Re: CFQ timer precision

Hi Jeff, On Mon 16-11-15 12:23:05, Jeff Moyer wrote: > Jens Axboe writes: > > On 11/16/2015 08:11 AM, Jan Kara wrote: > >> Hello, > >> > >> lately I was looking into a big performance hit we take when blkio > >> controller is enabled and jbd2 thread ends up in a different cgroup than > >> user

Re: CFQ timer precision

Hi Jeff, On Mon 16-11-15 12:23:05, Jeff Moyer wrote: > Jens Axboe writes: > > On 11/16/2015 08:11 AM, Jan Kara wrote: > >> Hello, > >> > >> lately I was looking into a big performance hit we take when blkio > >> controller is enabled and jbd2 thread ends up in a different

Re: CFQ timer precision

Hi Jan, Jens Axboe writes: > On 11/16/2015 08:11 AM, Jan Kara wrote: >> Hello, >> >> lately I was looking into a big performance hit we take when blkio >> controller is enabled and jbd2 thread ends up in a different cgroup than >> user process. E.g. dbench4 throughput drops from ~140 MB/s to

Re: CFQ timer precision

On 11/16/2015 08:11 AM, Jan Kara wrote: Hello, lately I was looking into a big performance hit we take when blkio controller is enabled and jbd2 thread ends up in a different cgroup than user process. E.g. dbench4 throughput drops from ~140 MB/s to ~20 MB/s. However artificial dbench4 is, this

CFQ timer precision

Hello, lately I was looking into a big performance hit we take when blkio controller is enabled and jbd2 thread ends up in a different cgroup than user process. E.g. dbench4 throughput drops from ~140 MB/s to ~20 MB/s. However artificial dbench4 is, this kind of drop will likely be clearly

Re: CFQ timer precision

On 11/16/2015 08:11 AM, Jan Kara wrote: Hello, lately I was looking into a big performance hit we take when blkio controller is enabled and jbd2 thread ends up in a different cgroup than user process. E.g. dbench4 throughput drops from ~140 MB/s to ~20 MB/s. However artificial dbench4 is, this

CFQ timer precision

Hello, lately I was looking into a big performance hit we take when blkio controller is enabled and jbd2 thread ends up in a different cgroup than user process. E.g. dbench4 throughput drops from ~140 MB/s to ~20 MB/s. However artificial dbench4 is, this kind of drop will likely be clearly

Re: CFQ timer precision

Hi Jan, Jens Axboe writes: > On 11/16/2015 08:11 AM, Jan Kara wrote: >> Hello, >> >> lately I was looking into a big performance hit we take when blkio >> controller is enabled and jbd2 thread ends up in a different cgroup than >> user process. E.g. dbench4 throughput drops