[Intel-gfx] [PATCH igt 4/9] igt/vc4_wait_bo: Add a test for VC4's wait-for-BO ioctl.

[Eric Anholt]

Daniel Stone <daniel at fooishbar.org> writes:

> Hi,
>
> On 3 February 2016 at 21:41, Eric Anholt <eric at anholt.net> wrote:
>> +               ret = ioctl(fd, DRM_IOCTL_VC4_WAIT_BO, &arg);
>> +               igt_assert(ret == -1 && errno == EINVAL);
>
> A couple of nitpicks: all these should be either do_ioctl() for
> success, or do_ioctl_err() for failure, which not only cuts down the
> number of lines a bit, but also shows you the exact condition which
> occurred (e.g. it'll show that errno was -EBUSY rather than expected
> -EINVAL without having to round-trip through the reporter). Similarly,
> all your igt_assert(x == y) should be igt_assert_eq(x, y), or any of
> the igt_assert_{eq,neq} variants, e.g. _u32 for comparing uint32_t.
> You can also use do_or_die(foo) as shorthand for igt_assert_eq(foo,
> 0).
>
> With those addressed:
> Reviewed-by: Daniel Stone <daniels at collabora.com>

I converted everything but one of the ones in wait_bo, and pushed the
updated series to the vc4 branch of my tree.  I'll give it another
couple of days in case anyone else has review to do on the series.

Thanks!
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 818 bytes
Desc: not available
URL: <http://lists.freedesktop.org/archives/intel-gfx/attachments/20160204/506a59a8/attachment.sig>