[igt-dev] [PATCH 4/7] lib/amdgpu: added dispatch tests for gfx and compute
vitaly.prosyak at amd.com
vitaly.prosyak at amd.com
Sat Aug 13 00:03:03 UTC 2022
From: Vitaly Prosyak <vitaly.prosyak at amd.com>
Signed-off-by: Vitaly Prosyak <vitaly.prosyak at amd.com>
Acked-by: Christian König <christian.koenig at amd.com>
---
lib/amdgpu/amd_dispatch.c | 559 ++++++++++++++++++++++++++++++++++++++
lib/amdgpu/amd_dispatch.h | 42 +++
lib/meson.build | 3 +-
tests/amdgpu/amd_basic.c | 18 ++
4 files changed, 621 insertions(+), 1 deletion(-)
create mode 100644 lib/amdgpu/amd_dispatch.c
create mode 100644 lib/amdgpu/amd_dispatch.h
diff --git a/lib/amdgpu/amd_dispatch.c b/lib/amdgpu/amd_dispatch.c
new file mode 100644
index 000000000..cf773c309
--- /dev/null
+++ b/lib/amdgpu/amd_dispatch.c
@@ -0,0 +1,559 @@
+/*
+ * Copyright 2014 Advanced Micro Devices, Inc.
+ * Copyright 2022 Advanced Micro Devices, Inc.
+ * *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ *
+ */
+#include <amdgpu.h>
+#include "amd_memory.h"
+#include "amd_dispatch.h"
+#include "amd_shared_dispatch.h"
+#include "amd_dispatch_helpers.h"
+#include "amd_PM4.h"
+#include "amd_ip_blocks.h"
+#include "amd_shaders.h"
+
+static void
+amdgpu_memset_dispatch_test(amdgpu_device_handle device_handle,
+ uint32_t ip_type, uint32_t ring,
+ uint32_t version)
+{
+ amdgpu_context_handle context_handle;
+ amdgpu_bo_handle bo_dst, bo_shader, bo_cmd, resources[3];
+ volatile unsigned char *ptr_dst;
+ void *ptr_shader;
+ uint32_t *ptr_cmd;
+ uint64_t mc_address_dst, mc_address_shader, mc_address_cmd;
+ amdgpu_va_handle va_dst, va_shader, va_cmd;
+ int i, r;
+ int bo_dst_size = 16384;
+ int bo_shader_size = 4096;
+ int bo_cmd_size = 4096;
+ struct amdgpu_cs_request ibs_request = {0};
+ struct amdgpu_cs_ib_info ib_info= {0};
+ amdgpu_bo_list_handle bo_list;
+ struct amdgpu_cs_fence fence_status = {0};
+ uint32_t expired;
+
+ struct amdgpu_cmd_base * base_cmd = get_cmd_base();
+
+ r = amdgpu_cs_ctx_create(device_handle, &context_handle);
+ igt_assert_eq(r, 0);
+
+ r = amdgpu_bo_alloc_and_map(device_handle, bo_cmd_size, 4096,
+ AMDGPU_GEM_DOMAIN_GTT, 0,
+ &bo_cmd, (void **)&ptr_cmd,
+ &mc_address_cmd, &va_cmd);
+ igt_assert_eq(r, 0);
+ memset(ptr_cmd, 0, bo_cmd_size);
+ base_cmd->attach_buf(base_cmd, ptr_cmd, bo_cmd_size);
+
+ r = amdgpu_bo_alloc_and_map(device_handle, bo_shader_size, 4096,
+ AMDGPU_GEM_DOMAIN_VRAM, 0,
+ &bo_shader, &ptr_shader,
+ &mc_address_shader, &va_shader);
+ igt_assert_eq(r, 0);
+ memset(ptr_shader, 0, bo_shader_size);
+
+ r = amdgpu_dispatch_load_cs_shader(ptr_shader, CS_BUFFERCLEAR, version);
+ igt_assert_eq(r, 0);
+
+ r = amdgpu_bo_alloc_and_map(device_handle, bo_dst_size, 4096,
