Date: Tue, 16 Jul 2019 13:41:06 -0700 From: John Baldwin <jhb@FreeBSD.org> To: Shawn Webb <shawn.webb@hardenedbsd.org>, Brooks Davis <brooks@FreeBSD.org> Cc: src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org Subject: Re: svn commit: r350049 - head/contrib/amd/amd Message-ID: <6ddf37a1-3b1a-2401-a338-36b6c8b9947d@FreeBSD.org> In-Reply-To: <20190716194454.odytcdjtlduliz6l@mutt-hbsd> References: <201907161603.x6GG38oZ020228@repo.freebsd.org> <20190716194454.odytcdjtlduliz6l@mutt-hbsd>
next in thread | previous in thread | raw e-mail | index | archive | help
On 7/16/19 12:44 PM, Shawn Webb wrote: > On Tue, Jul 16, 2019 at 04:03:08PM +0000, Brooks Davis wrote: >> Author: brooks >> Date: Tue Jul 16 16:03:08 2019 >> New Revision: 350049 >> URL: https://svnweb.freebsd.org/changeset/base/350049 >> >> Log: >> Fix two mismatches between function declaration and definition. >> >> In both cases, function pointer arguments were inconsistently declared >> and the result worked because of C's odd rules around function pointer >> (de)references. With a stricter compiler these fail to compile. > > And, with CFI applied to the kernel, would cause a panic. :) > > Good catch and thanks for the great work! How would an incorrect function prototype in userland cause a kernel panic? -- John Baldwin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?6ddf37a1-3b1a-2401-a338-36b6c8b9947d>