* [Intel-gfx] [PATCH 1/2] drm/i915/selftests: fixup igt_shrink_thp
@ 2021-07-29 13:34 Tvrtko Ursulin
2021-07-29 13:34 ` [Intel-gfx] [PATCH 2/2] drm/i915: Use Transparent Hugepages when IOMMU is enabled Tvrtko Ursulin
` (3 more replies)
0 siblings, 4 replies; 13+ messages in thread
From: Tvrtko Ursulin @ 2021-07-29 13:34 UTC (permalink / raw)
To: Intel-gfx; +Cc: Matthew Auld, dri-devel
From: Matthew Auld <matthew.auld@intel.com>
Since the object might still be active here, the shrink_all will simply
ignore it, which blows up in the test, since the pages will still be
there. Currently THP is disabled which should result in the test being
skipped, but if we ever re-enable THP we might start seeing the failure.
Fix this by forcing I915_SHRINK_ACTIVE.
Signed-off-by: Matthew Auld <matthew.auld@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
---
drivers/gpu/drm/i915/gem/selftests/huge_pages.c | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/i915/gem/selftests/huge_pages.c b/drivers/gpu/drm/i915/gem/selftests/huge_pages.c
index a094f3ce1a90..acc435f14ac9 100644
--- a/drivers/gpu/drm/i915/gem/selftests/huge_pages.c
+++ b/drivers/gpu/drm/i915/gem/selftests/huge_pages.c
@@ -1572,12 +1572,15 @@ static int igt_shrink_thp(void *arg)
goto out_put;
/*
- * Now that the pages are *unpinned* shrink-all should invoke
+ * Now that the pages are *unpinned* shrinking should invoke
* shmem to truncate our pages.
*/
- i915_gem_shrink_all(i915);
+ i915_gem_shrink(NULL, i915, -1UL, NULL,
+ I915_SHRINK_BOUND |
+ I915_SHRINK_UNBOUND |
+ I915_SHRINK_ACTIVE);
if (i915_gem_object_has_pages(obj)) {
- pr_err("shrink-all didn't truncate the pages\n");
+ pr_err("shrinking didn't truncate the pages\n");
err = -EINVAL;
goto out_put;
}
--
2.30.2
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
^ permalink raw reply related [flat|nested] 13+ messages in thread
* [Intel-gfx] [PATCH 2/2] drm/i915: Use Transparent Hugepages when IOMMU is enabled
2021-07-29 13:34 [Intel-gfx] [PATCH 1/2] drm/i915/selftests: fixup igt_shrink_thp Tvrtko Ursulin
@ 2021-07-29 13:34 ` Tvrtko Ursulin
2021-07-29 14:06 ` Daniel Vetter
2021-07-29 14:01 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for series starting with [1/2] drm/i915/selftests: fixup igt_shrink_thp Patchwork
` (2 subsequent siblings)
3 siblings, 1 reply; 13+ messages in thread
From: Tvrtko Ursulin @ 2021-07-29 13:34 UTC (permalink / raw)
To: Intel-gfx; +Cc: Eero Tamminen, dri-devel, Chris Wilson, Matthew Auld
From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Usage of Transparent Hugepages was disabled in 9987da4b5dcf
("drm/i915: Disable THP until we have a GPU read BW W/A"), but since it
appears majority of performance regressions reported with an enabled IOMMU
can be almost eliminated by turning them on, lets just do that.
To err on the side of safety we keep the current default in cases where
IOMMU is not active, and only when it is default to the "huge=within_size"
mode. Although there probably would be wins to enable them throughout,
more extensive testing across benchmarks and platforms would need to be
done.
With the patch and IOMMU enabled my local testing on a small Skylake part
shows OglVSTangent regression being reduced from ~14% (IOMMU on versus
IOMMU off) to ~2% (same comparison but with THP on).
v2:
* Add Kconfig dependency to transparent hugepages and some help text.
* Move to helper for easier handling of kernel build options.
v3:
* Drop Kconfig. (Daniel)
References: b901bb89324a ("drm/i915/gemfs: enable THP")
References: 9987da4b5dcf ("drm/i915: Disable THP until we have a GPU read BW W/A")
References: https://gitlab.freedesktop.org/drm/intel/-/issues/430
Co-developed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Eero Tamminen <eero.t.tamminen@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Cc: Daniel Vetter <daniel@ffwll.ch>
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> # v1
---
drivers/gpu/drm/i915/gem/i915_gemfs.c | 22 +++++++++++++++++++---
1 file changed, 19 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/i915/gem/i915_gemfs.c b/drivers/gpu/drm/i915/gem/i915_gemfs.c
index 5e6e8c91ab38..dbdbdc344d87 100644
--- a/drivers/gpu/drm/i915/gem/i915_gemfs.c
+++ b/drivers/gpu/drm/i915/gem/i915_gemfs.c
@@ -6,7 +6,6 @@
#include <linux/fs.h>
#include <linux/mount.h>
-#include <linux/pagemap.h>
#include "i915_drv.h"
#include "i915_gemfs.h"
@@ -15,6 +14,7 @@ int i915_gemfs_init(struct drm_i915_private *i915)
{
struct file_system_type *type;
struct vfsmount *gemfs;
+ char *opts;
type = get_fs_type("tmpfs");
if (!type)
@@ -26,10 +26,26 @@ int i915_gemfs_init(struct drm_i915_private *i915)
*
* One example, although it is probably better with a per-file
* control, is selecting huge page allocations ("huge=within_size").
- * Currently unused due to bandwidth issues (slow reads) on Broadwell+.
+ * However, we only do so to offset the overhead of iommu lookups
+ * due to bandwidth issues (slow reads) on Broadwell+.
*/
- gemfs = kern_mount(type);
+ opts = NULL;
+ if (intel_vtd_active()) {
+ if (IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE)) {
+ static char huge_opt[] = "huge=within_size"; /* r/w */
+
+ opts = huge_opt;
+ drm_info(&i915->drm,
+ "Transparent Hugepage mode '%s'\n",
+ opts);
+ } else {
+ drm_notice(&i915->drm,
+ "Transparent Hugepage support is recommended for optimal performance when IOMMU is enabled!\n");
+ }
+ }
+
+ gemfs = vfs_kern_mount(type, SB_KERNMOUNT, type->name, opts);
if (IS_ERR(gemfs))
return PTR_ERR(gemfs);
--
2.30.2
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
^ permalink raw reply related [flat|nested] 13+ messages in thread
* [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for series starting with [1/2] drm/i915/selftests: fixup igt_shrink_thp
2021-07-29 13:34 [Intel-gfx] [PATCH 1/2] drm/i915/selftests: fixup igt_shrink_thp Tvrtko Ursulin
2021-07-29 13:34 ` [Intel-gfx] [PATCH 2/2] drm/i915: Use Transparent Hugepages when IOMMU is enabled Tvrtko Ursulin
@ 2021-07-29 14:01 ` Patchwork
2021-07-29 14:29 ` [Intel-gfx] ✓ Fi.CI.BAT: success " Patchwork
2021-07-29 23:16 ` [Intel-gfx] ✓ Fi.CI.IGT: " Patchwork
3 siblings, 0 replies; 13+ messages in thread
From: Patchwork @ 2021-07-29 14:01 UTC (permalink / raw)
To: Tvrtko Ursulin; +Cc: intel-gfx
== Series Details ==
Series: series starting with [1/2] drm/i915/selftests: fixup igt_shrink_thp
URL : https://patchwork.freedesktop.org/series/93182/
State : warning
== Summary ==
$ dim checkpatch origin/drm-tip
1cac26e2f95f drm/i915/selftests: fixup igt_shrink_thp
054c89f72f0f drm/i915: Use Transparent Hugepages when IOMMU is enabled
-:6: ERROR:GIT_COMMIT_ID: Please use git commit description style 'commit <12+ chars of sha1> ("<title line>")' - ie: 'commit 9987da4b5dcf ("drm/i915: Disable THP until we have a GPU read BW W/A")'
#6:
Usage of Transparent Hugepages was disabled in 9987da4b5dcf
-:28: ERROR:GIT_COMMIT_ID: Please use git commit description style 'commit <12+ chars of sha1> ("<title line>")' - ie: 'commit b901bb89324a ("drm/i915/gemfs: enable THP")'
#28:
References: b901bb89324a ("drm/i915/gemfs: enable THP")
-:29: WARNING:COMMIT_LOG_LONG_LINE: Possible unwrapped commit description (prefer a maximum 75 chars per line)
#29:
References: 9987da4b5dcf ("drm/i915: Disable THP until we have a GPU read BW W/A")
-:29: ERROR:GIT_COMMIT_ID: Please use git commit description style 'commit <12+ chars of sha1> ("<title line>")' - ie: 'commit 9987da4b5dcf ("drm/i915: Disable THP until we have a GPU read BW W/A")'
#29:
References: 9987da4b5dcf ("drm/i915: Disable THP until we have a GPU read BW W/A")
-:75: WARNING:STATIC_CONST_CHAR_ARRAY: static char array declaration should probably be static const char
#75: FILE: drivers/gpu/drm/i915/gem/i915_gemfs.c:36:
+ static char huge_opt[] = "huge=within_size"; /* r/w */
total: 3 errors, 2 warnings, 0 checks, 42 lines checked
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [Intel-gfx] [PATCH 2/2] drm/i915: Use Transparent Hugepages when IOMMU is enabled
2021-07-29 13:34 ` [Intel-gfx] [PATCH 2/2] drm/i915: Use Transparent Hugepages when IOMMU is enabled Tvrtko Ursulin
@ 2021-07-29 14:06 ` Daniel Vetter
2021-09-03 12:47 ` Tvrtko Ursulin
0 siblings, 1 reply; 13+ messages in thread
From: Daniel Vetter @ 2021-07-29 14:06 UTC (permalink / raw)
To: Tvrtko Ursulin
Cc: intel-gfx, dri-devel, Chris Wilson, Matthew Auld, Eero Tamminen
On Thu, Jul 29, 2021 at 3:34 PM Tvrtko Ursulin
<tvrtko.ursulin@linux.intel.com> wrote:
>
> From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>
> Usage of Transparent Hugepages was disabled in 9987da4b5dcf
> ("drm/i915: Disable THP until we have a GPU read BW W/A"), but since it
> appears majority of performance regressions reported with an enabled IOMMU
> can be almost eliminated by turning them on, lets just do that.
>
> To err on the side of safety we keep the current default in cases where
> IOMMU is not active, and only when it is default to the "huge=within_size"
> mode. Although there probably would be wins to enable them throughout,
> more extensive testing across benchmarks and platforms would need to be
> done.
>
> With the patch and IOMMU enabled my local testing on a small Skylake part
> shows OglVSTangent regression being reduced from ~14% (IOMMU on versus
> IOMMU off) to ~2% (same comparison but with THP on).
>
> v2:
> * Add Kconfig dependency to transparent hugepages and some help text.
> * Move to helper for easier handling of kernel build options.
>
> v3:
> * Drop Kconfig. (Daniel)
>
> References: b901bb89324a ("drm/i915/gemfs: enable THP")
> References: 9987da4b5dcf ("drm/i915: Disable THP until we have a GPU read BW W/A")
> References: https://gitlab.freedesktop.org/drm/intel/-/issues/430
> Co-developed-by: Chris Wilson <chris@chris-wilson.co.uk>
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
> Cc: Matthew Auld <matthew.auld@intel.com>
> Cc: Eero Tamminen <eero.t.tamminen@intel.com>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
> Cc: Daniel Vetter <daniel@ffwll.ch>
> Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> # v1
On both patches: Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
> ---
> drivers/gpu/drm/i915/gem/i915_gemfs.c | 22 +++++++++++++++++++---
> 1 file changed, 19 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/gem/i915_gemfs.c b/drivers/gpu/drm/i915/gem/i915_gemfs.c
> index 5e6e8c91ab38..dbdbdc344d87 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gemfs.c
> +++ b/drivers/gpu/drm/i915/gem/i915_gemfs.c
> @@ -6,7 +6,6 @@
>
> #include <linux/fs.h>
> #include <linux/mount.h>
> -#include <linux/pagemap.h>
>
> #include "i915_drv.h"
> #include "i915_gemfs.h"
> @@ -15,6 +14,7 @@ int i915_gemfs_init(struct drm_i915_private *i915)
> {
> struct file_system_type *type;
> struct vfsmount *gemfs;
> + char *opts;
>
> type = get_fs_type("tmpfs");
> if (!type)
> @@ -26,10 +26,26 @@ int i915_gemfs_init(struct drm_i915_private *i915)
> *
> * One example, although it is probably better with a per-file
> * control, is selecting huge page allocations ("huge=within_size").
