> When burning a DVD-R with growisofs, the reported write speed is 4.00x, > but the measured speed is only 0.40x. DMA is on and -20 nicing > growisofs and mkisofs has no effect.
It's hardly an application problem, rather a kernel one. growisofs does work directly with ide-cd even under 2.4.x and I've received few reports from users who *successfully* use it without ide-scsi. I don't know if they are on this list though... > Using ide-scsi, I can easily get > 4.00x measured speed. However, until ide-scsi stops causing kernel > panics, I'd like to avoid it. Formally this is wrong list to report problems with Linux kernel, but it might make sense to cross-post the next/thorough report to both kernel development and this list. Once agian to make it clear. It makes sense to cross-post, because subscribers of this list might help to identify triggering factor, but you shouldn't expect actual solutions being suggested here. > My setup is as follows: > > Linux Kernel - 2.4.21 + bootsplash patch + updated lm-sensors, bttv, > and v4l2 + loopjari crypto > Processor - Athlon XP 1.26Ghz It's more important to state what chipset do you have than processor type. Complement report with lspci output and relevant portion from your .config (there is a number of options for different IDE controllers). > DVD+-RW Drive - Vendor: PIONEER Model: DVD-RW DVR-106D Rev: 1.07 > > For the ide-scsi, the last thing in the log is always of the form: > > [kernel] scsi : aborting command due to timeout : pid 555338, scsi0, > channel 0, id 0, lun 0 0x35 00 00 00 00 00 00 00 00 00 This alone doesn't indicate a kernel crash. If you really don't get any panic message on console (it doesn't necessarily end up in log file), then you should *explicitly* state that. If you get panic message, you should a) make sure kernel is compiled with CONFIG_FRAME_POINTER, b) collect the panic message, c) process it with ksymoops if you run 2.4.x. A. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
