Intel-XE Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Michal Wajdeczko <michal.wajdeczko@intel.com>
To: "Mallesh, Koujalagi" <mallesh.koujalagi@intel.com>,
	<intel-xe@lists.freedesktop.org>,
	Jani Nikula <jani.nikula@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Subject: Re: [PATCH v4 11/32] drm/xe/log: Index all SIGID printk messages
Date: Thu, 13 Aug 2026 14:54:30 +0200	[thread overview]
Message-ID: <e91dfe34-1e98-4f53-967b-a853a88427fb@intel.com> (raw)
In-Reply-To: <5f61f3de-357c-489a-9bb2-f3874a98573e@intel.com>



On 8/13/2026 2:31 PM, Mallesh, Koujalagi wrote:
> 
> On 13-08-2026 12:44 am, Michal Wajdeczko wrote:
>> When CONFIG_PRINTK_INDEX is enabled, it is expected that all device
>> level printk messages are indexed for audit. While usually this is
>> done automatically behind the scenes when code is using regular
>> dev_printk macros, since we are generating different dmesg messages
>> inside xe_log_emit() based on the severity, component and location,
>> we only get those entries in /sys/kernel/debug/printk/index/xe:
>>
>>    <3> drivers/gpu/drm/xe/xe_log.c:142 log_dmesg_vprintk "%s %s: [drm] *ERROR* %pV"
>>    <6> drivers/gpu/drm/xe/xe_log.c:140 log_dmesg_vprintk "%s %s: [drm] %pV"
>>
>> Explicitly generate printk index using dev_printk_index_emit() with
>> some generic prefix that includes the SIGID tag.
>>
>> Suggested-by: Jani Nikula <jani.nikula@intel.com>
>> Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
>> Cc: Jani Nikula <jani.nikula@intel.com>
>> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
>> ---
>>   drivers/gpu/drm/xe/xe_log.c |  8 ++++----
>>   drivers/gpu/drm/xe/xe_log.h | 14 +++++++++++---
>>   2 files changed, 15 insertions(+), 7 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/xe/xe_log.c b/drivers/gpu/drm/xe/xe_log.c
>> index 0b423ed121cd..50a9b35e5b3e 100644
>> --- a/drivers/gpu/drm/xe/xe_log.c
>> +++ b/drivers/gpu/drm/xe/xe_log.c
>> @@ -177,7 +177,7 @@ static void log_emit_dmesg(struct pci_dev *pdev, int cper_sev, enum xe_sigid sig
>>   }
>>     /**
>> - * xe_log_emit() - Emit a structured SIGID log entry
>> + * __xe_log_emit() - Emit a structured SIGID log entry
>>    * @pdev: the &pci_dev device
>>    * @cper_sev: CPER severity (CPER_SEV_FATAL, CPER_SEV_RECOVERABLE, ...)
>>    * @sigid: signature identifier, see &enum xe_sigid
>> @@ -206,9 +206,9 @@ static void log_emit_dmesg(struct pci_dev *pdev, int cper_sev, enum xe_sigid sig
>>    *   <3> xe 0000:03:00.0: [drm] *ERROR* SIGID=106 (-ETIMEDOUT) Engine 'rcs0' hung
>>    *   <6> xe 0000:03:00.0: [drm] SIGID=103 In survivability mode
>>    */
>> -void xe_log_emit(struct pci_dev *pdev, int cper_sev, enum xe_sigid sigid,
>> -         u32 component, u32 location, const void *data, size_t len,
>> -         const char *fmt, ...)
>> +void __xe_log_emit(struct pci_dev *pdev, int cper_sev, enum xe_sigid sigid,
>> +           u32 component, u32 location, const void *data, size_t len,
>> +           const char *fmt, ...)
>>   {
>>       struct va_format vaf;
>>       va_list args;
>> diff --git a/drivers/gpu/drm/xe/xe_log.h b/drivers/gpu/drm/xe/xe_log.h
>> index 0928b0866617..53fe2bb7ddd3 100644
>> --- a/drivers/gpu/drm/xe/xe_log.h
>> +++ b/drivers/gpu/drm/xe/xe_log.h
>> @@ -16,9 +16,17 @@
>>   struct pci_dev;
>>     __printf(8, 9)
>> -void xe_log_emit(struct pci_dev *pdev, int cper_sev, enum xe_sigid sigid,
>> -         u32 component, u32 location, const void *data, size_t len,
>> -         const char *fmt, ...);
>> +void __xe_log_emit(struct pci_dev *pdev, int cper_sev, enum xe_sigid sigid,
>> +           u32 component, u32 location, const void *data, size_t len,
>> +           const char *fmt, ...);
>> +
>> +#define __xe_log_emit_printk_index(fmt) \
>> +    dev_printk_index_emit(NULL, "[drm]%s SIGID=%u %s(%s)%s%s%s: " fmt);
> We need to change format for a. error path b. blob path c. plain or info path right?

well, that's not doable as final format/output depends on the
severity and data value, which could be non-const at compile
time, while printk-index requires this to emit right entry

I can change that to a something more generic (and unfriendly), like:

    dev_printk_index_emit(NULL, "%sSIGID=%u %s" fmt);

but that will catch and match all our outputs.

@Jani, are you OK with that?

>> +
>> +#define xe_log_emit(pdev, sev, sig, comp, loc, data, len, fmt, args...) ({        \
>> +    __xe_log_emit_printk_index(fmt);                        \
>> +    __xe_log_emit((pdev), (sev), (sig), (comp), (loc), (data), (len), fmt, ##args); \
>> +})
> 
> nit: use do{} while (0)
> 
> Reviewed-by: Mallesh Koujalagi <mallesh.koujalagi@intel.com>
> 
>>     #define xe_log_emit_fatal(pdev, sig, comp, loc, data, len, fmt, args...) \
>>       xe_log_emit((pdev), CPER_SEV_FATAL, (sig), (comp), (loc), \


  reply	other threads:[~2026-08-13 12:54 UTC|newest]

Thread overview: 109+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-08-12 19:14 [PATCH v4 00/32] drm/xe: Add structured SIGID error logging infrastructure Michal Wajdeczko
2026-08-12 19:14 ` [PATCH v4 01/32] drm/xe: Introduce xe_any helpers Michal Wajdeczko
2026-08-12 19:29   ` sashiko-bot
2026-08-13 16:51   ` Rodrigo Vivi
2026-08-13 23:37   ` Mallesh, Koujalagi
2026-08-12 19:14 ` [PATCH v4 02/32] drm/xe/log: Add structured SIGID error logging infrastructure Michal Wajdeczko
2026-08-13 13:33   ` Mallesh, Koujalagi
2026-08-13 13:57     ` Michal Wajdeczko
2026-08-13 13:42   ` Nilawar, Badal
2026-08-13 19:00     ` Rodrigo Vivi
2026-08-14 13:05       ` Nilawar, Badal
2026-08-14 13:54         ` Michal Wajdeczko
2026-08-13 20:38     ` Michal Wajdeczko
2026-08-14 13:19       ` Nilawar, Badal
2026-08-13 18:51   ` Rodrigo Vivi
2026-08-14 13:32   ` Nilawar, Badal
2026-08-14 14:00     ` Michal Wajdeczko
2026-08-12 19:14 ` [PATCH v4 03/32] drm/xe/log: Introduce structured component/location identifiers Michal Wajdeczko
2026-08-13 17:20   ` Rodrigo Vivi
2026-08-12 19:14 ` [PATCH v4 04/32] drm/xe/log: Add component/location decorations to dmesg Michal Wajdeczko
2026-08-14  0:00   ` Mallesh, Koujalagi
2026-08-12 19:14 ` [PATCH v4 05/32] drm/xe/log: Add SIGID log helpers for severity Michal Wajdeczko
2026-08-12 19:14 ` [PATCH v4 06/32] drm/xe/log: Add SIGID log helpers for location Michal Wajdeczko
2026-08-12 19:14 ` [PATCH v4 07/32] drm/xe/log: Add SIGID log helpers for location & severity Michal Wajdeczko
2026-08-12 19:14 ` [PATCH v4 08/32] drm/xe/log: Add SIGID log helpers for components Michal Wajdeczko
2026-08-12 19:14 ` [PATCH v4 09/32] drm/xe/log: Add SIGID log helpers for component & severity Michal Wajdeczko
2026-08-13  4:38   ` Mallesh, Koujalagi
2026-08-12 19:14 ` [PATCH v4 10/32] drm/xe/log: Add SIGID log helpers for errno-only Michal Wajdeczko
2026-08-12 19:14 ` [PATCH v4 11/32] drm/xe/log: Index all SIGID printk messages Michal Wajdeczko
2026-08-12 19:35   ` sashiko-bot
2026-08-13 12:31   ` Mallesh, Koujalagi
2026-08-13 12:54     ` Michal Wajdeczko [this message]
2026-08-13 13:29       ` Michal Wajdeczko
2026-08-12 19:14 ` [PATCH v4 12/32] drm/xe/log: Add hardware error signatures Michal Wajdeczko
2026-08-13  5:03   ` Mallesh, Koujalagi
2026-08-12 19:14 ` [PATCH v4 13/32] drm/xe/log: Extend components list with hardware items Michal Wajdeczko
2026-08-12 19:14 ` [PATCH v4 14/32] drm/xe/ras: Check RAS and LOG component definitions Michal Wajdeczko
2026-08-12 19:14 ` [PATCH v4 15/32] drm/xe/kunit: Setup driver data in the test device Michal Wajdeczko
2026-08-12 19:14 ` [PATCH v4 16/32] drm/xe/tests: Add Kunit tests for xe_log Michal Wajdeczko
2026-08-14 13:36   ` Nilawar, Badal
2026-08-14 14:11     ` Michal Wajdeczko
2026-08-12 19:14 ` [PATCH v4 17/32] drm/xe/tests: Add kunit tests for xe_any Michal Wajdeczko
2026-08-12 19:14 ` [PATCH v4 18/32] drm/xe: Report 'probe blocked' error using SIGID Michal Wajdeczko
2026-08-13  6:26   ` Mallesh, Koujalagi
2026-08-12 19:14 ` [PATCH v4 19/32] drm/xe: Report all probe errors " Michal Wajdeczko
2026-08-13  6:50   ` Mallesh, Koujalagi
2026-08-13  9:12     ` Michal Wajdeczko
2026-08-13  9:58       ` Mallesh, Koujalagi
2026-08-13 10:09         ` Michal Wajdeczko
2026-08-12 19:14 ` [PATCH v4 20/32] drm/xe/survivability: Report 'boot status' " Michal Wajdeczko
2026-08-13  8:38   ` Mallesh, Koujalagi
2026-08-13  9:28     ` Michal Wajdeczko
2026-08-13 10:07       ` Mallesh, Koujalagi
2026-08-13 10:18         ` Michal Wajdeczko
2026-08-14 10:46           ` Tauro, Riana
2026-08-14 13:38             ` Michal Wajdeczko
2026-08-12 19:14 ` [PATCH v4 21/32] drm/xe/survivability: Report 'sysfs failure' error " Michal Wajdeczko
2026-08-13  8:54   ` Mallesh, Koujalagi
2026-08-14 10:34   ` Tauro, Riana
2026-08-14 13:32     ` Michal Wajdeczko
2026-08-12 19:14 ` [PATCH v4 22/32] drm/xe/survivability: Report 'Boot Mode enabled' status " Michal Wajdeczko
2026-08-13 10:52   ` Mallesh, Koujalagi
2026-08-13 11:01     ` Michal Wajdeczko
2026-08-14  6:31       ` Tauro, Riana
2026-08-14 10:06         ` Michal Wajdeczko
2026-08-14 10:20           ` Tauro, Riana
2026-08-14 10:40             ` Michal Wajdeczko
2026-08-14 10:51               ` Tauro, Riana
2026-08-14 13:17                 ` Michal Wajdeczko
