 |
 |
| View previous topic :: View next topic |
| Author |
Message |
starkeeper
Guest
|
| Cant send more than 4Bytes with Code from ex_usb_hid via USB |
 Posted: Wed Dec 28, 2005 2:37 pm |
 |
|
Hi,
I'm trying to send 6 Bytes of Data with my PIC2550 over USB. I took the ex_usb_hid example and tuned it a Bit, to send 6Bytes. But only the first time 4Bytes arrive the Host PC the other 2 Bytes are lost. The other Times I get a timeout from my Software, or an device is not working error.
That is my Code:
| Code: |
#include <18F2550.h>
#fuses HSPLL,NOWDT,NOPROTECT,NOLVP,NODEBUG,USBDIV,PLL5,CPUDIV1,VREGEN
#use delay(clock=48000000)
#DEFINE USB_HID_DEVICE TRUE
#define USB_EP1_TX_ENABLE USB_ENABLE_INTERRUPT //turn on EP1 for IN bulk/interrupt transfers
#define USB_EP1_TX_SIZE 8 //allocate 8 bytes in the hardware for transmission
#define USB_EP1_RX_ENABLE USB_ENABLE_INTERRUPT //turn on EP1 for OUT bulk/interrupt transfers
#define USB_EP1_RX_SIZE 8 //allocate 8 bytes in the hardware for reception
#define USB_CON_SENSE_PIN PIN_C2
#include <pic18_usb.h> //Microchip 18Fxx5x hardware layer for usb.c
#include <usb_desc_hid.h> //USB Configuration and Device descriptors for this UBS device
#include <usb.c> //handles usb setup tokens and get descriptor reports
void main(void)
{
int8 out_data[20];
int8 in_data[2];
int16 send_timer=0;
usb_init();
while(TRUE)
{
usb_task();
if (usb_enumerated())
{
if(send_timer == 0)
{
send_timer=1000;
out_data[0]=0xAB;
out_data[1]=0xCD;
out_data[2]=0xEF;
out_data[3]=0xAB;
out_data[4]=0xCD;
out_data[5]=0xEF;
usb_put_packet(1, out_data, 6, USB_DTS_TOGGLE);
}
if (usb_kbhit(1))
{
usb_get_packet(1, in_data, 1);
}
send_timer=send_timer-1;
delay_ms(1);
}
}
}
|
|
|
 |
Darren Rook
Joined: 06 Sep 2003
Posts: 287
Location: Milwaukee, WI
|
|
| Posted: Wed Dec 28, 2005 3:47 pm |
|
|
Did you change the HID descriptor? Also check the max endpoint packet size (in the endpoint descriptors)
_________________
I came, I saw, I compiled. |
|
|
starkeeper
Guest
|
|
| Posted: Thu Dec 29, 2005 2:50 am |
|
|
| Darren Rook wrote: | | Did you change the HID descriptor? Also check the max endpoint packet size (in the endpoint descriptors) |
Well I'm using the standard HID-descriptor from the example:
| Code: |
#IFNDEF __USB_DESCRIPTORS__
#DEFINE __USB_DESCRIPTORS__
#include <usb.h>
const char USB_CLASS_SPECIFIC_DESC[] = {
6, 0, 255, // Usage Page = Vendor Defined
9, 1, // Usage = IO device
0xa1, 1, // Collection = Application
0x19, 1, // Usage minimum
0x29, 8, // Usage maximum
0x15, 0x80, // Logical minimum (-128)
0x25, 0x7F, // Logical maximum (127)
0x75, 8, // Report size = 8 (bits)
0x95, 2, // Report count = 16 bits (2 bytes)
0x81, 2, // Input (Data, Var, Abs)
0x19, 1, // Usage minimum
0x29, 8, // Usage maximum
0x75, 8, // Report size = 8 (bits)
0x95, 2, // Report count = 16 bits (2 bytes)
0x91, 2, // Output (Data, Var, Abs)
0xc0 // End Collection
};
const int16 USB_CLASS_SPECIFIC_DESC_LOOKUP[USB_NUM_CONFIGURATIONS][1] =
{
//config 1
//interface 0
0
};
const int16 USB_CLASS_SPECIFIC_DESC_LOOKUP_SIZE[USB_NUM_CONFIGURATIONS][1] =
{
//config 1
//interface 0
32
};
#DEFINE USB_TOTAL_CONFIG_LEN 41 //config+interface+class+endpoint+endpoint (2 endpoints)
const char USB_CONFIG_DESC[] = {
//IN ORDER TO COMPLY WITH WINDOWS HOSTS, THE ORDER OF THIS ARRAY MUST BE:
// config(s)
// interface(s)
// class(es)
// endpoint(s)
//config_descriptor for config index 1
USB_DESC_CONFIG_LEN, //length of descriptor size ==1
USB_DESC_CONFIG_TYPE, //constant CONFIGURATION (CONFIGURATION 0x02) ==2
USB_TOTAL_CONFIG_LEN,0, //size of all data returned for this config ==3,4
1, //number of interfaces this device supports ==5
0x01, //identifier for this configuration. (IF we had more than one configurations) ==6
0x00, //index of string descriptor for this configuration ==7
0xC0, //bit 6=1 if self powered, bit 5=1 if supports remote wakeup (we don't), bits 0-4 unused and bit7=1 ==8
