One Wire Library Issue

[heinrich321]

Please investigate possible issue with OneWire library on ATtiny3216. Pins toggling fine, but when using with OneWire, normal example such as here there is no output on the pin. Have tried multiple pins. :)

[SpenceKonde]

Please link to the version of OneWire you are using. You see how many issues I have open, and that doesn't even include the other core that I haven't released yet and have people asking me for! I do not have time to go chasing down libraries, make it easy for me.

[heinrich321]

Hey brother sure thing. Its the newest version 2.3.5 standard one wire lib. I tried it on all of the pins, let me know if I can provide any other info. In the meantime, ill do a bit-bang :)

[SpenceKonde]

Please confirm that this works: https://github.com/SpenceKonde/OneWire
If so will submit PR with that change to Paul.

[SpenceKonde]

Can't say I was much impressed by line 18 of: https://github.com/PaulStoffregen/OneWire/blob/master/util/OneWire_direct_gpio.h

[heinrich321]

Works like a charm! Thanks for that, pretty hardcore picking that up so fast! Cheers :)

[SpenceKonde]

Actually, @heinrich321 can you test one more thing?

1. Record sketch size with the version you say works from above.
2. Pull latest changes from
3. Test with those https://github.com/SpenceKonde/OneWire/
4. Comment here reporting whether it works, and whether the sketch size decreased (and by how much)

Thanks!

[heinrich321]

Hi @SpenceKonde First Fix

Sketch uses 6275 bytes (19%) of program storage space. Maximum is 32768 bytes.
Global variables use 204 bytes (9%) of dynamic memory, leaving 1844 bytes for local variables. Maximum is 2048 bytes.

(Works)

With the latest optimisation seems like theres a syntax error:

_Arduino: 1.8.13 Hourly Build 2020/06/03 04:33 (Windows 10), Board: "ATtiny3216/1616/1606/816/806/416/406, ATtiny3216, 20 MHz, 1.8V (5 MHz or less), Disabled, Disabled, EEPROM retained, Enabled (default timer), Closer to 5v"

C:\Users\heinr\Documents\Arduino\libraries\OneWire\OneWire.cpp: In member function 'uint8_t OneWire::reset()':
C:\Users\heinr\Documents\Arduino\libraries\OneWire\OneWire.cpp:178:36: error: expected ')' before ';' token

} while ( !DIRECT_READ(reg, mask));
^
In file included from C:\Users\heinr\Documents\Arduino\libraries\OneWire\OneWire.cpp:144:0:

C:\Users\heinr\Documents\Arduino\libraries\OneWire\util/OneWire_direct_gpio.h:22:64: error: lvalue required as left operand of assignment
#define DIRECT_WRITE_LOW(base, mask) ((*(base)+1)) &= ~(mask))
C:\Users\heinr\Documents\Arduino\libraries\OneWire\OneWire.cpp:181:2: note: in expansion of macro 'DIRECT_WRITE_LOW'_

[heinrich321]

I think there is just a few bracket errors, think I fixed it:

As well as de-referencing before incrementing on line 22

#define PIN_TO_BASEREG(pin) (((digitalPinToPort(pin))<<2))
#define DIRECT_READ(base, mask) ((*((base+2)) & (mask)) ? 1 : 0)
#define DIRECT_MODE_INPUT(base, mask) ((*(base)) &= ~(mask))
#define DIRECT_MODE_OUTPUT(base, mask) ((*(base)) |= (mask))
#define DIRECT_WRITE_LOW(base, mask) ((*((base)+1)) &= ~(mask))
#define DIRECT_WRITE_HIGH(base, mask) ((*((base)+1)) |= (mask))

[heinrich321]

Optimisation:

Fix 1 Sketch uses 6275 bytes (19%)
Fix 2 Sketch uses 6227 bytes (19%)

0.7% size reduction & working

[SpenceKonde]

Nice, thanks for going the extra mile and fixing my messed up parens too!

[SpenceKonde]

I test compiled it - no good, it gets an -fpermissive warning on my core and not with stock core. It needs an explicit cast in there somewhere. Can't have -fpermissive warnings; my core permits them, but I get flak over it frequently, and at least one relevant core does NOT permit them... Many people seem to really hate -fpermissive.

[SpenceKonde]

I can't for the life of me figure out why the stock core doesn't, honetly...

[SpenceKonde]

Okay - it now builds without the warning. I didn't screw anything up right? (I don't have hardware to verify behavior against)

[heinrich321]

Sorry been a bit hectic - will test tonight :) and let you know

[SpenceKonde]

PaulStoffregen/OneWire#94 PR submitted. Closing this one as my fork can be used by megaTinyCore users who need to interact with OneWire devices.

[BigGene]

Glad I found this post and fix. Your OneWire fork works properly with the ATtiny1614 on your development board :) What's the best way to get Paul to take action on the PR from last August?
I can also report that the ss_oled.h (ver 4.2.0), RF24.h (ver 1.4.0), and DallasTemperature.h (ver 3.9.0) libraries seem to be working. Thank you!

[SpenceKonde]

Hmm, seeing as I have been unsuccessful for 9 months, I can';t say I think I'm well placed to advise ;-)

[BigGene]

It looks like the OneWire library was updated to version 2.3.6. Do you know if that includes your fix for megaTinyCore users? I should be able to check it myself, but I don't want to mess up my current working setup. Sorry for being such a newb!

[SpenceKonde]

Does not look like he has addressed the issue.

[BigGene]

Thanks. That's what I was thinking :( On Tuesday, December 14, 2021, 08:41:51 AM EST, Spence Konde (aka Dr. Azzy) ***@***.***> wrote: Does not look like he has addressed the issue. — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe. Triage notifications on the go with GitHub Mobile for iOS or Android.