Messages in this thread | | | Date | Tue, 7 Nov 2023 17:07:03 +0900 | From | "Dae R. Jeong" <> | Subject | Re: Missing a write memory barrier in tls_init() |
| |
Hi, Jakub,
Thank you for your reply.
On Mon, Nov 06, 2023 at 02:36:59PM -0800, Jakub Kicinski wrote: > On Thu, 2 Nov 2023 16:11:29 +0900 Dae R. Jeong wrote: > > In addition, I believe the {tls_setsockopt, tls_getsockopt} > > implementation is fine because of the address dependency. I think > > load-load reordering is prohibited in this case so we don't need a > > read barrier. > > Sounds plausible, could you send a patch?
Sure. I am doing something else today, so I will send a patch tomorrow or the day after tomorrow.
> The smb_wmb() would be better placed in tls_init(), IMHO.
It sounds better. I will write a patch in that way.
Best regards, Dae R. Jeong
| |