答复: USB Boot

[YhLu]

OHCI usb boot ok now.

But it is some slow, only 173KB/S,

Full speed should be 12Mbps/8 = 1500KB/s

Regards

YH

Mounted fat
Image checksum: 000047da
Loading Linux version 2.6.5 (root@tst288xsuse9) #11 SMP Mon May 10 14:58:17
PDT 2004...
segment 1 addr:0x00010000 file:0x0000543c mem:0x0001a764 phdr:0xefe86654
loading... clearing... ok
segment 2 addr:0x00091000 file:0x00000000 mem:0x00000070 phdr:0xefe86674
loading... clearing... ok
segment 3 addr:0x00100000 file:0x001f5a40 mem:0x00700000 phdr:0xefe86694
loading... clearing... ok
segment 4 addr:0x00800000 file:0x005c418e mem:0x005c418e phdr:0xefe866b4
loading... clearing... ok
Loaded 8122378 bytes in 46722ms (173KB/s

-----邮件原件-----
发件人: YhLu 
发送时间: 2004年6月1日 17:53
收件人: Steven James
抄送: ron minnich; Jay Miller; [e-mail]; LinuxBIOS
主题: USB Boot

Some good news.

In OHCI, bulk transfer works well.

Status: it can read sector 0 and sector 68, after that it will reboot the
system.

Regards

YH
_______________________________________________
Linuxbios mailing list
[e-mail]
http://www.clustermatic.org/mailman/listinfo/linuxbios

YhLu  writes:
> OHCI usb boot ok now.
>
> But it is some slow, only 173KB/S,
>
> Full speed should be 12Mbps/8 = 1500KB/s

Congratulations.

Eric

USB Boot

[Peter Stuge]

On Tue, Jun 01, 2004 at 09:42:20PM -0700, YhLu wrote:
> OHCI usb boot ok now.

Agree with Eric, congratulations! :)
> But it is some slow, only 173KB/S,
>
> Full speed should be 12Mbps/8 = 1500KB/s

Actually not. 12Mbps is the speed on the USB wire and each of the
protocols in the stack and in the application causes some overhead,
as you know.

I don't know much specifically about USB storage but I'd guesstimate
the maximum possible throughput somewhere between 1000kb/s and
1400kb/s. The simplest optimization is to increase the block
transfer size to always try to transfer full 64kb, if that hasn't
been done already.

Great to have a complete USB boot solution anyway!

//Peter