Re: [PATCH 06/25] Add struct scratch_buffer and its internal helper functions

[Stefan Liebler <stli at linux dot vnet dot ibm dot com>]

On 04/07/2015 05:49 PM, Florian Weimer wrote:
> On 04/07/2015 04:33 PM, Florian Weimer wrote:
> > On 04/07/2015 03:18 PM, Stefan Liebler wrote:
> > > Hi Florian,
> > >
> > > i get a build error on s390x:
> > > gcc scratch_buffer_grow_preserve.c -c
> > > ...
> > > scratch_buffer_grow_preserve.c: In function
> > > â__libc_scratch_buffer_grow_preserveâ:
> > > scratch_buffer_grow_preserve.c:35:7: error: implicit declaration of
> > > function âmemcpyâ [-Werror=implicit-function-declaration]
> > >         memcpy (new_ptr, buffer->__space, buffer->length);
> > >         ^
> > > scratch_buffer_grow_preserve.c:35:7: error: incompatible implicit
> > > declaration of built-in function âmemcpyâ [-Werror]
> > > cc1: all warnings being treated as errors
> > >
> > > #include <string.h> in malloc/scratch_buffer_grow_preserve.c solves this
> > > error. Please update scratch_buffer_grow_preserve.c.
> >
> > Sorry about that.  I did build on ppc64, but building on s390x is
> > difficult for me due to the binutils 2.24 requirement.  I'm trying to
> > get a suitable machine reservation, but I'm not sure if that will succeed.
> >
> > Should I commit this fix blindly, without testing it on s390x?
>
> I found a machine to test this on, so I committed the attached patch.
Thanks.

I have another issue with tst-scratch_buffer.c on s390-32 where size_t 
is an unsigned long with only 4 bytes:
gcc tst-scratch_buffer.c -c
...
tst-scratch_buffer.c: In function âdo_testâ:
tst-scratch_buffer.c:133:8: error: large integer implicitly truncated to 
unsigned type [-Werror=overflow]
        && unchanged_array_size (&buf, 1ULL << 32, 0)
        ^
tst-scratch_buffer.c:134:8: error: large integer implicitly truncated to 
unsigned type [-Werror=overflow]
        && unchanged_array_size (&buf, 0, 1ULL << 32)))
        ^
cc1: all warnings being treated as errors

Can you change the test and use "1ULL << 31" or make the usage of "1ULL 
<< 32" conditionally?

Bye Stefan