From: "Dixit, Ashutosh" <ashutosh.dixit@intel.com>
To: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: <intel-xe@lists.freedesktop.org>
Subject: Re: [PATCH 5/9] drm/xe/vf: Don't initialize OA if VF
Date: Wed, 19 Jun 2024 15:57:42 -0700 [thread overview]
Message-ID: <855xu4r15l.wl-ashutosh.dixit@intel.com> (raw)
In-Reply-To: <73d602b1-6042-47b9-9b35-396ce72a3788@intel.com>
On Wed, 19 Jun 2024 15:40:43 -0700, Michal Wajdeczko wrote:
>
> On 20.06.2024 00:10, Dixit, Ashutosh wrote:
> > On Wed, 19 Jun 2024 14:56:32 -0700, Dixit, Ashutosh wrote:
> >>
> >> On Wed, 19 Jun 2024 14:45:53 -0700, Michal Wajdeczko wrote:
> >>>
> >>> We don't support Observation Architecture on the VF device.
> >>>
> >>> Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
> >>> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
> >>> Cc: Ashutosh Dixit <ashutosh.dixit@intel.com>
> >>> ---
> >>> drivers/gpu/drm/xe/xe_oa.c | 4 ++++
> >>> 1 file changed, 4 insertions(+)
> >>>
> >>> diff --git a/drivers/gpu/drm/xe/xe_oa.c b/drivers/gpu/drm/xe/xe_oa.c
> >>> index 34206e0b6a08..f54a97954452 100644
> >>> --- a/drivers/gpu/drm/xe/xe_oa.c
> >>> +++ b/drivers/gpu/drm/xe/xe_oa.c
> >>> @@ -35,6 +35,7 @@
> >>> #include "xe_perf.h"
> >>> #include "xe_pm.h"
> >>> #include "xe_sched_job.h"
> >>> +#include "xe_sriov.h"
> >>>
> >>> #define DEFAULT_POLL_FREQUENCY_HZ 200
> >>> #define DEFAULT_POLL_PERIOD_NS (NSEC_PER_SEC / DEFAULT_POLL_FREQUENCY_HZ)
> >>> @@ -2385,6 +2386,9 @@ int xe_oa_init(struct xe_device *xe)
> >>> if (XE_WARN_ON(!xe_device_uc_enabled(xe)) || XE_WARN_ON(GRAPHICS_VER(xe) < 12))
> >>> return 0;
> >>>
> >>> + if (IS_SRIOV_VF(xe))
> >>> + return 0;
> >>> +
> >>
> >> Do we want to add XE_WARN_ON's
> >
> > Though maybe not, if we don't want to see the WARN_ON when loading the VF driver...
>
> yep, we don't want WARN on every VF boot
>
> and TBH IMO above XE_WARN_ON are all wrong, as:
>
> - why WARN if someone used legal param xe.force_execlist ?
> we can rely on existing drm_dbg() when someone will attempt to use oa
>
> - why WARN if Xe driver by design does not support gen < 12 ?
> if you want to be paranoid just add xe_assert() to document that
I agree, best to get rid of these WARN_ON's (xe_assert will also cause a
crash which we don't want). The ioctl's have drm_dbg's which should be
sufficient as you say. I just sent a patch to do this:
https://patchwork.freedesktop.org/series/135081/
Could you please take a look at this. Thanks.
>
> >
> >> or combine this if statement with the previous one?
