Sony laptop sound problems with auich0 and openbsd 3.9, 4.0, 4.1-beta.
Hello list,
Having serious problems with Sony PCG-V505EX laptop.
Basically, sound doesn't work unless there is some activity (traffic) on
fxp0 or iwi0 interfaces. Even if there is some traffic, sound grinds to
a halt after few seconds. It doesn't matter what source sound is coming
from: cd, dvd, mp3, ogg.
First thing that comes to my mind is IRQ issue. there are 5 devices
using the same irq 9 by looking at dmesg. BIOS settings are really
limited on this laptop.
Same exact problem occurs when using 3.9 4.0 releases, with of course
GENERIC kernel.
Any help is greatly appreciated.
DMESG:
OpenBSD 4.1-beta (GENERIC) #3: Sun Feb 18 11:08:26 PST 2007
[EMAIL PROTECTED]:/usr/src/sys/arch/i386/compile/GENERIC
cpu0: Intel(R) Pentium(R) M processor 1500MHz ("GenuineIntel"
686-class) 1.49 GHz
cpu0:
FPU,V86,DE,PSE,TSC,MSR,MCE,CX8,SEP,MTRR,PGE,MCA,CMOV,PAT,CFLUSH,DS,ACPI,MMX,FXSR,SSE,SSE2,TM,SBF,EST,TM2
real mem = 535851008 (523292K)
avail mem = 480706560 (469440K)
using 4256 buffers containing 26914816 bytes (26284K) of memory
mainbus0 (root)
bios0 at mainbus0: AT/286+ BIOS, date 11/21/03, BIOS32 rev. 0 @ 0xfd751,
SMBIOS rev. 2.3 @ 0xd8010 (17 entries)
bios0: Sony Corporation PCG-V505EX(UC)
apm0 at bios0: Power Management spec V1.2
apm0: AC on, battery charge unknown
apm0: flags 30102 dobusy 0 doidle 1
pcibios0 at bios0: rev 2.1 @ 0xfd750/0x8b0
pcibios0: PCI IRQ Routing Table rev 1.0 @ 0xfdf30/176 (9 entries)
pcibios0: PCI Interrupt Router at 000:31:0 ("Intel 82371FB ISA" rev 0x00)
pcibios0: PCI bus #3 is the last bus
bios0: ROM list: 0xc/0x1 0xd8000/0x4000! 0xdc000/0x4000!
acpi at mainbus0 not configured
cpu0 at mainbus0
cpu0: Enhanced SpeedStep 1500 MHz (1484 mV): speeds: 1500, 1400, 1200,
1000, 800, 600 MHz
pci0 at mainbus0 bus 0: configuration mode 1 (no bios)
pchb0 at pci0 dev 0 function 0 "Intel 82855PE Hub" rev 0x03
ppb0 at pci0 dev 1 function 0 "Intel 82855PE AGP" rev 0x03
pci1 at ppb0 bus 1
vga1 at pci1 dev 0 function 0 "ATI Radeon Mobility 9200" rev 0x01
wsdisplay0 at vga1 mux 1: console (80x25, vt100 emulation)
wsdisplay0: screen 1-5 added (80x25, vt100 emulation)
uhci0 at pci0 dev 29 function 0 "Intel 82801DB USB" rev 0x03: irq 9
usb0 at uhci0: USB revision 1.0
uhub0 at usb0
uhub0: Intel UHCI root hub, rev 1.00/1.00, addr 1
uhub0: 2 ports with 2 removable, self powered
uhci1 at pci0 dev 29 function 1 "Intel 82801DB USB" rev 0x03: irq 9
usb1 at uhci1: USB revision 1.0
uhub1 at usb1
uhub1: Intel UHCI root hub, rev 1.00/1.00, addr 1
uhub1: 2 ports with 2 removable, self powered
uhci2 at pci0 dev 29 function 2 "Intel 82801DB USB" rev
0x03pci_intr_map: no mapping for pin C
: couldn't map interrupt
ehci0 at pci0 dev 29 function 7 "Intel 82801DB USB" rev
0x03pci_intr_map: no mapping for pin D
: couldn't map interrupt
ppb1 at pci0 dev 30 function 0 "Intel 82801BAM Hub-to-PCI" rev 0x83
pci2 at ppb1 bus 2
cbb0 at pci2 dev 5 function 0 "Ricoh 5C475 CardBus" rev 0xb8: irq 3
"Ricoh 5C551 Firewire" rev 0x00 at pci2 dev 5 function 1 not configured
fxp0 at pci2 dev 8 function 0 "Intel PRO/100 VE" rev 0x83, i82562: irq
9, address 08:00:46:cd:ab:1c
inphy0 at fxp0 phy 1: i82562ET 10/100 PHY, rev. 0
iwi0 at pci2 dev 11 function 0 "Intel PRO/Wireless 2200BG" rev 0x05: irq
9, address 00:0e:35:0d:38:65
cardslot0 at cbb0 slot 0 flags 0
cardbus0 at cardslot0: bus 3 device 0 cacheline 0x0, lattimer 0x40
pcmcia0 at cardslot0
ichpcib0 at pci0 dev 31 function 0 "Intel 82801DBM LPC" rev 0x03
pciide0 at pci0 dev 31 function 1 "Intel 82801DBM IDE" rev 0x03: DMA,
channel 0 configured to compatibility, channel 1 configured to compatibility
wd0 at pciide0 channel 0 drive 0:
16-sector PIO, LBA48, 57231MB, 117210240 sectors
wd0(pciide0:0:0): using PIO mode 4, Ultra-DMA mode 5
atapiscsi0 at pciide0 channel 1 drive 0
scsibus0 at atapiscsi0: 2 targets
cd0 at scsibus0 targ 0 lun 0: SCSI0
5/cdrom removable
cd0(pciide0:1:0): using PIO mode 4, Ultra-DMA mode 2
ichiic0 at pci0 dev 31 function 3 "Intel 82801DB SMBus" rev
0x03pci_intr_map: no mapping for pin B
: polling
iic0 at ichiic0
auich0 at pci0 dev 31 function 5 "Intel 82801DB AC97" rev 0x03: irq 9,
ICH4 AC97
ac97: codec id 0x594d4803 (Yamaha YMF753-S)
ac97: codec features 18 bit DAC, No 3D Stereo
audio0 at auich0
"Intel 82801DB Modem" rev 0x03 at pci0 dev 31 function 6 not configured
isa0 at ichpcib0
isadma0 at isa0
pckbc0 at isa0 port 0x60/5
pckbd0 at pckbc0 (kbd slot)
pckbc0: using irq 1 for kbd slot
wskbd0 at pckbd0: console keyboard, using wsdisplay0
pms0 at pckbc0 (aux slot)
pckbc0: using irq 12 for aux slot
wsmouse0 at pms0 mux 0
pcppi0 at isa0 port 0x61
midi0 at pcppi0:
spkr0 at pcppi0
npx0 at isa0 port 0xf0/16: reported by CPUID; using exception 16
biomask effd netmask effd ttymask
pctr: 686-class user-level performance counters enabled
mtrr: Pentium Pro MTRR support
dkcsum: wd0 matches BIOS drive 0x80
root on wd0a
rootdev=0x0 rrootdev=0x300 rawdev=0x302
# audioctl -a
name=ICH4 AC97
version=0x03
config=auich0
encodings=ulinear:8,mulaw:8*,alaw:8*,slinear:8*,sli
serial console on macbook?
can't install 4.0 or snapshots on my macbook due to what appear to be issues with the usb controller. ( lots of errors about the usb controller, and the keyboard is nonresponsive... no capslock light, no input ) does anyone have any ideas about how to capture the dmesg so i can submit? thanks, ben
Re: OpenBSD speed on desktops
* Travers Buda <[EMAIL PROTECTED]> [2007-02-18 14:42:34]: > * Jon Drews <[EMAIL PROTECTED]> [2007-02-18 11:17:08]: > > > On 2/17/07, R. Fumione <[EMAIL PROTECTED]> wrote: > > >Hello, > > > > > >I am using OpenBSD on server since few years now, and I am very happy > > >with it's easy maintenance and it's stability. I want to try on > > >desktop, and I am having trouble. > > > > > >Everything is much slower than existing Linux system. For example, > > >Firefox takes 3-5 seconds to start on Linux but ~10 seconds on > > >OpenBSD on same machine! > > > > So? For all practicality's sake, you're only starting firefox a > few times a day (in my normal usage.) Basically, once you start > getting around to about 10 seconds for a massive program to start > up, you're really not going to see any more efficiency in your work > by an increased speed-up. > > IMHO, more speed than today's modern sorts of computers (hammer, > core) is really not going to improve the user experience. Likewise, > a slight speed-up in the OS is really not going to do much for you. > > But, after all, you are getting something out of that minute launch > time disparity. The return is much greater than the cost. > > By the way... I'd imagine the slowness attributed to OpenBSD in > this case actually lies with this: > $ grep Os /usr/ports/www/mozilla-firefox/Makefile > $ --enable-optimize=-Os > And and thank god for it. I remember how > firefox totally hosed my memory on a bunch of linux systems with -O2. It > didn't matter if the box had 256 or a gig of ram, somehow firefox > managed to misuse all of it and played havoc with the swap--other > running applications suffered. All for 5 seconds faster startup. > > -- > Travers Buda -- Travers Buda
Re: is there [EMAIL PROTECTED] archive?
> I agree with scorch, how do we find out what hardware is working best > and most used with OpenBSD. Quite honestly anything I get just works. Really. Everything works. And if it does not work perfectly, I talk to the various developers and we try to get it to work better. I talk to the various developers in the same way as anyone can, and we focus on the various issues that show up, be they drivers or the base bus code. So that by the time you get to the point of worrying about what hardware works best, it should all work best. > Even we you cant release the dmesg reports, what about a statistics > page, something along the lines of, x amount of x mobos is used with > OpenBSD, and other hardware as well. > > would that be possible? No. No time. Sorry. Even if we had the stats, how do you know that most people in dmesg are reporting well-working machines, instead of say... maybe they send in a dmesg if it works poorly? Many of you entirely overestimate the quality of the dmesg output. Many don't say what the machine is in detail, or how well it works. We do not intend to change our policy. Sorry.
Re: is there [EMAIL PROTECTED] archive?
I agree with scorch, how do we find out what hardware is working best and most used with OpenBSD. Even we you cant release the dmesg reports, what about a statistics page, something along the lines of, x amount of x mobos is used with OpenBSD, and other hardware as well. would that be possible? On 18/02/07, Theo de Raadt <[EMAIL PROTECTED]> wrote: > > The dmesgs submitted to [EMAIL PROTECTED] are not publicly accessible. > > At some point in time this was discussed, but we can't do that, since > > we never told people that they would be published. So they remain > > accessible to developers only. They are consulted very often, so keep > > them coming in! > > > On this topic - should we resubmit dmesgs periodically as the machines > are updated to newer versions? DEFINATELY! There are developers in the group who read the dmesg output very closely, looking for issues that you, as a user, would not even notice ;) -- -Lawrence -Student ID 1028219 -CCNA
Re: is there [EMAIL PROTECTED] archive?
> > The dmesgs submitted to [EMAIL PROTECTED] are not publicly accessible. > > At some point in time this was discussed, but we can't do that, since > > we never told people that they would be published. So they remain > > accessible to developers only. They are consulted very often, so keep > > them coming in! > > > On this topic - should we resubmit dmesgs periodically as the machines > are updated to newer versions? DEFINATELY! There are developers in the group who read the dmesg output very closely, looking for issues that you, as a user, would not even notice ;)
Re: is there [EMAIL PROTECTED] archive?
> The dmesgs submitted to [EMAIL PROTECTED] are not publicly accessible. > At some point in time this was discussed, but we can't do that, since > we never told people that they would be published. So they remain > accessible to developers only. They are consulted very often, so keep > them coming in! > On this topic - should we resubmit dmesgs periodically as the machines are updated to newer versions? I've machines that date back to 2.7. The machines have been updated as new releases come out, but I haven't resubmitted dmesgs since they were built. -- Craig
Re: Serial console not working for IBM Aptiva
Thanks for the response, Nick, I'm almost there and just one further query:
On 18/02/07, Nick Holland <[EMAIL PROTECTED]> wrote:
[snip]
Been there, seen that, kicked self in butt once I realized the port I
was trying to use as a serial terminal was also configured as a serial
console for the terminal machine (D'oh!). Basically, the two machines
start bouncing back and forth off each other, and nothing works as
desired.
