:Hey,
:
:It is great if you are finding a solution for the VIA-chipset.
:Do you have any idea if it is a simular problem that I am experiencing?
:
:I am not enough into chip code to have a clue what exactly the patch
:is doing - but maybe it is just decreasing the load on the kernel/system in
:a
ndolences!
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
but
may also occur in earlier VIA chipsets.
What I would like to do is try forcing the DMA transfer rate to 66 MHz,
i.e. UDMA66 or UDMA33, to see if that solves the problem. Soren,
could you supply a patch that universally turns off higher UDMA modes?
Note that we aren't complaining about your patch or anything, you
are simply the closest thing we have to a VIA chip expert right now :-)
-Matt
Matthew Dillon
&l
:Anyways could I have you guys read out all 256 PCI regs from the
:main chip that is the 0x03051106 one ?
:
:-Søren
How is this done?
-Matt
Matthew Dillon
<[EM
Ok, there's packet loss. From this extract you can see
that the client receives through sequence 641, then the
next packet it receives starts at sequence 993.
15:28:09.879928 transwarp.tao.org.uk.telnet > genius.tao.org.uk.kpop: P 609:641(32)
ack 64 win 33304 (DF) [tos 0x10]
15:28
ppier.
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
:I'll try cutting some of the intermediate infrastructure out and see if
:that helps.
:
:> I recommend replacing the hubs with switches. The whole topology will
:> be happier.
:
:Of course. I've not found a well priced switch though. (on a budget).
:
:Thanks for helping me with this. I
don't even get 'late collision'
errors.
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
:I've connected the laptop to the both of the server's fxps, in the first
:case using the combinations of the upstairs and downstairs hub, and in
:the later case via the hub on the router on the other side. I get
:exactly the same behaviour. This to me says that it's not the hubs.
:To test it f
ndow advertisement on the client to deal with lots of small
packets but it would destroy performance for larger packets / data
streams.
-Matt
Matthew Dillon
<[EMAIL PROTE
:Hi,
:
:I have tried this too, it makes absoutely no difference at all. My mallocs
:fail after a certain no. of runs of my code(and there is no memory leak),
:and there was no difference by increasing MAXDSIZ/DFLDSIZ.
:
:Thanks,
:Anjali
MAXDSIZ and DFLDSIZ are associated with USER malloc()s,
ckets
will have terrible performance.
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
s you intend to cap the window size
at, say, 128 bytes, this will solve the ssh issue but destroy performance
for everything else going over the USB ethernet.
-Matt
Matthew D
:
:the driver should return ENOBUFS
:and TCP should repect that..
:I think it already does in fact..
How does the client's driver returning ENOBUFS prevent the server from
sending too many packets?
-Matt
To Unsubscribe: send mail to
server to force the server to reduce its
congestion window. New-Reno is a mess and I am not volunteering to
hack it up more.
-Matt
Matthew Dillon
<[EMAIL PROTECTED]
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
DELAY on FreeBSD and, in fact, I believe it didn't used to.
We should just turn it off.
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EM
:>TCP_NODELAY on FreeBSD and, in fact, I believe it didn't used to.
:>We should just turn it off.
:
: Nagle + delayed acks don't really get along all that well, and often
:result in character echoes lagging .25-.5 seconds behind the keystrokes (due
:to round trip + 200ms). This really
been an m_freem(). Could someone check that?
The patch is against -stable.
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
Inde
Kerneltrap did a good interview of me a week or so ago (one of several
interviews they are doing) and just published it:
http://kerneltrap.com/article.php?sid=459
-Matt
Matthew Dillon
:
:See the post below. Is there a way to get at this information in FreeBSD?
:
:Thanks,
:Jos
Well, ktrace -i will certainly follow children. In fact, ktrace can
attach to all current children (-d) of a process as well as attach to
new children. Yahoo found a few bugs in ktrace by r
:Here's what pstack does:
:
: pstackPrint a hex+symbolic stack trace for each lwp in
: each process.
:
:Solaris truss(1) has this:
:
: -lIncludes the id of the responsible lightweight
: process (LWP) with each line of trace output. If
:
56KBytes/sec over
:IEEE and he can do 20KBytes/sec over the original lines. With 1541 FLASH,
:the thoretical maximum is about the same as my IEEE transfers. In either
:case the transfer is 1000% better than Commodores, and the only holding
:factor is that the data must be read off disk before it c
ome references to it.
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
:Unix!
:
:I've got _all_ the original CBM stuff for the VIC-20 and C-64, hardware
:and hardcopy. Even some aftermarket FDDs.
:
:I poked a _lot_ of stuff from Compute!, including the assembler, and
:have several of their wire-bound books, too.
:
:I've got the 6502 monitor and 300bps modem cartridg
rcuit board,
turn it upside down, and apply a heat gun to the backside. Now *that*
is fun!
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAI
to hack
:virus code to make the MPSW fixup patch live across a reboot
:so that you could run the PC and Mac emulators, but it let
:you run SVR3.2 on the A1000, if you had the Supra SCSI drive
:and Zorro controller... ah, the first UNIX box I ever owned...).
:
:-- Terry
Ho!
:...
:> in a wire select to an unused bank, which meant the screen was spaghetti
:> on power-up until i LOAD'd a copy of the character set.
:
:UGH. You didn't load the RAM from the ROM at power on?!?
No extra rom slots. Had to load from tape or floppy.
:We had the "high resolution
Heya Alfred, Alan, hackers. Ok, I've been tracking down a bug with A
russian news admin and I believe I may have found it (he's testing it),
plus a few other bugs.
I would like a review of this (for -stable, but applies to -current too):
Patch section 1
Here we were
:Ian
Ok, cool. I'll get the commit gears started for the
first part of the patch.
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [E
:
:In message <[EMAIL PROTECTED]>, Matthew Dillon wri
:tes:
:>Ok, cool. I'll get the commit gears started for the
:>first part of the patch.
:
:FYI, I was able to reproduce this and confirm that the first part
:of your patch fixes it. All that it takes is for the mknod
a guinea pig. I would be MFC'ing based on what he actually
comitted to -current.
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAIL PROTECTE
What? You don't like directories named '...w^Ha^Hr^He^Hz^H^H^H' ?
I like it, but there are a few problems. What about underscore? And
will this mess up people using ftp outside the U.S.?
-Matt
es are aborted. Passing -1 aborts the entire buffer.
ioctl(fd, IOPTR_RCLEAR, 0);
Abort any previously reserved write buffer and force the OS
to unmap any cached memory space associated with readptr().
-Matt
:
:Julian Elischer wrote:
:> You can mmap() devices and you can mmap files..
:>
:> you cannot mmap FIFOs or sockets.
:>
:> for this reason I think that devices are still well represented by
:> vnodes. If we merged vnodes and vm objects,
:> then if devices were not vnodes, how would you represe
:> Hi Folks,
:>
:> I have an emergency and I need some quick help.
:>
:> I just ran boot0cfg to install boot0 as my boot code instead of the standard
:> mbr. This is on a dedicated FreeBSD system. Now the system boots and gives
:> me a choice F4 FreeBSD or F5 Disk 1 both selections hang. I had n
:Hi!
:
:Is it reasonable to administratively limit users' ability to call fork
:too other ? Users can take away too much CPU time even if you have
:limited them by login.conf 'cputime' limit - just forking lightweight
:processes too often.
:
:If it seems good thing to do, i'll try to code it and s
work on -current. Alfred has been
very patient I know.
It would also be nice if someone took a more detailed look at the
one 'XXX' comment I have in there for the LNC driver.
-Matt
:> >
:> > What disgusting code. I find it amazing that they didn't even stick in
:> > some peephole optimizer to at least limit it to one operation.
:>
:> It's clearly the result of work in progress :-).
:
:I see really cruddy stuff like this every time i do a gcc -S, don't
:they watch for and t
easonable display size at the full 30Hz frame
rate. Please feel free to email me simple patches.
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to
ems pretty comprehensive
though I am still attempting to figure out how to mmap() multiple
frames.
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAIL
Hey, great find! I see two places in that file where ICANON is
improperly tested against t_iflag instead of t_lflag.
Have you done any further testing using the t_lflag field instead
of t_iflag? Did it fix the crash you were getting?
-Mat
:
:Hi,
:
:I changed the NetBSD sources (which have the identical bug) and it
:now it works as it should. i.e. writes to the pseudo terminal block,
:don't overflow the real tty. I have not done any testing with FreeBSD.
:
:The bug is easy to reproduce and the fix is correspondingly easy to verify
Mike, did you get my email clarifying what the masking code does?
gettimeofday performance:
1 process / fast acpi timer 514000 calls/sec
1 process / dillon acpi timer 366284 calls/sec
(previously posted patch set)
1 process / dillon acpi timer
l-Henning Kamp | UNIX since Zilog Zeus 3.20
Ugh. Oh well. So much for the code being simple. I'll incorporate
a similar change which adjusts the mask dynamically in both directions.
-Matt
t. Be careful,
you can blow up the time counter if you set it to a bad value like 0
reasonable values: -1, -2, -4, -8, -16, -32, -64. If you set it too
strict (i.e. -1) it will auto fallback.
-Matt
cks or not.
The soluion is to not use ftruncate() to create such files but
to instead pre-create the file by filling it with zero's.
