Hi Tobias,
Am 29.03.22 um 09:14 schrieb Tobias Burnus:
Hi Harald,
On 28.03.22 22:03, Harald Anlauf via Fortran wrote:
All current cases of printing a HOST_WIDE_INT in gcc/fortran/ use
'sprintf', and I did not find any other use of %wd/%wu. So the
mentioned implementation is not really stressed yet... ;-)
That's all your fault ;-)
true; I'm pleading guilty for that one.
(Your commit
https://gcc.gnu.org/r12-3273-ge4cb3bb9ac11b4126ffa718287dd509a4b10a658
did remove the only user.)
I've now made good for it. ;-)
That's only a warning. Have you tried whether it works at runtime?
My bet is that it does!
Yes, it did work, it was just the warning alerting me ...
Question: Do you build with --disable-bootstrap ? Or do you do a proper
bootstrap?
... because I did build with --disable-bootstrap to save on time for
building the compiler on my local machine, and the system's default
gcc is older.
Can you check & try again? I don't mind getting a format warning with
GCC < GCC 12. But with GCC 12 compiled (either installed compiler or
when bootstrapping) it should compile without errors.
If you can confirm my suspicion, the patch LGTM.
I've just pushed that version as
https://gcc.gnu.org/g:0712f356374c2cf26015cccfa3141537e42cbb12
Sorry for the noise, and thanks for the review!
Harald
Tobias