Yup, makes perfect sense now.
[snip]
> pccom1: irq 3 already in use
> pccom2: irq 5 already in use
And that tells us more.
Note that it's pccom1 and pccom2, but no pccom0... pccom devices are
ISA devices, so they can't share IRQs with PCI devices (though, of
course, PCI devices can share IRQs with each other). So that's why
com0 didn't work, and the IRQ conflict won't make com1 and com2 work
very well...
Take a close look at your machine config (most likely, the BIOS setup
program), you will probably find non-standard configs for the two
serial ports. You might need to "tag" IRQ 3 as "Reserved for ISA" or
similar.
The Aptiva has an anaemic BIOS program, but by disabling one of the
two serial interfaces I now appear to have eliminated IRQ conflicts
and acquired a working serial console - BUT I lose nearly all of the
dmesg(8) and init(8) output at boot, with it being directed to the
screen instead. I also note that boot(8) tells me I have com0 and no
com1 (which is expected since I disabled it in the BIOS) whereas
dmesg(8) tells me I have pccom1 but no pccom0 and this seems a little
strange to me.
To clarify, boot(8) tells me I have com0 available at boot. So in
/etc/boot.conf I tell it:
set tty com0
and it switches to the console but all that is output to the console is:
OpenBSD/i386 BOOT 2.10
boot>
booting hd0a:/bsd: 4966344+867848 [52+255872+237161]=0x608d64
entry point at 0x100120
That's it for the output seen on the terminal, at this point the
dmesg(8) and init(8) output is directed to the screen. Then when
getty(8) is executed interactivity for _both_ the keyboard and the
serial console are restored.
Any further thoughts will be appreciated; dmesg(8) and ttys(5) are
included below:
# dmesg
OpenBSD 3.9 (GENERIC) #617: Thu Mar 2 02:26:48 MST 2006
[EMAIL PROTECTED]:/usr/src/sys/arch/i386/compile/GENERIC
cpu0: Intel Pentium III ("GenuineIntel" 686-class, 512KB L2 cache) 549 MHz
cpu0: FPU,V86,DE,PSE,TSC,MSR,PAE,MCE,CX8,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,SER,MMX
,FXSR,SSE
cpu0: disabling processor serial number
real mem = 66625536 (65064K)
avail mem = 53170176 (51924K)
using 838 buffers containing 3432448 bytes (3352K) of memory
mainbus0 (root)
bios0 at mainbus0: AT/286+(ff) BIOS, date 06/30/99, BIOS32 rev. 0 @ 0xf0210
pcibios0 at bios0: rev 2.1 @ 0xf0200/0xb00
pcibios0: PCI IRQ Routing Table rev 1.0 @ 0xf9e00/128 (6 entries)
pcibios0: PCI Interrupt Router at 000:07:0 ("Intel 82371AB PIIX4 ISA" rev 0x00)
pcibios0: PCI bus #1 is the last bus
bios0: ROM list: 0xc/0xa000
cpu0 at mainbus0
pci0 at mainbus0 bus 0: configuration mode 1 (no bios)
pchb0 at pci0 dev 0 function 0 "Intel 82443BX AGP" rev 0x03
ppb0 at pci0 dev 1 function 0 "Intel 82443BX AGP" rev 0x03
pci1 at ppb0 bus 1
vga1 at pci1 dev 0 function 0 "S3 Savage 4" rev 0x02
wsdisplay0 at vga1 mux 1: console (80x25, vt100 emulation)
wsdisplay0: screen 1-5 added (80x25, vt100 emulation)
pcib0 at pci0 dev 7 function 0 "Intel 82371AB PIIX4 ISA" rev 0x02
pciide0 at pci0 dev 7 function 1 "Intel 82371AB IDE" rev 0x01: DMA, channel 0 wi
red to compatibility, channel 1 wired to compatibility
wd0 at pciide0 channel 0 drive 0:
wd0: 16-sector PIO, LBA, 9770MB, 20010816 sectors
wd0(pciide0:0:0): using PIO mode 4, Ultra-DMA mode 2
atapiscsi0 at pciide0 channel 1 drive 0
scsibus0 at atapiscsi0: 2 targets
cd0 at scsibus0 targ 0 lun 0: SCSI0 5/cdrom removabl
e
cd0(pciide0:1:0): using PIO mode 4, DMA mode 2
uhci0 at pci0 dev 7 function 2 "Intel 82371AB USB" rev 0x01: irq 11
usb0 at uhci0: USB revision 1.0
uhub0 at usb0
uhub0: Intel UHCI root hub, rev 1.00/1.00, addr 1
uhub0: 2 ports with 2 removable, self powered
piixpm0 at pci0 dev 7 function 3 "Intel 82371AB Power" rev 0x02: polling
iic0 at piixpm0
eso0 at pci0 dev 11 function 0 "ESS SOLO-1 AudioDrive" rev 0x01: ES1946, irq 10
eso0: mapping Audio 1 DMA using VC I/O space at 0x8cc0
audio0 at eso0
opl0 at eso0: model OPL3
midi0 at opl0:
"Conexant 56k Winmodem" rev 0x08 at pci0 dev 13 function 0 not configured
sis0 at pci0 dev 14 function 0 "NS DP83815 10/100" rev 0x00, DP83815C: irq 10, a
ddress 00:a0:cc:74:48:46
nsphyter0 at sis0 phy 0: DP83815 10/100 PHY, rev. 1
rl0 at pci0 dev 16 function 0 "Accton MPX 5030/5038" rev 0x10: irq 9, address 00
:10:b5:08:5c:32
rlphy0 at rl0 phy 0: RTL internal PHY
isa0 at pcib0
isadma0 at isa0
pckbc0 at isa0 port 0x60/5
pckbd0 at pckbc0 (kbd slot)
pckbc0: using irq 1 for kbd slot
wskbd0 at pckbd0: console keyboard, using wsdisplay0
pcppi0 at isa0 port 0x61
midi1 at pcppi0:
spkr0 at pcppi0
npx0 at isa0 port 0xf0/16: using exception 16
pcc
Re: perl modules/chroot apache
Julien TOUCHE wrote: i know about mod_perl and i was considering it to replace perl in chroot. but as far as i know, it does not replace perl modules ... or i miss it ? You can load the modules on apache startup. They don't have to exist in chroot. http://www.apacheref.com/ref/mod_perl/PerlRequire.html Marc
Re: Supported GPS receivers
On 2/18/07, Phusion <[EMAIL PROTECTED]> wrote: What are the best supported GPS receivers under 4.0 release and stable code-base? I would like to find a GPS receiver (serial or USB) that is supported under OpenBSD that will work for OpenNTPD or NTP. As a time source, any receiver that outputs NMEA should work. If you're planning on using your gps for timing and positioning (ie. for kismet) then you probably want to consider using one of the many receiver types supported by gpsd. Personally, I use a uBlox ANTARIS TIM-LP (sold as the San Jose FV-25). There are many other sexy GPS receivers out there that would also work. USB-serial chips - either as a dongle with a DB9 on it, or embedded into GPS receivers - tend to not have control lines connected, so you can't get accurate pulse-per-second timing. For that you need a real serial port. If you don't have control lines then you need to wait for the navigation solution to come out, and that can