Re: [PATCH v3 4/4] selftests/nolibc: Add tests for strlcat() and strlcpy()

[Paul E. McKenney]

On Tue, Apr 23, 2024 at 06:28:31PM +0200, Willy Tarreau wrote:
> Hi Thomas,
> 
> On Tue, Apr 23, 2024 at 11:18:06AM +0200, Thomas Weißschuh wrote:
> > + Shuah and Paul, please see below.
> > 
> > So I picked this up and it got picked up by Shuah, but...
> > 
> > On 2024-02-18 16:51:06+0000, Rodrigo Campos wrote:
> > > I've verified that the tests matches libbsd's strlcat()/strlcpy()
> > > implementation.
> > > 
> > > Please note that as strlcat()/strlcpy() are not part of the libc, the
> > > tests are only compiled when using nolibc.
> > > 
> > > Signed-off-by: Rodrigo Campos <rodrigo@xxxxxxxxxxx>
> > > ---
> > >  tools/testing/selftests/nolibc/nolibc-test.c | 40 ++++++++++++++++++++
> > >  1 file changed, 40 insertions(+)
> > > 
> > > diff --git a/tools/testing/selftests/nolibc/nolibc-test.c b/tools/testing/selftests/nolibc/nolibc-test.c
> > > index 6ba4f8275ac4..d373fc14706c 100644
> > > --- a/tools/testing/selftests/nolibc/nolibc-test.c
> > > +++ b/tools/testing/selftests/nolibc/nolibc-test.c
> > > @@ -600,6 +600,25 @@ int expect_strne(const char *expr, int llen, const char *cmp)
> > >  	return ret;
> > >  }
> > >  
> > > +#define EXPECT_STRBUFEQ(cond, expr, buf, val, cmp)				\
> > > +	do { if (!(cond)) result(llen, SKIPPED); else ret += expect_str_buf_eq(expr, buf, val, llen, cmp); } while (0)
> > > +
> > > +static __attribute__((unused))
> > > +int expect_str_buf_eq(size_t expr, const char *buf, size_t val, int llen, const char *cmp)
> > > +{
> > > +	llen += printf(" = %lu <%s> ", expr, buf);
> > 
> > This introduces a compiler warning on 32bit:
> > 
> >     i386-linux-gcc -Os -fno-ident -fno-asynchronous-unwind-tables -std=c89 -W -Wall -Wextra -fno-stack-protector -m32 -mstack-protector-guard=global -fstack-protector-all  -o nolibc-test \
> >       -nostdlib -nostdinc -static -Isysroot/i386/include nolibc-test.c nolibc-test-linkage.c -lgcc
> >     nolibc-test.c: In function 'expect_str_buf_eq':
> >     nolibc-test.c:610:30: error: format '%lu' expects argument of type 'long unsigned int', but argument 2 has type 'size_t' {aka 'unsigned int'} [-Werror=format=]
> >       610 |         llen += printf(" = %lu <%s> ", expr, buf);
> >           |                            ~~^         ~~~~
> >           |                              |         |
> >           |                              |         size_t {aka unsigned int}
> >           |                              long unsigned int
> >           |                            %u
> > 
> > 
> > It is easy enough to fix through a cast to "unsigned long".
> 
> Yes, that's usually what I do as well with size_t everywhere as well.
> 
> > The original patch was already sent to Shuah and included in -next.
> > 
> > Shuah, Paul:
> > 
> > I'd like to rewrite the offending commit instead of having a fixup commit.
> > As it seems to me the original patch would only go into 6.10 anyways.
> > 
> > Any objections?
> 
> I, too, think it would be the cleanest history-wise, I just don't know
> if that's the easiest for Shuah if it requires to change already published
> commit IDs.

One way would be to resend the series and have Shuah replace what she
has with the new series.  But I don't know enough about Shuah's process
to judge.

							Thanx, Paul

> > Notes to self:
> > 
> > * Add flag to run-tests.sh to use -Werror
> > * Implement "%zu" in nolibc printf()
> 
> Could be nice indeed!
> 
> Thanks Thomas for taking care of this!
> Willy