svn commit: r287876 - in projects/iosched/sys/cam: . ata scsi

[Bruce Simpson]

On 17/09/15 15:16, Warner Losh wrote:
> This isn't even in -current yet, but this will work in 10.x. It's
> dependent, at the moment,
> on the I/O scheduler work I've done in this project branch. There's
> nothing intrinsic about
> that.

That is good news; the bug specifically affected NCQ'd TRIM commands. 
Performing TRIM manually is an admin headache and gets old, fast.

> I'm not sure I understand your comment about Linux or the self-booting
> stuff. The new
> firmware for the M500's that has working NCQ trim has existed for less
> than a week...

Crucial have had updates to fix the NCQ TRIM bug on other drives (e.g. 
MX100) for several months now.

My point is that Crucial have done the right thing in how they delivered 
the patch for these drives; it "just works" out of the box, but might 
need to be adapted to e.g. PXE based provisioning environments for 
automated rollout.

However, because it's a plain Linux based image, such mass updates are 
actually possible. From experience, updating firmware on e.g. HP gear is 
anything but straightforward. For example, I have had HP ROMPaqs fail to 
generate bootable USB media even when run natively under Windows.

There are loose guides "out there" about how to extract ROMPaqs and set 
up the correct FreeDOS bootable USB media under Linux, but these need to 
be adapted for each ROMPaq. That's really for HP to fix, but they could 
do worse than follow Crucial's lead here.

Bruce