Hi Guys,
OK - how about this reformulation of the pr61886 test ?
The patch changes references to __fread_chk with references to just
fread, which I assume will be present in all target runtime libraries. I
had to add some preprocessor trickery in order to ensure that
__USER_LABEL_PREFIX__ is correctly prepended to the assembler name of
the functions, but other than that the test remains the same. No linker
funny business this time.
I have tested this patch with an x86_64 native (whose C runtime does
include __fread_chk), an ARM cross compiler (using newlib, which does
not provide __fread_chk) and with a V850 cross compiler (which uses
newlib and which also defines __USER_LABEL_PREFIX to "_"). Prior to
Honza's r231548 patch all three toolchains fail this patched test. Using
today#s latest and greatest mainline gcc sources, all three toolchains
pass the patched test.
OK to apply ?
Cheers
Nick
gcc/testsuite/ChangeLog
2016-01-08 Nick Clifton <nickc@redhat.com>
PR target/68913
* gcc.dg/lto/pr61886_0.c: Rename the external function called
to fread so that it will be found in all target runtimes.