[PATCH v2 libdrm 6/7] tegra: Add gr2d-fill test

[Erik Faye-Lund]

On Fri, May 2, 2014 at 4:25 PM, Thierry Reding <thierry.reding at gmail.com> wrote:
> On Thu, Apr 10, 2014 at 07:28:16PM +0200, Erik Faye-Lund wrote:
>> On Wed, Apr 9, 2014 at 1:40 PM, Thierry Reding <thierry.reding at gmail.com> wrote:
>> > diff --git a/tests/tegra/gr2d-fill.c b/tests/tegra/gr2d-fill.c
>> > new file mode 100644
>> > index 000000000000..b6ba35a9d668
>> > --- /dev/null
>> > +++ b/tests/tegra/gr2d-fill.c
>> > @@ -0,0 +1,146 @@
>> > +/*
>> > + * Copyright © 2014 NVIDIA 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 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.
>> > + */
>> > +
>> > +#ifdef HAVE_CONFIG_H
>> > +#  include "config.h"
>> > +#endif
>> > +
>> > +#include <errno.h>
>> > +#include <fcntl.h>
>> > +#include <stdbool.h>
>> > +#include <stdint.h>
>> > +#include <stdio.h>
>> > +#include <stdlib.h>
>> > +#include <string.h>
>> > +#include <unistd.h>
>> > +
>> > +#include <sys/ioctl.h>
>> > +
>> > +#include "xf86drm.h"
>> > +#include "xf86drmMode.h"
>> > +#include "drm_fourcc.h"
>> > +
>> > +#include "drm-test-tegra.h"
>> > +#include "tegra.h"
>> > +
>> > +int main(int argc, char *argv[])
>> > +{
>> > +       uint32_t format = DRM_FORMAT_XRGB8888;
>> > +       struct drm_tegra_gr2d *gr2d;
>> > +       struct drm_framebuffer *fb;
>> > +       struct drm_screen *screen;
>> > +       unsigned int pitch, size;
>> > +       struct drm_tegra_bo *bo;
>> > +       struct drm_tegra *drm;
>> > +       uint32_t handle;
>> > +       int fd, err;
>> > +       void *ptr;
>> > +
>> > +       fd = drm_open(argv[1]);
>> > +       if (fd < 0) {
>> > +               fprintf(stderr, "failed to open DRM device %s: %s\n", argv[1],
>> > +                       strerror(errno));
>> > +               return 1;
>> > +       }
>>
>> I'm not quite sure I understand this part. Why would argv[1] be
>> anything else than NULL? Is this useful for manual debugging, perhaps?
>
> Yes. On newer Tegra generations the nouveau driver can create its own
> device files in /dev/dri and depending on device probe order, the card0
> device can be the wrong one. For such cases these test programs can be
> passed the path to the correct device via the command-line.
>
> Probably a better approach would be to search for a compatible device
> via udev matching or iterating over /dev/dri/card* and querying the
> driver name (which is what drm_tegra_new() already does). But that seems
> overkill currently. Perhaps when this turns into a more fully-fledged
> test suite that could be implemented more cleverly.
>
> For now I've addressed this by making the tests fallback to a default
> device (/dev/dri/card0) if no argument has been specified.
>

Thanks for the explanation, makes sense!

>> > +       err = drm_tegra_gr2d_fill(gr2d, fb, fb->width / 4, fb->height / 4,
>> > +                                 fb->width / 2, fb->height / 2, 0x00000000);
>> > +       if (err < 0) {
>> > +               fprintf(stderr, "failed to fill rectangle: %s\n",
>> > +                       strerror(-err));
>> > +               return 1;
>> > +       }
>> > +
>> > +       sleep(1);
>> > +
>>
>> Why do we need to sleep here?
>
> This is a visual test, so that one second gives me some time to see if
> the result looks as expected.
>
> I have tentative plans to implement a set of more automated tests, where
> an image is rendered and the checksum computed over the content can be
> compared with a known good reference checksum, but for now this is
> better than nothing.

Aha, makes sense. Thanks, looks good!