Re: [Musicpd-dev-team] aac fixes

2008-11-20 Thread Max Kellermann
On 2008/11/17 21:29, Kodest [EMAIL PROTECTED] wrote: I recompiled mpd and unfortunately those warnings are back again: decoder/aac_plugin.c: In function 'getAacFloatTotalTime': decoder/aac_plugin.c:259: warning: dereferencing type-punned pointer will break strict-aliasing rules

Re: [Musicpd-dev-team] aac fixes

2008-11-20 Thread Kodest
Hi, On 20 November 2008 14:59, Max Kellermann [EMAIL PROTECTED] wrote: What did configure say about broken libfaad headers on your machine? Look what config.log writes. config.log snippet: -- configure:9555: checking for broken libfaad headers configure:9578: gcc -std=gnu99 -c -pthread -g -O2

Re: [Musicpd-dev-team] aac fixes

2008-11-20 Thread Max Kellermann
On 2008/11/20 17:31, Kodest [EMAIL PROTECTED] wrote: And the following line comes up in the configure stdout: checking for broken libfaad headers... broken HAVE_FAAD_LONG is defined. So the following lines in the aac plugin are compiled which triggers the warnings: 259: unsigned long

Re: [Musicpd-dev-team] aac fixes

2008-11-20 Thread Kodest
Hi, On 20 November 2008 17:39, Max Kellermann [EMAIL PROTECTED] wrote: What type does NeAACDecInit2() expect on your machine? It's unsigned long *samplerate on the latest unmodified release (2.6.1), and it compiles without warnings here. I think you meant NeAACDecInit(), not NeAACDecInit2().

Re: [Musicpd-dev-team] aac fixes

2008-11-20 Thread Max Kellermann
On 2008/11/20 18:01, Kodest [EMAIL PROTECTED] wrote: I think you meant NeAACDecInit(), not NeAACDecInit2(). The mpd code calls faacDecInit() which is defined to NeAACDecInit() in neaacdec.h. Btw this doesn't matter while the signatures of these two functions are the same in this header for me: