Issue #973: Differentiate multiple USB programmers of the same VID/PI… #1588

MikeRaffa · 2023-12-04T06:42:28Z

…D (libusb or hidapi or libftdi)

Added code for USBasp to check for bus:device match using the -P option. Syntax is -P usb:: (same as current USBTiny implementation).

This also supports serial number check with the alternative -P usb:<serial_number> format (no ':').

In verbose mode, prints out bus/device/serial number for any found USBasps.

Only tested on Linux, but it works with HAVE_LIBUSB_1_0 on or off (there's some slightly different code in the two versions of usbOpenDevice()).

…e VID/PID (libusb or hidapi or libftdi) Added code for USBasp to check for bus:device match using the -P option. Syntax is -P usb:<bus>:<device> (same as current USBTiny implementation). This also supports serial number check with the alternative -P usb:<serial_number> format (no ':'). In verbose mode, prints out bus/device/serial number for any found USBasps. Only tested on Linux, but it works with HAVE_LIBUSB_1_0 on or off (there's some slightly different code in the two versions of usbOpenDevice()).

mcuee · 2023-12-04T06:49:37Z

Reference: #973

Differentiate multiple USB programmers of the same VID/PID (libusb or hidapi or libftdi) #973

stefanrueger · 2023-12-04T08:33:31Z

Thanks for submitting a PR!

I'll add a few tips on how to make the code easier to maintain (and read) and harden it against failure. Once that's been addressed it will need testing.

stefanrueger · 2023-12-04T08:35:18Z

src/usbasp.c

@@ -399,6 +399,37 @@ static int usbasp_transmit(const PROGRAMMER *pgm,
  return nbytes;
 }

+static int check_for_port_argument_match(const char *port, char *bus, char *device, char *serial_num) {


This function returns a 1 if the port arguments are not matched and 0 if they do. So I suggest renaming the function to port_argument_not_matched().

stefanrueger · 2023-12-04T08:41:54Z

src/usbasp.c

+        return 1;
+      }
+      return 0;
+    } else {


This is not wrong, but I am not in favour of elsing a return. It feels wrong.

stefanrueger · 2023-12-04T08:43:16Z

src/usbasp.c

+      return 0;
+    } else {
+      // serial number case
+      if (!str_eq(serial_num, port)) {


These five lines can be replaced by return !str_eq(serial_num, port);

... and actually, taking Jörg's comment about matching trailing SNs into account, it would be better to simply use

return !str_ends(serial_num, port);

or if you don't want an empty serial number by the user to match everything

return !(*port && str_ends(serial_num, port));

stefanrueger · 2023-12-04T08:53:26Z

src/usbasp.c

+    char *colon_pointer = strchr(port, ':');
+    if (colon_pointer) {
+      // Value contains ':' character. Compare with bus/device.
+      if (strncmp(port, bus, colon_pointer - port)) {


It is not wrong to use braces here, but it's unnecessary and costs an extra line. I generally prefer code to be crisp, to lessen the need for scrolling for maintenance. Those unnecessary braces appear a few times.

stefanrueger · 2023-12-04T08:58:43Z

src/usbasp.c

+        return 1;
+      }
+      port = colon_pointer + 1;
+      if (strcmp(port, device)) {


These 4 lines can be replaced by return !str_eq(port, device);

stefanrueger · 2023-12-04T09:03:06Z

src/usbasp.c

+              if(!str_eq(port, "usb")) {
+                // -P option given
+                libusb_get_string_descriptor_ascii(handle, descriptor.iSerialNumber, (unsigned char*)string, sizeof(string));
+                char bus_num[4];


Uhhm, bus_num needs to have space to hold the ASCII representation of any integer as we don't control what number range libusb_get_bus_number() returns. Worst case could mean 21 characters are needed.

stefanrueger · 2023-12-04T09:03:38Z

src/usbasp.c

+                libusb_get_string_descriptor_ascii(handle, descriptor.iSerialNumber, (unsigned char*)string, sizeof(string));
+                char bus_num[4];
+                sprintf(bus_num, "%d", libusb_get_bus_number(dev));
+                char dev_addr[4];


Same here: I suggest char dev_addr[21];

stefanrueger · 2023-12-04T09:09:35Z

Apart from a few niggles, the code looks good. Thank you @MikeRaffa. Don't forget to document the new superpower.

MikeRaffa · 2023-12-04T10:01:24Z

Thanks for the feedback! I'll try to look into all those items within the next couple of days.

dl8dtl · 2023-12-04T21:12:14Z

Question from me: has it been tested on at least two different OSes? It'd be interesting in particular to name some examples in the documentation.
(I'd appreciate if USBasps came with a kind of semi-random serial number to distinguish them, but I accept we won't be able to change the world. Alas.)

MikeRaffa · 2023-12-04T21:24:22Z

I've just tested this on Windows using MYSY2 Mingw64. Seems to work correctly. I am getting this error: "avrdude warning: cannot open USB device: Function not implemented", but avrdude is in fact communicating with the USBasp, so I'm not sure what's up with that.

Change sense of check_for_port_argument_match() to return nonzero when there is a match. Increase size of 'bus_num' and 'dev_addr' from 4 to 21. Various cleanup, formatting changes, and simplifications. Match trailing end of serial numbers.

mcuee · 2023-12-05T13:24:28Z

Working fine under Windows.

USBASP unit 1, snL 6666, connected to ATmega32A
USBASP unit 2, snL 8888, connected to ATmega328PB

PS C:\work\avr\avrdude_test\avrdude_bin> function prompt {$null}
PS>.\avrdude_pr1588 -C .\avrdude_pr1588.conf -c usbasp -p m328pb -P usb:8888
avrdude_pr1588: AVR device initialized and ready to accept instructions
avrdude_pr1588: device signature = 0x1e9516 (probably m328pb)

avrdude_pr1588 done.  Thank you.

PS>.\avrdude_pr1588 -C .\avrdude_pr1588.conf -c usbasp -p m32a -P usb:6666
avrdude_pr1588: AVR device initialized and ready to accept instructions
avrdude_pr1588: device signature = 0x1e9502 (probably m32a)

avrdude_pr1588 done.  Thank you.

mcuee · 2023-12-05T13:38:59Z

The PR is also good for Linux.

USBASP unit 1, snL 6666, connected to ATmega32A
USBASP unit 2, snL 8888, connected to ATmega328PB

mcuee@UbuntuSwift3 $ ./avrdude_pr1588 -C ./avrdude_pr1588.conf -c usbasp -p m328pb -P usb:8888
avrdude_pr1588: AVR device initialized and ready to accept instructions
avrdude_pr1588: device signature = 0x1e9516 (probably m328pb)

avrdude_pr1588 done.  Thank you.

mcuee@UbuntuSwift3 $ ./avrdude_pr1588 -C ./avrdude_pr1588.conf -c usbasp -p m32a -P usb:8888
avrdude_pr1588: AVR device initialized and ready to accept instructions
avrdude_pr1588: device signature = 0x1e9516 (probably m328pb)
avrdude_pr1588 error: expected signature for ATmega32A is 1E 95 02
               double check chip or use -F to override this check

avrdude_pr1588 done.  Thank you.

mcuee@UbuntuSwift3 $ ./avrdude_pr1588 -C ./avrdude_pr1588.conf -c usbasp -p m32a -P usb:6666
avrdude_pr1588: AVR device initialized and ready to accept instructions
avrdude_pr1588: device signature = 0x1e9502 (probably m32a)

avrdude_pr1588 done.  Thank you.

mcuee@UbuntuSwift3 $ ./avrdude_pr1588 -C ./avrdude_pr1588.conf -c usbasp -p m328pb -P usb:6666
avrdude_pr1588: AVR device initialized and ready to accept instructions
avrdude_pr1588: device signature = 0x1e9502 (probably m32)
avrdude_pr1588 error: expected signature for ATmega328PB is 1E 95 16
               double check chip or use -F to override this check

avrdude_pr1588 done.  Thank you.

stefanrueger · 2023-12-06T10:23:11Z

I've just tested this on Windows using MYSY2 Mingw64. Seems to work correctly. I am getting this error: "avrdude warning: cannot open USB device: Function not implemented", but avrdude is in fact communicating with the USBasp, so I'm not sure what's up with that.

