[PATCH 0/2] arm64: Add PMU node for APM X-Gene Storm SoC

[PATCH 0/2] arm64: Add PMU node for APM X-Gene Storm SoC

[Vinayak Kale]

This patch series adds PMU node for APM X-Gene's Potenza CPU.
Potenza CPU PMU is compatible with ARMv8-PMUv3.

Vinayak Kale (2):
  dt-bindings: Add Potenza PMU binding
  arm64: dts: Add PMU node for APM X-Gene Storm SOC

 Documentation/devicetree/bindings/arm/pmu.txt |    1 +
 arch/arm64/boot/dts/apm-storm.dtsi            |    5 +++++
 2 files changed, 6 insertions(+)

-- 
1.7.9.5

[PATCH 1/2] dt-bindings: Add Potenza PMU binding

[Vinayak Kale]

This patch documents the compatible string for APM X-Gene Potenza CPU's PMU.

Signed-off-by: Vinayak Kale <vkale@apm.com>
---
 Documentation/devicetree/bindings/arm/pmu.txt |    1 +
 1 file changed, 1 insertion(+)

diff --git a/Documentation/devicetree/bindings/arm/pmu.txt b/Documentation/devicetree/bindings/arm/pmu.txt
index 3e1e498..0246f81 100644
--- a/Documentation/devicetree/bindings/arm/pmu.txt
+++ b/Documentation/devicetree/bindings/arm/pmu.txt
@@ -7,6 +7,7 @@ representation in the device tree should be done as under:-
 Required properties:
 
 - compatible : should be one of
+	"apm,potenza-pmu"
 	"arm,armv8-pmuv3"
 	"arm,cortex-a15-pmu"
 	"arm,cortex-a9-pmu"
-- 
1.7.9.5

[PATCH 2/2] arm64: dts: Add PMU node for APM X-Gene Storm SOC

[Vinayak Kale]

This patch adds the PMU device tree node for APM X-Gene Storm SOC.

Please note that this patch has dependancy on a GIC driver patch [1] which is
yet to be approved by maintainers.

[1]- https://lkml.org/lkml/2014/2/27/605
(irqchip:gic: change access of gicc_ctrl register to read modify write)

Signed-off-by: Vinayak Kale <vkale@apm.com>
---
 arch/arm64/boot/dts/apm-storm.dtsi |    5 +++++
 1 file changed, 5 insertions(+)

diff --git a/arch/arm64/boot/dts/apm-storm.dtsi b/arch/arm64/boot/dts/apm-storm.dtsi
index d37d736..5edb16a 100644
--- a/arch/arm64/boot/dts/apm-storm.dtsi
+++ b/arch/arm64/boot/dts/apm-storm.dtsi
@@ -97,6 +97,11 @@
 		clock-frequency = <50000000>;
 	};
 
+	pmu {
+		compatible = "apm,potenza-pmu", "arm,armv8-pmuv3";
+		interrupts = <1 12 0xff04>;
+	};
+
 	soc {
 		compatible = "simple-bus";
 		#address-cells = <2>;
-- 
1.7.9.5

Re: [PATCH 2/2] arm64: dts: Add PMU node for APM X-Gene Storm SOC

[Ming Lei]

Hi Vinayak,

On Thu, Mar 20, 2014 at 5:47 PM, Vinayak Kale <vkale-qTEPVZfXA3Y@public.gmane.org> wrote:
> This patch adds the PMU device tree node for APM X-Gene Storm SOC.
>
> Please note that this patch has dependancy on a GIC driver patch [1] which is
> yet to be approved by maintainers.
>
> [1]- https://lkml.org/lkml/2014/2/27/605
> (irqchip:gic: change access of gicc_ctrl register to read modify write)
>
> Signed-off-by: Vinayak Kale <vkale-qTEPVZfXA3Y@public.gmane.org>
> ---
>  arch/arm64/boot/dts/apm-storm.dtsi |    5 +++++
>  1 file changed, 5 insertions(+)
>
> diff --git a/arch/arm64/boot/dts/apm-storm.dtsi b/arch/arm64/boot/dts/apm-storm.dtsi
> index d37d736..5edb16a 100644
> --- a/arch/arm64/boot/dts/apm-storm.dtsi
> +++ b/arch/arm64/boot/dts/apm-storm.dtsi
> @@ -97,6 +97,11 @@
>                 clock-frequency = <50000000>;
>         };
>
> +       pmu {
> +               compatible = "apm,potenza-pmu", "arm,armv8-pmuv3";
> +               interrupts = <1 12 0xff04>;
> +       };
> +
>         soc {
>                 compatible = "simple-bus";
>                 #address-cells = <2>;
> --
> 1.7.9.5

No matter if Feng Kan's gic patch is applied or not, perf(PMU)
still can't work well with your patch, and the failure[1] will be
triggered when 'perf top' is started.

