Hi Andre,
On 3 April 2020 11:04:17 PM IST, "André Przywara" <andre.przywara@xxxxxxx> wrote:
>On 03/04/2020 11:29, André Przywara wrote:
>
>(repost with CC: Actions Semi mailing list)
>
>> On 03/04/2020 11:02, Amit Tomer wrote:
>>
>> Hi,
>>
>>> While booting mainline Kernel on Cubieboard7-lite(based on S700),
>see
>>> the following logs
>>> that suggests S700 pin controller is not probed well.
>>>
>>> [ 0.257415] pinctrl-s700 e01b0000.pinctrl: can't request region
>for
>>> resource [mem 0xe01b0000-0xe01b0fff]
>>> [ 0.266902] pinctrl-s700: probe of e01b0000.pinctrl failed with
>error -16
>>>
>>> Register range defined for pin controller seems to be huge and
>causing
>>> this failure.
>>>
>>> reg = <0x0 0xe01b0000 0x0 0x1000>;
>>
>> To summarise the discussion I had with Amit over IRC:
>>
>> The "sps" power domain controller is overlapping with the pinctrl:
>> sps: power-controller@e01b0100 {
>> compatible = "actions,s700-sps";
>> reg = <0x0 0xe01b0100 0x0 0x100>;
>> #power-domain-cells = <1>;
>> };
>> pinctrl: pinctrl@e01b0000 {
>> compatible = "actions,s700-pinctrl";
>> reg = <0x0 0xe01b0000 0x0 0x1000>;
>> ...
>>
>> This is not the case on the S900, where most of the S700 support
>seemed
>> to be lifted from.
>>
>> Looking at the datasheet there is also other IP sharing the MMIO
>region
>> with the GPIO/pinctrl:
>> GPIO/MFP Controller (include PWM, SPS, RMU, AVS)
>>
>> Seems to be a pretty nasty hardware design, at least when it comes to
>> mapping this to the Linux:
>> - The PWM controller uses register offsets 0x50-0x60, and 0x78 - 0x80
>> - The SPS controller uses offsets 0x100 - 0x108
>> - The pinctrl uses 0x00 - 0x50, 0x60-0x78, 0x80-0x8c and 0x200 -
>0x254
>>
>> So at least the PWM sits in the middle of the pinctrl, in two chunks.
>>
>> What is the typical solution for those kind of problems? One device
>> exporting a regmap? Or using a MFD?
>
>
>Thought a bit about it, what about this:
>
>- We amend the pinctrl binding to also deal with *two* reg regions: the
>first for the actual pinctrl/GPIO part (0x0 - 0x100), the second for
>the
>interrupt part (starting at 0x200). It seems like we don't use the
>interrupt functionality at the moment, so that should not break
>anything.
>- The pinctrl driver exports a regmap for the PWM registers.
>- The (to-be-written) PWM driver and binding contain a reference to the
>pinctrl node, which is then used in the PWM driver to retrieve the
>regmap.
>This should cover the three devices:
>pinctrl: pinctrl@e01b0000 {
> compatible = "actions,s700-pinctrl";
> reg = <0x0 0xe01b0000 0x0 0x100>,
> <0x0 0xe01b0200 0x0 0x100>;
>...
>sps: power-controller@e01b0100 {
> compatible = "actions,s700-sps";
> reg = <0x0 0xe01b0100 0x0 0x100>;
>....
>pwm: pwm@e01b0050 {
> compatible = "TBD";
> reg = <0x0 0xe01b0050 0x0 0x10>,
> <0x0 0xe01b0078 0x0 0x08>;
> regmap = <&pinctrl>;
>...
>The PWM driver would need to deal with a failing
>devm_platform_ioremap_resource(), reverting to using the regmap then.
>Or
>maybe never try this in the first place, to avoid breaking the pinctrl
>driver when being probed first.
>
>This would allow to keep the pinctrl and sps drivers as they are, even
>allowing the new (fixed) DT to work with existing kernels.
>
>Does that sound feasible? Or is there a better solution?
This looks to me a hack. I did miss to spot the sps and gpio/pinctrl conflict for S700 before, sorry for that.
But for me, implementing gpio/pinctrl block as MFD seems to be a proper solution. S900 shares PWM with it, so definitely it can benefit from doing so.
Andreas, what do you think?
Thanks,
Mani
>
>Cheers,
>Andre
>
>>> Complete logs are below
>>>
>>> Starting kernel ...
>>>
>>> [ 0.000000] Booting Linux on physical CPU 0x0000000000
>[0x410fd034]
>>> [ 0.000000] Linux version 5.6.0-02330-g458ef2a-dirty
>>> (amit@amit-ThinkPad-X230) (gcc version 7.3.1 20180425
>>> [linaro-7.3-2018.05 revision
>d29120a424ecfbc167ef90065c0eeb7f91977701]
>>> (Linaro GCC 7.3-2018.05)) #12 SMP PREEMPT Fri Apr 3 14:19:08 IST
>2020
>>> [ 0.000000] Machine model: CubieBoard7
>>> [ 0.000000] earlycon: owl0 at MMIO 0x00000000e0126000 (options
>'')
>>> [ 0.000000] printk: bootconsole [owl0] enabled
>>> [ 0.000000] efi: UEFI not found.
>>> [ 0.000000] cma: Reserved 32 MiB at 0x000000003e000000
>>> [ 0.000000] NUMA: No NUMA configuration found
>>> [ 0.000000] NUMA: Faking a node at [mem
>>> 0x0000000000000000-0x000000003fffffff]
>>> [ 0.000000] NUMA: NODE_DATA [mem 0x3ddfc100-0x3ddfdfff]
>>> [ 0.000000] Zone ranges:
>>> [ 0.000000] DMA [mem
>0x0000000000000000-0x000000003fffffff]
>>> [ 0.000000] DMA32 empty
>>> [ 0.000000] Normal empty
>>> [ 0.000000] Movable zone start for each node
>>> [ 0.000000] Early memory node ranges
>>> [ 0.000000] node 0: [mem
>0x0000000000000000-0x000000001effffff]
>>> [ 0.000000] node 0: [mem
>0x0000000020000000-0x000000003fffffff]
>>> [ 0.000000] Initmem setup node 0 [mem
>0x0000000000000000-0x000000003fffffff]
>>> [ 0.000000] psci: probing for conduit method from DT.
>>> [ 0.000000] psci: PSCIv0.2 detected in firmware.
>>> [ 0.000000] psci: Using standard PSCI v0.2 function IDs
>>> [ 0.000000] psci: Trusted OS migration not required
>>> [ 0.000000] percpu: Embedded 23 pages/cpu s53272 r8192 d32744
>u94208
>>> [ 0.000000] Detected VIPT I-cache on CPU0
>>> [ 0.000000] CPU features: detected: ARM erratum 845719
>>> [ 0.000000] Built 1 zonelists, mobility grouping on. Total
>pages: 253952
>>> [ 0.000000] Policy zone: DMA
>>> [ 0.000000] Kernel command line: console=ttyOWL,115200
>>> earlycon=owl,0xe0126000 root=/dev/mmcblk0p2
>>> [ 0.000000] Dentry cache hash table entries: 131072 (order: 8,
>>> 1048576 bytes, linear)
>>> [ 0.000000] Inode-cache hash table entries: 65536 (order: 7,
>524288
>>> bytes, linear)
>>> [ 0.000000] mem auto-init: stack:off, heap alloc:off, heap
>free:off
>>> [ 0.000000] Memory: 951668K/1032192K available (12604K kernel
>code,
>>> 1948K rwdata, 6964K rodata, 5312K init, 462K bss, 47756K reserved,
>>> 32768K cma-reserved)
>>> [ 0.000000] SLUB: HWalign=64, Order=0-3, MinObjects=0, CPUs=4,
>Nodes=1
>>> [ 0.000000] rcu: Preemptible hierarchical RCU implementation.
>>> [ 0.000000] rcu: RCU event tracing is enabled.
>>> [ 0.000000] rcu: RCU restricting CPUs from NR_CPUS=256 to
>nr_cpu_ids=4.
>>> [ 0.000000] Tasks RCU enabled.
>>> [ 0.000000] rcu: RCU calculated value of scheduler-enlistment
>delay
>>> is 25 jiffies.
>>> [ 0.000000] rcu: Adjusting geometry for rcu_fanout_leaf=16,
>nr_cpu_ids=4
>>> [ 0.000000] NR_IRQS: 64, nr_irqs: 64, preallocated irqs: 0
>>> [ 0.000000] GIC: Using split EOI/Deactivate mode
>>> [ 0.000000] random: get_random_bytes called from
>>> start_kernel+0x2a8/0x444 with crng_init=0
>>> [ 0.000000] arch_timer: cp15 timer(s) running at 24.00MHz (phys).
