VERSION: bump for 23.3.0

ac/nir: fix partial mesh shader output writes on GFX11
Fixes dEQP-VK.mesh_shader.ext.smoke.monolithic.mesh_shader_triangle with nir_opt_combine_stores disabled. Signed-off-by: Rhys Perry <pendingchaos02@gmail.com> Reviewed-by: Timur Kristóf <timur.kristof@gmail.com> Fixes: 240e16fc8e ("ac/nir/ngg: Use attribute ring for mesh shader params.") Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25530> (cherry picked from commit 2d98236dd5)
2023-11-29 19:30:57 +00:00 · 2023-11-28 13:04:20 +00:00 · 2023-11-28 13:04:19 +00:00 · 2023-11-28 13:04:17 +00:00 · 2023-11-28 13:03:57 +00:00 · 2023-11-28 13:03:56 +00:00
10 changed files with 270 additions and 14 deletions
--- a/.pick_status.json
+++ b/.pick_status.json
@@ -1,4 +1,244 @@
 [
+    {
+        "sha": "f843b14c171299e1696ca6d971ccaa496f60c3ab",
+        "description": "d3d12: Fix hevc encoder 32-bit build (uint64_t -> size_t)",
+        "nominated": false,
+        "nomination_type": 3,
+        "resolution": 4,
+        "main_sha": null,
+        "because_sha": null,
+        "notes": null
+    },
+    {
+        "sha": "ae62fc01fa95d7121451ae10d1988a73ff645c9b",
+        "description": "d3d12: Fix h264 encoder 32-bit build (uint64_t -> size_t)",
+        "nominated": false,
+        "nomination_type": 3,
+        "resolution": 4,
+        "main_sha": null,
+        "because_sha": null,
+        "notes": null
+    },
+    {
+        "sha": "500955b6cbc367439549a92498201d34b1f3d5d9",
+        "description": "nak: Only insert barriers around ifs if they actually re-converge",
+        "nominated": false,
+        "nomination_type": 3,
+        "resolution": 4,
+        "main_sha": null,
+        "because_sha": null,
+        "notes": null
+    },
+    {
+        "sha": "804201a3d70af2c78612052ab3e9c13ee0bdbbcb",
+        "description": "nak: Run rustfmt",
+        "nominated": false,
+        "nomination_type": 3,
+        "resolution": 4,
+        "main_sha": null,
+        "because_sha": null,
+        "notes": null
+    },
+    {
+        "sha": "e93935dd0497c1908c5cb6558258266bba07230e",
+        "description": "nvk: Limit shader stages to supported stages",
+        "nominated": false,
+        "nomination_type": 1,
+        "resolution": 4,
+        "main_sha": null,
+        "because_sha": "c7c73d6d17b38a2f4c899a34614eb3750eea6034",
+        "notes": null
+    },
+    {
+        "sha": "6f9be9a2a055ecbf7850eccae2cb2c819469e241",
+        "description": "hasvk: ensure we reapply always pipeline dynamic state in runtime state",
+        "nominated": false,
+        "nomination_type": 3,
+        "resolution": 4,
+        "main_sha": null,
+        "because_sha": null,
+        "notes": null
+    },
+    {
+        "sha": "2d98236dd525622527df22179756f7b0f1466afa",
+        "description": "ac/nir: fix partial mesh shader output writes on GFX11",
+        "nominated": true,
+        "nomination_type": 1,
+        "resolution": 1,
+        "main_sha": null,
+        "because_sha": "240e16fc8e27ba76afa1bf5adbd6d70c680027ac",
+        "notes": null
+    },
+    {
+        "sha": "f59665bb62b579c6cc26155a26ce557f9f3357d2",
+        "description": "venus: Do not submit batch manually when no feedback is required",
+        "nominated": true,
