From: Jann Horn <jannh@google.com>
To: Josh Poimboeuf <jpoimboe@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>,
"the arch/x86 maintainers" <x86@kernel.org>,
kernel list <linux-kernel@vger.kernel.org>,
Ard Biesheuvel <ard.biesheuvel@linaro.org>,
Andy Lutomirski <luto@kernel.org>,
Steven Rostedt <rostedt@goodmis.org>,
Ingo Molnar <mingo@kernel.org>,
Thomas Gleixner <tglx@linutronix.de>,
Linus Torvalds <torvalds@linux-foundation.org>,
Masami Hiramatsu <mhiramat@kernel.org>,
Jason Baron <jbaron@akamai.com>, Jiri Kosina <jkosina@suse.cz>,
David Laight <David.Laight@aculab.com>,
Borislav Petkov <bp@alien8.de>, Julia Cartwright <julia@ni.com>,
Jessica Yu <jeyu@kernel.org>, "H. Peter Anvin" <hpa@zytor.com>,
Nadav Amit <namit@vmware.com>,
Rasmus Villemoes <linux@rasmusvillemoes.dk>,
Edward Cree <ecree@solarflare.com>,
Daniel Bristot de Oliveira <bristot@redhat.com>
Subject: Re: [PATCH v3 0/6] Static calls
Date: Mon, 17 Feb 2020 22:10:27 +0100 [thread overview]
Message-ID: <CAG48ez0owFet0E43UAGd7sV9Oi0yhVpWTmy4W+Vm5+0q=74-DA@mail.gmail.com> (raw)
In-Reply-To: <20190110205226.iburt6mrddsxnjpk@treble>
On Thu, Jan 10, 2019 at 9:52 PM Josh Poimboeuf <jpoimboe@redhat.com> wrote:
> On Thu, Jan 10, 2019 at 09:30:23PM +0100, Peter Zijlstra wrote:
> > On Wed, Jan 09, 2019 at 04:59:35PM -0600, Josh Poimboeuf wrote:
> > > With this version, I stopped trying to use text_poke_bp(), and instead
> > > went with a different approach: if the call site destination doesn't
> > > cross a cacheline boundary, just do an atomic write. Otherwise, keep
> > > using the trampoline indefinitely.
> >
> > > - Get rid of the use of text_poke_bp(), in favor of atomic writes.
> > > Out-of-line calls will be promoted to inline only if the call sites
> > > don't cross cache line boundaries. [Linus/Andy]
> >
> > Can we perserve why text_poke_bp() didn't work? I seem to have forgotten
> > again. The problem was poking the return address onto the stack from the
> > int3 handler, or something along those lines?
>
> Right, emulating a call instruction from the #BP handler is ugly,
> because you have to somehow grow the stack to make room for the return
> address. Personally I liked the idea of shifting the iret frame by 16
> bytes in the #DB entry code, but others hated it.
>
> So many bad-but-not-completely-unacceptable options to choose from.
Silly suggestion from someone who has skimmed the thread:
Wouldn't a retpoline-style trampoline solve this without needing
memory allocations? Let the interrupt handler stash the destination in
a percpu variable and clear IF in regs->flags. Something like:
void simulate_call(unsigned long target) {
__this_cpu_write(static_call_restore_if, (regs->flags & X86_EFLAGS_IF) != 0);
regs->flags &= ~X86_EFLAGS_IF;
__this_cpu_write(static_call_trampoline_source, regs->ip + 5);
__this_cpu_write(static_call_trampoline_target, target);
regs->ip = magic_static_call_trampoline;
}
magic_static_call_trampoline:
; set up return address for returning from target function
pushl PER_CPU_VAR(static_call_trampoline_source)
; set up retpoline-style return address
pushl PER_CPU_VAR(static_call_trampoline_target)
; restore flags if needed
cmp PER_CPU_VAR(static_call_restore_if), 0
je 1f
sti ; NOTE: percpu data must not be accessed past this point
1:
ret ; "return" to the call target
By using a return to implement the call, we don't need any scratch
registers for the call.
next prev parent reply other threads:[~2020-02-17 21:10 UTC|newest]
Thread overview: 90+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-01-09 22:59 [PATCH v3 0/6] Static calls Josh Poimboeuf
2019-01-09 22:59 ` [PATCH v3 1/6] compiler.h: Make __ADDRESSABLE() symbol truly unique Josh Poimboeuf
2019-01-09 22:59 ` [PATCH v3 2/6] static_call: Add basic static call infrastructure Josh Poimboeuf
2019-01-10 14:03 ` Edward Cree
2019-01-10 18:37 ` Josh Poimboeuf
2019-01-09 22:59 ` [PATCH v3 3/6] x86/static_call: Add out-of-line static call implementation Josh Poimboeuf
2019-01-10 0:16 ` Nadav Amit
2019-01-10 16:28 ` Josh Poimboeuf
2019-01-09 22:59 ` [PATCH v3 4/6] static_call: Add inline static call infrastructure Josh Poimboeuf
2019-01-09 22:59 ` [PATCH v3 5/6] x86/alternative: Use a single access in text_poke() where possible Josh Poimboeuf
2019-01-10 9:32 ` Nadav Amit
2019-01-10 17:20 ` Josh Poimboeuf
2019-01-10 17:29 ` Nadav Amit
2019-01-10 17:32 ` Steven Rostedt
2019-01-10 17:42 ` Sean Christopherson
2019-01-10 17:57 ` Steven Rostedt
2019-01-10 18:04 ` Sean Christopherson
2019-01-10 18:21 ` Josh Poimboeuf
2019-01-10 18:24 ` Steven Rostedt
2019-01-11 12:10 ` Alexandre Chartre
2019-01-11 15:28 ` Josh Poimboeuf
2019-01-11 16:46 ` Alexandre Chartre
2019-01-11 16:57 ` Josh Poimboeuf
2019-01-11 17:41 ` Jason Baron
2019-01-11 17:54 ` Nadav Amit
2019-01-15 11:10 ` Alexandre Chartre
2019-01-15 16:19 ` Steven Rostedt
2019-01-15 16:45 ` Alexandre Chartre
2019-01-11 0:59 ` hpa
2019-01-11 1:34 ` Sean Christopherson
2019-01-11 8:13 ` hpa
2019-01-09 22:59 ` [PATCH v3 6/6] x86/static_call: Add inline static call implementation for x86-64 Josh Poimboeuf
2019-01-10 1:21 ` [PATCH v3 0/6] Static calls Nadav Amit
2019-01-10 16:44 ` Josh Poimboeuf
2019-01-10 17:32 ` Nadav Amit
2019-01-10 18:18 ` Josh Poimboeuf
2019-01-10 19:45 ` Nadav Amit
2019-01-10 20:32 ` Josh Poimboeuf
2019-01-10 20:48 ` Nadav Amit
2019-01-10 20:57 ` Josh Poimboeuf
2019-01-10 21:47 ` Nadav Amit
2019-01-10 17:31 ` Linus Torvalds
2019-01-10 20:51 ` H. Peter Anvin
2019-01-10 20:30 ` Peter Zijlstra
2019-01-10 20:52 ` Josh Poimboeuf
2019-01-10 23:02 ` Linus Torvalds
2019-01-11 0:56 ` Andy Lutomirski
2019-01-11 1:47 ` Nadav Amit
2019-01-11 15:15 ` Josh Poimboeuf
2019-01-11 15:48 ` Nadav Amit
2019-01-11 16:07 ` Josh Poimboeuf
2019-01-11 17:23 ` Nadav Amit
2019-01-11 19:03 ` Linus Torvalds
2019-01-11 19:17 ` Nadav Amit
2019-01-11 19:23 ` hpa
2019-01-11 19:33 ` Nadav Amit
2019-01-11 19:34 ` Linus Torvalds
2019-01-13 0:34 ` hpa
2019-01-13 0:36 ` hpa
2019-01-11 19:39 ` Jiri Kosina
2019-01-14 2:31 ` H. Peter Anvin
2019-01-14 2:40 ` H. Peter Anvin
2019-01-14 20:11 ` Andy Lutomirski
2019-01-14 22:00 ` H. Peter Anvin
2019-01-14 22:54 ` H. Peter Anvin
2019-01-15 3:05 ` Andy Lutomirski
2019-01-15 5:01 ` H. Peter Anvin
2019-01-15 5:37 ` H. Peter Anvin
2019-01-14 23:27 ` Andy Lutomirski
2019-01-14 23:51 ` Nadav Amit
2019-01-15 2:28 ` hpa
2019-01-11 20:04 ` Josh Poimboeuf
2019-01-11 20:12 ` Linus Torvalds
2019-01-11 20:31 ` Josh Poimboeuf
2019-01-11 20:46 ` Linus Torvalds
2019-01-11 21:05 ` Andy Lutomirski
2019-01-11 21:10 ` Linus Torvalds
2019-01-11 21:32 ` Josh Poimboeuf
2019-01-14 12:28 ` Peter Zijlstra
2019-01-11 21:22 ` Josh Poimboeuf
2019-01-11 21:23 ` Josh Poimboeuf
2019-01-11 21:25 ` Josh Poimboeuf
2019-01-11 21:36 ` Nadav Amit
2019-01-11 21:41 ` Josh Poimboeuf
2019-01-11 21:55 ` Steven Rostedt
2019-01-11 21:59 ` Nadav Amit
2019-01-11 21:56 ` Nadav Amit
2019-01-12 23:54 ` Andy Lutomirski
2020-02-17 21:10 ` Jann Horn [this message]
2020-02-17 21:57 ` Steven Rostedt
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='CAG48ez0owFet0E43UAGd7sV9Oi0yhVpWTmy4W+Vm5+0q=74-DA@mail.gmail.com' \
--to=jannh@google.com \
--cc=David.Laight@aculab.com \
--cc=ard.biesheuvel@linaro.org \
--cc=bp@alien8.de \
--cc=bristot@redhat.com \
--cc=ecree@solarflare.com \
--cc=hpa@zytor.com \
--cc=jbaron@akamai.com \
--cc=jeyu@kernel.org \
--cc=jkosina@suse.cz \
--cc=jpoimboe@redhat.com \
--cc=julia@ni.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux@rasmusvillemoes.dk \
--cc=luto@kernel.org \
--cc=mhiramat@kernel.org \
--cc=mingo@kernel.org \
--cc=namit@vmware.com \
--cc=peterz@infradead.org \
--cc=rostedt@goodmis.org \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
--cc=x86@kernel.org \
/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.
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).