From: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>, intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH 17/31] drm/i915: Remove obsolete engine->gpu_caches_dirty
Date: Mon, 25 Jul 2016 12:14:44 +0300 [thread overview]
Message-ID: <1469438084.5495.34.camel@linux.intel.com> (raw)
In-Reply-To: <1469432687-22756-18-git-send-email-chris@chris-wilson.co.uk>
On ma, 2016-07-25 at 08:44 +0100, Chris Wilson wrote:
> Space for flushing the GPU cache prior to completing the request is
> preallocated and so cannot fail.
Patch title and commit message have some disconnect. Could you explain
in a bit more detail what made gpu_caches_dirty obsolete?
Also, worth mentioning that after this change legacy/execlist flushing
code is unified (could be split patch too? With your "Now that ..."
reasoning).
Somebody not reviewing the series linearly might feel lost.
> diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
> index ca1d4f573832..048050176ff9 100644
> --- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
> +++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
> @@ -998,10 +998,8 @@ i915_gem_execbuffer_move_to_gpu(struct drm_i915_gem_request *req,
> if (flush_domains & I915_GEM_DOMAIN_GTT)
> wmb();
>
> - /* Unconditionally invalidate gpu caches and ensure that we do flush
> - * any residual writes from the previous batch.
> - */
> - return intel_engine_invalidate_all_caches(req);
> + /* Unconditionally invalidate gpu caches and TLBs. */
A nitpick, but maybe s/gpu/GPU/
> + return req->engine->emit_flush(req, I915_GEM_GPU_DOMAINS, 0);
> }
>
> static bool
<SNIP>
> diff --git a/drivers/gpu/drm/i915/i915_gem_request.c b/drivers/gpu/drm/i915/i915_gem_request.c
> index 942b5b1f1602..7b772d914e23 100644
> --- a/drivers/gpu/drm/i915/i915_gem_request.c
> +++ b/drivers/gpu/drm/i915/i915_gem_request.c
> @@ -451,10 +451,9 @@ void __i915_add_request(struct drm_i915_gem_request *request,
> * what.
> */
> if (flush_caches) {
> - if (i915.enable_execlists)
> - ret = logical_ring_flush_all_caches(request);
> - else
> - ret = intel_engine_flush_all_caches(request);
> + ret = request->engine->emit_flush(request,
> + 0, I915_GEM_GPU_DOMAINS);
> +
> /* Not allowed to fail! */
> WARN(ret, "*_ring_flush_all_caches failed: %d!\n", ret);
Fix this message too.
Regards, Joonas
--
Joonas Lahtinen
Open Source Technology Center
Intel Corporation
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2016-07-25 9:14 UTC|newest]
Thread overview: 94+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-07-25 7:44 Refined set of intel_ringbuffer renames Chris Wilson
2016-07-25 7:44 ` [PATCH 01/31] drm/i915: Reduce breadcrumb lock coverage for intel_engine_enable_signaling() Chris Wilson
2016-07-26 4:37 ` Joonas Lahtinen
2016-07-25 7:44 ` [PATCH 02/31] drm/i915: Prefer list_first_entry_or_null Chris Wilson
2016-07-25 7:55 ` Joonas Lahtinen
2016-07-25 8:03 ` Chris Wilson
2016-07-25 10:01 ` Joonas Lahtinen
