You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, there are two methods for the RW1C bit: clear_xxx and set_0_xxx. The former sets the bit to 1, the latter sets it to 0.
Originally, only the former existed, because I thought there was no need to write 0 in the RW1C bit. However, as pointed out in #146, there are actually situations where it is necessary to write 0 to prevent the bit from being erased, which is why the latter was added in #148.
I didn't think anything when I merged #148, but later I thought this was obviously confusing. For example, the same clear_xxx method exists for RW bits, but this method sets the corresponding bit to 0.
Should I fix this and have the RW1C bits also have set_xxx/clear_xxx methods that actually set the bits to 1/0?
One problem is that the clear_xxx method will have the exact opposite meaning in a later version of this change, and thus must be clarified in the changelog.