firewire: optimize iso queueing by setting wake only after the last packet

When queueing iso packets, the run time is dominated by the two
MMIO accesses that set the DMA context's wake bit.  Because most
drivers submit packets in batches, we can save much time by
removing all but the last wakeup.

The internal kernel API is changed to require a call to
fw_iso_context_queue_flush() after a batch of queued packets.
The user space API does not change, so one call to
FW_CDEV_IOC_QUEUE_ISO must specify multiple packets to take
advantage of this optimization.

In my measurements, this patch reduces the time needed to queue
fifty skip packets from userspace to one sixth on a 2.5 GHz CPU,
or to one third at 800 MHz.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>

6 files changed, 32 insertions, 5 deletions

diff --git a/drivers/firewire/core-card.c b/drivers/firewire/core-card.c
index e119f1e6ba47..f05fc7bfceeb 100644
--- a/drivers/firewire/core-card.c
+++ b/drivers/firewire/core-card.c
@@ -630,6 +630,10 @@ static int dummy_queue_iso(struct fw_iso_context *ctx, struct fw_iso_packet *p,
         return -ENODEV;
 }
 
+static void dummy_flush_queue_iso(struct fw_iso_context *ctx)
+{
+}
+
 static const struct fw_card_driver dummy_driver_template = {
         .read_phy_reg           = dummy_read_phy_reg,
         .update_phy_reg         = dummy_update_phy_reg,
@@ -641,6 +645,7 @@ static const struct fw_card_driver dummy_driver_template = {
         .start_iso              = dummy_start_iso,
         .set_iso_channels       = dummy_set_iso_channels,
         .queue_iso              = dummy_queue_iso,
+        .flush_queue_iso        = dummy_flush_queue_iso,
 };
 
 void fw_card_release(struct kref *kref)
diff --git a/drivers/firewire/core-cdev.c b/drivers/firewire/core-cdev.c
index 2a3f1c4d6906..64768c2194f1 100644
--- a/drivers/firewire/core-cdev.c
+++ b/drivers/firewire/core-cdev.c
@@ -1107,6 +1107,7 @@ static int ioctl_queue_iso(struct client *client, union ioctl_arg *arg)
                 payload += u.packet.payload_length;
                 count++;
         }
+        fw_iso_context_queue_flush(ctx);
 
         a->size    -= uptr_to_u64(p) - a->packets;
         a->packets  = uptr_to_u64(p);
diff --git a/drivers/firewire/core-iso.c b/drivers/firewire/core-iso.c
index f872ede5af37..57c3973093ad 100644
--- a/drivers/firewire/core-iso.c
+++ b/drivers/firewire/core-iso.c
@@ -185,6 +185,12 @@ int fw_iso_context_queue(struct fw_iso_context *ctx,
 }
 EXPORT_SYMBOL(fw_iso_context_queue);
 
+void fw_iso_context_queue_flush(struct fw_iso_context *ctx)
+{
+        ctx->card->driver->flush_queue_iso(ctx);
+}
+EXPORT_SYMBOL(fw_iso_context_queue_flush);
+
 int fw_iso_context_stop(struct fw_iso_context *ctx)
 {
         return ctx->card->driver->stop_iso(ctx);
diff --git a/drivers/firewire/core.h b/drivers/firewire/core.h
index 25e729cde2f7..0fe4e4e6eda7 100644
--- a/drivers/firewire/core.h
+++ b/drivers/firewire/core.h
@@ -97,6 +97,8 @@ struct fw_card_driver {
                          struct fw_iso_buffer *buffer,
                          unsigned long payload);
 
+        void (*flush_queue_iso)(struct fw_iso_context *ctx);
+
         int (*stop_iso)(struct fw_iso_context *ctx);
 };
 
diff --git a/drivers/firewire/net.c b/drivers/firewire/net.c
index 3f04dd3681cf..b9762d07198d 100644
--- a/drivers/firewire/net.c
+++ b/drivers/firewire/net.c
@@ -881,7 +881,9 @@ static void fwnet_receive_broadcast(struct fw_iso_context *context,
 
         spin_unlock_irqrestore(&dev->lock, flags);
 
-        if (retval < 0)
+        if (retval >= 0)
+                fw_iso_context_queue_flush(dev->broadcast_rcv_context);
+        else
                 fw_error("requeue failed\n");
 }
 
diff --git a/drivers/firewire/ohci.c b/drivers/firewire/ohci.c
index f9f55703375e..438e6c831170 100644
--- a/drivers/firewire/ohci.c
+++ b/drivers/firewire/ohci.c
@@ -1192,9 +1192,6 @@ static void context_append(struct context *ctx,
         wmb(); /* finish init of new descriptors before branch_address update */
         ctx->prev->branch_address = cpu_to_le32(d_bus | z);
         ctx->prev = find_branch_descriptor(d, z);
-
-        reg_write(ctx->ohci, CONTROL_SET(ctx->regs), CONTEXT_WAKE);
-        flush_writes(ctx->ohci);
 }
 
 static void context_stop(struct context *ctx)
@@ -1348,8 +1345,12 @@ static int at_context_queue_packet(struct context *ctx,
 
         context_append(ctx, d, z, 4 - z);
 
-        if (!ctx->running)
+        if (ctx->running) {
+                reg_write(ohci, CONTROL_SET(ctx->regs), CONTEXT_WAKE);
+                flush_writes(ohci);
+        } else {
                 context_run(ctx, 0);
+        }
 
         return 0;
 }
@@ -3121,6 +3122,15 @@ static int ohci_queue_iso(struct fw_iso_context *base,
         return ret;
 }
 
+static void ohci_flush_queue_iso(struct fw_iso_context *base)
+{
+        struct context *ctx =
+                        &container_of(base, struct iso_context, base)->context;
+
+        reg_write(ctx->ohci, CONTROL_SET(ctx->regs), CONTEXT_WAKE);
+        flush_writes(ctx->ohci);
+}
+
 static const struct fw_card_driver ohci_driver = {
         .enable                 = ohci_enable,
         .read_phy_reg           = ohci_read_phy_reg,
@@ -3137,6 +3147,7 @@ static const struct fw_card_driver ohci_driver = {
         .free_iso_context       = ohci_free_iso_context,
         .set_iso_channels       = ohci_set_iso_channels,
         .queue_iso              = ohci_queue_iso,
+        .flush_queue_iso        = ohci_flush_queue_iso,
         .start_iso              = ohci_start_iso,
         .stop_iso               = ohci_stop_iso,
 };