On 05/10/2017 03:31 AM, Phil Elwell wrote: > On 10/05/2017 11:09, Marc Zyngier wrote: >> On 10/05/17 10:05, Phil Elwell wrote: >>> On 10/05/2017 09:55, Marc Zyngier wrote: >>>> On Wed, May 10 2017 at 9:27:10 am BST, Phil Elwell <p...@raspberrypi.org> >>>> wrote: >>>>> On 10/05/2017 08:42, Marc Zyngier wrote: >>>>>> On 09/05/17 20:02, Phil Elwell wrote: >>>>>>> On 09/05/2017 19:53, Marc Zyngier wrote: >>>>>>>> On 09/05/17 19:52, Phil Elwell wrote: >>>>>>>>> On 09/05/2017 19:14, Marc Zyngier wrote: >>>>>>>>>> On 09/05/17 19:08, Eric Anholt wrote: >>>>>>>>>>> Marc Zyngier <marc.zyng...@arm.com> writes: >>>>>>>>>>> >>>>>>>>>>>> On 09/05/17 17:59, Eric Anholt wrote: >>>>>>>>>>>>> Phil Elwell <p...@raspberrypi.org> writes: >>>>>>>>>>>>> >>>>>>>>>>>>>> In order to reduce power consumption and bus traffic, it is >>>>>>>>>>>>>> sensible >>>>>>>>>>>>>> for secondary cores to enter a low-power idle state when waiting >>>>>>>>>>>>>> to >>>>>>>>>>>>>> be started. The wfe instruction causes a core to wait until an >>>>>>>>>>>>>> event >>>>>>>>>>>>>> or interrupt arrives before continuing to the next instruction. >>>>>>>>>>>>>> The sev instruction sends a wakeup event to the other cores, so >>>>>>>>>>>>>> call >>>>>>>>>>>>>> it from bcm2836_smp_boot_secondary, the function that wakes up >>>>>>>>>>>>>> the >>>>>>>>>>>>>> waiting cores during booting. >>>>>>>>>>>>>> >>>>>>>>>>>>>> It is harmless to use this patch without the corresponding change >>>>>>>>>>>>>> adding wfe to the ARMv7/ARMv8-32 stubs, but if the stubs are >>>>>>>>>>>>>> updated >>>>>>>>>>>>>> and this patch is not applied then the other cores will sleep >>>>>>>>>>>>>> forever. >>>>>>>>>>>>>> >>>>>>>>>>>>>> See: https://github.com/raspberrypi/linux/issues/1989 >>>>>>>>>>>>>> >>>>>>>>>>>>>> Signed-off-by: Phil Elwell <p...@raspberrypi.org> >>>>>>>>>>>>>> --- >>>>>>>>>>>>>> drivers/irqchip/irq-bcm2836.c | 3 +++ >>>>>>>>>>>>>> 1 file changed, 3 insertions(+) >>>>>>>>>>>>>> >>>>>>>>>>>>>> diff --git a/drivers/irqchip/irq-bcm2836.c >>>>>>>>>>>>>> b/drivers/irqchip/irq-bcm2836.c >>>>>>>>>>>>>> index e10597c..6dccdf9 100644 >>>>>>>>>>>>>> --- a/drivers/irqchip/irq-bcm2836.c >>>>>>>>>>>>>> +++ b/drivers/irqchip/irq-bcm2836.c >>>>>>>>>>>>>> @@ -248,6 +248,9 @@ static int __init >>>>>>>>>>>>>> bcm2836_smp_boot_secondary(unsigned int cpu, >>>>>>>>>>>>>> writel(secondary_startup_phys, >>>>>>>>>>>>>> intc.base + LOCAL_MAILBOX3_SET0 + 16 * cpu); >>>>>>>>>>>>>> >>>>>>>>>>>>>> + dsb(sy); /* Ensure write has completed before waking >>>>>>>>>>>>>> the other CPUs */ >>>>>>>>>>>>>> + sev(); >>>>>>>>>>>>>> + >>>>>>>>>>>>>> return 0; >>>>>>>>>>>>>> } >>>>>>>>>>>>> >>>>>>>>>>>>> This is also the behavior that the standard arm64 spin-table >>>>>>>>>>>>> method has, >>>>>>>>>>>>> which we unfortunately can't quite use. >>>>>>>>>>>> >>>>>>>>>>>> And why is that so? Why do you have to reinvent the wheel (and >>>>>>>>>>>> hide the >>>>>>>>>>>> cloned wheel in an interrupt controller driver)? >>>>>>>>>>>> >>>>>>>>>>>> That doesn't seem right to me. >>>>>>>>>>> >>>>>>>>>>> The armv8 stubs (firmware-supplied code in the low page that do the >>>>>>>>>>> spinning) do actually implement arm64's spin-table method. It's the >>>>>>>>>>> armv7 stubs that use these registers in the irqchip instead of plain >>>>>>>>>>> addresses in system memory. >>>>>>>>>> >>>>>>>>>> Let's put ARMv7 aside for the time being. If your firmware already >>>>>>>>>> implements spin-tables, why don't you simply use that at least on >>>>>>>>>> arm64? >>>>>>>>> >>>>>>>>> We do. >>>>>>>> >>>>>>>> Obviously not the way it is intended if you have to duplicate the core >>>>>>>> architectural code in the interrupt controller driver, which couldn't >>>>>>>> care less. >>>>>>> >>>>>>> If we were using this method on arm64 then the other cores would not >>>>>>> start up >>>>>>> because armstub8.S has always included a wfe. Nothing in the commit >>>>>>> mentions >>>>>>> arm64 - this is an ARCH=arm fix. >>>>>> >>>>>> Thanks for the clarification, which you could have added to the commit >>>>>> message. >>>>>> >>>>>> The question still remains: why do we have CPU bring-up code in an >>>>>> interrupt controller, instead of having it in the architecture code? >>>>>> >>>>>> The RPi-2 is the *only* platform to have its SMP bringup code outside of >>>>>> arch/arm, so the first course of action would be to move that code where >>>>>> it belongs. >>>>> >>>>> You were CC'd on the commit (41f4988cc287e5f836d3f6620c9f900bc9b560e9) >>>>> that >>>>> introduced bcm2836_smp_boot_secondary - it seems strange to start >>>>> objecting >>>>> now. >>>> >>>> Well, I'm far from being perfect. If I had noticed it, I'd have NACKed >>>> it. >>>> >>>>> Yes, I think it is odd that it didn't go into arch/arm/mach-bcm, but in >>>>> the interests of making changes in small, independent steps, do you have a >>>>> problem with this commit? >>>> >>>> On its own, no. I'm just not keen on adding more unrelated stuff to this >>>> file, so let's start with dealing with the original bug, and you can >>>> then add this fix on top. >>> >>> That's an interesting use of the word "bug". From Wikipedia: >>> >>> "A software bug is an error, flaw, failure or fault in a computer program or >>> system that causes it to produce an incorrect or unexpected result, or to >>> behave in unintended ways." >> >> Whatever. Should I call it "pile of crap dumped in unsuitable locations" >> instead? What does Wikipedia says about it? >> >>> Although your concerns are valid, the faults you are objecting to are not >>> causing >>> a malfunction of any kind. If we were to update the RPi firmware before this >>> patch was merged then upstream users would be left with one wheel on their >>> wagon. >> >> And that'd be your problem, not mine. Look, you can argue around this >> all day, or you can fix this mess. Your choice. > > Is that the opinion of all here?
The choice of word here got largely out of the original topic and I surely did eat a ton of popcorn here. There are two things that need fixing, and the time line and process for fixing these is clear: - your bugfix (Phil) is something that should be applied now, and backported to -stable trees once the fix hits the irqchip tree (or Linus') - relocating the code that does the secondary boot out of drivers/irqchip/ into arch/arm/mach-bcm/ needs to happen (Marc), and this is 4.13 material, there is no urgency in doing this *right now*, but it needs to happen Does that work for everyone? -- Florian