The collected works of Y888099 - Page 2

I am building a new Xterm, made with Linux and X11R7, on a PowerPC embedded board.
Why? Because I like 'network computing' (NC) approach, and it's also an useful exercise for me.

I wish I had knowledge and budget for QNX,WxWorks, better kernels ...

Raion-Fox wrote: Much better for someone to make molds for injection molding than 3D printing.

Sure, but it's more expensive, especially for a low quantity production.

QNX is an interesting ucKernel, different from linux in many aspects, including the fact that it's not open-source and you need to complete a legal procedure before you can gain access to QNX source code. More specifically (as I have already tried to do, I have been explained that -> ) you need to contact either your QNX sales or technical support representative for the details of the procedure. There is absolutely no other way (including educational proposal) to gain access to the source code: this implies that you cannot gain access to the source code unless you are a legitimate QNX customer.

I am not one of their customer, I can't play with QNX, so here we are: someone so lucky to be in business with QNX? (automotive?)
Have you ever tried to port X11 on it? Have you ever heard of any X11-terminals based on QNX+X11? Tried it? Developed them?

Let me know

jan-jaap wrote:
X11 terminals are a thing of the nineties, when workstations were an expensive resource that had to be shared.

X11 terminals are still good when you have to leave them far from your eyes in public space, where everybody can put his/her hands on, and you want to be sure that nobody can do strange things with linux on his/her USBPendive.

I need to develop a new product. In the theory I am the hardware guy, the one who designs the PCB. In practice I am also the one who deals with software (and I am not so skilled, but hey? they say "stay in budget")

Elf wrote: Most people these days use RDP

Of course, but I can't use RDP, I need X11.

Elf wrote: Citrix thin client terminals for that sort of use case

I also need to integrate an ISO7816 SmartCard reader for authentication. The next generation of products will have encrypted smartcards (thanks God not contactless, the RFID technology goes more complex), and a picture of the owner printed on the card.

Currently I am working with AT24C128SC-cards (they are I2C EEprom chip on ISO7816 shape, directly wired without any protection) handled by TDA8029 chip. Basically it's a customized MPU on smd chip. Already designed a couple of PCBs and built a prototype, I am dealing with the software side.

Elf wrote: Linux based or custom OS

Well, I have some skills with linux, and I am putting a kernel on a powerpc-4xx prototype. I'd like to experiment other solutions, if possible. QNX sounds good, but, in order 'to stay in budget', I am afraid I am doomed 'to stay with linux'

jan-jaap wrote: special sauce

anyone has already tasted? hot pepper sauce?

So, I am building a customized rootfs for my Virtex-PPC405GP board, where I can't use glibc because it's too elephantic for the footprint I need to use.

Good news: luckily there are alternatives : { uclibc, musl }, see the comparing table, musl looks very good on the paper!

Bad news: unfortunately they both seem to be experimental ( it means bugs for sure ) and the README says "fonly for masochists"

These are extremely experimental! They are only for masochists who enjoy breaking stuff just so they can fix it. musl is extremely lean and fast, so there is a practical reason to work at it.

musl/readme.txt

so, I haven't decided yet what I see, the donut, or the hole

it works



UAT may be wrong? ?

I think when the kernel loads it has trouble passing the /dev along and therefore sees no devices.
But I still can't figure out what to do.

Any idea?

(linux 2.6.22)







edit:
oh, and also this this crap o_O

For a long time after bind mounts were introduced, the kernel associated mount options with devices rather than mount points, end of story. When you ran mount --bind, the kernel silently ignored all options since they couldn't be applied just to the bind mount.

Starting with kernel 2.6.26 (or earlier for distributions that patched the upstream kernel), bind mounts have a read-only status that is separate from the original mount. So read-only bind mounts do work. However, the support is not perfect, for example the kernel still ignores options passed to mount --bind. You can make a read-only bind mount by making a bind mount and mounting it read-only. This introduces a security problem in some scenarios (there is a small window of time during which the bind mount is writable).

Debian lenny has a patched 2.6.26 kernel that makes mount --bind -r create a read-only bind mount atomically. Ubuntu 10.04 doesn't include that patch.

