On 6/25/25 3:18 PM, Christoph Hellwig wrote:
> On Wed, Jun 25, 2025 at 03:14:51PM +0900, Damien Le Moal wrote:
>> On 6/25/25 3:12 PM, Christoph Hellwig wrote:
>>> On Wed, Jun 25, 2025 at 02:59:05PM +0900, Damien Le Moal wrote:
+bool bio_needs_zone_write_plugging(struct bio *bio)
>>>
>>> Can yo
On Wed, Jun 25, 2025 at 03:14:51PM +0900, Damien Le Moal wrote:
> On 6/25/25 3:12 PM, Christoph Hellwig wrote:
> > On Wed, Jun 25, 2025 at 02:59:05PM +0900, Damien Le Moal wrote:
> >> +bool bio_needs_zone_write_plugging(struct bio *bio)
> >
> > Can you use this in blk_zone_plug_bio instead of dupl
On 6/25/25 3:12 PM, Christoph Hellwig wrote:
> On Wed, Jun 25, 2025 at 02:59:05PM +0900, Damien Le Moal wrote:
>> +bool bio_needs_zone_write_plugging(struct bio *bio)
>
> Can you use this in blk_zone_plug_bio instead of duplicating the logic?
I thought about doing that, but we would still need to
On Wed, Jun 25, 2025 at 02:59:05PM +0900, Damien Le Moal wrote:
> +bool bio_needs_zone_write_plugging(struct bio *bio)
Can you use this in blk_zone_plug_bio instead of duplicating the logic?
I also wonder if we should only it it, as despite looking quite complex
it should compile down to just a f
In preparation for fixing device mapper zone write handling, introduce
the helper function bio_needs_zone_write_plugging() to test if a BIO
requires handling through zone write plugging. This function returns
true for any write operation to a zoned block device. For zone append
opertions, true is r
