diff options
| author | K. Y. Srinivasan <kys@microsoft.com> | 2014-04-08 18:45:53 -0700 |
|---|---|---|
| committer | Greg Kroah-Hartman <gregkh@linuxfoundation.org> | 2014-05-03 19:24:26 -0400 |
| commit | d3ba720dd58cdf6630fee4b89482c465d5ad0d0f (patch) | |
| tree | 03ea0aba6f0509bf7c501dee9ae6e931634f3c4b /drivers/hv/connection.c | |
| parent | e4d8270e604c3202131bac607969605ac397b893 (diff) | |
| download | op-kernel-dev-d3ba720dd58cdf6630fee4b89482c465d5ad0d0f.zip op-kernel-dev-d3ba720dd58cdf6630fee4b89482c465d5ad0d0f.tar.gz | |
Drivers: hv: Eliminate the channel spinlock in the callback path
By ensuring that we set the callback handler to NULL in the channel close
path on the same CPU that the channel is bound to, we can eliminate this lock
acquisition and release in a performance critical path.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Diffstat (limited to 'drivers/hv/connection.c')
| -rw-r--r-- | drivers/hv/connection.c | 11 |
1 files changed, 4 insertions, 7 deletions
diff --git a/drivers/hv/connection.c b/drivers/hv/connection.c index 2e7801a..df2363e 100644 --- a/drivers/hv/connection.c +++ b/drivers/hv/connection.c @@ -277,7 +277,6 @@ struct vmbus_channel *relid2channel(u32 relid) static void process_chn_event(u32 relid) { struct vmbus_channel *channel; - unsigned long flags; void *arg; bool read_state; u32 bytes_to_read; @@ -296,13 +295,12 @@ static void process_chn_event(u32 relid) /* * A channel once created is persistent even when there * is no driver handling the device. An unloading driver - * sets the onchannel_callback to NULL under the - * protection of the channel inbound_lock. Thus, checking - * and invoking the driver specific callback takes care of - * orderly unloading of the driver. + * sets the onchannel_callback to NULL on the same CPU + * as where this interrupt is handled (in an interrupt context). + * Thus, checking and invoking the driver specific callback takes + * care of orderly unloading of the driver. */ - spin_lock_irqsave(&channel->inbound_lock, flags); if (channel->onchannel_callback != NULL) { arg = channel->channel_callback_context; read_state = channel->batched_reading; @@ -327,7 +325,6 @@ static void process_chn_event(u32 relid) pr_err("no channel callback for relid - %u\n", relid); } - spin_unlock_irqrestore(&channel->inbound_lock, flags); } /* |
