Properly determine libusb read size for large reports (Fixes #274) by sudobash1 · Pull Request #728 · libusb/hidapi

sudobash1 · 2025-03-25T20:27:04Z

Summary

Derive the libusb interrupt-IN transfer size from the HID report descriptor so reports larger than wMaxPacketSize are delivered whole.
Preserve wMaxPacketSize as the minimum buffer size for devices that pad short reports, and fall back to it when a descriptor cannot be read or parsed safely.
Cache successfully fetched report descriptors for hid_get_report_descriptor().
Bound descriptor-controlled report sizes, handle read-thread allocation/submission failures safely, and skip automatic report-descriptor requests for non-HID vendor interfaces.
Add cross-platform parser tests using all 25 Windows _real.rpt_desc fixtures plus malformed, overflow, Report ID, Push/Pop, long-item, and boundary cases.

Verification

GCC C build with -Wall -Wextra -pedantic -Werror: 26 runnable tests passed; hardware-only DeviceIO test skipped.
GCC C++ backend build with -Wall -Wextra -Werror: 26 runnable tests passed; hardware-only DeviceIO test skipped.
AddressSanitizer + UndefinedBehaviorSanitizer build: 26 runnable tests passed; hardware-only DeviceIO test skipped.
Modeled FreeBSD CMake configuration preserves HIDAPI_WITH_TESTS=ON and registers all 26 libusb tests.

Fixes #274

Related to #796

Related to #833

Assisted-by: codex:gpt-5.6-sol
Assisted-by: claude-code:claude-fable-5

sudobash1 · 2025-03-25T22:08:47Z

PR updated to remove different signedness comparison compiler warning when compiling with -Wall.

Be-ing · 2025-03-26T01:07:51Z

Thank you for finally taking care of this longstanding bug.

Youw

Thanks for this, I'll perform a thorrow review a bit later.
@Be-ing can you confirm it fixes the issue with long reports too?

mcuee · 2025-03-27T05:12:38Z

BTW, there is a typo here.
@Youw and @JoergAtGithub.

hidapi/windows/test/data/046A_0011_0006_0001.pp_data

Line 4 in 0ab6c14

    
           dev->manufacturer_string = "dev->product_string      = "dev->release_number      = 0x0100

dev->manufacturer_string = "dev->product_string      = "dev->release_number      = 0x0100

JoergAtGithub · 2025-03-27T06:17:16Z

I put 046A_0011_0006_0001_real.rpt_desc into https://eleccelerator.com/usbdescreqparser/ and got:

0x05, 0x01,        // Usage Page (Generic Desktop Ctrls)
0x09, 0x06,        // Usage (Keyboard)
0xA1, 0x01,        // Collection (Application)
0x05, 0x07,        //   Usage Page (Kbrd/Keypad)
0x19, 0xE0,        //   Usage Minimum (0xE0)
0x29, 0xE7,        //   Usage Maximum (0xE7)
0x15, 0x00,        //   Logical Minimum (0)
0x25, 0x01,        //   Logical Maximum (1)
0x75, 0x01,        //   Report Size (1)
0x95, 0x08,        //   Report Count (8)
0x81, 0x02,        //   Input (Data,Var,Abs,No Wrap,Linear,Preferred State,No Null Position)
0x75, 0x08,        //   Report Size (8)
0x95, 0x01,        //   Report Count (1)
0x81, 0x03,        //   Input (Const,Var,Abs,No Wrap,Linear,Preferred State,No Null Position)
0x19, 0x00,        //   Usage Minimum (0x00)
0x29, 0xDD,        //   Usage Maximum (0xDD)
0x15, 0x00,        //   Logical Minimum (0)
0x26, 0xDD, 0x00,  //   Logical Maximum (221)
0x75, 0x08,        //   Report Size (8)
0x95, 0x06,        //   Report Count (6)
0x81, 0x00,        //   Input (Data,Array,Abs,No Wrap,Linear,Preferred State,No Null Position)
0x05, 0x08,        //   Usage Page (LEDs)
0x19, 0x01,        //   Usage Minimum (Num Lock)
0x29, 0x03,        //   Usage Maximum (Scroll Lock)
0x15, 0x00,        //   Logical Minimum (0)
0x25, 0x01,        //   Logical Maximum (1)
0x75, 0x01,        //   Report Size (1)
0x95, 0x03,        //   Report Count (3)
0x91, 0x02,        //   Output (Data,Var,Abs,No Wrap,Linear,Preferred State,No Null Position,Non-volatile)
0x75, 0x05,        //   Report Size (5)
0x95, 0x01,        //   Report Count (1)
0x91, 0x03,        //   Output (Const,Var,Abs,No Wrap,Linear,Preferred State,No Null Position,Non-volatile)
0xC0,              // End Collection

The InputReport contains:

8x1bit data
1x8bit constant padding
6x8bit data
This sums to 8byte, but InputReportByteLength is defined as follows:
Specifies the maximum size, in bytes, of all the input reports. Includes the report ID, which is prepended to the report data. If report ID is not used, the ID value is zero.
So we've to add a byte for the ReportID and are at 9 bytes.

JoergAtGithub · 2025-03-27T06:20:44Z

BTW, there is a typo here. @Youw and @JoergAtGithub.

hidapi/windows/test/data/046A_0011_0006_0001.pp_data

Line 4 in 0ab6c14

dev->manufacturer_string = "dev->product_string = "dev->release_number = 0x0100
dev->manufacturer_string = "dev->product_string      = "dev->release_number      = 0x0100

There is something missing, as this is autogenerated by pp_data_dump, I guess there was something in the manufacturer_string , that pp_data_dump couldn't handle. Could you please open a dedicated issue for this, as this is unrelated to this PR.

mcuee · 2025-03-27T07:34:55Z

There is something missing, as this is autogenerated by pp_data_dump, I guess there was something in the manufacturer_string , that pp_data_dump couldn't handle. Could you please open a dedicated issue for this, as this is unrelated to this PR.

pp_data_dump issue for 046A_0011_0006_0001 -- Lower Priority #729

sudobash1 · 2025-03-27T20:00:14Z

Thanks @JoergAtGithub for walking me through that. I was misreading the descriptor.

I have added tests for libusb using the same data as the windows tests. And I extended the functionality of the libusb method to be able to calculate the maximum output and feature report sizes as well. This has no functional use currently, but it lets us run three times as many tests since the pp_data files have all three max sizes available.

JoergAtGithub · 2025-03-27T20:10:00Z

And I extended the functionality of the libusb method to be able to calculate the maximum output and feature report sizes as well. This has no functional use currently, but it lets us run three times as many tests since the pp_data files have all three max sizes available.

Independent of this PR, I think it would generally make sense to store these 3 values in the device structure. On Windows we would have to use the values InputReportByteLength, OutputReportByteLength and FeatureReportByteLength from https://learn.microsoft.com/en-us/windows-hardware/drivers/ddi/hidpi/ns-hidpi-_hidp_caps . On all other backends we could use your parser to determine them - and with the testcase you proved that they are the same as on Windows.
In this way, we were able to programmatically determine the required buffer sizes for read/write operations on all platforms. @Youw What do you think about this?

Youw

I don't have a strong opinion on test implementation (I trust @JoergAtGithub on this one).
libusb implementation seem fine.

Lets make sure it runs with CI on Github Actions and we're good to go here.

Youw · 2025-03-28T10:39:49Z

And I extended the functionality of the libusb method to be able to calculate the maximum output and feature report sizes as well. This has no functional use currently, but it lets us run three times as many tests since the pp_data files have all three max sizes available.

Independent of this PR, I think it would generally make sense to store these 3 values in the device structure. On Windows we would have to use the values InputReportByteLength, OutputReportByteLength and FeatureReportByteLength from https://learn.microsoft.com/en-us/windows-hardware/drivers/ddi/hidpi/ns-hidpi-_hidp_caps . On all other backends we could use your parser to determine them - and with the testcase you proved that they are the same as on Windows. In this way, we were able to programmatically determine the required buffer sizes for read/write operations on all platforms. @Youw What do you think about this?

Lets continue here: #731

mcuee · 2025-03-28T14:46:09Z

This PR does not seem to work.

Test device is the same as the one used in Issue #274. The FW is a mod of Jan Axelson's FX2HID example and codes are included in the following #274 comment.

HID reports of more than 64 bytes and libusb backend #274 (comment)

I can reproduce the issue reported in #274 with hidapi git libusb backend, hidraw backend is okay.

mcuee@UbuntuSwift3 ~/build/hid/hidapitester (master)$ sudo ./hidapitester_hidraw_git --vidpid 0925:1234 --open --buflen 256 -l 129 --send-output 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128  --read-input
Opening device, vid/pid: 0x0925/0x1234
Writing output report of 129-bytes...wrote 129 bytes:
 00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F
 20 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F
 40 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F
 60 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F
 80
Reading 129-byte input report 0, 250 msec timeout...read 128 bytes:
 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F 20
 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F 40
 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F 60
 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F 80
 00
Closing device

mcuee@UbuntuSwift3 ~/build/hid/hidapitester (master)$ sudo ./hidapitester_libusb_git --vidpid 0925:1234 --open --buflen 256 -l 129 --send-output 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128  --read-input
Opening device, vid/pid: 0x0925/0x1234
Writing output report of 129-bytes...wrote 129 bytes:
 00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F
 20 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F
 40 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F
 60 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F
 80
Reading 129-byte input report 0, 250 msec timeout...read 64 bytes:
 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F 20
 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F 40
 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 00
Closing device

With this PR, no change in hidraw backend behavior, which is good.

mcuee@UbuntuSwift3 ~/build/hid/hidapitester (master)$ sudo ./hidapitester_hidraw_pr728 --vidpid 0925:1234 --open --buflen 256 -l 129 --send-output 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128  --read-input
Opening device, vid/pid: 0x0925/0x1234
Writing output report of 129-bytes...wrote 129 bytes:
 00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F
 20 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F
 40 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F
 60 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F
 80
Reading 129-byte input report 0, 250 msec timeout...read 128 bytes:
 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F 20
 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F 40
 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F 60
 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F 80
 00
Closing device

But the libusb backend fix is not working.

mcuee@UbuntuSwift3 ~/build/hid/hidapitester (master)$ make -f Makefile_pr728.libusb 
cc -I/usr/local/include/libusb-1.0 -I ../hidapi_pr728/hidapi -c ../hidapi_pr728/libusb/hid.c -o ../hidapi_pr728/libusb/hid.o
cc -I/usr/local/include/libusb-1.0 -I ../hidapi_pr728/hidapi -c hidapitester.c -o hidapitester.o
cc -I/usr/local/include/libusb-1.0 -I ../hidapi_pr728/hidapi ../hidapi_pr728/libusb/hid.o hidapitester.o -o hidapitester -L/usr/local/lib -lusb-1.0 -pthread

mcuee@UbuntuSwift3 ~/build/hid/hidapitester (master)$ mv hidapitester hidapitester_libusb_pr728 

mcuee@UbuntuSwift3 ~/build/hid/hidapitester (master)$ sudo ./hidapitester_libusb_pr728 --vidpid 0925:1234 --open --buflen 256 -l 128 --send-output 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128  --read-input
Opening device, vid/pid: 0x0925/0x1234
Writing output report of 128-bytes...wrote 128 bytes:
 00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F
 20 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F
 40 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F
 60 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F
Reading 128-byte input report 0, 250 msec timeout...read 0 bytes:
Closing device

mcuee · 2025-03-28T14:50:44Z

HID Report Descriptor.

ReportDscr:

	db 06h, 0A0h, 0FFh ;    Usage Page (FFA0h, vendor defined)
	db 09h, 01h     ;       Usage (vendor defined)
	db 0A1h, 01h    ;       Collection (Application)
	db 09h, 02h     ;       Usage (vendor defined)
	db 0A1h, 00h    ;       Collection (Physical)
	db 06h, 0A1h, 0FFh ;    Usage Page (vendor defined)

;; The Input report
	db 09h, 03h     ;       Usage (vendor defined)
	db 09h, 04h     ;       Usage (vendor defined)
	db 15h, 80h	;	Logical minimum (80h or -128)
	db 25h, 7Fh	;	Logical maximum (7Fh or 127)
	db 35h, 00h	;	Physical minimum (0)
	db 45h, 0FFh	;	Physical maximum (255)
	db 75h, 08h	;	Report size (8 bits)
	db 95h, 80h	;	Report count (128 fields)
	db 81h, 02h	;	Input (data, variable, absolute)

;; The Output report
	db 09h, 05h     ;       Usage (vendor defined)
	db 09h, 06h     ;       Usage (vendor defined)
	db 15h, 80h	;	Logical minimum (80h or -128)
	db 25h, 7Fh	;	Logical maximum (7Fh or 127)
	db 35h, 00h	;	Physical minimum (0)
	db 45h, 0FFh	;	Physical maximum (255)
	db 75h, 08h	;	Report size (8 bits)
	db 95h, 80h	;	Report count (128 fields)
	db 91h, 02h	;	Output (data, variable, absolute)

	db 0C0h         ;       End Collection (Physical)
	db 0C0h         ;       End Collection (Application)

ReportDscrEnd:

hidtest is okay.

mcuee@UbuntuSwift3 ~/build/hid/hidapi (master)$ sudo ./hidtest/hidtest-libusb 
hidapi test/example tool. Compiled with hidapi version 0.15.0, runtime version 0.15.0.
Compile-time version matches runtime version of hidapi.

