- From: Graham Steel <graham.steel@inria.fr>
- Date: Wed, 30 Apr 2014 09:05:06 +0100
- To: public-webcrypto-comments@w3.org
+1 to drop RSAES-PKCS1-v1_5 from (at least) unwrapKey Even with the most cryptic error messages, the behaviour of the application (whether it accepts the key or not) is still a small leak. The best *publicly known* attack against that oracle needs 12 million messages. Attacks only get better. > > On 30 April 2014 00:05, <bugzilla@jessica.w3.org> wrote: >> https://www.w3.org/Bugs/Public/show_bug.cgi?id=25431 >> >> Richard Barnes <rlb@ipv.sx> changed: >> >> What |Removed |Added >> ---------------------------------------------------------------------------- >> CC| |rlb@ipv.sx >> >> --- Comment #2 from Richard Barnes <rlb@ipv.sx> --- >> The changes needed to make this safe would make the API even more cryptic to >> devs than it already is. I would prefer to just drop RSAES-PKCS1-v1_5 from >> unwrapKey. >> >> -- >> You are receiving this mail because: >> You are on the CC list for the bug. >> > > > > -- > http://www.lsv.ens-cachan.fr/~steel/ -- http://www.lsv.ens-cachan.fr/~steel/
Received on Wednesday, 30 April 2014 08:22:51 UTC