Re: Encapsulate io_uring process count calculation

2025-06-04 Thread Michael Paquier
On Wed, Jun 04, 2025 at 11:12:07AM +0800, Japin Li wrote: > On Tue, 03 Jun 2025 at 13:27, Michael Paquier wrote: >> Sounds like a copy-pasto error to me, so good idea to reuse this >> routine. pgaio_uring_procs() is only used for shmem sizing and in a >> check at backend initialization currently.

Re: Encapsulate io_uring process count calculation

2025-06-03 Thread Japin Li
On Tue, 03 Jun 2025 at 13:27, Michael Paquier wrote: > On Tue, Jun 03, 2025 at 12:16:21PM +0800, Japin Li wrote: >> I've replaced the hardcoded arithmetic expression for calculating TotalProcs >> within the pgaio_uring_shmem_init() with a call to the dedicated >> pgaio_uring_procs() function. >>

Re: Encapsulate io_uring process count calculation

2025-06-02 Thread Michael Paquier
On Tue, Jun 03, 2025 at 12:16:21PM +0800, Japin Li wrote: > I've replaced the hardcoded arithmetic expression for calculating TotalProcs > within the pgaio_uring_shmem_init() with a call to the dedicated > pgaio_uring_procs() function. > > This can make the code's intent clearer. Simplifies futur