This happened with CONFIG_HIGHMEM enabled on a almost-vanilla 2.6.11... with that option disabled I don't see such problems (also tested on both the same exact kernel, and with nearby ones).
To sum up, there is a read() syscall, a copy_to_user() (which calls kmap()) with preemption disabled (in_atomic() == 1) and the might_sleep produces a warning... now, what's strange is that UML is always built with preemption disabled (will this mean in_atomic == 1 always?). And also I don't understand what changes with HIGHMEM... it seems that kmap always calls might_sleep(). Debug: sleeping function called from invalid context at arch/um/sys-i386/highmem.c:5 in_atomic():1, irqs_disabled():0 Call Trace: 0bdd77a0: [<08079cec>] __might_sleep+0xac/0xd0 0bdd77c0: [<0807676d>] kmap+0x1d/0x40 0bdd77d0: [<08061417>] do_op+0x37/0x70 0bdd77f0: [<080614ca>] do_buffer_op+0x7a/0x1a0 0bdd7800: [<08061780>] copy_chunk_to_user+0x0/0x40 0bdd7810: [<08061780>] copy_chunk_to_user+0x0/0x40 0bdd7830: [<0805f7e1>] setjmp_wrapper+0x51/0x60 0bdd783c: [<08237b45>] sigemptyset+0x25/0x40 0bdd7868: [<0805f7ac>] setjmp_wrapper+0x1c/0x60 0bdd78ac: [<08237b45>] sigemptyset+0x25/0x40 0bdd78c0: [<0805c17b>] change_signals+0x5b/0x90 0bdd7900: [<08061631>] buffer_op+0x41/0x80 0bdd7904: [<08061450>] do_buffer_op+0x0/0x1a0 0bdd7914: [<08061780>] copy_chunk_to_user+0x0/0x40 0bdd7930: [<0806183e>] copy_to_user_skas+0x7e/0xc0 0bdd7940: [<08061780>] copy_chunk_to_user+0x0/0x40 0bdd7960: [<080b053f>] file_read_actor+0xcf/0x150 0bdd7990: [<080b02a7>] do_generic_mapping_read+0x607/0x7d0 0bdd79b0: [<0805c219>] enable_mask+0x49/0x60 0bdd7a40: [<080b0770>] __generic_file_aio_read+0x1b0/0x220 0bdd7a58: [<080b0470>] file_read_actor+0x0/0x150 0bdd7a88: [<080ba2ae>] poison_obj+0x2e/0x60 0bdd7aa0: [<080b0953>] generic_file_read+0xd3/0x100 0bdd7b20: [<080a2fe0>] autoremove_wake_function+0x0/0x60 0bdd7b50: [<080e0534>] vfs_read+0xe4/0x180 0bdd7b80: [<080e0901>] sys_read+0x51/0x80 0bdd7bb0: [<08060c87>] execute_syscall_skas+0xd7/0x100 0bdd7bd0: [<08280e04>] schedule+0x3a4/0x6f0 0bdd7c20: [<0805cb79>] record_syscall_start+0x59/0x70 0bdd7c40: [<08060ce9>] handle_syscall+0x39/0x70 0bdd7c60: [<080600d1>] userspace+0x231/0x3b0 0bdd7c90: [<080779d9>] finish_task_switch+0x79/0x110 0bdd7cd0: [<080611e8>] force_flush_all_skas+0x38/0x40 0bdd7cf0: [<080608dc>] fork_handler+0xbc/0xd0 0bdd7d20: [<08237738>] __restore+0x0/0x8 0bdd7d60: [<08237ab1>] kill+0x11/0x20 -- Paolo Giarrusso, aka Blaisorblade Linux registered user n. 292729 http://www.user-mode-linux.org/~blaisorblade ------------------------------------------------------- SF email is sponsored by - The IT Product Guide Read honest & candid reviews on hundreds of IT Products from real users. Discover which products truly live up to the hype. Start reading now. http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click _______________________________________________ User-mode-linux-devel mailing list User-mode-linux-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
