Re: [PATCH] net: tls: fix possible race condition between do_tls_getsockopt_conf() and do_tls_setsockopt_conf()

[Sabrina Dubroca <sd@queasysnail.net>]

2023-02-24, 10:55:08 -0800, Jakub Kicinski wrote:
> On Fri, 24 Feb 2023 13:06:06 +0100 Florian Westphal wrote:
> > Hangyu Hua <hbh25y@gmail.com> wrote:
> > > ctx->crypto_send.info is not protected by lock_sock in
> > > do_tls_getsockopt_conf(). A race condition between do_tls_getsockopt_conf()
> > > and do_tls_setsockopt_conf() can cause a NULL point dereference or
> > > use-after-free read when memcpy.  
> > 
> > Its good practice to quote the relevant parts of the splat here.
> 
> Right, the bug and the fix seem completely bogus.
> Please make sure the bugs are real and the fixes you sent actually 
> fix them.

I suggested a change of locking in do_tls_getsockopt_conf this
morning [1]. The issue reported last seemed valid, but this patch is not
at all what I had in mind.
[1] https://lore.kernel.org/all/Y/ht6gQL+u6fj3dG@hog/

do_tls_setsockopt_conf fills crypto_info immediately from what
userspace gives us (and clears it on exit in case of failure), which
getsockopt could see since it's not locking the socket when it checks
TLS_CRYPTO_INFO_READY. So getsockopt would progress up to the point it
finally locks the socket, but if setsockopt failed, we could have
cleared TLS_CRYPTO_INFO_READY and freed iv/rec_seq.

-- 
Sabrina