VIA 686 vs cdrecord = slow!
Hello! During the past week I've updated to a new system that features an Athlon 1.33G processor and the VIA chipset. I did a direct copy of my Linux partitions from the HD of the old machine to the HD of this machine. On the old machine I had no problems writing CDs at 12X with my Sony CRX160E using either xcdroast or the command line. In this machine 12X fails every time. More testing revealed that 4X works every time. 8X is unreliable. During the one time 8X worked, I received the following message about 25% through the disk (a 16 track audio CD of about 72 minutes): Probable hardware bug: clock time configuration lost - probably a VIA 686a Probable hardware bug: restoring chip configuration The write process continued and completed normally. The failures generated similar output to the following: Waiting for reader process to fill input buffer ... input buffer ready. Starting new track at sector: 0 cdrecord: Input/output error. write_g1: scsi sendcmd: no error CDB: 2A 00 00 00 0D EC 00 00 1B 00 status: 0x2 (CHECK CONDITION) Sense Bytes: F0 00 03 00 00 0D EC 12 00 00 00 00 0C 09 00 00 Sense Key: 0x3 Medium Error, Segment 0 Sense Code: 0x0C Qual 0x09 (write error - loss of streaming) Fru 0x0 Sense flags: Blk 3564 (valid) cmd finished after 0.027s timeout 40s Sense Bytes: 70 00 00 00 00 00 00 12 00 00 00 00 00 00 00 00 00 00 write track data: error after 8382528 bytes Writing time: 10.710s Fixating... I'm curious if there are known bugs with the VIA chipset and whether additional testers are needed? Please let me know if I can help. 4X is okay, but I'd like 12X! I am using Debian Testing with a 2.4.18 kernel and cdrecord 1.10-2.4 packaged by Debian. - Nate >> -- Wireless | Amateur Radio Station N0NB | "We have awakened a Internet | [EMAIL PROTECTED] | sleeping giant and Location | Bremen, Kansas USA EM19ov | have instilled in him Amateur radio exams; ham radio; Linux info @ | a terrible resolve". http://www.qsl.net/n0nb/ | - Admiral Yamomoto -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: VIA 686 vs cdrecord = slow!
>From: Nate Bargmann <[EMAIL PROTECTED]> >During the past week I've updated to a new system that features an >Athlon 1.33G processor and the VIA chipset. I did a direct copy of my >Linux partitions from the HD of the old machine to the HD of this >machine. On the old machine I had no problems writing CDs at 12X with >my Sony CRX160E using either xcdroast or the command line. >In this machine 12X fails every time. More testing revealed that 4X >works every time. 8X is unreliable. During the one time 8X worked, I >received the following message about 25% through the disk (a 16 track >audio CD of about 72 minutes): >Probable hardware bug: clock time configuration lost - probably a VIA 686a >Probable hardware bug: restoring chip configuration >The write process continued and completed normally. >The failures generated similar output to the following: >Waiting for reader process to fill input buffer ... >input buffer ready. >Starting new track at sector: 0 >cdrecord: Input/output error. write_g1: scsi sendcmd: no error >CDB: 2A 00 00 00 0D EC 00 00 1B 00 >status: 0x2 (CHECK CONDITION) >Sense Bytes: F0 00 03 00 00 0D EC 12 00 00 00 00 0C 09 00 00 >Sense Key: 0x3 Medium Error, Segment 0 >Sense Code: 0x0C Qual 0x09 (write error - loss of streaming) Fru 0x0 >Sense flags: Blk 3564 (valid) >cmd finished after 0.027s timeout 40s A buffer underrun! Read README.ATAPI. If this does not help, ask the Linux kernel folks. Jörg EMail:[EMAIL PROTECTED] (home) Jörg Schilling D-13353 Berlin [EMAIL PROTECTED] (uni) If you don't have iso-8859-1 [EMAIL PROTECTED] (work) chars I am J"org Schilling URL: http://www.fokus.gmd.de/usr/schilling ftp://ftp.fokus.gmd.de/pub/unix -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: VIA 686 vs cdrecord = slow!
On Wed, Mar 06, 2002 at 10:43:33PM -0600, Nate Bargmann wrote: > Hello! > > During the past week I've updated to a new system that features an > Athlon 1.33G processor and the VIA chipset. I did a direct copy of my > Linux partitions from the HD of the old machine to the HD of this > machine. On the old machine I had no problems writing CDs at 12X with > my Sony CRX160E using either xcdroast or the command line. > > In this machine 12X fails every time. More testing revealed that 4X > works every time. 8X is unreliable. During the one time 8X worked, I > received the following message about 25% through the disk (a 16 track > audio CD of about 72 minutes): > > Probable hardware bug: clock time configuration lost - probably a VIA 686a > Probable hardware bug: restoring chip configuration > > The write process continued and completed normally. > > The failures generated similar output to the following: > > Waiting for reader process to fill input buffer ... > input buffer ready. > Starting new track at sector: 0 > cdrecord: Input/output error. write_g1: scsi sendcmd: no error > CDB: 2A 00 00 00 0D EC 00 00 1B 00 > status: 0x2 (CHECK CONDITION) > Sense Bytes: F0 00 03 00 00 0D EC 12 00 00 00 00 0C 09 00 00 > Sense Key: 0x3 Medium Error, Segment 0 > Sense Code: 0x0C Qual 0x09 (write error - loss of streaming) Fru 0x0 > Sense flags: Blk 3564 (valid) > cmd finished after 0.027s timeout 40s > > Sense Bytes: 70 00 00 00 00 00 00 12 00 00 00 00 00 00 00 00 00 00 > write track data: error after 8382528 bytes > Writing time: 10.710s > Fixating... > > I'm curious if there are known bugs with the VIA chipset and whether > additional testers are needed? Please let me know if I can help. 4X is > okay, but I'd like 12X! > > I am using Debian Testing with a 2.4.18 kernel and cdrecord 1.10-2.4 > packaged by Debian. You probably need DMA access on your HD to get enough read speed to read the disk. I have used 2.2 kernels on a via 686 south bridge and gotten about 3MB/sec, and after switching to a 2.4 kernel with support for that chipset, dma was enabled and I got 27MB/sec which would help a lot for doing fast writes, and also makes the system much much faster. Check /proc/ide/hda/settings, and make sure dma is currently on, and make sure the kernel is new enough to support your chipset fully. Len Sorensen -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: VIA 686 vs cdrecord = slow!
On Thu, 7 Mar 2002, Joerg Schilling wrote: > >From: Nate Bargmann <[EMAIL PROTECTED]> > > > >During the past week I've updated to a new system that features an > >Athlon 1.33G processor and the VIA chipset. I did a direct copy of my > >Linux partitions from the HD of the old machine to the HD of this > >machine. On the old machine I had no problems writing CDs at 12X with > >my Sony CRX160E using either xcdroast or the command line. > > >In this machine 12X fails every time. More testing revealed that 4X > >works every time. 8X is unreliable. During the one time 8X worked, I > >received the following message about 25% through the disk (a 16 track > >audio CD of about 72 minutes): > > >Probable hardware bug: clock time configuration lost - probably a VIA 686a > >Probable hardware bug: restoring chip configuration > A buffer underrun! > > Read README.ATAPI. > > If this does not help, ask the Linux kernel folks. I shudder to think that you really "did a direct copy" from one machine and are running the same kernel on a new machine with a different architecture. Try building a 2.4.17 or 2.4.19-pre2-ac2 kernel for this and see if the problem goes away. When you go to a bright shiney new machine, with a very recent chipset, going to a very recent kernel is highly recommended. I don't promise that this is fixed, but at least a kernel which fits your hardware will give you a fighting chance. I bet you're still running your old config files as well, doing whatever hardware setup was a good idea for the old machine... -- -bill davidsen ([EMAIL PROTECTED]) "The secret to procrastination is to put things off until the last possible moment - but no longer" -me -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: VIA 686 vs cdrecord = slow!
* Joerg Schilling <[EMAIL PROTECTED]> [2002 Mar 07 06:08 -0600]: > > A buffer underrun! Yup! Right on the money. A bit of research and I was able to turn my measly 1.9 MB/S throughput into 30.77 MB/S. > Read README.ATAPI. I about gave up on it until I reached the very end where buffer underruns were discussed. > If this does not help, ask the Linux kernel folks. I consulted the lkml archives and found wuite a bit on disk tuning. Thanks for the tip. Burning now works at 12X with my drive. - Nate >> -- Wireless | Amateur Radio Station N0NB | "We have awakened a Internet | [EMAIL PROTECTED] | sleeping giant and Location | Bremen, Kansas USA EM19ov | have instilled in him Amateur radio exams; ham radio; Linux info @ | a terrible resolve". http://www.qsl.net/n0nb/ | - Admiral Yamomoto -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Re: VIA 686 vs cdrecord = slow!
* Bill Davidsen <[EMAIL PROTECTED]> [2002 Mar 07 16:27 -0600]: > > I shudder to think that you really "did a direct copy" from one machine > and are running the same kernel on a new machine with a different > architecture. Try building a 2.4.17 or 2.4.19-pre2-ac2 kernel for this > and see if the problem goes away. Actually, it worked quite well as the previous system was quite generic and I'd never compiled chipset specific features in the 2.2.19 kernel. It was built for a K6 kernel and I really wondered if it would boot at all. But, Linux is quite forgiving, IMO. > When you go to a bright shiney new machine, with a very recent chipset, > going to a very recent kernel is highly recommended. I don't promise that > this is fixed, but at least a kernel which fits your hardware will give > you a fighting chance. I now have that kernel. :-) > I bet you're still running your old config files as well, doing > whatever hardware setup was a good idea for the old machine... I've slowly merged things over and now I performed the most dramatic tune-up and things are running nicely. Oh, and I'm back to torching CDs at 12X again. Oh, happy day! - Nate >> -- Wireless | Amateur Radio Station N0NB | "We have awakened a Internet | [EMAIL PROTECTED] | sleeping giant and Location | Bremen, Kansas USA EM19ov | have instilled in him Amateur radio exams; ham radio; Linux info @ | a terrible resolve". http://www.qsl.net/n0nb/ | - Admiral Yamomoto -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
