Received: from mnm [127.0.0.1]
by localhost with POP3 (fetchmail-5.9.0)
for akpm@localhost (single-drop); Wed, 06 Aug 2003 10:25:57 -0700 (PDT)
Received: by mangalore (mbox akpm)
(with Cubic Circle's cucipop (v1.31 1998/05/13) Thu Aug 7 03:24:09 2003)
X-From_: linux-kernel-owner+akpm=40zip.com.au@vger.kernel.org Thu Aug 7 03:22:00 2003
Return-Path: <linux-kernel-owner+akpm=40zip.com.au@vger.kernel.org>
Received: from vger.kernel.org (vger.kernel.org [67.72.78.212])
by mangalore.zipworld.com.au (8.12.3/8.12.3/Debian-6.4) with ESMTP id h76HLvLx021476
for <akpm@zip.com.au>; Thu, 7 Aug 2003 03:21:59 +1000
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
id S270804AbTHFRTo (ORCPT <rfc822;akpm@zip.com.au>);
Wed, 6 Aug 2003 13:19:44 -0400
Received: (majordomo@vger.kernel.org) by vger.kernel.org id S270832AbTHFRTo
(ORCPT <rfc822;linux-kernel-outgoing>);
Wed, 6 Aug 2003 13:19:44 -0400
Received: from vana.vc.cvut.cz ([147.32.240.58]:18311 "EHLO vana.vc.cvut.cz")
by vger.kernel.org with ESMTP id S270804AbTHFRT1 (ORCPT
<rfc822;linux-kernel@vger.kernel.org>);
Wed, 6 Aug 2003 13:19:27 -0400
Received: from vana.vc.cvut.cz (smmsp@localhost [127.0.0.1])
by vana.vc.cvut.cz (8.12.9/8.12.9/Debian-5) with ESMTP id h76HJMHb005114;
Wed, 6 Aug 2003 19:19:22 +0200
Received: (from root@localhost)
by vana.vc.cvut.cz (8.12.9/8.12.9/Debian-5) id h76HJLrW005106;
Wed, 6 Aug 2003 19:19:21 +0200
Date: Wed, 6 Aug 2003 19:19:21 +0200
From: Petr Vandrovec <vandrove@vc.cvut.cz>
To: torvalds@osdl.org
Cc: linux-kernel@vger.kernel.org
Subject: [RESEND, first time as BKPATCH] matroxfb updates
Message-ID: <20030806171921.GA4892@vana.vc.cvut.cz>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.5.4i
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk
X-Mailing-List: linux-kernel@vger.kernel.org
X-Spam-Status: No, hits=-15.8 required=6.0
tests=BAYES_01,PATCH_UNIFIED_DIFF,USER_AGENT_MUTT,X_MAILING_LIST
autolearn=ham version=2.53
X-Spam-Level:
X-Spam-Checker-Version: SpamAssassin 2.53 (1.174.2.15-2003-03-30-exp)
Linus, please do a
bk pull http://ncpfs.bkbits.net/matroxfb
I hope that this format is more acceptable to you than patches - and if nothing
else, it will make further resends simpler...
Thanks,
Petr Vandrovec
This will update the following files:
drivers/video/matrox/matroxfb_DAC1064.c | 2
drivers/video/matrox/matroxfb_Ti3026.c | 1
drivers/video/matrox/matroxfb_base.c | 67 +++++++++++++++++++++-----------
drivers/video/matrox/matroxfb_base.h | 2
drivers/video/matrox/matroxfb_crtc2.c | 7 ---
drivers/video/matrox/matroxfb_crtc2.h | 1
6 files changed, 51 insertions(+), 29 deletions(-)
through these ChangeSets:
<vandrove@vc.cvut.cz> (03/08/06 1.1122)
Add support for panning at vertical blanking to the matroxfb. Now mplayer output
looks much better on primary output (secondary output is always synced with vbl).
<vandrove@vc.cvut.cz> (03/08/06 1.1121)
matroxfb: Initialize fbcon's cmap.
<vandrove@vc.cvut.cz> (03/08/06 1.1120)
Remove write-only palette variable from matroxfb. Now it is possible to build
matroxfb without fbcon support.
