Skip to contentSkip to conversation

AMD & Fujitsu CFI driver problem

DISCUSSION Dmitry Konyshev 2 participants 3 messages Apr 21, 2004 Memory technology devices (MTD)

lore
Share & tools
lore
#0 lore raw

Hi there,

I have troubles with current AMD & Fujitsu CFI driver. When it works
with 2 x16 devices in 32-bit mode, it fails to correctly detect ending of erasure operation. Looks like that when one device reports the end of operation by setting all 16 its bits in 1, the other device is not done yet, and the code in chip_status() returns CHIP_TIMEDOUT because it doesn't consider such a situation.

I wonder if there's a fix for this issue already or anybody's gonna
fix it soon.

wbr, Dmitry

#1 lore raw
Quoted 3 lines · Dmitry Konyshev

I have troubles with current AMD & Fujitsu CFI driver. When it works
with 2 x16 devices in 32-bit mode, it fails to correctly detect ending of erasure operation. Looks like that when one device reports the end of operation by setting all 16 its bits in 1, the other device is not done yet, and the code in chip_status() returns CHIP_TIMEDOUT because it doesn't consider such a situation.

That's correct. chip_status() is broken for interleaved chips. Which is why the latest version (1.98) in CVS no longer uses it. Update and see if that fixes the issue.

David Vrabel

#2 lore raw
Quoted 2 lines · David Vrabel

That's correct. chip_status() is broken for interleaved chips. Which is why the latest version (1.98) in CVS no longer uses it. Update and see if that fixes the issue.

Thanks a lot, it actually fixes the issue.

wbr, Dmitry