[PATCH libdrm 1/4] tests/amdgpu: add memset dispatch test
Cui, Flora
Flora.Cui at amd.com
Fri Mar 15 08:50:54 UTC 2019
add memset dispatch test for gfx9
v2: disable dispatch test for other ASICs
Change-Id: I55a410a4ea0a6d19d1aca1f172e1dfcbbd1112fa
Signed-off-by: Flora Cui <flora.cui at amd.com>
Tested-by: Rui Teng <rui.teng at amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang at amd.com>
---
tests/amdgpu/amdgpu_test.c | 5 +
tests/amdgpu/basic_tests.c | 273 ++++++++++++++++++++++++++++++++++++++++++++-
2 files changed, 276 insertions(+), 2 deletions(-)
diff --git a/tests/amdgpu/amdgpu_test.c b/tests/amdgpu/amdgpu_test.c
index ebf4409..35324cb 100644
--- a/tests/amdgpu/amdgpu_test.c
+++ b/tests/amdgpu/amdgpu_test.c
@@ -436,6 +436,11 @@ static void amdgpu_disable_suites()
if (family_id < AMDGPU_FAMILY_VI || family_id > AMDGPU_FAMILY_RV)
if (amdgpu_set_test_active(BASIC_TESTS_STR, "Sync dependency Test", CU_FALSE))
fprintf(stderr, "test deactivation failed - %s\n", CU_get_error_msg());
+
+ /* This test was ran on GFX9 only */
+ if (family_id < AMDGPU_FAMILY_AI || family_id > AMDGPU_FAMILY_RV)
+ if (amdgpu_set_test_active(BASIC_TESTS_STR, "Dispatch Test", CU_FALSE))
+ fprintf(stderr, "test deactivation failed - %s\n", CU_get_error_msg());
}
/* The main() function for setting up and running the tests.
diff --git a/tests/amdgpu/basic_tests.c b/tests/amdgpu/basic_tests.c
index dbae4d5..fe6db09 100644
--- a/tests/amdgpu/basic_tests.c
+++ b/tests/amdgpu/basic_tests.c
@@ -49,6 +49,7 @@ static void amdgpu_userptr_test(void);
static void amdgpu_semaphore_test(void);
static void amdgpu_sync_dependency_test(void);
static void amdgpu_bo_eviction_test(void);
+static void amdgpu_dispatch_test(void);
static void amdgpu_command_submission_write_linear_helper(unsigned ip_type);
static void amdgpu_command_submission_const_fill_helper(unsigned ip_type);
@@ -70,6 +71,7 @@ CU_TestInfo basic_tests[] = {
{ "Command submission Test (SDMA)", amdgpu_command_submission_sdma },
{ "SW semaphore Test", amdgpu_semaphore_test },
{ "Sync dependency Test", amdgpu_sync_dependency_test },
+ { "Dispatch Test", amdgpu_dispatch_test },
CU_TEST_INFO_NULL,
};
#define BUFFER_SIZE (8 * 1024)
@@ -117,6 +119,7 @@ CU_TestInfo basic_tests[] = {
#define PACKET3(op, n) ((PACKET_TYPE3 << 30) | \
(((op) & 0xFF) << 8) | \
((n) & 0x3FFF) << 16)
+#define PACKET3_COMPUTE(op, n) PACKET3(op, n) | (1 << 1)
/* Packet 3 types */
#define PACKET3_NOP 0x10
@@ -245,8 +248,11 @@ CU_TestInfo basic_tests[] = {
#define PACKET3_SET_SH_REG_START 0x00002c00
#define PACKET3_DISPATCH_DIRECT 0x15
-
-
+#define PACKET3_EVENT_WRITE 0x46
+#define PACKET3_ACQUIRE_MEM 0x58
+#define PACKET3_SET_CONTEXT_REG 0x69
+#define PACKET3_SET_UCONFIG_REG 0x79
+#define PACKET3_DRAW_INDEX_AUTO 0x2D
/* gfx 8 */
#define mmCOMPUTE_PGM_LO 0x2e0c
#define mmCOMPUTE_PGM_RSRC1 0x2e12
@@ -286,6 +292,25 @@ static uint32_t shader_bin[] = {
#define CODE_OFFSET 512
#define DATA_OFFSET 1024
+enum cs_type {
+ CS_BUFFERCLEAR,
+};
+
+static const uint32_t bufferclear_cs_shader_gfx9[] = {
+ 0xD1FD0000, 0x04010C08, 0x7E020204, 0x7E040205,
+ 0x7E060206, 0x7E080207, 0xE01C2000, 0x80000100,
+ 0xBF810000
+};
+
+static const uint32_t bufferclear_cs_shader_registers_gfx9[][2] = {
+ {0x2e12, 0x000C0041}, //{ mmCOMPUTE_PGM_RSRC1, 0x000C0041 },
+ {0x2e13, 0x00000090}, //{ mmCOMPUTE_PGM_RSRC2, 0x00000090 },
+ {0x2e07, 0x00000040}, //{ mmCOMPUTE_NUM_THREAD_X, 0x00000040 },
+ {0x2e08, 0x00000001}, //{ mmCOMPUTE_NUM_THREAD_Y, 0x00000001 },
+ {0x2e09, 0x00000001}, //{ mmCOMPUTE_NUM_THREAD_Z, 0x00000001 }
+};
+
+static const uint32_t bufferclear_cs_shader_registers_num_gfx9 = 5;
int amdgpu_bo_alloc_and_map_raw(amdgpu_device_handle dev, unsigned size,
unsigned alignment, unsigned heap, uint64_t alloc_flags,
@@ -1883,3 +1908,247 @@ static void amdgpu_sync_dependency_test(void)
free(ibs_request.dependencies);
}
+
+static int amdgpu_dispatch_load_cs_shader(uint8_t *ptr,
+ int cs_type)
+{
+ uint32_t shader_size;
+ const uint32_t *shader;
+
+ switch (cs_type) {
+ case CS_BUFFERCLEAR:
+ shader = bufferclear_cs_shader_gfx9;
+ shader_size = sizeof(bufferclear_cs_shader_gfx9);
+ break;
+ default:
+ return -1;
+ break;
+ }
+
+ memcpy(ptr, shader, shader_size);
+ return 0;
+}
+
+static int amdgpu_dispatch_init(uint32_t *ptr, uint32_t ip_type)
+{
+ int i = 0;
+
+ /* Write context control and load shadowing register if necessary */
+ if (ip_type == AMDGPU_HW_IP_GFX) {
+ ptr[i++] = PACKET3(PKT3_CONTEXT_CONTROL, 1);
+ ptr[i++] = 0x80000000;
+ ptr[i++] = 0x80000000;
+ }
+
+ /* Issue commands to set default compute state. */
+ /* clear mmCOMPUTE_START_Z - mmCOMPUTE_START_X */
+ ptr[i++] = PACKET3_COMPUTE(PKT3_SET_SH_REG, 3);
+ ptr[i++] = 0x204;
+ i += 3;
+ /* clear mmCOMPUTE_RESOURCE_LIMITS */
+ ptr[i++] = PACKET3_COMPUTE(PKT3_SET_SH_REG, 1);
+ ptr[i++] = 0x215;
+ ptr[i++] = 0;
+ /* clear mmCOMPUTE_TMPRING_SIZE */
+ ptr[i++] = PACKET3_COMPUTE(PKT3_SET_SH_REG, 1);
+ ptr[i++] = 0x218;
+ ptr[i++] = 0;
+
+ return i;
+}
+
+static int amdgpu_dispatch_write_cumask(uint32_t *ptr)
+{
+ int i = 0;
+
+ /* Issue commands to set cu mask used in current dispatch */
+ /* set mmCOMPUTE_STATIC_THREAD_MGMT_SE1 - mmCOMPUTE_STATIC_THREAD_MGMT_SE0 */
+ ptr[i++] = PACKET3_COMPUTE(PKT3_SET_SH_REG, 2);
+ ptr[i++] = 0x216;
+ ptr[i++] = 0xffffffff;
+ ptr[i++] = 0xffffffff;
+ /* set mmCOMPUTE_STATIC_THREAD_MGMT_SE3 - mmCOMPUTE_STATIC_THREAD_MGMT_SE2 */
+ ptr[i++] = PACKET3_COMPUTE(PKT3_SET_SH_REG, 2);
+ ptr[i++] = 0x219;
+ ptr[i++] = 0xffffffff;
+ ptr[i++] = 0xffffffff;
+
+ return i;
+}
+
+static int amdgpu_dispatch_write2hw(uint32_t *ptr, uint64_t shader_addr)
+{
+ int i, j;
+
+ i = 0;
+
+ /* Writes shader state to HW */
+ /* set mmCOMPUTE_PGM_HI - mmCOMPUTE_PGM_LO */
+ ptr[i++] = PACKET3_COMPUTE(PKT3_SET_SH_REG, 2);
+ ptr[i++] = 0x20c;
+ ptr[i++] = (shader_addr >> 8);
+ ptr[i++] = (shader_addr >> 40);
+ /* write sh regs*/
+ for (j = 0; j < bufferclear_cs_shader_registers_num_gfx9; j++) {
+ ptr[i++] = PACKET3_COMPUTE(PKT3_SET_SH_REG, 1);
+ /* - Gfx9ShRegBase */
+ ptr[i++] = bufferclear_cs_shader_registers_gfx9[j][0] - 0x2c00;
+ ptr[i++] = bufferclear_cs_shader_registers_gfx9[j][1];
+ }
+
+ return i;
+}
+
+static void amdgpu_memset_dispatch_test(amdgpu_device_handle device_handle,
+ uint32_t ip_type,
+ uint32_t ring)
+{
+ 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;
+