# This is a BitKeeper generated patch for the following project:
# Project Name: Linux kernel tree
# This patch format is intended for GNU patch command version 2.5 or higher.
# This patch includes the following deltas:
# ChangeSet 1.1119 -> 1.1122
# drivers/video/matrox/matroxfb_Ti3026.c 1.7 -> 1.8
# drivers/video/matrox/matroxfb_crtc2.h 1.4 -> 1.5
# drivers/video/matrox/matroxfb_DAC1064.c 1.15 -> 1.16
# drivers/video/matrox/matroxfb_base.h 1.22 -> 1.24
# drivers/video/matrox/matroxfb_base.c 1.39 -> 1.42
# drivers/video/matrox/matroxfb_crtc2.c 1.24 -> 1.27
#
# The following is the BitKeeper ChangeSet Log
# --------------------------------------------
# 03/08/06 vandrove@vc.cvut.cz 1.1120
# Remove write-only palette variable from matroxfb. Now it is possible to build
# matroxfb without fbcon support.
# --------------------------------------------
# 03/08/06 vandrove@vc.cvut.cz 1.1121
# matroxfb: Initialize fbcon's cmap.
# --------------------------------------------
# 03/08/06 vandrove@vc.cvut.cz 1.1122
# Add support for panning at vertical blanking to the matroxfb. Now mplayer output
# looks much better on primary output (secondary output is always synced with vbl).
# --------------------------------------------
#
diff -Nru a/drivers/video/matrox/matroxfb_DAC1064.c b/drivers/video/matrox/matroxfb_DAC1064.c
--- a/drivers/video/matrox/matroxfb_DAC1064.c Wed Aug 6 19:16:11 2003
+++ b/drivers/video/matrox/matroxfb_DAC1064.c Wed Aug 6 19:16:11 2003
@@ -1036,6 +1036,7 @@
DAC1064_restore_1(PMINFO2);
matroxfb_vgaHWrestore(PMINFO2);
+ ACCESS_FBINFO(crtc1.panpos) = -1;
for (i = 0; i < 6; i++)
mga_setr(M_EXTVGA_INDEX, i, hw->CRTCEXT[i]);
DAC1064_restore_2(PMINFO2);
@@ -1062,6 +1063,7 @@
if (ACCESS_FBINFO(devflags.support32MB))
mga_setr(M_EXTVGA_INDEX, 8, hw->CRTCEXT[8]);
#endif
+ ACCESS_FBINFO(crtc1.panpos) = -1;
for (i = 0; i < 6; i++)
mga_setr(M_EXTVGA_INDEX, i, hw->CRTCEXT[i]);
DAC1064_restore_2(PMINFO2);
diff -Nru a/drivers/video/matrox/matroxfb_Ti3026.c b/drivers/video/matrox/matroxfb_Ti3026.c
--- a/drivers/video/matrox/matroxfb_Ti3026.c Wed Aug 6 19:16:11 2003
+++ b/drivers/video/matrox/matroxfb_Ti3026.c Wed Aug 6 19:16:11 2003
@@ -575,6 +575,7 @@
CRITBEGIN
+ ACCESS_FBINFO(crtc1.panpos) = -1;
for (i = 0; i < 6; i++)
mga_setr(M_EXTVGA_INDEX, i, hw->CRTCEXT[i]);
diff -Nru a/drivers/video/matrox/matroxfb_base.c b/drivers/video/matrox/matroxfb_base.c
--- a/drivers/video/matrox/matroxfb_base.c Wed Aug 6 19:16:11 2003
+++ b/drivers/video/matrox/matroxfb_base.c Wed Aug 6 19:16:11 2003
@@ -176,6 +176,27 @@
}
}
+static void matroxfb_crtc1_panpos(WPMINFO2) {
+ if (ACCESS_FBINFO(crtc1.panpos) >= 0) {
+ unsigned long flags;
+ int panpos;
+
+ matroxfb_DAC_lock_irqsave(flags);
+ panpos = ACCESS_FBINFO(crtc1.panpos);
+ if (panpos >= 0) {
+ unsigned int extvga_reg;
+
+ ACCESS_FBINFO(crtc1.panpos) = -1; /* No update pending anymore */
+ extvga_reg = mga_inb(M_EXTVGA_INDEX);
+ mga_setr(M_EXTVGA_INDEX, 0x00, panpos);
