Re: [ANN] LuaJIT x64 port sponsorship program

[Patrick Donnelly <batrick@...>]

On Wed, Feb 3, 2010 at 2:03 PM, Tony Finch <dot@dotat.at> wrote:
> On Wed, 3 Feb 2010, Patrick Donnelly wrote:
>> On Tue, Feb 2, 2010 at 12:30 PM, Mike Pall <mikelu-1002@mike.de> wrote:
>> >
>> > What really counts is commonly accepted usage. So, while I can buy
>> > the need for MAP_FAILED et al, any reasonable Lua/C binding ought
>> > to convert this to nil or throw an error.
>>
>> What about "sentinel" defined values like those in the socket
>> libraries. On my GNU system:
>>
>> #define SIG_ERR ((__sighandler_t)-1)    /* error return from signal */
>>
>> I'm not sure why someone would want to push a SIG_ERR lightuserdata
>> but I imagine it could come up.
>
> SIG_ERR ought to be converted to nil just like MAP_FAILED, and
> PTHREAD_CANCELLED too, I think. I don't know of any odd pointer values in
> the sockets API. There are some funny RTLD_* and fenv.h FE_* pointer
> values on some systems. And I think that's about it for core Unix APIs.

My brain skipped the MAP_FAILED discussion (despite quoting it). :S
Thanks for pointing that out! Sorry for the noise.

-- 
- Patrick Donnelly