On 9/19/22 6:40 PM, Leo Li wrote: > > >> -----Original Message----- >> From: Sean Anderson <sean.anderson@xxxxxxxx> >> Sent: Monday, September 19, 2022 5:24 PM >> To: Oleksij Rempel <linux@xxxxxxxxxxxxxxxx>; Pengutronix Kernel Team >> <kernel@xxxxxxxxxxxxxx>; linux-i2c@xxxxxxxxxxxxxxx; linux-arm-kernel >> <linux-arm-kernel@xxxxxxxxxxxxxxxxxxx>; Vinod Koul <vkoul@xxxxxxxxxx>; >> dmaengine@xxxxxxxxxxxxxxx >> Cc: Sumit Semwal <sumit.semwal@xxxxxxxxxx>; Christian König >> <christian.koenig@xxxxxxx>; Linux Kernel Mailing List <linux- >> kernel@xxxxxxxxxxxxxxx>; linux-media@xxxxxxxxxxxxxxx; dri- >> devel@xxxxxxxxxxxxxxxxxxxxx; linaro-mm-sig@xxxxxxxxxxxxxxxx; Joy Zou >> <joy.zou@xxxxxxx>; Peng Ma <peng.ma@xxxxxxx>; Robin Gong >> <yibin.gong@xxxxxxx>; Shawn Guo <shawnguo@xxxxxxxxxx>; Leo Li >> <leoyang.li@xxxxxxx> >> Subject: [BUG] ls1046a: eDMA does not transfer data from I2C >> >> Hi all, >> >> I discovered a bug in either imx_i2c or fsl-edma on the LS1046A where no >> data is read in i2c_imx_dma_read except for the last two bytes (which are >> not read using DMA). This is perhaps best illustrated with the following >> example: > > What is the kernel tree/tag that you are testing with? 5.15.40 Sorry, I forgot to put that in the first email. I can test with linux/master, but I reviewed the different commits and I did not think there would be any difference. I also cherry-picked the eDMA memcpy commit, but the dma test passed. --Sean
