From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
To: "Thomas Hellström (Intel)" <thomas_os@shipmail.org>,
intel-gfx@lists.freedesktop.org
Subject: Re: [Intel-gfx] [PATCH 00/19] drm/i915: Short-term pinning and async eviction.
Date: Thu, 16 Sep 2021 13:24:41 +0200 [thread overview]
Message-ID: <0848953e-b161-1bf4-cbe4-b37154359bc1@linux.intel.com> (raw)
In-Reply-To: <1baadb2d-6636-1fe1-a1a2-34600947c5e7@shipmail.org>
Op 16-09-2021 om 11:40 schreef Thomas Hellström (Intel):
>
> On 8/30/21 2:09 PM, Maarten Lankhorst wrote:
>> Remove some parts of the i915_vma api, ensure obj->vma always exists,
>> and finally force the object lock to be taken when calling i915_vma_unbind
>> is called.
>
> Should this be vma->obj?
>
Correct, missed that.
>
>>
>> With this, locking is a lot cleaner, and we no longer need all the if (!obj->vma) checks.
>> We kill off the locking around i915_vma->set/get pages, and kill off the short-term
>> pinning/unpinning in execbuf.
>>
>> After this, we go 1 step further, populate obj->moving where required, and add support
>> for async eviction/clearing.
>
> Does the async eviction / clearing depend on the vma cleanup? If not can we move that into a separate series?
It kind of does. We cannot always require vma->obj to be locked and remove vma locks if we don't have vma->obj all the time.
prev parent reply other threads:[~2021-09-16 11:24 UTC|newest]
Thread overview: 54+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-08-30 12:09 [Intel-gfx] [PATCH 00/19] drm/i915: Short-term pinning and async eviction Maarten Lankhorst
2021-08-30 12:09 ` [Intel-gfx] [PATCH 01/19] drm/i915: Move __i915_gem_free_object to ttm_bo_destroy Maarten Lankhorst
2021-09-16 9:43 ` Thomas Hellström (Intel)
2021-09-16 13:35 ` Maarten Lankhorst
2021-08-30 12:09 ` [Intel-gfx] [PATCH 02/19] drm/i915: Remove unused bits of i915_vma/active api Maarten Lankhorst
2021-09-08 1:37 ` Niranjana Vishwanathapura
2021-08-30 12:09 ` [Intel-gfx] [PATCH 03/19] drm/i915: Slightly rework EXEC_OBJECT_CAPTURE handling Maarten Lankhorst
2021-09-08 1:49 ` Niranjana Vishwanathapura
2021-08-30 12:09 ` [Intel-gfx] [PATCH 04/19] drm/i915: Remove gen6_ppgtt_unpin_all Maarten Lankhorst
2021-09-29 8:07 ` Matthew Auld
2021-08-30 12:09 ` [Intel-gfx] [PATCH 05/19] drm/i915: Create a dummy object for gen6 ppgtt Maarten Lankhorst
2021-08-30 12:09 ` [Intel-gfx] [PATCH 06/19] drm/i915: Create a full object for mock_ring Maarten Lankhorst
2021-08-30 12:09 ` [Intel-gfx] [PATCH 07/19] drm/i915: vma is always backed by an object Maarten Lankhorst
2021-08-31 9:18 ` Tvrtko Ursulin
2021-08-31 9:34 ` Maarten Lankhorst
2021-08-31 10:29 ` Tvrtko Ursulin
2021-09-03 9:31 ` Maarten Lankhorst
2021-09-03 10:48 ` Tvrtko Ursulin
2021-09-16 13:41 ` Maarten Lankhorst
2021-09-16 14:30 ` Tvrtko Ursulin
2021-09-16 15:05 ` Tvrtko Ursulin
2021-08-30 12:09 ` [Intel-gfx] [PATCH 08/19] drm/i915: Fix runtime pm handling in i915_gem_shrink Maarten Lankhorst
2021-09-08 1:12 ` Niranjana Vishwanathapura
2021-09-29 8:37 ` Matthew Auld
2021-09-29 8:40 ` Matthew Auld
2021-09-29 8:53 ` Maarten Lankhorst
2021-08-30 12:09 ` [Intel-gfx] [PATCH 09/19] drm/i915: Change shrink ordering to use locking around unbinding Maarten Lankhorst
2021-09-08 1:04 ` Niranjana Vishwanathapura
2021-08-30 12:09 ` [Intel-gfx] [PATCH 10/19] Move CONTEXT_VALID_BIT check Maarten Lankhorst
2021-09-08 19:45 ` Niranjana Vishwanathapura
2021-08-30 12:09 ` [Intel-gfx] [PATCH 11/19] drm/i915: Remove resv from i915_vma Maarten Lankhorst
2021-09-08 1:10 ` Niranjana Vishwanathapura
2021-08-30 12:09 ` [Intel-gfx] [PATCH 12/19] drm/i915: Remove pages_mutex and intel_gtt->vma_ops.set/clear_pages members Maarten Lankhorst
2021-08-30 12:10 ` [Intel-gfx] [PATCH 13/19] drm/i915: Take object lock in i915_ggtt_pin if ww is not set Maarten Lankhorst
2021-09-08 3:11 ` Niranjana Vishwanathapura
2021-09-16 13:54 ` Maarten Lankhorst
2021-08-30 12:10 ` [Intel-gfx] [PATCH 14/19] drm/i915: Add i915_vma_unbind_unlocked, and take obj lock for i915_vma_unbind Maarten Lankhorst
2021-09-03 9:33 ` Maarten Lankhorst
2021-08-30 12:10 ` [Intel-gfx] [PATCH 15/19] drm/i915: Remove support for unlocked i915_vma unbind Maarten Lankhorst
2021-09-08 3:35 ` Niranjana Vishwanathapura
2021-08-30 12:10 ` [Intel-gfx] [PATCH 16/19] drm/i915: Remove short-term pins from execbuf Maarten Lankhorst
2021-08-30 12:10 ` [Intel-gfx] [PATCH 17/19] drm/i915: Add functions to set/get moving fence Maarten Lankhorst
2021-09-08 4:08 ` Niranjana Vishwanathapura
2021-09-16 13:49 ` Maarten Lankhorst
2021-09-16 9:48 ` Thomas Hellström
2021-08-30 12:10 ` [Intel-gfx] [PATCH 18/19] drm/i915: Add support for asynchronous moving fence waiting Maarten Lankhorst
2021-09-16 10:01 ` Thomas Hellström (Intel)
2021-08-30 12:10 ` [Intel-gfx] [PATCH 19/19] drm/i915: Add accelerated migration to ttm Maarten Lankhorst
2021-09-16 11:19 ` Thomas Hellström (Intel)
2021-08-30 13:03 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for drm/i915: Short-term pinning and async eviction Patchwork
2021-08-30 13:05 ` [Intel-gfx] ✗ Fi.CI.SPARSE: " Patchwork
2021-08-30 13:32 ` [Intel-gfx] ✗ Fi.CI.BAT: failure " Patchwork
2021-09-16 9:40 ` [Intel-gfx] [PATCH 00/19] " Thomas Hellström (Intel)
2021-09-16 11:24 ` Maarten Lankhorst [this message]
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=0848953e-b161-1bf4-cbe4-b37154359bc1@linux.intel.com \
--to=maarten.lankhorst@linux.intel.com \
--cc=intel-gfx@lists.freedesktop.org \
--cc=thomas_os@shipmail.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