On Mon, 6 Mar 2023 at 13:23, Peter Maydell <peter.mayd...@linaro.org> wrote: > > On Mon, 20 Feb 2023 at 08:13, <qianfangui...@163.com> wrote: > > > > From: qianfan Zhao <qianfangui...@163.com> > > > > Next is an example when read/write trace enabled: > > > > allwinner_i2c_write write XADDR(0x04): 0x00 > > allwinner_i2c_write write CNTR(0x0c): 0x50 M_STP BUS_EN > > allwinner_i2c_write write CNTR(0x0c): 0xe4 A_ACK M_STA BUS_EN INT_EN > > allwinner_i2c_read read CNTR(0x0c): 0xcc A_ACK INT_FLAG BUS_EN INT_EN > > allwinner_i2c_read read STAT(0x10): 0x08 STAT_M_STA_TX > > This seems like overkill to me. We don't do it for > any other devices. If we did want to do it we should > do it by providing a generic mechanism somewhere that > it's easy for devices to use, not by implementing a lot > of string operations inside the code for this one device.
I disagree with this patch, but I've taken patches 2 and 3 into target-arm.next for 8.0. thanks -- PMM
