Hi I am testing main lined RT features in 2.6.39. When I am testing cyclictest with loading the device with hackbench. I am using ramdisk of size 125536. At around 3857615 cycles, cyclictest crashes due to data abort. When I increase ramdisk size cyclic test crash cycle count increases. Pasting the Back Trace [<c032b0ac>] (dump_stack+0x0/0x1c) from [<c009a870>] (dump_header+0x84/0x19c) [<c009a7ec>] (dump_header+0x0/0x19c) from [<c009ad0c>] (oom_kill_process+0x4c/0x288) [<c009acc0>] (oom_kill_process+0x0/0x288) from [<c009b218>] (out_of_memory+0x2d0/0x39c) [<c009af48>] (out_of_memory+0x0/0x39c) from [<c009e688>] (__alloc_pages_nodemask+0x4ac/0x620) [<c009e1dc>] (__alloc_pages_nodemask+0x0/0x620) from [<c00a04bc>] (__do_page_cache_readahead+0xe0/0x244) [<c00a03dc>] (__do_page_cache_readahead+0x0/0x244) from [<c00a064c>] (ra_submit+0x2c/0x34) [<c00a0620>] (ra_submit+0x0/0x34) from [<c00a08f8>] (ondemand_readahead+0x1c8/0x1d8) [<c00a0730>] (ondemand_readahead+0x0/0x1d8) from [<c00a0a20>] (page_cache_sync_readahead+0x70/0x78) [<c00a09b0>] (page_cache_sync_readahead+0x0/0x78) from [<c0098420>] (filemap_fault+0x14c/0x44c) [<c00982d4>] (filemap_fault+0x0/0x44c) from [<c00aee90>] (__do_fault+0x58/0x400) [<c00aee38>] (__do_fault+0x0/0x400) from [<c00aff28>] (handle_pte_fault+0x290/0x6b4) [<c00afc98>] (handle_pte_fault+0x0/0x6b4) from [<c00b08f4>] (handle_mm_fault+0xd8/0xec) [<c00b081c>] (handle_mm_fault+0x0/0xec) from [<c032fc58>] (do_page_fault+0x164/0x25c) [<c032faf4>] (do_page_fault+0x0/0x25c) from [<c00342e4>] (do_DataAbort+0x3c/0x9c) [<c00342a8>] (do_DataAbort+0x0/0x9c) from [<c032de64>] (ret_from_exception+0x0/0x10) --- Regards, Mugunthan V N. -- To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
