Out of memory error when trying to compile drizzletest.cc

[Lee Bieber]

As of build 1096 we still two machines (debian 32 bit and centos 32 bit) where the machine runs out of memory when trying to compile drizzletest.cc

From Jay:
OK, so, I've tracked down the issue I was having with the replication branch to a difference in behaviour in atomic<> variables on CentOS (osindriz03) vs. other machines...hoping someone has an answer.

Here is the code which is behaving differently depending on the platform:

log_offset is declared as drizzled::atomic<off_t>.

log_offset= 0;
uint64_t length= 42;

off_t cur_offset;

cur_offset= log_offset.fetch_and_add((sizeof(uint64_t) + length));

On ALL other platforms other than CentOS 5, the above code sets log_offset AND cur_offset equal to 50.

On CentOS, it sets log_offset to 50 and cur_offset to 0.

What does this mean? It means that the assignment to cur_offset is happening *before* the call to drizzled::atomic<off_t>::fetch_and_add().

Without resorting back to mutexes, I am uncertain how to solve this particular timing/atomics problem. I'm hoping someone out there has more experience dealing with this.

[Jay Pipes (jaypipes)]

The comment below is about an unrelated bug (also on CentOS, which is probably where the confusion arose). Assigning back to nobody for now. Devananda and I tracked it down to something in r1095 that produces the problem, so we should investigate from there.