On 7/1/22, Randy Dunlap <rdunlap@xxxxxxxxxxxxx> wrote: > > > On 7/1/22 15:20, Martin Fernandez wrote: >> Add a description to the MACRO_ARG_REUSE check. >> >> I feel like this is also a good place to put a workaround although I'm >> not sure if there is a cannonical way to solve those kinds of issues. > > canonical > > > The usual way in the kernel is to declare a local _x and local _y (for your > example below). See how it is done in include/linux/minmax.h for > min_not_zero(). >> >> Signed-off-by: Martin Fernandez <martin.fernandez@xxxxxxxxxxxxx> >> --- >> Documentation/dev-tools/checkpatch.rst | 15 +++++++++++++++ >> 1 file changed, 15 insertions(+) >> >> diff --git a/Documentation/dev-tools/checkpatch.rst >> b/Documentation/dev-tools/checkpatch.rst >> index b52452bc2963..43fa99f188f5 100644 >> --- a/Documentation/dev-tools/checkpatch.rst >> +++ b/Documentation/dev-tools/checkpatch.rst >> @@ -759,6 +759,21 @@ Indentation and Line Breaks >> Macros, Attributes and Symbols >> ------------------------------ >> >> + **ARG_REUSE** >> + Using the same argument multiple times in the macro definition >> + would lead to unwanted side-effects. >> + >> + For example, given a `min` macro defined like:: >> + >> + #define min(x, y) ((x) < (y) ? (x) : (y)) >> + >> + If you call it with `min(foo(x), 0)` would expand to:: >> + >> + foo(x) < 0 ? foo(x) : 0 >> + >> + If `foo` have side-effects or it's an expensive calculation the >> + results might not be what the user inteded. > > intended. > >> + >> **ARRAY_SIZE** >> The ARRAY_SIZE(foo) macro should be preferred over >> sizeof(foo)/sizeof(foo[0]) for finding number of elements in an Thank you for the feedback, I'll send a new one right away! > thanks. > -- > ~Randy >
