Tom Keiser wrote: > Does it even acquire a mutex most of the time? The Linux allocator is > based upon the Solaris slab allocator. That allocator just disables > preemption and tries to satisfy the allocations out of a cpu-local > slab, only grabbing a mutex if the local allocation fails. Our > home-grown allocators are going to need major surgery to compete with > such a highly-performant algorithm.
The overflow of the credentials allocation (over one page) was a performance hit also -- it causes the code to use vmalloc instead of kmalloc, and vmalloc has to set up page table entries. The patch I posted to keep it all on one page should eliminate this -- not to mention TLB thrashing. -- Joe Buehler _______________________________________________ OpenAFS-devel mailing list [email protected] https://lists.openafs.org/mailman/listinfo/openafs-devel
