In article <59236cc520chrisjohnson@spamcop.net>, News
<chrisjohnson@spamcop.net> wrote:
> In article <1a116a2359.chris@mytarbis.plus.com>, Chris Hughes
> <news13@noonehere.co.uk> wrote:
> > There are no scroll bars in Iris

> Indeed there are. Hover the mouse over the right side of the window and
> the scroll bars appear. They do not work in quite the same way as
> standard - if you click at the bottom, the page will immediately scroll
> to the bottom. 'Normal' operation just moves the window down one page,
> whereever you click.

Although it may be the 'norm' on other OSs I find the combination of "hide
the scrollbar when you aren't hovering the mouse over a very narrow region"
and "jump to where the mouse clicks" a real irritation when trying to read
a long page. The standard RO behaviour of a persistent wider scrollbar and
jumping a 'window size" amount is far more convenient.

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics https://www.st-andrews.ac.uk/~www_pa/Scots_Guide/intro/electron.htm
biog http://jcgl.orpheusweb.co.uk/history/ups_and_downs.html
Audio Misc http://www.audiomisc.co.uk/index.html

In article <5924406daenoise@audiomisc.co.uk>, Jim Lesurf
<noise@audiomisc.co.uk> wrote:

> Although it may be the 'norm' on other OSs I find the
> combination of "hide the scrollbar when you aren't
> hovering the mouse over a very narrow region" and "jump
> to where the mouse clicks" a real irritation when trying
> to read a long page. The standard RO behaviour of a
> persistent wider scrollbar and jumping a 'window size"
> amount is far more convenient.

In Mint you can over-ride that default behaviour. I know
you use something different but for me it's:

System settings->Themes->Settings and then:

Jump to position when clicking in trough (unset)

Use overlay scroll bars (set)

Override the current theme's scroll bar width (set)

This last brings up the slider for the width setting. Slide
to taste.

I hope this is useful.

John

--
John
newsmcc@blueyonder.co.uk
j dot mccartney atte blueyonder dot co dot uk

In article <5924406daenoise@audiomisc.co.uk>,
Jim Lesurf <noise@audiomisc.co.uk> wrote:

> Although it may be the 'norm' on other OSs I find the combination
> of "hide the scrollbar when you aren't hovering the mouse over a
> very narrow region" and "jump to where the mouse clicks" a real
> irritation when trying to read a long page. The standard RO
> behaviour of a persistent wider scrollbar and jumping a 'window
> size" amount is far more convenient.

A classic example of "style over function" it's never a good idea.

Function must come first.

Bob.

In article <59245fc20fnewsmcc@blueyonder.co.uk>, John
<newsmcc@blueyonder.co.uk> wrote:
> In article <5924406daenoise@audiomisc.co.uk>, Jim Lesurf
> <noise@audiomisc.co.uk> wrote:

> > Although it may be the 'norm' on other OSs I find the combination of
> > "hide the scrollbar when you aren't hovering the mouse over a very
> > narrow region" and "jump to where the mouse clicks" a real irritation
> > when trying to read a long page. The standard RO behaviour of a
> > persistent wider scrollbar and jumping a 'window size" amount is far
> > more convenient.

> In Mint you can over-ride that default behaviour. I know you use
> something different but for me it's:

> System settings->Themes->Settings and then:

> Jump to position when clicking in trough (unset)

> Use overlay scroll bars (set)

> Override the current theme's scroll bar width (set)

> This last brings up the slider for the width setting. Slide to taste.

> I hope this is useful.

Dunno. I'd need to check. I do use Mint - but xfce for the desktop, and
install ROX to use most of the time as the interface.

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics https://www.st-andrews.ac.uk/~www_pa/Scots_Guide/intro/electron.htm
biog http://jcgl.orpheusweb.co.uk/history/ups_and_downs.html
Audio Misc http://www.audiomisc.co.uk/index.html

The date being 29 Apr 2021, Bob Latham <bob@sick-of-spam.invalid> decided
to write:

> In article <5924406daenoise@audiomisc.co.uk>,
> Jim Lesurf <noise@audiomisc.co.uk> wrote:

>> Although it may be the 'norm' on other OSs I find the combination
>> of "hide the scrollbar when you aren't hovering the mouse over a
>> very narrow region" and "jump to where the mouse clicks" a real
>> irritation when trying to read a long page. The standard RO
>> behaviour of a persistent wider scrollbar and jumping a 'window
>> size" amount is far more convenient.

