diff --git a/security/integrity/ima/ima_kexec.c b/security/integrity/ima/ima_kexec.c
index 13753136f03f..419dc405c831 100644
--- a/security/integrity/ima/ima_kexec.c
+++ b/security/integrity/ima/ima_kexec.c
@@ -137,7 +137,7 @@ void ima_add_kexec_buffer(struct kimage *image)
/*
* Restore the measurement list from the previous kernel.
*/
-void ima_load_kexec_buffer(void)
+void __init ima_load_kexec_buffer(void)
{
void *kexec_buffer = NULL;
size_t kexec_buffer_size = 0;
--
2.35.1
Hi Stefan,
Thank you for the patch! Yet something to improve:
[auto build test ERROR on 03c765b0e3b4cb5063276b086c76f7a612856a9a]
url: https://github.com/intel-lab-lkp/linux/commits/Stefan-Berger/tpm-Preserve-TPM-measurement-log-across-kexec-ppc64/20220706-232658
base: 03c765b0e3b4cb5063276b086c76f7a612856a9a
config: s390-randconfig-r015-20220706 (https://download.01.org/0day-ci/archive/20220707/[email protected]/config)
compiler: clang version 15.0.0 (https://github.com/llvm/llvm-project f553287b588916de09c66e3e32bf75e5060f967f)
reproduce (this is a W=1 build):
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
chmod +x ~/bin/make.cross
# install s390 cross compiling tool for clang build
# apt-get install binutils-s390x-linux-gnu
# https://github.com/intel-lab-lkp/linux/commit/d5d97eac90ae272727cc71409e88360b7024b343
git remote add linux-review https://github.com/intel-lab-lkp/linux
git fetch --no-tags linux-review Stefan-Berger/tpm-Preserve-TPM-measurement-log-across-kexec-ppc64/20220706-232658
git checkout d5d97eac90ae272727cc71409e88360b7024b343
# save the config file
mkdir build_dir && cp config build_dir/.config
COMPILER_INSTALL_PATH=$HOME/0day COMPILER=clang make.cross W=1 O=build_dir ARCH=s390 SHELL=/bin/bash
If you fix the issue, kindly add following tag where applicable
Reported-by: kernel test robot <[email protected]>
All errors (new ones prefixed by >>):
/opt/cross/gcc-11.3.0-nolibc/s390x-linux/bin/s390x-linux-ld: kernel/kexec_file.o: in function `__se_sys_kexec_file_load':
>> kexec_file.c:(.text+0xee2): undefined reference to `tpm_add_kexec_buffer'
/opt/cross/gcc-11.3.0-nolibc/s390x-linux/bin/s390x-linux-ld: drivers/dma/fsl-edma.o: in function `fsl_edma_probe':
fsl-edma.c:(.text+0x1e4): undefined reference to `devm_ioremap_resource'
/opt/cross/gcc-11.3.0-nolibc/s390x-linux/bin/s390x-linux-ld: fsl-edma.c:(.text+0x32a): undefined reference to `devm_ioremap_resource'
/opt/cross/gcc-11.3.0-nolibc/s390x-linux/bin/s390x-linux-ld: drivers/dma/idma64.o: in function `idma64_platform_probe':
idma64.c:(.text+0xbe): undefined reference to `devm_ioremap_resource'
/opt/cross/gcc-11.3.0-nolibc/s390x-linux/bin/s390x-linux-ld: drivers/crypto/ccree/cc_driver.o: in function `ccree_probe':
cc_driver.c:(.text+0xd4c): undefined reference to `devm_ioremap_resource'
/opt/cross/gcc-11.3.0-nolibc/s390x-linux/bin/s390x-linux-ld: drivers/crypto/ccree/cc_debugfs.o: in function `cc_debugfs_init':
cc_debugfs.c:(.text+0x1b0): undefined reference to `debugfs_create_regset32'
/opt/cross/gcc-11.3.0-nolibc/s390x-linux/bin/s390x-linux-ld: cc_debugfs.c:(.text+0x390): undefined reference to `debugfs_create_regset32'
--
0-DAY CI Kernel Test Service
https://01.org/lkp
On Wed, Jul 06, 2022 at 11:23:28AM -0400, Stefan Berger wrote:
> Refactor IMA buffer related functions to make them reusable for carrying
> TPM logs across kexec.
>
> Signed-off-by: Stefan Berger <[email protected]>
> Cc: Rob Herring <[email protected]>
> Cc: Frank Rowand <[email protected]>
> Cc: Mimi Zohar <[email protected]>
>
> ---
> v5:
> - Rebased on Jonathan McDowell's commit "b69a2afd5afc x86/kexec: Carry
> forward IMA measurement log on kexec"
> v4:
> - Move debug output into setup_buffer()
> ---
> drivers/of/kexec.c | 125 ++++++++++++++++++++++++++-------------------
> 1 file changed, 73 insertions(+), 52 deletions(-)
>
> diff --git a/drivers/of/kexec.c b/drivers/of/kexec.c
> index 548dd5b1b5c1..404a86bb3978 100644
> --- a/drivers/of/kexec.c
> +++ b/drivers/of/kexec.c
> @@ -117,45 +117,56 @@ static int do_get_kexec_buffer(const void *prop, int len, unsigned long *addr,
> }
>
> #ifdef CONFIG_HAVE_IMA_KEXEC
> -/**
> - * ima_get_kexec_buffer - get IMA buffer from the previous kernel
> - * @addr: On successful return, set to point to the buffer contents.
> - * @size: On successful return, set to the buffer size.
> - *
> - * Return: 0 on success, negative errno on error.
> - */
> -int __init ima_get_kexec_buffer(void **addr, size_t *size)
> +static int get_kexec_buffer(const char *name, unsigned long *addr, size_t *size)
I think this new function can + should be marked __init as well; it's
only called from the existing IMA functions and your new
tpm_get_kexec_buffer which is already __init.
> {
> int ret, len;
> - unsigned long tmp_addr;
> unsigned long start_pfn, end_pfn;
> - size_t tmp_size;
> const void *prop;
>
> - prop = of_get_property(of_chosen, "linux,ima-kexec-buffer", &len);
> + prop = of_get_property(of_chosen, name, &len);
> if (!prop)
> return -ENOENT;
>
> - ret = do_get_kexec_buffer(prop, len, &tmp_addr, &tmp_size);
> + ret = do_get_kexec_buffer(prop, len, addr, size);
> if (ret)
> return ret;
>
> - /* Do some sanity on the returned size for the ima-kexec buffer */
> - if (!tmp_size)
> + /* Do some sanity on the returned size for the kexec buffer */
> + if (!*size)
> return -ENOENT;
>
> /*
> * Calculate the PFNs for the buffer and ensure
> * they are with in addressable memory.
> */
> - start_pfn = PHYS_PFN(tmp_addr);
> - end_pfn = PHYS_PFN(tmp_addr + tmp_size - 1);
> + start_pfn = PHYS_PFN(*addr);
> + end_pfn = PHYS_PFN(*addr + *size - 1);
> if (!page_is_ram(start_pfn) || !page_is_ram(end_pfn)) {
> - pr_warn("IMA buffer at 0x%lx, size = 0x%zx beyond memory\n",
> - tmp_addr, tmp_size);
> + pr_warn("%s buffer at 0x%lx, size = 0x%zx beyond memory\n",
> + name, *addr, *size);
> return -EINVAL;
> }
>
> + return 0;
> +}
> +
> +/**
> + * ima_get_kexec_buffer - get IMA buffer from the previous kernel
> + * @addr: On successful return, set to point to the buffer contents.
> + * @size: On successful return, set to the buffer size.
> + *
> + * Return: 0 on success, negative errno on error.
