mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Jiri Slaby <jslaby@suse.cz>
To: gregkh@linuxfoundation.org
Cc: alan@linux.intel.com, linux-serial@vger.kernel.org,
	linux-kernel@vger.kernel.org, jirislaby@gmail.com,
	Tony Luck <tony.luck@intel.com>,
	Fenghua Yu <fenghua.yu@intel.com>
Subject: [PATCH 46/68] TTY: simserial, remove some tty ops
Date: Mon,  5 Mar 2012 14:52:33 +0100	[thread overview]
Message-ID: <1330955575-26641-47-git-send-email-jslaby@suse.cz> (raw)
In-Reply-To: <1330955575-26641-1-git-send-email-jslaby@suse.cz>

All ->start, ->stop and ->wait_until_sent are empty and need not be
defined. The time to remove them is now.

Signed-off-by: Jiri Slaby <jslaby@suse.cz>
Cc: Tony Luck <tony.luck@intel.com>
Cc: Fenghua Yu <fenghua.yu@intel.com>
---
 arch/ia64/hp/sim/simserial.c |   39 +--------------------------------------
 1 file changed, 1 insertion(+), 38 deletions(-)

diff --git a/arch/ia64/hp/sim/simserial.c b/arch/ia64/hp/sim/simserial.c
index 909357e..d173dba 100644
--- a/arch/ia64/hp/sim/simserial.c
+++ b/arch/ia64/hp/sim/simserial.c
@@ -64,32 +64,7 @@ static struct console *console;
 
 extern struct console *console_drivers; /* from kernel/printk.c */
 
-/*
- * ------------------------------------------------------------
- * rs_stop() and rs_start()
- *
- * This routines are called before setting or resetting tty->stopped.
- * They enable or disable transmitter interrupts, as necessary.
- * ------------------------------------------------------------
- */
-static void rs_stop(struct tty_struct *tty)
-{
-#ifdef SIMSERIAL_DEBUG
-	printk("rs_stop: tty->stopped=%d tty->hw_stopped=%d tty->flow_stopped=%d\n",
-		tty->stopped, tty->hw_stopped, tty->flow_stopped);
-#endif
-
-}
-
-static void rs_start(struct tty_struct *tty)
-{
-#ifdef SIMSERIAL_DEBUG
-	printk("rs_start: tty->stopped=%d tty->hw_stopped=%d tty->flow_stopped=%d\n",
-		tty->stopped, tty->hw_stopped, tty->flow_stopped);
-#endif
-}
-
-static  void receive_chars(struct tty_struct *tty)
+static void receive_chars(struct tty_struct *tty)
 {
 	unsigned char ch;
 	static unsigned char seen_esc = 0;
@@ -406,7 +381,6 @@ static void rs_set_termios(struct tty_struct *tty, struct ktermios *old_termios)
 	if ((old_termios->c_cflag & CRTSCTS) &&
 	    !(tty->termios->c_cflag & CRTSCTS)) {
 		tty->hw_stopped = 0;
-		rs_start(tty);
 	}
 }
 /*
@@ -514,14 +488,6 @@ static void rs_close(struct tty_struct *tty, struct file * filp)
 }
 
 /*
- * rs_wait_until_sent() --- wait until the transmitter is empty
- */
-static void rs_wait_until_sent(struct tty_struct *tty, int timeout)
-{
-}
-
-
-/*
  * rs_hangup() --- called by tty_hangup() when a hangup is signaled.
  */
 static void rs_hangup(struct tty_struct *tty)
@@ -736,10 +702,7 @@ static const struct tty_operations hp_ops = {
 	.unthrottle = rs_unthrottle,
 	.send_xchar = rs_send_xchar,
 	.set_termios = rs_set_termios,
-	.stop = rs_stop,
-	.start = rs_start,
 	.hangup = rs_hangup,
-	.wait_until_sent = rs_wait_until_sent,
 	.proc_fops = &rs_proc_fops,
 };
 
-- 
1.7.9.2



  parent reply	other threads:[~2012-03-05 14:07 UTC|newest]

Thread overview: 84+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-03-05 13:51 [PATCH 00/68] TTY buffer in tty_port -- prep no. 1 Jiri Slaby
2012-03-05 13:51 ` [PATCH 01/68] USB: cdc-acm, use tty_standard_install Jiri Slaby
2012-03-05 14:03   ` Jiri Slaby
2012-03-05 14:14     ` Oliver Neukum
2012-03-05 23:39   ` Alan Cox
2012-03-05 13:51 ` [PATCH 02/68] TTY: tty_io, remove buffer re-assignments Jiri Slaby
2012-03-05 13:51 ` [PATCH 03/68] TTY: let alloc_tty_driver deduce the owner automatically Jiri Slaby
2012-03-05 13:51 ` [PATCH 04/68] TTY: remove minor_num from tty_driver Jiri Slaby
2012-03-05 13:51 ` [PATCH 05/68] TTY: remove re-assignments to tty_driver members Jiri Slaby
2012-03-06 23:49   ` Tilman Schmidt
2012-03-05 13:51 ` [PATCH 06/68] TTY: simplify tty_driver_lookup_tty a bit Jiri Slaby
2012-03-05 13:51 ` [PATCH 07/68] TTY: remove tty driver re-set from tty_reopen Jiri Slaby
2012-03-05 13:51 ` [PATCH 08/68] TTY: serial, simplify ASYNC_USR_MASK Jiri Slaby
2012-03-05 13:51 ` [PATCH 09/68] TTY: tty_driver, document tty->ops->shutdown limitation Jiri Slaby
2012-03-05 13:51 ` [PATCH 10/68] ALPHA: srmcons, use timer functions Jiri Slaby
2012-03-05 13:51 ` [PATCH 11/68] ALPHA: srmcons, fix racy singleton structure Jiri Slaby
2012-03-05 13:51 ` [PATCH 12/68] TTY: srmcons, convert to use tty_port Jiri Slaby
2012-03-05 13:52 ` [PATCH 13/68] TTY: serialP, remove DECLARE_WAITQUEUE check Jiri Slaby
2012-03-05 13:52 ` [PATCH 14/68] TTY: remove unneeded tty->index checks Jiri Slaby
2012-03-05 13:52 ` [PATCH 15/68] TTY: ipwireless, fix tty->index handling Jiri Slaby
2012-03-05 14:03   ` Jiri Kosina
2012-03-05 13:52 ` [PATCH 16/68] NET: pc300, do not zero global variables Jiri Slaby
2012-03-05 13:52 ` [PATCH 17/68] NET: pc300, show version info from module init Jiri Slaby
2012-03-05 13:52 ` [PATCH 18/68] XTENSA: iss/console, use setup_timer Jiri Slaby
2012-03-05 13:52 ` [PATCH 19/68] XTENSA: iss/console, fix potential deadlock Jiri Slaby
2012-03-05 13:52 ` [PATCH 20/68] TTY: iss/console, use tty_port Jiri Slaby
2012-03-05 13:52 ` [PATCH 21/68] TTY: serial, use atomic_inc_return in ioc4_serial Jiri Slaby
2012-03-05 13:52 ` [PATCH 22/68] TTY: serial, include pci.h in m32r_sio Jiri Slaby
2012-03-05 13:52 ` [PATCH 23/68] TTY: remove serialP.h inclusion from some files Jiri Slaby
2012-03-05 13:52 ` [PATCH 24/68] TTY: speakup, do not use serialP Jiri Slaby
2012-03-05 13:52 ` [PATCH 25/68] TTY: serialP, remove unused material Jiri Slaby
2012-03-05 13:52 ` [PATCH 26/68] TTY: amiserial, remove tasklet for tty_wakeup Jiri Slaby
2012-03-05 13:52 ` [PATCH 27/68] TTY: amiserial, use only one copy of async flags Jiri Slaby
2012-03-05 13:52 ` [PATCH 28/68] TTY: simserial, " Jiri Slaby
2012-03-05 13:52 ` [PATCH 29/68] TTY: simserial/amiserial, use one instance of other members Jiri Slaby
2012-03-05 13:52 ` [PATCH 30/68] TTY: simserial, remove support of shared interrupts Jiri Slaby
2012-03-05 13:52 ` [PATCH 31/68] TTY: simserial, remove IRQ_T Jiri Slaby
2012-03-05 13:52 ` [PATCH 32/68] TTY: amiserial, remove IRQ_ports Jiri Slaby
2012-03-05 13:52 ` [PATCH 33/68] TTY: serialP, merge serial_state and async_struct Jiri Slaby
2012-03-05 13:52 ` [PATCH 34/68] TTY: amiserial, simplify set_serial_info Jiri Slaby
2012-03-05 13:52 ` [PATCH 35/68] TTY: amiserial, pass tty down to functions Jiri Slaby
2012-03-05 13:52 ` [PATCH 36/68] TTY: simserial, " Jiri Slaby
2012-03-05 13:52 ` [PATCH 37/68] TTY: amiserial/simserial, use tty_port Jiri Slaby
2012-03-05 13:52 ` [PATCH 38/68] TTY: amiserial/simserial, use close delays from tty_port Jiri Slaby
