search

dioannidis / usbasp

usbasp improved firmware
GNU General Public License v2.0
73 stars 12 forks source link

Error in read after entering PE with slow clocks #22

Closed Franck78 closed 1 year ago

Franck78 commented 1 year ago

With firmware 1.11

random read somewhere !

between computer and usbasp or between usbasp and device programmed

avrdude: auto set sck period (because given equals null)
avrdude: AVR device initialized and ready to accept instructions

Reading | ################################################## | 100% 0.00s

avrdude: Device signature = 0x000307
avrdude: Expected signature for ATmega8 is 1E 93 07
         Double check chip, or use -F to override this check.

avrdude done.  Thank you.

some read: avrdude: Device signature = 0x010103 avrdude: Device signature = 0x013505 avrdude: Device signature = 0x00032f

Going back to fisch 1.05 and everything is ok.

dioannidis commented 1 year ago

Hi,

could you plz post which os, os version and avrdude version do you have ?

regards,

dioannidis commented 1 year ago

Hi,

I can't reproduced it in Windows 10 and in Debian 11 ( running in a VM ) .

regards,

Franck78 commented 1 year ago

avrdude: Version 6.3-20171130 (on the raspberry) for programming the USBASP

avrdude: Version 6.1, (openSUSE 42.3 Buildservice) for programming using the USBASP

Franck78 commented 1 year ago

that will be difficult to fix. it's a bug in reading the device I think. 'bit banging' bug.

Timing problem surely

dioannidis commented 1 year ago

Hi,

do you have the same problem if you connect the usbasp to rasp and use the 6.3 avrdude ?

If yes, I could try to reproduced it with my old raspberry pi 2 over the next few days .

EDIT: Also try to see if its stable in lower speeds using the -B parameter :

1      /* 500 Hz */
2      /*   1 kHz */
3      /*   2 kHz */
4      /*   4 kHz */
5      /*   8 kHz */
6      /*  16 kHz */
7      /*  32 kHz */
8      /*  93.75 kHz */
9      /* 187.5  kHz */
10    /* 375 kHz   */
11    /* 750 kHz   */
12    /* 1.5 MHz   */

regards,

dioannidis commented 1 year ago

Hi,

If yes, I could try to reproduced it with my old raspberry pi 2 over the next few days .

with raspberry's avrdude I cannot reproduced it either !

pi@raspberrypi:~ $ cat /etc/debian_version
10.12
pi@raspberrypi:~ $ avrdude -v

avrdude: Version 6.3-20171130
         Copyright (c) 2000-2005 Brian Dean, http://www.bdmicro.com/
         Copyright (c) 2007-2014 Joerg Wunsch

         System wide configuration file is "/etc/avrdude.conf"
         User configuration file is "/home/pi/.avrduderc"
         User configuration file does not exist or is not a regular file, skipping

avrdude: no programmer has been specified on the command line or the config file
         Specify a programmer using the -c option and try again

pi@raspberrypi:~ $ sudo lsusb -v

Bus 001 Device 004: ID 16c0:05dc Van Ooijen Technische Informatica shared ID for use with libusb
Device Descriptor:
  bLength                18
  bDescriptorType         1
  bcdUSB               2.01
  bDeviceClass            0
  bDeviceSubClass         0
  bDeviceProtocol         0
  bMaxPacketSize0         8
  idVendor           0x16c0 Van Ooijen Technische Informatica
  idProduct          0x05dc shared ID for use with libusb
  bcdDevice            1.11
  iManufacturer           1 www.fischl.de
  iProduct                2 USBasp
  iSerial                 3 0000
  bNumConfigurations      1

Testing 100 times using bash :

pi@raspberrypi:~ $ for i in {1..100}; do sudo avrdude -c usbasp -p m8; done &> usbasp_stability.dump
pi@raspberrypi:~ $ cat usbasp_stability.dump | grep Expected
pi@raspberrypi:~ $
pi@raspberrypi:~ $ cat usbasp_stability.dump | more

avrdude: AVR device initialized and ready to accept instructions

Reading | ################################################## | 100% 0.00s

avrdude: Device signature = 0x1e9307 (probably m8)

avrdude: safemode: Fuses OK (E:FF, H:C5, L:BF)

avrdude done.  Thank you.

avrdude: AVR device initialized and ready to accept instructions

Reading | ################################################## | 100% 0.00s

avrdude: Device signature = 0x1e9307 (probably m8)

avrdude: safemode: Fuses OK (E:FF, H:C5, L:BF)

avrdude done.  Thank you.