+ AMDGPU_GEM_DOMAIN_VRAM, 0,
+ &bo_dst, (void **)&ptr_dst,
+ &mc_address_dst, &va_dst);
+ igt_assert_eq(r, 0);
+ /// TODO helper function for this bloc
+ amdgpu_dispatch_init(ip_type, base_cmd, version);
+
+ /* Issue commands to set cu mask used in current dispatch */
+ amdgpu_dispatch_write_cumask(base_cmd, version);
+
+ /* Writes shader state to HW */
+ amdgpu_dispatch_write2hw(base_cmd, mc_address_shader, version);
+
+ /* Write constant data */
+ /* Writes the UAV constant data to the SGPRs. */
+ base_cmd->emit(base_cmd, PACKET3_COMPUTE(PKT3_SET_SH_REG, 4));
+ base_cmd->emit(base_cmd, 0x240);
+
+ base_cmd->emit(base_cmd, mc_address_dst);
+ base_cmd->emit(base_cmd, (mc_address_dst >> 32) | 0x100000);
+
+ base_cmd->emit(base_cmd, 0x400);
+ if (version == 9)
+ base_cmd->emit(base_cmd, 0x74fac);
+ else if (version == 10)
+ base_cmd->emit(base_cmd, 0x1104bfac);
+
+ /* Sets a range of pixel shader constants */
+ base_cmd->emit(base_cmd, PACKET3_COMPUTE(PKT3_SET_SH_REG, 4));
+ base_cmd->emit(base_cmd, 0x244);
+
+ base_cmd->emit(base_cmd, 0x22222222);
+ base_cmd->emit(base_cmd, 0x22222222);
+ base_cmd->emit(base_cmd, 0x22222222);
+ base_cmd->emit(base_cmd, 0x22222222);
+
+ /* clear mmCOMPUTE_RESOURCE_LIMITS */
+ base_cmd->emit(base_cmd, PACKET3_COMPUTE(PKT3_SET_SH_REG, 1));
+ base_cmd->emit(base_cmd, 0x215);
+ base_cmd->emit(base_cmd, 0);
+
+ /* dispatch direct command */
+ base_cmd->emit(base_cmd, PACKET3_COMPUTE(PACKET3_DISPATCH_DIRECT,3));
+ base_cmd->emit(base_cmd, 0x10);
+ base_cmd->emit(base_cmd, 1);
+ base_cmd->emit(base_cmd, 1);
+ base_cmd->emit(base_cmd, 1);
+
+ base_cmd->emit_aligned(base_cmd, 7, 0xffff1000);
+ resources[0] = bo_dst;
+ resources[1] = bo_shader;
+ resources[2] = bo_cmd;
+ r = amdgpu_bo_list_create(device_handle, 3, resources, NULL, &bo_list);
+ igt_assert_eq(r, 0);
+
+ ib_info.ib_mc_address = mc_address_cmd;
+ ib_info.size = base_cmd->cdw;
+ ibs_request.ip_type = ip_type;
+ ibs_request.ring = ring;
+ ibs_request.resources = bo_list;
+ ibs_request.number_of_ibs = 1;
+ ibs_request.ibs = &ib_info;
+ ibs_request.fence_info.handle = NULL;
+
+ /* submit CS */
+ r = amdgpu_cs_submit(context_handle, 0, &ibs_request, 1);
+ igt_assert_eq(r, 0);
+
+ r = amdgpu_bo_list_destroy(bo_list);
+ igt_assert_eq(r, 0);
+
+ fence_status.ip_type = ip_type;
+ fence_status.ip_instance = 0;
+ fence_status.ring = ring;
+ fence_status.context = context_handle;
+ fence_status.fence = ibs_request.seq_no;
+
+ /* wait for IB accomplished */
+ r = amdgpu_cs_query_fence_status(&fence_status,
+ AMDGPU_TIMEOUT_INFINITE,
+ 0, &expired);
+ igt_assert_eq(r, 0);
+ igt_assert_eq(expired, true);
+
+ /* verify if memset test result meets with expected */
+ i = 0;
+ while(i < bo_dst_size) {
+ igt_assert_eq(ptr_dst[i++], 0x22);
+ }
+
+ amdgpu_bo_unmap_and_free(bo_dst, va_dst, mc_address_dst, bo_dst_size);
+ amdgpu_bo_unmap_and_free(bo_shader, va_shader, mc_address_shader,
+ bo_shader_size);
+ amdgpu_bo_unmap_and_free(bo_cmd, va_cmd, mc_address_cmd, bo_cmd_size);
+ amdgpu_cs_ctx_free(context_handle);
+}
+
+void
+amdgpu_memcpy_dispatch_test(amdgpu_device_handle device_handle,
+ uint32_t ip_type, uint32_t ring, uint32_t version,
+ int hang)
+{
+ amdgpu_context_handle context_handle;
+ amdgpu_bo_handle bo_src, bo_dst, bo_shader, bo_cmd, resources[4];
+ volatile unsigned char *ptr_dst;
+ void *ptr_shader;
+ unsigned char *ptr_src;
+ uint32_t *ptr_cmd;
+ uint64_t mc_address_src, mc_address_dst, mc_address_shader, mc_address_cmd;
+ amdgpu_va_handle va_src, va_dst, va_shader, va_cmd;
+ int i, r;
+ int bo_dst_size = 16384;
+ int bo_shader_size = 4096;
+ int bo_cmd_size = 4096;
+ struct amdgpu_cs_request ibs_request = {0};
+ struct amdgpu_cs_ib_info ib_info= {0};
+ uint32_t expired, hang_state, hangs;
+ enum cs_type cs_type;
+ amdgpu_bo_list_handle bo_list;
+ struct amdgpu_cs_fence fence_status = {0};
+ struct amdgpu_cmd_base * base_cmd = get_cmd_base();
+
+ r = amdgpu_cs_ctx_create(device_handle, &context_handle);
+ igt_assert_eq(r, 0);
+
+ r = amdgpu_bo_alloc_and_map(device_handle, bo_cmd_size, 4096,
+ AMDGPU_GEM_DOMAIN_GTT, 0,
+ &bo_cmd, (void **)&ptr_cmd,
+ &mc_address_cmd, &va_cmd);
+ igt_assert_eq(r, 0);
+ memset(ptr_cmd, 0, bo_cmd_size);
+ base_cmd->attach_buf(base_cmd, ptr_cmd, bo_cmd_size);
+
+ r = amdgpu_bo_alloc_and_map(device_handle, bo_shader_size, 4096,
+ AMDGPU_GEM_DOMAIN_VRAM, 0,
+ &bo_shader, &ptr_shader,
+ &mc_address_shader, &va_shader);
+ igt_assert_eq(r, 0);
+ memset(ptr_shader, 0, bo_shader_size);
+
+ cs_type = hang ? CS_HANG : CS_BUFFERCOPY;
+ r = amdgpu_dispatch_load_cs_shader(ptr_shader, cs_type, version);
+ igt_assert_eq(r, 0);
+
+ r = amdgpu_bo_alloc_and_map(device_handle, bo_dst_size, 4096,
+ AMDGPU_GEM_DOMAIN_VRAM, 0,
+ &bo_src, (void **)&ptr_src,
+ &mc_address_src, &va_src);
+ igt_assert_eq(r, 0);
+
+ r = amdgpu_bo_alloc_and_map(device_handle, bo_dst_size, 4096,
+ AMDGPU_GEM_DOMAIN_VRAM, 0,
+ &bo_dst, (void **)&ptr_dst,
+ &mc_address_dst, &va_dst);
+ igt_assert_eq(r, 0);
+
+ ///TODO helper function for this bloc
+ amdgpu_dispatch_init(ip_type, base_cmd, version);
+ /* Issue commands to set cu mask used in current dispatch */
+ amdgpu_dispatch_write_cumask(base_cmd, version);
+ /* Writes shader state to HW */
+ amdgpu_dispatch_write2hw(base_cmd, mc_address_shader, version);
+ memset(ptr_src, 0x55, bo_dst_size);
+
+ /* Write constant data */
+ /* Writes the texture resource constants data to the SGPRs */
