| Message ID | 20220301094402.14992-6-dust.li@linux.alibaba.com (mailing list archive) |
|---|---|
| State | Not Applicable |
| Headers | show |
| Series | net/smc: some datapath performance optimizations | expand |
diff --git a/net/smc/smc_core.c b/net/smc/smc_core.c index 29525d03b253..1674b2549f8b 100644 --- a/net/smc/smc_core.c +++ b/net/smc/smc_core.c @@ -1988,7 +1988,7 @@ static struct smc_buf_desc *smc_buf_get_slot(int compressed_bufsize, */ static inline int smc_rmb_wnd_update_limit(int rmbe_size) { - return min_t(int, rmbe_size / 10, SOCK_MIN_SNDBUF / 2); + return max_t(int, rmbe_size / 10, SOCK_MIN_SNDBUF / 2); } /* map an rmb buf to a link */
rmbe_update_limit is used to limit announcing receive window updating too frequently. RFC7609 request a minimal increase in the window size of 10% of the receive buffer space. But current implementation used: min_t(int, rmbe_size / 10, SOCK_MIN_SNDBUF / 2) and SOCK_MIN_SNDBUF / 2 == 2304 Bytes, which is almost always less then 10% of the receive buffer space. This causes the receiver always sending CDC message to update its consumer cursor when it consumes more then 2K of data. And as a result, we may encounter something like "TCP silly window syndrome" when sending 2.5~8K message. This patch fixes this using max(rmbe_size / 10, SOCK_MIN_SNDBUF / 2). With this patch and SMC autocorking enabled, qperf 2K/4K/8K tcp_bw test shows 45%/75%/40% increase in throughput respectively. Signed-off-by: Dust Li <dust.li@linux.alibaba.com> --- net/smc/smc_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)