Re: sparsemem panic in 2.6.17-rc5-mm1 and -mm2

[Yasunori Goto <y-goto@jp.fujitsu.com>]

> I looked back into 2.6.15, 2.6.16. 
> It looks -mm's time of initialization of "total_memory" is not changed from them.
> (yes, Andrew's fix looks sane.)
> 
> I'm intersted in the following texts in the log.
> ==
> Node 0 DMA: 0*4kB 0*8kB 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 0kB
> Node 0 DMA32: empty
> Node 0 Normal: 0*4kB 0*8kB 0*16kB 0*32kB 0*64kB 0*128kB 0*256kB 0*512kB 0*1024kB 0*2048kB 0*4096kB = 0kB
> Node 0 HighMem: 1*4kB 1*8kB 1*16kB 1*32kB 1*64kB 1*128kB 0*256kB 0*512kB 1*1024kB 2*2048kB 3962*4096kB = 16233724kB
> Node 1 DMA: empty
> Node 1 DMA32: empty
> Node 1 Normal: empty
> Node 1 HighMem: 1*4kB 1*8kB 0*16kB 0*32kB 0*64kB 1*128kB 0*256kB 1*512kB 1*1024kB 0*2048kB 4065*4096kB = 16651916kB
> Node 2 DMA: empty
> Node 2 DMA32: empty
> Node 2 Normal: empty
> Node 2 HighMem: 1*4kB 1*8kB 0*16kB 0*32kB 0*64kB 1*128kB 0*256kB 1*512kB 1*1024kB 0*2048kB 4065*4096kB = 16651916kB
> Node 3 DMA: empty
> Node 3 DMA32: empty
> Node 3 Normal: empty
> Node 3 HighMem: 1*4kB 1*8kB 0*16kB 0*32kB 0*64kB 1*128kB 0*256kB 1*512kB 1*1024kB 0*2048kB 3811*4096kB = 15611532kB
> ==
> Looks 64GB memory. but there are only HIGHMEM, no NORMAL, DMA. so, shrink_zone() worked.

Its log shows there are some memory in DMA and NORMAL just immediately
before that.....

> Active:2 inactive:15 dirty:0 writeback:0 unstable:0 free:16287272 slab:1823 mapped:0 pagetables:0
> Node 0 DMA free:0kB min:0kB low:0kB high:0kB active:0kB inactive:0kB present:16384kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0 0 0
> Node 0 DMA32 free:0kB min:0kB low:0kB high:0kB active:0kB inactive:0kB present:0kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0 0 0
> Node 0 Normal free:0kB min:0kB low:0kB high:0kB active:0kB inactive:0kB present:385024kB pages_scanned:0 all_unreclaimable? no
lowmem_reserve[]: 0 0 0 0

It looks like that something wasted all of DMA(16MB) and NORMAL(385MB)
zone suddenly. Hmmm...

Bye.

-- 
Yasunori Goto 


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/