[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Normal buildtime for a kernel?



Giacomo Mulas schrieb:
Have a look at whether DMA is enabled in the disk driver you are using. If not, high disk throughput will cost you a lot of CPU power. You can find out by looking at the kernel messages at boot, for example on one of my boxes I see lines like

ata1: SATA max UDMA/133 cmd 0xFFFFFF0000004200 ctl 0xFFFFFF0000004238 bmdma 0x0 irq 18

which tells me that my serial ata disk is capable of (and using) dma.
Another thing which may make a difference is the io-scheduler you are using, if your computer is heavily loaded with other disk-intensive tasks when you are compiling your kernel (e.g. some cron jobs run extensive find commands). Be sure to compile the as or cfq io scheduler(s) in your kernel (or as modules) and to pass the "elevator=as" or "elevator=cfq" arguments
to your kernel boot sequence, this may help.

I'm using it as a desktop, so I guess the scheduling is not the problem.

SATA output:

libata version 1.10 loaded.
sata_sil version 0.8
ACPI: PCI Interrupt 0000:00:12.0[A] -> GSI 22 (level, low) -> IRQ 185
ata1: SATA max UDMA/100 cmd 0xFFFFC20000004080 ctl 0xFFFFC2000000408A bmdma 0xFFFFC20000004000 irq 185 ata2: SATA max UDMA/100 cmd 0xFFFFC200000040C0 ctl 0xFFFFC200000040CA bmdma 0xFFFFC20000004008 irq 185 ata1: dev 0 cfg 49:2f00 82:346b 83:7d01 84:4023 85:3469 86:3c01 87:4023 88:407f
ata1: dev 0 ATA, max UDMA/133, 488397168 sectors: lba48
ata1: dev 0 configured for UDMA/100
scsi0 : sata_sil
ata2: no device found (phy stat 00000000)
scsi1 : sata_sil
  Vendor: ATA       Model: ST3250823AS       Rev: 3.01
  Type:   Direct-Access                      ANSI SCSI revision: 05
ACPI: No ACPI bus support for 0:0:0:0
SCSI device sda: 488397168 512-byte hdwr sectors (250059 MB)
SCSI device sda: drive cache: write back
SCSI device sda: 488397168 512-byte hdwr sectors (250059 MB)
SCSI device sda: drive cache: write back
 sda: sda1 sda2 < sda5 sda6 >
Attached scsi disk sda at scsi0, channel 0, id 0, lun 0

What can you read out of this? (straight from dmesg)

Thanks and with kind regards,
Oliver Korpilla



Reply to: