Emulating Linux MIPS in Perl – Part 1: ELF loader

http://schplog.schmorp.de/2015-06-08-emulating-linux-mips-in-perl-1.html

by draegtun • 11 years ago

77 13 11 years ago

schplog.schmorp.de

TickleSteve• 11 years ago

Am I reading this correctly??

Emulating a whole processor to run a shell script on Windows.... Seriously??

Strikes me that someone didn't research the options correctly or just wants an excuse to play...

joosters2• 11 years ago

What's wrong with writing an emulator just for the fun of it?

Ultimatt• 11 years ago

Playing aka learning in this profession. Some still care to work towards being able to contribute to the "options" that are in play. This is how those developers and new options are born.

seba_dos1• 11 years ago

Am I reading this correctly??

Complaining about hackers being hackers on a site called Hacker News... Seriously??

_ak• 11 years ago

http://www.catb.org/jargon/html/H/hack-value.html

andrewchambers• 11 years ago

A tiny mips emulator I wrote which can boot a linux kernel:

https://github.com/andrewchambers/cmips

joosters2• 11 years ago

This series of posts is amazingly well-written, it explains the steps involved really well IMO, thoughtfully ordering the code snippets to make understanding each step simple.

As a perl user, I'm also highly biased as it's nice to see a perl article on HN once in a while :)

userbinator• 11 years ago

The next step along this path would be to run Perl inside this VM, and then run the emulator in it...

MIPS is certainly one of the simplest instruction sets. A while ago I decided to write one too, for fun, after having done ones for a few other CPUs (Z80, 6502, 8086), and MIPS was by far the simplest, almost to the point of being boring. Getting the branch delay slots right (see http://www.pagetable.com/?p=313 for some details) was the hardest part.

sklogic• 11 years ago

If you like challenge, MIPS can be exceptionally hard and complicated if you want to implement an OoO superscalar core.

TheLoneWolfling• 11 years ago

W.r.t. branch delay slots.

An interesting idea: instead of fixing a number of branch delay slots (or none), what about splitting branches? That is - you have a branch_commit instruction, and anything between the branch and the branch_commit is treated as (a) branch delay slot(s), with the processor either filling in pipeline bubbles or stalling the branch as necessary. If you want to get fancy about it you can start doing matched branch / commit pairs (like nesting brackets), or potentially even start dealing with general matching.

Although the decreased instruction density may kill the benefits.

chriswarbo• 11 years ago

I came across the phrase "XY problem" the other day http://www.perlmonks.org/?node=XY+Problem

You want to do X, and you think Y is the best way of doing so. Instead of asking about X, you ask about Y.

If we allow "ask" to include asking oneself, then this seems like a great example :)

kazinator• 11 years ago

Ironic; I put together a MIPS based Linux distro about eight years ago, and Perl was one of the items that wouldn't cross-compile to it.

How you get Perl on MIPS is by compiling on a MIPS machine, or in an emulated chroot environment (QEMU on x86, say). Now Perl can maybe supply that environment.

l1n• 11 years ago

Speaking of MIPS, has anyone had success emulating IRIX?

Emulating Linux MIPS in Perl – Part 1: ELF loader

Related Stories