Hi,
Could someone here give me some optimization advice/reference about
vnode locking? Recently we find that vnode lookup locking has
significant negative impact on I/O latency. We trace down to two
places in the source code which initiated the locking:
1. the vn_lock( ) in lookup( ), kern/vfs_lo
In section 2.2.3, /etc/sysctl.conf:
To our experience, it also helps to tune inode cache behavior by setting:
vfs.vmiodirenable="0"; (maybe)
vfs.nameileafonly="-1"
On a server with large memory, if apache 1.x is tested, maybe it is also necessary
to increase:
vm.max_proc_mmap
In section 2.2.4,
Not sure if somebody here has already noticed this problem before. I
have a small program using select() . Though it works correctly on 4.x,
some problem happens on 5.1. First of all, I notice that macros about
FD_xx have been moved into sys/select.h. Then FD_ZERO() seems to be
failed. My program l
But I have to admit that we didn't scale the demand too much, just up to
2000 connections. (but the FD set for the queue should be larger since
there're open files.)
- Yaoping Ruan
Computer Science Dept.
Princeton University
___
[EMAIL PROTECTE
t;
> On Sat, Oct 18, 2003 at 04:03:53PM -0400, Yaoping Ruan wrote:
>
> > I installed the 4.8 release on a new box with 4GB memory, and found
> > kernel panic when I tried to write date. But the system run great with
> > only 2GB memory. Is there any kernel compiling option
in open ()
#14 0xc038a06d in syscall2 ()
#15 0xc037ae85 in Xint0x80_syscall ()
#16 0x8048b3f in ?? ()
#17 0x80496f1 in ?? ()
#18 0x8048971 in ?? ()
Thanks in advance for any clue.
- Yaoping Ruan
___
[EMAIL PROTECTED] mailing list
http://lists.free
welcome.
Thanks a lot
- Yaoping Ruan
*
Computer Science Dept.
Princeton University
*
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-hackers
To unsubscribe, send any mail to
Sometime ago we posted a work on improving Web server performance on
FreeBSD. and got quite a few valuable feedback from the list members.
>From the discussion, we realize it would be great if part of the work,
the sendfile() optimization, could be merged into the mainline code. You
may want to tr
Hi,
Thank you so much for your constructive notes. We do feel confident to
merge some parts into the mainline code. Other parts may require further
discussion:
Bosko Milekic wrote:
> 1) User-visible DeBoxInfo structure has the magic number "5"
>PerSleepInfo structs and the magic numbe
Hi,
Thanks for your interests. Just clarify that it is not a patch to boost kernel
performance, but for the performance debugging tool we built, DeBox.
We should be able to release the patch in a few days.
[EMAIL PROTECTED] wrote:
> I'd be very interested in a copy of the patches, I have several
We have been working on improving Web server performance on
FreeBSD, and think you may be interested in the results and
techniques we used. Specifically, we focus on the SpecWeb99
benchmark and the Flash Web Server, and have roughly quadrupled
its performance. We did this by adding support for a ve
Greeting,
Could anybody give me some advise about the kernel profiling implementation in
FreeBSD. Specifically, I am
confused by:
1. Is the asm code in /sys/i386/isa/prof_machdep.c the entry code plugged into each
function? and it calls
_MCOUNT_DECL(frompc, selfpc) in /sys/libkern/mcount.c, which
rouble for just a single stack image, and kgdb is
not simple enough. Which source file(s) I should look at?
- Yaoping Ruan
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
13 matches
Mail list logo
