To work around a misbehavior of the compiler's ability to see into
composite flexible array structs (as detailed in the coming memcpy()
hardening series[1]), split the memcpy() of the header and the payload
so no false positive run-time overflow warning will be generated.
[1] https://lore.kernel.org/linux-hardening/[email protected]/
Cc: Johannes Berg <[email protected]>
Cc: "David S. Miller" <[email protected]>
Cc: Eric Dumazet <[email protected]>
Cc: Jakub Kicinski <[email protected]>
Cc: Paolo Abeni <[email protected]>
Cc: [email protected]
Cc: [email protected]
Signed-off-by: Kees Cook <[email protected]>
---
net/wireless/nl80211.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
index 2705e3ee8fc4..461897933e92 100644
--- a/net/wireless/nl80211.c
+++ b/net/wireless/nl80211.c
@@ -13171,7 +13171,8 @@ static int nl80211_parse_wowlan_tcp(struct cfg80211_registered_device *rdev,
wake_mask_size);
if (tok) {
cfg->tokens_size = tokens_size;
- memcpy(&cfg->payload_tok, tok, sizeof(*tok) + tokens_size);
+ cfg->payload_tok = *tok;
+ memcpy(cfg->payload_tok.token_stream, tok->token_stream, + tokens_size);
}
trig->tcp = cfg;
--
2.34.1
On Mon, Sep 26, 2022 at 05:39:03PM -0700, Kees Cook wrote:
> To work around a misbehavior of the compiler's ability to see into
> composite flexible array structs (as detailed in the coming memcpy()
> hardening series[1]), split the memcpy() of the header and the payload
> so no false positive run-time overflow warning will be generated.
>
> [1] https://lore.kernel.org/linux-hardening/[email protected]/
>
> Cc: Johannes Berg <[email protected]>
> Cc: "David S. Miller" <[email protected]>
> Cc: Eric Dumazet <[email protected]>
> Cc: Jakub Kicinski <[email protected]>
> Cc: Paolo Abeni <[email protected]>
> Cc: [email protected]
> Cc: [email protected]
> Signed-off-by: Kees Cook <[email protected]>
> ---
> net/wireless/nl80211.c | 3 ++-
> 1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
> index 2705e3ee8fc4..461897933e92 100644
> --- a/net/wireless/nl80211.c
> +++ b/net/wireless/nl80211.c
> @@ -13171,7 +13171,8 @@ static int nl80211_parse_wowlan_tcp(struct cfg80211_registered_device *rdev,
> wake_mask_size);
> if (tok) {
> cfg->tokens_size = tokens_size;
> - memcpy(&cfg->payload_tok, tok, sizeof(*tok) + tokens_size);
> + cfg->payload_tok = *tok;
> + memcpy(cfg->payload_tok.token_stream, tok->token_stream, + tokens_size);
There is a spurious '+' here............................................^^^
--
Gustavo
> }
>
> trig->tcp = cfg;
> --
> 2.34.1
>
On Mon, Sep 26, 2022 at 08:33:47PM -0500, Gustavo A. R. Silva wrote:
> On Mon, Sep 26, 2022 at 05:39:03PM -0700, Kees Cook wrote:
> > To work around a misbehavior of the compiler's ability to see into
> > composite flexible array structs (as detailed in the coming memcpy()
> > hardening series[1]), split the memcpy() of the header and the payload
> > so no false positive run-time overflow warning will be generated.
> >
> > [1] https://lore.kernel.org/linux-hardening/[email protected]/
> >
> > Cc: Johannes Berg <[email protected]>
> > Cc: "David S. Miller" <[email protected]>
> > Cc: Eric Dumazet <[email protected]>
> > Cc: Jakub Kicinski <[email protected]>
> > Cc: Paolo Abeni <[email protected]>
> > Cc: [email protected]
> > Cc: [email protected]
> > Signed-off-by: Kees Cook <[email protected]>
> > ---
> > net/wireless/nl80211.c | 3 ++-
> > 1 file changed, 2 insertions(+), 1 deletion(-)
> >
> > diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c
> > index 2705e3ee8fc4..461897933e92 100644
> > --- a/net/wireless/nl80211.c
> > +++ b/net/wireless/nl80211.c
> > @@ -13171,7 +13171,8 @@ static int nl80211_parse_wowlan_tcp(struct cfg80211_registered_device *rdev,
> > wake_mask_size);
> > if (tok) {
> > cfg->tokens_size = tokens_size;
> > - memcpy(&cfg->payload_tok, tok, sizeof(*tok) + tokens_size);
> > + cfg->payload_tok = *tok;
> > + memcpy(cfg->payload_tok.token_stream, tok->token_stream, + tokens_size);
>
> There is a spurious '+' here............................................^^^
HUnh. Well. I learned a new C-ism today. A leading "+" is just accepted.
I build-tested this, and it was happy. I just double-checked, and sure
enough, it is actually parsing that line. I will send a v2. Thanks for
catching that! ... That's such a weird typo we don't even have a
checkpatch rule for it. :P
--
Kees Cook