> A classic example of "style over function" it's never a good idea.

> Function must come first.

I've just noticed a very thin progress bar along the top of the window.
It's useful because the stop/refresh button isn't a good indicator of
whether Iris has finished processing (and it needs to flash if something
is still in progress).

Is there a repository for bug reports, suggestions and requests?

In article <b177ba2559.news@user.minijem.plus.com>, Richard Porter
<ricp@minijem.plus.com> wrote:
> The date being 29 Apr 2021, Bob Latham <bob@sick-of-spam.invalid>
> decided to write:

> > In article <5924406daenoise@audiomisc.co.uk>, Jim Lesurf
> > <noise@audiomisc.co.uk> wrote:

> >> Although it may be the 'norm' on other OSs I find the combination of
> >> "hide the scrollbar when you aren't hovering the mouse over a very
> >> narrow region" and "jump to where the mouse clicks" a real
> >> irritation when trying to read a long page. The standard RO
> >> behaviour of a persistent wider scrollbar and jumping a 'window
> >> size" amount is far more convenient.

> > A classic example of "style over function" it's never a good idea.

> > Function must come first.

> I've just noticed a very thin progress bar along the top of the window.
> It's useful because the stop/refresh button isn't a good indicator of
> whether Iris has finished processing (and it needs to flash if
> something is still in progress).

> Is there a repository for bug reports, suggestions and requests?

When I talked to Andrew about feedback for Iris, Pinboard etc he
replied...

send to andrew@riscosdev.com but try and collate stuff together,
and don't over-load me please! :) I'll do my best :)

--
Chris Newman

In message <5925be239acvjazz@waitrose.com>
Chris Newman <cvjazz@waitrose.com> wrote:

> In article <b177ba2559.news@user.minijem.plus.com>, Richard Porter
> <ricp@minijem.plus.com> wrote:
>> The date being 29 Apr 2021, Bob Latham <bob@sick-of-spam.invalid>
>> decided to write:

>>> In article <5924406daenoise@audiomisc.co.uk>, Jim Lesurf
>>> <noise@audiomisc.co.uk> wrote:

>>>> Although it may be the 'norm' on other OSs I find the combination of
>>>> "hide the scrollbar when you aren't hovering the mouse over a very
>>>> narrow region" and "jump to where the mouse clicks" a real
>>>> irritation when trying to read a long page. The standard RO
>>>> behaviour of a persistent wider scrollbar and jumping a 'window
>>>> size" amount is far more convenient.

>>> A classic example of "style over function" it's never a good idea.

>>> Function must come first.

>> I've just noticed a very thin progress bar along the top of the window.
>> It's useful because the stop/refresh button isn't a good indicator of
>> whether Iris has finished processing (and it needs to flash if
>> something is still in progress).

>> Is there a repository for bug reports, suggestions and requests?

> When I talked to Andrew about feedback for Iris, Pinboard etc he
> replied...

> send to andrew@riscosdev.com but try and collate stuff together,
> and don't over-load me please! :) I'll do my best :)

Everyone needs to remember Iris is still Beta software, i.e. not finished

They have a long list of things they want to improve, but this takes time
and money plus developer time of course. Its very important the above bit.

Collate issues - bug fixes are probably more important then new features.

Also remember this is an extremely big program and uses the WebKit stuff
used on all the other platforms so do not except everything to be totally
style guide. Be happy we have much better browser.

--
Chris Hughes

>> When I talked to Andrew about feedback for Iris, Pinboard etc he
>> replied...

>> send to andrew@riscosdev.com but try and collate stuff together,
>> and don't over-load me please! :) I'll do my best :)

> Everyone needs to remember Iris is still Beta software, i.e. not finished

> They have a long list of things they want to improve, but this takes time
> and money plus developer time of course. Its very important the above bit.

> Collate issues - bug fixes are probably more important then new features.

> Also remember this is an extremely big program and uses the WebKit stuff
> used on all the other platforms so do not except everything to be totally
> style guide. Be happy we have much better browser.

Well said Chris.

Everyone was hassling Andrew about getting "their free copy" and Andrew
rightly stated the reasons for a delay in doing that and also set out
expectations when it was released to a wider group.

People need to remember that this is a effectively a miniscule development
team and inundating them with feature requests etc now just diverts what
time they have to develop and support the product.

Yes feedback is good but just remember to save things up and not send in a
constant stream.

