[Bug 109304] GfxBench AztecRuins Vulkan version Segfault

bugzilla-daemon at freedesktop.org bugzilla-daemon at freedesktop.org
Fri Jan 11 09:41:41 UTC 2019


https://bugs.freedesktop.org/show_bug.cgi?id=109304

            Bug ID: 109304
           Summary: GfxBench AztecRuins Vulkan version Segfault
           Product: Mesa
           Version: git
          Hardware: Other
                OS: All
            Status: NEW
          Severity: normal
          Priority: medium
         Component: Drivers/Vulkan/intel
          Assignee: intel-3d-bugs at lists.freedesktop.org
          Reporter: eero.t.tamminen at intel.com
        QA Contact: intel-3d-bugs at lists.freedesktop.org
                CC: jason at jlekstrand.net

Setup:
* Ubuntu 18.04
* Latest Mesa from git
* Latest X or Weston from git
* drm-tip git kernel v4.20 or newer
* GfxBench v5 (X or Wayland build, GOLD2 or GOLD3 version)

Test-case:
* Run AztecRuins (normal or high) Vulkan version:
bin/testfw_app --gfx vulkan --gl_api vulkan --width 1920 --height 1080
--fullscreen 1 --test_id vulkan_5_normal

Result:
* Segfault when run from Gdb (abort when run without as GfxBench catches them
and aborts):
--------------------------------------
...
compile light_directional_cube.frag...
done

SPIR-V WARNING:
    In file
../../../../../source2/_repos/git___anongit_freedesktop_org_git_mesa_mesa/src/compiler/spirv/vtn_variables.c:2258
    OpStore of a sampler detected.  Doing on-the-fly copy propagation to
workaround the problem.
    9408 bytes into the SPIR-V binary

Thread 2 "testfw_app" received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7ffff3339700 (LWP 10251)]
--------------------------------------

This doesn't happen with GL or GLES versions, only Vulkan.  This regression
started between following commits:
* 2019-01-07 17:07:33 0cc01f45e7: glsl: correct typo in GLSL compilation error
message
* 2019-01-08 17:51:46 d0c6ef2793: nir: rename global/local to private/function
memory


Valgrind says following:
--------------------------------------
SPIR-V WARNING:
    In file src/compiler/spirv/vtn_variables.c:2258
    OpStore of a sampler detected.  Doing on-the-fly copy propagation to
workaround the problem.
    9408 bytes into the SPIR-V binary
==10457== Invalid read of size 4
==10457==    at 0xA54E458: try_lower_tex_ycbcr
(anv_nir_lower_ycbcr_textures.c:328)
==10457==    by 0xA54E458: anv_nir_lower_ycbcr_textures
(anv_nir_lower_ycbcr_textures.c:464)
==10457==    by 0xA55372F: anv_pipeline_lower_nir (anv_pipeline.c:500)
==10457==    by 0xA555FEE: anv_pipeline_compile_graphics (anv_pipeline.c:1034)
==10457==    by 0xA5577BB: anv_pipeline_init (anv_pipeline.c:1463)
==10457==    by 0xA676675: gen9_graphics_pipeline_create (genX_pipeline.c:1787)
==10457==    by 0xA676675: gen9_CreateGraphicsPipelines (genX_pipeline.c:2001)
==10457==    by 0x917FCC: ??? (in bin/testfw_app)
==10457==  Address 0x3c is not stack'd, malloc'd or (recently) free'd
==10457== 
==10457== 
==10457== Process terminating with default action of signal 6 (SIGABRT)
==10457==    at 0x7239E97: raise (raise.c:51)
==10457==    by 0x723B800: abort (abort.c:79)
==10457==    by 0x6872B5: ??? (in bin/testfw_app)
==10457==    by 0x623888F: ??? (in /lib/x86_64-linux-gnu/libpthread-2.27.so)
==10457==    by 0xA54E457: try_lower_tex_ycbcr
(anv_nir_lower_ycbcr_textures.c:396)
==10457==    by 0xA54E457: anv_nir_lower_ycbcr_textures
(anv_nir_lower_ycbcr_textures.c:464)
==10457==    by 0xA55372F: anv_pipeline_lower_nir (anv_pipeline.c:500)
==10457==    by 0xA555FEE: anv_pipeline_compile_graphics (anv_pipeline.c:1034)
==10457==    by 0xA5577BB: anv_pipeline_init (anv_pipeline.c:1463)
==10457==    by 0xA676675: gen9_graphics_pipeline_create (genX_pipeline.c:1787)
==10457==    by 0xA676675: gen9_CreateGraphicsPipelines (genX_pipeline.c:2001)
--------------------------------------

-- 
You are receiving this mail because:
You are the QA Contact for the bug.
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-3d-bugs/attachments/20190111/75e73cce/attachment-0001.html>


More information about the intel-3d-bugs mailing list