This message is from the T13 list server.
Referring to Jeff's recommendation:
1) Retry the cycle in PIO Mode [0], but leave normal
transfers at ATA UDMA Mode [N]
2) If you get more than 6 CRCs, then shift to UDMA Mode [N-1]
retry the cycle in PIO Mode [0], but normal
data transfers will now use mode [N-1]
3) If you get more than 6 CRCs, and you are already down
to UDMA Mode [0], then downshift to PIO Mode [0]
Why not switch to MWDMA ?
Several reasons:
MWDMA does not have CRC checking
UDMA Mode [0], generally has more than TWICE
the Valid window (setup + hold) over
MWDMA Mode [2] even though both have a sustained
transfer rate of 16 MB/s.
MKE's response:
I would suggest that if you are retrying each failure at PIO, you are just
holding off the inevitable. Assuming you don't do the PIO retry, if you
wait for 6 CRC errors you will have a good chance that the OS would have
already bug-checked (blue-screen, if you will) and is way to lenient.
Hale, I have seen successive transfers (same transfer retried) fail multiple
times, and some OS's will crash on the third retry. I would agree that it
is due to bad cabling, at least in the cases that I test under where I use
40-conductor cables to emulate the situation. MWDMA is a bad choice, as is
SWDMA too because it has no CRC checking and lots of people have poor MWDMA
implementations and many do not have SWDMA.
But why would you go all the way to PIO-0? Wouldn't it be prudent to go
from PIO-4, then step down to 0?
MKE.
-----Original Message-----
From: Hale Landis [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, December 11, 2001 7:14 PM
To: [EMAIL PROTECTED]
Subject: RE: [t13] UDMA 'recovery procedure'
This message is from the T13 list server.
On Tue, 11 Dec 2001 13:52:35 -0600, Wolford, Jeff wrote:
>1) retrying the cycle using the same transfer mode is NOT very useful
Just an observation... In many hundreds of hours of doing nothing but
random write/read testing using UDMA on properly configured systems I
have seen two (2!) CRC errors and in both cases the errors did not
happen on the immediate retry of the command. In my opinion if you
are seeing a lot of CRC errors then you have a badly configured
hardware environment, probably a bad cable. Switching to MW DMA or
PIO might just expose you to undetected data transfer errors.
*** Hale Landis *** [EMAIL PROTECTED] ***
*** Niwot, CO USA *** www.ata-atapi.com ***
Subscribe/Unsubscribe instructions can be found at www.t13.org.
Subscribe/Unsubscribe instructions can be found at www.t13.org.