Andrew and the team do need some downtime to recharge their batteries as
well like anyone else.

--
Experience the future using ARM Technology - ARMBook,BeagleBoard -xM,
PandaBoard,Raspberry Pi,iMX6/ARMX6,IGEPv5 & Titanium powered by RISC OS
5.28.

In article <82b9c02559.chris@mytarbis.plus.com>,
Chris Hughes <news13@noonehere.co.uk> wrote:

[Snip]

> Everyone needs to remember Iris is still Beta software, i.e. not
> finished

> They have a long list of things they want to improve, but this takes
> time and money plus developer time of course. Its very important the
> above bit.

> Collate issues - bug fixes are probably more important then new
> features.

> Also remember this is an extremely big program and uses the WebKit
> stuff used on all the other platforms so do not except everything to
> be totally style guide. Be happy we have much better browser.

Absolutely, and pretty much as stated in the ReadMe. At this stage the
ability to visit complex sites without having to use the Windoze machine
outweighs the style guide concerns.

For now I am more interested in Alignment Exceptions and whether they
should be 'Off' as recommended when Iris starts or 'On' which is the
default on this machine. Both options come with warnings, about Iris in
the first case being unstable or failing and in the second case about
older software crashing or corrupting data.

Given that, almost by definition, all software is 'older' than Iris I
ignore Iris's recommendation but would be interested in the thoughts and
recommendations of others? Does this hinge on the definition of 'older
software'?
B

--
_____________________________________________________________________

Brian Jordan
RISC OS 5.28 on Raspberry Pi
_____________________________________________________________________

In message <5925c8ba0dbrian.jordan9@btinternet.com>
Brian Jordan <brian.jordan9@btinternet.com> wrote:

> In article <82b9c02559.chris@mytarbis.plus.com>,
> Chris Hughes <news13@noonehere.co.uk> wrote:

> [Snip]

>> Everyone needs to remember Iris is still Beta software, i.e. not
>> finished

>> They have a long list of things they want to improve, but this takes
>> time and money plus developer time of course. Its very important the
>> above bit.

>> Collate issues - bug fixes are probably more important then new
>> features.

>> Also remember this is an extremely big program and uses the WebKit
>> stuff used on all the other platforms so do not except everything to
>> be totally style guide. Be happy we have much better browser.

> Absolutely, and pretty much as stated in the ReadMe. At this stage the
> ability to visit complex sites without having to use the Windoze machine
> outweighs the style guide concerns.

> For now I am more interested in Alignment Exceptions and whether they
> should be 'Off' as recommended when Iris starts or 'On' which is the
> default on this machine. Both options come with warnings, about Iris in
> the first case being unstable or failing and in the second case about
> older software crashing or corrupting data.

> Given that, almost by definition, all software is 'older' than Iris I
> ignore Iris's recommendation but would be interested in the thoughts and
> recommendations of others? Does this hinge on the definition of 'older
> software'?

I have been running with Alignment Exception Off now since I installed
this version of Iris, I used to have it on.

If you want to use Iris at the current time then you need to turn
Alignment Exception Off. Jeffery Lee in the ROOL forums has expected in
detail the various options.

By old software, I believe they mainly mean 26 bit software or 32 bit
software that has not been updated to use the new standards. If you have
lots of older software, then either ensure you are using the latest
possible version or possibly find a more modern alternative to reduce
possible issue

The Thumb2 part of the cross-platform WebKit requires the alignment
exceptions to be off at this time. As the ReadMe file says they will look
at this in the fullness of time.

--
Chris Hughes

In article <44e4c32559.dougjwebb@btinternet.com>, Doug Webb
<doug.j.webb@btinternet.com> wrote:

> >> When I talked to Andrew about feedback for Iris, Pinboard etc he
> >> replied...

> >> send to andrew@riscosdev.com but try and collate stuff together, and
> >> don't over-load me please! :) I'll do my best :)

> > Everyone needs to remember Iris is still Beta software, i.e. not
> > finished

> > They have a long list of things they want to improve, but this takes
> > time and money plus developer time of course. Its very important the
> > above bit.

> > Collate issues - bug fixes are probably more important then new
> > features.

> > Also remember this is an extremely big program and uses the WebKit
> > stuff used on all the other platforms so do not except everything to
> > be totally style guide. Be happy we have much better browser.

> Well said Chris.

> Everyone was hassling Andrew about getting "their free copy" and Andrew
> rightly stated the reasons for a delay in doing that and also set out
> expectations when it was released to a wider group.

> People need to remember that this is a effectively a miniscule
> development team and inundating them with feature requests etc now
> just diverts what time they have to develop and support the product.

> Yes feedback is good but just remember to save things up and not send
> in a constant stream.

> Andrew and the team do need some downtime to recharge their batteries
> as well like anyone else.

Absolutely. There's a balance to be achieved between not bombarding them
but making sure they are aware off problems in case they've missed them.

I happily paid for the download plus a small donation. I'm so glad to be
able to do my banking on Ridc OS and get into my router's set-up sans
Windows. Lots of things work already. Congratulations to the team.

--
Chris Newman

On 1 May, Brian Jordan wrote in message
<5925c8ba0dbrian.jordan9@btinternet.com>:

> For now I am more interested in Alignment Exceptions and whether they
> should be 'Off' as recommended when Iris starts or 'On' which is the
> default on this machine. Both options come with warnings, about Iris in
> the first case being unstable or failing and in the second case about
> older software crashing or corrupting data.

This is, indeed, an interesting question.

> Given that, almost by definition, all software is 'older' than Iris I
> ignore Iris's recommendation but would be interested in the thoughts and
> recommendations of others? Does this hinge on the definition of 'older
> software'?

Kind of.

If enabled, alignment exceptions are fired when something tries to do an
"unaligned load" or "unaligned store", which are a special kind of read from
or write to memory. The reason that the exceptions exist is that ARM changed
how unaligned memory access worked between ARMv5 and before (the Iyonix, and
Pi 1, which is actually ARMv6 in v5 compatibility mode) and ARMv7 and later
(pretty much every modern system apart from the aforementioned).

Anything compiled for an Iyonix and using these instructions would find that
the bytes they were reading or writing would be scrambled if used on an
ARMv7 processor, so the exceptions were provided to allow users to be
notified that things were going wrong. Despite claims to the contrary,
running with alignment exceptions on has always been advisable, to catch
software which never got updated post-Iyonix. Turning them off might make
some broken software usable, but personally I'd not be happy doing it on my
machine.

However, Iris is complied especially for ARMv7 and later systems, using a
modern compiler that understands the new unaligned load and store
instructions. It makes use of them to optimise the code and wring more
performance out of the system (not least because it uses Thumb instructions
in the Javascript JIT, which benefit from the changes).

So Iris needs alignment exceptions to be off, as it uses the instructions
correctly in their new form. It's probably the only piece of RISC OS
software to do so. Unfortunately, there's still a lot of old stuff out there
which uses the instructions incorrectly, in their old form, and expects them
to work as they did on the Iyonix.

As to what to do, I don't know. I'll probably continue to browse on Linux
and leave alignment exceptions on. If you wish to use Iris, you'll need to
turn them off, however. Were I to do that, I think I'd turn them back on for
a while when running new software for the first time, though.

Incidentally, this answers Chris N's question in the current Archive. Saying
something is "32-bit" is pretty meaningless these days, as it could be
"ARMv5" (Iyonix, and Pi 1), "ARMv7" (most later stuff) or "ARMv8" (Pi 3 and
4). The reason that alleged "32-bit" stuff is unstable on the Pi 4 is very
likely that it was 32-bitted for the Iyonix and never revisited since the
ARMv7 changes (such as the ones above) became a thing. That's why developers
who understand the issues tend to flag stuff as both "Iyonix" and "ARMv7"
(aside from one very specific issue, it seems that ARMv8 didn't really make
a lot of difference).

--
Steve Fryatt - Leeds, England

http://www.stevefryatt.org.uk/

In article <mpro.qsfo7e02qnaxg04da.news@stevefryatt.org.uk>, Steve Fryatt
<news@stevefryatt.org.uk> wrote:

> So Iris needs alignment exceptions to be off, as it uses the
> instructions correctly in their new form. It's probably the only piece
> of RISC OS software to do so. Unfortunately, there's still a lot of old
> stuff out there which uses the instructions incorrectly, in their old
> form, and expects them to work as they did on the Iyonix.

> As to what to do, I don't know. I'll probably continue to browse on
> Linux and leave alignment exceptions on. If you wish to use Iris,
> you'll need to turn them off, however. Were I to do that, I think I'd
> turn them back on for a while when running new software for the first
> time, though.

There is a sort of half-way house. If you have them on, when you load
Iris, it gives you the option to turn them off until the next re-boot.
I'm using that at the moment to see how it goes.

