[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <4A43E9F1.90209@cosmosbay.com>
Date: Thu, 25 Jun 2009 23:19:45 +0200
From: Eric Dumazet <dada1@...mosbay.com>
To: Paweł Staszewski <pstaszewski@...are.pl>
CC: Linux Network Development list <netdev@...r.kernel.org>
Subject: Re: rib_trie / Fix inflate_threshold_root. Now=15 size=11 bits
Paweł Staszewski a écrit :
> Hello ALL
>
> Some time ago i report this:
> http://bugzilla.kernel.org/show_bug.cgi?id=6648
>
> and now with 2.6.29 / 2.6.29.1 / 2.6.29.3 and 2.6.30 it back
> dmesg output:
> oprofile: using NMI interrupt.
> Fix inflate_threshold_root. Now=15 size=11 bits
> Fix inflate_threshold_root. Now=15 size=11 bits
> Fix inflate_threshold_root. Now=15 size=11 bits
> Fix inflate_threshold_root. Now=15 size=11 bits
> Fix inflate_threshold_root. Now=15 size=11 bits
> Fix inflate_threshold_root. Now=15 size=11 bits
> Fix inflate_threshold_root. Now=15 size=11 bits
> Fix inflate_threshold_root. Now=15 size=11 bits
> Fix inflate_threshold_root. Now=15 size=11 bits
> Fix inflate_threshold_root. Now=15 size=11 bits
> Fix inflate_threshold_root. Now=15 size=11 bits
> Fix inflate_threshold_root. Now=15 size=11 bits
> Fix inflate_threshold_root. Now=15 size=11 bits
> Fix inflate_threshold_root. Now=15 size=11 bits
> Fix inflate_threshold_root. Now=15 size=11 bits
Curious, you seem to hit an old alloc_pages limit()... (MAX_ORDER allocation)
Your root node has 2^18 = 262144 pointers of 8 bytes -> 2097152 bytes (+ header -> 4194304 bytes)
But since following commit, we should use vmalloc() so this PAGE_SIZE<<10) limit
should not anymore be applied.
Could you do a "cat /proc/vmallocinfo" just to check your big tnodes are vmalloced() ?
commit 15be75cdb5db442d0e33d37b20832b88f3ccd383
Author: Stephen Hemminger <shemminger@...tta.com>
Date: Thu Apr 10 02:56:38 2008 -0700
IPV4: fib_trie use vmalloc for large tnodes
Use vmalloc rather than alloc_pages to avoid wasting memory.
The problem is that tnode structure has a power of 2 sized array,
plus a header. So the current code wastes almost half the memory
allocated because it always needs the next bigger size to hold
that small header.
This is similar to an earlier patch by Eric, but instead of a list
and lock, I used a workqueue to handle the fact that vfree can't
be done in interrupt context.
Signed-off-by: Stephen Hemminger <shemminger@...tta.com>
Signed-off-by: David S. Miller <davem@...emloft.net>
>
> cat /proc/net/fib_triestat
> Basic info: size of leaf: 40 bytes, size of tnode: 56 bytes.
> Main:
> Aver depth: 2.28
> Max depth: 6
> Leaves: 276539
> Prefixes: 289922
> Internal nodes: 66762
> 1: 35046 2: 13824 3: 9508 4: 4897 5: 2331 6: 1149 7: 5
> 9: 1 18: 1
> Pointers: 691228
> Null ptrs: 347928
> Total size: 35709 kB
>
> Counters:
> ---------
> gets = 26276593
> backtracks = 547306
> semantic match passed = 26188746
> semantic match miss = 1117
> null node hit= 27285055
> skipped node resize = 0
>
> Local:
> Aver depth: 3.33
> Max depth: 4
> Leaves: 9
> Prefixes: 10
> Internal nodes: 8
> 1: 8
> Pointers: 16
> Null ptrs: 0
> Total size: 2 kB
>
> Counters:
> ---------
> gets = 26642350
> backtracks = 1282818
> semantic match passed = 18166
> semantic match miss = 0
> null node hit= 0
> skipped node resize = 0
>
>
>
> This machine is running bgpd with two bgp peers / full route table
>
> cat /proc/meminfo
> MemTotal: 12279032 kB
> MemFree: 11521920 kB
> Buffers: 80288 kB
> Cached: 34416 kB
> SwapCached: 0 kB
> Active: 286816 kB
> Inactive: 82024 kB
> Active(anon): 254296 kB
> Inactive(anon): 0 kB
> Active(file): 32520 kB
> Inactive(file): 82024 kB
> Unevictable: 0 kB
> Mlocked: 0 kB
> SwapTotal: 987988 kB
> SwapFree: 987988 kB
> Dirty: 1140 kB
> Writeback: 0 kB
> AnonPages: 254164 kB
> Mapped: 5440 kB
> Slab: 365084 kB
> SReclaimable: 28784 kB
> SUnreclaim: 336300 kB
> PageTables: 2104 kB
> NFS_Unstable: 0 kB
> Bounce: 0 kB
> WritebackTmp: 0 kB
> CommitLimit: 7127504 kB
> Committed_AS: 267704 kB
> VmallocTotal: 34359738367 kB
> VmallocUsed: 11824 kB
> VmallocChunk: 34359707815 kB
> HugePages_Total: 0
> HugePages_Free: 0
> HugePages_Rsvd: 0
> HugePages_Surp: 0
> Hugepagesize: 2048 kB
> DirectMap4k: 3392 kB
> DirectMap2M: 12578816 kB
>
>
> Interfaces mtu is1500
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists