Re: [PATCH-RFC] (was: Re: Sata Sil3512 bug?; Promise SATA300 TX4)

Re: [PATCH-RFC] (was: Re: Sata Sil3512 bug?; Promise SATA300 TX4)

[Mikael Pettersson]

On Sat, 27 Oct 2007 17:24:32 +0400, Alexander Sabourenkov wrote:
> There appears to be a hardware bug in that it chokes on scatterlist
> if the last item is larger than 164 bytes.
> 
> The patch that follows fixes my problem on 2.6.22.
> 
> I can't think of a way to avoid second pass over scatterlist without
> duplicating code (ata_qc_prep() and ata_fill_sg() from libata-core.c).
> 
> 
> --- a/drivers/ata/sata_promise.c	2007-07-09 03:32:17.000000000 +0400
> +++ b/drivers/ata/sata_promise.c	2007-10-27 17:20:03.000000000 +0400

Interesting, but can you please give more background information?
I.e., how did you determine the existence of this bug?

And please cc: the sata_promise maintainer on sata_promise patches.
(Hint: that's me)

And please choose a Subject: that makes it absolutely clear what
the post is about. "Sata Sil3512 bug" doesn't exactly sound like
something the sata_promise maintainer needs to look at.

Meanwhile I'll check the Promise driver(s) to see if there's
something about SG table formatting restrictions there.

/Mikael

Re: [PATCH-RFC] Promise SATA300 TX4

[Alexander Sabourenkov]

Hello.

> Interesting, but can you please give more background information?
> I.e., how did you determine the existence of this bug?

I was experiencing read/write errors, with data corruption.
Dmesgs and lspcis are at http://lxnt.info/linux/

I downloaded driver from the promise.com, put up 2.6.11 and compiled it.
It worked ok.

I then unsuccessfully tried to port the driver to 2.6.22.
While I failed, this gave me enough insights into ATA workings, that I
could read the CAM module that comes with the driver.

Looking through it I found the SG_COUNT_ASIC_BUG constant and
corresponding code. I decided to try and implement the workaround in
sata_promise.c - and this fixed the problem.

> And please cc: the sata_promise maintainer on sata_promise patches.
> (Hint: that's me)

Hmm. sata_promise.c says:

*  Maintained by:  Jeff Garzik <jgarzik@pobox.com>

> Meanwhile I'll check the Promise driver(s) to see if there's
> something about SG table formatting restrictions there.

Take a look at:

cam_ata.c:6190
cam_ata.c:6259

The fix is also conditioned on sg segment length == 0, which I did not
implement. Is this at all possible in libata ?

-- 

./lxnt

Re: [PATCH-RFC] Promise SATA300 TX4

[Jeff Garzik]

Alexander Sabourenkov wrote:
> Hmm. sata_promise.c says:
> 
> *  Maintained by:  Jeff Garzik <jgarzik@pobox.com>

That should be changed :)  I wrote the code but Mikael is now the 
primary maintainer.


>> Meanwhile I'll check the Promise driver(s) to see if there's
>> something about SG table formatting restrictions there.
> 
> Take a look at:
> 
> cam_ata.c:6190
> cam_ata.c:6259
> 
> The fix is also conditioned on sg segment length == 0, which I did not
> implement. Is this at all possible in libata ?

The block layer should never pass us a scatterlist element of length zero.

	Jeff