[adding potentially interested people] On Fri, Nov 7, 2014 at 3:58 PM, Andy Lutomirski <luto@xxxxxxxxxxxxxx> wrote: > The syscall exit asm is a big mess. There's a really fast path, some > kind of fast path code (with a hard-coded optimization for audit), and > the really slow path. The result is that it's very hard to work with > this code. There are some asm paths that are much slower than they > should be (context tracking is a major offender), but no one really > wants to add even more asm to speed them up. > > This series takes a different, unorthodox approach. Rather than trying > to avoid entering the very slow iret path, it adds a way back out of the > iret path. The result is a dramatic speedup for context tracking, user > return notification, and similar code, as the cost of a few lines of > tricky asm. Nonetheless, it's barely a net addition of asm code, > because we get to remove the fast path optimizations for audit and > rescheduling. > > Thoughts? If this works, it opens the door for a lot of further > consolidation of the exit code. > > Note: patch 1 in this series has been floating around on the list > for quite a while. It's mandatory for this series to work, because > the buglet that it fixes almost completely defeats the optimization > that I'm introducing. It turns out that sysret_audit may be rather buggy. I think it leaves edx and edi in a confused state, and it interacts badly with SCHEDULE_USER if context tracking is on. My preferred long-term solution is to delete sysret_audit entirely, which this patch set does. Can you (x86 people and people who, for reasons that escape me, enjoy reviewing this stuff) take a look? This clearly isn't 3.18 material, and it may want to soak in -next (can -tip do that? I can do it myself, I suppose), but it might also be a good idea to try to do this for 3.19 to get rid of sysret_audit. For those who haven't followed all the recent threads: the asm that's deleted in patch 3 currently has a nasty RCU + context tracking + audit bug that has become much easier to trigger as a result of the seccomp changes in 3.18. This isn't directly a bug in the seccomp changes -- it's just that the seccomp changes make it much easier to cause the offending ask to be executed. --Andy > > Andy Lutomirski (3): > x86_64,entry: Fix RCX for traced syscalls > x86_64,entry: Use sysret to return to userspace when possible > x86_64,entry: Remove the syscall exit audit and schedule optimizations > > arch/x86/kernel/entry_64.S | 103 ++++++++++++++++++++++++--------------------- > 1 file changed, 55 insertions(+), 48 deletions(-) > > -- > 1.9.3 > -- Andy Lutomirski AMA Capital Management, LLC -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