The failure happened during the 1st pmu overflow interrupt.

I test several mustang boards, and all has the problem with
upstream kernel.

Could you take a look the issue?

[1], perf segmentation log
[   98.862963] Bad mode in FIQ handler detected, code 0x92000007
[   98.868688] CPU: 2 PID: 3979 Comm: perf Not tainted 3.15.0-next-20140610+ #26
[   98.875787] task: ffffffc3eec4c080 ti: ffffffc3ea27c000 task.ti:
ffffffc3ea27c000
[   98.883232] PC is at 0x7fb177f3ec
[   98.886527] LR is at 0x45f280
[   98.889478] pc : [<0000007fb177f3ec>] lr : [<000000000045f280>]
pstate: 80000000
[   98.896834] sp : ffffffc3ea27fff0
[   98.900130] x29: 0000007ff2519ba0 x28: 0000000000521000
[   98.905434] x27: 0000000026dced90 x26: 0000000000000000
[   98.910738] x25: 0000000026dcd830 x24: 0000007ff251ada0
[   98.916041] x23: 0000000000000003 x22: 000000000000275a
[   98.921344] x21: 0000000026dcd390 x20: 0000000000011f0d
[   98.926648] x19: 0000000000011f0d x18: 0000007ff2519910
[   98.931952] x17: 0000007fb177f3a8 x16: 0000000000519000
[   98.937255] x15: 0000007fb11ce5a0 x14: 6632383931646363
[   98.942560] x13: 0000000001000000 x12: 0000000000000004
[   98.947865] x11: 0000000000000002 x10: 0101010101010101
[   98.953169] x9 : 0000000007cb19c0 x8 : 7f7f7f7f7f7f7f7f
[   98.958474] x7 : 0000007f96a1cc33 x6 : 3a043a1c11766963
[   98.963779] x5 : ffffffc0009b2648 x4 : 000000000000275a
[   98.969083] x3 : 000000000011f0d0 x2 : 0000007ff2519c90
[   98.974387] x1 : 000000000000275a x0 : 0000000026dced90
[   98.979692]
[   98.981177] Bad mode in Synchronous Abort handler detected, code 0x8600000f
[   98.988103] CPU: 2 PID: 3979 Comm: perf Not tainted 3.15.0-next-20140610+ #26
[   98.995200] task: ffffffc3eec4c080 ti: ffffffc3ea27c000 task.ti:
ffffffc3ea27c000
[   99.002642] PC is at 0x45f280
[   99.005592] LR is at 0x45f280
[   99.008542] pc : [<000000000045f280>] lr : [<000000000045f280>]
pstate: 600003c5
[   99.015897] sp : ffffffc3ea27fed0
[   99.019192] x29: 0000007ff2519ba0 x28: 0000000000521000
[   99.024495] x27: 0000000026dced90 x26: 0000000000000000
[   99.029799] x25: 0000000026dcd830 x24: 0000007ff251ada0
[   99.035103] x23: 0000000080000000 x22: 0000007fb177f3ec
[   99.040407] x21: ffffffc3ea27fff0 x20: 0000000000011f0d
[   99.045711] x19: 0000000000011f0d x18: 0000007ff2519910
[   99.051016] x17: 0000007fb177f3a8 x16: 0000000000519000
[   99.056320] x15: 0000007fb11ce5a0 x14: 3030303030303038
[   99.061623] x13: 203a657461747370 x12: 205d3e3038326635
[   99.066927] x11: 0000000000000002 x10: 0000000000000420
[   99.072232] x9 : 0000000000000421 x8 : 0000000000000006
[   99.077537] x7 : ffffffc0000812cc x6 : ffffffc3ea27fea0
[   99.082842] x5 : 00000000000003c0 x4 : 0000000000000001
[   99.088147] x3 : ffffffc3ea27fcf0 x2 : 00000000dead4ead
[   99.093452] x1 : 000000000000005e x0 : 0000000000000000
[   99.098757]
[   99.100240] Internal error: Oops - bad mode: 0 [#1] SMP
[   99.105436] Modules linked in:
[   99.108483] CPU: 2 PID: 3979 Comm: perf Not tainted 3.15.0-next-20140610+ #26
[   99.115581] task: ffffffc3eec4c080 ti: ffffffc3ea27c000 task.ti:
ffffffc3ea27c000
[   99.123024] PC is at 0x45f280
[   99.125974] LR is at 0x45f280
[   99.128923] pc : [<000000000045f280>] lr : [<000000000045f280>]
pstate: 600003c5
[   99.136279] sp : ffffffc3ea27fed0
[   99.139574] x29: 0000007ff2519ba0 x28: 0000000000521000
[   99.144875] x27: 0000000026dced90 x26: 0000000000000000
[   99.150179] x25: 0000000026dcd830 x24: 0000007ff251ada0
[   99.155483] x23: 0000000080000000 x22: 0000007fb177f3ec
[   99.160788] x21: ffffffc3ea27fff0 x20: 0000000000011f0d
[   99.166092] x19: 0000000000011f0d x18: 0000007ff2519910
[   99.171397] x17: 0000007fb177f3a8 x16: 0000000000519000
[   99.176702] x15: 0000007fb11ce5a0 x14: 3030303030303038
[   99.182007] x13: 203a657461747370 x12: 205d3e3038326635
[   99.187312] x11: 0000000000000002 x10: 0000000000000420
[   99.192616] x9 : 0000000000000421 x8 : 0000000000000006
[   99.197922] x7 : ffffffc0000812cc [   99.198976] Bad mode in FIQ
handler detected, code 0x56000000
[   99.198980] CPU: 1 PID: 1891 Comm: mongod Not tainted
3.15.0-next-20140610+ #26
[   99.198982] task: ffffffc3e9c12040 ti: ffffffc3ea200000 task.ti:
ffffffc3ea200000
[   99.198984] PC is at 0x7f93059410
[   99.198985] LR is at 0x90e120
[   99.198986] pc : [<0000007f93059410>] lr : [<000000000090e120>]
pstate: 60000000
[   99.198987] sp : ffffffc3ea203ff0
[   99.198988] x29: 0000007f72ffe2b0 x28: 0000000000000001
[   99.198990] x27: 0000007f72ffe3a0 x26: 000000000000000a
[   99.198992] x25: 0000007f600019d4 x24: 0000007f72ffe520
[   99.198994] x23: 0000007f72ffe440 x22: 0000000000000400
[   99.198996] x21: 0000007f72ffe448 x20: 0000007f600019d4
[   99.198998] x19: 0000007f72ffe3c0 x18: 0000007f72ffe010
[   99.199000] x17: 0000007f93059410 x16: 0000000000b9c078
[   99.199001] x15: 0001504794000000 x14: 6f697463656e6e6f
[   99.199003] x13: 6320726f6620676e x12: 0000000000000000
[   99.199005] x11: 000000000000001b x10: 00000000ffffffff
[   99.199007] x9 : fefefefefefefeff x8 : 0000000000000048
[   99.199008] x7 : 0000000000000000 x6 : 00000000ffffffbb
[   99.199010] x5 : 0000000000002710 x4 : 0000007f72ffe380
[   99.199012] x3 : 0000000000000000 x2 : 0000000000000000
[   99.199014] x1 : 0000007f72ffe3c0 x0 : 000000000000000b
[   99.199015]
[   99.199018] Bad mode in Synchronous Abort handler detected, code 0x8600000f
[   99.199020] CPU: 1 PID: 1891 Comm: mongod Not tainted
3.15.0-next-20140610+ #26
[   99.199022] task: ffffffc3e9c12040 ti: ffffffc3ea200000 task.ti:
ffffffc3ea200000
[   99.199022] PC is at 0x90e120
[   99.199023] LR is at 0x90e120
[   99.199024] pc : [<000000000090e120>] lr : [<000000000090e120>]
pstate: 600003c5
[   99.199025] sp : ffffffc3ea203ed0
[   99.199026] x29: 0000007f72ffe2b0 x28: 0000000000000001
[   99.199028] x27: 0000007f72ffe3a0 x26: 000000000000000a
[   99.199030] x25: 0000007f600019d4 x24: 0000007f72ffe520
[   99.199031] x23: 0000000060000000 x22: 0000007f93059410
[   99.199033] x21: ffffffc3ea203ff0 x20: 0000007f600019d4
[   99.199035] x19: 0000007f72ffe3c0 x18: 0000007f72ffe010
[   99.199036] x17: 0000007f93059410 x16: 0000000000b9c078
[   99.199038] x15: 0001504794000000 x14: 30303036203a6574
[   99.199040] x13: 61747370205d3e30 x12: 3231653039303030
[   99.199042] x11: ffffffc000981000 x10: ffffffc0009f14a0
[   99.199044] x9 : 00000000000f0e14 x8 : ffffffc000a013f4
[   99.199045] x7 : 0000000000000015 x6 : 0000000080000000
[   99.199047] x5 : 00000000000003c0 x4 : 0000000000000001
[   99.199049] x3 : ffffffc3ea203cf0 x2 : 00000000dead4ead
[   99.199050] x1 : 00000000000000c3 x0 : 0000000000000000
[   99.199052]

