Fwd: GCC 7.5 Status Report (2019-10-22)
chrisj at rtems.org
Thu Oct 24 05:10:51 UTC 2019
On 24/10/19 4:03 pm, Sebastian Huber wrote:
> On 24/10/2019 07:00, Chris Johns wrote:
>> On 23/10/19 5:18 pm, Sebastian Huber wrote:
>>> On 22/10/2019 14:45, Joel Sherrill wrote:
>>>> I would like to see most targets moved to newer major versions but I think at
>>>> least the PowerPC will want to bump for the upcoming gcc 7 release.
>>> Yes. For all targets except powerpc I would directly move to GCC 10.
>> I am not sure about a change without some evidence there are no issues. Building
>> on all supported hosts would be a good start, test results on hardware even
> If we are close to the RTEMS 5 release, then I would not change the GCC baseline.
I would like to start the process to converge on a release. I am attempting to
make space between now and the end of the year.
The final point may depend on the work you are doing and it's schedule.
>>>> Binutils has a new release and I think we also have changes to newlib to
>>>> pick up.
>>> Some problems on ARM were mentioned on the Binutils mailing list. I am not sure
>>> what the status is here.
>> Can binutils be upgraded?
> It seems the ARM issue turned out to be not a Binutils problem. So yes, we can
>>>> For newlib, I'm caught up at the moment. Not sure about Sebastian.
>>> I wait since one month for a Newlib patch approval.
>> Is this important to 5?
> Yes, it is quite important if we want to keep the libbsd up to date.
Is it just not review and so would a ping help?
More information about the devel