+ base_cmd->emit(base_cmd, PACKET3_COMPUTE(PKT3_SET_SH_REG, 4));
+ base_cmd->emit(base_cmd, 0x240);
+ base_cmd->emit(base_cmd, mc_address_src);
+
+ base_cmd->emit(base_cmd, (mc_address_src >> 32) | 0x100000);
+
+ base_cmd->emit(base_cmd, 0x400);
+ if (version == 9)
+ base_cmd->emit(base_cmd,0x74fac);
+ else if (version == 10)
+ base_cmd->emit(base_cmd,0x1104bfac);
+
+ /* Writes the UAV constant data to the SGPRs. */
+ base_cmd->emit(base_cmd, PACKET3_COMPUTE(PKT3_SET_SH_REG, 4));
+ base_cmd->emit(base_cmd, 0x244);
+ base_cmd->emit(base_cmd, mc_address_dst);
+ base_cmd->emit(base_cmd, (mc_address_dst >> 32) | 0x100000);
+ base_cmd->emit(base_cmd, 0x400);
+ if (version == 9)
+ base_cmd->emit(base_cmd, 0x74fac);
+ else if (version == 10)
+ base_cmd->emit(base_cmd, 0x1104bfac);
+
+ /* clear mmCOMPUTE_RESOURCE_LIMITS */
+ base_cmd->emit(base_cmd, PACKET3_COMPUTE(PKT3_SET_SH_REG, 1));
+ base_cmd->emit(base_cmd, 0x215);
+ base_cmd->emit(base_cmd, 0);
+
+ /* dispatch direct command */
+ base_cmd->emit(base_cmd, PACKET3_COMPUTE(PACKET3_DISPATCH_DIRECT,3));
+ base_cmd->emit(base_cmd, 0x10);
+ base_cmd->emit(base_cmd, 1);
+ base_cmd->emit(base_cmd, 1);
+ base_cmd->emit(base_cmd, 1);
+
+ base_cmd->emit_aligned(base_cmd, 7, 0xffff1000); /* type3 nop packet */
+
+ resources[0] = bo_shader;
+ resources[1] = bo_src;
+ resources[2] = bo_dst;
+ resources[3] = bo_cmd;
+
+ r = amdgpu_bo_list_create(device_handle, 4, resources, NULL, &bo_list);
+ igt_assert_eq(r, 0);
+
+ ib_info.ib_mc_address = mc_address_cmd;
+ ib_info.size = base_cmd->cdw;
+ ibs_request.ip_type = ip_type;
+ ibs_request.ring = ring;
+ ibs_request.resources = bo_list;
+ ibs_request.number_of_ibs = 1;
+ ibs_request.ibs = &ib_info;
+ ibs_request.fence_info.handle = NULL;
+ r = amdgpu_cs_submit(context_handle, 0, &ibs_request, 1);
+ igt_assert_eq(r, 0);
+
+ fence_status.ip_type = ip_type;
+ fence_status.ip_instance = 0;
+ fence_status.ring = ring;
+ fence_status.context = context_handle;
+ fence_status.fence = ibs_request.seq_no;
+
+ /* wait for IB accomplished */
+ r = amdgpu_cs_query_fence_status(&fence_status,
+ AMDGPU_TIMEOUT_INFINITE,
+ 0, &expired);
+
+ if (!hang) {
+ igt_assert_eq(r, 0);
+ igt_assert_eq(expired, true);
+
+ /* verify if memcpy test result meets with expected */
+ i = 0;
+ /*it works up to 12287 ? vs required 16384 for gfx 8*/
+ while(i < bo_dst_size) {
+ igt_assert_eq(ptr_dst[i], ptr_src[i]);
+ i++;
+ }
+ } else {
+ r = amdgpu_cs_query_reset_state(context_handle, &hang_state, &hangs);
+ igt_assert_eq(r, 0);
+ igt_assert_eq(hang_state, AMDGPU_CTX_UNKNOWN_RESET);
+ }
+
+ amdgpu_bo_list_destroy(bo_list);
+ amdgpu_bo_unmap_and_free(bo_src, va_src, mc_address_src, bo_dst_size);
+ amdgpu_bo_unmap_and_free(bo_dst, va_dst, mc_address_dst, bo_dst_size);
+ amdgpu_bo_unmap_and_free(bo_cmd, va_cmd, mc_address_cmd, bo_cmd_size);
+ amdgpu_bo_unmap_and_free(bo_shader, va_shader, mc_address_shader,
+ bo_shader_size);
+ amdgpu_cs_ctx_free(context_handle);
+}
+
+static void
+amdgpu_memcpy_dispatch_hang_slow_test(amdgpu_device_handle device_handle,
+ uint32_t ip_type, uint32_t ring,
+ int version, uint32_t gpu_reset_status_equel)
+{
+ amdgpu_context_handle context_handle;
+ amdgpu_bo_handle bo_src, bo_dst, bo_shader, bo_cmd, resources[4];
+ volatile unsigned char *ptr_dst;
+ void *ptr_shader;
+ unsigned char *ptr_src;
+ uint32_t *ptr_cmd;
+ uint64_t mc_address_src, mc_address_dst, mc_address_shader, mc_address_cmd;
+ amdgpu_va_handle va_src, va_dst, va_shader, va_cmd;
+ int r;
+
+ int bo_dst_size = 0x4000000;
+ int bo_shader_size = 0x400000;
+ int bo_cmd_size = 4096;
+
+ struct amdgpu_cs_request ibs_request = {0};
+ struct amdgpu_cs_ib_info ib_info= {0};
+ uint32_t hang_state, hangs, expired;
+ struct amdgpu_gpu_info gpu_info = {0};
+ amdgpu_bo_list_handle bo_list;
+ struct amdgpu_cs_fence fence_status = {0};
+
+ struct amdgpu_cmd_base * base_cmd = get_cmd_base();
+
+ r = amdgpu_query_gpu_info(device_handle, &gpu_info);
+ igt_assert_eq(r, 0);
+
+ r = amdgpu_cs_ctx_create(device_handle, &context_handle);
+ igt_assert_eq(r, 0);
+
+ r = amdgpu_bo_alloc_and_map(device_handle, bo_cmd_size, 4096,
+ AMDGPU_GEM_DOMAIN_GTT, 0,
+ &bo_cmd, (void **)&ptr_cmd,
+ &mc_address_cmd, &va_cmd);
+ igt_assert_eq(r, 0);
+ memset(ptr_cmd, 0, bo_cmd_size);
+ base_cmd->attach_buf(base_cmd, ptr_cmd, bo_cmd_size);
+
+ r = amdgpu_bo_alloc_and_map(device_handle, bo_shader_size, 4096,
+ AMDGPU_GEM_DOMAIN_VRAM, 0, &bo_shader,
+ &ptr_shader, &mc_address_shader, &va_shader);
+ igt_assert_eq(r, 0);
+ memset(ptr_shader, 0, bo_shader_size);
+
+ r = amdgpu_dispatch_load_cs_shader_hang_slow(ptr_shader,
+ gpu_info.family_id);
+ igt_assert_eq(r, 0);
+
+ r = amdgpu_bo_alloc_and_map(device_handle, bo_dst_size, 4096,
+ AMDGPU_GEM_DOMAIN_VRAM, 0, &bo_src,
+ (void **)&ptr_src, &mc_address_src, &va_src);
+ igt_assert_eq(r, 0);
+
+ r = amdgpu_bo_alloc_and_map(device_handle, bo_dst_size, 4096,
+ AMDGPU_GEM_DOMAIN_VRAM, 0, &bo_dst,
+ (void **)&ptr_dst, &mc_address_dst, &va_dst);
+ igt_assert_eq(r, 0);
+
+ memset(ptr_src, 0x55, bo_dst_size);
+
+ amdgpu_dispatch_init(ip_type, base_cmd, version );
+
+
+
+ /* Issue commands to set cu mask used in current dispatch */
+ amdgpu_dispatch_write_cumask(base_cmd, version);
+
+ /* Writes shader state to HW */
+ amdgpu_dispatch_write2hw(base_cmd, mc_address_shader, version);
+
+ /* Write constant data */
