Re: [RFC PATCH] nvme: Always invalidate whole cqes[] array

2021-03-19 Thread Tom Rini
On Mon, Feb 08, 2021 at 01:31:54PM +, Andre Przywara wrote: > At the moment nvme_read_completion_status() tries to invidate a single > member of the cqes[] array, which is shady as just a single entry is > not cache line aligned. > The structure is dictated by hardware, and with 16 bytes is sm

Re: [RFC PATCH] nvme: Always invalidate whole cqes[] array

2021-02-26 Thread Suniel Mahesh
On Fri, Feb 26, 2021 at 8:12 PM Jagan Teki wrote: > > Hi Suniel, > > On Mon, Feb 8, 2021 at 7:02 PM Andre Przywara wrote: > > > > At the moment nvme_read_completion_status() tries to invidate a single > > member of the cqes[] array, which is shady as just a single entry is > > not cache line alig

Re: [RFC PATCH] nvme: Always invalidate whole cqes[] array

2021-02-26 Thread Neil Armstrong
On 08/02/2021 14:31, Andre Przywara wrote: > At the moment nvme_read_completion_status() tries to invidate a single > member of the cqes[] array, which is shady as just a single entry is > not cache line aligned. > The structure is dictated by hardware, and with 16 bytes is smaller than > any cache

Re: [RFC PATCH] nvme: Always invalidate whole cqes[] array

2021-02-26 Thread Jagan Teki
Hi Suniel, On Mon, Feb 8, 2021 at 7:02 PM Andre Przywara wrote: > > At the moment nvme_read_completion_status() tries to invidate a single > member of the cqes[] array, which is shady as just a single entry is > not cache line aligned. > The structure is dictated by hardware, and with 16 bytes is

Re: [RFC PATCH] nvme: Always invalidate whole cqes[] array

2021-02-09 Thread Michael Nazzareno Trimarchi
Hi Jagan On Tue, Feb 9, 2021 at 12:28 PM Bin Meng wrote: > > On Mon, Feb 8, 2021 at 9:32 PM Andre Przywara wrote: > > > > At the moment nvme_read_completion_status() tries to invidate a single > > typo: invalidate > > > member of the cqes[] array, which is shady as just a single entry is > > not

Re: [RFC PATCH] nvme: Always invalidate whole cqes[] array

2021-02-09 Thread Bin Meng
On Mon, Feb 8, 2021 at 9:32 PM Andre Przywara wrote: > > At the moment nvme_read_completion_status() tries to invidate a single typo: invalidate > member of the cqes[] array, which is shady as just a single entry is > not cache line aligned. > The structure is dictated by hardware, and with 16 b