lkml.org 
[lkml]   [2022]   [Jan]   [6]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [RFC PATCH rdma-next 08/10] RDMA/rxe: Implement flush execution in responder side
On Thu, Jan 06, 2022 at 06:42:57AM +0000, lizhijian@fujitsu.com wrote:
>
>
> On 06/01/2022 08:28, Jason Gunthorpe wrote:
> > On Tue, Dec 28, 2021 at 04:07:15PM +0800, Li Zhijian wrote:
> >> + while (length > 0) {
> >> + va = (u8 *)(uintptr_t)buf->addr + offset;
> >> + bytes = buf->size - offset;
> >> +
> >> + if (bytes > length)
> >> + bytes = length;
> >> +
> >> + arch_wb_cache_pmem(va, bytes);
> > So why did we need to check that the va was pmem to call this?
> Sorry, i didn't get you.
>
> I didn't check whether va is pmem, since only MR registered with PERSISTENCE(only pmem can
> register this access flag) can reach here.

Yes, that is what I mean, why did we need to check anything to call
this API - it should work on any CPU mapped address.

Jason

\
 
 \ /
  Last update: 2022-01-06 18:35    [W:0.084 / U:0.112 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site