mailing list of musl libc
 help / color / mirror / code / Atom feed
From: Rich Felker <dalias@libc.org>
To: David Edelsohn <dje.gcc@gmail.com>
Cc: musl@lists.openwall.com, Florian Weimer <fweimer@redhat.com>,
	Stijn Tintel <stijn@linux-ipv6.be>
Subject: Re: [musl] [PATCH] ppc64: check for AltiVec in setjmp/longjmp
Date: Mon, 6 Dec 2021 20:39:32 -0500	[thread overview]
Message-ID: <20211207013930.GM7074@brightrain.aerifal.cx> (raw)
In-Reply-To: <CAGWvnymFQ69Eh4Ji0DP5qv7f6LhTyxC9m4OSfhPuD4B1yTsP-Q@mail.gmail.com>

On Mon, Dec 06, 2021 at 08:15:48PM -0500, David Edelsohn wrote:
> On Mon, Dec 6, 2021 at 7:59 PM Rich Felker <dalias@libc.org> wrote:
> >
> > On Tue, Dec 07, 2021 at 01:37:12AM +0100, Florian Weimer wrote:
> > > * Stijn Tintel:
> > >
> > > > diff --git a/src/setjmp/powerpc64/setjmp.s b/src/setjmp/powerpc64/setjmp.s
> > > > index 37683fda..32853693 100644
> > > > --- a/src/setjmp/powerpc64/setjmp.s
> > > > +++ b/src/setjmp/powerpc64/setjmp.s
> > > > @@ -69,7 +69,17 @@ __setjmp_toc:
> > > >     stfd 30, 38*8(3)
> > > >     stfd 31, 39*8(3)
> > > >
> > > > -   # 5) store vector registers v20-v31
> > > > +   # 5) store vector registers v20-v31 if hardware supports AltiVec
> > > > +   mflr 0
> > > > +   bl 1f
> > > > +   .hidden __hwcap
> > > > +   .long __hwcap-.
> > > > +1: mflr 4
> > >
> > > This de-balances the return stack and probably has quite severe
> > > performance impact.  The ISA manual says to use
> > >
> > >   bcl 20,31,$+4
> > >
> > > and you'll have to store the __hwcap offset somewhere else.
> >
> > To begin with, let's change the .s files to .S files and put the whole
> > branch logic inside #ifndef __ALTIVEC__ so that it does not impact
> > normal builds with an ISA level where Altivec can be assumed to be
> > present.
> >
> > I'm not sufficiently familiar with the PowerPC ISA to know how bcl
> > works, but if there's a less expensive solution along those lines
> > that's compatible with all ISA levels, by all means let's use it. The
> > same could be done for powerpc-sf (32-bit) and its SPE branches, too.
> 
> bl = branch and link
> bcl = branch conditional and link
> 
> link means place the next instruction address in the link register.
> Normally a branch and link would be used for a matching "return"
> instruction, but in this case it is being used to compute a position
> independent code address.  As Florian correctly points out, the "bl"
> will corrupt the link stack in the processor used to predict return
> addresses and the recommended sequence is the one that he suggests.
> 
> bcl 20,31,addr
> 
> which means branch always and, because the condition register bits are
> irrelevant, a special value that instructs the processor to not  push
> the address onto the link stack so that the "calls" and "returns"
> remain matched.

Thanks. Am I correct in understanding then that we don't need $+4, but
can instead use the 1f just as now, with inline .long __hwcap-. -- in
other words that "bcl 20,31," is a drop-in replacement for "bl"
without the link stack impact?

Rich

  reply	other threads:[~2021-12-07  1:39 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-12-06 23:43 Stijn Tintel
2021-12-07  0:37 ` Florian Weimer
2021-12-07  0:59   ` Rich Felker
2021-12-07  1:15     ` David Edelsohn
2021-12-07  1:39       ` Rich Felker [this message]
2021-12-07  1:44         ` David Edelsohn
2021-12-07 13:25           ` Rich Felker
2021-12-07 13:39             ` David Edelsohn
2021-12-07 14:43               ` Rich Felker
2021-12-07 14:48                 ` David Edelsohn
2021-12-07 18:39             ` Markus Wichmann
2021-12-07 18:57               ` David Edelsohn
2021-12-07 19:28               ` Florian Weimer
2021-12-07 20:15                 ` Markus Wichmann
2021-12-07 20:29                   ` Rich Felker
2021-12-08  5:02                     ` Markus Wichmann
2021-12-07 18:27           ` James Y Knight
2021-12-07 18:57             ` Markus Wichmann
2021-12-08  8:43     ` Stijn Tintel
2021-12-08 13:37       ` Rich Felker
2021-12-08 15:36         ` Rich Felker

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20211207013930.GM7074@brightrain.aerifal.cx \
    --to=dalias@libc.org \
    --cc=dje.gcc@gmail.com \
    --cc=fweimer@redhat.com \
    --cc=musl@lists.openwall.com \
    --cc=stijn@linux-ipv6.be \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
Code repositories for project(s) associated with this public inbox

	https://git.vuxu.org/mirror/musl/

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).