Most recent kernel where this bug did not occur: none Distribution: Gentoo Linux Hardware Environment: LG LS70 Express laptop with dual-layer burner Software Environment: dvd+rw-tools-5.21.4.10.8, kernel 2.6.13.2 Problem Description: DVD burning is very slow about 0.5x and there is high cpu usage (load average: 1.78, 1.41, 1.41) even the dma is on. This has been in discussion in Gentoo Forums, but no one have solution for this. Laptop is fairly new and using Intel 915GM chipset and has dual-layer burner (TSSTcorpCD/DVDW TS-L632B). I have tested many different type of kernel configurations and kernel versions 2.4.31 (ide-scsi), 2.6.10, 2.6.13 and 2.6.13-archck6.2 with same problem and trying to set different kombinations of settings with hdparm to the drive. No luck. Maybe there is problem in ide-cd or chipset driver (PIIx), don't know. I tried with 2.4.31 kernel ide-scsi burning also with no success, same problems. Normal CD's seems to be ok and working. Maybe this bug report is duplicate of #4617 but I don't want to decide that. Steps to reproduce: Try to burn DVD's with my hardware.
Created attachment 6226 [details] Burning log and info about my system
Created attachment 6227 [details] Nother burning log in single user mode
Created attachment 6228 [details] Third burning log when tried to set drive parameters
Created attachment 6229 [details] Output of dmesg
It would be useful if you could generate a kernel profile during the actual burning operation. See Documentation/basic_profiling.txt Thanks.
Created attachment 6274 [details] Captured profile Attached the "captured_profile" and this is what I did and outputs of cmd's (I stopped the burning process in 1.0% with CTRL+C). # readprofile -r # growisofs -Z /dev/hdb=image.iso Executing 'builtin_dd if=image.iso of=/dev/hdb obs=32k seek=0' /dev/hdb: "Current Write Speed" is 4.1x1385KBps. 655360/4696774656 ( 0.0%) @0.1x, remaining 716:34 3604480/4696774656 ( 0.1%) @0.6x, remaining 217:00 5210112/4696774656 ( 0.1%) @0.3x, remaining 195:06 7536640/4696774656 ( 0.2%) @0.5x, remaining 165:55 9207808/4696774656 ( 0.2%) @0.4x, remaining 169:41 11108352/4696774656 ( 0.2%) @0.4x, remaining 161:41 13041664/4696774656 ( 0.3%) @0.4x, remaining 155:37 16285696/4696774656 ( 0.3%) @0.7x, remaining 143:41 18022400/4696774656 ( 0.4%) @0.4x, remaining 142:47 19857408/4696774656 ( 0.4%) @0.4x, remaining 141:18 20774912/4696774656 ( 0.4%) @0.2x, remaining 150:03 22872064/4696774656 ( 0.5%) @0.4x, remaining 146:27 23953408/4696774656 ( 0.5%) @0.2x, remaining 149:33 25493504/4696774656 ( 0.5%) @0.3x, remaining 152:41 26836992/4696774656 ( 0.6%) @0.3x, remaining 153:42 28114944/4696774656 ( 0.6%) @0.3x, remaining 154:59 29753344/4696774656 ( 0.6%) @0.3x, remaining 156:51 31784960/4696774656 ( 0.7%) @0.4x, remaining 154:06 33751040/4696774656 ( 0.7%) @0.4x, remaining 151:58 35913728/4696774656 ( 0.8%) @0.5x, remaining 151:24 37388288/4696774656 ( 0.8%) @0.3x, remaining 151:37 39518208/4696774656 ( 0.8%) @0.5x, remaining 149:16 42270720/4696774656 ( 0.9%) @0.6x, remaining 146:48 43941888/4696774656 ( 0.9%) @0.4x, remaining 146:28 46202880/4696774656 ( 1.0%) @0.5x, remaining 144:16 /dev/hdb: flushing cache /dev/hdb: stopping de-icing /dev/hdb: writing lead-out /dev/hdb: reloading tray # readprofile -m /boot/System.map-2.6.14-rc3 > captured_profile
Is the problem still present in 2.6.15? > hda: FUJITSU MHV2060AT, ATA DISK drive > hdb: TSSTcorpCD/DVDW TS-L632B, ATAPI CD/DVD-ROM drive Placing HDD and CD on the same channel is not recommended (unfortunately we can do nothing about it because it is a laptop) however I don't know what the actual performance impact can be considered as normal. Maybe the chipset is in the combined mode (I'm just thinking loudly for now as this will need some reading of ICH6 datasheet and maybe adding debugging patch).
Hello Just to confirm that I have experienced the same bug here, on a compal d75 laptop with the same Intel 915GM chipset and DVD burner. I have the same issues with 2.6.8 Debian Sarge kernel and 2.6.15 on Etch. I have tried OpenBSD and it burns fine at 3x speed(just to test and discard a hardware failure). If you need further information I will be pleased to help you. Thank you very much. Ramiro ea1abz@gmail.com
Ramiro, Have you tested with latest kernel? Thanks.
Please reopen this bug if it's still present with kernel 2.6.22.