Re: OpenBSD 5.5 on mSATA SSD unit in PC Engines APU.1C - bad dir ino 2 at offset 0: mangled entry kernel panic
Does an install on usb or sdcard medium work for you? Am 11.06.2014 06:01 schrieb Chris Cappuccio ch...@nmedia.net: Chris Cappuccio [ch...@nmedia.net] wrote: Mattieu Baptiste [mattie...@gmail.com] wrote: Le 8 juin 2014 13:38, Nick Ryan n...@njryan.com a ??crit : I know it???s no consolation to you but using a Kingston 30 GB mSATA from amazon works perfectly. The APU is on the May bios and I???ve had no issues. Didn???t the PCEngines mSATA drive have problems in general? There???s a mention on here about issues with the a version - is that yours? http://pcengines.ch/msata16b.htm Theoritically, I should have the new firmware (that's what told my vendor). But it seems there are still problems with these. Thanks for the tip concerning the Kingston drive. I've been using the Sandisk X110 msata. Borat says great success! As soon as I open my mouth # tar xzpf base55.tgz ahci0: log page read failed, slot 31 was still active. ahci0: stopping the port, softreset slot 31 was still active. tar: Failed write to file ./usr/lib/libedit.so.5.1: Input/output error gzip: stdin: crc error It's actually the Plextor M5M seems to be ok, that's what is in the box I've been using more. Ironically the buggy SuperSpeed thing from PC Engines stock is also reliable for me, albiet slow. Another problem I noticed with the X110 msata was the drive not saving all data on reboot. Like 'reboot' and low and behold, filesystem dirty! Christ
Re: Vision 2020: Making OpenBSD the world's fastest OS
On 2014/06/11 00:11, Allan Streib wrote: Stuart Henderson s...@spacehopper.org writes: This came up before, and I replied: http://marc.info/?l=openbsd-portsm=139450013100779w=2 The knob change you are recommending here allows 2GB shared memory. Depends on the system but for some people this will be way too much. I doubled my kern.shminfo.shmall setting to 16384 and that alone seems to have made my web browsers much happier. Too soon to say for sure, but that's the first impression. To be clear, this can *only* affect programs which use SysV shared memory. Chromium uses it for some users but from what I understand they do random trials with different methods so some people may be using it and others not. Not sure about Firefox, but the process I have running at the moment isn't using any, you can use ipcs(1) to look.
Re: Vision 2020: Making OpenBSD the world's fastest OS
Hi STeve, In my experience, OpenBSD's I/O operation speed depends on what I do. When I run rsync over a folder with many sub-folders and over 10,000 files (small files and large files), Linux is faster than OpenBSD. Kind regards, Xianwen On Wed, Jun 11, 2014 at 12:40 AM, STeve Andre' and...@msu.edu wrote: On 06/10/14 01:17, Amit Kulkarni wrote: Lastly, I will remind you that the fastest OS compared to OpenBSD is very likely less than 15%. Say its 25% even, and you could get faster hardware to accomedate that. Come on, that is a false assertion. OpenBSD does have its warts, like everybody else out there. They are different warts compared to others. But IMHO running it slow with security is better than running it fast, and not paying attention to secuirty. It's false? You think OpenBSD is slower than 15%? I don't, based on a few tests run against some version of Debian. It was faster, both in terms of disk i/o and the running of a program that did a lot of computations with little output. It seemed to me to be less than 6%, using stopwatches but small enough to make me stop testing. But I think you agree with the general tone of this? --STeve Andre' -- Xianwen Chen | xchen.tk
Re: hplip
On Tue, Jun 10, 2014 at 01:08:49PM -0700 or thereabouts, patrick keshishian wrote: On 6/10/14, Maurice McCarthy m...@mythic-beasts.com wrote: What was confusing was that hp-setup said in was configured as hp:/usb/... whereas in fact it ends configured as hp:/net/... HP even say, somewhere, it is best to remove the usb cable after set up. I think I read in your original post that one of the first things you did was to set up the wireless interface of the printer (ignore my post if I am misremembering). I don't know about this specific printer, but my Brother printer has USB, Ethernet and Wifi interfaces, but, once Wifi interface is configured, for example, Ethernet port will not work. Until the printer is reconfigured/reset and Ethernet selected. I wouldn't be surprised if your printer acts similarly. i.e., abandoning other interfaces in favor of the configure one. See your printer's documentation, it should specify this info. Mine did. --patrick Thank you Patrick. I think you are on the money there. I was expecting a both/and result and got an either-or. Silly me. Moss
Re: Vision 2020: Making OpenBSD the world's fastest OS
- Original Nachricht Von: STeve Andre' and...@msu.edu An: OpenBSD Misc misc@openbsd.org Datum: 11.06.2014 00:40 Betreff: Re: Vision 2020: Making OpenBSD the world's fastest OS But I think you agree with the general tone of this? In some aspects OpenBSD is *very* slow. After booting the X Windows System appears without delay but then it takes 23 seconds until xdm screen opens (on a fast Dell E6540). Also after changing from X to a virtual console and back (in the test case xlock was active) X appears without delay but it takes 10 seconds until the keyboard works. There is no delay in this cases on NetBSD or Linux even on my very old Dell D830 and even with a nvidia card. What can be the reason for this large delays? --Carsten
Re: Vision 2020: Making OpenBSD the world's fastest OS
Great tips! For a fresh install of OpenBSD, enabling softupdates may also help a bit (http://www.openbsd.org/faq/faq14.html#SoftUpdates). I know it's trivial, but maybe it's not that obvious for newbies. Also, having a supported video card would help in some heavy desktop environments, like Xfce (the new radeon driver in 5.5 made quite the difference on my machine). Claudiu. To: str...@cs.indiana.edu CC: misc@openbsd.org Subject: Re: Vision 2020: Making OpenBSD the world's fastest OS From: pe...@bsdly.net Date: Mon, 9 Jun 2014 22:28:32 +0200 Allan Streib str...@cs.indiana.edu writes: Can you share what you changed in login.conf, and what problems were resolved as a result? I mucked around with increasing the shared memory limits, and in fact it helped certain browsers go from glacial response times to merely 'a tad slow at times, YMMW'. http://home.nuug.no/~peter/transition/bsdcan2014/desktop.html and the following slide has the meat, such as it is. There's more work to be done for any 'OpenBSD as the ultimate desktop' article, though. - Peter -- Peter N. M. Hansteen, member of the first RFC 1149 implementation team http://bsdly.blogspot.com/ http://www.bsdly.net/ http://www.nuug.no/ Remember to set the evil bit on all malicious network traffic delilah spamd[29949]: 85.152.224.147: disconnected after 42673 seconds.
Re: Vision 2020: Making OpenBSD the world's fastest OS
Not just fastest OS but The Best OS. On Wed, Jun 11, 2014 at 3:57 PM, Claudiu TÄnÄselia clau...@tanaselia.ro wrote: Great tips! For a fresh install of OpenBSD, enabling softupdates may also help a bit (http://www.openbsd.org/faq/faq14.html#SoftUpdates). I know it's trivial, but maybe it's not that obvious for newbies. Also, having a supported video card would help in some heavy desktop environments, like Xfce (the new radeon driver in 5.5 made quite the difference on my machine). Claudiu. To: str...@cs.indiana.edu CC: misc@openbsd.org Subject: Re: Vision 2020: Making OpenBSD the world's fastest OS From: pe...@bsdly.net Date: Mon, 9 Jun 2014 22:28:32 +0200 Allan Streib str...@cs.indiana.edu writes: Can you share what you changed in login.conf, and what problems were resolved as a result? I mucked around with increasing the shared memory limits, and in fact it helped certain browsers go from glacial response times to merely 'a tad slow at times, YMMW'. http://home.nuug.no/~peter/transition/bsdcan2014/desktop.html and the following slide has the meat, such as it is. There's more work to be done for any 'OpenBSD as the ultimate desktop' article, though. - Peter -- Peter N. M. Hansteen, member of the first RFC 1149 implementation team http://bsdly.blogspot.com/ http://www.bsdly.net/ http://www.nuug.no/ Remember to set the evil bit on all malicious network traffic delilah spamd[29949]: 85.152.224.147: disconnected after 42673 seconds.
Re: Vision 2020: Making OpenBSD the world's fastest OS
On 2014-06-11, Carsten Kunze carsten.ku...@arcor.de wrote: - Original Nachricht Von: STeve Andre' and...@msu.edu An: OpenBSD Misc misc@openbsd.org Datum: 11.06.2014 00:40 Betreff: Re: Vision 2020: Making OpenBSD the world's fastest OS But I think you agree with the general tone of this? In some aspects OpenBSD is *very* slow. After booting the X Windows System appears without delay but then it takes 23 seconds until xdm screen opens (on a fast Dell E6540). Also after changing from X to a virtual console and back (in the test case xlock was active) X appears without delay but it takes 10 seconds until the keyboard works. There is no delay in this cases on NetBSD or Linux even on my very old Dell D830 and even with a nvidia card. What can be the reason for this large delays? --Carsten This may be a hostname lookup issue. Is this slow too? $ getent hosts `hostname`
Re: restore: no memory to extend symbol table
On Tue, Jun 10, 2014 at 09:18:33PM +0200, Otto Moerbeek wrote: A ulimit -a reveals your data limit, which is likely smaller than 1GB. You could try ulimit -d unlimited Doh! That did the trick. No more error message, restore(8) finished its work and all is good. Thanks Otto! -- best regards q#
ftp-proxy and multiple nat-to addresses
Hi, I have pf setup which includes NAT and ftp-proxy for accessing FTP servers on the Internet, and it works fine. I would like to add multiple addresses to NAT pool, instead of just one as in current setup, but I am not sure if this is going to play well with ftp-proxy. If I remember well, in order for ftp-proxy to enable outbound FTP connections from NAT clients to Internet FTP servers, its source adress (-a flag) needs to be the same as the public address to which NAT clients are translated. Thank you in advance, -- Marko Cupać
Re: Vision 2020: Making OpenBSD the world's fastest OS
- Original Nachricht Von: Stuart Henderson s...@spacehopper.org An: misc@openbsd.org Datum: 11.06.2014 13:50 Betreff: Re: Vision 2020: Making OpenBSD the world's fastest OS This may be a hostname lookup issue. Is this slow too? $ getent hosts `hostname` No, it returns fast (but does not print anything).
Re: Vision 2020: Making OpenBSD the world's fastest OS
2014-06-11 10:58 GMT-03:00 Carsten Kunze carsten.ku...@arcor.de: - Original Nachricht Von: Stuart Henderson s...@spacehopper.org An: misc@openbsd.org Datum: 11.06.2014 13:50 Betreff: Re: Vision 2020: Making OpenBSD the world's fastest OS This may be a hostname lookup issue. Is this slow too? $ getent hosts `hostname` No, it returns fast (but does not print anything). What is the output from echo $?, after you run the getent ... command?
Re: [LaTeX] Missing enumitem.sty
Hi Antonie, Thank you. That really helped. By the way, pkglocate is not a standard system binary, is it? Does it come by with another package? Kind regards, Xianwen On Wed, Jun 11, 2014 at 7:23 AM, Antoine Jacoutot ajacou...@bsdfrog.org wrote: On Wed, Jun 11, 2014 at 01:08:19AM +0200, Xiánwén Chén wrote: Hi guys, I've installed texlive_base. And yet several files are missing, including enumitem.sty. Is there another package I can install? You probably want texlive_texmf-full: $ pkglocate enumitem.sty dblatex-0.3p8:textproc/dblatex:/usr/local/share/dblatex/latex/misc/enumitem.s ty texlive_texmf-full-2013:print/texlive/texmf,-full:/usr/local/share/texmf-dist /tex/latex/enumitem/enumitem.sty texlive_texmf-full-2013:print/texlive/texmf,-full:/usr/local/share/texmf-dist /tex/latex/interfaces/interfaces-enumitem.sty -- Antoine -- Xianwen Chen | xchen.tk
Re: ftp-proxy and multiple nat-to addresses
On 11.6.2014. 14:29, Marko Cupać wrote: Hi, I have pf setup which includes NAT and ftp-proxy for accessing FTP servers on the Internet, and it works fine. I would like to add multiple addresses to NAT pool, instead of just one as in current setup, but I am not sure if this is going to play well with ftp-proxy. If I remember well, in order for ftp-proxy to enable outbound FTP connections from NAT clients to Internet FTP servers, its source adress (-a flag) needs to be the same as the public address to which NAT clients are translated. Thank you in advance, hello, maybe this is what you need http://marc.info/?l=openbsd-miscm=133061681116026w=1
Re: Vision 2020: Making OpenBSD the world's fastest OS
- Original Nachricht Von: Rodrigo Mosconi open...@mosconi.mat.br An: Carsten Kunze carsten.ku...@arcor.de Datum: 11.06.2014 16:05 Betreff: Re: Vision 2020: Making OpenBSD the world's fastest OS What is the output from echo $?, after you run the getent ... command? $ getent hosts `hostname` $ echo $? 0
Re: [LaTeX] Missing enumitem.sty
On Wed, Jun 11, 2014 at 03:54:32PM +0200, Xiánwén Chén wrote: Hi Antonie, Thank you. That really helped. By the way, pkglocate is not a standard system binary, is it? Does it come by with another package? Yes, install pkglocatedb
Re: Vision 2020: Making OpenBSD the world's fastest OS
On 06/11/14 15:16, Carsten Kunze wrote: - Original Nachricht Von: Rodrigo Mosconi open...@mosconi.mat.br An: Carsten Kunze carsten.ku...@arcor.de Datum: 11.06.2014 16:05 Betreff: Re: Vision 2020: Making OpenBSD the world's fastest OS What is the output from echo $?, after you run the getent ... command? $ getent hosts `hostname` $ echo $? 0 `hostname` should be replaced with a host...eg: port:fred ~ getent hosts 'google.com' 173.194.41.160 google.com 173.194.41.163 google.com 173.194.41.168 google.com 173.194.41.167 google.com 173.194.41.161 google.com 173.194.41.165 google.com 173.194.41.164 google.com 173.194.41.169 google.com 173.194.41.174 google.com 173.194.41.162 google.com 173.194.41.166 google.com 2a00:1450:4009:809::1003google.com
Re: Vision 2020: Making OpenBSD the world's fastest OS
- Original Nachricht Von: Fred open...@crowsons.com An: Carsten Kunze carsten.ku...@arcor.de, misc@openbsd.org Datum: 11.06.2014 16:25 Betreff: Re: Vision 2020: Making OpenBSD the world's fastest OS `hostname` should be replaced with a host...eg: port:fred ~ getent hosts 'google.com' 173.194.41.160 google.com 173.194.41.163 google.com 173.194.41.168 google.com 173.194.41.167 google.com 173.194.41.161 google.com 173.194.41.165 google.com 173.194.41.164 google.com 173.194.41.169 google.com 173.194.41.174 google.com 173.194.41.162 google.com 173.194.41.166 google.com 2a00:1450:4009:809::1003google.com In this case I have similar output. Also without delay.
Re: Vision 2020: Making OpenBSD the world's fastest OS
On 06/11/14 15:25, Fred wrote: On 06/11/14 15:16, Carsten Kunze wrote: - Original Nachricht Von: Rodrigo Mosconi open...@mosconi.mat.br An: Carsten Kunze carsten.ku...@arcor.de Datum: 11.06.2014 16:05 Betreff: Re: Vision 2020: Making OpenBSD the world's fastest OS What is the output from echo $?, after you run the getent ... command? $ getent hosts `hostname` $ echo $? 0 `hostname` should be replaced with a host...eg: port:fred ~ getent hosts 'google.com' 173.194.41.160 google.com 173.194.41.163 google.com 173.194.41.168 google.com 173.194.41.167 google.com 173.194.41.161 google.com 173.194.41.165 google.com 173.194.41.164 google.com 173.194.41.169 google.com 173.194.41.174 google.com 173.194.41.162 google.com 173.194.41.166 google.com 2a00:1450:4009:809::1003google.com Sent to quickly :~( if `hostname` is not returning anything then the current system does not have an /etc/myname file. hth Fred
Re: Vision 2020: Making OpenBSD the world's fastest OS
On Wed, Jun 11, 2014 at 4:32 PM, Carsten Kunze carsten.ku...@arcor.de wrote: - Original Nachricht Von: Fred open...@crowsons.com An: Carsten Kunze carsten.ku...@arcor.de, misc@openbsd.org Datum: 11.06.2014 16:25 Betreff: Re: Vision 2020: Making OpenBSD the world's fastest OS `hostname` should be replaced with a host...eg: port:fred ~ getent hosts 'google.com' 173.194.41.160 google.com 173.194.41.163 google.com 173.194.41.168 google.com 173.194.41.167 google.com 173.194.41.161 google.com 173.194.41.165 google.com 173.194.41.164 google.com 173.194.41.169 google.com 173.194.41.174 google.com 173.194.41.162 google.com 173.194.41.166 google.com 2a00:1450:4009:809::1003google.com In this case I have similar output. Also without delay. Also: # cat /etc/myname and: # cat /etc/hosts ciao, David
Re: OpenBSD 5.5 on mSATA SSD unit in PC Engines APU.1C - bad dir ino 2 at offset 0: mangled entry kernel panic
Julian Andrej [j.and...@gmx.de] wrote: Does an install on usb or sdcard medium work for you? SD on APU is USB, and it works fine, although my SD card to test is so pathetically slow that 'noatime' on fliesystem mounts makes a noticeable difference. It seems like every bit of disk activity big or small has some large waiting time with this random, old SD card. Plextor M5M (msata) works fine so far. I have regular SATA X110 models that work on intel ahci, but the 32GB msatas are really funky on this amd ahci. Either I got a batch of bad ones or there are some workarounds necessary. Plextor M5M is already unavailable so I'm going to replace my bad X110s with M6M. A crapshoot. msata is just a smaller form factor for sata, so if sandisk's x110 part has a general compatibility issue, i'd imagine it would show up everywhere. Just to shut down the x110 msata properly, I have to do sync; sleep 1; sync; reboot
Re: Vision 2020: Making OpenBSD the world's fastest OS
- Original Nachricht Von: Fred open...@crowsons.com An: Carsten Kunze carsten.ku...@arcor.de, misc@openbsd.org Datum: 11.06.2014 16:28 Betreff: Re: Vision 2020: Making OpenBSD the world's fastest OS Sent to quickly :~( if `hostname` is not returning anything then the current system does not have an /etc/myname file. No, it has this file ... It contains a name of the form a.b.c where b.c is an offical domain name, but a is unkown to an external name server. (But a.b.c is known to our intranet name server.) --Carsten
Re: Vision 2020: Making OpenBSD the world's fastest OS
Carsten Kunze [carsten.ku...@arcor.de] wrote: - Original Nachricht Von: STeve Andre' and...@msu.edu An: OpenBSD Misc misc@openbsd.org Datum: 11.06.2014 00:40 Betreff: Re: Vision 2020: Making OpenBSD the world's fastest OS But I think you agree with the general tone of this? In some aspects OpenBSD is *very* slow. After booting the X Windows System appears without delay but then it takes 23 seconds until xdm screen opens (on a fast Dell E6540). Also after changing from X to a virtual console and back (in the test case xlock was active) X appears without delay but it takes 10 seconds until the keyboard works. I believe this is fixed in -current. ps/2 mouse driver issue compared to modern hardware
Re: Vision 2020: Making OpenBSD the world's fastest OS
- Original Nachricht Von: David Coppa dco...@gmail.com An: Carsten Kunze carsten.ku...@arcor.de Datum: 11.06.2014 16:35 Betreff: Re: Vision 2020: Making OpenBSD the world's fastest OS # cat /etc/myname It's a company hostname, I don't know, if I get legal issues It's like a146.b.com a and b are words with lowercase letters. b.com is known to external name servers. # cat /etc/hosts # $OpenBSD: hosts,v 1.12 2009/03/10 00:42:13 deraadt Exp $ # # Host Database # # RFC 1918 specifies that these networks are internal. # 10.0.0.0 10.255.255.255 # 172.16.0.0172.31.255.255 # 192.168.0.0 192.168.255.255 # 127.0.0.1 localhost ::1 localhost Cheers, Carsten
Re: Vision 2020: Making OpenBSD the world's fastest OS
On Wed, Jun 11, 2014 at 4:45 PM, Chris Cappuccio ch...@nmedia.net wrote: Carsten Kunze [carsten.ku...@arcor.de] wrote: - Original Nachricht Von: STeve Andre' and...@msu.edu An: OpenBSD Misc misc@openbsd.org Datum: 11.06.2014 00:40 Betreff: Re: Vision 2020: Making OpenBSD the world's fastest OS But I think you agree with the general tone of this? In some aspects OpenBSD is *very* slow. After booting the X Windows System appears without delay but then it takes 23 seconds until xdm screen opens (on a fast Dell E6540). Also after changing from X to a virtual console and back (in the test case xlock was active) X appears without delay but it takes 10 seconds until the keyboard works. I believe this is fixed in -current. ps/2 mouse driver issue compared to modern hardware Chris is right. It's a Dell, so you probably need this patch: http://www.openbsd.org/cgi-bin/cvsweb/src/sys/dev/pckbc/pms.c.diff?r1=1.49;r2=1.50 Ciao! David
Re: Vision 2020: Making OpenBSD the world's fastest OS
- Original Nachricht Von: Chris Cappuccio ch...@nmedia.net An: Carsten Kunze carsten.ku...@arcor.de Datum: 11.06.2014 16:45 Betreff: Re: Vision 2020: Making OpenBSD the world's fastest OS I believe this is fixed in -current. ps/2 mouse driver issue compared to modern hardware Ok, thanx for this!
Re: Vision 2020: Making OpenBSD the world's fastest OS
- Original Nachricht Von: David Coppa dco...@gmail.com An: Chris Cappuccio ch...@nmedia.net Datum: 11.06.2014 16:55 Betreff: Re: Vision 2020: Making OpenBSD the world's fastest OS Chris is right. It's a Dell, so you probably need this patch: http://www.openbsd.org/cgi-bin/cvsweb/src/sys/dev/pckbc/pms.c.diff?r1=1.49;r 2=1.50 Ciao! David Ok, my old laptop had this problem too, it had also been a Dell ;) Thank you all! Cheers, Carsten
Re: [LaTeX] Missing enumitem.sty
Thank you Marc. Kind regards, Xianwen On Wed, Jun 11, 2014 at 4:19 PM, Marc Espie es...@nerim.net wrote: On Wed, Jun 11, 2014 at 03:54:32PM +0200, Xiánwén Chén wrote: Hi Antonie, Thank you. That really helped. By the way, pkglocate is not a standard system binary, is it? Does it come by with another package? Yes, install pkglocatedb -- Xianwen Chen | xchen.tk
Ruby, Python programs are unusually slow
When I start Python and Ruby programs (and probably others too), they are very, very slow to do anything and do not respond to SIGINT or SIGTERM. For example, when I run rackup in an almost bare Rack project (in Ruby), I have to wait for about 10 minutes before the web server starts. The same is true when running python2.7 -m SimpleHTTPServer, which I'd expect to start serving almost immediately, but instead takes 10 minutes to show any sign of activity. I know this isn't a problem with my hardware (a ThinkPad T61) being slow, since it works fine on Debian and FreeBSD. Seeing as the problem is worst with programs that access the network, maybe the problem has something to do with that? I'd hate to have to switch to another OS to do work. dmesg is below. OpenBSD 5.5 (GENERIC.MP) #315: Wed Mar 5 09:37:46 MST 2014 dera...@amd64.openbsd.org:/usr/src/sys/arch/amd64/compile/GENERIC.MP real mem = 2087387136 (1990MB) avail mem = 2023268352 (1929MB) mainbus0 at root bios0 at mainbus0: SMBIOS rev. 2.4 @ 0xe0010 (73 entries) bios0: vendor LENOVO version 7LET44WW (1.14 ) date 06/27/2007 bios0: LENOVO 765912G acpi0 at bios0: rev 2 acpi0: sleep states S0 S3 S4 S5 acpi0: tables DSDT FACP SSDT ECDT TCPA APIC MCFG HPET SLIC BOOT ASF! SSDT SSDT SSDT SSDT acpi0: wakeup devices LID_(S3) SLPB(S3) LURT(S3) DURT(S3) IGBE(S4) EXP0(S4) EXP1(S4) EXP2(S4) EXP3(S4) EXP4(S4) PCI1(S4) USB0(S3) USB1(S3) USB2(S3) USB3(S3) USB4(S3) [...] acpitimer0 at acpi0: 3579545 Hz, 24 bits acpiec0 at acpi0 acpimadt0 at acpi0 addr 0xfee0: PC-AT compat