2GB process crashing on 2.4.14
Quoted 2 lines · Paul Sargent
So if I was hitting this limit then I should see no / very few gaps, in the /proc/<pid>/maps. Is that true?
It usually fails when malloc() hits your libraries. One solution is to recompile the kernel with a higher TASK_UNMAPPED_BASE (should be a sysctl, but is a fixed define currently). That would force the shared libraries to start at a higher address and give sbrk() more breathing space.
Another way is to use mallopt(M_MMAP_THRESHOLD, ..) and set a low mmap threshold. This allows malloc to use mmap earlier instead of sbrk() and skip the shared library area. It comes at a cost thought, malloc() tends to become more expensive.
With some more changes you can also force the user space to 3.5GB, at the cost of much less kernel memory. It usually makes sense to change TASK_UNMAPPED_BASE with this.
-Andi
P.S.: I'm pretty sure this is a FAQ.
Quoted 7 lines · Andi Kleen
Quoted 2 lines · Paul Sargent
So if I was hitting this limit then I should see no / very few gaps, in the /proc/<pid>/maps. Is that true?
It usually fails when
malloc()hits your libraries.
[remedies to increase available address space snipped]
Quoted 3 lines
-Andi
P.S.: I'm pretty sure this is a FAQ.
Yes, I think your right, this is. At least, I've come across it before, but I'm not convinced this is my problem in this case.
I think it's failing about 1GB before getting to the bottom of the libraries, but I'm going to monitor maps over the weekend to ensure it doesn't start doing anything wacky just before crashing.
Paul
Signature · 4 lines
Paul Sargent Tel: +44 (1784) 476669 Fax: +44 (1784) 470699 mailto: [email protected]