Skip to contentSkip to conversation

ARM: bcm281xx: Add timer driver (driver portion)

PATCH New Christian Daudt 6 participants 12 messages Mar 13, 2013 – Apr 2, 2013

Subsystems (6)
lore
Share & tools
lore
#0 lore raw

PATCH v7 1/2 ARM: bcm281xx: Add timer driver (driver portion)

This adds support for the Broadcom timer, used in the following SoCs:
BCM11130, BCM11140, BCM11351, BCM28145, BCM28155

Updates from V6:
- Split DT portion into a separate patch

Updates from V5:
- Rebase to latest arm-soc/for-next

Updates from V4:
- Switch code to use CLOCKSOURCE_OF_DECLARE

Updates from V3:
- Migrate to 3.9 timer framework updates

Updates from V2:
- prepend static fns + fields with kona_

Updates from V1:
- Rename bcm_timer.c to bcm_kona_timer.c
- Pull .h into bcm_kona_timer.c
- Make timers static
- Clean up comment block
- Switched to using clockevents_config_and_register
- Added an error to the get_timer loop if it repeats too much
- Added to Documentation/devicetree/bindings/arm/bcm/bcm,kona-timer.txt
- Added missing readl to timer_disable_and_clear

Note: bcm,kona-timer was kept as the 'compatible' field to make it specific enough for when there are multiple bcm timers (bcm,timer is too generic).

arch/arm/mach-bcm/Kconfig +1 -0
index bf02471..f112895 100644
@@ -6,6 +6,7 @@ config ARCH_BCM
 	select ARM_ERRATA_764369 if SMP
 	select ARM_GIC
 	select CPU_V7
+	select CLKSRC_OF
 	select GENERIC_CLOCKEVENTS
 	select GENERIC_TIME
 	select GPIO_BCM
arch/arm/mach-bcm/board_bcm.c +2 -5
index f0f9aba..2595935 100644
@@ -16,14 +16,11 @@
 #include <linux/device.h>
 #include <linux/platform_device.h>
 #include <linux/irqchip.h>
+#include <linux/clocksource.h>
 
 #include <asm/mach/arch.h>
 #include <asm/mach/time.h>
 
-static void timer_init(void)
-{
-}
-
 
 static void __init board_init(void)
 {
@@ -35,7 +32,7 @@ static const char * const bcm11351_dt_compat[] = { "bcm,bcm11351", NULL, };
 
 DT_MACHINE_START(BCM11351_DT, "Broadcom Application Processor")
 	.init_irq = irqchip_init,
-	.init_time = timer_init,
+	.init_time = clocksource_of_init,
 	.init_machine = board_init,
 	.dt_compat = bcm11351_dt_compat,
 MACHINE_END
drivers/clocksource/Makefile +1 -0
index 4d8283a..96e2531 100644
@@ -19,6 +19,7 @@ obj-$(CONFIG_ARCH_BCM2835)	+= bcm2835_timer.o
 obj-$(CONFIG_SUNXI_TIMER)	+= sunxi_timer.o
 obj-$(CONFIG_ARCH_TEGRA)	+= tegra20_timer.o
 obj-$(CONFIG_VT8500_TIMER)	+= vt8500_timer.o
+obj-$(CONFIG_ARCH_BCM)		+= bcm_kona_timer.o
 
 obj-$(CONFIG_ARM_ARCH_TIMER)		+= arm_arch_timer.o
 obj-$(CONFIG_CLKSRC_METAG_GENERIC)	+= metag_generic.o
drivers/clocksource/bcm_kona_timer.c new file +211 -0

211 lines not on this page · raw message

Signature · 1 line
1.7.10.4
#1 lore raw

PATCH v7 2/2 ARM: bcm281xx: Add timer driver (DT portion)

This adds support for the Broadcom timer, used in the following SoCs:
BCM11130, BCM11140, BCM11351, BCM28145, BCM28155

Updates from V6:
- Split DT portion into a separate patch

Updates from V5:
- Rebase to latest arm-soc/for-next

Updates from V4:
- Switch code to use CLOCKSOURCE_OF_DECLARE

Updates from V3:
- Migrate to 3.9 timer framework updates

Updates from V2:
- prepend static fns + fields with kona_

Updates from V1:
- Rename bcm_timer.c to bcm_kona_timer.c
- Pull .h into bcm_kona_timer.c
- Make timers static
- Clean up comment block
- Switched to using clockevents_config_and_register
- Added an error to the get_timer loop if it repeats too much
- Added to Documentation/devicetree/bindings/arm/bcm/bcm,kona-timer.txt
- Added missing readl to timer_disable_and_clear

Note: bcm,kona-timer was kept as the 'compatible' field to make it specific enough for when there are multiple bcm timers (bcm,timer is too generic).

