hum, do i get a speed ticket?
i did some tests before applying your patches:
lizard ./tbench 1 dev
.1 clients started
..+*
Throughput 6.10567 MB/sec (NB=7.63209 MB/sec 61.0567 MBit/sec)
lizard ./tbench 2 dev
..2 clients started
++**
Throughput 7.71796
I've fixed a couple of additional problems.
* tbench() assumes that accept() propogates the NODELAY tcp option.
It doesn't in FreeBSD. Er, it didn't in FreeBSD... my patch fixes
this.
* If the transwmitter sees a 0 window it stalls waiting for an ack.
However, if
That's reasonable with patch #2 and delayed acks turned off (assuming
this is a 100BaseTX network you are testing on). You should get
slightly better results with Patch #3 and delayed acks turned on.
-Matt
:hum, do i get a speed
That's reasonable with patch #2 and delayed acks turned off (assuming
this is a 100BaseTX network you are testing on). You should get
slightly better results with Patch #3 and delayed acks turned on.
-Matt
i did not apply any
It seems Richard Sharpe wrote:
Attached is the patch I am using, which is based on what Greg gave me.
It tries UDMA5 first, and steps down ...
The following patch is bogus, it doesn't set the chip to the prober
mode (always sets it to UDMA2), it just set the disk, this wont
work guys...
I
: slightly better results with Patch #3 and delayed acks turned on.
:
: -Matt
:i did not apply any patches - yet -, net.inet.tcp.newreno is 1, on both hosts.
:they are connected at 100mb, so the strange numbers are when the number of
:clients is
On Sun, 2 Dec 2001, Greg Lehey wrote:
Note that there are other chips out there which return the same PCI
information but which appear to be capable of ATA 100. I recently
gave a patch to Richard Sharpe (copied) which he says was able to get
his SiS 5591 to run at ATA 100. I'm still
It seems Zwane Mwaikambo wrote:
On Sun, 2 Dec 2001, Greg Lehey wrote:
Note that there are other chips out there which return the same PCI
information but which appear to be capable of ATA 100. I recently
gave a patch to Richard Sharpe (copied) which he says was able to get
his SiS
Hello,
Could someone review and commit the following patch:
Index: pthread.3
===
RCS file: /home/ncvs/src/share/man/man3/pthread.3,v
retrieving revision 1.20
diff -u -r1.20 pthread.3
--- pthread.3 1 Oct 2001 16:09:20 -
OK Matt, that last patch did the trick.
I am now getting 68 and 69Mb/s between my Linux system and the FreeBSD
system.
I have also tried the loopback interface, and I am getting 371Mb/s for 1 process,
dropping to about 320Mb/s for 5.
This seems like it is close to the limit for the machine I
Hi,
It seems like all of the issues uncovered have been fixed, so it seems like you cannot
use performance as a way to choose between FreeBSD and Linux any longer.
I will re-issue my report, but I do not have any more time to spend on this now for
several days.
I will most likely re-run
Its a 4.4-Stable box, actually it does the same thing on three different
4.4-Stable boxes. make world works, so does installworld, make release
barfs. It exists in my standard /usr/src, but not in the checked out
chrooted build tree.
On Sun, 2001-12-02 at 00:50, Makoto Matsushita wrote:
You
hola estoy buscando algo para gla gx28 gracias por informacion o donde
puedo dirigirme att mauro
To Unsubscribe: send mail to [EMAIL PROTECTED]
with unsubscribe freebsd-hackers in the body of the message
:
:OK Matt, that last patch did the trick.
:
:I am now getting 68 and 69Mb/s between my Linux system and the FreeBSD
:system.
Excellent!
:I have also tried the loopback interface, and I am getting 371Mb/s for 1 process,
:dropping to about 320Mb/s for 5.
Excellent!
:This seems like
On Sun, Dec 02, 2001 at 11:18:42AM -0800, Matthew Dillon wrote:
[...]
:Does the FreeBSD tcp stack do zero copy (page flip the data to
:userspace)? In the localhost case, it seems like there are two copies
:to/from userspace there.
:
:--
:Richard Sharpe, [EMAIL PROTECTED], LPIC-1
I'm writing a PCI character device driver and need some clairification.
As I see the FreeBSD driver structure there are really two components to a
device driver, there is the KLD component which contains the device methods
for probe, attach, detach, etc and handles the dynamic componet of the
On Sun, 2 Dec 2001, Matthew Dillon wrote:
Throughput 47.2446 MB/sec (NB=59.0558 MB/sec 472.446 MBit/sec) 20 procs
It seems to max-out at around 75,000 packets per second (input + output).
I doubt these results could be duplicated on anything but a DELL2550.
It dedicates an
Okay ive setup a /usr/cvs, exported CVSROOT /usr/cvs, cd'd into
/usr/cvs, cvs init, cd'd to /usr/src, cvs import src devel beta, all
went fine and created a /usr/cvs/src . cd'd to /usr/src
setenv CVSROOT /usr/cvs
setenv CHROOTDIR /usr/release
make release
churn churn churn
it builds a
On Sun, 2 Dec 2001, Lamont Granquist wrote:
:
:What is the remaining bottleneck in these tests? CPU? Interrupts? What
:would you need to do to get that closer to the theoretical limit
:(something around 920 Mbs for GigE IIRC)?
Well, for one thing, I'd imagine that per-byte and per-copy
I'm seeing data corruption on the Promise channel of a A7V133,
WITH the southbridge fix applied and NO sound card of any kind in
the system (the built-in is disabled in the bios; current wouldn't boot
at all with it enabled, when I bought the mb). All the cards are
video (a TNT2 card of some
:On Sun, 2 Dec 2001, Matthew Dillon wrote:
: Throughput 47.2446 MB/sec (NB=59.0558 MB/sec 472.446 MBit/sec) 20 procs
:
: It seems to max-out at around 75,000 packets per second (input + output).
:
: I doubt these results could be duplicated on anything but a DELL2550.
: It
Hello,
AFAIK it is possible to attach/detach an ata channel with atacontrol on
-CURRENT. I am looking for a way to do this on -STABLE, too. Particularly
I want to hot swap an ATA disk w/ an appropriate IDE drive bay. No RAID
is involved at the moment.
Currently the whole machine is to be shut
On Sun, 2 Dec 2001, Pete Carah wrote:
I'm seeing data corruption on the Promise channel of a A7V133,
WITH the southbridge fix applied and NO sound card of any kind in
the system (the built-in is disabled in the bios; current wouldn't boot
at all with it enabled, when I bought the mb).
On Sun, 2 Dec 2001, Matthew Dillon wrote:
This is connecting to inetd running a dd if=/dev/zero bs=32k on a
machine with the rfc sysctl's turned on and 262144 byte send and
receive buffers, without jumbo frames (my gigE switch doesn't support
them :-( ).
nice, 950 Mbs which
In message 000601c17b6b$7a89c190$037d6041@gandalf Dragon Fire writes:
: I'm writing a PCI character device driver and need some clairification.
:
: As I see the FreeBSD driver structure there are really two components to a
: device driver, there is the KLD component which contains the device
:nice, 950 Mbs which should be the theoretical maximum. what kind of CPUs
:do you have in there, and do you know how hard they were working?
:
These are 1.1 GHz duel Pentium III's. One of the cpu's is maxed out
at that transfer rate (this is -stable and the program is in the system
I am using stable not devfs for development.
In the Design and Implementation of the 4.4 BSD Operating System. Chapter 6
states
Device Drivers
A device is divided into three main sections
1. Autoconfiguration and initialization routines
2. Routines for servicing I/O requests (the top half)
3.
In message 003e01c17b9c$0e571ff0$037d6041@gandalf Dragon Fire writes:
: I am using stable not devfs for development.
:
: In the Design and Implementation of the 4.4 BSD Operating System. Chapter 6
: states
:
: Device Drivers
:
: A device is divided into three main sections
: 1.
Repeat again.
kerberus it builds a tree, cvs's the src goes into its build stage, and again,
kerberus fails with a fresh cvsup from today, which did a complete make world
kerberus kernel from the standard /usr/src, now can some one tell me what im
kerberus missing or doing wrong ?? and yes i
On Sun, Dec 02, 2001 at 12:10:53AM -0800, Matthew Dillon wrote:
...
There are still a couple of unresolved issues. I noticed that when
connecting locally TCP is non-optimal... when sending a 4100 byte
data block it sends two 1460 byte packets (maxseg), then one
1176 byte
Hello,
On the subject of tcp performance graphs, I have
taken a few tcpdumps of some pages, to understand the entire process in more
minute detail, and to co-relate the theory with the graphs in practice. I have
plotted a graph of cumulative data received at the client side vs.time
:curious, as the loopback's MTU is normally 16384.
:Also, any idea on where does the 4096 limit (1460*2+1176) come from ?
:
: cheers
: luigi
It comes from the size of an mbuf, which is 2K. If you are trying to
send 4100 bytes of data what winds up happening is this:
Does the FreeBSD tcp stack do zero copy (page flip the data to
userspace)? In the localhost case, it seems like there are two copies
to/from userspace there.
It has the ability to do it via sendfile() and a few other mechanisms, but
not as a normal part of typical read()/write().
Ahh, but
:
:Ahh, but there are patches floating around that do support zero-copy.
:Just ask Ken Merry and Drew Gallatin. I don't think they've been integrated
:due to lack of testing time, but they've existed for 2 or so years now.
:
http://people.freebsd.org/~ken/zero_copy/
--
Andrew R. Reiter
[EMAIL
34 matches
Mail list logo