[PULL] drm-xe-next-fixes
Hi Dave and Sima, It is worth to mention that although there's a fix on xe_exec ioctl, that should not cause any uapi change. It would only fix a potential crash in the case that xe_ioctl failed with a bad sync array as input. Thanks, Rodrigo. drm-xe-next-fixes-2024-07-18: - Xe_exec ioctl minor fix on sync entry cleanup upon error (Ashutosh) - SRIOV: limit VF LMEM provisioning (Michal) - Wedge mode fixes (Brost) The following changes since commit 478a52707b0abe98aac7f8c53ccddb759be66b06: Merge tag 'amd-drm-next-6.11-2024-07-12' of https://gitlab.freedesktop.org/agd5f/linux into drm-next (2024-07-18 09:20:00 +1000) are available in the Git repository at: https://gitlab.freedesktop.org/drm/xe/kernel.git tags/drm-xe-next-fixes-2024-07-18 for you to fetch changes up to 90936a0a4c54f0a1cdf4538f9128821ad70c36ab: drm/xe: Don't suspend device upon wedge (2024-07-18 10:25:37 -0400) - Xe_exec ioctl minor fix on sync entry cleanup upon error (Ashutosh) - SRIOV: limit VF LMEM provisioning (Michal) - Wedge mode fixes (Brost) Ashutosh Dixit (1): drm/xe/exec: Fix minor bug related to xe_sync_entry_cleanup Matthew Brost (2): drm/xe: Wedge the entire device drm/xe: Don't suspend device upon wedge Michal Wajdeczko (1): drm/xe/pf: Limit fair VF LMEM provisioning drivers/gpu/drm/xe/xe_device.c | 20 drivers/gpu/drm/xe/xe_exec.c | 14 +-- drivers/gpu/drm/xe/xe_gt.c | 15 drivers/gpu/drm/xe/xe_gt.h | 1 + drivers/gpu/drm/xe/xe_gt_sriov_pf_config.c | 1 + drivers/gpu/drm/xe/xe_guc.c| 16 + drivers/gpu/drm/xe/xe_guc.h| 1 + drivers/gpu/drm/xe/xe_guc_submit.c | 38 -- drivers/gpu/drm/xe/xe_guc_submit.h | 1 + drivers/gpu/drm/xe/xe_uc.c | 14 +++ drivers/gpu/drm/xe/xe_uc.h | 1 + 11 files changed, 102 insertions(+), 20 deletions(-)
Re: [PULL] drm-xe-next-fixes v2
On Thu, Jul 11, 2024 at 02:29:04PM -0700, Jordan Justen wrote: > On 2024-07-11 08:18:18, Rodrigo Vivi wrote: > > Hi Dave and Sima, > > > > This is a v2 of https://lore.kernel.org/intel-xe/zo2so4t32dxqy...@intel.com/ > > > > v2 - Removed Thomas' write-back caching mode patch since Lucas will > > propagete > > that through drm-xe-fixes towards 6.10. So we remove the amount of patch > > duplication. > > > > Again, it is important to highlight the uapi rename present in this > > pull-request. > > Mesa is aligned and waiting to merge their side: > > https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30027 > > > > Since this uapi was recently merged, after we get both sides propagate > > there won't be any kernel or mesa releases with the old bad naming. > > So we should be good. > > This looks to be a simple rename, and it appears that the binary > interface is functionally the same. So, even if there was a Mesa > release using the old header, it should function fine with the > interface to the kernel regardless of which header the kernel used. > > If the binary interface had changed, I'm not sure an argument of "no > kernel or Mesa releases have happened" would be a good way to justify > such a change. Luckily that is not the case here anyway. Agreed. If it was that drastic we should never do and this argument shouldn't apply. But it is not so transparent as a full rename because there's one sysfs file name that also changed with the rest of the renaming. :/ > > -Jordan
Re: [PULL] drm-xe-next-fixes v2
On 2024-07-11 08:18:18, Rodrigo Vivi wrote: > Hi Dave and Sima, > > This is a v2 of https://lore.kernel.org/intel-xe/zo2so4t32dxqy...@intel.com/ > > v2 - Removed Thomas' write-back caching mode patch since Lucas will propagete > that through drm-xe-fixes towards 6.10. So we remove the amount of patch > duplication. > > Again, it is important to highlight the uapi rename present in this > pull-request. > Mesa is aligned and waiting to merge their side: > https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30027 > > Since this uapi was recently merged, after we get both sides propagate > there won't be any kernel or mesa releases with the old bad naming. > So we should be good. This looks to be a simple rename, and it appears that the binary interface is functionally the same. So, even if there was a Mesa release using the old header, it should function fine with the interface to the kernel regardless of which header the kernel used. If the binary interface had changed, I'm not sure an argument of "no kernel or Mesa releases have happened" would be a good way to justify such a change. Luckily that is not the case here anyway. -Jordan
[PULL] drm-xe-next-fixes v2
Hi Dave and Sima, This is a v2 of https://lore.kernel.org/intel-xe/zo2so4t32dxqy...@intel.com/ v2 - Removed Thomas' write-back caching mode patch since Lucas will propagete that through drm-xe-fixes towards 6.10. So we remove the amount of patch duplication. Again, it is important to highlight the uapi rename present in this pull-request. Mesa is aligned and waiting to merge their side: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30027 Since this uapi was recently merged, after we get both sides propagate there won't be any kernel or mesa releases with the old bad naming. So we should be good. Thanks, Rodrigo. drm-xe-next-fixes-2024-07-11: UAPI Changes: - Rename xe perf layer as xe observation layer (Ashutosh) Driver Changes: - Drop trace_xe_hw_fence_free (Brost) The following changes since commit dbf35b4deabb5706e739cec7ce35c12631bb8c87: Merge tag 'drm-intel-next-2024-06-28' of https://gitlab.freedesktop.org/drm/i915/kernel into drm-next (2024-07-10 10:36:47 +0200) are available in the Git repository at: https://gitlab.freedesktop.org/drm/xe/kernel.git tags/drm-xe-next-fixes-2024-07-11 for you to fetch changes up to 26d289158e491d5ae34b10eab40b093a40fab7d7: drm/xe: Drop trace_xe_hw_fence_free (2024-07-11 09:54:24 -0400) UAPI Changes: - Rename xe perf layer as xe observation layer (Ashutosh) Driver Changes: - Drop trace_xe_hw_fence_free (Brost) Ashutosh Dixit (1): drm/xe/uapi: Rename xe perf layer as xe observation layer Matthew Brost (1): drm/xe: Drop trace_xe_hw_fence_free drivers/gpu/drm/xe/Makefile | 2 +- drivers/gpu/drm/xe/xe_device.c | 4 +- drivers/gpu/drm/xe/xe_device_types.h | 2 +- drivers/gpu/drm/xe/xe_gt_types.h | 2 +- drivers/gpu/drm/xe/xe_hw_fence.c | 1 - drivers/gpu/drm/xe/xe_module.c | 6 +-- drivers/gpu/drm/xe/xe_oa.c | 34 ++-- drivers/gpu/drm/xe/xe_observation.c | 93 drivers/gpu/drm/xe/xe_observation.h | 20 +++ drivers/gpu/drm/xe/xe_perf.c | 92 --- drivers/gpu/drm/xe/xe_perf.h | 20 --- drivers/gpu/drm/xe/xe_trace.h| 5 -- include/uapi/drm/xe_drm.h| 102 ++- 13 files changed, 190 insertions(+), 193 deletions(-) create mode 100644 drivers/gpu/drm/xe/xe_observation.c create mode 100644 drivers/gpu/drm/xe/xe_observation.h delete mode 100644 drivers/gpu/drm/xe/xe_perf.c delete mode 100644 drivers/gpu/drm/xe/xe_perf.h
Re: [PULL] drm-xe-next-fixes
On Wed, Jul 10, 2024 at 02:42:38PM -0500, Lucas De Marchi wrote: > On Tue, Jul 09, 2024 at 05:31:39PM GMT, Rodrigo Vivi wrote: > > Hi Dave and Sima, > > > > Here goes a very early drm-xe-next-fixes with 2 important fixes > > that are going to impact user space. > > > > 1. The first one is the rename of the OA stuff from the bad 'perf' > > name to the xe_observation. Although the rename in the uapi > > header is likely inoffensive because our UMDs always copy the > > header to their code, there's a sysfs filename change that is > > impacting mesa. > > > > For this one Mesa MR is ready and they are only waiting for this > > pull request to be picked by you so they can merge that to Mesa: > > https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30027 > > > > With both sides in place, there won't be any kernel version > > released with the bad naming and no Mesa released using that > > filename. This is the main reason that I'm sending this PR > > earlier than normal. > > > > 2. The second case, which also impact user space is the > > write-back caching mode for system memory on DGFX. In this case > > we introduce a limitation in the cache selection uapi that is > > transparent to UMDs. I mean, no change on their side is needed. > > Coherence is maintained with some know possible visible and > > acceptable/accepted differences in CPU access speed. > > but this commit is also going to drm-xe-fixes, so I don't think we > should have it in this pull. I'm looking at some changes to dim to avoid > this kind of problem in future. okay, let's avoid the churn then. Sima, Dave, please ignore this PR. I'm going to drop this patch and resend this PR with the other two. > > Lucas De Marchi > > > > > Thanks, > > Rodrigo. > > > > drm-xe-next-fixes-2024-07-09: > > UAPI Changes: > > - Rename xe perf layer as xe observation layer (Ashutosh) > > - Use write-back caching mode for system memory on DGFX (Thomas) > > > > Driver Changes: > > - Drop trace_xe_hw_fence_free (Brost) > > The following changes since commit 62a05f4ae9c1fb70bc75d494c9c1c373d2c2e374: > > > > Merge tag 'drm-msm-next-2024-07-04' of > > https://gitlab.freedesktop.org/drm/msm into drm-next (2024-07-05 12:45:41 > > +0200) > > > > are available in the Git repository at: > > > > https://gitlab.freedesktop.org/drm/xe/kernel.git > > tags/drm-xe-next-fixes-2024-07-09 > > > > for you to fetch changes up to 463108053c19f24fa228863824698d5ca72826b6: > > > > drm/xe: Drop trace_xe_hw_fence_free (2024-07-09 16:21:26 -0400) > > > > > > UAPI Changes: > > - Rename xe perf layer as xe observation layer (Ashutosh) > > - Use write-back caching mode for system memory on DGFX (Thomas) > > > > Driver Changes: > > - Drop trace_xe_hw_fence_free (Brost) > > > > > > Ashutosh Dixit (1): > > drm/xe/uapi: Rename xe perf layer as xe observation layer > > > > Matthew Brost (1): > > drm/xe: Drop trace_xe_hw_fence_free > > > > Thomas Hellström (1): > > drm/xe: Use write-back caching mode for system memory on DGFX > > > > drivers/gpu/drm/xe/Makefile | 2 +- > > drivers/gpu/drm/xe/xe_bo.c | 47 +-- > > drivers/gpu/drm/xe/xe_bo_types.h | 3 +- > > drivers/gpu/drm/xe/xe_device.c | 4 +- > > drivers/gpu/drm/xe/xe_device_types.h | 2 +- > > drivers/gpu/drm/xe/xe_gt_types.h | 2 +- > > drivers/gpu/drm/xe/xe_hw_fence.c | 1 - > > drivers/gpu/drm/xe/xe_module.c | 6 +- > > drivers/gpu/drm/xe/xe_oa.c | 34 +-- > > drivers/gpu/drm/xe/xe_observation.c | 93 + > > drivers/gpu/drm/xe/xe_observation.h | 20 +++ > > drivers/gpu/drm/xe/xe_perf.c | 92 - > > drivers/gpu/drm/xe/xe_perf.h | 20 --- > > drivers/gpu/drm/xe/xe_trace.h| 5 -- > > include/uapi/drm/xe_drm.h| 110 > > +++ > > 15 files changed, 227 insertions(+), 214 deletions(-) > > create mode 100644 drivers/gpu/drm/xe/xe_observation.c > > create mode 100644 drivers/gpu/drm/xe/xe_observation.h > > delete mode 100644 drivers/gpu/drm/xe/xe_perf.c > > delete mode 100644 drivers/gpu/drm/xe/xe_perf.h
Re: [PULL] drm-xe-next-fixes
On Tue, Jul 09, 2024 at 05:31:39PM GMT, Rodrigo Vivi wrote: Hi Dave and Sima, Here goes a very early drm-xe-next-fixes with 2 important fixes that are going to impact user space. 1. The first one is the rename of the OA stuff from the bad 'perf' name to the xe_observation. Although the rename in the uapi header is likely inoffensive because our UMDs always copy the header to their code, there's a sysfs filename change that is impacting mesa. For this one Mesa MR is ready and they are only waiting for this pull request to be picked by you so they can merge that to Mesa: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30027 With both sides in place, there won't be any kernel version released with the bad naming and no Mesa released using that filename. This is the main reason that I'm sending this PR earlier than normal. 2. The second case, which also impact user space is the write-back caching mode for system memory on DGFX. In this case we introduce a limitation in the cache selection uapi that is transparent to UMDs. I mean, no change on their side is needed. Coherence is maintained with some know possible visible and acceptable/accepted differences in CPU access speed. but this commit is also going to drm-xe-fixes, so I don't think we should have it in this pull. I'm looking at some changes to dim to avoid this kind of problem in future. Lucas De Marchi Thanks, Rodrigo. drm-xe-next-fixes-2024-07-09: UAPI Changes: - Rename xe perf layer as xe observation layer (Ashutosh) - Use write-back caching mode for system memory on DGFX (Thomas) Driver Changes: - Drop trace_xe_hw_fence_free (Brost) The following changes since commit 62a05f4ae9c1fb70bc75d494c9c1c373d2c2e374: Merge tag 'drm-msm-next-2024-07-04' of https://gitlab.freedesktop.org/drm/msm into drm-next (2024-07-05 12:45:41 +0200) are available in the Git repository at: https://gitlab.freedesktop.org/drm/xe/kernel.git tags/drm-xe-next-fixes-2024-07-09 for you to fetch changes up to 463108053c19f24fa228863824698d5ca72826b6: drm/xe: Drop trace_xe_hw_fence_free (2024-07-09 16:21:26 -0400) UAPI Changes: - Rename xe perf layer as xe observation layer (Ashutosh) - Use write-back caching mode for system memory on DGFX (Thomas) Driver Changes: - Drop trace_xe_hw_fence_free (Brost) Ashutosh Dixit (1): drm/xe/uapi: Rename xe perf layer as xe observation layer Matthew Brost (1): drm/xe: Drop trace_xe_hw_fence_free Thomas Hellström (1): drm/xe: Use write-back caching mode for system memory on DGFX drivers/gpu/drm/xe/Makefile | 2 +- drivers/gpu/drm/xe/xe_bo.c | 47 +-- drivers/gpu/drm/xe/xe_bo_types.h | 3 +- drivers/gpu/drm/xe/xe_device.c | 4 +- drivers/gpu/drm/xe/xe_device_types.h | 2 +- drivers/gpu/drm/xe/xe_gt_types.h | 2 +- drivers/gpu/drm/xe/xe_hw_fence.c | 1 - drivers/gpu/drm/xe/xe_module.c | 6 +- drivers/gpu/drm/xe/xe_oa.c | 34 +-- drivers/gpu/drm/xe/xe_observation.c | 93 + drivers/gpu/drm/xe/xe_observation.h | 20 +++ drivers/gpu/drm/xe/xe_perf.c | 92 - drivers/gpu/drm/xe/xe_perf.h | 20 --- drivers/gpu/drm/xe/xe_trace.h| 5 -- include/uapi/drm/xe_drm.h| 110 +++ 15 files changed, 227 insertions(+), 214 deletions(-) create mode 100644 drivers/gpu/drm/xe/xe_observation.c create mode 100644 drivers/gpu/drm/xe/xe_observation.h delete mode 100644 drivers/gpu/drm/xe/xe_perf.c delete mode 100644 drivers/gpu/drm/xe/xe_perf.h
[PULL] drm-xe-next-fixes
Hi Dave and Sima, Here goes a very early drm-xe-next-fixes with 2 important fixes that are going to impact user space. 1. The first one is the rename of the OA stuff from the bad 'perf' name to the xe_observation. Although the rename in the uapi header is likely inoffensive because our UMDs always copy the header to their code, there's a sysfs filename change that is impacting mesa. For this one Mesa MR is ready and they are only waiting for this pull request to be picked by you so they can merge that to Mesa: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/30027 With both sides in place, there won't be any kernel version released with the bad naming and no Mesa released using that filename. This is the main reason that I'm sending this PR earlier than normal. 2. The second case, which also impact user space is the write-back caching mode for system memory on DGFX. In this case we introduce a limitation in the cache selection uapi that is transparent to UMDs. I mean, no change on their side is needed. Coherence is maintained with some know possible visible and acceptable/accepted differences in CPU access speed. Thanks, Rodrigo. drm-xe-next-fixes-2024-07-09: UAPI Changes: - Rename xe perf layer as xe observation layer (Ashutosh) - Use write-back caching mode for system memory on DGFX (Thomas) Driver Changes: - Drop trace_xe_hw_fence_free (Brost) The following changes since commit 62a05f4ae9c1fb70bc75d494c9c1c373d2c2e374: Merge tag 'drm-msm-next-2024-07-04' of https://gitlab.freedesktop.org/drm/msm into drm-next (2024-07-05 12:45:41 +0200) are available in the Git repository at: https://gitlab.freedesktop.org/drm/xe/kernel.git tags/drm-xe-next-fixes-2024-07-09 for you to fetch changes up to 463108053c19f24fa228863824698d5ca72826b6: drm/xe: Drop trace_xe_hw_fence_free (2024-07-09 16:21:26 -0400) UAPI Changes: - Rename xe perf layer as xe observation layer (Ashutosh) - Use write-back caching mode for system memory on DGFX (Thomas) Driver Changes: - Drop trace_xe_hw_fence_free (Brost) Ashutosh Dixit (1): drm/xe/uapi: Rename xe perf layer as xe observation layer Matthew Brost (1): drm/xe: Drop trace_xe_hw_fence_free Thomas Hellström (1): drm/xe: Use write-back caching mode for system memory on DGFX drivers/gpu/drm/xe/Makefile | 2 +- drivers/gpu/drm/xe/xe_bo.c | 47 +-- drivers/gpu/drm/xe/xe_bo_types.h | 3 +- drivers/gpu/drm/xe/xe_device.c | 4 +- drivers/gpu/drm/xe/xe_device_types.h | 2 +- drivers/gpu/drm/xe/xe_gt_types.h | 2 +- drivers/gpu/drm/xe/xe_hw_fence.c | 1 - drivers/gpu/drm/xe/xe_module.c | 6 +- drivers/gpu/drm/xe/xe_oa.c | 34 +-- drivers/gpu/drm/xe/xe_observation.c | 93 + drivers/gpu/drm/xe/xe_observation.h | 20 +++ drivers/gpu/drm/xe/xe_perf.c | 92 - drivers/gpu/drm/xe/xe_perf.h | 20 --- drivers/gpu/drm/xe/xe_trace.h| 5 -- include/uapi/drm/xe_drm.h| 110 +++ 15 files changed, 227 insertions(+), 214 deletions(-) create mode 100644 drivers/gpu/drm/xe/xe_observation.c create mode 100644 drivers/gpu/drm/xe/xe_observation.h delete mode 100644 drivers/gpu/drm/xe/xe_perf.c delete mode 100644 drivers/gpu/drm/xe/xe_perf.h
