Skip to contentSkip to conversation

Understanding schedular and slab allocation

DISCUSSION Sven Krohlas 1 participant 1 message Dec 15, 2004 3 messages not in the archive

lore
Share & tools
lore

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

A message that is not in the archive (sent off-list or not yet received): [email protected]

#3 lore raw

Hi,

Quoted 3 lines
Quoted 2 lines

Btw: is anybody working on the slab allocator as described in Bonwicks 2001 paper?

Linux already fronts the slab allocator with per-cpu pools.

Well, I haven't had such a deep look at the 2001 version (my topic was the current implementation)...
I've seen the per-cpu pools. Bonwick talks in his paper of allocations in guaranteed constant time (using vmem). Is this goal already achived in Linux? There is a lock protecting the cache_chain, doesn't this hurt scalability?

And what about the other improvements?

-vmem for allocating general resources
-A user-space implementation (well I know, that's a bit off topic here..)

Are they already implemented/is anybody working on this or are there already known better solutions than Bonwicks suggestions?

Greetings,
Sven