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 - Valery Sedletski

Pages: 1 ... 16 17 [18] 19 20 ... 25
256
Applications / Re: VirtualBox v5.0.51 Testing
« on: April 17, 2018, 03:29:33 am »
2Mark Szkolnicki: Updating WinXP to win7 is possible, but note that win7 will require significantly more resources than winxp. The bare minimum for Win7 is 768 MB of RAM, but in case you disabled all "bells and whistles" like Aero and visual effects, and enable VT-x/AMD-V. Otherwize, it will run too slow. WinXP with 512 MB of RAM is an optimal variant even without HW virtualization. So, I'd not recommend to upgrade it to Win7.

> I can say that they do open (a hopeful start) and that I can access files from the host computer and others, through my peer network. I did try enabling the "shared clipboard" in VBOX through SETTINGS | GENERAL then the Advanced Tab and found that attempting to cut and paste between the host and the VM locked things up. Only tried "Bidirectional" at the moment (also has a Host to VM or VM to Host option), but, I expect, this being a developmental version that some code linking the OS and the VM clipboards is still needed.

Shared Folders should work with different guests (excluding OS/2 atm. I did not finished the shared folders IFS yet. But it currently allows navigating the file system and viewing the files. Though, it can crash the VM easily, so it is not stable yet, but partially working). So, host support for shared folders on OS/2 is working, but guest support is not yet.

Inversely, Shared clipboard host support is not yet finished, so it can easily hang your host system! So, beware. Support for shared folders on OS/2 guests should work, though.

2ivan: running two different VM's on top of OS/2 host is possible. At least, I ran two VM's, one with WinXP, another with Knoppix LiveCD (Linux). They are not prevent to run each other. Running a VBox inside VBox is not poossible, AFAIK, because VBox does not support nested virtualization. QEMU supports it. So, I was able to run VBox inside QEMU. It worked, but very slow. KolibriOS started for an hour, or so, in such VBox instance. (This is without HW virtualization/KVM. With KVM enabled, VBox runs relatively well inside QEMU).

257
Applications / Re: VirtualBox v5.0.51 Testing
« on: April 17, 2018, 03:10:36 am »
2David Graser: QSINIT is mostly compatible with os2ldr from OS/4, but it has more features

258
Applications / Re: VirtualBox problems
« on: April 15, 2018, 03:03:35 am »
2roberto: So, With VAL=2880 VBox starts ok, but you got error -1018? Unfortunately, no fix for this error, so far. On IBM kernels, CPU ID is got as Local APIC ID, which is not guaranteed to be in a contiguous range. So, when you got into the "hole", you got this error. The only means to get rid of this error is currently, install an OS/4 kernel. It can get CPU ID's reliably. (In the future, maybe, some workaround for IBM kernels is possible, but for now, use OS/4 kernel). Also note that if you'll get this error or not, depends on LAPIC ID's in your system. If they are contiguous, all is ok (like on all three of my machines). If not, you can get this error.

VT-x should be disabled in VBox settings, if you have IBM kernel!

259
Applications / Re: VirtualBox v5.0.51 Testing
« on: April 15, 2018, 02:24:56 am »
2Dave Yeo: Yes, good note! Modern videocards have too much on-board RAM. Mine, for example has 256 MB (it is Radeon 9600 XT), so I decreased memory mapping to 16 MB in SNAP, which is sufficient for 2D acceleration and my FullHD screen to work. This will greatly economize the kernel address space. The key for VBox to work is to get more available address space to userland (hence, increase VAL to 3072). But increasing VAL requires decreasing JFS cache, if it is too much, and THREADS (to e.g., 511)/PROCESSES (to e.g., 128). Also, you need to control how much free space remained in the kernel arena, with Theseus (see the end of System->Kernel Information->Kernel Memory usage report):

Quote
allocated committed   present  resident  Totals
 --------  --------  --------  --------
 38133000  12D93000  12D15000  111E3000  Total (in bytes)
   918732    308812    308308    280460  Total (in Kbytes)
  897.199   301.574   301.082   273.887  Total (in Mbytes)
< End of THESEUS4 (v 4.001.00) output @ 3:01:58 on 15/4/2018 >

You can see that here is 897.199 of 1024 MB is occupied. So, 100+ MB is free, which is sufficient. I have VAL=3072, PROCESSES=156, THREADS=511, and JFS cache set to 128 MB explicitly:

VIRTUALADDRESSLIMIT=3072
EARLYMEMINIT=TRUE
DLLBASING=OFF
PROCESSES=156
THREADS=511
ifs=d:\os2\jfs.ifs /autocheck:* /cache:128000 /lazy:8,60,5 /p1

PS: Also, you can try to decrease Video memory used  by your video card in BIOS, if you have such setting (in case you have an integrated video card).

2Dariusz Piatkowski: Documentation is good enough, at least, most needed configuration options are described ok. You can ask develpers, if in troubles, they are on this forum and on IRC.

Installation is simple, as any other OS/2 kernel. You have to copy kernel and os2ldr instead IBM's ones, or install QSINIT, which is better to use. Both QSINIT and OS/4 os2ldr support multibooting several kernels. (But there is a number of drivers supplied with the kernel, which should be copied to \os2\boot. These drivers have no name conflicts with existing drivers. AlexT wrote a short HOWTO for quick setup, which is included in latest kernels. Also, there is doscalls1.dll which is optional. It is compatible with IBM's version). Also, there is a multibooting feature of OS/4 os2ldr (and QSINIT works the same time, I use it). So, you can boot multiple kernels from the os2ldr menu (it is better described in QSINIT documentation). Also, OS/4 has built-in CONFIG.SYS editor (activated with Alt-E in loader menu), which is handy if you need to REM out or add something temporarily. It has a good logging feature (a central system log, available for reading via a COM port, or via "copy KERNLOG$ kernlog.txt" after the system booted successfully). Very handy for developers, to debug drivers. Also, there are some enhancements, like hi-res HPET timer support (no hi-res timer bug in apps like Mozilla, like ACPI.PSD has), support for CPU rendez-vous (running the code on all CPU's in parallel), which is used in VBox for VT-x/AMD-V support etc., etc.

Regarding ACPI, yes, OS/4 devs got boring patching each time every new version of ACPI.PSD, so they gave up. But they created their own ACPI module (which is not yet released, it is available for testers, and those who is not too lazy to debug it to make it working, with devs' help). So, ACPI.PSD should be not a problem too.

260
Applications / Re: VirtualBox v5.0.51 Testing
« on: April 15, 2018, 01:48:46 am »
2Doug Clark: VAL=1536 or 2048 may be too small for running VBox! You need VAL=3072. Please reread what has been written in this thread (and another one VBox thread) carefully. You don't need to switch off AMD-V/VT-x in BIOS settings -- only in VBox settings! Your system supports VT-x/AMD-V, but your kernel is not. That's because it is enabled on your system. My system does not support AMD-V, because it is an early Athlon 64 system, so it is turned off by default on my system. So please, turn the checkbox off.

You'd better not start old VPC VM's in VBox, better create a new one, because it has different hardware emulated.

> Unable to allocate memory for mode info cache!
> Fatal error in driver. Use Alt-Ctrl-Del to reboot.

Never seen this. Looks like you have insufficient kernel memory. Are you sure you have correct VAL and decreased JFS cache and THREADS/PROCESSES accordingly? Do you have EARLYMEMINIT=TRUE in CONFIG.SYS BTW?

> I also experience an error after VirtualBox.exe is running, when I start the VM. If I wait the 15 seconds suggested by Mark (I believe) the VM will start.

Why do you need this? You have -1018 error too?

> Does networking work in Vbox?

It works, via NAT, and via OS2TAP driver (but the latter is broken). NAT should work ok. Yes, probably, you have some network configuration error in WinXP. If your OS/2 network connection works, then it should woork in VBox too.

261
Applications / Re: VirtualBox problems
« on: April 12, 2018, 12:28:59 am »
Maybe, it depends on the JFS version, I'm not sure, but my version had too big cache by default. (I have no Arcaos, eCS 2.2 only, but I needed to set explicit JFS cache to have my system stable) After setting an explicit value, it become ok.

> At that I remember trying to increase the cache size to 128 MBs and the system becoming very unstable.

Are you sure you did all at the same time: Set explicit JFS cache, Set VAL and THREADS/PROCESSES as I said? Also, you should look into Theseus, to be sure: System->Kernel Information->Kernel Memory Usage. I have VAL=3072, so it should be 1 GB of kernel memory. In the above report I see Allocated: 872 MB, Committed: 295 MB ("Totals" at the bottom). So, 128 MB must be free. You can check it yourself -- some space should be free.

