a) Open-Drain output of the IRQ lane (see inside for possible caveats). b) Improvements for multi-instances on the same SPI CS. This chip is "special" because it supports several "instances" sharing the same SPI CS lane. Each chip has three HW pins for specifying an internal address. The first byte of each SPI transaction then distinguishes which chip we're talking to. So far, so good -- that's something that the driver already supports. The actual implementation simply instantiates several regmap and pinctrl instances sharing the same parent (the spi/spiX.Y dev). That leads to clashes as soon as more HW chips are present because of naming clashes within debugfs. Fixing this for regmap is pretty straightforward. The patch which implements something similar for the pinctrl subsystem is a bit more involved because it touches the pinctrl core. The resulting names are ugly and the `strcmp` in there is largely a hack in order to not force even more ugliness to the existing chips. Jan Kundrát (4): pinctrl: mcp23s08: IRQ behavior for open-drain interrupts pinctrl: mcp23s08: spi: Fix regmap debugfs entries pinctrl: mcp23s08: spi: Add HW address to gpio_chip.label pinctrl: mcp23s08: spi: Fix duplicate pinctrl debugfs entries .../bindings/pinctrl/pinctrl-mcp23s08.txt | 4 + drivers/pinctrl/core.c | 18 ++- drivers/pinctrl/pinctrl-mcp23s08.c | 131 +++++++++++++++------ 3 files changed, 114 insertions(+), 39 deletions(-) -- 2.14.3 -- To unsubscribe from this list: send the line "unsubscribe linux-spi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html