Degrading qemu performance in DooM Part II

Posted on by
4

ticks for demo (fewer is better)

VersionASM FixedDiv/FixedMulC
0.9.0259268
0.10.5299300
0.11.0316317
0.12.0289281
0.12.5290282
0.14.0282274
0.14.1280269
20180430192195
20190218187187

So after the last round, I went ahead and dug out my crap version, where I had just recently found a nice abs() fix for a FixedDiv issue that the old iD code suffers from, and re-built a version of DooM that both used the assembly fixed division, and another with the C version. To compile I used my old GCC 2.7.2.3 to build with the flags:

-m486 -msoft-float -ffast-math -O2 -fforce-addr -fomit-frame-pointer

So here we go using the versions of Qemu that I can build quickly with GCC 3.4.5 MinGW, along with the last two pre-built Win64 builds.

It’s kind of interesting just how close the performance is between the two versions.

Naturally the real test is to run it on actual hardware, and to try a few versions of Watcom C.

Maybe the real takeaway is that Qemu runs GCC built code better…?

Adding DOSBox’s MPU401 to Qemu 0.90

Posted on by
Reply

I thought this may be something cool, if not kind of pointless. Anyways the MPU401 UART can be run like a traditional serial port with an IRQ, in intelligent mode, or just as a ‘dumb’ device you can just bit bang to talk to MIDI devices. So while playing with DOSBox I thought it’d be fun to take it’s emulation and plug it into Qemu.

And this is the end result.

It’s far from perfect, when it works it does tend to work well, although it fails to work with things like Return to Zork, but it does work with DMX’s sound code in DooM and the MPU401 driver for Windows 3.1

While doing this I was originally struggling with mapping the IO ports. Qemu has some functions to map in the memory model to assign a function that will trap read/write space. In this case base is 0x330 the base of the MPU401 device.

register_ioport_write(base, 8, 1, mpu401_ioport_write, s); register_ioport_read(base, 8, 1, mpu401_ioport_read, s);

I was thinking that the port 0x331 needed to be mapped in the same way, but it turns out after looking through more of the source, it’s actually a word aligned access. So in that case you can use a switch to see which port is actually being accessed.

static uint32_t mpu401_ioport_read(void *opaque, uint32_t addr) { switch(addr&0xf) { case 0: return(MPU401_ReadData(addr,0)); break; case 1: return(MPU401_ReadStatus(addr,0)); break; default: return(0xff); break; } return(0xff); }

Pretty simple, right?

And from there it’s a matter of mapping the DOSBox MPU code, along with the Windows interface code. Since I’m not using intelligent or IRQ mode, I just amputated the code where applicable.

If anyone wants to look at what I did to merge into anything else (and probably do a better job!) it’s on sourceforge as mpu401.c.

Otherwise the binary is available on sourceforge:

Download Qemu090b

Proot, defeating security by ignoring it in user-space

Posted on by
1

From the proot project site :

PRoot is a user-space implementation of chroot, mount –bind, and binfmt_misc.

Android among other Linux systems creates a very restricted user mode where the end user is denied the root user privileges. This is annoying as to ‘root’ a phone can be incredibly complicated l, and beyond a normal user that wants to use their phone for more than some kind of cat video social machine (don’t tell me the incredible popularity of cat videos isn’t toxoplasmosis!).

Many new phone SOC’s are supporting external HDMI, and USB host capabilities allowing you to dock your phone and use it with a keyboard and mouse.

Well thanks to the app, aptly named UserLand, running a light weight Linux distro is just a few screen presses away!

What is cool is that by emulating a scant few system calls it makes the deployment quick and seemingly trivial. And a lot more lightweight compared to docker, User Mode Linux, or Qemu (in full system emulation). But it can invoke qemu to run foreign architecture binaries to give Intel users an Arm UserLand.

Yes, via VNC you can run X11! And yes on my phone it shows all 8 Cores.

Although I’ve been using qemu, UML and other strategies to sidestep restrictive environments, proot proves itself as an exciting new tool!

End of an era, Linux to Deprecate a.out support

Posted on by
5

From the post on kernel.org:

Linux supports ELF binaries for ~25 years now. a.out coredumping has bitrotten quite significantly and would need some fixing to get it into shape again but considering how even the toolchains cannot create a.out executables in its default configuration, let’s deprecate a.out support and remove it a couple of releases later, instead.

I can’t say I’m all that surprised, maintaining backwards compatibility has not really been a Linux thing, as most people are incapable of doing any troubleshooting in the myriad of hundreds to thousands of independent packages, and instead find it far easier to switch to a different distro entirely.

At the same time, the vast majority of Linux packages are available in source code, so re-building things as ELF most likely has happened in the last 25 years.

During the great ELF migration, it was a gigantic PITA having basically 2 copies of all the libraries as things were converted over, and a.out stuff quickly evaporated. For me, the beauty of a.out was for later kernels to be able to mount and run older stuff. But as we are in the era of both ‘cheap’ user mode kernels along with virtualization will the old executable format truly be missed?

Linux has survived the removal of native support for the 80386, and even the detection logic for the NexGen processors (yes they were real, and yes they did ship), so no doubt this further amputation won’t matter to the vast majority.

I have to wonder how long until the i386 32bit target is removed? Distros like Debian have long since removed support for 80486/80586 classed processors to bring the minimum up to requiring SSE-2 based instructions, and I can’t imagine anyone who is running a 32bit OS for their main OS in this day and era.

diff --git a/arch/x86/um/Kconfig b/arch/x86/um/Kconfig index f518b47..494eeb5 100644 --- a/arch/x86/um/Kconfig +++ b/arch/x86/um/Kconfig @@ -16,7 +16,6 @@ config 64BIT config X86_32 def_bool !64BIT - select HAVE_AOUT select ARCH_WANT_IPC_PARSE_VERSION select MODULES_USE_ELF_REL select CLONE_BACKWARDS

Farewell a.out

Update: from further along in the changelog

And Alan Cox points out that the a.out binary loader _could_ be done in user space if somebody wants to, but we might keep just the loader in the kernel if somebody really wants it, since the loader isn’t that big and has no really odd special cases like the core dumping does.

VC6 Ultimate updates

Posted on by
Reply

I’ve been trading emails with various people from the project after I had made my post, and helping them integrate more of Visual Studio 2003 into the project and working through a few issues to bring far better compatibility to VS 2003.

And the best part is being able to build projects in parallel!

10.2 seconds in parallel!

I haven’t ordered new processors, so the 2.1Ghz parts are… lacking. However being able to use all available cores makes building DOSBox pretty fast.

Restricting the build to a single process takes 1:13 while the full parallel build on this machine takes a mere 10 seconds!

So absolutely check out VC6Ultimate!

Visual Studio .net Enterprise Architect

I uh, also saw this on archive.org, which may help people looking for this stuff from the future as old tools get harder and harder to find. Especially bigger editions like the Enterprise Architect version.

Linux hits version 5.0!

Posted on by
1

YES OMG I can’t take it anymore.

So many emails/blogs/videos/messages/tweets/whatevers!

I get it, ok ok, Linus thought 4.21 seemed silly so went straight to 5.0. Contrary to the numbering it’s not a significant divergence.

There.

I said it.

Am I the only one being hit with 50,000 things about Linux 5.0?!

Degrading qemu performance in DooM

Posted on by
18

Quick table… its late.

I’m using MS-DOS 5 & this benchmark suite loaded into a VMDK, and ran a few tests to check performance numbers.

version2 3d bench3 chris bencha doom ticksc quake demo
0.9.0182257.426346.4
0.10.5204247.649141.7
0.12.5305.1296.747546.8
0.13.0305.3284.351145.7
0.14.1289.6285.350945.2
0.15.0213.4176.952015.3
1197.9152.154615.5
2.4.0.1392.1340.674324.2
3.1.50405462.3149733.3

I snagged 3.1.50 from https://qemu.weilnetz.de/w64/

better performance than v2, sure, but for interactive stuff.. not so much.

So what is really going on here? Why is 0.90 so much faster when it comes to doom, and how is it possible that it’s the slowest in raw CPU performance. And fastest at IO? It appears that the crux of the issue is simply how it handles its IO, heavily favoring device performance VS CPU.

I’ll have to follow up with more builds and reading release notes to see what changed between releases. And what was it exactly that broke between gcc 3 and 4, and why the rip had to be.

I still like 0.90, if anything for it’s ability to run NeXTSTEP and NetWare.

So I was building a Windows 2019 server

Posted on by
8

As I got a ‘totally legit’ serial code in my box of cereal.

After the install I thought it’d be fun to install the Linux Subsystem.

While following the powershell instructions here, I thought the list of quick links of distros to download was interesting:

That’s right ARM Linux userland! I still have high hopes for Windows on ARM (I have 2 Windows RT devices now!!) although I’m not holding my breath.

Maybe there will be some ARM boards that are suitable for the desktop that aren’t over 1k USD.. That’d be nice.

Interesting trivia is that the Linux Subsystem started it’s life on ARM as a way to run Android binaries on Windows Phone. And true to everything Microsoft does, it got to the point where it could start to run things (albeit poorly) and was summarily killed. Although it’s found life despite the original false start as a general ‘text mode’ subsystem for Windows.

However running Linux binaries on Windows currently just shows that NTOS isn’t as efficient as the Linux kernel when it comes to emulating the Linux ABI. Although this was the original ‘dream’ of the microkernel, and a POSIX subsystem for NT was always part of the original design, although it really was more of a checkbox for GSA contracts, and outside of being able to use pax & vi it really was handicapped by not having BSD extensions, and especially by not having any access to the TCP/IP stack.

EDIT*

I should add these notes from the future past for the future me, when messing around with Windows Server 2019 build 1809 when they finally brought the Linux Subsystem into the fold. Unpacking the distribution and running the ‘setup’ sets it up DIRECTLY into that directory. So put it where you want it.

When you mess that up, you have to use the wslconfig program!

Of note is:

wslconfig /list /all
Lists all distributions, including ones that aren’t currently usable. They may be in the process of installing, uninstalling, or are in a broken state.

wslconfig /unregister <DistributionName>
Unregisters the distribution from WSL so it can be reinstalled or cleaned up.

This way you can now clean up your mess, and get Linux installed right.

VC6 Ultimate

Posted on by
1

I saw this the other day, VC6 Ultimate. It’s an interesting ‘update’ on the old Visual C++ 6.0 product with an improved UI, along with updated compiler toolchain taken from later versions of Visual C++. Naturally something like this is 1000000% unofficial.

Features include:

  • Portable and compatible with Win7 / Win10
    bye bye regedit, hello .hjson setting file !
    also meaning it should not mess with your current install
  • More compatible compiler
    multicore version of VC7.1 compiler (It’s fast)
    you can compile with other compilers (64bit), but not debug yet
  • Real-time highlighting and diagnostics
    based on libclang 6.0 and compatible with VisualAssistX
  • Real multicursor editing
    search, sort, number, evaluate, etc. while in multicursor mode
  • Improved UX and UI
    32bit icons, dark skin, lot of visual hints
    multi-monitor friendly
    revamped dialogs (project settings, threads, breakpoints, …)
    searchable command palette
  • It’s free (as in free beer)
    ever had to pay for a birthday present ? 😉
  • Every change has a toggle
    only take what you like, but we can not check each combination
  • It’s an internal spare time project
    don’t expect everything to work in every setup, but feel free to reach out

Included in the bundle is the following compilers:

  • clang version 3.8.0 (branches/release_38)
    Target: i686-pc-windows-msvc
  • Microsoft (R) 32-bit C/C++ Optimizing Compiler Version 13.10.6030 for 80×86
    Copyright (C) Microsoft Corporation 1984-2002. All rights reserved.
  • Microsoft (R) C/C++ Optimizing Compiler Version 14.00.40310.41 for AMD64
    Copyright (C) Microsoft Corporation. All rights reserved.

It’s an interesting project, although I tried to re-build some Visual C++ 2003 projects and it bombed out. Maybe it’s just more geared towards VC 6 as indicated.

You can download it here

https://gitlab.com/VC6Ultimate/VC6Ultimate