From: Tomas Elf <tomas.elf@intel.com>
To: "John.C.Harrison@Intel.com" <John.C.Harrison@Intel.com>,
"Intel-GFX@Lists.FreeDesktop.Org"
<Intel-GFX@Lists.FreeDesktop.Org>
Subject: Re: [PATCH 07/59] drm/i915: Early alloc request in execbuff
Date: Tue, 31 Mar 2015 17:09:06 +0100 [thread overview]
Message-ID: <551AC6A2.60007@intel.com> (raw)
In-Reply-To: <1426768264-16996-8-git-send-email-John.C.Harrison@Intel.com>
On 19/03/2015 12:30, John.C.Harrison@Intel.com wrote:
> From: John Harrison <John.C.Harrison@Intel.com>
>
> Start of explicit request management in the execbuffer code path. This patch
> adds a call to allocate a request structure before all the actual hardware work
> is done. Thus guaranteeing that all that work is tagged by a known request. At
> present, nothing further is done with the request, the rest comes later in the
> series.
>
> The only noticable change is that failure to get a request (e.g. due to lack of
> memory) will be caught earlier in the sequence. It now occurs right at the start
> before any un-undoable work has been done.
>
> v2: Simplified the error handling path.
>
> For: VIZ-5115
> Signed-off-by: John Harrison <John.C.Harrison@Intel.com>
> ---
> drivers/gpu/drm/i915/i915_gem_execbuffer.c | 7 +++++++
> 1 file changed, 7 insertions(+)
>
> diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
> index 1bb237c..2504cfd 100644
> --- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
> +++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
> @@ -1594,10 +1594,16 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
> } else
> exec_start += i915_gem_obj_offset(batch_obj, vm);
>
> + /* Allocate a request for this batch buffer nice and early. */
> + ret = i915_gem_request_alloc(ring, ctx);
> + if (ret)
> + goto err_batch_unpin;
> +
> ret = dev_priv->gt.execbuf_submit(dev, file, ring, ctx, args,
> &eb->vmas, batch_obj, exec_start,
> dispatch_flags);
>
> +err_batch_unpin:
> /*
> * FIXME: We crucially rely upon the active tracking for the (ppgtt)
> * batch vma for correctness. For less ugly and less fragility this
> @@ -1606,6 +1612,7 @@ i915_gem_do_execbuffer(struct drm_device *dev, void *data,
> */
> if (dispatch_flags & I915_DISPATCH_SECURE)
> i915_gem_object_ggtt_unpin(batch_obj);
> +
> err:
> /* the request owns the ref now */
> i915_gem_context_unreference(ctx);
>
Reviewed-by: Tomas Elf <tomas.elf@intel.com>
Thanks,
Tomas
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2015-03-31 16:09 UTC|newest]
Thread overview: 122+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-03-19 12:30 [PATCH 00/59] Remove the outstanding_lazy_request John.C.Harrison
2015-03-19 12:30 ` [PATCH 01/59] drm/i915: Rename 'do_execbuf' to 'execbuf_submit' John.C.Harrison
2015-03-31 15:45 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 02/59] drm/i915: Make intel_logical_ring_begin() static John.C.Harrison
2015-03-31 15:47 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 03/59] drm/i915: Move common request allocation code into a common function John.C.Harrison
2015-03-31 15:48 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 04/59] drm/i915: Fix for ringbuf space wait in LRC mode John.C.Harrison
2015-03-19 14:56 ` Daniel, Thomas
2015-03-31 15:50 ` Tomas Elf
2015-04-01 5:56 ` Daniel Vetter
2015-04-01 12:00 ` John Harrison
2015-04-01 8:53 ` Chris Wilson
2015-03-19 12:30 ` [PATCH 05/59] drm/i915: Reserve ring buffer space for i915_add_request() commands John.C.Harrison
2015-03-20 15:13 ` Daniel Vetter
2015-03-20 15:55 ` John Harrison
2015-03-23 8:54 ` Daniel Vetter
2015-03-31 16:03 ` Chris Wilson
2015-03-20 16:19 ` John Harrison
2015-03-20 18:13 ` John Harrison
2015-03-31 15:58 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 06/59] drm/i915: i915_add_request must not fail John.C.Harrison
