Messages in this thread | | | Date | Sun, 19 May 2013 11:37:27 -0700 | From | Greg Kroah-Hartman <> | Subject | Re: [ 027/102] x86, vm86: fix VM86 syscalls: use SYSCALL_DEFINEx(...) |
| |
On Fri, May 17, 2013 at 04:51:33PM -0700, Greg Kroah-Hartman wrote: > On Fri, May 17, 2013 at 11:49:30PM +0100, Al Viro wrote: > > On Fri, May 17, 2013 at 02:35:42PM -0700, Greg Kroah-Hartman wrote: > > > 3.9-stable review patch. If anyone has any objections, please let me know. > > > > This is seriously wrong. For 3.9 you _need_ asmlinkage_protect() in that > > thing; by the time when that went into the tree, mainline already had > > it generated automatically by SYSCALL_DEFINE, so there was no point in > > that part of patch - the switch to SYSCALL_DEFINE alone did the job. > > For 3.9 it's very much needed; as the matter of fact, in 3.9 that commit > > is a no-op in the form you took. > > > > We can grab all prereqs into 3.9-stable (there's not that much of those), > > but that would be much more intrusive than the variant adding explicit > > asmlinkage_protect() in those two syscalls. > > Ok, Alexander was saying something was off here. > > Can someone send me just the needed patch to get this working properly, > and I will be glad to drop this one from the 3.9.x tree.
I've now dropped this, and will release without it.
greg k-h
| |