0x32, //maximum bus power required (maximum milliamperes/2) (0x32 = 100mA)
//interface descriptor 1
USB_DESC_INTERFACE_LEN, //length of descriptor =10
USB_DESC_INTERFACE_TYPE, //constant INTERFACE (INTERFACE 0x04) =11
0x00, //number defining this interface (IF we had more than one interface) ==12
0x00, //alternate setting ==13
2, //number of endpoins, except 0 (pic167xx has 3, but we dont have to use all). ==14
0x03, //class code, 03 = HID ==15
0x00, //subclass code //boot ==16
0x00, //protocol code ==17
0x00, //index of string descriptor for interface ==18
//class descriptor 1 (HID)
USB_DESC_CLASS_LEN, //length of descriptor ==19
USB_DESC_CLASS_TYPE, //dscriptor type (0x21 == HID) ==20
0x00,0x01, //hid class release number (1.0) (try 1.10) ==21,22
0x00, //localized country code (0 = none) ==23
0x01, //number of hid class descrptors that follow (1) ==24
0x22, //report descriptor type (0x22 == HID) ==25
USB_CLASS_SPECIFIC_DESC_LOOKUP_SIZE[0][0], 0x00, //length of report descriptor ==26,27
//endpoint descriptor
USB_DESC_ENDPOINT_LEN, //length of descriptor ==28
USB_DESC_ENDPOINT_TYPE, //constant ENDPOINT (ENDPOINT 0x05) ==29
0x81, //endpoint number and direction (0x81 = EP1 IN) ==30
0x03, //transfer type supported (0x03 is interrupt) ==31
USB_EP1_TX_SIZE,0x00, //maximum packet size supported ==32,33
#if USB_USE_FULL_SPEED
1, //polling interval, in ms. (cant be smaller than 10) ==34
#else
10, //polling interval, in ms. (cant be smaller than 10) ==34
#endif
//endpoint descriptor
USB_DESC_ENDPOINT_LEN, //length of descriptor ==35
USB_DESC_ENDPOINT_TYPE, //constant ENDPOINT (ENDPOINT 0x05) ==36
0x01, //endpoint number and direction (0x01 = EP1 OUT) ==37
0x03, //transfer type supported (0x03 is interrupt) ==38
USB_EP1_RX_SIZE,0x00, //maximum packet size supported ==39,40
#if USB_USE_FULL_SPEED
1
#else
10 //polling interval, in ms. (cant be smaller than 10) ==41
#endif
};
#define USB_NUM_HID_INTERFACES 1
#define USB_MAX_NUM_INTERFACES 1
const char USB_NUM_INTERFACES[USB_NUM_CONFIGURATIONS]={1};
const int16 USB_CLASS_DESCRIPTORS[USB_NUM_CONFIGURATIONS][1][1]=
{
//config 1
//interface 0
//class 1
18
};
#if (sizeof(USB_CONFIG_DESC) != USB_TOTAL_CONFIG_LEN)
#error USB_TOTAL_CONFIG_LEN not defined correctly
#endif
const char USB_DEVICE_DESC[USB_DESC_DEVICE_LEN] ={
//starts of with device configuration. only one possible
USB_DESC_DEVICE_LEN, //the length of this report ==1
0x01, //the constant DEVICE (DEVICE 0x01) ==2
0x10,0x01, //usb version in bcd (pic167xx is 1.1) ==3,4
0x00, //class code ==5
0x00, //subclass code ==6
0x00, //protocol code ==7
USB_MAX_EP0_PACKET_LENGTH, //max packet size for endpoint 0. (SLOW SPEED SPECIFIES 8) ==8
0x61,0x04, //vendor id (0x04D8 is Microchip, or is it 0x0461 ??)
0x20,0x00, //product id ==11,12 //don't use ffff says usb-by-example guy. oops
0x00,0x01, //device release number ==13,14
0x01, //index of string description of manufacturer. therefore we point to string_1 array (see below) ==15
0x02, //index of string descriptor of the product ==16
0x00, //index of string descriptor of serial number ==17
USB_NUM_CONFIGURATIONS //number of possible configurations ==18
};
char USB_STRING_DESC_OFFSET[]={0,4,12};
char const USB_STRING_DESC[]={
//string 0
4, //length of string index
USB_DESC_STRING_TYPE, //descriptor type 0x03 (STRING)
0x09,0x04, //Microsoft Defined for US-English
//string 1
8, //length of string index
USB_DESC_STRING_TYPE, //descriptor type 0x03 (STRING)
'C',0,
'C',0,
'S',0,
//string 2
26, //length of string index
USB_DESC_STRING_TYPE, //descriptor type 0x03 (STRING)
'C',0,
'C',0,
'S',0,
' ',0,
'H',0,
'I',0,
'D',0,
' ',0,
'D',0,
'e',0,
'm',0,
'o',0
};
#ENDIF
|
I thought that the packet size would be defined by theese canstants:
#define USB_EP1_TX_SIZE 8
#define USB_EP1_RX_SIZE 8
I have set them to 8 and I only want to send 6 Bytes.
What is wrong with that HID-Descriptor? |
|
|
Darren Rook
Joined: 06 Sep 2003
Posts: 287
Location: Milwaukee, WI
|
|
| Posted: Thu Dec 29, 2005 8:17 am |
|
|
What's wrong with the HID descriptor is that it's configured to send 2 byte report counts.
_________________
I came, I saw, I compiled. |
|
|
Guest
|
|
| Posted: Thu Dec 29, 2005 2:27 pm |
|
|
| Darren Rook wrote: | | What's wrong with the HID descriptor is that it's configured to send 2 byte report counts. |
| Code: |
const char USB_CLASS_SPECIFIC_DESC[] = {
6, 0, 255, // Usage Page = Vendor Defined
9, 1, // Usage = IO device
0xa1, 1, // Collection = Application
0x19, 1, // Usage minimum
0x29, 8, // Usage maximum
0x15, 0x80, // Logical minimum (-128)
0x25, 0x7F, // Logical maximum (127)
0x75, 8, // Report size = 8 (bits)
0x95, 6, // Report count = 16 bits (2 bytes)
0x81, 2, // Input (Data, Var, Abs)
0x19, 1, // Usage minimum
0x29, 8, // Usage maximum
0x75, 8, // Report size = 8 (bits)
0x95, 6, // Report count = 16 bits (2 bytes)
0x91, 2, // Output (Data, Var, Abs)
0xc0 // End Collection
}; |
I have set the report count to 6, now I can send 6Bytes. But the only way to see this is thru my USB-Sniffer SnoopyPro.
Any Software I tested says that my device has timeouts and needs to be reset. What else could be wrong? |
|
|
Darren Rook
Joined: 06 Sep 2003
Posts: 287
Location: Milwaukee, WI
|
|
| Posted: Thu Dec 29, 2005 8:19 pm |
|
|
Interesting.
You may want to change the output portion of the descriptor to specify 1 byte report count.
Did you change the endpoint descriptor's max endpoint size, or are they still using the constants USB_EP1_TX_SIZE and USB_EP1_RX_SIZE?
Also I believe on the PC side you have to ask for a 7 byte packet because the PC will append a byte to the beginning of the message which is the report ID. You are not using report IDs so your report IDs will always be 0.
This software, is this custom software you wrote?
_________________
I came, I saw, I compiled. |
|
|
Guest
|
|
| Posted: Fri Dec 30, 2005 3:32 am |
|
|
| Darren Rook wrote: | Interesting.
You may want to change the output portion of the descriptor to specify 1 byte report count.
Did you change the endpoint descriptor's max endpoint size, or are they still using the constants USB_EP1_TX_SIZE and USB_EP1_RX_SIZE?
Also I believe on the PC side you have to ask for a 7 byte packet because the PC will append a byte to the beginning of the message which is the report ID. You are not using report IDs so your report IDs will always be 0.
This software, is this custom software you wrote? |
Wow, that's it!
I had to set the packetsize on the PC side to 7. I don't want to know what the hell the USB-Developers thought by this.
Another problem is that I loose packets. I don't know if this is normal or even volitional. My Device sends answers to my PC, sometimes the PC is not fast enough to catch all packets.
For Instance the PIC sends all 5 seconds a packet. But If the PC requests a packet after 10 seconds, one packet is lost.
How can I prevent this phenomenon? Is this a PIC or a PC side problem? |
|
|
Darren Rook
Joined: 06 Sep 2003
Posts: 287
Location: Milwaukee, WI
|
|
| Posted: Sun Jan 01, 2006 10:31 pm |
|
|
| Anonymous wrote: |
Wow, that's it!
I had to set the packetsize on the PC side to 7. I don't want to know what the hell the USB-Developers thought by this.
|
It's actually not as strange as you think, you're just new to HID. If you look at ex_usb_kbmouse2.c it has an example of using multiple HID report IDs to send multiple packets of varying types.
| Quote: | Another problem is that I loose packets. I don't know if this is normal or even volitional. My Device sends answers to my PC, sometimes the PC is not fast enough to catch all packets.
For Instance the PIC sends all 5 seconds a packet. But If the PC requests a packet after 10 seconds, one packet is lost.
How can I prevent this phenomenon? Is this a PIC or a PC side problem? |
I have heard of this problem, but never heard of a solution (people too lazy to post their experiences after they've found a fix, I guess). I'm thinking it's because people are using ReadFile() and ReadFile() doesn't buffer, it only holds the latest packet.
_________________
I came, I saw, I compiled. |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
Powered by phpBB © 2001, 2005 phpBB Group
|
FAQ 
Search
Register
Profile 
Log in to check your private messages 
Log in
