Re: [PATCH V2 1/1] pmem: set QUEUE_FLAG_NOWAIT

2023-08-02 Thread Chaitanya Kulkarni
On 8/2/23 08:27, Jeff Moyer wrote: > Christoph Hellwig writes: > >> Given that pmem simply loops over an arbitrarily large bio I think >> we also need a threshold for which to allow nowait I/O. While it >> won't block for giant I/Os, doing all of them in the submitter >> context isn't exactly the

Re: [PATCH V2 1/1] pmem: set QUEUE_FLAG_NOWAIT

2023-08-02 Thread Chaitanya Kulkarni
On 8/2/23 08:36, Jens Axboe wrote: > On 8/2/23 6:30?AM, Christoph Hellwig wrote: >> Given that pmem simply loops over an arbitrarily large bio I think >> we also need a threshold for which to allow nowait I/O. While it >> won't block for giant I/Os, doing all of them in the submitter >> context is

Re: [PATCH V2 1/1] pmem: set QUEUE_FLAG_NOWAIT

2023-08-02 Thread Chaitanya Kulkarni
On 8/2/23 05:30, Christoph Hellwig wrote: > Given that pmem simply loops over an arbitrarily large bio I think > we also need a threshold for which to allow nowait I/O. While it > won't block for giant I/Os, doing all of them in the submitter > context isn't exactly the idea behind the nowait I/O.

Re: [PATCH V2 1/1] pmem: set QUEUE_FLAG_NOWAIT

2023-08-02 Thread Christoph Hellwig
On Wed, Aug 02, 2023 at 09:36:32AM -0600, Jens Axboe wrote: > You can do a LOT of looping over a giant bio and still come out way > ahead compared to needing to punt to a different thread. So I do think > it's the right choice. But I'm making assumptions here on what it looks > like, as I haven't s

Re: [PATCH V2 1/1] pmem: set QUEUE_FLAG_NOWAIT

2023-08-02 Thread Jens Axboe
On 8/2/23 6:30?AM, Christoph Hellwig wrote: > Given that pmem simply loops over an arbitrarily large bio I think > we also need a threshold for which to allow nowait I/O. While it > won't block for giant I/Os, doing all of them in the submitter > context isn't exactly the idea behind the nowait I/

Re: [PATCH V2 1/1] pmem: set QUEUE_FLAG_NOWAIT

2023-08-02 Thread Jeff Moyer
Christoph Hellwig writes: > Given that pmem simply loops over an arbitrarily large bio I think > we also need a threshold for which to allow nowait I/O. While it > won't block for giant I/Os, doing all of them in the submitter > context isn't exactly the idea behind the nowait I/O. > > I'm not r

Re: [PATCH V2 1/1] pmem: set QUEUE_FLAG_NOWAIT

2023-08-02 Thread Christoph Hellwig
Given that pmem simply loops over an arbitrarily large bio I think we also need a threshold for which to allow nowait I/O. While it won't block for giant I/Os, doing all of them in the submitter context isn't exactly the idea behind the nowait I/O. I'm not really sure what a good theshold would b