Re: [PATCH v5 2/3] random: introduce generic vDSO getrandom() implementation

[Jason A. Donenfeld]

On Sun, Nov 20, 2022 at 02:43:31AM +0100, Jason A. Donenfeld wrote:
> On Sun, Nov 20, 2022 at 01:53:53AM +0100, Jason A. Donenfeld wrote:
> > shouldn't fork or something, but that seems disappointing. Or more state
> > could be allocated in the zeroing region, to hold a chacha state, so
> > another 64 bytes, which would be sort of unfortunate. Or something else?
> > I'd be interested to hear your impression of this quandary.
> 
> Another 128 bytes, actually. And the current chacha in there isn't
> cleaning up its stack as one might hope. So maybe the cleanest solution
> would be to just bite the bullet and allocate another 128 bytes per
> state and make a mini chacha that operates over that? (And I guess hope
> it doesn't need to spill and such...)

I've got it implemented without using any stack now. Wasn't so bad. So
all of the additional concerns I added will be addressed in v+1.

Jason