On Tue, 12 Aug 2003, Petri Helenius wrote:
Is there way to view PCI / PCI-X configuration with bus-width and clock
values when FreeBSD has already booted?
If you know where in PCI config space to look you could use pciconf to
query it. Can't say I've heard of PCI speed negotiation
Is there way to view PCI / PCI-X configuration with bus-width and clock values
when FreeBSD has already booted?
Pete
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to [EMAIL
Related to the em driver, 82540M has not worked since sometime in
5.1-BETA time,
I filed a pr on that a few months ago but it seems the fault might be
with PCI IRQ routing,
not the em driver itself.
Pete
Hartmann, O. wrote:
On Wed, 20 Aug 2003, Colin Faber wrote:
Hi.
I first swapped the
My first try with ATAng on -CURRENT as of 18 hours ago ended up dying
around reboot. Other than this, it seems to work as expected.
I don't have WITNESS or INVARIANTS on this box, but I can recompile
kernel with them if it helps.
# reboot
Aug 28 09:24:35 rms21 reboot: rebooted by root
boot()
ATAng does not seem to detect my RAID set, while ATAold boots it
fine, first boot -v with ATAng kernel and then ATAold.
Any ideas if raid configs should be compatible between them? Should
make sense that they are so people upgrading don't get burned...
ata2-master: pio=0x0c wdma=0x22 udma=0x45
I also noticed that earlier ATAng spits out these messages; (cutpasting boot -v
is a little tedious, but I can copy it all if it helps)
The err=0x01 caught my eye.
Pete
atapci0: Promise PDC20619 UDMA133 controller port
0x1080-0x10ff,0x1480-0x148f,0x1000-0x107f mem
Soren Schmidt wrote:
It seems Petri Helenius wrote:
ATAng does not seem to detect my RAID set, while ATAold boots it
fine, first boot -v with ATAng kernel and then ATAold.
Any ideas if raid configs should be compatible between them? Should
make sense that they are so people upgrading don't get
Soren Schmidt wrote:
Oh, that one :) fixed...
-Søren
Do you have an idea why my SMP system would crash under heavy load, like
mysql shutdown, just a reboot, etc.?
The system did not have any issues before ATAng was committed.
This is one of the panics;
I have a DVD drive which is recognized by the BIOS as UDMA33 and also the
documentation states that UDMA33 is the fastest it goes, however with ATAng
the hw.ata.atapi_dma is gone and the kernel thinks PIO4 is the way to go;
acd0: DVDR HL-DT-ST DVDRAM GSA-4040B at ata7-master PIO4
it works alright
I´ve been getting these crashes on a DVD-RAM filesystem which gets
unmounted,
newfs´d, remounted and used. No specific pattern though but they are
quite frequent.
Not sure if these came after or before ATAng.
db trace
Debugger(c03e1565,0,c03f09e4,dd7309d0,100) at Debugger+0x55
Atacontrol detach / attach seems to panic if atapicam is complied in the
kernel.
Should this work or is the work to port this to ATAng still undergoing?
This is current from four days ago, just after Soren's ar DMA size commit.
# atacontrol detach 4
ad8: deleted from ar0 disk1
ar0: WARNING -
While fscking from previous panic in ufs2 code on -CURRENT from Sep 2, I
got this:
panic: ffs_copyonwrite: recursive call
cpuid = 1; lapic.id = 0100
Debugger(panic)
Stopped at Debugger+0x55: xchgl %ebx,in_Debugger.0
db trace
Debugger(c03e1565,100,c03eebf0,dd78eac8,100) at
While the machine was bg-fscking and building a new kernel, on -CURRENT from 2 Sep
after the ataraid transfer size patch, it fell over like this:
Fatal trap 12: page fault while in kernel mode
fault virtual address = 0xa2b22cf4
fault code = supervisor write, page not present
It seems that the same bug that initially stopped ataraid working
at all is there with rebuild:
kompak# atacontrol rebuild 0
ad4: FAILURE - oversized DMA transfer attempted 131072 65536
ad4: setting up DMA failed
kompak# atacontrol status 0
ar0: ATA RAID1 subdisks: ad4 ad8 status: REBUILDING 0%
The problem of not being able to reboot without a panic seems to persist on
late current with ATAng and SMP.
Pete
Sep 16 13:16:11 rms21 reboot: rebooted by pete
Sep 16 13:16:11 rms21 syslogd: exiting on signal 15
boot() called on cpu#0
Waiting (max 60 seconds) for system process `vnlru' to
Soren Schmidt wrote:
It seems Petri Helenius wrote:
The problem of not being able to reboot without a panic seems to persist on
late current with ATAng and SMP.
The below doesn't say much actually, but I can reboot to my hearts content
on my SMP box here with no problems. What else do you
Got the chatter below on a box which had some libkse and some mmap activity
when it got shutdown. Built from morning of 30th May sources.
Pete
May 30 08:56:08 kompak halt: hallted by root
ock order reversal
1st 0xc3335aa8 sigacts (sigacts) @ kern/subr_trap.c:248
2nd 0xc3347d88 process lock
Is there anyone actually successfully using raidframe and if yes, what kind
of hardware?
Same question goes for any recent SCSI RAID controllers supported
by FreeBSD.
I admit not having tried all combinations but it seems that using anything
else than simple ahc scsi stuff results in kernel panic
RAIDframe is non-functional in 5.1 and -current [kern/50541] and it would be
unwise to use it in 5.0 for anything other than experimentation. Hopefully it
will be fixed before 5.2.
Makes one wonder how broken code ever got into the tree in the first place...
Pete
-
From: Scott Long [EMAIL PROTECTED]
To: Petri Helenius [EMAIL PROTECTED]
Cc: Tim Robbins [EMAIL PROTECTED]; [EMAIL PROTECTED]
Sent: Sunday, June 01, 2003 11:20 PM
Subject: Re: raidframe
Petri Helenius wrote:
RAIDframe is non-functional in 5.1 and -current [kern/50541] and it would
If you rewind to last October, RAIDFrame worked well. Unfortunately,
some kernel interfaces changed in between now and then and RAIDFrame was
left behind. I am remis in not fixing it, but please understand that I
also have quite a few other responsibilities, and I get paid $0 to work
on
FreeBSD 5.x series is slowly progressing, but is nowhere near to
production quality. As the things are currently, you simply waste
your time.
This is only my opinion and I don't want to offend anyone.
IMO, software does not magically get better but it must be actively
being used and
replies
and without a management utility the card would be useless for me anyway.
I´ll ask them for another loaner if needed.
While we´re at it, what´s the utility command to turn off the alarm on
the card? It got annoying after a while practising pulling out drives :-)
Pete
Scott
Petri Helenius
I'm trying to upgrade RELENG_4 to RELENG_5_1 using source and make
buildworld compiles
a while and then stops with:
Question is if this is supported and if yes, what should I do differently?
Pete
cc -fpic -DPIC -O -pipe -mcpu=pentiumpro -DPTHREAD_KERNEL
There...
http://www.freebsd.org/cgi/query-pr.cgi?pr=i386/52835
Pete
- Original Message -
From: Robert Watson [EMAIL PROTECTED]
To: Petri Helenius [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]; [EMAIL PROTECTED]
Sent: Thursday, May 15, 2003 9:54 PM
Subject: Re: 5.1-BETA em
Could you file
The CPUs you see are the two logical cores of one of your CPUs.
The kernel HLTs one of them by default due to performance issues
when running certain loads on hypethreading.
Pete
- Original Message -
From: Brendon and Wendy [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Cc: [EMAIL PROTECTED]
I´m booting 5.1 install in IBM x342 with floppies (because the thing does not seem to
want
to boot from the CD-ROM) and when sysinstall goes into probing devices, I get a message
about / filesystem being full and I´m going nowhere without my init and then the
kernel
panics.
Any ideas how to
It seems that the ata driver still has some issues with configuring the
RAID devices. Sources cvsupped today.
Pete
Hit [Enter] to boot immediately, or any other key for command prompt.
Booting [/boot/kernel/kernel]...
Copyright (c) 1992-2003 The FreeBSD Project.
Copyright (c) 1979, 1980, 1983,
Is this normal that multiple installations of 5.1-RELEASE complain
when running bsdlabel:
# /dev/ad0s1:
8 partitions:
#size offsetfstype [fsize bsize bps/cpg]
a: 524288 634.2BSD 2048 16384 32776
b: 4194304 524351 swap
c: 39102273
Poul-Henning Kamp wrote:
In message [EMAIL PROTECTED], Petri Helenius writes:
Is this normal that multiple installations of 5.1-RELEASE complain
when running bsdlabel:
# /dev/ad0s1:
8 partitions:
#size offsetfstype [fsize bsize bps/cpg]
a: 524288 634.2BSD 2048
Just saw the support for the controller go in. Does the driver support any
RAID1 style mirroring?
Pete
___
[EMAIL PROTECTED] mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to [EMAIL PROTECTED]
[EMAIL PROTECTED]
It seems Petri Helenius wrote:
Just saw the support for the controller go in. Does the driver support any
RAID1 style mirroring?
Only our own ATA RAID (see atacontrol)..
I do have an Adaptec 1210 that has RAID capabilities and I have
deciphered their RAID config info
Soeren Schmidt wrote:
So that drives could be assigned to a mirror / stripe set before installing without
having to build the mirror on another machine first?
(in this case the BIOS does not help)
You cant boot from a stripe on a controller that doesn't have a BIOS
to do it.
Sure, but I
Christian Brueffer wrote:
On Sat, Aug 02, 2003 at 11:19:48AM +0300, Petri Helenius wrote:
I just wish there would either be more functionality in the twe driver
or somebody would come
out with 8-12 port SATA controller. Looking at the issues for example
the Adaptec SCSI RAIDs
have, it seems
When using console redirection on SE7505VB2 Intel board I seem to get
occasional
failures on starting the second core of the second CPU.
