flywithu/ntel-gpu-tools
1
0
Fork 0
mirror of https://github.com/tiagovignatti/intel-gpu-tools.git synced 2025-11-04 12:07:12 +00:00
Code Issues Releases Wiki Activity

igt/gem_concurrent_blit: Apply some fence pressure as well

Browse Source 
As before, we also want to race against access through the fence
registers. This overlaps slightly with gem_set_tiling_vs_blt, but the
different access pattern should make it useful.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
This commit is contained in:
Chris Wilson 2014-08-29 17:36:29 +01:00
parent d28b9d27bf
commit 86055df968
1 changed files with 75 additions and 27 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

102
tests/gem_concurrent_blit.c
View File

@ -100,7 +100,7 @@ prw_cmp_bo(drm_intel_bo *bo, uint32_t val, int width, int height)
}
static drm_intel_bo *
unmapped_create_bo(drm_intel_bufmgr *bufmgr, uint32_t val, int width, int height)
unmapped_create_bo(drm_intel_bufmgr *bufmgr, int width, int height)
{
drm_intel_bo *bo;
@ -135,13 +135,8 @@ gtt_cmp_bo(drm_intel_bo *bo, uint32_t val, int width, int height)
}
static drm_intel_bo *
gtt_create_bo(drm_intel_bufmgr *bufmgr, uint32_t val, int width, int height)
map_bo(drm_intel_bo *bo)
{
drm_intel_bo *bo;
bo = drm_intel_bo_alloc(bufmgr, "bo", 4*width*height, 0);
igt_assert(bo);
/* gtt map doesn't have a write parameter, so just keep the mapping
* around (to avoid the set_domain with the gtt write domain set) and
* manually tell the kernel when we start access the gtt. */
@ -150,6 +145,42 @@ gtt_create_bo(drm_intel_bufmgr *bufmgr, uint32_t val, int width, int height)
return bo;
}
static drm_intel_bo *
tile_bo(drm_intel_bo *bo, int width)
{
uint32_t tiling = I915_TILING_X;
uint32_t stride = width * 4;
do_or_die(drm_intel_bo_set_tiling(bo, &tiling, stride));
return bo;
}
static drm_intel_bo *
gtt_create_bo(drm_intel_bufmgr *bufmgr, int width, int height)
{
return map_bo(unmapped_create_bo(bufmgr, width, height));
}
static drm_intel_bo *
gttX_create_bo(drm_intel_bufmgr *bufmgr, int width, int height)
{
return tile_bo(gtt_create_bo(bufmgr, width, height), width);
}
static drm_intel_bo *
gpu_create_bo(drm_intel_bufmgr *bufmgr, int width, int height)
{
return unmapped_create_bo(bufmgr, width, height);
}
static drm_intel_bo *
gpuX_create_bo(drm_intel_bufmgr *bufmgr, int width, int height)
{
return tile_bo(gpu_create_bo(bufmgr, width, height), width);
}
static void
cpu_set_bo(drm_intel_bo *bo, uint32_t val, int width, int height)
{
@ -185,6 +216,9 @@ gpu_set_bo(drm_intel_bo *bo, uint32_t val, int width, int height)
struct drm_i915_gem_pwrite gem_pwrite;
struct drm_i915_gem_create create;
uint32_t buf[10], *b;
uint32_t tiling, swizzle;
drm_intel_bo_get_tiling(bo, &tiling, &swizzle);
memset(reloc, 0, sizeof(reloc));
memset(gem_exec, 0, sizeof(gem_exec));
@ -194,7 +228,12 @@ gpu_set_bo(drm_intel_bo *bo, uint32_t val, int width, int height)
*b++ = XY_COLOR_BLT_CMD_NOLEN |
((gen >= 8) ? 5 : 4) |
COLOR_BLT_WRITE_ALPHA | XY_COLOR_BLT_WRITE_RGB;
*b++ = 0xf0 << 16 | 1 << 25 | 1 << 24 | width << 2;
if (gen >= 4 && tiling) {
b[-1] |= XY_COLOR_BLT_TILED;
*b = width;
} else
*b = width << 2;
*b++ |= 0xf0 << 16 | 1 << 25 | 1 << 24;
*b++ = 0;
*b++ = height << 16 | width;
reloc[0].offset = (b - buf) * sizeof(uint32_t);
@ -222,9 +261,8 @@ gpu_set_bo(drm_intel_bo *bo, uint32_t val, int width, int height)
execbuf.buffers_ptr = (uintptr_t)gem_exec;
execbuf.buffer_count = 2;
execbuf.batch_len = (b - buf) * sizeof(buf[0]);
execbuf.flags = 1 << 11;
if (HAS_BLT_RING(devid))
execbuf.flags |= I915_EXEC_BLT;
if (gen >= 6)
execbuf.flags = I915_EXEC_BLT;
gem_pwrite.handle = gem_exec[1].handle;
gem_pwrite.offset = 0;
@ -248,8 +286,7 @@ gpu_cmp_bo(drm_intel_bo *bo, uint32_t val, int width, int height)
struct access_mode {
