Abstract and PDF here:
https://lpc.events/event/17/contributions/1468/
Video:
https://www.youtube.com/watch?v=3vL3gtAu84s
Patch 1 deals in detail documentation on minidump.
Patch 2-4 refactors minidump existing layout and separate it from remoteproc files.
Patch 6 is the Qualcomm APSS minidump driver.
Patch 7-10 Enable support to reserve dynamic ramoops and the support to
register ramoops region with minidump.
Detail about Minidump is discussed in documentation patch (1/10) and also briefly
discussed after below changelog.
Changes in v8:
- Addressed documentation comment made by Randy Dunlap.
- Rebased on linux-next tag next-20240130
Changes in v7:
- Addressed comment made by [Pavan.K] to use generic notifiers.
- Addresses comment made on Dynamic ramoops about error handling.
- Significant change minidump documentation suggested by [Bryan O'Donoghue]
- Added Reviewed by from [Bagas]
- Renamed ramoops notifiers.
Changes in v6: https://lore.kernel.org/lkml/[email protected]/
- Accumalated the feedback received on v5 and rebase v5 versions in v6.
- Removed the exported function as there is no current users of them.
- Applied [Pavan.K] suggestion on caller/callee placement of dynamic ramoops reserve memory.
- Addressed [krzysztof] comment on sizeof() and to have qcom_apss_md_table_exit().
- Addressed [Bagas.S] comment on minidump doc.
- Tried to implement [Kees] suggestion in slight different way with callback registration
with ramoops instead of pstore core.
Change in rebase v5: https://lore.kernel.org/lkml/[email protected]/
- Rebased it on latest tag available on linux-next
- Added missed Poovendhan sign-off on 15/17 and tested-by tag from
Kathiravan. Thanks to him for testing and reminding me of missing sign-off.
Changes in v5: https://lore.kernel.org/lkml/[email protected]/
- On suggestion from Pavan.k, to have single function call for minidump collection
from remoteproc driver, separated the logic to have separate minidump file called
qcom_rproc_minidump.c and also renamed the function from qcom_minidump() to
qcom_rproc_minidump(); however, dropped his suggestion about rework on lazy deletion
during region unregister in this series, will pursue it in next series.
- To simplify the minidump driver, removed the complication for frontend and different
backend from Greg suggestion, will pursue this once main driver gets mainlined.
- Move the dynamic ramoops region allocation from Device tree approach to command line
approch with the introduction command line parsing and memblock reservation during
early boot up; Not added documentation about it yet, will add if it gets positive
response.
- Exporting linux banner from kernel to make minidump build also as module, however,
minidump is a debug module and should be kernel built to get most debug information
from kernel.
- Tried to address comments given on dload patch series.
Changes in v4: https://lore.kernel.org/lkml/[email protected]/
- Redesigned the driver and divided the driver into front end and backend (smem) so
that any new backend can be attached easily to avoid code duplication.
- Patch reordering as per the driver and subsystem to easier review of the code.
- Removed minidump specific code from remoteproc to minidump smem based driver.
- Enabled the all the driver as modules.
- Address comments made on documentation and yaml and Device tree file [Krzysztof/Konrad]
- Address comments made qcom_pstore_minidump driver and given its Device tree
same set of properties as ramoops. [Luca/Kees]
- Added patch for MAINTAINER file.
- Include defconfig change as one patch as per [Krzysztof] suggestion.
- Tried to remove the redundant file scope variables from the module as per [Krzysztof] suggestion.
- Addressed comments made on dload mode patch v6 version
https://lore.kernel.org/lkml/[email protected]/
Changes in v3: https://lore.kernel.org/lkml/[email protected]/
- Addressed most of the comments by Srini on v2 and refactored the minidump driver.
- Added platform device support
- Unregister region support.
- Added update region for clients.
- Added pending region support.
- Modified the documentation guide accordingly.
- Added qcom_pstore_ramdump client driver which happen to add ramoops platform
device and also registers ramoops region with minidump.
- Added download mode patch series with this minidump series.
https://lore.kernel.org/lkml/[email protected]/
Changes in v2: https://lore.kernel.org/lkml/[email protected]/
- Addressed review comment made by [quic_tsoni/bmasney] to add documentation.
- Addressed comments made by [srinivas.kandagatla]
- Dropped pstore 6/6 from the last series, till i get conclusion to get pstore
region in minidump.
- Fixed issue reported by kernel test robot.
Changes in v1: https://lore.kernel.org/lkml/[email protected]/
Minidump is a best effort mechanism to collect useful and predefined data
for first level of debugging on end user devices running on Qualcomm SoCs.
It is built on the premise that System on Chip (SoC) or subsystem part of
SoC crashes, due to a range of hardware and software bugs.
Qualcomm devices in engineering mode provides a mechanism for generating
full system ramdumps for post mortem debugging. But in some cases it's
however not feasible to capture the entire content of RAM. The minidump
mechanism provides the means for selecting which snippets should be
included in the ramdump.
The core of SMEM based minidump feature is part of Qualcomm's boot
firmware code. It initializes shared memory (SMEM), which is a part of
DDR and allocates a small section of SMEM to minidump table i.e also
called global table of content (G-ToC). Each subsystem (APSS, ADSP, ...)
has their own table of segments to be included in the minidump and all
get their reference from G-ToC. Each segment/region has some details
like name, physical address and it's size etc. and it could be anywhere
scattered in the DDR.
Existing upstream Qualcomm remoteproc driver[1] already supports SMEM
based minidump feature for remoteproc instances like ADSP, MODEM, ...
where predefined selective segments of subsystem region can be dumped
as part of coredump collection which generates smaller size artifacts
compared to complete coredump of subsystem on crash.
[1]
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/tree/drivers/remoteproc/qcom_common.c#n142
In addition to managing and querying the APSS minidump description,
the Linux driver maintains a ELF header in a segment. This segment
gets updated with section/program header whenever a new entry gets
registered.
Support for Minidump enablement on Qualcomm SoCs is pursued separately and
can be done via below series of changes. For testing, these patches can be
applied
https://lore.kernel.org/lkml/[email protected]/
https://lore.kernel.org/lkml/[email protected]/
https://lore.kernel.org/lkml/[email protected]/
Testing of these patches has been done on sm8450 target after enabling kernel
config like CONFIG_PSTORE_RAM/CONFIG_PSTORE_CONSOLE and once the device boots
up. Below command can be executed from sysfs to enable minidump in the firmware.
echo mini > /sys/module/qcom_scm/parameters/download_mode
Try crashing it via devmem2 0xf11c000(this is known command to create xpu violation
and put the device crash dump mode) on command prompt.
Default storage type is set to via USB, so Minidump would be downloaded with the
help of x86_64 machine (running PCAT tool) attached to Qualcomm product which has
backed Minidump boot firmware support.
After that we will see a bunch of predefined registered region as binary blobs files
starts with md_* downloaded on the x86 machine at configured/default location in PCAT
tool from the product, more about this can be found in qualcomm minidump guide
patch.
Mukesh Ojha (10):
docs: qcom: Add qualcomm minidump guide
soc: qcom: Add qcom_rproc_minidump module
remoteproc: qcom_q6v5_pas: Use qcom_rproc_minidump()
remoteproc: qcom: Remove minidump related data from qcom_common.c
init: export linux_banner data variable
soc: qcom: Add Qualcomm APSS minidump kernel driver
MAINTAINERS: Add entry for minidump related files
pstore/ram: Add dynamic ramoops region support through commandline
pstore/ram: Add ramoops information notifier support
soc: qcom: register ramoops region with APSS minidump
Documentation/admin-guide/index.rst | 1 +
Documentation/admin-guide/qcom_minidump.rst | 318 +++++++++
Documentation/admin-guide/ramoops.rst | 23 +-
MAINTAINERS | 10 +
drivers/remoteproc/Kconfig | 1 +
drivers/remoteproc/qcom_common.c | 160 -----
drivers/remoteproc/qcom_q6v5_pas.c | 3 +-
drivers/soc/qcom/Kconfig | 23 +
drivers/soc/qcom/Makefile | 2 +
drivers/soc/qcom/qcom_minidump.c | 690 ++++++++++++++++++++
drivers/soc/qcom/qcom_minidump_internal.h | 74 +++
drivers/soc/qcom/qcom_rproc_minidump.c | 111 ++++
drivers/soc/qcom/smem.c | 20 +
fs/pstore/Kconfig | 15 +
fs/pstore/ram.c | 180 ++++-
include/linux/init.h | 3 +
include/linux/pstore_ram.h | 20 +
include/linux/soc/qcom/smem.h | 2 +
include/soc/qcom/qcom_minidump.h | 41 ++
init/main.c | 3 +
init/version-timestamp.c | 3 +
21 files changed, 1538 insertions(+), 165 deletions(-)
create mode 100644 Documentation/admin-guide/qcom_minidump.rst
create mode 100644 drivers/soc/qcom/qcom_minidump.c
create mode 100644 drivers/soc/qcom/qcom_minidump_internal.h
create mode 100644 drivers/soc/qcom/qcom_rproc_minidump.c
create mode 100644 include/soc/qcom/qcom_minidump.h
base-commit: 41d66f96d0f15a0a2ad6fa2208f6bac1a66cbd52
--
2.43.0.254.ga26002b62827
Now, as all the minidump specific data structure is moved to
minidump specific files and implementation wise qcom_rproc_minidump()
and qcom_minidump() exactly same and the name qcom_rproc_minidump
make more sense as it happen to collect the minidump for the
remoteproc processors. So, let's use qcom_rproc_minidump() and
we will be removing qcom_minidump() and minidump related stuff
from driver/remoteproc/qcom_common.c .
Signed-off-by: Mukesh Ojha <[email protected]>
---
drivers/remoteproc/Kconfig | 1 +
drivers/remoteproc/qcom_q6v5_pas.c | 3 ++-
2 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/remoteproc/Kconfig b/drivers/remoteproc/Kconfig
index 48845dc8fa85..cea960749e2c 100644
--- a/drivers/remoteproc/Kconfig
+++ b/drivers/remoteproc/Kconfig
@@ -166,6 +166,7 @@ config QCOM_PIL_INFO
config QCOM_RPROC_COMMON
tristate
+ select QCOM_RPROC_MINIDUMP
config QCOM_Q6V5_COMMON
tristate
diff --git a/drivers/remoteproc/qcom_q6v5_pas.c b/drivers/remoteproc/qcom_q6v5_pas.c
index a9dd58608052..21e2e8eb0b2f 100644
--- a/drivers/remoteproc/qcom_q6v5_pas.c
+++ b/drivers/remoteproc/qcom_q6v5_pas.c
@@ -25,6 +25,7 @@
#include <linux/soc/qcom/mdt_loader.h>
#include <linux/soc/qcom/smem.h>
#include <linux/soc/qcom/smem_state.h>
+#include <soc/qcom/qcom_minidump.h>
#include "qcom_common.h"
#include "qcom_pil_info.h"
@@ -131,7 +132,7 @@ static void adsp_minidump(struct rproc *rproc)
if (rproc->dump_conf == RPROC_COREDUMP_DISABLED)
return;
- qcom_minidump(rproc, adsp->minidump_id, adsp_segment_dump);
+ qcom_rproc_minidump(rproc, adsp->minidump_id, adsp_segment_dump);
}
static int adsp_pds_enable(struct qcom_adsp *adsp, struct device **pds,
--
2.43.0.254.ga26002b62827
As minidump specific data structure and functions move under
config QCOM_RPROC_MINIDUMP, so remove minidump specific data
from driver/remoteproc/qcom_common.c .
Signed-off-by: Mukesh Ojha <[email protected]>
---
drivers/remoteproc/qcom_common.c | 160 -------------------------------
1 file changed, 160 deletions(-)
diff --git a/drivers/remoteproc/qcom_common.c b/drivers/remoteproc/qcom_common.c
index 03e5f5d533eb..085fd73fa23a 100644
--- a/drivers/remoteproc/qcom_common.c
+++ b/drivers/remoteproc/qcom_common.c
@@ -17,7 +17,6 @@
#include <linux/rpmsg/qcom_smd.h>
#include <linux/slab.h>
#include <linux/soc/qcom/mdt_loader.h>
-#include <linux/soc/qcom/smem.h>
#include "remoteproc_internal.h"
#include "qcom_common.h"
@@ -26,61 +25,6 @@
#define to_smd_subdev(d) container_of(d, struct qcom_rproc_subdev, subdev)
#define to_ssr_subdev(d) container_of(d, struct qcom_rproc_ssr, subdev)
-#define MAX_NUM_OF_SS 10
-#define MAX_REGION_NAME_LENGTH 16
-#define SBL_MINIDUMP_SMEM_ID 602
-#define MINIDUMP_REGION_VALID ('V' << 24 | 'A' << 16 | 'L' << 8 | 'I' << 0)
-#define MINIDUMP_SS_ENCR_DONE ('D' << 24 | 'O' << 16 | 'N' << 8 | 'E' << 0)
-#define MINIDUMP_SS_ENABLED ('E' << 24 | 'N' << 16 | 'B' << 8 | 'L' << 0)
-
-/**
- * struct minidump_region - Minidump region
- * @name : Name of the region to be dumped
- * @seq_num: : Use to differentiate regions with same name.
- * @valid : This entry to be dumped (if set to 1)
- * @address : Physical address of region to be dumped
- * @size : Size of the region
- */
-struct minidump_region {
- char name[MAX_REGION_NAME_LENGTH];
- __le32 seq_num;
- __le32 valid;
- __le64 address;
- __le64 size;
-};
-
-/**
- * struct minidump_subsystem - Subsystem's SMEM Table of content
- * @status : Subsystem toc init status
- * @enabled : if set to 1, this region would be copied during coredump
- * @encryption_status: Encryption status for this subsystem
- * @encryption_required : Decides to encrypt the subsystem regions or not
- * @region_count : Number of regions added in this subsystem toc
- * @regions_baseptr : regions base pointer of the subsystem
- */
-struct minidump_subsystem {
- __le32 status;
- __le32 enabled;
- __le32 encryption_status;
- __le32 encryption_required;
- __le32 region_count;
- __le64 regions_baseptr;
-};
-
-/**
- * struct minidump_global_toc - Global Table of Content
- * @status : Global Minidump init status
- * @md_revision : Minidump revision
- * @enabled : Minidump enable status
- * @subsystems : Array of subsystems toc
- */
-struct minidump_global_toc {
- __le32 status;
- __le32 md_revision;
- __le32 enabled;
- struct minidump_subsystem subsystems[MAX_NUM_OF_SS];
-};
-
struct qcom_ssr_subsystem {
const char *name;
struct srcu_notifier_head notifier_list;
@@ -90,110 +34,6 @@ struct qcom_ssr_subsystem {
static LIST_HEAD(qcom_ssr_subsystem_list);
static DEFINE_MUTEX(qcom_ssr_subsys_lock);
-static void qcom_minidump_cleanup(struct rproc *rproc)
-{
- struct rproc_dump_segment *entry, *tmp;
-
- list_for_each_entry_safe(entry, tmp, &rproc->dump_segments, node) {
- list_del(&entry->node);
- kfree(entry->priv);
- kfree(entry);
- }
-}
-
-static int qcom_add_minidump_segments(struct rproc *rproc, struct minidump_subsystem *subsystem,
- void (*rproc_dumpfn_t)(struct rproc *rproc, struct rproc_dump_segment *segment,
- void *dest, size_t offset, size_t size))
-{
- struct minidump_region __iomem *ptr;
- struct minidump_region region;
- int seg_cnt, i;
- dma_addr_t da;
- size_t size;
- char *name;
-
- if (WARN_ON(!list_empty(&rproc->dump_segments))) {
- dev_err(&rproc->dev, "dump segment list already populated\n");
- return -EUCLEAN;
- }
-
- seg_cnt = le32_to_cpu(subsystem->region_count);
- ptr = ioremap((unsigned long)le64_to_cpu(subsystem->regions_baseptr),
- seg_cnt * sizeof(struct minidump_region));
- if (!ptr)
- return -EFAULT;
-
- for (i = 0; i < seg_cnt; i++) {
- memcpy_fromio(®ion, ptr + i, sizeof(region));
- if (le32_to_cpu(region.valid) == MINIDUMP_REGION_VALID) {
- name = kstrndup(region.name, MAX_REGION_NAME_LENGTH - 1, GFP_KERNEL);
- if (!name) {
- iounmap(ptr);
- return -ENOMEM;
- }
- da = le64_to_cpu(region.address);
- size = le64_to_cpu(region.size);
- rproc_coredump_add_custom_segment(rproc, da, size, rproc_dumpfn_t, name);
- }
- }
-
- iounmap(ptr);
- return 0;
-}
-
-void qcom_minidump(struct rproc *rproc, unsigned int minidump_id,
- void (*rproc_dumpfn_t)(struct rproc *rproc,
- struct rproc_dump_segment *segment, void *dest, size_t offset,
- size_t size))
-{
- int ret;
- struct minidump_subsystem *subsystem;
- struct minidump_global_toc *toc;
-
- /* Get Global minidump ToC*/
- toc = qcom_smem_get(QCOM_SMEM_HOST_ANY, SBL_MINIDUMP_SMEM_ID, NULL);
-
- /* check if global table pointer exists and init is set */
- if (IS_ERR(toc) || !toc->status) {
- dev_err(&rproc->dev, "Minidump TOC not found in SMEM\n");
- return;
- }
-
- /* Get subsystem table of contents using the minidump id */
- subsystem = &toc->subsystems[minidump_id];
-
- /**
- * Collect minidump if SS ToC is valid and segment table
- * is initialized in memory and encryption status is set.
- */
- if (subsystem->regions_baseptr == 0 ||
- le32_to_cpu(subsystem->status) != 1 ||
- le32_to_cpu(subsystem->enabled) != MINIDUMP_SS_ENABLED) {
- return rproc_coredump(rproc);
- }
-
- if (le32_to_cpu(subsystem->encryption_status) != MINIDUMP_SS_ENCR_DONE) {
- dev_err(&rproc->dev, "Minidump not ready, skipping\n");
- return;
- }
-
- /**
- * Clear out the dump segments populated by parse_fw before
- * re-populating them with minidump segments.
- */
- rproc_coredump_cleanup(rproc);
-
- ret = qcom_add_minidump_segments(rproc, subsystem, rproc_dumpfn_t);
- if (ret) {
- dev_err(&rproc->dev, "Failed with error: %d while adding minidump entries\n", ret);
- goto clean_minidump;
- }
- rproc_coredump_using_sections(rproc);
-clean_minidump:
- qcom_minidump_cleanup(rproc);
-}
-EXPORT_SYMBOL_GPL(qcom_minidump);
-
static int glink_subdev_start(struct rproc_subdev *subdev)
{
struct qcom_rproc_glink *glink = to_glink_subdev(subdev);
--
2.43.0.254.ga26002b62827
Add the entries into maintainer file for all the minidump related
modules.
Signed-off-by: Mukesh Ojha <[email protected]>
---
MAINTAINERS | 10 ++++++++++
1 file changed, 10 insertions(+)
diff --git a/MAINTAINERS b/MAINTAINERS
index c45587eb4c1d..a7eea6e64dce 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -18176,6 +18176,16 @@ S: Maintained
F: Documentation/devicetree/bindings/regulator/vqmmc-ipq4019-regulator.yaml
F: drivers/regulator/vqmmc-ipq4019-regulator.c
+QUALCOMM MINIDUMP DRIVER
+M: Mukesh Ojha <[email protected]>
+L: [email protected]
+S: Maintained
+F: Documentation/admin-guide/qcom_minidump.rst
+F: drivers/soc/qcom/qcom_minidump.c
+F: drivers/soc/qcom/qcom_ramoops_minidump.c
+F: drivers/soc/qcom/qcom_rproc_minidump.c
+
+
QUALCOMM NAND CONTROLLER DRIVER
M: Manivannan Sadhasivam <[email protected]>
L: [email protected]
--
2.43.0.254.ga26002b62827
Client like minidump is interested in knowing ramoops
individual zone addresses and their size so that it
could register them with its table.
Let's introduce a info notifier in ramoops which
gets called when ramoops driver probes successfully
and it passes the ramoops region information to the
passed callback by the client and If the call for
ramoops ready register comes after ramoops probe
than call the callback directly.
Signed-off-by: Mukesh Ojha <[email protected]>
---
fs/pstore/ram.c | 114 +++++++++++++++++++++++++++++++++++++
include/linux/pstore_ram.h | 15 +++++
2 files changed, 129 insertions(+)
diff --git a/fs/pstore/ram.c b/fs/pstore/ram.c
index 1faf0835700b..bd94c11e43ff 100644
--- a/fs/pstore/ram.c
+++ b/fs/pstore/ram.c
@@ -22,6 +22,8 @@
#include <linux/of_address.h>
#include <linux/memblock.h>
#include <linux/mm.h>
+#include <linux/mutex.h>
+#include <linux/notifier.h>
#include "internal.h"
#include "ram_internal.h"
@@ -101,6 +103,19 @@ struct ramoops_context {
unsigned int ftrace_read_cnt;
unsigned int pmsg_read_cnt;
struct pstore_info pstore;
+ struct blocking_notifier_head ramoops_notifiers;
+ bool ramoops_ready;
+ /*
+ * Lock to serialize access to ramoops_ready and to not
+ * miss any ongoing notifier registration while ramoops
+ * probe is in progress.
+ */
+ struct mutex lock;
+};
+
+struct ramoops_backend {
+ struct notifier_block nb;
+ int (*fn)(const char *name, int id, void *vaddr, phys_addr_t paddr, size_t size);
};
static struct platform_device *dummy;
@@ -501,6 +516,8 @@ static struct ramoops_context oops_cxt = {
.write_user = ramoops_pstore_write_user,
.erase = ramoops_pstore_erase,
},
+ .ramoops_notifiers = BLOCKING_NOTIFIER_INIT(oops_cxt.ramoops_notifiers),
+ .lock = __MUTEX_INITIALIZER(oops_cxt.lock),
};
static void ramoops_free_przs(struct ramoops_context *cxt)
@@ -666,6 +683,98 @@ static int ramoops_init_prz(const char *name,
return 0;
}
+static int __ramoops_info_notifier(struct ramoops_context *cxt, int (*fn)(const char *,
+ int, void *, phys_addr_t, size_t))
+{
+ struct persistent_ram_zone *prz;
+ int ret;
+ int i;
+
+ for (i = 0; i < cxt->max_dump_cnt; i++) {
+ prz = cxt->dprzs[i];
+ ret = fn("dmesg", i, prz->vaddr, prz->paddr, prz->size);
+ if (ret < 0)
+ goto err;
+ }
+
+ if (cxt->console_size) {
+ prz = cxt->cprz;
+ ret = fn("console", 0, prz->vaddr, prz->paddr, prz->size);
+ if (ret < 0)
+ goto err;
+ }
+
+ for (i = 0; i < cxt->max_ftrace_cnt; i++) {
+ prz = cxt->fprzs[i];
+ ret = fn("ftrace", i, prz->vaddr, prz->paddr, prz->size);
+ if (ret < 0)
+ goto err;
+ }
+
+ if (cxt->pmsg_size) {
+ prz = cxt->mprz;
+ ret = fn("pmsg", 0, prz->vaddr, prz->paddr, prz->size);
+ if (ret < 0)
+ goto err;
+ }
+
+err:
+ return ret;
+}
+
+static int ramoops_info_notifier(struct notifier_block *nb, unsigned long event,
+ void *data)
+{
+ struct ramoops_backend *b_info = container_of(nb, struct ramoops_backend, nb);
+ struct ramoops_context *cxt = data;
+
+ return __ramoops_info_notifier(cxt, b_info->fn);
+}
+
+void *register_ramoops_info_notifier(int (*fn)(const char *, int,
+ void *, phys_addr_t, size_t))
+{
+ struct ramoops_context *cxt = &oops_cxt;
+ struct ramoops_backend *b_info;
+
+ mutex_lock(&cxt->lock);
+ /*
+ * There is no need to register callback if ramoops probe
+ * is already done instead, call the callback directly
+ */
+ if (cxt->ramoops_ready) {
+ mutex_unlock(&cxt->lock);
+ __ramoops_info_notifier(cxt, fn);
+ return NULL;
+ }
+
+ b_info = kzalloc(sizeof(*b_info), GFP_KERNEL);
+ if (!b_info) {
+ b_info = ERR_PTR(-ENOMEM);
+ goto out;
+ }
+
+ b_info->fn = fn;
+ b_info->nb.notifier_call = ramoops_info_notifier;
+ blocking_notifier_chain_register(&cxt->ramoops_notifiers, &b_info->nb);
+
+out:
+ mutex_unlock(&cxt->lock);
+ return b_info;
+}
+EXPORT_SYMBOL_GPL(register_ramoops_info_notifier);
+
+void unregister_ramoops_info_notifier(void *b_info)
+{
+ struct ramoops_context *cxt = &oops_cxt;
+ struct ramoops_backend *tmp = b_info;
+
+ mutex_lock(&cxt->lock);
+ blocking_notifier_chain_unregister(&cxt->ramoops_notifiers, &tmp->nb);
+ mutex_unlock(&cxt->lock);
+}
+EXPORT_SYMBOL_GPL(unregister_ramoops_info_notifier);
+
/* Read a u32 from a dt property and make sure it's safe for an int. */
static int ramoops_parse_dt_u32(struct platform_device *pdev,
const char *propname,
@@ -915,6 +1024,11 @@ static int ramoops_probe(struct platform_device *pdev)
ramoops_pmsg_size = pdata->pmsg_size;
ramoops_ftrace_size = pdata->ftrace_size;
+ mutex_lock(&cxt->lock);
+ cxt->ramoops_ready = true;
+ mutex_unlock(&cxt->lock);
+ blocking_notifier_call_chain(&cxt->ramoops_notifiers, 0, cxt);
+
pr_info("using 0x%lx@0x%llx, ecc: %d\n",
cxt->size, (unsigned long long)cxt->phys_addr,
cxt->ecc_info.ecc_size);
diff --git a/include/linux/pstore_ram.h b/include/linux/pstore_ram.h
index 1efff7a38333..7e27cfc09243 100644
--- a/include/linux/pstore_ram.h
+++ b/include/linux/pstore_ram.h
@@ -39,6 +39,21 @@ struct ramoops_platform_data {
struct persistent_ram_ecc_info ecc_info;
};
+#if IS_ENABLED(CONFIG_PSTORE_RAM)
+void *register_ramoops_info_notifier(int (*fn)(const char *name, int id,
+ void *vaddr, phys_addr_t paddr,
+ size_t size));
+void unregister_ramoops_info_notifier(void *nb_cookie);
+#else
+static inline void *register_ramoops_info_notifier(int (*fn)(const char *name, int id,
+ void *vaddr, phys_addr_t paddr,
+ size_t size))
+{
+ return NULL;
+}
+static inline void unregister_ramoops_info_notifier(void *nb_cookie) { }
+#endif
+
#ifdef CONFIG_PSTORE_DYNAMIC_RAMOOPS
void __init setup_dynamic_ramoops(void);
#else
--
2.43.0.254.ga26002b62827
Add the qualcomm minidump guide for the users which tries to cover
the dependency, API use and the way to test and collect minidump
on Qualcomm supported SoCs.
Signed-off-by: Mukesh Ojha <[email protected]>
Reviewed-by: Bagas Sanjaya <[email protected]>
---
Documentation/admin-guide/index.rst | 1 +
Documentation/admin-guide/qcom_minidump.rst | 318 ++++++++++++++++++++
2 files changed, 319 insertions(+)
create mode 100644 Documentation/admin-guide/qcom_minidump.rst
diff --git a/Documentation/admin-guide/index.rst b/Documentation/admin-guide/index.rst
index fb40a1f6f79e..edab05fc4653 100644
--- a/Documentation/admin-guide/index.rst
+++ b/Documentation/admin-guide/index.rst
@@ -121,6 +121,7 @@ configure specific aspects of kernel behavior to your liking.
pm/index
pmf
pnp
+ qcom_minidump
rapidio
ras
rtc
diff --git a/Documentation/admin-guide/qcom_minidump.rst b/Documentation/admin-guide/qcom_minidump.rst
new file mode 100644
index 000000000000..2a057612422b
--- /dev/null
+++ b/Documentation/admin-guide/qcom_minidump.rst
@@ -0,0 +1,318 @@
+Qualcomm Minidump Feature
+=========================
+
+Introduction
+------------
+
+Minidump is a best effort mechanism to collect useful and predefined
+data for post-mortem debugging on a Qualcomm System on chip(SoCs).
+
+Minidump is built on the premise that a hardware or software component
+on the SoC has encountered an unexpected fault. This means that data
+collected by Minidump can not be assumed to be correct or Minidump
+collection itself could fail.
+
+Qualcomm SoCs in engineering mode provides mechanism for generating
+complete RAM dump for both kernel/non-kernel crashes for postmortem
+debugging however, on a end user product taking complete RAM dump at
+the time of failure has substantial storage requirement as well as it
+is time consuming to transfer them electronically. To encounter this
+problem, Minidump was introduced in Qualcomm boot firmware that provides
+a way to collect selected region in the final RAM dump which is less
+in size compared to complete RAM dump.
+
+Qualcomm SoCs contains Application Processor SubSystem(APSS) and its
+co-processor like Audio Digital Signal Process(ADSP), Compute DSP(CDSP),
+MODEM running their operating system or firmware can register their
+selected region in their respective table called SubSystem table of
+content (SS-ToC) and the addresses of these tables is further maintained
+in a separate table called Global Table of Content (G-ToC) inside separate
+region maintaied inside RAM called Shared memory(SMEM). More about shared
+memory can be found inside ``driver/soc/qcom/smem.c`` under doc section
+and it is briefly discussed in later section.
+
+It is to note that SubSystems, Remote processors and co-processors have
+same meaning in this document and been used interchangeably.
+
+High level design
+-----------------
+::
+
+ +-----------------------------------------------+
+ | RAM +-------------+ |
+ | | SS0-ToC| |
+ | +----------------+ +----------------+ | |
+ | |Shared memory | | SS1-ToC| | |
+ | |(SMEM) | | | | |
+ | | | +-->|--------+ | | |
+ | |G-ToC | | | SS-ToC \ | | |
+ | |+-------------+ | | | +-----------+ | | |
+ | ||-------------| | | | |-----------| | | |
+ | || SS0-ToC | | | +-|<|SS1 region1| | | |
+ | ||-------------| | | | | |-----------| | | |
+ | || SS1-ToC |-|>+ | | |SS1 region2| | | |
+ | ||-------------| | | | |-----------| | | |
+ | || SS2-ToC | | | | | ... | | | |
+ | ||-------------| | | | |-----------| | | |
+ | || ... | | |-|<|SS1 regionN| | | |
+ | ||-------------| | | | |-----------| | | |
+ | || SSn-ToC | | | | +-----------+ | | |
+ | |+-------------+ | | | | | |
+ | | | | |----------------| | |
+ | | | +>| regionN | | |
+ | | | | |----------------| | |
+ | +----------------+ | | | | |
+ | | |----------------| | |
+ | +>| region1 | | |
+ | |----------------| | |
+ | | | | |
+ | |----------------|-+ |
+ | | region5 | |
+ | |----------------| |
+ | | | |
+ | Region information +----------------+ |
+ | +---------------+ |
+ | |region name | |
+ | |---------------| |
+ | |region address | |
+ | |---------------| |
+ | |region size | |
+ | +---------------+ |
+ +-----------------------------------------------+
+
+G-ToC: Global table of contents
+SSX-ToC: SubSystem X table of contents
+ X is an integer in the range of 0 to 10
+ Older boot firmware has kept this limit to 10
+ however, in newer firmware this number is expected to change
+
+SSX-MSn: SubSystem memory segments numbered from 0 to n
+ For APSS, n is limited to 200 from older boot firmware
+
+ Older boot firmware statically allocates 300 as total number of
+ supported region across all SubSystem in Minidump table out of
+ which, APSS limit is kept to 201. In future, this limitation
+ from boot firmware might get removed by allocating the region
+ dynamically. APSS Minidump kernel driver keeping this limit to
+ 201 to be compatible with older boot firmware.
+
+SMEM is a section of RAM reserved by boot firmware and is the backbone of
+Minidump functionality to work. It is also a medium of inter processor
+communication and a way where boot firmware can prepare something for
+upcoming operating system usage.
+
+Qualcomm SoCs boot firmware must reserve an area of RAM as SMEM prior to
+handling over control to the run-time operating system. It creates SMEM
+partition for Minidump with ``SBL_MINIDUMP_SMEM_ID`` and creates an array
+of pointers called Global table of content (G-ToC) at the start of this
+partition. Each index of this array is uniquely assigned to each SubSystem
+like for APSS it is 0 while for ADSP, CDSP, MODEM it is 5, 7 and 3 respectively.
+points to their table of segments called SS-ToC to be included in the Minidump.
+
+From the diagram above, Global Table of Contents (G-ToC) enumerates a fixed
+size number of SubSystem Table of Contents (SS-ToC) structures. Each
+SS-ToC contains a list of SubSystem Memory Segments which are named
+according to the containing SS-ToC hence, SSX-MSn where "X" denotes the
+SubSystem index of the containing SSX-ToC and "n" denotes an individual
+Memory segment within the SubSystem. Hence, SS0-MS0 belongs to SS0-ToC
+whereas SS1-MS0 belongs to SS1-ToC. Segment structure contains name,
+base address, size of a Segment to be dumped.
+
+The Application Processor SubSystem (APSS) runs the Linux kernel and is
+therefore not responsible for assembling Minidump data. One of the other
+system agents in the SoC will be responsible for capturing the Minidump
+data during system reset. Typically one of the SoC Digital Signal
+Processors (DSP) will be used for this purpose. During reset, the DSP will
+walk the G-ToC, SSX-ToCs and SSX-MSns either., dump the regions as binary
+blob into storage or pushed outside to the attached host machine via USB
+(more described in Dump collection section below).
+
+Qualcomm Remote Processor Minidump support
+------------------------------------------
+
+Linux Kernel support recovery and coredump collection on remote processor
+failure through remoteproc framework and in this document, remote processors
+meant for ADSP, CDSP, MODEM etc. Qualcomm remoteproc driver has support for
+collecting Minidump for remote processors as well where each remote processor
+has their unique statically assigned descriptor in the G-ToC which is
+represented via ``minidump_id`` in ``driver/remoteproc/qcom_q6v5_pas.c``
+and it helps getting further information about valid registered region from
+firmware and later collecting via remoteproc coredump framework.
+
+Qualcomm APSS Minidump kernel driver concept
+--------------------------------------------
+
+Qualcomm APSS Minidump kernel driver adds the capability to add Linux
+region to be dumped as part of Minidump collection. Shared memory
+driver creates platform device for Minidump driver and on Minidump
+driver probe it gets the G-ToC address (``struct minidump_global_toc``)
+by querying Minidump SMEM ID ``SBL_MINIDUMP_SMEM_ID`` as one of parameter
+to ``qcom_smem_get`` function. Further, driver uses APSS Minidump unique
+descriptor or index i.e., 0 to get APSS SubSystem ToC and fills up the
+fields of ``struct minidump_subsystem`` and allocates memory for Segment
+array of structure ``struct minidump_region`` of size compatible with
+boot firmware (default size is 201). This really means that total 201
+APSS regions can be registered for APSS alone and the Minidump kernel
+driver provides ``qcom_minidump_region_register`` and
+``qcom_minidump_region_unregister`` function to register and unregister
+APSS minidump region. Example usage explained in later section.
+
+To simplify post-mortem debugging, APSS driver registers the first region
+as an ELF header that gets updated each time a new region gets registered.
+and rest 200 region can be used by other APSS Minidump driver client.
+
+The solution supports extracting the Minidump produced either over USB
+or stored to an attached storage device, if not configured default mode
+is USB more described in Dump collection section.
+
+How a kernel client driver can register region with minidump
+------------------------------------------------------------
+
+A client driver can use ``qcom_minidump_region_register`` API's to register
+and ``qcom_minidump_region_unregister`` to unregister their region from
+minidump driver.
+
+A client needs to fill their region by filling ``qcom_minidump_region``
+structure object which consists of the region name, region's virtual
+and physical address and its size.
+
+ .. code-block:: c
+
+ #include <soc/qcom/qcom_minidump.h>
+ [...]
+
+
+ [... inside a function ...]
+ struct qcom_minidump_region region;
+
+ [...]
+
+ client_mem_region = kzalloc(region_size, GFP_KERNEL);
+ if (!client_mem_region)
+ return -ENOMEM;
+
+ [... Just write a pattern ...]
+ memset(client_mem_region, 0xAB, region_size);
+
+ [... Fill up the region object ...]
+ strlcpy(region.name, "REGION_A", sizeof(region.name));
+ region.virt_addr = client_mem_region;
+ region.phys_addr = virt_to_phys(client_mem_region);
+ region.size = region_size;
+
+ ret = qcom_minidump_region_register(®ion);
+ if (ret < 0) {
+ pr_err("failed to add region in minidump: err: %d\n", ret);
+ return ret;
+ }
+
+ [...]
+
+
+Testing
+-------
+
+Existing Qualcomm SoCs already supports collecting complete RAM dump (also
+called full dump) can be configured by writing appropriate value to Qualcomm's
+top control and status register (tcsr) in ``driver/firmware/qcom_scm.c``.
+Complete RAM dump on system failure is where entire RAM snapshot is pushed out
+to Host computer attached to SoC via USB similar to one of the way will be
+used for Minidump described later in Dump collection section. Complete RAM
+dump entirely get controlled from Qualcomm boot firmware and is not related
+to Minidump or SMEM except the fact that same register is used to configure
+one of the mode.
+
+SCM device Tree bindings required to support download mode
+For example (sm8450) ::
+
+ / {
+
+ [...]
+
+ firmware {
+ scm: scm {
+ compatible = "qcom,scm-sm8450", "qcom,scm";
+ [... tcsr register ... ]
+ qcom,dload-mode = <&tcsr 0x13000>;
+
+ [...]
+ };
+ };
+
+ [...]
+
+ soc: soc@0 {
+
+ [...]
+
+ tcsr: syscon@1fc0000 {
+ compatible = "qcom,sm8450-tcsr", "syscon";
+ reg = <0x0 0x1fc0000 0x0 0x30000>;
+ };
+
+ [...]
+ };
+ [...]
+
+ };
+
+A kernel command line parameter is provided to facilitate selection of
+dump mode also called download mode. Boot firmware configures download
+mode to be full dump even before Linux boots up however, one need to pass
+``qcom_scm.download_mode="mini"`` to switch the default download mode
+to Minidump. Similarly ``"full"`` need to be passed to set the download
+mode to full dump and passing ``"full,mini"`` will set the download mode
+where both Minidump along with fulldump will be collected on system failure
+however, this mode will only work if dump need to collected via USB more
+about this described in Dump collection section.
+
+Writing to sysfs node can also be used to set the mode to minidump::
+
+ echo "mini" > /sys/module/qcom_scm/parameter/download_mode
+
+Once the download mode is set, any kind of crash will make the device collect
+respective dump as per the set download mode.
+
+Dump collection
+---------------
+::
+
+ +-----------+
+ | |
+ | | +------+
+ | | | |
+ | | +--+---+ Product(Qualcomm SoC)
+ +-----------+ |
+ |+++++++++++|<------------+
+ |+++++++++++| usb cable
+ +-----------+
+ x86_64 PC
+
+The solution supports a product running with Qualcomm SoC (where minidump)
+is supported from the firmware) connected to x86_64 host PC running PCAT
+tool. It supports downloading the minidump produced from product to the
+host PC over USB or to save the minidump to the product attached storage
+device(UFS/eMMC/SD Card) into minidump dedicated partition.
+
+By default, dumps are downloaded via USB to the attached x86_64 PC running
+PCAT (Qualcomm tool) software. Upon download, we will see a set of binary
+blobs starting with name ``md_*`` in PCAT configured directory in x86_64
+machine, so for above example from the client it will be ``md_REGION_A.BIN``.
+This binary blob depends on region content to determine whether it needs
+external parser support to get the content of the region, so for simple
+plain ASCII text we don't need any parsing and the content can be seen
+just opening the binary file.
+
+To collect the dump to attached storage type, one needs to write appropriate
+value to IMEM register, in that case dumps are collected in rawdump
+partition on the product device itself.
+
+One needs to read the entire rawdump partition and pull out content to
+save it onto the attached x86_64 machine over USB. Later, this rawdump
+can be passed to another tool (``dexter.exe`` [Qualcomm tool]) which
+converts this into the similar binary blobs which we have got it when
+download type was set to USB, i.e. a set of registered regions as blobs
+and their name starts with ``md_*``.
+
+Replacing the ``dexter.exe`` with some open source tool can be added as future
+scope of this document.
--
2.43.0.254.ga26002b62827
Add qcom_rproc_minidump module in a preparation to remove
minidump specific code from driver/remoteproc/qcom_common.c
and provide needed exported API, this as well helps to
abstract minidump specific data layout from qualcomm's
remoteproc driver.
It is just a copying of qcom_minidump() functionality from
driver/remoteproc/qcom_common.c into a separate file under
qcom_rproc_minidump().
Signed-off-by: Mukesh Ojha <[email protected]>
---
drivers/soc/qcom/Kconfig | 10 ++
drivers/soc/qcom/Makefile | 1 +
drivers/soc/qcom/qcom_minidump_internal.h | 64 +++++++++++++
drivers/soc/qcom/qcom_rproc_minidump.c | 111 ++++++++++++++++++++++
include/soc/qcom/qcom_minidump.h | 23 +++++
5 files changed, 209 insertions(+)
create mode 100644 drivers/soc/qcom/qcom_minidump_internal.h
create mode 100644 drivers/soc/qcom/qcom_rproc_minidump.c
create mode 100644 include/soc/qcom/qcom_minidump.h
diff --git a/drivers/soc/qcom/Kconfig b/drivers/soc/qcom/Kconfig
index c6ca4de42586..33584bbbfc0e 100644
--- a/drivers/soc/qcom/Kconfig
+++ b/drivers/soc/qcom/Kconfig
@@ -268,4 +268,14 @@ config QCOM_INLINE_CRYPTO_ENGINE
tristate
select QCOM_SCM
+config QCOM_RPROC_MINIDUMP
+ tristate "QCOM Remoteproc Minidump Support"
+ depends on ARCH_QCOM || COMPILE_TEST
+ depends on QCOM_SMEM
+ help
+ Enablement of core Minidump feature is controlled from boot firmware
+ side, so if it is enabled from firmware, this config allow Linux to
+ query predefined Minidump segments associated with the remote processor
+ and check its validity and end up collecting the dump on remote processor
+ crash during its recovery.
endmenu
diff --git a/drivers/soc/qcom/Makefile b/drivers/soc/qcom/Makefile
index 9268e15e1719..fa685e5a12a6 100644
--- a/drivers/soc/qcom/Makefile
+++ b/drivers/soc/qcom/Makefile
@@ -35,3 +35,4 @@ obj-$(CONFIG_QCOM_KRYO_L2_ACCESSORS) += kryo-l2-accessors.o
obj-$(CONFIG_QCOM_ICC_BWMON) += icc-bwmon.o
qcom_ice-objs += ice.o
obj-$(CONFIG_QCOM_INLINE_CRYPTO_ENGINE) += qcom_ice.o
+obj-$(CONFIG_QCOM_RPROC_MINIDUMP) += qcom_rproc_minidump.o
diff --git a/drivers/soc/qcom/qcom_minidump_internal.h b/drivers/soc/qcom/qcom_minidump_internal.h
new file mode 100644
index 000000000000..71709235b196
--- /dev/null
+++ b/drivers/soc/qcom/qcom_minidump_internal.h
@@ -0,0 +1,64 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2023 Qualcomm Innovation Center, Inc. All rights reserved.
+ */
+
+#ifndef _QCOM_MINIDUMP_INTERNAL_H_
+#define _QCOM_MINIDUMP_INTERNAL_H_
+
+#define MAX_NUM_OF_SS 10
+#define MAX_REGION_NAME_LENGTH 16
+#define SBL_MINIDUMP_SMEM_ID 602
+#define MINIDUMP_REGION_VALID ('V' << 24 | 'A' << 16 | 'L' << 8 | 'I' << 0)
+#define MINIDUMP_SS_ENCR_DONE ('D' << 24 | 'O' << 16 | 'N' << 8 | 'E' << 0)
+#define MINIDUMP_SS_ENABLED ('E' << 24 | 'N' << 16 | 'B' << 8 | 'L' << 0)
+
+/**
+ * struct minidump_region - Minidump region
+ * @name : Name of the region to be dumped
+ * @seq_num: : Use to differentiate regions with same name.
+ * @valid : This entry to be dumped (if set to 1)
+ * @address : Physical address of region to be dumped
+ * @size : Size of the region
+ */
+struct minidump_region {
+ char name[MAX_REGION_NAME_LENGTH];
+ __le32 seq_num;
+ __le32 valid;
+ __le64 address;
+ __le64 size;
+};
+
+/**
+ * struct minidump_subsystem - Subsystem's SMEM Table of content
+ * @status : Subsystem toc init status
+ * @enabled : if set to 1, this region would be copied during coredump
+ * @encryption_status: Encryption status for this subsystem
+ * @encryption_required : Decides to encrypt the subsystem regions or not
+ * @region_count : Number of regions added in this subsystem toc
+ * @regions_baseptr : regions base pointer of the subsystem
+ */
+struct minidump_subsystem {
+ __le32 status;
+ __le32 enabled;
+ __le32 encryption_status;
+ __le32 encryption_required;
+ __le32 region_count;
+ __le64 regions_baseptr;
+};
+
+/**
+ * struct minidump_global_toc - Global Table of Content
+ * @status : Global Minidump init status
+ * @md_revision : Minidump revision
+ * @enabled : Minidump enable status
+ * @subsystems : Array of subsystems toc
+ */
+struct minidump_global_toc {
+ __le32 status;
+ __le32 md_revision;
+ __le32 enabled;
+ struct minidump_subsystem subsystems[MAX_NUM_OF_SS];
+};
+
+#endif /* _QCOM_MINIDUMP_INTERNAL_H_ */
diff --git a/drivers/soc/qcom/qcom_rproc_minidump.c b/drivers/soc/qcom/qcom_rproc_minidump.c
new file mode 100644
index 000000000000..9bc84cc2536f
--- /dev/null
+++ b/drivers/soc/qcom/qcom_rproc_minidump.c
@@ -0,0 +1,111 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2023 Qualcomm Innovation Center, Inc. All rights reserved.
+ */
+#include <linux/io.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/remoteproc.h>
+#include <linux/soc/qcom/smem.h>
+#include <linux/string.h>
+#include <soc/qcom/qcom_minidump.h>
+
+#include "qcom_minidump_internal.h"
+
+static void qcom_minidump_cleanup(struct rproc *rproc)
+{
+ struct rproc_dump_segment *entry, *tmp;
+
+ list_for_each_entry_safe(entry, tmp, &rproc->dump_segments, node) {
+ list_del(&entry->node);
+ kfree(entry->priv);
+ kfree(entry);
+ }
+}
+
+static int qcom_add_minidump_segments(struct rproc *rproc, struct minidump_subsystem *subsystem,
+ void (*rproc_dumpfn_t)(struct rproc *rproc, struct rproc_dump_segment *segment,
+ void *dest, size_t offset, size_t size))
+{
+ struct minidump_region __iomem *ptr;
+ struct minidump_region region;
+ int seg_cnt, i;
+ dma_addr_t da;
+ size_t size;
+ char *name;
+
+ if (WARN_ON(!list_empty(&rproc->dump_segments))) {
+ dev_err(&rproc->dev, "dump segment list already populated\n");
+ return -EUCLEAN;
+ }
+
+ seg_cnt = le32_to_cpu(subsystem->region_count);
+ ptr = ioremap((unsigned long)le64_to_cpu(subsystem->regions_baseptr),
+ seg_cnt * sizeof(struct minidump_region));
+ if (!ptr)
+ return -EFAULT;
+
+ for (i = 0; i < seg_cnt; i++) {
+ memcpy_fromio(®ion, ptr + i, sizeof(region));
+ if (le32_to_cpu(region.valid) == MINIDUMP_REGION_VALID) {
+ name = kstrndup(region.name, MAX_REGION_NAME_LENGTH - 1, GFP_KERNEL);
+ if (!name) {
+ iounmap(ptr);
+ return -ENOMEM;
+ }
+ da = le64_to_cpu(region.address);
+ size = le64_to_cpu(region.size);
+ rproc_coredump_add_custom_segment(rproc, da, size, rproc_dumpfn_t, name);
+ }
+ }
+
+ iounmap(ptr);
+ return 0;
+}
+
+void qcom_rproc_minidump(struct rproc *rproc, unsigned int minidump_id,
+ void (*rproc_dumpfn_t)(struct rproc *rproc,
+ struct rproc_dump_segment *segment, void *dest, size_t offset,
+ size_t size))
+{
+ int ret;
+ struct minidump_subsystem *subsystem;
+ struct minidump_global_toc *toc;
+
+ /* Get Global minidump ToC*/
+ toc = qcom_smem_get(QCOM_SMEM_HOST_ANY, SBL_MINIDUMP_SMEM_ID, NULL);
+
+ /* check if global table pointer exists and init is set */
+ if (IS_ERR(toc) || !toc->status) {
+ dev_err(&rproc->dev, "Minidump TOC not found in SMEM\n");
+ return;
+ }
+
+ /* Get subsystem table of contents using the minidump id */
+ subsystem = &toc->subsystems[minidump_id];
+
+ /**
+ * Collect minidump if SS ToC is valid and segment table
+ * is initialized in memory and encryption status is set.
+ */
+ if (subsystem->regions_baseptr == 0 ||
+ le32_to_cpu(subsystem->status) != 1 ||
+ le32_to_cpu(subsystem->enabled) != MINIDUMP_SS_ENABLED ||
+ le32_to_cpu(subsystem->encryption_status) != MINIDUMP_SS_ENCR_DONE) {
+ dev_err(&rproc->dev, "Minidump not ready, skipping\n");
+ return;
+ }
+
+ ret = qcom_add_minidump_segments(rproc, subsystem, rproc_dumpfn_t);
+ if (ret) {
+ dev_err(&rproc->dev, "Failed with error: %d while adding minidump entries\n", ret);
+ goto clean_minidump;
+ }
+ rproc_coredump_using_sections(rproc);
+clean_minidump:
+ qcom_minidump_cleanup(rproc);
+}
+EXPORT_SYMBOL_GPL(qcom_rproc_minidump);
+
+MODULE_DESCRIPTION("Qualcomm remoteproc minidump(smem) helper module");
+MODULE_LICENSE("GPL");
diff --git a/include/soc/qcom/qcom_minidump.h b/include/soc/qcom/qcom_minidump.h
new file mode 100644
index 000000000000..cd87caef919d
--- /dev/null
+++ b/include/soc/qcom/qcom_minidump.h
@@ -0,0 +1,23 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2023 Qualcomm Innovation Center, Inc. All rights reserved.
+ */
+
+#ifndef _QCOM_MINIDUMP_H_
+#define _QCOM_MINIDUMP_H_
+
+struct rproc;
+struct rproc_dump_segment;
+
+#if IS_ENABLED(CONFIG_QCOM_RPROC_MINIDUMP)
+void qcom_rproc_minidump(struct rproc *rproc, unsigned int minidump_id,
+ void (*rproc_dumpfn_t)(struct rproc *rproc,
+ struct rproc_dump_segment *segment, void *dest, size_t offset,
+ size_t size));
+#else
+static inline void qcom_rproc_minidump(struct rproc *rproc, unsigned int minidump_id,
+ void (*rproc_dumpfn_t)(struct rproc *rproc,
+ struct rproc_dump_segment *segment, void *dest, size_t offset,
+ size_t size)) { }
+#endif /* CONFIG_QCOM_RPROC_MINIDUMP */
+#endif /* _QCOM_MINIDUMP_H_ */
--
2.43.0.254.ga26002b62827
Some debug loadable module like minidump is interested in knowing
the kernel version against which it is being build. Let's export
linux_banner.
Signed-off-by: Mukesh Ojha <[email protected]>
---
include/linux/init.h | 3 +++
init/version-timestamp.c | 2 ++
2 files changed, 5 insertions(+)
diff --git a/include/linux/init.h b/include/linux/init.h
index 58cef4c2e59a..d3d96188551c 100644
--- a/include/linux/init.h
+++ b/include/linux/init.h
@@ -145,6 +145,9 @@ extern char *saved_command_line;
extern unsigned int saved_command_line_len;
extern unsigned int reset_devices;
+/* Defined in init/version-timestamp.c */
+extern const char linux_banner[];
+
/* used by init/main.c */
void setup_arch(char **);
void prepare_namespace(void);
diff --git a/init/version-timestamp.c b/init/version-timestamp.c
index 043cbf80a766..a48f2c19e5d7 100644
--- a/init/version-timestamp.c
+++ b/init/version-timestamp.c
@@ -6,6 +6,7 @@
#include <linux/refcount.h>
#include <linux/uts.h>
#include <linux/utsname.h>
+#include <linux/init.h>
struct uts_namespace init_uts_ns = {
.ns.count = REFCOUNT_INIT(2),
@@ -28,3 +29,4 @@ struct uts_namespace init_uts_ns = {
const char linux_banner[] =
"Linux version " UTS_RELEASE " (" LINUX_COMPILE_BY "@"
LINUX_COMPILE_HOST ") (" LINUX_COMPILER ") " UTS_VERSION "\n";
+EXPORT_SYMBOL_GPL(linux_banner);
--
2.43.0.254.ga26002b62827
Minidump is a best effort mechanism to collect useful and predefined
data for first level of debugging on end user devices running on
Qualcomm SoCs. It is built on the premise that System on Chip (SoC)
or subsystem part of SoC crashes, due to a range of hardware and
software bugs. Hence, the ability to collect accurate data is only
a best-effort. The data collected could be invalid or corrupted,
data collection itself could fail, and so on.
Qualcomm devices in engineering mode provides a mechanism for
generating full system ramdumps for post mortem debugging. But in some
cases it's however not feasible to capture the entire content of RAM.
Minidump mechanism provides the means for selecting region should
be included in the ramdump. The solution supports extracting the
Ramdump/Minidump produced either over USB or stored to an attached
storage device.
The core of Minidump feature is part of Qualcomm's boot firmware code.
It initializes shared memory (SMEM), which is a part of RAM and
allocates a small section of it to Minidump table i.e also called
global table of content (G-ToC). Each subsystem (APSS, ADSP, ...) has
their own table of segments to be included in the Minidump, all
references from a descriptor in SMEM (G-ToC). Each segment/region has
some details like name, physical address and it's size etc. and it
could be anywhere scattered in the DDR.
To simplify post mortem debugging, it creates and maintain an ELF
header as first region that gets updated upon registration
of a new region.
Signed-off-by: Mukesh Ojha <[email protected]>
---
drivers/soc/qcom/Kconfig | 13 +
drivers/soc/qcom/Makefile | 1 +
drivers/soc/qcom/qcom_minidump.c | 606 ++++++++++++++++++++++
drivers/soc/qcom/qcom_minidump_internal.h | 10 +
drivers/soc/qcom/smem.c | 20 +
include/linux/soc/qcom/smem.h | 2 +
include/soc/qcom/qcom_minidump.h | 18 +
7 files changed, 670 insertions(+)
create mode 100644 drivers/soc/qcom/qcom_minidump.c
diff --git a/drivers/soc/qcom/Kconfig b/drivers/soc/qcom/Kconfig
index 33584bbbfc0e..956336dae931 100644
--- a/drivers/soc/qcom/Kconfig
+++ b/drivers/soc/qcom/Kconfig
@@ -278,4 +278,17 @@ config QCOM_RPROC_MINIDUMP
query predefined Minidump segments associated with the remote processor
and check its validity and end up collecting the dump on remote processor
crash during its recovery.
+
+config QCOM_MINIDUMP
+ tristate "QCOM APSS Minidump driver"
+ depends on ARCH_QCOM || COMPILE_TEST
+ depends on QCOM_SMEM
+ help
+ This config enables Linux core infrastructure for Application
+ processor subsystem (APSS) Minidump collection, i.e., it enables
+ Linux client drivers to register their internal data structures
+ and debug messages as part of the APSS Minidump table and when
+ the SoC has crashed, these selective regions will be dumped
+ instead of the entire RAM dump. This saves a significant amount
+ of time and/or storage space.
endmenu
diff --git a/drivers/soc/qcom/Makefile b/drivers/soc/qcom/Makefile
index fa685e5a12a6..c64082c33fc8 100644
--- a/drivers/soc/qcom/Makefile
+++ b/drivers/soc/qcom/Makefile
@@ -36,3 +36,4 @@ obj-$(CONFIG_QCOM_ICC_BWMON) += icc-bwmon.o
qcom_ice-objs += ice.o
obj-$(CONFIG_QCOM_INLINE_CRYPTO_ENGINE) += qcom_ice.o
obj-$(CONFIG_QCOM_RPROC_MINIDUMP) += qcom_rproc_minidump.o
+obj-$(CONFIG_QCOM_MINIDUMP) += qcom_minidump.o
diff --git a/drivers/soc/qcom/qcom_minidump.c b/drivers/soc/qcom/qcom_minidump.c
new file mode 100644
index 000000000000..c0f76a51d0e8
--- /dev/null
+++ b/drivers/soc/qcom/qcom_minidump.c
@@ -0,0 +1,606 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (c) 2023 Qualcomm Innovation Center, Inc. All rights reserved.
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/device.h>
+#include <linux/elf.h>
+#include <linux/err.h>
+#include <linux/errno.h>
+#include <linux/export.h>
+#include <linux/init.h>
+#include <linux/io.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/mod_devicetable.h>
+#include <linux/mutex.h>
+#include <linux/platform_device.h>
+#include <linux/printk.h>
+#include <linux/soc/qcom/smem.h>
+#include <linux/string.h>
+#include <soc/qcom/qcom_minidump.h>
+
+#include "qcom_minidump_internal.h"
+
+/**
+ * struct minidump_ss_data - Minidump subsystem private data
+ * @md_ss_toc: Application Subsystem TOC pointer
+ * @md_regions: Application Subsystem region base pointer
+ */
+struct minidump_ss_data {
+ struct minidump_subsystem *md_ss_toc;
+ struct minidump_region *md_regions;
+};
+
+/**
+ * struct minidump_elfhdr - Minidump table elf header
+ * @ehdr: elf main header
+ * @shdr: Section header
+ * @phdr: Program header
+ * @elf_offset: Section offset in elf
+ * @strtable_idx: String table current index position
+ */
+struct minidump_elfhdr {
+ struct elfhdr *ehdr;
+ struct elf_shdr *shdr;
+ struct elf_phdr *phdr;
+ size_t elf_offset;
+ size_t strtable_idx;
+};
+
+/**
+ * struct minidump - Minidump driver data information
+ * @elf: Minidump elf header
+ * @dev: Minidump backend device
+ * @apss_data: APSS driver data
+ * @md_lock: Lock to protect access to APSS minidump table
+ */
+struct minidump {
+ struct minidump_elfhdr elf;
+ struct device *dev;
+ struct minidump_ss_data *apss_data;
+ struct mutex md_lock;
+};
+
+/*
+ * In some of the Old Qualcomm devices, boot firmware statically allocates 300
+ * as total number of supported region (including all co-processors) in
+ * minidump table out of which linux was using 201. In future, this limitation
+ * from boot firmware might get removed by allocating the region dynamically.
+ * So, keep it compatible with older devices, we can keep the current limit for
+ * Linux to 201.
