linux-kernel - Re: [PATCH 2/4] msm: scm: Fix improper register assignment

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: <4D688AF1.1090607@codeaurora.org>
Date:	Fri, 25 Feb 2011 21:09:05 -0800
From:	Saravana Kannan <skannan@...eaurora.org>
To:	Will Deacon <will.deacon@....com>
CC:	Stephen Boyd <sboyd@...eaurora.org>, linux-arm-msm@...r.kernel.org,
	David Brown <davidb@...eaurora.org>,
	linux-kernel@...r.kernel.org, linux-arm-kernel@...ts.infradead.org
Subject: Re: [PATCH 2/4] msm: scm: Fix improper register assignment

On 02/25/2011 05:23 AM, Will Deacon wrote:
> On Thu, 2011-02-24 at 18:44 +0000, Stephen Boyd wrote:
>> Assign the registers used in the inline assembly immediately
>> before the inline assembly block. This ensures the compiler
>> doesn't optimize away dead register assignments when it
>> shouldn't.
>>
>> Signed-off-by: Stephen Boyd<sboyd@...eaurora.org>
>> ---
>>   arch/arm/mach-msm/scm.c |    7 +++++--
>>   1 files changed, 5 insertions(+), 2 deletions(-)
>>
>> diff --git a/arch/arm/mach-msm/scm.c b/arch/arm/mach-msm/scm.c
>> index ba57b5a..5eddf54 100644
>> --- a/arch/arm/mach-msm/scm.c
>> +++ b/arch/arm/mach-msm/scm.c
>> @@ -264,13 +264,16 @@ u32 scm_get_version(void)
>>   {
>>          int context_id;
>>          static u32 version = -1;
>> -       register u32 r0 asm("r0") = 0x1<<  8;
>> -       register u32 r1 asm("r1") = (u32)&context_id;
>> +       register u32 r0 asm("r0");
>> +       register u32 r1 asm("r1");
>>
>>          if (version != -1)
>>                  return version;
>>
>>          mutex_lock(&scm_lock);
>> +
>> +       r0 = 0x1<<  8;
>> +       r1 = (u32)&context_id;
>>          asm volatile(
>>                  __asmeq("%0", "r1")
>>                  __asmeq("%1", "r0")
>
>
> Whoa, have you seen the compiler `optimise' the original assignments
> away? Since there is a use in the asm block, the definition shouldn't
> be omitted. What toolchain are you using?
>

Yeah, Stephen and I spent quite a bit of time discussing this and 
experimenting to figure out what the heck GCC was doing. But it kept 
optimizing the fake code we put in trying to force GCC to use a specific 
register.

My hypothesis at this point is that the "register xx asm("rx")" 
declarations are just for giving a symbolic name to refer to the 
specific register in C code. I doesn't tell GCC to reserve away the 
register and make sure the value is preserved. And the assignments to 
these said variables seem to translate to a pure "mov rx, 5" kinda 
instruction with no further preservation of rx either.

That's the only hypothesis I/we could come up with as to how this got 
optimized away.

I would be great if someone explains the exact meaning of these 
"register asm" declarations and the assignments in C code.

-Saravana

-- 
Sent by an employee of the Qualcomm Innovation Center, Inc.
The Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum.
--
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/