From ecf8328e59447b83a1f79628487e0e9f8801db84 Mon Sep 17 00:00:00 2001
From: Clemens Ladisch <clemens@ladisch.de>
Date: Mon, 11 Apr 2011 09:56:12 +0200
Subject: [PATCH 1/9] firewire: ohci: do not start DMA contexts before link is
 enabled

OHCI 1.1 5.7.3 not only forbids enabling or starting any DMA contexts
before the linkEnable bit is set, but also explicitly warns of undefined
behaviour if this order is violated.

Don't violate it then.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
---
 drivers/firewire/ohci.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/drivers/firewire/ohci.c b/drivers/firewire/ohci.c
index f903d7b6f34a..dfb3cb774a79 100644
--- a/drivers/firewire/ohci.c
+++ b/drivers/firewire/ohci.c
@@ -2066,8 +2066,6 @@ static int ohci_enable(struct fw_card *card,
 
 	reg_write(ohci, OHCI1394_SelfIDBuffer, ohci->self_id_bus);
 	reg_write(ohci, OHCI1394_LinkControlSet,
-		  OHCI1394_LinkControl_rcvSelfID |
-		  OHCI1394_LinkControl_rcvPhyPkt |
 		  OHCI1394_LinkControl_cycleTimerEnable |
 		  OHCI1394_LinkControl_cycleMaster);
 
@@ -2094,9 +2092,6 @@ static int ohci_enable(struct fw_card *card,
 	reg_write(ohci, OHCI1394_FairnessControl, 0);
 	card->priority_budget_implemented = ohci->pri_req_max != 0;
 
-	ar_context_run(&ohci->ar_request_ctx);
-	ar_context_run(&ohci->ar_response_ctx);
-
 	reg_write(ohci, OHCI1394_PhyUpperBound, 0x00010000);
 	reg_write(ohci, OHCI1394_IntEventClear, ~0);
 	reg_write(ohci, OHCI1394_IntMaskClear, ~0);
@@ -2186,7 +2181,13 @@ static int ohci_enable(struct fw_card *card,
 	reg_write(ohci, OHCI1394_HCControlSet,
 		  OHCI1394_HCControl_linkEnable |
 		  OHCI1394_HCControl_BIBimageValid);
-	flush_writes(ohci);
+
+	reg_write(ohci, OHCI1394_LinkControlSet,
+		  OHCI1394_LinkControl_rcvSelfID |
+		  OHCI1394_LinkControl_rcvPhyPkt);
+
+	ar_context_run(&ohci->ar_request_ctx);
+	ar_context_run(&ohci->ar_response_ctx); /* also flushes writes */
 
 	/* We are ready to go, reset bus to finish initialization. */
 	fw_schedule_bus_reset(&ohci->card, false, true);

From da28947e7e3602669e27d5e9ce787436ed662fa4 Mon Sep 17 00:00:00 2001
From: Clemens Ladisch <clemens@ladisch.de>
Date: Mon, 11 Apr 2011 09:57:54 +0200
Subject: [PATCH 2/9] firewire: ohci: avoid separate DMA mapping for small AT
 payloads

For AT packet payloads of up to eight bytes, we have enough unused space
in the DMA descriptors list so that we can put a copy of the payload
there and thus avoid having to create a separate streaming DMA mapping
for the payload buffer.

In a CPU-bound microbenchmark that just sends 8-byte packets, bandwidth
was measured to increase by 5.7 %, from 1009 KB/s to 1067 KB/s.  In
practice, the only performance-sensitive usage of small asynchronous
packets is the SBP-2 driver's write to the ORB_POINTER register during
SCSI command submission.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
---
 drivers/firewire/ohci.c | 25 +++++++++++++++++--------
 1 file changed, 17 insertions(+), 8 deletions(-)

diff --git a/drivers/firewire/ohci.c b/drivers/firewire/ohci.c
index dfb3cb774a79..435ed800d03f 100644
--- a/drivers/firewire/ohci.c
+++ b/drivers/firewire/ohci.c
@@ -1218,6 +1218,7 @@ static void context_stop(struct context *ctx)
 }
 
 struct driver_data {
+	u8 inline_data[8];
 	struct fw_packet *packet;
 };
 
@@ -1301,20 +1302,28 @@ static int at_context_queue_packet(struct context *ctx,
 		return -1;
 	}
 
+	BUILD_BUG_ON(sizeof(struct driver_data) > sizeof(struct descriptor));
 	driver_data = (struct driver_data *) &d[3];
 	driver_data->packet = packet;
 	packet->driver_data = driver_data;
 
 	if (packet->payload_length > 0) {
-		payload_bus =
-			dma_map_single(ohci->card.device, packet->payload,
-				       packet->payload_length, DMA_TO_DEVICE);
-		if (dma_mapping_error(ohci->card.device, payload_bus)) {
-			packet->ack = RCODE_SEND_ERROR;
-			return -1;
+		if (packet->payload_length > sizeof(driver_data->inline_data)) {
+			payload_bus = dma_map_single(ohci->card.device,
+						     packet->payload,
+						     packet->payload_length,
+						     DMA_TO_DEVICE);
+			if (dma_mapping_error(ohci->card.device, payload_bus)) {
+				packet->ack = RCODE_SEND_ERROR;
+				return -1;
+			}
+			packet->payload_bus	= payload_bus;
+			packet->payload_mapped	= true;
+		} else {
+			memcpy(driver_data->inline_data, packet->payload,
+			       packet->payload_length);
+			payload_bus = d_bus + 3 * sizeof(*d);
 		}
-		packet->payload_bus	= payload_bus;
-		packet->payload_mapped	= true;
 
 		d[2].req_count    = cpu_to_le16(packet->payload_length);
 		d[2].data_address = cpu_to_le32(payload_bus);

From 0ff8fbc61727c926883eec381fbd3d32d1fab504 Mon Sep 17 00:00:00 2001
From: Clemens Ladisch <clemens@ladisch.de>
Date: Tue, 12 Apr 2011 07:54:59 +0200
Subject: [PATCH 3/9] firewire: ohci: optimize find_branch_descriptor()

When z==2, the condition "key == 2" is superfluous because it cannot
occur without "b == 3", as a descriptor with b!=3 and key==2 would be
an OUTPUT_MORE_IMMEDIATE descriptor which cannot be used alone.

Also remove magic numbers and needless computations on the b field.

Signed-off-by: Clemens Ladisch <clemens@ladisch.de>
Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
---
 drivers/firewire/ohci.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/drivers/firewire/ohci.c b/drivers/firewire/ohci.c
index 435ed800d03f..ce767a4882fb 100644
--- a/drivers/firewire/ohci.c
+++ b/drivers/firewire/ohci.c
@@ -1006,13 +1006,12 @@ static void ar_context_run(struct ar_context *ctx)
 
 static struct descriptor *find_branch_descriptor(struct descriptor *d, int z)
 {
-	int b, key;
+	__le16 branch;
 
-	b   = (le16_to_cpu(d->control) & DESCRIPTOR_BRANCH_ALWAYS) >> 2;
-	key = (le16_to_cpu(d->control) & DESCRIPTOR_KEY_IMMEDIATE) >> 8;
+	branch = d->control & cpu_to_le16(DESCRIPTOR_BRANCH_ALWAYS);
 
 	/* figure out which descriptor the branch address goes in */
-	if (z == 2 && (b == 3 || key == 2))
+	if (z == 2 && branch == cpu_to_le16(DESCRIPTOR_BRANCH_ALWAYS))
 		return d;
 	else
 		return d + z - 1;

From f30e6d3e419bfb5540fa82ba7eca01d578556e6b Mon Sep 17 00:00:00 2001
From: Stefan Richter <stefanr@s5r6.in-berlin.de>
Date: Fri, 22 Apr 2011 15:13:54 +0200
Subject: [PATCH 4/9] firewire: octlet AT payloads can be stack-allocated

We do not need slab allocations anymore in order to satisfy
streaming DMA mapping constraints, thanks to commit da28947e7e36
"firewire: ohci: avoid separate DMA mapping for small AT payloads".

(Besides, the slab-allocated buffers that firewire-core, firewire-sbp2,
and firedtv used to provide for 8-byte write and lock requests were
still not fully portable since they crossed cacheline boundaries or
shared a cacheline with unrelated CPU-accessed data.  snd-firewire-lib
got this aspect right by using an extra kmalloc/ kfree just for the
8-byte transaction buffer.)

This change replaces kmalloc'ed lock transaction scratch buffers in
firewire-core, firedtv, and snd-firewire-lib by local stack allocations.
Perhaps the most notable result of the change is simpler locking because
there is no need to serialize usages of preallocated per-device buffers
anymore.  Also, allocations and deallocations are simpler.

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Acked-by: Clemens Ladisch <clemens@ladisch.de>
---
 drivers/firewire/core-card.c             | 16 ++++++++--------
 drivers/firewire/core-cdev.c             |  4 +---
 drivers/firewire/core-iso.c              | 21 +++++++++++----------
 drivers/firewire/core-transaction.c      |  7 ++++---
 drivers/media/dvb/firewire/firedtv-avc.c | 15 +--------------
 include/linux/firewire.h                 |  3 +--
 sound/firewire/cmp.c                     |  3 +--
 sound/firewire/iso-resources.c           | 12 +++---------
 sound/firewire/iso-resources.h           |  1 -
 9 files changed, 30 insertions(+), 52 deletions(-)

diff --git a/drivers/firewire/core-card.c b/drivers/firewire/core-card.c
index 3c44fbc81acb..e119f1e6ba47 100644
--- a/drivers/firewire/core-card.c
+++ b/drivers/firewire/core-card.c
@@ -258,8 +258,7 @@ static void allocate_broadcast_channel(struct fw_card *card, int generation)
 
 	if (!card->broadcast_channel_allocated) {
 		fw_iso_resource_manage(card, generation, 1ULL << 31,
-				       &channel, &bandwidth, true,
-				       card->bm_transaction_data);
+				       &channel, &bandwidth, true);
 		if (channel != 31) {
 			fw_notify("failed to allocate broadcast channel\n");
 			return;
@@ -294,6 +293,7 @@ static void bm_work(struct work_struct *work)
 	bool root_device_is_cmc;
 	bool irm_is_1394_1995_only;
 	bool keep_this_irm;
+	__be32 transaction_data[2];
 
 	spin_lock_irq(&card->lock);
 
@@ -355,21 +355,21 @@ static void bm_work(struct work_struct *work)
 			goto pick_me;
 		}
 
-		card->bm_transaction_data[0] = cpu_to_be32(0x3f);
-		card->bm_transaction_data[1] = cpu_to_be32(local_id);
+		transaction_data[0] = cpu_to_be32(0x3f);
+		transaction_data[1] = cpu_to_be32(local_id);
 
 		spin_unlock_irq(&card->lock);
 
 		rcode = fw_run_transaction(card, TCODE_LOCK_COMPARE_SWAP,
 				irm_id, generation, SCODE_100,
 				CSR_REGISTER_BASE + CSR_BUS_MANAGER_ID,
-				card->bm_transaction_data, 8);
+				transaction_data, 8);
 
 		if (rcode == RCODE_GENERATION)
 			/* Another bus reset, BM work has been rescheduled. */
 			goto out;
 
-		bm_id = be32_to_cpu(card->bm_transaction_data[0]);
+		bm_id = be32_to_cpu(transaction_data[0]);
 
 		spin_lock_irq(&card->lock);
 		if (rcode == RCODE_COMPLETE && generation == card->generation)
@@ -490,11 +490,11 @@ static void bm_work(struct work_struct *work)
 		/*
 		 * Make sure that the cycle master sends cycle start packets.
 		 */
-		card->bm_transaction_data[0] = cpu_to_be32(CSR_STATE_BIT_CMSTR);
+		transaction_data[0] = cpu_to_be32(CSR_STATE_BIT_CMSTR);
 		rcode = fw_run_transaction(card, TCODE_WRITE_QUADLET_REQUEST,
 				root_id, generation, SCODE_100,
 				CSR_REGISTER_BASE + CSR_STATE_SET,
-				card->bm_transaction_data, 4);
+				transaction_data, 4);
 		if (rcode == RCODE_GENERATION)
 			goto out;
 	}
diff --git a/drivers/firewire/core-cdev.c b/drivers/firewire/core-cdev.c
index 62ac111af243..2a3f1c4d6906 100644
--- a/drivers/firewire/core-cdev.c
+++ b/drivers/firewire/core-cdev.c
@@ -141,7 +141,6 @@ struct iso_resource {
 	int generation;
 	u64 channels;
 	s32 bandwidth;
-	__be32 transaction_data[2];
 	struct iso_resource_event *e_alloc, *e_dealloc;
 };
 
@@ -1229,8 +1228,7 @@ static void iso_resource_work(struct work_struct *work)
 			r->channels, &channel, &bandwidth,
 			todo == ISO_RES_ALLOC ||
 			todo == ISO_RES_REALLOC ||
-			todo == ISO_RES_ALLOC_ONCE,
-			r->transaction_data);
+			todo == ISO_RES_ALLOC_ONCE);
 	/*
 	 * Is this generation outdated already?  As long as this resource sticks
 	 * in the idr, it will be scheduled again for a newer generation or at
diff --git a/drivers/firewire/core-iso.c b/drivers/firewire/core-iso.c
index 481056df9268..f872ede5af37 100644
--- a/drivers/firewire/core-iso.c
+++ b/drivers/firewire/core-iso.c
@@ -196,9 +196,10 @@ EXPORT_SYMBOL(fw_iso_context_stop);
  */
 
 static int manage_bandwidth(struct fw_card *card, int irm_id, int generation,
-			    int bandwidth, bool allocate, __be32 data[2])
+			    int bandwidth, bool allocate)
 {
 	int try, new, old = allocate ? BANDWIDTH_AVAILABLE_INITIAL : 0;
+	__be32 data[2];
 
 	/*
 	 * On a 1394a IRM with low contention, try < 1 is enough.
@@ -233,9 +234,10 @@ static int manage_bandwidth(struct fw_card *card, int irm_id, int generation,
 }
 
 static int manage_channel(struct fw_card *card, int irm_id, int generation,
-		u32 channels_mask, u64 offset, bool allocate, __be32 data[2])
+		u32 channels_mask, u64 offset, bool allocate)
 {
 	__be32 bit, all, old;
+	__be32 data[2];
 	int channel, ret = -EIO, retry = 5;
 
 	old = all = allocate ? cpu_to_be32(~0) : 0;
@@ -284,7 +286,7 @@ static int manage_channel(struct fw_card *card, int irm_id, int generation,
 }
 
 static void deallocate_channel(struct fw_card *card, int irm_id,
-			       int generation, int channel, __be32 buffer[2])
+			       int generation, int channel)
 {
 	u32 mask;
 	u64 offset;
@@ -293,7 +295,7 @@ static void deallocate_channel(struct fw_card *card, int irm_id,
 	offset = channel < 32 ? CSR_REGISTER_BASE + CSR_CHANNELS_AVAILABLE_HI :
 				CSR_REGISTER_BASE + CSR_CHANNELS_AVAILABLE_LO;
 
-	manage_channel(card, irm_id, generation, mask, offset, false, buffer);
+	manage_channel(card, irm_id, generation, mask, offset, false);
 }
 
 /**
@@ -322,7 +324,7 @@ static void deallocate_channel(struct fw_card *card, int irm_id,
  */
 void fw_iso_resource_manage(struct fw_card *card, int generation,
 			    u64 channels_mask, int *channel, int *bandwidth,
-			    bool allocate, __be32 buffer[2])
+			    bool allocate)
 {
 	u32 channels_hi = channels_mask;	/* channels 31...0 */
 	u32 channels_lo = channels_mask >> 32;	/* channels 63...32 */
@@ -335,11 +337,11 @@ void fw_iso_resource_manage(struct fw_card *card, int generation,
 	if (channels_hi)
 		c = manage_channel(card, irm_id, generation, channels_hi,
 				CSR_REGISTER_BASE + CSR_CHANNELS_AVAILABLE_HI,
-				allocate, buffer);
+				allocate);
 	if (channels_lo && c < 0) {
 		c = manage_channel(card, irm_id, generation, channels_lo,
 				CSR_REGISTER_BASE + CSR_CHANNELS_AVAILABLE_LO,
-				allocate, buffer);
+				allocate);
 		if (c >= 0)
 			c += 32;
 	}
@@ -351,14 +353,13 @@ void fw_iso_resource_manage(struct fw_card *card, int generation,
 	if (*bandwidth == 0)
 		return;
 
-	ret = manage_bandwidth(card, irm_id, generation, *bandwidth,
-			       allocate, buffer);
+	ret = manage_bandwidth(card, irm_id, generation, *bandwidth, allocate);
 	if (ret < 0)
 		*bandwidth = 0;
 
 	if (allocate && ret < 0) {
 		if (c >= 0)
-			deallocate_channel(card, irm_id, generation, c, buffer);
+			deallocate_channel(card, irm_id, generation, c);
 		*channel = ret;
 	}
 }
diff --git a/drivers/firewire/core-transaction.c b/drivers/firewire/core-transaction.c
index d00f8ce902cc..77275fdf6c1f 100644
--- a/drivers/firewire/core-transaction.c
+++ b/drivers/firewire/core-transaction.c
@@ -326,8 +326,8 @@ static int allocate_tlabel(struct fw_card *card)
  * It will contain tag, channel, and sy data instead of a node ID then.
  *
  * The payload buffer at @data is going to be DMA-mapped except in case of
- * quadlet-sized payload or of local (loopback) requests.  Hence make sure that
- * the buffer complies with the restrictions for DMA-mapped memory.  The
+ * @length <= 8 or of local (loopback) requests.  Hence make sure that the
+ * buffer complies with the restrictions of the streaming DMA mapping API.
  * @payload must not be freed before the @callback is called.
  *
  * In case of request types without payload, @data is NULL and @length is 0.
@@ -411,7 +411,8 @@ static void transaction_callback(struct fw_card *card, int rcode,
  *
  * Returns the RCODE.  See fw_send_request() for parameter documentation.
  * Unlike fw_send_request(), @data points to the payload of the request or/and
- * to the payload of the response.
+ * to the payload of the response.  DMA mapping restrictions apply to outbound
+ * request payloads of >= 8 bytes but not to inbound response payloads.
  */
 int fw_run_transaction(struct fw_card *card, int tcode, int destination_id,
 		       int generation, int speed, unsigned long long offset,
diff --git a/drivers/media/dvb/firewire/firedtv-avc.c b/drivers/media/dvb/firewire/firedtv-avc.c
index fc5ccd8c923a..21c52e3b522e 100644
--- a/drivers/media/dvb/firewire/firedtv-avc.c
+++ b/drivers/media/dvb/firewire/firedtv-avc.c
@@ -1320,14 +1320,10 @@ static int cmp_read(struct firedtv *fdtv, u64 addr, __be32 *data)
 {
 	int ret;
 
-	mutex_lock(&fdtv->avc_mutex);
-
 	ret = fdtv_read(fdtv, addr, data);
 	if (ret < 0)
 		dev_err(fdtv->device, "CMP: read I/O error\n");
 
-	mutex_unlock(&fdtv->avc_mutex);
-
 	return ret;
 }
 
@@ -1335,18 +1331,9 @@ static int cmp_lock(struct firedtv *fdtv, u64 addr, __be32 data[])
 {
 	int ret;
 
-	mutex_lock(&fdtv->avc_mutex);
-
-	/* data[] is stack-allocated and should not be DMA-mapped. */
-	memcpy(fdtv->avc_data, data, 8);
-
-	ret = fdtv_lock(fdtv, addr, fdtv->avc_data);
+	ret = fdtv_lock(fdtv, addr, data);
 	if (ret < 0)
 		dev_err(fdtv->device, "CMP: lock I/O error\n");
-	else
-		memcpy(data, fdtv->avc_data, 8);
-
-	mutex_unlock(&fdtv->avc_mutex);
 
 	return ret;
 }
diff --git a/include/linux/firewire.h b/include/linux/firewire.h
index c64f3680d4f1..de90e1ff8488 100644
--- a/include/linux/firewire.h
+++ b/include/linux/firewire.h
@@ -125,7 +125,6 @@ struct fw_card {
 	struct delayed_work bm_work; /* bus manager job */
 	int bm_retries;
 	int bm_generation;
-	__be32 bm_transaction_data[2];
 	int bm_node_id;
 	bool bm_abdicate;
 
@@ -447,6 +446,6 @@ int fw_iso_context_stop(struct fw_iso_context *ctx);
 void fw_iso_context_destroy(struct fw_iso_context *ctx);
 void fw_iso_resource_manage(struct fw_card *card, int generation,
 			    u64 channels_mask, int *channel, int *bandwidth,
-			    bool allocate, __be32 buffer[2]);
+			    bool allocate);
 
 #endif /* _LINUX_FIREWIRE_H */
diff --git a/sound/firewire/cmp.c b/sound/firewire/cmp.c
index 4a37f3a6fab9..14cacbc655dd 100644
--- a/sound/firewire/cmp.c
+++ b/sound/firewire/cmp.c
@@ -49,10 +49,9 @@ static int pcr_modify(struct cmp_connection *c,
 		      enum bus_reset_handling bus_reset_handling)
 {
 	struct fw_device *device = fw_parent_device(c->resources.unit);
-	__be32 *buffer = c->resources.buffer;
 	int generation = c->resources.generation;
 	int rcode, errors = 0;
-	__be32 old_arg;
+	__be32 old_arg, buffer[2];
 	int err;
 
 	buffer[0] = c->last_pcr_value;
diff --git a/sound/firewire/iso-resources.c b/sound/firewire/iso-resources.c
index 775dbd5f3445..bb9c0c1fb529 100644
--- a/sound/firewire/iso-resources.c
+++ b/sound/firewire/iso-resources.c
@@ -11,7 +11,6 @@
 #include <linux/jiffies.h>
 #include <linux/mutex.h>
 #include <linux/sched.h>
-#include <linux/slab.h>
 #include <linux/spinlock.h>
 #include "iso-resources.h"
 
@@ -25,10 +24,6 @@
  */
 int fw_iso_resources_init(struct fw_iso_resources *r, struct fw_unit *unit)
 {
-	r->buffer = kmalloc(2 * 4, GFP_KERNEL);
-	if (!r->buffer)
-		return -ENOMEM;
-
 	r->channels_mask = ~0uLL;
 	r->unit = fw_unit_get(unit);
 	mutex_init(&r->mutex);
@@ -44,7 +39,6 @@ int fw_iso_resources_init(struct fw_iso_resources *r, struct fw_unit *unit)
 void fw_iso_resources_destroy(struct fw_iso_resources *r)
 {
 	WARN_ON(r->allocated);
-	kfree(r->buffer);
 	mutex_destroy(&r->mutex);
 	fw_unit_put(r->unit);
 }
@@ -131,7 +125,7 @@ int fw_iso_resources_allocate(struct fw_iso_resources *r,
 
 	bandwidth = r->bandwidth + r->bandwidth_overhead;
 	fw_iso_resource_manage(card, r->generation, r->channels_mask,
-			       &channel, &bandwidth, true, r->buffer);
+			       &channel, &bandwidth, true);
 	if (channel == -EAGAIN) {
 		mutex_unlock(&r->mutex);
 		goto retry_after_bus_reset;
@@ -184,7 +178,7 @@ int fw_iso_resources_update(struct fw_iso_resources *r)
 	bandwidth = r->bandwidth + r->bandwidth_overhead;
 
 	fw_iso_resource_manage(card, r->generation, 1uLL << r->channel,
-			       &channel, &bandwidth, true, r->buffer);
+			       &channel, &bandwidth, true);
 	/*
 	 * When another bus reset happens, pretend that the allocation
 	 * succeeded; we will try again for the new generation later.
@@ -220,7 +214,7 @@ void fw_iso_resources_free(struct fw_iso_resources *r)
 	if (r->allocated) {
 		bandwidth = r->bandwidth + r->bandwidth_overhead;
 		fw_iso_resource_manage(card, r->generation, 1uLL << r->channel,
-				       &channel, &bandwidth, false, r->buffer);
+				       &channel, &bandwidth, false);
 		if (channel < 0)
 			dev_err(&r->unit->device,
 				"isochronous resource deallocation failed\n");
diff --git a/sound/firewire/iso-resources.h b/sound/firewire/iso-resources.h
index 3f0730e4d841..5a9af7c61657 100644
--- a/sound/firewire/iso-resources.h
+++ b/sound/firewire/iso-resources.h
@@ -24,7 +24,6 @@ struct fw_iso_resources {
 	unsigned int bandwidth_overhead;
 	int generation; /* in which allocation is valid */
 	bool allocated;
-	__be32 *buffer;
 };
 
 int fw_iso_resources_init(struct fw_iso_resources *r,

From 13882a82ee1646336c3996c93b4a560a55d2a419 Mon Sep 17 00:00:00 2001
From: Clemens Ladisch <clemens@ladisch.de>
Date: Mon, 2 May 2011 09:33:56 +0200
Subject: [PATCH 5/9] 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>
---
 drivers/firewire/core-card.c            |  5 +++++
 drivers/firewire/core-cdev.c            |  1 +
 drivers/firewire/core-iso.c             |  6 ++++++
 drivers/firewire/core.h                 |  2 ++
 drivers/firewire/net.c                  |  4 +++-
 drivers/firewire/ohci.c                 | 19 +++++++++++++++----
 drivers/media/dvb/firewire/firedtv-fw.c |  1 +
 include/linux/firewire.h                |  1 +
 sound/firewire/amdtp.c                  |  1 +
 9 files changed, 35 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,
 };
diff --git a/drivers/media/dvb/firewire/firedtv-fw.c b/drivers/media/dvb/firewire/firedtv-fw.c
index 8022b743af91..864b6274c729 100644
--- a/drivers/media/dvb/firewire/firedtv-fw.c
+++ b/drivers/media/dvb/firewire/firedtv-fw.c
@@ -125,6 +125,7 @@ static void handle_iso(struct fw_iso_context *context, u32 cycle,
 
 		i = (i + 1) & (N_PACKETS - 1);
 	}
+	fw_iso_context_queue_flush(ctx->context);
 	ctx->current_packet = i;
 }
 
diff --git a/include/linux/firewire.h b/include/linux/firewire.h
index de90e1ff8488..c0fb405bb435 100644
--- a/include/linux/firewire.h
+++ b/include/linux/firewire.h
@@ -440,6 +440,7 @@ int fw_iso_context_queue(struct fw_iso_context *ctx,
 			 struct fw_iso_packet *packet,
 			 struct fw_iso_buffer *buffer,
 			 unsigned long payload);
+void fw_iso_context_queue_flush(struct fw_iso_context *ctx);
 int fw_iso_context_start(struct fw_iso_context *ctx,
 			 int cycle, int sync, int tags);
 int fw_iso_context_stop(struct fw_iso_context *ctx);
diff --git a/sound/firewire/amdtp.c b/sound/firewire/amdtp.c
index b18140ff2b93..87657dd7714c 100644
--- a/sound/firewire/amdtp.c
+++ b/sound/firewire/amdtp.c
@@ -396,6 +396,7 @@ static void out_packet_callback(struct fw_iso_context *context, u32 cycle,
 
 	for (i = 0; i < packets; ++i)
 		queue_out_packet(s, ++cycle);
+	fw_iso_context_queue_flush(s->context);
 }
 
 static int queue_initial_skip_packets(struct amdtp_out_stream *s)

