On Nov 7, 2007 1:55 PM, Tejun Heo <htejun@xxxxxxxxx> wrote:Florian La Roche wrote:Hello all,Thanks. We're currently trying to find out what's actually going on
I've taking email addresses from the last NCQ blacklist changes going
into the kernel.
This Fujitsu drive also gives me spurious command completions. Detailed
output also available at https://bugzilla.redhat.com/show_bug.cgi?id=366181.
Let me know if you need more info or anything else.
--- drivers/ata/libata-core.c
+++ drivers/ata/libata-core.c
@@ -4222,6 +4222,7 @@
{ "WDC WD740ADFD-00NLR1", NULL, ATA_HORKAGE_NONCQ, },
{ "WDC WD3200AAJS-00RYA0", "12.01B01", ATA_HORKAGE_NONCQ, },
{ "FUJITSU MHV2080BH", "00840028", ATA_HORKAGE_NONCQ, },
+ { "FUJITSU MHW2160BJ G2", NULL, ATA_HORKAGE_NONCQ },
{ "ST9120822AS", "3.CLF", ATA_HORKAGE_NONCQ, },
{ "ST9160821AS", "3.CLF", ATA_HORKAGE_NONCQ, },
{ "ST9160821AS", "3.ALD", ATA_HORKAGE_NONCQ, },
with all these drives. At first, drives which got blacklisted aren't
many and made sense (had other problems with NCQ, etc..) but with new
generation drives from many vendors showing the same symptom, we aren't
too sure now.
Is there a way to tell whether Windows is using NCQ or not? I checked
the system log (or whatever it's called) on my notebook and is clean
but I'm not sure it's using NCQ (I don't even know if it'd log
spurious completions somewhere).