firewire: core: no need to track irq flags in bm_work
PATCH 1/2 firewire: core: no need to track irq flags in bm_work
This is a workqueue job and always entered with IRQs enabled.
--- drivers/firewire/core-card.c | 15 +++++++-------- 1 file changed, 7 insertions(+), 8 deletions(-)
drivers/firewire/core-card.c +7 -8
@@ -239,7 +239,6 @@ static void fw_card_bm_work(struct work_ struct fw_card *card = container_of(work, struct fw_card, work.work);struct fw_device *root_device, *irm_device;struct fw_node *root_node; -unsigned long flags;int root_id, new_root_id, irm_id, local_id;int gap_count, generation, grace, rcode;bool do_reset = false;
@@ -247,10 +246,10 @@ static void fw_card_bm_work(struct work_bool root_device_is_cmc;bool irm_is_1394_1995_only; -spin_lock_irqsave (&card->lock, flags); +spin_lock_irq (&card->lock);if (card->local_node ==NULL ) { -spin_unlock_irqrestore (&card->lock, flags); +spin_unlock_irq (&card->lock); goto out_put_card; }
@@ -305,7 +304,7 @@ static void fw_card_bm_work(struct work_ card->bm_transaction_data[0 ] =cpu_to_be32 (0x3f ); card->bm_transaction_data[1 ] =cpu_to_be32 (local_id); -spin_unlock_irqrestore (&card->lock, flags); +spin_unlock_irq (&card->lock); rcode = fw_run_transaction(card, TCODE_LOCK_COMPARE_SWAP, irm_id, generation, SCODE_100,
@@ -336,7 +335,7 @@ static void fw_card_bm_work(struct work_ goto out; } -spin_lock_irqsave (&card->lock, flags); +spin_lock_irq (&card->lock);if (rcode !=RCODE_COMPLETE ) {/*
@@ -355,7 +354,7 @@ static void fw_card_bm_work(struct work_* We weren't BM in the last generation, and the last * bus reset is less than 125ms ago. Reschedule this job. */ -spin_unlock_irqrestore (&card->lock, flags); +spin_unlock_irq (&card->lock);fw_schedule_bm_work (card,DIV_ROUND_UP (HZ ,8 )); goto out; }
@@ -378,7 +377,7 @@ static void fw_card_bm_work(struct work_* If we haven't probed this device yet, bail out now * and let's try again once that's done. */ -spin_unlock_irqrestore (&card->lock, flags); +spin_unlock_irq (&card->lock); goto out; }else if (root_device_is_cmc) {/*
@@ -416,7 +415,7 @@ static void fw_card_bm_work(struct work_ (card->gap_count != gap_count || new_root_id != root_id)) do_reset = true; -spin_unlock_irqrestore (&card->lock, flags); +spin_unlock_irq (&card->lock);if (do_reset) { fw_notify("phy config: card %d, new root=%x, gap_count=%d\n",
Signature · 3 lines
Stefan Richter -=====-==-=- -==- =-=-= http://arcgraph.de/sr/
PATCH 2/2 firewire: cdev: fix fw_cdev_event_bus_reset.bm_node_id
Fix an obscure ABI feature that is a bit of a hassle to implement. However, somebody put it into the ABI, so let's fill in a sensible value there.
--- drivers/firewire/core-card.c | 12 +++++++++--- drivers/firewire/core-cdev.c | 2 +- drivers/firewire/core-topology.c | 1 + include/linux/firewire-cdev.h | 5 +++++ include/linux/firewire.h | 1 + 5 files changed, 17 insertions(+), 4 deletions(-)
drivers/firewire/core-card.c +9 -3
@@ -239,7 +239,7 @@ static void fw_card_bm_work(struct work_ struct fw_card *card = container_of(work, struct fw_card, work.work);struct fw_device *root_device, *irm_device;struct fw_node *root_node; -int root_id, new_root_id, irm_id, local_id; +int root_id, new_root_id, irm_id, bm_id, local_id;int gap_count, generation, grace, rcode;bool do_reset = false;bool root_device_is_running;
@@ -315,9 +315,15 @@ static void fw_card_bm_work(struct work_/* Another bus reset, BM work has been rescheduled. */ goto out; -if (rcode ==RCODE_COMPLETE && - card->bm_transaction_data[0 ] !=cpu_to_be32 (0x3f )) { + bm_id =be32_to_cpu (card->bm_transaction_data[0 ]); +spin_lock_irq (&card->lock); +if (rcode ==RCODE_COMPLETE && generation == card->generation) + card->bm_node_id = + bm_id ==0x3f ? local_id :0xffc0 | bm_id; +spin_unlock_irq (&card->lock); + +if (rcode ==RCODE_COMPLETE && bm_id !=0x3f ) {/* Somebody else is BM. Only act as IRM. */ if (local_id == irm_id)allocate_broadcast_channel (card, generation);
drivers/firewire/core-cdev.c +1 -1
@@ -319,7 +319,7 @@ static void fill_bus_reset_event(struct event->generation = client->device->generation; event->node_id = client->device->node_id; event->local_node_id = card->local_node->node_id; - event->bm_node_id =0 ;/* FIXME: We don't track the BM. */ + event->bm_node_id = card->bm_node_id; event->irm_node_id = card->irm_node->node_id; event->root_node_id = card->root_node->node_id;
drivers/firewire/core-topology.c +1 -0
@@ -552,6 +552,7 @@ void fw_core_handle_bus_reset(struct fw_smp_wmb (); card->generation = generation; card->reset_jiffies = jiffies; + card->bm_node_id =0xffff ; card->bm_abdicate = bm_abdicate;fw_schedule_bm_work (card,0 );
include/linux/firewire-cdev.h +5 -0
@@ -71,6 +71,10 @@ struct fw_cdev_event_common {* This event is sent when the bus the device belongs to goes through a bus * reset. It provides information about the new bus configuration, such as * new node ID for this device, new root ID, and others. +* +* If @bm_node_id is 0xffff right after bus reset it can be reread by an +* %FW_CDEV_IOC_GET_INFO ioctl after bus manager selection was finished. +* Kernels with ABI version < 4 do not set @bm_node_id. */ struct fw_cdev_event_bus_reset {__u64 closure;
@@ -353,6 +357,7 @@ union fw_cdev_event {* 3 (2.6.34) - made &fw_cdev_get_cycle_timer reliable * - added %FW_CDEV_IOC_GET_CYCLE_TIMER2 * 4 (2.6.36) - added %FW_CDEV_EVENT_REQUEST2 +* - implemented &fw_cdev_event_bus_reset.bm_node_id */ #define FW_CDEV_VERSION 3/* Meaningless; don't use this macro. */
include/linux/firewire.h +1 -0
@@ -119,6 +119,7 @@ struct fw_card {int bm_retries;int bm_generation;__be32 bm_transaction_data[2 ]; +int bm_node_id;bool bm_abdicate;bool priority_budget_implemented;/* controller feature */
Signature · 3 lines
Stefan Richter -=====-==-=- -==- =-=-= http://arcgraph.de/sr/
Quoted 1 line · Stefan Richter
This is a workqueue job and always entered with IRQs enabled.
did you mean 'disabled' ?
Philippe
Quoted 24 lines
Signed-off-by Stefan Richter <[email protected]>--- drivers/firewire/core-card.c | 15 +++++++-------- 1 file changed, 7 insertions(+), 8 deletions(-)Index:
b/drivers/firewire/core-card.c
===================================================================drivers/firewire/core-card.c +1
-2@@ -239,7 +239,6 @@ static void fw_card_bm_work(struct work_ struct fw_card *card = container_of(work, struct fw_card, work.work);struct fw_device *root_device, *irm_device;struct fw_node *root_node; -unsigned long flags;int root_id, new_root_id, irm_id, local_id;int gap_count, generation, grace, rcode;bool do_reset = false;@@ -247,10 +246,10 @@ static void fw_card_bm_work(struct work_bool root_device_is_cmc;bool irm_is_1394_1995_only; -spin_lock_irqsave (&card->lock, flags); +spin_lock_irq (&card->lock);
Quoted 4 lines · Philippe De Muyter
Quoted 1 line · Stefan Richter
This is a workqueue job and always entered with IRQs enabled.
did you mean 'disabled' ?
