httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Bill Stoddard" <redd...@attglobal.net>
Subject Re: cvs commit: apache-2.0/src/modules/standard mod_asis.cmod_file_cache.c
Date Tue, 11 Jul 2000 01:38:57 GMT

>
> >   Log:
> >   Reimplement ap_send_fd. Eliminate ap_send_fd_length. If APR_HAS_SENDFILE is
> >   defined but ap_sendfile fails with APR_ENOTIMPL, the BUFF implementation
> >   of ap_send_fd will get a shot at serving the request.  This fix is
> >   required to get Apache working on 95/98 again and can also be useful on
> >   Unix systems where sendfile is available via a servicepack/fixpack/PTF
> >   on a particular level of the OS (e.g., AIX 4.3.2 base does not include
> >   sendfile but is is available with a PTF).
>
> I am very confused by this.  If the PTF is applied to the machine, then
> APR will detect sendfile won't it?  If it isn't applied, then the machine
> doesn't have sendfile.  If Apache is compiled, and then the PTF is
> applied, then Apache/APR will need to be re-compiled in order to pick it
> up.  How is this useful on a Unix machine?
>

As in the Windows case, we would do a runtime check.  Say I compile an Apache binary kit on
AIX
4.3.2 (with the sendfile PTF applied). The binary kit works for me, but not on a 4.3.2 machine
w/o
the PTF applied. So, would need to do a runtime check for the PTF and not call send_file (and
return
APR_ENOTIMPL) if the PTF is not available. That's what I'm talking about. Not sure it is worth
the
effort, but now it's at least possible to do runtime checks if needed.

Bill


Mime
View raw message