+ if (extvga_reg != 0x00) {
+ mga_outb(M_EXTVGA_INDEX, extvga_reg);
+ }
+ }
+ matroxfb_DAC_unlock_irqrestore(flags);
+ }
+}
+
static irqreturn_t matrox_irq(int irq, void *dev_id, struct pt_regs *fp)
{
u_int32_t status;
@@ -188,6 +209,7 @@
if (status & 0x20) {
mga_outl(M_ICLEAR, 0x20);
ACCESS_FBINFO(crtc1.vsync.cnt)++;
+ matroxfb_crtc1_panpos(PMINFO2);
wake_up_interruptible(&ACCESS_FBINFO(crtc1.vsync.wait));
handled = 1;
}
@@ -209,12 +231,13 @@
bm = 0x020;
if (!test_and_set_bit(0, &ACCESS_FBINFO(irq_flags))) {
- printk(KERN_DEBUG "matroxfb: enabling IRQ\n");
if (request_irq(ACCESS_FBINFO(pcidev)->irq, matrox_irq,
- SA_SHIRQ, "MGA Vertical Sync", MINFO)) {
+ SA_SHIRQ, "matroxfb", MINFO)) {
clear_bit(0, &ACCESS_FBINFO(irq_flags));
return -EINVAL;
}
+ /* Clear any pending field interrupts */
+ mga_outl(M_ICLEAR, bm);
mga_outl(M_IEN, mga_inl(M_IEN) | bm);
} else if (reenable) {
u_int32_t ien;
@@ -230,7 +253,8 @@
static void matroxfb_disable_irq(WPMINFO2) {
if (test_and_clear_bit(0, &ACCESS_FBINFO(irq_flags))) {
- printk(KERN_DEBUG "matroxfb: disabling IRQ\n");
+ /* Flush pending pan-at-vbl request... */
+ matroxfb_crtc1_panpos(PMINFO2);
if (ACCESS_FBINFO(devflags.accelerator) == FB_ACCEL_MATROX_MGAG400)
mga_outl(M_IEN, mga_inl(M_IEN) & ~0x220);
else
@@ -284,6 +308,9 @@
#ifdef CONFIG_FB_MATROX_32MB
unsigned int p3;
#endif
+ int vbl;
+ unsigned long flags;
+
CRITFLAGS
DBG(__FUNCTION__)
@@ -302,15 +329,26 @@
p3 = ACCESS_FBINFO(hw).CRTCEXT[8] = pos >> 21;
#endif
+ /* FB_ACTIVATE_VBL and we can acquire interrupts? Honor FB_ACTIVATE_VBL then... */
+ vbl = (var->activate & FB_ACTIVATE_VBL) && (matroxfb_enable_irq(PMINFO 0) == 0);
+
CRITBEGIN
+ matroxfb_DAC_lock_irqsave(flags);
mga_setr(M_CRTC_INDEX, 0x0D, p0);
mga_setr(M_CRTC_INDEX, 0x0C, p1);
#ifdef CONFIG_FB_MATROX_32MB
if (ACCESS_FBINFO(devflags.support32MB))
mga_setr(M_EXTVGA_INDEX, 0x08, p3);
#endif
- mga_setr(M_EXTVGA_INDEX, 0x00, p2);
+ if (vbl) {
+ ACCESS_FBINFO(crtc1.panpos) = p2;
+ } else {
+ /* Abort any pending change */
+ ACCESS_FBINFO(crtc1.panpos) = -1;
+ mga_setr(M_EXTVGA_INDEX, 0x00, p2);
+ }
+ matroxfb_DAC_unlock_irqrestore(flags);
update_crtc2(PMINFO pos);
@@ -627,11 +665,6 @@
if (regno >= ACCESS_FBINFO(curr.cmap_len))
return 1;
- ACCESS_FBINFO(palette[regno].red) = red;
- ACCESS_FBINFO(palette[regno].green) = green;
- ACCESS_FBINFO(palette[regno].blue) = blue;
- ACCESS_FBINFO(palette[regno].transp) = transp;
-
if (ACCESS_FBINFO(fbcon).var.grayscale) {
/* gray = 0.30*R + 0.59*G + 0.11*B */
red = green = blue = (red * 77 + green * 151 + blue * 28) >> 8;
@@ -748,19 +781,6 @@
else
ACCESS_FBINFO(curr.ydstorg.pixels) = (ydstorg * 8) / var->bits_per_pixel;
ACCESS_FBINFO(curr.final_bppShift) = matroxfb_get_final_bppShift(PMINFO var->bits_per_pixel);
- if (visual == MX_VISUAL_PSEUDOCOLOR) {
- int i;
-
- for (i = 0; i < 16; i++) {
- int j;
-
- j = color_table[i];
- ACCESS_FBINFO(palette[i].red) = default_red[j];
