Message ID | 20240129104327.10639-1-mdoucha@suse.cz |
---|---|
State | Accepted |
Headers | show |
Series | splice07: Whitelist /proc/self/maps on pipe write | expand |
Hi Martin, > Just like with /dev/zero, splicing /proc/self/maps to writable pipe > is expected to pass. After generic splice() read emulation was removed, > it was reimplemented in commit b0072734ffaa ("tty, proc, kernfs, random: > Use copy_splice_read()") Ah, very good catch, thank you. Kernel 6.5. Reviewed-by: Petr Vorel <pvorel@suse.cz> Kind regards, Petr
Hi!
Hopefully this is a last one...
Reviewed-by: Cyril Hrubis <chrubis@suse.cz>
Hi Martin, Cyril, > Hi! > Hopefully this is a last one... Great, merged. Thank you both! Kind regards, Petr > Reviewed-by: Cyril Hrubis <chrubis@suse.cz>
diff --git a/testcases/kernel/syscalls/splice/splice07.c b/testcases/kernel/syscalls/splice/splice07.c index 1fa13b3e6..9c23e0aac 100644 --- a/testcases/kernel/syscalls/splice/splice07.c +++ b/testcases/kernel/syscalls/splice/splice07.c @@ -40,6 +40,7 @@ static void check_splice(struct tst_fd *fd_in, struct tst_fd *fd_out) /* While these combinations succeeed */ case TST_FD_DEV_ZERO: case TST_FD_FILE: + case TST_FD_PROC_MAPS: case TST_FD_MEMFD: return; /* And this complains about socket not being connected */
Just like with /dev/zero, splicing /proc/self/maps to writable pipe is expected to pass. After generic splice() read emulation was removed, it was reimplemented in commit b0072734ffaa ("tty, proc, kernfs, random: Use copy_splice_read()") Signed-off-by: Martin Doucha <mdoucha@suse.cz> --- testcases/kernel/syscalls/splice/splice07.c | 1 + 1 file changed, 1 insertion(+)