PATCH: Export console functions for use by Software Suspend nice display

Hi.

Here's a second patch; this exports gotoxy, reset_terminal, hide_cursor, getconsxy and putconsxy for use in Software Suspend's nice display.

diff -ruN linux-2.6.1/drivers/char/vt.c software-suspend-linux-2.6.1-rev3/drivers/char/vt.c
@@ -149,13 +149,13 @@
 static void vc_init(unsigned int console, unsigned int rows,
 		    unsigned int cols, int do_clear);
 static void blank_screen(unsigned long dummy);
-static void gotoxy(int currcons, int new_x, int new_y);
+void gotoxy(int currcons, int new_x, int new_y);
 static void save_cur(int currcons);
-static void reset_terminal(int currcons, int do_clear);
+void reset_terminal(int currcons, int do_clear);
 static void con_flush_chars(struct tty_struct *tty);
 static void set_vesa_blanking(unsigned long arg);
 static void set_cursor(int currcons);
-static void hide_cursor(int currcons);
+void hide_cursor(int currcons);
 static void unblank_screen_t(unsigned long dummy);
 static void console_callback(void *ignored);
 

@@ -530,7 +530,7 @@
 		sw->con_putc(vc_cons[currcons].d, i, y, x);
 }
 
-static void hide_cursor(int currcons)
+void hide_cursor(int currcons)
 {
 	if (currcons == sel_cons)
 		clear_selection();

@@ -859,7 +859,7 @@
  * might also be negative. If the given position is out of
  * bounds, the cursor is placed at the nearest margin.
  */
-static void gotoxy(int currcons, int new_x, int new_y)
+void gotoxy(int currcons, int new_x, int new_y)
 {
 	int min_y, max_y;
 

@@ -1403,7 +1403,7 @@
 	ESpalette };
 
 /* console_sem is held (except via vc_init()) */
-static void reset_terminal(int currcons, int do_clear)
+void reset_terminal(int currcons, int do_clear)
 {
 	top		= 0;
 	bottom		= video_num_lines;

@@ -2996,13 +2996,13 @@
 	return screenpos(currcons, 2 * w_offset, viewed);
 }
 
-void getconsxy(int currcons, char *p)
+void getconsxy(int currcons, unsigned char *p)
 {
 	p[0] = x;
 	p[1] = y;
 }
 
-void putconsxy(int currcons, char *p)
+void putconsxy(int currcons, unsigned char *p)
 {
 	gotoxy(currcons, p[0], p[1]);
 	set_cursor(currcons);

My work on Software Suspend is graciously brought to you by
LinuxFund.org.

Why don't you open /dev/console on rootfs and use write(2)?

Really, swsusp shouldn't mess with console internals. And you don't even explain what "nice display" is supposed to mean.

Hi.

'Nice display' means that it gives a simple, clean display with progress bar etc. Al(?) Viro has informed me of the write call. I'll look it up and see how I can improve things. As I said to Mr Viro, I'm just a user who wanted Suspend to work better and ended up just about rewriting the thing. Please forgive my ignorance!

Regards,

Nigel

My work on Software Suspend is graciously brought to you by
LinuxFund.org.

Hi.

I'm not sure that write is what I want. At the very least, it will make the code harder to read and maintain. Here's a small portion of what I'm currently doing:

/* Print version */
posn[0] = (unsigned char) (0);
posn[1] = (unsigned char) (video_num_lines); putconsxy(suspend_console, posn); cond_console_print(swsusp_version, strlen(swsusp_version));

/* Print header */
posn[0] = (unsigned char) ((video_num_columns - 29) / 2); posn[1] = (unsigned char) ((video_num_lines / 3) -4); putconsxy(suspend_console, posn);

The output looks something like this:
-----

S U S P E N D T O D I S K

Writing caches...
[-------- 120/640MB ]

2.0-rc4
-----

Bootsplash is also supported, so there's an even nicer version :>

Regards,

Nigel

My work on Software Suspend is graciously brought to you by
LinuxFund.org.

Sure it is. Write some wrapper functions that blat out the ANSI cursor control sequences, just like anyone who does this from userspace without using libcurses.

Daniel Jacobowitz
MontaVista Software                         Debian GNU/Linux Developer

Also, by using write instead of blasting the low level code, you will not have to worry about locking issues. (The way you tap the low level stuff should require at least the console semaphore held, write to /dev/console don't)

Ben

Hi.

Locking is not an issue. This is suspend-to-disk. Everything else is stopped while we're working.

By the way, am I understanding the suggestion correctly? Do you (collective) mean getting a fd for /dev/console from within kernel code and using that? I've been looking at the way printk works and wondering if con->write is equivalent (once I find the right console to write to)?

Regards,

Nigel

My work on Software Suspend is graciously brought to you by
LinuxFund.org.

No. You can still get a printk from irq...

Yes. get an fd and write to it. grep for write to find other uses :)

You may have to be a bit careful about what context you are in, but I suppose it's whatever userland process triggered the sleep in the first place, no ? For load, it's probably whatever process did swapon ?

In both cases, it should be fine.

Ben.

Sorry to reply to myself.

I don't think I worded that last message right... I didn't mean that locking isn't an issue at all, just that we don't have to worry about another process writing to the console while we are.

Nigel

My work on Software Suspend is graciously brought to you by
LinuxFund.org.

Hi.

Mmm. But we also adjust the console loglevel depending upon the verbosity of debugging info required. (And reset it when we're done). When the nice display is on, the loglevel is 0 or 1. If a driver wants to printk at KERN_EMERG or KERN_ALERT then, well the display should get messed up a little :>

I did. Didn't find much. I'll look again. Perhaps I'm just blind :>

No. There's a kernel daemon - kswsuspd - that does all the hard work. The userland process that initiates the suspend sleeps in wait_for_completion until post resume. At boot time, we initiate the resume via an init call. Still fine?

Regards,

Nigel
.

My work on Software Suspend is graciously brought to you by
LinuxFund.org.