On 13/09 09.33, Sunanda Menon wrote: > Bjorn , > > Not sure why you say the patch fails ?This patch has been in use for a > while now and I can see the patched file in the source .
OK, the way you invoke the patch is doesn't fail outright, as it did when I tried it in a different context. However it does hint at a problem: --- patching file mysql-5.0.45/innobase/include/univ.i Hunk #1 succeeded at 58 with fuzz 2. --- Note the "fuzz 2". Here's the patch file: --- --- mysql-5.0.45/innobase/include/univ.i 2007-11-03 10:59:11.653348000 -0700 +++ univ.i.new 2007-11-03 11:00:48.048649000 -0700 @@ -58,7 +58,7 @@ /* We only try to do explicit inlining of functions with gcc, Microsoft Visual C++ and Sun Compilers */ -#if !defined(__GNUC__) && !defined(__WIN__) +#if !defined(__GNUC__) && !defined(__WIN__) && !defined(__SUNWSPRO_C) #undef UNIV_MUST_NOT_INLINE /* Remove compiler warning */ #define UNIV_MUST_NOT_INLINE #endif --- It changes the "#if !defined..." line, but if you look at the context, you see that the comment in the lines above has been taken from the *new* version of the file, and does not match the original file. That's why my attempt to patch failed. And if you look at the resulting file in your workspace, you'll see that it still has the old version of the comment which does not refer to "Sun Compilers". My version of the patch file (make sure to take the second one) changes both the comment *and* the #if line, as I'm sure was th eoriginal intention. -- Bjorn Munch Sun Microsystems Trondheim, Norway http://sun.com/postgresql/
