From 7bfd186021e68844ac629ddfc6b67a3658f82f3f Mon Sep 17 00:00:00 2001 From: Reza Jahanbakhshi Date: Thu, 8 Dec 2022 12:58:21 +0100 Subject: [PATCH] added i915 intel gallium driver added patches to fix a hang where the HW is hung on a PIPE_CONTROL after a GPGPU_WALKER --- .SRCINFO | 8 +- ...INTERFACE_DESCRIPTOR_LOAD-reemit-aft.patch | 40 +++++++++ ...mask-the-bottom-bits-of-the-sampler-.patch | 83 +++++++++++++++++++ PKGBUILD | 10 ++- 4 files changed, 138 insertions(+), 3 deletions(-) create mode 100644 0001-anv-force-MEDIA_INTERFACE_DESCRIPTOR_LOAD-reemit-aft.patch create mode 100644 0002-intel-fs-always-mask-the-bottom-bits-of-the-sampler-.patch diff --git a/.SRCINFO b/.SRCINFO index ee44928..736aa69 100644 --- a/.SRCINFO +++ b/.SRCINFO @@ -1,6 +1,6 @@ pkgbase = mesa-git pkgdesc = an open-source implementation of the OpenGL specification, git version - pkgver = 23.0.0_devel.162765.4ceaed7839a.d41d8cd98f00b204e9800998ecf8427e + pkgver = 23.0.0_devel.163974.bafbe7c23a1.d41d8cd98f00b204e9800998ecf8427e pkgrel = 1 url = https://www.mesa3d.org arch = x86_64 @@ -62,10 +62,16 @@ pkgbase = mesa-git conflicts = vulkan-swrast conflicts = mesa-libgl source = mesa::git+https://gitlab.freedesktop.org/mesa/mesa.git#branch=main + source = 0001-anv-force-MEDIA_INTERFACE_DESCRIPTOR_LOAD-reemit-aft.patch + source = 0002-intel-fs-always-mask-the-bottom-bits-of-the-sampler-.patch source = LICENSE md5sums = SKIP + md5sums = 102af8525e4c1f266cc54b038d1b9314 + md5sums = 094f900983f68bec0325bd29d4789ad5 md5sums = 5c65a0fe315dd347e09b1f2826a1df5a sha512sums = SKIP + sha512sums = 4ff7c359f08aedb1fb6eb2c6bb34bafa399edfa7916cd6cce7844bb38795e84c0265324e3fc7d37237d4824f029cb04da176bcf476785fd2e35a8b0ce4f8a394 + sha512sums = 5dd0cb8affa9cfe6e7d94f59b8e23727036fd8ab76938321f8d266315f30611584da6f6277fe2aa920130483302adab5e57e2bc08f1bd3c62ea57b3e4b007305 sha512sums = 25da77914dded10c1f432ebcbf29941124138824ceecaf1367b3deedafaecabc082d463abcfa3d15abff59f177491472b505bcb5ba0c4a51bb6b93b4721a23c2 pkgname = mesa-git diff --git a/0001-anv-force-MEDIA_INTERFACE_DESCRIPTOR_LOAD-reemit-aft.patch b/0001-anv-force-MEDIA_INTERFACE_DESCRIPTOR_LOAD-reemit-aft.patch new file mode 100644 index 0000000..19c21a0 --- /dev/null +++ b/0001-anv-force-MEDIA_INTERFACE_DESCRIPTOR_LOAD-reemit-aft.patch @@ -0,0 +1,40 @@ +From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001 +From: Lionel Landwerlin +Date: Sun, 12 Jun 2022 23:59:05 +0300 +Subject: [PATCH] anv: force MEDIA_INTERFACE_DESCRIPTOR_LOAD reemit after + 3D->GPGPU switch + +Seems to fix a hang in Age of Empire 4 where the HW is hung on a +PIPE_CONTROL after a GPGPU_WALKER but no +MEDIA_INTERFACE_DESCRIPTOR_LOAD was emitted since the switch from 3D +to GPGPU. + +This would happen in the following case : + + vkCmdBindPipeline(COMPUTE, cs_pipeline); + vkCmdDispatch(...); + vkCmdBindPipeline(GRAPHICS, gfx_pipeline); + vkCmdDraw(...); + vkCmdDispatch(...); + +Signed-off-by: Lionel Landwerlin +--- + src/intel/vulkan/genX_cmd_buffer.c | 5 +++++ + 1 file changed, 5 insertions(+) + +diff --git a/src/intel/vulkan/genX_cmd_buffer.c b/src/intel/vulkan/genX_cmd_buffer.c +index 1aa997f2ade2..2e9979c8f63e 100644 +--- a/src/intel/vulkan/genX_cmd_buffer.c ++++ b/src/intel/vulkan/genX_cmd_buffer.c +@@ -5955,6 +5955,11 @@ genX(flush_pipeline_select)(struct anv_cmd_buffer *cmd_buffer, + } + #endif + ++#if GFX_VERx10 == 120 ++ if (pipeline == _3D) ++ cmd_buffer->state.compute.pipeline_dirty = true; ++#endif ++ + /* From "BXML » GT » MI » vol1a GPU Overview » [Instruction] + * PIPELINE_SELECT [DevBWR+]": + * diff --git a/0002-intel-fs-always-mask-the-bottom-bits-of-the-sampler-.patch b/0002-intel-fs-always-mask-the-bottom-bits-of-the-sampler-.patch new file mode 100644 index 0000000..ef12fda --- /dev/null +++ b/0002-intel-fs-always-mask-the-bottom-bits-of-the-sampler-.patch @@ -0,0 +1,83 @@ +From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001 +From: Lionel Landwerlin +Date: Sat, 25 Jun 2022 23:38:45 +0300 +Subject: [PATCH] intel/fs: always mask the bottom bits of the sampler extended + descriptor + +Fixes a hang in Age Of Empire 4. The HW is hang with the sampler input +unit busy. Replaying on simulation showed the extended message length +in the extended descriptor is invalid. Since the Anv ensures the input +is correct in anv_surface_state_to_handle(), the likely reason for +this issue is the use of VK_VALVE_mutable_descriptor_type and the +application leaving a previous value for a different descriptor type. + +Signed-off-by: Lionel Landwerlin +--- + src/intel/compiler/brw_fs.cpp | 2 +- + .../compiler/brw_lower_logical_sends.cpp | 20 +++++++++++++++---- + 2 files changed, 17 insertions(+), 5 deletions(-) + +diff --git a/src/intel/compiler/brw_fs.cpp b/src/intel/compiler/brw_fs.cpp +index 624454676031..061eb7d603bb 100644 +--- a/src/intel/compiler/brw_fs.cpp ++++ b/src/intel/compiler/brw_fs.cpp +@@ -4439,7 +4439,7 @@ brw_fb_write_msg_control(const fs_inst *inst, + return mctl; + } + +- /** ++/** + * Predicate the specified instruction on the sample mask. + */ + void +diff --git a/src/intel/compiler/brw_lower_logical_sends.cpp b/src/intel/compiler/brw_lower_logical_sends.cpp +index 1ff064d342ae..90cb00daeb9b 100644 +--- a/src/intel/compiler/brw_lower_logical_sends.cpp ++++ b/src/intel/compiler/brw_lower_logical_sends.cpp +@@ -1117,30 +1117,42 @@ lower_sampler_logical_send_gfx7(const fs_builder &bld, fs_inst *inst, opcode op, + inst->src[1] = brw_imm_ud(0); + } else if (surface_handle.file != BAD_FILE) { + /* Bindless surface */ ++ const fs_builder ubld = bld.group(1, 0).exec_all(); + assert(devinfo->ver >= 9); + inst->desc = brw_sampler_desc(devinfo, + GFX9_BTI_BINDLESS, + sampler.file == IMM ? sampler.ud % 16 : 0, + msg_type, + simd_mode, + 0 /* return_format unused on gfx7+ */); + + /* For bindless samplers, the entire address is included in the message + * header so we can leave the portion in the message descriptor 0. + */ + if (sampler_handle.file != BAD_FILE || sampler.file == IMM) { + inst->src[0] = brw_imm_ud(0); + } else { +- const fs_builder ubld = bld.group(1, 0).exec_all(); + fs_reg desc = ubld.vgrf(BRW_REGISTER_TYPE_UD); + ubld.SHL(desc, sampler, brw_imm_ud(8)); + inst->src[0] = desc; + } + +- /* We assume that the driver provided the handle in the top 20 bits so +- * we can use the surface handle directly as the extended descriptor. ++ /* We previously assumed that the driver provided the handle in the top ++ * 20 bits (leaving the bottom 12 bits at 0). But with extensions like ++ * VK_VALVE_mutable_descriptor_type, the application is more in control ++ * of the content of VkDescriptors which is where we store ++ * surface/sampler offsets. We experience GPU hangs because the ++ * application left an invalid value in the descriptor (probably used ++ * for another descriptor type than sampler) and the lower 12bits of the ++ * surface handle overlapping with the extended descriptor length make ++ * the HW hang. The following AND() clears those bits and fixes a hang ++ * in Age Of Empire 4. + */ +- inst->src[1] = retype(surface_handle, BRW_REGISTER_TYPE_UD); ++ fs_reg ex_desc = ubld.vgrf(BRW_REGISTER_TYPE_UD); ++ ubld.AND(ex_desc, ++ retype(surface_handle, BRW_REGISTER_TYPE_UD), ++ brw_imm_ud(INTEL_MASK(31, 12))); ++ inst->src[1] = component(ex_desc, 0); + } else { + /* Immediate portion of the descriptor */ + inst->desc = brw_sampler_desc(devinfo, diff --git a/PKGBUILD b/PKGBUILD index dd196e5..e439d71 100644 --- a/PKGBUILD +++ b/PKGBUILD @@ -12,7 +12,7 @@ pkgname=mesa-git pkgdesc="an open-source implementation of the OpenGL specification, git version" -pkgver=23.0.0_devel.162765.4ceaed7839a.d41d8cd98f00b204e9800998ecf8427e +pkgver=23.0.0_devel.163975.a3249415e23.5269a95f00c4d6964d487d9dbd94f62b pkgrel=1 arch=('x86_64') makedepends=('git' 'python-mako' 'xorgproto' @@ -26,10 +26,16 @@ conflicts=('mesa' 'opencl-mesa' 'vulkan-intel' 'vulkan-radeon' 'vulkan-mesa-laye url="https://www.mesa3d.org" license=('custom') source=('mesa::git+https://gitlab.freedesktop.org/mesa/mesa.git#branch=main' + '0001-anv-force-MEDIA_INTERFACE_DESCRIPTOR_LOAD-reemit-aft.patch' + '0002-intel-fs-always-mask-the-bottom-bits-of-the-sampler-.patch' 'LICENSE') md5sums=('SKIP' + '102af8525e4c1f266cc54b038d1b9314' + '094f900983f68bec0325bd29d4789ad5' '5c65a0fe315dd347e09b1f2826a1df5a') sha512sums=('SKIP' + '4ff7c359f08aedb1fb6eb2c6bb34bafa399edfa7916cd6cce7844bb38795e84c0265324e3fc7d37237d4824f029cb04da176bcf476785fd2e35a8b0ce4f8a394' + '5dd0cb8affa9cfe6e7d94f59b8e23727036fd8ab76938321f8d266315f30611584da6f6277fe2aa920130483302adab5e57e2bc08f1bd3c62ea57b3e4b007305' '25da77914dded10c1f432ebcbf29941124138824ceecaf1367b3deedafaecabc082d463abcfa3d15abff59f177491472b505bcb5ba0c4a51bb6b93b4721a23c2') # NINJAFLAGS is an env var used to pass commandline options to ninja @@ -118,7 +124,7 @@ build () { -D b_ndebug=true \ -D b_lto=false \ -D platforms=x11,wayland \ - -D gallium-drivers=r300,r600,radeonsi,nouveau,virgl,svga,swrast,iris,crocus,zink,d3d12 \ + -D gallium-drivers=r300,r600,radeonsi,nouveau,virgl,svga,swrast,i915,iris,crocus,zink,d3d12 \ -D vulkan-drivers=amd,intel,swrast,virtio-experimental,intel_hasvk \ -D vulkan-layers=device-select,overlay \ -D dri3=enabled \