On 16/12/2020 19:02, Dave Nadler wrote:
> On 12/14/2020 3:07 AM, David Brown wrote:
>> On 12/12/2020 16:48, Dave Nadler wrote:
>>> Anybody ever posted bug reports on GCC as built+distributed by ARM?
>>> I posted what I thought was a decently written (and very real) bug
>>> report, but I have never received any response at all:
>>> https://bugs.launchpad.net/gcc-arm-embedded/+bug/1905459
>>> Any ideas?
>>> Thanks,
>>> Best Regards, Dave
>>
>> I have only reported one bug, and you can see how well it has been
>> answered:
>>
>> <https://bugs.launchpad.net/gcc-arm-embedded/+bug/1646883>
>>
>> Bugs and questions /do/ get answered and fixed on the tracker (I've made
>> a few helpful comments myself to some).&nbsp; But I don't believe the
>> responsible people have anything close to the time and manpower needed
>> to handle the support.&nbsp; They can fix some of the easy stuff.&nbsp; Your issue
>> here - clearly written and important though it is - is far from easy
>> stuff.
>>
>> It is good to make these bug reports - it stops things getting
>> forgotten, and it makes it much easier for other people to find out
>> about them.&nbsp; However, it is normally best to find a workaround for them
>> rather than waiting for the bug to be fixed.
> 
> I'm confused - 3 years on, you did the patch yourself?

No.  The patch was done in mainline gcc by a gcc developer from ARM,
after it was mentioned (by me) in gcc help mailing list discussion on a
related topic.  I merely reported it back to the gcc arm embedded bug
report.  You can follow the links from that report, if you are
interested.  (Perhaps - based on that experience - you might have luck
mentioning it on the gcc help mailing list.  Along with the gcc
developers' mailing list, that's where most of the actual gcc developers
hang out - even the ones employed by ARM.)

> 
> In the bug I'm troubled by, there is *no coding required*
> (unless you count creating a module with 6 empty functions coding).
> It DOES require messing with the build configuration to use the no-op
> module to build the ARM library...
> 
> I guess I could try do it myself, but I fear days of entanglement
> with baroque layered build components, plus days trying to set up a
> suitable build VM...
> 
> Thanks,
> Best Regards, Dave

On 12/14/2020 3:07 AM, David Brown wrote:
> On 12/12/2020 16:48, Dave Nadler wrote:
>> Anybody ever posted bug reports on GCC as built+distributed by ARM?
>> I posted what I thought was a decently written (and very real) bug
>> report, but I have never received any response at all:
>> https://bugs.launchpad.net/gcc-arm-embedded/+bug/1905459
>> Any ideas?
>> Thanks,
>> Best Regards, Dave
> 
> I have only reported one bug, and you can see how well it has been answered:
> 
> <https://bugs.launchpad.net/gcc-arm-embedded/+bug/1646883>
> 
> Bugs and questions /do/ get answered and fixed on the tracker (I've made
> a few helpful comments myself to some).  But I don't believe the
> responsible people have anything close to the time and manpower needed
> to handle the support.  They can fix some of the easy stuff.  Your issue
> here - clearly written and important though it is - is far from easy stuff.
> 
> It is good to make these bug reports - it stops things getting
> forgotten, and it makes it much easier for other people to find out
> about them.  However, it is normally best to find a workaround for them
> rather than waiting for the bug to be fixed.

I'm confused - 3 years on, you did the patch yourself?

In the bug I'm troubled by, there is *no coding required*
(unless you count creating a module with 6 empty functions coding).
It DOES require messing with the build configuration to use the no-op 
module to build the ARM library...

I guess I could try do it myself, but I fear days of entanglement
with baroque layered build components, plus days trying to set up a
suitable build VM...

Thanks,
Best Regards, Dave

On 12/12/2020 16:48, Dave Nadler wrote:
> Anybody ever posted bug reports on GCC as built+distributed by ARM?
> I posted what I thought was a decently written (and very real) bug
> report, but I have never received any response at all:
> https://bugs.launchpad.net/gcc-arm-embedded/+bug/1905459
> Any ideas?
> Thanks,
> Best Regards, Dave

I have only reported one bug, and you can see how well it has been answered:

<https://bugs.launchpad.net/gcc-arm-embedded/+bug/1646883>

Bugs and questions /do/ get answered and fixed on the tracker (I've made
a few helpful comments myself to some).  But I don't believe the
responsible people have anything close to the time and manpower needed
to handle the support.  They can fix some of the easy stuff.  Your issue
here - clearly written and important though it is - is far from easy stuff.

It is good to make these bug reports - it stops things getting
forgotten, and it makes it much easier for other people to find out
about them.  However, it is normally best to find a workaround for them
rather than waiting for the bug to be fixed.

On Sat, 12 Dec 2020 10:48:59 -0500, Dave Nadler <drn@nadler.com>
wrote:

>Anybody ever posted bug reports on GCC as built+distributed by ARM?
>I posted what I thought was a decently written (and very real) bug 
>report, but I have never received any response at all:
>https://bugs.launchpad.net/gcc-arm-embedded/+bug/1905459
>Any ideas?
>Thanks,
>Best Regards, Dave

You didn't provide a sample program that exhibits the problem.  And it
has been only a few weeks (in holiday season) - perhaps no one from
the development team has looked at the report.

George

On 12/12/2020 04:48 PM, Dave Nadler wrote:
> Anybody ever posted bug reports on GCC as built+distributed by ARM?
> I posted what I thought was a decently written (and very real) bug report, but I have never received any response at all:
> https://bugs.launchpad.net/gcc-arm-embedded/+bug/1905459
> Any ideas?
> Thanks,
> Best Regards, Dave
You have to wait 10-20 years. By then it's either fixed or disappeared. 
If it disappeared, resubmit and wait some more.