2015-03-19 12:30 ` [PATCH 07/59] drm/i915: Early alloc request in execbuff John.C.Harrison
2015-03-31 16:09 ` Tomas Elf [this message]
2015-03-19 12:30 ` [PATCH 08/59] drm/i915: Set context in request from creation even in legacy mode John.C.Harrison
2015-03-31 16:10 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 09/59] drm/i915: Merged the many do_execbuf() parameters into a structure John.C.Harrison
2015-03-31 16:16 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 10/59] drm/i915: Simplify i915_gem_execbuffer_retire_commands() parameters John.C.Harrison
2015-03-19 12:30 ` [PATCH 11/59] drm/i915: Update alloc_request to return the allocated request John.C.Harrison
2015-03-31 16:20 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 12/59] drm/i915: Add request to execbuf params and add explicit cleanup John.C.Harrison
2015-03-19 12:30 ` [PATCH 13/59] drm/i915: Update the dispatch tracepoint to use params->request John.C.Harrison
2015-03-19 12:30 ` [PATCH 14/59] drm/i915: Update move_to_gpu() to take a request structure John.C.Harrison
2015-03-19 12:30 ` [PATCH 15/59] drm/i915: Update execbuffer_move_to_active() " John.C.Harrison
2015-03-19 12:30 ` [PATCH 16/59] drm/i915: Add flag to i915_add_request() to skip the cache flush John.C.Harrison
2015-03-31 16:32 ` Tomas Elf
2015-04-01 5:59 ` Daniel Vetter
2015-04-01 8:52 ` Chris Wilson
2015-03-19 12:30 ` [PATCH 17/59] drm/i915: Update i915_gpu_idle() to manage its own request John.C.Harrison
2015-03-19 12:30 ` [PATCH 18/59] drm/i915: Split i915_ppgtt_init_hw() in half - generic and per ring John.C.Harrison
2015-03-31 16:34 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 19/59] drm/i915: Moved the for_each_ring loop outside of i915_gem_context_enable() John.C.Harrison
2015-03-19 12:30 ` [PATCH 20/59] drm/i915: Don't tag kernel batches as user batches John.C.Harrison
2015-03-31 16:35 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 21/59] drm/i915: Add explicit request management to i915_gem_init_hw() John.C.Harrison
2015-03-31 16:38 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 22/59] drm/i915: Update ppgtt_init_ring() & context_enable() to take requests John.C.Harrison
2015-03-31 16:38 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 23/59] drm/i915: Update i915_switch_context() to take a request structure John.C.Harrison
2015-03-19 12:30 ` [PATCH 24/59] drm/i915: Update do_switch() " John.C.Harrison
2015-03-31 16:40 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 25/59] drm/i915: Update deferred context creation to do explicit request management John.C.Harrison
2015-03-31 16:43 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 26/59] drm/i915: Update init_context() to take a request structure John.C.Harrison
2015-03-19 12:30 ` [PATCH 27/59] drm/i915: Update render_state_init() " John.C.Harrison
2015-03-19 12:30 ` [PATCH 28/59] drm/i915: Update i915_gem_object_sync() " John.C.Harrison
2015-03-31 16:53 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 29/59] drm/i915: Update overlay code to do explicit request management John.C.Harrison
2015-03-31 16:53 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 30/59] drm/i915: Update queue_flip() to take a request structure John.C.Harrison
2015-03-31 16:54 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 31/59] drm/i915: Update add_request() " John.C.Harrison
2015-03-19 12:30 ` [PATCH 32/59] drm/i915: Update [vma|object]_move_to_active() to take request structures John.C.Harrison
2015-03-19 12:30 ` [PATCH 33/59] drm/i915: Update l3_remap to take a request structure John.C.Harrison
2015-03-19 12:30 ` [PATCH 34/59] drm/i915: Update mi_set_context() " John.C.Harrison
2015-03-19 12:30 ` [PATCH 35/59] drm/i915: Update a bunch of execbuffer helpers to take request structures John.C.Harrison
