I've been noticing some nasty problems with 2.4.0-test11-pre7, constantly
getting out of memory problems running megahal ( a conversation generator),
I think it has a memory leak.. nothing else seems to cause this.
Nov 29 21:36:41 intolerance kernel: kernel BUG at page_alloc.c:84!
Nov 29 21:36:41 intolerance kernel: invalid operand: 0000
Nov 29 21:36:41 intolerance kernel: CPU: 0
Nov 29 21:36:41 intolerance kernel: EIP: 0010:[__free_pages_ok+73/824]
Nov 29 21:36:41 intolerance kernel: EFLAGS: 00010282
Nov 29 21:36:41 intolerance kernel: eax: 0000001f ebx: c100c89c ecx: c0257828 edx: 00000000
Nov 29 21:36:41 intolerance kernel: esi: 000003ce edi: c8c22098 ebp: 00000000 esp: cd80de68
Nov 29 21:36:41 intolerance kernel: ds: 0018 es: 0018 ss: 0018
Nov 29 21:36:41 intolerance kernel: Process megahal (pid: 31863, stackpage=cd80d000)
Nov 29 21:36:41 intolerance kernel: Stack: c0211c72 c0211e60 00000054 c100c89c 000003ce c8c22098 ce67a0c8 c1044010
Nov 29 21:36:41 intolerance kernel: c02588e0 00000207 ffffffff 0000317e c012a847 c012ac7b 00400000 000003ce
Nov 29 21:36:41 intolerance kernel: c011fa5a c100c89c cabbdfa0 08050000 cec55960 024a3000 c8c22098 09800000
Nov 29 21:36:41 intolerance kernel: Call Trace: [tvecs+8030/43772] [tvecs+8524/43772] [__free_pages+19/20] [free_page_and_swap_cache+131/136] [zap_page_range+374/504] [exit_mmap+201/288] [mmput+21/44]
Nov 29 21:36:41 intolerance kernel: [do_exit+165/508] [do_signal+544/636] [do_page_fault+0/988] [update_process_times+29/144] [update_wall_time+11/60] [timer_bh+36/604] [timer_interrupt+95/264] [bh_action+27/96]
Nov 29 21:36:41 intolerance kernel: [tasklet_hi_action+60/96] [do_softirq+63/100] [do_IRQ+161/176] [error_code+52/60] [signal_return+20/24]
Nov 29 21:36:41 intolerance kernel: Code: 0f 0b 83 c4 0c 89 f6 89 d8 2b 05 38 c8 2b c0 69 c0 f1 f0 f0
Does that backtrace provide anything useful?
--
Stephen