[PULL] drm-xe-next-fixes
Hi, Dave & Sima This weeks -next-fixes. Two fixes breifly described below. Driver Changes: - Use ordered WQ for G2H handler. (Matthew Brost) - Use flexible-array rather than zero-sized (Lucas De Marchi) Thanks, Thomas The following changes since commit 3bc8848bb7f7478e6806e4522b06b63f40a53e1e: drm/xe: Merge 16021540221 and 18034896535 WAs (2024-05-02 11:29:42 +0200) are available in the Git repository at: https://gitlab.freedesktop.org/drm/xe/kernel.git tags/drm-xe-next-fixes-2024-05-09-1 for you to fetch changes up to d69c3d4b53829097b8948d6791ea32c07de3faab: drm/xe/ads: Use flexible-array (2024-05-09 17:51:46 +0200) Driver Changes: - Use ordered WQ for G2H handler. (Matthew Brost) - Use flexible-array rather than zero-sized (Lucas De Marchi) Lucas De Marchi (1): drm/xe/ads: Use flexible-array Matthew Brost (1): drm/xe: Use ordered WQ for G2H handler drivers/gpu/drm/xe/xe_guc_ads.c | 2 +- drivers/gpu/drm/xe/xe_guc_ct.c | 5 + drivers/gpu/drm/xe/xe_guc_ct.h | 2 +- drivers/gpu/drm/xe/xe_guc_ct_types.h | 2 ++ 4 files changed, 9 insertions(+), 2 deletions(-)
[PULL] drm-xe-next-fixes
Dave, Sima This week's small set of fixes for drm-next. drm-xe-next-fixes-2024-05-02: Driver Changes: - Fix for a backmerge going slightly wrong. - An UAF fix - Avoid a WA error on LNL. Thanks, Thomas The following changes since commit 4a56c0ed5aa0bcbe1f5f7d755fb1fe1ebf48ae9c: Merge tag 'amd-drm-next-6.10-2024-04-26' of https://gitlab.freedesktop.org/agd5f/linux into drm-next (2024-04-30 14:43:00 +1000) are available in the Git repository at: https://gitlab.freedesktop.org/drm/xe/kernel.git tags/drm-xe-next-fixes-2024-05-02 for you to fetch changes up to 3bc8848bb7f7478e6806e4522b06b63f40a53e1e: drm/xe: Merge 16021540221 and 18034896535 WAs (2024-05-02 11:29:42 +0200) Driver Changes: - Fix for a backmerge going slightly wrong. - An UAF fix - Avoid a WA error on LNL. Lucas De Marchi (1): drm/xe: Merge 16021540221 and 18034896535 WAs Matthew Auld (1): drm/xe/vm: prevent UAF in rebind_work_func() Thomas Hellström (1): drm/xe: Fix unexpected backmerge results drivers/gpu/drm/xe/xe_vm.c | 16 ++-- drivers/gpu/drm/xe/xe_vm_types.h | 4 drivers/gpu/drm/xe/xe_wa.c | 7 +-- 3 files changed, 15 insertions(+), 12 deletions(-)
[PULL] drm-xe-next-fixes
Hi Dave and Sima, Here are the drm-xe fixes for the 6.9 cycle. Just 3 fixes: one trivial fix for error path handling, one to avoid pinning all VMAs in drm_gpuvm_exec_lock() when num_batch_buffer passed to xe_exec is 0 and the other to allow userspace to free userptr while still having bindings. Lucas De Marchi drm-xe-next-fixes-2024-03-14: Driver changes: - Invalidate userptr VMA on page pin fault, allowing userspace to free userptr while still having bindings - Fail early on sysfs file creation error - Skip VMA pinning on xe_exec with num_batch_buffer == 0 The following changes since commit e62d2e00780b4a465c77d2229837495fcbc480d3: drm/xe: Replace 'grouped target' in Makefile with pattern rule (2024-03-04 08:41:28 -0600) are available in the Git repository at: https://gitlab.freedesktop.org/drm/xe/kernel.git tags/drm-xe-next-fixes-2024-03-14 for you to fetch changes up to dd8a07f06dfd946e0eea1a3323d52e7c28a6ed80: drm/xe: Skip VMAs pin when requesting signal to the last XE_EXEC (2024-03-14 14:29:42 -0500) Driver changes: - Invalidate