All of lore.kernel.org
 help / color / mirror / Atom feed
From: John Harrison <John.C.Harrison@Intel.com>
To: "Intel-GFX@Lists.FreeDesktop.Org" <intel-gfx@lists.freedesktop.org>
Subject: Re: [RFC 00/12] Convert requests to use struct fence
Date: Mon, 23 Nov 2015 11:38:48 +0000	[thread overview]
Message-ID: <5652FAC8.6090406@Intel.com> (raw)
In-Reply-To: <1448278471-31181-1-git-send-email-John.C.Harrison@Intel.com>

Oops, forgot to update the tag. These should be PATCH not RFC.

On 23/11/2015 11:34, John.C.Harrison@Intel.com wrote:
> From: John Harrison <John.C.Harrison@Intel.com>
>
> There is a construct in the linux kernel called 'struct fence' that is
> intended to keep track of work that is executed on hardware. I.e. it
> solves the basic problem that the drivers 'struct
> drm_i915_gem_request' is trying to address. The request structure does
> quite a lot more than simply track the execution progress so is very
> definitely still required. However, the basic completion status side
> could be updated to use the ready made fence implementation and gain
> all the advantages that provides.
>
> Using the struct fence object also has the advantage that the fence
> can be used outside of the i915 driver (by other drivers or by
> userland applications). That is the basis of the dma-buff
> synchronisation API and allows asynchronous tracking of work
> completion. In this case, it allows applications to be signalled
> directly when a batch buffer completes without having to make an IOCTL
> call into the driver.
>
> This is work that was planned since the conversion of the driver from
> being seqno value based to being request structure based. This patch
> series does that work.
>
> An IGT test to exercise the fence support from user land is in
> progress and will follow. Android already makes extensive use of
> fences for display composition. Real world linux usage is planned in
> the form of Jesse's page table sharing / bufferless execbuf support.
> There is also a plan that Wayland (and others) could make use of it in
> a similar manner to Android.
>
> v2: Updated for review comments by various people and to add support
> for Android style 'native sync'.
>
> v3: Updated from review comments by Tvrtko Ursulin. Also moved sync
> framework out of staging and improved request completion handling.
>
> [Patches against drm-intel-nightly tree fetched 17/11/2015]
>
> John Harrison (9):
>    staging/android/sync: Move sync framework out of staging
>    drm/i915: Convert requests to use struct fence
>    drm/i915: Removed now redudant parameter to i915_gem_request_completed()
>    drm/i915: Add per context timelines to fence object
>    drm/i915: Delay the freeing of requests until retire time
>    drm/i915: Interrupt driven fences
>    drm/i915: Updated request structure tracing
>    drm/i915: Add sync framework support to execbuff IOCTL
>    drm/i915: Cache last IRQ seqno to reduce IRQ overhead
>
> Maarten Lankhorst (1):
>    staging/android/sync: add sync_fence_create_dma
>
> Peter Lawthers (1):
>    android/sync: Fix reversed sense of signaled fence
>
> Tvrtko Ursulin (1):
>    staging/android/sync: Support sync points created from dma-fences
>
>   drivers/android/Kconfig                    |  28 ++
>   drivers/android/Makefile                   |   2 +
>   drivers/android/sw_sync.c                  | 260 ++++++++++
>   drivers/android/sw_sync.h                  |  59 +++
>   drivers/android/sync.c                     | 734 +++++++++++++++++++++++++++++
>   drivers/android/sync.h                     | 387 +++++++++++++++
>   drivers/android/sync_debug.c               | 256 ++++++++++
>   drivers/android/trace/sync.h               |  82 ++++
>   drivers/gpu/drm/i915/Kconfig               |   3 +
>   drivers/gpu/drm/i915/i915_debugfs.c        |   7 +-
>   drivers/gpu/drm/i915/i915_drv.h            |  75 +--
>   drivers/gpu/drm/i915/i915_gem.c            | 437 ++++++++++++++++-
>   drivers/gpu/drm/i915/i915_gem_context.c    |  15 +-
>   drivers/gpu/drm/i915/i915_gem_execbuffer.c |  95 +++-
>   drivers/gpu/drm/i915/i915_irq.c            |   2 +-
>   drivers/gpu/drm/i915/i915_trace.h          |  13 +-
>   drivers/gpu/drm/i915/intel_display.c       |   4 +-
>   drivers/gpu/drm/i915/intel_lrc.c           |  13 +
>   drivers/gpu/drm/i915/intel_pm.c            |   6 +-
>   drivers/gpu/drm/i915/intel_ringbuffer.c    |   5 +
>   drivers/gpu/drm/i915/intel_ringbuffer.h    |   9 +
>   drivers/staging/android/Kconfig            |  28 --
>   drivers/staging/android/Makefile           |   2 -
>   drivers/staging/android/sw_sync.c          | 260 ----------
>   drivers/staging/android/sw_sync.h          |  59 ---
>   drivers/staging/android/sync.c             | 729 ----------------------------
>   drivers/staging/android/sync.h             | 356 --------------
>   drivers/staging/android/sync_debug.c       | 254 ----------
>   drivers/staging/android/trace/sync.h       |  82 ----
>   drivers/staging/android/uapi/sw_sync.h     |  32 --
>   drivers/staging/android/uapi/sync.h        |  97 ----
>   include/uapi/Kbuild                        |   1 +
>   include/uapi/drm/i915_drm.h                |  16 +-
>   include/uapi/sync/Kbuild                   |   3 +
>   include/uapi/sync/sw_sync.h                |  32 ++
>   include/uapi/sync/sync.h                   |  97 ++++
>   36 files changed, 2569 insertions(+), 1971 deletions(-)
>   create mode 100644 drivers/android/sw_sync.c
>   create mode 100644 drivers/android/sw_sync.h
>   create mode 100644 drivers/android/sync.c
>   create mode 100644 drivers/android/sync.h
>   create mode 100644 drivers/android/sync_debug.c
>   create mode 100644 drivers/android/trace/sync.h
>   delete mode 100644 drivers/staging/android/sw_sync.c
>   delete mode 100644 drivers/staging/android/sw_sync.h
>   delete mode 100644 drivers/staging/android/sync.c
>   delete mode 100644 drivers/staging/android/sync.h
>   delete mode 100644 drivers/staging/android/sync_debug.c
>   delete mode 100644 drivers/staging/android/trace/sync.h
>   delete mode 100644 drivers/staging/android/uapi/sw_sync.h
>   delete mode 100644 drivers/staging/android/uapi/sync.h
>   create mode 100644 include/uapi/sync/Kbuild
>   create mode 100644 include/uapi/sync/sw_sync.h
>   create mode 100644 include/uapi/sync/sync.h
>

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

  parent reply	other threads:[~2015-11-23 11:38 UTC|newest]

Thread overview: 92+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-11-29  8:47 i915_wait_request scaling Chris Wilson
2015-11-29  8:47 ` [PATCH 01/15] drm/i915: Break busywaiting for requests on pending signals Chris Wilson
2015-11-30 10:01   ` Tvrtko Ursulin
2015-11-29  8:48 ` [PATCH 02/15] drm/i915: Limit the busy wait on requests to 10us not 10ms! Chris Wilson
2015-11-30 10:02   ` Tvrtko Ursulin
2015-11-30 10:08     ` Chris Wilson
2015-11-29  8:48 ` [PATCH 03/15] drm/i915: Only spin whilst waiting on the current request Chris Wilson
2015-11-30 10:06   ` Tvrtko Ursulin
2015-12-01 15:47     ` Dave Gordon
2015-12-01 15:58       ` Chris Wilson
2015-12-01 16:44         ` Dave Gordon
2015-12-03  8:52           ` Daniel Vetter
2015-11-29  8:48 ` [PATCH 04/15] drm/i915: Cache the reset_counter for the request Chris Wilson
2015-12-01  8:31   ` Daniel Vetter
2015-12-01  8:47     ` Chris Wilson
2015-12-01  9:15       ` Chris Wilson
2015-12-01 11:05         ` [PATCH 1/3] drm/i915: Hide the atomic_read(reset_counter) behind a helper Chris Wilson
2015-12-01 11:05           ` [PATCH 2/3] drm/i915: Store the reset counter when constructing a request Chris Wilson
2015-12-03  8:59             ` Daniel Vetter
2015-12-01 11:05           ` [PATCH 3/3] drm/i915: Prevent leaking of -EIO from i915_wait_request() Chris Wilson
2015-12-03  9:14             ` Daniel Vetter
2015-12-03  9:41               ` Chris Wilson
2015-12-11  9:02               ` Chris Wilson
2015-12-11 16:46                 ` Daniel Vetter
2015-12-03  8:57           ` [PATCH 1/3] drm/i915: Hide the atomic_read(reset_counter) behind a helper Daniel Vetter
2015-12-03  9:02             ` Chris Wilson
2015-12-03  9:20               ` Daniel Vetter
2015-11-29  8:48 ` [PATCH 05/15] drm/i915: Suppress error message when GPU resets are disabled Chris Wilson
2015-12-01  8:30   ` Daniel Vetter
2015-11-29  8:48 ` [PATCH 06/15] drm/i915: Delay queuing hangcheck to wait-request Chris Wilson
2015-11-29  8:48 ` [PATCH 07/15] drm/i915: Check the timeout passed to i915_wait_request Chris Wilson
2015-11-30 10:14   ` Tvrtko Ursulin
2015-11-30 10:19     ` Chris Wilson
2015-11-30 10:27       ` Tvrtko Ursulin
2015-11-30 10:22   ` Chris Wilson
2015-11-30 10:28     ` Tvrtko Ursulin
2015-11-29  8:48 ` [PATCH 08/15] drm/i915: Slaughter the thundering i915_wait_request herd Chris Wilson
2015-11-30 10:53   ` Chris Wilson
2015-11-30 12:09     ` Tvrtko Ursulin
2015-11-30 12:38       ` Chris Wilson
2015-11-30 13:33         ` Tvrtko Ursulin
2015-11-30 14:30           ` Chris Wilson
2015-11-30 12:05   ` Tvrtko Ursulin
2015-11-30 12:30     ` Chris Wilson
2015-11-30 13:32       ` Tvrtko Ursulin
2015-11-30 14:18         ` Chris Wilson
2015-12-01 17:06           ` Dave Gordon
2015-11-30 14:26         ` Chris Wilson
2015-11-30 14:34   ` [PATCH v4] " Chris Wilson
2015-11-30 16:30     ` Chris Wilson
2015-11-30 16:40       ` Chris Wilson
2015-12-01 18:34     ` Dave Gordon
2015-12-03 16:22       ` [PATCH v7] " Chris Wilson
2015-12-07 15:08         ` Tvrtko Ursulin
2015-12-08 10:44           ` Chris Wilson
2015-12-08 14:03             ` Tvrtko Ursulin
2015-12-08 14:33               ` Chris Wilson
2015-11-23 11:34                 ` [RFC 00/12] Convert requests to use struct fence John.C.Harrison
2015-11-23 11:34                   ` [RFC 01/12] staging/android/sync: Support sync points created from dma-fences John.C.Harrison
2015-11-23 13:29                     ` Maarten Lankhorst
2015-11-23 13:31                     ` [Intel-gfx] " Tvrtko Ursulin
2015-11-23 11:34                   ` [RFC 02/12] staging/android/sync: add sync_fence_create_dma John.C.Harrison
2015-11-23 13:27                     ` Maarten Lankhorst
2015-11-23 13:38                       ` John Harrison
2015-11-23 13:44                         ` Tvrtko Ursulin
2015-11-23 13:48                           ` Maarten Lankhorst
2015-11-23 11:34                   ` [RFC 03/12] staging/android/sync: Move sync framework out of staging John.C.Harrison
2015-11-23 11:34                   ` [RFC 04/12] drm/i915: Convert requests to use struct fence John.C.Harrison
2015-11-23 11:34                   ` [RFC 05/12] drm/i915: Removed now redudant parameter to i915_gem_request_completed() John.C.Harrison
2015-11-23 11:34                   ` [RFC 06/12] drm/i915: Add per context timelines to fence object John.C.Harrison
2015-11-23 11:34                   ` [RFC 07/12] drm/i915: Delay the freeing of requests until retire time John.C.Harrison
2015-11-23 11:34                   ` [RFC 08/12] drm/i915: Interrupt driven fences John.C.Harrison
2015-12-11 12:17                     ` Tvrtko Ursulin
2015-11-23 11:34                   ` [RFC 09/12] drm/i915: Updated request structure tracing John.C.Harrison
2015-11-23 11:34                   ` [RFC 10/12] android/sync: Fix reversed sense of signaled fence John.C.Harrison
2015-11-23 11:34                   ` [RFC 11/12] drm/i915: Add sync framework support to execbuff IOCTL John.C.Harrison
2015-11-23 11:34                   ` [RFC 12/12] drm/i915: Cache last IRQ seqno to reduce IRQ overhead John.C.Harrison
2015-11-23 11:38                   ` John Harrison [this message]
2015-12-08 14:53                   ` [PATCH v7] drm/i915: Slaughter the thundering i915_wait_request herd Dave Gordon
2015-11-30 15:45   ` [PATCH] drm/i915: Convert trace-irq to the breadcrumb waiter Chris Wilson
2015-11-29  8:48 ` [PATCH 09/15] drm/i915: Separate out the seqno-barrier from engine->get_seqno Chris Wilson
2015-11-29  8:48 ` [PATCH 10/15] drm/i915: Remove the lazy_coherency parameter from request-completed? Chris Wilson
2015-11-29  8:48 ` [PATCH 11/15] drm/i915: Use HWS for seqno tracking everywhere Chris Wilson
2015-11-29  8:48 ` [PATCH 12/15] drm/i915: Reduce seqno/irq barrier to a clflush on legacy gen6+ Chris Wilson
2015-11-29  8:48 ` [PATCH 13/15] drm/i915: Stop setting wraparound seqno on initialisation Chris Wilson
2015-12-01 16:57   ` Dave Gordon
2015-12-04  9:36     ` Daniel Vetter
2015-12-04  9:51       ` Chris Wilson
2015-11-29  8:48 ` [PATCH 14/15] drm/i915: Only query timestamp when measuring elapsed time Chris Wilson
2015-11-30 10:19   ` Tvrtko Ursulin
2015-11-30 14:31     ` Chris Wilson
2015-11-29  8:48 ` [PATCH 15/15] drm/i915: On GPU reset, set the HWS breadcrumb to the last seqno Chris Wilson

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=5652FAC8.6090406@Intel.com \
    --to=john.c.harrison@intel.com \
    --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.