Potential fix for 0-length usb packets seen on OSX

[proxmark3-svn] / armsrc / lfops.c

diff --git a/armsrc/lfops.c b/armsrc/lfops.c
index 188d7280f7deab98d763dc8198a180e0a2ed8c19..733bc953ed04a8dd85129c87f9af6d027f8bfcd4 100644 (file)
--- a/armsrc/lfops.c
+++ b/armsrc/lfops.c
@@ -399,14 +399,10 @@ void SimulateTagLowFrequency(int period, int gap, int ledcontrol)
  #define OPEN_COIL()           HIGH(GPIO_SSC_DOUT)
 
        i = 0;
-       byte_t rx[sizeof(UsbCommand)]; // Storage for usb_read call in loop
        for(;;) {
                //wait until SSC_CLK goes HIGH
                while(!(AT91C_BASE_PIOA->PIO_PDSR & GPIO_SSC_CLK)) {
-                       // Craig Young - Adding a usb_read() here to avoid abort on empty UsbCommand
-                       // My OS X client does this preventing simulation.
-                       // Performance hit should be non-existent since the read is only performed if usb_poll is true
-                       if(BUTTON_PRESS() || (usb_poll() && usb_read(rx,sizeof(UsbCommand)))) {
+                       if(BUTTON_PRESS() || (usb_poll_validate_length() )) {
                                DbpString("Stopped");
                                return;
                        }