Skip to contentSkip to conversation

slab: introduce kmem_cache_zalloc allocator

PATCH Bodo Eggert 4 participants 4 messages Mar 20, 2006 – Mar 21, 2006 4 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]

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

#4 lore raw
Quoted 7 lines · Balbir Singh
Quoted 4 lines

No, no, no! I am introducing kmem_cache_zalloc() because there are existing users in the tree. I plan to kill the slab wrappers from XFS completely which is why I need this. We already have object constructors for what you're describing.

Ok, please keep the interface - build kmem_cache_zalloc() on top of what I suggest.

The benefit of using *zalloc is the ability to skip the memset by using pre-zeroed memory or to use more efficient ways of zeroing a page. Having to check the value of a parameter wouldn't help.

Signature · 2 lines
Ich danke GMX dafür, die Verwendung meiner Adressen mittels per SPF
verbreiteten Lügen zu sabotieren.
#5 lore raw
Quoted 6 lines
Quoted 2 lines

Ok, please keep the interface - build kmem_cache_zalloc() on top of what I suggest.

The benefit of using *zalloc is the ability to skip the memset by using pre-zeroed memory or to use more efficient ways of zeroing a page. Having to check the value of a parameter wouldn't help.

Hmm... the current patch directly does a memset(). Are you talking about possible optimizations to kmem_cache_zalloc()?

Balbir

#6 lore raw
Quoted 9 lines · Balbir Singh
Quoted 6 lines
Quoted 2 lines

Ok, please keep the interface - build kmem_cache_zalloc() on top of what I suggest.

The benefit of using *zalloc is the ability to skip the memset by using pre-zeroed memory or to use more efficient ways of zeroing a page. Having to check the value of a parameter wouldn't help.

Hmm... the current patch directly does a memset(). Are you talking about possible optimizations to kmem_cache_zalloc()?

Yes. At least that's what I understand from the whole zalloc process.

Signature · 2 lines
Top 100 things you don't want the sysadmin to say:
4. This won't affect what you're doing.
#7 lore raw
Quoted 1 line · Bodo Eggert

Yes. At least that's what I understand from the whole zalloc process.

Zalloc does two things: (1) reduce human error and (2) shrink kernel text size. Pre-zeroing is a possible optimization but needs to be measured. It is not necessarily a positive gain due to cache effects.

Pekka