The fuse filesystem bindfs generalizes the effect of mount --bind. It supports read-only bind mounts and many other permission and ownership changes. It is not fully equivalent to mount --bind, however. For example, reading from a read-only bind mount never updates the access time of a file, but might for a bindfs -p a-w fuse mount.

found two decent stage3: stage3-powerpc-g3-uclibc-2005, stage3-powerpc-g3-2017

2005: it works, no defects found, but libraries and dependencies are too old for modern stuff
2017: strange behavior, it's defined "experimental, not stable", and so it goes

Also, "genius" in IBM have changed (again) the uImage-format and now it's not clear what I need to bootstrap a modern kernel

praetor242 wrote: nekoware GCC doesn't work completely in 6.5.22, which kinda screws you if you have an Indy

So IP28 does. Both Indy and Indigo2.

vishnu wrote: His email is s c o t t at tringali dot org, as of yesterday he's still active on the nedit mailing list (to which I subscribe).

nedit also has a lot of problems on X11-pseudocolor-terminals(1)

• v5.7 crashes
• v5.6 crashes
• v5.5 crashes
• v5.4 doesn't crash but it's damn slow to start.

and why does it crash in the first place? Why doesn't show up anything useful instead?
Any dbg_out(stderr, ErrorMessage)? Never heard before? bah

None of them reports a clue.
Need to write to someone.

(1) tested on
HP Envizex
Tektronix XP200s

So, one of the more challenging (I don't want to say very annoying, but it's so) aspects of porting Linux (and U-Boot) to your new board is the recent (>=2.6.25) requirement for a "device tree blob".

DTB is also referred to as a flat device tree, device tree binary, or simply device tree.

At the beginning it was a piece of cake: easy life, everything worked out of the shell. There was an elf-loader, able to understand the entry point, where the ram begins, ends, where it should place things, the load-point. And it was so fab, since the job was done silently.

Then they wanted to improve, because life could be boring without imprecations, so they wanted to introduce the so called "new native image format", mainly used by U-Boot-new-world, still used in embedded.

The first dot uImage (now it's called U-Boot-old-world) didn't add anything, any boot code, any table code. It just wraped a compressed kernel (objcopy -o binary kernel.elf) in the uImage data structure. It was easy, like a piece of cake, and I personally loved it.

But they wanted to "improve", so the next generation started requiring a version of U-Boot that was able to pass a device tree to the kernel at boot, and it was the beginning of the end. Everything is still broken since then.

Porting a new kernel to an old board is a challenge.

[code]
# tftp $load_addr T211-RC2.uImage
# bootm
## Booting image at 00800000 ...
Image Name: experiment#211, kernel-v4.9.16-RC2
Created: 2017-05-08 9:01:17 UTC
Image Type: PowerPC Linux Kernel Image (gzip compressed)
Data Size: 2262336 Bytes = 2.2 MB
Load Address: 00500000
Entry Point: 005000a4
Verifying Checksum ... OK
Uncompressing Kernel Image ... OK
Memory

Hakimoto

Do not double post.

The only difference between this post and the other, that I can see, is that this one here is unfinished.

You do have the possibility of editing posts, so why create a second new post instead?

Locking.

This forum has a bug: if you put arrows inside a [ code ] ... [ /code ] session, bad things happen!
In first place, post gets corrupted.

Hakimoto wrote: What kind of "arrows"

Ones used on UNIX's shell to redirect the stdout/stdin, usually also used to indicate if something is greater, or less than.
It seems this forum reacts in bad way, which is odd if we consider that code /code should consider the text as *text*.

I happen to have PHP3 forum, it doesn't happen on my pages.
May be an hidden BBCode? A problem with the PHP parser?



This was the point, in my previous message, where things went wrong.

Excellent !

Doesn't it run on Irix-6.5, right? A pity for that, but well done!

Good questions ...

Tried exactly the same, here it destroyed the post

I got:

[ code ]
Memory


I am on Chrome Version 49.0.2623.112
Windows XP/32bit

japes wrote: But I cycled back to wanting an Octane2 gain.

Why? Any hidden plan ?

Krokodil wrote: There has to be a software depot. Somewhere.

Viewed from outer space , planet Earth provides a pleasing blue and white view. I want to believe there has to be brains depot. Somewhere, out of there. May be on Alpha Centauri, the closest star system to the Solar System, being 4.37 light-years

edit:
removed

josehill wrote: what are you using to post to the forum?

Chrome Version 49.0.2623.112
Windows XP/32bit

josehill wrote: What character encoding are you using?

UTF-8

Firefox v52.1.0 is also affected

[code]
Memory

jan-jaap wrote: You might as well ask for the release of IRIX source.

lol

Raion-Fox wrote: Windows sucks at localization

It's English, but this version of Windows(1) appears of having been modified when the previous owner installed Altium-R14, which is known to be invasive. Might be the same reason for Xilinx ISE, OrCAD, TinaDesign ... all the software I installed after the purchase. All of these applications required additional packages and RegBin modification. I set them English, but Altium came already installed when I bought the workstation, the previous owner is French. Perhaps it's a clue.

I need to try from a different PC.

(1) System Properties says "Windows XP, manufactured by IBM, supported by Altium Limited", never opened this panel, never noticed before. WTF?

RIP.

Raion-Fox wrote: I don't mourn MIPS, it's a dirty architecture with nasty register spill tendency, similar to SuperH.

Well, MIPS is not a dirty architecture, why dirty? it's a clean and interesting architecture, from the first R2K to the last MIPS64, and the last MIPS32R2 fixes a lot of things, making assembly more friendly.

I have been programming assembly MIPS for years (now you can do it with Microchip PIC32), what I don't really like is .. the MIPS-cache! Cache is still terrible on MIPS. Even worse than on PowerPC.

The nasty register spill tendency is natural on RISC, and as every RISC, MIPS suffers of being bloated in code density. It's the price to pay for a simple decode unit. But the internal design along the datapath is simple enough to have fun on your own HDL implementation. Can't you do it so simple with other RISC.

p.s.
never programmed SuperHitachi, but I have an old SH3 chip in my old GPS. It comes with an external flash, SOIC chip, it can be unsoldered and wired out to a ROM emulator. There is also a serial port, and 512Kbyte of RAM.

It sounds funny. Just need to find the occasion (time) to hack it

kramlq wrote: The only thing I didn't like was that if you needed to write code across many vendors MIPS processors, there was little on chip support for explicit feature detection.

Do you remember IDT-MIPS (R3K)? I remember (back to 90s) a lot of assembly code just to identify their CPUs in order to handle them properly. Cache was terrible also because of different behaviors on different vendor/id. So, it was a lot of code, full of #ifdef and switch cases

Raion-Fox wrote: The MIPS code I've dealt with is dirty, the register spill problem was worse than other RISCs

It sounds a problem specific to the compiler you used (gcc? with silly flags?)

Raion-Fox wrote: and at the end of the day, the ISA never lived up to its potential.

The MIPS's ISA makes easier to introduce your own defined COP (COP0 is reserved as Exception, COP1 is reserved as FPU) than on other RISCs.

edit: removed

Hakimoto wrote: OK, I think then we can all agree that the bug is not with the forum software...

Not sure, it doesn't happen on other forums, e.g. on EEvBlogs where I have never experimented strange behaviors when put things between [ code ] ... [ / code ]. Including bash and C code, which happens to have a lot of arrows.

btw, to cut the story short with Nekochan, I have already set up a "filter script" (based on 'sed') removing all the poisoned patterns from my edit-board. In case of need I will use the attach function to attach a file.

robespierre wrote: Much more elegant than anything Altera or Xilinx has come out with in their SoCs.

Indeed!

p.s.
different context, but on FPGa+SoC hybrid I have to say: I don't like AMBA!
(it's not CPU's ISA, it's a connection bus, mainly used with ARM, in fpga coSoC solutions)
It goes too damn complex

josehill wrote: Are you using a "code=php" tag instead of a simple "code" tag?

Simple "code" tag.

Toxiccameron wrote: I did email and ask Autodesk if they have older versions of Maya available for sgi computers and they just said that they don't have it because it's so old.

The same happens with Wolfram Mathematica. SGI-Irix is NoGo.

With binclock ngde



you can display a binary clock in your terminal !!!
(or build a matrix of LEDs, with those '1's light on)

edit:
removed

I need qty=2. Let me know

edit:
Moved to linux-kernel || linux-pci mail list.

edit:
removed

dexter1 wrote: specific questions about software please ask them here

So, the interest here is: I want to buy/sell/trade.
Everything else is a waste of time. Understood.

pentium wrote: then gave it 12000 volts

I did something similar with a couple of mini-EATs

uunix wrote: my last purchase from the US incurred a large postage fee

Customs? How much is it? Like 20-23% of the declared value?