[PATCH v4] tests/exynos: add fimg2d performance analysis
Hyungwon Hwang
human.hwang at samsung.com
Mon Nov 9 18:59:46 PST 2015
Hello Tobias,
On Mon, 09 Nov 2015 10:47:13 +0100
Tobias Jakobi <tjakobi at math.uni-bielefeld.de> wrote:
> Hello Hyungwon,
>
>
> Hyungwon Hwang wrote:
> > Hello,
> >
> > I think this patch should update .gitignore, not for adding the
> > built binary to untracked file list.
> Thanks!
>
>
> > Also, I want to make clear about the purpose of this test program.
> > What do you want to get after this test? This program runs G2D with
> > randomly chosen number of pixel and shows the elapsed time to do
> > that. I run it on my board. But I could not find any meaning of the
> > test. If you just want to know the execution time of solid fill,
> > what about get the width and height from user and run the same tests
> > iteratively for more accurate result? Or at least, increasing
> > number of pixels?
> The test is to measure the dependency between amount of pixels the G2D
> has to process and the amount of time for the G2D to process such
> pixels.
>
> It's exactly what a performance test should do, measure the time it
> takes for a certain workload to complete.
>
> In particular the test wants to answer the question if the dependency
> stated above is of linear type.
>
> Of course it's not, since we have setup time, so at least it should be
> affine linear. But even that is not true, since you see subtle
> 'branching' when doing high density plots (that's why I added export
> of the data to Mathematica).
>
>
> What you ask for (user input) is in fact already implemented. The user
> can specify the buffer width and height, which in turn limits the size
> of the rectangle that is solid filled.
>
> If you want smaller rectangles filled, decrease buffer width and
> height, if you want bigger ones filled, increase.
>
>
> The second purpose is to stress test the G2D, as already indicated in
> the commit description. The G2D can be overclocked quite a lot under
> certain conditions. With increase MIF/INT voltages I can run it with
> 400MHz instead of the 200MHz defaults. The application can now be used
> to check stability. E.g. if voltages are too low the system can
> quickly lock-up.
>
> In particular one could also check how processing time depends on the
> clock rate of the G2D. One interesting question here is how memory
> bandwidth limits us.
>
>
>
> With best wishes,
> Tobias
Yes. I agree with the broad view. Please see the below, I run the test
2 times in a row.
root at localhost:~# ./exynos_fimg2d_perf -i 10 -w 1024 -h 1024
exynos/fimg2d: G2D version (4.1).
starting simple G2D performance test
buffer width = 1024, buffer height = 1024, iterations = 10
num_pixels = 136000, usecs = 236000
num_pixels = 8492, usecs = 47083
num_pixels = 100688, usecs = 200042
num_pixels = 141312, usecs = 216667
num_pixels = 39962, usecs = 92708
num_pixels = 95046, usecs = 156542
num_pixels = 2562, usecs = 34666
num_pixels = 176485, usecs = 326916
num_pixels = 17760, usecs = 56625
num_pixels = 1625, usecs = 31833
starting multi G2D performance test (batch size = 3)
buffer width = 1024, buffer height = 1024, iterations = 10
num_pixels = 245180, usecs = 385083
num_pixels = 276320, usecs = 398625
num_pixels = 196807, usecs = 356666
num_pixels = 305540, usecs = 420458
num_pixels = 65978, usecs = 120250
num_pixels = 265028, usecs = 379417
num_pixels = 139079, usecs = 213667
num_pixels = 24970, usecs = 67625
num_pixels = 46808, usecs = 114125
num_pixels = 100804, usecs = 179750
root at localhost:~# ./exynos_fimg2d_perf -i 10 -w 1024 -h 1024
exynos/fimg2d: G2D version (4.1).
