svn commit: r255031 - head/sys/dev/uart

[Marcel Moolenaar]

Author: marcel
Date: Thu Aug 29 16:26:04 2013
New Revision: 255031
URL: http://svnweb.freebsd.org/changeset/base/255031

Log:
  Work-around a timing problem with the ITE IT8513E now that the core
  calls ns8250_bus_ipend() almost immediately after ns8250_bus_attach().
  As it appears, a line break condition is being signalled for almost
  all received characters due to this. A delay of 150ms seems enough
  to allow the H/W to settle and to avoid the problem.
  More analysis is needed, but for now a regression has been addressed.
  
  Reported by: kevlo@
  Tested by: kevlo@

Modified:
  head/sys/dev/uart/uart_dev_ns8250.c

Modified: head/sys/dev/uart/uart_dev_ns8250.c
==============================================================================
--- head/sys/dev/uart/uart_dev_ns8250.c	Thu Aug 29 15:59:05 2013	(r255030)
+++ head/sys/dev/uart/uart_dev_ns8250.c	Thu Aug 29 16:26:04 2013	(r255031)
@@ -453,7 +453,19 @@ ns8250_bus_attach(struct uart_softc *sc)
 	ns8250->ier |= ns8250->ier_rxbits;
 	uart_setreg(bas, REG_IER, ns8250->ier);
 	uart_barrier(bas);
-	
+
+	/*
+	 * Timing of the H/W access was changed with r253161 of uart_core.c
+	 * It has been observed that an ITE IT8513E would signal a break
+	 * condition with pretty much every character it received, unless
+	 * it had enough time to settle between ns8250_bus_attach() and
+	 * ns8250_bus_ipend() -- which it accidentally had before r253161.
+	 * It's not understood why the UART chip behaves this way and it
+	 * could very well be that the DELAY make the H/W work in the same
+	 * accidental manner as before. More analysis is warranted, but
+	 * at least now we fixed a known regression.
+	 */
+	DELAY(150);
 	return (0);
 }