On 28/8/2023 11:30 am, Alex White wrote: > This patch addresses an issue seen with MicroBlaze in the dl07, dl08, > and dl09 tests by removing the use of floats. The issue relates to the > fact that the MicroBlaze BSPs use software floating point routines. Any > float passed to `printf` is converted to a double automatically because > of default argument promotion with variadic functions. This conversion > is handled by the `__extendsfdf2` libgcc function. The problem is that > the main test executable does not use floating point code, so the > `__extendsfdf2` function is not linked in. This causes a problem when > loading the test objects. > > This patch removes the use of floats so that BSPs that use software > floating point routines can still run the tests.
The floats need to stay to make sure there are no issues with this type. I see the tests with floats as valid. Is the only missing function `__extendsfdf2`? Chris _______________________________________________ devel mailing list devel@rtems.org http://lists.rtems.org/mailman/listinfo/devel
