On 26 October 2007 23:46, David Miller wrote:
> From: "Dave Korn"
> Date: Fri, 26 Oct 2007 17:35:44 +0100
>
>> On 26 October 2007 17:28, Andrew Haley wrote:
>>
>>> Richard Guenther writes:
>>> > >
>>> > > This is legal POSIX threads code: counter is not accessed when we do
>>> > > not hold the mutex. According to POSIX we do not have to declare
>>> > > volatile memory that we only access when we hold a mutex. >
>>> > I hope we're not trying to support such w/o volatile counter.
>>>
>>> I think we have to: not just for POSIX, but for the Linux kernel too.
>>>
>>> > Whatever POSIX says, this would pessimize generic code too much.
>>>
>>> We don't have to do it for non-threaded code.
>>
>> I certainly won't object to any move to prohibit the
>> read-conditional-add-write (and related) optimisation(s) when compiling
>> with an option that explicitly specifies that we are compiling
>> multi-threaded code.
>
> What about signals?
Heh, you got me there. What about signals?
> Those are just another asynchronous context with similar issues.
Sure. Lump it all in with threaded in the above statement.
> Please don't point me at standards documents in your response,
> that is not what truly matters here.
Let me generalise the above statement:
>> I certainly won't object to any move to prohibit
... any optimisations whatsoever, including but not limited to ...
>> read-conditional-add-write (and related) optimisation(s) when compiling
>> with an option that explicitly specifies that we are compiling
... any kind of asynchronous or ...
>> multi-threaded code.
cheers,
DaveK
--
Can't think of a witty .sigline today....