Device Found
  type: 0925 1234
  path: 3-2.1:1.0
  serial_number: (null)
  Manufacturer: CYPRESS
  Product:      EZ-USB FX2 HID USBHIDIO
  Release:      0
  Interface:    0
  Usage (page): 0x0 (0x0)
  Bus type: 1 (USB)

  Report Descriptor: (52 bytes)
0x06, 0xa0, 0xff, 0x09, 0x01, 0xa1, 0x01, 0x09, 0x02, 0xa1, 
0x00, 0x06, 0xa1, 0xff, 0x09, 0x03, 0x09, 0x04, 0x15, 0x80, 
0x25, 0x7f, 0x35, 0x00, 0x45, 0xff, 0x75, 0x08, 0x95, 0x80, 
0x81, 0x02, 0x09, 0x05, 0x09, 0x06, 0x15, 0x80, 0x25, 0x7f, 
0x35, 0x00, 0x45, 0xff, 0x75, 0x08, 0x95, 0x80, 0x91, 0x02, 
0xc0, 0xc0, 
unable to open device

sudobash1 · 2025-03-29T18:20:37Z

If I have time, I'll try to address @JoergAtGithub and @Youw's comments on Monday.

@mcuee I tried feeding the report descriptor that you sent into get_max_report_size and it returned 129, which seems correct. I'll try to look into this on Monday as well.

I did notice a typo in your test. When you ran ./hidapitester_libusb_pr728, you specified -l 128 which should have been -l 129, but I think it should have returned data anyway.

mcuee · 2025-03-30T00:19:38Z

Let me try again.

mcuee@UbuntuSwift3 ~/build/cyusb/fxload_libusb (master)$ sudo ./fxload -v -t fx2lp -I ./fx2hid.hex -D /dev/bus/usb/003/019 
microcontroller type: fx2lp
single stage:  load on-chip memory
open RAM hexfile image ./fx2hid.hex
stop CPU
write on-chip, addr 0x0600 len  234 (0x00ea)
write on-chip, addr 0x09e5 len   10 (0x000a)
write on-chip, addr 0x0380 len  428 (0x01ac)
write on-chip, addr 0x0080 len  768 (0x0300)
write on-chip, addr 0x0033 len    3 (0x0003)
write on-chip, addr 0x09ff len    7 (0x0007)
write on-chip, addr 0x07b8 len   92 (0x005c)
write on-chip, addr 0x0851 len   59 (0x003b)
write on-chip, addr 0x05fe len    2 (0x0002)
write on-chip, addr 0x0a07 len    4 (0x0004)
write on-chip, addr 0x09ef len    8 (0x0008)
write on-chip, addr 0x09b0 len   18 (0x0012)
write on-chip, addr 0x09f7 len    8 (0x0008)
write on-chip, addr 0x09c2 len   18 (0x0012)
write on-chip, addr 0x0a0b len    6 (0x0006)
write on-chip, addr 0x091d len   40 (0x0028)
write on-chip, addr 0x096a len   24 (0x0018)
write on-chip, addr 0x06ea len   22 (0x0016)
write on-chip, addr 0x099a len   22 (0x0016)
write on-chip, addr 0x088c len   54 (0x0036)
write on-chip, addr 0x0982 len   24 (0x0018)
write on-chip, addr 0x0814 len   61 (0x003d)
write on-chip, addr 0x0a11 len   36 (0x0024)
write on-chip, addr 0x08f1 len   44 (0x002c)
write on-chip, addr 0x08c2 len   47 (0x002f)
write on-chip, addr 0x0945 len   20 (0x0014)
write on-chip, addr 0x05b8 len   70 (0x0046)
write on-chip, addr 0x0959 len   17 (0x0011)
write on-chip, addr 0x0043 len    3 (0x0003)
write on-chip, addr 0x0053 len    3 (0x0003)
write on-chip, addr 0x0700 len  184 (0x00b8)
write on-chip, addr 0x0000 len    3 (0x0003)
write on-chip, addr 0x052c len   12 (0x000c)
write on-chip, addr 0x09d4 len   17 (0x0011)
write on-chip, addr 0x0538 len  128 (0x0080)
write on-chip, addr 0x0a06 len    1 (0x0001)
... WROTE: 2497 bytes, 36 segments, avg 69
reset CPU

mcuee@UbuntuSwift3 ~/build/hid/hidapitester (master)$ sudo ./hidapitester_libusb_pr728 --vidpid 0925:1234 --open --buflen 256 -l 129 --send-output 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128  --read-input 
Opening device, vid/pid: 0x0925/0x1234
Writing output report of 129-bytes...wrote 129 bytes:
 00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F
 20 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F
 40 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F
 60 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F
 80
Reading 129-byte input report 0, 250 msec timeout...read 0 bytes:
Closing device

mcuee@UbuntuSwift3 ~/build/hid/hidapitester (master)$ sudo ./hidapitester_hidraw_pr728 --vidpid 0925:1234 --open --buflen 256 -l 129 --send-output 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128  --read-input 
Opening device, vid/pid: 0x0925/0x1234
Writing output report of 129-bytes...wrote 129 bytes:
 00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F
 20 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F
 40 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F
 60 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F
 80
Reading 129-byte input report 0, 250 msec timeout...read 128 bytes:
 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F 20
 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F 40
 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F 60
 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F 80
 00
Closing device

mcuee · 2025-03-30T03:11:55Z

Same problem under FreeBSD 14.1 Release.

This is with a physical machine, Chuwi mini PC, Intel J4125 CPU, 8GB/256GB configuration.

Without this PR, hidapi git has the problem mentioned in #274.