@MikeRaffa Try running avrdude with -vvv so that the warning tells you the source file and line number of the message. Then you have a sporting chance to see why that warning might come about.

stefanrueger · 2023-12-06T10:24:28Z

src/usbasp.c

+      return str_eq(port, device);
+    }
+    // serial number case
+    return (*port && str_ends(serial_num, port));


Outer parentheses not needed

stefanrueger · 2023-12-06T10:26:01Z

src/usbasp.c

+                sprintf(bus_num, "%d", libusb_get_bus_number(dev));
+                char dev_addr[21];
+                sprintf(dev_addr, "%d", libusb_get_device_address(dev));
+                if (!check_for_port_argument_match(port, bus_num, dev_addr, string)) {


Unnecessary braces

stefanrueger · 2023-12-06T10:26:27Z

src/usbasp.c

+                    // -P option given
+                    usb_get_string_simple(handle, dev->descriptor.iSerialNumber,
+                                          string, sizeof(string));
+                    if (!check_for_port_argument_match(port, bus->dirname, dev->filename, string)) {


Unnecessary braces

stefanrueger · 2023-12-06T10:29:35Z

src/usbasp.c

@@ -399,6 +399,28 @@ static int usbasp_transmit(const PROGRAMMER *pgm,
  return nbytes;
 }

+static int check_for_port_argument_match(const char *port, char *bus, char *device, char *serial_num) {
+
+  if (verbose) {


pmsg_notice() only outputs sth if verbose is one or greater. Hence, the if test is unnecessary

Remove some unnecessary braces and parens.

MikeRaffa · 2023-12-07T01:10:56Z

This is the output I got with -vvv on Windows. It is correctly reading the signature from the MCU so I'm not sure what to make of that. Maybe something to do with Windows libusb setup? I was getting the same behavior without my changes. Note: this was run with the latest pushed version so the line number (457) reflects that.

$ ./build_mingw64_nt-10.0-22631/src/avrdude.exe -p m128 -c USBasp -vvv -P usb:3

avrdude: Version 7.2-20231201 (5501c63a)
         Copyright the AVRDUDE authors;
         see https://github.com/avrdudes/avrdude/blob/main/AUTHORS

         System wide configuration file is I:\stuff\msys64\home\raffa\avrdude\build_mingw64_nt-10.0-22631\src\avrdude.conf

         Using port            : usb:3
         Using programmer      : usbasp
avrdude: usbasp_open("usb:3")
avrdude usbOpenDevice() [usbasp.c:457] warning: cannot open USB device: Function not implemented
avrdude usbOpenDevice() [usbasp.c:457] warning: cannot open USB device: Function not implemented
avrdude: seen device from vendor >www.fischl.de<
avrdude: seen product >USBasp<
avrdude: usbOpenDevice(): found USBasp, bus:device: 2:32, serial_number: 3
         AVR Part              : ATmega128
         Programming modes     : ISP, HVPP, JTAG, JTAGmkI, SPM

         Memory         Size  Pg size
         ----------------------------
         eeprom         4096        8
         flash        131072      256
         efuse             1        1
         hfuse             1        1
         lfuse             1        1
         lock              1        1
         calibration       4        1
         signature         3        1
         io              224        1
         sram           4096        1

