Debian has an unreleased port underway to amd64 (x86-64). One of our users reports that Xprt crashes when compiled and run in 64-bit mode, see http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=278510 Could the bug be related to the compiler warnings?, such as: cfb8segC.c: I funktion `cfb32SegmentSS1RectCopy': cfb8segC.c:476: varning: cast from pointer to integer of different size gdb says: (gdb) r Starting program: /usr/bin/Xprt Program received signal SIGSEGV, Segmentation fault. 0x0000002a95869f11 in free () from /lib/libc.so.6 (gdb) bt #0 0x0000002a95869f11 in free () from /lib/libc.so.6 #1 0x0000000000406645 in AlterSaveSetForClient () #2 0x00000000004a11ea in MediumDiscreteSizeListParse () #3 0x000000000042c53f in WaitForSomething () #4 0x000000000041ff64 in Dispatch () #5 0x000000000040e8ac in main () A section of the strace looks like: open("/usr/X11R6/lib/X11/fonts/100dpi/fonts.alias", O_RDONLY) = 3 fstat(3, {st_mode=S_IFREG|0644, st_size=3302, ...}) = 0 fstat(3, {st_mode=S_IFREG|0644, st_size=3302, ...}) = 0 mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x2a955a4000 read(3, "!! fonts.alias -- automatically "..., 4096) = 3302 read(3, "", 4096) = 0 close(3) = 0 munmap(0x2a955a4000, 4096) = 0 brk(0) = 0xbd1000 brk(0xbfa000) = 0xbfa000 brk(0) = 0xbfa000 brk(0) = 0xbfa000 brk(0xbea000) = 0xbea000 brk(0) = 0xbea000 open("/usr/X11R6/lib/X11/fonts/misc/6x13-ISO8859-1.pcf.gz", O_RDONLY) = 3 read(3, "\37\213\10\0f\233ZA\0\3\355\234\177x\24\307y\307\277\2"..., 8192) = 4366 read(3, "", 8192) = 0 close(3) = 0 open("/usr/X11R6/lib/X11/fonts/misc/cursor.pcf.gz", O_RDONLY) = 3 read(3, "\37\213\10\0o\233ZA\0\3\355\233\177pT\327u\307\317JB\22"..., 8192) = 5094 read(3, "", 8192) = 0 close(3) = 0 mmap(NULL, 233472, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x2a955a4000 munmap(0x2a955a4000, 233472) = 0 mmap(NULL, 233472, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) = 0x2a955a4000 munmap(0x2a955a4000, 233472) = 0 --- SIGSEGV (Segmentation fault) @ 0 (0) --- +++ killed by SIGSEGV +++
Created attachment 1371 [details] [review] Xalloc declarations wrong This fast fix allows us to get slightly longer, but Xprt now crashes at another stage - any ideas? I'm running in a gcc 3.4 environment, btw. Program received signal SIGSEGV, Segmentation fault. 0x0000000000407b79 in LookupClient (rid=5161623, client=0x7fbffff6f0) at dixutils.c:316 316 return clients[clientIndex]; (gdb) bt #0 0x0000000000407b79 in LookupClient (rid=5161623, client=0x7fbffff6f0) at dixutils.c:316 #1 0x0000000000407e9a in BlockHandler (pTimeout=0x7fbffff578, pReadmask=0xb50dc0) at dixutils.c:431 #2 0x0000000000444d87 in WaitForSomething (pClientsReady=0x7fbffff5b0) at WaitFor.c:304 #3 0x0000000000430f0a in Dispatch () at dispatch.c:253 #4 0x000000000041379f in main (argc=10, argv=0x7fbffffb18) at main.c:364
Please ignore my "Xalloc" patch - it looks like the segfault occurs at different places, depending on the compiler used. (I'm using gcc-3.4.3.)
The new experimental debian-version, 0.1.0.alpha1-1, works under amd64 with mozilla 1.7.3 , so the amd64-problem seems to be fixed in the 010+ cvs-tree!
(In reply to comment #3) > The new experimental debian-version, 0.1.0.alpha1-1, works under > amd64 with mozilla 1.7.3 , so the amd64-problem seems to be fixed in > the 010+ cvs-tree! I got the same confirmation from Suse/AMD64 so I am marking this issue as FIXED.
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.