I have 3500 MB of physical memory available to OS/2 on this system. Some systems have less than 3072 physical memory available to OS/2. If so, then it maybe less kernel memory available. So, you'll need to tailor it to your system further.

262
Applications / Re: VirtualBox v5.0.51 Testing
« on: April 10, 2018, 11:43:44 pm »
2Mark Szkolnicki: Changing JFS cache to a lesser value can only cause it to perform slower. So, I doubt that if will have catastrophic consequences ;) It if set too large, it can remain insufficient memory for other needs. But if set lo a lower value, then no harm. And with VAL=3072, and JFS cache not set explicitly, it will be, most probably, still insufficient memory for kernel and drivers, so, it may be some weird system behaviour. So, I don't expect anything better with VAL/PROCESSES/THREADS set, but JFS cache not set. Error -1018 will not go away until you use IBM's kernel, and I doubt that waiting 15 seconds would be of any help. Maybe, pressing "Cancel" or "Continue" will help temporarily, but I'm not sure, as I had no such error on my machines.

Available shared memory is memory in lower shared memory arena. It does not change when you change VAL. VAL can add more high shared or private memory to userland address space if VAL is moved up to the maximum of 3072 MB, so that, VBox can allocate more high private memory for VM's.

263
Applications / Re: VirtualBox v5.0.51 Testing
« on: April 10, 2018, 12:37:57 am »
2Mark Szkolnicki:

> Regarding physical media support - I have not ran into any difficulties at all at this testing phase - all files and installations proceeded without any errors, and I can only report that all physical optical media has booted correctly and flawlessly, as of now, for the optical drives tested.

Yes, reading the attached media from within the VM is good, once booted. But if you try to boot from a physical floppy, or physical CD, you can encounter some problems (possibly, VM will crash).

> The VirtualBox package I downloaded was vbox-os2-i386-v5.0.51.zip dated 22/3/2018 following the link Martin created. Might there be another package floating about or that you are using? Or perhaps a global setting that enables or disables this feature, causing what I have observed?

No newer version was released. so far. I'm not aware of any global version, which disables it for all VM's. Just disable it for every VM created for now. Possibly, I'll try to disable it globally if required features are missing on your kernel, in the future.

Quote
Regarding your P.S. question - The VBox topic has generated a lot of interest, based on the number of views.

I started the new thread as one of the big problems we always have in the OS/2 / eCS / ArcaOS community is getting people to test new items or contribute to the conversation. The focus should not only be on the problems which developers and the highly technical or experienced users discuss, but also how to use and test the software properly, to allow other useful or important feedback to be obtained.. Perhaps, with a little encouragement, especially with a lot of the new users coming on-board with the release of ArcaOS, We can get a lot more testing done, which would certainly help you and others advance development of what I personally consider an important piece of software that could help my business, as VirtualPC definitely needs a more robust and flexible follow-on for our community.

"Problems" in the title always get people to shy away from testing or even asking questions, thinking their questions may be dumb, untechnical or not pertinent. We will also seriously need some documentation for VirtualBox, as the solutions we are coming up with need to be captured to allow people to consistently install and use it successfully, especially if you or others coming up with new drops in future.

Hm, I don't think the word "Problems" will get people shy away from the thread. But indeed "Testing" is better name than "Problems", I agree.

PS: BTW, in the end of older thread I added some CONFIG.SYS options to try for you. You said that you had to close Seamonkey, to run VBox successfully. I'd suspect that if you try these options, you will be able to run both at the same time (if you'll make everything correctly)

Quote
Hi Valery!

I'm not sure how many in the community are using OS/4 consistently or not at the moment.

A big number, in fact.

Quote
I myself do not use it, and I suspect the majority of users here are using a version of the IBM kernel, especially with the ArcaOS release, so you may find many testing reports varying significantly from what you may see on OS/4.

I actually would like to test it (OS/4) at some point, but I suspect if something varies significantly from what you have experienced, testing on OS/4, the difference between kernels and structures is probably the culprit. (including the slowdowns I mentioned previously with AMD-V enabled on the IBM kernel or physical media being unstable).

That why wide testing could be very important, as this VirtualBox development continues.

Indeed, not so big difference. OS/4 kernels contain some enhancements. For example, consistent hi-res timer support, based on HPET timer; Special KEE's for consistently getting curent CPU number (no -1018 error), support for CPU rendez-vous (which allows to use HW virtualization, etc., etc.)

