From: Stephen Rothwell <sfr@canb•auug.org.au>
To: Dave Airlie <airlied@redhat•com>, DRI <dri-devel@lists•freedesktop.org>
Cc: Francois Dugast <francois.dugast@intel•com>,
Himal Prasad Ghimiray <himal.prasad.ghimiray@intel•com>,
Linux Kernel Mailing List <linux-kernel@vger•kernel.org>,
Linux Next Mailing List <linux-next@vger•kernel.org>,
Lucas De Marchi <lucas.demarchi@intel•com>,
Matthew Brost <matthew.brost@intel•com>
Subject: linux-next: manual merge of the drm tree with Linus' tree
Date: Wed, 21 May 2025 14:13:18 +1000 [thread overview]
Message-ID: <20250521141318.59a76e45@canb.auug.org.au> (raw)
[-- Attachment #1: Type: text/plain, Size: 3085 bytes --]
Hi all,
Today's linux-next merge of the drm tree got a conflict in:
drivers/gpu/drm/xe/xe_svm.c
between commit:
794f5493f518 ("drm/xe: Strict migration policy for atomic SVM faults")
from Linus' tree and commit:
80bcbdfc8cf5 ("drm/xe/svm: Add stats for SVM page faults")
from the drm tree.
I fixed it up (see below) and can carry the fix as necessary. This
is now fixed as far as linux-next is concerned, but any non trivial
conflicts should be mentioned to your upstream maintainer when your tree
is submitted for merging. You may also want to consider cooperating
with the maintainer of the conflicting tree to minimise any particularly
complex conflicts.
--
Cheers,
Stephen Rothwell
diff --cc drivers/gpu/drm/xe/xe_svm.c
index 975094c1a582,d25f02c8d7fc..000000000000
--- a/drivers/gpu/drm/xe/xe_svm.c
+++ b/drivers/gpu/drm/xe/xe_svm.c
@@@ -654,18 -650,12 +659,19 @@@ void xe_svm_fini(struct xe_vm *vm
}
static bool xe_svm_range_is_valid(struct xe_svm_range *range,
- struct xe_tile *tile)
+ struct xe_tile *tile,
+ bool devmem_only)
{
- return (range->tile_present & ~range->tile_invalidated) & BIT(tile->id);
+ /*
+ * Advisory only check whether the range currently has a valid mapping,
+ * READ_ONCE pairs with WRITE_ONCE in xe_pt.c
+ */
+ return ((READ_ONCE(range->tile_present) &
+ ~READ_ONCE(range->tile_invalidated)) & BIT(tile->id)) &&
+ (!devmem_only || xe_svm_range_in_vram(range));
}
+ #if IS_ENABLED(CONFIG_DRM_XE_DEVMEM_MIRROR)
static struct xe_vram_region *tile_to_vr(struct xe_tile *tile)
{
return &tile->mem.vram;
@@@ -727,37 -717,15 +733,45 @@@ unlock
return err;
}
+ #else
+ static int xe_svm_alloc_vram(struct xe_vm *vm, struct xe_tile *tile,
+ struct xe_svm_range *range,
+ const struct drm_gpusvm_ctx *ctx)
+ {
+ return -EOPNOTSUPP;
+ }
+ #endif
+static bool supports_4K_migration(struct xe_device *xe)
+{
+ if (xe->info.vram_flags & XE_VRAM_FLAGS_NEED64K)
+ return false;
+
+ return true;
+}
+
+static bool xe_svm_range_needs_migrate_to_vram(struct xe_svm_range *range,
+ struct xe_vma *vma)
+{
+ struct xe_vm *vm = range_to_vm(&range->base);
+ u64 range_size = xe_svm_range_size(range);
+
+ if (!range->base.flags.migrate_devmem)
+ return false;
+
+ if (xe_svm_range_in_vram(range)) {
+ drm_dbg(&vm->xe->drm, "Range is already in VRAM\n");
+ return false;
+ }
+
+ if (range_size <= SZ_64K && !supports_4K_migration(vm->xe)) {
+ drm_dbg(&vm->xe->drm, "Platform doesn't support SZ_4K range migration\n");
+ return false;
+ }
+
+ return true;
+}
+
/**
* xe_svm_handle_pagefault() - SVM handle page fault
* @vm: The VM.