userptr VMA on page pin fault, allowing userspace to free userptr while still having bindings - Fail early on sysfs file creation error - Skip VMA pinning on xe_exec with num_batch_buffer == 0 Himal Prasad Ghimiray (1): drm/xe: Return if kobj creation is failed José Roberto de Souza (1): drm/xe: Skip VMAs pin when requesting signal to the last XE_EXEC Matthew Brost (1): drm/xe: Invalidate userptr VMA on page pin fault drivers/gpu/drm/xe/xe_exec.c | 41 drivers/gpu/drm/xe/xe_gt_pagefault.c | 4 ++-- drivers/gpu/drm/xe/xe_trace.h| 2 +- drivers/gpu/drm/xe/xe_vm.c | 32 +--- drivers/gpu/drm/xe/xe_vm_types.h | 7 ++ drivers/gpu/drm/xe/xe_vram_freq.c| 4 +++- 6 files changed, 55 insertions(+), 35 deletions(-)
[PULL] drm-xe-next-fixes
Hi Dave and Sima, A few xe fixes for 6.9. drm-xe-next-fixes-2024-03-04: Driver Changes: - Fix kunit link failure with built-in xe - Fix one more 32-bit build failure with ARM compiler - Fix initialization order of topology struct - Cleanup unused fields in struct xe_vm - Fix xe_vm leak when handling page fault on a VM not in fault mode - Drop use of "grouped target" feature in Makefile since that's only available in make >= 4.3 thanks, Lucas De Marchi The following changes since commit c6d6a82d8a9f8f9326b760accaa532b839b80140: Merge tag 'drm-misc-next-fixes-2024-02-29' of https://anongit.freedesktop.org/git/drm/drm-misc into drm-next (2024-03-01 19:38:13 +1000) are available in the Git repository at: ssh://g...@gitlab.freedesktop.org/drm/xe/kernel.git tags/drm-xe-next-fixes-2024-03-04 for you to fetch changes up to e62d2e00780b4a465c77d2229837495fcbc480d3: drm/xe: Replace 'grouped target' in Makefile with pattern rule (2024-03-04 08:41:28 -0600) Driver Changes: - Fix kunit link failure with built-in xe - Fix one more 32-bit build failure with ARM compiler - Fix initialization order of topology struct - Cleanup unused fields in struct xe_vm - Fix xe_vm leak when handling page fault on a VM not in fault mode - Drop use of "grouped target" feature in Makefile since that's only available in make >= 4.3 Arnd Bergmann (2): drm/xe/kunit: fix link failure with built-in xe drm/xe/xe2: fix 64-bit division in pte_update_size Dafna Hirschfeld (1): drm/xe: Replace 'grouped target' in Makefile with pattern rule Matthew Brost (1): drm/xe: Fix ref counting leak on page fault Mika Kuoppala (1): drm/xe: Remove obsolete async_ops from struct xe_vm Zhanjun Dong (1): drm/xe/guc: Fix missing topology init drivers/gpu/drm/xe/Kconfig | 1 + drivers/gpu/drm/xe/Kconfig.debug | 1 - drivers/gpu/drm/xe/Makefile | 9 ++--- drivers/gpu/drm/xe/xe_gt.c | 3 +-- drivers/gpu/drm/xe/xe_gt_pagefault.c | 6 -- drivers/gpu/drm/xe/xe_migrate.c | 2 +- drivers/gpu/drm/xe/xe_vm_types.h | 24 7 files changed, 13 insertions(+), 33 deletions(-)
[PULL] drm-xe-next-fixes
Hi Maintainers, Various fixes for the Xe driver, as described below, for -rc1. Thanks, Thomas The following changes since commit 315acff5196f4e2f84a2a2d093000e0c6b0b4d1c: drm/xe: Fix warning on impossible condition (2023-12-26 12:53:26 -0500) are available in the Git repository at: https://gitlab.freedesktop.org/drm/xe/kernel.git tags/drm-xe-next-fixes-2024-01-16 for you to fetch changes up to bf3ff145df184698a8a80b33265064638572366f: drm/xe: display support should not depend on EXPERT (2024-01-15 21:40:32 +0100) Driver Changes: - Fix for definition of wakeref_t - Fix for an error code aliasing - Fix for VM_UNBIND_ALL in the case there are no bound VMAs - Fixes for a number of __iomem address space mismatches reported by sparse - Fixes for the assignment of exec_queue priority - A Fix for skip_guc_pc not taking effect - Workaround for a build problem on GCC 11 - A couple of fixes for error paths - Fix a Flat CCS compression metadata copy issue - Fix a misplace array bounds checking - Don't have display support depend on EXPERT (as discussed on IRC) Brian Welty (3): drm/xe: Fix guc_exec_queue_set_priority drm/xe: Fix modifying exec_queue priority in xe_migrate_init drm/xe: Fix bounds checking in __xe_bo_placement_for_flags() Dan Carpenter (3): drm/xe/device: clean up on error in probe() drm/xe/selftests: Fix an error pointer dereference bug drm/xe: unlock on error path in xe_vm_add_compute_exec_queue() Jani Nikula (1): drm/xe: display support should not depend on EXPERT José Roberto de Souza (1): drm/xe: Fix definition of intel_wakeref_t Matthew Brost (1): drm/xe: Fix exec IOCTL long running exec queue ring full condition Paul E. McKenney (1): drm/xe: Fix build bug for GCC 11 Thomas Hellström (6): drm/xe/vm: Fix an error path drm/xe: Use __iomem for the regs pointer drm/xe: Annotate xe_mem_region::mapping with __iomem drm/xe: Annotate multiple mmio pointers with __iomem drm/xe: Annotate xe_ttm_stolen_mgr::mapping with __iomem drm/xe/migrate: Fix CCS copy for small VRAM copy chunks Vinay Belgaumkar (1): drm/xe: Check skip_guc_pc before setting SLPC flag drivers/gpu/drm/xe/Kconfig | 2 +- drivers/gpu/drm/xe/Makefile| 1 - .../gpu/drm/xe/compat-i915-headers/intel_wakeref.h | 2 +- drivers/gpu/drm/xe/tests/xe_bo.c | 5 +- drivers/gpu/drm/xe/tests/xe_migrate.c | 2 +- drivers/gpu/drm/xe/xe_bo.c | 16 +-- drivers/gpu/drm/xe/xe_device.c | 2 +- drivers/gpu/drm/xe/xe_device_types.h | 8 +- drivers/gpu/drm/xe/xe_exec.c | 7 +- drivers/gpu/drm/xe/xe_exec_queue.c | 5 + drivers/gpu/drm/xe/xe_exec_queue_types.h | 6 +- drivers/gpu/drm/xe/xe_gt_freq.c| 3 + drivers/gpu/drm/xe/xe_guc.c| 7 +- drivers/gpu/drm/xe/xe_guc_submit.c | 7 +- drivers/gpu/drm/xe/xe_migrate.c| 133 + drivers/gpu/drm/xe/xe_mmio.c | 2 +- drivers/gpu/drm/xe/xe_ttm_stolen_mgr.c | 4 +- drivers/gpu/drm/xe/xe_vm.c | 15 ++- 18 files changed, 136 insertions(+), 91 deletions(-)
[PULL] drm-xe-next-fixes
Hi Dave and Sima, Here goes the fixes that I had promised last week. With these in place we should be good now with the all yes config W=1 and different compilers. Thanks for already include that one that disables the 32-bit. I had just noticed when I was trying to cherry-pick it to the -next-fixes branch. Thanks, Rodrigo. The following changes since commit 92242716ee92d2aa3c38c736b53d8910d443566d: Merge tag 'drm-habanalabs-next-2023-12-19' of https://git.kernel.org/pub/scm/linux/kernel/git/ogabbay/linux into drm-next (2023-12-22 14:52:04 +1000) are available in the Git repository at: https://gitlab.freedesktop.org/drm/xe/kernel.git tags/drm-xe-next-fixes-2023-12-26 for you to fetch changes up to 315acff5196f4e2f84a2a2d093000e0c6b0b4d1c: drm/xe: Fix warning on impossible condition (2023-12-26 12:53:26 -0500) - Fix couple unfined behavior cases to calm UBSAN and clang (Matt Brost, Lucas) Lucas De Marchi (1): drm/xe: Fix warning on impossible condition Matthew Brost (1): drm/xe: Fix UBSAN splat in add_preempt_fences() drivers/gpu/drm/xe/xe_vm.c | 3 +++ drivers/gpu/drm/xe/xe_wait_user_fence.c | 1 + 2 files changed, 4 insertions(+)
