public inbox for intel-gfx@lists.freedesktop.org
 help / color / mirror / Atom feed
From: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>, intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH 70/73] drm/i915: Move obj->active:5 to obj->flags
Date: Mon, 01 Aug 2016 15:46:53 +0300	[thread overview]
Message-ID: <1470055613.4174.17.camel@linux.intel.com> (raw)
In-Reply-To: <1470042681-25318-71-git-send-email-chris@chris-wilson.co.uk>

On ma, 2016-08-01 at 10:11 +0100, Chris Wilson wrote:
> We are motivated to avoid using a bitfield for obj->active for a couple
> of reasons. Firstly, we wish to document our lockless read of obj->active
> using READ_ONCE inside i915_gem_busy_ioctl() and that requires an
> integral type (i.e. not a bitfield). Secondly, gcc produces abysmal code
> when presented with a bitfield and that shows up high on the profiles of
> request tracking (mainly due to excess memory traffic as it converts
> the bitfield to a register and back and generates frequent AGI in the
> process).
> 
> v2: BIT, break up a long line in compute the other engines, new paint
> for i915_gem_object_is_active (now i915_gem_object_get_active).
> 
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

Patch reads much better now,

Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>

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

  reply	other threads:[~2016-08-01 12:46 UTC|newest]

Thread overview: 94+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-08-01  9:10 A few bug fixes leading to exporting prime fences [mostly reviewed] Chris Wilson
2016-08-01  9:10 ` [PATCH 01/73] drm/i915: Unify intel_logical_ring_emit and intel_ring_emit Chris Wilson
2016-08-01  9:10 ` [PATCH 02/73] drm/i915: Rename request->ringbuf to request->ring Chris Wilson
2016-08-01  9:10 ` [PATCH 03/73] drm/i915: Rename intel_context[engine].ringbuf Chris Wilson
2016-08-01  9:10 ` [PATCH 04/73] drm/i915: Rename struct intel_ringbuffer to struct intel_ring Chris Wilson
2016-08-01  9:10 ` [PATCH 05/73] drm/i915: Rename residual ringbuf parameters Chris Wilson
2016-08-01  9:10 ` [PATCH 06/73] drm/i915: Rename intel_pin_and_map_ring() Chris Wilson
2016-08-01  9:10 ` [PATCH 07/73] drm/i915: Remove obsolete engine->gpu_caches_dirty Chris Wilson
2016-08-01  9:10 ` [PATCH 08/73] drm/i915: Reduce engine->emit_flush() to a single mode parameter Chris Wilson
2016-08-01  9:10 ` [PATCH 09/73] drm/i915: Simplify request_alloc by returning the allocated request Chris Wilson
2016-08-01  9:10 ` [PATCH 10/73] drm/i915: Unify legacy/execlists emission of MI_BATCHBUFFER_START Chris Wilson
2016-08-01  9:10 ` [PATCH 11/73] drm/i915: Remove intel_ring_get_tail() Chris Wilson
2016-08-01  9:10 ` [PATCH 12/73] drm/i915: Convert engine->write_tail to operate on a request Chris Wilson
2016-08-01  9:10 ` [PATCH 13/73] drm/i915: Move the modulus for ring emission to the register write Chris Wilson
2016-08-01 10:07   ` Joonas Lahtinen
2016-08-01 10:15     ` Chris Wilson
2016-08-01 14:28       ` Joonas Lahtinen
2016-08-01 16:17         ` Chris Wilson
2016-08-01 16:23           ` Chris Wilson
2016-08-01 16:32         ` Chris Wilson
2016-08-02  9:42           ` Dave Gordon
2016-08-02 10:14             ` Chris Wilson
2016-08-01 10:21     ` Chris Wilson
2016-08-01  9:10 ` [PATCH 14/73] drm/i915: Unify request submission Chris Wilson