But then somehow this PR does not work.

mcuee@freebsd14:~/build/hidapi_pr730 $ uname -a
FreeBSD freebsd14 14.1-RELEASE FreeBSD 14.1-RELEASE releng/14.1-n267679-10e31f0946d8 GENERIC amd64

mcuee@freebsd14:~ $ sudo lsusb
Bus /dev/usb Device /dev/ugen0.5: ID 04b4:8613 Cypress Semiconductor Corp. CY7C68013 EZ-USB FX2 USB 2.0 Development Kit
Bus /dev/usb Device /dev/ugen0.6: ID 04d8:003f Microchip Technology, Inc. 
Bus /dev/usb Device /dev/ugen0.4: ID 13d3:3503 IMC Networks 
Bus /dev/usb Device /dev/ugen0.3: ID 1ea7:0064 SHARKOON Technologies GmbH 2.4GHz Wireless rechargeable vertical mouse [More&Better]
Bus /dev/usb Device /dev/ugen0.2: ID 04f2:0760 Chicony Electronics Co., Ltd Acer KU-0760 Keyboard
Bus /dev/usb Device /dev/ugen0.1: ID 0000:0000  

mcuee@freebsd14:~/build/hidapi_pr730 $ fxload 
no device specified!
usage: fxload [-vV] [-B backend] [-l] [-t type] [-D devpath]
		[-I firmware_hexfile] [-s loader] [-c config_byte]
		[-L link] [-m mode]
... [-D devpath] overrides DEVICE= in env
... device types:  one of an21, fx, fx2, fx2lp, fx3
... at least one of -I, -L, -m is required

mcuee@freebsd14:~/build/hidapi_pr730 $ fxload -t fx2lp -I ./fx2hid.hex -D vid=0x04b4,pid=0x8613

mcuee@freebsd14:~/build/hidapi_pr730 $ lsusb
Bus /dev/usb Device /dev/ugen0.6: ID 04d8:003f Microchip Technology, Inc. 

mcuee@freebsd14:~/build/hidapi_pr730 $ sudo chmod 666 /dev/ugen0.5

mcuee@freebsd14:~/build/hidapi_pr730 $ lsusb
Bus /dev/usb Device /dev/ugen0.5: ID 0925:1234 Lakeview Research 
Bus /dev/usb Device /dev/ugen0.6: ID 04d8:003f Microchip Technology, Inc. 

mcuee@freebsd14:~/build/hidapitester $ ./hidapitester_libusb_git --vidpid 0925:1234 --open --buflen 256 -l 129 --send-output 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128  --read-input
Opening device, vid/pid: 0x0925/0x1234
Writing output report of 129-bytes...wrote 129 bytes:
 00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F
 20 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F
 40 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F
 60 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F
 80
Reading up to 129-byte input report, 250 msec timeout...read 64 bytes:
 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F 60
 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F 80
 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 00
Closing device

mcuee@freebsd14:~/build/hidapitester $ ./hidapitester_libusb_pr728 --vidpid 0925:1234 --open --buflen 256 -l 129 --send-output 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128  --read-input
Opening device, vid/pid: 0x0925/0x1234
Writing output report of 129-bytes...wrote 129 bytes:
 00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F
 20 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F
 40 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F
 60 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F
 80
Reading up to 129-byte input report, 250 msec timeout...read 0 bytes:
Closing device

mcuee · 2025-04-01T00:01:36Z

If I use the original fx2hid example (loop back of two bytes output report and input report), it seems to me this PR works fine. But that just means this PR has no regression.
http://janaxelson.com/hidpage.htm
http://janaxelson.com/files/fx2hid.zip

mcuee@freebsd14:~/build/hidapitester $ sudo ./hidapitester_libusb_pr728 --vidpid 0925:1234 --open --buflen 3 -l 3 --send-output 0,1,2 --read-input
Opening device, vid/pid: 0x0925/0x1234
Writing output report of 3-bytes...wrote 3 bytes:
 00 01 02
Reading up to 3-byte input report, 250 msec timeout...read 2 bytes:
 01 02 00
Closing device

mcuee@freebsd14:~/build/hidapitester $ sudo ./hidapitester_libusb_git --vidpid 0925:1234 --open --buflen 3 -l 3 --send-output 0,1,2 --read-input
Opening device, vid/pid: 0x0925/0x1234
Writing output report of 3-bytes...wrote 3 bytes:
 00 01 02
Reading up to 3-byte input report, 250 msec timeout...read 2 bytes:
 01 02 00
Closing device

mcuee · 2025-12-30T02:03:37Z

@sudobash1

Just wondering if you have any updates on this one. I think it is pretty close now.

mcuee · 2026-03-20T01:46:06Z

Just update the branch to trigger a new build.

It will be good to move this PR forward. May need more testing and reviews.

mcuee · 2026-04-19T07:57:27Z

@Youw

Just wondering if it is worth asking Github Copilot to review this PR. It seems to be pretty close to be okay.

Youw

Coming back to this after dropping it for a while — did a spec/kernel cross-check pass and a code re-read. Leaving hardware testing to those with the devices.

Algorithm validation

get_max_report_size matches HID 1.11 (§6.2.2.7, §8 for Report ID semantics) and is structurally identical to Linux's report-descriptor parsing in drivers/hid/hid-core.c:

hid_add_field(): report->size += parser->global.report_size * parser->global.report_count;
hid_register_report(): if (id != 0) report_enum->numbered = 1;
On receive, one byte is consumed as the ID prefix iff numbered — which is exactly the + report_id_used semantic here.

The FX2 128-byte fix is the subtle one: the hang wasn't the size math, it was the unconditional +1 making the buffer a non-multiple of wMaxPacketSize and suppressing the libusb short-packet terminator. Good catch on making it conditional on the Report ID tag actually being used.

Reusing windows/test/data/*.pp_data for a cross-backend parity test is a nice consistency proof.

Suggestions before merge

Scope of "Fixes #274" — this resolves the long-report truncation aspect, but #274 also asks for Full-Speed misconfiguration detection / reporting to the app. Consider softening to "Partially fixes" or adding a note.
Platform length asymmetry that @mcuee observed on Windows (libusb wrote 130 / read 129 vs native HID wrote 129 / read 128) — worth a sentence in the PR body pointing to #731 so users don't re-file it as a regression.
Inline nits on specific lines below.

Algorithm + spec alignment LGTM on the core change.

Extended version:

Comparison with the spec and other implementations
HID 1.11 spec (§6.2.2.7 & §5.6 / §8): "If a Report ID tag is used anywhere in Report descriptor, all data reports for the device are preceded by a single byte ID field." Report ID = 0 is a placeholder meaning "no report IDs". Reports are sized as Σ(report_count × report_size) bits, rounded up to bytes. The PR's algorithm and the conditional +report_id_used byte match the spec exactly.