I meant enabled.
[...]
Quoted 6 lines
drivers/firewire/core-card.c static void fw_card_bm_work(struct work_ · line 249
Quoted 6 lines
drivers/firewire/core-card.c +1
-1@@ -247,10 +246,10 @@ static void fw_card_bm_work(struct work_bool root_device_is_cmc;bool irm_is_1394_1995_only; -spin_lock_irqsave (&card->lock, flags); +spin_lock_irq (&card->lock);
- spin_lock + spin_unlock don't influence whether IRQs on the current CPU are on or off.
- spin_lock_irq + spin_unlock_irq always switch IRQs on the current CPU off and back on. This is necessary if the lock could also be taken by an IRQ handler. (Well, card->lock is actually only taken by process contexts and by tasklets. Seems we could switch to spin_lock_bh + spin_unlock_bh for card->lock everywhere in the firewire stack.)
- spin_lock_irqsave + spin_unlock_irqrestore switch IRQs on the current CPU off and back on only if used while IRQs are enabled; if used while local IRQs are already disabled they leave them disabled.
http://lwn.net/images/pdf/LDD3/ch05.pdf#page=14
Therefore some people prefer to use the safer spin_lock_irqsave()/ spin_unlock_irqrestore() everywhere. However, their downsides are the need to track IRQ state flags, and --- subjectively --- that their appearance in the code could create an impression to a casual reader that this code was meant to be able to run in IRQs-on context as well as in IRQs-off context. fw_card_bm_work() however definitely requires to be called with IRQs on, notably to be able to wait for IEEE 1394 transactions to complete.
Signature · 3 lines
Stefan Richter -=====-==-=- -==- =-==- http://arcgraph.de/sr/
Hello Stephan,
Quoted 33 lines · Stefan Richter
Quoted 4 lines · Philippe De Muyter
Quoted 1 line · Stefan Richter
This is a workqueue job and always entered with IRQs enabled.
did you mean 'disabled' ?
I meant enabled.
[...]
Quoted 6 lines
drivers/firewire/core-card.c static void fw_card_bm_work(struct work_ · line 249
Quoted 6 lines
drivers/firewire/core-card.c +1
-1@@ -247,10 +246,10 @@ static void fw_card_bm_work(struct work_bool root_device_is_cmc;bool irm_is_1394_1995_only; -spin_lock_irqsave (&card->lock, flags); +spin_lock_irq (&card->lock);-
spin_lock+spin_unlockdon't influence whether IRQs on the current CPU are on or off.-
spin_lock_irq+spin_unlock_irqalways switch IRQs on the current CPU off and back on. This is necessary if the lock could also be taken by an IRQ handler. (Well,card->lockis actually only taken by process contexts and by tasklets. Seems we could switch tospin_lock_bh+spin_unlock_bhforcard->lockeverywhere in the firewire stack.)-
spin_lock_irqsave+spin_unlock_irqrestoreswitch IRQs on the current CPU off and back on only if used while IRQs are enabled; if used while local IRQs are already disabled they leave them disabled.
Quoted 8 lines
Therefore some people prefer to use the safer spin_lock_irqsave()/
spin_unlock_irqrestore()everywhere. However, their downsides are the need to track IRQ state flags, and --- subjectively --- that their appearance in the code could create an impression to a casual reader that this code was meant to be able to run in IRQs-on context as well as in IRQs-off context.fw_card_bm_work()however definitely requires to be called with IRQs on, notably to be able to wait for IEEE 1394 transactions to complete.
Thanks for the clear explanation, and sorry for your wasted time.
Philippe
Quoted 1 line · Philippe De Muyter
Thanks for the clear explanation, and sorry for your wasted time.
Oh, no problem. It helps that somebody actually looks at these patches and asks if something looks illogical. I for one am on record to have posted patches that got things backwards. :-)
Signature · 3 lines
Stefan Richter -=====-==-=- -==- =-==- http://arcgraph.de/sr/