[Jamin] x86_64 problem

From: John Rigg ^lt;jam_at_sound-man.co.uk>
Date: 06/11/06 13:39 EDT
Message-Id: <E1FpTuY-0001ee-OA@localhost.localdomain>
I just compiled jamin-0.95.0 after I found that the Debian-packaged
version (in Debian testing) hangs for 20 or 30 seconds with a blank
window on start up.
The compiled version does exactly the same. There's also something
strange happening in the xruns display, again in both versions.
The count keeps increasing even when there's no audio being processed.
jackd doesn't display any xruns and there isn't anything audibly wrong
(I managed to master a CD for a client with it behaving like this).

gcc produced _lots_ of warnings, but at the top of the list were
the following:

compressor-ui.c:133: warning: cast to pointer from integer of different size
compressor-ui.c:134: warning: cast to pointer from integer of different size
compressor-ui.c:135: warning: cast to pointer from integer of different size
compressor-ui.c:136: warning: cast to pointer from integer of different size
compressor-ui.c:137: warning: cast to pointer from integer of different size
compressor-ui.c:138: warning: cast to pointer from integer of different size
compressor-ui.c:147: warning: cast from pointer to integer of different size
geq.c:71: warning: cast to pointer from integer of different size
geq.c:221: warning: cast from pointer to integer of different size

As this is x86_64, pointers (64 bits) shouldn't be cast to int (32 bits) and
vice versa. Using long int (64 bits) would be correct and would still
work on x86 where pointer and long int are both 32 bits (and AFAIK
the same applies to 32 and 64-bit PPC).

The rest of the gcc warnings (hundreds of them) were of the following form:

state.c:503: warning: pointer targets in passing argument 1 of 'xmlNewDoc' differ in signedness
state.c:504: warning: pointer targets in passing argument 3 of 'xmlNewDocRawNode' differ in signedness

I'm not sure if the warnings are anything to do with the GUI hang or the
xrun count, but it looks a little complicated to fix. I started to look
at the pointer size issue but gave up when I saw the size of the macro
defining connect_scale in compressor-ui.c. Has anyone else experienced
these problems on AMD64?

John



_______________________________________________
Jamin-devel mailing list
Jamin-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/jamin-devel
Received on Sun Jun 11 13:23:21 2006

This archive was generated by hypermail 2.1.8 : 06/11/06 13:23 EDT