Hello, This patch set adds the support for the i2c gpio expander pca9505 used on the JTAG/GPIO box which can be connected to the Mirabox.
To be able to use the pca9505 I had to do several changes in the driver. Indeed, until now the pca953x driver accessed all the bank of a given register in a single command using only a 32 bits variable. This expander comes with 40 GPIOs which no more fits in a 32 variable. This patch set makes the accesses to the registers more generic by relying on an array of u8 variables. This fits exactly the way the registers are represented in the hardware. Once the per-bank representation was added, it was easier to introduce helpers to access to a single register of a bank instead of reading or writing all the banks for a given register. As the GPIO API allows only the accesses to a single GPIO at a time there was no point to read and write all the other banks. Hence it should help to decrease the latency especially for the pca9505. However as the block GPIO API from Roland Stigge is incoming I kept the helpers used to access all the banks in the same time. I had to make some modifications in the arguments that these functions received, so it will have a conflict here. Currently my patch set is based on v3.8-rc4, but I am willing to rebase onto gpio-for-next once the GPIO block will be merged into it. After the feedback from Maxime Ripard on the second version, I continue to abuse of his good will as I didn't get any information on the Mirabox about the interrupt pin. Thanks to him I managed to fix the bug related to the interrupt part. Now the IRQ part have been also validated! Thanks again to Maxime for his help. I have also updated the branch gpio-pca9505 available at: https://github.com/MISL-EBU-System-SW/mainline-public.git Linus, now you can apply the first two patches on your git tree. I kept your acked-by for the 3rd one, as I didn't change anything for this one. Thanks, Changelog: V1->V2: - Fix the way to calculate the shift used to apply to register to access a given bank. - Fix all the pending issue in the IRQ part which appeared once I have enable CONFIG_GPIO_PCA953X_IRQ! V2->V3: - Rebased on v3.8-rc4 - Fix the interrupt handler and not try to call an handler if there is no irq pending on a gpio bank Gregory CLEMENT (3): gpio: pca953x: make the register access by GPIO bank gpio: pca953x: add support for pca9505 arm: mvebu: enable gpio expander over i2c on Mirabox platform arch/arm/boot/dts/armada-370-mirabox.dts | 10 ++ drivers/gpio/gpio-pca953x.c | 283 +++++++++++++++++++----------- 2 files changed, 187 insertions(+), 106 deletions(-) -- 1.7.9.5 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
