Poster | Thread |
kas1e
| |
New mini update from Hyperion part #2 Posted on 18-May-2021 6:19:16
| | [ #1 ] |
|
|
|
Elite Member |
Joined: 11-Jan-2004 Posts: 3549
From: Russia | | |
|
| @All Another update !
This time is a necessary bugfix for an installer binary done by our Broadblues (Andy Broad). Bug can manifest in different conditons, but were found when we testing installation of latest Enhancer. Then it was reported to hyperion's bugzilla and fixed by Andy. And thanks to Simon it now available for all. Last edited by kas1e on 18-May-2021 at 06:26 AM.
_________________ Join us to improve dopus5! zerohero's mirror of os4/os3 crosscompiler suites |
|
Status: Offline |
|
|
Trixie
| |
Re: New mini update from Hyperion part #2 Posted on 18-May-2021 9:48:07
| | [ #2 ] |
|
|
|
Amiga Developer Team |
Joined: 1-Sep-2003 Posts: 2090
From: Czech Republic | | |
|
| @kas1e
Good to see updates coming!
_________________ The Rear Window blog
AmigaOne X5000/020 @ 2GHz / 4GB RAM / Radeon RX 560 / ESI Juli@ / AmigaOS 4.1 Final Edition SAM440ep-flex @ 667MHz / 1GB RAM / Radeon 9250 / AmigaOS 4.1 Final Edition |
|
Status: Offline |
|
|
kas1e
| |
Re: New mini update from Hyperion part #2 Posted on 18-May-2021 11:15:24
| | [ #3 ] |
|
|
|
Elite Member |
Joined: 11-Jan-2004 Posts: 3549
From: Russia | | |
|
| |
Status: Offline |
|
|
virgolamobile
| |
Re: New mini update from Hyperion part #2 Posted on 18-May-2021 11:57:29
| | [ #4 ] |
|
|
|
Regular Member |
Joined: 23-Feb-2004 Posts: 192
From: Somewhere in Northern Italy | | |
|
| @kas1e @Trixie
Great! Finally from words to actions! Thanks to all people involved!
I am unsure if I want to know the reasons causing Hyperion to be more active recently, but the most important thing is the result |
|
Status: Offline |
|
|
BigD
| |
Re: New mini update from Hyperion part #2 Posted on 18-May-2021 13:15:38
| | [ #5 ] |
|
|
|
Elite Member |
Joined: 11-Aug-2005 Posts: 7322
From: UK | | |
|
| @virgolamobile
Quote:
I am unsure if I want to know the reasons causing Hyperion to be more active recently... |
Do you think it's simply a use it or use it approach? Plus there's lower risk as they don't have to pay the workers on the Classic side of things! Maybe the Classic OS profits fund the NG development?Last edited by BigD on 18-May-2021 at 01:16 PM.
_________________ "Art challenges technology. Technology inspires the art." John Lasseter, Co-Founder of Pixar Animation Studios |
|
Status: Offline |
|
|
NutsAboutAmiga
| |
Re: New mini update from Hyperion part #2 Posted on 18-May-2021 18:02:18
| | [ #6 ] |
|
|
|
Elite Member |
Joined: 9-Jun-2004 Posts: 12818
From: Norway | | |
|
| |
Status: Offline |
|
|
Chris_Y
| |
Re: New mini update from Hyperion part #2 Posted on 18-May-2021 23:48:26
| | [ #7 ] |
|
|
|
Elite Member |
Joined: 21-Jun-2003 Posts: 3203
From: Beds, UK | | |
|
| @NutsAboutAmiga
Quote:
NutsAboutAmiga wrote: @BigD
My guess is updating AmigaOS3.2 fixing bugs can directly benefit AmigaOS4.1 update 2 as well.
|
It would be nice if that were true, but I've yet to see any evidence that OS3.1.4 and up bugfixes have made it into OS4.
Some OS4 stuff has been backported, but other improvements listed for OS3.2 appear mainly to have been written from scratch (the env-handler stuff sounds completely different to OS4), and haven't made it back to OS4 if they are improved past what they are currently in that branch (images in AmigaGuides, Pointer Prefs). I have little faith smaller bugfixes are treated any differently.