2026-08-14 12:40         ` Mallesh, Koujalagi
2026-08-12 19:14 ` [PATCH v4 23/32] drm/xe/survivability: Report 'Runtime " Michal Wajdeczko
2026-08-13 11:40   ` Mallesh, Koujalagi
2026-08-13 12:46     ` Michal Wajdeczko
2026-08-13 13:16       ` Mallesh, Koujalagi
2026-08-14  7:06       ` Tauro, Riana
2026-08-14 10:18         ` Michal Wajdeczko
2026-08-14 10:27           ` Tauro, Riana
2026-08-14 13:26             ` Michal Wajdeczko
2026-08-14 13:00         ` Mallesh, Koujalagi
2026-08-14 13:46           ` Michal Wajdeczko
2026-08-12 19:14 ` [PATCH v4 24/32] drm/xe: Report 'device wedged' errors " Michal Wajdeczko
2026-08-12 22:28   ` Rodrigo Vivi
2026-08-13 11:56   ` Mallesh, Koujalagi
2026-08-12 19:14 ` [PATCH v4 25/32] drm/xe/pcode: Report 'Mailbox failed' error " Michal Wajdeczko
2026-08-13 12:10   ` Bhadane, Dnyaneshwar
2026-08-13 12:35     ` Michal Wajdeczko
2026-08-12 19:14 ` [PATCH v4 26/32] drm/xe/pcode: Report 'timeout, retrying' " Michal Wajdeczko
2026-08-12 19:51   ` sashiko-bot
2026-08-13 16:46   ` Umesh Nerlige Ramappa
2026-08-13 18:42     ` Rodrigo Vivi
2026-08-12 19:14 ` [PATCH v4 27/32] drm/xe/pcode: Report 'initialization timedout' " Michal Wajdeczko
2026-08-12 19:49   ` sashiko-bot
2026-08-13 18:40   ` Rodrigo Vivi
2026-08-12 19:14 ` [PATCH v4 28/32] drm/xe/guc: Report 'GuC mmio' errors " Michal Wajdeczko
2026-08-12 19:14 ` [PATCH v4 29/32] drm/xe/gt: Report 'reset failed' " Michal Wajdeczko
2026-08-12 19:14 ` [PATCH v4 30/32] drm/xe/gt: Report 'Fault response' pagefault error " Michal Wajdeczko
2026-08-13 18:37   ` Rodrigo Vivi
2026-08-13 19:46     ` Michal Wajdeczko
2026-08-13 22:12       ` Daniele Ceraolo Spurio
2026-08-14 10:30         ` Michal Wajdeczko
2026-08-14 12:49           ` Rodrigo Vivi
2026-08-12 19:14 ` [PATCH v4 31/32] drm/xe/gt: Report 'Queue full' " Michal Wajdeczko
2026-08-13 17:25   ` Rodrigo Vivi
2026-08-12 19:14 ` [PATCH v4 32/32] drm/xe/pci: Report 'cannot re-enable' " Michal Wajdeczko
2026-08-13 17:21   ` Rodrigo Vivi
2026-08-12 19:22 ` ✗ CI.checkpatch: warning for drm/xe: Add structured SIGID error logging infrastructure (rev4) Patchwork
2026-08-12 19:24 ` ✓ CI.KUnit: success " Patchwork
2026-08-12 20:34 ` ✗ Xe.CI.BAT: failure " Patchwork
2026-08-13  2:47 ` ✗ Xe.CI.FULL: " Patchwork

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=e91dfe34-1e98-4f53-967b-a853a88427fb@intel.com \
    --to=michal.wajdeczko@intel.com \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=jani.nikula@intel.com \
    --cc=mallesh.koujalagi@intel.com \
    --cc=rodrigo.vivi@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox