Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Tue, 14 Apr 2020 20:31:17 -0500
From:   Josh Poimboeuf <jpoimboe@redhat.com>
To:     Joe Lawrence <joe.lawrence@redhat.com>
Cc:     live-patching@vger.kernel.org, linux-kernel@vger.kernel.org,
        Peter Zijlstra <peterz@infradead.org>,
        Jessica Yu <jeyu@kernel.org>
Subject: Re: [PATCH 0/7] livepatch,module: Remove .klp.arch and
 module_disable_ro()
Message-ID: <20200415013117.rc7vlidmo4okzypl@treble>
References: <cover.1586881704.git.jpoimboe@redhat.com>
 <187a2ccd-1d04-54db-2fd3-8c4ca6872830@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <187a2ccd-1d04-54db-2fd3-8c4ca6872830@redhat.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Tue, Apr 14, 2020 at 08:57:15PM -0400, Joe Lawrence wrote:
> On 4/14/20 12:28 PM, Josh Poimboeuf wrote:
> > Better late than never, these patches add simplifications and
> > improvements for some issues Peter found six months ago, as part of his
> > non-writable text code (W^X) cleanups.
> > 
> > Highlights:
> > 
> > - Remove the livepatch arch-specific .klp.arch sections, which were used
> >    to do paravirt patching and alternatives patching for livepatch
> >    replacement code.
> > 
> > - Add support for jump labels in patched code.
> 
> Re: jump labels and late-module patching support...
> 
> Is there still an issue of a non-exported static key defined in a
> to-be-patched module referenced and resolved via klp-relocation when the
> livepatch module is loaded first?  (Basically the same case I asked Petr
> about in his split livepatch module PoC. [1])
> 
> Or should we declare this an invalid klp-relocation use case and force the
> livepatch author to use static_key_enabled()?
> 
> [1] https://lore.kernel.org/lkml/20200407205740.GA17061@redhat.com/

Right, if the static key lives in a module, then it's still not possible
for a jump label to use it.  I added a check in kpatch-build to block
that case and suggest static_key_enabled() instead.

I don't know what the solution is, other than getting rid of late module
patching.

I confess I haven't looked at Petr's patches due to other distractions,
but I plan to soon.

-- 
Josh