> - * Currently unused due to bandwidth issues (slow reads) on Broadwell+.
> + * However, we only do so to offset the overhead of iommu lookups
> + * due to bandwidth issues (slow reads) on Broadwell+.
> */
>
> - gemfs = kern_mount(type);
> + opts = NULL;
> + if (intel_vtd_active()) {
> + if (IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE)) {
> + static char huge_opt[] = "huge=within_size"; /* r/w */
> +
> + opts = huge_opt;
> + drm_info(&i915->drm,
> + "Transparent Hugepage mode '%s'\n",
> + opts);
> + } else {
> + drm_notice(&i915->drm,
> + "Transparent Hugepage support is recommended for optimal performance when IOMMU is enabled!\n");
> + }
> + }
> +
> + gemfs = vfs_kern_mount(type, SB_KERNMOUNT, type->name, opts);
> if (IS_ERR(gemfs))
> return PTR_ERR(gemfs);
>
> --
> 2.30.2
>
--
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
^ permalink raw reply [flat|nested] 13+ messages in thread
* [Intel-gfx] ✓ Fi.CI.BAT: success for series starting with [1/2] drm/i915/selftests: fixup igt_shrink_thp
2021-07-29 13:34 [Intel-gfx] [PATCH 1/2] drm/i915/selftests: fixup igt_shrink_thp Tvrtko Ursulin
2021-07-29 13:34 ` [Intel-gfx] [PATCH 2/2] drm/i915: Use Transparent Hugepages when IOMMU is enabled Tvrtko Ursulin
2021-07-29 14:01 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for series starting with [1/2] drm/i915/selftests: fixup igt_shrink_thp Patchwork
@ 2021-07-29 14:29 ` Patchwork
2021-07-29 23:16 ` [Intel-gfx] ✓ Fi.CI.IGT: " Patchwork
3 siblings, 0 replies; 13+ messages in thread
From: Patchwork @ 2021-07-29 14:29 UTC (permalink / raw)
To: Tvrtko Ursulin; +Cc: intel-gfx
[-- Attachment #1.1: Type: text/plain, Size: 3722 bytes --]
== Series Details ==
Series: series starting with [1/2] drm/i915/selftests: fixup igt_shrink_thp
URL : https://patchwork.freedesktop.org/series/93182/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_10418 -> Patchwork_20741
====================================================
Summary
-------
**SUCCESS**
No regressions found.
External URL: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/index.html
Known issues
------------
Here are the changes found in Patchwork_20741 that come from known issues:
### IGT changes ###
#### Issues hit ####
* igt@amdgpu/amd_basic@semaphore:
- fi-bdw-5557u: NOTRUN -> [SKIP][1] ([fdo#109271]) +29 similar issues
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/fi-bdw-5557u/igt@amdgpu/amd_basic@semaphore.html
* igt@core_hotunplug@unbind-rebind:
- fi-bdw-5557u: NOTRUN -> [WARN][2] ([i915#3718])
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/fi-bdw-5557u/igt@core_hotunplug@unbind-rebind.html
* igt@i915_pm_rpm@basic-rte:
- fi-bdw-5557u: NOTRUN -> [FAIL][3] ([i915#579])
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/fi-bdw-5557u/igt@i915_pm_rpm@basic-rte.html
* igt@kms_chamelium@dp-crc-fast:
- fi-bdw-5557u: NOTRUN -> [SKIP][4] ([fdo#109271] / [fdo#111827]) +8 similar issues
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/fi-bdw-5557u/igt@kms_chamelium@dp-crc-fast.html
#### Possible fixes ####
* igt@gem_exec_suspend@basic-s0:
- fi-tgl-1115g4: [FAIL][5] ([i915#1888]) -> [PASS][6]
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/fi-tgl-1115g4/igt@gem_exec_suspend@basic-s0.html
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/fi-tgl-1115g4/igt@gem_exec_suspend@basic-s0.html
* igt@kms_chamelium@common-hpd-after-suspend:
- fi-kbl-7500u: [DMESG-FAIL][7] ([i915#165]) -> [PASS][8]
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/fi-kbl-7500u/igt@kms_chamelium@common-hpd-after-suspend.html
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/fi-kbl-7500u/igt@kms_chamelium@common-hpd-after-suspend.html
{name}: This element is suppressed. This means it is ignored when computing
the status of the difference (SUCCESS, WARNING, or FAILURE).
[fdo#109271]: https://bugs.freedesktop.org/show_bug.cgi?id=109271
[fdo#111827]: https://bugs.freedesktop.org/show_bug.cgi?id=111827
[i915#165]: https://gitlab.freedesktop.org/drm/intel/issues/165
[i915#1888]: https://gitlab.freedesktop.org/drm/intel/issues/1888
[i915#1982]: https://gitlab.freedesktop.org/drm/intel/issues/1982
[i915#3303]: https://gitlab.freedesktop.org/drm/intel/issues/3303
[i915#3718]: https://gitlab.freedesktop.org/drm/intel/issues/3718
[i915#579]: https://gitlab.freedesktop.org/drm/intel/issues/579
Participating hosts (41 -> 34)
------------------------------
Missing (7): fi-ilk-m540 fi-hsw-4200u fi-bsw-cyan bat-adlp-4 fi-ctg-p8600 fi-bdw-samus bat-jsl-1
Build changes
-------------
* Linux: CI_DRM_10418 -> Patchwork_20741
CI-20190529: 20190529
CI_DRM_10418: e8b9eb4ffe9ad59eeda3c5318282fc2d7acbe7da @ git://anongit.freedesktop.org/gfx-ci/linux
IGT_6156: 354638455597ac48ac433606b24b82a4d7b65d20 @ https://gitlab.freedesktop.org/drm/igt-gpu-tools.git
Patchwork_20741: 054c89f72f0f9f04e8377c7f9f5c66a6a7d047f8 @ git://anongit.freedesktop.org/gfx-ci/linux
== Linux commits ==
054c89f72f0f drm/i915: Use Transparent Hugepages when IOMMU is enabled
1cac26e2f95f drm/i915/selftests: fixup igt_shrink_thp
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/index.html
[-- Attachment #1.2: Type: text/html, Size: 4414 bytes --]
[-- Attachment #2: Type: text/plain, Size: 160 bytes --]
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
^ permalink raw reply [flat|nested] 13+ messages in thread
* [Intel-gfx] ✓ Fi.CI.IGT: success for series starting with [1/2] drm/i915/selftests: fixup igt_shrink_thp
2021-07-29 13:34 [Intel-gfx] [PATCH 1/2] drm/i915/selftests: fixup igt_shrink_thp Tvrtko Ursulin
` (2 preceding siblings ...)
2021-07-29 14:29 ` [Intel-gfx] ✓ Fi.CI.BAT: success " Patchwork
@ 2021-07-29 23:16 ` Patchwork
3 siblings, 0 replies; 13+ messages in thread
From: Patchwork @ 2021-07-29 23:16 UTC (permalink / raw)
To: Tvrtko Ursulin; +Cc: intel-gfx
[-- Attachment #1.1: Type: text/plain, Size: 30290 bytes --]
== Series Details ==
Series: series starting with [1/2] drm/i915/selftests: fixup igt_shrink_thp
URL : https://patchwork.freedesktop.org/series/93182/
State : success
== Summary ==
CI Bug Log - changes from CI_DRM_10418_full -> Patchwork_20741_full
====================================================
Summary
-------
**SUCCESS**
No regressions found.
Possible new issues
-------------------
Here are the unknown changes that may have been introduced in Patchwork_20741_full:
### IGT changes ###
#### Suppressed ####
The following results come from untrusted machines, tests, or statuses.
They do not affect the overall result.
* igt@gen7_exec_parse@basic-allowed:
- {shard-rkl}: [SKIP][1] ([fdo#109289]) -> [TIMEOUT][2]
[1]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-1/igt@gen7_exec_parse@basic-allowed.html
[2]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-2/igt@gen7_exec_parse@basic-allowed.html
* igt@kms_busy@extended-pageflip-hang-newfb:
- {shard-rkl}: NOTRUN -> [SKIP][3]
[3]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-5/igt@kms_busy@extended-pageflip-hang-newfb.html
* igt@kms_cursor_crc@pipe-a-cursor-64x21-random:
- {shard-rkl}: [SKIP][4] ([fdo#112022]) -> [TIMEOUT][5]
[4]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-1/igt@kms_cursor_crc@pipe-a-cursor-64x21-random.html
[5]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-2/igt@kms_cursor_crc@pipe-a-cursor-64x21-random.html
* igt@kms_draw_crc@draw-method-rgb565-blt-ytiled:
- {shard-rkl}: [SKIP][6] ([fdo#111314]) -> [DMESG-WARN][7]
[6]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-5/igt@kms_draw_crc@draw-method-rgb565-blt-ytiled.html
[7]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-6/igt@kms_draw_crc@draw-method-rgb565-blt-ytiled.html
* igt@kms_frontbuffer_tracking@psr-1p-primscrn-shrfb-plflip-blt:
- {shard-rkl}: [SKIP][8] ([i915#1849]) -> [TIMEOUT][9] +1 similar issue
[8]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-1/igt@kms_frontbuffer_tracking@psr-1p-primscrn-shrfb-plflip-blt.html
[9]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-2/igt@kms_frontbuffer_tracking@psr-1p-primscrn-shrfb-plflip-blt.html
* igt@runner@aborted:
- {shard-rkl}: ([FAIL][10], [FAIL][11], [FAIL][12], [FAIL][13], [FAIL][14]) ([i915#3002] / [i915#3810] / [i915#3811] / [i915#3871]) -> ([FAIL][15], [FAIL][16], [FAIL][17], [FAIL][18], [FAIL][19]) ([i915#2029] / [i915#3002] / [i915#3811] / [i915#3871])
[10]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-2/igt@runner@aborted.html
[11]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-2/igt@runner@aborted.html
[12]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-1/igt@runner@aborted.html
[13]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-1/igt@runner@aborted.html
[14]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-6/igt@runner@aborted.html
[15]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-1/igt@runner@aborted.html
[16]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-6/igt@runner@aborted.html
[17]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-6/igt@runner@aborted.html
[18]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-5/igt@runner@aborted.html
[19]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-6/igt@runner@aborted.html
Known issues
------------
Here are the changes found in Patchwork_20741_full that come from known issues:
### IGT changes ###
#### Issues hit ####
* igt@gem_ctx_persistence@legacy-engines-hostile-preempt:
- shard-snb: NOTRUN -> [SKIP][20] ([fdo#109271] / [i915#1099]) +2 similar issues
[20]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-snb7/igt@gem_ctx_persistence@legacy-engines-hostile-preempt.html
* igt@gem_exec_fair@basic-pace@bcs0:
- shard-iclb: [PASS][21] -> [FAIL][22] ([i915#2842])
[21]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-iclb7/igt@gem_exec_fair@basic-pace@bcs0.html
[22]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-iclb1/igt@gem_exec_fair@basic-pace@bcs0.html
* igt@gem_exec_fair@basic-pace@vcs1:
- shard-iclb: NOTRUN -> [FAIL][23] ([i915#2842])
[23]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-iclb1/igt@gem_exec_fair@basic-pace@vcs1.html
* igt@gem_exec_fair@basic-throttle@rcs0:
- shard-glk: [PASS][24] -> [FAIL][25] ([i915#2842])
[24]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-glk7/igt@gem_exec_fair@basic-throttle@rcs0.html
[25]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-glk9/igt@gem_exec_fair@basic-throttle@rcs0.html
* igt@gem_exec_schedule@semaphore-codependency:
- shard-snb: NOTRUN -> [SKIP][26] ([fdo#109271]) +137 similar issues
[26]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-snb5/igt@gem_exec_schedule@semaphore-codependency.html
* igt@gem_huc_copy@huc-copy:
- shard-apl: NOTRUN -> [SKIP][27] ([fdo#109271] / [i915#2190])
[27]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-apl2/igt@gem_huc_copy@huc-copy.html
- shard-kbl: NOTRUN -> [SKIP][28] ([fdo#109271] / [i915#2190])
[28]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-kbl3/igt@gem_huc_copy@huc-copy.html
* igt@gem_pread@exhaustion:
- shard-apl: NOTRUN -> [WARN][29] ([i915#2658]) +1 similar issue
[29]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-apl2/igt@gem_pread@exhaustion.html
* igt@gem_render_copy@x-tiled-to-vebox-yf-tiled:
- shard-kbl: NOTRUN -> [SKIP][30] ([fdo#109271]) +114 similar issues
[30]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-kbl3/igt@gem_render_copy@x-tiled-to-vebox-yf-tiled.html
* igt@gem_softpin@noreloc-s3:
- shard-apl: [PASS][31] -> [DMESG-WARN][32] ([i915#180])
[31]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-apl1/igt@gem_softpin@noreloc-s3.html
[32]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-apl3/igt@gem_softpin@noreloc-s3.html
* igt@i915_pm_lpsp@kms-lpsp@kms-lpsp-dp:
- shard-apl: NOTRUN -> [SKIP][33] ([fdo#109271] / [i915#1937])
[33]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-apl8/igt@i915_pm_lpsp@kms-lpsp@kms-lpsp-dp.html
* igt@i915_pm_rpm@basic-rte:
- shard-apl: NOTRUN -> [FAIL][34] ([i915#579])
[34]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-apl6/igt@i915_pm_rpm@basic-rte.html
* igt@i915_suspend@sysfs-reader:
- shard-apl: NOTRUN -> [DMESG-WARN][35] ([i915#180])
[35]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-apl6/igt@i915_suspend@sysfs-reader.html
* igt@kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-180-async-flip:
- shard-skl: NOTRUN -> [FAIL][36] ([i915#3722])
[36]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-skl7/igt@kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-180-async-flip.html
* igt@kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-180-hflip:
- shard-apl: NOTRUN -> [SKIP][37] ([fdo#109271] / [i915#3777]) +1 similar issue
[37]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-apl6/igt@kms_big_fb@y-tiled-max-hw-stride-32bpp-rotate-180-hflip.html
* igt@kms_big_fb@yf-tiled-max-hw-stride-64bpp-rotate-0:
- shard-apl: NOTRUN -> [SKIP][38] ([fdo#109271]) +251 similar issues
[38]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-apl8/igt@kms_big_fb@yf-tiled-max-hw-stride-64bpp-rotate-0.html
* igt@kms_chamelium@hdmi-hpd-enable-disable-mode:
- shard-snb: NOTRUN -> [SKIP][39] ([fdo#109271] / [fdo#111827]) +8 similar issues
[39]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-snb5/igt@kms_chamelium@hdmi-hpd-enable-disable-mode.html
* igt@kms_chamelium@hdmi-mode-timings:
- shard-kbl: NOTRUN -> [SKIP][40] ([fdo#109271] / [fdo#111827]) +4 similar issues
[40]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-kbl4/igt@kms_chamelium@hdmi-mode-timings.html
* igt@kms_chamelium@vga-hpd:
- shard-apl: NOTRUN -> [SKIP][41] ([fdo#109271] / [fdo#111827]) +22 similar issues
[41]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-apl2/igt@kms_chamelium@vga-hpd.html
* igt@kms_color@pipe-a-ctm-red-to-blue:
- shard-skl: [PASS][42] -> [DMESG-WARN][43] ([i915#1982]) +1 similar issue
[42]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-skl6/igt@kms_color@pipe-a-ctm-red-to-blue.html
[43]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-skl5/igt@kms_color@pipe-a-ctm-red-to-blue.html
* igt@kms_color_chamelium@pipe-invalid-degamma-lut-sizes:
- shard-skl: NOTRUN -> [SKIP][44] ([fdo#109271] / [fdo#111827]) +3 similar issues
[44]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-skl10/igt@kms_color_chamelium@pipe-invalid-degamma-lut-sizes.html
* igt@kms_content_protection@srm:
- shard-kbl: NOTRUN -> [TIMEOUT][45] ([i915#1319])
[45]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-kbl3/igt@kms_content_protection@srm.html
* igt@kms_cursor_legacy@pipe-d-torture-bo:
- shard-apl: NOTRUN -> [SKIP][46] ([fdo#109271] / [i915#533]) +1 similar issue
[46]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-apl2/igt@kms_cursor_legacy@pipe-d-torture-bo.html
* igt@kms_dither@fb-8bpc-vs-panel-8bpc@edp-1-pipe-a:
- shard-iclb: [PASS][47] -> [SKIP][48] ([i915#3788])
[47]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-iclb3/igt@kms_dither@fb-8bpc-vs-panel-8bpc@edp-1-pipe-a.html
[48]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-iclb2/igt@kms_dither@fb-8bpc-vs-panel-8bpc@edp-1-pipe-a.html
* igt@kms_flip@2x-plain-flip-fb-recreate-interruptible:
- shard-skl: NOTRUN -> [SKIP][49] ([fdo#109271]) +77 similar issues
[49]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-skl10/igt@kms_flip@2x-plain-flip-fb-recreate-interruptible.html
* igt@kms_flip@flip-vs-suspend-interruptible@a-dp1:
- shard-kbl: [PASS][50] -> [DMESG-WARN][51] ([i915#180]) +7 similar issues
[50]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-kbl3/igt@kms_flip@flip-vs-suspend-interruptible@a-dp1.html
[51]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-kbl7/igt@kms_flip@flip-vs-suspend-interruptible@a-dp1.html
* igt@kms_flip@plain-flip-ts-check@a-edp1:
- shard-skl: [PASS][52] -> [FAIL][53] ([i915#2122])
[52]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-skl8/igt@kms_flip@plain-flip-ts-check@a-edp1.html
[53]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-skl2/igt@kms_flip@plain-flip-ts-check@a-edp1.html
* igt@kms_pipe_crc_basic@compare-crc-sanitycheck-pipe-d:
- shard-skl: NOTRUN -> [SKIP][54] ([fdo#109271] / [i915#533])
[54]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-skl7/igt@kms_pipe_crc_basic@compare-crc-sanitycheck-pipe-d.html
* igt@kms_pipe_crc_basic@hang-read-crc-pipe-d:
- shard-kbl: NOTRUN -> [SKIP][55] ([fdo#109271] / [i915#533])
[55]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-kbl3/igt@kms_pipe_crc_basic@hang-read-crc-pipe-d.html
* igt@kms_pipe_crc_basic@suspend-read-crc-pipe-b:
- shard-skl: [PASS][56] -> [FAIL][57] ([i915#53])
[56]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-skl10/igt@kms_pipe_crc_basic@suspend-read-crc-pipe-b.html
[57]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-skl8/igt@kms_pipe_crc_basic@suspend-read-crc-pipe-b.html
* igt@kms_plane_alpha_blend@pipe-a-alpha-basic:
- shard-apl: NOTRUN -> [FAIL][58] ([fdo#108145] / [i915#265]) +2 similar issues
[58]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-apl2/igt@kms_plane_alpha_blend@pipe-a-alpha-basic.html
* igt@kms_plane_alpha_blend@pipe-b-alpha-transparent-fb:
- shard-skl: NOTRUN -> [FAIL][59] ([i915#265])
[59]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-skl10/igt@kms_plane_alpha_blend@pipe-b-alpha-transparent-fb.html
* igt@kms_plane_alpha_blend@pipe-c-coverage-7efc:
- shard-skl: [PASS][60] -> [FAIL][61] ([fdo#108145] / [i915#265]) +1 similar issue
[60]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-skl4/igt@kms_plane_alpha_blend@pipe-c-coverage-7efc.html
[61]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-skl3/igt@kms_plane_alpha_blend@pipe-c-coverage-7efc.html
* igt@kms_plane_scaling@scaler-with-clipping-clamping@pipe-c-scaler-with-clipping-clamping:
- shard-apl: NOTRUN -> [SKIP][62] ([fdo#109271] / [i915#2733])
[62]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-apl2/igt@kms_plane_scaling@scaler-with-clipping-clamping@pipe-c-scaler-with-clipping-clamping.html
* igt@kms_psr2_sf@cursor-plane-update-sf:
- shard-skl: NOTRUN -> [SKIP][63] ([fdo#109271] / [i915#658]) +1 similar issue
[63]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-skl7/igt@kms_psr2_sf@cursor-plane-update-sf.html
* igt@kms_psr2_sf@overlay-plane-update-sf-dmg-area-1:
- shard-kbl: NOTRUN -> [SKIP][64] ([fdo#109271] / [i915#658]) +1 similar issue
[64]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-kbl1/igt@kms_psr2_sf@overlay-plane-update-sf-dmg-area-1.html
* igt@kms_psr2_sf@overlay-primary-update-sf-dmg-area-3:
- shard-apl: NOTRUN -> [SKIP][65] ([fdo#109271] / [i915#658]) +3 similar issues
[65]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-apl2/igt@kms_psr2_sf@overlay-primary-update-sf-dmg-area-3.html
* igt@kms_psr2_su@page_flip:
- shard-iclb: [PASS][66] -> [SKIP][67] ([fdo#109642] / [fdo#111068] / [i915#658])
[66]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-iclb2/igt@kms_psr2_su@page_flip.html
[67]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-iclb4/igt@kms_psr2_su@page_flip.html
* igt@kms_psr@psr2_primary_page_flip:
- shard-iclb: [PASS][68] -> [SKIP][69] ([fdo#109441]) +1 similar issue
[68]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-iclb2/igt@kms_psr@psr2_primary_page_flip.html
[69]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-iclb4/igt@kms_psr@psr2_primary_page_flip.html
* igt@kms_selftest@all@damage_iter_no_damage:
- shard-snb: NOTRUN -> [INCOMPLETE][70] ([i915#3871])
[70]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-snb5/igt@kms_selftest@all@damage_iter_no_damage.html
* igt@kms_sysfs_edid_timing:
- shard-skl: NOTRUN -> [FAIL][71] ([IGT#2])
[71]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-skl7/igt@kms_sysfs_edid_timing.html
* igt@kms_vblank@pipe-b-ts-continuation-suspend:
- shard-kbl: [PASS][72] -> [INCOMPLETE][73] ([i915#155] / [i915#2828])
[72]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-kbl6/igt@kms_vblank@pipe-b-ts-continuation-suspend.html
[73]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-kbl2/igt@kms_vblank@pipe-b-ts-continuation-suspend.html
* igt@perf@polling:
- shard-skl: [PASS][74] -> [FAIL][75] ([i915#1542])
[74]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-skl8/igt@perf@polling.html
[75]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-skl9/igt@perf@polling.html
* igt@sysfs_clients@sema-10:
- shard-kbl: NOTRUN -> [SKIP][76] ([fdo#109271] / [i915#2994]) +1 similar issue
[76]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-kbl3/igt@sysfs_clients@sema-10.html
- shard-apl: NOTRUN -> [SKIP][77] ([fdo#109271] / [i915#2994]) +1 similar issue
[77]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-apl2/igt@sysfs_clients@sema-10.html
* igt@sysfs_clients@sema-25:
- shard-skl: NOTRUN -> [SKIP][78] ([fdo#109271] / [i915#2994])
[78]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-skl10/igt@sysfs_clients@sema-25.html
#### Possible fixes ####
* igt@fbdev@eof:
- {shard-rkl}: [SKIP][79] ([i915#2582]) -> [PASS][80]
[79]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-1/igt@fbdev@eof.html
[80]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-6/igt@fbdev@eof.html
* igt@gem_ctx_persistence@legacy-engines-hang@render:
- {shard-rkl}: [FAIL][81] ([i915#2410]) -> [PASS][82]
[81]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-5/igt@gem_ctx_persistence@legacy-engines-hang@render.html
[82]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-1/igt@gem_ctx_persistence@legacy-engines-hang@render.html
* igt@gem_eio@hibernate:
- {shard-rkl}: [INCOMPLETE][83] ([i915#3833]) -> [PASS][84]
[83]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-1/igt@gem_eio@hibernate.html
[84]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-1/igt@gem_eio@hibernate.html
* igt@gem_eio@unwedge-stress:
- shard-tglb: [TIMEOUT][85] ([i915#2369] / [i915#3063] / [i915#3648]) -> [PASS][86]
[85]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-tglb2/igt@gem_eio@unwedge-stress.html
[86]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-tglb8/igt@gem_eio@unwedge-stress.html
- shard-iclb: [TIMEOUT][87] ([i915#2369] / [i915#2481] / [i915#3070]) -> [PASS][88]
[87]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-iclb7/igt@gem_eio@unwedge-stress.html
[88]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-iclb8/igt@gem_eio@unwedge-stress.html
* igt@gem_exec_fair@basic-none-rrul@rcs0:
- shard-glk: [FAIL][89] ([i915#2842]) -> [PASS][90] +1 similar issue
[89]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-glk7/igt@gem_exec_fair@basic-none-rrul@rcs0.html
[90]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-glk5/igt@gem_exec_fair@basic-none-rrul@rcs0.html
* igt@gem_exec_fair@basic-pace-share@rcs0:
- shard-tglb: [FAIL][91] ([i915#2842]) -> [PASS][92] +2 similar issues
[91]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-tglb1/igt@gem_exec_fair@basic-pace-share@rcs0.html
[92]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-tglb2/igt@gem_exec_fair@basic-pace-share@rcs0.html
* igt@gem_exec_suspend@basic-s0:
- {shard-rkl}: [FAIL][93] ([fdo#103375]) -> [PASS][94] +1 similar issue
[93]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-6/igt@gem_exec_suspend@basic-s0.html
[94]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-5/igt@gem_exec_suspend@basic-s0.html
* igt@gem_mmap_gtt@cpuset-big-copy:
- {shard-rkl}: [FAIL][95] ([i915#307]) -> [PASS][96]
[95]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-1/igt@gem_mmap_gtt@cpuset-big-copy.html
[96]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-2/igt@gem_mmap_gtt@cpuset-big-copy.html
* igt@i915_pm_rpm@sysfs-read:
- {shard-rkl}: [SKIP][97] ([i915#3844] / [i915#579]) -> [PASS][98]
[97]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-6/igt@i915_pm_rpm@sysfs-read.html
[98]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-5/igt@i915_pm_rpm@sysfs-read.html
* igt@i915_selftest@live@gt_pm:
- {shard-rkl}: [DMESG-FAIL][99] ([i915#1021]) -> [PASS][100]
[99]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-6/igt@i915_selftest@live@gt_pm.html
[100]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-5/igt@i915_selftest@live@gt_pm.html
* igt@kms_big_fb@linear-32bpp-rotate-180:
- shard-glk: [DMESG-WARN][101] ([i915#118] / [i915#95]) -> [PASS][102]
[101]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-glk8/igt@kms_big_fb@linear-32bpp-rotate-180.html
[102]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-glk9/igt@kms_big_fb@linear-32bpp-rotate-180.html
* igt@kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-180-hflip-async-flip:
- {shard-rkl}: [SKIP][103] ([i915#3721]) -> [PASS][104] +3 similar issues
[103]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-5/igt@kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-180-hflip-async-flip.html
[104]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-6/igt@kms_big_fb@x-tiled-max-hw-stride-32bpp-rotate-180-hflip-async-flip.html
* igt@kms_ccs@pipe-a-crc-primary-rotation-180-y_tiled_gen12_rc_ccs:
- {shard-rkl}: [FAIL][105] ([i915#3678]) -> [PASS][106] +1 similar issue
[105]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-5/igt@kms_ccs@pipe-a-crc-primary-rotation-180-y_tiled_gen12_rc_ccs.html
[106]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-6/igt@kms_ccs@pipe-a-crc-primary-rotation-180-y_tiled_gen12_rc_ccs.html
* igt@kms_color@pipe-a-ctm-0-75:
- {shard-rkl}: [SKIP][107] ([i915#1149] / [i915#1849]) -> [PASS][108] +2 similar issues
[107]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-1/igt@kms_color@pipe-a-ctm-0-75.html
[108]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-6/igt@kms_color@pipe-a-ctm-0-75.html
* igt@kms_concurrent@pipe-a:
- {shard-rkl}: [SKIP][109] ([i915#1845]) -> [PASS][110] +17 similar issues
[109]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-5/igt@kms_concurrent@pipe-a.html
[110]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-6/igt@kms_concurrent@pipe-a.html
* igt@kms_cursor_crc@pipe-a-cursor-128x42-sliding:
- {shard-rkl}: [SKIP][111] ([fdo#112022]) -> [PASS][112] +8 similar issues
[111]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-5/igt@kms_cursor_crc@pipe-a-cursor-128x42-sliding.html
[112]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-6/igt@kms_cursor_crc@pipe-a-cursor-128x42-sliding.html
* igt@kms_cursor_crc@pipe-a-cursor-suspend:
- shard-skl: [INCOMPLETE][113] ([i915#2828] / [i915#300]) -> [PASS][114]
[113]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-skl2/igt@kms_cursor_crc@pipe-a-cursor-suspend.html
[114]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-skl10/igt@kms_cursor_crc@pipe-a-cursor-suspend.html
* igt@kms_cursor_crc@pipe-c-cursor-suspend:
- shard-kbl: [DMESG-WARN][115] ([i915#180]) -> [PASS][116] +5 similar issues
[115]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-kbl4/igt@kms_cursor_crc@pipe-c-cursor-suspend.html
[116]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-kbl3/igt@kms_cursor_crc@pipe-c-cursor-suspend.html
* igt@kms_cursor_legacy@flip-vs-cursor-atomic:
- shard-skl: [FAIL][117] ([i915#2346]) -> [PASS][118]
[117]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-skl3/igt@kms_cursor_legacy@flip-vs-cursor-atomic.html
[118]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-skl6/igt@kms_cursor_legacy@flip-vs-cursor-atomic.html
* igt@kms_cursor_legacy@flip-vs-cursor-crc-atomic:
- {shard-rkl}: [SKIP][119] ([fdo#111825]) -> [PASS][120] +5 similar issues
[119]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-5/igt@kms_cursor_legacy@flip-vs-cursor-crc-atomic.html
[120]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-6/igt@kms_cursor_legacy@flip-vs-cursor-crc-atomic.html
* igt@kms_flip@flip-vs-suspend@a-dp1:
- shard-apl: [DMESG-WARN][121] ([i915#180]) -> [PASS][122] +2 similar issues
[121]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-apl6/igt@kms_flip@flip-vs-suspend@a-dp1.html
[122]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-apl3/igt@kms_flip@flip-vs-suspend@a-dp1.html
* igt@kms_frontbuffer_tracking@fbcpsr-1p-indfb-fliptrack-mmap-gtt:
- {shard-rkl}: [SKIP][123] ([i915#1849] / [i915#3180]) -> [PASS][124]
[123]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-1/igt@kms_frontbuffer_tracking@fbcpsr-1p-indfb-fliptrack-mmap-gtt.html
[124]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-6/igt@kms_frontbuffer_tracking@fbcpsr-1p-indfb-fliptrack-mmap-gtt.html
* igt@kms_frontbuffer_tracking@psr-1p-primscrn-spr-indfb-draw-mmap-gtt:
- {shard-rkl}: [SKIP][125] ([i915#1849]) -> [PASS][126] +23 similar issues
[125]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-5/igt@kms_frontbuffer_tracking@psr-1p-primscrn-spr-indfb-draw-mmap-gtt.html
[126]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-6/igt@kms_frontbuffer_tracking@psr-1p-primscrn-spr-indfb-draw-mmap-gtt.html
* igt@kms_hdr@bpc-switch-dpms:
- shard-skl: [FAIL][127] ([i915#1188]) -> [PASS][128] +1 similar issue
[127]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-skl8/igt@kms_hdr@bpc-switch-dpms.html
[128]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-skl4/igt@kms_hdr@bpc-switch-dpms.html
* igt@kms_plane_multiple@atomic-pipe-c-tiling-y:
- {shard-rkl}: [SKIP][129] ([i915#3558]) -> [PASS][130] +2 similar issues
[129]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-1/igt@kms_plane_multiple@atomic-pipe-c-tiling-y.html
[130]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-6/igt@kms_plane_multiple@atomic-pipe-c-tiling-y.html
* igt@kms_psr@cursor_plane_move:
- {shard-rkl}: [SKIP][131] ([i915#1072]) -> [PASS][132] +3 similar issues
[131]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-1/igt@kms_psr@cursor_plane_move.html
[132]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-6/igt@kms_psr@cursor_plane_move.html
* igt@kms_psr@psr2_cursor_blt:
- shard-iclb: [SKIP][133] ([fdo#109441]) -> [PASS][134] +3 similar issues
[133]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-iclb4/igt@kms_psr@psr2_cursor_blt.html
[134]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-iclb2/igt@kms_psr@psr2_cursor_blt.html
* igt@perf@polling-parameterized:
- shard-skl: [FAIL][135] ([i915#1542]) -> [PASS][136]
[135]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-skl8/igt@perf@polling-parameterized.html
[136]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-skl9/igt@perf@polling-parameterized.html
* igt@sysfs_preempt_timeout@timeout@rcs0:
- {shard-rkl}: [FAIL][137] ([i915#2821]) -> [PASS][138] +3 similar issues
[137]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-rkl-1/igt@sysfs_preempt_timeout@timeout@rcs0.html
[138]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-rkl-6/igt@sysfs_preempt_timeout@timeout@rcs0.html
#### Warnings ####
* igt@i915_pm_dc@dc3co-vpb-simulation:
- shard-iclb: [SKIP][139] ([i915#588]) -> [SKIP][140] ([i915#658])
[139]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-iclb2/igt@i915_pm_dc@dc3co-vpb-simulation.html
[140]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-iclb4/igt@i915_pm_dc@dc3co-vpb-simulation.html
* igt@i915_pm_rc6_residency@rc6-fence:
- shard-iclb: [WARN][141] ([i915#2684]) -> [WARN][142] ([i915#1804] / [i915#2684])
[141]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-iclb1/igt@i915_pm_rc6_residency@rc6-fence.html
[142]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-iclb3/igt@i915_pm_rc6_residency@rc6-fence.html
* igt@kms_psr2_sf@primary-plane-update-sf-dmg-area-1:
- shard-iclb: [SKIP][143] ([i915#2920]) -> [SKIP][144] ([i915#658]) +1 similar issue
[143]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-iclb2/igt@kms_psr2_sf@primary-plane-update-sf-dmg-area-1.html
[144]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-iclb4/igt@kms_psr2_sf@primary-plane-update-sf-dmg-area-1.html
* igt@kms_psr2_sf@primary-plane-update-sf-dmg-area-3:
- shard-iclb: [SKIP][145] ([i915#658]) -> [SKIP][146] ([i915#2920])
[145]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-iclb6/igt@kms_psr2_sf@primary-plane-update-sf-dmg-area-3.html
[146]: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/shard-iclb2/igt@kms_psr2_sf@primary-plane-update-sf-dmg-area-3.html
* igt@runner@aborted:
- shard-kbl: ([FAIL][147], [FAIL][148], [FAIL][149], [FAIL][150], [FAIL][151], [FAIL][152], [FAIL][153], [FAIL][154], [FAIL][155], [FAIL][156], [FAIL][157], [FAIL][158]) ([i915#1436] / [i915#180] / [i915#1814] / [i915#2426] / [i915#2505] / [i915#3002] / [i915#3363] / [i915#602]) -> ([FAIL][159], [FAIL][160], [FAIL][161], [FAIL][162], [FAIL][163], [FAIL][164], [FAIL][165], [FAIL][166], [FAIL][167]) ([i915#1436] / [i915#180] / [i915#1814] / [i915#2426] / [i915#2505] / [i915#3002] / [i915#3363])
[147]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-kbl2/igt@runner@aborted.html
[148]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/shard-kbl4/igt@runner@aborted.html
[149]: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_10418/sha
== Logs ==
For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_20741/index.html
[-- Attachment #1.2: Type: text/html, Size: 33782 bytes --]
[-- Attachment #2: Type: text/plain, Size: 160 bytes --]
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [Intel-gfx] [PATCH 2/2] drm/i915: Use Transparent Hugepages when IOMMU is enabled
2021-07-29 14:06 ` Daniel Vetter
@ 2021-09-03 12:47 ` Tvrtko Ursulin
2021-09-07 8:42 ` Daniel Vetter
0 siblings, 1 reply; 13+ messages in thread
From: Tvrtko Ursulin @ 2021-09-03 12:47 UTC (permalink / raw)
To: Daniel Vetter
Cc: intel-gfx, dri-devel, Tvrtko Ursulin, Chris Wilson,
Joonas Lahtinen, Matthew Auld, Eero Tamminen, Rodrigo Vivi
On 29/07/2021 15:06, Daniel Vetter wrote:
> On Thu, Jul 29, 2021 at 3:34 PM Tvrtko Ursulin
> <tvrtko.ursulin@linux.intel.com> wrote:
>>
>> From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>>
>> Usage of Transparent Hugepages was disabled in 9987da4b5dcf
>> ("drm/i915: Disable THP until we have a GPU read BW W/A"), but since it
>> appears majority of performance regressions reported with an enabled IOMMU
>> can be almost eliminated by turning them on, lets just do that.
>>
>> To err on the side of safety we keep the current default in cases where
>> IOMMU is not active, and only when it is default to the "huge=within_size"
>> mode. Although there probably would be wins to enable them throughout,
>> more extensive testing across benchmarks and platforms would need to be
>> done.
>>
>> With the patch and IOMMU enabled my local testing on a small Skylake part
>> shows OglVSTangent regression being reduced from ~14% (IOMMU on versus
>> IOMMU off) to ~2% (same comparison but with THP on).
>>
>> v2:
>> * Add Kconfig dependency to transparent hugepages and some help text.
>> * Move to helper for easier handling of kernel build options.
>>
>> v3:
>> * Drop Kconfig. (Daniel)
>>
>> References: b901bb89324a ("drm/i915/gemfs: enable THP")
>> References: 9987da4b5dcf ("drm/i915: Disable THP until we have a GPU read BW W/A")
>> References: https://gitlab.freedesktop.org/drm/intel/-/issues/430
>> Co-developed-by: Chris Wilson <chris@chris-wilson.co.uk>
>> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
>> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
>> Cc: Matthew Auld <matthew.auld@intel.com>
>> Cc: Eero Tamminen <eero.t.tamminen@intel.com>
>> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
>> Cc: Daniel Vetter <daniel@ffwll.ch>
>> Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>> Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> # v1
>
> On both patches: Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Eero's testing results at
https://gitlab.freedesktop.org/drm/intel/-/issues/430 are looking good -
seem to show this to be a net win for at least Gen9 and Gen12 platforms.
Is the ack enough to merge in this case or I should look for an r-b as well?
Regards,
Tvrtko
>> ---
>> drivers/gpu/drm/i915/gem/i915_gemfs.c | 22 +++++++++++++++++++---
>> 1 file changed, 19 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/i915/gem/i915_gemfs.c b/drivers/gpu/drm/i915/gem/i915_gemfs.c
>> index 5e6e8c91ab38..dbdbdc344d87 100644
>> --- a/drivers/gpu/drm/i915/gem/i915_gemfs.c
>> +++ b/drivers/gpu/drm/i915/gem/i915_gemfs.c
>> @@ -6,7 +6,6 @@
>>
>> #include <linux/fs.h>
>> #include <linux/mount.h>
>> -#include <linux/pagemap.h>
>>
>> #include "i915_drv.h"
>> #include "i915_gemfs.h"
>> @@ -15,6 +14,7 @@ int i915_gemfs_init(struct drm_i915_private *i915)
>> {
>> struct file_system_type *type;
>> struct vfsmount *gemfs;
>> + char *opts;
>>
>> type = get_fs_type("tmpfs");
>> if (!type)
>> @@ -26,10 +26,26 @@ int i915_gemfs_init(struct drm_i915_private *i915)
>> *
>> * One example, although it is probably better with a per-file
>> * control, is selecting huge page allocations ("huge=within_size").
>> - * Currently unused due to bandwidth issues (slow reads) on Broadwell+.
>> + * However, we only do so to offset the overhead of iommu lookups
>> + * due to bandwidth issues (slow reads) on Broadwell+.
>> */
>>
>> - gemfs = kern_mount(type);
>> + opts = NULL;
>> + if (intel_vtd_active()) {
>> + if (IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE)) {
>> + static char huge_opt[] = "huge=within_size"; /* r/w */
>> +
>> + opts = huge_opt;
>> + drm_info(&i915->drm,
>> + "Transparent Hugepage mode '%s'\n",
>> + opts);
>> + } else {
>> + drm_notice(&i915->drm,
>> + "Transparent Hugepage support is recommended for optimal performance when IOMMU is enabled!\n");
>> + }
>> + }
>> +
>> + gemfs = vfs_kern_mount(type, SB_KERNMOUNT, type->name, opts);
>> if (IS_ERR(gemfs))
>> return PTR_ERR(gemfs);
>>
>> --
>> 2.30.2
>>
>
>
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [Intel-gfx] [PATCH 2/2] drm/i915: Use Transparent Hugepages when IOMMU is enabled
2021-09-03 12:47 ` Tvrtko Ursulin
@ 2021-09-07 8:42 ` Daniel Vetter
2021-09-07 9:34 ` Tvrtko Ursulin
0 siblings, 1 reply; 13+ messages in thread
From: Daniel Vetter @ 2021-09-07 8:42 UTC (permalink / raw)
To: Tvrtko Ursulin
Cc: Daniel Vetter, intel-gfx, dri-devel, Tvrtko Ursulin,
Chris Wilson, Joonas Lahtinen, Matthew Auld, Eero Tamminen,
Rodrigo Vivi
On Fri, Sep 03, 2021 at 01:47:52PM +0100, Tvrtko Ursulin wrote:
>
> On 29/07/2021 15:06, Daniel Vetter wrote:
> > On Thu, Jul 29, 2021 at 3:34 PM Tvrtko Ursulin
> > <tvrtko.ursulin@linux.intel.com> wrote:
> > >
> > > From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> > >
> > > Usage of Transparent Hugepages was disabled in 9987da4b5dcf
> > > ("drm/i915: Disable THP until we have a GPU read BW W/A"), but since it
> > > appears majority of performance regressions reported with an enabled IOMMU
> > > can be almost eliminated by turning them on, lets just do that.
> > >
> > > To err on the side of safety we keep the current default in cases where
> > > IOMMU is not active, and only when it is default to the "huge=within_size"
> > > mode. Although there probably would be wins to enable them throughout,
> > > more extensive testing across benchmarks and platforms would need to be
> > > done.
> > >
> > > With the patch and IOMMU enabled my local testing on a small Skylake part
> > > shows OglVSTangent regression being reduced from ~14% (IOMMU on versus
> > > IOMMU off) to ~2% (same comparison but with THP on).
> > >
> > > v2:
> > > * Add Kconfig dependency to transparent hugepages and some help text.
> > > * Move to helper for easier handling of kernel build options.
> > >
> > > v3:
> > > * Drop Kconfig. (Daniel)
> > >
> > > References: b901bb89324a ("drm/i915/gemfs: enable THP")
> > > References: 9987da4b5dcf ("drm/i915: Disable THP until we have a GPU read BW W/A")
> > > References: https://gitlab.freedesktop.org/drm/intel/-/issues/430
> > > Co-developed-by: Chris Wilson <chris@chris-wilson.co.uk>
> > > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> > > Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
> > > Cc: Matthew Auld <matthew.auld@intel.com>
> > > Cc: Eero Tamminen <eero.t.tamminen@intel.com>
> > > Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> > > Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
> > > Cc: Daniel Vetter <daniel@ffwll.ch>
> > > Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> > > Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> # v1
> >
> > On both patches: Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
>
> Eero's testing results at
> https://gitlab.freedesktop.org/drm/intel/-/issues/430 are looking good -
> seem to show this to be a net win for at least Gen9 and Gen12 platforms.
>
> Is the ack enough to merge in this case or I should look for an r-b as well?
Since your back to defacto v1 with the 2nd patch I think you have full r-b
already. So more than enough I think.
Please do record the relative perf numbers from Eero in that issue in the
commit message so that we have that on the git log record too. It's easier
to find there than following the link and finding the right comment in the
issue.
Thanks, Daniel
>
> Regards,
>
> Tvrtko
>
> > > ---
> > > drivers/gpu/drm/i915/gem/i915_gemfs.c | 22 +++++++++++++++++++---
> > > 1 file changed, 19 insertions(+), 3 deletions(-)
> > >
> > > diff --git a/drivers/gpu/drm/i915/gem/i915_gemfs.c b/drivers/gpu/drm/i915/gem/i915_gemfs.c
> > > index 5e6e8c91ab38..dbdbdc344d87 100644
> > > --- a/drivers/gpu/drm/i915/gem/i915_gemfs.c
> > > +++ b/drivers/gpu/drm/i915/gem/i915_gemfs.c
> > > @@ -6,7 +6,6 @@
> > >
> > > #include <linux/fs.h>
> > > #include <linux/mount.h>
> > > -#include <linux/pagemap.h>
> > >
> > > #include "i915_drv.h"
> > > #include "i915_gemfs.h"
> > > @@ -15,6 +14,7 @@ int i915_gemfs_init(struct drm_i915_private *i915)
> > > {
> > > struct file_system_type *type;
> > > struct vfsmount *gemfs;
> > > + char *opts;
> > >
> > > type = get_fs_type("tmpfs");
> > > if (!type)
> > > @@ -26,10 +26,26 @@ int i915_gemfs_init(struct drm_i915_private *i915)
> > > *
> > > * One example, although it is probably better with a per-file
> > > * control, is selecting huge page allocations ("huge=within_size").
> > > - * Currently unused due to bandwidth issues (slow reads) on Broadwell+.
> > > + * However, we only do so to offset the overhead of iommu lookups
> > > + * due to bandwidth issues (slow reads) on Broadwell+.
> > > */
> > >
> > > - gemfs = kern_mount(type);
> > > + opts = NULL;
> > > + if (intel_vtd_active()) {
> > > + if (IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE)) {
> > > + static char huge_opt[] = "huge=within_size"; /* r/w */
> > > +
> > > + opts = huge_opt;
> > > + drm_info(&i915->drm,
> > > + "Transparent Hugepage mode '%s'\n",
> > > + opts);
> > > + } else {
> > > + drm_notice(&i915->drm,
> > > + "Transparent Hugepage support is recommended for optimal performance when IOMMU is enabled!\n");
> > > + }
> > > + }
> > > +
> > > + gemfs = vfs_kern_mount(type, SB_KERNMOUNT, type->name, opts);
> > > if (IS_ERR(gemfs))
> > > return PTR_ERR(gemfs);
> > >
> > > --
> > > 2.30.2
> > >
> >
> >
--
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [Intel-gfx] [PATCH 2/2] drm/i915: Use Transparent Hugepages when IOMMU is enabled
2021-09-07 8:42 ` Daniel Vetter
@ 2021-09-07 9:34 ` Tvrtko Ursulin
0 siblings, 0 replies; 13+ messages in thread
From: Tvrtko Ursulin @ 2021-09-07 9:34 UTC (permalink / raw)
To: Daniel Vetter
Cc: intel-gfx, dri-devel, Tvrtko Ursulin, Chris Wilson,
Joonas Lahtinen, Matthew Auld, Eero Tamminen, Rodrigo Vivi
On 07/09/2021 09:42, Daniel Vetter wrote:
> On Fri, Sep 03, 2021 at 01:47:52PM +0100, Tvrtko Ursulin wrote:
>>
>> On 29/07/2021 15:06, Daniel Vetter wrote:
>>> On Thu, Jul 29, 2021 at 3:34 PM Tvrtko Ursulin
>>> <tvrtko.ursulin@linux.intel.com> wrote:
>>>>
>>>> From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>>>>
>>>> Usage of Transparent Hugepages was disabled in 9987da4b5dcf
>>>> ("drm/i915: Disable THP until we have a GPU read BW W/A"), but since it
>>>> appears majority of performance regressions reported with an enabled IOMMU
>>>> can be almost eliminated by turning them on, lets just do that.
>>>>
>>>> To err on the side of safety we keep the current default in cases where
>>>> IOMMU is not active, and only when it is default to the "huge=within_size"
>>>> mode. Although there probably would be wins to enable them throughout,
>>>> more extensive testing across benchmarks and platforms would need to be
>>>> done.
>>>>
>>>> With the patch and IOMMU enabled my local testing on a small Skylake part
>>>> shows OglVSTangent regression being reduced from ~14% (IOMMU on versus
>>>> IOMMU off) to ~2% (same comparison but with THP on).
>>>>
>>>> v2:
>>>> * Add Kconfig dependency to transparent hugepages and some help text.
>>>> * Move to helper for easier handling of kernel build options.
>>>>
>>>> v3:
>>>> * Drop Kconfig. (Daniel)
>>>>
>>>> References: b901bb89324a ("drm/i915/gemfs: enable THP")
>>>> References: 9987da4b5dcf ("drm/i915: Disable THP until we have a GPU read BW W/A")
>>>> References: https://gitlab.freedesktop.org/drm/intel/-/issues/430
>>>> Co-developed-by: Chris Wilson <chris@chris-wilson.co.uk>
>>>> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
>>>> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
>>>> Cc: Matthew Auld <matthew.auld@intel.com>
>>>> Cc: Eero Tamminen <eero.t.tamminen@intel.com>
>>>> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>>>> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
>>>> Cc: Daniel Vetter <daniel@ffwll.ch>
>>>> Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>>>> Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> # v1
>>>
>>> On both patches: Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
>>
>> Eero's testing results at
>> https://gitlab.freedesktop.org/drm/intel/-/issues/430 are looking good -
>> seem to show this to be a net win for at least Gen9 and Gen12 platforms.
