Bug 44436

Summary: [i965] SIGBUS when trying glCopyTexture2D from fbo into a depth texture with max supported texture size
Product: Mesa Reporter: Shuang He <shuang.he>
Component: Drivers/DRI/i965Assignee: Ian Romanick <idr>
Status: VERIFIED DUPLICATE QA Contact:
Severity: normal    
Priority: medium    
Version: git   
Hardware: Other   
OS: Linux (All)   
Whiteboard:
i915 platform: i915 features:
Attachments: New piglit case to reproduce this regression

Description Shuang He 2012-01-03 22:14:57 UTC
Created attachment 55098 [details]
New piglit case to reproduce this regression

System Environment:
--------------------------
 Libdrm:		(master)2.4.29-17-ga9dd34a7ee9d03d357e15f045ab85a12f6f6e4b8
 Mesa:		(master)b5fd0e04a70d37a1bef9144096890e7940f9a200
 Xserver:		(master)xorg-server-1.11.99.901
 Xf86_video_intel:		(master)2.17.0-262-g770a953ff03bb8328c3f29e274d225528840f30c
 Cairo:		(master)6816f70da54e730e88b98c8592aa6fc84e082928
 Libva:		(vaapi-ext)4aeaa296febf2f71200ff30380902e2c80cbf679
 Libva_intel_driver:		(vaapi-ext)f0358f252e13a619550d37e5a720f9894c7aa18c
 Kernel:	(drm-intel-next) 097354eb14fa94d31a09c64d640643f58e4a5a9a

Bug detailed description:
-------------------------
This issue happens on Sandybridge and Ironlake
SIGBUS is triggered when trying glCopyTexture2D from fbo into a depth texture with max supported texture size
Related Oglconform test case are (GL30)fbo(scenario.shadowMaps)
backtrace is as following by the time of SIGBUG:
(gdb) bt
#0  unpack_uint_z_X8_Z24 (src=0xa6d9c000, dst=0x841e1b0, n=100) at main/format_unpack.c:2021
#1  0xb7a95ec4 in _mesa_unpack_uint_z_row (format=MESA_FORMAT_X8_Z24, n=100, src=0xa6d9c000, dst=0x841e1b0) at main/format_unpack.c:2074
#2  0xb7abf7d6 in fast_read_depth_pixels (ctx=0x80c4e80, x=<value optimized out>, y=0, width=100, height=100, type=5125, pixels=0x841e1b0, packing=0xbfffeb28) at main/readpix.c:84
#3  read_depth_pixels (ctx=0x80c4e80, x=<value optimized out>, y=0, width=100, height=100, type=5125, pixels=0x841e1b0, packing=0xbfffeb28) at main/readpix.c:125
#4  0xb7abffdd in _mesa_readpixels (ctx=0x80c4e80, x=0, y=0, width=100, height=100, format=6402, type=5125, packing=0x80d38fc, pixels=<value optimized out>) at main/readpix.c:564
#5  0xb79d867e in intelReadPixels (ctx=0x80c4e80, x=0, y=0, width=100, height=100, format=6402, type=5125, pack=0x80d38fc, pixels=0x841e1b0) at intel_pixel_read.c:200
#6  0xb7b94c79 in copy_tex_sub_image (ctx=0x80c4e80, dims=2, target=3553, level=0, xoffset=0, yoffset=0, zoffset=0, x=0, y=0, width=100, height=100) at drivers/common/meta.c:3149
#7  0xb79d98e4 in intelCopyTexSubImage2D (ctx=0x80c4e80, target=3553, level=0, xoffset=0, yoffset=0, x=0, y=0, width=100, height=100) at intel_tex_copy.c:208
#8  0xb7ad5542 in copyteximage (ctx=0x80c4e80, dims=2, target=3553, level=0, internalFormat=6402, x=0, y=0, width=100, height=100, border=0) at main/teximage.c:2837
#9  0x0806bd8a in piglit_display () at /home/shuang/piglit/tests/fbo/fbo-copyteximage-simple.c:90
#10 0x0806c413 in display () at /home/shuang/piglit/tests/util/piglit-framework.c:56
#11 0xb7eed5b6 in ?? () from /usr/lib/libglut.so.3
#12 0xb7ef13f8 in fgEnumWindows () from /usr/lib/libglut.so.3
#13 0xb7eedb13 in glutMainLoopEvent () from /usr/lib/libglut.so.3
#14 0xb7eee395 in glutMainLoop () from /usr/lib/libglut.so.3
#15 0x0806cb09 in main (argc=1, argv=0xbffff134) at /home/shuang/piglit/tests/util/piglit-framework.c:294


Reproduce steps:
1. compile and run attached new piglit test case


EXPECTED OUTCOME:
===================
get pass

ACTUAL OUTCOME:
===================
get SIGBUS
Comment 1 Chris Wilson 2012-01-04 01:35:17 UTC

*** This bug has been marked as a duplicate of bug 37871 ***
Comment 2 Shuang He 2012-01-04 04:26:08 UTC
verified, and the attached test case seems to reproduce the issue very easy on my side on one sandybridge

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.