From: Peter Zijlstra <peterz@infradead.org>
To: Ingo Molnar <mingo@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>,
LKML <linux-kernel@vger.kernel.org>,
x86@kernel.org, Juergen Gross <jgross@suse.com>,
Andi Kleen <ak@linux.intel.com>
Subject: Re: x86/paravirt: Detect over-sized patching bugs in paravirt_patch_call()
Date: Thu, 25 Apr 2019 14:30:45 +0200 [thread overview]
Message-ID: <20190425123045.GD14281@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <20190425114004.GN4038@hirez.programming.kicks-ass.net>
On Thu, Apr 25, 2019 at 01:40:04PM +0200, Peter Zijlstra wrote:
> So this then fails to patch the immediate; but the compiler emitted:
>
> 175: ff 25 00 00 00 00 jmpq *0x0(%rip) # 17b <syscall_return_via_sysret+0x75>
> 177: R_X86_64_PC32 pv_ops+0xfc
>
> and pv_ops+0xfc is (+4 because of reloc magic):
>
> void (*usergs_sysret64)(void); /* 0x100 0x8 */
>
> which defaults to:
>
> arch/x86/kernel/paravirt.c: .cpu.usergs_sysret64 = native_usergs_sysret64,
>
> which in turn reads like:
>
> 0000000000000000 <native_usergs_sysret64>:
> 0: 0f 01 f8 swapgs
> 3: 48 0f 07 sysretq
>
> So I _really_ don't understand how:
>
> > Which with the vanilla kernel crashes on boot much, much later:
> >
> > [ 2.478026] PANIC: double fault, error_code: 0x0
>
> happens.
Just spoke to Juergen on IRC; the most likely explanation is that
(because of PTI) the pv_ops table isn't mapped in the user page-tables
and therefore the indirect jump comes unstuck.
next prev parent reply other threads:[~2019-04-25 12:30 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-04-24 13:41 [patch 0/3] x86/paravirt: Rework paravirt patching Thomas Gleixner
2019-04-24 13:41 ` [patch 1/3] x86/paravirt: Remove bogus extern declarations Thomas Gleixner
2019-04-25 7:31 ` [tip:x86/paravirt] " tip-bot for Thomas Gleixner
2019-05-24 7:58 ` tip-bot for Thomas Gleixner
2019-04-24 13:41 ` [patch 2/3] x86/paravirt: Unify 32/64 bit patch code Thomas Gleixner
2019-04-25 7:32 ` [tip:x86/paravirt] " tip-bot for Thomas Gleixner
2019-05-24 8:00 ` [tip:x86/paravirt] x86/paravirt: Unify the 32/64 bit paravirt patching code tip-bot for Thomas Gleixner
2019-04-24 13:41 ` [patch 3/3] x86/paravirt: Replace paravirt patch asm magic Thomas Gleixner
2019-04-25 6:52 ` Ingo Molnar
2019-04-25 7:22 ` Thomas Gleixner
2019-04-25 7:46 ` Juergen Gross
2019-04-25 8:10 ` [PATCH] x86/paravirt: Match paravirt patchlet field definition ordering to initialization ordering Ingo Molnar
2019-04-25 9:17 ` [PATCH] x86/paravirt: Detect oversized patching bugs as they happen and BUG_ON() to avoid later crashes Ingo Molnar
2019-04-25 9:21 ` Peter Zijlstra
2019-04-25 9:50 ` x86/paravirt: Detect over-sized patching bugs in paravirt_patch_call() Ingo Molnar
2019-04-25 10:22 ` Peter Zijlstra
2019-04-25 10:57 ` Ingo Molnar
2019-04-25 11:30 ` Juergen Gross
2019-04-25 12:30 ` Juergen Gross
2019-04-25 11:40 ` Peter Zijlstra
2019-04-25 12:30 ` Peter Zijlstra [this message]
2019-05-24 7:59 ` [tip:x86/paravirt] " tip-bot for Ingo Molnar
2019-05-24 7:58 ` [tip:x86/paravirt] x86/paravirt: Detect over-sized patching bugs in paravirt_patch_insns() tip-bot for Ingo Molnar
2019-05-24 8:01 ` [tip:x86/paravirt] x86/paravirt: Match paravirt patchlet field definition ordering to initialization ordering tip-bot for Ingo Molnar
2019-04-25 8:08 ` [patch 3/3] x86/paravirt: Replace paravirt patch asm magic Peter Zijlstra
2019-04-25 8:19 ` Peter Zijlstra
2019-04-25 9:20 ` Ingo Molnar
2019-05-24 8:00 ` [tip:x86/paravirt] x86/paravirt: Replace the " tip-bot for Thomas Gleixner
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=20190425123045.GD14281@hirez.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=ak@linux.intel.com \
--cc=jgross@suse.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=tglx@linutronix.de \
--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