You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@httpd.apache.org by Joe Orton <jo...@redhat.com> on 2005/09/16 18:39:16 UTC
Re: svn commit: r289593 - /httpd/httpd/branches/2.2.x/os/unix/os.h
On Fri, Sep 16, 2005 at 04:07:04PM -0000, William Rowe wrote:
> Author: wrowe
> Date: Fri Sep 16 09:07:01 2005
> New Revision: 289593
>
> URL: http://svn.apache.org/viewcvs?rev=289593&view=rev
> Log:
>
> I'd prefer, before 2.2-final, to have AP_NEED_SET_MUTEX_PERMS defined
> across all architectures as 1|0. Comments?
-0, it's a bikeshed which doesn't need repainting.
Re: svn commit: r289593 - /httpd/httpd/branches/2.2.x/os/unix/os.h
Posted by "William A. Rowe, Jr." <wr...@rowe-clan.net>.
Joe Orton wrote:
>
> Sounds reasonable, +0 then but please wait until it's been reviewed
> before putting it on the 2.2.x branch.
2.2.x branch is CTR, no?
But I indended to ask before moving forward with the matching #define's
of 0 value and changing the various tests throughout httpd to test #if
in place of #ifdef. Whatever I commit to trunk I'll be backporting, we
won't be able to change the meaning once we ship 2.2.0. That's why I
asked the question in the commit message.
The first commit fixed breakage against using the value of that macro
and otherwise was a noop to httpd.
Bill
Re: svn commit: r289593 - /httpd/httpd/branches/2.2.x/os/unix/os.h
Posted by Joe Orton <jo...@redhat.com>.
On Fri, Sep 16, 2005 at 12:15:50PM -0500, William Rowe wrote:
> Joe Orton wrote:
> >On Fri, Sep 16, 2005 at 04:07:04PM -0000, William Rowe wrote:
> >> I'd prefer, before 2.2-final, to have AP_NEED_SET_MUTEX_PERMS defined
> >> across all architectures as 1|0. Comments?
> >
> >-0, it's a bikeshed which doesn't need repainting.
>
> Actually, if you are the author of mod_foo (mod_jk, for example) and you
> intend to compile against either httpd-2.0 or 2.2, it's quite useful.
>
> I suppose it's possible to add all the versioning tests and use the
> platform flags for 2.0 builds; but simply trusting that the value of
> AP_NEED_SET_MUTEX_PERMS is correct, if defined, will simplify the rest
> of the logic. E.g...
Sounds reasonable, +0 then but please wait until it's been reviewed
before putting it on the 2.2.x branch.
joe
Re: svn commit: r289593 - /httpd/httpd/branches/2.2.x/os/unix/os.h
Posted by "William A. Rowe, Jr." <wr...@rowe-clan.net>.
Joe Orton wrote:
> On Fri, Sep 16, 2005 at 04:07:04PM -0000, William Rowe wrote:
>
>>Author: wrowe
>>Date: Fri Sep 16 09:07:01 2005
>>New Revision: 289593
>>
>>URL: http://svn.apache.org/viewcvs?rev=289593&view=rev
>>Log:
>>
>> I'd prefer, before 2.2-final, to have AP_NEED_SET_MUTEX_PERMS defined
>> across all architectures as 1|0. Comments?
>
> -0, it's a bikeshed which doesn't need repainting.
Actually, if you are the author of mod_foo (mod_jk, for example) and you
intend to compile against either httpd-2.0 or 2.2, it's quite useful.
I suppose it's possible to add all the versioning tests and use the
platform flags for 2.0 builds; but simply trusting that the value of
AP_NEED_SET_MUTEX_PERMS is correct, if defined, will simplify the rest
of the logic. E.g...
#ifdef AP_NEED_SET_MUTEX_PERMS
# define JK_NEED_SET_MUTEX_PERMS AP_NEED_SET_MUTEX_PERMS
#else
/* A special case for httpd-2.0 */
# if !defined(OS2) && !defined(WIN32) && !defined(BEOS) && !defined(NETWARE)
# define JK_NEED_SET_MUTEX_PERMS 1
# else
# define JK_NEED_SET_MUTEX_PERMS 0
# endif
#endif
Bill