llvm dev - Oct 2020 - [RFC] Backend for Motorola 6800 series CPU (M68k)

Its awesome to see so much progress on this!

A very minor question - why is it called M680x0 and not M68K given 
that's what the target arch/triple is and how its usually referred to?

Sorry for the bikeshedding....

Simon.

On 30/09/2020 21:14, Min-Yih Hsu via llvm-dev wrote:
> Hi All,
>
> I've composed a draft roadmap for this new target. I've decided to
try
> Github's "Projects" feature, as it provides a clearer view to
see all
> the blockers and action items, IMAO. Here is the link:
> https://github.com/M680x0/M680x0-mono-repo/projects 
> <https://github.com/M680x0/M680x0-mono-repo/projects>
>
> Currently I only created two major milestones: Becoming an 
> experimental target and becoming an official target. For each 
> milestone, I've listed the expected features, estimated time frame 
> (though I'm not really confident on that), and most importantly, the 
> blockers for the milestone.
> Fortunately, all of our essential features are complete (e.g. ISel, 
> MC), so there will be more house cleaning tasks and bug fixing than 
> adding new features in the second milestone.
>
> In addition to the aforementioned two milestones, I've added "be
able
> to run toy programs" as a separate milestone to accommodate some more 
> urgent tasks right now. More specifically, driver problems that make 
> Clang unable to leverage all our components.
>
> Thank you!
> Min
>
> On Tue, Sep 29, 2020 at 11:27 AM Renato Golin <rengolin at gmail.com 
> <mailto:rengolin at gmail.com>> wrote:
>
>     On Tue, 29 Sep 2020 at 18:53, John Paul Adrian Glaubitz
>     <glaubitz at physik.fu-berlin.de
>     <mailto:glaubitz at physik.fu-berlin.de>> wrote:
>
>         So, shall we setup a server for that or is there some existing
>         infrastructure
>         from LLVM that is used in this case?
>
>
>     Unfortunately, we don't have a centralised infrastructure like
>     GCC. Each target community is responsible for maintaining their
>     own buildbots.
>
>     All we provide is the "build master", which aggregates all
builds,
>     email when there are regressions, etc.
>
>     http://llvm.org/docs/HowToAddABuilder.html
>     <http://llvm.org/docs/HowToAddABuilder.html>
>
>     It should be trivial to copy & paste an existing bot config and
>     change to add the experimental hardware (once it's in the tree).
>
>     Pay attention to the version of buildbot you install, as using an
>     even slightly different version can cause weird errors.
>
>     (I'm sure you understand how many times we tried to move on for
>     the past 10 years... :)
>
>
>         As far as I understand, 90% of the tests already pass
>         (according to Min).
>
>
>     Awesome! We have buildbots with test-suite running, you can copy
>     those, too. There were some QEMU bots in the past, I'm not sure
>     they're around, but the infra to run them should still be there.
>
>     cheers,
>     --renato
>
>
>
> -- 
> Min-Yih Hsu
> Ph.D Student in ICS Department, University of California, Irvine (UCI).
>
> _______________________________________________
> cfe-dev mailing list
> cfe-dev at lists.llvm.org
> https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20201001/3a3b5d20/attachment.html>

> On Oct 1, 2020, at 2:38 AM, Simon Pilgrim via cfe-dev <cfe-dev at
lists.llvm.org> wrote:
> 
> Its awesome to see so much progress on this!
> 
> 
Thank you!
> A very minor question - why is it called M680x0 and not M68K given
that's what the target arch/triple is and how its usually referred to?
> 
> 
I think it’s a legit question, IIRC Jessica also had the same question in one of
the patches here. I think the primary reason is that we want to conform with
GCC’s target triple, which came before our work.

Also I think in our patch, m680x0 is the only valid target triple.