+ /* Writes the texture resource constants data to the SGPRs */
+ base_cmd->emit(base_cmd, PACKET3_COMPUTE(PKT3_SET_SH_REG, 4));
+ base_cmd->emit(base_cmd, 0x240);
+ base_cmd->emit(base_cmd, mc_address_src);
+ base_cmd->emit(base_cmd, (mc_address_src >> 32) | 0x100000);
+ base_cmd->emit(base_cmd, 0x400000);
+ if (version == 9)
+ base_cmd->emit(base_cmd, 0x74fac);
+ else if (version == 10)
+ base_cmd->emit(base_cmd, 0x1104bfac);
+
+
+ /* Writes the UAV constant data to the SGPRs. */
+ base_cmd->emit(base_cmd, PACKET3_COMPUTE(PKT3_SET_SH_REG, 4));
+ base_cmd->emit(base_cmd, 0x244);
+ base_cmd->emit(base_cmd, mc_address_dst);
+ base_cmd->emit(base_cmd, (mc_address_dst >> 32) | 0x100000);
+ base_cmd->emit(base_cmd, 0x400000);
+ if (version == 9)
+ base_cmd->emit(base_cmd, 0x74fac);
+ else if (version == 10)
+ base_cmd->emit(base_cmd, 0x1104bfac);
+
+
+ /* clear mmCOMPUTE_RESOURCE_LIMITS */
+ base_cmd->emit(base_cmd, PACKET3_COMPUTE(PKT3_SET_SH_REG, 1));
+ base_cmd->emit(base_cmd, 0x215);
+ base_cmd->emit(base_cmd, 0);
+
+ /* dispatch direct command */
+
+ base_cmd->emit(base_cmd, PACKET3_COMPUTE(PACKET3_DISPATCH_DIRECT, 3));
+ base_cmd->emit(base_cmd, 0x10000);
+ base_cmd->emit(base_cmd, 1);
+ base_cmd->emit(base_cmd, 1);
+ base_cmd->emit(base_cmd, 1);
+
+ base_cmd->emit_aligned(base_cmd, 7, 0xffff1000); /* type3 nop packet */
+
+ resources[0] = bo_shader;
+ resources[1] = bo_src;
+ resources[2] = bo_dst;
+ resources[3] = bo_cmd;
+ r = amdgpu_bo_list_create(device_handle, 4, resources, NULL, &bo_list);
+ igt_assert_eq(r, 0);
+
+ ib_info.ib_mc_address = mc_address_cmd;
+ ib_info.size = base_cmd->cdw;
+ ibs_request.ip_type = ip_type;
+ ibs_request.ring = ring;
+ ibs_request.resources = bo_list;
+ ibs_request.number_of_ibs = 1;
+ ibs_request.ibs = &ib_info;
+ ibs_request.fence_info.handle = NULL;
+ r = amdgpu_cs_submit(context_handle, 0, &ibs_request, 1);
+ igt_assert_eq(r, 0);
+
+ fence_status.ip_type = ip_type;
+ fence_status.ip_instance = 0;
+ fence_status.ring = ring;
+ fence_status.context = context_handle;
+ fence_status.fence = ibs_request.seq_no;
+
+ /* wait for IB accomplished */
+ r = amdgpu_cs_query_fence_status(&fence_status,
+ AMDGPU_TIMEOUT_INFINITE,
+ 0, &expired);
+
+ r = amdgpu_cs_query_reset_state(context_handle, &hang_state, &hangs);
+ igt_assert_eq(r, 0);
+ igt_assert_eq(hang_state, gpu_reset_status_equel);
+
+ r = amdgpu_bo_list_destroy(bo_list);
+ igt_assert_eq(r, 0);
+
+ amdgpu_bo_unmap_and_free(bo_src, va_src, mc_address_src, bo_dst_size);
+ amdgpu_bo_unmap_and_free(bo_dst, va_dst, mc_address_dst, bo_dst_size);
+ amdgpu_bo_unmap_and_free(bo_cmd, va_cmd, mc_address_cmd, bo_cmd_size);
+ amdgpu_bo_unmap_and_free(bo_shader, va_shader, mc_address_shader,
+ bo_shader_size);
+ amdgpu_cs_ctx_free(context_handle);
+ free_cmd_base(base_cmd);
+}
+
+void
+amdgpu_dispatch_hang_slow_helper(amdgpu_device_handle device_handle,
+ uint32_t ip_type)
+{
+ int r;
+ struct drm_amdgpu_info_hw_ip info;
+ uint32_t ring_id, version;
+
+ r = amdgpu_query_hw_ip_info(device_handle, ip_type, 0, &info);
+ igt_assert_eq(r, 0);
+ if (!info.available_rings)
+ printf("SKIP ... as there's no ring for ip %d\n", ip_type);
+
+ version = info.hw_ip_version_major;
+ if (version != 9 && version != 10) {
+ printf("SKIP ... unsupported gfx version %d\n", version);
+ return;
+ }
+ //TODO IGT
+ //if (version < 9)
+ // version = 9;
+ for (ring_id = 0; (1 << ring_id) & info.available_rings; ring_id++) {
+ amdgpu_memcpy_dispatch_test(device_handle, ip_type,
+ ring_id, version, 0);
+ amdgpu_memcpy_dispatch_hang_slow_test(device_handle, ip_type,
+ ring_id, version,
+ ip_type == AMDGPU_HW_IP_GFX ?
+ AMDGPU_CTX_UNKNOWN_RESET:AMDGPU_CTX_NO_RESET);
+
+ /* TODO Why reset status is differant? */
+ amdgpu_memcpy_dispatch_test(device_handle, ip_type, ring_id,
+ version, 0);
+ }
+}
+
+void amdgpu_gfx_dispatch_test(amdgpu_device_handle device_handle, uint32_t ip_type)
+{
+ int r;
+ struct drm_amdgpu_info_hw_ip info;
+ uint32_t ring_id, version;
+
+ r = amdgpu_query_hw_ip_info(device_handle, AMDGPU_HW_IP_GFX, 0, &info);
+ igt_assert_eq(r, 0);
+ if (!info.available_rings)
+ printf("SKIP ... as there's no graphics ring\n");
+
+ version = info.hw_ip_version_major;
+ if (version != 9 && version != 10) {
+ printf("SKIP ... unsupported gfx version %d\n", version);
+ return;
+ }
+ if (version < 9)
+ version = 9;
+
+ for (ring_id = 0; (1 << ring_id) & info.available_rings; ring_id++) {
+ amdgpu_memset_dispatch_test(device_handle, ip_type, ring_id,
+ version);
+ amdgpu_memcpy_dispatch_test(device_handle, ip_type, ring_id,
+ version, 0);
+ }
+}
+
diff --git a/lib/amdgpu/amd_dispatch.h b/lib/amdgpu/amd_dispatch.h
new file mode 100644
index 000000000..1905a5562
--- /dev/null
+++ b/lib/amdgpu/amd_dispatch.h
@@ -0,0 +1,42 @@
+/*
+ * Copyright 2022 Advanced Micro Devices, Inc.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a
+ * copy of this software and associated documentation files (the "Software"),
+ * to deal in the Software without restriction, including without limitation
+ * the rights to use, copy, modify, merge, publish, distribute, sublicense,
+ * and/or sell copies of the Software, and to permit persons to whom the
+ * Software is furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR
+ * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
+ * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR
+ * OTHER DEALINGS IN THE SOFTWARE.
+ *
+ *
+ */
+#ifndef AMD_DISPATCH_H
+#define AMD_DISPATCH_H
+
+#include <amdgpu.h>
+
+void amdgpu_gfx_dispatch_test(amdgpu_device_handle device_handle,
+ uint32_t ip_type);
+
+void amdgpu_memcpy_dispatch_test(amdgpu_device_handle device_handle,
+ uint32_t ip_type,
+ uint32_t ring,
+ uint32_t version,
+ int hang);
+
+void amdgpu_dispatch_hang_slow_helper(amdgpu_device_handle device_handle,
+ uint32_t ip_type);
+
+
+#endif
diff --git a/lib/meson.build b/lib/meson.build
index 956c73095..9bb0c8410 100644
--- a/lib/meson.build
+++ b/lib/meson.build
@@ -135,7 +135,8 @@ if libdrm_amdgpu.found()
'amdgpu/amd_shaders.c',
'amdgpu/amd_gfx_v8_0.c',
'amdgpu/amd_gfx_v9_0.c',
- 'amdgpu/amd_dispatch_helpers.c'
+ 'amdgpu/amd_dispatch_helpers.c',
+ 'amdgpu/amd_dispatch.c'
]
endif
diff --git a/tests/amdgpu/amd_basic.c b/tests/amdgpu/amd_basic.c
index 68a8ac9de..9be034975 100644
--- a/tests/amdgpu/amd_basic.c
+++ b/tests/amdgpu/amd_basic.c
@@ -30,6 +30,7 @@
#include "lib/amdgpu/amd_compute.h"
#include "lib/amdgpu/amd_gfx.h"
#include "lib/amdgpu/amd_shaders.h"
+#include "lib/amdgpu/amd_dispatch.h"
#define BUFFER_SIZE (8 * 1024)
@@ -625,6 +626,18 @@ amdgpu_sync_dependency_test(amdgpu_device_handle device_handle)
free_cmd_base(base);
}
+static void
+amdgpu_gfx_dispatch_test_gfx(amdgpu_device_handle device_handle)
+{
+ amdgpu_gfx_dispatch_test(device_handle, AMDGPU_HW_IP_GFX);
+}
+
+static void
+amdgpu_gfx_dispatch_test_compute(amdgpu_device_handle device_handle)
+{
+ amdgpu_gfx_dispatch_test(device_handle, AMDGPU_HW_IP_COMPUTE);
+}
+
igt_main
{
amdgpu_device_handle device;
@@ -678,6 +691,11 @@ igt_main
igt_subtest("sync_dependency_test")
amdgpu_sync_dependency_test(device);
+ igt_subtest("amdgpu_gfx_dispatch_test_compute")
+ amdgpu_gfx_dispatch_test_compute(device);
+
+ igt_subtest("amdgpu_gfx_dispatch_test_gfx")
+ amdgpu_gfx_dispatch_test_gfx(device);
igt_fixture {
amdgpu_device_deinitialize(device);
--
2.25.1
More information about the igt-dev
mailing list