> + */
> +int __init ima_get_kexec_buffer(void **addr, size_t *size)
> +{
> + int ret;
> + unsigned long tmp_addr;
> + size_t tmp_size;
> +
> + ret = get_kexec_buffer("linux,ima-kexec-buffer", &tmp_addr, &tmp_size);
> + if (ret)
> + return ret;
> +
> *addr = __va(tmp_addr);
> *size = tmp_size;
>
> @@ -188,72 +199,82 @@ int __init ima_free_kexec_buffer(void)
> }
> #endif
>
> -/**
> - * remove_ima_buffer - remove the IMA buffer property and reservation from @fdt
> - *
> - * @fdt: Flattened Device Tree to update
> - * @chosen_node: Offset to the chosen node in the device tree
> - *
> - * The IMA measurement buffer is of no use to a subsequent kernel, so we always
> - * remove it from the device tree.
> - */
> -static void remove_ima_buffer(void *fdt, int chosen_node)
> +static int remove_buffer(void *fdt, int chosen_node, const char *name)
> {
> int ret, len;
> unsigned long addr;
> size_t size;
> const void *prop;
>
> - if (!IS_ENABLED(CONFIG_HAVE_IMA_KEXEC))
> - return;
> -
> - prop = fdt_getprop(fdt, chosen_node, "linux,ima-kexec-buffer", &len);
> + prop = fdt_getprop(fdt, chosen_node, name, &len);
> if (!prop)
> - return;
> + return -ENOENT;
>
> ret = do_get_kexec_buffer(prop, len, &addr, &size);
> - fdt_delprop(fdt, chosen_node, "linux,ima-kexec-buffer");
> + fdt_delprop(fdt, chosen_node, name);
> if (ret)
> - return;
> + return ret;
>
> ret = fdt_find_and_del_mem_rsv(fdt, addr, size);
> if (!ret)
> - pr_debug("Removed old IMA buffer reservation.\n");
> + pr_debug("Remove old %s buffer reserveration", name);
> + return ret;
> }
>
> -#ifdef CONFIG_IMA_KEXEC
> /**
> - * setup_ima_buffer - add IMA buffer information to the fdt
> - * @image: kexec image being loaded.
> - * @fdt: Flattened device tree for the next kernel.
> - * @chosen_node: Offset to the chosen node.
> + * remove_ima_buffer - remove the IMA buffer property and reservation from @fdt
> *
> - * Return: 0 on success, or negative errno on error.
> + * @fdt: Flattened Device Tree to update
> + * @chosen_node: Offset to the chosen node in the device tree
> + *
> + * The IMA measurement buffer is of no use to a subsequent kernel, so we always
> + * remove it from the device tree.
> */
> -static int setup_ima_buffer(const struct kimage *image, void *fdt,
> - int chosen_node)
> +static void remove_ima_buffer(void *fdt, int chosen_node)
> +{
> + if (!IS_ENABLED(CONFIG_HAVE_IMA_KEXEC))
> + return;
> +
> + remove_buffer(fdt, chosen_node, "linux,ima-kexec-buffer");
> +}
> +
> +#ifdef CONFIG_IMA_KEXEC
> +static int setup_buffer(void *fdt, int chosen_node, const char *name,
> + phys_addr_t addr, size_t size)
> {
> int ret;
>
> - if (!image->ima_buffer_size)
> + if (!size)
> return 0;
>
> ret = fdt_appendprop_addrrange(fdt, 0, chosen_node,
> - "linux,ima-kexec-buffer",
> - image->ima_buffer_addr,
> - image->ima_buffer_size);
> + name, addr, size);
> if (ret < 0)
> return -EINVAL;
>
> - ret = fdt_add_mem_rsv(fdt, image->ima_buffer_addr,
> - image->ima_buffer_size);
> + ret = fdt_add_mem_rsv(fdt, addr, size);
> if (ret)
> return -EINVAL;
>
> - pr_debug("IMA buffer at 0x%pa, size = 0x%zx\n",
> - &image->ima_buffer_addr, image->ima_buffer_size);
> + pr_debug("%s at 0x%pa, size = 0x%zx\n", name, &addr, size);
>
> return 0;
> +
> +}
> +
> +/**
> + * setup_ima_buffer - add IMA buffer information to the fdt
> + * @image: kexec image being loaded.
> + * @fdt: Flattened device tree for the next kernel.
> + * @chosen_node: Offset to the chosen node.
