Re: [PATCH] signal: restore the override_rlimit logic

On Sun, Nov 03, 2024 at 05:50:49PM +0100, Oleg Nesterov wrote:

On 11/02, Alexey Gladkov wrote:

...

+Cc Oleg Nesterov.

Well, I tend to agree with Roman and his patch looks good to me.

Thanks, Oleg!

But it seems that the change in inc_rlimit_get_ucounts() can be a bit simpler and more readable, see below.

Eric suggested the same approach earlier in this thread. I personally don't have a strong preference here or actually I slightly prefer my own version because this comparison to LONG_MAX looks confusing to me. But if you have a strong preference, I'm happy to send out v2. Please, let me know.

Thanks!