Revision 9aea5a65aa7a1af9a4236dfaeb0088f1624f9919 authored by Roland McGrath on 08 September 2010, 02:37:06 UTC, committed by Linus Torvalds on 10 September 2010, 15:10:26 UTC
An execve with a very large total of argument/environment strings can take a really long time in the execve system call. It runs uninterruptibly to count and copy all the strings. This change makes it abort the exec quickly if sent a SIGKILL. Note that this is the conservative change, to interrupt only for SIGKILL, by using fatal_signal_pending(). It would be perfectly correct semantics to let any signal interrupt the string-copying in execve, i.e. use signal_pending() instead of fatal_signal_pending(). We'll save that change for later, since it could have user-visible consequences, such as having a timer set too quickly make it so that an execve can never complete, though it always happened to work before. Signed-off-by: Roland McGrath <roland@redhat.com> Reviewed-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
1 parent 7993bc1
| File | Mode | Size |
|---|---|---|
| Kconfig | -rw-r--r-- | 5.5 KB |
| Makefile | -rw-r--r-- | 790 bytes |
| README.dino | -rw-r--r-- | 1.0 KB |
| asp.c | -rw-r--r-- | 3.5 KB |
| ccio-dma.c | -rw-r--r-- | 47.7 KB |
| ccio-rm-dma.c | -rw-r--r-- | 5.1 KB |
| dino.c | -rw-r--r-- | 30.8 KB |
| eisa.c | -rw-r--r-- | 11.6 KB |
| eisa_eeprom.c | -rw-r--r-- | 2.9 KB |
| eisa_enumerator.c | -rw-r--r-- | 11.5 KB |
| gsc.c | -rw-r--r-- | 5.9 KB |
| gsc.h | -rw-r--r-- | 1.5 KB |
| hppb.c | -rw-r--r-- | 2.8 KB |
| iommu-helpers.h | -rw-r--r-- | 4.5 KB |
| iosapic.c | -rw-r--r-- | 27.6 KB |
| iosapic_private.h | -rw-r--r-- | 5.7 KB |
| lasi.c | -rw-r--r-- | 6.2 KB |
| lba_pci.c | -rw-r--r-- | 46.6 KB |
| led.c | -rw-r--r-- | 19.7 KB |
| pdc_stable.c | -rw-r--r-- | 30.2 KB |
| power.c | -rw-r--r-- | 7.3 KB |
| sba_iommu.c | -rw-r--r-- | 57.8 KB |
| superio.c | -rw-r--r-- | 14.1 KB |
| wax.c | -rw-r--r-- | 3.2 KB |
Computing file changes ...
