Recent Posts

Topic: Aldershot / What's your preferred language?

Yes, kawai means cute, and often in a twee way, though not always. Understanding the Japanese language barely helps. Understanding the Japanese mind is a lifelong mission! ;)

Topic: General / OLED

Should that be “1-8 down”?

Ummm… Well…. Err…. <facepalm>

Is it worth making the various brightness change reason codes of OLED_Op into a single reason code with another register specifying the actual brightness?

That is “doable”, yes.

I appreciate probably only certain brightnesses are available, but even so, it might be neater, perhaps?

256 levels, apparently. Though as I think I pointed out, a lot of them look alike. For instance, toggling between 75% and 100%, can’t really tell any difference.

Is your implementation able to be adapted to multi-colour displays in the future, do you think?

No. It is pretty much hard wired to “black and white”. Though, there’s no reason a different module (“OLED4”?) couldn’t be produced which caters for the four colour types? One of the problems is that my module will only work with the 128×64,1bpp displays. The 1bpp was a design decision, and the 128×64 is because there is no mechanism for probing the display geometry.¹

I noticed some similar ones to the one you have that do two colours at least

Depends what you mean. There is a blue/yellow type. This isn’t a two colour display, it is a blue display with a strip that is yellow instead.

Like this one: http://www.sunrom.com/media/images/1258/1258b.jpg

Anyway, looks like a lot of fun! ;)

’Tis, yes! :-)

@ Raik: The technology may be similar, but yours looks like a full colour display. Nice, but not quite the same thing.

¹ If somebody wants to post me a dirt cheap 128×32 (or other different size) display that runs off an SSD1306 at 3.3V via IIC, I’ll look at adding support for it.

Topic: Aldershot / What's your preferred language?

My mind boggles. Perhaps if I understood Japanese some meaning would emerge. In the era when I was born there were things called flower fairies, which were the current expression of twee sentimentality. Perhaps Rick can tell us if that is what kawai means? The danger with such personalizations of programming languages (or of insurance companies or of toilet brushes or . . . ) is that it opens the gates for personalizations of a different kind, say with cartoon characters (Java as Marge Simpson), celebrated actresses (Erlang as Greta Garbo), and so on ad nauseam. What puzzles me in this instance is whether any rationale was used in the association of language with picture. I am fumbling in the dark here. Age thing – sorry!

Topic: General / OLED

Is this a comparable display I use for may HandPi ?

Topic: General / OLED

The API looks nice, Rick. I noticed a couple of things:

a. the home position is 1,1 (not 0,0 as in RISC OS), thus character positions are 1-16 across and 1-7 down.

Should that be “1-8 down”?

Is it worth making the various brightness change reason codes of OLED_Op into a single reason code with another register specifying the actual brightness? I appreciate probably only certain brightnesses are available, but even so, it might be neater, perhaps?

Is your implementation able to be adapted to multi-colour displays in the future, do you think? I noticed some similar ones to the one you have that do two colours at least (you’d think 3 + black really, wouldn’t you?).

Anyway, looks like a lot of fun! ;)

Topic: Bugs / Keyboard keypress repeat and keyboard stop working on Raspberry Pi

We need you there!

I know, I’m trying to figure that one out at the minute, couldn’t have come at a worse time!

Topic: Aldershot / What's your preferred language?

I’m worried now! I use C# at work and I shudder to think how that will be interpreted around here … I am a 水瓶座 though!

私は日本語が下手です :(

Topic: Bugs / Keyboard keypress repeat and keyboard stop working on Raspberry Pi

I won’t be doing anything for the forceable future. All development is on hold whilst I help a company out, unfortunately it involves lots of travel, so I’m not going to have any free time.

Hope you’re still going to make it to ROUGOL later today? We need you there!

Topic: Bugs / Keyboard keypress repeat and keyboard stop working on Raspberry Pi

For who ever looks at this: the Foundations latest USB is discussed at http://www.raspberrypi.org/forums/viewtopic.php?f=28&t=70437

Interesting; I didn’t know they were still making significant changes to the driver. But since that new version is still in development, I think I’ll play it safe and go with the original FIQ fix, at least for now.

Topic: Code review / OS_ValidateAddress shortcomings

looking at the documention am I correct in assuming that if it returns R1=0 the range is invalid?

Correct.

You may also want to change “it does NOT issue Service_ValidateAddress”

OS_ValidateAddress will issue Service_ValidateAddress, but OS_Memory 24 won’t.

One thing I’d like to see at some point, is a means to determine is an address/range is executable (and include ROM). This would be useful for an Abort handler to determine if it should action and return or clean up.

Yeah, that’ll be coming once I get nonexecutable pages/DAs working.

Thanks for checking ADFFS.

No problem!

Topic: Code review / OS_ValidateAddress shortcomings

I’ve gone with the path of least resistance and made our new OS_ValidateAddress do the same. This means things like ADFFS should continue to work correctly (e.g. your *LOAD into screen memory seemed to work OK).

Sounds sensible. looking at the documention am I correct in assuming that if it returns R1=0 the range is invalid? You may also want to change “it does NOT issue Service_ValidateAddress”

One thing I’d like to see at some point, is a means to determine is an address/range is executable (and include ROM). This would be useful for an Abort handler to determine if it should action and return or clean up.

Thanks for checking ADFFS.

Topic: General / Calling C code from vectors

I’m not going to pursue the Nisis/Aiptek module after all. Further experimentation has shown that it is impossible, on my T-6000U at least, to tell the state of the pen’s buttons.

Topic: Aldershot / What's your preferred language?

http://i.4cdn.org/g/1398022264381.png

You appear to have been captioned as Perl :)

Topic: Bugs / Keyboard keypress repeat and keyboard stop working on Raspberry Pi

Are you still planning on looking at this, Jon?

I won’t be doing anything for the forceable future. All development is on hold whilst I help a company out, unfortunately it involves lots of travel, so I’m not going to have any free time.

Topic: General / OLED

Doing a lot of file copying/counting with CJELED running will also give the I2C bus a good workout!

Topic: Aldershot / What's your preferred language?

Pics are too big for forum, so links instead…

http://i.4cdn.org/g/1398022264381.png

http://i.4cdn.org/g/1398023617144.png

Topic: Bugs / Keyboard keypress repeat and keyboard stop working on Raspberry Pi

For who ever looks at this: the Foundations latest USB is discussed at http://www.raspberrypi.org/forums/viewtopic.php?f=28&t=70437

Topic: Bugs / Keyboard keypress repeat and keyboard stop working on Raspberry Pi

I thought we were only supposed to have bounce in the mosh pit?

Okay, okay, I’m getting my coat…

Topic: Bugs / Keyboard keypress repeat and keyboard stop working on Raspberry Pi

Pressing SELECT to act as a ‘y’ keypress causes everything to go flashing by as I clicked once and the computer obviously saw “yyyyyyyyyyyyy”.

Always labelled as “key bounce” issues. Of course you could have bounce on the mouse events.

Having said that I had key repetition and mouse event repetition issues on this machine back then (last year – New Year) but not on current releases. Bear in mind that this machine is actually a W7 laptop running RPCEmu.

I initially thought it was a mouse button problem.

Topic: General / OLED

In a nice coincidence of timing, I have only just this minute uploaded a copy of the module documentation:

http://www.heyrick.co.uk/random/OLED_!Help.txt
(simple text file)

Apart from the one SWI that I didn’t plan on doing for the first release, the module is finished. All the _kernel_swi() calls are now assembler routines, some stuff optimised, etc etc.

Can’t release it yet, though. Still waiting on an official allocation because this build recycles the MIDI SWI chunk (for want of something else to use). I ought to have the who how-to thingy on my blog in a few days¹.

I guess I could bash hardware to speed up the bus, but its really the domain of the HAL. It probably should figure it is time to move on from 100kHz in this day and age.
The modification is a simple thing to do with a RISC OS ROM and Zap – only one word needs to be changed (could probably write a little tool to do it automatically). Details are at the end of the help file, if you are interested.