>>
>> Is the ack enough to merge in this case or I should look for an r-b as well?
>
> Since your back to defacto v1 with the 2nd patch I think you have full r-b
> already. So more than enough I think.
Just in case you missed it, v1 had Kconfig. But it's the same spirit so
probably indeed fine as you say.
> Please do record the relative perf numbers from Eero in that issue in the
> commit message so that we have that on the git log record too. It's easier
> to find there than following the link and finding the right comment in the
> issue.
Will do.
Regards,
Tvrtko
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [Intel-gfx] [PATCH 2/2] drm/i915: Use Transparent Hugepages when IOMMU is enabled
2021-07-29 12:21 ` Tvrtko Ursulin
@ 2021-07-29 12:28 ` Daniel Vetter
0 siblings, 0 replies; 13+ messages in thread
From: Daniel Vetter @ 2021-07-29 12:28 UTC (permalink / raw)
To: Tvrtko Ursulin
Cc: intel-gfx, dri-devel, Chris Wilson, Matthew Auld, Eero Tamminen
On Thu, Jul 29, 2021 at 2:21 PM Tvrtko Ursulin
<tvrtko.ursulin@linux.intel.com> wrote:
> On 29/07/2021 13:07, Daniel Vetter wrote:
> > On Thu, Jul 29, 2021 at 1:19 PM Tvrtko Ursulin
> > <tvrtko.ursulin@linux.intel.com> wrote:
> >>
> >> From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> >>
> >> Usage of Transparent Hugepages was disabled in 9987da4b5dcf
> >> ("drm/i915: Disable THP until we have a GPU read BW W/A"), but since it
> >> appears majority of performance regressions reported with an enabled IOMMU
> >> can be almost eliminated by turning them on, lets do that by adding a
> >> couple of Kconfig options.
> >>
> >> To err on the side of safety we keep the current default in cases where
> >> IOMMU is not active, and only when it is default to the "huge=within_size"
> >> mode. Although there probably would be wins to enable them throughout,
> >> more extensive testing across benchmarks and platforms would need to be
> >> done.
> >>
> >> With the patch and IOMMU enabled my local testing on a small Skylake part
> >> shows OglVSTangent regression being reduced from ~14% to ~2%.
> >
> > I guess the 14% regression is iommu disabled vs iommu enabled? Would
> > be good to clarify that.
>
> Should be clear from the first paragraph above - "...majority of
> performance regressions reported with an _enabled_ IOMMU can be almost
> eliminated...".
Yeah I inferred, but might be good to hammer that in by repeating, like
"reduced from 14% (for IOMMU on vs off case) to 2% (IOMMU on with THP
enabled vs IOMMU off with THP disabled)"
> >> v2:
> >> * Add Kconfig dependency to transparent hugepages and some help text.
> >
> > Uh I'm really not a huge fan of Kconfig for everything, especially for
> > tuning stuff. Maybe if there's a need a module param for debugging,
> > but otherwise can't we just pick the right default?
>
> Kconfig is picking the right default so I do not see a problem by
> allowing override from a deep enough menu. But I also do not feel so
> strongly about bikeshedding this to no kconfig, or a module param, or
> whatever - there are votes for all three options already, as usual. Main
> problem I have is actually..
Yeah that's pretty much what Kconfig is abused for: Everyone brings
their bikeshed because they're not quite happy, and it gets "resolved"
by Kconfigs to give everyone what they want. It just leads to
combinatorial explosion that no on tests. Hence unless we have a
demonstrated benefit of the choices there's going to be one default,
and you get to decide (which you did).
> > And it very much sounds like the right default here is "enable it
> > unconditionally if we have iommu support".
>
> .. about this - who knows? I will remind you of a certain VLK-20150
> which I thought was very important for going forward but was falling on
> deaf ears for years. As such I am waiting for Eero to come back and
> improvise some unofficial testing. It's extra bewildering to me given
> how we had the facility and then shut it down just like that.
Oh sure the general performance tuning is terrible, and also the
specific case of when THP. But we're looking the very specific case of
"IOMMU is enabled and it sucks away perf", and it looks like enabling
THP is the answer. So let's just do that.
Ofc we don't have full perf data, but we never have that even with a
nice perf lab (there's always more to benchmark than there's machine
time), so just doing as good as we can is imo perfectly fine enough.
You've put in the work (at least a bit), you get to pick the default
until we find something new.
-Daniel
>
> Regards,
>
> Tvrtko
>
> > -Daniel
> >
> >> * Move to helper for easier handling of kernel build options.
> >>
> >> References: b901bb89324a ("drm/i915/gemfs: enable THP")
> >> References: 9987da4b5dcf ("drm/i915: Disable THP until we have a GPU read BW W/A")
> >> References: https://gitlab.freedesktop.org/drm/intel/-/issues/430
> >> Co-developed-by: Chris Wilson <chris@chris-wilson.co.uk>
> >> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> >> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
> >> Cc: Matthew Auld <matthew.auld@intel.com>
> >> Cc: Eero Tamminen <eero.t.tamminen@intel.com>
> >> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> >> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
> >> Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> >> Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> # v1
> >> ---
> >> drivers/gpu/drm/i915/Kconfig.profile | 73 +++++++++++++++++++++++++++
> >> drivers/gpu/drm/i915/gem/i915_gemfs.c | 27 ++++++++--
> >> 2 files changed, 97 insertions(+), 3 deletions(-)
> >>
> >> diff --git a/drivers/gpu/drm/i915/Kconfig.profile b/drivers/gpu/drm/i915/Kconfig.profile
> >> index 39328567c200..d49ee794732f 100644
> >> --- a/drivers/gpu/drm/i915/Kconfig.profile
> >> +++ b/drivers/gpu/drm/i915/Kconfig.profile
> >> @@ -119,3 +119,76 @@ config DRM_I915_TIMESLICE_DURATION
> >> /sys/class/drm/card?/engine/*/timeslice_duration_ms
> >>
> >> May be 0 to disable timeslicing.
> >> +
> >> +choice
> >> + prompt "Transparent Hugepage Support (native)"
> >> + default DRM_I915_THP_NATIVE_NEVER
> >> + depends on TRANSPARENT_HUGEPAGE
> >> + help
> >> + Select the preferred method for allocating from Transparent Hugepages
> >> + when IOMMU is not enabled.
> >> +
> >> + config DRM_I915_THP_NATIVE_NEVER
> >> + bool "Never"
> >> + help
> >> + Disable using THP for system memory allocations, individually
> >> + allocating each 4K chunk as a separate page. It is unlikely that such
> >> + individual allocations will return contiguous memory.
> >> +
> >> + config DRM_I915_THP_NATIVE_WITHIN
> >> + bool "Within size"
> >> + help
> >> + Allocate whole 2M superpages while those chunks do not exceed the
> >> + object size. The remainder of the object will be allocated from 4K
> >> + pages. No overallocation.
> >> +
> >> + config DRM_I915_THP_NATIVE_ALWAYS
> >> + bool "Always"
> >> + help
> >> + Allocate the whole object using 2M superpages, even if the object does
> >> + not require an exact number of superpages.
> >> +
> >> +endchoice
> >> +
> >> +config DRM_I915_THP_NATIVE
> >> + string
> >> + default "always" if DRM_I915_THP_NATIVE_ALWAYS
> >> + default "within_size" if DRM_I915_THP_NATIVE_WITHIN
> >> + default "never" if DRM_I915_THP_NATIVE_NEVER
> >> +
> >> +choice
> >> + prompt "Transparent Hugepage Support (IOMMU)"
> >> + default DRM_I915_THP_IOMMU_WITHIN if TRANSPARENT_HUGEPAGE=y
> >> + default DRM_I915_THP_IOMMU_NEVER if TRANSPARENT_HUGEPAGE=n
> >> + depends on TRANSPARENT_HUGEPAGE
> >> + help
> >> + Select the preferred method for allocating from Transparent Hugepages
> >> + with IOMMU active.
> >> +
> >> + config DRM_I915_THP_IOMMU_NEVER
> >> + bool "Never"
> >> + help
> >> + Disable using THP for system memory allocations, individually
> >> + allocating each 4K chunk as a separate page. It is unlikely that such
> >> + individual allocations will return contiguous memory.
> >> +
> >> + config DRM_I915_THP_IOMMU_WITHIN
> >> + bool "Within size"
> >> + help
> >> + Allocate whole 2M superpages while those chunks do not exceed the
> >> + object size. The remainder of the object will be allocated from 4K
> >> + pages. No overallocation.
> >> +
> >> + config DRM_I915_THP_IOMMU_ALWAYS
> >> + bool "Always"
> >> + help
> >> + Allocate the whole object using 2M superpages, even if the object does
> >> + not require an exact number of superpages.
> >> +
> >> +endchoice
> >> +
> >> +config DRM_I915_THP_IOMMU
> >> + string
> >> + default "always" if DRM_I915_THP_IOMMU_ALWAYS
> >> + default "within_size" if DRM_I915_THP_IOMMU_WITHIN
> >> + default "never" if DRM_I915_THP_IOMMU_NEVER
> >> diff --git a/drivers/gpu/drm/i915/gem/i915_gemfs.c b/drivers/gpu/drm/i915/gem/i915_gemfs.c
> >> index 5e6e8c91ab38..871cbfb02fdf 100644
> >> --- a/drivers/gpu/drm/i915/gem/i915_gemfs.c
> >> +++ b/drivers/gpu/drm/i915/gem/i915_gemfs.c
> >> @@ -11,6 +11,26 @@
> >> #include "i915_drv.h"
> >> #include "i915_gemfs.h"
> >>
> >> +#if defined(CONFIG_DRM_I915_THP_NATIVE) && defined(CONFIG_DRM_I915_THP_IOMMU)
> >> +static char *gemfd_mount_opts(struct drm_i915_private *i915)
> >> +{
> >> + static char thp_native[] = "huge=" CONFIG_DRM_I915_THP_NATIVE;
> >> + static char thp_iommu[] = "huge=" CONFIG_DRM_I915_THP_IOMMU;
> >> + char *opts;
> >> +
> >> + opts = intel_vtd_active() ? thp_iommu : thp_native;
> >> + drm_info(&i915->drm, "Transparent Hugepage mode '%s'", opts);
> >> +
> >> + return opts;
> >> +}
> >> +#else
> >> +static char *gemfd_mount_opts(struct drm_i915_private *i915)
> >> +{
> >> + return NULL;
> >> +}
> >> +#endif
> >> +
> >> +
> >> int i915_gemfs_init(struct drm_i915_private *i915)
> >> {
> >> struct file_system_type *type;
> >> @@ -26,10 +46,11 @@ int i915_gemfs_init(struct drm_i915_private *i915)
> >> *
> >> * One example, although it is probably better with a per-file
> >> * control, is selecting huge page allocations ("huge=within_size").
> >> - * Currently unused due to bandwidth issues (slow reads) on Broadwell+.
> >> + * However, we only do so to offset the overhead of iommu lookups
> >> + * due to bandwidth issues (slow reads) on Broadwell+.
> >> */
> >> -
> >> - gemfs = kern_mount(type);
> >> + gemfs = vfs_kern_mount(type, SB_KERNMOUNT, type->name,
> >> + gemfd_mount_opts(i915));
> >> if (IS_ERR(gemfs))
> >> return PTR_ERR(gemfs);
> >>
> >> --
> >> 2.30.2
> >>
> >
> >
--
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [Intel-gfx] [PATCH 2/2] drm/i915: Use Transparent Hugepages when IOMMU is enabled
2021-07-29 12:07 ` Daniel Vetter
@ 2021-07-29 12:21 ` Tvrtko Ursulin
2021-07-29 12:28 ` Daniel Vetter
0 siblings, 1 reply; 13+ messages in thread
From: Tvrtko Ursulin @ 2021-07-29 12:21 UTC (permalink / raw)
To: Daniel Vetter
Cc: intel-gfx, dri-devel, Chris Wilson, Matthew Auld, Eero Tamminen
On 29/07/2021 13:07, Daniel Vetter wrote:
> On Thu, Jul 29, 2021 at 1:19 PM Tvrtko Ursulin
> <tvrtko.ursulin@linux.intel.com> wrote:
>>
>> From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>>
>> Usage of Transparent Hugepages was disabled in 9987da4b5dcf
>> ("drm/i915: Disable THP until we have a GPU read BW W/A"), but since it
>> appears majority of performance regressions reported with an enabled IOMMU
>> can be almost eliminated by turning them on, lets do that by adding a
>> couple of Kconfig options.
>>
>> To err on the side of safety we keep the current default in cases where
>> IOMMU is not active, and only when it is default to the "huge=within_size"
>> mode. Although there probably would be wins to enable them throughout,
>> more extensive testing across benchmarks and platforms would need to be
>> done.
>>
>> With the patch and IOMMU enabled my local testing on a small Skylake part
>> shows OglVSTangent regression being reduced from ~14% to ~2%.
>
> I guess the 14% regression is iommu disabled vs iommu enabled? Would
> be good to clarify that.
Should be clear from the first paragraph above - "...majority of
performance regressions reported with an _enabled_ IOMMU can be almost
eliminated...".
>
>> v2:
>> * Add Kconfig dependency to transparent hugepages and some help text.
>
> Uh I'm really not a huge fan of Kconfig for everything, especially for
> tuning stuff. Maybe if there's a need a module param for debugging,
> but otherwise can't we just pick the right default?
Kconfig is picking the right default so I do not see a problem by
allowing override from a deep enough menu. But I also do not feel so
strongly about bikeshedding this to no kconfig, or a module param, or
whatever - there are votes for all three options already, as usual. Main
problem I have is actually..
> And it very much sounds like the right default here is "enable it
> unconditionally if we have iommu support".
.. about this - who knows? I will remind you of a certain VLK-20150
which I thought was very important for going forward but was falling on
deaf ears for years. As such I am waiting for Eero to come back and
improvise some unofficial testing. It's extra bewildering to me given
how we had the facility and then shut it down just like that.
Regards,
Tvrtko
> -Daniel
>
>> * Move to helper for easier handling of kernel build options.
>>
>> References: b901bb89324a ("drm/i915/gemfs: enable THP")
>> References: 9987da4b5dcf ("drm/i915: Disable THP until we have a GPU read BW W/A")
>> References: https://gitlab.freedesktop.org/drm/intel/-/issues/430
>> Co-developed-by: Chris Wilson <chris@chris-wilson.co.uk>
>> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
>> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
>> Cc: Matthew Auld <matthew.auld@intel.com>
>> Cc: Eero Tamminen <eero.t.tamminen@intel.com>
>> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
>> Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>> Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> # v1
>> ---
>> drivers/gpu/drm/i915/Kconfig.profile | 73 +++++++++++++++++++++++++++
>> drivers/gpu/drm/i915/gem/i915_gemfs.c | 27 ++++++++--
>> 2 files changed, 97 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/i915/Kconfig.profile b/drivers/gpu/drm/i915/Kconfig.profile
>> index 39328567c200..d49ee794732f 100644
>> --- a/drivers/gpu/drm/i915/Kconfig.profile
>> +++ b/drivers/gpu/drm/i915/Kconfig.profile
>> @@ -119,3 +119,76 @@ config DRM_I915_TIMESLICE_DURATION
>> /sys/class/drm/card?/engine/*/timeslice_duration_ms
>>
>> May be 0 to disable timeslicing.
>> +
>> +choice
>> + prompt "Transparent Hugepage Support (native)"
>> + default DRM_I915_THP_NATIVE_NEVER
>> + depends on TRANSPARENT_HUGEPAGE
>> + help
>> + Select the preferred method for allocating from Transparent Hugepages
>> + when IOMMU is not enabled.
>> +
>> + config DRM_I915_THP_NATIVE_NEVER
>> + bool "Never"
>> + help
>> + Disable using THP for system memory allocations, individually
>> + allocating each 4K chunk as a separate page. It is unlikely that such
>> + individual allocations will return contiguous memory.
>> +
>> + config DRM_I915_THP_NATIVE_WITHIN
>> + bool "Within size"
>> + help
>> + Allocate whole 2M superpages while those chunks do not exceed the
>> + object size. The remainder of the object will be allocated from 4K
>> + pages. No overallocation.
>> +
>> + config DRM_I915_THP_NATIVE_ALWAYS
>> + bool "Always"
>> + help
>> + Allocate the whole object using 2M superpages, even if the object does
>> + not require an exact number of superpages.
>> +
>> +endchoice
>> +
>> +config DRM_I915_THP_NATIVE
>> + string
>> + default "always" if DRM_I915_THP_NATIVE_ALWAYS
>> + default "within_size" if DRM_I915_THP_NATIVE_WITHIN
>> + default "never" if DRM_I915_THP_NATIVE_NEVER
>> +
>> +choice
>> + prompt "Transparent Hugepage Support (IOMMU)"
>> + default DRM_I915_THP_IOMMU_WITHIN if TRANSPARENT_HUGEPAGE=y
>> + default DRM_I915_THP_IOMMU_NEVER if TRANSPARENT_HUGEPAGE=n
>> + depends on TRANSPARENT_HUGEPAGE
>> + help
>> + Select the preferred method for allocating from Transparent Hugepages
>> + with IOMMU active.
>> +
>> + config DRM_I915_THP_IOMMU_NEVER
>> + bool "Never"
>> + help
>> + Disable using THP for system memory allocations, individually
>> + allocating each 4K chunk as a separate page. It is unlikely that such
>> + individual allocations will return contiguous memory.
>> +
>> + config DRM_I915_THP_IOMMU_WITHIN
>> + bool "Within size"
>> + help
>> + Allocate whole 2M superpages while those chunks do not exceed the
>> + object size. The remainder of the object will be allocated from 4K
>> + pages. No overallocation.
>> +
>> + config DRM_I915_THP_IOMMU_ALWAYS
>> + bool "Always"
>> + help
>> + Allocate the whole object using 2M superpages, even if the object does
>> + not require an exact number of superpages.
>> +
>> +endchoice
>> +
>> +config DRM_I915_THP_IOMMU
>> + string
>> + default "always" if DRM_I915_THP_IOMMU_ALWAYS
>> + default "within_size" if DRM_I915_THP_IOMMU_WITHIN
>> + default "never" if DRM_I915_THP_IOMMU_NEVER
>> diff --git a/drivers/gpu/drm/i915/gem/i915_gemfs.c b/drivers/gpu/drm/i915/gem/i915_gemfs.c
>> index 5e6e8c91ab38..871cbfb02fdf 100644
>> --- a/drivers/gpu/drm/i915/gem/i915_gemfs.c
>> +++ b/drivers/gpu/drm/i915/gem/i915_gemfs.c
>> @@ -11,6 +11,26 @@
>> #include "i915_drv.h"
>> #include "i915_gemfs.h"
>>
>> +#if defined(CONFIG_DRM_I915_THP_NATIVE) && defined(CONFIG_DRM_I915_THP_IOMMU)
>> +static char *gemfd_mount_opts(struct drm_i915_private *i915)
>> +{
>> + static char thp_native[] = "huge=" CONFIG_DRM_I915_THP_NATIVE;
>> + static char thp_iommu[] = "huge=" CONFIG_DRM_I915_THP_IOMMU;
>> + char *opts;
>> +
>> + opts = intel_vtd_active() ? thp_iommu : thp_native;
>> + drm_info(&i915->drm, "Transparent Hugepage mode '%s'", opts);
>> +
>> + return opts;
>> +}
>> +#else
>> +static char *gemfd_mount_opts(struct drm_i915_private *i915)
>> +{
>> + return NULL;
>> +}
>> +#endif
>> +
>> +
>> int i915_gemfs_init(struct drm_i915_private *i915)
>> {
>> struct file_system_type *type;
>> @@ -26,10 +46,11 @@ int i915_gemfs_init(struct drm_i915_private *i915)
>> *
>> * One example, although it is probably better with a per-file
>> * control, is selecting huge page allocations ("huge=within_size").
>> - * Currently unused due to bandwidth issues (slow reads) on Broadwell+.
>> + * However, we only do so to offset the overhead of iommu lookups
>> + * due to bandwidth issues (slow reads) on Broadwell+.
>> */
>> -
>> - gemfs = kern_mount(type);
>> + gemfs = vfs_kern_mount(type, SB_KERNMOUNT, type->name,
>> + gemfd_mount_opts(i915));
>> if (IS_ERR(gemfs))
>> return PTR_ERR(gemfs);
>>
>> --
>> 2.30.2
>>
>
>
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [Intel-gfx] [PATCH 2/2] drm/i915: Use Transparent Hugepages when IOMMU is enabled
2021-07-29 11:18 ` [Intel-gfx] [PATCH 2/2] drm/i915: Use Transparent Hugepages when IOMMU is enabled Tvrtko Ursulin
@ 2021-07-29 12:07 ` Daniel Vetter
2021-07-29 12:21 ` Tvrtko Ursulin
0 siblings, 1 reply; 13+ messages in thread
From: Daniel Vetter @ 2021-07-29 12:07 UTC (permalink / raw)
To: Tvrtko Ursulin
Cc: intel-gfx, dri-devel, Chris Wilson, Matthew Auld, Eero Tamminen
On Thu, Jul 29, 2021 at 1:19 PM Tvrtko Ursulin
<tvrtko.ursulin@linux.intel.com> wrote:
>
> From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>
> Usage of Transparent Hugepages was disabled in 9987da4b5dcf
> ("drm/i915: Disable THP until we have a GPU read BW W/A"), but since it
> appears majority of performance regressions reported with an enabled IOMMU
> can be almost eliminated by turning them on, lets do that by adding a
> couple of Kconfig options.
>
> To err on the side of safety we keep the current default in cases where
> IOMMU is not active, and only when it is default to the "huge=within_size"
> mode. Although there probably would be wins to enable them throughout,
> more extensive testing across benchmarks and platforms would need to be
> done.
>
> With the patch and IOMMU enabled my local testing on a small Skylake part
> shows OglVSTangent regression being reduced from ~14% to ~2%.
I guess the 14% regression is iommu disabled vs iommu enabled? Would
be good to clarify that.
> v2:
> * Add Kconfig dependency to transparent hugepages and some help text.
Uh I'm really not a huge fan of Kconfig for everything, especially for
tuning stuff. Maybe if there's a need a module param for debugging,
but otherwise can't we just pick the right default?
And it very much sounds like the right default here is "enable it
unconditionally if we have iommu support".
-Daniel
> * Move to helper for easier handling of kernel build options.
>
> References: b901bb89324a ("drm/i915/gemfs: enable THP")
> References: 9987da4b5dcf ("drm/i915: Disable THP until we have a GPU read BW W/A")
> References: https://gitlab.freedesktop.org/drm/intel/-/issues/430
> Co-developed-by: Chris Wilson <chris@chris-wilson.co.uk>
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
> Cc: Matthew Auld <matthew.auld@intel.com>
> Cc: Eero Tamminen <eero.t.tamminen@intel.com>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
> Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> # v1
> ---
> drivers/gpu/drm/i915/Kconfig.profile | 73 +++++++++++++++++++++++++++
> drivers/gpu/drm/i915/gem/i915_gemfs.c | 27 ++++++++--
> 2 files changed, 97 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/Kconfig.profile b/drivers/gpu/drm/i915/Kconfig.profile
> index 39328567c200..d49ee794732f 100644
> --- a/drivers/gpu/drm/i915/Kconfig.profile
> +++ b/drivers/gpu/drm/i915/Kconfig.profile
> @@ -119,3 +119,76 @@ config DRM_I915_TIMESLICE_DURATION
> /sys/class/drm/card?/engine/*/timeslice_duration_ms
>
> May be 0 to disable timeslicing.
> +
> +choice
> + prompt "Transparent Hugepage Support (native)"
> + default DRM_I915_THP_NATIVE_NEVER
> + depends on TRANSPARENT_HUGEPAGE
> + help
> + Select the preferred method for allocating from Transparent Hugepages
> + when IOMMU is not enabled.
> +
> + config DRM_I915_THP_NATIVE_NEVER
> + bool "Never"
> + help
> + Disable using THP for system memory allocations, individually
> + allocating each 4K chunk as a separate page. It is unlikely that such
> + individual allocations will return contiguous memory.
> +
> + config DRM_I915_THP_NATIVE_WITHIN
> + bool "Within size"
> + help
> + Allocate whole 2M superpages while those chunks do not exceed the
> + object size. The remainder of the object will be allocated from 4K
> + pages. No overallocation.
> +
> + config DRM_I915_THP_NATIVE_ALWAYS
> + bool "Always"
> + help
> + Allocate the whole object using 2M superpages, even if the object does
> + not require an exact number of superpages.
