Fyi, hybrid mutex support in BDB 4.6.21 is still not a good idea on 2.6 Linux systems.
Howard Chu wrote:
Fyi, hybrid mutex support in BDB 4.6.21 is still not a good idea on 2.6 Linux systems.
And just to be clear, because I probably didn't include enough context above - the "fix" for the poor locking behavior in 4.6.19 was just the addition of a yield() call in the spinlock loop, and yield() is known to have very poor behavior on 2.6 Linux NPTL systems.