Re: [PATCH][next] loop: Fix missing max_active argument in alloc_workqueue call

On 3/19/21 3:59 AM, Krzysztof Kozlowski wrote:

On 18/03/2021 21:42, Jens Axboe wrote:

quoted

On 3/18/21 2:24 PM, Colin Ian King wrote:

quoted

On 18/03/2021 20:12, Jens Axboe wrote:

quoted

On 3/18/21 9:16 AM, Colin King wrote:

quoted

From: Colin Ian King <redacted>

The 3rd argument to alloc_workqueue should be the max_active count,
however currently it is the lo->lo_number that is intended for the
loop%d number. Fix this by adding in the missing max_active count.

Dan, please fold this (or something similar) in when you're redoing the
series.

Appreciate this fix being picked up. Are we going to lose the SoB?

If it's being redone, would be silly to have that error in there. Do
we have a tag that's appropriate for this? I often wonder when I'm
folding in a fix. Ala Fixes-by: or something like that.

Why it is being redone if it was put into next? And even then, several
other maintainers just apply a fix on top (I think Andrew Morton, Greg
KH, Mark Brown) to avoid rebasing, preserve the history and also give
credits to the fixer.

linux-next doesn't have continual history, and I rebase my for-next
all the time. Since the series was going to be re-done and applied to
a different tree even, it would be silly to retain a bug _just_ so that
we can have credits to the fixer separately. For this case, it's
rebased anyway, and there's honestly not any history to preserve here.
The only downside is losing the fixer attribution, which I do agree is
annoying and hence why I asked/lobbied in a fixes-by kind of tag.

-- 
Jens Axboe