_________________ "Miracles we do at once, the impossible takes a little longer" - AJS on Hyperion Avatar is Tabitha by Eric W Schwartz |
|
Status: Offline |
|
|
klx300r
| |
Re: New mini update from Hyperion part #2 Posted on 19-May-2021 2:05:17
| | [ #8 ] |
|
|
|
Elite Member |
Joined: 4-Mar-2008 Posts: 3837
From: Toronto, Canada | | |
|
| @kas1e
great updates for my favourite OS on both 68k and PPC..thanks to all involved! _________________ ____________________________ c64-2sids, A1000, A1200T-060@50(finally working!),A4000-CSMKIII ! My Master Miggies- Amiga 1000 & AmigaOne X1000 ! mancave-ramblings X1000 I BELIEVE |
|
Status: Offline |
|
|
Steady
| |
Re: New mini update from Hyperion part #2 Posted on 19-May-2021 2:08:54
| | [ #9 ] |
|
|
|
Regular Member |
Joined: 1-Nov-2004 Posts: 211
From: Melbourne, OZ | | |
|
| @kas1e
Nice to see some fixes coming through. Thanks OS4 devs! |
|
Status: Offline |
|
|
kas1e
| |
Re: New mini update from Hyperion part #2 Posted on 19-May-2021 9:30:54
| | [ #10 ] |
|
|
|
Elite Member |
Joined: 11-Jan-2004 Posts: 3549
From: Russia | | |
|
| @Chris_Y
Quote:
It would be nice if that were true, but I've yet to see any evidence that OS3.1.4 and up bug fixes have made it into OS4.
Some OS4 stuff has been backported, but other improvements listed for OS3.2 appear mainly to have been written from scratch (the env-handler stuff sounds completely different to OS4), and haven't made it back to OS4 if they are improved past what they are currently in that branch (images in AmigaGuides, Pointer Prefs). I have little faith smaller bug fixes are treated any differently.
|
Env-handler for OS4 (together with app-handler, new dos.library, etc) all was written from scratch of late by Colin, and they far better than those ones in OS3.2, so there no need to worry about them.
But in others, those developers who still left working under Ben's control, working for free, in their free time, on the things they want. And I didn't see any of them are very energetic with time and motivation. It's just luck for Ben ppls working for free for him on OS3.x.
There is no more profit for him on the OS4 front, so we can be 99% sure that he will spend shit on paying developers. But from another side, strange why he just didn't sell out AOS4 components, he owns completely to someone who in interest still. Year, or two, and no one will give him for OS4 ownership a shit.
And probably no one will be in will to buy anything as well from him, he will cheat everyone, so later to take the rights back :)
_________________ Join us to improve dopus5! zerohero's mirror of os4/os3 crosscompiler suites |
|
Status: Offline |
|
|
Chris_Y
| |
Re: New mini update from Hyperion part #2 Posted on 19-May-2021 20:27:21
| | [ #11 ] |
|
|
|
Elite Member |
Joined: 21-Jun-2003 Posts: 3203
From: Beds, UK | | |
|
| @kas1e
Quote:
kas1e wrote:
Env-handler for OS4 (together with app-handler, new dos.library, etc) all was written from scratch of late by Colin, and they far better than those ones in OS3.2, so there no need to worry about them.
|
I wasn't worrying about them, it just seems that there's two branches that work differently, when really we should be moving OS3.2 and OS4 towards being more compatible. Fortunately the env stuff isn't relevant from an API compatibility perspective, just user experience might be different.
That was just one example where features and fixes have been done independently rather than OS3.2 and OS4 feeding updates into one another for the benefit of both (and not doing the same work twice).
_________________ "Miracles we do at once, the impossible takes a little longer" - AJS on Hyperion Avatar is Tabitha by Eric W Schwartz |
|
Status: Offline |
|
|
redfox
| |
Re: New mini update from Hyperion part #2 Posted on 19-May-2021 22:47:32
| | [ #12 ] |
|
|
|
Elite Member |
Joined: 7-Mar-2003 Posts: 2066
From: Canada | | |
|
| Updated the system Installer utility via AmiUpdate. It's nice to be using AmiUpdate again for these mini updates.
Thanks to Andy Broad, Trixie, kas1e and Simon.
