Re: [HACKERS] planner fails on HEAD

2011-12-05 Thread Tom Lane
Pavel Stehule writes: > 2011/12/4 Tom Lane : >> Is this x86? I can't reproduce it on x86_64. > yes, this is x86 platform > uname -a > Linux nemesis 2.6.35.14-106.fc14.i686.PAE #1 SMP Wed Nov 23 13:39:51 > UTC 2011 i686 i686 i386 GNU/Linux I reproduced this with gcc 4.6.0 on Fedora 15 x86, too.

Re: [HACKERS] planner fails on HEAD

2011-12-05 Thread Pavel Stehule
2011/12/5 Merlin Moncure : > On Mon, Dec 5, 2011 at 12:17 PM, Pavel Stehule > wrote: >> Hello >> >> 2011/12/4 Tom Lane : >>> Pavel Stehule writes: it looks like gcc bug - gcc 4.5.1 20100924 (Red Hat 4.5.1) It was configured just with --enable-debug and --enable-cassert >>> >>> Is this

Re: [HACKERS] planner fails on HEAD

2011-12-05 Thread Merlin Moncure
On Mon, Dec 5, 2011 at 12:17 PM, Pavel Stehule wrote: > Hello > > 2011/12/4 Tom Lane : >> Pavel Stehule writes: >>> it looks like gcc bug - gcc 4.5.1 20100924 (Red Hat 4.5.1) It was >>> configured just with --enable-debug and --enable-cassert >> >> Is this x86?  I can't reproduce it on x86_64. >>

Re: [HACKERS] planner fails on HEAD

2011-12-05 Thread Tom Lane
Merlin Moncure writes: > On Sun, Dec 4, 2011 at 4:55 PM, Tom Lane wrote: >> Pavel Stehule writes: > it looks like gcc bug - gcc 4.5.1 20100924 (Red Hat 4.5.1) It was > configured just with --enable-debug and --enable-cassert >> >> Is this x86?  I can't reproduce it on x86_64. > reading all the

Re: [HACKERS] planner fails on HEAD

2011-12-05 Thread Merlin Moncure
On Sun, Dec 4, 2011 at 4:55 PM, Tom Lane wrote: > Pavel Stehule writes: >> it looks like gcc bug - gcc 4.5.1 20100924 (Red Hat 4.5.1) It was >> configured just with --enable-debug and --enable-cassert > > Is this x86?  I can't reproduce it on x86_64. reading all the comments in the gcc bug repor

Re: [HACKERS] planner fails on HEAD

2011-12-04 Thread Tom Lane
Pavel Stehule writes: > it looks like gcc bug - gcc 4.5.1 20100924 (Red Hat 4.5.1) It was > configured just with --enable-debug and --enable-cassert Is this x86? I can't reproduce it on x86_64. It's fairly easy to get a set of values such that innerstartsel *should* equal innerendsel; but if on

Re: [HACKERS] planner fails on HEAD

2011-12-04 Thread Pavel Stehule
2011/12/4 Tom Lane : > Pavel Stehule writes: >> 2011/12/4 Tom Lane : >>> [ scratches head ... ]  Given that it got past the previous assertions, >>> surely that ought to be impossible.  Could we see the values of >>> cost_mergejoin's local variables, please? > >> It is strange > >> when I put a fp

Re: [HACKERS] planner fails on HEAD

2011-12-04 Thread Tom Lane
Pavel Stehule writes: > 2011/12/4 Tom Lane : >> [ scratches head ... ] Given that it got past the previous assertions, >> surely that ought to be impossible. Could we see the values of >> cost_mergejoin's local variables, please? > It is strange > when I put a fprintf(stderr, "const literal")

Re: [HACKERS] planner fails on HEAD

2011-12-04 Thread Pavel Stehule
2011/12/4 Tom Lane : > Pavel Stehule writes: >> #3  0x083a1dfe in ExceptionalCondition (conditionName=0x8505474 >> "!(innerstartsel <= innerendsel)", errorType=0x83db178 >> "FailedAssertion", fileName=0x8505140 "costsize.c", lineNumber=1937) >> at assert.c:57 > > [ scratches head ... ]  Given that

Re: [HACKERS] planner fails on HEAD

2011-12-04 Thread Tom Lane
Pavel Stehule writes: > #3 0x083a1dfe in ExceptionalCondition (conditionName=0x8505474 > "!(innerstartsel <= innerendsel)", errorType=0x83db178 > "FailedAssertion", fileName=0x8505140 "costsize.c", lineNumber=1937) > at assert.c:57 [ scratches head ... ] Given that it got past the previous asse

Re: [HACKERS] planner fails on HEAD

2011-12-03 Thread Pavel Stehule
a plan for modified query is ohs=# explain analyze SELECT object_id, inserted, 'ASSIGN_RSLT', order_id, 2, seqnum, rejected_flat_file_id, true FROM ( SELECT q.object_id, fe.inserted, q.order_id, q.seqnum, q