--
Chris Newman

In article <5925db9de3cvjazz@waitrose.com>,
Chris Newman <cvjazz@waitrose.com> wrote:
> In article <mpro.qsfo7e02qnaxg04da.news@stevefryatt.org.uk>, Steve Fryatt
> <news@stevefryatt.org.uk> wrote:

> > So Iris needs alignment exceptions to be off, as it uses the
> > instructions correctly in their new form. It's probably the only piece
> > of RISC OS software to do so. Unfortunately, there's still a lot of old
> > stuff out there which uses the instructions incorrectly, in their old
> > form, and expects them to work as they did on the Iyonix.

> > As to what to do, I don't know. I'll probably continue to browse on
> > Linux and leave alignment exceptions on. If you wish to use Iris,
> > you'll need to turn them off, however. Were I to do that, I think I'd
> > turn them back on for a while when running new software for the first
> > time, though.

> There is a sort of half-way house. If you have them on, when you load
> Iris, it gives you the option to turn them off until the next re-boot.
> I'm using that at the moment to see how it goes.

for the benefit of a bear with little brain, where do I find the control
for this?

--
from KT24 in Surrey, England
"I'd rather die of exhaustion than die of boredom" Thomas Carlyle

In article <5925df23eccharles@candehope.me.uk>,
charles <charles@candehope.me.uk> wrote:
> for the benefit of a bear with little brain, where do I find the control
> for this?

Configure > CPU

In article <5925df23eccharles@candehope.me.uk>, charles
<charles@candehope.me.uk> wrote:
> In article <5925db9de3cvjazz@waitrose.com>, Chris Newman
> <cvjazz@waitrose.com> wrote:
> > In article <mpro.qsfo7e02qnaxg04da.news@stevefryatt.org.uk>, Steve
> > Fryatt <news@stevefryatt.org.uk> wrote:

> > > So Iris needs alignment exceptions to be off, as it uses the
> > > instructions correctly in their new form. It's probably the only
> > > piece of RISC OS software to do so. Unfortunately, there's still a
> > > lot of old stuff out there which uses the instructions incorrectly,
> > > in their old form, and expects them to work as they did on the Iyonix.

> > > As to what to do, I don't know. I'll probably continue to browse on
> > > Linux and leave alignment exceptions on. If you wish to use Iris,
> > > you'll need to turn them off, however. Were I to do that, I think I'd
> > > turn them back on for a while when running new software for the first
> > > time, though.

> > There is a sort of half-way house. If you have them on, when you load
> > Iris, it gives you the option to turn them off until the next re-boot.
> > I'm using that at the moment to see how it goes.

> for the benefit of a bear with little brain, where do I find the control
> for this?

Following up my own message. Found it.

Charles

--
from KT24 in Surrey, England
"I'd rather die of exhaustion than die of boredom" Thomas Carlyle

In article <5925df23eccharles@candehope.me.uk>,
charles <charles@candehope.me.uk> wrote:
> for the benefit of a bear with little brain, where do I find the
> control for this?

If you launch Iris with AEs on, a warning box will be displayed,
giving you the choice of turning AEs off or not.

--
Chris Johnson

In article <5925db9de3cvjazz@waitrose.com>, Chris Newman
<cvjazz@waitrose.com> wrote:
> In article <mpro.qsfo7e02qnaxg04da.news@stevefryatt.org.uk>, Steve
> Fryatt <news@stevefryatt.org.uk> wrote:

> > As to what to do, I don't know. I'll probably continue to browse
> > on Linux and leave alignment exceptions on. If you wish to use
> > Iris, you'll need to turn them off, however. Were I to do that, I
> > think I'd turn them back on for a while when running new software
> > for the first time, though.

> There is a sort of half-way house. If you have them on, when you
> load Iris, it gives you the option to turn them off until the next
> re-boot. I'm using that at the moment to see how it goes.

I normally run with Alignment Exceptions On, and I have never seen any
such message from Iris. Even running with v1.010 (4 April 2021).

--
Martin Avison
Note that unfortunately this email address will become invalid
without notice if (when) any spam is received.

In message <4d2ad02559.chris@mytarbis.plus.com>
Chris Hughes <news13@noonehere.co.uk> wrote:

> I have been running with Alignment Exception Off now since I installed
> this version of Iris, I used to have it on.

My take on it would be that if you have had alignment exceptions On for a
long time and not had any errors, then all the software you are using is
ok and switching them Off now will be fine.

Bryan.

--
RISC OS User Group Of London - http://www.rougol.jellybaby.net/
RISC OS London Show - http://www.riscoslondonshow.co.uk/

In message <5925e441efNews03@avisoft.f9.co.uk>
Martin <News03@avisoft.f9.co.uk> wrote:

> I normally run with Alignment Exceptions On, and I have never seen any
> such message from Iris. Even running with v1.010 (4 April 2021).

I thnk the warning was only added to the version that was made more
generally available, it wasn't in the release to ROD subscribers.

Bryan.

--
RISC OS User Group Of London - http://www.rougol.jellybaby.net/
RISC OS London Show - http://www.riscoslondonshow.co.uk/

In article <cd17f02559.bryan@helpful-demon.co.uk>,
Bryan Hogan <spam@nowhere.invalid> wrote:
> In message <5925e441efNews03@avisoft.f9.co.uk>
> Martin <News03@avisoft.f9.co.uk> wrote:

> > I normally run with Alignment Exceptions On, and I have never
> > seen any such message from Iris. Even running with v1.010 (4
> > April 2021).

> I thnk the warning was only added to the version that was made more
> generally available, it wasn't in the release to ROD subscribers.

Which is why it is so important to state the version of a program you
are referring to in posts, particularly when they are changing
rapidly.

--
Martin Avison
Note that unfortunately this email address will become invalid
without notice if (when) any spam is received.

In article <82b9c02559.chris@mytarbis.plus.com>, Chris Hughes
<news13@noonehere.co.uk> wrote:

> Everyone needs to remember Iris is still Beta software, i.e. not finished

> They have a long list of things they want to improve, but this takes
> time and money plus developer time of course. Its very important the
> above bit.

> Collate issues - bug fixes are probably more important then new features.

> Also remember this is an extremely big program and uses the WebKit stuff
> used on all the other platforms so do not except everything to be
> totally style guide. Be happy we have much better browser.

That is why I've been avoiding raising 'issues' with Andrew. However they
do exist, so leave me at times with the question: which ones do the
developer(s) already know about?... particularly if no-one else has
reported them.

Making comments and reporting things isn't the same as demanding a quick
fix and perfection.

Jim

--
Please use the address on the audiomisc page if you wish to email me.
Electronics https://www.st-andrews.ac.uk/~www_pa/Scots_Guide/intro/electron.htm
biog http://jcgl.orpheusweb.co.uk/history/ups_and_downs.html
Audio Misc http://www.audiomisc.co.uk/index.html

The date being 1 May 2021, Chris Hughes <news13@noonehere.co.uk> decided
to write:

> Everyone needs to remember Iris is still Beta software, i.e. not finished

Yes, but the whole point of a Beta release is that you flush out the bugs
before it goes on general release. Obviously crashes and serious failures
take priority over wish lists.

Richard

On 01/05/2021 15:57, Steve Fryatt wrote:
> Incidentally, this answers Chris N's question in the current Archive. Saying
> something is "32-bit" is pretty meaningless these days, as it could be
> "ARMv5" (Iyonix, and Pi 1), "ARMv7" (most later stuff) or "ARMv8" (Pi 3 and
> 4). The reason that alleged "32-bit" stuff is unstable on the Pi 4 is very
> likely that it was 32-bitted for the Iyonix and never revisited since the
> ARMv7 changes (such as the ones above) became a thing. That's why developers
> who understand the issues tend to flag stuff as both "Iyonix" and "ARMv7"
> (aside from one very specific issue, it seems that ARMv8 didn't really make
> a lot of difference).

I'm finding a lot of stuff which has problems on the Iyonix was not 32
bitted at all - someone just bodged a 32 bit header on the unmodified 26
bit code, and because it didn't fall over immediately, thought "job done".

Some of those bodges fell over on ARMv6 and ARMv7, *all* of them fall
over on ARMv8 capable processors, as you cannot use any 26 bit only
instruction in any processor mode.

---druck

On 03/05/2021 21:57, druck wrote:
> I'm finding a lot of stuff which has problems on the Iyonix was not 32
> bitted at all - someone just bodged a 32 bit header on the unmodified 26
> bit code, and because it didn't fall over immediately, thought "job done".

That should read ..."a lot of stuff which has problems on the Pi 4"...
...."didn't fall over immediately on the Iyonix,"...

---druck