From: Alexander Potapenko Subject: Re: [PATCH 6/7] md/raid10, LLVM: get rid of variable length array Date: Fri, 17 Mar 2017 13:31:23 +0100 Message-ID: References: <20170317001520.85223-1-md@google.com> <20170317001520.85223-7-md@google.com> <20170317120837.pr74cv3xuj7qpoin@hirez.programming.kicks-ass.net> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Cc: Michael Davidson , Michal Marek , Thomas Gleixner , Ingo Molnar , "H. Peter Anvin" , Herbert Xu , "David S. Miller" , Shaohua Li , Dmitry Vyukov , Matthias Kaehlcke , x86@kernel.org, linux-kbuild@vger.kernel.org, LKML , linux-crypto@vger.kernel.org, linux-raid@vger.kernel.org To: Peter Zijlstra Return-path: In-Reply-To: <20170317120837.pr74cv3xuj7qpoin@hirez.programming.kicks-ass.net> Sender: linux-raid-owner@vger.kernel.org List-Id: linux-crypto.vger.kernel.org On Fri, Mar 17, 2017 at 1:08 PM, Peter Zijlstra wrot= e: > On Thu, Mar 16, 2017 at 05:15:19PM -0700, Michael Davidson wrote: >> Replace a variable length array in a struct by allocating >> the memory for the entire struct in a char array on the stack. >> >> Signed-off-by: Michael Davidson >> --- >> drivers/md/raid10.c | 9 ++++----- >> 1 file changed, 4 insertions(+), 5 deletions(-) >> >> diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c >> index 063c43d83b72..158ebdff782c 100644 >> --- a/drivers/md/raid10.c >> +++ b/drivers/md/raid10.c >> @@ -4654,11 +4654,10 @@ static int handle_reshape_read_error(struct mdde= v *mddev, >> /* Use sync reads to get the blocks from somewhere else */ >> int sectors =3D r10_bio->sectors; >> struct r10conf *conf =3D mddev->private; >> - struct { >> - struct r10bio r10_bio; >> - struct r10dev devs[conf->copies]; >> - } on_stack; >> - struct r10bio *r10b =3D &on_stack.r10_bio; >> + char on_stack_r10_bio[sizeof(struct r10bio) + >> + conf->copies * sizeof(struct r10dev)] >> + __aligned(__alignof__(struct r10bio)); >> + struct r10bio *r10b =3D (struct r10bio *)on_stack_r10_bio; >> int slot =3D 0; >> int idx =3D 0; >> struct bio_vec *bvec =3D r10_bio->master_bio->bi_io_vec; > > > That's disgusting. Why not fix LLVM to support this? IIUC there's only a handful of VLAIS instances in LLVM code, why not just drop them for the sake of better code portability? (To quote Linus, "this feature is an abomination": https://lkml.org/lkml/2013/9/23/500) --=20 Alexander Potapenko Software Engineer Google Germany GmbH Erika-Mann-Stra=C3=9Fe, 33 80636 M=C3=BCnchen Gesch=C3=A4ftsf=C3=BChrer: Matthew Scott Sucherman, Paul Terence Manicle Registergericht und -nummer: Hamburg, HRB 86891 Sitz der Gesellschaft: Hamburg