Documentation/devicetree/bindings/arm/bcm/bcm,kona-timer.txt new file +19 -0
new file mode 100644
index 0000000..59fa6e6
@@ -0,0 +1,19 @@
+Broadcom Kona Family timer
+-----------------------------------------------------
+This timer is used in the following Broadcom SoCs:
+ BCM11130, BCM11140, BCM11351, BCM28145, BCM28155
+
+Required properties:
+- compatible : "bcm,kona-timer"
+- reg : Register range for the timer
+- interrupts : interrupt for the timer
+- clock-frequency: frequency that the clock operates
+
+Example:
+	timer at 35006000 {
+		compatible = "bcm,kona-timer";
+		reg = <0x35006000 0x1000>;
+		interrupts = <0x0 7 0x4>;
+		clock-frequency = <32768>;
+	};
+
arch/arm/boot/dts/bcm11351.dtsi +8 -0
index ad13588..8f71f40 100644
@@ -47,4 +47,12 @@
 		    cache-unified;
 		    cache-level = <2>;
 	};
+
+	timer at 35006000 {
+		compatible = "bcm,kona-timer";
+		reg = <0x35006000 0x1000>;
+		interrupts = <0x0 7 0x4>;
+		clock-frequency = <32768>;
+	};
+
 };
Signature · 1 line
1.7.10.4
#2 lore raw

PATCH v7 2/2 ARM: bcm281xx: Add timer driver (DT portion)

Quoted 80 lines · Christian Daudt

This adds support for the Broadcom timer, used in the following SoCs:
BCM11130, BCM11140, BCM11351, BCM28145, BCM28155

Updates from V6:
- Split DT portion into a separate patch

Updates from V5:
- Rebase to latest arm-soc/for-next

Updates from V4:
- Switch code to use CLOCKSOURCE_OF_DECLARE

Updates from V3:
- Migrate to 3.9 timer framework updates

Updates from V2:
- prepend static fns + fields with kona_

Updates from V1:
- Rename bcm_timer.c to bcm_kona_timer.c
- Pull .h into bcm_kona_timer.c
- Make timers static
- Clean up comment block
- Switched to using clockevents_config_and_register
- Added an error to the get_timer loop if it repeats too much
- Added to Documentation/devicetree/bindings/arm/bcm/bcm,kona-timer.txt
- Added missing readl to timer_disable_and_clear

Note: bcm,kona-timer was kept as the 'compatible' field to make it specific enough for when there are multiple bcm timers (bcm,timer is too generic).

Documentation/devicetree/bindings/arm/bcm/bcm,kona-timer.txt new file +19 -0
new file mode 100644
index 0000000..59fa6e6
@@ -0,0 +1,19 @@
+Broadcom Kona Family timer
+-----------------------------------------------------
+This timer is used in the following Broadcom SoCs:
+ BCM11130, BCM11140, BCM11351, BCM28145, BCM28155
+
+Required properties:
+- compatible : "bcm,kona-timer"
+- reg : Register range for the timer
+- interrupts : interrupt for the timer
+- clock-frequency: frequency that the clock operates
+
+Example:
+       timer at 35006000 {
+               compatible = "bcm,kona-timer";
+               reg = <0x35006000 0x1000>;
+               interrupts = <0x0 7 0x4>;
+               clock-frequency = <32768>;
+       };
+
arch/arm/boot/dts/bcm11351.dtsi +8 -1
index ad13588..8f71f40 100644
@@ -47,4 +47,12 @@
                    cache-unified;
                    cache-level = <2>;
        };
+
+       timer at 35006000 {
+               compatible = "bcm,kona-timer";
+               reg = <0x35006000 0x1000>;
+               interrupts = <0x0 7 0x4>;
+               clock-frequency = <32768>;
+       };
+
 };
--
Quoted 3 lines

1.7.10.4

Hi Grant,
In discussion with Olof @ Connect, he suggested that it would be best
to send dt changes as a standalone patch to devicetree-discuss list to get it applied on its own. Can you pls apply this patch ?