2016-07-25 7:44 ` [PATCH 03/31] drm/i915: Only clear the client pointer when tearing down the file Chris Wilson
2016-07-25 8:15 ` Joonas Lahtinen
2016-07-25 7:44 ` [PATCH 04/31] drm/i915: Only drop the batch-pool's object reference Chris Wilson
2016-07-25 8:38 ` Joonas Lahtinen
2016-07-25 8:44 ` Chris Wilson
2016-07-25 10:43 ` Joonas Lahtinen
2016-07-25 7:44 ` [PATCH 05/31] drm/i915/cmdparser: Remove stray intel_engine_cs *ring Chris Wilson
2016-07-25 8:40 ` Joonas Lahtinen
2016-07-25 9:06 ` [PATCH v2] " Chris Wilson
2016-07-25 11:01 ` Joonas Lahtinen
2016-07-25 11:12 ` Chris Wilson
2016-07-25 7:44 ` [PATCH 06/31] drm/i915: Use engine to refer to the user's BSD intel_engine_cs Chris Wilson
2016-07-25 8:42 ` Joonas Lahtinen
2016-07-25 7:44 ` [PATCH 07/31] drm/i915: Avoid using intel_engine_cs *ring for GPU error capture Chris Wilson
2016-07-25 7:44 ` [PATCH 08/31] drm/i915: Remove stray intel_engine_cs ring identifiers from i915_gem.c Chris Wilson
2016-07-25 8:45 ` Joonas Lahtinen
2016-07-25 8:49 ` Chris Wilson
2016-07-26 15:12 ` Dave Gordon
2016-07-25 7:44 ` [PATCH 09/31] drm/i915: Update a couple of hangcheck comments to talk about engines Chris Wilson
2016-07-25 8:46 ` Joonas Lahtinen
2016-07-25 7:44 ` [PATCH 10/31] drm/i915: Unify intel_logical_ring_emit and intel_ring_emit Chris Wilson
2016-07-25 7:44 ` [PATCH 11/31] drm/i915: Rename request->ringbuf to request->ring Chris Wilson
2016-07-25 7:44 ` [PATCH 12/31] drm/i915: Rename backpointer from intel_ringbuffer to intel_engine_cs Chris Wilson
2016-07-25 8:49 ` Joonas Lahtinen
2016-07-25 9:10 ` Chris Wilson
2016-07-25 7:44 ` [PATCH 13/31] drm/i915: Rename intel_context[engine].ringbuf Chris Wilson
2016-07-25 7:44 ` [PATCH 14/31] drm/i915: Rename struct intel_ringbuffer to struct intel_ring Chris Wilson
2016-07-25 7:44 ` [PATCH 15/31] drm/i915: Rename residual ringbuf parameters Chris Wilson
2016-07-25 8:58 ` Joonas Lahtinen
2016-07-25 7:44 ` [PATCH 16/31] drm/i915: Rename intel_pin_and_map_ring() Chris Wilson
2016-07-25 7:44 ` [PATCH 17/31] drm/i915: Remove obsolete engine->gpu_caches_dirty Chris Wilson
2016-07-25 9:14 ` Joonas Lahtinen [this message]
2016-07-25 9:24 ` Chris Wilson
2016-07-27 9:49 ` Dave Gordon
2016-07-27 10:00 ` Chris Wilson
2016-07-27 11:18 ` Dave Gordon
2016-07-27 11:26 ` Joonas Lahtinen
2016-07-27 10:53 ` [PATCH] drm/i915: Reduce engine->emit_flush() to a single mode parameter Chris Wilson
2016-07-28 7:11 ` Joonas Lahtinen
2016-07-28 8:37 ` Chris Wilson
2016-07-28 10:03 ` Joonas Lahtinen
2016-07-28 14:57 ` Dave Gordon
2016-07-25 7:44 ` [PATCH 18/31] drm/i915: Simplify request_alloc by returning the allocated request Chris Wilson
2016-07-25 9:18 ` Joonas Lahtinen
2016-07-27 11:08 ` Dave Gordon
2016-07-27 15:28 ` Chris Wilson
2016-07-28 12:48 ` Dave Gordon
2016-07-28 15:10 ` Chris Wilson
2016-07-28 15:20 ` Dave Gordon
2016-07-25 7:44 ` [PATCH 19/31] drm/i915: Unify legacy/execlists emission of MI_BATCHBUFFER_START Chris Wilson
2016-07-25 7:44 ` [PATCH 20/31] drm/i915: Remove intel_ring_get_tail() Chris Wilson
2016-07-25 9:43 ` Joonas Lahtinen
2016-07-25 7:44 ` [PATCH 21/31] drm/i915: Convert engine->write_tail to operate on a request Chris Wilson
2016-07-27 11:53 ` Dave Gordon
2016-07-27 12:29 ` Chris Wilson
2016-07-28 15:05 ` Dave Gordon
2016-07-28 15:09 ` Chris Wilson
2016-07-27 12:30 ` Chris Wilson
2016-07-28 6:41 ` Joonas Lahtinen
2016-07-28 7:12 ` Chris Wilson
2016-07-28 7:52 ` Joonas Lahtinen
2016-07-28 9:16 ` [PATCH 1/2] " Chris Wilson
2016-07-28 9:16 ` [PATCH 2/2] drm/i915: Move the modulus for ring emission to the register write Chris Wilson
2016-07-28 9:59 ` Joonas Lahtinen
2016-07-28 15:16 ` Dave Gordon
2016-07-25 7:44 ` [PATCH 22/31] drm/i915: Unify request submission Chris Wilson
2016-07-25 9:49 ` Joonas Lahtinen
2016-07-25 7:44 ` [PATCH 23/31] drm/i915/lrc: Update function names to match request flow Chris Wilson
2016-07-25 9:50 ` Joonas Lahtinen
2016-07-25 7:44 ` [PATCH 24/31] drm/i915: Stop passing caller's num_dwords to engine->semaphore.signal() Chris Wilson
2016-07-25 9:53 ` Joonas Lahtinen
2016-07-25 7:44 ` [PATCH 25/31] drm/i915: Reuse legacy breadcrumbs + tail emission Chris Wilson
2016-07-28 15:23 ` Dave Gordon
2016-07-28 15:29 ` Chris Wilson
2016-07-28 15:33 ` Dave Gordon
2016-07-25 7:44 ` [PATCH 26/31] drm/i915/ringbuffer: Specialise SNB+ request emission for semaphores Chris Wilson
2016-07-25 9:55 ` Joonas Lahtinen
2016-07-25 7:44 ` [PATCH 27/31] drm/i915: Remove duplicate golden render state init from execlists Chris Wilson
2016-07-25 7:44 ` [PATCH 28/31] drm/i915: Refactor golden render state emission to unconfuse gcc Chris Wilson
2016-07-25 9:59 ` Joonas Lahtinen
2016-07-25 7:44 ` [PATCH 29/31] drm/i915: Unify legacy/execlists submit_execbuf callbacks Chris Wilson
2016-07-25 7:44 ` [PATCH 30/31] drm/i915: Simplify calling engine->sync_to Chris Wilson
2016-07-25 7:44 ` [PATCH 31/31] drm/i915: Rename engine->semaphore.sync_to, engine->sempahore.signal locals Chris Wilson
2016-07-25 8:28 ` ✓ Ro.CI.BAT: success for series starting with [01/31] drm/i915: Reduce breadcrumb lock coverage for intel_engine_enable_signaling() Patchwork
2016-07-25 9:32 ` ✓ Ro.CI.BAT: success for series starting with [01/31] drm/i915: Reduce breadcrumb lock coverage for intel_engine_enable_signaling() (rev2) Patchwork
2016-07-27 11:00 ` ✗ Ro.CI.BAT: failure for series starting with [01/31] drm/i915: Reduce breadcrumb lock coverage for intel_engine_enable_signaling() (rev3) Patchwork
2016-07-28 9:20 ` ✗ Ro.CI.BAT: failure for series starting with [01/31] drm/i915: Reduce breadcrumb lock coverage for intel_engine_enable_signaling() (rev5) 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=1469438084.5495.34.camel@linux.intel.com \
--to=joonas.lahtinen@linux.intel.com \
--cc=chris@chris-wilson.co.uk \
--cc=intel-gfx@lists.freedesktop.org \
/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.