- ACCESS_FBINFO(palette[i].green) = default_grn[j];
- ACCESS_FBINFO(palette[i].blue) = default_blu[j];
- }
- }
-
{ struct my_timming mt;
struct matrox_hw_state* hw;
int out;
@@ -1698,6 +1718,7 @@
/* after __init time we are like module... no logo */
ACCESS_FBINFO(fbcon.flags) = hotplug ? FBINFO_FLAG_MODULE : FBINFO_FLAG_DEFAULT;
ACCESS_FBINFO(video.len_usable) &= PAGE_MASK;
+ fb_alloc_cmap(&ACCESS_FBINFO(fbcon.cmap), 256, 1);
#ifndef MODULE
/* mode database is marked __init!!! */
@@ -1990,6 +2011,7 @@
ACCESS_FBINFO(irq_flags) = 0;
init_waitqueue_head(&ACCESS_FBINFO(crtc1.vsync.wait));
init_waitqueue_head(&ACCESS_FBINFO(crtc2.vsync.wait));
+ ACCESS_FBINFO(crtc1.panpos) = -1;
err = initMatrox2(PMINFO b);
if (!err) {
@@ -2483,6 +2505,7 @@
EXPORT_SYMBOL(matroxfb_register_driver);
EXPORT_SYMBOL(matroxfb_unregister_driver);
EXPORT_SYMBOL(matroxfb_wait_for_sync);
+EXPORT_SYMBOL(matroxfb_enable_irq);
/*
* Overrides for Emacs so that we follow Linus's tabbing style.
diff -Nru a/drivers/video/matrox/matroxfb_base.h b/drivers/video/matrox/matroxfb_base.h
--- a/drivers/video/matrox/matroxfb_base.h Wed Aug 6 19:16:11 2003
+++ b/drivers/video/matrox/matroxfb_base.h Wed Aug 6 19:16:11 2003
@@ -457,6 +457,7 @@
struct matrox_vsync vsync;
unsigned int pixclock;
int mnp;
+ int panpos;
} crtc1;
struct {
struct matrox_vsync vsync;
@@ -595,7 +596,6 @@
dll:1;
} memory;
} values;
- struct { unsigned red, green, blue, transp; } palette[256];
u_int32_t cmap[17];
};
diff -Nru a/drivers/video/matrox/matroxfb_crtc2.c b/drivers/video/matrox/matroxfb_crtc2.c
--- a/drivers/video/matrox/matroxfb_crtc2.c Wed Aug 6 19:16:11 2003
+++ b/drivers/video/matrox/matroxfb_crtc2.c Wed Aug 6 19:16:11 2003
@@ -33,10 +33,6 @@
if (regno >= 16)
return 1;
- m2info->palette[regno].red = red;
- m2info->palette[regno].blue = blue;
- m2info->palette[regno].green = green;
- m2info->palette[regno].transp = transp;
if (m2info->fbcon.var.grayscale) {
/* gray = 0.30*R + 0.59*G + 0.11*B */
red = green = blue = (red * 77 + green * 151 + blue * 28) >> 8;
@@ -152,7 +148,7 @@
mga_outl(0x3C10, tmp);
ACCESS_FBINFO(hw).crtc2.ctl = tmp;
- tmp = 0x0FFF0000; /* line compare */
+ tmp = mt->VDisplay << 16; /* line compare */
if (mt->sync & FB_SYNC_HOR_HIGH_ACT)
tmp |= 0x00000100;
if (mt->sync & FB_SYNC_VERT_HIGH_ACT)
@@ -609,6 +605,7 @@
m2info->fbcon.flags = FBINFO_FLAG_DEFAULT;
m2info->fbcon.currcon = -1;
m2info->fbcon.pseudo_palette = m2info->cmap;
+ fb_alloc_cmap(&m2info->fbcon.cmap, 256, 1);
if (mem < 64)
mem *= 1024;
diff -Nru a/drivers/video/matrox/matroxfb_crtc2.h b/drivers/video/matrox/matroxfb_crtc2.h
--- a/drivers/video/matrox/matroxfb_crtc2.h Wed Aug 6 19:16:11 2003
+++ b/drivers/video/matrox/matroxfb_crtc2.h Wed Aug 6 19:16:11 2003
@@ -30,7 +30,6 @@
int interlaced:1;
u_int32_t cmap[17];
- struct { unsigned red, green, blue, transp; } palette[17];
};
#endif /* __MATROXFB_CRTC2_H__ */
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/