cpu0 at mainbus0: apid 0 (boot processor) cpu0: Intel(R) Core(TM)2 Duo CPU T7300 @ 2.00GHz, 2194.88 MHz cpu0: FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CFLUSH,DS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE,SSE3,DTES64,MWAIT,DS-CPL,VMX,EST,TM2,SSSE3,CX16,xTPR,PDCM,NXE,LONG,LAHF,PERF cpu0: 4MB 64b/line 16-way L2 cache cpu0: smt 0, core 0, package 0 mtrr: Pentium Pro MTRR support, 8 var ranges, 88 fixed ranges cpu0: apic clock running at 199MHz cpu0: mwait min=64, max=64, C-substates=0.2.2.2.2, IBE cpu1 at mainbus0: apid 1 (application processor) cpu1: Intel(R) Core(TM)2 Duo CPU T7300 @ 2.00GHz, 1995.01 MHz cpu1: FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CFLUSH,DS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE,SSE3,DTES64,MWAIT,DS-CPL,VMX,EST,TM2,SSSE3,CX16,xTPR,PDCM,NXE,LONG,LAHF,PERF cpu1: 4MB 64b/line 16-way L2 cache cpu1: smt 0, core 1, package 0 ioapic0 at mainbus0: apid 1 pa 0xfec0, version 20, 24 pins ioapic0: misconfigured as apic 2, remapped to apid 1 acpimcfg0 at acpi0 addr 0xf000, bus 0-63 acpihpet0 at acpi0: 14318179 Hz acpiprt0 at acpi0: bus 0 (PCI0) acpiprt1 at acpi0: bus -1 (AGP_) acpiprt2 at acpi0: bus 2 (EXP0) acpiprt3 at acpi0: bus 3 (EXP1) acpiprt4 at acpi0: bus 4 (EXP2) acpiprt5 at acpi0: bus 5 (EXP3) acpiprt6 at acpi0: bus 13 (EXP4) acpiprt7 at acpi0: bus 21 (PCI1) acpicpu0 at acpi0: C3, C2, C1, PSS acpicpu1 at acpi0: C3, C2, C1, PSS acpipwrres0 at acpi0: PUBS, resource for USB0, USB2, USB4, EHC0, EHC1 acpitz0 at acpi0: critical temperature is 127 degC acpitz1 at acpi0: critical temperature is 100 degC acpibtn0 at acpi0: LID_ acpibtn1 at acpi0: SLPB acpibat0 at acpi0: BAT0 model 42T5225 serial 160 type LION oem Panasonic acpibat1 at acpi0: BAT1 not present acpiac0 at acpi0: AC unit online acpithinkpad0 at acpi0 acpidock0 at acpi0: GDCK not docked (0) cpu0: Enhanced SpeedStep 2194 MHz: speeds: 2001, 2000, 1600, 1200, 800 MHz pci0 at mainbus0 bus 0 pchb0 at pci0 dev 0 function 0 Intel GM965 Host rev 0x0c vga1 at pci0 dev 2 function 0 Intel GM965 Video rev 0x0c intagp0 at vga1 agp0 at intagp0: aperture at 0xe000, size 0x1000 inteldrm0 at vga1 drm0 at inteldrm0 inteldrm0: 1280x800 wsdisplay0 at vga1 mux 1: console (std, vt100 emulation) wsdisplay0: screen 1-5 added (std, vt100 emulation) Intel GM965 Video rev 0x0c at pci0 dev 2 function 1 not configured em0 at pci0 dev 25 function 0 Intel ICH8 IGP M AMT rev 0x03: msi, address 00:15:58:c6:fd:70 uhci0 at pci0 dev 26 function 0 Intel 82801H USB rev 0x03: apic 1 int 20 uhci1 at pci0 dev 26 function 1 Intel 82801H USB rev 0x03: apic 1 int 21 ehci0 at pci0 dev 26 function 7 Intel 82801H USB rev 0x03: apic 1 int 22 usb0 at ehci0: USB revision 2.0 uhub0 at usb0 Intel EHCI root hub rev 2.00/1.00 addr 1 azalia0 at pci0 dev 27 function 0 Intel 82801H HD Audio rev 0x03: msi azalia0: codecs: Analog Devices AD1984, Conexant/0x2bfa, using Analog Devices AD1984 audio0 at azalia0 ppb0 at pci0 dev 28 function 0 Intel 82801H PCIE rev 0x03: msi pci1 at ppb0 bus 2 ppb1 at pci0 dev 28 function 1 Intel 82801H PCIE rev 0x03: msi pci2 at ppb1 bus 3 iwn0 at pci2 dev 0 function 0 Intel Wireless WiFi Link 4965 rev 0x61: msi, MIMO 2T3R, MoW2, address 00:13:e8:73:04:85 ppb2 at pci0 dev 28 function 2 Intel 82801H PCIE rev 0x03: msi pci3 at ppb2 bus 4 ppb3 at pci0 dev 28 function 3 Intel 82801H PCIE rev 0x03: msi pci4 at ppb3 bus 5 ppb4 at pci0 dev 28 function 4 Intel 82801H PCIE rev 0x03:
Duplicating a disk
To duplicate a disk I used the following: dd if=/dev/rsd2c of=/dev/rsd3c bs=32M seek=1 skip=1 conv=noerror the bs=32M was picked because it was a large size, and the machine has lots of free memory. Watching the machine I could see the disk activity lights blinking alternately about once a second and looks like, what I would expect, that dd does blocking I/O. Is there any method of coping a disk or partition, or even a file that uses non-blocking I/O? Such a method should cut the time down by half. Also for dd the block size has always been a puzzle. Asking google gives various opinion, only agreeing that the number should be a power of two. I have always had the believe that a bigger size is never hurts as long as there is free memory available on the system. Would there not be a method for dd to calculate what an optimal block size would be given the free memory and devices used.
Re: Duplicating a disk
On Wed, Jun 11, 2014 at 05:45:30PM +, Peter Fraser wrote: To duplicate a disk I used the following: dd if=/dev/rsd2c of=/dev/rsd3c bs=32M seek=1 skip=1 conv=noerror the bs=32M was picked because it was a large size, and the machine has lots of free memory. Watching the machine I could see the disk activity lights blinking alternately about once a second and looks like, what I would expect, that dd does blocking I/O. Is there any method of coping a disk or partition, or even a file that uses non-blocking I/O? Such a method should cut the time down by half. Also for dd the block size has always been a puzzle. Asking google gives various opinion, only agreeing that the number should be a power of two. I have always had the believe that a bigger size is never hurts as long as there is free memory available on the system. Would there not be a method for dd to calculate what an optimal block size would be given the free memory and devices used. It could very well be that large block sizes are the cause of lack of parallelism. Try smaller ones, e.g. 64k. But if dd writes and waits for the write to finish, it would not matter much. -Otto
Re: Duplicating a disk
To duplicate a disk I used the following: dd if=/dev/rsd2c of=/dev/rsd3c bs=32M seek=1 skip=1 conv=noerror the bs=32M was picked because it was a large size, and the machine has lots of free memory. Watching the machine I could see the disk activity lights blinking alternately about once a second and looks like, what I would expect, that dd does blocking I/O. Is there any method of coping a disk or partition, or even a file that uses non-blocking I/O? Such a method should cut the time down by half. Also for dd the block size has always been a puzzle. Asking google gives various opinion, only agreeing that the number should be a power of two. I have always had the believe that a bigger size is never hurts as long as there is free memory available on the system. Would there not be a method for dd to calculate what an optimal block size would be given the free memory and devices used. It could very well be that large block sizes are the cause of lack of parallelism. Try smaller ones, e.g. 64k. But if dd writes and waits for the write to finish, it would not matter much. This is similar to the netcat conversation yesterday.
Re: Duplicating a disk
On Wed, Jun 11, 2014 at 17:45, Peter Fraser wrote: To duplicate a disk I used the following: dd if=/dev/rsd2c of=/dev/rsd3c bs=32M seek=1 skip=1 conv=noerror the bs=32M was picked because it was a large size, and the machine has lots of free memory. Watching the machine I could see the disk activity lights blinking alternately about once a second and looks like, what I would expect, that dd does blocking I/O. Is there any method of coping a disk or partition, or even a file that uses non-blocking I/O? There was a change made a few years ago to limit the number of pending writes, which improves interactive response. It's not supposed to seriously impair write performance, but I've had a few worries that maybe it does. A workaround may be to use two dd processes; something like: dd if=/dev/rsd0c bs=64k | dd of=/dev/rsd1c bs=64 (The change was originally made due to a very similar situation as you're doing, but with slower disks. If you copy a huge file to a slow USB flash drive, the kernel would buffer the entire file for writing and eventually you'd be unable to do anything else until the file was written out. After the change, writes are effectively restricted to disk speed, but this may be causing dd to stall when it could be reading.)
Re: Ruby, Python programs are unusually slow
On Wed, Jun 11, 2014 at 05:52:12PM +0100, Kaashif Hymabaccus wrote: [...] I know this isn't a problem with my hardware (a ThinkPad T61) being slow, [...] Definitely, my R61 starts a Python HTTP server almost instantly. Seeing as the problem is worst with programs that access the network, maybe the problem has something to do with that? [...] Could be a problem with name resolution. Do you have an entry for your hostname in /etc/hosts? How quick is name resolution in general, i.e. via something like host localhost host `hostname -s` host openbsd.org -- Gregor Best
Re: cursor problems with radeondrm(4) framebuffer - HD3200
On Thu, May 22, 2014 at 9:48 PM, Andrew Daugherity andrew.daugher...@gmail.com wrote: 2) The cursor completely blocks out whatever letter it is positioned over (command editing, vi, etc.)... I also noticed that my laptop does not show highlighting in man pages -- everything is the same standard light-grey color. Probably a related issue. This an MSI Wind U230L netbook, with Athlon Neo MV40 CPU + RS780/HD3200 graphics. X works fine, at least what little bit I've tested. [ dmesg in original message] Anybody have any ideas on how to debug this further, or possible fixes? I did try the latest snapshot as of my message (i.e. ca. May 20), with no difference. -Andrew
Re: Duplicating a disk
It was pointed out to me that linux's dd has a oflag=nonblock and a iflag=nonblock option to invoke non-blocking I/O. I don't know why linux allows non-blocking I/O on the input file. It never makes any sense to have non-blocking I/O in the input file, The read has to complete before the write can take place. All it would do is complicate the program flow. -Original Message- From: owner-m...@openbsd.org [mailto:owner-m...@openbsd.org] On Behalf Of Ted Unangst Sent: Wednesday, June 11, 2014 2:11 PM To: Peter Fraser Cc: 'misc@openbsd.org' Subject: Re: Duplicating a disk On Wed, Jun 11, 2014 at 17:45, Peter Fraser wrote: To duplicate a disk I used the following: dd if=/dev/rsd2c of=/dev/rsd3c bs=32M seek=1 skip=1 conv=noerror the bs=32M was picked because it was a large size, and the machine has lots of free memory. Watching the machine I could see the disk activity lights blinking alternately about once a second and looks like, what I would expect, that dd does blocking I/O. Is there any method of coping a disk or partition, or even a file that uses non-blocking I/O?
Re: Duplicating a disk
On 2014-06-11, Peter Fraser p...@thinkage.ca wrote: To duplicate a disk I used the following: dd if=/dev/rsd2c of=/dev/rsd3c bs=32M seek=1 skip=1 conv=noerror Why are you skipping the first 32M? Is there any method of coping a disk or partition, or even a file that uses non-blocking I/O? You could use buffer(1) from ports/misc/buffer. I used this to keep tape drives streaming... It's been a while. I only find myself copying disks/partitions/large file trees at the filesystem level; dump|restore or tar|tar seem to keep source and destination pretty busy, but then again, speed is usually not of the essence. Also for dd the block size has always been a puzzle. For accessing a raw device you want it to be a multiple of the sector size of the device (512 bytes for most disks) and there is usually no point in making it bigger than MAXPHYS (64k on OpenBSD), i.e., the maximal size of a single I/O transfer the kernel handles; larger reads or writes will be broken up into multiple transfers. -- Christian naddy Weisgerber na...@mips.inka.de
Re: Ruby, Python programs are unusually slow
On Wed, Jun 11, 2014 at 08:42:14PM +0200, Gregor Best wrote: On Wed, Jun 11, 2014 at 05:52:12PM +0100, Kaashif Hymabaccus wrote: [...] I know this isn't a problem with my hardware (a ThinkPad T61) being slow, [...] Definitely, my R61 starts a Python HTTP server almost instantly. Seeing as the problem is worst with programs that access the network, maybe the problem has something to do with that? [...] Could be a problem with name resolution. Do you have an entry for your hostname in /etc/hosts? How quick is name resolution in general, i.e. via something like host localhost host `hostname -s` host openbsd.org -- Gregor Best Localhost was taking a very, very long time (almost exactly the 10 minutes the web server was taking to start) to resolve for some reason, so I changed the nameservers in /etc/resolv.conf and everything seems to work fine. I even upgraded to -current to see if it'd fix the problem! I feel silly now, the problem was so obvious. Thanks.
Re: Duplicating a disk
On 06/11/14 15:55, Christian Weisgerber wrote: On 2014-06-11, Peter Fraser p...@thinkage.ca wrote: ... Also for dd the block size has always been a puzzle. For accessing a raw device you want it to be a multiple of the sector size of the device (512 bytes for most disks) and there is usually no point in making it bigger than MAXPHYS (64k on OpenBSD), i.e., the maximal size of a single I/O transfer the kernel handles; larger reads or writes will be broken up into multiple transfers. I've heard this a number of times...and yet my testing on hardware I've had in front of me (i.e., your throughput may vary) has shown that bs=1M does give substantially better throughput when zeroing disks than 32k, and last time I did extensive testing in this, sizes larger than 1MB give even better throughput, though the return gets very small after around 1MB -- so I usually use 1MB so a pkill -INFO dd will give me an indication of the progress in easy to read terms, which I find more useful than a 1% reduction in time. I'm just reporting an observation, not explaining it. :) Nick.