--- redfox
Last edited by redfox on 19-May-2021 at 10:49 PM. Last edited by redfox on 19-May-2021 at 10:48 PM.
|
|
Status: Offline |
|
|
matthey
| |
Re: New mini update from Hyperion part #2 Posted on 20-May-2021 0:23:14
| | [ #13 ] |
|
|
|
Elite Member |
Joined: 14-Mar-2007 Posts: 2008
From: Kansas | | |
|
| Chris_Y Quote:
Some OS4 stuff has been backported, but other improvements listed for OS3.2 appear mainly to have been written from scratch (the env-handler stuff sounds completely different to OS4), and haven't made it back to OS4 if they are improved past what they are currently in that branch (images in AmigaGuides, Pointer Prefs). I have little faith smaller bugfixes are treated any differently.
|
kas1e Quote:
Env-handler for OS4 (together with app-handler, new dos.library, etc) all was written from scratch of late by Colin, and they far better than those ones in OS3.2, so there no need to worry about them.
|
Let's take a closer look at the AmigaOS_3.2-FAQ.txt for how the env-handler/happy-env feature is supported.
AmigaOS_3.2-FAQ.txt Quote:
13.4 * Where can I find the env-handler/happy-env feature I read about? How do I use it?
The RAM-Disk now supports "external hardlinks". You can create, for RAM: only, hardlinks to files and directories *outside* of RAM:.
A practical example:
MakeLink RAM:ENV to ENVARC: FORCE
This creates a hardlink from "RAM:ENV" to the ENVARC: assign, which lives outside of RAM:. Clearly, a contradiction in terms, but wait...
Whenever you now access a file in that "hardlinked" directory, RAM: will automatically retrieve that file from the external resource.
So what the line above does is that it performs a "copy on demand" from ENVARC: to ENV:
And that is exactly the purpose: Integrate env-handling features into RAM, and avoid unnecessary copying of files into RAM: in the Startup-sequence script.
We already incorporated the changes necessary in the Startup-sequence to work right out of the box. This feature ends up saving a discrete amount of memory and makes the system boot a tiny bit faster.
|
I believe the first part of the AmigaOS_3.2-FAQ.txt was written by ThoR but this part of the text was written by someone with less technical knowledge. The "external hardlinks" described are actually soft links or symbolic links. The AmigaOS has supported soft links since kickstart 2.0 although there have been many bugs including in various C: commands. The 68k ram disk did not support soft links until AmigaOS 3.9 and it looks like the AmigaOS 3.2 ram disk supports them too. The "MakeLink RAM:ENV to ENVARC: FORCE" command may be suggesting new behavior though. Is RAM:ENV checked for a file before following the soft link to ENVARC: if the file is missing? Would this new behavior make env-handler/happy-env programs unnecessary?
|
|
Status: Offline |
|
|
Chris_Y
| |
Re: New mini update from Hyperion part #2 Posted on 20-May-2021 11:03:17
| | [ #14 ] |
|
|
|
Elite Member |
Joined: 21-Jun-2003 Posts: 3203
From: Beds, UK | | |
|
| @matthey
Quote:
Is RAM:ENV checked for a file before following the soft link to ENVARC: if the file is missing? |
That's the impression I get.
Quote:
Would this new behavior make env-handler/happy-env programs unnecessary?
|
Yes, if it works as described, however "copy on demand" isn't how softlinks are supposed to work!
I think having a dedicated handler for ENV is a neater solution.
_________________ "Miracles we do at once, the impossible takes a little longer" - AJS on Hyperion Avatar is Tabitha by Eric W Schwartz |
|
Status: Offline |
|
|
Gregor
| |
Re: New mini update from Hyperion part #2 Posted on 20-May-2021 15:04:40
| | [ #15 ] |
|
|
|
Regular Member |
Joined: 12-Sep-2011 Posts: 212
From: Unknown | | |
|
| In my system (AmigaOS 4.1FEu2, X5000) AmiUpdate does not see at all that Installer update... I have in the 'servers' menu 'www.amiupdate.net' and 'update.amigaos.net', and both are selected.
Here is the au.log which appeared in RAM disk:
16:54:06 Program started
16:54:06 Started system scan
16:54:06 Found host machine as: X5020
16:54:06 Synchronising with server: 'www.amiupdate.net'...
16:54:06 Trying to connect to: www.amiupdate.net on port: 80
16:54:06 Connected
16:54:06 Redirection required, following link....
16:54:06 Trying to connect to: codebench.co.uk on port: 80
16:54:06 Connected
16:54:06 Failed to get filesize from header, Content-Length: tag missing
16:54:06 Trying to connect to: codebench.co.uk on port: 80
16:54:06 Connected
16:54:07 Synchronisation successful
16:54:07 This update has a listed requirement with an invalid syntax (Amiga PPC and AmigaOS4 final edition Update 2), checking aborted 16:54:08 Synchronising with server: 'update.amigaos.net'...
16:54:08 Trying to connect to: www2.hyperion-entertainment.com on port: 80
16:54:08 Connected
16:54:08 User and Pass found in sitelist
16:54:08 Trying to connect to: www2.hyperion-entertainment.com on port: 80
16:54:08 Connected
16:54:08 Failed to get filesize from header, Content-Length: tag missing
16:54:08 Trying to connect to: www2.hyperion-entertainment.com on port: 80
16:54:08 Connected
16:54:09 Synchronisation successful
16:54:09 Scan completed
16:54:09 Waiting for user input...
16:54:13 Closing program |
|
Status: Offline |
|
|
matthey
| |
Re: New mini update from Hyperion part #2 Posted on 20-May-2021 16:45:12
| | [ #16 ] |
|
|
|
Elite Member |
Joined: 14-Mar-2007 Posts: 2008
From: Kansas | | |
|
| Chris_Y Quote:
Yes, if it works as described, however "copy on demand" isn't how softlinks are supposed to work!
I think having a dedicated handler for ENV is a neater solution.
|
I'm not sure the author of the AmigaOS 3.2 FAQ section about the ENV handler has the technical knowledge to understand what is going on under the hood. The "copy on demand" may not even be correct and does seem like extra unnecessary work for a softlink.
xxx Prefs startup: read from ENV:xxx.prefs xxx Prefs use: write to ENV:xxx.prefs xxx Prefs save: write to ENVARC:xxx.prefs and ENV:xxx.prefs
MakeLink RAM:ENV to ENVARC: FORCE
The non-technical author is describing a caching system for soft link directories which copies accessed files from the soft linked "to" location to the "from" location. This does *not* seem elegant and I doubt that it would be ThoR approved which is why I wonder if the author understands what is going on under the hood.
xxx Prefs startup: read from ENV:xxx.prefs 1. if !exists(ENV:xxx.prefs) copy(ENVARC:xxx.prefs to ENV:xxx.prefs) 2. read(ENV:xxx.prefs)
I believe the more elegant way of implementing directory soft links would be what I described earlier.
xxx Prefs startup: read from ENV:xxx.prefs 1. if exists(ENV:xxx.prefs) read(ENV:xxx.prefs) else read(ENVARC:xxx.prefs)
Now there is no copy on demand and only prefs which "use" or "save" settings are located in memory. Both the copy on demand and env-handler methods load accessed xxx.prefs files into ENV: taking up memory and leave them there even when there are no changes made in the prefs. The env-handler method takes extra memory for the handler itself and I suspect that file/directory notification is more CPU intensive. IPrefs is already using notification to watch for ENV: changes which may be convenient but is likely more processor intensive than sending a message to a public port.
As for how soft links are "supposed" to work, it varies by OS.
https://en.wikipedia.org/wiki/Symbolic_link
It would be good to have more discussion before changing behavior and better documentation after changing behavior.
Last edited by matthey on 21-May-2021 at 05:08 AM. Last edited by matthey on 21-May-2021 at 04:50 AM. Last edited by matthey on 20-May-2021 at 04:47 PM.
|
|
Status: Offline |
|
|
kolla
| |
Re: New mini update from Hyperion part #2 Posted on 20-May-2021 20:15:54
| | [ #17 ] |
|
|
|
Elite Member |
Joined: 21-Aug-2003 Posts: 2894
From: Trondheim, Norway | | |
|
| @matthey
Quote:
xxx Prefs startup: read from ENV:xxx.prefs 1. if exists(ENV:xxx.prefs) read(ENV:xxx.prefs) else read(ENVARC:xxx.prefs)
|
Yeah, if only we had sources and compilers for all software that ever use env:/envarc:
No really, a dedicated ENV-handler is the only way - I know I will have breakage with the new "clever" softlink method, as I already use softlinks on RAM: for totally different purposes. Sadly, discussing any sort of issues with Thomas Richter is pointless, he has made up his mind and anyone who contradict him need psychiatric help.Last edited by kolla on 20-May-2021 at 08:18 PM.
