Sorry, a mistake: > Ok, I've tested the simple example of a SRF from > http://www.postgresql.org/docs/8.4/interactive/xfunc-c.html (section > 34.9.10) and the application crash in the same way. When tries to > access SRF_IS_FIRSTCALL(), or any other PostgreSQL macro like > SRF_FIRST_CALL_INIT(), PG_GETARG_UINT32(0), etc. But in the rest of my > functions (non-SRF functions), everything works fine with these macros > and the rest of the code. The rest of my functions, non-SRF functions, don't use these macros, of course. But work fine. Only the SRF-function crash. Thanks and sorry for being annoying. On Fri, Apr 16, 2010 at 12:51 PM, Jorge Arevalo <jorgearevalo@xxxxxxxxxxxx> wrote: > On Thu, Apr 15, 2010 at 8:16 PM, Jorge Arevalo > <jorgearevalo@xxxxxxxxxxxx> wrote: >> On Tue, Apr 13, 2010 at 10:04 PM, Tom Lane <tgl@xxxxxxxxxxxxx> wrote: >>> Jorge Arevalo <jorgearevalo@xxxxxxxxxxxx> writes: >>>> I'm sorry, because this may be a simple question: I'm programming a C >>>> function that returns a set (PostgreSQL 8.4). The function crash, and >>>> I'm trying to >>>> debug it. I've read: >>> >>>> http://www.postgresql.org/docs/8.4/interactive/xfunc-c.html >>>> http://www.postgresql.org/docs/8.4/interactive/error-message-reporting.html >>>> http://www.postgresql.org/docs/current/static/runtime-config-logging.html >>> >>>> but I can't see any of my log messages. >>> >>> Seems likely to me that the function is crashing before it gets to any >>> of the elog calls; which would most likely be a problem in linking or in >>> declaring the function. Did you remember to use the >>> PG_FUNCTION_INFO_V1() macro? Try writing a trivial function that only >>> does an elog and see what it takes to get that to work. >>> >>> regards, tom lane >>> >> >> Yes, I put the PG_FUNCTION_INFO_V1(myFunction) macro. >> >> I solved some small bugs, and I checked the function is called, but I >> get a segmentation fault when try to call SRF_IS_FIRSTCALL. I think >> this has no much sense, and the problem keeps hidden for another >> reason. >> >> I'm going to continue debugging, but if helps, I provide the >> interesting part of log file. >> >> Thanks! >> >> Best regards, >> Jorge >> > > Ok, I've tested the simple example of a SRF from > http://www.postgresql.org/docs/8.4/interactive/xfunc-c.html (section > 34.9.10) and the application crash in the same way. When tries to > access SRF_IS_FIRSTCALL(), or any other PostgreSQL macro like > SRF_FIRST_CALL_INIT(), PG_GETARG_UINT32(0), etc. But in the rest of my > functions (non-SRF functions), everything works fine with these macros > and the rest of the code. > > I provide another log file, with the results of the simple example > execution. Another signal 11: Segmentation Fault. caused by line with > SRF_IS_FIRSTCALL(). > > Debugging... > -- Sent via pgsql-general mailing list (pgsql-general@xxxxxxxxxxxxxx) To make changes to your subscription: http://www.postgresql.org/mailpref/pgsql-general
