gro_cells: reduce number of synchronize_net() calls
Subsystems (2)
PATCH 1/1 gro_cells: reduce number of synchronize_net() calls
From: Eric Dumazet <[email protected]>
After cited commit, gro_cells_destroy() became damn slow on hosts with a lot of cores.
This is because we have one additional synchronize_net() per cpu as stated in the changelog.
gro_cells_init() is setting NAPI_STATE_NO_BUSY_POLL, and this was enough to not have one synchronize_net() call per netif_napi_del()
We can factorize all the synchronize_net() to a single one, right before freeing per-cpu memory.
--- net/core/gro_cells.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
net/core/gro_cells.c +6 -1
@@ -99,9 +99,14 @@ void gro_cells_destroy(struct gro_cells *gcells)struct gro_cell *cell =per_cpu_ptr (gcells->cells, i);napi_disable (&cell->napi); -netif_napi_del (&cell->napi); +__netif_napi_del (&cell->napi);__skb_queue_purge (&cell->napi_skbs); } +/* This barrier is needed because netpoll could access dev->napi_list +* under rcu protection. +*/ +synchronize_net (); +free_percpu (gcells->cells); gcells->cells =NULL ; }
Signature · 1 line
2.29.2.454.gaff20da3a2-goog
Quoted 8 lines · On Tue, 24 Nov 2020 12:38:22 -0800 Eric Dumazet wrote:
From: Eric Dumazet <[email protected]>
After cited commit,
gro_cells_destroy()became damn slow on hosts with a lot of cores.This is because we have one additional
synchronize_net()per cpu as stated in the changelog.
Quoted 8 lines
gro_cells_init()is settingNAPI_STATE_NO_BUSY_POLL, and this was enough to not have onesynchronize_net()call pernetif_napi_del()We can factorize all the
synchronize_net()to a single one, right before freeing per-cpu memory.Fixes 5198d545dba8 ("net: remove napi_hash_del() from driver-facing API")Signed-off-by Eric Dumazet <[email protected]>
Applied, thank you!