Re: [PATCH v2 0/9] i2c: move handling of suspended adapters to the core
From: Geert Uytterhoeven
Date: Wed Dec 26 2018 - 06:02:15 EST
On Sat, Dec 22, 2018 at 9:26 PM Wolfram Sang
> Here is the new version without specific I2C helpers but using the
> 'is_suspended' flag from the PM core. I didn't like messing with the
> flag directly, so I did a helper in patch 1. So far, I like the
> approach. The diffstat looks nice, and i2c-rcar.c and i2c-sh_mobile.c
> rejected rightfully too later transfers without further modifications.
> Tested on a Renesas Lager board (R-Car H2).
> I dropped a few Tested-by tags because I think this approach is too
> different from V1 to keep them. I hope you guys can have a look again.
> Thanks for all the testing, so far!
> A branch can be found here:
> git://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux.git renesas/i2c/reject-when-suspended
> If this series is acceptable, I'd suggest to take it via my i2c tree
> after rc1. And then I'll provide an immutable branch for the PM tree to
> pick. Let me know if this works for you.
This breaks resume of the CS2000 clock driver on Salvator-X(S) (and
presumable ULCB, too):
Accessing already suspended I2C/SMBus adapter
WARNING: CPU: 1 PID: 1186 at drivers/i2c/i2c-core-smbus.c:551
Modules linked in:
CPU: 1 PID: 1186 Comm: s2idle Not tainted
Hardware name: Renesas Salvator-X board based on r8a7795 ES1.x (DT)
pstate: 60400005 (nZCv daif +PAN -UAO)
pc : __i2c_smbus_xfer+0x38/0x66c
lr : __i2c_smbus_xfer+0x38/0x66c
sp : ffffff8013413880
x29: ffffff8013413880 x28: ffffffc6f78b4820
x27: 0000000000000010 x26: ffffff8010cf6178
x25: ffffff8013413976 x24: 0000000000000002
x23: 0000000000000016 x22: ffffffc6f7872088
x21: 0000000000000000 x20: 000000000000004f
x19: ffffffc6f7872088 x18: 000000000000000a
x17: 0000000000000000 x16: 0000000000000000
x15: 0000000000029c80 x14: 0720072007200720
x13: 0720072007200720 x12: 0720072007200720
x11: 0720072007200720 x10: 0720072007200720
x9 : ffffff801100e6d0 x8 : 6461207375424d53
x7 : ffffff8011c825c8 x6 : ffffff8011c82000
x5 : 0000000000000000 x4 : ffffff8013414000
x3 : ffffff80134136e0 x2 : 00000046ee875000
x1 : 82c6d7c720d64000 x0 : 0000000000000000
The CS2000 driver uses SET_LATE_SYSTEM_SLEEP_PM_OPS().
Suspend/resume of the BD9571 driver works fine, as that one uses
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds