On 05/21/2013 11:22 AM, Martin Storsjö wrote:
> On Tue, 21 May 2013, Luca Barbato wrote:
> 
>> On 05/21/2013 10:43 AM, Martin Storsjö wrote:
>>> ---
>>>  configure |    3 +++
>>>  1 file changed, 3 insertions(+)
>>>
>>> diff --git a/configure b/configure
>>> index c84050c..792b77b 100755
>>> --- a/configure
>>> +++ b/configure
>>> @@ -3196,6 +3196,9 @@ elif check_func_headers stdlib.h _get_doserrno;
>>> then
>>>                                   vsnprintf=avpriv_vsnprintf
>>>  elif check_cpp_condition stddef.h "defined __KLIBC__"; then
>>>      libc_type=klibc
>>> +elif check_cpp_condition sys/cdefs.h "defined __BIONIC__"; then
>>> +    libc_type=bionic
>>> +    add_compat strtod.o strtod=avpriv_strtod
>>>  fi
>>>
>>>  test -n "$libc_type" && enable $libc_type
>>
>> Upstream won't fix that?
> 
> They might, but it's not very high on their list of priorities. The
> source for the strtod function is from a pretty old netbsd version - I
> tried hacking in rudimentary support for a better strtod(), but they
> said they'd rather try to resync with the newer upstream netbsd version
> where strtod was implemented more completely than deviate more by adding
> another not-totally complete implementation. But the netbsd libc had
> changed pretty much before that function was improved in the netbsd
> tree, so it's not dead simple to cherrypick that particular fix from
> there either.

Patch ok.

_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to