--------- 98 times -----------

regards,

dioannidis commented 1 year ago

@mcuee

If you have avrdude 6.1 in linux, can you please test to see if you can reproduce this behavior ?

Franck78 commented 1 year ago

USBASP under fisch1.05

plugged on my PI_1, I can read one time. Next time is error. I have to unplug the usb. With my laptop Opensuse, no need to unplug, it works fine.

USBASP under fw1.11

same random reading . Device id changing all the the time. But code read seems similar. And no need to unplug it...

root@rpi:~# avrdude -v -cusbasp -pm8

avrdude: Version 6.3-20171130
         Copyright (c) 2000-2005 Brian Dean, http://www.bdmicro.com/
         Copyright (c) 2007-2014 Joerg Wunsch

         System wide configuration file is "/etc/avrdude.conf"
         User configuration file is "/root/.avrduderc"
         User configuration file does not exist or is not a regular file, skipping

         Using Port                    : usb
         Using Programmer              : usbasp
         AVR Part                      : ATmega8
         Chip Erase delay              : 10000 us
         PAGEL                         : PD7
         BS2                           : PC2
         RESET disposition             : dedicated
         RETRY pulse                   : SCK
         serial program mode           : yes
         parallel program mode         : yes
         Timeout                       : 200
         StabDelay                     : 100
         CmdexeDelay                   : 25
         SyncLoops                     : 32
         ByteDelay                     : 0
         PollIndex                     : 3
         PollValue                     : 0x53                                                                                                                           
         Memory Detail                 :                                                                                                                                

                                  Block Poll               Page                       Polled                                                                            
           Memory Type Mode Delay Size  Indx Paged  Size   Size #Pages MinW  MaxW   ReadBack                                                                            
           ----------- ---- ----- ----- ---- ------ ------ ---- ------ ----- ----- ---------                                                                            
           eeprom         4    20   128    0 no        512    4      0  9000  9000 0xff 0xff
           flash         33    10    64    0 yes      8192   64    128  4500  4500 0xff 0x00
           lfuse          0     0     0    0 no          1    0      0  2000  2000 0x00 0x00
           hfuse          0     0     0    0 no          1    0      0  2000  2000 0x00 0x00
           lock           0     0     0    0 no          1    0      0  2000  2000 0x00 0x00
           calibration    0     0     0    0 no          4    0      0     0     0 0x00 0x00
           signature      0     0     0    0 no          3    0      0     0     0 0x00 0x00

         Programmer Type : usbasp
         Description     : USBasp, http://www.fischl.de/usbasp/

avrdude: auto set sck period (because given equals null)
avrdude: AVR device initialized and ready to accept instructions

Reading | ################################################## | 100% 0.00s

avrdude: Device signature = 0x013707
avrdude: Expected signature for ATmega8 is 1E 93 07
         Double check chip, or use -F to override this check.

