If you have some CD-RW (usually bundled with your CD/DVD writer), you can use them. It should allow you to do ~1000 tests of different ROS builds per CD.DarkSnake-Kobra wrote:I'd like to test ReactOS on a spare computer, but I don't have many cd's left(rather use them for something important).
Request for testing on real hardware
Moderator: Moderator Team
Re: Request for testing on real hardware
-
- Posts: 29
- Joined: Sat Apr 10, 2010 3:24 am
Re: Request for testing on real hardware
I can give ReactOS a try again on my pentium II. Unfortunately last time the RAM was a major hindrance. Perhaps this time it'll work better. I'll post my specs later and try to start testing tomorrow.
Re: Request for testing on real hardware
What was your problem with RAM?
-
- Posts: 29
- Joined: Sat Apr 10, 2010 3:24 am
Re: Request for testing on real hardware
The machine only has 64mb. Last time I tested it I'd be fortunate if I was able to get it to start up without something crashing one out of like every four or five times. I was told by one of the developers (I forgot who exactly) that while ReactOS should theoretically work with this amount, it just isn't able to run stably with such little ram in its current state or something along the lines of that.Haos wrote:What was your problem with RAM?
-
- Posts: 19
- Joined: Sat Oct 08, 2011 10:14 pm
- Location: U.S.A
Re: Request for testing on real hardware
Don't have any of those ether. Well my dad gave me the rest of his so I can try that. Leaving town tomorrow so I'll have to check this out later. Might try and pickup so CD-RW next time I get more. Normally I buy the cheap CD-R 100 packs.fred02 wrote:If you have some CD-RW (usually bundled with your CD/DVD writer), you can use them. It should allow you to do ~1000 tests of different ROS builds per CD.
Re: Request for testing on real hardware
I would really like to test it on my reactos-test machine, but because of bug 5857 (mshtml) i can't install it.
Re: Request for testing on real hardware
There is an ACPI fixed Feature Button, an ACPI Hardware and an ACPI power button deviceaicommander wrote:
Could you check in device manager and tell me if you see "ACPI Fixed Feature Button", "Power Button", or "Sleep Button" under "System devices"? Thanks for the report on USB. I'm going to look into the problem on some real hardware when I get the chance. It's working on QEMU with emulated USB keyboard and mouse.
Re: Request for testing on real hardware
0.3.12 is the last you can try with your limited amount of system memory. See [http://www.reactos.org/bugzilla/attachment.cgi?id=6769] (bug 6526). Unfortunately that doesn't allow you to test latest trunk (hang at MSHTML, 2nd phase, if only 64MB). It also shows virtual memory usage (through pagefile) isn't properly implemented (otherwise 64MB wouldn't be an issue at all). If the build machines automatically tested at 64MB instead of 128MB+ some stuff might get caught/fixed a tad faster.clancg2003 wrote: The machine only has 64mb. Last time I tested it I'd be fortunate if I was able to get it to start up without something crashing one out of like every four or five times. I was told by one of the developers (I forgot who exactly) that while ReactOS should theoretically work with this amount, it just isn't able to run stably with such little ram in its current state or something along the lines of that.
(for ReactOS itself, not for applications).
Perhaps the LiveCD is an option, haven't tried that for ages.
Re: Request for testing on real hardware
Hm Rbuild versions of trunk might work for a 64MB situation, they don't contain MSHTML (Wine Gecko). Anyway, goodluck testing. Maybe I'll test as well, but lacking any harddisk with FAT partitions, so sticking to LiveCD likely.
Re: Request for testing on real hardware
Read comment #65 and you know that I have!Have you tried the trick from comment #63 in that bug report?
Mhhh, what I haven't tried is the debug-build on real hardware with this option - maybe tomorrow
Re: Request for testing on real hardware
Just tested rev 54190 in QEMU, showing that phase 1 can be completed with 64MB, as well as phase 2 (despite the unaligned timezone text). However phase 3 is severely broken.
Have fun with http://iso.reactos.org/bootcd/bootcd-54191-dbg.7z or newer, on real hardware. One of these days a 'good enough working trunk' might exist
Have fun with http://iso.reactos.org/bootcd/bootcd-54191-dbg.7z or newer, on real hardware. One of these days a 'good enough working trunk' might exist
Re: Request for testing on real hardware
Just tried LiveCD (dbg), it seems to use a non-ACPI HAL by default as Shutdown wasn't working (which bootcd does work with). 64MB doesn't seem to be enough, and there's tons of error messages showing in QEMU's console log.
(plus missing desktop icons).
(plus missing desktop icons).
Re: Request for testing on real hardware
well... 64 MB is intended to use with bootcd, not livecd. Obviously, livecd eats quite more RAM than the installed version
Re: Request for testing on real hardware
ah I ment tons of errors even with 128MB, for LiveCD.
Anyway, if bootCD is able to get within 64MB for all phases again, we're getting somewhere as then behaviour on real hardware can be compared to behaviour inside various emulators / virtualisation environments.
Anyway, if bootCD is able to get within 64MB for all phases again, we're getting somewhere as then behaviour on real hardware can be compared to behaviour inside various emulators / virtualisation environments.
Who is online
Users browsing this forum: Google [Bot] and 64 guests