Yossi Itigin

Port CI port range fixes to v1.13.x

1

## Why Port CI fixes of failing to start TCP listener because the port is in use from master branch ## How - Port #8423 - Port #8436 - Fixed...

UCM/CUDA/TEST: Install memory hooks for async Cuda allocations

6

## Why As discussed in #7194 and #7110 , need to add memory hooks support for cuda async allocations. Without this, applications using these allocations may fail to detect Cuda...

SHMEM/BASE: added missing initializers

1

Port https://github.com/open-mpi/ompi/pull/10822 to v4.1.x

UCP/PROTO: Move non-rndv protocols to use probe() instead of init()

1

## Why Next phase of adding protocol variants, moving from init() to probe()

## Why Add memory hooks for VMM allocations Potential fix for internal issue [4598755 ](https://nvbugspro.nvidia.com/bug/4598755)

## Why Allow keeping same request size if increasing max lanes

UCP/PROTO/TEST: Add protocol performance data strucutre

3

## Why Unite common code for protocol performance manipulations @ivankochin @brminich

UCP/PROTO/PERF: Add function to aggregate multiple performances

2

## Why Support combining multiple protocol parts