From: "Zebediah Figura (she/her)" <zfigura@codeweavers.com>
Subject: Re: [PATCH vkd3d v2 1/7] tests: Separate the compile_shader function from the shader_test parser.
Message-Id: <068e8292-5512-f67c-284b-f323aa1a35ca@codeweavers.com>
Date: Mon, 17 Jan 2022 17:21:37 -0600
In-Reply-To: <CAOsNvwxWeynAFy2pW8JT_GwM7gKAZ0nrSWcmSsxd5KP2QTVUoA@mail.gmail.com>
References: <20220114233308.49491-1-zfigura@codeweavers.com> <CAOsNvwxWeynAFy2pW8JT_GwM7gKAZ0nrSWcmSsxd5KP2QTVUoA@mail.gmail.com>

On 1/17/22 08:29, Henri Verbeet wrote:
> On Sat, 15 Jan 2022 at 00:33, Zebediah Figura <zfigura@codeweavers.com> wrote:
>>
>> In essence, the goal is to separate the d3d12 backend. Just separate the compiler first, though, to make the diff easier to read.
>>
>> NOTE TO SELF: Generate this diff with -C
>>
>> Signed-off-by: Zebediah Figura <zfigura@codeweavers.com>
>> ---
>>   Makefile.am                  |  41 +-
>>   include/private/vkd3d_test.h |  19 +-
>>   tests/shader_runner.c        | 862 +++++++++++++++++++++++++++++++++++
>>   tests/shader_runner.h        |  29 ++
>>   tests/shader_runner_d3d12.c  | 847 +---------------------------------
>>   5 files changed, 947 insertions(+), 851 deletions(-)
>>   create mode 100644 tests/shader_runner.c
>>   create mode 100644 tests/shader_runner.h
>>
> This patch introduces warnings for "make crosstest":
> 
>      <vkd3d>/tests/shader_runner.c:463:38: warning: format ‘%x’ expects
> argument of type ‘unsigned int’, but argument 4 has type ‘HRESULT’
> {aka ‘long int’} [-Wformat=]
>                           ok(hr == E_FAIL, "Got unexpected hr %#x.\n", hr);

These warnings are basically an existing problem with the MinGW build of 
vkd3d, which is probably why I disregarded them.

I guess the best solution here is to define HRESULT as int always.