* Status of MTD for CFI
@ 2000-06-29 19:45 Sébastien Côté
2000-06-29 21:03 ` David Woodhouse
0 siblings, 1 reply; 7+ messages in thread
From: Sébastien Côté @ 2000-06-29 19:45 UTC (permalink / raw)
To: mtd
Hi,
I'd like to know the status of MTD for CFI devices. If I have a flash
connected to the ISA bus, will I be able to access it with MTD and
install JFFS on it ? If not, what needs to be done ?
--
Sébastien Côté
To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: Status of MTD for CFI
2000-06-29 19:45 Status of MTD for CFI Sébastien Côté
@ 2000-06-29 21:03 ` David Woodhouse
2000-06-30 13:09 ` Sébastien Côté
2000-07-03 23:12 ` Jason Gunthorpe
0 siblings, 2 replies; 7+ messages in thread
From: David Woodhouse @ 2000-06-29 21:03 UTC (permalink / raw)
To: Sébastien Côté; +Cc: mtd
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: TEXT/PLAIN; charset=UTF8, Size: 1720 bytes --]
On Thu, 29 Jun 2000, [iso-8859-1] Sébastien Côté wrote:
> Hi,
>
> I'd like to know the status of MTD for CFI devices. If I have a flash
> connected to the ISA bus, will I be able to access it with MTD and
> install JFFS on it ? If not, what needs to be done ?
If it's a 16-bit device which is capable of the Intel Extended command
set, all you need to do is provide a map driver which provides
read16() and write16() functions for your board. See nora.c, vmax301.c and
octagon-5066.c for examples.
If it's not a 16-bit device, or if it doesn't talk the Intel command set,
then you'll actualy need to code up read/write/erase support for it too.
See cfi_cmdset_0001.c for a vague idea.
If it's something that cfi_cmdset_0001.c will drive (i.e. 16-bit Intel
device), you need to check out the version from the 26th - while it was at
least working, albeit busy waiting for erases and writes. I've since
broken it to get it slightly more well-behaved, and haven't finished that
task yet.
You'll want revision 1.8 of 2000/06/26 16:18:58 - the revision before the
one with the changelog entry that says "First attempt at doing locking and
per-chip access. I'll eat my hat if this works - but it compiles!"
It's work in progress at the moment - but I've had an FTL root filesystem
running off it, and there's no real reason why I shouldn't have had
JFFS-root too if the embedded board that I'm working on wasn't such a PITA
to use.
I'm hoping to get the current CVS code in working order again tomorrow.
Jgg is apparently looking at support for JEDEC flash chips which aren't
CFI-compliant.
--
dwmw2
To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: Status of MTD for CFI
2000-06-29 21:03 ` David Woodhouse
@ 2000-06-30 13:09 ` Sébastien Côté
2000-06-30 13:25 ` David Woodhouse
2000-07-03 23:12 ` Jason Gunthorpe
1 sibling, 1 reply; 7+ messages in thread
From: Sébastien Côté @ 2000-06-30 13:09 UTC (permalink / raw)
To: David Woodhouse; +Cc: mtd
David Woodhouse wrote:
>
> On Thu, 29 Jun 2000, [iso-8859-1] Sébastien Côté wrote:
>
> > Hi,
> >
> > I'd like to know the status of MTD for CFI devices. If I have a flash
> > connected to the ISA bus, will I be able to access it with MTD and
> > install JFFS on it ? If not, what needs to be done ?
>
> If it's a 16-bit device which is capable of the Intel Extended command
> set, all you need to do is provide a map driver which provides
> read16() and write16() functions for your board. See nora.c, vmax301.c and
> octagon-5066.c for examples.
I've got an Intel StrataFlash so it's both 16 bits and CFI compliant. I
quickly checked the files you mentioned and this all seems very simple.
But I've got a few questions : what is nora.c for (from the name, I
guess it's a NOR flash and from the code I guess it's CFI but that's all
I know)? and what does SBC stand for ?
--
Sébastien Côté
To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: Status of MTD for CFI
2000-06-30 13:09 ` Sébastien Côté
@ 2000-06-30 13:25 ` David Woodhouse
0 siblings, 0 replies; 7+ messages in thread
From: David Woodhouse @ 2000-06-30 13:25 UTC (permalink / raw)
To: Sébastien Côté; +Cc: mtd
scote1@Matrox.COM said:
> : what is nora.c for (from the name, I guess it's a NOR flash and from
> the code I guess it's CFI but that's all I know)?
nora.c is misnamed. It's just a 'map' driver for memory chips which are
mapped directly into the CPU's VM, which is what happens on a lot of ARM
boards. It's not really CFI-specific at all, except that it happens to call
the cfi_probe function and none other at the moment.
By changing a couple of defines for the address and size, you can make it
handle memory chips mapped anywhere in VM rather than just where my
particular board happens to have them.
> and what does SBC stand for ?
Single Board Computer.
scote1@Matrox.COM said:
> I've got an Intel StrataFlash so it's both 16 bits and CFI compliant.
> I quickly checked the files you mentioned and this all seems very
> simple.
It just got less simple. We now make an attempt to do proper async handling
of flash chips, and I'm just about to test it. If it works, I'll try to
build up the courage to shift the erase stuff into a timer, and enable
suspension of in-progress erases.
--
dwmw2
To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: Status of MTD for CFI
2000-06-29 21:03 ` David Woodhouse
2000-06-30 13:09 ` Sébastien Côté
@ 2000-07-03 23:12 ` Jason Gunthorpe
2000-07-04 7:31 ` David Woodhouse
1 sibling, 1 reply; 7+ messages in thread
From: Jason Gunthorpe @ 2000-07-03 23:12 UTC (permalink / raw)
To: David Woodhouse; +Cc: Sébastien Côté, mtd
On Thu, 29 Jun 2000, David Woodhouse wrote:
> Jgg is apparently looking at support for JEDEC flash chips which aren't
> CFI-compliant.
I popped my current codes up at http://amber.deltatee.com/~jgg/mtd/jedec/
I've never actually tried it in the kernel yet, but the logic all appears
correct and it is in the right sort of format, have been using it to flash
kernels into the SIMM for booting this board. It needs testing and fixing
to support 8bit interfaces, I have to steal an octagon board so I can test
that.
If someone desperately needs this then they are certainly welcome to
fiddle with it.
It supports most of the old style AMD uniform sector size chips, maybe
others. Works on my 16 chip banked flash SIMM, which is the most
complicated situation possible!
Anyhow, I am going on a trip soon to Florida and then up to OLS.. after
that I need to do a bunch of VHDL, but I should have time to finish it
off and write a JFFS bootloader.
Thanks,
Jason
To unsubscribe, send "unsubscribe mtd" to majordomo@infradead.org
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2000-07-04 15:40 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2000-06-29 19:45 Status of MTD for CFI Sébastien Côté
2000-06-29 21:03 ` David Woodhouse
2000-06-30 13:09 ` Sébastien Côté
2000-06-30 13:25 ` David Woodhouse
2000-07-03 23:12 ` Jason Gunthorpe
2000-07-04 7:31 ` David Woodhouse
2000-07-04 15:39 ` Jason Gunthorpe
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox