On Wed, Jun 07, 2017 at 12:46:52PM -0600, Shuah Khan wrote:
> Hi David,
>
> On 05/16/2017 08:12 PM, David Gibson wrote:
> > On Tue, May 16, 2017 at 10:08:08AM -0600, Shuah Khan wrote:
> >> Fix the following compile error found on odroid-xu4:
> >>
> >> checks.c: In function ‘check_simple_bus_reg’:
> >> checks.c:876:41: error: format ‘%lx’ expects argument of type
> >> ‘long unsigned int’, but argument 4 has type
> >> ‘uint64_t{aka long long unsigned int}’ [-Werror=format=]
> >> snprintf(unit_addr, sizeof(unit_addr), "%lx", reg);
> >> ^
> >> checks.c:876:41: error: format ‘%lx’ expects argument of type
> >> ‘long unsigned int’, but argument 4 has type
> >> ‘uint64_t {aka long long unsigned int}’ [-Werror=format=]
> >> cc1: all warnings being treated as errors
> >> Makefile:304: recipe for target 'checks.o' failed
> >> make: *** [checks.o] Error 1
> >>
> >> Signed-off-by: Shuah Khan <shuahkh@xxxxxxxxxxxxxxx>
> >
> > The bug is real, but the fix is only correct for some platforms (just
> > as the original would only compile for some platforms). I've adjusted
> > to %zx which should be right for all platforms and applied.
> This patch with %zx made it into Linux 4.12-rc4 and I am seeing these
> warnings:
>
> HOSTCC scripts/dtc/checks.o
> scripts/dtc/checks.c: In function ‘check_simple_bus_reg’:
> scripts/dtc/checks.c:876:41: warning: format ‘%zx’ expects argument of type ‘size_t’, but argument 4 has type ‘uint64_t {aka long long unsigned int}’ [-Wformat=]
> snprintf(unit_addr, sizeof(unit_addr), "%zx", reg);
> ^
> scripts/dtc/checks.c:876:41: warning: format ‘%zx’ expects argument of type ‘size_t’, but argument 4 has type ‘uint64_t {aka long long unsigned int}’ [-Wformat=]
> HOSTLD scripts/dtc/dtc
> CHK scripts/mod/devicetable-offsets.h
>
>
> build doesn't fail, which is good. This probably needs fixing,
> but since my fix using %llx fails on some platforms, I am just
> reporting the problem, instead of attempting to fix.
*facepalm*
Sorry. I "corrected" your patch from one wrong thing to a different
wrong thing. Apparently I thought the parameter was of type size_t,
rather than uint64_t.
I've now fixed this in my master branch. This time for sure. I hope.
--
David Gibson | I'll have my music baroque, and my code
david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_
| _way_ _around_!
http://www.ozlabs.org/~dgibson
Attachment:
signature.asc
Description: PGP signature
