Messages in this thread | | | Date | Fri, 19 Apr 2024 14:05:28 +0200 | From | "Jason A. Donenfeld" <> | Subject | Re: [PATCH v2] random: Fix the issue of '_might_sleep' function running in an atomic contex |
| |
On Fri, Apr 19, 2024 at 06:54:53PM +0800, Guoyong Wang wrote: > In the case that a delay is acceptable for 'crng_set_ready', it can be > deferred to a workqueue in order to accommodate different contexts. > > Signed-off-by: Guoyong Wang <guoyong.wang@mediatek.com> > --- > v2: Compared to version 1, version 2 has removed the definition of > 'execute_in_non_atomic_context' and always uses a workqueue to execute > 'crng_set_ready'. > > Send out the patch again for further discussion. > > [1]: https://patchwork.kernel.org/patch/13595066
Sorry, but this doesn't make sense in terms of how kernel development is typically done:
1) Before you sent this, I told you there was no need for a v2. 2) The removal snippet at the bottom applies to your previous v1, not to upstream code. 3) The other snippet is taken verbatim from a patch that I posted. 4) That snippet is already upstream: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=e871abcda3b67d0820b4182ebe93435624e9c6a4
There is no further work to do on this. Thank you for reporting the bug - very appreciated - and for discussing the fix. But now we have a fix. So everything is all set and there's nothing else to do and everyone can be happy.
Jason
| |