linux-kernel - Re: [PATCH] x86/ucode: replace redundant string literals

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: <5679947402000078000C2691@prv-mh.provo.novell.com>
Date:	Tue, 22 Dec 2015 10:20:36 -0700
From:	"Jan Beulich" <JBeulich@...e.com>
To:	"Borislav Petkov" <bp@...e.de>
Cc:	<mingo@...e.hu>, <tglx@...utronix.de>,
	<linux-kernel@...r.kernel.org>, <hpa@...or.com>
Subject: Re: [PATCH] x86/ucode: replace redundant string literals

>>> On 22.12.15 at 18:07, <bp@...e.de> wrote:
> On Tue, Dec 22, 2015 at 08:40:43AM -0700, Jan Beulich wrote:
>> This doesn't just eliminate needless redundancy
>> (plus avoid a possible disconnect if one string instance gets changed
>> without the other(s)),
> 
> This argument is bogus. We will never ever change a user-visible command
> line option. Ever.
> 
>> but also eliminates a warning some gcc versions emit ("array access
>> beyond array bounds", observed with 4.3.4) in the 32-bit case.
> 
> Now that I'm interested in - how exactly do you trigger this? gcc
> version, etc?

As said - 4.3.4 (I only now realize that this could also have been a
kernel version).

>> --- 4.4-rc6/arch/x86/kernel/cpu/microcode/core.c
>> +++ 4.4-rc6-x86-ucode-early-string/arch/x86/kernel/cpu/microcode/core.c
>> @@ -83,13 +83,11 @@ static bool __init check_loader_disabled
>>  {
>>  #ifdef CONFIG_X86_32
>>  	const char *cmdline = (const char *)__pa_nodebug(boot_command_line);
>> -	const char *opt	    = "dis_ucode_ldr";
>> -	const char *option  = (const char *)__pa_nodebug(opt);
>> +	const char *option  = (const char *)__pa_nodebug(__setup_str_disable_loader);
>>  	bool *res = (bool *)__pa_nodebug(&dis_ucode_ldr);
>> -
>>  #else /* CONFIG_X86_64 */
>>  	const char *cmdline = boot_command_line;
>> -	const char *option  = "dis_ucode_ldr";
>> +	const char *option  = __setup_str_disable_loader;
>>  	bool *res = &dis_ucode_ldr;
>>  #endif
> 
> I don't like it: it is not clear at a glance that this __setup_str*
> magic gets generated from the __setup macro. In addition, this code is
> as unreadable as it is now - your patch makes it even more cryptic.
> 
> I much prefer the redundancy.

And the compiler warning? At the very least these should then be
const char __initconst [].

Jan

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/