Jump to content
w1k

write disabled?

Recommended Posts

did you update RespeQt, and then this happened?

 

You're running on windows? what version?

 

What kind of SIO2PC-USB? FTDI based?

 

What kind of handshaking?

Edited by Joey Z

Share this post


Link to post
Share on other sites

I using SIO2PC USB, i cant write to disk

 

 

Could you please switch the "Backend" to "Standard Backend" and re-check?

It could be a problem related to the "Qt Backend".

Share this post


Link to post
Share on other sites

@w1k

thanks for the pointing the error.

I can confirm that this is a problem with the "Qt Backend" - I have never tested writing to a disk before.

Share this post


Link to post
Share on other sites

quick test here shows no issues, let me try a longer write though....

 

EDIT: no issue formatting a disk here and copying a 293 sector file to it. I am running at 1X though, and I am using linux. I am using the Qt serial port backend.

Edited by Joey Z

Share this post


Link to post
Share on other sites

quick test here shows no issues, let me try a longer write though....

 

EDIT: no issue formatting a disk here and copying a 293 sector file to it. I am running at 1X though, and I am using linux. I am using the Qt serial port backend.

 

I was fooling with the Qt backend on my RPi setup last week (so Raspian Jessie Linux) and got some strange timeouts a couple times in general use. Disabling emulation and re-enabling it seemed to clear them up but I later changed over to the standard backend and had no further problems. I will say though that I did no real troubleshooting, nor did I attempt to isolate the issues.

​Gives me a good excuse to fool around tonight.

 

EDIT: Okay, interesting bug with the Qt backend ... I'm using an SIO2PC device from Ray (FTDI based), connected through my ICD P:R: Connection. Using the Qt backend I can create an instant crash by selecting "Use Non-Standard POKEY Divisor" or whatever the exact phrasing is in the dialog, and then power-cycling my system for a cold boot. However, if the system is already up and running and I check the box, I can then do read/write operations between multiple drives - even between the flash-based IDE hard disk in my SIDE2 cart and a large SDX-formatted HD image in RespeQt - with no trouble. But as soon as I do a system power cycle, RespeQt crashes on me.

 

By contrast, the Standard backend works fine all the way up and down the speed range - no crash with a cold boot using non-standard POKEY divisors.

​This is using a build I made last weekend on my RPi from Joey's Main branch.

Edited by DrVenkman

Share this post


Link to post
Share on other sites

The reason for introducing Qt Backend was that somebody reported me that his Arduino Micro Pro with the following sketch (acting as SIO2PC):

void setup()
{
   Serial.begin(19200);
   Serial1.begin(19200);
}

void loop()
{
  if (Serial1.available())
    Serial.write(Serial1.read());
  if (Serial.available())
    Serial1.write(Serial.read());
}

doesn't work with RespeQt under Windows (although works perfectly well under Linux).

He told me that PUTTY and Qt terminal (example code from Qt) show properly data coming from the ATARI.

I was curious and promised to help and ordered a Micro Pro board.

While I was waiting for delivery I implemented the Qt backend (as an experimental feature).

The real reason why RespeQt didn't work with Micro Pro under Windows, was the serial driver.

It expected the asserted DTR line and RespeQt was resetting the DTR (to low state).

A change for the DTR line helped.

However I already had the Qt backend code and decided to add it to Github.

Now it looks like it was too early.

I examined the w1k problem with a logic analyzer and realized that the Windows implementation of the QSerialPort seems to be too slow for the SIO protocol.

In this case I see no reason to keep it in RespeQt and I propose to remove it.

  • Like 1

Share this post


Link to post
Share on other sites

Just out of curiosity - did anyone duplicate my immediate crash with the Qt Backend by enabling non-standard POKEY divisors and a cold boot of the host Atari?

Share this post


Link to post
Share on other sites

so, whats next? its a bug or problem with my sio2pc?

Share this post


Link to post
Share on other sites

Binaries have been compiled and are now available at the link.

Edited by Joey Z
  • Like 1

Share this post


Link to post
Share on other sites

Did a binary for Ubuntu ever get compiled that fixes the write issue? Presently I'm just using the one that was done for Windows in VirtualBox under a Ubuntu distribution, but it would be nice to have it work natively under Linux as well.

 

- Michael

Share this post


Link to post
Share on other sites

Weirdness. I compile every release for Raspian right on my RPi boxes and haven’t ever had an issue writing to disk images.

Share this post


Link to post
Share on other sites

Weirdness. I compile every release for Raspian right on my RPi boxes and haven’t ever had an issue writing to disk images.

 

The last time I ran a compiled version on Linux was R3 I believe, and it was still cranky as could be about writing to the disk (required dropping down to SIO stock speed), but it's always been fine when reading which luckily is most of the time. However on that same computer if I run WinXP in VirtualBox with any version of RespeQt I can write at Divisor zero all day long. So getting back to my original question, is there a compiled version beyond R3 that might work better for me under Linux? Or if the latest version of the sources has been corrected for this, I guess I could compile that myself, but I just wanted to double check before doing that only to discover that it still has problems.

 

- Michael

Share this post


Link to post
Share on other sites

So getting back to my original question, is there a compiled version beyond R3 that might work better for me under Linux? Or if the latest version of the sources has been corrected for this, I guess I could compile that myself, but I just wanted to double check before doing that only to discover that it still has problems.

 

The original problem was related to the "Qt Backend" (implementation of the serial port access based on the Qt Serial Port library).

This was an experimental feature and was completely removed from RespeQt due to various issues.

So, you can compile RespeQt R4 by yourself.

Share this post


Link to post
Share on other sites

 

The original problem was related to the "Qt Backend" (implementation of the serial port access based on the Qt Serial Port library).

This was an experimental feature and was completely removed from RespeQt due to various issues.

So, you can compile RespeQt R4 by yourself.

 

Thank you :) . I'll go grab it.

 

Edit: Picked up a compiled version HERE and it works just fine now when writing at divisor zero in Ubuntu.

 

- Michael

  • Like 2

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...