linux-kernel - Re: [PATCH] livepatch: Move error print out of lock protection in klp_enable

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <65fe1978-60da-5bd4-9559-fddec13f03bf@redhat.com>
Date:   Thu, 1 Sep 2022 09:24:07 -0400
From:   Joe Lawrence <joe.lawrence@...hat.com>
To:     Zhen Lei <thunder.leizhen@...wei.com>
Cc:     Josh Poimboeuf <jpoimboe@...nel.org>,
        Jiri Kosina <jikos@...nel.org>,
        Miroslav Benes <mbenes@...e.cz>,
        Petr Mladek <pmladek@...e.com>, linux-kernel@...r.kernel.org,
        live-patching@...r.kernel.org
Subject: Re: [PATCH] livepatch: Move error print out of lock protection in
 klp_enable_patch()

On 8/31/22 10:27 PM, Zhen Lei wrote:
> The patch->mod is not a protected object of mutex_lock(&klp_mutex). Since
> it's in the error handling branch, it might not be helpful to reduce lock
> conflicts, but it can reduce some code size.
> 
> Before:
>    text    data     bss     dec     hex filename
>   10330     464       8   10802    2a32 kernel/livepatch/core.o
> 
> After:
>    text    data     bss     dec     hex filename
>   10307     464       8   10779    2a1b kernel/livepatch/core.o
> 

Is a size change expected, or is it just compiler fall out from
shuffling the code around a little bit?

I see some arches do a little better, some a little worse with gcc-9.3.0
cross compilers:

Before
------
   text    data     bss     dec     hex filename
   8490     600       8    9098    238a arm64/kernel/livepatch/core.o
   9424     680       8   10112    2780 s390/kernel/livepatch/core.o
   9802     228       4   10034    2732 ppc32/kernel/livepatch/core.o
  13746     456       8   14210    3782 ppc64le/kernel/livepatch/core.o
  10443     464       8   10915    2aa3 x86_64/kernel/livepatch/core.o


After
-----
   text    data     bss     dec     hex filename
   8514     600       8    9122    23a2 arm64/kernel/livepatch/core.o
   9424     680       8   10112    2780 s390/kernel/livepatch/core.o
   9818     228       4   10050    2742 ppc32/kernel/livepatch/core.o
  13762     456       8   14226    3792 ppc64le/kernel/livepatch/core.o
  10446     464       8   10918    2aa6 x86_64/kernel/livepatch/core.o

In which case, I'd just omit the size savings from the commit msg.

> Signed-off-by: Zhen Lei <thunder.leizhen@...wei.com>
> ---
>  kernel/livepatch/core.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/kernel/livepatch/core.c b/kernel/livepatch/core.c
> index 42f7e716d56bf72..cb7abc821a50584 100644
> --- a/kernel/livepatch/core.c
> +++ b/kernel/livepatch/core.c
> @@ -1041,9 +1041,9 @@ int klp_enable_patch(struct klp_patch *patch)
>  	mutex_lock(&klp_mutex);
>  
>  	if (!klp_is_patch_compatible(patch)) {
> +		mutex_unlock(&klp_mutex);
>  		pr_err("Livepatch patch (%s) is not compatible with the already installed livepatches.\n",
>  			patch->mod->name);
> -		mutex_unlock(&klp_mutex);
>  		return -EINVAL;
>  	}
>  
> 

That said, I don't see anything obviously wrong about the change (we
don't need to sync our error msgs, right?) so:

Acked-by: Joe Lawrence <joe.lawrence@...hat.com>

-- 
Joe