This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
Other format: | [Raw text] |
On 05/12/16 09:01, Christophe Lyon wrote:
On 2 December 2016 at 17:35, Thomas Preudhomme <thomas.preudhomme@foss.arm.com> wrote:Hi, The logic to make -mthumb optional for Thumb-only targets was designed to only apply when no mode is specified either at compile time or when the toolchain was configured (using --with-mode). The dg-skip-if in the testcases optional_thumb-1 and optional_thumb-2 catch the former case but not the latter. Unfortunately there is no dejagnu procedure to check the presence of an option in the configure line used for the compiler. This patch marks the two tests as XFAIL since that will at least allow to catch a change from PASS to XFAIL, thus making the test still useful. It also fixes a codestyle issue in the target triplet of the dg-skip-if directives of all the optional_thumb-* testcases.So, this makes the test XPASS when no --with-mode or --with-mode=thumb is used.
Yes. At least we can still see change from XPASS to XFAIL and not let people worry if they build GCC with --with-mode=* and the test fails.
Isn't it possible to add an dejagnu effective target to detect this configure option? I guess you looked at that but the effort is not worth it?
Since we can run the compiler to compile testcases it should be possible to run it with -v and then check the output for something. However that would involve more code so would take me more time to come up with a solution and I want to have a solution for the short term.
Best regards, Thomas
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |