Re: [PATCH v3 3/4] block: loop: use kthread_work

2015-05-12 Thread Ming Lei
On Tue, May 12, 2015 at 3:25 PM, Christoph Hellwig wrote: > On Mon, May 11, 2015 at 09:12:56PM +0800, Ming Lei wrote: >> For non-AIO case, single thread has been used for long long time, >> and it was just converted to work in v4.0, which has caused performance >> regression for fedora live bootin

Re: [PATCH v3 3/4] block: loop: use kthread_work

2015-05-12 Thread Christoph Hellwig
On Mon, May 11, 2015 at 09:12:56PM +0800, Ming Lei wrote: > For non-AIO case, single thread has been used for long long time, > and it was just converted to work in v4.0, which has caused performance > regression for fedora live booting already. In discussion[1], even though > submitting I/O via wo

Re: [PATCH v3 3/4] block: loop: use kthread_work

2015-05-11 Thread Ming Lei
On Mon, May 11, 2015 at 3:20 PM, Christoph Hellwig wrote: > On Thu, May 07, 2015 at 06:32:03PM +0800, Ming Lei wrote: >> > I can't really parse this, what's the specific advantage here? >> >> Patch 4's commit log provides the test data. >> >> >From the data, it is observed that one thread is enoug

Re: [PATCH v3 3/4] block: loop: use kthread_work

2015-05-11 Thread Christoph Hellwig
On Thu, May 07, 2015 at 06:32:03PM +0800, Ming Lei wrote: > > I can't really parse this, what's the specific advantage here? > > Patch 4's commit log provides the test data. > > >From the data, it is observed that one thread is enough to get > similar throughput with previous one which submits IO

Re: [PATCH v3 3/4] block: loop: use kthread_work

2015-05-07 Thread Ming Lei
On Thu, May 7, 2015 at 3:17 PM, Christoph Hellwig wrote: > On Thu, May 07, 2015 at 01:08:25AM +0800, Ming Lei wrote: >> The following patch will use dio/aio to submit IO to backing file, >> then it isn't good to schedule IO concurrently from work, so >> use kthread_work. > > I can't really parse t

Re: [PATCH v3 3/4] block: loop: use kthread_work

2015-05-07 Thread Christoph Hellwig
On Thu, May 07, 2015 at 01:08:25AM +0800, Ming Lei wrote: > The following patch will use dio/aio to submit IO to backing file, > then it isn't good to schedule IO concurrently from work, so > use kthread_work. I can't really parse this, what's the specific advantage here? -- To unsubscribe from th

[PATCH v3 3/4] block: loop: use kthread_work

2015-05-06 Thread Ming Lei
The following patch will use dio/aio to submit IO to backing file, then it isn't good to schedule IO concurrently from work, so use kthread_work. Signed-off-by: Ming Lei --- drivers/block/loop.c | 79 drivers/block/loop.h | 10 +++ 2 files