Also, on the same board, including device ichsmb on kernel config will
result in
hang on boot 100%.
The sources are from yesterday.
/boot/kernel/acpi.ko
I'm not aware of any registers in the standard PCI config space that
will tell you the speed of the bus. Some PCI devices will make that
information available, but not in a standard way. The BIOS of some
higher-end systems might also tell you this information.
How about the busmaster
True, those parameters are available, but the original question was
about reporting the bus width and frequency, which are not available.
How can these parameters be displayed?
(whether the original question was only for the specifically mentioned
parameters is actually matter of semantics,
I'm trying to create a raid set but after hitting the end of
initializing a raid5
stripe the following is displayed on the console:
raid0: node (Rod) returned fail, rolling backward
raid0: IO Error. Marking /dev/da2s1e as failed.
raid0: node (Rod) returned fail, rolling backward
raid0: IO
I seem to want to quite the opposite what other people would like to happen,
we have a supermicro P4DPR board with two Xeon´s. Since lately the kernel seems to
want to launch the virtual cores regardless of BIOS setting of HyperThreading being
[Disabled].
Any ideas how to disable the cores since
disabling with a kernel variable?
Pete
On Mon, 3 Mar 2003, Petri Helenius wrote:
I seem to want to quite the opposite what other people would like to happen,
we have a supermicro P4DPR board with two Xeon´s. Since lately the kernel seems to
want to launch the virtual cores regardless of BIOS
Thanks, that´s greatly appreciated.
Pete
On 03-Mar-2003 Petri Helenius wrote:
After some reading around I ended up putting a return; on top of
the mptable_hyperthread_fixup function and my performance is back.
Setting the sysctl helps some but it does not really fix the performance
This does look odd... maybe there's a leak somewhere... does in use
go back down to a much lower number eventually? What kind of test are
you running? in pool means that that's the number in the cache
while in use means that that's the number out of the cache
currently being used
Yeah, it kinda sucks but I'm not sure how it works... when are the
mbufs freed? If they're all freed in a continous for loop that kinda
sucks.
I think there is nothing really special about the driver there? The mbufs
are allocated in the driver and then freed when other parts in the
While you are there debugging mbuf issues, you might also want to try
this patch:
Didn´t run profiling yet, but judging from the CPU utilization, this did not change
the whole picture a lot (dunno why it should since CPU is mostly spent freeing the
mbufs,
not allocating them)
Pete
i have yet to see a cisco ios image supporting ipv6 that was usable
in production environment. and i have tried hard.
This is getting OT but on the subject of repelling users, they´re probably
trying hard to repel their users to the vendor J boxen.
but i will admit to not having seen apollo
Any comments on the high cpu consumption of mb_free? Or any other places
where I should look to improve performance?
What do you mean high cpu consumption? The common case of mb_free()
is this:
According to profiling mb_free takes 18.9% of all time consumed in kernel and is
almost
I did some profiling on -CURRENT from a few days back, and I think I haven´t
figured the new tunables out or the code is not doing what it´s supposed to
or I´m asking more than it is supposed to do but it seems that mb_free
is being quite wasteful...
Any pointers to how the new high/low
Yes, it's normal. The commit log clearly states that the new
watermarks do nothing for now. I have a patch that changes that but I
haven't committed it yet because I left for vacation last Sunday and I
only returned early this Monday. Since then, I've been too busy to
clean up
There's probably a tightloop of frees going on somewhere. It's tough
for me to analyze this as I cannot reproduce it. Have you tried
running your tests over loopback to see if the same thing happens?
What is the definition of tightloop? The received packet mbufs are freed
when the
M. Warner Losh wrote:
ISA support is not obsolete. All new PCs still have ISA busses. They
might not have ISA Expansion Bus Slots, but they all[*] still connect
their serial ports, parallel ports, and mouse/keyboard ports via ISA.
Not to mention i8254 which gets to be major pain if ACPI
Terry Lambert wrote:
Ah. You are receiver livelocked. Try enabling polling; it will
help up to the first stall barrier (NETISR not getting a chance
to run protocol processing to completion because of interrupt
overhead); there are two other stall barriers after that, and
another in user space
You can get to this same point in -CURRENT, if you are using up to
date sources, by enabling direct dispatch, which disables NETISR.
This will help somewhat more than polling, since it will remove the
normal timer latency between receipt of a packet, and processing of
the packet through the
If you are asking for paper references, then I can at least tell
you where to start; go to: http://citeseer.nj.nec.com/cs and look
for Jeff Mogul, DEC Western Research Laboratories, Mohit
Aron, Peter Druschel, Sally Floyd, Van Jacobson, SCALA,
TCP Rate halving, Receiver Livelock, RICE
53 matches
Mail list logo