Fixes for IPVS for v3.11
* Fix overflow in LBLC and LBLCR schedulers.
This resolves a regression introduced by commit b552f7e3a9524abcbcdf
("ipvs: unify the formula to estimate the overhead of processing
connections").
This problem was introduced in v2.6.38
* Correct stats calculation in the absence of CPU 0 in the possible mask
This resolves a regression introduced by commit 17fc9963f837ef1
("IPVS: netns, ip_vs_stats and its procfs").
This problem was introduced in v2.6.38
* Make the service replacement more robust
This resolves a regression introduced by commit 578bc3ef1e473a
("ipvs: reorganize dest trash")
This problem was introduced in 3.9
* Do not use dest after ip_vs_dest_put in LBLCR
This resolves a regression introduced by commit c5549571f975ab
("ipvs: convert lblcr scheduler to rcu")
This problem was introduced in 3.9
* Do not use dest after ip_vs_dest_put in LBLC
This resolves a regression introduced by commit 2a4ffb70eef39
("ipvs: convert lblc scheduler to rcu")
This problem was introduced in 3.9
ipvs: fix overflow on dest weight multiply
Schedulers such as lblc and lblcr require the weight to be as high as the
maximum number of active connections. In commit b552f7e3a9524abcbcdf
("ipvs: unify the formula to estimate the overhead of processing
connections"), the consideration of inactconns and activeconns was cleaned
up to always count activeconns as 256 times more important than inactconns.
In cases where 3000 or more connections are expected, a weight of 3000 *
256 * 3000 connections overflows the 32-bit signed result used to determine
if rescheduling is required.
On amd64, this merely changes the multiply and comparison instructions to
64-bit. On x86, a 64-bit result is already present from imull, so only
a few more comparison instructions are emitted.
Signed-off-by: Simon Kirby <sim@hostway.ca>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Simon Horman <horms@verge.net.au>
6 files changed