Windows (HIDP_CAPS): InputReportByteLength/OutputReportByteLength/FeatureReportByteLength — these always include the Report-ID byte (value 0 when unused). That's the source of the read/write-length platform asymmetry and is exactly what follow-up issue #731 is about. The PR's test case subtracts 1 byte from these values when ReportID == 0x00 to reconcile with the libusb convention (see parse_max_input_report_size in max_input_report_size_test.c).

Linux kernel (drivers/hid/hid-core.c): Uses the same algorithm. hid_add_field() accumulates report->size += parser->global.report_size * parser->global.report_count. In hid_register_report(), if (id != 0) report_enum->numbered = 1;. In hid_report_raw_event, when numbered is true the first data byte is consumed as the ID (cdata++; csize--;). Size is bounded by max_buffer_size - 1 bits when numbered. This is structurally identical to what the PR does — so the PR's approach is directly validated by the reference kernel implementation.

hidapi/linux (hidraw backend): Avoids parsing entirely — delegates to the kernel: read(dev->device_handle, data, length) with user-supplied length, because the kernel has already sized the per-report buffer. That's why hidraw "just works" for the 128-byte FX2 firmware and why the libusb backend is the outlier.

Strengths
Correct algorithm. Matches HID 1.11 and Linux kernel's parsing exactly, including the subtle conditional Report-ID byte.
Good test coverage via reuse of 25 real-world Windows pp_data fixtures — this is a nice cross-backend consistency proof.
Safe fallback to input_ep_max_packet_size when the descriptor is corrupt/unparseable — no regression risk for devices already working.
Fixes the classic "exact multiple of max packet size hangs" libusb gotcha — the conditional +1 was the critical insight; without it devices with payloads that happen to be multiples of 64 would hang forever waiting for a short packet.
Remaining concerns worth flagging on review
Re-read of report descriptor. hidapi_initialize_device now calls hid_get_report_descriptor_libusb during init (new control transfer on every open). This is a second descriptor fetch in addition to any caller-driven hid_get_report_descriptor(). Consider caching.
HID_API_MAX_REPORT_DESCRIPTOR_SIZE buffer on stack (4096 bytes typically). libusb/hid.c:1314 in the diff allocates this on the stack — acceptable, but worth noting for embedded targets.
cur_size reset timing. In the parser, cur_size is reset on each Report ID encounter. For descriptors that interleave INPUT/OUTPUT items between Report ID changes this is correct; just note that the max_size comparison happens only at Report-ID boundaries and at the end — which matches the spec semantics where each Report ID opens a fresh report namespace.
Platform-length asymmetry (mcuee's Windows observation) is intentionally out of scope — tracked in #731. Worth linking from the PR description.
PR description still says "Fixes #274" but #274 has broader asks (detection of USB Full-Speed misconfiguration, speed reporting to app). This PR fixes the "long reports get truncated" aspect; the speed/error-reporting aspect remains open.
Typo in test comment (.dpt_desc → .rpt_desc) in libusb/test/CMakeLists.txt.
Bottom line
The PR is technically sound, aligned with HID 1.11 and the Linux kernel reference, and has been hardware-validated on Linux, FreeBSD and Android. The earlier "doesn't work" reports were a spec-compliance bug in the Report-ID handling that sudobash1 fixed. Primary blockers now are administrative: final Youw approval + merge, plus perhaps addressing the minor concerns above. The broader cross-platform length-consistency question is correctly deferred to #731.

Drafted with Claude Code.

Youw · 2026-04-19T20:01:01Z

This is the result of the review by Claude Opus 4.7 (locally).

JoergAtGithub · 2026-04-19T21:48:37Z

PR description still says "Fixes #274" but #274 has broader asks (detection of USB Full-Speed misconfiguration, speed reporting to app). This PR fixes the "long reports get truncated" aspect; the speed/error-reporting aspect remains open.

I think this statement is not correct. The USB standard specifies the max read size per speed category. With this PR every read size should work and therefore every USB speed.

mcuee · 2026-04-24T10:30:43Z

This is the result of the review by Claude Opus 4.7 (locally).

@Youw

Nice reviews. Just want to check if I read the comments correctly.

This PR is in general okay.
The platform related length discrepancy can be delayed to Issue Store Input/Output/Feature reports sizes for all platforms #731.

In that case, I think this PR can be merged.

mcuee · 2026-04-25T04:38:25Z

Latest test results with hidapi git.

This PR is OK.

Windows

PS C:\work\libusb\hidapitester> .\hidapitester_git_winapi.exe --vidpid 0925:1234 --open --buflen 256 -l 129 --send-output 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128  --read-input
Opening device, vid/pid: 0x0925/0x1234
Writing output report of 129-bytes...wrote 129 bytes:
 00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F
 20 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F
 40 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F
 60 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F
 80
Reading up to 129-byte input report, 250 msec timeout...read 128 bytes:
 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F 20
 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F 40
 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F 60
 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F 80
 00
Closing device

PS C:\work\libusb\hidapitester> .\hidapitester_git_libusb.exe --vidpid 0925:1234 --open --buflen 256 -l 129 --send-output 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128  --read-input
Opening device, vid/pid: 0x0925/0x1234
Writing output report of 129-bytes...wrote 129 bytes:
 00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F
 20 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F
 40 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F
 60 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F
 80
Reading up to 129-byte input report, 250 msec timeout...read 64 bytes:
 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F 20
 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F 40
 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 00
Closing device

PS C:\work\libusb\hidapitester> .\hidapitester_pr728_libusb_new.exe --vidpid 0925:1234 --open --buflen 256 -l 129 --send-output 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128  --read-input
Opening device, vid/pid: 0x0925/0x1234
Writing output report of 129-bytes...wrote 129 bytes:
 00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F
 20 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F
 40 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F
 60 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F
 80
Reading up to 129-byte input report, 250 msec timeout...read 128 bytes:
 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F 20
 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F 40
 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F 60
 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F 80
 00
Closing device

FreeBSD

mcuee@freebsdvm:~/build/hidapitester $ sudo ./hidapitester_git_libusb --vidpid 0925:1234 --open --buflen 256 -l 129 --send-output 0,1,2,3,4,5,6,7,8,
9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58
,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,
106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128  --read-input
Opening device, vid/pid: 0x0925/0x1234
Writing output report of 129-bytes...wrote 129 bytes:
 00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F
 20 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F
 40 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F
 60 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F
 80
Reading up to 129-byte input report, 250 msec timeout...read 64 bytes:
 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F 20
 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F 40
 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 00
Closing device

mcuee@freebsdvm:~/build/hidapitester $ sudo ./hidapitester_pr728_libusb --vidpid 0925:1234 --open --buflen 256 -l 129 --send-output 0,1,2,3,4,5,6,7,
8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,
58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,10
5,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128  --read-input
Opening device, vid/pid: 0x0925/0x1234
Writing output report of 129-bytes...wrote 129 bytes:
 00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F
 20 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F
 40 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F
 60 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F
 80
Reading up to 129-byte input report, 250 msec timeout...read 128 bytes:
 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F 20
 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F 40
 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F 60
 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F 80
 00
Closing device

mcuee@freebsdvm:~/build/hidapitester $ sudo usbconfig
ugen1.1: <EHCI root HUB Intel> at usbus1, cfg=0 md=HOST spd=HIGH (480Mbps) pwr=SAVE (0mA)
ugen0.1: <OHCI root HUB Apple> at usbus0, cfg=0 md=HOST spd=FULL (12Mbps) pwr=SAVE (0mA)
ugen1.2: <EZ-USB FX2 HID USBHIDIO Lakeview Research> at usbus1, cfg=0 md=HOST spd=HIGH (480Mbps) pwr=ON (80mA)

mcuee@freebsdvm:~/build/hidapitester $ sudo lsusb -vvv -d 0925:1234

Bus /dev/usb Device /dev/ugen1.2: ID 0925:1234 Lakeview Research
Device Descriptor:
  bLength                18
  bDescriptorType         1
  bcdUSB               2.00
  bDeviceClass            0 (Defined at Interface level)
  bDeviceSubClass         0
  bDeviceProtocol         0
  bMaxPacketSize0        64
  idVendor           0x0925 Lakeview Research
  idProduct          0x1234
  bcdDevice            0.00
  iManufacturer           1 CYPRESS
  iProduct                2 EZ-USB FX2 HID USBHIDIO
  iSerial                 0
  bNumConfigurations      1
  Configuration Descriptor:
    bLength                 9
    bDescriptorType         2
    wTotalLength           41
    bNumInterfaces          1
    bConfigurationValue     1
    iConfiguration          3
    bmAttributes         0x80
      (Bus Powered)
    MaxPower               80mA
    Interface Descriptor:
      bLength                 9
      bDescriptorType         4
      bInterfaceNumber        0
      bAlternateSetting       0
      bNumEndpoints           2
      bInterfaceClass         3 Human Interface Device
      bInterfaceSubClass      0 No Subclass
      bInterfaceProtocol      0 None
      iInterface              0
        HID Device Descriptor:
          bLength                 9
          bDescriptorType        33
          bcdHID               1.10
          bCountryCode            0 Not supported
          bNumDescriptors         1
          bDescriptorType        34 Report
          wDescriptorLength      52
          Report Descriptor: (length is 52)
            Item(Global): Usage Page, data= [ 0xa0 0xff ] 65440
                            (null)
            Item(Local ): Usage, data= [ 0x01 ] 1
                            (null)
            Item(Main  ): Collection, data= [ 0x01 ] 1
                            Application
            Item(Local ): Usage, data= [ 0x02 ] 2
                            (null)
            Item(Main  ): Collection, data= [ 0x00 ] 0
                            Physical
            Item(Global): Usage Page, data= [ 0xa1 0xff ] 65441
                            (null)
            Item(Local ): Usage, data= [ 0x03 ] 3
                            (null)
            Item(Local ): Usage, data= [ 0x04 ] 4
                            (null)
            Item(Global): Logical Minimum, data= [ 0x80 ] 128
            Item(Global): Logical Maximum, data= [ 0x7f ] 127
            Item(Global): Physical Minimum, data= [ 0x00 ] 0
            Item(Global): Physical Maximum, data= [ 0xff ] 255
            Item(Global): Report Size, data= [ 0x08 ] 8
            Item(Global): Report Count, data= [ 0x80 ] 128
            Item(Main  ): Input, data= [ 0x02 ] 2
                            Data Variable Absolute No_Wrap Linear
                            Preferred_State No_Null_Position Non_Volatile Bitfield
            Item(Local ): Usage, data= [ 0x05 ] 5
                            (null)
            Item(Local ): Usage, data= [ 0x06 ] 6
                            (null)
            Item(Global): Logical Minimum, data= [ 0x80 ] 128
            Item(Global): Logical Maximum, data= [ 0x7f ] 127
            Item(Global): Physical Minimum, data= [ 0x00 ] 0
            Item(Global): Physical Maximum, data= [ 0xff ] 255
            Item(Global): Report Size, data= [ 0x08 ] 8
            Item(Global): Report Count, data= [ 0x80 ] 128
            Item(Main  ): Output, data= [ 0x02 ] 2
                            Data Variable Absolute No_Wrap Linear
                            Preferred_State No_Null_Position Non_Volatile Bitfield
            Item(Main  ): End Collection, data=none
            Item(Main  ): End Collection, data=none
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x81  EP 1 IN
        bmAttributes            3
          Transfer Type            Interrupt
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0040  1x 64 bytes
        bInterval               5
      Endpoint Descriptor:
        bLength                 7
        bDescriptorType         5
        bEndpointAddress     0x01  EP 1 OUT
        bmAttributes            3
          Transfer Type            Interrupt
          Synch Type               None
          Usage Type               Data
        wMaxPacketSize     0x0040  1x 64 bytes
        bInterval               5
Device Qualifier (for other device speed):
  bLength                10
  bDescriptorType         6
  bcdUSB               2.00
  bDeviceClass            0 (Defined at Interface level)
  bDeviceSubClass         0
  bDeviceProtocol         0
  bMaxPacketSize0         4
  bNumConfigurations      1
can't get debug descriptor: Input/output error
Device Status:     0x0002
  (Bus Powered)
  Remote Wakeup Enabled

mcuee · 2026-04-25T09:09:27Z

Latest results under Linux.

This PR is okay.

mcuee  …/hidapitester   main ?    v15.2.0  17:06  sudo ./hidapitester_git_hidraw --vidpid 0925:1234 --open --buflen 256 -l 129 --send-output 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128  --read-input
Opening device, vid/pid: 0x0925/0x1234
Writing output report of 129-bytes...wrote 129 bytes:
 00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F
 20 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F
 40 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F
 60 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F
 80
Reading up to 129-byte input report, 250 msec timeout...read 128 bytes:
 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F 20
 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F 40
 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F 60
 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F 80
 00
Closing device

mcuee  …/hidapitester   main ?    v15.2.0  17:07  sudo ./hidapitester_git_libusb --vidpid 0925:1234 --open --buflen 256 -l 129 --send-output 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128  --read-input
Opening device, vid/pid: 0x0925/0x1234
Writing output report of 129-bytes...wrote 129 bytes:
 00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F
 20 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F
 40 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F
 60 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F
 80
Reading up to 129-byte input report, 250 msec timeout...read 64 bytes:
 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F 20
 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F 40
 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 00
Closing device

mcuee  …/hidapitester   main ?    v15.2.0  17:07  sudo ./hidapitester_pr728_libusb --vidpid 0925:1234 --open --buflen 256 -l 129 --send-output 0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128  --read-input
Opening device, vid/pid: 0x0925/0x1234
Writing output report of 129-bytes...wrote 129 bytes:
 00 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F
 20 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F
 40 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F
 60 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F
 80
Reading up to 129-byte input report, 250 msec timeout...read 128 bytes:
 01 02 03 04 05 06 07 08 09 0A 0B 0C 0D 0E 0F 10 11 12 13 14 15 16 17 18 19 1A 1B 1C 1D 1E 1F 20
 21 22 23 24 25 26 27 28 29 2A 2B 2C 2D 2E 2F 30 31 32 33 34 35 36 37 38 39 3A 3B 3C 3D 3E 3F 40
 41 42 43 44 45 46 47 48 49 4A 4B 4C 4D 4E 4F 50 51 52 53 54 55 56 57 58 59 5A 5B 5C 5D 5E 5F 60
 61 62 63 64 65 66 67 68 69 6A 6B 6C 6D 6E 6F 70 71 72 73 74 75 76 77 78 79 7A 7B 7C 7D 7E 7F 80
 00
Closing device

Youw · 2026-04-25T11:44:20Z

This PR is okay.

There is still a number of unresolved comments across the code. Need to either handle or a reason to dismiss, before merge this. (Or move out into separate issue/ticket.

mcuee · 2026-05-01T04:34:44Z

No regression for the following HID device under Linux (with non-Zero Input/Output/Feature reports IDs).
https://github.com/mcuee/picusb/tree/master/lvr_genhid_mod

mcuee  …/hidapitester   main ?    v15.2.0  12:31  sudo ./hidapitester_pr728_libusb --vidpid 0925:7001 --open --buflen 64 -l 3 --send-output 2,5,6  --read-input
Opening device, vid/pid: 0x0925/0x7001
Writing output report of 3-bytes...wrote 3 bytes:
 02 05 06
Reading up to 3-byte input report, 250 msec timeout...read 3 bytes:
 01 05 06
Closing device

mcuee  …/hidapitester   main ?    v15.2.0  12:31  sudo ./hidapitester_pr728_libusb --vidpid 0925:7001 --open --buflen 64 -l 3 --send-output 2,5,6  --read-input
Opening device, vid/pid: 0x0925/0x7001
Writing output report of 3-bytes...wrote 3 bytes:
 02 05 06
Reading up to 3-byte input report, 250 msec timeout...read 3 bytes:
 01 05 06
Closing device

mcuee  …/hidapitester   main ?    v15.2.0  12:31  sudo ./hidapitester_pr728_libusb --vidpid 0925:7001 --open --buflen 64 -l 3 --send-feature 3,5,6  --read-feature 4
Opening device, vid/pid: 0x0925/0x7001
Writing 3-byte feature report...wrote 3 bytes:
 03 05 06
Reading 3-byte feature report, report_id 4...read 3 bytes:
 04 05 06
Closing device

mcuee  …/hidapitester   main ?    v15.2.0  12:32  sudo ./hidapitester_pr728_libusb --vidpid 0925:7001 --open --buflen 64 -l 3 --send-feature 3,5,6  --read-feature 4
Opening device, vid/pid: 0x0925/0x7001
Writing 3-byte feature report...wrote 3 bytes:
 03 05 06
Reading 3-byte feature report, report_id 4...read 3 bytes:
 04 05 06
Closing device

mcuee · 2026-05-01T05:08:56Z

No regression under Windows either.

Minor modification needs to be used to get hidapitester to build with hidapi git or this PR, using libusb backend.

PS C:\work\libusb\hidapi_pr728_new> git diff
diff --git a/libusb/hid.c b/libusb/hid.c
index aabe462..fc2f7e5 100644
--- a/libusb/hid.c
+++ b/libusb/hid.c
@@ -34,8 +34,8 @@
 #include <unistd.h>
 #include <sys/types.h>
 #include <sys/stat.h>
-#include <sys/ioctl.h>
-#include <sys/utsname.h>
+//#include <sys/ioctl.h>
+//#include <sys/utsname.h>
 #include <fcntl.h>
 #include <wchar.h>

Using WinUSB driver.

$ ./hidapitester_pr728_libusb.exe --vidpid 0925:7001 --open --buflen 64 -l 3 --send-output 2,5,6  --read-input
Opening device, vid/pid: 0x0925/0x7001
Writing output report of 3-bytes...wrote 3 bytes:
 02 05 06
Reading up to 3-byte input report, 250 msec timeout...read 3 bytes:
 01 05 06
Closing device

$ ./hidapitester_pr728_libusb.exe --vidpid 0925:7001 --open --buflen 64 -l 3 --send-output 2,5,6  --read-input
Opening device, vid/pid: 0x0925/0x7001
Writing output report of 3-bytes...wrote 3 bytes:
 02 05 06
Reading up to 3-byte input report, 250 msec timeout...read 3 bytes:
 01 05 06
Closing device

$ ./hidapitester_pr728_libusb.exe --vidpid 0925:7001 --open --buflen 64 -l 3 --send-output 3,5,6  --read-feature 4
Opening device, vid/pid: 0x0925/0x7001
Writing output report of 3-bytes...wrote 3 bytes:
 03 05 06
Reading 3-byte feature report, report_id 4...read 3 bytes:
 04 05 06
Closing device

$ ./hidapitester_pr728_libusb.exe --vidpid 0925:7001 --open --buflen 64 -l 3 --send-output 3,5,6  --read-feature 4
Opening device, vid/pid: 0x0925/0x7001
Writing output report of 3-bytes...wrote 3 bytes:
 03 05 06
Reading 3-byte feature report, report_id 4...read 3 bytes:
 04 05 06
Closing device

JoergAtGithub · 2026-05-06T21:16:05Z

@sudobash1 It would be great if you could address the lost open comments in the code review.

JoergAtGithub · 2026-07-17T18:32:59Z

@sudobash1 It would be great if you could address the last remaining comments in this PR.

Test real report descriptors, distinguish malformed descriptors, cache descriptor data, and run the libusb tests in CI. Assisted-by: codex:gpt-5.6-sol

Bring PR libusb#728 onto the current base and retain the libusb report-size tests alongside the new virtual-device test suite. Assisted-by: codex:gpt-5.6-sol

Preserve endpoint-packet compatibility, bound descriptor-derived buffers, unwind read-thread initialization failures, avoid non-HID descriptor requests, and expand cross-platform parser coverage. Assisted-by: codex:gpt-5.6-sol Assisted-by: claude-code:claude-fable-5

Resolve the libusb CMake conflict by retaining upstream's generalized pkg-config generation together with the report-descriptor parser tests. Assisted-by: codex:gpt-5.6-sol

Youw

After latest fixes from codex - this needs to be tested

sudobash1 mentioned this pull request Mar 25, 2025

HID reports of more than 64 bytes and libusb backend #274

Open

sudobash1 force-pushed the libusb_input_size branch from 068fd62 to 52c6cd9 Compare March 25, 2025 22:06

mcuee added bug Something isn't working libusb Related to libusb backend labels Mar 26, 2025

Youw reviewed Mar 26, 2025

View reviewed changes

Comment thread libusb/hid.c Outdated

JoergAtGithub reviewed Mar 26, 2025

View reviewed changes

Comment thread libusb/hid.c Outdated

mcuee mentioned this pull request Mar 27, 2025

pp_data_dump issue for 046A_0011_0006_0001 -- Lower Priority #729

Closed

sudobash1 force-pushed the libusb_input_size branch 2 times, most recently from 2708264 to 3710895 Compare March 27, 2025 19:42

JoergAtGithub suggested changes Mar 27, 2025

View reviewed changes

Comment thread libusb/test/CMakeLists.txt Outdated

Comment thread libusb/hid.c Outdated

Youw reviewed Mar 28, 2025

View reviewed changes

Comment thread CMakeLists.txt Outdated

Youw reviewed Mar 28, 2025

View reviewed changes

Comment thread libusb/test/CMakeLists.txt Outdated

Youw reviewed Mar 28, 2025

View reviewed changes

Comment thread libusb/test/CMakeLists.txt

Youw reviewed Mar 28, 2025

View reviewed changes

Youw mentioned this pull request Mar 28, 2025

Store Input/Output/Feature reports sizes for all platforms #731

Open

mcuee force-pushed the libusb_input_size branch from a4c4da9 to e19ba07 Compare December 30, 2025 02:03

Merge branch 'master' into libusb_input_size

6eedce1

Merge branch 'master' into libusb_input_size

cda57eb

Youw reviewed Apr 19, 2026

View reviewed changes

Comment thread libusb/test/CMakeLists.txt Outdated

Comment thread libusb/hid.c Outdated

Youw mentioned this pull request Apr 25, 2026

Move the test data to <root>/test_data ? #796

Open

auxcorelabs mentioned this pull request Apr 25, 2026

Add hid_set_num_input_buffers() API #787

Open

Merge branch 'master' into libusb_input_size

7ebb679

mcuee added 2 commits May 20, 2026 09:41

Merge branch 'master' into libusb_input_size

1ab6b3f

Merge branch 'master' into libusb_input_size

dffba86

mcuee mentioned this pull request Jun 21, 2026

PICKit 4, SNAP, Atmel ICE and Power Debugger only work with hidapi but not libusb avrdudes/avrdude#1221

Open

Youw added 3 commits July 30, 2026 17:13

Address libusb report-size review feedback

39e58ae

Test real report descriptors, distinguish malformed descriptors, cache descriptor data, and run the libusb tests in CI. Assisted-by: codex:gpt-5.6-sol

Merge master into libusb_input_size

62c717d

Bring PR libusb#728 onto the current base and retain the libusb report-size tests alongside the new virtual-device test suite. Assisted-by: codex:gpt-5.6-sol

Youw added the ci-virtual-device Run the virtual HID device CI jobs (Windows driver + libusb raw-gadget) label Jul 30, 2026

Merge master into libusb_input_size

580996e

Resolve the libusb CMake conflict by retaining upstream's generalized pkg-config generation together with the report-descriptor parser tests. Assisted-by: codex:gpt-5.6-sol

Youw reviewed Jul 30, 2026

View reviewed changes

Youw mentioned this pull request Jul 30, 2026

Consolidate HID report descriptor parsing across backends #833

Open

Uh oh!

Conversation

sudobash1 commented Mar 25, 2025 • edited by Youw Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Verification

Uh oh!

sudobash1 commented Mar 25, 2025

Uh oh!

Be-ing commented Mar 26, 2025

Uh oh!

Youw left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

mcuee commented Mar 27, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

JoergAtGithub commented Mar 27, 2025

Uh oh!

JoergAtGithub commented Mar 27, 2025

Uh oh!

mcuee commented Mar 27, 2025

Uh oh!

Uh oh!

Uh oh!

sudobash1 commented Mar 27, 2025

Uh oh!

JoergAtGithub commented Mar 27, 2025

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Youw left a comment

Choose a reason for hiding this comment

Uh oh!

Youw commented Mar 28, 2025

Uh oh!

mcuee commented Mar 28, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

mcuee commented Mar 28, 2025

Uh oh!

sudobash1 commented Mar 29, 2025

Uh oh!

mcuee commented Mar 30, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

mcuee commented Mar 30, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

mcuee commented Apr 1, 2025

Uh oh!

mcuee commented Dec 30, 2025

Uh oh!

mcuee commented Mar 20, 2026

Uh oh!

mcuee commented Apr 19, 2026

Uh oh!

Youw left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Algorithm validation

Suggestions before merge

Uh oh!

Uh oh!

Uh oh!

Youw commented Apr 19, 2026

Uh oh!

JoergAtGithub commented Apr 19, 2026

Uh oh!

mcuee commented Apr 24, 2026

Uh oh!

mcuee commented Apr 25, 2026

Uh oh!

mcuee commented Apr 25, 2026

Uh oh!

Youw commented Apr 25, 2026

Uh oh!

sudobash1 commented Mar 25, 2025 •

edited by Youw

Loading

mcuee commented Mar 27, 2025 •

edited

Loading

mcuee commented Mar 28, 2025 •

edited

Loading

mcuee commented Mar 30, 2025 •

edited

Loading

mcuee commented Mar 30, 2025 •

edited

Loading

Youw left a comment •

edited

Loading

mcuee commented May 1, 2026 •

edited

Loading