Thanks,
csd

#3 lore raw

PATCH v7 2/2 ARM: bcm281xx: Add timer driver (DT portion)

Quoted 4 lines · Christian Daudt

Hi Grant,
In discussion with Olof @ Connect, he suggested that it would be best
to send dt changes as a standalone patch to devicetree-discuss list to get it applied on its own. Can you pls apply this patch ?

Actually, I wanted them split but since these change the SoC dtsi it should probably still go through arm-soc. I've applied it to our next/soc branch now.

Based on the thread, it looks like the main driver is picked up by John and will go in that through his tree.

Thanks for your patience on dealing with this!

-Olof

#4 lore raw

PATCH v7 2/2 ARM: bcm281xx: Add timer driver (DT portion)

In reply to Christian Daudt
Quoted 28 lines · Christian Daudt

This adds support for the Broadcom timer, used in the following SoCs:
BCM11130, BCM11140, BCM11351, BCM28145, BCM28155

Updates from V6:
- Split DT portion into a separate patch

Updates from V5:
- Rebase to latest arm-soc/for-next

Updates from V4:
- Switch code to use CLOCKSOURCE_OF_DECLARE

Updates from V3:
- Migrate to 3.9 timer framework updates

Updates from V2:
- prepend static fns + fields with kona_

Updates from V1:
- Rename bcm_timer.c to bcm_kona_timer.c
- Pull .h into bcm_kona_timer.c
- Make timers static
- Clean up comment block
- Switched to using clockevents_config_and_register
- Added an error to the get_timer loop if it repeats too much
- Added to Documentation/devicetree/bindings/arm/bcm/bcm,kona-timer.txt
- Added missing readl to timer_disable_and_clear

Quoted 8 lines

Note: bcm,kona-timer was kept as the 'compatible' field to make it specific enough for when there are multiple bcm timers (bcm,timer is too generic).

Thanks, applied to next/soc.

-Olof

#5 lore raw

PATCH v7 1/2 ARM: bcm281xx: Add timer driver (driver portion)

In reply to Christian Daudt
Quoted 2 lines · Christian Daudt

This adds support for the Broadcom timer, used in the following SoCs:
BCM11130, BCM11140, BCM11351, BCM28145, BCM28155

[snip]

Quoted 4 lines

Hey Olof,
So Christian mentioned you were hoping I'd pull these in through my
tree, but I'd really rather these go via the arch-soc tree, since that is more likely where they will be tested and more intelligently reviewed.

I've mentioned before my distaste for the drivers/clocksource directory becoming a dumping ground for arch specific (for the most part arm) clocksource, and more recently clockevent, drivers. I initially wanted the drivers/clocksource directory to be only for cross-arch clocksources, and I fear we'll end up with something like the drivers/rtc directory where its not entirely clear what hardware you actually need for a given driver (might as well be rtc-<sha1>.c ;).

That said, I realize I've been overruled on this. :)

But I'd really rather the patch flow for arch specific clocksource and clockevent drivers go through the arch tree, since there's a better chance folks will be building and testing against other arch specific changes that might cause problems. There's just no way for me to be able to intelligently review these sorts of changes.

Now, if there are changes to the clocksource core code, then I definitely want to be in the path there. Additionally meta-drivers like mmio, I should probably be more involved with, so they can be more properly integrated into the clocksource core. Also for drivers that are actually arch shared (ie: things like hpet/acpi_pm where ia64 and x86 share them).

But if its arch specific for hardware I don't have, I'd really prefer the arch maintainer be the upstream path.

Thomas: Am I being too obstinate here? If not, should we drop "F:
drivers/clocksource" from the MAINTAINERS entry?

Christian: Sorry for the confusion on all this!

thanks
-john

#6 lore raw

PATCH v7 1/2 ARM: bcm281xx: Add timer driver (driver portion)

Quoted 9 lines · John Stultz
Quoted 2 lines · Christian Daudt

This adds support for the Broadcom timer, used in the following SoCs:
BCM11130, BCM11140, BCM11351, BCM28145, BCM28155

[snip]

Quoted 4 lines
Quoted 8 lines

Hey Olof,
So Christian mentioned you were hoping I'd pull these in through my tree,
but I'd really rather these go via the arch-soc tree, since that is more likely where they will be tested and more intelligently reviewed.

I've mentioned before my distaste for the drivers/clocksource directory becoming a dumping ground for arch specific (for the most part arm) clocksource, and more recently clockevent, drivers. I initially wanted the

OTOH, if they are all in one place we have a better chance to analyze them in bulk, find similarities and patterns for consolidation. And it's simpler for the developer of a new SoC support to find the matching driver for his IP block instead of copying stuff from one mach directory to the next. We've been there :)

Quoted 5 lines

But I'd really rather the patch flow for arch specific clocksource and clockevent drivers go through the arch tree, since there's a better chance folks will be building and testing against other arch specific changes that might cause problems. There's just no way for me to be able to intelligently review these sorts of changes.

I agree with the build and test part, but you can still review the code in general - correctness of the particular hw details aside. I just opened a random one there and found:

static cycle_t vt8500_timer_read(struct clocksource *cs)

{
        int loops = msecs_to_loops(10);
        writel(3, regbase + TIMER_CTRL_VAL);
        while ((readl((regbase + TIMER_AS_VAL)) & TIMER_COUNT_R_ACTIVE)
                                                && --loops)
                cpu_relax();
        return readl(regbase + TIMER_COUNT_VAL);

You don't need particular hw knowledge to see that this looks more than fishy at least without comments describing the hardware designers brainfart.

Quoted 11 lines

Now, if there are changes to the clocksource core code, then I definitely want to be in the path there. Additionally meta-drivers like mmio, I should probably be more involved with, so they can be more properly integrated into the clocksource core. Also for drivers that are actually arch shared (ie:
things like hpet/acpi_pm where ia64 and x86 share them).

But if its arch specific for hardware I don't have, I'd really prefer the arch maintainer be the upstream path.

Thomas: Am I being too obstinate here? If not, should we drop "F:
drivers/clocksource" from the MAINTAINERS entry?

Maybe we should move the ARM specific ones into
drivers/clocksource/arm/ ?

Thanks,

tglx

#7 lore raw

PATCH v7 1/2 ARM: bcm281xx: Add timer driver (driver portion)

Quoted 5 lines · Thomas Gleixner
Quoted 5 lines

But if its arch specific for hardware I don't have, I'd really prefer the arch maintainer be the upstream path.

Thomas: Am I being too obstinate here? If not, should we drop "F:
drivers/clocksource" from the MAINTAINERS entry?

The idea of moving drivers out of arch/* into drivers/* is definitely to have someone who understands the subsystem act as the gatekeeper. There should be very little architecture specific knowledge in those drivers, and I think it's more important to ensure that they are following a sensible understanding of how timekeeping is done than that they are following specific architecture maintainer's preferences.

Quoted 2 lines

Maybe we should move the ARM specific ones into
drivers/clocksource/arm/ ?

About half the IP blocks we use on ARM are also used on at least one ARM64/AVR32/MIPS/PowerPC/x86/SH/Hexagon/c6x/etc part. Grouping them by which CPU architecture first starts using them or happens to be more popular at the time does not seem too helpful here.

Maybe it's better to have a subdirectory for those clock sources that are used on any SoC, or have subdirectories based on the company that created that part, as we do for ethernet drivers. I wouldn't bother with that until there are a couple of dozen different clock source drivers.

Arnd

#8 lore raw

PATCH v7 1/2 ARM: bcm281xx: Add timer driver (driver portion)

Quoted 22 lines · Arnd Bergmann
Quoted 5 lines · Thomas Gleixner
Quoted 5 lines

But if its arch specific for hardware I don't have, I'd really prefer the arch maintainer be the upstream path.

Thomas: Am I being too obstinate here? If not, should we drop "F:
drivers/clocksource" from the MAINTAINERS entry?

The idea of moving drivers out of arch/* into drivers/* is definitely to have someone who understands the subsystem act as the gatekeeper. There should be very little architecture specific knowledge in those drivers, and I think it's more important to ensure that they are following a sensible understanding of how timekeeping is done than that they are following specific architecture maintainer's preferences.

Quoted 2 lines

Maybe we should move the ARM specific ones into
drivers/clocksource/arm/ ?

About half the IP blocks we use on ARM are also used on at least one ARM64/AVR32/MIPS/PowerPC/x86/SH/Hexagon/c6x/etc part. Grouping them by which CPU architecture first starts using them or happens to be more popular at the time does not seem too helpful here.

Quoted 5 lines

Maybe it's better to have a subdirectory for those clock sources that are used on any SoC, or have subdirectories based on the company that created that part, as we do for ethernet drivers. I wouldn't bother with that until there are a couple of dozen different clock source drivers.

So it seems the (weak...) consensus is that it should go through the clocksource tree. John, can you please apply the patch ?

Thanks,
csd

#9 lore raw

PATCH v7 1/2 ARM: bcm281xx: Add timer driver (driver portion)

Quoted 5 lines · Christian Daudt
Quoted 4 lines · Arnd Bergmann
Quoted 3 lines · Thomas Gleixner
Quoted 3 lines

But if its arch specific for hardware I don't have, I'd really prefer the arch maintainer be the upstream path.

Quoted 8 lines
Quoted 8 lines
Quoted 2 lines
Quoted 2 lines

Thomas: Am I being too obstinate here? If not, should we drop "F:
drivers/clocksource" from the MAINTAINERS entry?

The idea of moving drivers out of arch/* into drivers/* is definitely to have someone who understands the subsystem act as the gatekeeper. There should be very little architecture specific knowledge in those drivers, and I think it's more important to ensure that they are following a sensible understanding of how timekeeping is done than that they are following specific architecture maintainer's preferences.

Sorry for the slow reply here.

Arnd: So I think I better understand the crux of the debate:

I'm unfamiliar with the hardware specifics, so I'm pushing the gatekeeping to the SoC maintainers, where as the SoC maintainer (dealing with tons of different SoC's), you're also unfamiliar with the hardware specifics, and would like to push the gatekeeping of the subsystem specific functionality to those maintainers.

Given that for each SoC there's tons of subsystem specific functionality that you're having to deal with, I can sympathize with your need to offload some of the gatekeeping.

Quoted 7 lines
Quoted 7 lines
Quoted 2 lines

Maybe we should move the ARM specific ones into
drivers/clocksource/arm/ ?

About half the IP blocks we use on ARM are also used on at least one ARM64/AVR32/MIPS/PowerPC/x86/SH/Hexagon/c6x/etc part. Grouping them by which CPU architecture first starts using them or happens to be more popular at the time does not seem too helpful here.

Quoted 5 lines
Quoted 5 lines

Maybe it's better to have a subdirectory for those clock sources that are used on any SoC, or have subdirectories based on the company that created that part, as we do for ethernet drivers. I wouldn't bother with that until there are a couple of dozen different clock source drivers.

So having had a few days to think about this, I think what usually rubs me the wrong way when I get driver/clocksource submissions, is that for 99% of it, they *aren't clocksource drivers*. Most of the code is *clockevent* driver logic, and then maybe 1-5 lines of actual clocksource code.

Now, I know the reason for this is often the clocksource and clockevent drivers are backed by the same hardware, and since there's no clockevent directory, might as well have it all in a single file somewhere. But mixing the different subsystem drivers together causes some of the maintenance confusion here.

So instead of creating drivers/clocksource/arch/ directories, what I'd propose is we create a drivers/clockevent directory to handle the actual clockevent code. I think this would better delineate the lines of responsibility on the gatekeeper side (that being Thomas or maybe someone else who has an interest in the subtleties of how various hardware timers are be broken-by-design ;), and I'd be much happier taking clocksource code where I felt I had a reasonable chance of noticing bugs.

Thomas: Not that you need more to maintain, but does this seem semi-reasonable? Do we need to find someone else to help here?

That said, at the end of the day, if I take a bad drivers/clocksource patch, what breaks won't be the timekeeping core, it will be an SoC board. So I'll have to really rely on the original clocksource driver authors to help vet incoming patches. This is where I think having the SoC tree as a central point for SoC patches has and advantage, as its less likely breakage will sneak upstream via a subsystem tree. But understanding the need for review help, I think I'm ok with taking on more clocksource specific review.

Quoted 2 lines

So it seems the (weak...) consensus is that it should go through the clocksource tree. John, can you please apply the patch ?