avrdude: Device signature = 0x133732
avrdude: Device signature = 0x130307
avrdude: Device signature = 0x010301
avrdude: Device signature = 0x010304
avrdude: Device signature = 0x013704
avrdude: Device signature = 0x010104
avrdude: Device signature = 0x010107
avrdude: Device signature = 0x000101
avrdude: Device signature = 0x133501
avrdude: Device signature = 0x013504
avrdude: Device signature = 0x010304
avrdude: Device signature = 0x010301
avrdude: Device signature = 0x130304
avrdude: Device signature = 0x003734
avrdude: Device signature = 0x130301
avrdude: Device signature = 0x130132
avrdude: Device signature = 0x130107
avrdude: Device signature = 0x130301
avrdude: Device signature = 0x013701
avrdude: Device signature = 0x130307
avrdude: Device signature = 0x010107
avrdude: Device signature = 0x010107
avrdude: Device signature = 0x013734
avrdude: Device signature = 0x130301
avrdude: Device signature = 0x000332
avrdude: Device signature = 0x010134
avrdude: Device signature = 0x013704
avrdude: Device signature = 0x010304
avrdude: Device signature = 0x133734
avrdude: Device signature = 0x010101
avrdude: Device signature = 0x130301
avrdude: Device signature = 0x010132
avrdude: Device signature = 0x010307
avrdude: Device signature = 0x010101
avrdude: Device signature = 0x130101
avrdude: Device signature = 0x130304
avrdude: Device signature = 0x130332
avrdude: Device signature = 0x010101
avrdude: Device signature = 0x003532
avrdude: Device signature = 0x130104
avrdude: Device signature = 0x013501
avrdude: Device signature = 0x003734
avrdude: Device signature = 0x013701
avrdude: Device signature = 0x130307
avrdude: Device signature = 0x003704
avrdude: Device signature = 0x010301
avrdude: Device signature = 0x013501
avrdude: Device signature = 0x013532
avrdude: Device signature = 0x010307
avrdude: Device signature = 0x010334
avrdude: Device signature = 0x010304
avrdude: Device signature = 0x000307
avrdude: Device signature = 0x130104
avrdude: Device signature = 0x010134
avrdude: Device signature = 0x013503
avrdude: Device signature = 0x010101
avrdude: Device signature = 0x010307
avrdude: Device signature = 0x010307
avrdude: Device signature = 0x000334
avrdude: Device signature = 0x370104
avrdude: Device signature = 0x010107
avrdude: Device signature = 0x010301
avrdude: Device signature = 0x010301
avrdude: Device signature = 0x010134
avrdude: Device signature = 0x010101
avrdude: Device signature = 0x013532
avrdude: Device signature = 0x013531
avrdude: Device signature = 0x130304
avrdude: Device signature = 0x130132
avrdude: Device signature = 0x003701
avrdude: Device signature = 0x133707
avrdude: Device signature = 0x010303
avrdude: Device signature = 0x130303
avrdude: Device signature = 0x370304
avrdude: Device signature = 0x133704
avrdude: Device signature = 0x130107
avrdude: Device signature = 0x013701
avrdude: Device signature = 0x130301
avrdude: Device signature = 0x010304
avrdude: Device signature = 0x010304
avrdude: Device signature = 0x000301
avrdude: Device signature = 0x010301
avrdude: Device signature = 0x010334
avrdude: Device signature = 0x003501
avrdude: Device signature = 0x010301
avrdude: Device signature = 0x010101
avrdude: Device signature = 0x130304
avrdude: Device signature = 0x130101
avrdude: Device signature = 0x010301
avrdude: Device signature = 0x010301
avrdude: Device signature = 0x003703
avrdude: Device signature = 0x370301
avrdude: Device signature = 0x010301
avrdude: Device signature = 0x130107
avrdude: Device signature = 0x000307
avrdude: Device signature = 0x013501
avrdude: Device signature = 0x010332
avrdude: Device signature = 0x013501
avrdude: Device signature = 0x010101
avrdude: Device signature = 0x130307
avrdude: Device signature = 0x013504
avrdude: Device signature = 0x000301
avrdude: Device signature = 0x010301
avrdude: Device signature = 0x013734

signature, not so random, 100 times


root@rpi:~# cat /etc/debian_version
10.13

Bus 001 Device 010: ID 16c0:05dc Van Ooijen Technische Informatica shared ID for use with libusb
Device Descriptor:
  bLength                18
  bDescriptorType         1
  bcdUSB               2.01
  bDeviceClass            0 
  bDeviceSubClass         0 
  bDeviceProtocol         0 
  bMaxPacketSize0         8
  idVendor           0x16c0 Van Ooijen Technische Informatica
  idProduct          0x05dc shared ID for use with libusb
  bcdDevice            1.11
  iManufacturer           1 www.fischl.de
  iProduct                2 USBasp
  iSerial                 3 (error)
  bNumConfigurations      1

iSerial 3 (error) !!??

avrdude -v -cusbasp -pm8  -B10
avrdude: set SCK frequency to 93750 Hz
avrdude: AVR device initialized and ready to accept instructions

Reading | ################################################## | 100% 0.01s

avrdude: Device signature = 0x1e9307 (probably m8)
avrdude: safemode: lfuse reads as C4
avrdude: safemode: hfuse reads as D9

avrdude: safemode: lfuse reads as C4
avrdude: safemode: hfuse reads as D9
avrdude: safemode: Fuses OK (E:FF, H:D9, L:C4)

reading correctly with -B1 .. B7 ... -B10

B11, random read

mcuee commented 1 year ago

@Franck78

Can you try latest avrdude 7.1 release or even avrdude git main? There are many fixes in avrdude 7.1 and latest git main which may help.

For avrdude project, we will not support older releases if the latest release fixes the issue.

dioannidis commented 1 year ago

Hi,

@Franck78

Can you please try with this fw http://www.nephelae.eu/USBasp_ATMega8_12MHz_TPI_HIDUART_SNWRITE_fix_tm.zip ? ( no need to flash eeprom again )

@mcuee

I think I know what might be the problem ...

regards,

Franck78 commented 1 year ago

no noticeable change with nephelae. Still need -B10

I will try latter with avrdude 7.

dioannidis commented 1 year ago

Hi,

@Franck78, @mcuee

IMHO its not a problem of the firmware or avrdude, it's the connection between the USBasp and your board / mcu which cannot communicate at 750 kHz and above for some reason .

This fw uses automatic SCK speed negotiation and a higher default SCK clock speed. When a speed is found where the target responds, for subsequent communications with the target, the speed is doubled, since the slowest speed is only needed the first time the PE command is received after power-up.

When you used -B 10, USBasp enters programming mode at 375 kHz and then doubles it to 750 kHz for the rest of the communications . This worked well and it's the first time I got a report that this fails . Anyway lets see if I'm wrong ...

EDIT : Here is a fw version that you can try www.nephelae.eu/USBasp_ATMega8_12MHz_TPI_HIDUART_SNWRITE_no_double_speed.zip .

regards,

mcuee commented 1 year ago

@dioannidis

I believe your analysis is good. I can not reproduce the issue under Ubuntu 20.04, with either avrdude 6.1 release or 6.3 release.

mcuee@UbuntuSwift3:~/build/avr/avrdude_release/avrdude-6.1$ ./avrdude -C ./avrdude.conf -p m8

avrdude: no programmer has been specified on the command line or the config file
         Specify a programmer using the -c option and try again

mcuee@UbuntuSwift3:~/build/avr/avrdude_release/avrdude-6.1$ ./avrdude -C ./avrdude.conf -c usbasp -p m8

avrdude: AVR device initialized and ready to accept instructions

Reading | ################################################## | 100% 0.00s

avrdude: Device signature = 0x1e9307

avrdude: safemode: Fuses OK (E:FF, H:D9, L:9F)

avrdude done.  Thank you.

mcuee@UbuntuSwift3:~/build/avr/avrdude_release/avrdude-6.1$ cd ..
mcuee@UbuntuSwift3:~/build/avr/avrdude_release$ cd avrdude-6.3/
mcuee@UbuntuSwift3:~/build/avr/avrdude_release/avrdude-6.3$ ./avrdude -C ./avrdude.conf -c usbasp -p m8

avrdude: AVR device initialized and ready to accept instructions

Reading | ################################################## | 100% 0.00s

avrdude: Device signature = 0x1e9307 (probably m8)

avrdude: safemode: Fuses OK (E:FF, H:D9, L:9F)

avrdude done.  Thank you.

Same for 7.1 release and 7.0 release.

mcuee@UbuntuSwift3:~/build/avr/avrdude_release/avrdude-7.1/build_linux/src$ ./avrdude -c usbasp -p m8a

avrdude: AVR device initialized and ready to accept instructions
avrdude: device signature = 0x1e9307 (probably m8a)

avrdude done.  Thank you.

mcuee@UbuntuSwift3:~/build/avr/avrdude_release/avrdude-7.0_cmake/build_linux/src$ ./avrdude -c usbasp -p m8a

avrdude: AVR device initialized and ready to accept instructions

Reading | ################################################## | 100% 0.00s

avrdude: Device signature = 0x1e9307 (probably m8a)

avrdude done.  Thank you.
dioannidis commented 1 year ago

Hi,

@Franck78

Can you test this fw www.nephelae.eu/USBasp_ATMega8_12MHz_TPI_HIDUART_SNWRITE_2hid_fix_2.zip ?

( maybe similar with #23 )

regards,

Franck78 commented 1 year ago

Working now with 2hid_fix2 and no '-B10'

another bug/request for that ?

mai 07 11:34:18 : hid-generic 0003:16C0:05DC.0008: hiddev96,hidraw1: USB HID v1.01 Device [www.fischl.de USBasp] on usb-0000:00:1d.0-1.5/input1
mai 07 11:34:18 : hid-generic 0003:16C0:05DC.0009: hiddev97,hidraw2: USB HID v1.01 Device [www.fischl.de USBasp] on usb-0000:00:1d.0-1.5/input2

I was secretly expecting a serial device as 'ttyACM0' ;)

dioannidis commented 1 year ago

Hi,

@Franck78, @mcuee

Working now with 2hid_fix2 and no '-B10'

Great ! I'll update the 1.11 pre release with the new binaries .

Thank you very much for your feedback and for testing !

another bug/request for that ? I was secretly expecting a serial device as 'ttyACM0' ;)

Unfortunately, the specs for USB 1.1 for low speed devices doesn't support bulk transfers which the USB CDC needs and besides that you'll need to modify your hardware ( USBasp ) to indicate that is a full speed USB device .

You can use the usbasphiduart utility or you can implement the UART HID protocol your self .

regards,

Franck78 commented 1 year ago

moving the 'R3' from 'd-' to 'd+' is not the problem on the cheap chineese usbasp. The question is:

Can the USB stack inside the atmega8 implement the full speed when hardware says 'full speed' !

dioannidis commented 1 year ago

Hi,

Can the USB stack inside the atmega8 implement the full speed when hardware says 'full speed' !

The V-USB driver which USBasp is based on doesn't support full speed .

And there are other issues i.e. Low Speed is 1.5 MHz but Full Speed is 12 MHz . The atmega8 max clock is 16 MHz and ( if you a have a USBisp ) the atmega88 max clock is 20MHz . Even if you change the crystals, those speeds are not enough for sampling the usb signals AND decode the USB packets .

regards,

mcuee commented 1 year ago

Can the USB stack inside the atmega8 implement the full speed when hardware says 'full speed' !

As mentioned by @dioannidis, V-USB is based on FW and cannot support full-speed USB.

If you want to have proper USB to Serial support inside FW (using USB CDC-ACM), then you need to use AVR chips with native USB support, like ATmega32U4 or things like that. For example, Arduino is using ATmeag16U2 for the USB-DFU+USB-Serial in the Arduino Uno/Mega. https://github.com/arduino/ArduinoCore-avr/tree/master/firmwares/atmegaxxu2

Franck78 commented 1 year ago

yep that was what I was thinking, with 12MHz crystal, they better be virtuose assembler writers.

All my 'confusion' is from #21 first comment: why did it appear as 'full-speed' when hardware says low-speed. Anyway, fixed.

Franck78 commented 1 year ago

FYI:

I took a supposed brand new sealed ATMEGA8 for programming It needed a -B7 to read correctly. Without that, same bad reading.

With -B7 I got the fuse in it

ller2:~> avrdude -v   -cusbasp -pm8 -U lfuse:w:0xc4:m -B7

avrdude: Version 6.1, (openSUSE Buildservice)
         Copyright (c) 2000-2005 Brian Dean, http://www.bdmicro.com/
         Copyright (c) 2007-2014 Joerg Wunsch

         System wide configuration file is "/etc/avrdude.conf"
         User configuration file is "/home/fbourdonnec/.avrduderc"
         User configuration file does not exist or is not a regular file, skipping

         Using Port                    : usb
         Using Programmer              : usbasp
         Setting bit clk period        : 7.0
         AVR Part                      : ATmega8
         Chip Erase delay              : 10000 us
         PAGEL                         : PD7
         BS2                           : PC2
         RESET disposition             : dedicated
         RETRY pulse                   : SCK
         serial program mode           : yes
         parallel program mode         : yes
         Timeout                       : 200
         StabDelay                     : 100
         CmdexeDelay                   : 25
         SyncLoops                     : 32
         ByteDelay                     : 0
         PollIndex                     : 3
         PollValue                     : 0x53
         Memory Detail                 :

                                  Block Poll               Page                       Polled
           Memory Type Mode Delay Size  Indx Paged  Size   Size #Pages MinW  MaxW   ReadBack
           ----------- ---- ----- ----- ---- ------ ------ ---- ------ ----- ----- ---------
           eeprom         4    20   128    0 no        512    4      0  9000  9000 0xff 0xff
           flash         33    10    64    0 yes      8192   64    128  4500  4500 0xff 0x00
           lfuse          0     0     0    0 no          1    0      0  2000  2000 0x00 0x00
           hfuse          0     0     0    0 no          1    0      0  2000  2000 0x00 0x00
           lock           0     0     0    0 no          1    0      0  2000  2000 0x00 0x00
           calibration    0     0     0    0 no          4    0      0     0     0 0x00 0x00
           signature      0     0     0    0 no          3    0      0     0     0 0x00 0x00

         Programmer Type : usbasp
         Description     : USBasp, http://www.fischl.de/usbasp/

avrdude: set SCK frequency to 93750 Hz
avrdude: AVR device initialized and ready to accept instructions

Reading | ################################################## | 100% 0.00s

avrdude: Device signature = 0x1e9307
avrdude: safemode: lfuse reads as E1
avrdude: safemode: hfuse reads as D9
avrdude: reading input file "0xc1"
avrdude: writing lfuse (1 bytes):

Writing | ################################################## | 100% 0.01s

avrdude: 1 bytes of lfuse written
avrdude: verifying lfuse memory against 0xc4:
avrdude: load data lfuse data from input file 0xc4:
avrdude: input file 0xc4 contains 1 bytes
avrdude: reading on-chip lfuse data:

Reading | ################################################## | 100% 0.00s

avrdude: verifying ...
avrdude: 1 bytes of lfuse verified

avrdude: safemode: lfuse reads as C4
avrdude: safemode: hfuse reads as D9
avrdude: safemode: Fuses OK (E:FF, H:D9, L:C4)

avrdude done.  Thank you.

C4 is the Lfuse that is in the chip that originaly prompted this report.

and it works

avrdude: auto set sck period (because given equals null)
avrdude: AVR device initialized and ready to accept instructions

Reading | ################################################## | 100% 0.00s

avrdude: Device signature = 0x1e9307
avrdude: safemode: lfuse reads as C4
avrdude: safemode: hfuse reads as D9

avrdude: safemode: lfuse reads as C4
avrdude: safemode: hfuse reads as D9
avrdude: safemode: Fuses OK (E:FF, H:D9, L:C4)

avrdude done.  Thank you.
dioannidis commented 1 year ago

Hi,

@mcuee

avrdude: Device signature = 0x1e9307 avrdude: safemode: lfuse reads as E1 avrdude: safemode: hfuse reads as D9 avrdude: reading input file "0xc1"

tested an atmega8a with the above fuses and I can reproduce the issue .

regards,

dioannidis commented 1 year ago

Hi,

@Franck78 , @mcuee

as one can use the -B option and don't currently have time I'll take a look at this next week

regards,

dioannidis commented 1 year ago

@Franck78

When you have time can you test this fix https://github.com/dioannidis/usbasp/releases/download/v1.11/Firmware_ATMega8_12MHz_TPI_HIDUART_SNWRITE_FIX_LOW_SPEED.zip ?

Tested with an atmega8 with hfuse 0xD9 and lfuse 0xE1

regards,

dioannidis commented 1 year ago

@Franck78

The v1.11 is released

regards