From 6ea9e7bbfc389a12d52646449a201fe933ccd663 Mon Sep 17 00:00:00 2001
From: Stefan Richter <stefanr@s5r6.in-berlin.de>
Date: Wed, 13 Oct 2010 13:39:46 +0200
Subject: [PATCH 6/9] firewire: core: use non-reentrant workqueue with rescuer

firewire-core manages the following types of work items:

fw_card.br_work:
  - resets the bus on a card and possibly sends a PHY packet before that
  - does not sleep for long or not at all
  - is scheduled via fw_schedule_bus_reset() by
      - firewire-ohci's pci_probe method
      - firewire-ohci's set_config_rom method, called by kernelspace
        protocol drivers and userspace drivers which add/remove
	Configuration ROM descriptors
      - userspace drivers which use the bus reset ioctl
      - itself if the last reset happened less than 2 seconds ago

fw_card.bm_work:
  - performs bus management duties
  - usually does not (but may in corner cases) sleep for long
  - is scheduled via fw_schedule_bm_work() by
      - firewire-ohci's self-ID-complete IRQ handler tasklet
      - firewire-core's fw_device.work instances whenever the root node
        device was (successfully or unsuccessfully) discovered,
	refreshed, or rediscovered
      - itself in case of resource allocation failures or in order to
        obey the 125ms bus manager arbitration interval

fw_device.work:
  - performs node probe, update, shutdown, revival, removal; including
    kernel driver probe, update, shutdown and bus reset notification to
    userspace drivers
  - usually sleeps moderately long, in corner cases very long
  - is scheduled by
      - firewire-ohci's self-ID-complete IRQ handler tasklet via the
        core's fw_node_event
      - firewire-ohci's pci_remove method via core's fw_destroy_nodes/
        fw_node_event
      - itself during retries, e.g. while a node is powering up

iso_resource.work:
  - accesses registers at the Isochronous Resource Manager node
  - usually does not (but may in corner cases) sleep for long
  - is scheduled via schedule_iso_resource() by
      - the owning userspace driver at addition and removal of the
        resource
      - firewire-core's fw_device.work instances after bus reset
      - itself in case of resource allocation if necessary to obey the
        1000ms reallocation period after bus reset

fw_card.br_work instances should not, and instances of the others must
not, be executed in parallel by multiple CPUs -- but were not protected
against that.  Hence allocate a non-reentrant workqueue for them.

fw_device.work may be used in the memory reclaim path in case of SBP-2
device updates.  Hence we need a workqueue with rescuer and cannot use
system_nrt_wq.

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
Reviewed-by: Tejun Heo <tj@kernel.org>
---
 drivers/firewire/core-card.c        |  6 +++---
 drivers/firewire/core-cdev.c        |  2 +-
 drivers/firewire/core-device.c      | 30 ++++++++++++++++++-----------
 drivers/firewire/core-transaction.c | 12 +++++++++++-
 drivers/firewire/core.h             |  2 ++
 5 files changed, 36 insertions(+), 16 deletions(-)

diff --git a/drivers/firewire/core-card.c b/drivers/firewire/core-card.c
index f05fc7bfceeb..bb8c4d22b03e 100644
--- a/drivers/firewire/core-card.c
+++ b/drivers/firewire/core-card.c
@@ -228,8 +228,8 @@ void fw_schedule_bus_reset(struct fw_card *card, bool delayed, bool short_reset)
 
 	/* Use an arbitrary short delay to combine multiple reset requests. */
 	fw_card_get(card);
-	if (!schedule_delayed_work(&card->br_work,
-				   delayed ? DIV_ROUND_UP(HZ, 100) : 0))
+	if (!queue_delayed_work(fw_wq, &card->br_work,
+				delayed ? DIV_ROUND_UP(HZ, 100) : 0))
 		fw_card_put(card);
 }
 EXPORT_SYMBOL(fw_schedule_bus_reset);
@@ -241,7 +241,7 @@ static void br_work(struct work_struct *work)
 	/* Delay for 2s after last reset per IEEE 1394 clause 8.2.1. */
 	if (card->reset_jiffies != 0 &&
 	    time_before64(get_jiffies_64(), card->reset_jiffies + 2 * HZ)) {
-		if (!schedule_delayed_work(&card->br_work, 2 * HZ))
+		if (!queue_delayed_work(fw_wq, &card->br_work, 2 * HZ))
 			fw_card_put(card);
 		return;
 	}
diff --git a/drivers/firewire/core-cdev.c b/drivers/firewire/core-cdev.c
index 64768c2194f1..aa1131d26e30 100644
--- a/drivers/firewire/core-cdev.c
+++ b/drivers/firewire/core-cdev.c
@@ -149,7 +149,7 @@ static void release_iso_resource(struct client *, struct client_resource *);
 static void schedule_iso_resource(struct iso_resource *r, unsigned long delay)
 {
 	client_get(r->client);
-	if (!schedule_delayed_work(&r->work, delay))
+	if (!queue_delayed_work(fw_wq, &r->work, delay))
 		client_put(r->client);
 }
 
diff --git a/drivers/firewire/core-device.c b/drivers/firewire/core-device.c
index 9a262439e3a7..ef900d923f15 100644
--- a/drivers/firewire/core-device.c
+++ b/drivers/firewire/core-device.c
@@ -725,6 +725,14 @@ struct fw_device *fw_device_get_by_devt(dev_t devt)
 	return device;
 }
 
+struct workqueue_struct *fw_wq;
+
+static void fw_schedule_device_work(struct fw_device *device,
+				    unsigned long delay)
+{
+	queue_delayed_work(fw_wq, &device->work, delay);
+}
+
 /*
  * These defines control the retry behavior for reading the config
  * rom.  It shouldn't be necessary to tweak these; if the device
@@ -750,7 +758,7 @@ static void fw_device_shutdown(struct work_struct *work)
 	if (time_before64(get_jiffies_64(),
 			  device->card->reset_jiffies + SHUTDOWN_DELAY)
 	    && !list_empty(&device->card->link)) {
-		schedule_delayed_work(&device->work, SHUTDOWN_DELAY);
+		fw_schedule_device_work(device, SHUTDOWN_DELAY);
 		return;
 	}
 
@@ -862,7 +870,7 @@ static int lookup_existing_device(struct device *dev, void *data)
 		fw_notify("rediscovered device %s\n", dev_name(dev));
 
 		PREPARE_DELAYED_WORK(&old->work, fw_device_update);
-		schedule_delayed_work(&old->work, 0);
+		fw_schedule_device_work(old, 0);
 
 		if (current_node == card->root_node)
 			fw_schedule_bm_work(card, 0);
@@ -953,7 +961,7 @@ static void fw_device_init(struct work_struct *work)
 		if (device->config_rom_retries < MAX_RETRIES &&
 		    atomic_read(&device->state) == FW_DEVICE_INITIALIZING) {
 			device->config_rom_retries++;
-			schedule_delayed_work(&device->work, RETRY_DELAY);
+			fw_schedule_device_work(device, RETRY_DELAY);
 		} else {
 			if (device->node->link_on)
 				fw_notify("giving up on config rom for node id %x\n",
@@ -1019,7 +1027,7 @@ static void fw_device_init(struct work_struct *work)
 			   FW_DEVICE_INITIALIZING,
 			   FW_DEVICE_RUNNING) == FW_DEVICE_GONE) {
 		PREPARE_DELAYED_WORK(&device->work, fw_device_shutdown);
-		schedule_delayed_work(&device->work, SHUTDOWN_DELAY);
+		fw_schedule_device_work(device, SHUTDOWN_DELAY);
 	} else {
 		if (device->config_rom_retries)
 			fw_notify("created device %s: GUID %08x%08x, S%d00, "
@@ -1098,7 +1106,7 @@ static void fw_device_refresh(struct work_struct *work)
 		if (device->config_rom_retries < MAX_RETRIES / 2 &&
 		    atomic_read(&device->state) == FW_DEVICE_INITIALIZING) {
 			device->config_rom_retries++;
-			schedule_delayed_work(&device->work, RETRY_DELAY / 2);
+			fw_schedule_device_work(device, RETRY_DELAY / 2);
 
 			return;
 		}
@@ -1131,7 +1139,7 @@ static void fw_device_refresh(struct work_struct *work)
 		if (device->config_rom_retries < MAX_RETRIES &&
 		    atomic_read(&device->state) == FW_DEVICE_INITIALIZING) {
 			device->config_rom_retries++;
-			schedule_delayed_work(&device->work, RETRY_DELAY);
+			fw_schedule_device_work(device, RETRY_DELAY);
 
 			return;
 		}
@@ -1158,7 +1166,7 @@ static void fw_device_refresh(struct work_struct *work)
  gone:
 	atomic_set(&device->state, FW_DEVICE_GONE);
 	PREPARE_DELAYED_WORK(&device->work, fw_device_shutdown);
-	schedule_delayed_work(&device->work, SHUTDOWN_DELAY);
+	fw_schedule_device_work(device, SHUTDOWN_DELAY);
  out:
 	if (node_id == card->root_node->node_id)
 		fw_schedule_bm_work(card, 0);
@@ -1214,7 +1222,7 @@ void fw_node_event(struct fw_card *card, struct fw_node *node, int event)
 		 * first config rom scan half a second after bus reset.
 		 */
 		INIT_DELAYED_WORK(&device->work, fw_device_init);
-		schedule_delayed_work(&device->work, INITIAL_DELAY);
+		fw_schedule_device_work(device, INITIAL_DELAY);
 		break;
 
 	case FW_NODE_INITIATED_RESET:
@@ -1230,7 +1238,7 @@ void fw_node_event(struct fw_card *card, struct fw_node *node, int event)
 			    FW_DEVICE_RUNNING,
 			    FW_DEVICE_INITIALIZING) == FW_DEVICE_RUNNING) {
 			PREPARE_DELAYED_WORK(&device->work, fw_device_refresh);
-			schedule_delayed_work(&device->work,
+			fw_schedule_device_work(device,
 				device->is_local ? 0 : INITIAL_DELAY);
 		}
 		break;
@@ -1245,7 +1253,7 @@ void fw_node_event(struct fw_card *card, struct fw_node *node, int event)
 		device->generation = card->generation;
 		if (atomic_read(&device->state) == FW_DEVICE_RUNNING) {
 			PREPARE_DELAYED_WORK(&device->work, fw_device_update);
-			schedule_delayed_work(&device->work, 0);
+			fw_schedule_device_work(device, 0);
 		}
 		break;
 
@@ -1270,7 +1278,7 @@ void fw_node_event(struct fw_card *card, struct fw_node *node, int event)
 		if (atomic_xchg(&device->state,
 				FW_DEVICE_GONE) == FW_DEVICE_RUNNING) {
 			PREPARE_DELAYED_WORK(&device->work, fw_device_shutdown);
-			schedule_delayed_work(&device->work,
+			fw_schedule_device_work(device,
 				list_empty(&card->link) ? 0 : SHUTDOWN_DELAY);
 		}
 		break;
diff --git a/drivers/firewire/core-transaction.c b/drivers/firewire/core-transaction.c
index 77275fdf6c1f..d4c28a217b2c 100644
--- a/drivers/firewire/core-transaction.c
+++ b/drivers/firewire/core-transaction.c
@@ -36,6 +36,7 @@
 #include <linux/string.h>
 #include <linux/timer.h>
 #include <linux/types.h>
+#include <linux/workqueue.h>
 
 #include <asm/byteorder.h>
 
@@ -1213,13 +1214,21 @@ static int __init fw_core_init(void)
 {
 	int ret;
 
+	fw_wq = alloc_workqueue(KBUILD_MODNAME,
+				WQ_NON_REENTRANT | WQ_MEM_RECLAIM, 0);
+	if (!fw_wq)
+		return -ENOMEM;
+
 	ret = bus_register(&fw_bus_type);
-	if (ret < 0)
+	if (ret < 0) {
+		destroy_workqueue(fw_wq);
 		return ret;
+	}
 
 	fw_cdev_major = register_chrdev(0, "firewire", &fw_device_ops);
 	if (fw_cdev_major < 0) {
 		bus_unregister(&fw_bus_type);
+		destroy_workqueue(fw_wq);
 		return fw_cdev_major;
 	}
 
@@ -1235,6 +1244,7 @@ static void __exit fw_core_cleanup(void)
 {
 	unregister_chrdev(fw_cdev_major, "firewire");
 	bus_unregister(&fw_bus_type);
+	destroy_workqueue(fw_wq);
 	idr_destroy(&fw_device_idr);
 }
 
diff --git a/drivers/firewire/core.h b/drivers/firewire/core.h
index 0fe4e4e6eda7..00ea7730c6a7 100644
--- a/drivers/firewire/core.h
+++ b/drivers/firewire/core.h
@@ -140,6 +140,8 @@ void fw_cdev_handle_phy_packet(struct fw_card *card, struct fw_packet *p);
 extern struct rw_semaphore fw_device_rwsem;
 extern struct idr fw_device_idr;
 extern int fw_cdev_major;
+struct workqueue_struct;
+extern struct workqueue_struct *fw_wq;
 
 struct fw_device *fw_device_get_by_devt(dev_t devt);
 int fw_device_set_broadcast_channel(struct device *dev, void *gen);

From b75ca5ea8e439893121ad80406a3c04c4b7612ab Mon Sep 17 00:00:00 2001
From: Stefan Richter <stefanr@s5r6.in-berlin.de>
Date: Fri, 22 Apr 2011 12:21:44 +0200
Subject: [PATCH 7/9] firewire: sbp2: omit Scsi_Host lock from queuecommand

firewire-sbp2 already takes care for internal serialization where
required (ORB list accesses), and it does not use cmd->serial_number
internally.  Hence it is safe to not grab the shost lock around
queuecommand.

While we are at housekeeping, drop a redundant struct member:
sbp2_command_orb.done is set once in a hot path and dereferenced once in
a hot path.  We can as well dereference sbp2_command_orb.cmd->scsi_done
instead.

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
---
 drivers/firewire/sbp2.c | 20 ++++++--------------
 1 file changed, 6 insertions(+), 14 deletions(-)

diff --git a/drivers/firewire/sbp2.c b/drivers/firewire/sbp2.c
index 77ed589b360d..cc002e92275a 100644
--- a/drivers/firewire/sbp2.c
+++ b/drivers/firewire/sbp2.c
@@ -125,9 +125,6 @@ MODULE_PARM_DESC(workarounds, "Work around device bugs (default = 0"
 	", override internal blacklist = " __stringify(SBP2_WORKAROUND_OVERRIDE)
 	", or a combination)");
 
-/* I don't know why the SCSI stack doesn't define something like this... */
-typedef void (*scsi_done_fn_t)(struct scsi_cmnd *);
-
 static const char sbp2_driver_name[] = "sbp2";
 
