diff options
| author | Heiko Carstens <heiko.carstens@de.ibm.com> | 2006-11-02 22:06:58 -0800 |
|---|---|---|
| committer | Linus Torvalds <torvalds@g5.osdl.org> | 2006-11-03 12:27:55 -0800 |
| commit | 87c2b7c045a44f6c1c7af23e64f2b286e6f7130a (patch) | |
| tree | 42c59705e0b04b8cd4770556f59f65228b62feaa /fs | |
| parent | 7870db4c7fa1b03fec133c4f4e67fdaa04c5ac15 (diff) | |
| download | linux-87c2b7c045a44f6c1c7af23e64f2b286e6f7130a.tar.bz2 | |
[PATCH] sys_pselect7 vs compat_sys_pselect7 uaccess error handling
758333458aa719bfc26ec16eafd4ad3a9e96014d fixes the not checked copy_to_user
return value of compat_sys_pselect7. I ran into this too because of an old
source tree, but my fix would look quite a bit different to Andi's fix.
The reason is that the compat function IMHO should behave the very same as
the non-compat function if possible. Since sys_pselect7 does not return
-EFAULT in this specific case, change the compat code so it behaves like
sys_pselect7.
Cc: David Woodhouse <dwmw2@infradead.org>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Diffstat (limited to 'fs')
| -rw-r--r-- | fs/compat.c | 20 |
1 files changed, 17 insertions, 3 deletions
diff --git a/fs/compat.c b/fs/compat.c index 50624d4a70c6..8d0a0018a7d2 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -1835,9 +1835,12 @@ asmlinkage long compat_sys_pselect7(int n, compat_ulong_t __user *inp, } while (!ret && !timeout && tsp && (ts.tv_sec || ts.tv_nsec)); - if (ret == 0 && tsp && !(current->personality & STICKY_TIMEOUTS)) { + if (tsp) { struct compat_timespec rts; + if (current->personality & STICKY_TIMEOUTS) + goto sticky; + rts.tv_sec = timeout / HZ; rts.tv_nsec = (timeout % HZ) * (NSEC_PER_SEC/HZ); if (rts.tv_nsec >= NSEC_PER_SEC) { @@ -1846,8 +1849,19 @@ asmlinkage long compat_sys_pselect7(int n, compat_ulong_t __user *inp, } if (compat_timespec_compare(&rts, &ts) >= 0) rts = ts; - if (copy_to_user(tsp, &rts, sizeof(rts))) - ret = -EFAULT; + if (copy_to_user(tsp, &rts, sizeof(rts))) { +sticky: + /* + * If an application puts its timeval in read-only + * memory, we don't want the Linux-specific update to + * the timeval to cause a fault after the select has + * completed successfully. However, because we're not + * updating the timeval, we can't restart the system + * call. + */ + if (ret == -ERESTARTNOHAND) + ret = -EINTR; + } } if (ret == -ERESTARTNOHAND) { |