Updated patch with more tests and a first attempt at doc updates. As the commit message and doc now point out, using WaitForLockersMultiple() makes for a behavior difference with actually locking multiple tables, in that the combined set of conflicting locks is obtained only once for all tables, rather than obtaining conflicts and locking / waiting for just the first table and then obtaining conflicts and locking / waiting for the second table, etc. This is definitely desirable for my use case, but maybe these kinds of differences illustrate the potential awkwardness of extending LOCK?
Thanks again for any and all feedback!
v2-0001-Add-WAIT-ONLY-option-to-LOCK-statement.patch
Description: Binary data