         Variants         Package  F max   T range         V range
         ----------------------------------------------------------------
         ATmega128-16AN   TQFP64   16 MHz  [-40 C, 105 C]  [4.5 V, 5.5 V]
         ATmega128-16ANR  TQFP64   16 MHz  [-40 C, 105 C]  [4.5 V, 5.5 V]
         ATmega128-16AU   TQFP64   16 MHz  [-40 C,  85 C]  [4.5 V, 5.5 V]
         ATmega128-16AUR  TQFP64   16 MHz  [-40 C,  85 C]  [4.5 V, 5.5 V]
         ATmega128-16MN   MLF64    16 MHz  [-40 C, 105 C]  [4.5 V, 5.5 V]
         ATmega128-16MNR  MLF64    16 MHz  [N/A,     N/A]  [4.5 V, 5.5 V]
         ATmega128-16MU   MLF64    16 MHz  [-40 C,  85 C]  [4.5 V, 5.5 V]
         ATmega128-16MUR  MLF64    16 MHz  [-40 C,  85 C]  [4.5 V, 5.5 V]
         ATmega128L-8AN   TQFP64   8 MHz   [-40 C, 105 C]  [3 V,   5.5 V]
         ATmega128L-8ANR  TQFP64   8 MHz   [-40 C, 105 C]  [3 V,   5.5 V]
         ATmega128L-8AU   TQFP64   8 MHz   [-40 C,  85 C]  [2.7 V, 5.5 V]
         ATmega128L-8AUR  TQFP64   8 MHz   [-40 C,  85 C]  [2.7 V, 5.5 V]
         ATmega128L-8MN   MLF64    8 MHz   [-40 C, 105 C]  [3 V,   5.5 V]
         ATmega128L-8MNR  MLF64    8 MHz   [N/A,     N/A]  [2.7 V, 5.5 V]
         ATmega128L-8MU   MLF64    8 MHz   [-40 C,  85 C]  [2.7 V, 5.5 V]
         ATmega128L-8MUR  MLF64    8 MHz   [-40 C,  85 C]  [2.7 V, 5.5 V]

         Programmer Type       : usbasp
         Description           : USBasp ISP and TPI programmer
avrdude: usbasp_initialize()
avrdude: usbasp_spi_set_sck_period(0)
avrdude: auto set sck period (because given equals null)
avrdude: usbasp_program_enable()
avrdude: AVR device initialized and ready to accept instructions
Reading |                                                    | 0% 0.00 s
avrdude: usbasp_spi_cmd(0x30, 0x00, 0x00, 0x00) => 0x00, 0x30, 0x00, 0x1e
avrdude: usbasp_spi_cmd(0x30, 0x00, 0x01, 0x00) => 0x00, 0x30, 0x00, 0x97
Reading | #################                                  | 33% 0.01 s
avrdude: usbasp_spi_cmd(0x30, 0x00, 0x02, 0x00) => 0x00, 0x30, 0x00, 0x02
Reading | ################################################## | 100% 0.02 s
avrdude: device signature = 0x1e9702 (probably m128)
avrdude: usbasp_close()

avrdude done.  Thank you.

stefanrueger · 2023-12-11T12:30:09Z

avrdude: usbasp_open("usb:3")
avrdude usbOpenDevice() [usbasp.c:457] warning: cannot open USB device: Function not implemented
avrdude usbOpenDevice() [usbasp.c:457] warning: cannot open USB device: Function not implemented

Interesting. The key seems to be why line 457 of usbasp.c is visited twice... Any idea?

MikeRaffa · 2023-12-11T18:03:05Z

I was able to get it run cleanly (without that error) by changing the driver to libusbK with the "Zadig" tool.

$ ./avrdude.exe -p m128 -c USBasp -P usb:2 -vvv

avrdude: Version 7.2-20231201 (5501c63a)
         Copyright the AVRDUDE authors;
         see https://github.com/avrdudes/avrdude/blob/main/AUTHORS

         System wide configuration file is I:\stuff\msys64\home\raffa\avrdude\build_mingw64_nt-10.0-22631\src\avrdude.conf

         Using port            : usb:2
         Using programmer      : usbasp
avrdude: usbasp_open("usb:2")
avrdude: seen device from vendor >www.fischl.de<
avrdude: seen product >USBasp<
avrdude: usbOpenDevice(): found USBasp, bus:device: 2:37, serial_number: 2
         AVR Part              : ATmega128
         Programming modes     : ISP, HVPP, JTAG, JTAGmkI, SPM

         Memory         Size  Pg size
         ----------------------------
         eeprom         4096        8
         flash        131072      256
         efuse             1        1
         hfuse             1        1
         lfuse             1        1
         lock              1        1
         calibration       4        1
         signature         3        1
         io              224        1
         sram           4096        1

