Re: [mpich-discuss] Core dump in parse_args
I tried -configfile option and it further crashed with a memory failure. The configfile itself has 1199 words/23777 characters. Overall it has 4 lines for 4 partitions. *** Error in `vmpiexec': double free or corruption (out): 0x00000000007001e0 *** ======= Backtrace: ========= /lib64/libc.so.6(+0x81329)[0x7ffff7206329] vmpiexec[0x4262a9] vmpiexec[0x4220ed] vmpiexec[0x409d23] vmpiexec[0x4035da] /lib64/libc.so.6(__libc_start_main+0xf5)[0x7ffff71a7555] vmpiexec[0x404c05] ======= Memory map: ======== 00400000-0047c000 r-xp 00000000 00:77 2085572302 ./mpich2/linux_x86_64/bin/mpiexec.hydra 0067b000-0067e000 rw-p 0007b000 00:77 2085572302 ./mpich2/linux_x86_64/bin/mpiexec.hydra 0067e000-00bd9000 rw-p 00000000 00:00 0 [heap] 7ffff0000000-7ffff0021000 rw-p 00000000 00:00 0 7ffff0021000-7ffff4000000 ---p 00000000 00:00 0 7ffff6b45000-7ffff6b6a000 r-xp 00000000 08:02 45522 /usr/lib64/liblzma.so.5.2.2 7ffff6b6a000-7ffff6d69000 ---p 00025000 08:02 45522 /usr/lib64/liblzma.so.5.2.2 7ffff6d69000-7ffff6d6a000 r--p 00024000 08:02 45522 /usr/lib64/liblzma.so.5.2.2 7ffff6d6a000-7ffff6d6b000 rw-p 00025000 08:02 45522 /usr/lib64/liblzma.so.5.2.2 7ffff6d6b000-7ffff6d80000 r-xp 00000000 08:02 447697 /usr/lib64/libz.so.1.2.7 7ffff6d80000-7ffff6f7f000 ---p 00015000 08:02 447697 /usr/lib64/libz.so.1.2.7 7ffff6f7f000-7ffff6f80000 r--p 00014000 08:02 447697 /usr/lib64/libz.so.1.2.7 7ffff6f80000-7ffff6f81000 rw-p 00015000 08:02 447697 /usr/lib64/libz.so.1.2.7 7ffff6f81000-7ffff6f83000 r-xp 00000000 08:02 45003 /usr/lib64/libdl-2.17.so 7ffff6f83000-7ffff7183000 ---p 00002000 08:02 45003 /usr/lib64/libdl-2.17.so 7ffff7183000-7ffff7184000 r--p 00002000 08:02 45003 /usr/lib64/libdl-2.17.so Regards Tarun From: Kumar, Tarun (DI SW ICS DVT RD QSCE) Sent: Monday, September 18, 2023 3:16 PM To: Zhou, Hui <[email protected]>; [email protected]; Thakur, Rajeev <[email protected]> Cc: Raffenetti, Ken <[email protected]> Subject: RE: [mpich-discuss] Core dump in parse_args Thanks. Good to know. I will explore further. Regards Tarun From: Zhou, Hui <[email protected]<mailto:[email protected]>> Sent: Monday, September 18, 2023 2:34 PM To: [email protected]<mailto:[email protected]>; Thakur, Rajeev <[email protected]<mailto:[email protected]>> Cc: Raffenetti, Ken <[email protected]<mailto:[email protected]>>; Kumar, Tarun (DI SW ICS DVT RD QSCE) <[email protected]<mailto:[email protected]>> Subject: Re: [mpich-discuss] Core dump in parse_args Hydra supports -configfile {name} option, so one could try putting all his commandline arguments in a file and pass that to hydra. Hui ________________________________ From: Raffenetti, Ken via discuss <[email protected]<mailto:[email protected]>> Sent: Monday, September 18, 2023 4:22 PM To: [email protected]<mailto:[email protected]> <[email protected]<mailto:[email protected]>>; Thakur, Rajeev <[email protected]<mailto:[email protected]>> Cc: Raffenetti, Ken <[email protected]<mailto:[email protected]>>; [email protected]<mailto:[email protected]> <[email protected]<mailto:[email protected]>> Subject: Re: [mpich-discuss] Core dump in parse_args I’m not aware of a hard-coded limit, but I am able to reproduce a segfault reliably by passing 1000 arguments a toy executable. Will update with more info when we have it. Ken From: "kumar.tarun--- via discuss" <[email protected]<mailto:[email protected]>> Reply-To: "[email protected]<mailto:[email protected]>" <[email protected]<mailto:[email protected]>> Date: Friday, September 15, 2023 at 4:57 PM To: "Thakur, Rajeev" <[email protected]<mailto:[email protected]>>, "[email protected]<mailto:[email protected]>" <[email protected]<mailto:[email protected]>> Cc: "[email protected]<mailto:[email protected]>" <[email protected]<mailto:[email protected]>> Subject: Re: [mpich-discuss] Core dump in parse_args Thanks Rajeev for your reply. I don’t have the exact number as of now. It’s definitely more than 1000. Is there a hard coded limit? Regards Tarun From: Thakur, Rajeev <[email protected]<mailto:[email protected]>> Sent: Friday, September 15, 2023 2:40 PM To: [email protected]<mailto:[email protected]> Cc: Kumar, Tarun (DI SW ICS DVT RD QSCE) <[email protected]<mailto:[email protected]>> Subject: Re: [mpich-discuss] Core dump in parse_args How many arguments? Rajeev From: "kumar.tarun--- via discuss" <[email protected]<mailto:[email protected]>> Reply-To: "[email protected]<mailto:[email protected]>" <[email protected]<mailto:[email protected]>> Date: Friday, September 15, 2023 at 4:22 PM To: "[email protected]<mailto:[email protected]>" <[email protected]<mailto:[email protected]>> Cc: "[email protected]<mailto:[email protected]>" <[email protected]<mailto:[email protected]>> Subject: [mpich-discuss] Core dump in parse_args Hi, I recently encountered this crash where mpiexec when executed from a bash script and with a large number of arguments crashes. The core dump is as follows: #0 0x0000000000408ec3 in parse_args () #1 0x0000000000409f26 in HYD_uii_mpx_get_parameters () #2 0x000000000040397a in main () If I reduce the number of arguments then I don’t see the crash. Is it a known issue? Regards Tarun
participants (1)
-
kumar.tarun@siemens.com