On 2019-02-27 18:55, Robert Shearman wrote: > On 26/02/2019 11:54, Peter Rosin wrote: >> On 2019-02-25 16:51, Robert Shearman wrote: >>> From: Robert Shearman <robert.shearman@xxxxxxx> >>> >>> The behaviour, by default, to not deselect after each transfer is >>> unsafe when there is a device with an address that conflicts with >>> another device on another pca954x mux on the same parent bus, and it >>> may not be convenient to use devicetree to set the deselect mux, >>> e.g. when running on x86_64 when ACPI is used to discover most of the >>> device hierarchy. >>> >>> Therefore, provide the ability to set the idle state behaviour using a >>> new sysfs file, idle_state as a complement to the method of >>> instantiating the device via sysfs. The possible behaviours are >>> disconnect, i.e. to deselect all channels from the mux, as-is (the >>> default), i.e. leave the last channel selected, and set a >>> predetermined channel. >>> >>> The current behaviour of leaving the channel as-is after each >>> transaction is preserved. >>> >>> Signed-off-by: Robert Shearman <robert.shearman@xxxxxxx> >>> --- >>> .../ABI/testing/sysfs-bus-i2c-devices-pca954x | 20 +++++ >> >> Should not ABI/... patches go to someone? Who? Did you use >> get_maintainer.pl? I think you should at least include the main kernel >> list? Maybe someone filters messages looking for ABI changes? > > get_maintainer.pl doesn't recommend anything more than usual with the > presence of the ABI changes, but I can certainly copy LKML on the next > version of the series. Please also Cc Greg as hinted by Wolfram, that seems like a good choice. >> >>> drivers/i2c/muxes/i2c-mux-pca954x.c | 84 +++++++++++++++++-- >>> 2 files changed, 97 insertions(+), 7 deletions(-) >>> create mode 100644 Documentation/ABI/testing/sysfs-bus-i2c-devices-pca954x >>> >>> diff --git a/Documentation/ABI/testing/sysfs-bus-i2c-devices-pca954x b/Documentation/ABI/testing/sysfs-bus-i2c-devices-pca954x >>> new file mode 100644 >>> index 000000000000..809545a5f9e4 >>> --- /dev/null >>> +++ b/Documentation/ABI/testing/sysfs-bus-i2c-devices-pca954x >>> @@ -0,0 +1,20 @@ >>> +What: /sys/bus/i2c/.../idle_state >>> +Date: January 2019 >>> +KernelVersion: 5.2 >>> +Contact: Robert Shearman <robert.shearman@xxxxxxx> >>> +Description: >>> + Value that can be written to control the behaviour of >>> + the multiplexer on idle. Possible values: >>> + -2 - disconnect on idle, i.e. deselect the last used >>> + channel, which is useful when there is a device >>> + with an address that conflicts with another >>> + device on another pca954x mux on the same parent >> >> Nit: It doesn't need to be a PCA954x mux, it could be any mux. > > Ok, will fix. I'll remove the reference to mux here since it could be > any device and needn't be behind a mux. Oh, but some kind of mux is needed for the situation to occur. If there are no muxes forming an adapter tree, the I2C core will prevent address clashes. So, you need to have the same I2C address on two different branches of the tree, and they cannot be branches on the same mux. In summary, the minimum tree displaying the problem has two parallel muxes. That is, if IIUC... Cheers, Peter
