Geoffrey Young wrote:
-#define MP_dINTERP_PUTBACK(interp) dNOOP
+#define MP_INTERP_PUTBACK(interp) NOOP
ah, I missed the dNOOP change and only saw the name change :)
I'm looking forward for mandrake cooker gcc picking up the changes
from the end of july where the gcc team has added a new flag
John Peacock wrote:
Stas Bekman wrote:
Good, but it's not the real fix, can you please revert that patch and
try a new patch at the end of this file? Thanks.
Reverted, patched, compiled, tested, installed! I really wish I had
sent my message yesterday, so I wouldn't have flopped around quite
Stas Bekman wrote:
Good, but it's not the real fix, can you please revert that patch and
try a new patch at the end of this file? Thanks.
Reverted, patched, compiled, tested, installed! I really wish I had sent my
message yesterday, so I wouldn't have flopped around quite so much on my own...
-#define MP_dINTERP_PUTBACK(interp) dNOOP
+#define MP_INTERP_PUTBACK(interp) NOOP
ah, I missed the dNOOP change and only saw the name change :)
I'm looking forward for mandrake cooker gcc picking up the changes from
the end of july where the gcc team has added a new flag that will
complain abou
Geoffrey Young wrote:
-#ifdef USE_ITHREADS
-MP_dINTERP_PUTBACK(interp);
-#endif
+MP_INTERP_PUTBACK(interp);
yeah, that makes sense - the macro is already a no-op for non-threaded
perls.
but why the name change? it parallels MP_dINTERP_SELECT as it stands now.
That's exactly the mistak
-#ifdef USE_ITHREADS
-MP_dINTERP_PUTBACK(interp);
-#endif
+MP_INTERP_PUTBACK(interp);
yeah, that makes sense - the macro is already a no-op for non-threaded perls.
but why the name change? it parallels MP_dINTERP_SELECT as it stands now.
--Geoff
John Peacock wrote:
Stas Bekman wrote:
Yup, I know, Carl has reported this earlier, but never followed up on
my suggestion. Please try this patch:
Excellent! That gets it building and all the tests pass!
Good, but it's not the real fix, can you please revert that patch and try a
new patch at
Stas Bekman wrote:
Yup, I know, Carl has reported this earlier, but never followed up on my
suggestion. Please try this patch:
Excellent! That gets it building and all the tests pass!
because the -Wl is the cc option that should have been stripped before
passing the -E to ld. :~(
Yup, that's
John Peacock wrote:
I am building a new server (FreeBSD 4.8-RELEASE) and wanted to get
mod_perl2 working. If I use the /usr/ports/www/mod_perl2 and
/usr/ports/lang/perl5.8, I can build mod_perl and it works like a
charm. However, nothing I do with with Perl5.8.1_rc5 works.
With the mod_perl2
I am building a new server (FreeBSD 4.8-RELEASE) and wanted to get mod_perl2
working. If I use the /usr/ports/www/mod_perl2 and /usr/ports/lang/perl5.8, I
can build mod_perl and it works like a charm. However, nothing I do with with
Perl5.8.1_rc5 works.
With the mod_perl2 CVS, I get this erro
10 matches
Mail list logo
