This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [PATCH] fix for aarch64 sim tbnz bug
- From: Jim Wilson <jim dot wilson at linaro dot org>
- To: Nick Clifton <nickc at redhat dot com>
- Cc: "gdb-patches at sourceware dot org" <gdb-patches at sourceware dot org>
- Date: Mon, 12 Dec 2016 10:38:10 -0800
- Subject: Re: [PATCH] fix for aarch64 sim tbnz bug
- Authentication-results: sourceware.org; auth=none
- References: <CABXYE2W++FWtcyNq4FAroJ2+VNJOHWs7-sdHMBm=6kOwSkD5Ug@mail.gmail.com> <CAJaTeTpn=Y_PYzufM1pkMbnp0J9jqoYYx=17UQJauY9VLq5n0g@mail.gmail.com> <CABXYE2WEVB=Bt=+uX7Sx6a9ZE4etGwFODLKnyeOzW52P+Rhe3w@mail.gmail.com>
ping
https://sourceware.org/ml/gdb-patches/2016-12/msg00128.html
On Sun, Dec 4, 2016 at 7:43 PM, Jim Wilson <jim.wilson@linaro.org> wrote:
> On Fri, Dec 2, 2016 at 1:31 AM, Mike Frysinger <vapier@gentoo.org> wrote:
>> can we please start getting tests added to sim too ? using gcc
>> indirectly to validate the sim is a bit un
>
> The GCC testsuite is the part that I care most about, but it does make
> sense for me to be adding simulator tests. There are currently no
> interesting aarch64 tests, so it isn't clear if there is any
> particular style I should be using. So I just chose a style that
> seemed OK to me. I have tests for the last two simulator patches I
> contributed. I verified that the tests work with current sources and
> fail without my two patches.
>
> I had to make a few fixes to the existing testsuite.inc file. I moved
> .Lpass and .Lfail out of the pass and fail macros, as otherwise I end
> up with duplicate definitions. I also changed fail to return
> non-zero, because it should. I find it confusing to manually run a
> testcase and see it print fail, and then return a zero exit code.
>
> Jim
>
> PS set_flags_for_add64 is broken, which causes adds and cmn to fail.
> I have a fix, and this will be my next aarch64 sim patch, but I
> haven't written a testcase yet.