Re: [HACKERS] Parallel Bitmap scans a bit broken

On Thu, Mar 16, 2017 at 5:14 PM, Dilip Kumar <dilipbalaut@gmail.com> wrote:
> pg_atomic_write_u32_impl(val=0) at generic.h:57, queue = 'com.apple.main-thread', stop reason = EXC_BAD_ACCESS
(code=1,address=0x0)
 
>>>   * frame #0: 0x0000000100caf314 postgres`tbm_prepare_shared_iterate [inlined] pg_atomic_write_u32_impl(val=0) at
generic.h:57[opt]
 
>>>     frame #1: 0x0000000100caf314 postgres`tbm_prepare_shared_iterate [inlined] pg_atomic_init_u32_impl(val_=0) at
generic.h:163[opt]
 
>>>     frame #2: 0x0000000100caf314 postgres`tbm_prepare_shared_iterate [inlined] pg_atomic_init_u32(val=0) + 17 at
atomics.h:237[opt]
 
>
> By looking at the call stack I got the problem location.  I am
> reviewing other parts of the code if there are the similar mistake at
> other places. Soon I will post the patch.  Thanks for the help.

Based on the call stack I have tried to fix the issue. The problem is
there was some uninitialized pointer access (in some special cases
i.e. TBM_EMPTY when pagetable is not created at all).

 fix_tbm_empty.patch have fixed some of them but induced one which you
are seeing in your call stack.

Hopefully, this time I got it correct.  Since I am unable to reproduce
the issue so I will again need your help in verifying the fix.

-- 
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com

-- 
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers