[Bug 103241] Anv crashes when using 64-bit vertex inputs
bugzilla-daemon at freedesktop.org
bugzilla-daemon at freedesktop.org
Thu Oct 12 11:57:03 UTC 2017
https://bugs.freedesktop.org/show_bug.cgi?id=103241
Bug ID: 103241
Summary: Anv crashes when using 64-bit vertex inputs
Product: Mesa
Version: git
Hardware: All
OS: All
Status: NEW
Severity: normal
Priority: medium
Component: Drivers/Vulkan/intel
Assignee: intel-3d-bugs at lists.freedesktop.org
Reporter: joseph.kucia at gmail.com
QA Contact: intel-3d-bugs at lists.freedesktop.org
CC: jason at jlekstrand.net
"elem_count" isn't calculated correctly for 64bit vertex inputs in
emit_vertex_input(). This leads to crash/assert() in the driver code.
The problem is that vs_prog_data->inputs_read may contain only parts of
varyings mark as used (see try_mask_partial_io() in
src/compiler/nir/nir_gather_info.c). On the other hand, for
vs_prog_data->double_inputs_read whole variables are always marked as used. The
element count is computed as _builtin_popcount(elements) -
_builtin_popcount(elements_double) / 2, but popcount(elements_double) / 2 >
popcount(elements) when only portions of double variables are mark as used.
This can be reproduced using the
VkPositiveLayerTest.CreatePipeline64BitAttributesPositive test from
Vulkan-LoaderAndValidationLayers:
./vk_layer_validation_tests
--gtest_filter='VkPositiveLayerTest.CreatePipeline64BitAttributesPositive'
Note: Google Test filter =
VkPositiveLayerTest.CreatePipeline64BitAttributesPositive
[==========] Running 1 test from 1 test case.
[----------] Global test environment set-up.
[----------] 1 test from VkPositiveLayerTest
[ RUN ] VkPositiveLayerTest.CreatePipeline64BitAttributesPositive
vk_layer_validation_tests: ./genxml/gen9_pack.h:66: __gen_uint: Assertion `v <=
max' failed.
Aborted
--
You are receiving this mail because:
You are the assignee for the bug.
You are the QA Contact for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-3d-bugs/attachments/20171012/ac5e6b9b/attachment.html>
More information about the intel-3d-bugs
mailing list