2015-03-19 12:30 ` [PATCH 36/59] drm/i915: Update workarounds_emit() " John.C.Harrison
2015-03-19 12:30 ` [PATCH 37/59] drm/i915: Update flush_all_caches() " John.C.Harrison
2015-03-19 12:30 ` [PATCH 38/59] drm/i915: Update switch_mm() to take a request structure John.C.Harrison
2015-03-31 16:57 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 39/59] drm/i915: Update ring->flush() to take a requests structure John.C.Harrison
2015-03-31 16:59 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 40/59] drm/i915: Update some flush helpers to take request structures John.C.Harrison
2015-03-19 12:30 ` [PATCH 41/59] drm/i915: Update ring->emit_flush() to take a request structure John.C.Harrison
2015-03-19 12:30 ` [PATCH 42/59] drm/i915: Update ring->add_request() " John.C.Harrison
2015-03-19 12:30 ` [PATCH 43/59] drm/i915: Update ring->emit_request() " John.C.Harrison
2015-03-31 17:01 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 44/59] drm/i915: Update ring->dispatch_execbuffer() " John.C.Harrison
2015-03-19 12:30 ` [PATCH 45/59] drm/i915: Update ring->emit_bb_start() " John.C.Harrison
2015-03-19 12:30 ` [PATCH 46/59] drm/i915: Update ring->sync_to() " John.C.Harrison
2015-03-31 17:03 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 47/59] drm/i915: Update ring->signal() " John.C.Harrison
2015-03-19 12:30 ` [PATCH 48/59] drm/i915: Update cacheline_align() " John.C.Harrison
2015-03-19 12:30 ` [PATCH 49/59] drm/i915: Update intel_ring_begin() " John.C.Harrison
2015-03-31 17:04 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 50/59] drm/i915: Update intel_logical_ring_begin() " John.C.Harrison
2015-03-31 17:05 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 51/59] drm/i915: Add *_ring_begin() to request allocation John.C.Harrison
2015-03-20 15:23 ` Daniel Vetter
2015-03-20 15:30 ` Chris Wilson
2015-03-20 16:09 ` John Harrison
2015-03-23 9:10 ` Daniel Vetter
2015-03-31 17:17 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 52/59] drm/i915: Remove the now obsolete intel_ring_get_request() John.C.Harrison
2015-03-19 12:30 ` [PATCH 53/59] drm/i915: Remove the now obsolete 'outstanding_lazy_request' John.C.Harrison
2015-03-31 18:01 ` Tomas Elf
2015-03-19 12:30 ` [PATCH 54/59] drm/i915: Move the request/file and request/pid association to creation time John.C.Harrison
2015-03-31 18:07 ` Tomas Elf
2015-03-19 12:31 ` [PATCH 55/59] drm/i915: Remove fallback poll for ring buffer space John.C.Harrison
2015-03-19 15:00 ` Daniel, Thomas
2015-03-19 15:16 ` Jani Nikula
2015-03-19 16:33 ` John Harrison
2015-03-19 17:29 ` Daniel Vetter
2015-03-31 18:13 ` Tomas Elf
2015-04-01 6:02 ` Daniel Vetter
2015-04-01 8:51 ` Chris Wilson
2015-03-19 12:31 ` [PATCH 56/59] drm/i915: Remove 'faked' request from LRC submission John.C.Harrison
2015-03-19 15:02 ` Daniel, Thomas
2015-03-31 18:14 ` Tomas Elf
2015-03-19 12:31 ` [PATCH 57/59] drm/i915: Update a bunch of LRC functions to take requests John.C.Harrison
2015-03-31 18:18 ` Tomas Elf
2015-03-19 12:31 ` [PATCH 58/59] drm/i915: Remove the now obsolete 'i915_gem_check_olr()' John.C.Harrison
2015-03-31 18:18 ` Tomas Elf
2015-03-19 12:31 ` [PATCH 59/59] drm/i915: Remove the almost obsolete i915_gem_object_flush_active() John.C.Harrison
2015-03-31 18:32 ` Tomas Elf
2015-04-01 6:06 ` Daniel Vetter
2015-05-28 20:02 ` [PATCH 00/59] Remove the outstanding_lazy_request Jesse Barnes
2015-05-28 21:20 ` Chris Wilson
2015-05-29 14:37 ` Jesse Barnes
2015-05-29 18:07 ` Chris Wilson
2015-05-29 11:00 ` John Harrison
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=551AC6A2.60007@intel.com \
--to=tomas.elf@intel.com \
--cc=Intel-GFX@Lists.FreeDesktop.Org \
--cc=John.C.Harrison@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