         Variants         Package  F max   T range         V range
         ----------------------------------------------------------------
         ATmega128-16AN   TQFP64   16 MHz  [-40 C, 105 C]  [4.5 V, 5.5 V]
         ATmega128-16ANR  TQFP64   16 MHz  [-40 C, 105 C]  [4.5 V, 5.5 V]
         ATmega128-16AU   TQFP64   16 MHz  [-40 C,  85 C]  [4.5 V, 5.5 V]
         ATmega128-16AUR  TQFP64   16 MHz  [-40 C,  85 C]  [4.5 V, 5.5 V]
         ATmega128-16MN   MLF64    16 MHz  [-40 C, 105 C]  [4.5 V, 5.5 V]
         ATmega128-16MNR  MLF64    16 MHz  [N/A,     N/A]  [4.5 V, 5.5 V]
         ATmega128-16MU   MLF64    16 MHz  [-40 C,  85 C]  [4.5 V, 5.5 V]
         ATmega128-16MUR  MLF64    16 MHz  [-40 C,  85 C]  [4.5 V, 5.5 V]
         ATmega128L-8AN   TQFP64   8 MHz   [-40 C, 105 C]  [3 V,   5.5 V]
         ATmega128L-8ANR  TQFP64   8 MHz   [-40 C, 105 C]  [3 V,   5.5 V]
         ATmega128L-8AU   TQFP64   8 MHz   [-40 C,  85 C]  [2.7 V, 5.5 V]
         ATmega128L-8AUR  TQFP64   8 MHz   [-40 C,  85 C]  [2.7 V, 5.5 V]
         ATmega128L-8MN   MLF64    8 MHz   [-40 C, 105 C]  [3 V,   5.5 V]
         ATmega128L-8MNR  MLF64    8 MHz   [N/A,     N/A]  [2.7 V, 5.5 V]
         ATmega128L-8MU   MLF64    8 MHz   [-40 C,  85 C]  [2.7 V, 5.5 V]
         ATmega128L-8MUR  MLF64    8 MHz   [-40 C,  85 C]  [2.7 V, 5.5 V]

         Programmer Type       : usbasp
         Description           : USBasp ISP and TPI programmer
avrdude: usbasp_initialize()
avrdude: usbasp_spi_set_sck_period(0)
avrdude: auto set sck period (because given equals null)
avrdude: usbasp_program_enable()
avrdude: AVR device initialized and ready to accept instructions
Reading |                                                    | 0% 0.00 s
avrdude: usbasp_spi_cmd(0x30, 0x00, 0x00, 0x00) => 0x00, 0x30, 0x00, 0x1e
avrdude: usbasp_spi_cmd(0x30, 0x00, 0x01, 0x00) => 0x00, 0x30, 0x00, 0x97
Reading | #################                                  | 33% 0.02 s
avrdude: usbasp_spi_cmd(0x30, 0x00, 0x02, 0x00) => 0x00, 0x30, 0x00, 0x02
Reading | ################################################## | 100% 0.03 s
avrdude: device signature = 0x1e9702 (probably m128)
avrdude: usbasp_close()

avrdude done.  Thank you.

mcuee · 2023-12-15T05:24:27Z

I was able to get it run cleanly (without that error) by changing the driver to libusbK with the "Zadig" tool.

Interesting.

Zadig 2.8 comes with libusb0.sys 1.2.7.3 version, can you try that as well? I suspect that you were using libusb0.sys 1.0.26 version.
https://github.com/pbatard/libwdi/releases/tag/v1.5.0

You can also try WinUSB and I believe it should work as well.

Or you can use the MSVC build which use avrdude's own avrdude-libusb implementation. MinGW build uses libusb-1.0 (eg: USBASP) or libusb-1.0+libusb-compat-0.1 (other programmers).

BTW, latest libusb-win32 version is 1.3.0.1 and you need to use a customized version of Zadig to install it.
https://github.com/mcuee/libusb-win32/releases/tag/snapshot_1.3.0.1

MikeRaffa · 2023-12-17T02:44:27Z

It also ran cleanly with the WinUSB from Zadig 2.8.

mcuee · 2023-12-17T07:38:08Z

It also ran cleanly with the WinUSB from Zadig 2.8.

@MikeRaffa
Thanks. Have you tried to use libusb0.sys 1.2.7.3 from Zadig 2.8?

MikeRaffa · 2023-12-17T08:27:11Z

Worked with that one as well.

mcuee · 2023-12-17T14:23:24Z

Worked with that one as well.

Thanks. Then it should be fine. There are some issues with the older libusb0.sys 1.2.6.0 version with some avrdude programmers. It is great that you can get libusb0.sys 1.2.7.3, libusbK.sys and WinUSB working fine.

mcuee · 2023-12-17T14:25:59Z

This PR looks good to me based on my test results.

mcuee added the enhancement New feature or request label Dec 4, 2023

stefanrueger reviewed Dec 4, 2023

View reviewed changes

stefanrueger linked an issue Dec 4, 2023 that may be closed by this pull request

Differentiate multiple USB programmers of the same VID/PID (libusb or hidapi or libftdi) #973

Closed

MikeRaffa added 2 commits December 4, 2023 17:20

Updates based on feedback from stefanrueger.

fa2e69d

Change sense of check_for_port_argument_match() to return nonzero when there is a match. Increase size of 'bus_num' and 'dev_addr' from 4 to 21. Various cleanup, formatting changes, and simplifications. Match trailing end of serial numbers.

Documentation for Issue avrdudes#973 changes.

e351dc3

stefanrueger reviewed Dec 6, 2023

View reviewed changes

Remove check for 'verbose' on call to pmsg_notice().

5dbbb4e

Remove some unnecessary braces and parens.

stefanrueger merged commit 79f7035 into avrdudes:main Dec 18, 2023
10 checks passed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Issue #973: Differentiate multiple USB programmers of the same VID/PI… #1588

Issue #973: Differentiate multiple USB programmers of the same VID/PI… #1588

MikeRaffa commented Dec 4, 2023

mcuee commented Dec 4, 2023

stefanrueger commented Dec 4, 2023

stefanrueger Dec 4, 2023

stefanrueger Dec 4, 2023

stefanrueger Dec 4, 2023

stefanrueger Dec 4, 2023 •

edited

stefanrueger Dec 4, 2023

stefanrueger Dec 4, 2023

stefanrueger Dec 4, 2023

stefanrueger Dec 4, 2023

stefanrueger commented Dec 4, 2023

MikeRaffa commented Dec 4, 2023

dl8dtl commented Dec 4, 2023

MikeRaffa commented Dec 4, 2023

mcuee commented Dec 5, 2023

mcuee commented Dec 5, 2023 •

edited

stefanrueger commented Dec 6, 2023

stefanrueger Dec 6, 2023

stefanrueger Dec 6, 2023

stefanrueger Dec 6, 2023

stefanrueger Dec 6, 2023

MikeRaffa commented Dec 7, 2023 •

edited by mcuee

stefanrueger commented Dec 11, 2023

MikeRaffa commented Dec 11, 2023 •

edited by mcuee

mcuee commented Dec 15, 2023 •

edited

MikeRaffa commented Dec 17, 2023

mcuee commented Dec 17, 2023

MikeRaffa commented Dec 17, 2023

mcuee commented Dec 17, 2023

mcuee commented Dec 17, 2023

Issue #973: Differentiate multiple USB programmers of the same VID/PI… #1588

Issue #973: Differentiate multiple USB programmers of the same VID/PI… #1588

Conversation

MikeRaffa commented Dec 4, 2023

mcuee commented Dec 4, 2023

stefanrueger commented Dec 4, 2023

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

stefanrueger Dec 4, 2023 • edited

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

stefanrueger commented Dec 4, 2023

MikeRaffa commented Dec 4, 2023

dl8dtl commented Dec 4, 2023

MikeRaffa commented Dec 4, 2023

mcuee commented Dec 5, 2023

mcuee commented Dec 5, 2023 • edited

stefanrueger commented Dec 6, 2023

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

MikeRaffa commented Dec 7, 2023 • edited by mcuee

stefanrueger commented Dec 11, 2023

MikeRaffa commented Dec 11, 2023 • edited by mcuee

mcuee commented Dec 15, 2023 • edited

MikeRaffa commented Dec 17, 2023

mcuee commented Dec 17, 2023

MikeRaffa commented Dec 17, 2023

mcuee commented Dec 17, 2023

mcuee commented Dec 17, 2023

stefanrueger Dec 4, 2023 •

edited

mcuee commented Dec 5, 2023 •

edited

MikeRaffa commented Dec 7, 2023 •

edited by mcuee

MikeRaffa commented Dec 11, 2023 •

edited by mcuee

mcuee commented Dec 15, 2023 •

edited