From 74b6fd3cdb6c4c7cb67e40e9ed27be158ad3d97e Mon Sep 17 00:00:00 2001 From: Yousaf Kaukab Date: Mar 21 2024 15:18:39 +0000 Subject: Merge remote-tracking branch 'origin/users/pjakobsson/SLE15-SP5/for-next' into SLE15-SP5 Pull drm patches metadata update from Patrik Jakobsson --- diff --git a/patches.suse/0549-drm-amdgpu-enable-Vangogh-VCN-indirect-sram-mode.patch b/patches.suse/0549-drm-amdgpu-enable-Vangogh-VCN-indirect-sram-mode.patch index 1ad0b48..1113f38 100644 --- a/patches.suse/0549-drm-amdgpu-enable-Vangogh-VCN-indirect-sram-mode.patch +++ b/patches.suse/0549-drm-amdgpu-enable-Vangogh-VCN-indirect-sram-mode.patch @@ -5,6 +5,7 @@ Subject: drm/amdgpu: enable Vangogh VCN indirect sram mode Git-commit: 9a8cc8cabc1e351614fd7f9e774757a5143b6fe8 Patch-mainline: v6.1-rc8 References: jsc#PED-1166 jsc#PED-1168 jsc#PED-1170 jsc#PED-1218 jsc#PED-1220 jsc#PED-1222 jsc#PED-1223 jsc#PED-1225 jsc#PED-2849 +Alt-commit: 82132ecc54322c58dd86012ebe9733d6b1f73bea So that uses PSP to initialize HW. diff --git a/patches.suse/1393-drm-i915-gt-Ignore-TLB-invalidations-on-idle-engines.patch b/patches.suse/1393-drm-i915-gt-Ignore-TLB-invalidations-on-idle-engines.patch index fb19365..8386029 100644 --- a/patches.suse/1393-drm-i915-gt-Ignore-TLB-invalidations-on-idle-engines.patch +++ b/patches.suse/1393-drm-i915-gt-Ignore-TLB-invalidations-on-idle-engines.patch @@ -5,6 +5,7 @@ Subject: drm/i915/gt: Ignore TLB invalidations on idle engines MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit +Alt-commit: 4bedceaed1ae1172cfe72d3ff752b3a1d32fe4d9 Git-commit: db100e28fdf026a1fc10657c5170bb1e65663805 Patch-mainline: v6.0-rc2 References: jsc#PED-1166 jsc#PED-1168 jsc#PED-1170 jsc#PED-1218 jsc#PED-1220 jsc#PED-1222 jsc#PED-1223 jsc#PED-1225 jsc#PED-2849 diff --git a/patches.suse/1394-drm-i915-gt-Invalidate-TLB-of-the-OA-unit-at-TLB-inv.patch b/patches.suse/1394-drm-i915-gt-Invalidate-TLB-of-the-OA-unit-at-TLB-inv.patch index 4aca4e3..494087d 100644 --- a/patches.suse/1394-drm-i915-gt-Invalidate-TLB-of-the-OA-unit-at-TLB-inv.patch +++ b/patches.suse/1394-drm-i915-gt-Invalidate-TLB-of-the-OA-unit-at-TLB-inv.patch @@ -5,6 +5,7 @@ Subject: drm/i915/gt: Invalidate TLB of the OA unit at TLB invalidations MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit +Alt-commit: dfc83de118ff7930acc9a4c8dfdba7c153aa44d6 Git-commit: 180abeb2c5032704787151135b6a38c6b71295a6 Patch-mainline: v6.0-rc2 References: jsc#PED-1166 jsc#PED-1168 jsc#PED-1170 jsc#PED-1218 jsc#PED-1220 jsc#PED-1222 jsc#PED-1223 jsc#PED-1225 jsc#PED-2849 diff --git a/patches.suse/1396-drm-i915-gt-Batch-TLB-invalidations.patch b/patches.suse/1396-drm-i915-gt-Batch-TLB-invalidations.patch index b115701..76eb0e2 100644 --- a/patches.suse/1396-drm-i915-gt-Batch-TLB-invalidations.patch +++ b/patches.suse/1396-drm-i915-gt-Batch-TLB-invalidations.patch @@ -5,6 +5,7 @@ Subject: drm/i915/gt: Batch TLB invalidations Git-commit: 59eda6ce824e95b98c45628fe6c0adb9130c6df2 Patch-mainline: v6.0-rc2 References: jsc#PED-1166 jsc#PED-1168 jsc#PED-1170 jsc#PED-1218 jsc#PED-1220 jsc#PED-1222 jsc#PED-1223 jsc#PED-1225 jsc#PED-2849 +Alt-commit: 5d36acb7198b0e5eb88e6b701f9ad7b9448f8df9 Invalidate TLB in batches, in order to reduce performance regressions. diff --git a/patches.suse/1398-drm-i915-pass-a-pointer-for-tlb-seqno-at-vma_invalid.patch b/patches.suse/1398-drm-i915-pass-a-pointer-for-tlb-seqno-at-vma_invalid.patch index c455afb..7e7aded 100644 --- a/patches.suse/1398-drm-i915-pass-a-pointer-for-tlb-seqno-at-vma_invalid.patch +++ b/patches.suse/1398-drm-i915-pass-a-pointer-for-tlb-seqno-at-vma_invalid.patch @@ -5,6 +5,7 @@ Subject: drm/i915: pass a pointer for tlb seqno at vma_invalidate_tlb() Git-commit: 9d50bff40e3e366886ec37299fc317edf84be0c9 Patch-mainline: v6.0-rc2 References: jsc#PED-1166 jsc#PED-1168 jsc#PED-1170 jsc#PED-1218 jsc#PED-1220 jsc#PED-1222 jsc#PED-1223 jsc#PED-1225 jsc#PED-2849 +Alt-commit: 3d037d99e61a1e7a3ae3d214146d88db349dd19f WRITE_ONCE() should happen at the original var, not on a local copy of it. diff --git a/patches.suse/1536-drm-i915-guc-clear-stalled-request-after-a-reset.patch b/patches.suse/1536-drm-i915-guc-clear-stalled-request-after-a-reset.patch index 272e85b..aa41f70 100644 --- a/patches.suse/1536-drm-i915-guc-clear-stalled-request-after-a-reset.patch +++ b/patches.suse/1536-drm-i915-guc-clear-stalled-request-after-a-reset.patch @@ -5,6 +5,8 @@ Subject: drm/i915/guc: clear stalled request after a reset Git-commit: de2228c04150df8632ad22ee490de2ed579f64e8 Patch-mainline: v6.0-rc4 References: jsc#PED-1166 jsc#PED-1168 jsc#PED-1170 jsc#PED-1218 jsc#PED-1220 jsc#PED-1222 jsc#PED-1223 jsc#PED-1225 jsc#PED-2849 +Alt-commit: 4595a25443447b9542b2a5ee7961eb290e94b496 +Alt-commit: f922fbb0f2ad1fd3e3186f39c46673419e6d9281 If the GuC CTs are full and we need to stall the request submission while waiting for space, we save the stalled request and where the stall diff --git a/patches.suse/1585-drm-i915-slpc-Let-s-fix-the-PCODE-min-freq-table-set.patch b/patches.suse/1585-drm-i915-slpc-Let-s-fix-the-PCODE-min-freq-table-set.patch index 13f4206..0c1ce63 100644 --- a/patches.suse/1585-drm-i915-slpc-Let-s-fix-the-PCODE-min-freq-table-set.patch +++ b/patches.suse/1585-drm-i915-slpc-Let-s-fix-the-PCODE-min-freq-table-set.patch @@ -5,6 +5,7 @@ Subject: drm/i915/slpc: Let's fix the PCODE min freq table setup for SLPC Git-commit: e1cab970574c001d83e59ca8388c474a57a1afb6 Patch-mainline: v6.0-rc5 References: jsc#PED-1166 jsc#PED-1168 jsc#PED-1170 jsc#PED-1218 jsc#PED-1220 jsc#PED-1222 jsc#PED-1223 jsc#PED-1225 jsc#PED-2849 +Alt-commit: 018a7bdbb090b9155a6509a0d1a684db4afaa5b1 We need to inform PCODE of a desired ring frequencies so PCODE update the memory frequencies to us. rps->min_freq and rps->max_freq are the diff --git a/patches.suse/1625-drm-i915-vdsc-Set-VDSC-PIC_HEIGHT-before-using-for-D.patch b/patches.suse/1625-drm-i915-vdsc-Set-VDSC-PIC_HEIGHT-before-using-for-D.patch index ec486d8..9b13cef 100644 --- a/patches.suse/1625-drm-i915-vdsc-Set-VDSC-PIC_HEIGHT-before-using-for-D.patch +++ b/patches.suse/1625-drm-i915-vdsc-Set-VDSC-PIC_HEIGHT-before-using-for-D.patch @@ -5,6 +5,7 @@ Subject: drm/i915/vdsc: Set VDSC PIC_HEIGHT before using for DP DSC Git-commit: 0785691f5711a8f210bb15a5177c2999ebd3702e Patch-mainline: v6.0-rc6 References: jsc#PED-1166 jsc#PED-1168 jsc#PED-1170 jsc#PED-1218 jsc#PED-1220 jsc#PED-1222 jsc#PED-1223 jsc#PED-1225 jsc#PED-2849 +Alt-commit: e72df53dcb01ec58e0410da353551adf94c8d0f1 Currently, pic_height of vdsc_cfg structure is being used to calculate slice_height, before it is set for DP. diff --git a/patches.suse/1631-drm-i915-gem-Really-move-i915_gem_context.link-under.patch b/patches.suse/1631-drm-i915-gem-Really-move-i915_gem_context.link-under.patch index 327df6b..8677bca 100644 --- a/patches.suse/1631-drm-i915-gem-Really-move-i915_gem_context.link-under.patch +++ b/patches.suse/1631-drm-i915-gem-Really-move-i915_gem_context.link-under.patch @@ -5,6 +5,7 @@ Subject: drm/i915/gem: Really move i915_gem_context.link under ref protection Git-commit: d119888b09bd567e07c6b93a07f175df88857e02 Patch-mainline: v6.0-rc7 References: jsc#PED-1166 jsc#PED-1168 jsc#PED-1170 jsc#PED-1218 jsc#PED-1220 jsc#PED-1222 jsc#PED-1223 jsc#PED-1225 jsc#PED-2849 +Alt-commit: ad3aa7c31efa5a09b0dba42e66cfdf77e0db7dc2 i915_perf assumes that it can use the i915_gem_context reference to protect its i915->gem.contexts.list iteration. However, this requires diff --git a/patches.suse/1639-drm-amd-Don-t-reset-dGPUs-if-the-system-is-going-to-.patch b/patches.suse/1639-drm-amd-Don-t-reset-dGPUs-if-the-system-is-going-to-.patch index 44e7e0c..e79bacb 100644 --- a/patches.suse/1639-drm-amd-Don-t-reset-dGPUs-if-the-system-is-going-to-.patch +++ b/patches.suse/1639-drm-amd-Don-t-reset-dGPUs-if-the-system-is-going-to-.patch @@ -5,6 +5,7 @@ Subject: drm/amd: Don't reset dGPUs if the system is going to s2idle Git-commit: 7123d39dc24dcd21ff23d75f46f926b15269b9da Patch-mainline: v5.18 References: jsc#PED-1166 jsc#PED-1168 jsc#PED-1170 jsc#PED-1218 jsc#PED-1220 jsc#PED-1222 jsc#PED-1223 jsc#PED-1225 +Alt-commit: 0223e516470aa0589da6c03e6d177c10594cabbd An A+A configuration on ASUS ROG Strix G513QY proves that the ASIC reset for handling aborted suspend can't work with s2idle. diff --git a/patches.suse/1644-i915-guc-reset-Make-__guc_reset_context-aware-of-gui.patch b/patches.suse/1644-i915-guc-reset-Make-__guc_reset_context-aware-of-gui.patch index 4a75d29..e700b6a 100644 --- a/patches.suse/1644-i915-guc-reset-Make-__guc_reset_context-aware-of-gui.patch +++ b/patches.suse/1644-i915-guc-reset-Make-__guc_reset_context-aware-of-gui.patch @@ -5,6 +5,7 @@ Subject: i915/guc/reset: Make __guc_reset_context aware of guilty engines Git-commit: 89e96d822bd51f7afe2d3e95a34099480b5c3d55 Patch-mainline: v5.18 References: jsc#PED-1166 jsc#PED-1168 jsc#PED-1170 jsc#PED-1218 jsc#PED-1220 jsc#PED-1222 jsc#PED-1223 jsc#PED-1225 +Alt-commit: 303760aa914b7f5ac9602dbb4b471a2ad52eeb3e There are 2 ways an engine can get reset in i915 and the method of reset affects how KMD labels a context as guilty/innocent. diff --git a/patches.suse/1855-drm-i915-ttm-fix-sg_table-construction.patch b/patches.suse/1855-drm-i915-ttm-fix-sg_table-construction.patch index e230cc0..2c553d2 100644 --- a/patches.suse/1855-drm-i915-ttm-fix-sg_table-construction.patch +++ b/patches.suse/1855-drm-i915-ttm-fix-sg_table-construction.patch @@ -5,6 +5,7 @@ Subject: drm/i915/ttm: fix sg_table construction MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit +Alt-commit: bc99f1209f19fefa3ee11e77464ccfae541f4291 Git-commit: aff1e0b09b54b64944b7fe32997229552737b9e9 Patch-mainline: v5.19-rc7 References: jsc#PED-1166 jsc#PED-1168 jsc#PED-1170 jsc#PED-1218 jsc#PED-1220 jsc#PED-1222 jsc#PED-1223 jsc#PED-1225 diff --git a/patches.suse/1859-drm-i915-selftests-fix-subtraction-overflow-bug.patch b/patches.suse/1859-drm-i915-selftests-fix-subtraction-overflow-bug.patch index 8753940..91d919a 100644 --- a/patches.suse/1859-drm-i915-selftests-fix-subtraction-overflow-bug.patch +++ b/patches.suse/1859-drm-i915-selftests-fix-subtraction-overflow-bug.patch @@ -5,6 +5,7 @@ Subject: drm/i915/selftests: fix subtraction overflow bug Git-commit: 333991c4e66b3d4b5613315f18016da80344f659 Patch-mainline: v5.19-rc7 References: jsc#PED-1166 jsc#PED-1168 jsc#PED-1170 jsc#PED-1218 jsc#PED-1220 jsc#PED-1222 jsc#PED-1223 jsc#PED-1225 +Alt-commit: ab3edc679c552a466e4bf0b11af3666008bd65a2 On some machines hole_end can be small enough to cause subtraction overflow. On the other side (addr + 2 * min_alignment) can overflow diff --git a/patches.suse/1866-drm-i915-ttm-fix-32b-build.patch b/patches.suse/1866-drm-i915-ttm-fix-32b-build.patch index 11cbb9c..b6c35b3 100644 --- a/patches.suse/1866-drm-i915-ttm-fix-32b-build.patch +++ b/patches.suse/1866-drm-i915-ttm-fix-32b-build.patch @@ -5,6 +5,7 @@ Subject: drm/i915/ttm: fix 32b build MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit +Alt-commit: 9306b2b2dfce6931241ef804783692cee526599c Git-commit: ced7866db39fc5c59ee05e154d4abc0977a17f6b Patch-mainline: v5.19-rc7 References: jsc#PED-1166 jsc#PED-1168 jsc#PED-1170 jsc#PED-1218 jsc#PED-1220 jsc#PED-1222 jsc#PED-1223 jsc#PED-1225 diff --git a/patches.suse/Revert-drm-amd-display-increased-min_dcfclk_mhz-and-.patch b/patches.suse/Revert-drm-amd-display-increased-min_dcfclk_mhz-and-.patch index 6d4ccd3..c34942e 100644 --- a/patches.suse/Revert-drm-amd-display-increased-min_dcfclk_mhz-and-.patch +++ b/patches.suse/Revert-drm-amd-display-increased-min_dcfclk_mhz-and-.patch @@ -5,6 +5,7 @@ Subject: [PATCH] Revert "drm/amd/display: increased min_dcfclk_mhz and min_fclk_ Git-commit: a538dabf772c169641e151834e161e241802ab33 Patch-mainline: v6.8-rc5 References: git-fixes +Alt-commit: 194bef0cc1f5ce5e2ca84d366c74be2bd9736aa3 [why]: This reverts commit 2ff33c759a4247c84ec0b7815f1f223e155ba82a. diff --git a/patches.suse/Revert-drm-amd-flush-any-delayed-gfxoff-on-suspend-e.patch b/patches.suse/Revert-drm-amd-flush-any-delayed-gfxoff-on-suspend-e.patch index 80db889..5e62794 100644 --- a/patches.suse/Revert-drm-amd-flush-any-delayed-gfxoff-on-suspend-e.patch +++ b/patches.suse/Revert-drm-amd-flush-any-delayed-gfxoff-on-suspend-e.patch @@ -5,6 +5,7 @@ Subject: [PATCH] Revert "drm/amd: flush any delayed gfxoff on suspend entry" Git-commit: 916361685319098f696b798ef1560f69ed96e934 Patch-mainline: v6.8-rc5 References: git-fixes +Alt-commit: ce311df91d73eaddc5489d4d63fb96c21e80f7cf commit ab4750332dbe ("drm/amdgpu/sdma5.2: add begin/end_use ring callbacks") caused GFXOFF control to be used more heavily and the diff --git a/patches.suse/Revert-drm-amd-pm-resolve-reboot-exception-for-si-ol.patch b/patches.suse/Revert-drm-amd-pm-resolve-reboot-exception-for-si-ol.patch index 5979b9d..12a0860 100644 --- a/patches.suse/Revert-drm-amd-pm-resolve-reboot-exception-for-si-ol.patch +++ b/patches.suse/Revert-drm-amd-pm-resolve-reboot-exception-for-si-ol.patch @@ -5,6 +5,7 @@ Subject: [PATCH] Revert "drm/amd/pm: resolve reboot exception for si oland" Git-commit: 955558030954b9637b41c97b730f9b38c92ac488 Patch-mainline: v6.8-rc7 References: git-fixes +Alt-commit: 1cb96a8a59c67bbada098e945b97c633e1f2c0aa This reverts commit e490d60a2f76bff636c68ce4fe34c1b6c34bbd86. diff --git a/patches.suse/drm-amd-Disable-ASPM-for-VI-w-all-Intel-systems.patch b/patches.suse/drm-amd-Disable-ASPM-for-VI-w-all-Intel-systems.patch index 5136433..1a01f01 100644 --- a/patches.suse/drm-amd-Disable-ASPM-for-VI-w-all-Intel-systems.patch +++ b/patches.suse/drm-amd-Disable-ASPM-for-VI-w-all-Intel-systems.patch @@ -5,6 +5,7 @@ Subject: [PATCH] drm/amd: Disable ASPM for VI w/ all Intel systems Git-commit: 64ffd2f1d00c6235dabe9704bbb0d9ce3e28147f Patch-mainline: v6.6 References: git-fixes +Alt-commit: e5f52a84bf0a817016ecd13e320fe3c3c807a83c Originally we were quirking ASPM disabled specifically for VI when used with Alder Lake, but it appears to have problems with Rocket diff --git a/patches.suse/drm-amd-Fix-detection-of-_PR3-on-the-PCIe-root-port.patch b/patches.suse/drm-amd-Fix-detection-of-_PR3-on-the-PCIe-root-port.patch index ffbac18..7794947 100644 --- a/patches.suse/drm-amd-Fix-detection-of-_PR3-on-the-PCIe-root-port.patch +++ b/patches.suse/drm-amd-Fix-detection-of-_PR3-on-the-PCIe-root-port.patch @@ -5,6 +5,7 @@ Subject: [PATCH] drm/amd: Fix detection of _PR3 on the PCIe root port Git-commit: 134b8c5d8674e7cde380f82e9aedfd46dcdd16f7 Patch-mainline: v6.6-rc5 References: git-fixes +Alt-commit: c4c8955b8acb4d88d2ca02a7dc6010e5f0c5288d On some systems with Navi3x dGPU will attempt to use BACO for runtime PM but fails to resume properly. This is because on these systems diff --git a/patches.suse/drm-amd-display-Add-smu-write-msg-id-fail-retry-proc.patch b/patches.suse/drm-amd-display-Add-smu-write-msg-id-fail-retry-proc.patch index 3fd0b26..72fa751 100644 --- a/patches.suse/drm-amd-display-Add-smu-write-msg-id-fail-retry-proc.patch +++ b/patches.suse/drm-amd-display-Add-smu-write-msg-id-fail-retry-proc.patch @@ -5,6 +5,7 @@ Subject: [PATCH] drm/amd/display: Add smu write msg id fail retry process Git-commit: 72105dcfa3d12b5af49311f857e3490baa225135 Patch-mainline: v6.6-rc1 References: git-fixes +Alt-commit: 302be1cb9f4b02995f3b10c50494d5eb8fdaf5c1 A benchmark stress test (12-40 machines x 48hours) found that DCN315 has cases where DC writes to an indirect register to set the smu clock msg diff --git a/patches.suse/drm-amd-display-Fix-memory-leak-in-dm_sw_fini.patch b/patches.suse/drm-amd-display-Fix-memory-leak-in-dm_sw_fini.patch index faeea77..5bf0896 100644 --- a/patches.suse/drm-amd-display-Fix-memory-leak-in-dm_sw_fini.patch +++ b/patches.suse/drm-amd-display-Fix-memory-leak-in-dm_sw_fini.patch @@ -5,6 +5,7 @@ Subject: [PATCH] drm/amd/display: Fix memory leak in dm_sw_fini() Git-commit: bae67893578d608e35691dcdfa90c4957debf1d3 Patch-mainline: v6.8-rc6 References: git-fixes +Alt-commit: 145242ed6f3f9b0f89f6a51cfceda3d430605d2a After destroying dmub_srv, the memory associated with it is not freed, causing a memory leak: diff --git a/patches.suse/drm-amd-display-Fix-possible-NULL-dereference-on-dev.patch