]> asedeno.scripts.mit.edu Git - linux.git/commit
PCI: imx6: Add reset-gpio-active-high boolean property to DT
authorPetr Štetiar <ynezz@true.cz>
Wed, 20 Apr 2016 00:42:07 +0000 (19:42 -0500)
committerBjorn Helgaas <bhelgaas@google.com>
Wed, 20 Apr 2016 00:42:07 +0000 (19:42 -0500)
commit3ea8529acc30467331b7b1f5da6cf668fac091c7
tree2d08f8d8e392a3badf9e25e1f78932498b0c3d16
parent4f6926e9fdf94208dc37da0cd17835328fa4dbba
PCI: imx6: Add reset-gpio-active-high boolean property to DT

Currently the reset-gpio DT property which controls the PCI bus device
reset signal defaults to active-low reset sequence (L=reset state,
H=operation state) plus the code in reset function isn't GPIO polarity
aware - it doesn't matter if the defined reset-gpio is active-low or
active-high, it will always result into active-low reset sequence.

I've tried to fix it properly and change the reset-gpio reset sequence to
be polarity-aware, but this patch has been accepted and then reverted as it
has introduced few backward incompatible issues:

1. Some DTBs, for example, imx6qdl-sabresd, don't define reset-gpio
polarity correctly:

  reset-gpio = <&gpio7 12 0>;

which means that it's defined as active-high, but in reality it's
active-low; thus it wouldn't work without a DTS fix.

2. The logic in the reset function is inverted:

gpio_set_value_cansleep(imx6_pcie->reset_gpio, 0)
msleep(100);
gpio_set_value_cansleep(imx6_pcie->reset_gpio, 1);

so even if some of the i.MX6 boards had reset-gpio polarity defined
correctly in their DTSes, they would stop working.

As we can't break old DTBs, we can't fix them, so we need to introduce this
new DT reset-gpio-active-high boolean property so we can support boards
with active-high reset sequence.

This active-high reset sequence is for example needed on Apalis SoMs, where
GPIO1_IO28, used to PCIe reset is not connected directly to PERST# PCIe
signal, but it's ORed with RESETBMCU coming off the PMIC, and thus is
inverted, active-high.

Tested-by: Tim Harvey <tharvey@gateworks.com> # Gateworks Ventana boards (which have active-low PERST#)
Signed-off-by: Petr Štetiar <ynezz@true.cz>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Lucas Stach <l.stach@pengutronix.de>
Acked-by: Rob Herring <robh@kernel.org>
Documentation/devicetree/bindings/pci/fsl,imx6q-pcie.txt
drivers/pci/host/pci-imx6.c