Re: [PATCH V3 0/9] arm64: imx8mn: Enable more imx8m Nano functions

[Adam Ford]

On Tue, Nov 16, 2021 at 12:27 PM Tim Harvey <tharvey@xxxxxxxxxxxxx> wrote:
>
> On Tue, Nov 16, 2021 at 10:04 AM Adam Ford <aford173@xxxxxxxxx> wrote:
> >
> > On Tue, Nov 16, 2021 at 11:57 AM Tim Harvey <tharvey@xxxxxxxxxxxxx> wrote:
> > >
> > > On Thu, Nov 4, 2021 at 9:18 AM Adam Ford <aford173@xxxxxxxxx> wrote:
> > > >
> > > > The i.MX8M Nano is similar to the i.MX8M Mini in some ways, but very
> > > > different in others.  With the blk-ctrl driver for Mini in place,
> > > > this series expands the blk-ctrl driver to support the Nano which
> > > > opens the door for additional functions in the future.  As part of
> > > > this series, it also addresses some issues in the GPCv2 driver and
> > > > finally adds support for enabling USB and GPU.
> > > >
> > > > V3:  Fixes an the yaml example
> > > > V2:  Fixes the clock count in the blk-ctrl
> > > >
> > > > Adam Ford (9):
> > > >   soc: imx: gpcv2: keep i.MX8MN gpumix bus clock enabled
> > > >   soc: imx: gpcv2: Add dispmix and mipi domains to imx8mn
> > > >   dt-bindings: power: imx8mn: add defines for DISP blk-ctrl domains
> > > >   dt-bindings: soc: add binding for i.MX8MN DISP blk-ctrl
> > > >   soc: imx: imx8m-blk-ctrl: add i.MX8MN DISP blk-ctrl
> > > >   arm64: dts: imx8mn: add GPC node
> > > >   arm64: dts: imx8mn: put USB controller into power-domains
> > > >   arm64: dts: imx8mn: add DISP blk-ctrl
> > > >   arm64: dts: imx8mn: Enable GPU
> > > >
> > > >  .../soc/imx/fsl,imx8mn-disp-blk-ctrl.yaml     |  97 +++++++++++++++++
> > > >  arch/arm64/boot/dts/freescale/imx8mn.dtsi     | 103 ++++++++++++++++++
> > > >  drivers/soc/imx/gpcv2.c                       |  26 +++++
> > > >  drivers/soc/imx/imx8m-blk-ctrl.c              |  75 ++++++++++++-
> > > >  include/dt-bindings/power/imx8mn-power.h      |   5 +
> > > >  5 files changed, 305 insertions(+), 1 deletion(-)
> > > >  create mode 100644 Documentation/devicetree/bindings/soc/imx/fsl,imx8mn-disp-blk-ctrl.yaml
> > > >
> > >
> > > Adam,
> > >
> > > Thanks for the patches. I'm not sure how best to test this but on an
> > > imx8mm-venice-gw7902 which has USB, but no display.
> > >
> > > I find that if DRM_ETNAVIV is enabled I hang at 'etnaviv etnaviv:
> > > bound 38000000.gpu (ops 0xffff800010964748)'.
> >
> > Thanks for testing this.
> >
> > Does your board send power to the GPU?  I recall someone somewhere
> > didn't power theirGPU, but I can't remember who and/or what board was
> > being discussed.
>
> Yes, the imx8mm-venice-gw7902 does have the GPU powered (the
> imx8mm-venice-gw7901 was the one that did not).

I cannot replicate your issue.  I applied the patch series to
5.16-rc1, and it's still working for me.

root@beacon-imx8mn-kit:~# dmesg |grep viv
[   11.323660] etnaviv etnaviv: bound 38000000.gpu (ops gpu_ops [etnaviv])
[   11.480543] etnaviv-gpu 38000000.gpu: model: GC7000, revision: 6203
[   11.747576] [drm] Initialized etnaviv 1.3.0 20151214 for etnaviv on minor 0

After booting, I connected a USB drive and it enumerated.

[  152.328228] ci_hdrc ci_hdrc.0: new USB bus registered, assigned bus number 1
[  152.351885] ci_hdrc ci_hdrc.0: USB 2.0 started, EHCI 1.00
[  152.363859] hub 1-0:1.0: USB hub found
[  152.371560] hub 1-0:1.0: 1 port detected
[  153.075902] usb 1-1: new high-speed USB device number 2 using ci_hdrc
[  153.244124] usb-storage 1-1:1.0: USB Mass Storage device detected
[  153.263651] scsi host0: usb-storage 1-1:1.0
[  154.277997] scsi 0:0:0:0: Direct-Access     SanDisk  Cruzer
  1.00 PQ: 0 ANSI: 6
[  154.292357] sd 0:0:0:0: [sda] 247529472 512-byte logical blocks:
(127 GB/118 GiB)
[  154.309246] sd 0:0:0:0: [sda] Write Protect is off
[  154.319237] sd 0:0:0:0: [sda] Write cache: disabled, read cache:
enabled, doesn't support DPO or FUA
[  154.359113]  sda: sda1
[  154.366975] sd 0:0:0:0: [sda] Attached SCSI removable disk

root@beacon-imx8mn-kit:~# lsusb
Bus 001 Device 002: ID 0781:5530 SanDisk Corp. Cruzer
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub

I am not sure where to go from here.

>
> Tim
>
> >
> > I'll run some more tests on the latest 5.16-rc1 to see if I can
> > replicate your issue.
> >
> > adam
> > >
> > > If DRM_ETNAVIV is not enabled:
> > > - boots fine
> > > - usb works
> > > - soft reboot works (does not hang)
> >
> > At least we have some progress.  :-)
> >
> > adam
> > >
> > > Best regards,
> > >
> > > Tim