On 1/14/21 4:18 PM, H.J. Lu via Gcc-patches wrote:
> On Thu, Jan 14, 2021 at 6:51 AM Uros Bizjak <ubiz...@gmail.com> wrote:
>>
>> On Thu, Jan 14, 2021 at 3:05 PM H.J. Lu <hjl.to...@gmail.com> wrote:
>>>
>>> -fcf-protection with CF_BRANCH inserts ENDBR32 at function entries.
>>> ENDBR32 is NOP only on 64-bit processors and 32-bit TARGET_CMOVE
>>> processors. Issue an error for -fcf-protection with CF_BRANCH when
>>> compiling for 32-bit non-TARGET_CMOVE targets.
>>>
>>> gcc/
>>>
>>> PR target/98667
>>> * config/i386/i386-options.c (ix86_option_override_internal):
>>> Issue an error for -fcf-protection with CF_BRANCH when compiling
>>> for 32-bit non-TARGET_CMOVE targets.
>>>
>>> gcc/testsuite/
>>>
>>> PR target/98667
>>> * gcc.target/i386/pr98667-1.c: New file.
>>> * gcc.target/i386/pr98667-2.c: Likewise.
>>> * gcc.target/i386/pr98667-3.c: Likewise.
>>> ---
>>> gcc/config/i386/i386-options.c | 9 ++++++++-
>>> gcc/testsuite/gcc.target/i386/pr98667-1.c | 9 +++++++++
>>> gcc/testsuite/gcc.target/i386/pr98667-2.c | 9 +++++++++
>>> gcc/testsuite/gcc.target/i386/pr98667-3.c | 7 +++++++
>>> 4 files changed, 33 insertions(+), 1 deletion(-)
>>> create mode 100644 gcc/testsuite/gcc.target/i386/pr98667-1.c
>>> create mode 100644 gcc/testsuite/gcc.target/i386/pr98667-2.c
>>> create mode 100644 gcc/testsuite/gcc.target/i386/pr98667-3.c
>>>
>>> diff --git a/gcc/config/i386/i386-options.c b/gcc/config/i386/i386-options.c
>>> index 4e0165ff32c..1489871b36f 100644
>>> --- a/gcc/config/i386/i386-options.c
>>> +++ b/gcc/config/i386/i386-options.c
>>> @@ -3016,8 +3016,15 @@ ix86_option_override_internal (bool main_args_p,
>>> }
>>>
>>> if (opts->x_flag_cf_protection != CF_NONE)
>>> - opts->x_flag_cf_protection
>>> + {
>>> + if ((opts->x_flag_cf_protection & CF_BRANCH) == CF_BRANCH
>>> + && !TARGET_64BIT
>>> + && !TARGET_CMOVE)
>>
>> You need TARGET_CMOV (note, no E) here. Also, please put both tests on one
>> line.
>>
>> LGTM with the above change.
>
> This is the patch I am checking in.
I might be doing something wrong, but this breaks the -m32 multilib build for
me. looking at a trunk 20210110 build, -fcf-protection -mshstk are passed to the
-m32 build as well, and now errors out.
Matthias
