Thread
-
RE: [PORTS] Re: [HACKERS] no answer to Solaris 2.6 failure to bu
Roman Volkoff <rv@tlg.khv.ru> — 1998-07-02T17:09:26Z
unsubscribe
-
Re: [HACKERS] no answer to Solaris 2.6 failure to build 6.3.2?
Tom Lane <tgl@sss.pgh.pa.us> — 1998-07-03T01:23:21Z
De Clarke <de@ucolick.org> writes: > I still cannot get postgres 6.3.2 libpq to build under > Solaris 2.6 ... this means I can't test any of my apps > using postgres on my deployment platform. can anyone > help? > gcc version 2.7.2.3 > fe-connect.c: In function `PQconnectdb': > fe-connect.c:204: warning: implicit declaration of function `strdup' > fe-connect.c:204: warning: assignment makes pointer from integer without a cast [ snip a whole lot of similar errors, all apparently arising from the lack of prototypes for strdup() and other functions... ] gcc is unhappy because it hasn't seen any declaration for strdup, and later fdopen, strcasecmp, etc. All the other complaints follow from that. Either Solaris 2.6 has incredibly brain-damaged system include files, or (more likely) you have a misconfigured gcc that is not reading the correct version of <stdio.h>, <string.h>, etc. One way that that can happen is if you try to copy a gcc installation from another system rather than configuring and compiling it on exactly the target system. (gcc tends to like to make "patched" copies of some of the system include files, and if those don't match up with the real ones you are in deep trouble.) Surely libpq is not the only area where things are failing to build because of these problems? Or is that the only subdirectory you have tried to build? regards, tom lane
-
Re: [HACKERS] no answer to Solaris 2.6 failure to build 6.3.2?
Maarten Boekhold <maartenb@dutepp2.et.tudelft.nl> — 1998-07-03T07:44:24Z
On Thu, 2 Jul 1998, Tom Lane wrote: > De Clarke <de@ucolick.org> writes: > > I still cannot get postgres 6.3.2 libpq to build under > > Solaris 2.6 ... this means I can't test any of my apps > > using postgres on my deployment platform. can anyone > > help? > > > gcc version 2.7.2.3 > > > fe-connect.c: In function `PQconnectdb': > > fe-connect.c:204: warning: implicit declaration of function `strdup' > > fe-connect.c:204: warning: assignment makes pointer from integer without a cast > > [ snip a whole lot of similar errors, all apparently arising from the > lack of prototypes for strdup() and other functions... ] > > gcc is unhappy because it hasn't seen any declaration for strdup, and > later fdopen, strcasecmp, etc. All the other complaints follow from > that. > > Either Solaris 2.6 has incredibly brain-damaged system include files, > or (more likely) you have a misconfigured gcc that is not reading the > correct version of <stdio.h>, <string.h>, etc. One way that that can > happen is if you try to copy a gcc installation from another system > rather than configuring and compiling it on exactly the target system. > (gcc tends to like to make "patched" copies of some of the system > include files, and if those don't match up with the real ones you are > in deep trouble.) Isn't it possible to just run 'fixincludes'? This is what's done when you install gcc from scratch. Don't actually know where to find fixincludes though. Maarten _____________________________________________________________________________ | TU Delft, The Netherlands, Faculty of Information Technology and Systems | | Department of Electrical Engineering | | Computer Architecture and Digital Technique section | | M.Boekhold@et.tudelft.nl | -----------------------------------------------------------------------------