Commit 19453f34 authored Nov 24, 2021 by Christophe JAILLET Committed by Jason Gunthorpe Nov 25, 2021

IB/mthca: Use non-atomic bitmap functions when possible in 'mthca_allocator.c'

The accesses to the 'alloc->table' bitmap are protected by the
'alloc->lock' spinlock, so no concurrent accesses can happen.

So prefer the non-atomic '__[set|clear]_bit()' functions to save a few
cycles.

Link: https://lore.kernel.org/r/5f909ca1284fa4d2cf13952b08b9e303b656c968.1637785902.git.christophe.jaillet@wanadoo.fr


Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

parent a277f383

drivers/infiniband/hw/mthca/mthca_allocator.c

+2 −2

Original line number	Diff line number	Diff line
		@@ -51,7 +51,7 @@ u32 mthca_alloc(struct mthca_alloc *alloc)
		}

		if (obj < alloc->max) {
		set_bit(obj, alloc->table);
		__set_bit(obj, alloc->table);
		obj \|= alloc->top;
		} else
		obj = -1;
		@@ -69,7 +69,7 @@ void mthca_free(struct mthca_alloc *alloc, u32 obj)

		spin_lock_irqsave(&alloc->lock, flags);

		clear_bit(obj, alloc->table);
		__clear_bit(obj, alloc->table);
		alloc->last = min(alloc->last, obj);
		alloc->top = (alloc->top + alloc->max) & alloc->mask;