> + *
> + * Return: 0 on success, or negative errno on error.
> + */
> +static int setup_ima_buffer(const struct kimage *image, void *fdt,
> + int chosen_node)
> +{
> + return setup_buffer(fdt, chosen_node, "linux,ima-kexec-buffer",
> + image->ima_buffer_addr, image->ima_buffer_size);
> }
> #else /* CONFIG_IMA_KEXEC */
> static inline int setup_ima_buffer(const struct kimage *image, void *fdt,
> --
> 2.35.1
>
On 7/7/22 10:47, Jonathan McDowell wrote:
> On Wed, Jul 06, 2022 at 11:23:28AM -0400, Stefan Berger wrote:
>> Refactor IMA buffer related functions to make them reusable for carrying
>> TPM logs across kexec.
>>
>> Signed-off-by: Stefan Berger <[email protected]>
>> Cc: Rob Herring <[email protected]>
>> Cc: Frank Rowand <[email protected]>
>> Cc: Mimi Zohar <[email protected]>
>>
>> ---
>> v5:
>> - Rebased on Jonathan McDowell's commit "b69a2afd5afc x86/kexec: Carry
>> forward IMA measurement log on kexec"
>> v4:
>> - Move debug output into setup_buffer()
>> ---
>> drivers/of/kexec.c | 125 ++++++++++++++++++++++++++-------------------
>> 1 file changed, 73 insertions(+), 52 deletions(-)
>>
>> diff --git a/drivers/of/kexec.c b/drivers/of/kexec.c
>> index 548dd5b1b5c1..404a86bb3978 100644
>> --- a/drivers/of/kexec.c
>> +++ b/drivers/of/kexec.c
>> @@ -117,45 +117,56 @@ static int do_get_kexec_buffer(const void *prop, int len, unsigned long *addr,
>> }
>>
>> #ifdef CONFIG_HAVE_IMA_KEXEC
>> -/**
>> - * ima_get_kexec_buffer - get IMA buffer from the previous kernel
>> - * @addr: On successful return, set to point to the buffer contents.
>> - * @size: On successful return, set to the buffer size.
>> - *
>> - * Return: 0 on success, negative errno on error.
>> - */
>> -int __init ima_get_kexec_buffer(void **addr, size_t *size)
>> +static int get_kexec_buffer(const char *name, unsigned long *addr, size_t *size)
>
> I think this new function can + should be marked __init as well; it's
> only called from the existing IMA functions and your new
> tpm_get_kexec_buffer which is already __init.
Yes, right. Thanks.
Stefan
On Wed, Jul 06, 2022 at 11:23:27AM -0400, Stefan Berger wrote:
> Simplify tpm_read_log_of() by moving reusable parts of the code into
> an inline function that makes it commonly available so it can be
> used also for kexec support. Call the new of_tpm_get_sml_parameters()
> function from the TPM Open Firmware driver.
>
> Signed-off-by: Stefan Berger <[email protected]>
> Cc: Jarkko Sakkinen <[email protected]>
> Cc: Jason Gunthorpe <[email protected]>
> Cc: Rob Herring <[email protected]>
> Cc: Frank Rowand <[email protected]>
>
> ---
> v4:
> - converted to inline function
> ---
> drivers/char/tpm/eventlog/of.c | 31 +++++--------------------------
> include/linux/tpm.h | 27 +++++++++++++++++++++++++++
> 2 files changed, 32 insertions(+), 26 deletions(-)
>
> diff --git a/drivers/char/tpm/eventlog/of.c b/drivers/char/tpm/eventlog/of.c
> index a9ce66d09a75..f9462d19632e 100644
> --- a/drivers/char/tpm/eventlog/of.c
> +++ b/drivers/char/tpm/eventlog/of.c
> @@ -12,6 +12,7 @@
>
> #include <linux/slab.h>
> #include <linux/of.h>
> +#include <linux/tpm.h>
> #include <linux/tpm_eventlog.h>
>
> #include "../tpm.h"
> @@ -20,11 +21,10 @@
> int tpm_read_log_of(struct tpm_chip *chip)
> {
> struct device_node *np;
> - const u32 *sizep;
> - const u64 *basep;
> struct tpm_bios_log *log;
> u32 size;
> u64 base;
> + int ret;
>
> log = &chip->log;
> if (chip->dev.parent && chip->dev.parent->of_node)
> @@ -35,30 +35,9 @@ int tpm_read_log_of(struct tpm_chip *chip)
> if (of_property_read_bool(np, "powered-while-suspended"))
> chip->flags |= TPM_CHIP_FLAG_ALWAYS_POWERED;
>
> - sizep = of_get_property(np, "linux,sml-size", NULL);
> - basep = of_get_property(np, "linux,sml-base", NULL);
> - if (sizep == NULL && basep == NULL)
> - return -ENODEV;
> - if (sizep == NULL || basep == NULL)
> - return -EIO;
> -
> - /*
> - * For both vtpm/tpm, firmware has log addr and log size in big
> - * endian format. But in case of vtpm, there is a method called
> - * sml-handover which is run during kernel init even before
> - * device tree is setup. This sml-handover function takes care
> - * of endianness and writes to sml-base and sml-size in little
> - * endian format. For this reason, vtpm doesn't need conversion
> - * but physical tpm needs the conversion.
> - */
> - if (of_property_match_string(np, "compatible", "IBM,vtpm") < 0 &&
> - of_property_match_string(np, "compatible", "IBM,vtpm20") < 0) {
> - size = be32_to_cpup((__force __be32 *)sizep);
> - base = be64_to_cpup((__force __be64 *)basep);
> - } else {
> - size = *sizep;
> - base = *basep;
> - }
> + ret = of_tpm_get_sml_parameters(np, &base, &size);
> + if (ret < 0)
> + return ret;
>
> if (size == 0) {
> dev_warn(&chip->dev, "%s: Event log area empty\n", __func__);
> diff --git a/include/linux/tpm.h b/include/linux/tpm.h
> index dfeb25a0362d..b3dff255bc58 100644
> --- a/include/linux/tpm.h
> +++ b/include/linux/tpm.h
> @@ -460,4 +460,31 @@ static inline struct tpm_chip *tpm_default_chip(void)
> return NULL;
> }
> #endif
> +
> +#ifdef CONFIG_OF
> +static inline int of_tpm_get_sml_parameters(struct device_node *np,
> + u64 *base, u32 *size)
> +{
> + const u32 *sizep;
> + const u64 *basep;
> +
> + sizep = of_get_property(np, "linux,sml-size", NULL);
> + basep = of_get_property(np, "linux,sml-base", NULL);
> + if (sizep == NULL && basep == NULL)
> + return -ENODEV;
> + if (sizep == NULL || basep == NULL)
> + return -EIO;
> +
> + if (of_property_match_string(np, "compatible", "IBM,vtpm") < 0 &&
> + of_property_match_string(np, "compatible", "IBM,vtpm20") < 0) {
> + *size = be32_to_cpup((__force __be32 *)sizep);
> + *base = be64_to_cpup((__force __be64 *)basep);
> + } else {
> + *size = *sizep;
> + *base = *basep;
> + }
> + return 0;
> +}
> +#endif
> +
> #endif
> --
> 2.35.1
>
Reviewed-by: Jarkko Sakkinen <[email protected]>
BR, Jarkko
On 7/10/22 23:04, Jarkko Sakkinen wrote:
> On Wed, Jul 06, 2022 at 11:23:27AM -0400, Stefan Berger wrote:
>> Simplify tpm_read_log_of() by moving reusable parts of the code into
>> an inline function that makes it commonly available so it can be
>> used also for kexec support. Call the new of_tpm_get_sml_parameters()
>> function from the TPM Open Firmware driver.
