Since switching from bfd to lld linker there are no warnings about
insecure libc functions anymore. To avoid false positives, change
the regress result from FAILED to EXPECTED_FAIL if lld is used.
When lld will support this feature, the regress tests can be enabled
again.
OK deraadt@

Skip the gcc tests on aarch64, the compiler does not exist there.

Adapt more places where binutils-2.17 has changed the format of the
linker warning massages.

Format of warning messages has changed in binutils-2.17. Remove
variable line number in text segment. Adapt expected gcc linker
output in regress test.

- adjust getcwd-1.c.exp.gcc{3,4} after the getwd linker warning got added
- adjust gcc-builtins to use the same approach as gcc-bounds
- enable gcc-bounds and gcc-builtins in the default regression suite
now that everything passes

Remove support for COMPILER_VERSION == gcc2.
Change the logic depending upon COMPILER_VERSION everywhere, to assume gcc4
is the norm and to explicitely test for gcc3 when a different behaviour
is required.
No functional change intended. Be sure to `make install' in share/mk before
attempting to do anything.

add failing regress test for gcc3 -Wbounded for 0-length arrays, spotted by nicm@

Switch to using COMPILER_VERSION instead of USE_GCC3, allows for more flexible
complier picking, eg supporting gcc2, gcc3, and gcc4. based on diff by Marco,
with fixes from espie@.
ok espie@ general mumbling of approval of others.

add another regress for the recent -Wbounded fixes

check for variable sized array in format arguments

NOOBJ=1, needed since gcc-bounds was connected to regress, noticed by david@

regress check for macro presence/failure in bounded attrib declaration

fill in the blanks for gcc2

Lots more regression tests to test builtins for gcc3 and various icky
corner cases. need to separate gcc2/gcc3 output expectations since
the error formats have diverged.

add a few regression tests for the recently fixed md functions

no longer warn on '0' bound size, add new test for the negative case

- fix makefile targets so c++ targets do get checked correctly
- update sscanf-cpp regress results to current error format
- add checks for variable length arrays to make sure they are ignored

various regression tests for -Wbounded

Skip the gcc tests on aarch64, the compiler does not exist there.

Adapt more places where binutils-2.17 has changed the format of the
linker warning massages.

Format of warning messages has changed in binutils-2.17. Remove
variable line number in text segment. Adapt expected gcc linker
output in regress test.

- adjust getcwd-1.c.exp.gcc{3,4} after the getwd linker warning got added
- adjust gcc-builtins to use the same approach as gcc-bounds
- enable gcc-bounds and gcc-builtins in the default regression suite
now that everything passes

Remove support for COMPILER_VERSION == gcc2.
Change the logic depending upon COMPILER_VERSION everywhere, to assume gcc4
is the norm and to explicitely test for gcc3 when a different behaviour
is required.
No functional change intended. Be sure to `make install' in share/mk before
attempting to do anything.

add failing regress test for gcc3 -Wbounded for 0-length arrays, spotted by nicm@

Switch to using COMPILER_VERSION instead of USE_GCC3, allows for more flexible
complier picking, eg supporting gcc2, gcc3, and gcc4. based on diff by Marco,
with fixes from espie@.
ok espie@ general mumbling of approval of others.

add another regress for the recent -Wbounded fixes

check for variable sized array in format arguments

NOOBJ=1, needed since gcc-bounds was connected to regress, noticed by david@

regress check for macro presence/failure in bounded attrib declaration

fill in the blanks for gcc2

Lots more regression tests to test builtins for gcc3 and various icky
corner cases. need to separate gcc2/gcc3 output expectations since
the error formats have diverged.

add a few regression tests for the recently fixed md functions

no longer warn on '0' bound size, add new test for the negative case

- fix makefile targets so c++ targets do get checked correctly
- update sscanf-cpp regress results to current error format
- add checks for variable length arrays to make sure they are ignored

various regression tests for -Wbounded

Adapt more places where binutils-2.17 has changed the format of the
linker warning massages.

Format of warning messages has changed in binutils-2.17. Remove
variable line number in text segment. Adapt expected gcc linker
output in regress test.

- adjust getcwd-1.c.exp.gcc{3,4} after the getwd linker warning got added
- adjust gcc-builtins to use the same approach as gcc-bounds
- enable gcc-bounds and gcc-builtins in the default regression suite
now that everything passes

Remove support for COMPILER_VERSION == gcc2.
Change the logic depending upon COMPILER_VERSION everywhere, to assume gcc4
is the norm and to explicitely test for gcc3 when a different behaviour
is required.
No functional change intended. Be sure to `make install' in share/mk before
attempting to do anything.

add failing regress test for gcc3 -Wbounded for 0-length arrays, spotted by nicm@

Switch to using COMPILER_VERSION instead of USE_GCC3, allows for more flexible
complier picking, eg supporting gcc2, gcc3, and gcc4. based on diff by Marco,
with fixes from espie@.
ok espie@ general mumbling of approval of others.

add another regress for the recent -Wbounded fixes

check for variable sized array in format arguments

NOOBJ=1, needed since gcc-bounds was connected to regress, noticed by david@

regress check for macro presence/failure in bounded attrib declaration

fill in the blanks for gcc2

Lots more regression tests to test builtins for gcc3 and various icky
corner cases. need to separate gcc2/gcc3 output expectations since
the error formats have diverged.

add a few regression tests for the recently fixed md functions

no longer warn on '0' bound size, add new test for the negative case

- fix makefile targets so c++ targets do get checked correctly
- update sscanf-cpp regress results to current error format
- add checks for variable length arrays to make sure they are ignored

various regression tests for -Wbounded