Questions about bringing up MT7623 board

[John Crispin]

On 11/02/18 18:54, Kristian Evensen wrote:
> Hi,
>
>
> On Sun, 11 Feb 2018 at 18:45, John Crispin <john at phrozen.org 
> <mailto:john at phrozen.org>> wrote:
>
>
>
>     On 11/02/18 17:09, Kristian Evensen wrote:
>     > Hi,
>     >
>     > On Sun, 11 Feb 2018 at 15:46, John Crispin <john at phrozen.org
>     <mailto:john at phrozen.org>
>     > <mailto:john at phrozen.org <mailto:john at phrozen.org>>> wrote:
>     >
>     >
>     >
>     >     On 11/02/18 15:01, Kristian Evensen wrote:
>     >     > Hi,
>     >     >
>     >     > It turns out that enabling early printk in the kernel
>     config did the
>     >     > trick, thank you very much for pointing me to this error.
>     I also
>     >     > changed my defconfig a bit, I merged the two OpenWRT 4.9
>     configs
>     >     > (generic and mediatek), created a new defconfig and used
>     that as a
>     >     > starting point (make ..._defconfig). Now I can finally see
>     what is
>     >     > going on and I see that the boot progresses quite far. My
>     boot now
>     >     > gets to the point where it looks for the root filesystem,
>     but here I
>     >     > have hit another problem. The kernel only produces the
>     following
>     >     > error:
>     >     >
>     >     > [    2.037354] mmc0: new high speed MMC card at address 0001
>     >     > [    2.045252] mmcblk0: mmc0:0001 008G70 7.28 GiB
>     >     > [    2.053533] mmcblk0boot0: mmc0:0001 008G70 partition 1
>     4.00 MiB
>     >     > [    2.053723] mmcblk0boot1: mmc0:0001 008G70 partition 2
>     4.00 MiB
>     >     > [    2.053910] mmcblk0rpmb: mmc0:0001 008G70 partition 3
>     4.00 MiB
>     >     > [    2.055456]  mmcblk0: p2 p3
>     >     > [    2.056986] xhci-mtk 1a1c0000.usb: 1a1c0000.usb supply
>     vbus not
>     >     > found, using dummy regulator
>     >     > [    2.153894] vpa: disabling
>     >     > [    2.156608] vmc: disabling
>     >     > [    2.159296] vmch: disabling
>     >     > [    2.162103] vgp1: disabling
>     >     > [    2.164897] vcamaf: disabling
>     >     > [    2.168241] VFS: Cannot open root device "(null)" or
>     >     > unknown-block(0,0): error -6
>     >     > [    2.175712] Please append a correct "root=" boot
>     option; here are
>     >     > the available partitions:
>     >     > [    2.184031] b300         7634944 mmcblk0
>     >     > [    2.184036]  driver: mmcblk
>     >     > [    2.190775]   b302         1048576 mmcblk0p2 00000000-02
>     >     > [    2.190779]
>     >     > [    2.197563]   b303         2097152 mmcblk0p3 00000000-03
>     >     > [    2.197566]
>     >     > [    2.204332] b318            4096 mmcblk0rpmb
>     >     > [    2.204335]  (driver?)
>     >     > [    2.210988] b310            4096 mmcblk0boot1
>     >     > [    2.210992]  (driver?)
>     >     > [    2.217747] b308            4096 mmcblk0boot0
>     >     > [    2.217750]  (driver?)
>     >     > [    2.224504] Kernel panic - not syncing: VFS: Unable to
>     mount root
>     >     > fs on unknown-block(0,0)
>     >     >
>     >     > My rootfs, etc, is stored on mmcblk0 and the partitioning is
>     >     specified
>     >     > in the boot arguments (which are:
>     >     > block2mtd.block2mtd=/dev/mmcblk0,65536,eMMC,5
>     >     >
>     >
>      mtdparts=eMMC:256k(mbr)ro,512k(uboot)ro,256k(config)ro,256k(factory)ro,32M(kernel),32M(recovery),1024M(rootfs),2048M(usrdata),-(bmtpool)
>     >     >   rootfstype=squashfs,jffs2). I have doublecheked, and I can't
>     >     find any
>     >     > missing MTD-options compared to the 4.9 config. Could it
>     be that new
>     >     > options have been added and have to be specified? For example,
>     >     > MTD_BLOCK2MTD and MTD_CMDLINE_PARTS is set.
>     >
>     >     block2mtd is a owrt specific driver, it wont work on on vanilla.
>     >
>     >
>     > Thanks for letting me know. I was not aware and just assumed the
>     > block2mtd would work as in owrt. Perhaps a better starting point for
>     > my work would be to try to hack together 4.14 support for
>     mediatek in
>     > OpenWRT. Installing a «normal» rootfs is a bit of a pain on this
>     > board. Another option would be to move the mtd-patches I guess. Will
>     > try both and see how it goes :)
>     >
>     > BR,
>     > Kristian
>     >
>
>     I have 4.14 support in my staging tree alread, i'll psh it to owrt the
>     next couple of days
>
>          John
>
>
> Great news, thanks for your effort! Would it be possible for you to 
> push it to your staging tree on openwrt.org <http://openwrt.org>, so 
> that I can pull and start testing with my board?
>
> BR,
> Kristian
>
>


will do in the morning ....
     John