Quote
As I said in my previous post I think we in the community need some formalized "prototype" and "final" documentation to distribute out with the OS/2 VirtualBox even if it comes in two flavours - OS/2 / eCS / ArcaOS and another for OS/4, as YMMV.

I don't think that we need two separate documentation files. Only some special notes. I think, README.OS2 file is sufficient. It did not updated since version 5.0.6. It should be enhanced more, of course, and contain OS/2-specific notes. Otherwise, documentation from Sun/Oracle applies.

264
Applications / Re: VirtualBox v5.0.51 Testing
« on: April 09, 2018, 11:17:29 pm »
2Mark Szkolnicki:

Ah? Yes, just tried on an IBM kernel: It enables AMD-V/VT-x by default. Not sure why, but ok, just disable it for now.

265
Applications / Re: VirtualBox v5.0.51 Testing
« on: April 09, 2018, 10:38:43 pm »
> "PIIX3 cannot attach drive to Secondary Master VER_IO_NOT_READY".

Did you removed the CD from your CD/DVD drive? If so, insert it back, or detach from "Media" menu. Note that Physical media support is unstable, yet. For example, you can easily fail to boot from physical media.

> Also. as a test, created a VDI based Windows 2000 VM as I happened to have one lying around. The virtualization was extremely slow (took about 3 hours to install the OS, while I was doing other things. Again don't know if that is typical of installation on VBox or not, but worth noting.

You need to use HW virtualization, to run heavy-weight OS-es (but it currently does not work on IBM kernels, so OS/4 kernels only for now)

> It seems "enable VT-x / AMD-V" are enabled by default in the settings. I was not sure about that previously in my posts to Valery in the thread "VirtualBox Problems" as the original VM I had created (Windows XP) was done under Windows 7 and Windows VirtualBox v5.0.22, so I wasn't sure if I had fiddled with the settings at some point.

No, I just created the new VM from scratch, VT-x/AMD-V is disabled. In any case, disable it if it enabled, for some weird reason (unless you have one of the latest OS/4 kernels).

win2000 may run ok without VT-x enabled, but win7 or ubuntu are slow like hell, and work conveniently only with VT-x enabled

PS: Why did you created the second topic? Is not the previous one good enough?

266
Applications / Re: VirtualBox problems
« on: April 09, 2018, 09:54:17 pm »
2Dave Yeo:

> You can use http://www.gus.biysk.ru/os4/tools/highmem.exe to set (some) DLLs to load high and save on low memory. Works mostly well with the newest kernel included with ArcaOS. With SeaMonkey closed, try running himem -b *dll in the SM program directory and retest your shared memory use. Sometimes, especially with older kernels, your better to use himem -c to only load the code high. Run himem -? to see all options including how to unmark. Experiment, don't load system DLLs high and if things start crashing, unmark.

Yes, and I have a highmem.cmd script included in the VBox distribution, which starts highmem.exe whith required parameters, to load some segments into high memory. You can easily cause your system to hang, or some API's to fail if you try to load some DLL's both code and data segments to high memory, so, different DLL require different highmem.exe options.

For FireFox/Seamonkey it is sufficient to load only code segments of XUL.DLL high. No other DLL's are required to mark for SM/FF to be working ok.

Also, there is extpack.cmd which is useful to run if you need the VNC extpack to be working. OS/2 uses 8.3 DLL names, so, some DLL names should be shortened, that is what extpack.cmd is doing with VNC extpack DLL, and one more extpack, called DTrace.

267
Applications / Re: VirtualBox problems
« on: April 09, 2018, 09:33:27 pm »
2Mark Szkolnicki:

> First in regard to one of your previous questions the 1018 error could be "-1018" (See attached JPG). When I saw a "dash" after "Unknown error", I normally assume in english it was a pause between two phrases, not a "minus" mark.

So yes, it is -1018, which means VERR_INVALID_CPU_ID, that is what I suggested. Probably, it will be some workaround for that in the future, but now only this way, until your LAPIC ID's are consecutive.

> VIRTUALADDRESSLIMIT=2560

It's too little! I have VIRTUALADDRESSLIMIT=3072 on all my three machines, and everything is working. Please try to change VIRTUALADDRESSLIMIT to 3072, and THREADS to 511, PROCESSES to 128 explicitly. Also, it is desirable to change JFS cache to some lesser explicit value, like:

VIRTUALADDRESSLIMIT=3072
THREADS=511
PROCESSES=156
ifs=d:\os2\jfs.ifs /autocheck:* /cache:128000 /lazy:8,60,5 /p1

-- Here, JFS cache is limited to 128 MB, threads is limited to 511 threads in the system, PROCESSES are limited to 156 processes max. THREADS and PROCESSES eat up some structures per each thread/process, and they consume some hundreds of MB in the system arena, which could be too much. JFS cache is limited to 20% of available RAM, which could be too much too.  (20% from 3 GB is 614 MB, which is a large % from 1 GB of kernel memory, if VIRTUALADDRESSLIMIT=3072, so JFS cache should be decreased to say, 128 MB. The exact number should be determined in Theseus). THREADS/PROCESSES/JFS cache consume memory above VIRTUALADDRESS limit (kernel memory), while VBox consumes memory below it. So, if you have too little VIRTUALADDRESSLIMIT, it can be not sufficient for VBOX. If you have VIRTUALADDRESSLIMIT near to 3072, then only 1024 MB above it is available for kernel and drivers. So, if THREADS/PROCESSES/JFS cache are not limited explicitly, you can easily run out of kernel memory.

PS: How much memory you assigned to your VM? 330 MB?

268
Applications / Re: VirtualBox problems
« on: April 03, 2018, 08:17:34 pm »
2 Mark Szkolnicki:

Quote
As mentioned last night, I was able to get a VBOX VM to open consistently after a 15 second delay after the program initially opens, or after closing the 1018 error windows. However, trying it this morning, I was having the same difficulties, even after a 15 second delay, until I realized I had Seamonkey already open in the background (Seamonkey 2.35).

After closing it, I was able to start a VM, and, keeping VBOX running, to download some Windows XP update files, and then reopen and use Seamonkey normally.

However, closing Seamonkey and restarting a second time with VBOX running and reopen Seamonkey it (Seamonkey) would not open until i shut down VBOX

So yes, these are some memory-related problems. Could you, please, show me your config.sys? Also, how many physical RAM is available to OS/2 on your machine? What does XCenter "sentinel memory monitor" show?

269
Applications / Re: VirtualBox problems
« on: April 03, 2018, 07:26:30 pm »
2Dave Yeo:

Thanks for the log! So, it works with SET LIBCX_HIGHMEM=2 for you? That's what I talked about. Leave it permanently in your config.sys then (or only in VBox environment). Unfortunately, it works fine on my machine, so I don't experience this, so I cannot play with it and see potential workarounds. VirtualBox.exe starts fine on my machine without SET LIBCX_HIGHMEM=1. Only VBoxSDL.exe crashes without it. So, I'd recommend for now to leave SET LIBCX_HIGHMEM=2 in CONFIG.SYS permanently.

Also, there can be problems with insufficient address space on your machine. How large is VIRTUALADDRESSLIMIT on your machine? Did you tried to load VBox DLL's high? There is highmem.cmd for that purpose (you need to have highmem.exe on PATH).

PS Unfortunantely, these problems are likely to happen on each memory-hungry program like VBox, Mozilla, Qt or OpenOffice. OS/2 is a 32-bit OS. Additionally, it has special address space layout with Private/Shared/High Private/High Shared arenas, limiting each other. So, you can easily run out of address space, so no wonder.

270
Applications / Re: VirtualBox problems
« on: April 03, 2018, 10:20:59 am »
@Mark Szkolnicki:

> error 1018

%define VERR_INVALID_CPU_ID    (-1018)

Are you sure it is 1018, not -1018?

This means that current CPU ID is invalid. Unfortunately, there is no reliable way to determine the current CPU ID for ring0 on IBM's kernels. Currently, LAPIC ID's are used to emulate getting the CPU ID, but they are not guaranteed to be consecutive, which is assumed. If your current CPU number gets into the "hole" between valid CPU numbers, this happens. So, you're unlucky. The only way to avoid this error is currently to use OS/4 kernel. Unlike IBM kernels, it has a consistent way of getting CPU ID's which are guaranteed to be consecutive (which is assumed in VBox). So, you can try using the OS/4 kernel (if you're experienced enough)

2Dave Yeo: Do you have SET NSPR_OS2_NO_HIRES_TIMER=1 in your config.sys? If so, could you repeat the test with it disabled (just set it to an empty line in OS/2 window before running the test)?

Pages: 1 ... 16 17 [18] 19 20 ... 25