Thank you
Min
> Sorry for the bikeshedding....
> 
> Simon.
> 
> On 30/09/2020 21:14, Min-Yih Hsu via llvm-dev wrote:
>> Hi All,
>> 
>> I've composed a draft roadmap for this new target. I've decided
to try Github's "Projects" feature, as it provides a clearer view
to see all the blockers and action items, IMAO. Here is the link:
>> https://github.com/M680x0/M680x0-mono-repo/projects
<https://github.com/M680x0/M680x0-mono-repo/projects>
>> 
>> Currently I only created two major milestones: Becoming an experimental
target and becoming an official target. For each milestone, I've listed the
expected features, estimated time frame (though I'm not really confident on
that), and most importantly, the blockers for the milestone.
>> Fortunately, all of our essential features are complete (e.g. ISel,
MC), so there will be more house cleaning tasks and bug fixing than adding new
features in the second milestone.
>> 
>> In addition to the aforementioned two milestones, I've added
"be able to run toy programs" as a separate milestone to accommodate
some more urgent tasks right now. More specifically, driver problems that make
Clang unable to leverage all our components.
>> 
>> Thank you!
>> Min
>> 
>> On Tue, Sep 29, 2020 at 11:27 AM Renato Golin <rengolin at gmail.com
<mailto:rengolin at gmail.com>> wrote:
>> On Tue, 29 Sep 2020 at 18:53, John Paul Adrian Glaubitz <glaubitz at
physik.fu-berlin.de <mailto:glaubitz at physik.fu-berlin.de>> wrote:
>> So, shall we setup a server for that or is there some existing
infrastructure
>> from LLVM that is used in this case?
>> 
>> Unfortunately, we don't have a centralised infrastructure like GCC.
Each target community is responsible for maintaining their own buildbots.
>> 
>> All we provide is the "build master", which aggregates all
builds, email when there are regressions, etc.
>> 
>> http://llvm.org/docs/HowToAddABuilder.html
<http://llvm.org/docs/HowToAddABuilder.html>
>> 
>> It should be trivial to copy & paste an existing bot config and
change to add the experimental hardware (once it's in the tree).
>> 
>> Pay attention to the version of buildbot you install, as using an even
slightly different version can cause weird errors.
>> 
>> (I'm sure you understand how many times we tried to move on for the
past 10 years... :)
>> 
>> 
>> As far as I understand, 90% of the tests already pass (according to
Min).
>> 
>> Awesome! We have buildbots with test-suite running, you can copy those,
too. There were some QEMU bots in the past, I'm not sure they're around,
but the infra to run them should still be there.
>> 
>> cheers,
>> --renato
>> 
>> 
>> -- 
>> Min-Yih Hsu
>> Ph.D Student in ICS Department, University of California, Irvine (UCI).
>> 
>> 
>> _______________________________________________
>> cfe-dev mailing list
>> cfe-dev at lists.llvm.org <mailto:cfe-dev at lists.llvm.org>
>> https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
<https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev>
> _______________________________________________
> cfe-dev mailing list
> cfe-dev at lists.llvm.org
> https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20201001/b1654e0c/attachment.html>

> On 2 Oct 2020, at 00:02, Min-Yih Hsu via llvm-dev <llvm-dev at
lists.llvm.org> wrote:
> 
> 
> 
>> On Oct 1, 2020, at 2:38 AM, Simon Pilgrim via cfe-dev <cfe-dev at
lists.llvm.org> wrote:
>> 
>> Its awesome to see so much progress on this!
>> 
>> 
> Thank you!
>> A very minor question - why is it called M680x0 and not M68K given
that's what the target arch/triple is and how its usually referred to?
>> 
>> 
> I think it’s a legit question, IIRC Jessica also had the same question in
one of the patches here. I think the primary reason is that we want to conform
with GCC’s target triple, which came before our work.
> 
> Also I think in our patch, m680x0 is the only valid target triple.
I continue to believe M680x0 is a mistake. Your own patch series uses
"[M68K]". The triple name is m68k-linux-gnu etc. Everyone calls it
either m68k or just 68k.. As for GCC there are only four instances of m680x0: 3
in install.texi and 1 in m68k.md, but the port is gcc/config/m68k. So please
let's not use this name for LLVM; it's longer, more confusing and just
generally clumsy in comparison to the de-facto standard that everyone knows
(even this thread's subject has "(M68k)" in the subject,
acknowledging that that's a name people recognise).

m680x0 is definitely not a valid GNU target triple though; config.sub will let
you have m68010 etc instead of m68k, but not a literal x in there.

Jess
> Thank you
> Min
>> Sorry for the bikeshedding....
>> 
>> Simon.
>> 
>> On 30/09/2020 21:14, Min-Yih Hsu via llvm-dev wrote:
>>> Hi All,
>>> 
>>> I've composed a draft roadmap for this new target. I've
decided to try Github's "Projects" feature, as it provides a
clearer view to see all the blockers and action items, IMAO. Here is the link:
>>> https://github.com/M680x0/M680x0-mono-repo/projects
>>> 
>>> Currently I only created two major milestones: Becoming an
experimental target and becoming an official target. For each milestone,
I've listed the expected features, estimated time frame (though I'm not
really confident on that), and most importantly, the blockers for the milestone.
>>> Fortunately, all of our essential features are complete (e.g. ISel,
MC), so there will be more house cleaning tasks and bug fixing than adding new
features in the second milestone.
>>> 
>>> In addition to the aforementioned two milestones, I've added
"be able to run toy programs" as a separate milestone to accommodate
some more urgent tasks right now. More specifically, driver problems that make
Clang unable to leverage all our components.
>>> 
>>> Thank you!
>>> Min
>>> 
>>> On Tue, Sep 29, 2020 at 11:27 AM Renato Golin <rengolin at
gmail.com> wrote:
>>> On Tue, 29 Sep 2020 at 18:53, John Paul Adrian Glaubitz
<glaubitz at physik.fu-berlin.de> wrote:
>>> So, shall we setup a server for that or is there some existing
infrastructure
>>> from LLVM that is used in this case?
>>> 
>>> Unfortunately, we don't have a centralised infrastructure like
GCC. Each target community is responsible for maintaining their own buildbots.
>>> 
>>> All we provide is the "build master", which aggregates
all builds, email when there are regressions, etc.
>>> 
>>> http://llvm.org/docs/HowToAddABuilder.html
>>> 
>>> It should be trivial to copy & paste an existing bot config and
change to add the experimental hardware (once it's in the tree).
>>> 
>>> Pay attention to the version of buildbot you install, as using an
even slightly different version can cause weird errors.
>>> 
>>> (I'm sure you understand how many times we tried to move on for
the past 10 years... :)
>>> 
>>> 
>>> As far as I understand, 90% of the tests already pass (according to
Min).
>>> 
>>> Awesome! We have buildbots with test-suite running, you can copy
those, too. There were some QEMU bots in the past, I'm not sure they're
around, but the infra to run them should still be there.
>>> 
>>> cheers,
>>> --renato
>>> 
>>> 
>>> -- 
>>> Min-Yih Hsu
>>> Ph.D Student in ICS Department, University of California, Irvine
(UCI).
>>> 
>>> 
>>> _______________________________________________
>>> cfe-dev mailing list
>>> 
>>> cfe-dev at lists.llvm.org
>>> https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
>> _______________________________________________
>> cfe-dev mailing list
>> cfe-dev at lists.llvm.org
>> https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev
> 
> _______________________________________________
> LLVM Developers mailing list
> llvm-dev at lists.llvm.org
> https://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev

On 01/10/2020 10:38, Simon Pilgrim via llvm-dev wrote:
> Its awesome to see so much progress on this!
> 
> A very minor question - why is it called M680x0 and not M68K given 
> that's what the target arch/triple is and how its usually referred to?
> 
> Sorry for the bikeshedding....
> 
Because it came in multiple versions? M68K expands to M68000, which was 
the original chip. I think the most popular version was the M68040, so 
M680x0 is unambiguously the family of chips - M68K can mean either the 
original, or the family. I've always known the family as M680x0.

Just another case where what people say and what they mean isn't 
necessarily the same, or clear ...

Cheers,
Wol

On Sat, Oct 10, 2020 at 11:04:21PM +0100, antlists via llvm-dev
wrote:
> On 01/10/2020 10:38, Simon Pilgrim via llvm-dev wrote:
> > Its awesome to see so much progress on this!
> > 
> > A very minor question - why is it called M680x0 and not M68K given
> > that's what the target arch/triple is and how its usually referred
to?
> > 
> > Sorry for the bikeshedding....
> > 
> Because it came in multiple versions? M68K expands to M68000, which was the
> original chip. I think the most popular version was the M68040, so M680x0
is
> unambiguously the family of chips - M68K can mean either the original, or
> the family. I've always known the family as M680x0.
Noone uses M68k to mean the original nowadays. Very few people care
about the original at all nowadays. So yes, please just use M68k.

Joerg