take 250 - 300mS. From there, the nmea(4) line discipline waits for the GPRMC sentence, but often GPGGA comes out first. More delay, but it's not predictable, because the delay depends on the transmission time of variable length sentences. The soft timestamping uses the first '$' it sees. Finally, some receivers output GPGGA, GPGLL, GPGSA, GPGSV, GPRMC - in that order. GPGSV is extremely variable, both in length and number of sentences transmitted. During seconds which GPGSV is supposed to output, GPRMC can be delayed up to 750mS after the start of the second, which is why the soft timestamp just goes with the first '$' each second. I added support for the nmea(4) line discipline and line timestamping to gpsd-2.34; this version is in the current ports tree. On my white-box pentium 4, once the kernel has calibrated the local oscillator, openntpd using my gps as the time source claims that it needs to do about 2uS of adjtime every 90 minutes or so. CK -- GDB has a 'break' feature; why doesn't it have 'fix' too?
Supported GPS receivers
What are the best supported GPS receivers under 4.0 release and stable code-base? I would like to find a GPS receiver (serial or USB) that is supported under OpenBSD that will work for OpenNTPD or NTP. Phusion
Re: is there [EMAIL PROTECTED] archive?
[EMAIL PROTECTED] wrote: Theo de Raadt wrote: And they will NEVER receive as many dmesg's as we do, because our we promise privacy... There's no middle ground. Our developers never look at those, so they do not serve our purposes. We don't want to distract our users from sending in reports which actually serve our developers, and thus make better code. Sorry, but I do not agree with your direction on this. VERY OFTEN we do not document something because it hurts other things. i follow your point on privacy. but not the latter part (isn't this exactly what you/we are concerned about from vendors - lack of disclosure of useful information, due to "hurting commercial interests"? more to the point, i wish to buy "stuff that works well" under openbsd, & clearly this is linked to the policies of vendors on disclosure, and developer interest. now if i can only guess at equipment that was fully documented, & followed up with support for the developers, then how can i allow my $$ to reward vendors with those policies, that help you write better code that we enjoy using? if i follow you correctly - #1 there's a strict requirement for privacy for dmesg@ - fair enough. i've sent a line to nick@ to clarify this in the FAQ as i've not seen it mentioned elsewhere. #2 you don't want to suggest to people that they can _also_ send a dmesg somewhere else as this might confuse them, & you won't get bug reports. i would hope somebody able to file the bug reports you ask for is able to manage CC'ing a dmesg to somewhere else as well without getting lost on the way. if I can't easily identify "stuff that works well" then I end up spending 500$ on a mobo destined for dust-ware, instead of on your admittedly great OS. & since my first CD was a 2.8 one, i'm loath to waste my money on junk kit that doesn't run well on openbsd. the http://www.openbsd.org/i386.html page doesn't include any info on motherboards, although it does cover the peripheral side pretty well. i'd be happy enough with a few notes on mobos that work here - but where would i as a user go to get this information? if you think it's our business to send in updates for this page, i'll do so - but i think i am not a good person to tell when a mobo is fully supported, or if the h/w vendor was helpful. if you stick to refusing #2, well, as always, it's your OS, & your call as developers. but i think that you are reducing your support base if there is no reference point for "stuff that works well". a+ scorch
Re: is there [EMAIL PROTECTED] archive?
Theo de Raadt wrote: i'm looking for new mobos (both embedded & "normal") & wondered if there's any way to search through sumbitted [EMAIL PROTECTED] (you do all send in your dmesg don't you?) to see what people ran into previously. gmane & marc have proved reasonably light on Sorry, but I have thought about this and do not plan to change our policy . not asking you to :-) it's a good reason. i'll send something to Nick for the FAQ. the only thing i found was http://www.nycbug.org/?NAV=dmesgd;f_bsd=OpenBSD which was not extensive enough - but a nice interface! And they will NEVER receive as many dmesg's as we do, because our we promise privacy... There's no middle ground. i & many others would be happy to have a public dmesg. if nycbug have a mail-accessible interface, & are happy to share, could we add a note to either afterboot or FAQ to point people there to upload one for reference? i don't see why there shouldn't be a choice for something so useful. a+ scorch out of the frying pan & into the fire
Re: Wanted: RALink based (2501/2600) pcmcia-Card with the possibility to add an external Antenna
On Feb 18, 2007, at 11:01 AM, mniche-news wrote: Hi Sebastian, Which brand of Ral card that you want? I need to implement hostap mode for 802.11 b/g, do you have any recommanded PCI card that works better with hostap? Should you want to stay with pcmcia-Card form factor from RALink 2501/2600 based, do you know any specific brand? I would prefer PCI if possible but if none then I can use PCMCIA cards too. I plan to buy 40 pieces and I am in Taiwan now for this project. Should you be able to identify any robust ral card (PCI or PCMCIA) then we can buy together too. http://www.newegg.com/Product/Product.asp?Item=N82E16833315041 This is what I bought for my obsd router: ral0 at pci0 dev 9 function 0 "Ralink RT2561S" rev 0x00: irq 5, address 00:00:00:00:00:00 ral0: MAC/BBP RT2561C, RF RT2527 Works like a charm. Shutdown, install and turn on PC, issue ifconfig options for hostap and bam wireless internet (oh and add PF rules too).
