tech-kern archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: UVM patch for PR port-xen/45975

Chuck Silvers <> wrote:
>  - it would be better to remove the mappings before freeing the page
>    on all platforms, not just xen.  that's just a good practice.
>    if you're concerned about a potential performance impact,
>    you could process the pages in batches of 16 or so using a little
>    array on the stack, which should handle most operations in one batch.
>    I did a quick experiment just now and found that most calls
>    to uvm_km_pgremove_intrsafe() are for 4 pages or less.
>    (I wonder why those aren't using some higher-level cache?)
>    about 10% of the calls were for 16 pages, and none were larger.

Yes.  Also, one should be careful about potential race condition - that is,
TLB invalidations must happen *before* VA becomes available for use by other
processes.  Otherwise, memory might be allocated with that VA range and used
while TLB caches have stale entries i.e. pointing to a wrong physical page.

>  - the calls to pmap_update() should be moved along with the calls to
>    pmap_kremove().

Due to deferred TLB invalidations (or flush), we want to call pmap_update()
at the point when all pmap_kremove() calls weere "collected".


Home | Main Index | Thread Index | Old Index