[   99.438632] x6 : ffffffc3ea27fea0
[   99.442208] x5 : 00000000000003c0 x4 : 0000000000000001
[   99.447512] x3 : ffffffc3ea27fcf0 x2 : 00000000dead4ead
[   99.452816] x1 : 000000000000005e x0 : 0000000000000000
[   99.458121]
[   99.459602] Process perf (pid: 3979, stack limit = 0xffffffc3ea27c058)
[   99.466095] Stack: (0xffffffc3ea27fed0 to 0xffffffc3ea280000)
[   99.471812] fec0:                                     26dced90
00000000 0000275a 00000000
[   99.479948] fee0: f2519c90 0000007f 0011f0d0 00000000 0000275a
00000000 009b2648 ffffffc0
[   99.488083] ff00: 11766963 3a043a1c 96a1cc33 0000007f 7f7f7f7f
7f7f7f7f 07cb19c0 00000000
[   99.496217] ff20: 01010101 01010101 00000002 00000000 00000004
00000000 01000000 00000000
[   99.504352] ff40: 31646363 66323839 b11ce5a0 0000007f 00519000
00000000 b177f3a8 0000007f
[   99.512486] ff60: f2519910 0000007f 00011f0d 00000000 00011f0d
00000000 26dcd390 00000000
[   99.520620] ff80: 0000275a 00000000 00000003 00000000 f251ada0
0000007f 26dcd830 00000000
[   99.528755] ffa0: 00000000 00000000 26dced90 00000000 00521000
00000000 f2519ba0 0000007f
[   99.536890] ffc0: 0045f280 00000000 ea27fff0 ffffffc3 b177f3ec
0000007f 80000000 00000000
[   99.545024] ffe0: 00000000 00000000 ffffffff ffffffff eda1d668
ffffffbc eda1d6a0 ffffffbc
[   99.553158] Call trace:
[   99.555592] Code: aa1b03e0 2a1603e1 910383a2 97fed479 (6b16029f)
[   99.561655] Internal error: Oops - bad mode: 0 [#2] SMP
[   99.561687] ---[ end trace 8a63b73eb5685762 ]---
[   99.561915] Bad mode in FIQ handler detected, code 0x56000000
[   99.561918] CPU: 0 PID: 1580 Comm: in:imklog Tainted: G      D
 3.15.0-next-20140610+ #26
[   99.561920] task: ffffffc3ef270000 ti: ffffffc3ea094000 task.ti:
ffffffc3ea094000
[   99.561921] PC is at 0x40d950
[   99.561922] LR is at 0x442b90
[   99.561924] pc : [<000000000040d950>] lr : [<0000000000442b90>]
pstate: 60000000
[   99.561925] sp : ffffffc3ea097ff0
[   99.561927] x29: 0000007fb5c42e80 x28: 0000000000001f9f
[   99.561929] x27: 0000007fb64bef28 x26: 0000007fb64d0520
[   99.561931] x25: 0000007fb5c43fdf x24: 0000000000465a20
[   99.561932] x23: 0000000000465ad0 x22: 0000000000000000
[   99.561934] x21: 0000000000483940 x20: 000000000047c000
[   99.561936] x19: 000000003d9a40d0 x18: 000000000000000d
[   99.561937] x17: 0000007fb672fa2c x16: 000000000047d268
[   99.561939] x15: ffffffffffffffed x14: 000000000000000d
[   99.561941] x13: fffffffffffffe09 x12: 0000000000000018
[   99.561943] x11: 0101010101010101 x10: 7f7f7f7f7fffff7f
[   99.561944] x9 : 0000000000000000 x8 : 0000000000000000
[   99.561946] x7 : 0000000000000000 x6 : 0000007fb5c64890
[   99.561948] x5 : 0000000000000000 x4 : 0000000000000001
[   99.561949] x3 : 00000000ffffffff x2 : 0000000000000000
[   99.561951] x1 : 0000000000000000 x0 : 000000003d9a4410
[   99.561951]
[   99.561955] Bad mode in Synchronous Abort handler detected, code 0x8600000f
[   99.561957] CPU: 0 PID: 1580 Comm: in:imklog Tainted: G      D
 3.15.0-next-20140610+ #26
[   99.561958] task: ffffffc3ef270000 ti: ffffffc3ea094000 task.ti:
ffffffc3ea094000