+ */
+#define MAX_NUM_ENTRIES 201
+#define MAX_STRTBL_SIZE (MAX_NUM_ENTRIES * MAX_REGION_NAME_LENGTH)
+
+static struct elf_shdr *elf_shdr_entry_addr(struct elfhdr *ehdr, int idx)
+{
+ struct elf_shdr *eshdr = (struct elf_shdr *)((size_t)ehdr + ehdr->e_shoff);
+
+ return &eshdr[idx];
+}
+
+static struct elf_phdr *elf_phdr_entry_addr(struct elfhdr *ehdr, int idx)
+{
+ struct elf_phdr *ephdr = (struct elf_phdr *)((size_t)ehdr + ehdr->e_phoff);
+
+ return &ephdr[idx];
+}
+
+static char *elf_str_table_start(struct elfhdr *ehdr)
+{
+ struct elf_shdr *eshdr;
+
+ if (ehdr->e_shstrndx == SHN_UNDEF)
+ return NULL;
+
+ eshdr = elf_shdr_entry_addr(ehdr, ehdr->e_shstrndx);
+
+ return (char *)ehdr + eshdr->sh_offset;
+}
+
+static char *elf_lookup_string(struct minidump *md, struct elfhdr *ehdr, int offset)
+{
+ char *strtab = elf_str_table_start(ehdr);
+
+ if (!strtab || (md->elf.strtable_idx < offset))
+ return NULL;
+
+ return strtab + offset;
+}
+
+static unsigned int append_str_to_strtable(struct minidump *md, const char *name)
+{
+ char *strtab = elf_str_table_start(md->elf.ehdr);
+ unsigned int old_idx = md->elf.strtable_idx;
+ unsigned int ret;
+
+ if (!strtab || !name)
+ return 0;
+
+ ret = old_idx;
+ old_idx += strscpy((strtab + old_idx), name, MAX_REGION_NAME_LENGTH);
+ md->elf.strtable_idx = old_idx + 1;
+
+ return ret;
+}
+
+static int qcom_md_clear_elfheader(struct minidump *md,
+ const struct qcom_minidump_region *region)
+{
+ struct elfhdr *ehdr = md->elf.ehdr;
+ struct elf_shdr *shdr;
+ struct elf_shdr *tmp_shdr;
+ struct elf_phdr *phdr;
+ struct elf_phdr *tmp_phdr;
+ unsigned int phidx;
+ unsigned int shidx;
+ unsigned int len;
+ unsigned int i;
+ char *shname;
+
+ for (i = 0; i < ehdr->e_phnum; i++) {
+ phdr = elf_phdr_entry_addr(ehdr, i);
+ if (phdr->p_paddr == region->phys_addr &&
+ phdr->p_memsz == region->size)
+ break;
+ }
+
+ if (i == ehdr->e_phnum) {
+ dev_err(md->dev, "Cannot find program header entry in elf\n");
+ return -EINVAL;
+ }
+
+ phidx = i;
+ for (i = 0; i < ehdr->e_shnum; i++) {
+ shdr = elf_shdr_entry_addr(ehdr, i);
+ shname = elf_lookup_string(md, ehdr, shdr->sh_name);
+ if (shname && !strcmp(shname, region->name) &&
+ shdr->sh_addr == (elf_addr_t)region->virt_addr &&
+ shdr->sh_size == region->size)
+ break;
+ }
+
+ if (i == ehdr->e_shnum) {
+ dev_err(md->dev, "Cannot find section header entry in elf\n");
+ return -EINVAL;
+ }
+
+ shidx = i;
+ if (shdr->sh_offset != phdr->p_offset) {
+ dev_err(md->dev, "Invalid entry details for region: %s\n", region->name);
+ return -EINVAL;
+ }
+
+ /* Clear name in string table */
+ len = strlen(shname) + 1;
+ memmove(shname, shname + len, md->elf.strtable_idx - shdr->sh_name - len);
+ md->elf.strtable_idx -= len;
+
+ /* Clear program header */
+ tmp_phdr = elf_phdr_entry_addr(ehdr, phidx);
+ for (i = phidx; i < ehdr->e_phnum - 1; i++) {
+ tmp_phdr = elf_phdr_entry_addr(ehdr, i + 1);
+ phdr = elf_phdr_entry_addr(ehdr, i);
+ memcpy(phdr, tmp_phdr, sizeof(*phdr));
+ phdr->p_offset = phdr->p_offset - region->size;
+ }
+ memset(tmp_phdr, 0, sizeof(*tmp_phdr));
+ ehdr->e_phnum--;
+
+ /* Clear section header */
+ tmp_shdr = elf_shdr_entry_addr(ehdr, shidx);
+ for (i = shidx; i < ehdr->e_shnum - 1; i++) {
+ tmp_shdr = elf_shdr_entry_addr(ehdr, i + 1);
+ shdr = elf_shdr_entry_addr(ehdr, i);
+ memcpy(shdr, tmp_shdr, sizeof(*shdr));
+ shdr->sh_offset -= region->size;
+ shdr->sh_name -= len;
+ }
+
+ memset(tmp_shdr, 0, sizeof(*tmp_shdr));
+ ehdr->e_shnum--;
+ md->elf.elf_offset -= region->size;
+
+ return 0;
+}
+
+static void qcom_md_update_elfheader(struct minidump *md,
+ const struct qcom_minidump_region *region)
+{
+ struct elfhdr *ehdr = md->elf.ehdr;
+ struct elf_shdr *shdr;
+ struct elf_phdr *phdr;
+
+ shdr = elf_shdr_entry_addr(ehdr, ehdr->e_shnum++);
+ phdr = elf_phdr_entry_addr(ehdr, ehdr->e_phnum++);
+
+ shdr->sh_type = SHT_PROGBITS;
+ shdr->sh_name = append_str_to_strtable(md, region->name);
+ shdr->sh_addr = (elf_addr_t)region->virt_addr;
+ shdr->sh_size = region->size;
+ shdr->sh_flags = SHF_WRITE;
+ shdr->sh_offset = md->elf.elf_offset;
+ shdr->sh_entsize = 0;
+
+ phdr->p_type = PT_LOAD;
+ phdr->p_offset = md->elf.elf_offset;
+ phdr->p_vaddr = (elf_addr_t)region->virt_addr;
+ phdr->p_paddr = region->phys_addr;
+ phdr->p_filesz = phdr->p_memsz = region->size;
+ phdr->p_flags = PF_R | PF_W;
+ md->elf.elf_offset += shdr->sh_size;
+}
+
+static void qcom_md_add_region(struct minidump_ss_data *mdss_data,
+ const struct qcom_minidump_region *region)
+{
+ struct minidump_subsystem *mdss_toc = mdss_data->md_ss_toc;
+ struct minidump_region *mdr;
+ unsigned int region_cnt;
+
+ region_cnt = le32_to_cpu(mdss_toc->region_count);
+ mdr = &mdss_data->md_regions[region_cnt];
+ strscpy(mdr->name, region->name, sizeof(mdr->name));
+ mdr->address = cpu_to_le64(region->phys_addr);
+ mdr->size = cpu_to_le64(region->size);
+ mdr->valid = cpu_to_le32(MINIDUMP_REGION_VALID);
+ region_cnt++;
+ mdss_toc->region_count = cpu_to_le32(region_cnt);
+}
+
+static int qcom_md_get_region_index(struct minidump_ss_data *mdss_data,
+ const struct qcom_minidump_region *region)
+{
+ struct minidump_subsystem *mdss_toc = mdss_data->md_ss_toc;
+ struct minidump_region *mdr;
+ unsigned int i;
+ unsigned int count;
+
+ count = le32_to_cpu(mdss_toc->region_count);
+ for (i = 0; i < count; i++) {
+ mdr = &mdss_data->md_regions[i];
+ if (!strcmp(mdr->name, region->name))
+ return i;
+ }
+
+ return -ENOENT;
+}
+
+static int qcom_md_region_unregister(struct minidump *md,
+ const struct qcom_minidump_region *region)
+{
+ struct minidump_ss_data *mdss_data = md->apss_data;
+ struct minidump_subsystem *mdss_toc = mdss_data->md_ss_toc;
+ struct minidump_region *mdr;
+ unsigned int region_cnt;
+ unsigned int idx;
+ int ret;
+
+ ret = qcom_md_get_region_index(mdss_data, region);
+ if (ret < 0) {
+ dev_err(md->dev, "%s region is not present\n", region->name);
+ return ret;
+ }
+
+ idx = ret;
+ mdr = &mdss_data->md_regions[0];
+ region_cnt = le32_to_cpu(mdss_toc->region_count);
+ /*
+ * Left shift all the regions exist after this removed region
+ * index by 1 to fill the gap and zero out the last region
+ * present at the end.
+ */
+ memmove(&mdr[idx], &mdr[idx + 1], (region_cnt - idx - 1) * sizeof(*mdr));
+ memset(&mdr[region_cnt - 1], 0, sizeof(*mdr));
+ region_cnt--;
+ mdss_toc->region_count = cpu_to_le32(region_cnt);
+
+ return 0;
+}
+
+static int qcom_md_region_register(struct minidump *md,
+ const struct qcom_minidump_region *region)
+{
+ struct minidump_ss_data *mdss_data = md->apss_data;
+ struct minidump_subsystem *mdss_toc = mdss_data->md_ss_toc;
+ unsigned int num_region;
+ int ret;
+
+ ret = qcom_md_get_region_index(mdss_data, region);
+ if (ret >= 0) {
+ dev_info(md->dev, "%s region is already registered\n", region->name);
+ return -EEXIST;
+ }
+
+ /* Check if there is a room for a new entry */
+ num_region = le32_to_cpu(mdss_toc->region_count);
+ if (num_region >= MAX_NUM_ENTRIES) {
+ dev_err(md->dev, "maximum region limit %u reached\n", num_region);
+ return -ENOSPC;
+ }
+
+ qcom_md_add_region(mdss_data, region);
+
+ return 0;
+}
+
+static bool qcom_minidump_valid_region(const struct qcom_minidump_region *region)
+{
+ return region &&
+ strnlen(region->name, MAX_NAME_LENGTH) < MAX_NAME_LENGTH &&
+ region->virt_addr &&
+ region->size &&
+ IS_ALIGNED(region->size, 4);
+}
+
+/**
+ * qcom_minidump_region_register() - Register region in APSS Minidump table.
+ * @region: minidump region.
+ *
+ * Return: On success, it returns 0 and negative error value on failure.
+ */
+static int qcom_minidump_region_register(const struct qcom_minidump_region *region)
+{
+ struct minidump *md;
+ int ret;
+
+ md = qcom_smem_minidump_ready();
+ if (!md)
+ return -EPROBE_DEFER;
+
+ if (!qcom_minidump_valid_region(region))
+ return -EINVAL;
+
+ mutex_lock(&md->md_lock);
+ ret = qcom_md_region_register(md, region);
+ if (ret)
+ goto unlock;
+
+ qcom_md_update_elfheader(md, region);
+unlock:
+ mutex_unlock(&md->md_lock);
+ return ret;
+}
+
+/**
+ * qcom_minidump_region_unregister() - Unregister region from APSS Minidump table.
+ * @region: minidump region.
+ *
+ * Return: On success, it returns 0 and negative error value on failure.
+ */
+static int qcom_minidump_region_unregister(const struct qcom_minidump_region *region)
+{
+ struct minidump *md;
+ int ret;
+
+ md = qcom_smem_minidump_ready();
+ if (!md)
+ return -EPROBE_DEFER;
+
+ if (!qcom_minidump_valid_region(region))
+ return -EINVAL;
+
+ mutex_lock(&md->md_lock);
+ ret = qcom_md_region_unregister(md, region);
+ if (ret)
+ goto unlock;
+
+ ret = qcom_md_clear_elfheader(md, region);
+unlock:
+ mutex_unlock(&md->md_lock);
+ return ret;
+}
+
+static int qcom_md_add_elfheader(struct minidump *md)
+{
+ struct qcom_minidump_region elfregion;
+ struct elfhdr *ehdr;
+ struct elf_shdr *shdr;
+ struct elf_phdr *phdr;
+ unsigned int elfh_size;
+ unsigned int strtbl_off;
+ unsigned int phdr_off;
+ unsigned int banner_len;
+ char *banner;
+
+ banner_len = strlen(linux_banner);
+ /*
+ * Header buffer contains:
+ * ELF header, (MAX_NUM_ENTRIES + 4) of Section and Program ELF headers,
+ * where, 4 additional entries, one for empty header, one for string table
+ * one for minidump table and one for linux banner.
+ *
+ * Linux banner is stored in minidump to aid post mortem tools to determine
+ * the kernel version.
+ */
+ elfh_size = sizeof(*ehdr);
+ elfh_size += MAX_STRTBL_SIZE;
+ elfh_size += banner_len + 1;
+ elfh_size += ((sizeof(*shdr) + sizeof(*phdr)) * (MAX_NUM_ENTRIES + 4));
+ elfh_size = ALIGN(elfh_size, 4);
+
+ md->elf.ehdr = devm_kzalloc(md->dev, elfh_size, GFP_KERNEL);
+ if (!md->elf.ehdr)
+ return -ENOMEM;
+
+ ehdr = md->elf.ehdr;
+ /* Assign Section/Program headers offset */
+ md->elf.shdr = shdr = (struct elf_shdr *)(ehdr + 1);
+ md->elf.phdr = phdr = (struct elf_phdr *)(shdr + MAX_NUM_ENTRIES);
+ phdr_off = sizeof(*ehdr) + (sizeof(*shdr) * MAX_NUM_ENTRIES);
+
+ memcpy(ehdr->e_ident, ELFMAG, SELFMAG);
+ ehdr->e_ident[EI_CLASS] = ELF_CLASS;
+ ehdr->e_ident[EI_DATA] = ELF_DATA;
+ ehdr->e_ident[EI_VERSION] = EV_CURRENT;
+ ehdr->e_ident[EI_OSABI] = ELF_OSABI;
+ ehdr->e_type = ET_CORE;
+ ehdr->e_machine = ELF_ARCH;
+ ehdr->e_version = EV_CURRENT;
+ ehdr->e_ehsize = sizeof(*ehdr);
+ ehdr->e_phoff = phdr_off;
+ ehdr->e_phentsize = sizeof(*phdr);
+ ehdr->e_shoff = sizeof(*ehdr);
+ ehdr->e_shentsize = sizeof(*shdr);
+ ehdr->e_shstrndx = 1;
+
+ md->elf.elf_offset = elfh_size;
+ /*
+ * The zeroth index of the section header is reserved and is rarely used.
+ * Set the section header as null (SHN_UNDEF) and move to the next one.
+ * 2nd Section is String table.
+ */
+ md->elf.strtable_idx = 1;
+ strtbl_off = sizeof(*ehdr) + ((sizeof(*phdr) + sizeof(*shdr)) * MAX_NUM_ENTRIES);
+ shdr++;
+ shdr->sh_type = SHT_STRTAB;
+ shdr->sh_offset = (elf_addr_t)strtbl_off;
+ shdr->sh_size = MAX_STRTBL_SIZE;
+ shdr->sh_entsize = 0;
+ shdr->sh_flags = 0;
+ shdr->sh_name = append_str_to_strtable(md, "STR_TBL");
+ shdr++;
+
+ /* 3rd Section is Linux banner */
+ banner = (char *)ehdr + strtbl_off + MAX_STRTBL_SIZE;
+ memcpy(banner, linux_banner, banner_len);
+
+ shdr->sh_type = SHT_PROGBITS;
+ shdr->sh_offset = (elf_addr_t)(strtbl_off + MAX_STRTBL_SIZE);
+ shdr->sh_size = banner_len + 1;
+ shdr->sh_addr = (elf_addr_t)linux_banner;
+ shdr->sh_entsize = 0;
+ shdr->sh_flags = SHF_WRITE;
+ shdr->sh_name = append_str_to_strtable(md, "linux_banner");
+
+ phdr->p_type = PT_LOAD;
+ phdr->p_offset = (elf_addr_t)(strtbl_off + MAX_STRTBL_SIZE);
+ phdr->p_vaddr = (elf_addr_t)linux_banner;
+ phdr->p_paddr = virt_to_phys(linux_banner);
+ phdr->p_filesz = phdr->p_memsz = banner_len + 1;
+ phdr->p_flags = PF_R | PF_W;
+
+ /*
+ * Above are some prdefined sections/program header used
+ * for debug, update their count here.
+ */
+ ehdr->e_phnum = 1;
+ ehdr->e_shnum = 3;
+
+ /* Register ELF header as first region */
+ strscpy(elfregion.name, "KELF_HEADER", sizeof(elfregion.name));
+ elfregion.virt_addr = md->elf.ehdr;
+ elfregion.phys_addr = virt_to_phys(md->elf.ehdr);
+ elfregion.size = elfh_size;
+
+ return qcom_md_region_register(md, &elfregion);
+}
+
+static int qcom_apss_md_table_init(struct minidump *md,
+ struct minidump_subsystem *mdss_toc)
+{
+ struct minidump_ss_data *mdss_data;
+
+ mdss_data = devm_kzalloc(md->dev, sizeof(*mdss_data), GFP_KERNEL);
+ if (!mdss_data)
+ return -ENOMEM;
+
+ mdss_data->md_ss_toc = mdss_toc;
+ mdss_data->md_regions = devm_kcalloc(md->dev, MAX_NUM_ENTRIES,
+ sizeof(*mdss_data->md_regions),
+ GFP_KERNEL);
+ if (!mdss_data->md_regions)
+ return -ENOMEM;
+
+ mdss_toc = mdss_data->md_ss_toc;
+ mdss_toc->regions_baseptr = cpu_to_le64(virt_to_phys(mdss_data->md_regions));
+ mdss_toc->enabled = cpu_to_le32(MINIDUMP_SS_ENABLED);
+ mdss_toc->status = cpu_to_le32(1);
+ mdss_toc->region_count = cpu_to_le32(0);
+
+ /* Tell bootloader not to encrypt the regions of this subsystem */
+ mdss_toc->encryption_status = cpu_to_le32(MINIDUMP_SS_ENCR_DONE);
+ mdss_toc->encryption_required = cpu_to_le32(MINIDUMP_SS_ENCR_NOTREQ);
+
+ md->apss_data = mdss_data;
+
+ return 0;
+}
+
+static void qcom_apss_md_table_exit(struct minidump_ss_data *mdss_data)
+{
+ memset(mdss_data->md_ss_toc, cpu_to_le32(0), sizeof(*mdss_data->md_ss_toc));
+}
+
+static int qcom_apss_minidump_probe(struct platform_device *pdev)
+{
+ struct minidump_global_toc *mdgtoc;
+ struct minidump *md;
+ size_t size;
+ int ret;
+
+ md = devm_kzalloc(&pdev->dev, sizeof(*md), GFP_KERNEL);
+ if (!md)
+ return -ENOMEM;
+
+ md->dev = &pdev->dev;
+ mdgtoc = qcom_smem_get(QCOM_SMEM_HOST_ANY, SBL_MINIDUMP_SMEM_ID, &size);
+ if (IS_ERR(mdgtoc)) {
+ ret = PTR_ERR(mdgtoc);
+ return dev_err_probe(md->dev, ret,
+ "Couldn't find minidump smem item\n");
+ }
+
+ if (size < sizeof(*mdgtoc) || !mdgtoc->status) {
+ ret = -EINVAL;
+ return dev_err_probe(md->dev, ret,
+ "minidump table is not initialized\n");
+ }
+
+ mutex_init(&md->md_lock);
+ ret = qcom_apss_md_table_init(md, &mdgtoc->subsystems[MINIDUMP_APSS_DESC]);
+ if (ret)
+ return dev_err_probe(md->dev, ret,
+ "apss minidump initialization failed\n");
+
+ /* First entry would be ELF header */
+ ret = qcom_md_add_elfheader(md);
+ if (ret) {
+ qcom_apss_md_table_exit(md->apss_data);
+ return dev_err_probe(md->dev, ret, "Failed to add elf header\n");
+ }
+
+ platform_set_drvdata(pdev, md);
+
+ return ret;
+}
+
+static void qcom_apss_minidump_remove(struct platform_device *pdev)
+{
+ struct minidump *md = platform_get_drvdata(pdev);
+
+ qcom_apss_md_table_exit(md->apss_data);
+}
+
+static const struct platform_device_id qcom_minidump_id_table[] = {
+ { .name = "qcom_minidump_smem" },
+ {}
+};
+MODULE_DEVICE_TABLE(platform, qcom_minidump_id_table);
+
+static struct platform_driver qcom_minidump_driver = {
+ .probe = qcom_apss_minidump_probe,
+ .remove_new = qcom_apss_minidump_remove,
+ .driver = {
+ .name = "qcom_minidump_smem",
+ },
+ .id_table = qcom_minidump_id_table,
+};
+
+module_platform_driver(qcom_minidump_driver);
+
+MODULE_DESCRIPTION("Qualcomm APSS minidump driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/soc/qcom/qcom_minidump_internal.h b/drivers/soc/qcom/qcom_minidump_internal.h
index 71709235b196..a2aebe5b690a 100644
--- a/drivers/soc/qcom/qcom_minidump_internal.h
+++ b/drivers/soc/qcom/qcom_minidump_internal.h
@@ -9,10 +9,20 @@
#define MAX_NUM_OF_SS 10
#define MAX_REGION_NAME_LENGTH 16
#define SBL_MINIDUMP_SMEM_ID 602
+
#define MINIDUMP_REGION_VALID ('V' << 24 | 'A' << 16 | 'L' << 8 | 'I' << 0)
+#define MINIDUMP_REGION_INVALID ('I' << 24 | 'N' << 16 | 'V' << 8 | 'A' << 0)
+#define MINIDUMP_REGION_INIT ('I' << 24 | 'N' << 16 | 'I' << 8 | 'T' << 0)
+#define MINIDUMP_REGION_NOINIT 0
+
+#define MINIDUMP_SS_ENCR_REQ (0 << 24 | 'Y' << 16 | 'E' << 8 | 'S' << 0)
+#define MINIDUMP_SS_ENCR_NOTREQ (0 << 24 | 0 << 16 | 'N' << 8 | 'R' << 0)
+#define MINIDUMP_SS_ENCR_START ('S' << 24 | 'T' << 16 | 'R' << 8 | 'T' << 0)
#define MINIDUMP_SS_ENCR_DONE ('D' << 24 | 'O' << 16 | 'N' << 8 | 'E' << 0)
#define MINIDUMP_SS_ENABLED ('E' << 24 | 'N' << 16 | 'B' << 8 | 'L' << 0)
+#define MINIDUMP_APSS_DESC 0
+
/**
* struct minidump_region - Minidump region
* @name : Name of the region to be dumped
diff --git a/drivers/soc/qcom/smem.c b/drivers/soc/qcom/smem.c
index 7191fa0c087f..f7632624886f 100644
--- a/drivers/soc/qcom/smem.c
+++ b/drivers/soc/qcom/smem.c
@@ -270,6 +270,7 @@ struct smem_region {
* @partitions: list of partitions of current processor/host
* @item_count: max accepted item number
* @socinfo: platform device pointer
+ * @minidump: minidump platform device pointer
* @num_regions: number of @regions
* @regions: list of the memory regions defining the shared memory
*/
@@ -280,6 +281,7 @@ struct qcom_smem {
u32 item_count;
struct platform_device *socinfo;
+ struct platform_device *minidump;
struct smem_ptable *ptable;
struct smem_partition global_partition;
struct smem_partition partitions[SMEM_HOST_COUNT];
@@ -795,6 +797,15 @@ int qcom_smem_get_soc_id(u32 *id)
}
EXPORT_SYMBOL_GPL(qcom_smem_get_soc_id);
+void *qcom_smem_minidump_ready(void)
+{
+ if (__smem && __smem->minidump)
+ return platform_get_drvdata(__smem->minidump);
+ else
+ return NULL;
+}
+EXPORT_SYMBOL_GPL(qcom_smem_minidump_ready);
+
static int qcom_smem_get_sbl_version(struct qcom_smem *smem)
{
struct smem_header *header;
@@ -1173,11 +1184,20 @@ static int qcom_smem_probe(struct platform_device *pdev)
if (IS_ERR(smem->socinfo))
dev_dbg(&pdev->dev, "failed to register socinfo device\n");
+ smem->minidump = platform_device_register_data(&pdev->dev, "qcom_minidump_smem",
+ PLATFORM_DEVID_NONE, NULL,
+ 0);
+ if (IS_ERR(smem->minidump)) {
+ dev_dbg(&pdev->dev, "failed to register minidump device\n");
+ smem->minidump = NULL;
+ }
+
return 0;
}
static void qcom_smem_remove(struct platform_device *pdev)
{
+ platform_device_unregister(__smem->minidump);
platform_device_unregister(__smem->socinfo);
hwspin_lock_free(__smem->hwlock);
diff --git a/include/linux/soc/qcom/smem.h b/include/linux/soc/qcom/smem.h
index a36a3b9d4929..08288360a55e 100644
--- a/include/linux/soc/qcom/smem.h
+++ b/include/linux/soc/qcom/smem.h
@@ -14,4 +14,6 @@ phys_addr_t qcom_smem_virt_to_phys(void *p);
int qcom_smem_get_soc_id(u32 *id);
+void *qcom_smem_minidump_ready(void);
+
#endif
diff --git a/include/soc/qcom/qcom_minidump.h b/include/soc/qcom/qcom_minidump.h
index cd87caef919d..9fdf9e9d0af3 100644
--- a/include/soc/qcom/qcom_minidump.h
+++ b/include/soc/qcom/qcom_minidump.h
@@ -6,6 +6,24 @@
#ifndef _QCOM_MINIDUMP_H_
#define _QCOM_MINIDUMP_H_
+#define MAX_NAME_LENGTH 12
+
+/**
+ * struct qcom_minidump_region - APSS Minidump region information
+ *
+ * @name: Entry name, Minidump will dump binary with this name.
+ * @virt_addr: Virtual address of the entry.
+ * @phys_addr: Physical address of the entry to dump.
+ * @size: Number of byte to dump from @address location,
+ * and it should be 4 byte aligned.
+ */
+struct qcom_minidump_region {
+ char name[MAX_NAME_LENGTH];
+ void *virt_addr;
+ phys_addr_t phys_addr;
+ size_t size;
+};
+
struct rproc;
struct rproc_dump_segment;
--
2.43.0.254.ga26002b62827
Register ramoops region with APSS minidump via ramoops
info notifier so that these region gets captured on
system crash.
Signed-off-by: Mukesh Ojha <[email protected]>
---
drivers/soc/qcom/qcom_minidump.c | 84 ++++++++++++++++++++++++++++++++
1 file changed, 84 insertions(+)
diff --git a/drivers/soc/qcom/qcom_minidump.c b/drivers/soc/qcom/qcom_minidump.c
index c0f76a51d0e8..7ca43d938b13 100644
--- a/drivers/soc/qcom/qcom_minidump.c
+++ b/drivers/soc/qcom/qcom_minidump.c
@@ -18,8 +18,10 @@
#include <linux/mutex.h>
#include <linux/platform_device.h>
#include <linux/printk.h>
+#include <linux/pstore_ram.h>
#include <linux/soc/qcom/smem.h>
#include <linux/string.h>
+#include <linux/workqueue.h>
#include <soc/qcom/qcom_minidump.h>
#include "qcom_minidump_internal.h"
@@ -56,12 +58,22 @@ struct minidump_elfhdr {
* @dev: Minidump backend device
* @apss_data: APSS driver data
* @md_lock: Lock to protect access to APSS minidump table
+ * @work: Minidump work for any required execution in process context.
+ * @nb_cookie: Save the cookie, will be used for unregistering the callback.
*/
struct minidump {
struct minidump_elfhdr elf;
struct device *dev;
struct minidump_ss_data *apss_data;
struct mutex md_lock;
+ struct work_struct work;
+ void *nb_cookie;
+};
+
+static LIST_HEAD(apss_md_rlist);
+struct md_region_list {
+ struct qcom_minidump_region md_region;
+ struct list_head list;
};
/*
@@ -530,6 +542,58 @@ static int qcom_apss_md_table_init(struct minidump *md,
return 0;
}
+static int register_ramoops_region(const char *name, int id, void *vaddr,
+ phys_addr_t paddr, size_t size)
+{
+ struct qcom_minidump_region *md_region;
+ struct md_region_list *mdr_list;
+ int ret;
+
+ mdr_list = kzalloc(sizeof(*mdr_list), GFP_KERNEL);
+ if (!mdr_list)
+ return -ENOMEM;
+
+ md_region = &mdr_list->md_region;
+ scnprintf(md_region->name, sizeof(md_region->name), "K%s%d", name, id);
+ md_region->virt_addr = vaddr;
+ md_region->phys_addr = paddr;
+ md_region->size = size;
+ ret = qcom_minidump_region_register(md_region);
+ if (ret < 0) {
+ pr_err("failed to register region in minidump: err: %d\n", ret);
+ return ret;
+ }
+
+ list_add(&mdr_list->list, &apss_md_rlist);
+
+ return 0;
+}
+
+static void register_ramoops_minidump_cb(struct work_struct *work)
+{
+ struct minidump *md = container_of(work, struct minidump, work);
+
+ md->nb_cookie = register_ramoops_info_notifier(register_ramoops_region);
+ if (IS_ERR_OR_NULL(md->nb_cookie)) {
+ pr_err("Fail to register ramoops info notifier\n");
+ md->nb_cookie = NULL;
+ }
+}
+
+static void qcom_ramoops_minidump_unregister(void)
+{
+ struct md_region_list *mdr_list;
+ struct md_region_list *tmp;
+
+ list_for_each_entry_safe(mdr_list, tmp, &apss_md_rlist, list) {
+ struct qcom_minidump_region *region;
+
+ region = &mdr_list->md_region;
+ qcom_minidump_region_unregister(region);
+ list_del(&mdr_list->list);
+ }
+}
+
static void qcom_apss_md_table_exit(struct minidump_ss_data *mdss_data)
{
memset(mdss_data->md_ss_toc, cpu_to_le32(0), sizeof(*mdss_data->md_ss_toc));
@@ -575,6 +639,22 @@ static int qcom_apss_minidump_probe(struct platform_device *pdev)
platform_set_drvdata(pdev, md);
+ /*
+ * Use separate context for registering ramoops region via workqueue
+ * as minidump probe can get called in same context of platform device
+ * register call from smem driver and further call to qcom_smem_minidump_ready()
+ * can return -EPROBE_DEFER as __smem->minidump is not yet initialized because
+ * of same context and it can only initialized after return from probe.
