Teensyduino 1.54 Beta #4

Status
Not open for further replies.

Paul

Administrator
Staff member
Here is a fourth beta test for Teensyduino 1.54.


Edit: Links removed. Please use 1.54-beta9
https://forum.pjrc.com/threads/67252-Teensyduino-1-54-Beta-9


Changes since Teensyduino 1.54-beta3:

Update NativeEthernet
Audio: Improve simultaneous WAV playing & add SimultaneousPlay test
SD default to 16 MHz SPI clock
Serial Monitor fix double spaced lines copied to clipboard
Serial Monitor avoid auto horizontal scroll on very long lines
Serial Monitor fix memory leak on MacOS
 
Installed over beta3 on a Win10x64 without issue.

Tested SD default to 16Mhz SPI Clock:
a. BUILTIN_SDCARD on T4.1 Listfiles.ino worked at 16Mhz no problem as well as the ILI9341_t3_Slideshow. Both tested with a 512Gb SD Card.
b, Tested a T3,2 with an external SDcard reader with a 32Gb SDcard and ran the ILI9341_t3_Slideshow. Images displayed correctly on a ILI9341 and no issue with reading SDcard

Serial Monitor fix double spaced lines copied to clipboard:
Code:
Initializing SD card...initialization done.

LOG_0816.csv		4364

OV7670.bmp		153666

System Volume Information/

	WPSettings.dat		12

	IndexerVolumeGuid		76

mtpindex.dat		0

datalog.txt		5363

LOG_0001.csv		86201

LOG_0000-afternno.csv		121978

LOG_00001.csv		65318

LOG_62620.csv		183998

perf.jpg		34910

squirrel_dither.jpg		39057

demo.jpg		40582

num3.wav		289750

num4.wav		260436

num5.wav		276426

num6.wav		260818

num1.wav		300028

num2.wav		266528

done!
Looks like still getting the extra lines on windows 10 while SerMon shows no extra lines
 
Tried the Linux Arm 64, it downloads, but if you try to run it, it gets a Bus error (core dumped) error.

Downloaded twice. Actually I think maybe three times. Tried wget, but for some reason I ran the beta 2 version? So I tried loading up desktop with firefox to download... Got the crash, In process of wget again

Still failed

Should mention: This is an RPI4-8 running Ubuntu 64 bit... The Beta 2 ran.
 
Oh, not good. I had problems with my Jetson TX2 used the build that code, crashing during the build. It seemed to be working this time, but the truth is I haven't tested it in any way. That TX2 doesn't even have a monitor or keyboard connected right now. Might be a couple weeks until I can pull it out onto a workbench and try to figure out what's going wrong.
 
Understood - I thought I would try it out again. More curious to see if by some magic the inability to have the Arduino IDE launch the Teensy app in a working state to be able to program a teensy was resolved. I think the last details about this was in the beta 2 thread. (https://forum.pjrc.com/threads/64023-Teensyduino-1-54-Beta-2?p=257054&viewfull=1#post257054)

My guess the Jetson is quicker to do the stuff, but my setup for this is as I mentioned an RPI4-8 (the newest), with firmware that allows you to boot off of USB. I then have an SSD drive connected up to boot. At one point I had the Raspian OS 64 bit Beta installed. Not what state that is in now.

Currently I installed the Ubuntu Server and then added the GUI...

@mjs513 has a setup as well.
 
Serial Monitor fix double spaced lines copied to clipboard:

I retested this again on Windows 10 and I can't reproduce the problem.

capture.png

Any chance there might have been an instance of teensy_ports.exe running while you installed? The installer is supposed to stop if Windows reports any error writing, so I can't understand how, but maybe teensy_ports.exe didn't get updated?
 
I installed Beta4 on W10/64....

Reprogrammed T3.2 with listfiles example with external reader:
Code:
Initializing SD card...initialization done.

System Volume Information/

	IndexerVolumeGuid		76

	WPSettings.dat		12

LaikH.JPG		2801

hallow.JPG		21026

DRAKE.JPG		15775

done!
screenshot.jpg

screenshot2.jpg

Update: If instead I choose the board under Serial ports and not under Teensy ports, than it Looks like I get single spacing...
Code:
Initializing SD card...initialization done.
System Volume Information/
	IndexerVolumeGuid		76
	WPSettings.dat		12
LaikH.JPG		2801
hallow.JPG		21026
DRAKE.JPG		15775
done!
 
I didn't really know where to post about it, but I made a new thread regarding an issue that I have since Teensyduino 1.52 and which is still present in this version. I tried to address that already in the betathread of 1.52 but wasn't really able to explain myself.

I tried to list all the information and knowledge that I have regarding this problem, and would love if we could solve that! Here is the link to the thread

Thanks for all the great work and help.
 
I installed Beta4 on W10/64....

Reprogrammed T3.2 with listfiles example with external reader:
Code:
Initializing SD card...initialization done.

System Volume Information/

	IndexerVolumeGuid		76

	WPSettings.dat		12

LaikH.JPG		2801

hallow.JPG		21026

DRAKE.JPG		15775

done!
View attachment 22309

View attachment 22310

Update: If instead I choose the board under Serial ports and not under Teensy ports, than it Looks like I get single spacing...
Code:
Initializing SD card...initialization done.
System Volume Information/
	IndexerVolumeGuid		76
	WPSettings.dat		12
LaikH.JPG		2801
hallow.JPG		21026
DRAKE.JPG		15775
done!

I installed clean copy of Arduino-1.8.13 along with beta4. Can confirm what @KurtE posted on the SerMon issue with Win10x64

As long as you "choosing the board under Serial ports and not under Teensy ports, than it Looks like I get single spacing...". If you choose Serial under TeensyPorts you get the double space. Tested with Builtin_sdcard on a T4.1
 
Just thought I would mention I again tried the SD Card list program (listfiles) External SPI reader works on both T3.2 and T4 (Sparkfun unit...)
Neither one will read the Samsung 32 EVO disk... Fails to initialize.

BUT if I build T4 with BUILTIN_SDCARD - it works with the EVO card...
 
@Paul and others...

And again something completely random...

Over the last few days, been hit again with people needing to increase the RX and/or TX buffers of Serial ports to get there stuff to work. Fixed it for T4... Wondering if worth it for T3.x/LC -

So far I hacked up the same type changes for Serial 1 (have not done 2-8 yet) as not sure if this is something desired or not

I pushed up the changes into the branch: https://github.com/KurtE/cores/tree/T3X_SerialX_AddMemory

Again only Serial1 done so far. Question is, is it worth completing?

Also would probably need some help testing once completed.
 
Win 10 install of TD 1.54 b4 no problem here - compiled the active sketch Audio w/SD on T_3.2 as before.

@KurtE - speaking of Random. Question: Did the rewrite of UART_Serial# result in no alloc of buffers for unused Serial#'s?

If not: With the Dynamic alloc of UART Serial# buffer working. Could it only alloc() the default buffer on Serial#.begin() [when a null *buff?] And if larger buffer wanted then do the ..._AddMemory() before .begin() to get a single buffer of desired size - and resulting in a non-null *buff when .begin() happens?
 
Just thought I would mention I again tried the SD Card list program (listfiles) External SPI reader works on both T3.2 and T4 (Sparkfun unit...)
Neither one will read the Samsung 32 EVO disk... Fails to initialize.

BUT if I build T4 with BUILTIN_SDCARD - it works with the EVO card...

Quick update here. I have been outputting some debug information in the code and output:
Code:
Initializing SD card...CardBegin Error: 12

SDFAT Cardbegin failed
initialization failed!
done!
Press any key to rerun

I believe that error code is:0X12,SD_CARD_ERROR_CMD58 - Read OCR register

Which in SdSpiCard.cpp there is code:
Code:
  // if SD2 read OCR register to check for SDHC card
  if (type() == SD_CARD_TYPE_SD2) {
    if (cardCommand(CMD58, 0)) {
      error(SD_CARD_ERROR_CMD58);
      goto fail;
    }
    if ((spiReceive() & 0XC0) == 0XC0) {

So it looks like this command is failing...
 
Win 10 install of TD 1.54 b4 no problem here - compiled the active sketch Audio w/SD on T_3.2 as before.

@KurtE - speaking of Random. Question: Did the rewrite of UART_Serial# result in no alloc of buffers for unused Serial#'s?

If not: With the Dynamic alloc of UART Serial# buffer working. Could it only alloc() the default buffer on Serial#.begin() [when a null *buff?] And if larger buffer wanted then do the ..._AddMemory() before .begin() to get a single buffer of desired size - and resulting in a non-null *buff when .begin() happens?

It should be setup now that only those Serial objects that are used are brought into the sketch. including the event objects and the like.

Earlier when we discussed the idea of adding memory to the buffers. There were those that did not want to use malloc or the like, but instead have static buffers as to make it easier to see if their sketch is running low on memory. At one point I thought about adding simple addMemory...(nullptr, 128) it would do the malloc for you... But I am pretty sure punted as the caller could have done that as well.
 
The card that was giving me occasional startup trouble was also a Samsung "EVO Select U1". It was 128GB. I sent a package to Bill Greiman with hardware for testing and I included that troublesome card, so it's no longer here.

I just looked through my collection of cards. The Samsung ones I have are "EVO Plus U1" or "EVO Select U3". I currently don't have any "EVO Select U1" cards. Will get 2 of those from Amazon soon.

Info I can find online suggests U1 means 10 Mbyte/sec and U3 means 30 Mbyte/sec. I have no idea what the difference between Plus vs Select is, other than the color. But I did find at least some conversation claiming there is a substantial difference between "EVO+" and "EVO Plus".
 
Looks like mb-me32g on the back of the card.

As for the real differences the one that I don't have says(about twice as expensive): Up to 80MB/s transfer speed
The one I have says: Ultra-fast read write speeds: Up to 95MB/S Read and 20MB/S Write Speeds;
 
Status
Not open for further replies.
Back
Top