Fast sigblock (AKA rtld speedup)

[Konstantin Belousov]

On Fri, Jan 11, 2013 at 06:23:44PM +0800, David Xu wrote:
> On 2013/01/11 17:54, Konstantin Belousov wrote:
> > On Mon, Jan 07, 2013 at 08:22:35PM +0200, Konstantin Belousov wrote:
> >> Below is the forward of the patch for which I failed to obtain a private
> >> review. Might be, the list generates more responses.
> >>
> >> Our rtld has a performance bootleneck, typically exposed by the images
> >> with the lot of the run-time relocation processing, and by the C++
> >> exception handling. We block the signals delivery during the rtld
> >> performing the relocations, as well as for the dl_iterate_phdr(3) (the
> >> later is used for finding the dwarf unwinding tables).
> >>
> >> The signal blocking is needed to allow the rtld to resolve the symbols
> >> for the signal handlers in the safe way, but also causes 2 syscalls
> >> overhead per each rtld entry.
> >>
> >> The proposed approach allows to shave off those two syscalls, doubling
> >> the FreeBSD performance for the (silly) throw/catch C++ microbenchmark.
> >>
> >> Date: Mon, 13 Aug 2012 15:26:00 +0300
> >> From: Konstantin Belousov <kostikbel at gmail.com>
> >>
> >> ...
> >>
> >> The basic idea is to implement sigprocmask() as single write into usermode
> >> address. If kernel needs to calculate the signal mask for current thread,
> >> it takes into the consideration non-zero value of the word at the agreed
> >> address. Also, usermode is informed about signals which were put on hold
> >> due to fast sigblock active.
> >>
> >> As I said, on my measurements in microbenchmark that did throw/catch in
> >> a loop, I see equal user and system time spent for unpatched system, and
> >> same user time with zero system time on patched system.
> >>
> >> Patch can be improved further, e.g. it would be nice to allow rtld to fall
> >> back to sigprocmask(2) if kernel does not support fast sigblock, to prevent
> >> flag day. Also, the mask enforced by fast sigblock can be made configurable.
> >>
> >> Note that libthr already blocks signals by catching them, and not using rtld
> >> service in the first line handler. I tried to make the change in the spirit
> >> of libthr interceptors, but handoff to libthr appears too complicated to
> >> work. In fact, libthr can be changed to start using fast sigblock instead
> >> of wrapping sigaction, but this is out of scope of the proposal right now.
> >>
> >> Please comment.
> >
> > So there were no overly negative comments, and thanks to Alfred and David
> > for useful notes.
> >
> > The patch at
> > http://people.freebsd.org/~kib/misc/rtld-sigblock.2.patch
> > is the commit candidate. Now kernel informs rtld about supprt for
> > fast_sigblock with new auxv flag. Rtld can operate on old (and possibly
> > future) kernels without fast_sigblock, rtld checks the auxv for
> > presence of the ELF_BSDF_FASTSIGBLK flag before use.
> >
> 
> The patch looks fine to me. Sorry, I still have a question:
> if I found a process does not responsing to a signal, now should we
> check both its signal mask and fast block value ? but is there such
> a tool available ? for example, in gdb.

Fair enough.

I added the facility to export the address of the fast sigblock word
to the tools, and implemented the procstat extension to print the address
with -j. You can indeed use gdb to look up the content of the word
when debugging, after getting it address from procstat.

http://people.freebsd.org/~kib/misc/rtld-sigblock.3.patch
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 834 bytes
Desc: not available
URL: <http://lists.freebsd.org/pipermail/freebsd-toolchain/attachments/20130111/7e35f3ff/attachment.sig>