http://bugs.winehq.com/show_bug.cgi?id=1423
------- Additional Comments From z_god@wanadoo.nl 2003-28-06 11:09 ------- Bug comments restored from Gmane.org:
I'm having trouble with this patch and Red Hat 9: <http://www. winehq.com/hypermail/wine-cvs/2003/04/0224.html>. Multithreaded apps started crashing deep down in the guts of gethostbyname_r whenever it's called from multiple threads at once. The crash typically looks like this:
Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 1118186784 (LWP 27326)] 0x4206c909 in fgets_unlocked () from /lib/tls/libc.so.6 (gdb) bt #0 0x4206c909 in fgets_unlocked () from /lib/tls/libc.so.6 #1 0x4023e1e4 in _nss_files_gethostbyname_r () from /lib/libnss_files.so.2 #2 0x420f2b3c in gethostbyname_r@@GLIBC_2.1.2 () from /lib/tls/libc.so.6 #3 0x40cf77c7 in __ws_gethostbyname (name=0x402900cf "media. monster.com", dup_flag=1) at socket.c:3044 #4 0x40cf78f4 in WS_gethostbyname (name=0x402900cf "media. monster.com") at socket.c:3085 (gdb) info thr * 8 Thread 1118186784 (LWP 27326) 0x4206c909 in fgets_unlocked () from /lib/tls/libc.so.6 7 Thread 1115761952 (LWP 27325) 0xffffe002 in ?? () 6 Thread 1113337120 (LWP 27324) 0xffffe002 in ?? () 5 Thread 1110912288 (LWP 27323) 0x004168ff in ?? () 2 Thread 1100885280 (LWP 27314) 0xffffe002 in ?? () 1 Thread 1074023232 (LWP 27305) 0xffffe002 in ?? () (gdb) thr 7 [Switching to thread 7 (Thread 1115761952 (LWP 27325))]#0 0xffffe002 in ?? () (gdb) bt #0 0xffffe002 in ?? () #1 0x41c5f6d4 in send_dg () from /lib/libresolv.so.2 #2 0x41c5ecb5 in __libc_res_nsend () from /lib/libresolv.so.2 #3 0x41c5dc07 in __libc_res_nquery () from /lib/libresolv.so.2 #4 0x41c5e35b in __libc_res_nquerydomain () from /lib/libresolv. so.2 #5 0x41c5df22 in __libc_res_nsearch () from /lib/libresolv.so.2 #6 0x40d0805f in _nss_dns_gethostbyname2_r () from /lib/libnss_dns.so.2 #7 0x40d08293 in _nss_dns_gethostbyname_r () from /lib/libnss_dns.so.2 #8 0x420f2b3c in gethostbyname_r@@GLIBC_2.1.2 () from /lib/tls/libc.so.6 #9 0x40cf77c7 in __ws_gethostbyname (name=0x40295b4f "media. monster.com", dup_flag=1) at socket.c:3044 #10 0x40cf78f4 in WS_gethostbyname (name=0x40295b4f "media. monster.com") at socket.c:3085
Reversing patch #7969 seems to fix the problem. To reproduce the crash, run the local web proxy Proxomitron <http://www. proxomitron.org/> and try loading sites with lots of images. I'm running Red Hat 9 with glibc 2.3.2-27.9 and the --with-nptl configure option (wine fully rebuilt after installing RH9).
------- Additional Comments From prupe <at> myrealbox.com 2003-05-03 14:36 ------- After rebuilding from the latest CVS, I can't get this crash to happen any more. Marking as fixed.