Re: OpenBSD speed on desktops
> > Oh really, "it has been stated". By who? Overall, I doubt that all > > of our security technologies add more than about 2% of a performance > > hit. Even a 'make build' on most architectures did not add that. I > > think you need to go back and read my slides again. Spreading lies > > about 5-10% performance hits is just not kind to our efforts. > > I've reread the slides again. I stand corrected when it comes to w^x & > propolice, but I'm still not in the clear when it comes to randomized > malloc & mmap. The slides from bsdcan 2004 state: "still failry > expensive", the slides from opencon 2005 no longer mention anything > about performance. Well, we never measured it again. Because we didn't feel any slowdown or feel any effect. Otto did speed something up a few weeks ago, but these are totally minor effects, honestly. But since we didn't bother measuring it, we should probably all assume a 10% slowdown. That's easier. It explains everything, including spring coming earlier every year. In the future, if you don't measure it yourself, please just withhold comment.
Re: Wanted: RALink based (2501/2600) pcmcia-Card with the possibility to add an external Antenna
Hi Sebastian, Which brand of Ral card that you want? I need to implement hostap mode for 802.11 b/g, do you have any recommanded PCI card that works better with hostap? Should you want to stay with pcmcia-Card form factor from RALink 2501/2600 based, do you know any specific brand? I would prefer PCI if possible but if none then I can use PCMCIA cards too. I plan to buy 40 pieces and I am in Taiwan now for this project. Should you be able to identify any robust ral card (PCI or PCMCIA) then we can buy together too. Thanks, Kevin > -Original Message- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] > On Behalf Of Sebastian Rother > Sent: Thursday, January 11, 2007 9:06 PM > To: misc@openbsd.org > Subject: Wanted: RALink based (2501/2600) pcmcia-Card with > the possibility to add an external Antenna > > Hello everybody, > > I`m looking for a RALink 2501/2600 based pcmcia-Card with the > possibility to connect a Yagi-Antenna. Unfortunaly I wasn`t able to > find anything after a hour of googling. > It would be good if the card would at leats provide 100mW. > > The 2501-Chipset would be nice because of the "a" possibility > but if somebody knows a 2500 based Card wich matchs my > specifications I > would be happy either. > > Once somebody said "Support Asia"... > Well I try... but it seams In Germany or Europe nobody ever > sold such a > pcmica-Card. > > If somebody knows where to get such Cards please do let me know! :) > > > Kind regards, > Sebastian
Re: OpenBSD speed on desktops
On 2/17/07, R. Fumione <[EMAIL PROTECTED]> wrote: Hello, I am using OpenBSD on server since few years now, and I am very happy with it's easy maintenance and it's stability. I want to try on desktop, and I am having trouble. Everything is much slower than existing Linux system. For example, Firefox takes 3-5 seconds to start on Linux but ~10 seconds on OpenBSD on same machine! Hi: Slowdowns for large applications for Firefox and Gimp can beaccompanied by the following warning: Gdk-WARNING**: shmget failed: error 28 (no space left on device) This can fixed by setting sysctl kern.shminfo.shmseg=128 sysctl kern.shminfo.shmall=32768 In /etc/sysctl.conf See: Re: dillo - Gdk-ERROR ? http://monkey.org/openbsd/archive/ports/0309/msg00164.html Another cause of the slowdowns mayy be that /etc/login.conf class "default" does not not allow enough files to be open at the same time. in /etc/login.conf change: :openfiles-cur=64:\ to :openfiles-cur=256:\ I have used OpenBSD as a desktop for several years and the slowdowns are not caused by a defect in the OS. In fact I use 4.0 on an ancient Pentium I with 96 MB of ram and it's load speed is satisfactory. -- Kind regards, Jonathan
Re: perl webapp and threads
for the archives, a small guide to thread perl in chroot + modules get latest perl use a non-root account with rights on chosen destination dir (to avoid any conflict with system perl) $ cd perl-x.x.x $ rm -f config.sh Policy.sh $ sh Configure -Dinstallprefix='/var/www/perl5web' -Dusethreads -Duseperlio -Duselargefiles -Dopenbsd_distribution=define -des $ make $ make test # can ignore (1 test failed for me) $ make install $ export PERL5LIB=/var/www/perl5web/lib/perl5/5.8.8/OpenBSD.i386-openbsd-thread-multi:/var/www/perl5web/lib/perl5/5.8.8 $ /var/www/perl5web/bin/perl -V|more# to check all is ok next edit /var/www/perl5web/lib/perl5/5.8.8/OpenBSD.i386-openbsd-thread-multi/Config.pm and check all paths (replace system one with ones you choose) ex: /usr/lib -> /var/www/perl5web/lib /usr/local -> /var/www/usr/perl5web # for modules last, you can compile your modules. set PERL5LIB like above and use the web compiled perl. ex: $ export PERL5LIB=/var/www/perl5web/lib/perl5/5.8.8/OpenBSD.i386-openbsd-thread-multi:/var/www/pe rl5web/lib/perl5/5.8.8 $ /var/www/perl5web/bin/perl Makefile.PL PREFIX=/var/www/usr/perl5web $ make $ make install Regards Julien note: to keep your security level good, you can systrace apache, use mod_security and restrict perl in apache config.
Re: OpenBSD speed on desktops
On Sun, 18 Feb 2007 10:03:37 -0700 Theo de Raadt <[EMAIL PROTECTED]> wrote: > Oh really, "it has been stated". By who? Overall, I doubt that all > of our security technologies add more than about 2% of a performance > hit. Even a 'make build' on most architectures did not add that. I > think you need to go back and read my slides again. Spreading lies > about 5-10% performance hits is just not kind to our efforts. I've reread the slides again. I stand corrected when it comes to w^x & propolice, but I'm still not in the clear when it comes to randomized malloc & mmap. The slides from bsdcan 2004 state: "still failry expensive", the slides from opencon 2005 no longer mention anything about performance. // nick
Re: OpenBSD speed on desktops
> On last thing that might add to openbsd's startup overhead is the > aggresive security stance. I don't know if library randomization has > anything to do with it, but w^x & propolice have been stated to give a > 5% to 10% performance impact in certain cases. I've noticed this mostly > in applications that map & unmap a lot of memory. Oh really, "it has been stated". By who? Overall, I doubt that all of our security technologies add more than about 2% of a performance hit. Even a 'make build' on most architectures did not add that. I think you need to go back and read my slides again. Spreading lies about 5-10% performance hits is just not kind to our efforts.
Re: OpenBSD speed on desktops
Joachim Schipper wrote: > > Since you didn't mention what you are using at the moment, I can't very > well tell you to switch to a lighter window manager, can I? Ion *is* > nice, though... ;-) > > ion whips a "giraffe's ass with a belt from a balcony" [0]. [0] wesley willis ( http://en.wikipedia.org/wiki/Wesley_Willis ) cheers, jake > Joachim
Re: missing isakmpd.fifo
Matthew Closson wrote: > Hello, > > I am experiencing the same problem. I am testing it to see if I can > find what is causing it. I am running OpenBSD 4.0-stable and I went > to add a new tunnel today and was greeted with a message the > isakmpd.fifo did not exist. I have isakmpd enabled in /etc/rc.conf > with flags -K. Even though I do not specify a location on the command > line of isakmpd for the fifo to occur, it does exist in fact when the > process is launched and sometime later dies off. This is what I found > today: > > # echo ike esp from 172.31.33.0/24 to 10.9.9.0/24 peer aaa.bbb.ccc.ddd > psk "" | ipsecctl -f - > ipsecctl: ike_ipsec_establish: open(/var/run/isakmpd.fifo): No such > file or directory > > Where as before the exact same command has worked fine. > hmmm. one of my 4.0-release hosts running isakmpd had something similar happen a few days ago: it was using isakmpd -K + ipsecctl -f /etc/ipsec.conf and then, AFAICT, isakmpd died and i got the same isakmpd.fifo error. once i restarted isakmpd and issued the ipsecctl -f /etc/ipsec.conf, all was restored. cheers, jake > Thanks, > > -Matt-
Re: USB nics to give away
On Sun, Feb 18, 2007 at 02:40:43PM +0100, Paul de Weerd wrote: > On Sun, Feb 18, 2007 at 11:21:14PM +1100, Jonathan Gray wrote: > | On Sun, Feb 18, 2007 at 10:14:07AM +0100, Paul de Weerd wrote: > | > Hi all, > | > > | > I have two USB nics (one wired, one wireless) to give away. They were > | > cheap, so I bought them to see if they were supported. Since they're > | > not, maybe some developer can use them. I know nothing about > | > documentation for these things, but this is what dmesg and usbdevs -v > | > say : > | > | Ah, but perhaps it is easy to add support. > | > | > > | > # dmesg # for the wireless part > | > ugen0 at uhub2 port 3 > | > ugen0: Tenda.. 54M USB Wireless NIC, rev 2.00/0.01, addr 2 > | > > | > # usbdevs -v # for the wireless part > | > port 3 addr 2: high speed, self powered, config 1, 54M USB Wireless > NIC(0x1fab), Tenda..(0x1286), rev 0.0 > | > 1 > | > | Need more information on this. > | It does not appear to be the Tenda TWL542U, it does not appear in the list > | of ids in the windows driver. > > Stoopid me, I did not include the modelnumber found on the thing > itself. This is a Tenda TWL541U. Windows driver online at > http://www.tenda.com.cn/Wireless/en/TWL541U.rar but it did not give me > many clues (no .sys or .inf files, only .dll, .tlb, .rgs and .exe). In this case: Extract the rar file Extract the installshield archive (I use a small program I based on the internals of http://www.synce.org/index.php/Orange) Extract the installshield cabinet files with unshield from ports .inf contains things like: %MRVL833810.DeviceDesc% = W833810USB.ndi.NT,USB\VID_1286&PID_1FAB MRVL833810.DeviceDesc = "Tenda TWL541U Wireless LAN Client Adapter - USB" So it is a Marvell device, but Marvell are so misguided as to not even list their own products on their webpage. But all it not lost, want.html contains: "Marvell based SDIO Wireless cards and Marvell based USB adaptors, for example the Xbox360 wireless adaptor needed in Switzerland. Contact [EMAIL PROTECTED] or [EMAIL PROTECTED]" > | Google leads me to believe this is a udav(4) device. > | Please try the following diff: > > Your diff works great : great, comitted to -current, slightly different manpage tidbit. > Thanks for your help, the device is still available if you (or another > developer) wants it. I already have a udav device but they aren't terribly common so if another developer wants to take it to play with great. People shouldn't expect terribly high speeds out of these things though as they aren't USB2 devices.
Re: OpenOSPFd and kernel routing table
And the output of the forwarding database with the unexpected nexthop of the local interface instead of the opsf-neighbor: # ospfctl sh fib flags: * = valid, O = OSPF, C = Connected, S = Static Flags Destination Nexthop C 195.140.212.0/24 link#22 I killed the ospfd-process and started it again, and the fib looks better: # ospfctl sh fib flags: * = valid, O = OSPF, C = Connected, S = Static Flags Destination Nexthop *O 195.140.212.0/24 10.0.0.2 Hm, killing and restarting ospfd isn't the way I want to do it everytime the state of the carp-interface changes :-) Falk
OpenOSPFd and kernel routing table
Hello,
I just set up OSPF to talk within two OpenBSD-Boxes (Pinky and Brain, do
you remember? :-). On both machines I configured a carp interface to
provide a default gateway for my local VLANs. The configuration was very
easy and intuitive. OSPF is redistributing routes only for
carp-interfaces which are in master-state, this is exact the way I was
looking for.
But if I take down the actual carp-master interface (195.140.212.1/24),
the (until now) backup interface take over the master-state, ospfd is
redistributing the changed routing information, I can see it in the
output of "ospfctl sh rib".
But if I take a look at "opsfctl sh fib" there is still the local link
and not the new route via the ospf-neighbor, as I expected.
"fib-update" should be on by default, but I tried to write it explicitly
into the configuration file, but it's not getting better.
This is my configuration on both sides (only the router-id is adapted):
# global configuration
router-id 194.9.86.1
fib-update yes
# route redistribution
redistribute connected
redistribute static
# areas
area 0.0.0.0 {
interface lo1
interface em1 {
metric 10
auth-type crypt
auth-md-keyid 1
auth-md 1"foobar"
}
}
The correct output of the routing information:
# ospfctl sh rib
Destination Nexthop Path TypeType Cost
Uptime
194.9.86.2 10.0.0.2 Intra-Area Router10
00:22:17
10.0.0.0/30 10.0.0.1 Intra-Area Network 10
00:22:24
194.9.86.2/3210.0.0.2 Intra-Area Network 10
00:22:17
195.140.212.0/24 10.0.0.2 Type 1 ext Network 110
00:21:45
195.140.212.3/32 10.0.0.2 Type 1 ext Network 110
00:22:17
195.140.213.3/32 10.0.0.2 Type 1 ext Network 110
00:22:17
And the output of the forwarding database with the unexpected nexthop of
the local interface instead of the opsf-neighbor:
# ospfctl sh fib
flags: * = valid, O = OSPF, C = Connected, S = Static
Flags Destination Nexthop
*C 10.0.0.0/30 link#2
*S 127.0.0.0/8 127.0.0.1
*C 127.0.0.1/8 link#0
* 127.0.0.1/32 127.0.0.1
*C 192.168.13.0/24 link#1
* 194.9.86.1/32194.9.86.1
*O 194.9.86.2/3210.0.0.2
C 195.140.212.0/24 link#22
*C 195.140.212.2/32 link#16
*O 195.140.212.3/32 10.0.0.2
*C 195.140.213.0/24 link#19
*C 195.140.213.2/32 link#17
*O 195.140.213.3/32 10.0.0.2
*C 195.225.134.4/30 link#3
*S 224.0.0.0/4 127.0.0.1
Any hints?
Thanks,
Falk
Re: USB nics to give away
On Sun, Feb 18, 2007 at 11:21:14PM +1100, Jonathan Gray wrote: | On Sun, Feb 18, 2007 at 10:14:07AM +0100, Paul de Weerd wrote: | > Hi all, | > | > I have two USB nics (one wired, one wireless) to give away. They were | > cheap, so I bought them to see if they were supported. Since they're | > not, maybe some developer can use them. I know nothing about | > documentation for these things, but this is what dmesg and usbdevs -v | > say : | | Ah, but perhaps it is easy to add support. | | > | > # dmesg # for the wireless part | > ugen0 at uhub2 port 3 | > ugen0: Tenda.. 54M USB Wireless NIC, rev 2.00/0.01, addr 2 | > | > # usbdevs -v # for the wireless part | > port 3 addr 2: high speed, self powered, config 1, 54M USB Wireless NIC(0x1fab), Tenda..(0x1286), rev 0.0 | > 1 | | Need more information on this. | It does not appear to be the Tenda TWL542U, it does not appear in the list | of ids in the windows driver. Stoopid me, I did not include the modelnumber found on the thing itself. This is a Tenda TWL541U. Windows driver online at http://www.tenda.com.cn/Wireless/en/TWL541U.rar but it did not give me many clues (no .sys or .inf files, only .dll, .tlb, .rgs and .exe). | A few ways to figure out what it is: | * Extract the .sys and .inf files from the windows driver, run strings/read for hints | * Open up device and look Nothing interesting. The radiopart is shielded with a little metal casing, I did not pry this open. | * Put FCC id into https://gullfoss2.fcc.gov/prod/oet/cf/eas/reports/GenericSearch.cfm | look at internal photos for chip No FCC id in or on this thing. | > | > # dmesg # for the wired part | > ugen0 at uhub0 port 2 | > ugen0: ShanTou ST268 USB NIC, rev 1.10/1.01, addr 2 | > | > # usbdevs -v # for the wired part | > port 2 addr 2: full speed, power 144 mA, config 1, ST268 USB NIC(0x0268), ShanTou(0x0a46), rev 1.01 | | Google leads me to believe this is a udav(4) device. | Please try the following diff: Your diff works great : udav0 at uhub0 port 1 udav0: ShanTou ST268 USB NIC, rev 1.10/1.01, addr 2 address 00:60:6e:30:e5:4d amphy0 at udav0 phy 0: DM9601 10/100 PHY, rev. 0 I've not done very extensive testing yet, but I can send and receive packets on this interface just fine. Here's the missing manpage bit : Index: udav.4 === RCS file: /cvs/src/share/man/man4/udav.4,v retrieving revision 1.8 diff -u -r1.8 udav.4 --- udav.4 19 Jan 2007 08:40:53 - 1.8 +++ udav.4 18 Feb 2007 13:39:22 - @@ -51,6 +51,7 @@ .Bl -tag -width Dv -offset indent -compact .It Tn Corega FEther USB-TXC .It Tn HenTong WK-668 +.It Tn ShanTou ST268 USB NIC .El .Pp For more information on configuring this device, see Thanks for your help, the device is still available if you (or another developer) wants it. Cheers, Paul 'WEiRD' de Weerd -- >[<++>-]<+++.>+++[<-->-]<.>+++[<+ +++>-]<.>++[<>-]<+.--.[-] http://www.weirdnet.nl/
Re: USB nics to give away
On Sun, Feb 18, 2007 at 10:14:07AM +0100, Paul de Weerd wrote:
> Hi all,
>
> I have two USB nics (one wired, one wireless) to give away. They were
> cheap, so I bought them to see if they were supported. Since they're
> not, maybe some developer can use them. I know nothing about
> documentation for these things, but this is what dmesg and usbdevs -v
> say :
Ah, but perhaps it is easy to add support.
>
> # dmesg # for the wireless part
> ugen0 at uhub2 port 3
> ugen0: Tenda.. 54M USB Wireless NIC, rev 2.00/0.01, addr 2
>
> # usbdevs -v # for the wireless part
> port 3 addr 2: high speed, self powered, config 1, 54M USB Wireless
> NIC(0x1fab), Tenda..(0x1286), rev 0.0
> 1
Need more information on this.
It does not appear to be the Tenda TWL542U, it does not appear in the list
of ids in the windows driver.
A few ways to figure out what it is:
* Extract the .sys and .inf files from the windows driver, run strings/read for
hints
* Open up device and look
* Put FCC id into
https://gullfoss2.fcc.gov/prod/oet/cf/eas/reports/GenericSearch.cfm
look at internal photos for chip
>
> # dmesg # for the wired part
> ugen0 at uhub0 port 2
> ugen0: ShanTou ST268 USB NIC, rev 1.10/1.01, addr 2
>
> # usbdevs -v # for the wired part
> port 2 addr 2: full speed, power 144 mA, config 1, ST268 USB NIC(0x0268),
> ShanTou(0x0a46), rev 1.01
Google leads me to believe this is a udav(4) device.
Please try the following diff:
You'll have to run 'make' in sys/dev/usb/ after applying it,
then build a new kernel.
Index: sys/dev/usb/usbdevs
===
RCS file: /cvs/src/sys/dev/usb/usbdevs,v
retrieving revision 1.262
diff -u -p -r1.262 usbdevs
--- sys/dev/usb/usbdevs 17 Feb 2007 02:17:18 - 1.262
+++ sys/dev/usb/usbdevs 18 Feb 2007 12:06:40 -
@@ -366,6 +366,7 @@ vendor CSR 0x0a12 Cambridge Silicon Rad
vendor TREK0x0a16 Trek Technology
vendor ASAHIOPTICAL0x0a17 Asahi Optical
vendor BOCASYSTEMS 0x0a43 Boca Systems
+vendor SHANTOU 0x0a46 ShanTou
vendor BROADCOM0x0a5c Broadcom
vendor GREENHOUSE 0x0a6b GREENHOUSE
vendor GEOCAST 0x0a79 Geocast Network Systems
@@ -1970,6 +1971,9 @@ product SERVERWORKS HUB 0x Root Hub
/* SGI products */
product SGI SN1_L1_SC 0x1234 SN1 L1 System Controller
+
+/* ShanTou products */
+product SHANTOU ST268 0x0268 ST268
/* Shark products */
product SHARK PA 0x0400 Pocket Adapter
Index: sys/dev/usb/if_udav.c
===
RCS file: /cvs/src/sys/dev/usb/if_udav.c,v
retrieving revision 1.19
diff -u -p -r1.19 if_udav.c
--- sys/dev/usb/if_udav.c 29 Sep 2006 08:43:07 - 1.19
+++ sys/dev/usb/if_udav.c 18 Feb 2007 12:06:42 -
@@ -148,7 +148,8 @@ static const struct udav_type {
} udav_devs [] = {
{{ USB_VENDOR_COREGA, USB_PRODUCT_COREGA_FETHER_USB_TXC }, 0 },
{{ USB_VENDOR_DAVICOM, USB_PRODUCT_DAVICOM_DM9601 }, 0 },
- {{ USB_VENDOR_DAVICOM, USB_PRODUCT_DAVICOM_WK668 }, 0 }
+ {{ USB_VENDOR_DAVICOM, USB_PRODUCT_DAVICOM_WK668 }, 0 },
+ {{ USB_VENDOR_SHANTOU, USB_PRODUCT_SHANTOU_ST268 }, 0 }
};
#define udav_lookup(v, p) ((struct udav_type *)usb_lookup(udav_devs, v, p))
Re: OpenBSD speed on desktops
On Sat, 17 Feb 2007 22:06:43 +0100 Joachim Schipper <[EMAIL PROTECTED]> wrote: Since prebind has already been explained in detail, I want to add that does indeed work, but if you use it on your ports it will invalidate all of the hashes used by pkg_add (which is most likely one of the issues theo mentioned). With prebinding my firefox starts in 4 seconds or so, half of what it needs without prebinding. > Another aspect is that Linux is much more aggressive in caching data > from disk; if the amount of data read, the amount of work done in > between, and the amount of RAM is such that Linux can get most data > from its memory cache while OpenBSD has to read most of it from disk, > Linux will be a *lot* faster. Of course, you would only see this > effect if you started Firefox twice without doing much in between. We're all hoping for UBC to come back in a working form, but hopefully some are doing the actual work :) If your box has memory to spare it will infact load firefox a lot faster the second time, if it still has the libraries cached in memory. A fixed size of memory is reserved for filesystem caching. What linux does (and UBC) is remove this fixed limit and let you use all your memory for buffer cache when it's not mapped to another application. > Both of those could explain why FF loads slower. If either of those is > the big culprit, though, FF should run just as fast (slow) as it ever > did, and since you're not likely to start it that often, I'd be > inclined to say it isn't that big an issue. On last thing that might add to openbsd's startup overhead is the aggresive security stance. I don't know if library randomization has anything to do with it, but w^x & propolice have been stated to give a 5% to 10% performance impact in certain cases. I've noticed this mostly in applications that map & unmap a lot of memory. I'm using openbsd on my systems, desktops & laptops included, since release 2.7. It might not be equal to a current linux kernel performance wise, but it's not lagging that much behind. I'll take the cleanness, easy of use & stability any day over a 10% performance difference. And that's not even going into the free code debate, it's hard to get more free than openbsd. // nick
USB nics to give away
Hi all, I have two USB nics (one wired, one wireless) to give away. They were cheap, so I bought them to see if they were supported. Since they're not, maybe some developer can use them. I know nothing about documentation for these things, but this is what dmesg and usbdevs -v say : # dmesg # for the wireless part ugen0 at uhub2 port 3 ugen0: Tenda.. 54M USB Wireless NIC, rev 2.00/0.01, addr 2 # usbdevs -v # for the wireless part port 3 addr 2: high speed, self powered, config 1, 54M USB Wireless NIC(0x1fab), Tenda..(0x1286), rev 0.0 1 # dmesg # for the wired part ugen0 at uhub0 port 2 ugen0: ShanTou ST268 USB NIC, rev 1.10/1.01, addr 2 # usbdevs -v # for the wired part port 2 addr 2: full speed, power 144 mA, config 1, ST268 USB NIC(0x0268), ShanTou(0x0a46), rev 1.01 If you want either or both, let me know. They're small and lightweight and I can ship them worldwide. Cheers, Paul 'WEiRD' de Weerd -- >[<++>-]<+++.>+++[<-->-]<.>+++[<+ +++>-]<.>++[<>-]<+.--.[-] http://www.weirdnet.nl/
