Messages in this thread | | | Date | Fri, 25 Nov 2022 14:28:30 +0000 | Subject | Re: [PATCH v6 1/5] jump_label: Prevent key->enabled int overflow | From | Dmitry Safonov <> |
| |
On 11/25/22 07:59, Peter Zijlstra wrote: > On Wed, Nov 23, 2022 at 05:38:55PM +0000, Dmitry Safonov wrote: >> 1. With CONFIG_JUMP_LABEL=n static_key_slow_inc() doesn't have any >> protection against key->enabled refcounter overflow. >> 2. With CONFIG_JUMP_LABEL=y static_key_slow_inc_cpuslocked() >> still may turn the refcounter negative as (v + 1) may overflow. >> >> key->enabled is indeed a ref-counter as it's documented in multiple >> places: top comment in jump_label.h, Documentation/staging/static-keys.rst, >> etc. >> >> As -1 is reserved for static key that's in process of being enabled, >> functions would break with negative key->enabled refcount: >> - for CONFIG_JUMP_LABEL=n negative return of static_key_count() >> breaks static_key_false(), static_key_true() >> - the ref counter may become 0 from negative side by too many >> static_key_slow_inc() calls and lead to use-after-free issues. >> >> These flaws result in that some users have to introduce an additional >> mutex and prevent the reference counter from overflowing themselves, >> see bpf_enable_runtime_stats() checking the counter against INT_MAX / 2. >> >> Prevent the reference counter overflow by checking if (v + 1) > 0. >> Change functions API to return whether the increment was successful. >> >> Signed-off-by: Dmitry Safonov <dima@arista.com> >> Acked-by: Jakub Kicinski <kuba@kernel.org> > > This looks good to me: > > Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Thank you, Peter!
> What is the plan for merging this? I'm assuming it would want to go > through the network tree, but as already noted earlier it depends on a > patch I have in tip/locking/core. > > Now I checked, tip/locking/core is *just* that one patch, so it might be > possible to merge that branch and this series into the network tree and > note that during the pull request to Linus.
I initially thought it has to go through tip trees because of the dependence, but as you say it's just one patch.
I was also asked by Jakub on v4 to wait for Eric's Ack/Review, so once I get a go from him, I will send all 6 patches for inclusion into -net tree, if that will be in time before the merge window.
Thanks again for the review and ack, Dmitry
| |