+ *
+ * qcom_apss_minidump_probe()
+ * register_ramoops_minidump_cb()
+ * register_ramoops_region()
+ * qcom_minidump_region_register()
+ * qcom_smem_minidump_ready()
+ */
+ INIT_WORK(&md->work, register_ramoops_minidump_cb);
+ schedule_work(&md->work);
+
return ret;
}
@@ -582,6 +662,10 @@ static void qcom_apss_minidump_remove(struct platform_device *pdev)
{
struct minidump *md = platform_get_drvdata(pdev);
+ flush_work(&md->work);
+ qcom_ramoops_minidump_unregister();
+ if (md->nb_cookie)
+ unregister_ramoops_info_notifier(md->nb_cookie);
qcom_apss_md_table_exit(md->apss_data);
}
--
2.43.0.254.ga26002b62827
The reserved memory region for ramoops is assumed to be at a fixed
and known location when read from the devicetree. This may not be
required for something like Qualcomm's minidump which is interested
in knowing addresses of ramoops region but it does not put hard
requirement of address being fixed as most of its SoC does not
support warm reset and does not use pstorefs at all instead it has
firmware way of collecting ramoops region if it gets to know the
address and register it with apss minidump table which is sitting
in shared memory region in DDR and firmware will have access to
these table during reset and collects it on crash of SoC.
So, add the support of reserving ramoops region to be dynamically
allocated early during boot if it is request through command line
via 'dyn_ramoops_size=<size>' and fill up reserved resource structure
and export the structure, so that it can be read by ramoops driver.
Signed-off-by: Mukesh Ojha <[email protected]>
---
Documentation/admin-guide/ramoops.rst | 23 +++++++++-
fs/pstore/Kconfig | 15 ++++++
fs/pstore/ram.c | 66 +++++++++++++++++++++++++--
include/linux/pstore_ram.h | 5 ++
init/main.c | 3 ++
5 files changed, 108 insertions(+), 4 deletions(-)
diff --git a/Documentation/admin-guide/ramoops.rst b/Documentation/admin-guide/ramoops.rst
index e9f85142182d..6de61002f9e9 100644
--- a/Documentation/admin-guide/ramoops.rst
+++ b/Documentation/admin-guide/ramoops.rst
@@ -33,6 +33,13 @@ memory are implementation defined, and won't work on many ARMs such as omaps.
Setting ``mem_type=2`` attempts to treat the memory region as normal memory,
which enables full cache on it. This can improve the performance.
+Ramoops supports its memory to be allocated dynamically during early boot
+for plaforms that do not have support for warm boot i.e., no assurance
+that Ram content will be preserved across boot and for these platforms
+giving static Ramoops memory is not necessary as it has separate backend
+mechanism to retrieve ramoops content on system failure. More about
+how to enable Dynamic ramoops in ``Setting the parameters`` A.b section.
+
The memory area is divided into ``record_size`` chunks (also rounded down to
power of two) and each kmesg dump writes a ``record_size`` chunk of
information.
@@ -59,7 +66,7 @@ Setting the parameters
Setting the ramoops parameters can be done in several different manners:
- A. Use the module parameters (which have the names of the variables described
+ A.a Use the module parameters (which have the names of the variables described
as before). For quick debugging, you can also reserve parts of memory during
boot and then use the reserved memory for ramoops. For example, assuming a
machine with > 128 MB of memory, the following kernel command line will tell
@@ -68,6 +75,20 @@ Setting the ramoops parameters can be done in several different manners:
mem=128M ramoops.mem_address=0x8000000 ramoops.ecc=1
+ A.b Ramoops memory can be also be dynamically reserved by Kernel and in such
+ scenario ``mem_address`` i.e., Ramoops base address can be anywhere in the RAM
+ instead of being fixed and predefined. A separate command line option
+ ``dyn_ramoops_size=<size>`` and kernel config CONFIG_PSTORE_DYNAMIC_RAMOOPS
+ are provided to facilitate Dynamic Ramoops memory reservation during early boot.
+ The command line option and the config should only be used in the presence of
+ separate backend which knows how to recover Dynamic Ramoops region otherwise
+ regular ramoops functionality will be impacted.
+ ``mem_size`` should not be used if Dynamic Ramoops support is requested and if
+ both are given ``mem_size`` value is overwritten with ``dyn_ramoops_size`` value
+ i.e., Dynamic Ramoops takes precedence::
+
+ dyn_ramoops_size=2M ramoops.console_size=2097152
+
B. Use Device Tree bindings, as described in
``Documentation/devicetree/bindings/reserved-memory/ramoops.yaml``.
For example::
diff --git a/fs/pstore/Kconfig b/fs/pstore/Kconfig
index 3acc38600cd1..b8bdbd2f0e73 100644
--- a/fs/pstore/Kconfig
+++ b/fs/pstore/Kconfig
@@ -81,6 +81,21 @@ config PSTORE_RAM
For more information, see Documentation/admin-guide/ramoops.rst.
+config PSTORE_DYNAMIC_RAMOOPS
+ bool "Reserve ramoops region dynamically"
+ select PSTORE_RAM
+ help
+ This enables the dynamic reservation of ramoops region for a special case
+ where there is no need to access the logs from pstorefs on next boot;
+ instead there is separate backend mechanism like minidump present which has
+ awareness about the dynamic ramoops region and can recover the logs. This is
+ enabled via command line parameter dyn_ramoops_size=<size> and should not be
+ used in absence of separate backend which knows how to recover this dynamic
+ region.
+
+ Note whenever this config is selected ramoops driver will be built statically
+ into kernel.
+
config PSTORE_ZONE
tristate
depends on PSTORE
diff --git a/fs/pstore/ram.c b/fs/pstore/ram.c
index 88b34fdbf759..1faf0835700b 100644
--- a/fs/pstore/ram.c
+++ b/fs/pstore/ram.c
@@ -20,6 +20,7 @@
#include <linux/compiler.h>
#include <linux/of.h>
#include <linux/of_address.h>
+#include <linux/memblock.h>
#include <linux/mm.h>
#include "internal.h"
@@ -103,6 +104,59 @@ struct ramoops_context {
};
static struct platform_device *dummy;
+static struct resource dyn_ramoops_res = {
+ .name = "ramoops",
+ .start = 0,
+ .end = 0,
+ .flags = IORESOURCE_BUSY | IORESOURCE_SYSTEM_RAM,
+ .desc = IORES_DESC_NONE,
+};
+static int dyn_ramoops_size;
+
+#ifdef CONFIG_PSTORE_DYNAMIC_RAMOOPS
+static int __init parse_dyn_ramoops_size(char *p)
+{
+ char *tmp;
+
+ dyn_ramoops_size = memparse(p, &tmp);
+ if (p == tmp) {
+ pr_err("ramoops: memory size expected\n");
+ dyn_ramoops_size = 0;
+ return -EINVAL;
+ }
+
+ return 0;
+}
+early_param("dyn_ramoops_size", parse_dyn_ramoops_size);
+
+/*
+ * setup_dynamic_ramoops() - Reserve memory for dynamic ramoops
+ *
+ * Enables dynamic reserve memory support for ramoops through
+ * command line.
+ */
+void __init setup_dynamic_ramoops(void)
+{
+ unsigned long long ramoops_base;
+ unsigned long long ramoops_size;
+
+ if (!dyn_ramoops_size)
+ return;
+
+ ramoops_base = memblock_phys_alloc_range(dyn_ramoops_size, SMP_CACHE_BYTES,
+ 0, MEMBLOCK_ALLOC_NOLEAKTRACE);
+ if (!ramoops_base) {
+ pr_err("cannot allocate ramoops dynamic memory (size:0x%llx).\n",
+ ramoops_size);
+ dyn_ramoops_size = 0;
+ return;
+ }
+
+ dyn_ramoops_res.start = ramoops_base;
+ dyn_ramoops_res.end = ramoops_base + dyn_ramoops_size - 1;
+ insert_resource(&iomem_resource, &dyn_ramoops_res);
+}
+#endif
static int ramoops_pstore_open(struct pstore_info *psi)
{
@@ -915,13 +969,19 @@ static void __init ramoops_register_dummy(void)
/*
* Prepare a dummy platform data structure to carry the module
- * parameters. If mem_size isn't set, then there are no module
- * parameters, and we can skip this.
+ * parameters.
+ *
+ * dyn_ramoops_size takes precedence over mem_size if it is
+ * enabled and valid.
*/
- if (!mem_size)
+ if (!dyn_ramoops_size && !mem_size)
return;
pr_info("using module parameters\n");
+ if (dyn_ramoops_size) {
+ mem_size = dyn_ramoops_size;
+ mem_address = dyn_ramoops_res.start;
+ }
memset(&pdata, 0, sizeof(pdata));
pdata.mem_size = mem_size;
diff --git a/include/linux/pstore_ram.h b/include/linux/pstore_ram.h
index 9d65ff94e216..1efff7a38333 100644
--- a/include/linux/pstore_ram.h
+++ b/include/linux/pstore_ram.h
@@ -39,4 +39,9 @@ struct ramoops_platform_data {
struct persistent_ram_ecc_info ecc_info;
};
+#ifdef CONFIG_PSTORE_DYNAMIC_RAMOOPS
+void __init setup_dynamic_ramoops(void);
+#else
+static inline void __init setup_dynamic_ramoops(void) {}
+#endif
#endif
diff --git a/init/main.c b/init/main.c
index ef3ce41b8fc5..3d74241bcb2b 100644
--- a/init/main.c
+++ b/init/main.c
@@ -99,6 +99,8 @@
#include <linux/init_syscalls.h>
#include <linux/stackdepot.h>
#include <linux/randomize_kstack.h>
+#include <linux/moduleloader.h>
+#include <linux/pstore_ram.h>
#include <net/net_namespace.h>
#include <asm/io.h>
@@ -890,6 +892,7 @@ void start_kernel(void)
pr_notice("%s", linux_banner);
early_security_init();
setup_arch(&command_line);
+ setup_dynamic_ramoops();
setup_boot_config();
setup_command_line(command_line);
setup_nr_cpu_ids();
--
2.43.0.254.ga26002b62827
Hi @Kees/@Tony/@Guilherme,
Wanted to get your early feedback, especially on 8th and 9th patch.
please suggest.
-Mukesh
On 1/31/2024 4:38 PM, Mukesh Ojha wrote:
> The reserved memory region for ramoops is assumed to be at a fixed
> and known location when read from the devicetree. This may not be
> required for something like Qualcomm's minidump which is interested
> in knowing addresses of ramoops region but it does not put hard
> requirement of address being fixed as most of its SoC does not
> support warm reset and does not use pstorefs at all instead it has
> firmware way of collecting ramoops region if it gets to know the
> address and register it with apss minidump table which is sitting
> in shared memory region in DDR and firmware will have access to
> these table during reset and collects it on crash of SoC.
>
> So, add the support of reserving ramoops region to be dynamically
> allocated early during boot if it is request through command line
> via 'dyn_ramoops_size=<size>' and fill up reserved resource structure
> and export the structure, so that it can be read by ramoops driver.
>
> Signed-off-by: Mukesh Ojha <[email protected]>
> ---
> Documentation/admin-guide/ramoops.rst | 23 +++++++++-
> fs/pstore/Kconfig | 15 ++++++
> fs/pstore/ram.c | 66 +++++++++++++++++++++++++--
> include/linux/pstore_ram.h | 5 ++
> init/main.c | 3 ++
> 5 files changed, 108 insertions(+), 4 deletions(-)
>
> diff --git a/Documentation/admin-guide/ramoops.rst b/Documentation/admin-guide/ramoops.rst
> index e9f85142182d..6de61002f9e9 100644
> --- a/Documentation/admin-guide/ramoops.rst
> +++ b/Documentation/admin-guide/ramoops.rst
> @@ -33,6 +33,13 @@ memory are implementation defined, and won't work on many ARMs such as omaps.
> Setting ``mem_type=2`` attempts to treat the memory region as normal memory,
> which enables full cache on it. This can improve the performance.
>
> +Ramoops supports its memory to be allocated dynamically during early boot
> +for plaforms that do not have support for warm boot i.e., no assurance
> +that Ram content will be preserved across boot and for these platforms
> +giving static Ramoops memory is not necessary as it has separate backend
> +mechanism to retrieve ramoops content on system failure. More about
> +how to enable Dynamic ramoops in ``Setting the parameters`` A.b section.
> +
> The memory area is divided into ``record_size`` chunks (also rounded down to
> power of two) and each kmesg dump writes a ``record_size`` chunk of
> information.
> @@ -59,7 +66,7 @@ Setting the parameters
>
> Setting the ramoops parameters can be done in several different manners:
>
> - A. Use the module parameters (which have the names of the variables described
> + A.a Use the module parameters (which have the names of the variables described
> as before). For quick debugging, you can also reserve parts of memory during
> boot and then use the reserved memory for ramoops. For example, assuming a
> machine with > 128 MB of memory, the following kernel command line will tell
> @@ -68,6 +75,20 @@ Setting the ramoops parameters can be done in several different manners:
>
> mem=128M ramoops.mem_address=0x8000000 ramoops.ecc=1
>
> + A.b Ramoops memory can be also be dynamically reserved by Kernel and in such
> + scenario ``mem_address`` i.e., Ramoops base address can be anywhere in the RAM
> + instead of being fixed and predefined. A separate command line option
> + ``dyn_ramoops_size=<size>`` and kernel config CONFIG_PSTORE_DYNAMIC_RAMOOPS
> + are provided to facilitate Dynamic Ramoops memory reservation during early boot.
> + The command line option and the config should only be used in the presence of
> + separate backend which knows how to recover Dynamic Ramoops region otherwise
> + regular ramoops functionality will be impacted.
> + ``mem_size`` should not be used if Dynamic Ramoops support is requested and if
> + both are given ``mem_size`` value is overwritten with ``dyn_ramoops_size`` value
> + i.e., Dynamic Ramoops takes precedence::
> +
> + dyn_ramoops_size=2M ramoops.console_size=2097152
> +
> B. Use Device Tree bindings, as described in
> ``Documentation/devicetree/bindings/reserved-memory/ramoops.yaml``.
> For example::
> diff --git a/fs/pstore/Kconfig b/fs/pstore/Kconfig
> index 3acc38600cd1..b8bdbd2f0e73 100644
> --- a/fs/pstore/Kconfig
> +++ b/fs/pstore/Kconfig
> @@ -81,6 +81,21 @@ config PSTORE_RAM
>
> For more information, see Documentation/admin-guide/ramoops.rst.
>
> +config PSTORE_DYNAMIC_RAMOOPS
> + bool "Reserve ramoops region dynamically"
> + select PSTORE_RAM
> + help
> + This enables the dynamic reservation of ramoops region for a special case
> + where there is no need to access the logs from pstorefs on next boot;
> + instead there is separate backend mechanism like minidump present which has
> + awareness about the dynamic ramoops region and can recover the logs. This is
> + enabled via command line parameter dyn_ramoops_size=<size> and should not be
> + used in absence of separate backend which knows how to recover this dynamic
> + region.
> +
> + Note whenever this config is selected ramoops driver will be built statically
> + into kernel.
> +
> config PSTORE_ZONE
> tristate
> depends on PSTORE
> diff --git a/fs/pstore/ram.c b/fs/pstore/ram.c
> index 88b34fdbf759..1faf0835700b 100644
> --- a/fs/pstore/ram.c
> +++ b/fs/pstore/ram.c
> @@ -20,6 +20,7 @@
> #include <linux/compiler.h>
> #include <linux/of.h>
> #include <linux/of_address.h>
> +#include <linux/memblock.h>
> #include <linux/mm.h>
>
> #include "internal.h"
> @@ -103,6 +104,59 @@ struct ramoops_context {
> };
>
> static struct platform_device *dummy;
> +static struct resource dyn_ramoops_res = {
> + .name = "ramoops",
> + .start = 0,
> + .end = 0,
> + .flags = IORESOURCE_BUSY | IORESOURCE_SYSTEM_RAM,
> + .desc = IORES_DESC_NONE,
> +};
> +static int dyn_ramoops_size;
> +
> +#ifdef CONFIG_PSTORE_DYNAMIC_RAMOOPS
> +static int __init parse_dyn_ramoops_size(char *p)
> +{
> + char *tmp;
> +
> + dyn_ramoops_size = memparse(p, &tmp);
> + if (p == tmp) {
> + pr_err("ramoops: memory size expected\n");
> + dyn_ramoops_size = 0;
> + return -EINVAL;
> + }
> +
> + return 0;
> +}
> +early_param("dyn_ramoops_size", parse_dyn_ramoops_size);
> +
> +/*
> + * setup_dynamic_ramoops() - Reserve memory for dynamic ramoops
> + *
> + * Enables dynamic reserve memory support for ramoops through
> + * command line.
> + */
> +void __init setup_dynamic_ramoops(void)
> +{
> + unsigned long long ramoops_base;
> + unsigned long long ramoops_size;
> +
> + if (!dyn_ramoops_size)
> + return;
> +
> + ramoops_base = memblock_phys_alloc_range(dyn_ramoops_size, SMP_CACHE_BYTES,
> + 0, MEMBLOCK_ALLOC_NOLEAKTRACE);
> + if (!ramoops_base) {
> + pr_err("cannot allocate ramoops dynamic memory (size:0x%llx).\n",
> + ramoops_size);
> + dyn_ramoops_size = 0;
> + return;
> + }
> +
> + dyn_ramoops_res.start = ramoops_base;
> + dyn_ramoops_res.end = ramoops_base + dyn_ramoops_size - 1;
> + insert_resource(&iomem_resource, &dyn_ramoops_res);
> +}
> +#endif
>
> static int ramoops_pstore_open(struct pstore_info *psi)
> {
> @@ -915,13 +969,19 @@ static void __init ramoops_register_dummy(void)
>
> /*
> * Prepare a dummy platform data structure to carry the module
> - * parameters. If mem_size isn't set, then there are no module
> - * parameters, and we can skip this.
> + * parameters.
> + *
> + * dyn_ramoops_size takes precedence over mem_size if it is
> + * enabled and valid.
> */
> - if (!mem_size)
> + if (!dyn_ramoops_size && !mem_size)
> return;
>
> pr_info("using module parameters\n");
> + if (dyn_ramoops_size) {
> + mem_size = dyn_ramoops_size;
> + mem_address = dyn_ramoops_res.start;
> + }
>
> memset(&pdata, 0, sizeof(pdata));
> pdata.mem_size = mem_size;
> diff --git a/include/linux/pstore_ram.h b/include/linux/pstore_ram.h
> index 9d65ff94e216..1efff7a38333 100644
> --- a/include/linux/pstore_ram.h
> +++ b/include/linux/pstore_ram.h
> @@ -39,4 +39,9 @@ struct ramoops_platform_data {
> struct persistent_ram_ecc_info ecc_info;
> };
>
> +#ifdef CONFIG_PSTORE_DYNAMIC_RAMOOPS
> +void __init setup_dynamic_ramoops(void);
> +#else
> +static inline void __init setup_dynamic_ramoops(void) {}
> +#endif
> #endif
> diff --git a/init/main.c b/init/main.c
> index ef3ce41b8fc5..3d74241bcb2b 100644
> --- a/init/main.c
> +++ b/init/main.c
> @@ -99,6 +99,8 @@
> #include <linux/init_syscalls.h>
> #include <linux/stackdepot.h>
> #include <linux/randomize_kstack.h>
> +#include <linux/moduleloader.h>
> +#include <linux/pstore_ram.h>
> #include <net/net_namespace.h>
>
> #include <asm/io.h>
> @@ -890,6 +892,7 @@ void start_kernel(void)
> pr_notice("%s", linux_banner);
> early_security_init();
> setup_arch(&command_line);
> + setup_dynamic_ramoops();
> setup_boot_config();
> setup_command_line(command_line);
> setup_nr_cpu_ids();
I would really appreciate if i get review on this series..
Thank you..
-Mukesh
On 1/31/2024 4:38 PM, Mukesh Ojha wrote:
> Abstract and PDF here:
> https://lpc.events/event/17/contributions/1468/
>
> Video:
> https://www.youtube.com/watch?v=3vL3gtAu84s
>
> Patch 1 deals in detail documentation on minidump.
> Patch 2-4 refactors minidump existing layout and separate it from remoteproc files.
> Patch 6 is the Qualcomm APSS minidump driver.
> Patch 7-10 Enable support to reserve dynamic ramoops and the support to
> register ramoops region with minidump.
>
> Detail about Minidump is discussed in documentation patch (1/10) and also briefly
> discussed after below changelog.
>
> Changes in v8:
> - Addressed documentation comment made by Randy Dunlap.
> - Rebased on linux-next tag next-20240130
>
> Changes in v7:
> - Addressed comment made by [Pavan.K] to use generic notifiers.
> - Addresses comment made on Dynamic ramoops about error handling.
> - Significant change minidump documentation suggested by [Bryan O'Donoghue]
> - Added Reviewed by from [Bagas]
> - Renamed ramoops notifiers.
>
> Changes in v6: https://lore.kernel.org/lkml/[email protected]/
> - Accumalated the feedback received on v5 and rebase v5 versions in v6.
> - Removed the exported function as there is no current users of them.
> - Applied [Pavan.K] suggestion on caller/callee placement of dynamic ramoops reserve memory.
> - Addressed [krzysztof] comment on sizeof() and to have qcom_apss_md_table_exit().
> - Addressed [Bagas.S] comment on minidump doc.
> - Tried to implement [Kees] suggestion in slight different way with callback registration
> with ramoops instead of pstore core.
>
> Change in rebase v5: https://lore.kernel.org/lkml/[email protected]/
> - Rebased it on latest tag available on linux-next
> - Added missed Poovendhan sign-off on 15/17 and tested-by tag from
> Kathiravan. Thanks to him for testing and reminding me of missing sign-off.
>
> Changes in v5: https://lore.kernel.org/lkml/[email protected]/
> - On suggestion from Pavan.k, to have single function call for minidump collection
> from remoteproc driver, separated the logic to have separate minidump file called
> qcom_rproc_minidump.c and also renamed the function from qcom_minidump() to
> qcom_rproc_minidump(); however, dropped his suggestion about rework on lazy deletion
> during region unregister in this series, will pursue it in next series.
>
> - To simplify the minidump driver, removed the complication for frontend and different
> backend from Greg suggestion, will pursue this once main driver gets mainlined.
>
> - Move the dynamic ramoops region allocation from Device tree approach to command line
> approch with the introduction command line parsing and memblock reservation during
> early boot up; Not added documentation about it yet, will add if it gets positive
> response.
>
> - Exporting linux banner from kernel to make minidump build also as module, however,
> minidump is a debug module and should be kernel built to get most debug information
> from kernel.
>
> - Tried to address comments given on dload patch series.
>
> Changes in v4: https://lore.kernel.org/lkml/[email protected]/
> - Redesigned the driver and divided the driver into front end and backend (smem) so
> that any new backend can be attached easily to avoid code duplication.
> - Patch reordering as per the driver and subsystem to easier review of the code.
> - Removed minidump specific code from remoteproc to minidump smem based driver.
> - Enabled the all the driver as modules.
> - Address comments made on documentation and yaml and Device tree file [Krzysztof/Konrad]
> - Address comments made qcom_pstore_minidump driver and given its Device tree
> same set of properties as ramoops. [Luca/Kees]
> - Added patch for MAINTAINER file.
> - Include defconfig change as one patch as per [Krzysztof] suggestion.
> - Tried to remove the redundant file scope variables from the module as per [Krzysztof] suggestion.
> - Addressed comments made on dload mode patch v6 version
> https://lore.kernel.org/lkml/[email protected]/
>
> Changes in v3: https://lore.kernel.org/lkml/[email protected]/
> - Addressed most of the comments by Srini on v2 and refactored the minidump driver.
> - Added platform device support
> - Unregister region support.
> - Added update region for clients.
> - Added pending region support.
> - Modified the documentation guide accordingly.
> - Added qcom_pstore_ramdump client driver which happen to add ramoops platform
> device and also registers ramoops region with minidump.
> - Added download mode patch series with this minidump series.
> https://lore.kernel.org/lkml/[email protected]/
>
> Changes in v2: https://lore.kernel.org/lkml/[email protected]/
> - Addressed review comment made by [quic_tsoni/bmasney] to add documentation.
> - Addressed comments made by [srinivas.kandagatla]
> - Dropped pstore 6/6 from the last series, till i get conclusion to get pstore
> region in minidump.
> - Fixed issue reported by kernel test robot.
>
> Changes in v1: https://lore.kernel.org/lkml/[email protected]/
>
> Minidump is a best effort mechanism to collect useful and predefined data
> for first level of debugging on end user devices running on Qualcomm SoCs.
> It is built on the premise that System on Chip (SoC) or subsystem part of
> SoC crashes, due to a range of hardware and software bugs.
>
> Qualcomm devices in engineering mode provides a mechanism for generating
> full system ramdumps for post mortem debugging. But in some cases it's
> however not feasible to capture the entire content of RAM. The minidump
> mechanism provides the means for selecting which snippets should be
> included in the ramdump.
>
> The core of SMEM based minidump feature is part of Qualcomm's boot
> firmware code. It initializes shared memory (SMEM), which is a part of
> DDR and allocates a small section of SMEM to minidump table i.e also
> called global table of content (G-ToC). Each subsystem (APSS, ADSP, ...)
> has their own table of segments to be included in the minidump and all
> get their reference from G-ToC. Each segment/region has some details
> like name, physical address and it's size etc. and it could be anywhere
> scattered in the DDR.
>
> Existing upstream Qualcomm remoteproc driver[1] already supports SMEM
> based minidump feature for remoteproc instances like ADSP, MODEM, ...
> where predefined selective segments of subsystem region can be dumped
> as part of coredump collection which generates smaller size artifacts
> compared to complete coredump of subsystem on crash.
>
> [1]
> https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/tree/drivers/remoteproc/qcom_common.c#n142
>
> In addition to managing and querying the APSS minidump description,
> the Linux driver maintains a ELF header in a segment. This segment
> gets updated with section/program header whenever a new entry gets
> registered.
>
> Support for Minidump enablement on Qualcomm SoCs is pursued separately and
> can be done via below series of changes. For testing, these patches can be
> applied
>
> https://lore.kernel.org/lkml/[email protected]/
>
> https://lore.kernel.org/lkml/[email protected]/
> https://lore.kernel.org/lkml/[email protected]/
>
> Testing of these patches has been done on sm8450 target after enabling kernel
> config like CONFIG_PSTORE_RAM/CONFIG_PSTORE_CONSOLE and once the device boots
> up. Below command can be executed from sysfs to enable minidump in the firmware.
>
> echo mini > /sys/module/qcom_scm/parameters/download_mode
>
> Try crashing it via devmem2 0xf11c000(this is known command to create xpu violation
> and put the device crash dump mode) on command prompt.
>
> Default storage type is set to via USB, so Minidump would be downloaded with the
> help of x86_64 machine (running PCAT tool) attached to Qualcomm product which has
> backed Minidump boot firmware support.
>
> After that we will see a bunch of predefined registered region as binary blobs files
> starts with md_* downloaded on the x86 machine at configured/default location in PCAT
> tool from the product, more about this can be found in qualcomm minidump guide
> patch.
>
> Mukesh Ojha (10):
> docs: qcom: Add qualcomm minidump guide
> soc: qcom: Add qcom_rproc_minidump module
> remoteproc: qcom_q6v5_pas: Use qcom_rproc_minidump()
> remoteproc: qcom: Remove minidump related data from qcom_common.c
> init: export linux_banner data variable
> soc: qcom: Add Qualcomm APSS minidump kernel driver
> MAINTAINERS: Add entry for minidump related files
> pstore/ram: Add dynamic ramoops region support through commandline
> pstore/ram: Add ramoops information notifier support
> soc: qcom: register ramoops region with APSS minidump
>
> Documentation/admin-guide/index.rst | 1 +
> Documentation/admin-guide/qcom_minidump.rst | 318 +++++++++
> Documentation/admin-guide/ramoops.rst | 23 +-
> MAINTAINERS | 10 +
> drivers/remoteproc/Kconfig | 1 +
> drivers/remoteproc/qcom_common.c | 160 -----
> drivers/remoteproc/qcom_q6v5_pas.c | 3 +-
> drivers/soc/qcom/Kconfig | 23 +
> drivers/soc/qcom/Makefile | 2 +
> drivers/soc/qcom/qcom_minidump.c | 690 ++++++++++++++++++++
> drivers/soc/qcom/qcom_minidump_internal.h | 74 +++
> drivers/soc/qcom/qcom_rproc_minidump.c | 111 ++++
> drivers/soc/qcom/smem.c | 20 +
> fs/pstore/Kconfig | 15 +
> fs/pstore/ram.c | 180 ++++-
> include/linux/init.h | 3 +
> include/linux/pstore_ram.h | 20 +
> include/linux/soc/qcom/smem.h | 2 +
> include/soc/qcom/qcom_minidump.h | 41 ++
> init/main.c | 3 +
> init/version-timestamp.c | 3 +
> 21 files changed, 1538 insertions(+), 165 deletions(-)
> create mode 100644 Documentation/admin-guide/qcom_minidump.rst
> create mode 100644 drivers/soc/qcom/qcom_minidump.c
> create mode 100644 drivers/soc/qcom/qcom_minidump_internal.h
> create mode 100644 drivers/soc/qcom/qcom_rproc_minidump.c
> create mode 100644 include/soc/qcom/qcom_minidump.h
>
>
> base-commit: 41d66f96d0f15a0a2ad6fa2208f6bac1a66cbd52