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: <BLU437-SMTP36FDD8606D9D645C6428697B10@phx.gbl>
Date:	Wed, 24 Sep 2014 08:42:37 -0400
From:	John David Anglin <dave.anglin@...l.net>
To:	nick <yocto6@...il.com>
CC:	Guenter Roeck <linux@...ck-us.net>, jejb@...isc-linux.org,
	deller@....de, akpm@...ux-foundation.org, mgorman@...e.de,
	julia.lawall@...6.fr, himangi774@...il.com,
	linux-parisc@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] parisc:Remove unnecessary FIXMES in init.c

On 24-Sep-14, at 7:42 AM, nick wrote:

>
>
> On 14-09-24 01:30 AM, Guenter Roeck wrote:
>> On Tue, Sep 23, 2014 at 09:49:43PM -0400, Nicholas Krause wrote:
>>> This removes the two fixmes in the file, init.c for compiler hints
>>> for comments related to compiler hints in linux_gateway_page_addr
>>> and map_hpux_gateway_page to change from FIXME to HINT in order
>>> for people reading this code to understand that these are compiler
>>> hints.
>>>
>>
>> "* const" tells the compiler that a pointer is a constant.
>> It doesn't tell the compiler that the data shall not be treated
>> as DP-relative.
>>
>> FIXME as in: We should find a better way to tell the compiler to
>> not treat this data as DP-relative.
>>
>> Can you please stop this ?
>>
>> Guenter
>>
>>> Signed-off-by: Nicholas Krause <yocto6@...il.com>
>>> ---
>>> arch/parisc/mm/init.c | 4 ++--
>>> 1 file changed, 2 insertions(+), 2 deletions(-)
>>>
>>> diff --git a/arch/parisc/mm/init.c b/arch/parisc/mm/init.c
>>> index 0bef864..668102e 100644
>>> --- a/arch/parisc/mm/init.c
>>> +++ b/arch/parisc/mm/init.c
>>> @@ -733,7 +733,7 @@ static void __init pagetable_init(void)
>>> static void __init gateway_init(void)
>>> {
>>> 	unsigned long linux_gateway_page_addr;
>>> -	/* FIXME: This is 'const' in order to trick the compiler
>>> +	/* HINT: This is 'const' in order to trick the compiler
>>> 	   into not treating it as DP-relative data. */
>>> 	extern void * const linux_gateway_page;
>>>
>>> @@ -761,7 +761,7 @@ map_hpux_gateway_page(struct task_struct *tsk,  
>>> struct mm_struct *mm)
>>> 	unsigned long start_pte;
>>> 	unsigned long address;
>>> 	unsigned long hpux_gw_page_addr;
>>> -	/* FIXME: This is 'const' in order to trick the compiler
>>> +	/* HINT: This is 'const' in order to trick the compiler
>>> 	   into not treating it as DP-relative data. */
>>> 	extern void * const hpux_gateway_page;
>>>
>>> -- 
>>> 1.9.1
>>>
>>> --
>>> 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/
>>>
>>>
> Guenter,
> I asked the maintainers and I got,
> On 9/23/2014 4:48 PM, Helge Deller wrote:
>> The FIXME: is misleading. It should maybe better read "HINT:" or  
>> something like that.
>> Trivially it's basically just casting the symbol so that the compiler
>> will produce the correct code later on.
>> There is nothing you could try to "fix".
> Why not just remove "FIXME: " from these comments?


The wording of the comment is not good.  You can't "trick" the  
compiler into not
treating a constant as DP-relative data.  Whether or not a constant is  
loaded as an
immediate depends on its value.

Dave
--
John David Anglin	dave.anglin@...l.net



--
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/

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