Debian SID kernel doesn't boot on PowerBook 3400c

[Christophe Leroy]

Stan Johnson <userm57 at yahoo.com> a écrit :

> Hello,
>
> The current Debian SID kernel will not boot on a PowerBook 3400c running
> the latest version of Debian SID. If booted using the BootX extension,
> the kernel hangs immediately:
>
> "Welcome to Linux, kernel 5.10.0-8-powerpc"
>
> If booted from Mac OS, the Mac OS screen hangs.
>
> Booting also hangs if the "No video driver" option is selected in BootX,
> "No video driver" causes "video=ofonly" to be passed to the kernel.
>
> This is the current command line that I'm using in BootX:
> root=/dev/sda13 video=chips65550:vmode:14,cmode:16
>
> Kernel v5.9 works as expected.
>
> The config file I'm using is attached.
>
> Here are the results of a git bisect, marking v5.9 as "good" and the
> most current kernel as "bad":
>
> $ cd linux
> $ git remote update
> $ git bisect reset
> $ git bisect start
> $ git bisect bad
> $ git bisect good v5.9
>
> Note: "bad" -> hangs at boot; "good" -> boots to login prompt
>
>  1) 5.11.0-rc5-pmac-00034-g684da7628d9 (bad)
>  2) 5.10.0-rc3-pmac-00383-gbb9dd3ce617 (good)
>  3) 5.10.0-pmac-06637-g2911ed9f47b (good)
>     Note: I had to disable SMP to build this kernel.
>  4) 5.10.0-pmac-10584-g9805529ec54 (good)
>     Note: I had to disable SMP to build this kernel.
>  5) 5.10.0-pmac-12577-g8552d28e140 (bad)
>  6) 5.10.0-pmac-11576-g8a5be36b930 (bad)
>  7) 5.10.0-pmac-11044-gbe695ee29e8 (good)
>     Note: I had to disable SMP to build this kernel.
>  8) 5.10.0-rc2-pmac-00288-g59d512e4374 (bad)
>  9) 5.10.0-rc2-pmac-00155-gc3d35ddd1ec (good)
> 10) 5.10.0-rc2-pmac-00221-g7049b288ea8 (good)
> 11) 5.10.0-rc2-pmac-00254-g4b74a35fc7e (bad)
> 12) 5.10.0-rc2-pmac-00237-ged22bb8d39f (good)
> 13) 5.10.0-rc2-pmac-00245-g87b57ea7e10 (good)
> 14) 5.10.0-rc2-pmac-00249-gf10881a46f8 (bad)
> 15) 5.10.0-rc2-pmac-00247-gf8a4b277c3c (good)
> 16) 5.10.0-rc2-pmac-00248-gdb972a3787d (bad)
>
> db972a3787d12b1ce9ba7a31ec376d8a79e04c47 is the first bad commit

Not sure this is really the root of the problem.

Can you try again without CONFIG_VMAP_STACK ?

Thanks
Christophe


> commit db972a3787d12b1ce9ba7a31ec376d8a79e04c47
> Author: Christophe Leroy <christophe.leroy at csgroup.eu>
> Date:   Tue Dec 8 05:24:19 2020 +0000
>
>     powerpc/powermac: Fix low_sleep_handler with CONFIG_VMAP_STACK
>
>     low_sleep_handler() can't restore the context from standard
>     stack because the stack can hardly be accessed with MMU OFF.
>
>     Store everything in a global storage area instead of storing
>     a pointer to the stack in that global storage area.
>
>     To avoid a complete churn of the function, still use r1 as
>     the pointer to the storage area during restore.
>
>     Fixes: cd08f109e262 ("powerpc/32s: Enable CONFIG_VMAP_STACK")
>     Reported-by: Giuseppe Sacco <giuseppe at sguazz.it>
>     Signed-off-by: Christophe Leroy <christophe.leroy at csgroup.eu>
>     Tested-by: Giuseppe Sacco <giuseppe at sguazz.it>
>     Signed-off-by: Michael Ellerman <mpe at ellerman.id.au>
>     Link:
> https://lore.kernel.org/r/e3e0d8042a3ba75cb4a9546c19c408b5b5b28994.1607404931.git.christophe.leroy@csgroup.eu
>
> :040000 040000 d5039513d19748fc13712a2c67ae034371b95fe7
> cbbdbdc4b05c713ea2577674260fd37e71306cc0 M	arch
>
> Please let me know if you need more information.
>
> -Stan Johnson