b/patches.suse/drm-amd-display-Fix-possible-NULL-dereference-on-dev.patch index 063a0c4..c754ddb 100644 --- a/patches.suse/drm-amd-display-Fix-possible-NULL-dereference-on-dev.patch +++ b/patches.suse/drm-amd-display-Fix-possible-NULL-dereference-on-dev.patch @@ -5,6 +5,7 @@ Subject: [PATCH] drm/amd/display: Fix possible NULL dereference on device remove Git-commit: a589fa17cc4456df75f16fa3b49e8da0112e5100 Patch-mainline: v6.8-rc5 References: git-fixes +Alt-commit: 10cd2bf92b4cf64a9c044f2c9774461e526d526a As part of a cleanup amdgpu_dm_fini() function, which is typically called when a device is being shut down or a driver is being unloaded diff --git a/patches.suse/drm-amd-display-Fix-possible-buffer-overflow-in-find.patch b/patches.suse/drm-amd-display-Fix-possible-buffer-overflow-in-find.patch index 9e90cad..57d4bc6 100644 --- a/patches.suse/drm-amd-display-Fix-possible-buffer-overflow-in-find.patch +++ b/patches.suse/drm-amd-display-Fix-possible-buffer-overflow-in-find.patch @@ -5,6 +5,7 @@ Subject: [PATCH] drm/amd/display: Fix possible buffer overflow in 'find_dcfclk_f Git-commit: ccc514b7e7acbd301219cbaec0fc0bfe5741acee Patch-mainline: v6.8-rc5 References: git-fixes +Alt-commit: 9ab1a996dc7f492d5dea51b7c163c92a524eaf00 when 'find_dcfclk_for_voltage()' function is looping over VG_NUM_SOC_VOLTAGE_LEVELS (which is 8), but the size of the DcfClocks diff --git a/patches.suse/drm-amd-display-Remove-wait-while-locked.patch b/patches.suse/drm-amd-display-Remove-wait-while-locked.patch index 98599c2..9e471dc 100644 --- a/patches.suse/drm-amd-display-Remove-wait-while-locked.patch +++ b/patches.suse/drm-amd-display-Remove-wait-while-locked.patch @@ -5,6 +5,7 @@ Subject: [PATCH] drm/amd/display: Remove wait while locked Git-commit: 5a3ccb1400339268c5e3dc1fa044a7f6c7f59a02 Patch-mainline: v6.6-rc1 References: git-fixes +Alt-commit: d12f00c91fdfe3e50747f9e7e229fd8ede16b632 [Why] We wait for mpc idle while in a locked state, leading to potential diff --git a/patches.suse/drm-amd-display-enable-cursor-degamma-for-DCN3-DRM-l.patch b/patches.suse/drm-amd-display-enable-cursor-degamma-for-DCN3-DRM-l.patch index a991369..7fd7348 100644 --- a/patches.suse/drm-amd-display-enable-cursor-degamma-for-DCN3-DRM-l.patch +++ b/patches.suse/drm-amd-display-enable-cursor-degamma-for-DCN3-DRM-l.patch @@ -5,6 +5,7 @@ Subject: [PATCH] drm/amd/display: enable cursor degamma for DCN3+ DRM legacy gam Git-commit: 57a943ebfcdb4a97fbb409640234bdb44bfa1953 Patch-mainline: v6.6-rc1 References: git-fixes +Alt-commit: fabd2165d11649ecca5012d786a62ac149e9d83f For DRM legacy gamma, AMD display manager applies implicit sRGB degamma using a pre-defined sRGB transfer function. It works fine for DCN2 diff --git a/patches.suse/drm-amd-display-fix-access-hdcp_workqueue-assert.patch b/patches.suse/drm-amd-display-fix-access-hdcp_workqueue-assert.patch index dcbc1b1..9734bbe 100644 --- a/patches.suse/drm-amd-display-fix-access-hdcp_workqueue-assert.patch +++ b/patches.suse/drm-amd-display-fix-access-hdcp_workqueue-assert.patch @@ -8,6 +8,7 @@ Content-transfer-encoding: 8bit Git-commit: 3cf7cd3f770a0b89dc5f06e19edb52e65b93b214 Patch-mainline: v6.4-rc1 References: git-fixes +Alt-commit: cdff36a0217aadf5cbc167893ad1c0da869619cb [Why] hdcp are enabled for asics from raven. for old asics which hdcp are not enabled, hdcp_workqueue are null. some diff --git a/patches.suse/drm-amd-display-fix-the-white-screen-issue-when-64GB.patch b/patches.suse/drm-amd-display-fix-the-white-screen-issue-when-64GB.patch index 42cb75c..0610e8f 100644 --- a/patches.suse/drm-amd-display-fix-the-white-screen-issue-when-64GB.patch +++ b/patches.suse/drm-amd-display-fix-the-white-screen-issue-when-64GB.patch @@ -5,6 +5,7 @@ Subject: [PATCH] drm/amd/display: fix the white screen issue when >= 64GB DRAM Git-commit: ef064187a9709393a981a56cce1e31880fd97107 Patch-mainline: v6.6-rc2 References: git-fixes +Alt-commit: a4246c63516600ce6feb4e2ee2124b8796f7a664 Dropping bit 31:4 of page table base is wrong, it makes page table base points to wrong address if phys addr is beyond 64GB; dropping diff --git a/patches.suse/drm-amd-display-prevent-potential-division-by-zero-e.patch b/patches.suse/drm-amd-display-prevent-potential-division-by-zero-e.patch index bbc3dc2..eee953c 100644 --- a/patches.suse/drm-amd-display-prevent-potential-division-by-zero-e.patch +++ b/patches.suse/drm-amd-display-prevent-potential-division-by-zero-e.patch @@ -5,6 +5,7 @@ Subject: [PATCH] drm/amd/display: prevent potential division by zero errors Git-commit: 07e388aab042774f284a2ad75a70a194517cdad4 Patch-mainline: v6.6-rc1 References: git-fixes +Alt-commit: 084f658ece139645d203fa09c77c7f96cb849bb7 There are two places in apply_below_the_range() where it's possible for a divide by zero error to occur. So, to fix this make sure the divisor diff --git a/patches.suse/drm-amd-display-register-edp_backlight_control-for-D.patch b/patches.suse/drm-amd-display-register-edp_backlight_control-for-D.patch index 7915c34..4008f42 100644 --- a/patches.suse/drm-amd-display-register-edp_backlight_control-for-D.patch +++ b/patches.suse/drm-amd-display-register-edp_backlight_control-for-D.patch @@ -5,6 +5,7 @@ Subject: [PATCH] drm/amd/display: register edp_backlight_control() for DCN301 Git-commit: 1611917f39bee1abfc01501238db8ac19649042d Patch-mainline: v6.6-rc1 References: git-fixes +Alt-commit: ac0ec1c7d1f0d017d0ea44954026d2f138c581e4 As made mention of in commit 099303e9a9bd ("drm/amd/display: eDP intermittent black screen during PnP"), we need to turn off the diff --git a/patches.suse/drm-amd-pm-parse-pp_handle-under-appropriate-conditi.patch b/patches.suse/drm-amd-pm-parse-pp_handle-under-appropriate-conditi.patch index a098e96..2284504 100644 --- a/patches.suse/drm-amd-pm-parse-pp_handle-under-appropriate-conditi.patch +++ b/patches.suse/drm-amd-pm-parse-pp_handle-under-appropriate-conditi.patch @@ -5,6 +5,7 @@ Subject: [PATCH] drm/amd/pm: parse pp_handle under appropriate conditions Git-commit: 58d9b9a14b47c2a3da6effcbb01607ad7edc0275 Patch-mainline: v6.4-rc2 References: git-fixes +Alt-commit: 9d7a348d368ccd940242d4aa68292cf665f6d474 amdgpu_dpm_is_overdrive_supported is a common API across all asics, so we should cast pp_handle into correct structure diff --git a/patches.suse/drm-amdgpu-Correct-the-power-calcultion-for-Renior-C.patch b/patches.suse/drm-amdgpu-Correct-the-power-calcultion-for-Renior-C.patch index d7ccf82..01148b1 100644 --- a/patches.suse/drm-amdgpu-Correct-the-power-calcultion-for-Renior-C.patch +++ b/patches.suse/drm-amdgpu-Correct-the-power-calcultion-for-Renior-C.patch @@ -5,6 +5,7 @@ Subject: [PATCH] drm/amdgpu: Correct the power calcultion for Renior/Cezanne. Git-commit: c7bae4aaa5609c1fa9761c35dbcc5fcc92915222 Patch-mainline: v6.2-rc5 References: git-fixes +Alt-commit: 29c472b28ba2426839af1185b5692bf70a014a6a From smu firmware,the value of power is transferred in units of watts. diff --git a/patches.suse/drm-amdgpu-Fix-missing-error-code-in-gmc_v6-7-8-9_0_.patch b/patches.suse/drm-amdgpu-Fix-missing-error-code-in-gmc_v6-7-8-9_0_.patch index 8bd07cb..24fb302 100644 --- a/patches.suse/drm-amdgpu-Fix-missing-error-code-in-gmc_v6-7-8-9_0_.patch +++ b/patches.suse/drm-amdgpu-Fix-missing-error-code-in-gmc_v6-7-8-9_0_.patch @@ -8,6 +8,7 @@ Content-transfer-encoding: 8bit Git-commit: 16da399091dca3d1e48109086403587af37cc196 Patch-mainline: v6.8-rc3 References: git-fixes +Alt-commit: fa8a91b0e52c37627d01395baf5e7aae21dce5bf Return 0 for success scenairos in 'gmc_v6/7/8/9_0_hw_init()' diff --git a/patches.suse/drm-amdgpu-display-Initialize-gamma-correction-mode-.patch b/patches.suse/drm-amdgpu-display-Initialize-gamma-correction-mode-.patch index f5341c0..a8efce2 100644 --- a/patches.suse/drm-amdgpu-display-Initialize-gamma-correction-mode-.patch +++ b/patches.suse/drm-amdgpu-display-Initialize-gamma-correction-mode-.patch @@ -5,6 +5,7 @@ Subject: [PATCH] drm/amdgpu/display: Initialize gamma correction mode variable i Git-commit: a82197e3a5f45450cbaf92095d8a51249dc44c79 Patch-mainline: v6.8-rc5 References: git-fixes +Alt-commit: 615fb058726dcf6248fd8c90a4ad71f32031dd48 The dcn30_get_gamcor_current() function is responsible for determining the current gamma correction mode used by the display controller. diff --git a/patches.suse/drm-amdgpu-dm-dp_mst-Don-t-grab-mst_mgr-lock-when-co.patch b/patches.suse/drm-amdgpu-dm-dp_mst-Don-t-grab-mst_mgr-lock-when-co.patch index 0701596..79d4fac 100644 --- a/patches.suse/drm-amdgpu-dm-dp_mst-Don-t-grab-mst_mgr-lock-when-co.patch +++ b/patches.suse/drm-amdgpu-dm-dp_mst-Don-t-grab-mst_mgr-lock-when-co.patch @@ -5,6 +5,7 @@ Subject: [PATCH] drm/amdgpu/dm/dp_mst: Don't grab mst_mgr->lock when computing D Git-commit: 14b651b22224251b35618259da714adb0b5f10ee Patch-mainline: v6.2-rc1 References: bsc#1206843 +Alt-commit: d60b82aa4d67b2e6cf0364947a008bb7255ca4da Now that we've fixed the issue with using the incorrect topology manager, we're actually grabbing the topology manager's lock - and consequently diff --git a/patches.suse/drm-amdgpu-dm-mst-Use-the-correct-topology-mgr-point.patch b/patches.suse/drm-amdgpu-dm-mst-Use-the-correct-topology-mgr-point.patch index 0338f87..1468a0e 100644 --- a/patches.suse/drm-amdgpu-dm-mst-Use-the-correct-topology-mgr-point.patch +++ b/patches.suse/drm-amdgpu-dm-mst-Use-the-correct-topology-mgr-point.patch @@ -5,6 +5,7 @@ Subject: [PATCH] drm/amdgpu/dm/mst: Use the correct topology mgr pointer in amdg Git-commit: d3e2c664ec9a3b16a28b558d6d1acde6d997ac04 Patch-mainline: v6.2-rc1 References: bsc#1206843 +Alt-commit: dfbc00410c48a9896d4a65600be7137202517780 This bug hurt me. Basically, it appears that we've been grabbing the entirely wrong mutex in the MST DSC computation code for amdgpu! While diff --git a/patches.suse/drm-amdgpu-fix-Null-pointer-dereference-error-in-amd.patch b/patches.suse/drm-amdgpu-fix-Null-pointer-dereference-error-in-amd.patch index f3ac093..173c13c 100644 --- a/patches.suse/drm-amdgpu-fix-Null-pointer-dereference-error-in-amd.patch +++ b/patches.suse/drm-amdgpu-fix-Null-pointer-dereference-error-in-amd.patch @@ -5,6 +5,7 @@ Subject: [PATCH] drm/amdgpu: fix Null pointer dereference error in amdgpu_device Git-commit: 2a1eb1a343208ce7d6839b73d62aece343e693ff Patch-mainline: v6.4-rc6 References: git-fixes +Alt-commit: cbb63eccc05626d0d111b335e44f111a3bb92871 Use the function of amdgpu_bo_vm_destroy to handle the resource release of shadow bo. During the amdgpu_mes_self_test, shadow bo released, but diff --git a/patches.suse/drm-amdgpu-gfx10-Disable-gfxoff-before-disabling-pow.patch b/patches.suse/drm-amdgpu-gfx10-Disable-gfxoff-before-disabling-pow.patch index 74d082c..a4ea636 100644 --- a/patches.suse/drm-amdgpu-gfx10-Disable-gfxoff-before-disabling-pow.patch +++ b/patches.suse/drm-amdgpu-gfx10-Disable-gfxoff-before-disabling-pow.patch @@ -5,6 +5,7 @@ Subject: [PATCH] drm/amdgpu/gfx10: Disable gfxoff before disabling powergating. Git-commit: 8173cab3368a13cdc3cad0bd5cf14e9399b0f501 Patch-mainline: v6.4-rc3 References: git-fixes +Alt-commit: a39b52c838127a42c57d5e080dbd5204770aa80e Otherwise we get a full system lock (looks like a FW mess). diff --git a/patches.suse/drm-amdgpu-nv-Apply-ASPM-quirk-on-Intel-ADL-AMD-Navi.patch b/patches.suse/drm-amdgpu-nv-Apply-ASPM-quirk-on-Intel-ADL-AMD-Navi.patch index 5007ac3..77b2bc9 100644 --- a/patches.suse/drm-amdgpu-nv-Apply-ASPM-quirk-on-Intel-ADL-AMD-Navi.patch +++ b/patches.suse/drm-amdgpu-nv-Apply-ASPM-quirk-on-Intel-ADL-AMD-Navi.patch @@ -5,6 +5,7 @@ Subject: [PATCH] drm/amdgpu/nv: Apply ASPM quirk on Intel ADL + AMD Navi Git-commit: 2b072442f4962231a8516485012bb2d2551ef2fe Patch-mainline: v6.3-rc4 References: bsc#1206843 +Alt-commit: 3ad5dcfe00a22404dfa2168f37fecd500ab26a6d S2idle resume freeze can be observed on Intel ADL + AMD WX5500. This is caused by commit 0064b0ce85bb ("drm/amd/pm: enable ASPM by default"). diff --git a/patches.suse/drm-amdgpu-vcn-Disable-indirect-SRAM-on-Vangogh-brok.patch b/patches.suse/drm-amdgpu-vcn-Disable-indirect-SRAM-on-Vangogh-brok.patch index d5295e6..ec4f752 100644 --- a/patches.suse/drm-amdgpu-vcn-Disable-indirect-SRAM-on-Vangogh-brok.patch +++ b/patches.suse/drm-amdgpu-vcn-Disable-indirect-SRAM-on-Vangogh-brok.patch @@ -5,6 +5,7 @@ Subject: [PATCH] drm/amdgpu/vcn: Disable indirect SRAM on Vangogh broken BIOSes Git-commit: 542a56e8eb4467ae654eefab31ff194569db39cd Patch-mainline: v6.3-rc3 References: git-fixes +Alt-commit: 1aff0a5d71d23be6658f893c88c6a9791202bcb1 The VCN firmware loading path enables the indirect SRAM mode if it's advertised as supported. We might have some cases of FW issues that diff --git a/patches.suse/drm-bridge-sii902x-Fix-probing-race-issue.patch b/patches.suse/drm-bridge-sii902x-Fix-probing-race-issue.patch index 5cc9afc..85bdeeb 100644 --- a/patches.suse/drm-bridge-sii902x-Fix-probing-race-issue.patch +++ b/patches.suse/drm-bridge-sii902x-Fix-probing-race-issue.patch @@ -5,6 +5,7 @@ Subject: drm/bridge: sii902x: Fix probing race issue Git-commit: 08ac6f132dd77e40f786d8af51140c96c6d739c9 Patch-mainline: v6.8-rc2 References: bsc#1220736 CVE-2024-26607 +Alt-commit: dffdfb8f5de13a8cacf9589ceb70fcf1442d4803 A null pointer dereference crash has been observed rarely on TI platforms using sii9022 bridge: diff --git a/patches.suse/drm-display-dp_mst-Fix-drm_dp_mst_add_affected_dsc_c.patch b/patches.suse/drm-display-dp_mst-Fix-drm_dp_mst_add_affected_dsc_c.patch index 2fa1dcd..4cf3071 100644 --- a/patches.suse/drm-display-dp_mst-Fix-drm_dp_mst_add_affected_dsc_c.patch +++ b/patches.suse/drm-display-dp_mst-Fix-drm_dp_mst_add_affected_dsc_c.patch @@ -5,6 +5,7 @@ Subject: [PATCH] drm/display/dp_mst: Fix drm_dp_mst_add_affected_dsc_crtcs() ret Git-commit: 2f3a1273862cb82cca227630cc7f04ce0c94b6bb Patch-mainline: v6.1-rc7 References: git-fixes +Alt-commit: a3ae99598b9f89dd5ac9d2900fb7ffb3dcfa3d35 Looks like that we're accidentally dropping a pretty important return code here. For some reason, we just return -EINVAL if we fail to get the MST diff --git a/patches.suse/drm-i915-Fix-negative-value-passed-as-remaining-time.patch b/patches.suse/drm-i915-Fix-negative-value-passed-as-remaining-time.patch index 6899bc5..050fdbf 100644 --- a/patches.suse/drm-i915-Fix-negative-value-passed-as-remaining-time.patch +++ b/patches.suse/drm-i915-Fix-negative-value-passed-as-remaining-time.patch @@ -6,6 +6,7 @@ Git-commit: a8899b8728013c7b2456f0bfa20e5fea85ee0fd1 Patch-mainline: v6.1-rc8 Alt-commit: f235dbd5b768e238d365fd05d92de5a32abc1c1f References: git-fixes +Alt-commit: 6a5347293992e0412bc748dae11228a7081393fa Commit b97060a99b01 ("drm/i915/guc: Update intel_gt_wait_for_idle to work with GuC") extended the API of intel_gt_retire_requests_timeout() with an diff --git a/patches.suse/drm-i915-Never-return-0-if-not-all-requests-retired.patch b/patches.suse/drm-i915-Never-return-0-if-not-all-requests-retired.patch index e2909cb..de7757f 100644 --- a/patches.suse/drm-i915-Never-return-0-if-not-all-requests-retired.patch +++ b/patches.suse/drm-i915-Never-return-0-if-not-all-requests-retired.patch @@ -6,6 +6,7 @@ Git-commit: 12b8b046e4c9de40fa59b6f067d6826f4e688f68 Patch-mainline: v6.1-rc8 Alt-commit: f301a29f143760ce8d3d6b6a8436d45d3448cde6 References: git-fixes +Alt-commit: 35aba5f51a39fb95351844ffb14ec02b8970e19f Users of intel_gt_retire_requests_timeout() expect 0 return value on success. However, we have no protection from passing back 0 potentially diff --git a/patches.suse/drm-i915-dp-Fix-passing-the-correct-DPCD_REV-for-drm.patch b/patches.suse/drm-i915-dp-Fix-passing-the-correct-DPCD_REV-for-drm.patch index aa9a85b..993dfee 100644 --- a/patches.suse/drm-i915-dp-Fix-passing-the-correct-DPCD_REV-for-drm.patch +++ b/patches.suse/drm-i915-dp-Fix-passing-the-correct-DPCD_REV-for-drm.patch @@ -6,6 +6,7 @@ Git-commit: 2bd7a06a1208aaacb4e7a2a5436c23bce8d70801 Patch-mainline: v6.7 Alt-commit: 3ee302ec22d6e1d7d1e6d381b0d507ee80f2135c References: git-fixes +Alt-commit: 3d9e9020b92288871b02f194c3ec88e03a1afa88 Using link_status to get DPCD_REV fails when disabling/defaulting phy pattern. Use intel_dp->dpcd to access DPCD_REV correctly.