On Tue, Jun 09, 2020 at 03:22:31AM +0000, Aisheng Dong wrote: > > From: Dan Carpenter <dan.carpenter@xxxxxxxxxx> > > Sent: Monday, June 8, 2020 10:49 PM > > > > On Mon, Jun 08, 2020 at 02:06:35PM +0000, Aisheng Dong wrote: > > > > From: haibo.chen@xxxxxxx <haibo.chen@xxxxxxx> > > > > Sent: Monday, June 8, 2020 6:00 PM > > > > > > > > This patch block system booting, find on imx7d-sdb board. > > > > From the dts we can see, iomux and iomux_lpsr share the memory > > > > region [0x30330000-0x3033ffff], so will trigger the following issue: > > > > > > > > [ 0.179561] imx7d-pinctrl 302c0000.iomuxc-lpsr: initialized IMX pinctrl > > > > driver > > > > [ 0.191742] imx7d-pinctrl 30330000.pinctrl: can't request region for > > > > resource [mem 0x30330000-0x3033ffff] > > > > [ 0.191842] imx7d-pinctrl: probe of 30330000.pinctrl failed with error > > -16 > > > > > > > > This reverts commit ba403242615c2c99e27af7984b1650771a2cc2c9. > > > > Btw, the `git revert` command really sets you up for failure by generating a patch > > in the wrong format. You did well to write a good commit message. I would > > probably also change the subject, the From: > > header and add a Fixes tag and a Signed-off by. The Fixes tag should > > be: > > > > Fixes: ba403242615c ("pinctrl: freescale: imx: Use 'devm_of_iomap()' to avoid a > > resource leak in case of error in 'imx_pinctrl_probe()'") > > > > By searching the kernel log, it seems most people didn't add Fixes tag for a Revert patch. > But anyway, I'm fine to add it. Yeah. It's really complicated to get revert patches right. The revert command was created 15 years ago and it doesn't match what we expect from commits today. Commit 40da7d9a93c8 ("NTB: Revert the change to use the NTB device dev for DMA allocations") is an example of a well written revert commit. I'm sort of surprised that patches where the subject starts with Revert don't break Greg's email sorting scripts. regards, dan carpenter