We decided with AMAT to remove gres/mps percent specification.

The GRES Links info was added to v18.08, but the unpack was
missing that field in v19.05.
bug 6337

Issuing job cancel on the wrong job ID. Also increase sleep in a
test to better try to overlap the jobs.

Support GRES types that include numbers (e.g. "--gres=gpu:123g:2").
Also change number suffix processing to use common function
See commit 396e82a167
bug 6345

See commit 396e82a167
bug 6345

See commit 396e82a167
bug 6345

Based on logic previously in src/plugins/burst_buffer/common/burst_buffer_common.c
bug 6345

If a Unix socket path exceeds the limit, emit errors and exit.
Fix memory leak on socket() failure.

Bug 5386

Need to easily detect real memory leaks

Forgot to free a list

The user can run multiple jobs on different nodes at the same time
or any number of non-gres/mps jobs on the same node. Different users
can also use MPS at the same time on any given node.

bug 6313

This is useful for testing and debugging.

Bug 6321

This reverts commit 4dc5839f.

Causes build failure on CentOS 7.6. Introduced in f7da0636.

Bug 6319.

Bug 5916.

As pthread_rwlock_t is a struct, we cannot use this syntax
to initialize all elements of the array to the same value
as we would if these were any basic type.

Instead, move the initialization to a dedicated loop to ensure
all locks are initialized correctly.

This does not seem to have caused any issues on Linux platforms,
but macOS at least does notice this mistake.

Bug 6319.

Add logic in read_config.c to handle old config option name.

Bug 5916.

This disables the test of tracking GPU/MPS by Type.

bug 5840

Add and comment out some debug messages to accrue time accounting.

Bug 6096

Added a new DebugFlag=accrue to show debug messages related to accrue time
count or other related stuff.

Bug 6096

Make sure count used is that for a single GPU rather than a total MPS
over multiple GPUs

Coverity CID 190461

This would be triggered by the gres.conf on the compute node having a
higher count of GRES defined with "File" parameters than the node has
defined in slurm.conf.

This also adds documentation and some tests. The actual scheduling
logic is still not written.