+        "nomination_type": 1,
+        "resolution": 1,
+        "main_sha": null,
+        "because_sha": "a55d26b566f1ba67f770f689e4de9b0f70c0d47b",
+        "notes": null
+    },
+    {
+        "sha": "1924cdc2898b0a14a8afd13db3a06dab9e1d032e",
+        "description": "d3d12: Fix multidimensional array ordering",
+        "nominated": true,
+        "nomination_type": 1,
+        "resolution": 1,
+        "main_sha": null,
+        "because_sha": "a6740ee7a4de17ce05fa10c872d6a6f08b39b77c",
+        "notes": null
+    },
+    {
+        "sha": "da3f3a46b19ade5de11d964cba1f495c5c2c9615",
+        "description": "ci: uprev vkd3d-proton to 2.11",
+        "nominated": false,
+        "nomination_type": 3,
+        "resolution": 4,
+        "main_sha": null,
+        "because_sha": null,
+        "notes": null
+    },
+    {
+        "sha": "cf510e38a51250702ec90bece7a9d397669ae28f",
+        "description": "intel/ci: fix .hasvk-manual-rules",
+        "nominated": true,
+        "nomination_type": 1,
+        "resolution": 1,
+        "main_sha": null,
+        "because_sha": "570acf56554a619ca2bf6ed697c876c6b12d7b31",
+        "notes": null
+    },
+    {
+        "sha": "19420731123e850e10c68eb3cd672b00561c48d7",
+        "description": "intel/perf: fix regex escaping",
+        "nominated": true,
+        "nomination_type": 1,
+        "resolution": 1,
+        "main_sha": null,
+        "because_sha": "aa04b47c6e43e6cc35611a43c1712ef750f161d9",
+        "notes": null
+    },
+    {
+        "sha": "1492d24f89f6e156d83d6f579d1d308eaae996ef",
+        "description": "lp: make sure 0xff is unsigned before shifting it past signed int range",
+        "nominated": false,
+        "nomination_type": 3,
+        "resolution": 4,
+        "main_sha": null,
+        "because_sha": null,
+        "notes": null
+    },
+    {
+        "sha": "023fa0aa5d3ab9baa4a9061d9d7c74e20b723d1f",
+        "description": "etnaviv: Mark etna_rs_gen_clear_surface(..) private",
+        "nominated": false,
+        "nomination_type": 3,
+        "resolution": 4,
+        "main_sha": null,
+        "because_sha": null,
+        "notes": null
+    },
+    {
+        "sha": "9342544ca5c9ec2d7c100fe80f3cb6ac41547231",
+        "description": "etnaviv: rs: Call etna_rs_gen_clear_surface(..) when needed",
+        "nominated": false,
+        "nomination_type": 3,
+        "resolution": 4,
+        "main_sha": null,
+        "because_sha": null,
+        "notes": null
+    },
+    {
+        "sha": "945288ffaecac106c978d10cd4d8512fa2992c47",
+        "description": "radeonsi: check sctx->tess_rings is valid before using it",
+        "nominated": true,
+        "nomination_type": 1,
+        "resolution": 1,
+        "main_sha": null,
+        "because_sha": "c89ca3b47f11ce2c2e6953d37590021e89c1d119",
+        "notes": null
+    },
+    {
+        "sha": "b6e98677c38a479aa187bc48a1acd5bd8049d4a1",
+        "description": "nir/print: print PATCH0 and VARn_16BIT names instead of numbers for TCS and TES",
+        "nominated": false,
+        "nomination_type": 3,
+        "resolution": 4,
+        "main_sha": null,
+        "because_sha": null,
+        "notes": null
+    },
+    {
+        "sha": "5c8730ebe880d05a36462904e401ed02afae7c4f",
+        "description": "nir: don't declare illegal varyings in nir_create_passthrough_tcs",
+        "nominated": false,
+        "nomination_type": 3,
+        "resolution": 4,
+        "main_sha": null,
+        "because_sha": null,
+        "notes": null
+    },
+    {
+        "sha": "7a9b73fcb81126faa92397d82b627db64e1b8e0f",
+        "description": "nir: fix gathering TESS_LEVEL_INNER/OUTER usage with lowered IO",
+        "nominated": true,
+        "nomination_type": 1,
+        "resolution": 1,
+        "main_sha": null,
+        "because_sha": "10be706778bd670197a66765c550cbb3a0cfda6d",
+        "notes": null
+    },
+    {
+        "sha": "827bbe48298b7e32d46c1b9b777511109762f305",
+        "description": "ci: use released version of meson",
+        "nominated": false,
+        "nomination_type": 3,
+        "resolution": 4,
+        "main_sha": null,
+        "because_sha": null,
+        "notes": null
+    },
+    {
+        "sha": "698344b93c49a9f3a257a0ef4546edf5cd3a9130",
+        "description": "d3d12/driconf: Force on ARB_texture_view for Blender",
+        "nominated": false,
+        "nomination_type": 3,
+        "resolution": 4,
+        "main_sha": null,
+        "because_sha": null,
+        "notes": null
+    },
+    {
+        "sha": "9feecda201d632ec0300d98b9f89a10d309d3f03",
+        "description": "docs: add another -rc",
+        "nominated": false,
+        "nomination_type": 3,
+        "resolution": 4,
+        "main_sha": null,
+        "because_sha": null,
+        "notes": null
+    },
+    {
+        "sha": "69d1e29dc318bb0f1c395c9a9ba1a94056d4dbef",
+        "description": "docs: update calendar for 23.0.0-rc5",
+        "nominated": false,
+        "nomination_type": 3,
+        "resolution": 4,
+        "main_sha": null,
+        "because_sha": null,
+        "notes": null
+    },
+    {
+        "sha": "28ae3210e1697c0506ca2581ebfceac22992da8f",
+        "description": "nvk: Wire up MESA_VK_VERSION_OVERRIDE",
+        "nominated": false,
+        "nomination_type": 3,
+        "resolution": 4,
+        "main_sha": null,
+        "because_sha": null,
+        "notes": null
+    },
    {
        "sha": "b07a58157d0b110dbc09a42cffe7046c3200dd3b",
        "description": "radeonsi: remove the LAYER output if the framebuffer state has only 1 layer",
--- a/2
+++ b/2
@@ -1 +1 @@
-23.3.0-rc5
+23.3.0
--- a/src/amd/common/ac_nir_lower_ngg.c
+++ b/src/amd/common/ac_nir_lower_ngg.c
@@ -3903,6 +3903,7 @@ ms_store_arrayed_output_intrin(nir_builder *b,
      nir_def *soffset = nir_load_ring_attr_offset_amd(b);
      nir_store_buffer_amd(b, store_val, ring, base_addr_off, soffset, arr_index,
                           .base = const_off + param_offset * 16,
+                           .write_mask = write_mask,
                           .memory_modes = nir_var_shader_out,
                           .access = ACCESS_COHERENT | ACCESS_IS_SWIZZLED_AMD);
   } else if (out_mode == ms_out_mode_var) {
--- a/src/compiler/nir/nir_gather_info.c
+++ b/src/compiler/nir/nir_gather_info.c
@@ -410,10 +410,16 @@ gather_intrinsic_info(nir_intrinsic_instr *instr, nir_shader *shader,
 {
   uint64_t slot_mask = 0;
   uint16_t slot_mask_16bit = 0;
+   bool is_patch_special = false;

   if (nir_intrinsic_infos[instr->intrinsic].index_map[NIR_INTRINSIC_IO_SEMANTICS] > 0) {
      nir_io_semantics semantics = nir_intrinsic_io_semantics(instr);

+      is_patch_special = semantics.location == VARYING_SLOT_TESS_LEVEL_INNER ||
+                         semantics.location == VARYING_SLOT_TESS_LEVEL_OUTER ||
+                         semantics.location == VARYING_SLOT_BOUNDING_BOX0 ||
+                         semantics.location == VARYING_SLOT_BOUNDING_BOX1;
+
      if (semantics.location >= VARYING_SLOT_PATCH0 &&
          semantics.location <= VARYING_SLOT_PATCH31) {
         /* Generic per-patch I/O. */
@@ -516,7 +522,8 @@ gather_intrinsic_info(nir_intrinsic_instr *instr, nir_shader *shader,
   case nir_intrinsic_load_input_vertex:
   case nir_intrinsic_load_interpolated_input:
      if (shader->info.stage == MESA_SHADER_TESS_EVAL &&
-          instr->intrinsic == nir_intrinsic_load_input) {
+          instr->intrinsic == nir_intrinsic_load_input &&
+          !is_patch_special) {
         shader->info.patch_inputs_read |= slot_mask;
         if (!nir_src_is_const(*nir_get_io_offset_src(instr)))
            shader->info.patch_inputs_read_indirectly |= slot_mask;
@@ -541,7 +548,8 @@ gather_intrinsic_info(nir_intrinsic_instr *instr, nir_shader *shader,
   case nir_intrinsic_load_per_vertex_output:
   case nir_intrinsic_load_per_primitive_output:
      if (shader->info.stage == MESA_SHADER_TESS_CTRL &&
-          instr->intrinsic == nir_intrinsic_load_output) {
+          instr->intrinsic == nir_intrinsic_load_output &&
+          !is_patch_special) {
         shader->info.patch_outputs_read |= slot_mask;
         if (!nir_src_is_const(*nir_get_io_offset_src(instr)))
            shader->info.patch_outputs_accessed_indirectly |= slot_mask;
@@ -575,7 +583,8 @@ gather_intrinsic_info(nir_intrinsic_instr *instr, nir_shader *shader,
   case nir_intrinsic_store_per_vertex_output:
   case nir_intrinsic_store_per_primitive_output:
      if (shader->info.stage == MESA_SHADER_TESS_CTRL &&
-          instr->intrinsic == nir_intrinsic_store_output) {
+          instr->intrinsic == nir_intrinsic_store_output &&
+          !is_patch_special) {
         shader->info.patch_outputs_written |= slot_mask;
         if (!nir_src_is_const(*nir_get_io_offset_src(instr)))
            shader->info.patch_outputs_accessed_indirectly |= slot_mask;
--- a/src/gallium/drivers/d3d12/d3d12_bufmgr.h
+++ b/src/gallium/drivers/d3d12/d3d12_bufmgr.h
@@ -71,7 +71,7 @@ struct d3d12_bo {
   uint8_t local_reference_mask[16];

   d3d12_context_state_table_entry local_context_states[16];
-   uint8_t local_reference_state[8][16];
+   uint8_t local_reference_state[16][8];
 };

 struct d3d12_buffer {
--- a/src/gallium/drivers/radeonsi/si_shader_info.c
+++ b/src/gallium/drivers/radeonsi/si_shader_info.c
@@ -617,7 +617,7 @@ void si_nir_scan_shader(struct si_screen *sscreen, const struct nir_shader *nir,
   }

   /* tess factors are loaded as input instead of system value */
-   info->reads_tess_factors = nir->info.patch_inputs_read &
+   info->reads_tess_factors = nir->info.inputs_read &
      (BITFIELD64_BIT(VARYING_SLOT_TESS_LEVEL_INNER) |
       BITFIELD64_BIT(VARYING_SLOT_TESS_LEVEL_OUTER));

--- a/src/gallium/drivers/radeonsi/si_state_shaders.cpp
+++ b/src/gallium/drivers/radeonsi/si_state_shaders.cpp
@@ -4274,8 +4274,16 @@ static void si_set_patch_vertices(struct pipe_context *ctx, uint8_t patch_vertic
   if (sctx->patch_vertices != patch_vertices) {
      sctx->patch_vertices = patch_vertices;
      si_update_tess_in_out_patch_vertices(sctx);
-      if (sctx->shader.tcs.current)
-         si_update_tess_io_layout_state(sctx);
+      if (sctx->shader.tcs.current) {
+         /* Update the io layout now if possible,
+          * otherwise make sure it's done by si_update_shaders.
+          */
+         if (sctx->tess_rings)
+            si_update_tess_io_layout_state(sctx);
+         else
+            sctx->do_update_shaders = true;
+      }
+
   }
 }

--- a/src/intel/ci/gitlab-ci-inc.yml
+++ b/src/intel/ci/gitlab-ci-inc.yml
@@ -118,7 +118,7 @@
    - !reference [.vulkan-manual-rules, rules]
    - changes:
      - src/intel/**/*
-      when: on_success
+      when: manual