_________________ B5D6A1D019D5D45BCC56F4782AC220D8B3E2A6CC |
|
Status: Offline |
|
|
matthey
| |
Re: New mini update from Hyperion part #2 Posted on 20-May-2021 21:22:26
| | [ #18 ] |
|
|
|
Elite Member |
Joined: 14-Mar-2007 Posts: 2008
From: Kansas | | |
|
| kolla Quote:
Yeah, if only we had sources and compilers for all software that ever use env:/envarc:
|
AmigaOS 3.2 should be out now so perhaps someone has tested it and can say which behavior it uses (copy on demand as the FAQ states, the more elegant method I refer to or something else). Also, is it breaking any 3rd party prefs?
kolla Quote:
No really, a dedicated ENV-handler is the only way - I know I will have breakage with the new "clever" softlink method, as I already use softlinks on RAM: for totally different purposes. Sadly, discussing any sort of issues with Thomas Richter is pointless, he has made up his mind and anyone who contradict him need psychiatric help. |
Have you been using soft links of directories (MakeLink from and to both specify directories with FORCE option designating a soft link)?
ThoR could be more open minded but generally makes logical decisions. I would describe him as an eccentric genius. Olsen may be the development team leader and is more practical and has better social skills including communication. Together they make a very good core team. I would rather not jump to conclusions and would like to hear their reasoning on any behavior changes. I doubt there are very many users using directory soft links on the 68k due to the numerous bugs in the past. It would be nice to do away with the env-handler saving memory and processing power if there was a clever and elegant way to do it with a minor behavior change resulting in a minimum amount of incompatibility. The 68k AmigaOS is going to be very handicapped with enhancements if 100% compatibility has to be maintained.
|
|
Status: Offline |
|
|
abalaban
| |
Re: New mini update from Hyperion part #2 Posted on 20-May-2021 21:55:43
| | [ #19 ] |
|
|
|
Super Member |
Joined: 1-Oct-2004 Posts: 1114
From: France | | |
|
| @kolla
No really, a dedicated ENV-handler is the only way - I know I will have breakage with the new "clever" softlink method, as I already use softlinks on RAM: for totally different purposes. Sadly, discussing any sort of issues with Thomas Richter is pointless, he has made up his mind and anyone who contradict him need psychiatric help.[/quote]
And what if we for example say that the features of any ENV-handler have been integrated into the RAM-handler and activated by the presence of a softlink to a directory in RAM? I do not even see the need for the "copy on demand": just use the original file in its original location when none is present in the RAM, and create one in RAM when writing, next time that local file will be used instead of the original one.
I do not forsee breakage for tons of software as this does not need to change how softlinks work OS-wise, only those in ram may be interpreted differently by the handler, and no guarantee has ever been given that all handlers will resolve softlinks similarly (if only they support them at all). _________________ AOS 4.1 : I dream it, Hyperion did it ! Now dreaming AOS 4.2... Thank you to all devs involved for this great job ! |
|
Status: Offline |
|
|
Chris_Y
| |
Re: New mini update from Hyperion part #2 Posted on 20-May-2021 22:49:31
| | [ #20 ] |
|
|
|
Elite Member |
Joined: 21-Jun-2003 Posts: 3203
From: Beds, UK | | |
|
| I can see an issue if you save to a softlinked directory in RAM:, expecting it to write to the original location... however it only writes it to RAM:. This could cause confusion (and loss of data)
OK, so the chances of somebody creating a directory softlink in RAM: and then using it to read/write files is slim, but it's still unexpected behaviour.
Software breakage though... if it is broken by this, it was likely broken by HappyENV years ago. Not likely.
Last edited by Chris_Y on 20-May-2021 at 10:52 PM. Last edited by Chris_Y on 20-May-2021 at 10:50 PM.
_________________ "Miracles we do at once, the impossible takes a little longer" - AJS on Hyperion Avatar is Tabitha by Eric W Schwartz |
|
Status: Offline |
|
|