2016-08-01 14:30   ` Joonas Lahtinen
2016-08-01 17:17   ` [PATCH v2] " Chris Wilson
2016-08-01  9:10 ` [PATCH 15/73] drm/i915/lrc: Update function names to match request flow Chris Wilson
2016-08-01  9:10 ` [PATCH 16/73] drm/i915: Stop passing caller's num_dwords to engine->semaphore.signal() Chris Wilson
2016-08-01  9:10 ` [PATCH 17/73] drm/i915: Reuse legacy breadcrumbs + tail emission Chris Wilson
2016-08-01  9:10 ` [PATCH 18/73] drm/i915/ringbuffer: Specialise SNB+ request emission for semaphores Chris Wilson
2016-08-01  9:10 ` [PATCH 19/73] drm/i915: Remove duplicate golden render state init from execlists Chris Wilson
2016-08-01  9:10 ` [PATCH 20/73] drm/i915: Refactor golden render state emission to unconfuse gcc Chris Wilson
2016-08-01  9:10 ` [PATCH 21/73] drm/i915: Unify legacy/execlists submit_execbuf callbacks Chris Wilson
2016-08-01  9:10 ` [PATCH 22/73] drm/i915: Simplify calling engine->sync_to Chris Wilson
2016-08-01  9:10 ` [PATCH 23/73] drm/i915: Rename engine->semaphore.sync_to, engine->sempahore.signal locals Chris Wilson
2016-08-01  9:10 ` [PATCH 24/73] drm/i915: Amalgamate GGTT/ppGTT vma debug list walkers Chris Wilson
2016-08-01  9:10 ` [PATCH 25/73] drm/i915: Split early global GTT initialisation Chris Wilson
2016-08-01  9:10 ` [PATCH 26/73] drm/i915: Store owning file on the i915_address_space Chris Wilson
2016-08-01  9:10 ` [PATCH 27/73] drm/i915: Count how many VMA are bound for an object Chris Wilson
2016-08-01  9:10 ` [PATCH 28/73] drm/i915: Be more careful when unbinding vma Chris Wilson
2016-08-01  9:10 ` [PATCH 29/73] drm/i915: Kill drop_pages() Chris Wilson
2016-08-01  9:10 ` [PATCH 30/73] drm/i915: Introduce i915_gem_active for request tracking Chris Wilson
2016-08-01  9:10 ` [PATCH 31/73] drm/i915: Prepare i915_gem_active for annotations Chris Wilson
2016-08-01  9:10 ` [PATCH 32/73] drm/i915: Mark up i915_gem_active for locking annotation Chris Wilson
2016-08-01  9:10 ` [PATCH 33/73] drm/i915: Refactor blocking waits Chris Wilson
2016-08-01  9:10 ` [PATCH 34/73] drm/i915: Rename request->list to link for consistency Chris Wilson
2016-08-01  9:10 ` [PATCH 35/73] drm/i915: Remove obsolete i915_gem_object_flush_active() Chris Wilson
2016-08-01  9:10 ` [PATCH 36/73] drm/i915: Refactor activity tracking for requests Chris Wilson
2016-08-01 12:52   ` Joonas Lahtinen
2016-08-01  9:10 ` [PATCH 37/73] drm/i915: Track requests inside each intel_ring Chris Wilson
2016-08-01  9:10 ` [PATCH 38/73] drm/i915: Convert intel_overlay to request tracking Chris Wilson
2016-08-01  9:10 ` [PATCH 39/73] drm/i915: Move the special case wait-request handling to its one caller Chris Wilson
2016-08-01  9:10 ` [PATCH 40/73] drm/i915: Disable waitboosting for a saturated engine Chris Wilson
2016-08-01  9:10 ` [PATCH 41/73] drm/i915: s/__i915_wait_request/i915_wait_request/ Chris Wilson
2016-08-01  9:10 ` [PATCH 42/73] drm/i915: Double check activity before relocations Chris Wilson
2016-08-01  9:10 ` [PATCH 43/73] drm/i915: Move request list retirement to i915_gem_request.c Chris Wilson
2016-08-01  9:10 ` [PATCH 44/73] drm/i915: i915_vma_move_to_active prep patch Chris Wilson
2016-08-01  9:10 ` [PATCH 45/73] drm/i915: Track active vma requests Chris Wilson
2016-08-01  9:10 ` [PATCH 46/73] drm/i915: Release vma when the handle is closed Chris Wilson
2016-08-01 11:26   ` Joonas Lahtinen
2016-08-01  9:10 ` [PATCH 47/73] drm/i915: Mark the context and address space as closed Chris Wilson
2016-08-01  9:10 ` [PATCH 48/73] Revert "drm/i915: Clean up associated VMAs on context destruction" Chris Wilson
2016-08-01  9:10 ` [PATCH 49/73] drm/i915: Combine loops within i915_gem_evict_something Chris Wilson
2016-08-01  9:10 ` [PATCH 50/73] drm/i915: Remove surplus drm_device parameter to i915_gem_evict_something() Chris Wilson
2016-08-01  9:10 ` [PATCH 51/73] drm/i915: Double check the active status on the batch pool Chris Wilson
2016-08-01  9:11 ` [PATCH 52/73] drm/i915: Remove request retirement before each batch Chris Wilson
2016-08-01  9:11 ` [PATCH 53/73] drm/i915: Remove i915_gem_execbuffer_retire_commands() Chris Wilson
2016-08-01  9:11 ` [PATCH 54/73] drm/i915: Fix up vma alignment to be u64 Chris Wilson
2016-08-01 12:21   ` Joonas Lahtinen
2016-08-01  9:11 ` [PATCH 55/73] drm/i915: Pad GTT views of exec objects up to user specified size Chris Wilson
2016-08-01  9:11 ` [PATCH 56/73] drm/i915: Reduce WARN(i915_gem_valid_gtt_space) to a debug-only check Chris Wilson
2016-08-01  9:11 ` [PATCH 57/73] drm/i915: Split insertion/binding of an object into the VM Chris Wilson
2016-08-01  9:11 ` [PATCH 58/73] drm/i915: Convert 4096 alignment request to 0 for drm_mm allocations Chris Wilson
2016-08-01  9:11 ` [PATCH 59/73] drm/i915: Update the GGTT size/alignment query functions Chris Wilson
2016-08-01 12:27   ` Joonas Lahtinen
2016-08-01  9:11 ` [PATCH 60/73] drm/i915: Update i915_gem_get_ggtt_size/_alignment to use drm_i915_private Chris Wilson
2016-08-01 12:30   ` Joonas Lahtinen
2016-08-01  9:11 ` [PATCH 61/73] drm/i915: Record allocated vma size Chris Wilson
2016-08-01 12:36   ` Joonas Lahtinen
2016-08-01 12:44     ` Chris Wilson
2016-08-01  9:11 ` [PATCH 62/73] drm/i915: Wrap vma->pin_count accessors with small inline helpers Chris Wilson
2016-08-01  9:11 ` [PATCH 63/73] drm/i915: Start passing around i915_vma from execbuffer Chris Wilson
2016-08-01  9:11 ` [PATCH 64/73] drm/i915: Combine all i915_vma bitfields into a single set of flags Chris Wilson
2016-08-01  9:11 ` [PATCH 65/73] drm/i915: Make i915_vma_pin() small and inline Chris Wilson
2016-08-01  9:11 ` [PATCH 66/73] drm/i915: Remove highly confusing i915_gem_obj_ggtt_pin() Chris Wilson
2016-08-01  9:11 ` [PATCH 67/73] drm/i915: Make fb_tracking.lock a spinlock Chris Wilson
2016-08-01  9:11 ` [PATCH 68/73] drm/i915: Use atomics to manipulate obj->frontbuffer_bits Chris Wilson
2016-08-01  9:11 ` [PATCH 69/73] drm/i915: Use dev_priv consistently through the intel_frontbuffer interface Chris Wilson
2016-08-01  9:11 ` [PATCH 70/73] drm/i915: Move obj->active:5 to obj->flags Chris Wilson
2016-08-01 12:46   ` Joonas Lahtinen [this message]
2016-08-01  9:11 ` [PATCH 71/73] drm/i915: Move i915_gem_object_wait_rendering() Chris Wilson
2016-08-01  9:11 ` [PATCH 72/73] drm/i915: Enable lockless lookup of request tracking via RCU Chris Wilson
2016-08-01  9:11 ` [PATCH 73/73] drm/i915: Export our request as a dma-buf fence on the reservation object Chris Wilson
2016-08-01 11:45 ` ✗ Ro.CI.BAT: failure for series starting with [01/73] drm/i915: Unify intel_logical_ring_emit and intel_ring_emit 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=1470055613.4174.17.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox