This is a note to let you know that I've just added the patch titled net/sonic: Avoid needless receive descriptor EOL flag updates to the 5.4-stable tree which can be found at: http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary The filename of the patch is: net-sonic-avoid-needless-receive-descriptor-eol-flag-updates.patch and it can be found in the queue-5.4 subdirectory. If you, or anyone else, feels it should not be added to the stable tree, please let <stable@xxxxxxxxxxxxxxx> know about it. >From eaabfd19b2c787bbe88dc32424b9a43d67293422 Mon Sep 17 00:00:00 2001 From: Finn Thain <fthain@xxxxxxxxxxxxxxxxxxx> Date: Thu, 23 Jan 2020 09:07:26 +1100 Subject: net/sonic: Avoid needless receive descriptor EOL flag updates From: Finn Thain <fthain@xxxxxxxxxxxxxxxxxxx> commit eaabfd19b2c787bbe88dc32424b9a43d67293422 upstream. The while loop in sonic_rx() traverses the rx descriptor ring. It stops when it reaches a descriptor that the SONIC has not used. Each iteration advances the EOL flag so the SONIC can keep using more descriptors. Therefore, the while loop has no definite termination condition. The algorithm described in the National Semiconductor literature is quite different. It consumes descriptors up to the one with its EOL flag set (which will also have its "in use" flag set). All freed descriptors are then returned to the ring at once, by adjusting the EOL flags (and link pointers). Adopt the algorithm from datasheet as it's simpler, terminates quickly and avoids a lot of pointless descriptor EOL flag changes. Fixes: efcce839360f ("[PATCH] macsonic/jazzsonic network drivers update") Tested-by: Stan Johnson <userm57@xxxxxxxxx> Signed-off-by: Finn Thain <fthain@xxxxxxxxxxxxxxxxxxx> Signed-off-by: David S. Miller <davem@xxxxxxxxxxxxx> Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> --- drivers/net/ethernet/natsemi/sonic.c | 21 +++++++++++++++------ 1 file changed, 15 insertions(+), 6 deletions(-) --- a/drivers/net/ethernet/natsemi/sonic.c +++ b/drivers/net/ethernet/natsemi/sonic.c @@ -436,6 +436,7 @@ static void sonic_rx(struct net_device * struct sonic_local *lp = netdev_priv(dev); int status; int entry = lp->cur_rx; + int prev_entry = lp->eol_rx; while (sonic_rda_get(dev, entry, SONIC_RD_IN_USE) == 0) { struct sk_buff *used_skb; @@ -516,13 +517,21 @@ static void sonic_rx(struct net_device * /* * give back the descriptor */ - sonic_rda_put(dev, entry, SONIC_RD_LINK, - sonic_rda_get(dev, entry, SONIC_RD_LINK) | SONIC_EOL); sonic_rda_put(dev, entry, SONIC_RD_IN_USE, 1); - sonic_rda_put(dev, lp->eol_rx, SONIC_RD_LINK, - sonic_rda_get(dev, lp->eol_rx, SONIC_RD_LINK) & ~SONIC_EOL); - lp->eol_rx = entry; - lp->cur_rx = entry = (entry + 1) & SONIC_RDS_MASK; + + prev_entry = entry; + entry = (entry + 1) & SONIC_RDS_MASK; + } + + lp->cur_rx = entry; + + if (prev_entry != lp->eol_rx) { + /* Advance the EOL flag to put descriptors back into service */ + sonic_rda_put(dev, prev_entry, SONIC_RD_LINK, SONIC_EOL | + sonic_rda_get(dev, prev_entry, SONIC_RD_LINK)); + sonic_rda_put(dev, lp->eol_rx, SONIC_RD_LINK, ~SONIC_EOL & + sonic_rda_get(dev, lp->eol_rx, SONIC_RD_LINK)); + lp->eol_rx = prev_entry; } /* * If any worth-while packets have been received, netif_rx() Patches currently in stable-queue which might be from fthain@xxxxxxxxxxxxxxxxxxx are queue-5.4/net-sonic-fix-receive-buffer-handling.patch queue-5.4/net-sonic-use-mmio-accessors.patch queue-5.4/net-sonic-prevent-tx-watchdog-timeout.patch queue-5.4/net-sonic-clear-interrupt-flags-immediately.patch queue-5.4/net-sonic-fix-command-register-usage.patch queue-5.4/net-sonic-fix-cam-initialization.patch queue-5.4/net-sonic-fix-interface-error-stats-collection.patch queue-5.4/net-sonic-quiesce-sonic-before-re-initializing-descriptor-memory.patch queue-5.4/net-sonic-add-mutual-exclusion-for-accessing-shared-state.patch queue-5.4/net-sonic-fix-receive-buffer-replenishment.patch queue-5.4/net-sonic-avoid-needless-receive-descriptor-eol-flag-updates.patch queue-5.4/net-sonic-improve-receive-descriptor-status-flag-check.patch