Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Messages - Dariusz Piatkowski

Pages: [1] 2 3 ... 78
1
Storage / Re: JFS failing in a disastrous way - without the user knowing
« on: December 02, 2022, 02:39:41 pm »
Lars, everyone...

Sorry, but this is ridiculous.
...
But I am not the master of disaster. That is AN. They have the source code. And of course, it's closed source

Ahh, you see, and therein lies part of the challenge that causes majority of us to take a more pragmatic approach to dealing with these issues.

Yes, you are absolutely right about this being ridiculous, however, what other choice do we really have? There certainly isn't much ability for any of us to action it, even folks like yourself who have the technical know-how are finding themselves having their hands tied.

Sooo...instead of continuing the risk exposure I will do the best I can, with the tools/controls available to me today, to avoid that risk.

2
Storage / Re: JFS failing in a disastrous way - without the user knowing
« on: December 01, 2022, 06:17:54 pm »
If you fear that something changes your config.xxx and you overlook to check the changes before reboot you can also start a rexx script from your startup.cmd which parses your current config.xxx...

You bring up a reason for why for every install I actually save a copy of CONFIG.SYS first and then afterwards compare the new copy to that backup.

This of course could be automated just as you say with a little REXX script very easily.

3
Storage / Re: JFS failing in a disastrous way - without the user knowing
« on: November 30, 2022, 02:11:56 pm »
Oh dude...I feel your pain!!!

1st of all, glad that you recovered, but yeah, as I read your post I likewise had the tell-tale feeling of "butterflies" in my stomach, old memories came back here.

A few years back I made the move from HPFS386 to JFS, but since this was NOT a fresh install I managed to convert stuff (standard XCOPY partition move approach, etc.) while UNICODE.SYS was NOT preceeding the JFS call in CONFIG.SYS. Bottom line being the NEW system was quite usable, appeared to work just fine, but soon enough strange things started to happen and eventually I had a major JFS problem on my hands.

In this case the root cause was my unfamiliarity with true JFS requirements and the fact that I missed tossing UNICODE.SYS into my CONFIG.SYS ahead of the JFS IFS call and ahead of the partition move process itself.

Your root cause is quite different from that, but at the time of my troubles I do recall thinking to myself: "well, if the darn thing is THAT important why isn't the IFS at least shouting - should really scream at me though - when it finds it missing?".

4
Programming / Re: Building Qt5
« on: November 24, 2022, 03:02:55 am »
Small clarification.

Programs crash happens not because of "store the registers during context switches".
It happens because  programs detects AVX support by OS and thus uses AVX branches of  their code.

And as you rightly said, our set (GCC + linker) is not able to  generate correct AVX code. But there is hope.  :)

OK, the BOLD highlight is mine...and that's also the focus point: so enlighten us, what exactly does that mean? (is someone else working on getting the GCC thing fixed-up for us? I ask b/c AFAIK Paul is the only one who actively maintains and ports the latest releases...sooo....????)

5
Web applications / Re: Our FF and Sync, anyone tried it?
« on: November 24, 2022, 02:59:30 am »
Thanks guys!

Dave,
Yeah, I suspect this is ultimately going to be proven out through pure testing.

Neil,
Do you recall by any chance what some of the misses were? Is it, as Dave pointed out, the difference in the Sync versioning between our 45.9x release and the most recent Win stuff (which I think is 106.0.5, at least on my Win2k box here)?

6
Web applications / Our FF and Sync, anyone tried it?
« on: November 23, 2022, 04:43:47 pm »
Our FF has been a great browser for some time, but here at least it's finally starting to show it's age and given that some of the key sites I use (banking, etc.) are now problematic I'm looking for alternative solutions.

Until a replacement browser is up and running, and i'm talking as in PRODUCTION ready stuff, I will continue to rely on Win hosted FF that's accessible through RDP.

OK, so that works here just fine, but now I need to figure out how best to sync-up all my information: bookmarks, ID/PASS combos, etc.

So I've done a little reading on the Mozilla FF Sync process, the logic, security implementation, etc. Seems like a viable alternative to manually attempting to keep all this synchronized across multiple devices...however, none of it will mean anything if our OS/2 FF Sync functionality is so badly out of date that it's practially unusable.

Therefore, I'm curious if anyone has tried this? Does it flat-out "work", or are there some glitches?

Next step here will be to set up some test accounts and FF profiles to give this a whirl...but I figured I might as well get a little smarter on what this means for us.

Thanks!

7
Storage / Re: HPFS and JFS
« on: November 21, 2022, 05:39:58 am »
Plain HPFS is no comparison to JFS.

Now toss HPFS386 into the mix and you will fare better (in fact some tests did have HPFS386 performing faster, but I think that was only the case for non-SMP systems and some network driven access), but still HPFS won't deliver the options that JFS has.

I for one got tired of having multiple HPFS 64G partitions kicking around...and when CHKDSK runs...oh man, you best come back another day...LOL, yeah, it sucked!

