2.18.1-1

Add support for IB SHARP to NVLS (NVLink SHARP algorithm).
Add NVLS+Tree algorithm.
Add support for memory management using cuMem* functions.
Use all NICs for Send/Receive operations on systems with more than
one NIC per GPU (#804).
Add ncclCommSplit primitive, with resource sharing option in config.
Fix alltoallv hang (#788)
Increase number of channels on H100 when we're not limited by NVLink.
Improve error reporting in case of IB failure, printing local and
remote ID (#779).
Add build option to allow compilation against RDMA includes instead
of dynamically loading IB verbs symbols (#802).
Fix context creation for progress thread (#803).
NET/IB: add option to use multiple QPs in round-robin mode.
Fix tree performance issue when NVB is disabled on HCM topologies.

src/include/align.h

@@ -13,6 +13,9 @@
 #define ROUNDUP(x, y) \
     (DIVUP((x), (y))*(y))
 
+#define ALIGN_POWER(x, y) \
+    ((x) > (y) ? ROUNDUP(x, y) : ((y)/((y)/(x))))
+
 #define ALIGN_SIZE(size, align) \
   size = ((size + (align) - 1) / (align)) * (align);