Hi Tobias,


> On Feb 1, 2021, at 11:12 PM, Tobias Pape <Das.Linux at gmx.de> wrote:
> 
> 
> 
>> On 2. Feb 2021, at 04:12, Eliot Miranda <eliot.miranda at gmail.com> wrote:
>> 
>> 
>> 
>>> On Sun, Jan 31, 2021 at 11:56 PM Bruce O'Neel <bruce.oneel at pckswarms.ch> wrote:
>>> Good news/bad news.
>>> 
>>> The good news is that Pulse sound driver can be solved with:
>>> 
>>> 1.  apt install libpulse-dev
>>> 2.  Make sure you say yes to clean, and rebuild.
>>> 
>>> 
>>> First, thank you Bruce!
>>> 
>>> Second:
>>> 
>>> $ git pull
>>> $ grep -i libpulse build.linux*/HowToBuild 
>>> $ 
>>> 
>>> *PLEASE* (pretty please wit bells on, I'm *begging you*) make sure that build instructions are added to the various HowToBuild files.  One cannot expect people to be able to build the VM if we don't maintain the documents.
>> 
>> If we have more than one build-instruction document, we have too many.

I disagree.  We have one build reasons per platform that we build on.  Every platform is different. There is a little duplication but the structure we have now means the build documents are situated in the most relevant place.


>> One and only one HowTo, INSTALL or the like in the main directory must suffice.
>> Aint nobody read more, whether we insist or not.
>> 
>> Preferably, also only one command (or command chain) to build should exist, but that's another story I'm not opening for debate here.
>> 
>> :)
>> 
>> -t
>> 
>> 
>>> Further, having a discussion here is fine, but one should at least cc the opensmalltalk-vm / vm-dev list.  That;'s the proper venue for discussing VM builds/bugs et al.
>>> 
>>> 
>>> The bad news is that while sound might "work" in the sense that sounds will now be produced we seem to have a problem with the sounds skipping and having pops in them.  That's still being worked on....
>>> 
>>> cheers
>>> 
>>> bruce
>>> 
>>> 
>>> 
>>> 01 February 2021 08:36 Jim Rosenberg <jr at amanue.com> wrote:
>>> On Mon, Jan 18, 2021 at 07:07:25AM -0500, I wrote:
>>>> Ouch! I have a lot of artistic work which is developed in Squeak on
>>>> Linux -- currently only up to 4.3 (which is working fine for me, so I
>>>> haven't upgraded). I'm trying to run one of my 4.3 images on a
>>>> Raspberry Pi 3B+ which is at Raspbian 10 (buster), and with the current
>>>> armv6 squeak VM, getting the message
>>>> 
>>>> This interpreter (vers. 6521) cannot read image file (vers. 6504).
>>>> 
>>>> Suggestions?
>> 
>> --On Monday, January 18, 2021 09:15:46 AM -0500 "David T. Lewis"
>> wrote:
>> 
>>> If you do not mind installing development tools on your Raspberry Pi, then
>>> the best thing to do is compile the VM yourself. Instructions for doing
>>> this are at http://wiki.squeak.org/squeak/6354
>>> 
>>> If you have any difficulty building your VM, please ask for help. And
>>> if it works without problems, please report back so we know.
>> 
>> It built without a hitch, and worked fine with all but one of the images I
>> tested it on. The image that caused me trouble uses sound. Everything else
>> worked fine, but the default build seemed not to support sound at all. The
>> only drivers were vm-sound-null and vm-sound-custom.
>> 
>> I've had no luck with Linux sound on my desktop using anything other than
>> vm-sound-pulse -- that works fine. squeak -h didn't show vm-sound-pulse as
>> an available driver oh the version I built on the Raspberry Pi, so I set
>> out to fix that. After installing the pulse development library, configure
>> found it, and now I have an image on which everything just works. Thanks
>> for the help!!
>> 
>> lit-archive 82% squeak -version
>> 4.19.5-3796 #1 Wed 20 Jan 2021 08:45:38 AM EST /usr/bin/cc
>> Linux lit-archive 5.4.83-v7+ #1379 SMP Mon Dec 14 13:08:57 GMT 2020 armv7l
>> GNU/Linux
>> plugin path: /usr/local/lib/squeak/4.19.5-3796 [default:
>> /usr/local/lib/squeak/4.19.5-3796/]
>> 
>> * * *
>> 
>>> You can find a precompiled VM for ARM v61 at http://squeakvm.org/unix/
>>> which may work, although it is out of date so I am not sure if it will
>>> run on your Pi.
>> 
>> Results here are not so happy. The display driver is marked in red on that
>> web page as "experimental"; on my images it mangles bitmaps in SketchMorphs
>> in a way that is completely unacceptable. It looks almost as though there
>> was some kind of attempt to do after-the-fact anti-aliasing that just went
>> haywire on my graphics.
>> 
>> -Thanks, Jim
> 
> 
> 
>