>>
>> Signed-off-by: Stefan Berger <[email protected]>
>> Cc: Jarkko Sakkinen <[email protected]>
>> Cc: Jason Gunthorpe <[email protected]>
>> Cc: Rob Herring <[email protected]>
>> Cc: Frank Rowand <[email protected]>
>>
>> ---
>> v4:
>> - converted to inline function
>> ---
>> drivers/char/tpm/eventlog/of.c | 31 +++++--------------------------
>> include/linux/tpm.h | 27 +++++++++++++++++++++++++++
>> 2 files changed, 32 insertions(+), 26 deletions(-)
>>
>> diff --git a/drivers/char/tpm/eventlog/of.c b/drivers/char/tpm/eventlog/of.c
>> index a9ce66d09a75..f9462d19632e 100644
>> --- a/drivers/char/tpm/eventlog/of.c
>> +++ b/drivers/char/tpm/eventlog/of.c
>> @@ -12,6 +12,7 @@
>>
>> #include <linux/slab.h>
>> #include <linux/of.h>
>> +#include <linux/tpm.h>
>> #include <linux/tpm_eventlog.h>
>>
>> #include "../tpm.h"
>> @@ -20,11 +21,10 @@
>> int tpm_read_log_of(struct tpm_chip *chip)
>> {
>> struct device_node *np;
>> - const u32 *sizep;
>> - const u64 *basep;
>> struct tpm_bios_log *log;
>> u32 size;
>> u64 base;
>> + int ret;
>>
>> log = &chip->log;
>> if (chip->dev.parent && chip->dev.parent->of_node)
>> @@ -35,30 +35,9 @@ int tpm_read_log_of(struct tpm_chip *chip)
>> if (of_property_read_bool(np, "powered-while-suspended"))
>> chip->flags |= TPM_CHIP_FLAG_ALWAYS_POWERED;
>>
>> - sizep = of_get_property(np, "linux,sml-size", NULL);
>> - basep = of_get_property(np, "linux,sml-base", NULL);
>> - if (sizep == NULL && basep == NULL)
>> - return -ENODEV;
>> - if (sizep == NULL || basep == NULL)
>> - return -EIO;
>> -
>> - /*
>> - * For both vtpm/tpm, firmware has log addr and log size in big
>> - * endian format. But in case of vtpm, there is a method called
>> - * sml-handover which is run during kernel init even before
>> - * device tree is setup. This sml-handover function takes care
>> - * of endianness and writes to sml-base and sml-size in little
>> - * endian format. For this reason, vtpm doesn't need conversion
>> - * but physical tpm needs the conversion.
>> - */
Jarkko,
can I apply your R-b tag after putting this lost comment back into the
inline function?
Stefan
>> - if (of_property_match_string(np, "compatible", "IBM,vtpm") < 0 &&
>> - of_property_match_string(np, "compatible", "IBM,vtpm20") < 0) {
>> - size = be32_to_cpup((__force __be32 *)sizep);
>> - base = be64_to_cpup((__force __be64 *)basep);
>> - } else {
>> - size = *sizep;
>> - base = *basep;
>> - }
>> + ret = of_tpm_get_sml_parameters(np, &base, &size);
>> + if (ret < 0)
>> + return ret;
>>
>> if (size == 0) {
>> dev_warn(&chip->dev, "%s: Event log area empty\n", __func__);
>> diff --git a/include/linux/tpm.h b/include/linux/tpm.h
>> index dfeb25a0362d..b3dff255bc58 100644
>> --- a/include/linux/tpm.h
>> +++ b/include/linux/tpm.h
>> @@ -460,4 +460,31 @@ static inline struct tpm_chip *tpm_default_chip(void)
>> return NULL;
>> }
>> #endif
>> +
>> +#ifdef CONFIG_OF
>> +static inline int of_tpm_get_sml_parameters(struct device_node *np,
>> + u64 *base, u32 *size)
>> +{
>> + const u32 *sizep;
>> + const u64 *basep;
>> +
>> + sizep = of_get_property(np, "linux,sml-size", NULL);
>> + basep = of_get_property(np, "linux,sml-base", NULL);
>> + if (sizep == NULL && basep == NULL)
>> + return -ENODEV;
>> + if (sizep == NULL || basep == NULL)
>> + return -EIO;
>> +
>> + if (of_property_match_string(np, "compatible", "IBM,vtpm") < 0 &&
>> + of_property_match_string(np, "compatible", "IBM,vtpm20") < 0) {
>> + *size = be32_to_cpup((__force __be32 *)sizep);
>> + *base = be64_to_cpup((__force __be64 *)basep);
>> + } else {
>> + *size = *sizep;
>> + *base = *basep;
>> + }
>> + return 0;
>> +}
>> +#endif
>> +
>> #endif
>> --
>> 2.35.1
>>
>
>
> Reviewed-by: Jarkko Sakkinen <[email protected]>
>
> BR, Jarkko
>
> _______________________________________________
> kexec mailing list
> [email protected]
> http://lists.infradead.org/mailman/listinfo/kexec
On Fri, Jul 15, 2022 at 12:01:36PM -0400, Stefan Berger wrote:
>
>
> On 7/10/22 23:04, Jarkko Sakkinen wrote:
> > On Wed, Jul 06, 2022 at 11:23:27AM -0400, Stefan Berger wrote:
> > > Simplify tpm_read_log_of() by moving reusable parts of the code into
> > > an inline function that makes it commonly available so it can be
> > > used also for kexec support. Call the new of_tpm_get_sml_parameters()
> > > function from the TPM Open Firmware driver.
> > >
> > > Signed-off-by: Stefan Berger <[email protected]>
> > > Cc: Jarkko Sakkinen <[email protected]>
> > > Cc: Jason Gunthorpe <[email protected]>
> > > Cc: Rob Herring <[email protected]>
> > > Cc: Frank Rowand <[email protected]>
> > >
> > > ---
> > > v4:
> > > - converted to inline function
> > > ---
> > > drivers/char/tpm/eventlog/of.c | 31 +++++--------------------------
> > > include/linux/tpm.h | 27 +++++++++++++++++++++++++++
> > > 2 files changed, 32 insertions(+), 26 deletions(-)
> > >
> > > diff --git a/drivers/char/tpm/eventlog/of.c b/drivers/char/tpm/eventlog/of.c
> > > index a9ce66d09a75..f9462d19632e 100644
> > > --- a/drivers/char/tpm/eventlog/of.c
> > > +++ b/drivers/char/tpm/eventlog/of.c
> > > @@ -12,6 +12,7 @@
> > > #include <linux/slab.h>
> > > #include <linux/of.h>
> > > +#include <linux/tpm.h>
> > > #include <linux/tpm_eventlog.h>
> > > #include "../tpm.h"
> > > @@ -20,11 +21,10 @@
> > > int tpm_read_log_of(struct tpm_chip *chip)
> > > {
> > > struct device_node *np;
> > > - const u32 *sizep;
> > > - const u64 *basep;
> > > struct tpm_bios_log *log;
> > > u32 size;
> > > u64 base;
> > > + int ret;
> > > log = &chip->log;
> > > if (chip->dev.parent && chip->dev.parent->of_node)
> > > @@ -35,30 +35,9 @@ int tpm_read_log_of(struct tpm_chip *chip)
> > > if (of_property_read_bool(np, "powered-while-suspended"))
> > > chip->flags |= TPM_CHIP_FLAG_ALWAYS_POWERED;
> > > - sizep = of_get_property(np, "linux,sml-size", NULL);
> > > - basep = of_get_property(np, "linux,sml-base", NULL);
> > > - if (sizep == NULL && basep == NULL)
> > > - return -ENODEV;
> > > - if (sizep == NULL || basep == NULL)
> > > - return -EIO;
> > > -
> > > - /*
> > > - * For both vtpm/tpm, firmware has log addr and log size in big
> > > - * endian format. But in case of vtpm, there is a method called
> > > - * sml-handover which is run during kernel init even before
> > > - * device tree is setup. This sml-handover function takes care
> > > - * of endianness and writes to sml-base and sml-size in little
> > > - * endian format. For this reason, vtpm doesn't need conversion
> > > - * but physical tpm needs the conversion.
> > > - */
>
>
> Jarkko,
>
> can I apply your R-b tag after putting this lost comment back into the
> inline function?
> Stefan
If there's a change I want to see the updated patch first.
BR, Jarkko