 # ruleset to trigger on changes affecting either anv or iris, for jobs using both (piglit, skqp)
 .intel-rules:
--- a/src/intel/perf/gen_perf.py
+++ b/src/intel/perf/gen_perf.py
@@ -251,10 +251,10 @@ hw_vars["$QueryMode"] = "perf->sys_vars.query_mode"
 def resolve_variable(name, set, allow_counters):
    if name in hw_vars:
        return hw_vars[name]
-    m = re.search('\$GtSlice([0-9]+)$', name)
+    m = re.search(r'\$GtSlice([0-9]+)$', name)
    if m:
        return 'intel_device_info_slice_available(&perf->devinfo, {0})'.format(m.group(1))
-    m = re.search('\$GtSlice([0-9]+)XeCore([0-9]+)$', name)
+    m = re.search(r'\$GtSlice([0-9]+)XeCore([0-9]+)$', name)
    if m:
        return 'intel_device_info_subslice_available(&perf->devinfo, {0}, {1})'.format(m.group(1), m.group(2))
    if allow_counters and name in set.counter_vars:
--- a/src/virtio/vulkan/vn_queue.c
+++ b/src/virtio/vulkan/vn_queue.c
@@ -1405,9 +1405,7 @@ vn_QueueBindSparse(VkQueue queue,
   /* if feedback isn't used in the batch, can directly submit */
   if (!submit.has_feedback_fence && !submit.has_feedback_semaphore &&
       !submit.has_feedback_query) {
-      result = vn_queue_bind_sparse_submit(&submit);
-      if (result != VK_SUCCESS)
-         return result;
+      return vn_queue_bind_sparse_submit(&submit);
   }

   for (uint32_t i = 0; i < submit.batch_count; i++) {
Author	SHA1	Message	Date
Eric Engestrom	1fbdd37d4c	VERSION: bump for 23.3.0	2023-11-29 19:30:57 +00:00
Rhys Perry	421a8aaff1	ac/nir: fix partial mesh shader output writes on GFX11 Fixes dEQP-VK.mesh_shader.ext.smoke.monolithic.mesh_shader_triangle with nir_opt_combine_stores disabled. Signed-off-by: Rhys Perry <pendingchaos02@gmail.com> Reviewed-by: Timur Kristóf <timur.kristof@gmail.com> Fixes: `240e16fc8e` ("ac/nir/ngg: Use attribute ring for mesh shader params.") Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/25530> (cherry picked from commit `2d98236dd5`)	2023-11-28 13:04:20 +00:00
Mary Guillemard	dced0dba02	venus: Do not submit batch manually when no feedback is required This fixes hangs with Zink on piglit spec@arb_sparse_buffer tests caused by the double submission. Fixes: `a55d26b566` ("venus: add back sparse binding support") Signed-off-by: Mary Guillemard <mary.guillemard@collabora.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26375> (cherry picked from commit `f59665bb62`)	2023-11-28 13:04:19 +00:00
Jesse Natalie	3ff4768692	d3d12: Fix multidimensional array ordering Apparently my C multimensional array syntax was rusty. Fixes: `a6740ee7` ("d3d12: Fix indexing of local_reference_state") Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26362> (cherry picked from commit `1924cdc289`)	2023-11-28 13:04:17 +00:00
Eric Engestrom	54f93b1788	intel/ci: fix .hasvk-manual-rules Fixes: `570acf5655` ("ci: Add a manual full and 1/10th hasvk CTS runs.") Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26259> (cherry picked from commit `cf510e38a5`)	2023-11-28 13:03:57 +00:00
Eric Engestrom	2525258917	intel/perf: fix regex escaping `\$` is interpreted before being passed to `re.search()`, but luckily for us the escape is also invalid and because of that, python 3.12+ warns us about it. Use a raw string instead, so that the `\` is passed untouched to `re.search()`. Fixes: `aa04b47c6e` ("intel/perf: add support for GtSlice/GtSliceXDualsubsliceY variables") Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26355> (cherry picked from commit `1942073112`)	2023-11-28 13:03:56 +00:00
Marek Olšák	94ad18d1e7	nir: fix gathering TESS_LEVEL_INNER/OUTER usage with lowered IO Those varyings shouldn't flag patch_inputs_read/patch_outputs_written. Fixes: `10be706778` - nir: gather indirect info from lowered IO intrinsics Reviewed-by: Timur Kristóf <timur.kristof@gmail.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26275> (cherry picked from commit `7a9b73fcb8`)	2023-11-28 13:03:52 +00:00
Pierre-Eric Pelloux-Prayer	3ed7b95a55	radeonsi: check sctx->tess_rings is valid before using it Fixes: `c89ca3b47f` ("radeonsi: change si_emit_derived_tess_state into a state atom") Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/10015 Reviewed-by: Marek Olšák <marek.olsak@amd.com> Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/26190> (cherry picked from commit `945288ffae`)	2023-11-28 13:03:48 +00:00
Eric Engestrom	dd56f591e5	.pick_status.json: Update to `f843b14c17`	2023-11-28 11:22:43 +00:00