On Wed, Apr 07, 2010 at 10:54:58AM +0800, Taras Glek wrote:Fair enough.
On 04/06/2010 07:24 PM, Wu Fengguang wrote:Boot time user space readahead can do better than kernel heuristic
Hi Taras,That's a slightly different usecase. I'd rather have all large apps
On Tue, Apr 06, 2010 at 05:51:35PM +0800, Johannes Weiner wrote:
On Mon, Apr 05, 2010 at 03:43:02PM -0700, Taras Glek wrote:How about improve Fedora (and other distros) to preload Mozilla (and
Hello,
I am working on improving Mozilla startup times. It turns out that page
faults(caused by lack of cooperation between user/kernelspace) are the
main cause of slow startup. I need some insights from someone who
understands linux vm behavior.
other apps the user run at the previous boot) with fadvise() at boot
time? This sounds like the most reasonable option.
startup as efficiently as possible without any hacks. Though until we
get there, we'll be using all of the hacks we can.
readahead in several ways:
- it can collect better knowledge on which files/pages will be used
which lead to high readahead hit ratio and less cache consumption
- it can submit readahead requests for many files in parallel,
which enables queuing (elevator, NCQ etc.) optimizations
So I won't call it dirty hack :)
That's excellent.Sure. It will do read-around from current fault offset - 64kb to +64kb.As for the kernel readahead, I have a patchset to increase defaultYes. Is the current readahead really doing read-around(ie does it read
mmap read-around size from 128kb to 512kb (except for small memory
systems). This should help your case as well.
pages before the one being faulted)? From what I've seen, having the
The first part of the file lists sections in a file and their hex offset+size.dynamic linker read binary sections backwards causes faults.There are too many data in
http://sourceware.org/bugzilla/show_bug.cgi?id=11447
http://people.mozilla.com/~tglek/startup/systemtap_graphs/ld_bug/report.txt
Can you show me the relevant lines? (wondering if I can ever find such lines..)
Released it yesterday. Hopefully other bloated binaries will benefit from this too.550 Can't openglibc-2.3.90-ld.so-madvise.diff in
This is interesting. I wonder how SuSE implements the policy.Current Situation:
The dynamic linker mmap()s executable and data sections of our
executable but it doesn't call madvise().
By default page faults trigger 131072byte reads. To make matters worse,
the compile-time linker + gcc lay out code in a manner that does not
correspond to how the resulting executable will be executed(ie the
layout is basically random). This means that during startup 15-40mb
binaries are read in basically random fashion. Even if one orders the
binary optimally, throughput is still suboptimal due to the puny readahead.
IO Hints:
Fortunately when one specifies madvise(WILLNEED) pagefaults trigger 2mb
reads and a binary that tends to take 110 page faults(ie program stops
execution and waits for disk) can be reduced down to 6. This has the
potential to double application startup of large apps without any clear
downsides.
Suse ships their glibc with a dynamic linker patch to fadvise()
dynamic libraries(not sure why they switched from doing madvise
before).
Do you have the patch or some strace output that demonstrates the
fadvise() call?
http://www.rpmseek.com/rpm/glibc-2.4-31.12.3.src.html?hl=com&cba=0:G:0:3732595:0:15:0:
/pub/linux/distributions/suse/pub/suse/update/10.1/rpm/src/glibc-2.4-31.12.3.src.rpm:
No such file or directory
OK I give up.
As I recall they just fadvise the filedescriptor before accessing it.Obviously this is a bit risky for small memory systems..
Right.Program page faults may exhibit random behavior once they've started.Program page faults are inherently random, so the straightforwardI filed a glibc bug about this atIt will throttle based on memory pressure. In idle situations it will
http://sourceware.org/bugzilla/show_bug.cgi?id=11431 . Uli commented
with his concern about wasting memory resources. What is the impact of
madvise(WILLNEED) or the fadvise equivalent on systems under memory
pressure? Does the kernel simply start ignoring these hints?
eat your file cache, however, to satisfy the request.
Now, the file cache should be much bigger than the amount of unneeded
pages you prefault with the hint over the whole library, so I guess the
benefit of prefaulting the right pages outweighs the downside of evicting
some cache for unused library pages.
Still, it's a workaround for deficits in the demand-paging/readahead
heuristics and thus a bit ugly, I feel. Maybe Wu can help.
solution would be to increase the mmap read-around size (for desktops
with reasonable large memory), rather than to improve program layout
or readahead heuristics :)
During startup page-in pattern of over-engineered OO applications isAgreed.
very predictable. Programs are laid out based on compilation units,
which have no relation to how they are executed. Another problem is that
any large old application will have lots of code that is either rarely
executed or completely dead. Random sprinkling of live code among mostly
unneeded code is a problem.
I'm able to reduce startup pagefaults by 2.5x and mem usage by a few MBThat's great. When will we enjoy your research fruits? :)
with proper binary layout. Even if one lays out a program wrongly, the
worst-case pagein pattern will be pretty similar to what it is by default.