All of lore.kernel.org
 help / color / mirror / Atom feed
From: Matthew Auld <matthew.auld@intel.com>
To: "Thomas Hellström" <thomas.hellstrom@linux.intel.com>,
	intel-gfx@lists.freedesktop.org, dri-devel@lists.freedesktop.org
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Subject: Re: [PATCH v3 07/12] drm/i915/gt: Pipelined page migration
Date: Tue, 15 Jun 2021 15:45:10 +0100	[thread overview]
Message-ID: <cc1f996f-e8fe-7e74-5bc9-c7d61ebe8322@intel.com> (raw)
In-Reply-To: <20210614162612.294869-8-thomas.hellstrom@linux.intel.com>

On 14/06/2021 17:26, Thomas Hellström wrote:
> From: Chris Wilson <chris@chris-wilson.co.uk>
> 
> If we pipeline the PTE updates and then do the copy of those pages
> within a single unpreemptible command packet, we can submit the copies
> and leave them to be scheduled without having to synchronously wait
> under a global lock. In order to manage migration, we need to
> preallocate the page tables (and keep them pinned and available for use
> at any time), causing a bottleneck for migrations as all clients must
> contend on the limited resources. By inlining the ppGTT updates and
> performing the blit atomically, each client only owns the PTE while in
> use, and so we can reschedule individual operations however we see fit.
> And most importantly, we do not need to take a global lock on the shared
> vm, and wait until the operation is complete before releasing the lock
> for others to claim the PTE for themselves.
> 
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Co-developed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
> Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>


WARNING: multiple messages have this Message-ID (diff)
From: Matthew Auld <matthew.auld@intel.com>
To: "Thomas Hellström" <thomas.hellstrom@linux.intel.com>,
	intel-gfx@lists.freedesktop.org, dri-devel@lists.freedesktop.org
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Subject: Re: [Intel-gfx] [PATCH v3 07/12] drm/i915/gt: Pipelined page migration
Date: Tue, 15 Jun 2021 15:45:10 +0100	[thread overview]
Message-ID: <cc1f996f-e8fe-7e74-5bc9-c7d61ebe8322@intel.com> (raw)
In-Reply-To: <20210614162612.294869-8-thomas.hellstrom@linux.intel.com>

On 14/06/2021 17:26, Thomas Hellström wrote:
> From: Chris Wilson <chris@chris-wilson.co.uk>
> 
> If we pipeline the PTE updates and then do the copy of those pages
> within a single unpreemptible command packet, we can submit the copies
> and leave them to be scheduled without having to synchronously wait
> under a global lock. In order to manage migration, we need to
> preallocate the page tables (and keep them pinned and available for use
> at any time), causing a bottleneck for migrations as all clients must
> contend on the limited resources. By inlining the ppGTT updates and
> performing the blit atomically, each client only owns the PTE while in
> use, and so we can reschedule individual operations however we see fit.
> And most importantly, we do not need to take a global lock on the shared
> vm, and wait until the operation is complete before releasing the lock
> for others to claim the PTE for themselves.
> 
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Co-developed-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
> Signed-off-by: Thomas Hellström <thomas.hellstrom@linux.intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>

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

  reply	other threads:[~2021-06-15 14:45 UTC|newest]

Thread overview: 44+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-14 16:26 [PATCH v3 00/12] i915 TTM sync accelerated migration and clear Thomas Hellström
2021-06-14 16:26 ` [Intel-gfx] " Thomas Hellström
2021-06-14 16:26 ` [PATCH v3 01/12] drm/i915: Reference objects on the ww object list Thomas Hellström
2021-06-14 16:26   ` [Intel-gfx] " Thomas Hellström
2021-06-14 16:26 ` [PATCH v3 02/12] drm/i915: Break out dma_resv ww locking utilities to separate files Thomas Hellström
2021-06-14 16:26   ` [Intel-gfx] " Thomas Hellström
2021-06-14 16:26 ` [PATCH v3 03/12] drm/i915: Introduce a ww transaction helper Thomas Hellström
2021-06-14 16:26   ` [Intel-gfx] " Thomas Hellström
2021-06-14 16:26 ` [PATCH v3 04/12] drm/i915/gt: Add an insert_entry for gen8_ppgtt Thomas Hellström
2021-06-14 16:26   ` [Intel-gfx] " Thomas Hellström
2021-06-14 16:26 ` [PATCH v3 05/12] drm/i915/gt: Add a routine to iterate over the pagetables of a GTT Thomas Hellström
2021-06-14 16:26   ` [Intel-gfx] " Thomas Hellström
2021-06-14 16:26 ` [PATCH v3 06/12] drm/i915/gt: Export the pinned context constructor and destructor Thomas Hellström
2021-06-14 16:26   ` [Intel-gfx] " Thomas Hellström
2021-06-14 16:26 ` [PATCH v3 07/12] drm/i915/gt: Pipelined page migration Thomas Hellström
2021-06-14 16:26   ` [Intel-gfx] " Thomas Hellström
2021-06-15 14:45   ` Matthew Auld [this message]
2021-06-15 14:45     ` Matthew Auld
2021-06-14 16:26 ` [PATCH v3 08/12] drm/i915/gt: Pipelined clear Thomas Hellström
2021-06-14 16:26   ` [Intel-gfx] " Thomas Hellström
2021-06-15 14:47   ` Matthew Auld
2021-06-15 14:47     ` [Intel-gfx] " Matthew Auld
2021-06-14 16:26 ` [PATCH v3 09/12] drm/i915/gt: Setup a default migration context on the GT Thomas Hellström
2021-06-14 16:26   ` [Intel-gfx] " Thomas Hellström
2021-06-14 16:26 ` [PATCH v3 10/12] drm/i915/ttm: accelerated move implementation Thomas Hellström
2021-06-14 16:26   ` [Intel-gfx] " Thomas Hellström
2021-06-14 17:55   ` Thomas Hellström
2021-06-14 17:55     ` [Intel-gfx] " Thomas Hellström
2021-06-15 10:06     ` [PATCH v3] " Ramalingam C
2021-06-15 10:06       ` [Intel-gfx] " Ramalingam C
2021-06-14 16:26 ` [PATCH v3 11/12] drm/i915/gem: Zap the client blt code Thomas Hellström
2021-06-14 16:26   ` [Intel-gfx] " Thomas Hellström
2021-06-14 16:33   ` Matthew Auld
2021-06-14 16:33     ` [Intel-gfx] " Matthew Auld
2021-06-14 16:40     ` Thomas Hellström
2021-06-14 16:40       ` [Intel-gfx] " Thomas Hellström
2021-06-14 16:26 ` [PATCH v3 12/12] drm/i915/gem: Zap the i915_gem_object_blt code Thomas Hellström
2021-06-14 16:26   ` [Intel-gfx] " Thomas Hellström
2021-06-14 16:43   ` Matthew Auld
2021-06-14 16:43     ` [Intel-gfx] " Matthew Auld
2021-06-15  0:55 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for i915 TTM sync accelerated migration and clear Patchwork
2021-06-15  1:24 ` [Intel-gfx] ✓ Fi.CI.BAT: success " Patchwork
2021-06-15 10:19 ` [Intel-gfx] ✓ Fi.CI.IGT: " Patchwork
2021-06-15 11:45 ` [Intel-gfx] ✗ Fi.CI.BUILD: failure for i915 TTM sync accelerated migration and clear (rev2) 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=cc1f996f-e8fe-7e74-5bc9-c7d61ebe8322@intel.com \
    --to=matthew.auld@intel.com \
    --cc=chris@chris-wilson.co.uk \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=thomas.hellstrom@linux.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 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.