Re: Normal buildtime for a kernel?
Lennart Sorensen schrieb:
On Wed, Mar 23, 2005 at 03:32:15PM +0100, Oliver Korpilla wrote:
Yeah, but slower??
Well at least very similar speeds seem not too unreasonable. I am not
sure if the 64bit pointers would hurt the performance of gcc in any way.
It does have to move more bytes around than a 32bit machine. it is
certainly faster at many tasks.
I don't know if this problem wouldn't be handled by the bus architecture
of the AMD64 itself. Remember it has also more regs and an integrated
memory controller, that should actually give a boost.
I already do -j 3.
I wonder if 3 is overkill.
I did with it and without it - it's barely noticeable.
Well what does hdparm -t /dev/sda say your disk speed is able to do?
What kind of HD and what size?
250GB Seagate SATA.
# hdparm -v /dev/sda
/dev/sda:
IO_support = 0 (default 16-bit)
readonly = 0 (off)
readahead = 256 (on)
geometry = 30401/255/63, sectors = 250059350016, start = 0
That doesn't say much interesting for sata drives. -t would have been
more interesting.
hdparm -t /dev/sda
/dev/sda:
Timing buffered disk reads: 98 MB in 3.06 seconds = 32.01 MB/sec
HDIO_DRIVE_CMD(null) (wait for flush complete) failed: Inappropriate
ioctl for device
root@bender:/home/korpo/download# hdparm -t /dev/sda
/dev/sda:
Timing buffered disk reads: 98 MB in 3.01 seconds = 32.52 MB/sec
HDIO_DRIVE_CMD(null) (wait for flush complete) failed: Inappropriate
ioctl for device
root@bender:/home/korpo/download# hdparm -t /dev/sda
/dev/sda:
Timing buffered disk reads: 98 MB in 3.04 seconds = 32.26 MB/sec
HDIO_DRIVE_CMD(null) (wait for flush complete) failed: Inappropriate
ioctl for device
Thanks and with kind regards,
Oliver Korpilla
Reply to: