On Tue, Nov 22, 2016 at 2:48 PM, Matt Turner <matts...@gmail.com> wrote:
> On 11/22, Aaron Watry wrote: > >> disk_cache_get returns void*, but we were storing/comparing a char*. >> >> Signed-off-by: Aaron Watry <awa...@gmail.com> >> --- >> Note that this did, and still, segfaults for me when I actually run it... >> > > Strange. It passes for me. > I guess that I should have clarified, this is also in the "building a 32-bit mesa on 64-bit host" scenario of my other two patches. If I enable the shader cache on a 64-bit build, it works fine. I'm not sure which scenario you're building for... What I'm getting in my cross-compile scenario is: /home/me/src/mesa/bin/test-driver: line 107: 10596 Segmentation fault (core dumped) "$@" > $log_file 2>&1 FAIL: glsl/tests/cache-test If I try to run the test directly, I get: Error: Test 'disk_cache_create with no environment variables' failed: Result=NULL, but expected something else. Segmentation fault (core dumped) Valgrind says: ==26330== Invalid read of size 4 ==26330== at 0x804A366: disk_cache_destroy (disk_cache.c:318) ==26330== by 0x8048FA3: test_disk_cache_create (cache_test.c:143) ==26330== by 0x8048FA3: main (cache_test.c:391) ==26330== Address 0x8 is not stack'd, malloc'd or (recently) free'd WIth that being said, I'm not seeing the warning that I thought I saw before... I'm going to go back to the drawing board with this one and see if I can address the segfault (looks like a failure to determine the user's home directory for my environment, and I'm not sure why that's happening for the 32-bit case). The warnings that I thought I was getting before are no longer present, and as you noted below should be allowed. It's possible I saw it when building with clang instead of gcc, but I'm not sure anymore (I've been carrying this patch locally for a while). --Aaron > > >> But at least the compiler is no longer complaining about the type >> mismatch. >> src/compiler/glsl/tests/cache_test.c | 6 +++--- >> 1 file changed, 3 insertions(+), 3 deletions(-) >> >> diff --git a/src/compiler/glsl/tests/cache_test.c >> b/src/compiler/glsl/tests/cache_test.c >> index 0ef05aa..c663e54 100644 >> --- a/src/compiler/glsl/tests/cache_test.c >> +++ b/src/compiler/glsl/tests/cache_test.c >> @@ -208,14 +208,14 @@ test_put_and_get(void) >> _mesa_sha1_compute(blob, sizeof(blob), blob_key); >> >> /* Ensure that disk_cache_get returns nothing before anything is >> added. */ >> - result = disk_cache_get(cache, blob_key, &size); >> + result = (char*) disk_cache_get(cache, blob_key, &size); >> > > void * is implicitly convertable to char *, so I think the code is fine > as is. Maybe if I saw the compiler warning. > > "result" could be declared void * to begin with, and that would be fine. > I just tested it, and it seems to work. > > Also, please use (char *) for casts and not (char*) in Mesa. > But, what I've also been seeing is sporadic warnings in the nir control flow tests regarding signed/unsigned comparisons.
_______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
