On 26/10/16 08:18, Robert Richter wrote: > On 21.10.16 13:52:43, Marc Zyngier wrote: >> On 21/10/16 13:07, Andrew Jones wrote: >>> On Fri, Oct 21, 2016 at 12:57:37PM +0100, Peter Maydell wrote: >>>> On 21 October 2016 at 12:49, Andrew Jones <drjones@xxxxxxxxxx> wrote: >>>>> I also read the register before writing it and saw it was 3. I tried >>>>> writing 3 instead of 0 to see what would happen, but the failure >>>>> persisted. I did read back the register after writing it to confirm the >>>>> change took affect. >>>> >>>> So what does it read back as after you write 0? The GICv3 spec >>>> says it can't read back as zero... >>>> >>> >>> I read back zero >>> >>> pre-read bpr1=3 >>> post-read bpr1=0 >>> FAIL: gicv3: ipi: self: Timed-out (5s). ACKS: missing=1 extra=0 unexpected=0 >> >> Gah... I guess we'll have to either roll a dice, or get someone from >> Cavium to tell us what's happening here. In the meantime, can you give >> the following patch a go? It doesn't fire on my FSL box, but everything >> hunky dory on it so far... > > "The minimum binary value programmable is BP_MIN. Attempt to > set it to a lower value automatically sets it to the minimum value > BP_MIN." > > The min value is 3 in this level, thus it should return 3 for [2:0] of > BPR1. Exactly (though there is a number of other rules such as taking ICC_CTLR_EL1.CBPR into account). And all this should reflect in VMCR_EL2.VBPR1. > I am going to reproduce this. Thanks, M. -- Jazz is not dead. It just smells funny... -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
