Hi Jean Thanks very much for reviewing this patch, you have given me some useful insights. Please consider the revised patch which addresses the issues you raise. Jean Delvare wrote: >Hi Peter, > > >>Avoid addressing self when sending a Master address. Follows instruction >>in Intel 80331/80321 manuals. >>Ignoring this worked previously on 80321, but causes a hang on i2cdetect >>on 80331 >> >> > >Can't you simply disable the slave address? You don't seem to use it >anyway. > > Unfortunately there doesn't seem to be any way to disable the slave address. >BTW, you must be using an old version of i2cdetect, as the slave address >appears to be forced to 0x02 in the i2c-iop3xx driver (not a smart >choice...) and i2cdetect does no more scan this address by default. > > OK, new patch leaves the slave address at default 0. I note that i2cdetect now scans from 0x3, but I'd still prefer to prevent writes to the device's own address since this wedges the unit. >>+#define I2C_ERR_WRITEMYSAR (I2C_ERR+10) >> >> > >Please use tabs, not spaces, for alignment. > > After forcing emacs to cooperate, tabs it is. Thanks Peter. -- Peter Milne Peter.Milne at d-tacq.com D-TACQ Solutions Ltd www.d-tacq.com -------------- next part -------------- A non-text attachment was scrubbed... Name: i2c-iop3xx.2.patch Type: text/x-patch Size: 2430 bytes Desc: not available Url : http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20060618/4d1092ad/attachment.bin