JFS is pretty steady here, that's with a massive 1G cache, still multiple partitions, but I used those for a nightly backup, so should anything go wrong, there is an immediate recovery option available.

8
Hardware / Re: Printer - last page of print job missing...ugh?
« on: November 10, 2022, 03:18:05 pm »
Ian and Per,

Sorry guys, I should have gone slower on my end testing and explaining.

My description left out an important fact (as I was trying this with multiple different PDFs, and unfortunately mixed up some of the results): this particular document that I shared, I can NOT print pg.5, which also happened to be the LAST page I was attempting to print b/c I was skipping printing of pg.6, which contains just the disclaimer.

This pattern occured in a number of PDFs (all published by the same bank, so presumably same PDF publishing app, I haven't checked all though).

I am going to re-visit later today, maybe it's just ODD or EVEN number of pages??? Seems crazy, or maybe there is something off with the page content that's throwing the printer off?

I was able to print that very same page from GSView to my ePDF printer though...so it's not like the OS/2 printing sub-system is failing...there seems to be something in the PostScript itself that's now causing the issue.

9
Hardware / Printer - last page of print job missing...ugh?
« on: November 10, 2022, 04:34:20 am »
So this is an entirely new occurrence: the Financial Market Overview/Commentary articles that my bank publishes all of a sudden - when printed by GSView - are always dropping the last page...as in: it does NOT print! Neither does attempting to print just that SINGLE page. Printer's "receiving data" LED flashes, but nothing ever prints.

Printer in question is my Brother HL-5470DW, PostScript, I'm using the current drivers that have been adjusted with a custom PPD (to use all printer functions). Regardless, these articles have been printed numerous times before and never had this issue. Granted, they were the past articles so maybe the Bank changed something which is generating the PDFs themselves, I simply cannot tell.

Other apps seem to print fine, other PDFs, etc...but I find this really weird. I've attached a sample of such a PDF...is anyone else having a problem printing it?

No printer/GSview/system changes have occurred recently, as far as I know anyways!!!

All ears here...any ideas what to look for/at in order to debug this further?

10
Setup & Installation / MARKEXE vs EXECMODE - differences?
« on: October 15, 2022, 05:47:06 pm »
Hi everyone,

I've had a persistent problem with some apps putting a complete soft-lock of my machine. This normally happens during application start-up, but occasionally it would also happen throughout execution. I know, it sounds weird, but it's absolutely consistent and in appears to be related to an underlying component DLL, although I'm still working through pegging that detail.

Anyways, so to avoid that dreaded outcome I found that the best way to deal with this problem is to start up the "offending" app with all CPU cores but core_0 disabled. I do this with the handy setproc utiility.

Alright...but doing this manually is a PITA, so I have a quick SMPOFF/ON wrapper *.cmd template that I have been using for such apps.

However, it would seem to me that perhaps flagging to offending EXE as SMP-unsafe would provide a better result. Well, I tried this but with rather bleak results, or to be more precise, no results. The lock still occurs, and maybe it's because it's a DLL issue as opposed to the main EXE (therefore parent vs child execution path issue).

But...if my understanding of how these two utilities control the SMP flags is incorrect, well, that may have a key bearing on what I'm seeing.

They both enable you to flag a particular EXE as MPU unsafe, and yet for exmaple flagging a sample EXE with execmode doesn't seem to show the correct status when querying it up in markexe, so what gives?

I've got some examples here to illustarte this, but for starters I'm looking for a general understanding of what's different in terms of how these two utilities mark a EXE?

11
Storage / Re: Cloning a drive
« on: October 14, 2022, 11:09:48 pm »
Has anyone successfully cloned a drive to a larger drive with ArcaOS installed on it?...

Are you looking to preserve the ArcaOS and somehow "migrate" your existing app install over to that new ArcaOS partition?

If that's the case, well, you simply can't do that, not for probably vast majority of apps. Simply moving them over to the ArcaOS partition won't bring all the other settings that may have been stored in the OS/2 INI files (OS2SYS and OS2).

ArcaN was contemplating developing a migration path, but I'm fairly certain that is absolutely the lowest priority in terms of all the things they want to do.

12
Multimedia / Re: Updated Uniaud32 build
« on: September 30, 2022, 02:50:55 pm »
Mark, everyone...

Heh Tom!

command prompt line for me to disable ......... unimix -id20 -val0=0

So the problem is indeed auto-mute being enabled for this chipset by default.

However still one little drawback to solve - I have to re-enable that every time I reboot warm or cold. Do you know of a way of setting unimix to automatically apply that fix upon reboot?

Just put that unimix command in the file startup.cmd , and it will be run every time you start your system.

I was thinking about that after I posted, but haven't worked much with Startup.CMD in the past - will have a look at that tonight or tomorrow.

But thanks for the tip and Best!

Mark

Tom's suggestion is a great one, you can take that a step further and basically control the FULL behaviour of the audio card as far as the various individual controls go.

Basically, manually adjust the controls to what you want them to be first. By the sounds of it you already know how to pull up the different values and set them up.

Once done, you then save the complete set of settings to a file:

'unimix -savesomefilename'

The very next time you want to restore ALL of these settings you execute:

'unimix -loadsomefilename'

I have that load statement in my startup.cmd, which is a super easy way to re-set all the controls to a pre-defined set of values.

This is how you can handle this particular settings, as well as any others that can be adjusted by referencing their ID. In my case, I have some 53 controllable IDs ('unimix -names').

13
Applications / Re: ColorWorks ArcaOS
« on: September 30, 2022, 03:03:07 am »
It's been a while since I used it here, but what image are you trying to open? ...what format is what I'm really asking...

14
Programming / Re: RUX function failure - but on multiple tries only...
« on: September 30, 2022, 02:33:59 am »
Hi Remy,

If you would like be sure that all RXU functions you are going to call are loaded, you can check each of them at init and if some are missing, load them then instead of just testing 'rxuinit'

...so do you mean for example, instead of calling this RXU.DLL entry-point:

Code: [Select]
...
CALL RxFuncAdd 'rxuinit','RXU','rxuinit'
CALL rxuinit
...

call the RxFuncAdd with a specific function from that REXX DLL that I want to use, such as:

Code: [Select]
CALL RxFuncAdd 'RxStartSession','RXU','RxStartSession'
CALL rxuinit

The way I'm loading the DLL is as per the instructions in the DLL INF flie:

Quote
How to register the functions:

In order to use the functions, you have to register them with Rexx like this:

  call rxfuncadd 'rxuinit','rxu','rxuinit'
  call rxuinit
...

For what it's worth, here is the matrix I put together to document the various Load/un-load APIs for specific REXX DLLs:

Code: [Select]
LIBRARY NAME DLL LOAD UNLOAD VERSION API NOTES
=============== =============== =============== =============== =============== ===================================================================

RXU RXU RxuInit RxuTerm RxQuery Version API returns a whole bunch of stuff, need to parse first

REXXUTIL REXXUTIL SysLoadFuncs SysDropFuncs SysUtilVersion

RXUTILEX RXUTILEX Sys2LoadFuncs Sys2DropFuncs Sys2Version

RXEXTRAS RXEXTRAS RxExtra(LOAD) RxExtra(DROP) RxExtra(LOAD)

RXLVM RXLVM RxLvmLoadFuncs RxLvmDropFuncs RxLvmVersion Load and Unload seems to have an issue, fails on repeated attempts

15
Programming / Re: RUX function failure - but on multiple tries only...
« on: September 30, 2022, 02:20:55 am »
Thanks for all the replies guys!

So a clarification first as I went back through my code to remind myself just what exactly I clubbered together!!!

Well, turns out my initial aim was rather high-enough, meaning:

1) first I check whether I need to use a particular API (that's the rxu_api_required variable)
- this is pre-defined at the start of the REXX script, it's a generic framework I use in all my scripts, so different REXX DLLs may be loaded by different scripts

2) if I need to use a particular REXX DLL API, I then check to see if it's already been loaded into memory (that's the rxu_api variable)

3) at the very end of my code I call the DLL unload procedure where I:
- check to see if all the flagged DLLs have been previously loaded into memory (by examining the rxu_api variable)
- only if one was not previously loaded do I actually un-load it

Sooo...weird, but this was rock-solid, never any problems...then all of a sudden...BAM!!!

I booted to my MAINT partition and did a CHKDSK on my boot drive, found it to be DIRTY, so I ran with a /F parameter. That claimed to correct some things, no messages about lost dirs or files though.

I booted up, re-tested, but still the same failure.

I thought I would re-install the RXU RPM package, but strangely I'm seeing this in ANPM:

Quote
----------[ 29 Sep 2022 20:03:25 ]----------
Executing: @python2.7 G:\UTIL\ANPM\scripts\yum_reinstall.py rxu
Enabling temporary repository arcanoae-sub
Enabling temporary repository arcanoae-exp
Enabling temporary repository arcanoae-stage
Running Transaction Check
Non-fatal POSTIN scriptlet failure in rpm package rxu-1.a.0-3.oc00.pentium4
ERROR: wpi4rpm: Please, close WarpIN and reinstall "rxu" again.


ERROR: wpi4rpm: Please, close WarpIN and reinstall "rxu" again.


warning: %post(rxu-1.a.0-3.oc00.pentium4) scriptlet failed, exit status 253

Return code: 0
----------[ 29 Sep 2022 20:03:40 ]----------

Not sure if the mirror is down or something like that (as I do get a warning from ANPM itself), but I'll re-try later.

For now I have taken the advice to NOT un-load the DLL, so that's been commented out and things appear to work.

Pages: [1] 2 3 ... 78