Does anybody know if there are any patent applications for any of the
"existing" RCU-based reference counting techniques? In particular, any
counting algorithms that have fine granularity and do not use any atomic-ops
and/or membars... The existing RCU counting implementations that I have seen
use a per-object counter adjusted with CAS/membar, or actually defer count
adjustments until after sync epoch (rcu-grace)... Pretty expensive imho...
The reason I ask is because I have a patent application out on another
technique that gets around SMR's atomic-ops/membars and most of the caveats
that come along with RCU read-side "critical-regions"... I was just
wondering if anybody might be working on something similar... I remember
scouring through RCU/SMR-based ref-counting patents and applications during
the
patent-feasibility study I did a couple of years ago... I did find anything
that was similar to my technique...
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
[Index of Archives]
[Kernel Newbies]
[Netfilter]
[Bugtraq]
[Photo]
[Stuff]
[Gimp]
[Yosemite News]
[MIPS Linux]
[ARM Linux]
[Linux Security]
[Linux RAID]
[Video 4 Linux]
[Linux for the blind]
[Linux Resources]