FS#43458 - gcc internal compiler error (probably out of memory, but doesn't say)
Attached to Project:
Arch Linux
Opened by Wendell Smith (wendell) - Thursday, 15 January 2015, 15:59 GMT
Last edited by Allan McRae (Allan) - Thursday, 15 January 2015, 23:23 GMT
Opened by Wendell Smith (wendell) - Thursday, 15 January 2015, 15:59 GMT
Last edited by Allan McRae (Allan) - Thursday, 15 January 2015, 23:23 GMT
|
Details
Description:
g++ crashes with the following error: "g++: internal compiler error: Killed (program cc1plus) Please submit a full bug report, with preprocessed source if appropriate. See <https://bugs.archlinux.org/> for instructions." I'm pretty sure that's because it ran out of memory, but it would be nice if the error message actually said that. The command failed twice; when I quit some other memory hogs (firefox, sublime text editor, etc.), it succeeded. Anyways, it said "please submit a full bug report", so here I am. I don't exactly know how to get the full preprocessed source, but if that would actually be helpful, I can figure it out (especially if you give me some advice) - let me know. Additional info: * gcc 4.9.2-2 All packages up to date as of Jan. 15, 2015 * config and/or log files etc. This is a 64-bit Arch build inside a VirtualBox on a MacBook Pro, with 4096 Mb of RAM allocated for Arch (8192 Mb on the actual machine). Steps to reproduce: I was compiling a library I wrote called parm ( https://github.com/wackywendell/parm ), commit 3f749d. It requires swig, and the command was 'make py' from the root directory. It only crashed with several other applications open. |
This task depends upon
Closed by Allan McRae (Allan)
Thursday, 15 January 2015, 23:23 GMT
Reason for closing: Upstream
Additional comments about closing: Talk to upstream about getting this fixed.
Thursday, 15 January 2015, 23:23 GMT
Reason for closing: Upstream
Additional comments about closing: Talk to upstream about getting this fixed.
Comment by
Dave Reisner (falconindy) -
Thursday, 15 January 2015, 17:28 GMT
If you ran out of memory, there's probably other logs saying as
much (e.g. if the cc1plus program was killed by the kernel's OOM
killer). You can't necessarily know why a program received
SIGKILL, since the victim has no chance to catch the signal.