lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening PHC | |
Open Source and information security mailing list archives
| ||
|
Date: Wed, 18 May 2022 14:21:13 +0800 From: "Guozihua (Scott)" <guozihua@...wei.com> To: "Gustavo A. R. Silva" <gustavoars@...nel.org> CC: <keyrings@...r.kernel.org>, <linux-crypto@...r.kernel.org>, <davem@...emloft.net>, <dhowells@...hat.com>, <herbert@...dor.apana.org.au>, <linux-hardening@...r.kernel.org> Subject: Re: [PATCH v2] crypto: Use struct_size() helper in kmalloc() On 2022/5/18 12:55, Gustavo A. R. Silva wrote: > On Wed, May 18, 2022 at 08:56:39AM +0800, GUO Zihua wrote: >> Make use of struct_size() heler for structures containing flexible array >> member instead of sizeof() which prevents potential issues as well as >> addressing the following sparse warning: >> >> crypto/asymmetric_keys/asymmetric_type.c:155:23: warning: using sizeof >> on a flexible structure >> crypto/asymmetric_keys/asymmetric_type.c:247:28: warning: using sizeof >> on a flexible structure > > The warnings above are not silenced with this patch as struct_size() > internally use sizeof on the struct-with-flex-array. > > However, the use of struct_size() instead of the open-coded expressions > in the calls to kmalloc() is correct. > >> >> Reference: https://github.com/KSPP/linux/issues/174 > > I updated this issue on the issue tracker. Please, from now on, just > use that issue as a list of potential open-coded instances to be > audited. :) > >> >> Signed-off-by: GUO Zihua <guozihua@...wei.com> > > Reviewed-by: Gustavo A. R. Silva <gustavoars@...nel.org> > > Thanks > -- > Gustavo > >> >> --- >> >> v2: >> Use size_add() helper following Kees Cook's suggestion. >> --- >> crypto/asymmetric_keys/asymmetric_type.c | 4 ++-- >> 1 file changed, 2 insertions(+), 2 deletions(-) >> >> diff --git a/crypto/asymmetric_keys/asymmetric_type.c b/crypto/asymmetric_keys/asymmetric_type.c >> index 41a2f0eb4ce4..e020222b1fe5 100644 >> --- a/crypto/asymmetric_keys/asymmetric_type.c >> +++ b/crypto/asymmetric_keys/asymmetric_type.c >> @@ -152,7 +152,7 @@ struct asymmetric_key_id *asymmetric_key_generate_id(const void *val_1, >> { >> struct asymmetric_key_id *kid; >> >> - kid = kmalloc(sizeof(struct asymmetric_key_id) + len_1 + len_2, >> + kid = kmalloc(struct_size(kid, data, size_add(len_1, len_2)), >> GFP_KERNEL); >> if (!kid) >> return ERR_PTR(-ENOMEM); >> @@ -244,7 +244,7 @@ struct asymmetric_key_id *asymmetric_key_hex_to_key_id(const char *id) >> if (asciihexlen & 1) >> return ERR_PTR(-EINVAL); >> >> - match_id = kmalloc(sizeof(struct asymmetric_key_id) + asciihexlen / 2, >> + match_id = kmalloc(struct_size(match_id, data, asciihexlen / 2), >> GFP_KERNEL); >> if (!match_id) >> return ERR_PTR(-ENOMEM); >> -- >> 2.36.0 >> > . Thanks for the review Gustavo. I'll send a v3 patch and update the commit message. Maybe refer to the mail from Linus instead. -- Best GUO Zihua
Powered by blists - more mailing lists