11 Oct
2007
11 Oct
'07
6:44 a.m.
On Thu, 2007-10-11 at 11:52 +0700, Patrick Shirkey wrote:
On Wed, 2007-10-10 at 06:37 -0400, Forest Bond wrote:
if you use your own compiled code for the copy, you can open the
destination files with O_SYNC. from man 2 open:
O_SYNC The file is opened for synchronous I/O. Any write()s on
the
resulting file descriptor will block the calling process
until
the data has been physically written to the underlying
hardware.
See RESTRICTIONS below, though.
this avoids having the entire mount be done with the sync option.
Hi,
On Wed, Oct 10, 2007 at 09:51:35AM +0700, Patrick Shirkey wrote:
For the purpose of this script I am happy with just write but I will be
needing read in the near future.
FYI, I can connect 200 usb-storage devices to a dual cpu intel with 4 GB
RAM and an extra pci-usb card on the same irq as the internal usb bus.
Basically I am uploading content to a couple of hundred mp3/multi media
players at the same time and this script is used by other people with
less Linux knowledge than I so I'm trying to make the progress bar as
accurate as possible. Otherwise there is a significant lag between when
the progress bar says it is finished to when the transfer is actually
finished. Speed is also a major issue here so I cannot use the sync
option as that causes way too much delay compared to async.
Can anyone here shed a little more light on how to monitor the disk
activity while transferring data? On Tue, 2007-10-09 at 09:16 -0400, Forest Bond
wrote:
But only for writes, no? On Tue, Oct 09, 2007 at 01:11:38PM +0200, Lars
Luthman wrote:
> Slightly-slower-hack: add the 'sync' option to whatever script or config
> file is used to mount the device (/etc/fstab, udev rules, KDE service
> menu etc). That way all writes will be synchronised automatically.
It sounds like this doesn't matter much for you, but keep in mind that -o
sync doesn't cause reads to be unbuffered, so don't go trying to benchmark
your flash devices that way!
Hmm, I need speed and accuracy.