@@@ -790,7 -754,7 +804,8 @@@ int xe_svm_handle_pagefault(struct xe_v
struct drm_gpusvm_range *r;
struct drm_exec exec;
struct dma_fence *fence;
+ int migrate_try_count = ctx.devmem_only ? 3 : 1;
+ struct xe_tile *tile = gt_to_tile(gt);
ktime_t end = 0;
int err;
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]
next reply other threads:[~2025-05-21 4:13 UTC|newest]
Thread overview: 159+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-05-21 4:13 Stephen Rothwell [this message]
-- strict thread matches above, loose matches on Subject: below --
2026-01-05 2:16 linux-next: manual merge of the drm tree with Linus' tree Stephen Rothwell
2026-01-05 7:07 ` Thomas Zimmermann
2026-01-06 21:35 ` Stephen Rothwell
2025-12-04 3:06 Stephen Rothwell
2025-12-04 7:35 ` Thomas Zimmermann
2025-06-06 2:08 Stephen Rothwell
2025-05-21 4:10 Stephen Rothwell
2025-03-24 3:06 Stephen Rothwell
2025-03-03 2:12 Stephen Rothwell
2024-11-11 2:04 Stephen Rothwell
2024-10-28 2:49 Stephen Rothwell
2024-09-16 4:15 Stephen Rothwell
2024-09-16 12:58 ` Tobias Jakobi
2024-09-17 20:24 ` Alex Deucher
2024-09-09 3:11 Stephen Rothwell
2024-03-04 1:08 Stephen Rothwell
2024-03-04 0:47 Stephen Rothwell
2024-01-08 0:14 Stephen Rothwell
2023-06-19 1:17 Stephen Rothwell
2022-11-30 22:57 Stephen Rothwell
2022-11-27 23:58 Stephen Rothwell
2022-11-29 9:07 ` Geert Uytterhoeven
2022-11-17 2:13 Stephen Rothwell
2022-09-19 0:58 Stephen Rothwell
2022-09-19 7:58 ` Geert Uytterhoeven
2022-09-19 14:23 ` Nathan Chancellor
2022-02-28 3:17 Stephen Rothwell
2021-10-29 0:48 Stephen Rothwell
2021-10-29 6:52 ` Joonas Lahtinen
2021-10-11 0:37 Stephen Rothwell
2021-08-23 2:41 Stephen Rothwell
2021-08-24 0:12 ` Stephen Rothwell
2021-08-12 1:20 Stephen Rothwell
2021-07-01 0:52 Stephen Rothwell
2021-03-29 2:14 Stephen Rothwell
2021-03-30 7:36 ` Geert Uytterhoeven
2021-03-30 23:41 ` Stephen Rothwell
2021-02-01 1:30 Stephen Rothwell
2021-02-14 22:07 ` Stephen Rothwell
2021-01-18 0:56 Stephen Rothwell
2020-10-02 3:42 Stephen Rothwell
2020-05-25 3:51 Stephen Rothwell
2020-03-23 0:50 Stephen Rothwell
2019-11-13 1:38 Stephen Rothwell
2019-11-13 1:13 Stephen Rothwell
2019-11-13 0:58 Stephen Rothwell
2019-11-13 0:50 Stephen Rothwell
2019-11-13 0:46 Stephen Rothwell
2019-11-13 0:40 Stephen Rothwell
2019-08-26 3:12 Stephen Rothwell
2019-06-11 2:19 Stephen Rothwell
2018-06-04 3:09 Stephen Rothwell
2018-03-26 3:38 Stephen Rothwell
2018-03-22 6:37 Stephen Rothwell
2018-03-26 3:45 ` Stephen Rothwell
2018-03-13 1:15 Stephen Rothwell
2018-02-18 23:10 Stephen Rothwell
2018-02-20 20:15 ` Rodrigo Vivi
2018-02-18 23:02 Stephen Rothwell
2018-01-15 1:08 Stephen Rothwell
2017-11-23 23:37 Stephen Rothwell
2017-10-30 18:36 Mark Brown
2017-10-30 18:26 Mark Brown
2017-08-28 2:32 Stephen Rothwell
2017-07-31 2:24 Stephen Rothwell
2017-07-24 2:06 Stephen Rothwell
2017-07-24 2:09 ` Stephen Rothwell
2017-06-05 2:49 Stephen Rothwell
2017-04-05 0:30 Stephen Rothwell
2017-02-23 23:17 Stephen Rothwell
2016-09-28 1:57 Stephen Rothwell
2016-09-20 2:42 Stephen Rothwell
2016-09-20 12:25 ` Philipp Zabel
2016-09-05 3:58 Stephen Rothwell
2016-09-05 3:51 Stephen Rothwell
2016-07-29 1:45 Stephen Rothwell
2016-05-09 1:28 Stephen Rothwell
2016-05-09 1:27 Stephen Rothwell
2016-03-17 0:45 Stephen Rothwell
2016-03-17 1:43 ` Luis R. Rodriguez
2016-02-29 2:59 Stephen Rothwell
2015-12-07 3:38 Stephen Rothwell
2015-11-01 8:11 Stephen Rothwell
2015-07-27 1:14 Stephen Rothwell
2015-06-21 3:50 Stephen Rothwell
2015-06-22 7:20 ` Christian König
2015-05-11 3:07 Stephen Rothwell
2015-04-07 5:22 Stephen Rothwell
2015-01-29 2:17 Stephen Rothwell
2015-01-29 8:19 ` Oded Gabbay
2015-01-22 2:22 Stephen Rothwell
2015-01-22 2:31 ` Dave Airlie
2015-01-22 2:57 ` Stephen Rothwell
2015-01-22 2:18 Stephen Rothwell
2015-01-22 2:15 Stephen Rothwell
2014-12-05 4:46 Stephen Rothwell
2014-12-05 4:39 Stephen Rothwell
2014-11-17 3:11 Stephen Rothwell
2014-11-17 9:47 ` Thierry Reding
2014-06-05 3:59 Stephen Rothwell
2014-06-05 3:54 Stephen Rothwell
2014-03-31 2:54 Stephen Rothwell
2014-03-24 2:01 Stephen Rothwell
2014-03-24 8:49 ` Jani Nikula
2013-12-19 2:09 Stephen Rothwell
2013-12-19 1:59 Stephen Rothwell
2013-11-07 2:21 Stephen Rothwell
2013-11-07 2:10 Stephen Rothwell
2013-10-28 5:58 Stephen Rothwell
2013-10-28 5:28 Stephen Rothwell
2013-10-28 5:11 Stephen Rothwell
2013-07-29 2:58 Stephen Rothwell
2013-07-26 3:01 Stephen Rothwell
2013-06-24 2:56 Stephen Rothwell
2013-06-17 3:09 Stephen Rothwell
2013-02-13 4:48 Stephen Rothwell
2012-11-30 1:22 Stephen Rothwell
2012-10-04 2:00 Stephen Rothwell
2012-10-04 1:54 Stephen Rothwell
2012-09-17 4:53 Stephen Rothwell
2012-08-24 2:20 Stephen Rothwell
2012-08-24 2:19 Stephen Rothwell
2012-08-24 7:13 ` Jani Nikula
2012-08-24 7:12 ` Dave Airlie
2012-08-24 20:29 ` Sedat Dilek
2012-07-19 1:38 Stephen Rothwell
2012-07-16 3:00 Stephen Rothwell
2012-06-29 3:30 Stephen Rothwell
2012-06-29 9:01 ` Daniel Vetter
2012-05-14 4:08 Stephen Rothwell
2012-05-07 3:45 Stephen Rothwell
2012-05-07 8:06 ` Daniel Vetter
2012-05-03 3:07 Stephen Rothwell
2012-04-13 3:38 Stephen Rothwell
2012-04-13 8:59 ` Daniel Vetter
2012-03-05 3:47 Stephen Rothwell
2011-12-28 2:53 Stephen Rothwell
2011-12-28 2:44 Stephen Rothwell
2011-12-21 3:18 Stephen Rothwell
2011-12-12 2:06 Stephen Rothwell
2011-03-09 3:41 Stephen Rothwell
2010-10-15 1:54 Stephen Rothwell
2010-10-03 23:30 Stephen Rothwell
2010-10-03 23:30 Stephen Rothwell
2010-04-28 3:58 Stephen Rothwell
2010-04-27 2:51 Stephen Rothwell
2010-04-12 5:06 Stephen Rothwell
2009-10-06 2:28 Stephen Rothwell
2009-10-08 2:35 ` Stephen Rothwell
2009-10-08 3:38 ` Dave Airlie
2009-10-08 3:44 ` Stephen Rothwell
2009-09-08 5:01 Stephen Rothwell
2009-08-20 5:27 Stephen Rothwell
2009-08-20 5:30 ` Dave Airlie
2009-08-20 6:43 ` Stephen Rothwell
2009-08-20 5:07 Stephen Rothwell
2009-08-05 3:44 Stephen Rothwell
2009-02-10 6:22 Stephen Rothwell
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=20250521141318.59a76e45@canb.auug.org.au \
--to=sfr@canb$(echo .)auug.org.au \
--cc=airlied@redhat$(echo .)com \
--cc=dri-devel@lists$(echo .)freedesktop.org \
--cc=francois.dugast@intel$(echo .)com \
--cc=himal.prasad.ghimiray@intel$(echo .)com \
--cc=linux-kernel@vger$(echo .)kernel.org \
--cc=linux-next@vger$(echo .)kernel.org \
--cc=lucas.demarchi@intel$(echo .)com \
--cc=matthew.brost@intel$(echo .)com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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