Messages in this thread | | | Date | Tue, 12 Oct 1999 21:46:33 +0200 | From | Manfred Spraul <> | Subject | Re: vma_list_sem |
| |
Alexander Viro wrote: > b) correct code should not be punished. Ever. ASSERT is wrong. >
I'll remove the code from load_elf_binary() once I'm sure that the down(mmap_sem) are really superflous. IIRC you agreed that load_elf_library() needs the locking. Possible problems for load_elf_binary(): * swap_out() could find that process. [seems safe] * swap_out() could execute "kill_proc(pid, SIGBUS, 1)" [not yet checked] * what about sys_ptrace()? [looks dangerous]
I really don't like the idea that a structure which can be found via linked lists has no proper locking. Obviously, locking would be superflous if noone has access to the "struct mm" pointer, but the pointer can be eg. found by find_task_by_pid()->mm.
> Some of missing pieces (modulo binfmt-related stuff): I'll add them, thanks,
Manfred
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |