Aarch64 number bug

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

Aarch64 number bug

KenDickey
 

16r80000000 < 16r8000000000000000. "==> false"

FYI,
-KenD
-KenD
Reply | Threaded
Open this post in threaded view
|

Re: Aarch64 number bug

Eliot Miranda-2
 


> On Jun 16, 2021, at 4:43 PM, [hidden email] wrote:
>
> 
> 16r80000000 < 16r8000000000000000. "==> false"

If the community valued its primary execution technology then it would be the case that a few members were competent enough to startup the simulation environment and debug this.  But the community doesn’t value it, takes it for granted, puts zero effort into moving it forward and expects the original authors to fix all the problems.  So when the principal authors have pressing demands on their time, intimately involved with ensuring funding for euro in the execution technology, funding and effort the “community” appears incapable of contributing to, then the “community” can wait.

>
> FYI,
> -KenD
Reply | Threaded
Open this post in threaded view
|

Re: Aarch64 number bug

marcel.taeumel
 
Hi Eliot.

> ...  then it would be the case that a few members were competent enough to startup the simulation environment and debug this.

This would be nice indeed. It's not that hard to prepare a VMMaker image on your local machine. This bug could be the perfect opportunity to get started. :-) The short feedback loops using the VM simulator are really awesome! Plus, any questions or thoughts on the VM simulation and debugging tools are welcomed on this list and sure be answered quickly.

@Everybody: Take your chances! Learn something new. It's definitely worth it.

> ...  then the “community” can wait.

And while doing so, it still makes sense to collect information about this issue here on the list. ^__^

Best,
Marcel

Am 18.06.2021 09:23:06 schrieb Eliot Miranda <[hidden email]>:




> On Jun 16, 2021, at 4:43 PM, [hidden email] wrote:
>
> 
> 16r80000000 < 16r8000000000000000. "==> false"

If the community valued its primary execution technology then it would be the case that a few members were competent enough to startup the simulation environment and debug this. But the community doesn’t value it, takes it for granted, puts zero effort into moving it forward and expects the original authors to fix all the problems. So when the principal authors have pressing demands on their time, intimately involved with ensuring funding for euro in the execution technology, funding and effort the “community” appears incapable of contributing to, then the “community” can wait.

>
> FYI,
> -KenD
Reply | Threaded
Open this post in threaded view
|

Re: Aarch64 number bug

KenDickey
In reply to this post by KenDickey
 
> "If the community valued its primary execution technology then it would
> be the case that a few members were competent enough to startup the
> simulation environment and debug this. "


Unfortunately, my attempt to build a VM simulator on aarch64 fails due
to a divide by zero bug.

Perhaps because
   16r8000000000000000 printStringHex. ==> '0'


You may be over-estimating my ability to debug this sort of thing.

-------------------

In the wider context, my suggestion is to declare, say the first week in
July, as "VM Appreciation Week".

The observation being that OpenSmalltalk is common to Cuis, Squeak, and
Pharo.  OpenSmalltalk is open source. OpenSmalltalk is a _Commons_.  A
"commons" in our case is a shared resource which is managed by its
users.

To be better managers (or managers at all, rather than just
freeloaders), we should take some time to experience it. [a] Build and
run a/the simulator, [b] Build a VM.

Participation means knowing the resource in order to be able to sustain
it.
-------------------

OK, so raising awareness to become visible in the right way.

There are a few speed bumps to be overcome first.

One desires that Joe or Jane coder be able to do [a] and [b] above.

Building VMMaker on Linux/x86-64 I get a large number of merge dialogs
and then:
vvv===vvv
This virtual machine is too old to support correct versions of the
FloatArray>>at:[put:] primitives 238 and 239.
FloatArray subclasses will not behave correctly and FloatArray[64]Test
tests will fail.
Please upgrade your VM.  You may continue and upgrade later or abort and
upgrade now.
^^^===^^^

But if I try to build a VM, I run into a number of problems (notes
attached; the "included too late" bug shows up in the arm build as
well).

It is difficult to see today how Joe or Jane can build a VM if like me
their knowledge of build mechanics is fairly shallow.

Like you all, I have a very busy life.  Among other things, given a
destabalized climate I am now tending my garden to raise food.

On the other hand, I am happy to spend time supporting the OpenSmalltalk
Commons as, yes, this _is_ important to me.

I don't feel that I have the skillset to fix the current speed bumps,
but I certainly would like to declare a "VM Appreciation Week" sometime
soon and welcome a discussion on how we engender a better stewardship of
our commons..

-KenD


ZeroDivide.png (372K) Download Attachment
build-bug.txt (30K) Download Attachment
-KenD