Re: [GIT PULL] Block fixes for 4.19-final

2018-10-19 Thread Greg Kroah-Hartman
On Fri, Oct 19, 2018 at 09:52:39AM -0600, Jens Axboe wrote: > Hi Greg, > > Two small fixes that should go into this release. Please pull! > > > git://git.kernel.dk/linux-block.git tags/for-linus-20181019 > Now merged, thanks. greg k-h

Re: [PATCH v2] block: Bundle together NONE and NOT_SET writehint types

2018-10-19 Thread Jens Axboe
On 10/19/18 5:19 AM, Avri Altman wrote: > There are 6 writehint types, but the size of the write_hints array > in the request queue is BLK_MAX_WRITE_HINTS = 5, > which causes the EXTREME type to be ignored when iterating > over the hints. > > We only have effectively 5 hints. > Bundle together

Re: [PATCH] block: setup bounce bio_sets properly

2018-10-19 Thread Jens Axboe
On 10/19/18 3:10 AM, Ming Lei wrote: > On Thu, Oct 18, 2018 at 03:24:47PM -0600, Jens Axboe wrote: >> We're only setting up the bounce bio sets if we happen >> to need bouncing for regular HIGHMEM, not if we only need >> it for ISA devices. >> >> Reported-by: Ondrej Zary >> Tested-by: Ondrej Zary

[GIT PULL] Block fixes for 4.19-final

2018-10-19 Thread Jens Axboe
Hi Greg, Two small fixes that should go into this release. Please pull! git://git.kernel.dk/linux-block.git tags/for-linus-20181019 Jens Axboe (1): Merge branch 'nvme-4.19' of git://git.infradead.org/nvme into for-linus

Re: [GIT PULL] nvme updates for Linux 4.20

2018-10-19 Thread Jens Axboe
On 10/19/18 9:35 AM, Christoph Hellwig wrote: > The second batch of updates for Linux 4.20: > - lot of fixes for issues found by static type checkers from Bart > - two small fixes from Keith > - fabrics cleanups in preparation of the TCP transport from Sagi > - more cleanups from Chaitanya

[GIT PULL] nvme updates for Linux 4.20

2018-10-19 Thread Christoph Hellwig
The second batch of updates for Linux 4.20: - lot of fixes for issues found by static type checkers from Bart - two small fixes from Keith - fabrics cleanups in preparation of the TCP transport from Sagi - more cleanups from Chaitanya The following changes since commit

Re: [PATCH v2] block: BFQ default for single queue devices

2018-10-19 Thread Bryan Gurney
On Fri, Oct 19, 2018 at 9:44 AM, Johannes Thumshirn wrote: > On 19/10/18 15:36, Bryan Gurney wrote: >> I like this idea. I don't have enough experience to write this patch >> myself, but I imagine something like adding "bool is_zoned_aware" to >> "struct elevator_type", and setting that true

Re: [PATCH v2] block: BFQ default for single queue devices

2018-10-19 Thread Johannes Thumshirn
On 19/10/18 15:36, Bryan Gurney wrote: > I like this idea. I don't have enough experience to write this patch > myself, but I imagine something like adding "bool is_zoned_aware" to > "struct elevator_type", and setting that true only for the schedulers > that are currently zoned-device aware

Re: [PATCH v2] block: BFQ default for single queue devices

2018-10-19 Thread Bryan Gurney
On Fri, Oct 19, 2018 at 4:42 AM, Linus Walleij wrote: > > On Wed, Oct 17, 2018 at 4:59 PM Bryan Gurney wrote: > > > I feel strongly about the prevention of users running into errors > > because of an incorrect scheduler default, because I encountered that > > situation three times in my testing

[PATCH v2] block: Bundle together NONE and NOT_SET writehint types

2018-10-19 Thread Avri Altman
There are 6 writehint types, but the size of the write_hints array in the request queue is BLK_MAX_WRITE_HINTS = 5, which causes the EXTREME type to be ignored when iterating over the hints. We only have effectively 5 hints. Bundle together NONE and NOT_SET, since they end up in the same bucket

Re: [PATCH v2] block: BFQ default for single queue devices

2018-10-19 Thread Paolo Valente
> Il giorno 15 ott 2018, alle ore 20:26, Paolo Valente > ha scritto: > > ... >> This kind of policy does not belong in the kernel, at least >> not in the current form. If we had some sort of "enable best >> options for a desktop" then it could fall under that umbrella. >> > > I don't think

Re: [PATCH v2] block: BFQ default for single queue devices

2018-10-19 Thread Oleksandr Natalenko
Hi. On 19.10.2018 10:33, Linus Walleij wrote: > + /* > + * Zoned devices must use a deadline scheduler because currently > + * that is the only scheduler respecting zoned writes. > + */ > + if (blk_queue_is_zoned(q)) > + policy = "mq-deadline"; > + else if

Re: [PATCH] block: setup bounce bio_sets properly

2018-10-19 Thread Ming Lei
On Thu, Oct 18, 2018 at 03:24:47PM -0600, Jens Axboe wrote: > We're only setting up the bounce bio sets if we happen > to need bouncing for regular HIGHMEM, not if we only need > it for ISA devices. > > Reported-by: Ondrej Zary > Tested-by: Ondrej Zary > Signed-off-by: Jens Axboe > > diff

Re: [PATCH v2] block: BFQ default for single queue devices

2018-10-19 Thread Linus Walleij
On Wed, Oct 17, 2018 at 4:59 PM Bryan Gurney wrote: > I feel strongly about the prevention of users running into errors > because of an incorrect scheduler default, because I encountered that > situation three times in my testing with zoned block devices. The > switch to SCSI_MQ would resolve

Re: [PATCH v2] block: BFQ default for single queue devices

2018-10-19 Thread Linus Walleij
On Mon, Oct 15, 2018 at 4:32 PM Oleksandr Natalenko wrote: > On 15.10.2018 16:10, Linus Walleij wrote: > > + /* > > + * Zoned devices must use a deadline scheduler because currently > > + * that is the only scheduler respecting zoned writes. > > + */ > > + if

Re: [PATCH v2] block: BFQ default for single queue devices

2018-10-19 Thread Pavel Machek
Hi! > >> Which is also the approach that I've been advocating for here, instead > >> of a kernel patch... > > > > I know you've been advocating the use of udev for IO scheduler selection. > > But do you want to force everybody to use udev? And for people who build > > their own (usually small)