+ r = amdgpu_cs_ctx_create(device_handle, &context_handle);
+ CU_ASSERT_EQUAL(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);
+ CU_ASSERT_EQUAL(r, 0);
+
+ 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);
+ CU_ASSERT_EQUAL(r, 0);
+
+ r = amdgpu_dispatch_load_cs_shader(ptr_shader, CS_BUFFERCLEAR);
+ CU_ASSERT_EQUAL(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);
+ CU_ASSERT_EQUAL(r, 0);
+
+ i = 0;
+ i += amdgpu_dispatch_init(ptr_cmd + i, ip_type);
+
+ /* Issue commands to set cu mask used in current dispatch */
+ i += amdgpu_dispatch_write_cumask(ptr_cmd + i);
+
+ /* Writes shader state to HW */
+ i += amdgpu_dispatch_write2hw(ptr_cmd + i, mc_address_shader);
+
+ /* Write constant data */
+ /* Writes the UAV constant data to the SGPRs. */
+ ptr_cmd[i++] = PACKET3_COMPUTE(PKT3_SET_SH_REG, 4);
+ ptr_cmd[i++] = 0x240;
+ ptr_cmd[i++] = mc_address_dst;
+ ptr_cmd[i++] = (mc_address_dst >> 32) | 0x100000;
+ ptr_cmd[i++] = 0x400;
+ ptr_cmd[i++] = 0x74fac;
+
+ /* Sets a range of pixel shader constants */
+ ptr_cmd[i++] = PACKET3_COMPUTE(PKT3_SET_SH_REG, 4);
+ ptr_cmd[i++] = 0x244;
+ ptr_cmd[i++] = 0x22222222;
+ ptr_cmd[i++] = 0x22222222;
+ ptr_cmd[i++] = 0x22222222;
+ ptr_cmd[i++] = 0x22222222;
+
+ /* dispatch direct command */
+ ptr_cmd[i++] = PACKET3_COMPUTE(PACKET3_DISPATCH_DIRECT, 3);
+ ptr_cmd[i++] = 0x10;
+ ptr_cmd[i++] = 1;
+ ptr_cmd[i++] = 1;
+ ptr_cmd[i++] = 1;
+
+ while (i & 7)
+ ptr_cmd[i++] = 0xffff1000; /* type3 nop packet */
+
+ resources[0] = bo_dst;
+ resources[1] = bo_shader;
+ resources[2] = bo_cmd;
+ r = amdgpu_bo_list_create(device_handle, 3, resources, NULL, &bo_list);
+ CU_ASSERT_EQUAL(r, 0);
+
+ ib_info.ib_mc_address = mc_address_cmd;
+ ib_info.size = i;
+ 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);
+ CU_ASSERT_EQUAL(r, 0);
+
+ r = amdgpu_bo_list_destroy(bo_list);
+ CU_ASSERT_EQUAL(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);
+ CU_ASSERT_EQUAL(r, 0);
+ CU_ASSERT_EQUAL(expired, true);
+
+ /* verify if memset test result meets with expected */
+ i = 0;
+ while(i < bo_dst_size) {
+ CU_ASSERT_EQUAL(ptr_dst[i++], 0x22);
+ }
+
+ r = amdgpu_bo_unmap_and_free(bo_dst, va_dst, mc_address_dst, bo_dst_size);
+ CU_ASSERT_EQUAL(r, 0);
+
+ r = amdgpu_bo_unmap_and_free(bo_shader, va_shader, mc_address_shader, bo_shader_size);
+ CU_ASSERT_EQUAL(r, 0);
+
+ r = amdgpu_bo_unmap_and_free(bo_cmd, va_cmd, mc_address_cmd, bo_cmd_size);
+ CU_ASSERT_EQUAL(r, 0);
+
+ r = amdgpu_cs_ctx_free(context_handle);
+ CU_ASSERT_EQUAL(r, 0);
+}
+
+static void amdgpu_dispatch_test(void)
+{
+ int r;
+ struct drm_amdgpu_info_hw_ip info;
+ uint32_t ring_id;
+
+ r = amdgpu_query_hw_ip_info(device_handle, AMDGPU_HW_IP_COMPUTE, 0, &info);
+ CU_ASSERT_EQUAL(r, 0);
+
+ for (ring_id = 0; (1 << ring_id) & info.available_rings; ring_id++)
+ amdgpu_memset_dispatch_test(device_handle, AMDGPU_HW_IP_COMPUTE, ring_id);
+
+ r = amdgpu_query_hw_ip_info(device_handle, AMDGPU_HW_IP_GFX, 0, &info);
+ CU_ASSERT_EQUAL(r, 0);
+
+ for (ring_id = 0; (1 << ring_id) & info.available_rings; ring_id++)
+ amdgpu_memset_dispatch_test(device_handle, AMDGPU_HW_IP_GFX, ring_id);
+}
--
2.7.4
More information about the amd-gfx
mailing list