We use ACPI_HEST_NOTIFY_SOFTWARE_DELEGATED Notify type for ras event.
In ATF:
ehf_activate_priority()
dispatch sdei()
ehf_deactivate_priority()
If ras error severity is fatal, panic was called in sdei(),
ehf_deactivate_priority was not called. we should restore interrupt before panic
otherwise kdump will trigger error.
Signed-off-by: Liguang Zhang <[email protected]>
---
drivers/acpi/apei/ghes.c | 25 +++++++++++++++++++++----
drivers/firmware/arm_sdei.c | 14 ++++++++++++++
include/linux/arm_sdei.h | 2 ++
3 files changed, 37 insertions(+), 4 deletions(-)
diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c
index 0c8330ed1ffd..4f734c60987c 100644
--- a/drivers/acpi/apei/ghes.c
+++ b/drivers/acpi/apei/ghes.c
@@ -141,6 +141,7 @@ static unsigned long ghes_estatus_pool_size_request;
static struct ghes_estatus_cache *ghes_estatus_caches[GHES_ESTATUS_CACHES_SIZE];
static atomic_t ghes_estatus_cache_alloced;
+static bool ghes_sdei_callback;
static int ghes_panic_timeout __read_mostly = 30;
static void __iomem *ghes_map(u64 pfn, enum fixed_addresses fixmap_idx)
@@ -837,18 +838,30 @@ static void ghes_estatus_cache_add(
rcu_read_unlock();
}
+static void sdei_api_restore_ras(void)
+{
+ /* reboot to log the error! */
+ if (!panic_timeout)
+ panic_timeout = ghes_panic_timeout;
+ panic("Fatal hardware error!");
+}
+
static void __ghes_panic(struct ghes *ghes,
struct acpi_hest_generic_status *estatus,
u64 buf_paddr, enum fixed_addresses fixmap_idx)
{
+ int err;
+
__ghes_print_estatus(KERN_EMERG, ghes->generic, estatus);
ghes_clear_estatus(ghes, estatus, buf_paddr, fixmap_idx);
- /* reboot to log the error! */
- if (!panic_timeout)
- panic_timeout = ghes_panic_timeout;
- panic("Fatal hardware error!");
+ if (ghes_sdei_callback) {
+ err = sdei_api_event_complete_and_resume((unsigned long)sdei_api_restore_ras);
+ if (err)
+ sdei_api_restore_ras();
+ } else
+ sdei_api_restore_ras();
}
static int ghes_proc(struct ghes *ghes)
@@ -1224,7 +1237,9 @@ static int ghes_sdei_normal_callback(u32 event_num, struct pt_regs *regs,
int err;
raw_spin_lock(&ghes_notify_lock_sdei_normal);
+ ghes_sdei_callback = true;
err = __ghes_sdei_callback(ghes, FIX_APEI_GHES_SDEI_NORMAL);
+ ghes_sdei_callback = false;
raw_spin_unlock(&ghes_notify_lock_sdei_normal);
return err;
@@ -1238,7 +1253,9 @@ static int ghes_sdei_critical_callback(u32 event_num, struct pt_regs *regs,
int err;
raw_spin_lock(&ghes_notify_lock_sdei_critical);
+ ghes_sdei_callback = true;
err = __ghes_sdei_callback(ghes, FIX_APEI_GHES_SDEI_CRITICAL);
+ ghes_sdei_callback = false;
raw_spin_unlock(&ghes_notify_lock_sdei_critical);
return err;
diff --git a/drivers/firmware/arm_sdei.c b/drivers/firmware/arm_sdei.c
index a7e762c352f9..1af6b6b55c57 100644
--- a/drivers/firmware/arm_sdei.c
+++ b/drivers/firmware/arm_sdei.c
@@ -473,6 +473,20 @@ static int sdei_api_event_unregister(u32 event_num)
0, 0, 0, NULL);
}
+int sdei_api_event_complete_and_resume(u64 addr)
+{
+ int err;
+
+ err = invoke_sdei_fn(SDEI_1_0_FN_SDEI_EVENT_COMPLETE_AND_RESUME, addr,
+ 0, 0, 0, 0, NULL);
+ if (err && err != -EIO) {
+ pr_warn_once("failed to restore CPU[%u]: %d\n", smp_processor_id(), err);
+ return err;
+ }
+
+ return 0;
+}
+
/* Called directly by the hotplug callbacks */
static void _local_event_unregister(void *data)
{
diff --git a/include/linux/arm_sdei.h b/include/linux/arm_sdei.h
index 0a241c5c911d..b6d347085834 100644
--- a/include/linux/arm_sdei.h
+++ b/include/linux/arm_sdei.h
@@ -46,9 +46,11 @@ int sdei_unregister_ghes(struct ghes *ghes);
/* For use by arch code when CPU hotplug notifiers are not appropriate. */
int sdei_mask_local_cpu(void);
int sdei_unmask_local_cpu(void);
+int sdei_api_event_complete_and_resume(u64 addr);
#else
static inline int sdei_mask_local_cpu(void) { return 0; }
static inline int sdei_unmask_local_cpu(void) { return 0; }
+int sdei_api_event_complete_and_resume(u64 addr) { return 0; }
#endif /* CONFIG_ARM_SDE_INTERFACE */
--
2.19.1.6.gb485710b
On Mon, Oct 11, 2021 at 11:10:28PM +0800, Liguang Zhang wrote:
> We use ACPI_HEST_NOTIFY_SOFTWARE_DELEGATED Notify type for ras event.
>
> In ATF:
> ehf_activate_priority()
> dispatch sdei()
> ehf_deactivate_priority()
>
> If ras error severity is fatal, panic was called in sdei(),
> ehf_deactivate_priority was not called. we should restore interrupt before panic
> otherwise kdump will trigger error.
I have *absolutely* no clue what this commit message is trying to tell
me - sorry you'd have to try again. Maybe structuring it this way, would
help:
--
Problem is A.
It happens because of B.
Fix it by doing C.
(Potentially do D).
For more detailed info, see
Documentation/process/submitting-patches.rst, Section "2) Describe your
changes".
Also, to the tone, from Documentation/process/submitting-patches.rst:
"Describe your changes in imperative mood, e.g. "make xyzzy do frotz"
instead of "[This patch] makes xyzzy do frotz" or "[I] changed xyzzy
to do frotz", as if you are giving orders to the codebase to change
its behaviour."
Also, do not talk about what your patch does - that should hopefully be
visible in the diff itself. Rather, talk about *why* you're doing what
you're doing.
--
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquette
Hi Liguang,
Thank you for the patch! Perhaps something to improve:
[auto build test WARNING on rafael-pm/linux-next]
[also build test WARNING on linux/master linus/master v5.15-rc5 next-20211011]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch]
url: https://github.com/0day-ci/linux/commits/Liguang-Zhang/ACPI-APEI-restore-interrupt-before-panic-in-sdei-flow/20211011-231126
base: https://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm.git linux-next
config: x86_64-randconfig-a012-20211011 (attached as .config)
compiler: gcc-9 (Debian 9.3.0-22) 9.3.0
reproduce (this is a W=1 build):
# https://github.com/0day-ci/linux/commit/abb332bc84212b2ef288bb3e7ad24a04cd4853e6
git remote add linux-review https://github.com/0day-ci/linux
git fetch --no-tags linux-review Liguang-Zhang/ACPI-APEI-restore-interrupt-before-panic-in-sdei-flow/20211011-231126
git checkout abb332bc84212b2ef288bb3e7ad24a04cd4853e6
# save the attached .config to linux build tree
make W=1 ARCH=x86_64
If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <[email protected]>
All warnings (new ones prefixed by >>):
In file included from drivers/acpi/apei/ghes.c:20:
>> include/linux/arm_sdei.h:53:5: warning: no previous prototype for 'sdei_api_event_complete_and_resume' [-Wmissing-prototypes]
53 | int sdei_api_event_complete_and_resume(u64 addr) { return 0; }
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
vim +/sdei_api_event_complete_and_resume +53 include/linux/arm_sdei.h
44
45 #ifdef CONFIG_ARM_SDE_INTERFACE
46 /* For use by arch code when CPU hotplug notifiers are not appropriate. */
47 int sdei_mask_local_cpu(void);
48 int sdei_unmask_local_cpu(void);
49 int sdei_api_event_complete_and_resume(u64 addr);
50 #else
51 static inline int sdei_mask_local_cpu(void) { return 0; }
52 static inline int sdei_unmask_local_cpu(void) { return 0; }
> 53 int sdei_api_event_complete_and_resume(u64 addr) { return 0; }
54 #endif /* CONFIG_ARM_SDE_INTERFACE */
55
56
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/[email protected]