 /*
@@ -314,7 +311,6 @@ struct sbp2_command_orb {
 		u8 command_block[SBP2_MAX_CDB_SIZE];
 	} request;
 	struct scsi_cmnd *cmd;
-	scsi_done_fn_t done;
 	struct sbp2_logical_unit *lu;
 
 	struct sbp2_pointer page_table[SG_ALL] __attribute__((aligned(8)));
@@ -1398,7 +1394,7 @@ static void complete_command_orb(struct sbp2_orb *base_orb,
 	sbp2_unmap_scatterlist(device->card->device, orb);
 
 	orb->cmd->result = result;
-	orb->done(orb->cmd);
+	orb->cmd->scsi_done(orb->cmd);
 }
 
 static int sbp2_map_scatterlist(struct sbp2_command_orb *orb,
@@ -1463,7 +1459,8 @@ static int sbp2_map_scatterlist(struct sbp2_command_orb *orb,
 
 /* SCSI stack integration */
 
-static int sbp2_scsi_queuecommand_lck(struct scsi_cmnd *cmd, scsi_done_fn_t done)
+static int sbp2_scsi_queuecommand(struct Scsi_Host *shost,
+				  struct scsi_cmnd *cmd)
 {
 	struct sbp2_logical_unit *lu = cmd->device->hostdata;
 	struct fw_device *device = target_device(lu->tgt);
@@ -1477,7 +1474,7 @@ static int sbp2_scsi_queuecommand_lck(struct scsi_cmnd *cmd, scsi_done_fn_t done
 	if (cmd->sc_data_direction == DMA_BIDIRECTIONAL) {
 		fw_error("Can't handle DMA_BIDIRECTIONAL, rejecting command\n");
 		cmd->result = DID_ERROR << 16;
-		done(cmd);
+		cmd->scsi_done(cmd);
 		return 0;
 	}
 
@@ -1490,11 +1487,8 @@ static int sbp2_scsi_queuecommand_lck(struct scsi_cmnd *cmd, scsi_done_fn_t done
 	/* Initialize rcode to something not RCODE_COMPLETE. */
 	orb->base.rcode = -1;
 	kref_init(&orb->base.kref);
-
-	orb->lu   = lu;
-	orb->done = done;
-	orb->cmd  = cmd;
-
+	orb->lu = lu;
+	orb->cmd = cmd;
 	orb->request.next.high = cpu_to_be32(SBP2_ORB_NULL);
 	orb->request.misc = cpu_to_be32(
 		COMMAND_ORB_MAX_PAYLOAD(lu->tgt->max_payload) |
@@ -1529,8 +1523,6 @@ static int sbp2_scsi_queuecommand_lck(struct scsi_cmnd *cmd, scsi_done_fn_t done
 	return retval;
 }
 
-static DEF_SCSI_QCMD(sbp2_scsi_queuecommand)
-
 static int sbp2_scsi_slave_alloc(struct scsi_device *sdev)
 {
 	struct sbp2_logical_unit *lu = sdev->hostdata;

From 81bf52d8622f05cfe89893fd5c1101efd85f855b Mon Sep 17 00:00:00 2001
From: Stefan Richter <stefanr@s5r6.in-berlin.de>
Date: Sun, 1 May 2011 21:06:42 +0200
Subject: [PATCH 8/9] firewire: sbp2: octlet AT payloads can be stack-allocated

We do not need slab allocations for ORB pointer write transactions
anymore in order to satisfy streaming DMA mapping constraints, thanks to
commit da28947e7e36 "firewire: ohci: avoid separate DMA mapping for
small AT payloads".

(Besides, the slab-allocated buffers that firewire-sbp2 used to provide
for 8-byte write requests were still not fully portable since they
shared a cacheline with unrelated CPU-accessed data.)

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
---
 drivers/firewire/sbp2.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/drivers/firewire/sbp2.c b/drivers/firewire/sbp2.c
index cc002e92275a..2aafc614ae14 100644
--- a/drivers/firewire/sbp2.c
+++ b/drivers/firewire/sbp2.c
@@ -258,7 +258,6 @@ struct sbp2_orb {
 	struct kref kref;
 	dma_addr_t request_bus;
 	int rcode;
-	struct sbp2_pointer pointer;
 	void (*callback)(struct sbp2_orb * orb, struct sbp2_status * status);
 	struct list_head link;
 };
@@ -490,10 +489,11 @@ static void sbp2_send_orb(struct sbp2_orb *orb, struct sbp2_logical_unit *lu,
 			  int node_id, int generation, u64 offset)
 {
 	struct fw_device *device = target_device(lu->tgt);
+	struct sbp2_pointer orb_pointer;
 	unsigned long flags;
 
-	orb->pointer.high = 0;
-	orb->pointer.low = cpu_to_be32(orb->request_bus);
+	orb_pointer.high = 0;
+	orb_pointer.low = cpu_to_be32(orb->request_bus);
 
 	spin_lock_irqsave(&device->card->lock, flags);
 	list_add_tail(&orb->link, &lu->orb_list);
@@ -504,7 +504,7 @@ static void sbp2_send_orb(struct sbp2_orb *orb, struct sbp2_logical_unit *lu,
 
 	fw_send_request(device->card, &orb->t, TCODE_WRITE_BLOCK_REQUEST,
 			node_id, generation, device->max_speed, offset,
-			&orb->pointer, 8, complete_transaction, orb);
+			&orb_pointer, 8, complete_transaction, orb);
 }
 
 static int sbp2_cancel_orbs(struct sbp2_logical_unit *lu)

From 105e53f863c04e1d9e5bb34bf753c9fdbce6a60c Mon Sep 17 00:00:00 2001
From: Stefan Richter <stefanr@s5r6.in-berlin.de>
Date: Sun, 1 May 2011 20:50:31 +0200
Subject: [PATCH 9/9] firewire: sbp2: parallelize login, reconnect, logout

The struct sbp2_logical_unit.work items can all be executed in parallel
but are not reentrant.  Furthermore, reconnect or re-login work must be
executed in a WQ_MEM_RECLAIM workqueue.

Hence replace the old single-threaded firewire-sbp2 workqueue by a
concurrency-managed but non-reentrant workqueue with rescuer.
firewire-core already maintains one, hence use this one.

In earlier versions of this change, I observed occasional failures of
parallel INQUIRY to an Initio INIC-2430 FireWire 800 to dual IDE bridge.
More testing indicates that parallel INQUIRY is not actually a problem,
but too quick successions of logout and login + INQUIRY, e.g. a quick
sequence of cable plugout and plugin, can result in failed INQUIRY.
This does not seem to be something that should or could be addressed by
serialization.

Another dual-LU device to which I currently have access to, an
OXUF924DSB FireWire 800 to dual SATA bridge with firmware from MacPower,
has been successfully tested with this too.

This change is beneficial to environments with two or more FireWire
storage devices, especially if they are located on the same bus.
Management tasks that should be performed as soon and as quickly as
possible, especially reconnect, are no longer held up by tasks on other
devices that may take a long time, especially login with INQUIRY and sd
or sr driver probe.

Signed-off-by: Stefan Richter <stefanr@s5r6.in-berlin.de>
---
 drivers/firewire/core-card.c        |  4 ++--
 drivers/firewire/core-cdev.c        |  2 +-
 drivers/firewire/core-device.c      |  5 +++--
 drivers/firewire/core-transaction.c | 12 ++++++------
 drivers/firewire/core.h             |  2 --
 drivers/firewire/sbp2.c             |  9 +--------
 include/linux/firewire.h            |  2 ++
 7 files changed, 15 insertions(+), 21 deletions(-)

diff --git a/drivers/firewire/core-card.c b/drivers/firewire/core-card.c
index bb8c4d22b03e..29d2423fae6d 100644
--- a/drivers/firewire/core-card.c
+++ b/drivers/firewire/core-card.c
@@ -228,7 +228,7 @@ void fw_schedule_bus_reset(struct fw_card *card, bool delayed, bool short_reset)
 
 	/* Use an arbitrary short delay to combine multiple reset requests. */
 	fw_card_get(card);
-	if (!queue_delayed_work(fw_wq, &card->br_work,
+	if (!queue_delayed_work(fw_workqueue, &card->br_work,
 				delayed ? DIV_ROUND_UP(HZ, 100) : 0))
 		fw_card_put(card);
 }
@@ -241,7 +241,7 @@ static void br_work(struct work_struct *work)
 	/* Delay for 2s after last reset per IEEE 1394 clause 8.2.1. */
 	if (card->reset_jiffies != 0 &&
 	    time_before64(get_jiffies_64(), card->reset_jiffies + 2 * HZ)) {
-		if (!queue_delayed_work(fw_wq, &card->br_work, 2 * HZ))
+		if (!queue_delayed_work(fw_workqueue, &card->br_work, 2 * HZ))
 			fw_card_put(card);
 		return;
 	}
diff --git a/drivers/firewire/core-cdev.c b/drivers/firewire/core-cdev.c
index aa1131d26e30..b1c11775839c 100644
--- a/drivers/firewire/core-cdev.c
+++ b/drivers/firewire/core-cdev.c
@@ -149,7 +149,7 @@ static void release_iso_resource(struct client *, struct client_resource *);
 static void schedule_iso_resource(struct iso_resource *r, unsigned long delay)
 {
 	client_get(r->client);
-	if (!queue_delayed_work(fw_wq, &r->work, delay))
+	if (!queue_delayed_work(fw_workqueue, &r->work, delay))
 		client_put(r->client);
 }
 
diff --git a/drivers/firewire/core-device.c b/drivers/firewire/core-device.c
index ef900d923f15..95a471401892 100644
--- a/drivers/firewire/core-device.c
+++ b/drivers/firewire/core-device.c
@@ -725,12 +725,13 @@ struct fw_device *fw_device_get_by_devt(dev_t devt)
 	return device;
 }
 
-struct workqueue_struct *fw_wq;
+struct workqueue_struct *fw_workqueue;
+EXPORT_SYMBOL(fw_workqueue);
 
 static void fw_schedule_device_work(struct fw_device *device,
 				    unsigned long delay)
 {
-	queue_delayed_work(fw_wq, &device->work, delay);
+	queue_delayed_work(fw_workqueue, &device->work, delay);
 }
 
 /*
diff --git a/drivers/firewire/core-transaction.c b/drivers/firewire/core-transaction.c
index d4c28a217b2c..334b82a3542c 100644
--- a/drivers/firewire/core-transaction.c
+++ b/drivers/firewire/core-transaction.c
@@ -1214,21 +1214,21 @@ static int __init fw_core_init(void)
 {
 	int ret;
 
-	fw_wq = alloc_workqueue(KBUILD_MODNAME,
-				WQ_NON_REENTRANT | WQ_MEM_RECLAIM, 0);
-	if (!fw_wq)
+	fw_workqueue = alloc_workqueue("firewire",
+				       WQ_NON_REENTRANT | WQ_MEM_RECLAIM, 0);
+	if (!fw_workqueue)
 		return -ENOMEM;
 
 	ret = bus_register(&fw_bus_type);
 	if (ret < 0) {
-		destroy_workqueue(fw_wq);
+		destroy_workqueue(fw_workqueue);
 		return ret;
 	}
 
 	fw_cdev_major = register_chrdev(0, "firewire", &fw_device_ops);
 	if (fw_cdev_major < 0) {
 		bus_unregister(&fw_bus_type);
-		destroy_workqueue(fw_wq);
+		destroy_workqueue(fw_workqueue);
 		return fw_cdev_major;
 	}
 
@@ -1244,7 +1244,7 @@ static void __exit fw_core_cleanup(void)
 {
 	unregister_chrdev(fw_cdev_major, "firewire");
 	bus_unregister(&fw_bus_type);
-	destroy_workqueue(fw_wq);
+	destroy_workqueue(fw_workqueue);
 	idr_destroy(&fw_device_idr);
 }
 
diff --git a/drivers/firewire/core.h b/drivers/firewire/core.h
index 00ea7730c6a7..0fe4e4e6eda7 100644
--- a/drivers/firewire/core.h
+++ b/drivers/firewire/core.h
@@ -140,8 +140,6 @@ void fw_cdev_handle_phy_packet(struct fw_card *card, struct fw_packet *p);
 extern struct rw_semaphore fw_device_rwsem;
 extern struct idr fw_device_idr;
 extern int fw_cdev_major;
-struct workqueue_struct;
-extern struct workqueue_struct *fw_wq;
 
 struct fw_device *fw_device_get_by_devt(dev_t devt);
 int fw_device_set_broadcast_channel(struct device *dev, void *gen);
diff --git a/drivers/firewire/sbp2.c b/drivers/firewire/sbp2.c
index 2aafc614ae14..41841a3e3f99 100644
--- a/drivers/firewire/sbp2.c
+++ b/drivers/firewire/sbp2.c
@@ -826,8 +826,6 @@ static void sbp2_target_put(struct sbp2_target *tgt)
 	kref_put(&tgt->kref, sbp2_release_target);
 }
 
-static struct workqueue_struct *sbp2_wq;
-
 /*
  * Always get the target's kref when scheduling work on one its units.
  * Each workqueue job is responsible to call sbp2_target_put() upon return.
@@ -835,7 +833,7 @@ static struct workqueue_struct *sbp2_wq;
 static void sbp2_queue_work(struct sbp2_logical_unit *lu, unsigned long delay)
 {
 	sbp2_target_get(lu->tgt);
-	if (!queue_delayed_work(sbp2_wq, &lu->work, delay))
+	if (!queue_delayed_work(fw_workqueue, &lu->work, delay))
 		sbp2_target_put(lu->tgt);
 }
 
@@ -1645,17 +1643,12 @@ MODULE_ALIAS("sbp2");
 
 static int __init sbp2_init(void)
 {
-	sbp2_wq = create_singlethread_workqueue(KBUILD_MODNAME);
-	if (!sbp2_wq)
-		return -ENOMEM;
-
 	return driver_register(&sbp2_driver.driver);
 }
 
 static void __exit sbp2_cleanup(void)
 {
 	driver_unregister(&sbp2_driver.driver);
-	destroy_workqueue(sbp2_wq);
 }
 
 module_init(sbp2_init);
diff --git a/include/linux/firewire.h b/include/linux/firewire.h
index c0fb405bb435..5e6f42789afe 100644
--- a/include/linux/firewire.h
+++ b/include/linux/firewire.h
@@ -449,4 +449,6 @@ void fw_iso_resource_manage(struct fw_card *card, int generation,
 			    u64 channels_mask, int *channel, int *bandwidth,
 			    bool allocate);
 
+extern struct workqueue_struct *fw_workqueue;
+
 #endif /* _LINUX_FIREWIRE_H */