mutter, mutter

The FAQ says:
Quote:
The last four ide_data_test lines should have the same wrote and read values if things are working correctly. For example, here is a screen showing a normal 'good' boot:

Code:

ide_data_test: wrote 0x5555 read 0x0080
ide_data_test: wrote 0x0000 read 0x0000
ide_data_test: wrote 0xffff read 0xffff
ide_data_test: wrote 0xaaaa read 0xaaaa
ide_data_test: wrote 0x5555 read 0x5555
hda: IBM-DJSA-230, ATA DISK drive


And yet JeffreyB has:
Code:

ide_data_test: wrote 0x0000 read 0xffa5
ide_data_test: wrote 0xffff read 0xffa5
ide_data_test: wrote 0xaaaa read 0xaaa5
ide_data_test: wrote 0x5555 read 0x55a5
hda: ST9100823A, ATA DISK drive


and both our empegs say:
Code:

ide_data_test: wrote 0x0000 read 0xffff
ide_data_test: wrote 0xffff read 0xffff
ide_data_test: wrote 0xaaaa read 0xaaaa
ide_data_test: wrote 0x5555 read 0x5555
hda: FUJITSU MHL2300AT, ATA DISK drive


It's easy to miss the bit of the FAQ that suggests that this is only a problem if you're having HD problems.

So may I ask:
* why do I get wrote 0x0000 read 0xffff on both my machines?
* should the FAQ emphasise that ide_data_test is not an error but only a diagnostic for other symptoms. In other words don't start pulling your ide cables apart just to cure odd ide_data_test messages?

Possibly replace:

However, if you are having hard disk trouble, these messages can be used to help diagnose the problem.

with:

However, if (and only if) you are having hard disk trouble, these messages can be used to help diagnose the problem. You should not treat these messages as a problem even if the last few wrote/read pairs show different values and certainly shouldn't start playing with IDE cables (and potentially causing real damage) without real symptoms.

I read Marks posts on the subject and looked at the code in ide-probe but was none the wiser.
I also changed the udelay(1) to udelay(2) and added in an extra call to ide_data_test(drive, 0x0000) after the others.

I always get
ide_data_test: wrote 0x0000 read 0xffff
but now I get more random values between the outw() and inw() so I'm putting it back --- eek!
_________________________
LittleBlueThing Running twin 30's