| Message ID | 20190909073117.20625-1-paolo.valente@linaro.org |
|---|---|
| Headers | show |
| Series | block, bfq: remove bfq prefix from cgroups filenames | expand |
News of this change? Can we have it (or the solution with the symlinks if you prefer it) for 5.4? Thanks, Paolo > Il giorno 9 set 2019, alle ore 09:31, Paolo Valente <paolo.valente@linaro.org> ha scritto: > > Hi Jens, > now that BFQ's weight interface has been fixed [1], can we proceed > with this change? > > In addition to acking this solution, in [2] Tejun already suggested a > reduced version of the present patch. In Tejun's version, only > bfq.weight is changed. But I guess that legacy code may use also some > of the other bfq parameters in cgroups, without the bfq prefix. Apart > from that, any version is ok for me, provided that it solves the > current confusing situation for userspace [3]. > > Thanks, > Paolo > > [1] https://lkml.org/lkml/2019/8/27/1716 > [2] https://www.mail-archive.com/linux-block@vger.kernel.org/msg35823.html > [3] https://github.com/systemd/systemd/issues/7057 > > Angelo Ruocco (1): > block, bfq: delete "bfq" prefix from cgroup filenames > > block/bfq-cgroup.c | 46 +++++++++++++++++++++++----------------------- > 1 file changed, 23 insertions(+), 23 deletions(-) > > -- > 2.20.1
On 9/16/19 8:56 AM, Paolo Valente wrote: > News of this change? Can we have it (or the solution with the > symlinks if you prefer it) for 5.4? Coordinate with Tejun and bundle the stuff we need into a series, we can definitely put that in 5.4. I did send out the initial pull request for block, but I've got a few things lined up for a secondary pull later this week. -- Jens Axboe
> Il giorno 16 set 2019, alle ore 17:01, Jens Axboe <axboe@kernel.dk> ha scritto: > > On 9/16/19 8:56 AM, Paolo Valente wrote: >> News of this change? Can we have it (or the solution with the >> symlinks if you prefer it) for 5.4? > > Coordinate with Tejun and bundle the stuff we need into a series, Ok. Tejun, could you put your switch-off-io-cost code into a standalone patch, so that I can put it together with this one in a complete series? Thanks, Paolo > we > can definitely put that in 5.4. I did send out the initial pull request > for block, but I've got a few things lined up for a secondary pull > later this week. > > -- > Jens Axboe >
Hello, Paolo. On Mon, Sep 16, 2019 at 05:07:29PM +0200, Paolo Valente wrote: > Tejun, could you put your switch-off-io-cost code into a standalone > patch, so that I can put it together with this one in a complete > series? It was more of a proof-of-concept / example, so the note in the email that the code is free to be modified / used any way you see fit. That said, if you like it as it is, I can surely prep it as a standalone patch. Thanks. -- tejun
> Il giorno 16 set 2019, alle ore 17:16, Tejun Heo <tj@kernel.org> ha scritto: > > Hello, Paolo. > > On Mon, Sep 16, 2019 at 05:07:29PM +0200, Paolo Valente wrote: >> Tejun, could you put your switch-off-io-cost code into a standalone >> patch, so that I can put it together with this one in a complete >> series? > > It was more of a proof-of-concept / example, so the note in the email > that the code is free to be modified / used any way you see fit. That > said, if you like it as it is, I can surely prep it as a standalone > patch. > AFAICT your proposal contains no evident error. Plus, no one seems to have complained about the idea (regardless from the exact implementation). So I guess the best next step is to go for it. Thanks, Paolo > Thanks. > > -- > tejun
On 9/16/19 9:21 AM, Paolo Valente wrote: > > >> Il giorno 16 set 2019, alle ore 17:16, Tejun Heo <tj@kernel.org> ha scritto: >> >> Hello, Paolo. >> >> On Mon, Sep 16, 2019 at 05:07:29PM +0200, Paolo Valente wrote: >>> Tejun, could you put your switch-off-io-cost code into a standalone >>> patch, so that I can put it together with this one in a complete >>> series? >> >> It was more of a proof-of-concept / example, so the note in the email >> that the code is free to be modified / used any way you see fit. That >> said, if you like it as it is, I can surely prep it as a standalone >> patch. >> > > AFAICT your proposal contains no evident error. Plus, no one seems to > have complained about the idea (regardless from the exact > implementation). So I guess the best next step is to go for it. Not filling me with a lot of confidence that you actually tested it? -- Jens Axboe
> Il giorno 16 set 2019, alle ore 18:01, Jens Axboe <axboe@kernel.dk> ha scritto: > > On 9/16/19 9:21 AM, Paolo Valente wrote: >> >> >>> Il giorno 16 set 2019, alle ore 17:16, Tejun Heo <tj@kernel.org> ha scritto: >>> >>> Hello, Paolo. >>> >>> On Mon, Sep 16, 2019 at 05:07:29PM +0200, Paolo Valente wrote: >>>> Tejun, could you put your switch-off-io-cost code into a standalone >>>> patch, so that I can put it together with this one in a complete >>>> series? >>> >>> It was more of a proof-of-concept / example, so the note in the email >>> that the code is free to be modified / used any way you see fit. That >>> said, if you like it as it is, I can surely prep it as a standalone >>> patch. >>> >> >> AFAICT your proposal contains no evident error. Plus, no one seems to >> have complained about the idea (regardless from the exact >> implementation). So I guess the best next step is to go for it. > > Not filling me with a lot of confidence that you actually tested it? > Tested it too. Waiting for Tejun's patch to re-submit it with mine. Thanks, Paolo > -- > Jens Axboe
On Mon, Sep 16, 2019 at 06:45:03PM +0200, Paolo Valente wrote:
> Tested it too. Waiting for Tejun's patch to re-submit it with mine.
http://lkml.kernel.org/r/20190917151308.GH3084169@devbig004.ftw2.facebook.com
Thanks.
--
tejun