> >>
> >> In any case:
> >>
> >> Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
>
> thanks
>
> >>
> >>
> >>> oa->xe = xe;
> >>> oa->oa_formats = oa_formats;
> >>>
> >>> --
> >>> 2.43.0
> >>>
next prev parent reply other threads:[~2024-06-19 22:57 UTC|newest]
Thread overview: 61+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-06-19 21:45 [PATCH 0/9] Adapt the driver to boot on the VF device Michal Wajdeczko
2024-06-19 21:45 ` [PATCH 1/9] drm/xe/vf: Disable features that do not apply to VFs Michal Wajdeczko
2024-06-20 9:17 ` Piotr Piórkowski
2024-06-20 10:05 ` Michal Wajdeczko
2024-06-20 10:01 ` [PATCH v2 " Michal Wajdeczko
2024-06-20 10:06 ` Piotr Piórkowski
2024-06-19 21:45 ` [PATCH 2/9] drm/xe/vf: Don't run any save-restore RTP actions if VF Michal Wajdeczko
2024-06-20 9:20 ` Piotr Piórkowski
2024-06-20 17:30 ` Matt Roper
2024-06-19 21:45 ` [PATCH 3/9] drm/xe/vf: Don't apply tile workarounds " Michal Wajdeczko
2024-06-20 9:21 ` Piotr Piórkowski
2024-06-20 17:31 ` Matt Roper
2024-06-19 21:45 ` [PATCH 4/9] drm/xe/vf: Don't change hwe IRQ masks if using memory IRQs Michal Wajdeczko
2024-06-20 9:29 ` Piotr Piórkowski
2024-06-19 21:45 ` [PATCH 5/9] drm/xe/vf: Don't initialize OA if VF Michal Wajdeczko
2024-06-19 21:56 ` Dixit, Ashutosh
2024-06-19 22:10 ` Dixit, Ashutosh
2024-06-19 22:40 ` Michal Wajdeczko
2024-06-19 22:57 ` Dixit, Ashutosh [this message]
2024-06-19 23:01 ` Michal Wajdeczko
2024-06-19 21:45 ` [PATCH 6/9] drm/xe/vf: Don't support gtidle " Michal Wajdeczko
2024-06-20 9:38 ` Piotr Piórkowski
2024-06-19 21:45 ` [PATCH 7/9] drm/xe/vf: Don't use register based TLB invalidation " Michal Wajdeczko
2024-06-19 23:13 ` Matthew Brost
2024-06-19 23:25 ` Michal Wajdeczko
2024-06-20 10:07 ` Piotr Piórkowski
2024-06-19 21:45 ` [PATCH 8/9] drm/xe/vf: Skip engine ring enabling " Michal Wajdeczko
2024-06-20 10:03 ` Piotr Piórkowski
2024-06-19 21:45 ` [PATCH 9/9] drm/xe/vf: Custom HuC initialization " Michal Wajdeczko
2024-06-19 23:11 ` Matthew Brost
2024-06-19 23:14 ` Michal Wajdeczko
2024-06-19 23:23 ` Matthew Brost
2024-06-19 23:34 ` Michal Wajdeczko
2024-06-20 1:59 ` Matthew Brost
2024-06-19 21:51 ` ✓ CI.Patch_applied: success for Adapt the driver to boot on the VF device Patchwork
2024-06-19 21:51 ` ✓ CI.checkpatch: " Patchwork
2024-06-19 21:52 ` ✓ CI.KUnit: " Patchwork
2024-06-19 22:04 ` ✓ CI.Build: " Patchwork
2024-06-19 22:06 ` ✗ CI.Hooks: failure " Patchwork
2024-06-19 22:07 ` ✓ CI.checksparse: success " Patchwork
2024-06-19 22:22 ` ✗ CI.BAT: failure " Patchwork
2024-06-20 6:28 ` Michal Wajdeczko
2024-06-20 6:16 ` ✗ CI.FULL: " Patchwork
2024-06-20 6:28 ` ✓ CI.Patch_applied: success for Adapt the driver to boot on the VF device (rev2) Patchwork
2024-06-20 6:28 ` ✓ CI.checkpatch: " Patchwork
2024-06-20 6:29 ` ✓ CI.KUnit: " Patchwork
2024-06-20 6:41 ` ✓ CI.Build: " Patchwork
2024-06-20 6:43 ` ✗ CI.Hooks: failure " Patchwork
2024-06-20 6:44 ` ✓ CI.checksparse: success " Patchwork
2024-06-20 7:01 ` ✓ CI.BAT: " Patchwork
2024-06-20 8:32 ` ✗ CI.FULL: failure " Patchwork
2024-06-20 10:09 ` ✓ CI.Patch_applied: success for Adapt the driver to boot on the VF device (rev3) Patchwork
2024-06-20 10:09 ` ✓ CI.checkpatch: " Patchwork
2024-06-20 10:10 ` ✓ CI.KUnit: " Patchwork
2024-06-20 10:22 ` ✓ CI.Build: " Patchwork
2024-06-20 10:24 ` ✗ CI.Hooks: failure " Patchwork
2024-06-20 10:26 ` ✓ CI.checksparse: success " Patchwork
2024-06-20 10:52 ` ✗ CI.BAT: failure " Patchwork
2024-06-20 14:29 ` Michal Wajdeczko
2024-06-20 13:01 ` ✗ CI.FULL: " Patchwork
2024-06-20 14:32 ` Michal Wajdeczko
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=855xu4r15l.wl-ashutosh.dixit@intel.com \
--to=ashutosh.dixit@intel.com \
--cc=intel-xe@lists.freedesktop.org \
--cc=michal.wajdeczko@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