Christian: So thanks again for the ping here. So I think I'd prefer Thomas to be the one to apply these sorts (ie: clockevent) of patches in the future, but this time around I'll do it because its been unfair to make you the monkey-in-the-middle as we've tossed the responsibility around (and my stuff goes through Thomas anyway).

thanks
-john

#10 lore raw

PATCH v7 1/2 ARM: bcm281xx: Add timer driver (driver portion)

Quoted 27 lines · John Stultz
Quoted 7 lines · Christian Daudt
Quoted 7 lines
Quoted 2 lines

Maybe we should move the ARM specific ones into
drivers/clocksource/arm/ ?

About half the IP blocks we use on ARM are also used on at least one ARM64/AVR32/MIPS/PowerPC/x86/SH/Hexagon/c6x/etc part. Grouping them by which CPU architecture first starts using them or happens to be more popular at the time does not seem too helpful here.

Quoted 5 lines
Quoted 5 lines

Maybe it's better to have a subdirectory for those clock sources that are used on any SoC, or have subdirectories based on the company that created that part, as we do for ethernet drivers. I wouldn't bother with that until there are a couple of dozen different clock source drivers.

So having had a few days to think about this, I think what usually rubs me the wrong way when I get driver/clocksource submissions, is that for 99% of it, they *aren't clocksource drivers*. Most of the code is *clockevent* driver logic, and then maybe 1-5 lines of actual clocksource code.

Now, I know the reason for this is often the clocksource and clockevent drivers are backed by the same hardware, and since there's no clockevent directory, might as well have it all in a single file somewhere. But mixing the different subsystem drivers together causes some of the maintenance confusion here.

Quoted 8 lines

So instead of creating drivers/clocksource/arch/ directories, what I'd propose is we create a drivers/clockevent directory to handle the actual clockevent code. I think this would better delineate the lines of responsibility on the gatekeeper side (that being Thomas or maybe someone else who has an interest in the subtleties of how various hardware timers are be broken-by-design ;), and I'd be much happier taking clocksource code where I felt I had a reasonable chance of noticing bugs.

Yes, this sounds like a good idea.

Quoted 11 lines

Thomas: Not that you need more to maintain, but does this seem semi-reasonable? Do we need to find someone else to help here?

That said, at the end of the day, if I take a bad drivers/clocksource patch, what breaks won't be the timekeeping core, it will be an SoC board. So I'll have to really rely on the original clocksource driver authors to help vet incoming patches. This is where I think having the SoC tree as a central point for SoC patches has and advantage, as its less likely breakage will sneak upstream via a subsystem tree. But understanding the need for review help, I think I'm ok with taking on more clocksource specific review.

I think the important point for you to realize is that you don't need to worry about breaking a specific SoC. The patches will come from someone who is using that hardware in the end. What we need from clock{source,event} subsystem maintainers is perspective of how things fit into the subsystem and to provide review like:
- this driver is not using CLOCKSOURCE_OF_DECLARE when it should
- that driver only handles clockevent, don't put it into drivers/clocksource
- I don't like your coding style, it doesn't fit in with the other drivers in this subsystem.
- what you do can be implemented much simpler using the generic infrastructure introduced by that earlier driver.

Arnd

Applied
#11 lore raw

PATCH v7 1/2 ARM: bcm281xx: Add timer driver (driver portion)

In reply to Christian Daudt
Quoted 28 lines · Christian Daudt

This adds support for the Broadcom timer, used in the following SoCs:
BCM11130, BCM11140, BCM11351, BCM28145, BCM28155

Updates from V6:
- Split DT portion into a separate patch

Updates from V5:
- Rebase to latest arm-soc/for-next

Updates from V4:
- Switch code to use CLOCKSOURCE_OF_DECLARE

Updates from V3:
- Migrate to 3.9 timer framework updates

Updates from V2:
- prepend static fns + fields with kona_

Updates from V1:
- Rename bcm_timer.c to bcm_kona_timer.c
- Pull .h into bcm_kona_timer.c
- Make timers static
- Clean up comment block
- Switched to using clockevents_config_and_register
- Added an error to the get_timer loop if it repeats too much
- Added to Documentation/devicetree/bindings/arm/bcm/bcm,kona-timer.txt
- Added missing readl to timer_disable_and_clear

Quoted 8 lines

Note: bcm,kona-timer was kept as the 'compatible' field to make it specific enough for when there are multiple bcm timers (bcm,timer is too generic).

Applied to my fortglx/3.10/time branch.

thanks
-john