¹ Rick-days, not calendar days. ;-)

Topic: Bugs / Keyboard keypress repeat and keyboard stop working on Raspberry Pi

Here’s one for you to ponder.

From time to time I get the keeeeeeeeeeyboard repeat problem. Standard generic keyboard.

But – I have written a demo of my OLED driver that uses OS_Confirm to wait for me (because it gives an easy choice of mouse click or keypress). And what happened on a couple of occasions? Yup – you guessed it. Pressing SELECT to act as a ‘y’ keypress causes everything to go flashing by as I clicked once and the computer obviously saw “yyyyyyyyyyyyy”.

You can now wipe up that coffee you just spat across the desk…
…cos…
…yeah…
…I thought the same thing.

[note: (older) self-built ROM with sources from mid-January (’cos unpacking them TakesAges so I only do it every-so-often)]

Topic: Bugs / Keyboard keypress repeat and keyboard stop working on Raspberry Pi

Are you still planning on looking at this, Jon? I’ve got a free block of time coming up shortly so should be able to look at it myself.

Topic: General / OLED

Very nice, Rick. Can the IIC bus rate be jacked up by poking some registers at runtime instead of compiling a new ROM?

Topic: Community Support / Boot problem

Therefore, it might have spun up and been ‘seen’ by the machine, but attempts to access it will fail because… it is… not ready. ;-)

Which is why I thought maybe there was a slightly iffy contact round the drive. A pullup/down on the ready sense line. Gary, that’s the links on the drive and the termination for the SCSI cable (normally at the opposite end of the cable to the host card) that need a lift and reseat.

Topic: Code review / OS_ValidateAddress shortcomings

My OS_Memory 24 implementation is now in CVS. After looking over ROL’s docs a second time it looks like their OS_ValidateAddress call still issues Service_ValdiateAddress, so for now I’ve gone with the path of least resistance and made our new OS_ValidateAddress do the same. This means things like ADFFS should continue to work correctly (e.g. your *LOAD into screen memory seemed to work OK).

FYI, our Service_ValidateAddress will report the following as being valid:

• Any areas which have any of bits 0-3 set on return from OS_Memory 24 (i.e. completely user/svc readable or writable), and which aren’t (partially) physically mapped. This will cover all RAM+ROM areas, including the bits OS_ValidateAddress previously didn’t report on.
• Any areas which are completely physically mapped and are completely readable+writable in user mode – i.e. screen memory on Iyonix/Pi. In the future I might change it to something a little different depending on how Service_ValidateAddress, etc. evolves. E.g. on RISC OS Select it looks like the job of reporting screen memory as valid was farmed out to the LegacyScreen module.
• Anything reported as valid by Service_ValidateAddress

And the OS_Memory 24 implementation only checks the page tables for the following situations:

• Sparse dynamic areas – L2PT checked to see which pages are mapped in
• L2PT itself – RISC OS only allocates pages to L2PT as necessary (a full L2PT would use 4MB of RAM), so L2PT is checked here to see which L2PT pages are mapped in
• IO memory (OS_Memory 13) – L1PT is checked to determine the access flags
• Physical access region (OS_Memory 14, 15) – L1PT checked to see if it’s mapped in or not

For everything else, OS_Memory 24 relies on the dynamic area list and a bunch of fixed address ranges. It also assumes that there aren’t any dynamic areas (apart from the system heap) which were created above the IO memory region (this would only be possible with code which manually specifies an address for the DA). Dynamic areas created within application space should work fine however.

After checking the OS_Memory 24 results against the page tables (and fixing a couple of other kernel bugs in the process!) it looks like it’s correctly reporting all memory regions, apart from anything that was mapped in manually via direct page table manipulation/OS_SetMemMapEntries.

As far as my original goal of making the kernel less likely to get stuck in infinite abort loops, I’ve now updated the kernel to check the error handler + buffer addresses and the exception register dump buffer using OS_Memory 24. The recovery isn’t always very graceful, but it should be better than getting a hard crash when something goes wrong.