> +
> +endchoice
> +
> +config DRM_I915_THP_NATIVE
> + string
> + default "always" if DRM_I915_THP_NATIVE_ALWAYS
> + default "within_size" if DRM_I915_THP_NATIVE_WITHIN
> + default "never" if DRM_I915_THP_NATIVE_NEVER
> +
> +choice
> + prompt "Transparent Hugepage Support (IOMMU)"
> + default DRM_I915_THP_IOMMU_WITHIN if TRANSPARENT_HUGEPAGE=y
> + default DRM_I915_THP_IOMMU_NEVER if TRANSPARENT_HUGEPAGE=n
> + depends on TRANSPARENT_HUGEPAGE
> + help
> + Select the preferred method for allocating from Transparent Hugepages
> + with IOMMU active.
> +
> + config DRM_I915_THP_IOMMU_NEVER
> + bool "Never"
> + help
> + Disable using THP for system memory allocations, individually
> + allocating each 4K chunk as a separate page. It is unlikely that such
> + individual allocations will return contiguous memory.
> +
> + config DRM_I915_THP_IOMMU_WITHIN
> + bool "Within size"
> + help
> + Allocate whole 2M superpages while those chunks do not exceed the
> + object size. The remainder of the object will be allocated from 4K
> + pages. No overallocation.
> +
> + config DRM_I915_THP_IOMMU_ALWAYS
> + bool "Always"
> + help
> + Allocate the whole object using 2M superpages, even if the object does
> + not require an exact number of superpages.
> +
> +endchoice
> +
> +config DRM_I915_THP_IOMMU
> + string
> + default "always" if DRM_I915_THP_IOMMU_ALWAYS
> + default "within_size" if DRM_I915_THP_IOMMU_WITHIN
> + default "never" if DRM_I915_THP_IOMMU_NEVER
> diff --git a/drivers/gpu/drm/i915/gem/i915_gemfs.c b/drivers/gpu/drm/i915/gem/i915_gemfs.c
> index 5e6e8c91ab38..871cbfb02fdf 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gemfs.c
> +++ b/drivers/gpu/drm/i915/gem/i915_gemfs.c
> @@ -11,6 +11,26 @@
> #include "i915_drv.h"
> #include "i915_gemfs.h"
>
> +#if defined(CONFIG_DRM_I915_THP_NATIVE) && defined(CONFIG_DRM_I915_THP_IOMMU)
> +static char *gemfd_mount_opts(struct drm_i915_private *i915)
> +{
> + static char thp_native[] = "huge=" CONFIG_DRM_I915_THP_NATIVE;
> + static char thp_iommu[] = "huge=" CONFIG_DRM_I915_THP_IOMMU;
> + char *opts;
> +
> + opts = intel_vtd_active() ? thp_iommu : thp_native;
> + drm_info(&i915->drm, "Transparent Hugepage mode '%s'", opts);
> +
> + return opts;
> +}
> +#else
> +static char *gemfd_mount_opts(struct drm_i915_private *i915)
> +{
> + return NULL;
> +}
> +#endif
> +
> +
> int i915_gemfs_init(struct drm_i915_private *i915)
> {
> struct file_system_type *type;
> @@ -26,10 +46,11 @@ int i915_gemfs_init(struct drm_i915_private *i915)
> *
> * One example, although it is probably better with a per-file
> * control, is selecting huge page allocations ("huge=within_size").
> - * Currently unused due to bandwidth issues (slow reads) on Broadwell+.
> + * However, we only do so to offset the overhead of iommu lookups
> + * due to bandwidth issues (slow reads) on Broadwell+.
> */
> -
> - gemfs = kern_mount(type);
> + gemfs = vfs_kern_mount(type, SB_KERNMOUNT, type->name,
> + gemfd_mount_opts(i915));
> if (IS_ERR(gemfs))
> return PTR_ERR(gemfs);
>
> --
> 2.30.2
>
--
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
^ permalink raw reply [flat|nested] 13+ messages in thread
* [Intel-gfx] [PATCH 2/2] drm/i915: Use Transparent Hugepages when IOMMU is enabled
2021-07-29 11:18 [Intel-gfx] [PATCH 1/2] " Tvrtko Ursulin
@ 2021-07-29 11:18 ` Tvrtko Ursulin
2021-07-29 12:07 ` Daniel Vetter
0 siblings, 1 reply; 13+ messages in thread
From: Tvrtko Ursulin @ 2021-07-29 11:18 UTC (permalink / raw)
To: Intel-gfx; +Cc: Eero Tamminen, dri-devel, Chris Wilson, Matthew Auld
From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Usage of Transparent Hugepages was disabled in 9987da4b5dcf
("drm/i915: Disable THP until we have a GPU read BW W/A"), but since it
appears majority of performance regressions reported with an enabled IOMMU
can be almost eliminated by turning them on, lets do that by adding a
couple of Kconfig options.
To err on the side of safety we keep the current default in cases where
IOMMU is not active, and only when it is default to the "huge=within_size"
mode. Although there probably would be wins to enable them throughout,
more extensive testing across benchmarks and platforms would need to be
done.
With the patch and IOMMU enabled my local testing on a small Skylake part
shows OglVSTangent regression being reduced from ~14% to ~2%.
v2:
* Add Kconfig dependency to transparent hugepages and some help text.
* Move to helper for easier handling of kernel build options.
References: b901bb89324a ("drm/i915/gemfs: enable THP")
References: 9987da4b5dcf ("drm/i915: Disable THP until we have a GPU read BW W/A")
References: https://gitlab.freedesktop.org/drm/intel/-/issues/430
Co-developed-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Cc: Eero Tamminen <eero.t.tamminen@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> # v1
---
drivers/gpu/drm/i915/Kconfig.profile | 73 +++++++++++++++++++++++++++
drivers/gpu/drm/i915/gem/i915_gemfs.c | 27 ++++++++--
2 files changed, 97 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/i915/Kconfig.profile b/drivers/gpu/drm/i915/Kconfig.profile
index 39328567c200..d49ee794732f 100644
--- a/drivers/gpu/drm/i915/Kconfig.profile
+++ b/drivers/gpu/drm/i915/Kconfig.profile
@@ -119,3 +119,76 @@ config DRM_I915_TIMESLICE_DURATION
/sys/class/drm/card?/engine/*/timeslice_duration_ms
May be 0 to disable timeslicing.
+
+choice
+ prompt "Transparent Hugepage Support (native)"
+ default DRM_I915_THP_NATIVE_NEVER
+ depends on TRANSPARENT_HUGEPAGE
+ help
+ Select the preferred method for allocating from Transparent Hugepages
+ when IOMMU is not enabled.
+
+ config DRM_I915_THP_NATIVE_NEVER
+ bool "Never"
+ help
+ Disable using THP for system memory allocations, individually
+ allocating each 4K chunk as a separate page. It is unlikely that such
+ individual allocations will return contiguous memory.
+
+ config DRM_I915_THP_NATIVE_WITHIN
+ bool "Within size"
+ help
+ Allocate whole 2M superpages while those chunks do not exceed the
+ object size. The remainder of the object will be allocated from 4K
+ pages. No overallocation.
+
+ config DRM_I915_THP_NATIVE_ALWAYS
+ bool "Always"
+ help
+ Allocate the whole object using 2M superpages, even if the object does
+ not require an exact number of superpages.
+
+endchoice
+
+config DRM_I915_THP_NATIVE
+ string
+ default "always" if DRM_I915_THP_NATIVE_ALWAYS
+ default "within_size" if DRM_I915_THP_NATIVE_WITHIN
+ default "never" if DRM_I915_THP_NATIVE_NEVER
+
+choice
+ prompt "Transparent Hugepage Support (IOMMU)"
+ default DRM_I915_THP_IOMMU_WITHIN if TRANSPARENT_HUGEPAGE=y
+ default DRM_I915_THP_IOMMU_NEVER if TRANSPARENT_HUGEPAGE=n
+ depends on TRANSPARENT_HUGEPAGE
+ help
+ Select the preferred method for allocating from Transparent Hugepages
+ with IOMMU active.
+
+ config DRM_I915_THP_IOMMU_NEVER
+ bool "Never"
+ help
+ Disable using THP for system memory allocations, individually
+ allocating each 4K chunk as a separate page. It is unlikely that such
+ individual allocations will return contiguous memory.
+
+ config DRM_I915_THP_IOMMU_WITHIN
+ bool "Within size"
+ help
+ Allocate whole 2M superpages while those chunks do not exceed the
+ object size. The remainder of the object will be allocated from 4K
+ pages. No overallocation.
+
+ config DRM_I915_THP_IOMMU_ALWAYS
+ bool "Always"
+ help
+ Allocate the whole object using 2M superpages, even if the object does
+ not require an exact number of superpages.
+
+endchoice
+
+config DRM_I915_THP_IOMMU
+ string
+ default "always" if DRM_I915_THP_IOMMU_ALWAYS
+ default "within_size" if DRM_I915_THP_IOMMU_WITHIN
+ default "never" if DRM_I915_THP_IOMMU_NEVER
diff --git a/drivers/gpu/drm/i915/gem/i915_gemfs.c b/drivers/gpu/drm/i915/gem/i915_gemfs.c
index 5e6e8c91ab38..871cbfb02fdf 100644
--- a/drivers/gpu/drm/i915/gem/i915_gemfs.c
+++ b/drivers/gpu/drm/i915/gem/i915_gemfs.c
@@ -11,6 +11,26 @@
#include "i915_drv.h"
#include "i915_gemfs.h"
+#if defined(CONFIG_DRM_I915_THP_NATIVE) && defined(CONFIG_DRM_I915_THP_IOMMU)
+static char *gemfd_mount_opts(struct drm_i915_private *i915)
+{
+ static char thp_native[] = "huge=" CONFIG_DRM_I915_THP_NATIVE;
+ static char thp_iommu[] = "huge=" CONFIG_DRM_I915_THP_IOMMU;
+ char *opts;
+
+ opts = intel_vtd_active() ? thp_iommu : thp_native;
+ drm_info(&i915->drm, "Transparent Hugepage mode '%s'", opts);
+
+ return opts;
+}
+#else
+static char *gemfd_mount_opts(struct drm_i915_private *i915)
+{
+ return NULL;
+}
+#endif
+
+
int i915_gemfs_init(struct drm_i915_private *i915)
{
struct file_system_type *type;
@@ -26,10 +46,11 @@ int i915_gemfs_init(struct drm_i915_private *i915)
*
* One example, although it is probably better with a per-file
* control, is selecting huge page allocations ("huge=within_size").
- * Currently unused due to bandwidth issues (slow reads) on Broadwell+.
+ * However, we only do so to offset the overhead of iommu lookups
+ * due to bandwidth issues (slow reads) on Broadwell+.
*/
-
- gemfs = kern_mount(type);
+ gemfs = vfs_kern_mount(type, SB_KERNMOUNT, type->name,
+ gemfd_mount_opts(i915));
if (IS_ERR(gemfs))
return PTR_ERR(gemfs);
--
2.30.2
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
^ permalink raw reply related [flat|nested] 13+ messages in thread
end of thread, other threads:[~2021-09-07 9:34 UTC | newest]
Thread overview: 13+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-07-29 13:34 [Intel-gfx] [PATCH 1/2] drm/i915/selftests: fixup igt_shrink_thp Tvrtko Ursulin
2021-07-29 13:34 ` [Intel-gfx] [PATCH 2/2] drm/i915: Use Transparent Hugepages when IOMMU is enabled Tvrtko Ursulin
2021-07-29 14:06 ` Daniel Vetter
2021-09-03 12:47 ` Tvrtko Ursulin
2021-09-07 8:42 ` Daniel Vetter
2021-09-07 9:34 ` Tvrtko Ursulin
2021-07-29 14:01 ` [Intel-gfx] ✗ Fi.CI.CHECKPATCH: warning for series starting with [1/2] drm/i915/selftests: fixup igt_shrink_thp Patchwork
2021-07-29 14:29 ` [Intel-gfx] ✓ Fi.CI.BAT: success " Patchwork
2021-07-29 23:16 ` [Intel-gfx] ✓ Fi.CI.IGT: " Patchwork
-- strict thread matches above, loose matches on Subject: below --
2021-07-29 11:18 [Intel-gfx] [PATCH 1/2] " Tvrtko Ursulin
2021-07-29 11:18 ` [Intel-gfx] [PATCH 2/2] drm/i915: Use Transparent Hugepages when IOMMU is enabled Tvrtko Ursulin
2021-07-29 12:07 ` Daniel Vetter
2021-07-29 12:21 ` Tvrtko Ursulin
2021-07-29 12:28 ` Daniel Vetter
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).