[   99.561959] PC is at 0x442b90
[   99.561960] LR is at 0x442b90
[   99.561961] pc : [<0000000000442b90>] lr : [<0000000000442b90>]
pstate: 600003c5
[   99.561961] sp : ffffffc3ea097ed0
[   99.561963] x29: 0000007fb5c42e80 x28: 0000000000001f9f
[   99.561965] x27: 0000007fb64bef28 x26: 0000007fb64d0520
[   99.561967] x25: 0000007fb5c43fdf x24: 0000000000465a20
[   99.561969] x23: 0000000060000000 x22: 000000000040d950
[   99.561970] x21: ffffffc3ea097ff0 x20: 000000000047c000
[   99.561972] x19: 000000003d9a40d0 x18: 000000000000000d
[   99.561974] x17: 0000007fb672fa2c x16: 000000000047d268
[   99.561976] x15: ffffffffffffffed x14: 30303036203a6574
[   99.561977] x13: 61747370205d3e30 x12: 3962323434303030
[   99.561979] x11: ffffffc000981000 x10: ffffffc0009f14a0
[   99.561981] x9 : 00000000000efb2c x8 : ffffffc000a026dc
[   99.561982] x7 : 0000007fb672fadc x6 : ffffffffffffffff
[   99.561984] x5 : 00000000000003c0 x4 : 0000000000000001
[   99.561986] x3 : ffffffc3ea097cf0 x2 : 00000000dead4ead
[   99.561987] x1 : 00000000000000b7 x0 : 0000000000000000
[   99.561988]
[   99.810579] Modules linked in:
[   99.813626] CPU: 1 PID: 1891 Comm: mongod Tainted: G      D
3.15.0-next-20140610+ #26
[   99.821933] task: ffffffc3e9c12040 ti: ffffffc3ea200000 task.ti:
ffffffc3ea200000
[   99.829375] PC is at 0x90e120
[   99.832325] LR is at 0x90e120
[   99.835275] pc : [<000000000090e120>] lr : [<000000000090e120>]
pstate: 600003c5
[   99.842630] sp : ffffffc3ea203ed0
[   99.845925] x29: 0000007f72ffe2b0 x28: 0000000000000001
[   99.851229] x27: 0000007f72ffe3a0 x26: 000000000000000a
[   99.856532] x25: 0000007f600019d4 x24: 0000007f72ffe520
[   99.861835] x23: 0000000060000000 x22: 0000007f93059410
[   99.867139] x21: ffffffc3ea203ff0 x20: 0000007f600019d4
[   99.872444] x19: 0000007f72ffe3c0 x18: 0000007f72ffe010
[   99.877749] x17: 0000007f93059410 x16: 0000000000b9c078
[   99.883054] x15: 0001504794000000 x14: 30303036203a6574
[   99.888359] x13: 61747370205d3e30 x12: 3231653039303030
[   99.893663] x11: ffffffc000981000 x10: ffffffc0009f14a0
[   99.898968] x9 : 00000000000f0e14 x8 : ffffffc000a013f4
[   99.904273] x7 : 0000000000000015 x6 : 0000000080000000
[   99.909578] x5 : 00000000000003c0 x4 : 0000000000000001
[   99.914882] x3 : ffffffc3ea203cf0 x2 : 00000000dead4ead
[   99.920186] x1 : 00000000000000c3 x0 : 0000000000000000
[   99.925490]
[   99.926971] Process mongod (pid: 1891, stack limit = 0xffffffc3ea200058)
[   99.933636] Stack: (0xffffffc3ea203ed0 to 0xffffffc3ea204000)
[   99.939353] 3ec0:                                     0000000b
00000000 72ffe3c0 0000007f
[   99.947489] 3ee0: 00000000 00000000 00000000 00000000 72ffe380
0000007f 00002710 00000000
[   99.955623] 3f00: ffffffbb 00000000 00000000 00000000 00000048
00000000 fefefeff fefefefe
[   99.963758] 3f20: ffffffff 00000000 0000001b 00000000 00000000
00000000 6620676e 6320726f
[   99.971893] 3f40: 656e6e6f 6f697463 94000000 00015047 00b9c078
00000000 93059410 0000007f
[   99.980028] 3f60: 72ffe010 0000007f 72ffe3c0 0000007f 600019d4
0000007f 72ffe448 0000007f
[   99.988162] 3f80: 00000400 00000000 72ffe440 0000007f 72ffe520
0000007f 600019d4 0000007f
[   99.996298] 3fa0: 0000000a 00000000 72ffe3a0 0000007f 00000001
00000000 72ffe2b0 0000007f
[  100.004432] 3fc0: 0090e120 00000000 ea203ff0 ffffffc3 93059410
0000007f 60000000 00000000
[  100.012568] 3fe0: 0000000b 00000000 00000048 00000000 020004c0
00000000 0a480040 00000000
[  100.020701] Call trace:
[  100.023133] Code: 910343a4 f9006ba2 f9006fa5 97f14ff9 (6b1f001f)
[  100.029194] Internal error: Oops - bad mode: 0 [#3] SMP
[  100.029215] ---[ end trace 8a63b73eb5685763 ]---
[  100.038982] Modules linked in:
[  100.042031] CPU: 0 PID: 1580 Comm: in:imklog Tainted: G      D
 3.15.0-next-20140610+ #26
[  100.050598] task: ffffffc3ef270000 ti: ffffffc3ea094000 task.ti:
ffffffc3ea094000
[  100.058041] PC is at 0x442b90
[  100.060991] LR is at 0x442b90
[  100.063941] pc : [<0000000000442b90>] lr : [<0000000000442b90>]
pstate: 600003c5
[  100.071296] sp : ffffffc3ea097ed0
[  100.074591] x29: 0000007fb5c42e80 x28: 0000000000001f9f
[  100.079894] x27: 0000007fb64bef28 x26: 0000007fb64d0520
[  100.085198] x25: 0000007fb5c43fdf x24: 0000000000465a20
[  100.090503] x23: 0000000060000000 x22: 000000000040d950
[  100.095806] x21: ffffffc3ea097ff0 x20: 000000000047c000
[  100.101111] x19: 000000003d9a40d0 x18: 000000000000000d
[  100.106416] x17: 0000007fb672fa2c x16: 000000000047d268
[  100.111722] x15: ffffffffffffffed x14: 30303036203a6574
[  100.117025] x13: 61747370205d3e30 x12: 3962323434303030
[  100.122329] x11: ffffffc000981000 x10: ffffffc0009f14a0
[  100.127633] x9 : 00000000000efb2c x8 : ffffffc000a026dc
[  100.132935] x7 : 0000007fb672fadc x6 : ffffffffffffffff
[  100.138240] x5 : 00000000000003c0 x4 : 0000000000000001
[  100.143545] x3 : ffffffc3ea097cf0 x2 : 00000000dead4ead
[  100.148850] x1 : 00000000000000b7 x0 : 0000000000000000
[  100.154154]
[  100.155635] Process in:imklog (pid: 1580, stack limit = 0xffffffc3ea094058)
[  100.162559] Stack: (0xffffffc3ea097ed0 to 0xffffffc3ea098000)
[  100.168275] 7ec0:                                     3d9a4410
00000000 00000000 00000000
[  100.176410] 7ee0: 00000000 00000000 ffffffff 00000000 00000001
00000000 00000000 00000000
[  100.184544] 7f00: b5c64890 0000007f 00000000 00000000 00000000
00000000 00000000 00000000
[  100.192678] 7f20: 7fffff7f 7f7f7f7f 01010101 01010101 00000018
00000000 fffffe09 ffffffff
[  100.200813] 7f40: 0000000d 00000000 ffffffed ffffffff 0047d268
00000000 b672fa2c 0000007f
[  100.208948] 7f60: 0000000d 00000000 3d9a40d0 00000000 0047c000
00000000 00483940 00000000
[  100.217082] 7f80: 00000000 00000000 00465ad0 00000000 00465a20
00000000 b5c43fdf 0000007f
[  100.225216] 7fa0: b64d0520 0000007f b64bef28 0000007f 00001f9f
00000000 b5c42e80 0000007f
[  100.233352] 7fc0: 00442b90 00000000 ea097ff0 ffffffc3 0040d950
00000000 60000000 00000000
[  100.241486] 7fe0: 3d9a421c 00000000 00000062 00000000 00000000
00000000 00000000 00000000
[  100.249619] Call trace:
[  100.252051] Code: d00001d4 97ffee28 f9408a60 97ff2b71 (b94067a0)
[  100.258146] ---[ end trace 8a63b73eb5685764 ]---


