Each SCTP association can have up to 65535 input and output streams.
For each stream type an array of sctp_stream_in or sctp_stream_out
structures is allocated using kmalloc_array() function. This function
allocates physically contiguous memory regions, so this can lead
to allocation of memory regions of very high order, i.e.:
sizeof(struct sctp_stream_out) == 24,
((65535 * 24) / 4096) == 383 memory pages (4096 byte per page),
which means 9th memory order.
This can lead to a memory allocation failures on the systems
under a memory stress.
We actually do not need these arrays of memory to be physically
contiguous. Possible simple solution would be to use kvmalloc()
instread of kmalloc() as kvmalloc() can allocate physically scattered
pages if contiguous pages are not available. But the problem
is that the allocation can happed in a softirq context with
GFP_ATOMIC flag set, and kvmalloc() cannot be used in this scenario.
So the other possible solution is to use flexible arrays instead of
contiguios arrays of memory so that the memory would be allocated
on a per-page basis.
This patchset replaces kvmalloc() with flex_array usage.
It consists of two parts:
* First patch is preparatory - it mechanically wraps all direct
access to assoc->stream.out[] and assoc->stream.in[] arrays
with SCTP_SO() and SCTP_SI() wrappers so that later a direct
array access could be easily changed to an access to a
flex_array (or any other possible alternative).
* Second patch replaces kmalloc_array() with flex_array usage.
Oleg Babin (2):
net/sctp: Make wrappers for accessing in/out streams
net/sctp: Replace in/out stream arrays with flex_array
include/net/sctp/structs.h | 31 ++++----
net/sctp/chunk.c | 6 +-
net/sctp/outqueue.c | 11 +--
net/sctp/socket.c | 4 +-
net/sctp/stream.c | 165 +++++++++++++++++++++++++++++--------------
net/sctp/stream_interleave.c | 20 +++---
net/sctp/stream_sched.c | 13 ++--
net/sctp/stream_sched_prio.c | 22 +++---
net/sctp/stream_sched_rr.c | 8 +--
9 files changed, 175 insertions(+), 105 deletions(-)
v2 changes:
sctp_stream_in() users are updated to provide stream as an argument,
sctp_stream_{in,out}_ptr() are now just sctp_stream_{in,out}().
Performance results:
====================
* Kernel: v4.18-rc6 - stock and with 2 patches from Oleg (earlier in this
thread)
* Node: CPU (8 cores): Intel(R) Xeon(R) CPU E31230 @ 3.20GHz
RAM: 32 Gb
* netperf: taken from https://github.com/HewlettPackard/netperf.git,
compiled from sources with sctp support
* netperf server and client are run on the same node
* ip link set lo mtu 1500
The script used to run tests:
# cat run_tests.sh
#!/bin/bash
for test in SCTP_STREAM SCTP_STREAM_MANY SCTP_RR SCTP_RR_MANY; do
echo "TEST: $test";
for i in `seq 1 3`; do
echo "Iteration: $i";
set -x
netperf -t $test -H localhost -p 22222 -S 200000,200000 -s 200000,200000 \
-l 60 -- -m 1452;
set +x
done
done
================================================
Results (a bit reformatted to be more readable):
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. 10^6bits/sec
v4.18-rc7 v4.18-rc7 + fixes
TEST: SCTP_STREAM
212992 212992 1452 60.21 1125.52 1247.04
212992 212992 1452 60.20 1376.38 1149.95
212992 212992 1452 60.20 1131.40 1163.85
TEST: SCTP_STREAM_MANY
212992 212992 1452 60.00 1111.00 1310.05
212992 212992 1452 60.00 1188.55 1130.50
212992 212992 1452 60.00 1108.06 1162.50
===========
Local /Remote
Socket Size Request Resp. Elapsed Trans.
Send Recv Size Size Time Rate
bytes Bytes bytes bytes secs. per sec
v4.18-rc7 v4.18-rc7 + fixes
TEST: SCTP_RR
212992 212992 1 1 60.00 45486.98 46089.43
212992 212992 1 1 60.00 45584.18 45994.21
212992 212992 1 1 60.00 45703.86 45720.84
TEST: SCTP_RR_MANY
212992 212992 1 1 60.00 40.75 40.77
212992 212992 1 1 60.00 40.58 40.08
212992 212992 1 1 60.00 39.98 39.97
--
2.15.1
