From: Matthew Brost <matthew.brost@intel.com>
To: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: intel-xe@lists.freedesktop.org, dri-devel@lists.freedesktop.org
Subject: Re: [Intel-xe] [PATCH 03/14] drm/xe: Do not take any action if our device was removed.
Date: Tue, 2 May 2023 15:40:50 +0000 [thread overview]
Message-ID: <ZFEvAuZRXl5U6ntH@DUT025-TGLU.fm.intel.com> (raw)
In-Reply-To: <20230426205713.512695-4-rodrigo.vivi@intel.com>
On Wed, Apr 26, 2023 at 04:57:02PM -0400, Rodrigo Vivi wrote:
> Unfortunately devcoredump infrastructure does not provide and
> interface for us to force the device removal upon the pci_remove
> time of our device.
>
> The devcoredump is linked at the device level, so when in use
> it will prevent the module removal, but it doesn't prevent the
> call of the pci_remove callback. This callback cannot fail
> anyway and we end up clearing and freeing the entire pci device.
>
> Hence, after we removed the pci device, we shouldn't allow any
> read or free operations to avoid segmentation fault.
>
> Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
> ---
> drivers/gpu/drm/xe/xe_devcoredump.c | 19 ++++++++++++++++---
> 1 file changed, 16 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpu/drm/xe/xe_devcoredump.c b/drivers/gpu/drm/xe/xe_devcoredump.c
> index d9531183f03a..a08929c01b75 100644
> --- a/drivers/gpu/drm/xe/xe_devcoredump.c
> +++ b/drivers/gpu/drm/xe/xe_devcoredump.c
> @@ -42,6 +42,11 @@
> * hang capture.
> */
>
> +static struct xe_device *coredump_to_xe(const struct xe_devcoredump *coredump)
> +{
> + return container_of(coredump, struct xe_device, devcoredump);
Confused how still would ever return NULL, can you explain?
Matt
> +}
> +
> static ssize_t xe_devcoredump_read(char *buffer, loff_t offset,
> size_t count, void *data, size_t datalen)
> {
> @@ -51,6 +56,10 @@ static ssize_t xe_devcoredump_read(char *buffer, loff_t offset,
> struct drm_print_iterator iter;
> struct timespec64 ts;
>
> + /* Our device is gone already... */
> + if (!data || !coredump_to_xe(coredump))
> + return -ENODEV;
> +
> iter.data = buffer;
> iter.offset = 0;
> iter.start = offset;
> @@ -80,12 +89,16 @@ static ssize_t xe_devcoredump_read(char *buffer, loff_t offset,
> static void xe_devcoredump_free(void *data)
> {
> struct xe_devcoredump *coredump = data;
> - struct xe_device *xe = container_of(coredump, struct xe_device,
> - devcoredump);
> +
> + /* Our device is gone. Nothing to do... */
> + if (!data || !coredump_to_xe(coredump))
> + return;
> +
> mutex_lock(&coredump->lock);
>
> coredump->faulty_engine = NULL;
> - drm_info(&xe->drm, "Xe device coredump has been deleted.\n");
> + drm_info(&coredump_to_xe(coredump)->drm,
> + "Xe device coredump has been deleted.\n");
>
> mutex_unlock(&coredump->lock);
> }
> --
> 2.39.2
>
next prev parent reply other threads:[~2023-05-02 15:41 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-04-26 20:56 [Intel-xe] [PATCH 00/14] Introduce xe_devcoredump Rodrigo Vivi
2023-04-26 20:57 ` [Intel-xe] [PATCH 01/14] drm/xe: Fix print of RING_EXECLIST_SQ_CONTENTS_HI Rodrigo Vivi
2023-04-26 21:40 ` Lucas De Marchi
2023-04-26 21:59 ` Rodrigo Vivi
2023-04-26 20:57 ` [Intel-xe] [PATCH 02/14] drm/xe: Introduce the dev_coredump infrastructure Rodrigo Vivi
2023-04-27 8:28 ` Thomas Hellström
2023-05-02 7:57 ` Matthew Brost
2023-05-02 18:06 ` Rodrigo Vivi
2023-05-02 20:29 ` Matthew Brost
2023-05-02 7:55 ` Jani Nikula
2023-05-02 17:25 ` Rodrigo Vivi
2023-04-26 20:57 ` [Intel-xe] [PATCH 03/14] drm/xe: Do not take any action if our device was removed Rodrigo Vivi
2023-05-02 15:40 ` Matthew Brost [this message]
2023-05-02 17:21 ` Rodrigo Vivi
2023-05-02 23:06 ` Matthew Brost
2023-04-26 20:57 ` [Intel-xe] [PATCH 04/14] drm/xe: Extract non mapped regions out of GuC CTB into its own struct Rodrigo Vivi
2023-05-02 5:12 ` Matthew Brost
2023-04-26 20:57 ` [Intel-xe] [PATCH 05/14] drm/xe: Convert GuC CT print to snapshot capture and print Rodrigo Vivi
2023-05-02 5:27 ` Matthew Brost
2023-04-26 20:57 ` [Intel-xe] [PATCH 06/14] drm/xe: Add GuC CT snapshot to xe_devcoredump Rodrigo Vivi
2023-05-02 14:55 ` Matthew Brost
2023-04-26 20:57 ` [Intel-xe] [PATCH 07/14] drm/xe: Introduce guc_submit_types.h with relevant structs Rodrigo Vivi
2023-05-02 7:44 ` Matthew Brost
2023-04-26 20:57 ` [Intel-xe] [PATCH 08/14] drm/xe: Convert GuC Engine print to snapshot capture and print Rodrigo Vivi
2023-05-02 15:01 ` Matthew Brost
2023-04-26 20:57 ` [Intel-xe] [PATCH 09/14] drm/xe: Add GuC Submit Engine snapshot to xe_devcoredump Rodrigo Vivi
2023-05-02 15:03 ` Matthew Brost
2023-04-26 20:57 ` [Intel-xe] [PATCH 10/14] drm/xe: Convert Xe HW Engine print to snapshot capture and print Rodrigo Vivi
2023-05-02 15:20 ` Matthew Brost
2023-04-26 20:57 ` [Intel-xe] [PATCH 11/14] drm/xe: Add HW Engine snapshot to xe_devcoredump Rodrigo Vivi
2023-05-02 15:30 ` Matthew Brost
2023-04-26 20:57 ` [Intel-xe] [PATCH 12/14] drm/xe: Limit CONFIG_DRM_XE_SIMPLE_ERROR_CAPTURE to itself Rodrigo Vivi
2023-05-02 15:35 ` Matthew Brost
2023-04-26 20:57 ` [Intel-xe] [PATCH 13/14] drm/xe: Convert VM print to snapshot capture and print Rodrigo Vivi
2023-05-02 7:50 ` Matthew Brost
2023-05-02 8:07 ` Matthew Brost
2023-04-26 20:57 ` [Intel-xe] [PATCH 14/14] drm/xe: Add VM snapshot to xe_devcoredump Rodrigo Vivi
2023-05-02 15:38 ` Matthew Brost
2023-04-26 21:01 ` [Intel-xe] ✓ CI.Patch_applied: success for Introduce xe_devcoredump Patchwork
2023-04-26 21:02 ` [Intel-xe] ✓ CI.KUnit: " Patchwork
2023-04-26 21:06 ` [Intel-xe] ✓ CI.Build: " Patchwork
2023-04-26 21:29 ` [Intel-xe] ○ CI.BAT: info " Patchwork
2023-05-02 8:11 ` [Intel-xe] [PATCH 00/14] " Matthew Brost
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=ZFEvAuZRXl5U6ntH@DUT025-TGLU.fm.intel.com \
--to=matthew.brost@intel.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=intel-xe@lists.freedesktop.org \
--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