Thanks,
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 2/2] arm64: dts: Add PMU node for APM X-Gene Storm SOC

[Jon Masters]

On 06/23/2014 10:47 PM, Ming Lei wrote:

> On Thu, Mar 20, 2014 at 5:47 PM, Vinayak Kale <vkale-qTEPVZfXA3Y@public.gmane.org> wrote:
>> This patch adds the PMU device tree node for APM X-Gene Storm SOC.
>>
>> Please note that this patch has dependancy on a GIC driver patch [1] which is
>> yet to be approved by maintainers.
>>
>> [1]- https://lkml.org/lkml/2014/2/27/605
>> (irqchip:gic: change access of gicc_ctrl register to read modify write)

> No matter if Feng Kan's gic patch is applied or not, perf(PMU)
> still can't work well with your patch, and the failure[1] will be
> triggered when 'perf top' is started.
> 
> The failure happened during the 1st pmu overflow interrupt.
> 
> I test several mustang boards, and all has the problem with
> upstream kernel.

Note that the GIC bypass register config is set by the firmware and this
only preserves the original setting, so you could be running an older
firmware (such as U-Boot). On our (UEFI-enabled systems, complete with
correct patch) perf top runs correctly without any such problems.

Jon.

--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 2/2] arm64: dts: Add PMU node for APM X-Gene Storm SOC

[Ming Lei]

On Tue, Jun 24, 2014 at 11:42 AM, Jon Masters <jcm-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> wrote:
> On 06/23/2014 10:47 PM, Ming Lei wrote:
>
>> On Thu, Mar 20, 2014 at 5:47 PM, Vinayak Kale <vkale-qTEPVZfXA3Y@public.gmane.org> wrote:
>>> This patch adds the PMU device tree node for APM X-Gene Storm SOC.
>>>
>>> Please note that this patch has dependancy on a GIC driver patch [1] which is
>>> yet to be approved by maintainers.
>>>
>>> [1]- https://lkml.org/lkml/2014/2/27/605
>>> (irqchip:gic: change access of gicc_ctrl register to read modify write)
>
>> No matter if Feng Kan's gic patch is applied or not, perf(PMU)
>> still can't work well with your patch, and the failure[1] will be
>> triggered when 'perf top' is started.
>>
>> The failure happened during the 1st pmu overflow interrupt.
>>
>> I test several mustang boards, and all has the problem with
>> upstream kernel.
>
> Note that the GIC bypass register config is set by the firmware and this
> only preserves the original setting, so you could be running an older
> firmware (such as U-Boot). On our (UEFI-enabled systems, complete with
> correct patch) perf top runs correctly without any such problems.

