Re: [v2,3/9] powerpc/powernv: Add icp_native_cause_ipi_rm

2015-12-06 Thread Michael Ellerman
Subject would be better as "powerpc/xics". On Wed, 2015-25-11 at 23:44:50 UTC, "Suresh E. Warrier" wrote: > Function to cause an IPI. Requires kvm_hstate.xics_phys > to be initialized with physical address of XICS. Please expand the change log a bit, this is a bit terse. > diff --git a/arch/powe

Re: [v2,2/9] powerpc/smp: Add smp_muxed_ipi_set_message

2015-12-06 Thread Michael Ellerman
On Wed, 2015-25-11 at 23:44:49 UTC, "Suresh E. Warrier" wrote: > smp_muxed_ipi_message_pass() invokes smp_ops->cause_ipi, which > updates the MFFR through an ioremapped address, to cause the > IPI. Because of this real mode callers cannot call > smp_muxed_ipi_message_pass() for IPI messaging. You'

Re: [PATCH v7 45/50] drivers/of: Avoid recursively calling unflatten_dt_node()

2015-12-06 Thread Guenter Roeck
On 12/06/2015 06:33 PM, Rob Herring wrote: On Sun, Dec 6, 2015 at 8:21 PM, Guenter Roeck wrote: On 12/06/2015 03:54 PM, Benjamin Herrenschmidt wrote: On Sun, 2015-12-06 at 14:28 -0600, Rob Herring wrote: Do you plan to respin the OF parts at least soon? There's another problem Guenter foun

Re: [PATCH v7 45/50] drivers/of: Avoid recursively calling unflatten_dt_node()

2015-12-06 Thread Rob Herring
On Sun, Dec 6, 2015 at 8:21 PM, Guenter Roeck wrote: > On 12/06/2015 03:54 PM, Benjamin Herrenschmidt wrote: >> >> On Sun, 2015-12-06 at 14:28 -0600, Rob Herring wrote: >>> >>> >>> Do you plan to respin the OF parts at least soon? There's another >>> problem Guenter found that of_fdt_unflatten_tre

Re: [PATCH v7 45/50] drivers/of: Avoid recursively calling unflatten_dt_node()

2015-12-06 Thread Guenter Roeck
On 12/06/2015 03:54 PM, Benjamin Herrenschmidt wrote: On Sun, 2015-12-06 at 14:28 -0600, Rob Herring wrote: Do you plan to respin the OF parts at least soon? There's another problem Guenter found that of_fdt_unflatten_tree is not re-entrant due to "depth" being static and this series fixes that

[PATCH] powernv/opal-irqchip: Fix double endian conversion

2015-12-06 Thread Alistair Popple
The OPAL event calls return a mask of events that are active in big endian format. This is checked when unmasking the events in the irqchip by comparison with a cached value. The cached value was stored in big endian format but should've been converted to CPU endian first. This bug leads to OPAL ev

Re: more POLL... fun

2015-12-06 Thread Michael Ellerman
On Fri, 2015-12-04 at 17:30 +, Al Viro wrote: > On Fri, Dec 04, 2015 at 06:13:54PM +0100, Arnd Bergmann wrote: > > Yes, I think that would be best. Do you want me to send that patch, or do > > you prefer to do it yourself? In theory that patch should also go into > > stable > > kernels, but I

Re: [PATCH v7 45/50] drivers/of: Avoid recursively calling unflatten_dt_node()

2015-12-06 Thread Benjamin Herrenschmidt
On Sun, 2015-12-06 at 14:28 -0600, Rob Herring wrote: > > Do you plan to respin the OF parts at least soon? There's another > problem Guenter found that of_fdt_unflatten_tree is not re-entrant due > to "depth" being static and this series fixes that. So I'd rather > apply this and avoid adding a m

[PATCH v2] selftests/powerpc: Add script to test HMI functionality

2015-12-06 Thread Daniel Axtens
HMIs (Hypervisor Management|Maintenance Interrupts) are a class of interrupt on POWER systems. HMI support has traditionally been exceptionally difficult to test, however Skiboot ships a tool that, with the correct magic numbers, will inject them. This, therefore, is a first pass at a script to i

Re: [PATCH] of: Fix comparison of reserved memory regions

2015-12-06 Thread Michael Ellerman
On Sun, 2015-12-06 at 14:31 -0600, Rob Herring wrote: > On Sat, Dec 5, 2015 at 5:43 AM, Michael Ellerman > wrote: > > On 5 December 2015 04:07:39 GMT+11:00, Mitchel Humpherys > > wrote: > > > On Wed, Nov 18 2015 at 09:46:38 PM, Michael Ellerman > > > wrote: > > > > Fix it by just doing the sta

Re: crypto/nx842: Ignore queue overflow informative error

2015-12-06 Thread Daniel Axtens
Haren Myneni writes: > NX842 coprocessor sets bit 3 if queue is overflow. It is just for > information to the user. So the driver prints this informative message > and ignores it. What queue, and what happens when the queue overflows? It seems like *something* would need to be done, somewhere, b

Re: [PATCH v7 45/50] drivers/of: Avoid recursively calling unflatten_dt_node()

2015-12-06 Thread Guenter Roeck
On 12/06/2015 12:28 PM, Rob Herring wrote: +Guenter On Wed, Nov 4, 2015 at 7:12 AM, Gavin Shan wrote: In current implementation, unflatten_dt_node() is called recursively to unflatten device nodes in FDT blob. It's stress to limited stack capacity. This avoids calling the function recursively

Re: [PATCH] of: Fix comparison of reserved memory regions

2015-12-06 Thread Rob Herring
On Sat, Dec 5, 2015 at 5:43 AM, Michael Ellerman wrote: > > > On 5 December 2015 04:07:39 GMT+11:00, Mitchel Humpherys > wrote: >>On Wed, Nov 18 2015 at 09:46:38 PM, Michael Ellerman >> wrote: >>> In order to check for overlapping reserved memory regions, we first >>need >>> to sort the array of

Re: [PATCH v7 45/50] drivers/of: Avoid recursively calling unflatten_dt_node()

2015-12-06 Thread Rob Herring
+Guenter On Wed, Nov 4, 2015 at 7:12 AM, Gavin Shan wrote: > In current implementation, unflatten_dt_node() is called recursively > to unflatten device nodes in FDT blob. It's stress to limited stack > capacity. > > This avoids calling the function recursively, meaning the device > nodes are unfl