starting simple G2D performance test
buffer width = 1024, buffer height = 1024, iterations = 10
num_pixels = 18676, usecs = 95541
num_pixels = 117056, usecs = 218875
num_pixels = 80784, usecs = 137209
num_pixels = 427, usecs = 33209
num_pixels = 238044, usecs = 403041
num_pixels = 4392, usecs = 37709
num_pixels = 19880, usecs = 59750
num_pixels = 3666, usecs = 36542
num_pixels = 4630, usecs = 36166
num_pixels = 70834, usecs = 125917
starting multi G2D performance test (batch size = 3)
buffer width = 1024, buffer height = 1024, iterations = 10
num_pixels = 216516, usecs = 347042
num_pixels = 242863, usecs = 422417
num_pixels = 28176, usecs = 72292
num_pixels = 110713, usecs = 179167
num_pixels = 292266, usecs = 431750
num_pixels = 274127, usecs = 392833
num_pixels = 291659, usecs = 415875
num_pixels = 140202, usecs = 218833
num_pixels = 122400, usecs = 193084
num_pixels = 168647, usecs = 251375
As you said, I can adjust the buffer width and height. But because the
program choose the number of pixel to process randomly, I can't compare
the result after I modified something (clock, or something else like
you mentioned). Also I can figure out the tendency between the number
of pixels and the processing time after I draw the graph. But it is too
hard not by doing that, because the number of pixels are not in
increasing order at least. I think that this program will be more useful
if the user can control the whole situation more tightly. But if you
need this kind of test, then it's OK to me.
Best regards,
Hyungwon Hwang
>
>
> >
> >
> > Best regards,
> > Hyungwon Hwang
> >
> >
> > On Mon, 02 Nov 2015 10:52:09 +0100
> > Tobias Jakobi <tjakobi at math.uni-bielefeld.de> wrote:
> >
> >> Currently only fast solid color clear performance is measured.
> >> A large buffer is allocated and solid color clear operations
> >> are executed on it with randomly chosen properties (position
> >> and size of the region, clear color). Execution time is
> >> measured and output together with the amount of pixels
> >> processed.
> >>
> >> The 'simple' variant only executes one G2D command buffer at
> >> a time, while the 'multi' variant executes multiple ones. This
> >> can be used to measure setup/exec overhead.
> >>
> >> The test also serves a stability check. If clocks/voltages are
> >> too high or low respectively, the test quickly reveals this.
> >>
> >> v2: Add GPLv2 header, argument handling and documentation.
> >> Tool is only installed when requested.
> >> v3: Free images array in fimg2d_perf_multi() as pointed out
> >> by Hyungwon Hwang.
> >> v4: Include header for error numbers (fixes build).
> >>
> >> Signed-off-by: Tobias Jakobi <tjakobi at math.uni-bielefeld.de>
> >> ---
> >> tests/exynos/Makefile.am | 19 ++-
> >> tests/exynos/exynos_fimg2d_perf.c | 327
> >> ++++++++++++++++++++++++++++++++++++++ 2 files changed, 344
> >> insertions(+), 2 deletions(-) create mode 100644
> >> tests/exynos/exynos_fimg2d_perf.c
> >>
> >> diff --git a/tests/exynos/Makefile.am b/tests/exynos/Makefile.am
> >> index b21d016..e82d199 100644
> >> --- a/tests/exynos/Makefile.am
> >> +++ b/tests/exynos/Makefile.am
> >> @@ -5,16 +5,31 @@ AM_CFLAGS = \
> >> -I $(top_srcdir)/exynos \
> >> -I $(top_srcdir)
> >>
> >> +bin_PROGRAMS =
> >> +noinst_PROGRAMS =
> >> +
> >> if HAVE_LIBKMS
> >> if HAVE_INSTALL_TESTS
> >> -bin_PROGRAMS = \
> >> +bin_PROGRAMS += \
> >> exynos_fimg2d_test
> >> else
> >> -noinst_PROGRAMS = \
> >> +noinst_PROGRAMS += \
> >> exynos_fimg2d_test
> >> endif
> >> endif
> >>
> >> +if HAVE_INSTALL_TESTS
> >> +bin_PROGRAMS += \
> >> + exynos_fimg2d_perf
> >> +else
> >> +noinst_PROGRAMS += \
> >> + exynos_fimg2d_perf
> >> +endif
> >> +
> >> +exynos_fimg2d_perf_LDADD = \
> >> + $(top_builddir)/libdrm.la \
> >> + $(top_builddir)/exynos/libdrm_exynos.la
> >> +
> >> exynos_fimg2d_test_LDADD = \
> >> $(top_builddir)/libdrm.la \
> >> $(top_builddir)/libkms/libkms.la \
> >> diff --git a/tests/exynos/exynos_fimg2d_perf.c
> >> b/tests/exynos/exynos_fimg2d_perf.c new file mode 100644
> >> index 0000000..1699bba
> >> --- /dev/null
> >> +++ b/tests/exynos/exynos_fimg2d_perf.c
> >> @@ -0,0 +1,327 @@
> >> +/*
> >> + * Copyright (C) 2015 - Tobias Jakobi
> >> + *
> >> + * This is free software: you can redistribute it and/or modify
> >> + * it under the terms of the GNU General Public License as
> >> published
> >> + * by the Free Software Foundation, either version 2 of the
> >> License,
> >> + * or (at your option) any later version.
> >> + *
> >> + * It is distributed in the hope that it will be useful, but
> >> + * WITHOUT ANY WARRANTY; without even the implied warranty of
> >> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> >> + * GNU General Public License for more details.
> >> + * You should have received a copy of the GNU General Public
> >> License
> >> + * along with it. If not, see <http://www.gnu.org/licenses/>.
> >> + */
> >> +
> >> +#include <stdlib.h>
> >> +#include <stdio.h>
> >> +#include <time.h>
> >> +#include <getopt.h>
> >> +#include <errno.h>
> >> +
> >> +#include <xf86drm.h>
> >> +
> >> +#include "exynos_drm.h"
> >> +#include "exynos_drmif.h"
> >> +#include "exynos_fimg2d.h"
> >> +
> >> +static int output_mathematica = 0;
> >> +
> >> +static int fimg2d_perf_simple(struct exynos_bo *bo, struct
> >> g2d_context *ctx,
> >> + unsigned buf_width, unsigned buf_height,
> >> unsigned iterations) +{
> >> + struct timespec tspec = { 0 };
> >> + struct g2d_image img = { 0 };
> >> +
> >> + unsigned long long g2d_time;
> >> + unsigned i;
> >> + int ret = 0;
> >> +
> >> + img.width = buf_width;
> >> + img.height = buf_height;
> >> + img.stride = buf_width * 4;
> >> + img.color_mode = G2D_COLOR_FMT_ARGB8888 | G2D_ORDER_AXRGB;
> >> + img.buf_type = G2D_IMGBUF_GEM;
> >> + img.bo[0] = bo->handle;
> >> +
> >> + srand(time(NULL));
> >> +
> >> + printf("starting simple G2D performance test\n");
> >> + printf("buffer width = %u, buffer height = %u, iterations
> >> = %u\n",
> >> + buf_width, buf_height, iterations);
> >> +
> >> + if (output_mathematica)
> >> + putchar('{');
> >> +
> >> + for (i = 0; i < iterations; ++i) {
> >> + unsigned x, y, w, h;
> >> +
> >> + x = rand() % buf_width;
> >> + y = rand() % buf_height;
> >> +
> >> + if (x == (buf_width - 1))
> >> + x -= 1;
> >> + if (y == (buf_height - 1))
> >> + y -= 1;
> >> +
> >> + w = rand() % (buf_width - x);
> >> + h = rand() % (buf_height - y);
> >> +
> >> + if (w == 0) w = 1;
> >> + if (h == 0) h = 1;
> >> +
> >> + img.color = rand();
> >> +
> >> + ret = g2d_solid_fill(ctx, &img, x, y, w, h);
> >> +
> >> + clock_gettime(CLOCK_MONOTONIC, &tspec);
> >> +
> >> + if (ret == 0)
> >> + ret = g2d_exec(ctx);
> >> +
> >> + if (ret != 0) {
> >> + fprintf(stderr, "error: iteration %u
> >> failed (x = %u, y = %u, w = %u, h = %u)\n",
> >> + i, x, y, w, h);
> >> + break;
> >> + } else {
> >> + struct timespec end = { 0 };
> >> + clock_gettime(CLOCK_MONOTONIC, &end);
> >> +
> >> + g2d_time = (end.tv_sec - tspec.tv_sec) *
> >> 1000000000ULL;
> >> + g2d_time += (end.tv_nsec - tspec.tv_nsec);
> >> +
> >> + if (output_mathematica) {
> >> + if (i != 0) putchar(',');
> >> + printf("{%u,%llu}", w * h,
> >> g2d_time);
> >> + } else {
> >> + printf("num_pixels = %u, usecs =
> >> %llu\n", w * h, g2d_time);
> >> + }
> >> + }
> >> + }
> >> +
> >> + if (output_mathematica)
> >> + printf("}\n");
> >> +
> >> + return ret;
