[PATCH] n_r3964 mod_timer() fix
Stephan Brodkorb [Mon, 17 Oct 2005 23:43:30 +0000 (16:43 -0700)]
Since Revision 1.10 was released the n_r3964 module wasn't able to receive any
data.  The reason for that behavior is because there were some wrong calls of
mod_timer(...) in the function receive_char (...).  This patch should fix this
problem and was successfully tested with talking to some kuka industrial
robots.

Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

drivers/char/n_r3964.c

index 97d6dc2..853c98c 100644 (file)
@@ -695,7 +695,7 @@ static void receive_char(struct r3964_info *pInfo, const unsigned char c)
             {
                TRACE_PE("IDLE - got STX but no space in rx_queue!");
                pInfo->state=R3964_WAIT_FOR_RX_BUF;
-              mod_timer(&pInfo->tmr, R3964_TO_NO_BUF);
+              mod_timer(&pInfo->tmr, jiffies + R3964_TO_NO_BUF);
                break;
             }
 start_receiving:
@@ -705,7 +705,7 @@ start_receiving:
             pInfo->last_rx = 0;
             pInfo->flags &= ~R3964_ERROR;
             pInfo->state=R3964_RECEIVING;
-           mod_timer(&pInfo->tmr, R3964_TO_ZVZ);
+           mod_timer(&pInfo->tmr, jiffies + R3964_TO_ZVZ);
            pInfo->nRetry = 0;
             put_char(pInfo, DLE);
             flush(pInfo);
@@ -732,7 +732,7 @@ start_receiving:
                if(pInfo->flags & R3964_BCC)
                {
                   pInfo->state = R3964_WAIT_FOR_BCC;
-                 mod_timer(&pInfo->tmr, R3964_TO_ZVZ);
+                 mod_timer(&pInfo->tmr, jiffies + R3964_TO_ZVZ);
                }
                else 
                {
@@ -744,7 +744,7 @@ start_receiving:
                pInfo->last_rx = c;
 char_to_buf:
                pInfo->rx_buf[pInfo->rx_position++] = c;
-              mod_timer(&pInfo->tmr, R3964_TO_ZVZ);
+              mod_timer(&pInfo->tmr, jiffies + R3964_TO_ZVZ);
             }
          }
         /* else: overflow-msg? BUF_SIZE>MTU; should not happen? */