On 21/03/13 11:06, Mark Rutland wrote: > Hi Rob, > > (adding Marc to Cc as he may have comments). > > On Wed, Mar 20, 2013 at 10:34:35PM +0000, Rob Herring wrote: >> From: Rob Herring <rob.herring@xxxxxxxxxxx> >> >> This converts arm and arm64 to use CLKSRC_OF DT based initialization for >> the arch timer. A new function arch_timer_arch_init is added to allow for >> arch specific setup. >> >> This has a side effect of enabling sched_clock on omap5 and exynos5. There >> should not be any reason not to use the arch timers for sched_clock. > > Nice! I was just about to post a (slightly updated) version of Thomas Abraham's > arch_timer clocksource_of_init patch, but this seems much more comprehensive. > > I have some other arch_timer patches which may clash, but they could be rebased > atop of this. > >> >> Signed-off-by: Rob Herring <rob.herring@xxxxxxxxxxx> >> Cc: Russell King <linux@xxxxxxxxxxxxxxxx> >> Cc: Kukjin Kim <kgene.kim@xxxxxxxxxxx> >> Cc: Tony Lindgren <tony@xxxxxxxxxxx> >> Cc: Simon Horman <horms@xxxxxxxxxxxx> >> Cc: Magnus Damm <magnus.damm@xxxxxxxxx> >> Cc: Catalin Marinas <catalin.marinas@xxxxxxx> >> Cc: Will Deacon <will.deacon@xxxxxxx> >> Cc: John Stultz <john.stultz@xxxxxxxxxx> >> Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx> >> Cc: linux-samsung-soc@xxxxxxxxxxxxxxx >> Cc: linux-omap@xxxxxxxxxxxxxxx >> Cc: linux-sh@xxxxxxxxxxxxxxx >> --- >> This is dependent on my CLKSRC_OF clean-up in arm-soc, my 64-bit sched_clock >> support series, and Arnd's default machine descriptor patch (for default >> clocksource_of_init call). This is only compile tested on arm. >> >> The full series (including sp804 work) is available here: >> git://sources.calxeda.com/kernel/linux.git arm-timers >> >> Rob >> > > [...] > >> diff --git a/arch/arm/mach-vexpress/v2m.c b/arch/arm/mach-vexpress/v2m.c >> index d0ad789..6215717 100644 >> --- a/arch/arm/mach-vexpress/v2m.c >> +++ b/arch/arm/mach-vexpress/v2m.c >> @@ -1,6 +1,7 @@ >> /* >> * Versatile Express V2M Motherboard Support >> */ >> +#include <linux/clocksource.h> >> #include <linux/device.h> >> #include <linux/amba/bus.h> >> #include <linux/amba/mmci.h> >> @@ -23,7 +24,6 @@ >> #include <linux/regulator/machine.h> >> #include <linux/vexpress.h> >> >> -#include <asm/arch_timer.h> >> #include <asm/mach-types.h> >> #include <asm/sizes.h> >> #include <asm/mach/arch.h> >> @@ -446,10 +446,7 @@ static void __init v2m_dt_timer_init(void) >> irq_of_parse_and_map(node, 0)); >> } >> >> - arch_timer_of_register(); >> - >> - if (arch_timer_sched_clock_init() != 0) >> - versatile_sched_clock_init(vexpress_get_24mhz_clock_base(), >> + versatile_sched_clock_init(vexpress_get_24mhz_clock_base(), >> 24000000); >> } >> > > On TC2 this series leads to using the vexpress 24MHz clock as the sched clock > in preference to the architected timer: > > Architected local timer running at 24.00MHz (virt). > Switching to timer-based delay loop > Registered arch_counter_get_cntvct+0x0/0x14 as sched_clock source > sched_clock: 32 bits at 24MHz, resolution 41ns, wraps every 178956ms > Registered versatile_read_sched_clock+0x0/0x28 as sched_clock source > > As they both have the same frequency, neither overrides the other, and > whichever gets registered last is used as the sched_clock. As accesses to the > architected timer are going to have a much lower overhead, this isn't very nice > (and it could be better to use it even if it had a lower frequency). Indeed. And if I look at it with my KVM hat on, it is even worse (the guest will exit all the way to platform emulation in userspace on each sched_clock read - a sure performance killer). Of course, emulating a VE is not the best option, but that's all we have so far when using QEMU as platform emulation. M. -- Jazz is not dead. It just smells funny... -- To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