> >> +}
> >> +
> >> +static int fimg2d_perf_multi(struct exynos_bo *bo, struct
> >> g2d_context *ctx,
> >> + unsigned buf_width, unsigned buf_height,
> >> unsigned iterations, unsigned batch) +{
> >> + struct timespec tspec = { 0 };
> >> + struct g2d_image *images;
> >> +
> >> + unsigned long long g2d_time;
> >> + unsigned i, j;
> >> + int ret = 0;
> >> +
> >> + images = calloc(batch, sizeof(struct g2d_image));
> >> + if (images == NULL) {
> >> + fprintf(stderr, "error: failed to allocate G2D
> >> images.\n");
> >> + return -ENOMEM;
> >> + }
> >> +
> >> + for (i = 0; i < batch; ++i) {
> >> + images[i].width = buf_width;
> >> + images[i].height = buf_height;
> >> + images[i].stride = buf_width * 4;
> >> + images[i].color_mode = G2D_COLOR_FMT_ARGB8888 |
> >> G2D_ORDER_AXRGB;
> >> + images[i].buf_type = G2D_IMGBUF_GEM;
> >> + images[i].bo[0] = bo->handle;
> >> + }
> >> +
> >> + srand(time(NULL));
> >> +
> >> + printf("starting multi G2D performance test (batch size =
> >> %u)\n", batch);
> >> + printf("buffer width = %u, buffer height = %u, iterations
> >> = %u\n",
> >> + buf_width, buf_height, iterations);
> >> +
> >> + if (output_mathematica)
> >> + putchar('{');
> >> +
> >> + for (i = 0; i < iterations; ++i) {
> >> + unsigned num_pixels = 0;
> >> +
> >> + for (j = 0; j < batch; ++j) {
> >> + unsigned x, y, w, h;
> >> +
> >> + x = rand() % buf_width;
> >> + y = rand() % buf_height;
> >> +
> >> + if (x == (buf_width - 1))
> >> + x -= 1;
> >> + if (y == (buf_height - 1))
> >> + y -= 1;
> >> +
> >> + w = rand() % (buf_width - x);
> >> + h = rand() % (buf_height - y);
> >> +
> >> + if (w == 0) w = 1;
> >> + if (h == 0) h = 1;
> >> +
> >> + images[j].color = rand();
> >> +
> >> + num_pixels += w * h;
> >> +
> >> + ret = g2d_solid_fill(ctx, &images[j], x,
> >> y, w, h);
> >> + if (ret != 0)
> >> + break;
> >> + }
> >> +
> >> + clock_gettime(CLOCK_MONOTONIC, &tspec);
> >> +
> >> + if (ret == 0)
> >> + ret = g2d_exec(ctx);
> >> +
> >> + if (ret != 0) {
> >> + fprintf(stderr, "error: iteration %u
> >> failed (num_pixels = %u)\n", i, num_pixels);
> >> + break;
> >> + } else {
> >> + struct timespec end = { 0 };
> >> + clock_gettime(CLOCK_MONOTONIC, &end);
> >> +
> >> + g2d_time = (end.tv_sec - tspec.tv_sec) *
> >> 1000000000ULL;
> >> + g2d_time += (end.tv_nsec - tspec.tv_nsec);
> >> +
> >> + if (output_mathematica) {
> >> + if (i != 0) putchar(',');
> >> + printf("{%u,%llu}", num_pixels,
> >> g2d_time);
> >> + } else {
> >> + printf("num_pixels = %u, usecs =
> >> %llu\n", num_pixels, g2d_time);
> >> + }
> >> + }
> >> + }
> >> +
> >> + if (output_mathematica)
> >> + printf("}\n");
> >> +
> >> + free(images);
> >> +
> >> + return ret;
> >> +}
> >> +
> >> +static void usage(const char *name)
> >> +{
> >> + fprintf(stderr, "usage: %s [-ibwh]\n\n", name);
> >> +
> >> + fprintf(stderr, "\t-i <number of iterations>\n");
> >> + fprintf(stderr, "\t-b <size of a batch> (default =
> >> 3)\n\n"); +
> >> + fprintf(stderr, "\t-w <buffer width> (default = 4096)\n");
> >> + fprintf(stderr, "\t-h <buffer height> (default =
> >> 4096)\n\n"); +
> >> + fprintf(stderr, "\t-M <enable Mathematica styled
> >> output>\n"); +
> >> + exit(0);
> >> +}
> >> +
> >> +int main(int argc, char **argv)
> >> +{
> >> + int fd, ret, c, parsefail;
> >> +
> >> + struct exynos_device *dev;
> >> + struct g2d_context *ctx;
> >> + struct exynos_bo *bo;
> >> +
> >> + unsigned int iters = 0, batch = 3;
> >> + unsigned int bufw = 4096, bufh = 4096;
> >> +
> >> + ret = 0;
> >> + parsefail = 0;
> >> +
> >> + while ((c = getopt(argc, argv, "i:b:w:h:M")) != -1) {
> >> + switch (c) {
> >> + case 'i':
> >> + if (sscanf(optarg, "%u", &iters) != 1)
> >> + parsefail = 1;
> >> + break;
> >> + case 'b':
> >> + if (sscanf(optarg, "%u", &batch) != 1)
> >> + parsefail = 1;
> >> + break;
> >> + case 'w':
> >> + if (sscanf(optarg, "%u", &bufw) != 1)
> >> + parsefail = 1;
> >> + break;
> >> + case 'h':
> >> + if (sscanf(optarg, "%u", &bufh) != 1)
> >> + parsefail = 1;
> >> + break;
> >> + case 'M':
> >> + output_mathematica = 1;
> >> + break;
> >> + default:
> >> + parsefail = 1;
> >> + break;
> >> + }
> >> + }
> >> +
> >> + if (parsefail || (argc == 1) || (iters == 0))
> >> + usage(argv[0]);
> >> +
> >> + if (bufw < 2 || bufw > 4096 || bufh < 2 || bufh > 4096) {
> >> + fprintf(stderr, "error: buffer width/height should
> >> be in the range 2 to 4096.\n");
> >> + ret = -1;
> >> +
> >> + goto out;
> >> + }
> >> +
> >> + if (bufw == 0 || bufh == 0) {
> >> + fprintf(stderr, "error: buffer width/height should
> >> be non-zero.\n");
> >> + ret = -1;
> >> +
> >> + goto out;
> >> + }
> >> +
> >> + fd = drmOpen("exynos", NULL);
> >> + if (fd < 0) {
> >> + fprintf(stderr, "error: failed to open drm\n");
> >> + ret = -1;
> >> +
> >> + goto out;
> >> + }
> >> +
> >> + dev = exynos_device_create(fd);
> >> + if (dev == NULL) {
> >> + fprintf(stderr, "error: failed to create
> >> device\n");
> >> + ret = -2;
> >> +
> >> + goto fail;
> >> + }
> >> +
> >> + ctx = g2d_init(fd);
> >> + if (ctx == NULL) {
> >> + fprintf(stderr, "error: failed to init G2D\n");
> >> + ret = -3;
> >> +
> >> + goto g2d_fail;
> >> + }
> >> +
> >> + bo = exynos_bo_create(dev, bufw dsa* bufh * 4, 0);
> >> + if (bo == NULL) {
> >> + fprintf(stderr, "error: failed to create bo\n");
> >> + ret = -4;
> >> +
> >> + goto bo_fail;
> >> + }
> >> +
> >> + ret = fimg2d_perf_simple(bo, ctx, bufw, bufh, iters);
> >> +
> >> + if (ret == 0)
> >> + ret = fimg2d_perf_multi(bo, ctx, bufw, bufh,
> >> iters, batch); +
> >> + exynos_bo_destroy(bo);
> >> +
> >> +bo_fail:
> >> + g2d_fini(ctx);
> >> +
> >> +g2d_fail:
> >> + exynos_device_destroy(dev);
> >> +
> >> +fail:
> >> + drmClose(fd);
> >> +
> >> +out:
> >> + return ret;
> >> +}
> >
>
More information about the dri-devel
mailing list