2012-03-05 13:52 ` [PATCH 39/68] TTY: amiserial/simserial, use count " Jiri Slaby
2012-03-05 13:52 ` [PATCH 40/68] TTY: amiserial/simserial, use flags " Jiri Slaby
2012-03-05 13:52 ` [PATCH 41/68] TTY: simserial, remove static initialization Jiri Slaby
2012-03-05 13:52 ` [PATCH 42/68] TTY: simserial, remove tmp_buf Jiri Slaby
2012-03-05 13:52 ` [PATCH 43/68] TTY: simserial, stop using serial_state->{line,icount} Jiri Slaby
2012-03-05 13:52 ` [PATCH 44/68] TTY: simserial no longer needs serialP Jiri Slaby
2012-03-05 13:52 ` [PATCH 45/68] TTY: simserial, define local tty_port pointer Jiri Slaby
2012-03-05 13:52 ` Jiri Slaby [this message]
2012-03-05 13:52 ` [PATCH 48/68] TTY: simserial, use tty_port_close_start Jiri Slaby
2012-03-05 13:52 ` [PATCH 49/68] TTY: simserial, properly refcount tty_port->tty Jiri Slaby
2012-03-05 13:52 ` [PATCH 50/68] TTY: simserial, use tty_port_open Jiri Slaby
2012-03-05 13:52 ` [PATCH 51/68] TTY: simserial, use tty_port_hangup Jiri Slaby
2012-03-05 13:52 ` [PATCH 52/68] TTY: simserial, remove useless comments Jiri Slaby
2012-03-05 13:52 ` [PATCH 53/68] TTY: simserial, fix includes Jiri Slaby
2012-03-05 13:52 ` [PATCH 54/68] TTY: simserial, reindent some code Jiri Slaby
2012-03-05 13:52 ` [PATCH 55/68] TTY: simserial, final cleanup Jiri Slaby
2012-03-05 13:52 ` [PATCH 56/68] TTY: amiserial, define local tty_port pointer Jiri Slaby
2012-03-05 13:52 ` [PATCH 57/68] TTY: amiserial, stop using serial_state->{irq,type,line} Jiri Slaby
2012-03-05 13:52 ` [PATCH 58/68] TTY: amiserial no longer needs serialP Jiri Slaby
2012-03-05 13:52 ` [PATCH 59/68] TTY: amiserial, provide carrier helpers Jiri Slaby
2012-03-05 13:52 ` [PATCH 60/68] TTY: amiserial, use tty_port_block_til_ready Jiri Slaby
2012-03-05 13:52 ` [PATCH 61/68] TTY: amiserial, use tty_port_close_end Jiri Slaby
2012-03-05 13:52 ` [PATCH 62/68] TTY: amiserial, use tty_port_close_start Jiri Slaby
2012-03-05 13:52 ` [PATCH 63/68] TTY: pdc_cons, fix racy tty test Jiri Slaby
2012-03-05 13:52 ` [PATCH 64/68] TTY: pdc_cons, fix open vs timer race Jiri Slaby
2012-03-05 13:52 ` [PATCH 65/68] TTY: pdc_cons, fix open vs pdc_console_tty_driver race Jiri Slaby
2012-03-05 13:52 ` [PATCH 66/68] TTY: pdc_cons, use tty_port Jiri Slaby
2012-03-05 13:52 ` [PATCH 67/68] TTY: isdn/gigaset, do not set tty->driver_data to NULL Jiri Slaby
2012-03-05 17:05   ` Tilman Schmidt
2012-03-05 17:12     ` Jiri Slaby
2012-03-05 13:52 ` [PATCH 68/68] TTY: isdn/gigaset, use tty_port Jiri Slaby
2012-03-05 17:08   ` Tilman Schmidt
2012-03-08 19:50 ` [PATCH 00/68] TTY buffer in tty_port -- prep no. 1 Greg KH
2012-03-08 20:01   ` [PATCH 1/4] [IA64] hpsim, fix SAL handling in fw-emu Jiri Slaby
2012-03-08 20:01     ` [PATCH 2/4] [IA64] simserial, include some headers Jiri Slaby
2012-03-08 20:01     ` [PATCH 3/4] [IA64] hpsim, initialize chip for assigned irqs Jiri Slaby
2012-03-08 20:01     ` [PATCH 4/4] [IA64] simserial, bail out when request_irq fails Jiri Slaby
2012-03-08 20:29     ` [PATCH 1/4] [IA64] hpsim, fix SAL handling in fw-emu Greg KH
2012-03-08 20:51 ` [PATCH 00/68] TTY buffer in tty_port -- prep no. 1 Greg KH
2012-03-18  8:56 ` Geert Uytterhoeven

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1330955575-26641-47-git-send-email-jslaby@suse.cz \
    --to=jslaby@suse.cz \
    --cc=alan@linux.intel.com \
    --cc=fenghua.yu@intel.com \
    --cc=gregkh@linuxfoundation.org \
    --cc=jirislaby@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-serial@vger.kernel.org \
    --cc=tony.luck@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®