From: Chris Wilson <chris@chris-wilson.co.uk>
To: Jason Ekstrand <jason@jlekstrand.net>
Cc: Intel GFX <intel-gfx@lists.freedesktop.org>
Subject: Re: [PATCH 3/4] drm/i915: Flush all user surfaces prior to first use
Date: Sat, 20 Jul 2019 11:49:21 +0100 [thread overview]
Message-ID: <156361976192.24728.12555142067859007832@skylake-alporthouse-com> (raw)
In-Reply-To: <CAOFGe94HiMuuauogvemDhfaoOD_jkracT4bJX6q28L32jUYYCA@mail.gmail.com>
Quoting Jason Ekstrand (2019-07-19 23:55:03)
> Just to be clear, is this just adding a CLFLUSH or is it actually changing the
> default caching state of buffers from CACHED to NONE? If it's actually
> changing the default state, that's going to break userspace badly.
What userspace is actually broken? You don't use set-domain, preferring
to do your own cache management and MOCS; where you do default to PTE is
where we expect the buffers to be uncached anyway. In effect, it is only
swap in of the objects that the kernel will insert an extra clflush for
you.
I'm looking for the negative impact of this, and not finding much
repercussion at all. At the end of the day, we have no choice but to
close such a hole; the only question is how much leeway to give and how
to mitigate such issues in future interfaces.
-Chris
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2019-07-20 10:49 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-07-18 14:54 [PATCH 1/4] drm/i915: Drop wmb() inside pread_gtt Chris Wilson
2019-07-18 14:54 ` [PATCH 2/4] drm/i915: Use maximum write flush for pwrite_gtt Chris Wilson
2019-07-18 18:28 ` Ville Syrjälä
2019-07-18 19:19 ` Chris Wilson
2019-07-19 0:45 ` Sasha Levin
2019-07-18 14:54 ` [PATCH 3/4] drm/i915: Flush all user surfaces prior to first use Chris Wilson
2019-07-19 10:01 ` Joonas Lahtinen
2019-07-19 10:18 ` Lionel Landwerlin
2019-07-19 10:21 ` Chris Wilson
2019-07-19 22:55 ` Jason Ekstrand
2019-07-20 10:49 ` Chris Wilson [this message]
2019-07-24 20:37 ` Francisco Jerez
2019-11-12 9:38 ` Chris Wilson
2019-11-12 9:38 ` [Intel-gfx] " Chris Wilson
2019-11-12 19:58 ` Francisco Jerez
2019-11-12 19:58 ` [Intel-gfx] " Francisco Jerez
2019-07-18 14:54 ` [PATCH 4/4] drm/i915: Flush stale cachelines on set-cache-level Chris Wilson
2019-07-19 10:03 ` Joonas Lahtinen
2019-07-19 10:03 ` Joonas Lahtinen
2019-11-12 9:09 ` [Intel-gfx] " Daniel Vetter
2019-11-12 9:09 ` Daniel Vetter
2019-11-12 9:09 ` Daniel Vetter
2019-11-12 9:42 ` Chris Wilson
2019-11-12 9:42 ` Chris Wilson
2019-11-12 10:57 ` Daniel Vetter
2019-11-12 10:57 ` Daniel Vetter
2019-11-12 12:08 ` Daniel Vetter
2019-11-12 12:08 ` Daniel Vetter
2019-07-18 15:35 ` ✗ Fi.CI.CHECKPATCH: warning for series starting with [1/4] drm/i915: Drop wmb() inside pread_gtt Patchwork
2019-07-18 16:22 ` ✓ Fi.CI.BAT: success " Patchwork
2019-07-18 17:29 ` ✓ Fi.CI.IGT: " Patchwork
2019-07-18 18:23 ` [PATCH 1/4] " Ville Syrjälä
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=156361976192.24728.12555142067859007832@skylake-alporthouse-com \
--to=chris@chris-wilson.co.uk \
--cc=intel-gfx@lists.freedesktop.org \
--cc=jason@jlekstrand.net \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.