i am trying to ask this again: i supply an apache 2.2 module to my customers. naturally, i need to build my module for each OS, cpu architecture & bit scheme. So, if my customer has Apache 2.2 on solaris sparc 32 bit - that's what i need to build.
however, from what i just discovered, there's another factor - LFS support. My module needs to be compiled with the LFS enabled or disabled - according to the specific apache of the specific customer. The problem is that i can't know what LFS support the apache was compiled with. This really shouts "bug" to me, but i would really like to know - who's bug and how to fix it. for example, if the packages of apache 2.2.0 with APR 1.2.2 on sunfreeware.com have a bug that was later fixed - then i can tell my customer to upgrade to a later version. ??? On Mon, Nov 21, 2011 at 11:44 PM, Asaf Dalet <asaf.da...@gmail.com> wrote: > this means that apache modules for 2.2.0 32 bits, on solaris 9 are not > compatible across differemt ./configure options. > > which actually means that if i want to check if my module will run on a > given apache - i cannot do it (there's no way to know a priori the size of > apr_off_t which apache was compiled with). am i right? > > > > by any chance, does anyone know what are the ./configure options that > control the size of apr_off_t, (if there are any)? > > > > thanks > > asaf > > On Mon, Nov 21, 2011 at 7:23 PM, William A. Rowe Jr. > <wr...@rowe-clan.net>wrote: > >> On 11/21/2011 9:48 AM, Asaf Dalet wrote: >> >>> Hi all, >>> some interesting news: >>> I downloaded apr-1.2.2-sol9-sparc-local from sunfreeware.com < >>> http://sunfreeware.com> and >>> >>> installed it. >>> the file apr/include/apr-1/apr.h contains this line: >>> typedef long apr_off_t; >>> >>> does this mean apr 1.2.2 does not support LFS or is it some bug in this >>> precompiled APR? >>> appreciate any comment >>> >> >> It means either 1) the builder deliberately forced this behavior >> through options to ./configure, or 2) apr fails to detect that >> LFS is available in the builder's environment. I suspect 1. >> >> >
