Re: After 7.3->8.1, moused on serial mouse draws 100% CPU
Hi Michael,
* Michael Sperber wrote:
> select(1024,{2},0x0,0x0,{0.24 }) = 1 (0x1)
> read(2,0xbfbfe4ed,1) ERR#35 'Resource temporarily
> unavailable'
Thanks for the debug info. Sorry for not asking this earlier, but can
you also give me the output of `pstat -t` for cuau0?
It's a bit strange the read() on the TTY returns EAGAIN, since this
doesn't seem to be generated by the TTY layer itself. It also seems
moused is pretty badly written, looking at how it operates on fd_sets.
--
Ed Schouten
WWW: http://80386.nl/
pgp12YUqDgUNH.pgp
Description: PGP signature
Re: Atheros AR2427 in FreeBSD 8.1
Oh! Thank you! I already lost hope that I can properly run its Wi-Fi card. Regards, Roman. Hi! Right, I'm now ready to start looking at the AR2427. I can't guarantee i'll get it -stable- (as that relies mostly on me getting the AR9280/AR9285 stable; that'll take some time) but I'll at least try to fix the above issue. That issue you've just noted is because the EEPROM endian-ness has been incorrectly detected. (See the invalid channel messages just before the EEPROM starts dumping out "no power set for x/y"; then the invalid txtime messages?) I've seen that in my local tree. I have an AR2427 in this netbook of mine, so I'll figure out what's busting up the detection and update my development HAL. I'll let you guys know when that's done. ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org" ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: After 7.3->8.1, moused on serial mouse draws 100% CPU
Ed Schouten writes:
> * Michael Sperber wrote:
>> select(1024,{2},0x0,0x0,{0.24 }) = 1 (0x1)
>> read(2,0xbfbfe4ed,1) ERR#35 'Resource temporarily
>> unavailable'
>
> Thanks for the debug info. Sorry for not asking this earlier, but can
> you also give me the output of `pstat -t` for cuau0?
This?
ttyu0 179200 1562 20160 1790 217 0 0 ICOolZ
--
Regards,
Mike
___
freebsd-stable@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-stable
To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
if_epair as module
Anybody uses if_epair compiled as _module_ on any platform other than amd64? If yes, could you please respond to me in private? Big thanks in advance. -- Andriy Gapon ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
SuperMicro i7 (UP) - very slow performance
I have a Supermicro with the C7X58 motherboard and an i7 930 cpu, and it is nowhere near the performance it should be. A buildworld just took 22.5 hours! br...@tahiti[~]>uname -a FreeBSD tahiti.bryce.net 8.1-STABLE FreeBSD 8.1-STABLE #0: Tue Sep 7 22:45:38 CDT 2010 r...@tahiti.bryce.net:/usr/obj/usr/src/sys/GENERIC amd64 I have disabled "Legacy USB Support" in the BIOS and that helped, but I'm not finding any other setting that are getting things where they need to be. I have tested the two system drives independently (currently a zfs mirror), so it is not likely to be an hdd issue. Here's the verbose dmesg boot details - http://www.bryce.net/files/dmesg.boot And, the IPMI ASL in case that is of any value - http://www.bryce.net/files/tahiti.asl Currently, I'm not running powerd, performance is not better with it running. r...@tahiti[/usr/src]#cat /boot/loader.conf ahci_load="YES" coretemp_load="YES" zfs_load="YES" vfs.root.mountfrom="zfs:system" #vfs.zfs.prefetch_disable=1 kern.maxfiles=16384 # async i/o aio_load="YES" # VirtualBox #vboxdrv_load="YES" # SMB #ichsmb_load="YES" #smb_load="YES" # Power Saving #kern.hz="100" # Disable APIC subsystem - no longer needed when disabling lapic below #hint.apic.0.disabled="1" # Disable local APIC (LAPIC) timer - for C3 state #hint.apic.0.clock="0" # Avoid 128 interrupts/sec per core, at cost of scheduling precision #hint.atrtc.0.clock="0" # Disable throttle control (and rely on EIST) hint.p4tcc.0.disabled="1" hint.acpi_throttle.0.disabled="1" Thanks in advance for your time! ::Bryce:: ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: SuperMicro i7 (UP) - very slow performance
On Sat, Sep 18, 2010 at 08:32:32AM -0500, Bryce Edwards wrote: > I have a Supermicro with the C7X58 motherboard and an i7 930 cpu, and > it is nowhere near the performance it should be. A buildworld just > took 22.5 hours! I use 5046A-XB with i7-930 as home workstation, running latest RELENG_8, and I do not have the problem you noted. My BIOS is v1.1, USB legacy is enabled. I did noted one issue with hw, built-in firewire controller generated too high interrupt rate, so I usually do not load firewire.ko unless needed. > > br...@tahiti[~]>uname -a > FreeBSD tahiti.bryce.net 8.1-STABLE FreeBSD 8.1-STABLE #0: Tue Sep 7 > 22:45:38 CDT 2010 > r...@tahiti.bryce.net:/usr/obj/usr/src/sys/GENERIC amd64 > > I have disabled "Legacy USB Support" in the BIOS and that helped, but > I'm not finding any other setting that are getting things where they > need to be. > > I have tested the two system drives independently (currently a zfs > mirror), so it is not likely to be an hdd issue. > > Here's the verbose dmesg boot details - http://www.bryce.net/files/dmesg.boot > > And, the IPMI ASL in case that is of any value - > http://www.bryce.net/files/tahiti.asl > > Currently, I'm not running powerd, performance is not better with it running. > > r...@tahiti[/usr/src]#cat /boot/loader.conf > ahci_load="YES" > coretemp_load="YES" > > zfs_load="YES" > vfs.root.mountfrom="zfs:system" > #vfs.zfs.prefetch_disable=1 > > kern.maxfiles=16384 > > # async i/o > aio_load="YES" > > # VirtualBox > #vboxdrv_load="YES" > > # SMB > #ichsmb_load="YES" > #smb_load="YES" > > # Power Saving > #kern.hz="100" > > # Disable APIC subsystem - no longer needed when disabling lapic below > #hint.apic.0.disabled="1" > > # Disable local APIC (LAPIC) timer - for C3 state > #hint.apic.0.clock="0" > > # Avoid 128 interrupts/sec per core, at cost of scheduling precision > #hint.atrtc.0.clock="0" > > # Disable throttle control (and rely on EIST) > hint.p4tcc.0.disabled="1" > hint.acpi_throttle.0.disabled="1" > > Thanks in advance for your time! > > ::Bryce:: > ___ > freebsd-stable@freebsd.org mailing list > http://lists.freebsd.org/mailman/listinfo/freebsd-stable > To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org" pgpSYvjBzVbps.pgp Description: PGP signature
Re: problem pkg_add kdehier4-1.0.6.tbz
> Von: Alberto Villa > Betreff: Re: problem pkg_add kdehier4-1.0.6.tbz > An: freebsd-stable@freebsd.org > CC: "rolle" > Datum: Samstag, 18. September, 2010 02:27 Uhr > On Friday 17 September 2010 19:02:51 > rolle wrote: > > Today I tried to update my notebook with pkg_upgrade > tool, written by > a > > German bsdforen.de member. I ran into problems > installing > "kdehier4-1.0.6" > > package. > > ports/UPDATING has the solution for this > -- > Alberto Villa, FreeBSD Committer > http://people.FreeBSD.org/~avilla > Everything worked like expected, thx for your help. ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send any mail to "freebsd-stable-unsubscr...@freebsd.org"
Re: Freebsd 8.1 + xorg + radeonhd hang
On 2010-Sep-15 22:29:46 +0200, Eivind E wrote: >That has crossed my mind aswell, the only thing which makes me doubt >it is that after updating X number of months ago (probably about a >year and a half), it started to work with no problems whatsoever. >Now, after upgrading again, it has stopped working. Can I suggest adding a "Virtual 1280 1024" (or similar) to the 'SubSection "Display"' of 'Section "Screen"'. I ran into some wierd problems with multiple screens and found that the virtual size isn't correctly initialised unless it's explicitly specified. Also, specifying EXA acceleration is fairly mandatory: The default XAA acceleration is broken. -- Peter Jeremy pgpuqhaZGDnMJ.pgp Description: PGP signature
