Bug 50315 - [r600g+llvm] Piglit failures with "EE r600_asm.c:119 r600_bytecode_get_num_operands - Need instruction operand number for 0x42."
Summary: [r600g+llvm] Piglit failures with "EE r600_asm.c:119 r600_bytecode_get_num_op...
Status: RESOLVED FIXED
Alias: None
Product: Mesa
Classification: Unclassified
Component: Mesa core (show other bugs)
Version: git
Hardware: x86-64 (AMD64) All
: medium normal
Assignee: mesa-dev
QA Contact:
URL:
Whiteboard:
Keywords: regression
Depends on:
Blocks:
 
Reported: 2012-05-24 09:29 UTC by Kai
Modified: 2012-06-03 11:18 UTC (History)
1 user (show)

See Also:
i915 platform:
i915 features:


Attachments
[PATCH] r600g: add RECIP_INT, PRED_SETE_INT to r600_bytecode_get_num_operands (1.43 KB, patch)
2012-05-24 09:48 UTC, Vadim Girlin
Details | Splinter Review

Description Kai 2012-05-24 09:29:26 UTC
The same as bug 50312, except the tests are not in state fail, only warn and the error:
> EE r600_asm.c:119 r600_bytecode_get_num_operands - Need instruction operand number for 0x42.

The tests that have state warn and show the error given above are:
glean/glsl1-discard statement in for loop
glean/glsl1-do-loop
glean/glsl1-do-loop with break
glean/glsl1-do-loop with continue and break
glean/glsl1-for-loop with continue
shaders/glsl-copy-propagation-loop-1
shaders/glsl-fs-discard-04
shaders/glsl-fs-loop
shaders/glsl-fs-loop-300
shaders/glsl-fs-loop-continue
shaders/glsl-fs-loop-nested
shaders/glsl-fs-loop-redundant-condition
shaders/glsl-fs-unroll-out-param
shaders/glsl-fs-unroll-side-effect
shaders/glsl-vs-copy-propagation-1
shaders/glsl-vs-loop
shaders/glsl-vs-loop-300
shaders/glsl-vs-loop-continue
shaders/glsl-vs-loop-nested
shaders/glsl-vs-loop-redundant-condition
spec/glsl-1.30/execution/fs-discard-exit-1
spec/glsl-1.30/execution/fs-discard-exit-2

Please see bug 50312 for the hardware and stack specification.
Comment 1 Vadim Girlin 2012-05-24 09:48:06 UTC
Created attachment 62060 [details] [review]
[PATCH] r600g: add RECIP_INT, PRED_SETE_INT to r600_bytecode_get_num_operands

This patch should handle both instructions 0x42 and 0x77
Comment 2 Kai 2012-05-24 10:46:54 UTC
Yes, applying attachment 62060 [details] [review] fixes /this/ bug completely, please see bug 50312, comment #1 for the changes there.
Comment 3 Tom Stellard 2012-06-03 11:18:37 UTC
Fixed in git commit 482041a53862f45afd88536b5a615501cbb6ce25


Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.