void (*set_bo)(drm_intel_bo *bo, uint32_t val, int w, int h);
void (*cmp_bo)(drm_intel_bo *bo, uint32_t val, int w, int h);
drm_intel_bo *(*create_bo)(drm_intel_bufmgr *bufmgr,
uint32_t val, int width, int height);
drm_intel_bo *(*create_bo)(drm_intel_bufmgr *bufmgr, int width, int height);
const char *name;
};
@ -260,14 +297,17 @@ struct access_mode access_modes[] = {
.create_bo = unmapped_create_bo, .name = "cpu" },
{ .set_bo = gtt_set_bo, .cmp_bo = gtt_cmp_bo,
.create_bo = gtt_create_bo, .name = "gtt" },
{ .set_bo = gtt_set_bo, .cmp_bo = gtt_cmp_bo,
.create_bo = gttX_create_bo, .name = "gttX" },
{ .set_bo = gpu_set_bo, .cmp_bo = gpu_cmp_bo,
.create_bo = unmapped_create_bo, .name = "gpu" },
.create_bo = gpu_create_bo, .name = "gpu" },
{ .set_bo = gpu_set_bo, .cmp_bo = gpu_cmp_bo,
.create_bo = gpuX_create_bo, .name = "gpuX" },
};
#define MAX_NUM_BUFFERS 1024
int num_buffers = MAX_NUM_BUFFERS;
drm_intel_bufmgr *bufmgr;
int width = 512, height = 512;
const int width = 512, height = 512;
igt_render_copyfunc_t rendercopy;
typedef void (*do_copy)(drm_intel_bo *dst, drm_intel_bo *src);
@ -277,16 +317,19 @@ static void render_copy_bo(drm_intel_bo *dst, drm_intel_bo *src)
struct igt_buf d = {
.bo = dst,
.size = width * height * 4,
.tiling = I915_TILING_NONE,
.num_tiles = width * height * 4,
.stride = width * 4,
}, s = {
.bo = src,
.size = width * height * 4,
.tiling = I915_TILING_NONE,
.num_tiles = width * height * 4,
.stride = width * 4,
};
uint32_t swizzle;
drm_intel_bo_get_tiling(dst, &d.tiling, &swizzle);
drm_intel_bo_get_tiling(src, &s.tiling, &swizzle);
igt_require(rendercopy);
rendercopy(batch, NULL,
&s, 0, 0,
@ -296,7 +339,10 @@ static void render_copy_bo(drm_intel_bo *dst, drm_intel_bo *src)
static void blt_copy_bo(drm_intel_bo *dst, drm_intel_bo *src)
{
intel_copy_bo(batch, dst, src, width*height*4);
intel_blt_copy(batch,
src, 0, 0, 4*width,
dst, 0, 0, 4*width,
width, height, 32);
}
static void do_overwrite_source(struct access_mode *mode,
@ -392,6 +438,7 @@ static void run_forked(struct access_mode *mode,
do_copy do_copy_func)
{
const int old_num_buffers = num_buffers;
drm_intel_bufmgr *bufmgr;
num_buffers /= 16;
num_buffers += 2;
@ -400,12 +447,12 @@ static void run_forked(struct access_mode *mode,
/* recreate process local variables */
bufmgr = drm_intel_bufmgr_gem_init(fd, 4096);
drm_intel_bufmgr_gem_enable_reuse(bufmgr);
batch = intel_batchbuffer_alloc(bufmgr, intel_get_drm_devid(fd));
batch = intel_batchbuffer_alloc(bufmgr, devid);
for (int i = 0; i < num_buffers; i++) {
src[i] = mode->create_bo(bufmgr, i, width, height);
dst[i] = mode->create_bo(bufmgr, ~i, width, height);
src[i] = mode->create_bo(bufmgr, width, height);
dst[i] = mode->create_bo(bufmgr, width, height);
}
dummy = mode->create_bo(bufmgr, 0, width, height);
dummy = mode->create_bo(bufmgr, width, height);
for (int loop = 0; loop < 10; loop++)
do_test_func(mode, src, dst, dummy, do_copy_func);
/* as we borrow the fd, we need to reap our bo */
@ -460,17 +507,18 @@ static void
run_modes(struct access_mode *mode)
{
drm_intel_bo *src[MAX_NUM_BUFFERS], *dst[MAX_NUM_BUFFERS], *dummy = NULL;
drm_intel_bufmgr *bufmgr;
igt_fixture {
bufmgr = drm_intel_bufmgr_gem_init(fd, 4096);
drm_intel_bufmgr_gem_enable_reuse(bufmgr);
batch = intel_batchbuffer_alloc(bufmgr, intel_get_drm_devid(fd));
batch = intel_batchbuffer_alloc(bufmgr, devid);
for (int i = 0; i < num_buffers; i++) {
src[i] = mode->create_bo(bufmgr, i, width, height);
dst[i] = mode->create_bo(bufmgr, ~i, width, height);
src[i] = mode->create_bo(bufmgr, width, height);
dst[i] = mode->create_bo(bufmgr, width, height);
}
dummy = mode->create_bo(bufmgr, 0, width, height);
dummy = mode->create_bo(bufmgr, width, height);
}
run_basic_modes(mode, src, dst, dummy, "", run_single);