Hi Chun-Yeow, On Tue, Nov 22, 2011 at 10:00 PM, Yeoh Chun-Yeow <[email protected]> wrote: > Hi, Thomas and all > > I have setup the secured mesh (after patching to allow hw crypto in mesh) > as follow (not multihop case testing, 1 hop scenario): > > Mesh Node (Mesh Gateway) <----------> Mesh Node A > <----------> Mesh Node B > > Both of the Mesh Node A and B are directly linked to Mesh Gateway. When Mesh > Node A generates the TCP stream towards Mesh Gateway and later on Mesh Node > B generates the TCP stream towards Mesh Gateway as well, the Mesh Node B > seems fail to push the throughput towards Mesh Gateway. The following > errors happen on Mesh Node B: > > # iperf -c 10.44.29.245 -i 1 -t 100 -p 5002 > ------------------------------------------------------------ > Client connecting to 10.44.29.245, TCP port 5002 > TCP window size: 16.0 KByte (default) > ------------------------------------------------------------ > [ 3] local 10.44.29.154 port 50127 connected with 10.44.29.245 port 5002 > <7>ath: tx queue 2 (3069858), link a30697f4 > <7>ath: TX complete: skb: 83a043b0 > [ ID] Interval Transfer Bandwidth > [ 3] 0.0- 1.0 sec 3.25 MBytes 27.3 Mbits/sec > <7>ath: transmitting packet, skb: 83a04cb0 > [ 3] 1.0- 2.0 sec 3.25 MBytes 27.3 Mbits/sec > <7>ath: TX complete: skb: 822d6230 > [ 3] 2.0- 3.0 sec 3.50 MBytes 29.4 Mbits/sec > <7>ath: tx queue 2 (30660e0), link a306c1c0 > <7>ath: link[2] (a30637dc)=30626ac (a30626ac) > [ 3] 3.0- 4.0 sec 3.38 MBytes 28.3 Mbits/sec > kworker/u:2: page allocation failure. order:0, mode:0x20 > Call Trace: > [<802599d4>] dump_stack+0x8/0x34 > [<800b0074>] __alloc_pages_nodemask+0x628/0x698 > [<800d21b8>] new_slab+0x8c/0x21c > [<800d2d04>] __slab_alloc.clone.69.clone.74+0x120/0x1ac > [<800d2dd4>] kmem_cache_alloc+0x44/0xa8 > [<830aa6f8>] ieee80211_mesh_init_sdata+0x774/0x8e4 [mac80211] > > Mem-Info: > Normal per-cpu: > CPU 0: hi: 18, btch: 3 usd: 2 > active_anon:404 inactive_anon:14 isolated_anon:0 > active_file:187 inactive_file:362 isolated_file:0 > unevictable:0 dirty:0 writeback:0 unstable:0 > free:96 slab_reclaimable:152 slab_unreclaimable:13253 > mapped:251 shmem:25 pagetables:58 bounce:0 > Normal free:384kB min:1016kB low:1268kB high:1524kB active_anon:1616kB > inactive_anon:56kB active_file:748kB inactive_file:1448kB unevictable:0kB > isolated(anon):0kB isolated(file):0kB present:65024kB mlocked:0kB dirty:0kB > writeback:0kB mapped:1004kB shmem:100kB slab_reclaimable:608kB > slab_unreclaimable:53012kB kernel_stack:344kB pagetables:232kB unstable:0kB > bounce:0kB writeback_tmp:0kB pages_scanned:0 all_unreclaimable? no > lowmem_reserve[]: 0 0 > Normal: 0*4kB 0*8kB 0*16kB 0*32kB 0*64kB 1*128kB 1*256kB 0*512kB 0*1024kB > 0*2048kB 0*4096kB = 384kB > 574 total pagecache pages > 0 pages in swap cache > Swap cache stats: add 0, delete 0, find 0/0 > Free swap = 0kB > Total swap = 0kB > 16384 pages RAM > 919 pages reserved > 1155 pages shared > 14896 pages non-shared > SLUB: Unable to allocate memory on node -1 (gfp=0x20) > cache: kmalloc-128, object size: 128, buffer size: 128, default order: 0, > min order: 0 > node 0: slabs: 0, objs: 0, free: 0 > kworker/u:2: page allocation failure. order:0, mode:0x20 > Call Trace: > [<802599d4>] dump_stack+0x8/0x34 > [<800b0074>] __alloc_pages_nodemask+0x628/0x698 > [<800d21b8>] new_slab+0x8c/0x21c > [<800d2d04>] __slab_alloc.clone.69.clone.74+0x120/0x1ac > [<800d2dd4>] kmem_cache_alloc+0x44/0xa8 > [<830aa6f8>] ieee80211_mesh_init_sdata+0x774/0x8e4 [mac80211 > > Do you all have any ideas?
I don't have time to reproduce this scenario right now, but that call trace certainly looks odd. Failed to allocate the mesh interface (again??)? Is there any other relevant dmesg output you can provide? Does a simple ping from B work when A is iperfing? Try 'cat /sys/kernel/debug/ieee80211/phy0/queues' (or whatever phy your meshif is on of course). These should be stopped this is a contention issue. Thomas _______________________________________________ Devel mailing list [email protected] http://open80211s.com/mailman/listinfo/devel