>>> [ 0.000000] clocksource: arch_sys_counter: mask: 0xffffffffffffff
>>> max_cycles: 0x588fe9dc0, max_idle_ns: 440795202592 ns
>>> [ 0.000000] sched_clock: 56 bits at 24MHz, resolution 41ns, wraps
>>> every 4398046511097ns
>>> [ 0.008163] sched_clock: 32 bits at 24MHz, resolution 41ns, wraps
>>> every 89478484971ns
>>> [ 0.015969] clocksource: timer: mask: 0xffffffff max_cycles:
>>> 0xffffffff, max_idle_ns: 79635851949 ns
>>> [ 0.025300] Console: colour dummy device 80x25
>>> [ 0.029821] Calibrating delay loop (skipped), value calculated
>>> using timer frequency.. 48.00 BogoMIPS (lpj=96000)
>>> [ 0.040069] pid_max: default: 32768 minimum: 301
>>> [ 0.044785] LSM: Security Framework initializing
>>> [ 0.049465] Mount-cache hash table entries: 2048 (order: 2, 16384
>>> bytes, linear)
>>> [ 0.056857] Mountpoint-cache hash table entries: 2048 (order: 2,
>>> 16384 bytes, linear)
>>> [ 0.066496] rcu: Hierarchical SRCU implementation.
>>> [ 0.072035] EFI services will not be available.
>>> [ 0.076832] smp: Bringing up secondary CPUs ...
>>> [ 0.082027] Detected VIPT I-cache on CPU1
>>> [ 0.082093] CPU1: Booted secondary processor 0x0000000001
>[0x410fd034]
>>> [ 0.082792] Detected VIPT I-cache on CPU2
>>> [ 0.082834] CPU2: Booted secondary processor 0x0000000002
>[0x410fd034]
>>> [ 0.083500] Detected VIPT I-cache on CPU3
>>> [ 0.083540] CPU3: Booted secondary processor 0x0000000003
>[0x410fd034]
>>> [ 0.083627] smp: Brought up 1 node, 4 CPUs
>>> [ 0.119242] SMP: Total of 4 processors activated.
>>> [ 0.123938] CPU features: detected: 32-bit EL0 Support
>>> [ 0.129087] CPU features: detected: CRC32 instructions
>>> [ 0.143064] CPU: All CPU(s) started at EL2
>>> [ 0.147207] alternatives: patching kernel code
>>> [ 0.153187] devtmpfs: initialized
>>> [ 0.158590] KASLR disabled due to lack of seed
>>> [ 0.163472] clocksource: jiffies: mask: 0xffffffff max_cycles:
>>> 0xffffffff, max_idle_ns: 7645041785100000 ns
>>> [ 0.173211] futex hash table entries: 1024 (order: 4, 65536
>bytes, linear)
>>> [ 0.181589] pinctrl core: initialized pinctrl subsystem
>>> [ 0.188064] thermal_sys: Registered thermal governor 'step_wise'
>>> [ 0.188068] thermal_sys: Registered thermal governor
>'power_allocator'
>>> [ 0.194183] DMI not present or invalid.
>>> [ 0.205285] NET: Registered protocol family 16
>>> [ 0.211593] DMA: preallocated 256 KiB pool for atomic allocations
>>> [ 0.217696] audit: initializing netlink subsys (disabled)
>>> [ 0.223266] audit: type=2000 audit(0.144:1): state=initialized
>>> audit_enabled=0 res=1
>>> [ 0.231024] cpuidle: using governor menu
>>> [ 0.235203] hw-breakpoint: found 6 breakpoint and 4 watchpoint
>registers.
>>> [ 0.242077] ASID allocator initialised with 65536 entries
>>> [ 0.248650] Serial: AMBA PL011 UART driver
>>> [ 0.257415] pinctrl-s700 e01b0000.pinctrl: can't request region
>for
>>> resource [mem 0xe01b0000-0xe01b0fff]
>>> [ 0.266902] pinctrl-s700: probe of e01b0000.pinctrl failed with
>error -16
>>>
>>
--
Sent from my Android device with K-9 Mail. Please excuse my brevity.
_______________________________________________
linux-actions mailing list
linux-actions@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/linux-actions
