linux-kernel - Re: [patch] kexec: potetially using uninitialized variable

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: <20160311153819.GA14320@mhuang-ThinkPad-T440s>
Date:	Fri, 11 Mar 2016 23:38:19 +0800
From:	Minfei Huang <mnfhuang@...il.com>
To:	walter harms <wharms@....de>
Cc:	Dan Carpenter <dan.carpenter@...cle.com>, xlpang@...hat.com,
	Andrew Morton <akpm@...ux-foundation.org>,
	Vivek Goyal <vgoyal@...hat.com>,
	Dave Young <dyoung@...hat.com>, Ingo Molnar <mingo@...nel.org>,
	Toshi Kani <toshi.kani@....com>,
	Mimi Zohar <zohar@...ux.vnet.ibm.com>,
	linux-kernel@...r.kernel.org, kernel-janitors@...r.kernel.org
Subject: Re: [patch] kexec: potetially using uninitialized variable

On 03/11/16 at 10:47am, walter harms wrote:
> 
> 
> Am 11.03.2016 10:19, schrieb Dan Carpenter:
> > On Fri, Mar 11, 2016 at 04:52:43PM +0800, Xunlei Pang wrote:
> >> Hi Dan,
> >>
> >> On 2016/03/11 at 16:07, Dan Carpenter wrote:
> >>> At the end of the function we check if "ret" has a negative error code,
> >>> but it seems possible that it is uninitialized.
> >>>
> >>> Fixes: 12db5562e035 ('kexec: load and relocate purgatory at kernel load time')
> >>> Signed-off-by: Dan Carpenter <dan.carpenter@...cle.com>
> >>>
> >>> diff --git a/kernel/kexec_file.c b/kernel/kexec_file.c
> >>> index 503bc2d..63d1af3 100644
> >>> --- a/kernel/kexec_file.c
> >>> +++ b/kernel/kexec_file.c
> >>> @@ -795,7 +795,7 @@ out:
> >>>  
> >>>  static int kexec_apply_relocations(struct kimage *image)
> >>>  {
> >>> -	int i, ret;
> >>> +	int i, ret = 0;
> >>>  	struct purgatory_info *pi = &image->purgatory_info;
> >>>  	Elf_Shdr *sechdrs = pi->sechdrs;
> >>>  
> >>
> >> Look further, there is a condition at the beginning of the for loop:
> >>  
> >>
> >>         if (sechdrs[i].sh_type != SHT_RELA &&
> >>             sechdrs[i].sh_type != SHT_REL)
> >>             continue;
> >>
> >> So, I think that's ok, but I don't konw if GCC is smart enough not to throw warnings.
> > 
> > Ah, right...
> > 
> > This wasn't a GCC warning.  GCC misses a lot of uninitialized variable
> > bugs so I'm doing this with Smatch.
> > 
> > Anyway, I'll patch this up in Smatch to not warn about this.
> > 
> 
> I am not so sure about this. the point should be that the reviewer can read it easily
> not if gcc complains or not.

Hi, All.

I think we can modify the logic a bit to make code simple. Thus gcc will
not complain about any more, and the logic is earier.

Following is a draft patch.

diff --git a/kernel/kexec_file.c b/kernel/kexec_file.c
index 007b791..7144e3b 100644
--- a/kernel/kexec_file.c
+++ b/kernel/kexec_file.c
@@ -887,7 +887,7 @@ static int kexec_apply_relocations(struct kimage *image)
 		if (sechdrs[i].sh_type == SHT_RELA)
 			ret = arch_kexec_apply_relocations_add(pi->ehdr,
 							       sechdrs, i);
-		else if (sechdrs[i].sh_type == SHT_REL)
+		else
 			ret = arch_kexec_apply_relocations(pi->ehdr,
 							   sechdrs, i);
 		if (ret)


> 
> just my 2 cents,
> 
> re,
>  wh
> 
>