-Matt
Matthew Dillon
<[EMA
lution sort
object->memq (it's the actual page queue that is the problem,
not the object queue). Looking at it some more I believe
this may be a viable solution. I am going to work something
up.
-Matt
Ok, here is a test patch. Now, there are some instructions to go along
with this patch, so continue reading.
I have implemented two optimizations. You can turn either or both (or
neither) on with a sysctl. I would like those interested to test all
four combinations. Be sur
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
words, msync() should probably not change the
:LRU order.
:
:-- Terry
No, it has no effect on the vm_page's act_count.
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
object->memq (which is also a possibility at some
later time).
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
of applications.
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
: vm.msync_flush_flags
: | 0 | 1 | 2 | 3 |
:---+---+---+---+---|
:write |
The vnode recycle code is currently not able to recycle vnodes in an
LRU fashion because we cannot move vnodes within the mount's vnodelist
without causing a number of filesystems, including UFS, to lose track
when scanning the mount's vnodelist and start doing loop-restarts,
r
g
written out sequentially. But this is just one particular feature
under the specified conditions so you can't really use it to justify a
general comparison.
-Matt
Matthew Dillon
it implements a very good interleaving
algorithm.
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
:Am I the only one who saw that he attached it to his 1st mail?
:
:Here you go:
:
map.c (which tracks MMU mappings), and
vm/vm_pageout.c and vm/vm_object.c (so you can see how object
flushing works.. I suggest reading the vm_object_page_clean()
procedure).
-Matt
Matthew Dillon
:>ptr = (char *) malloc(MALLOC_SIZE);
:>bzero(ptr, MALLOC_SIZE);
:
:The bzero is unnecessary on FreeBSD. Allocated pages start out
:zero'ed. Part of the performance issue might be that FreeBSD is
:being asked to zero the pages twice, instead of once.
malloc() does not guarentee a z
s you also deal with fork()
or you will share what is supposed to be process-private memory
across the fork().
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
:ehlo.
Oops. I'm sorry! This was an email all the from last october! I didn't
mean to resurrect the thread :-)
-Matt
:ehlo.
:
: I was told that diff format I used is unappropriate for most cases,
: so I redo it in unified (-u) format.
:
: Purpose: t
mp; 0' for 8 to me. NBBY is 8,
fs_frag of 8, results in 0xFF >> 0 which is 0xFF, not 0.
The original code looks correct. Try this:
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
:Doug White wrote:
:> I've been asked several times about how to get CPU speed information for
:> inventory purposes.
:>
:> People would really like the speed number printed on the chip, not what
:> it's currently running at, if that's retrievable :)
:
:Can't mask the speed number.
:
:Chips with
:Content-Type: text/plain; charset=us-ascii
:
:I submitted a bug report (PR# bin/31933) in November that's never been looked
:at. I even enclosed a patch that fixes the bug. Is there anything that I can
:do to get the bug looked at sooner rather than later?
:
:- --
:Ted Cabeen http://
:
:The user should never use '--'. The code just assigns the initial username or
:uid to the '--' argument so that it is associated with an argument. I wanted
:an argument that wouldn't be accidentially tried by the user thus causing
:inconsistencies if they used both the '--' argument and a le
While working on PR 36504 I noted that there are situations where
the system cannot continue if zalloc() fails.
Rather then force the caller of zalloc() to check for NULL in these
cases, I think it makes sense to add another flag, ZONE_PANICFAIL,
allowing a zone to automatical
(note: this kassert is wrong):
:+ KASSERT(item == NULL, ("zitems unexpectedly NULL"));
should be:
:+ KASSERT(item != NULL, ("zitems unexpectedly NULL"));
-Matt
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe f
Well, if UMA is going to eventually remove kmem_map and
just use kernel_map (or vise-versa), then the issue comes
down to why kernel_map->lock is lockmgr lock in the
first place, instead of a mutex.
I think the reason is historical. Basically kernel_map was
originally a
bsolete and
can be removed.
I will remove it now.
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
:>If you aren't using VLAN tagging, you shouldn't care.
:
: No, that is absolutely not correct. The checksum problems happend in many
:situations, depending on the chipset and other factors. The problem that
:resulted in the commit to disable the receive hardware checksum was caused
:by small p
,
:-- Terry
It was on incoming checksums I believe.
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
.
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
:Im pretty sure this has to be related to the Broadcom Cards.
:
:I tested it again this morning from the Compaq rack with the bge card
:in, same problem. tried doing cp -r /usr/ports /data1-home/ (nfs), copied
:about 1.3M and then froze. Even cd'ing to the directory to try and do
:an ls caused my
:
:Already running the card and switch port in 100BaseTX FDX (forced) :)
:
:Would use GigE if the switch supported it tho
Ack. Just rip the damn thing out and put in a normal 100BaseTX card,
then (if you haven't already). The whole system will probably be
happier.
I think this has gotten off-topic. I am going to go back and respond
to the original posting.
When I was doing BEST Internet we had very similar problems with our
mail servers. We were constantly cpu-bound. I was constantly fiddling
with it (sendmail in this case). For mon
:
:Get this guy outta here!
Firstly, Chris, I understand your frustration. But, please, do *NOT*
Cc a random ISP ( [EMAIL PROTECTED] ) in email that you also
send to a FreeBSD list, let alone several FreeBSD lists!
This particular spammer is operating from 213.96.224.148, which
:
:Hi Matt and everyone,
:
:I remember that Matt had mentioned adding diffs of the commits to the
:commit-mail, which would make it possible to quickly view what changed
:without leaving the MUA and manually searching for the proper CVS
:diff/rdiff incantation to view the diff.
:
:The idea was re
You can filter commit messages by creating a filter control
file on freefall.freebsd.org in /home/dillon/filters.
Create the file in /home/dillon/filters. Note that the
only file the filter will recognize is your username. If the program
becomes universal we will create a
add a manual page, and fix reported bugs first )
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebs
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
:I've just seen Matt Dillon's email on this topic
:([EMAIL PROTECTED]). I immediately thought
:about the overlap between two projects.
:
:FreshSource (http://www.FreshSource.org) will allow yo
night.
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
Ok, I've made some improvements to dfilter.
* I cleaned up the README file (/home/dillon/dfilter/README on freefall),
getting rid of variables that are not yet supported.
* If the program cannot parse your filter it should now send you an
email with the error. (You can
:| If none of the features are that useful to you, then by all means stick
:| with 4.7 or whatever we're up to when 5.0-release comes out. I'm sure
:
:Well, I definitely plan to move to 5.x. The question is when. When it
:will be ready for general use, and when I will be able to learn all the
sure :-).
0xdeadc162 - 0xdeadc0de would be the structural offset relative to
the pointer. The pointer itself is probably sitting at 0xdeadc0de.
-Matt
Matthew Dillon
but be warned
the full-sized photos are 2MB ea.
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-
icult!).
It's up at:
http://apollo.backplane.com/USENIX2002/FriB/
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAIL PROTECTED]
w
).
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
occur HERE XX */
}
}
The goto restart condition should occur one level up, as I show in
the comment.
Could someone take a look at this and tell me if I am blowing smoke?
-Matt
ting. The only reason it works at all
is because sys/queue.h does not clear out the pointers in the node
that was just removed. The code is just plain wrong, though, because
the queue mechanisms make no such (documented) guarentee.
-Matt
reads have
considerably more blocking states in -current then processes do
in -stable. Either that or there are races in -current that we don't
know about.
-Matt
:> I just added debug code in the TAILQ code that sets the forward pointor
:> to -1. Since Matt had this it's possible that this is what hit him?
:
:Most definitely. If you reset tqe_next, the code stops working. Try
:the patch below.
:--
:Jonathan
:
:Index: kern_synch.c
:
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
that we
can gdb the core against).
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
:questions:
:
:1) How do I give you an entire `ps` output from DDB ? Is there a way to
:output it to a floppy or something ? Or are you suggesting to copy down
:by hand ~1000 lines of ps output ?
If you have a couple of machines you can use a null-modem cable and
make the target machin
ectly.
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
:
:This is also the _default_ for how solaris manages sysv segments,
:although it would be nice if we could get the OBJT_PHYS stuff to
:use 4meg pages (unless someone already did that?)...
:
:Anyhow, I'm glad we corrected your misconception and we now have
:a more accurate understanding of how th
post-fork modifications (like when you mmap() or get a SysV
shared segment). The rmap patches are roughly equivalent to our
i386 pmap code and allow Rik to implement page queues and proper page
aging.
-Matt
: http://old.lwn.net/2002/0124/kernel.php3
: "What Rik van Riel is up to."
:
:FWIW: In the original Article:
:
: http://lwn.net/Articles/3327/
:
:they say they've gon to a 3 level page table scheme for the Hammer
:port.
:
:-- Terry
They are still using a 2 level page table
:[commenting live from ottawa]
:
:On Wed, Jun 26, 2002 at 10:11:54AM -0700, Matthew Dillon wrote:
:> They are still using a 2 level page table scheme in the linux
:> kernel. That is, they didn't try to integrate the 3-level hardware
:> in the hammer with the
:[commenting live from ottawa]
Pictures! We want pictures!
-Matt
Matthew Dillon
<[EMAIL PROTECTED]>
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "
701 - 800 of 1441 matches
Mail list logo