From: Andy Lutomirski Subject: Re: [PATCH V2] Fix a sleep-in-atomic bug in shash_setkey_unaligned Date: Tue, 3 Oct 2017 09:46:04 -0700 Message-ID: References: <1506997522-26684-1-git-send-email-baijiaju1990@163.com> <20171003052643.GB22750@gondor.apana.org.au> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Cc: Andy Lutomirski , Jia-Ju Bai , "David S. Miller" , Neil Horman , vyasevich@gmail.com, Kalle Valo , Linux Crypto Mailing List , Network Development , linux-sctp@vger.kernel.org, Linux Wireless List To: Herbert Xu Return-path: In-Reply-To: <20171003052643.GB22750@gondor.apana.org.au> Sender: netdev-owner@vger.kernel.org List-Id: linux-crypto.vger.kernel.org On Mon, Oct 2, 2017 at 10:26 PM, Herbert Xu wrote: > On Mon, Oct 02, 2017 at 09:18:24PM -0700, Andy Lutomirski wrote: >> > On Oct 2, 2017, at 7:25 PM, Jia-Ju Bai wrote: >> > >> > The SCTP program may sleep under a spinlock, and the function call path is: >> > sctp_generate_t3_rtx_event (acquire the spinlock) >> > sctp_do_sm >> > sctp_side_effects >> > sctp_cmd_interpreter >> > sctp_make_init_ack >> > sctp_pack_cookie >> > crypto_shash_setkey >> > shash_setkey_unaligned >> > kmalloc(GFP_KERNEL) >> >> I'm going to go out on a limb here: why on Earth is out crypto API so >> full of indirection that we allocate memory at all here? > > The crypto API operates on a one key per-tfm basis. So normally > tfm allocation and key setting is done once only and not done on > the data path. > > I have looked at the SCTP code and it appears to fit this paradigm. > That is, we should be able to allocate the tfm and set the key when > the key is actually generated via get_random_bytes, rather than every > time the key is used which is not only a waste but as you see runs > into API issues. It's a waste because it loses a pre-computation advantage. The fact that it has memory allocation issues is crypto API's fault, full stop. There is no legit reason to need to allocate anything.