We are using U-Boot 2013.04-mustang_sw_1.11.06-beta_rc (Feb 03 2014 -
09:10:17), looks not very old.

Do you know how to check if the GIC bypass register config is correct
inside kernel?

Thanks,
--
Ming Lei
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 2/2] arm64: dts: Add PMU node for APM X-Gene Storm SOC

[Ming Lei]

Hi Jon and Vinayak,

On Tue, Jun 24, 2014 at 11:42 AM, Jon Masters <jcm-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> wrote:
> On 06/23/2014 10:47 PM, Ming Lei wrote:
>
>> On Thu, Mar 20, 2014 at 5:47 PM, Vinayak Kale <vkale-qTEPVZfXA3Y@public.gmane.org> wrote:
>>> This patch adds the PMU device tree node for APM X-Gene Storm SOC.
>>>
>>> Please note that this patch has dependancy on a GIC driver patch [1] which is
>>> yet to be approved by maintainers.
>>>
>>> [1]- https://lkml.org/lkml/2014/2/27/605
>>> (irqchip:gic: change access of gicc_ctrl register to read modify write)
>
>> No matter if Feng Kan's gic patch is applied or not, perf(PMU)
>> still can't work well with your patch, and the failure[1] will be
>> triggered when 'perf top' is started.
>>
>> The failure happened during the 1st pmu overflow interrupt.
>>
>> I test several mustang boards, and all has the problem with
>> upstream kernel.
>
> Note that the GIC bypass register config is set by the firmware and this
> only preserves the original setting, so you could be running an older
> firmware (such as U-Boot). On our (UEFI-enabled systems, complete with
> correct patch) perf top runs correctly without any such problems.

I am wondering if it is related with boot or firmware. We have a board
which is running APM internal 3.12 based kernel, and perf(pmu) works
well. But if I just replace the 3.12 based kernel with upstream kernel
and keep others(boot, kernel parameter, fs, ....) not changed, then
the problem still can happen.

Thanks,
--
Ming Lei
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 2/2] arm64: dts: Add PMU node for APM X-Gene Storm SOC

[Dann Frazier]

On Tue, Jun 24, 2014 at 1:00 AM, Ming Lei <ming.lei-Z7WLFzj8eWMS+FvcfC7Uqw@public.gmane.org> wrote:
> Hi Jon and Vinayak,
>
> On Tue, Jun 24, 2014 at 11:42 AM, Jon Masters <jcm-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> wrote:
>> On 06/23/2014 10:47 PM, Ming Lei wrote:
>>
>>> On Thu, Mar 20, 2014 at 5:47 PM, Vinayak Kale <vkale-qTEPVZfXA3Y@public.gmane.org> wrote:
>>>> This patch adds the PMU device tree node for APM X-Gene Storm SOC.
>>>>
>>>> Please note that this patch has dependancy on a GIC driver patch [1] which is
>>>> yet to be approved by maintainers.
>>>>
>>>> [1]- https://lkml.org/lkml/2014/2/27/605
>>>> (irqchip:gic: change access of gicc_ctrl register to read modify write)
>>
>>> No matter if Feng Kan's gic patch is applied or not, perf(PMU)
>>> still can't work well with your patch, and the failure[1] will be
>>> triggered when 'perf top' is started.
>>>
>>> The failure happened during the 1st pmu overflow interrupt.
>>>
>>> I test several mustang boards, and all has the problem with
>>> upstream kernel.
>>
>> Note that the GIC bypass register config is set by the firmware and this
>> only preserves the original setting, so you could be running an older
>> firmware (such as U-Boot). On our (UEFI-enabled systems, complete with
>> correct patch) perf top runs correctly without any such problems.
>
> I am wondering if it is related with boot or firmware. We have a board
> which is running APM internal 3.12 based kernel, and perf(pmu) works
> well. But if I just replace the 3.12 based kernel with upstream kernel
> and keep others(boot, kernel parameter, fs, ....) not changed, then
> the problem still can happen.

fyi, newer firmware is confirmed to fix this issue on u-boot based platforms.
Tested-by: dann frazier <dann.frazier-Z7WLFzj8eWMS+FvcfC7Uqw@public.gmane.org>
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html