diff options
| author | Rick Farrington | 2018-08-28 20:32:55 +0200 |
|---|---|---|
| committer | David S. Miller | 2018-08-30 04:47:52 +0200 |
| commit | b943f17e06493fd2c7fd00743093ad5dcdb90e7f (patch) | |
| tree | c11e87c4f0720b3b86a99c01a19b24b58f162606 /drivers/net/ethernet/cavium/liquidio/request_manager.c | |
| parent | liquidio: remove unnecessary delay when processing IQ responses (diff) | |
| download | kernel-qcow2-linux-b943f17e06493fd2c7fd00743093ad5dcdb90e7f.tar.gz kernel-qcow2-linux-b943f17e06493fd2c7fd00743093ad5dcdb90e7f.tar.xz kernel-qcow2-linux-b943f17e06493fd2c7fd00743093ad5dcdb90e7f.zip | |
liquidio: fix race condition in instruction completion processing
In lio_enable_irq, the pkt_in_done count register was being cleared to
zero. However, there could be some completed instructions which were not
yet processed due to budget and limit constraints.
So, only write this register with the number of actual completions
that were processed.
Signed-off-by: Rick Farrington <ricardo.farrington@cavium.com>
Signed-off-by: Felix Manlunas <felix.manlunas@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'drivers/net/ethernet/cavium/liquidio/request_manager.c')
| -rw-r--r-- | drivers/net/ethernet/cavium/liquidio/request_manager.c | 2 |
1 files changed, 2 insertions, 0 deletions
diff --git a/drivers/net/ethernet/cavium/liquidio/request_manager.c b/drivers/net/ethernet/cavium/liquidio/request_manager.c index 0a06fbb4ff8a..5de5ce9a8f54 100644 --- a/drivers/net/ethernet/cavium/liquidio/request_manager.c +++ b/drivers/net/ethernet/cavium/liquidio/request_manager.c @@ -123,6 +123,7 @@ int octeon_init_instr_queue(struct octeon_device *oct, iq->do_auto_flush = 1; iq->db_timeout = (u32)conf->db_timeout; atomic_set(&iq->instr_pending, 0); + iq->pkts_processed = 0; /* Initialize the spinlock for this instruction queue */ spin_lock_init(&iq->lock); @@ -495,6 +496,7 @@ octeon_flush_iq(struct octeon_device *oct, struct octeon_instr_queue *iq, lio_process_iq_request_list(oct, iq, 0); if (inst_processed) { + iq->pkts_processed += inst_processed; atomic_sub(inst_processed, &iq->instr_pending); iq->stats.instr_processed += inst_processed; } |