Summary: | Konsole crash upon right click | ||
---|---|---|---|
Product: | [Applications] konsole | Reporter: | p.giarrusso |
Component: | general | Assignee: | Konsole Developer <konsole-devel> |
Status: | RESOLVED DUPLICATE | ||
Severity: | crash | CC: | shlomif |
Priority: | NOR | ||
Version: | unspecified | ||
Target Milestone: | --- | ||
Platform: | Unlisted Binaries | ||
OS: | Linux | ||
Latest Commit: | Version Fixed In: | ||
Sentry Crash Report: |
Description
p.giarrusso
2010-10-17 13:19:58 UTC
I don't know if it's a priority for the konsole devs, but many applications will crash if your system runs out of memory. So I suggest that this bug will be closed. This bug is not about out-of-memory, it happened in that context, but it's unlikely to be related. The stack traces does not suggest that a newly allocated pointer was NULL and was not checked, and that's the only obvious way in which an out-of-memory can crash an app. I don't see why detach() should allocate memory - of course, if you looked up in the sources, you can point it out (but then you should write that in the first place). Moreover, on Linux malloc() doesn't return NULL anyway if the system is low on memory, in most cases. By default, the memory allocation primitives like brk() (and mmap("/dev/zero") and friends) (on which always succeeds but memory is not physically allocated; writing to a blank page will look for memory, and if this fails, even after trying very hard, the Linux kernel kills some process using lots of memory - and it still doesn't use SIGSEGV. This policy can be changed by writing into /proc/sys/kernel/overcommit_memory, but I did not do this; see Documentation/vm/overcommit-accounting in the source tree of the Linux kernel for more info. I know this because I used to be a kernel hacker. Before an application runs out of memory, it will be OOM killed by the kernel. No application "crashes" when there is no memory. I doubt this is the issue. *** This bug has been marked as a duplicate of bug 233377 *** |