[Intel-gfx] [PATCH i-g-t] tests: add slice power programming test
Lionel Landwerlin
lionel.g.landwerlin at intel.com
Fri Sep 1 17:13:11 UTC 2017
Verifies that the kernel programs slices correctly by reading the
value of PWR_CLK_STATE register.
Signed-off-by: Lionel Landwerlin <lionel.g.landwerlin at intel.com>
---
tests/Makefile.sources | 1 +
tests/ctx_rpcs.c | 235 +++++++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 236 insertions(+)
create mode 100644 tests/ctx_rpcs.c
diff --git a/tests/Makefile.sources b/tests/Makefile.sources
index 9553e4d9..4f39302a 100644
--- a/tests/Makefile.sources
+++ b/tests/Makefile.sources
@@ -171,6 +171,7 @@ TESTS_progs = \
core_getversion \
core_prop_blob \
core_setmaster_vs_auth \
+ ctx_rpcs \
drm_import_export \
drm_read \
drm_vma_limiter \
diff --git a/tests/ctx_rpcs.c b/tests/ctx_rpcs.c
new file mode 100644
index 00000000..c9aa29a8
--- /dev/null
+++ b/tests/ctx_rpcs.c
@@ -0,0 +1,235 @@
+/*
+ * Copyright © 2017 Intel Corporation
+ *
+ * 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 (including the next
+ * paragraph) 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 AUTHORS OR COPYRIGHT HOLDERS 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.
+ *
+ * Authors:
+ * Lionel Landwerlin <lionel.g.landwerlin at intel.com>
+ *
+ */
+
+#define _GNU_SOURCE
+#include "igt.h"
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <unistd.h>
+#include <fcntl.h>
+#include <signal.h>
+#include <errno.h>
+#include <time.h>
+#include <sys/wait.h>
+
+#include "intel_bufmgr.h"
+
+#define MI_STORE_REGISTER_MEM (0x24 << 23)
+
+#define GEN8_R_PWR_CLK_STATE 0x20C8
+#define GEN8_RPCS_ENABLE (1 << 31)
+#define GEN8_RPCS_S_CNT_ENABLE (1 << 18)
+#define GEN8_RPCS_S_CNT_SHIFT 15
+#define GEN8_RPCS_S_CNT_MASK (0x7 << GEN8_RPCS_S_CNT_SHIFT)
+#define GEN8_RPCS_SS_CNT_ENABLE (1 << 11)
+#define GEN8_RPCS_SS_CNT_SHIFT 8
+#define GEN8_RPCS_SS_CNT_MASK (0x7 << GEN8_RPCS_SS_CNT_SHIFT)
+#define GEN8_RPCS_EU_MAX_SHIFT 4
+#define GEN8_RPCS_EU_MAX_MASK (0xf << GEN8_RPCS_EU_MAX_SHIFT)
+#define GEN8_RPCS_EU_MIN_SHIFT 0
+#define GEN8_RPCS_EU_MIN_MASK (0xf << GEN8_RPCS_EU_MIN_SHIFT)
+
+
+#ifndef I915_CONTEXT_PARAM_SSEU
+
+#define I915_CONTEXT_PARAM_SSEU 0x6
+
+union drm_i915_gem_context_param_sseu {
+ struct {
+ __u8 slice_mask;
+ __u8 subslice_mask;
+ __u8 min_eu_per_subslice;
+ __u8 max_eu_per_subslice;
+ } packed;
+ __u64 value;
+};
+
+#endif /* I915_CONTEXT_PARAM_SSEU */
+
+static int drm_fd;
+static int devid;
+
+static uint32_t
+read_rpcs_reg(drm_intel_bufmgr *bufmgr,
+ drm_intel_context *context)
+{
+ struct intel_batchbuffer *batch;
+ drm_intel_bo *bo;
+ uint32_t rpcs;
+ int ret;
+
+ batch = intel_batchbuffer_alloc(bufmgr, devid);
+ igt_assert(batch);
+
+ intel_batchbuffer_set_context(batch, context);
+
+ bo = drm_intel_bo_alloc(bufmgr, "target bo", 4096, 4096);
+ igt_assert(bo);
+
+ BEGIN_BATCH(3, 1);
+ OUT_BATCH(MI_STORE_REGISTER_MEM | (4 - 2));
+ OUT_BATCH(GEN8_R_PWR_CLK_STATE);
+ OUT_RELOC(bo, I915_GEM_DOMAIN_INSTRUCTION, I915_GEM_DOMAIN_INSTRUCTION, 0);
+ ADVANCE_BATCH();
+
+ intel_batchbuffer_flush_on_ring(batch, I915_EXEC_RENDER);
+
+ drm_intel_bo_wait_rendering(bo);
+
+ ret = drm_intel_bo_map(bo, false /* write enable */);
+ igt_assert_eq(ret, 0);
+
+ rpcs = *((uint32_t *) bo->virtual);
+
+ drm_intel_bo_unmap(bo);
+ drm_intel_bo_unreference(bo);
+
+ intel_batchbuffer_free(batch);
+
+ return rpcs;
+}
+
+
+static uint32_t
+read_slice_count(drm_intel_bufmgr *bufmgr,
+ drm_intel_context *context)
+{
+ return (read_rpcs_reg(bufmgr, context) & GEN8_RPCS_S_CNT_MASK)
+ >> GEN8_RPCS_S_CNT_SHIFT;
+}
+
+static void
+context_set_slice_count(drm_intel_context *context, uint32_t mask)
+{
+ struct drm_i915_gem_context_param arg;
+ union drm_i915_gem_context_param_sseu *sseu;
+ uint32_t context_id;
+ int ret;
+
+ ret = drm_intel_gem_context_get_id(context, &context_id);
+ igt_assert_eq(ret, 0);
+
+ memset(&arg, 0, sizeof(arg));
+ arg.ctx_id = context_id;
+ arg.param = I915_CONTEXT_PARAM_SSEU;
+ sseu = (union drm_i915_gem_context_param_sseu *) &arg.value;
+
+ do_ioctl(drm_fd, DRM_IOCTL_I915_GEM_CONTEXT_GETPARAM, &arg);
+
+ sseu->packed.slice_mask = mask;
+
+ do_ioctl(drm_fd, DRM_IOCTL_I915_GEM_CONTEXT_SETPARAM, &arg);
+}
+
+static void rpcs_program_gt(uint32_t gt)
+{
+ drm_intel_bufmgr *bufmgr;
+ drm_intel_context *context1, *context2;
+
+
+ bufmgr = drm_intel_bufmgr_gem_init(drm_fd, 4096);
+ igt_assert(bufmgr);
+
+ context1 = drm_intel_gem_context_create(bufmgr);
+ igt_assert(context1);
+
+ context2 = drm_intel_gem_context_create(bufmgr);
+ igt_assert(context2);
+
+ switch (gt) {
+ case 3:
+ context_set_slice_count(context1, 0x1);
+ context_set_slice_count(context2, 0x3);
+
+ igt_assert_eq(1, read_slice_count(bufmgr, context1));
+ igt_assert_eq(2, read_slice_count(bufmgr, context2));
+
+ context_set_slice_count(context1, 0x3);
+ context_set_slice_count(context2, 0x1);
+
+ igt_assert_eq(2, read_slice_count(bufmgr, context1));
+ igt_assert_eq(1, read_slice_count(bufmgr, context2));
+ break;
+
+ case 4:
+ context_set_slice_count(context1, 0x1);
+ context_set_slice_count(context2, 0x7);
+
+ igt_assert_eq(3, read_slice_count(bufmgr, context2));
+ igt_assert_eq(1, read_slice_count(bufmgr, context1));
+
+ context_set_slice_count(context1, 0x5);
+
+ igt_assert_eq(2, read_slice_count(bufmgr, context1));
+ break;
+
+ default:
+ igt_assert(0);
+ break;
+ }
+
+ drm_intel_gem_context_destroy(context1);
+ drm_intel_gem_context_destroy(context2);
+
+ drm_intel_bufmgr_destroy(bufmgr);
+}
+
+igt_main
+{
+ igt_fixture {
+ /* Use drm_open_driver to verify device existence */
+ drm_fd = drm_open_driver(DRIVER_INTEL);
+ igt_require_gem(drm_fd);
+
+ devid = intel_get_drm_devid(drm_fd);
+
+ /* Skip on Atom platforms (can you even slice?). */
+ igt_skip_on(IS_CHERRYVIEW(devid) || IS_BROXTON(devid) || IS_GEMINILAKE(devid));
+
+ /* We can only program slice count from Gen8. */
+ igt_skip_on(intel_gen(devid) < 8);
+ }
+
+ fprintf(stderr, "devid=%x\n", devid);
+ fprintf(stderr, "gt=%u\n", intel_gt(devid));
+
+ igt_subtest("rpcs-program-gt3") {
+ /* intel_gt() returns (GT - 1). */
+ igt_skip_on(intel_gt(devid) < 2);
+ rpcs_program_gt(3);
+ }
+
+ igt_subtest("rpcs-program-gt4") {
+ igt_skip_on(intel_gt(devid) < 3);
+ rpcs_program_gt(4);
+ }
+
+ igt_fixture {
+ close(drm_fd);
+ }
+}
--
2.14.1
More information about the Intel-gfx
mailing list