Messages in this thread | | | From | Mark Hagger <> | Subject | Re: problem with linux 2.2.15pre4 + Rik's VM fix | Date | Thu, 27 Jan 2000 16:43:58 +0000 |
| |
Hi,
I'm running linux-2.2.15pre4 with Rik's VM fix and get messages
gfp called by non-running (1) task from c011e896!
Output of grep c011e /System.map gives
nc14$ grep -i c011e /System.map c011e254 t put_page c011e27c T put_dirty_page c011e400 t do_wp_page c011e600 t partial_clear c011e6b8 T vmtruncate c011e79c t do_swap_page c011e868 t do_anonymous_page c011e8ec t do_no_page c011e9e0 T handle_mm_fault c011ebac T make_pages_present c011ec0c T vm_enough_memory c011ec8c T sys_brk c011edac T do_mmap
This gfp stuff occures pretty much as soon as the machine boots and starts to handle any network traffic.
I had wondered if this particular VM problem might explain why under high network traffic, CPU load, (and swapping) machines would go into infinite loops of network hangs, where sysrq's show repeated buffer allocation failures. I'm running some destruction tests now to see if I still get this hangup under the new code.
Mark
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |