aboutsummaryrefslogtreecommitdiffstats
path: root/Documentation/fb
diff options
context:
space:
mode:
authorKrzysztof Helt <krzysztof.h1@wp.pl>2009-03-31 18:25:40 -0400
committerLinus Torvalds <torvalds@linux-foundation.org>2009-04-01 11:59:33 -0400
commitddb53d48da5b0e691f35e703ac29118747f86c99 (patch)
treebdf0f4a6f3e74a87e5b2ccb9530f708eed6258d2 /Documentation/fb
parentec549a0fdc32171b26677f1ef0b5309faa743362 (diff)
fbdev: remove cyblafb driver
A tridentfb driver has all the functionality of the cyblafb driver without the bugs of the latter. Changes to the tridentfb driver: - FBINFO_READS_FAST added to the tridentfb. The cyblafb used a blitter for scrolling which is faster than color expansion on Cyberblade chipsets. The blitter is slower on a discrete Blade3D core. Use the blitter for scrolling in the tridentfb only for integrated Blade3D cores. Now, scrolling speed is about equal for the tridentfb and the cyblafb. - a copyright notice addition is done on request of Jani Monoses (the first author of the tridentfb). Tested on AGP Blade3D card and PCChips M787CLR motherboard: VIA C3 cpu + VT8601 north bridge (aka Cyberblade/i1). Signed-off-by: Krzysztof Helt <krzysztof.h1@wp.pl> Cc: "Jani Monoses" <jani@ubuntu.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'Documentation/fb')
-rw-r--r--Documentation/fb/00-INDEX2
-rw-r--r--Documentation/fb/cyblafb/bugs13
-rw-r--r--Documentation/fb/cyblafb/credits7
-rw-r--r--Documentation/fb/cyblafb/documentation17
-rw-r--r--Documentation/fb/cyblafb/fb.modes154
-rw-r--r--Documentation/fb/cyblafb/performance79
-rw-r--r--Documentation/fb/cyblafb/todo31
-rw-r--r--Documentation/fb/cyblafb/usage217
-rw-r--r--Documentation/fb/cyblafb/whatsnew29
-rw-r--r--Documentation/fb/cyblafb/whycyblafb85
10 files changed, 0 insertions, 634 deletions
diff --git a/Documentation/fb/00-INDEX b/Documentation/fb/00-INDEX
index caabbd395e61..a618fd99c9f0 100644
--- a/Documentation/fb/00-INDEX
+++ b/Documentation/fb/00-INDEX
@@ -11,8 +11,6 @@ aty128fb.txt
11 - info on the ATI Rage128 frame buffer driver. 11 - info on the ATI Rage128 frame buffer driver.
12cirrusfb.txt 12cirrusfb.txt
13 - info on the driver for Cirrus Logic chipsets. 13 - info on the driver for Cirrus Logic chipsets.
14cyblafb/
15 - directory with documentation files related to the cyblafb driver.
16deferred_io.txt 14deferred_io.txt
17 - an introduction to deferred IO. 15 - an introduction to deferred IO.
18fbcon.txt 16fbcon.txt
diff --git a/Documentation/fb/cyblafb/bugs b/Documentation/fb/cyblafb/bugs
deleted file mode 100644
index 9443a6d72cdd..000000000000
--- a/Documentation/fb/cyblafb/bugs
+++ /dev/null
@@ -1,13 +0,0 @@
1Bugs
2====
3
4I currently don't know of any bug. Please do send reports to:
5 - linux-fbdev-devel@lists.sourceforge.net
6 - Knut_Petersen@t-online.de.
7
8
9Untested features
10=================
11
12All LCD stuff is untested. If it worked in tridentfb, it should work in
13cyblafb. Please test and report the results to Knut_Petersen@t-online.de.
diff --git a/Documentation/fb/cyblafb/credits b/Documentation/fb/cyblafb/credits
deleted file mode 100644
index 0eb3b443dc2b..000000000000
--- a/Documentation/fb/cyblafb/credits
+++ /dev/null
@@ -1,7 +0,0 @@
1Thanks to
2=========
3 * Alan Hourihane, for writing the X trident driver
4 * Jani Monoses, for writing the tridentfb driver
5 * Antonino A. Daplas, for review of the first published
6 version of cyblafb and some code
7 * Jochen Hein, for testing and a helpfull bug report
diff --git a/Documentation/fb/cyblafb/documentation b/Documentation/fb/cyblafb/documentation
deleted file mode 100644
index bb1aac048425..000000000000
--- a/Documentation/fb/cyblafb/documentation
+++ /dev/null
@@ -1,17 +0,0 @@
1Available Documentation
2=======================
3
4Apollo PLE 133 Chipset VT8601A North Bridge Datasheet, Rev. 1.82, October 22,
52001, available from VIA:
6
7 http://www.viavpsd.com/product/6/15/DS8601A182.pdf
8
9The datasheet is incomplete, some registers that need to be programmed are not
10explained at all and important bits are listed as "reserved". But you really
11need the datasheet to understand the code. "p. xxx" comments refer to page
12numbers of this document.
13
14XFree/XOrg drivers are available and of good quality, looking at the code
15there is a good idea if the datasheet does not provide enough information
16or if the datasheet seems to be wrong.
17
diff --git a/Documentation/fb/cyblafb/fb.modes b/Documentation/fb/cyblafb/fb.modes
deleted file mode 100644
index fe0e5223ba86..000000000000
--- a/Documentation/fb/cyblafb/fb.modes
+++ /dev/null
@@ -1,154 +0,0 @@
1#
2# Sample fb.modes file
3#
4# Provides an incomplete list of working modes for
5# the cyberblade/i1 graphics core.
6#
7# The value 4294967256 is used instead of -40. Of course, -40 is not
8# a really reasonable value, but chip design does not always follow
9# logic. Believe me, it's ok, and it's the way the BIOS does it.
10#
11# fbset requires 4294967256 in fb.modes and -40 as an argument to
12# the -t parameter. That's also not too reasonable, and it might change
13# in the future or might even be differt for your current version.
14#
15
16mode "640x480-50"
17 geometry 640 480 2048 4096 8
18 timings 47619 4294967256 24 17 0 216 3
19endmode
20
21mode "640x480-60"
22 geometry 640 480 2048 4096 8
23 timings 39682 4294967256 24 17 0 216 3
24endmode
25
26mode "640x480-70"
27 geometry 640 480 2048 4096 8
28 timings 34013 4294967256 24 17 0 216 3
29endmode
30
31mode "640x480-72"
32 geometry 640 480 2048 4096 8
33 timings 33068 4294967256 24 17 0 216 3
34endmode
35
36mode "640x480-75"
37 geometry 640 480 2048 4096 8
38 timings 31746 4294967256 24 17 0 216 3
39endmode
40
41mode "640x480-80"
42 geometry 640 480 2048 4096 8
43 timings 29761 4294967256 24 17 0 216 3
44endmode
45
46mode "640x480-85"
47 geometry 640 480 2048 4096 8
48 timings 28011 4294967256 24 17 0 216 3
49endmode
50
51mode "800x600-50"
52 geometry 800 600 2048 4096 8
53 timings 30303 96 24 14 0 136 11
54endmode
55
56mode "800x600-60"
57 geometry 800 600 2048 4096 8
58 timings 25252 96 24 14 0 136 11
59endmode
60
61mode "800x600-70"
62 geometry 800 600 2048 4096 8
63 timings 21645 96 24 14 0 136 11
64endmode
65
66mode "800x600-72"
67 geometry 800 600 2048 4096 8
68 timings 21043 96 24 14 0 136 11
69endmode
70
71mode "800x600-75"
72 geometry 800 600 2048 4096 8
73 timings 20202 96 24 14 0 136 11
74endmode
75
76mode "800x600-80"
77 geometry 800 600 2048 4096 8
78 timings 18939 96 24 14 0 136 11
79endmode
80
81mode "800x600-85"
82 geometry 800 600 2048 4096 8
83 timings 17825 96 24 14 0 136 11
84endmode
85
86mode "1024x768-50"
87 geometry 1024 768 2048 4096 8
88 timings 19054 144 24 29 0 120 3
89endmode
90
91mode "1024x768-60"
92 geometry 1024 768 2048 4096 8
93 timings 15880 144 24 29 0 120 3
94endmode
95
96mode "1024x768-70"
97 geometry 1024 768 2048 4096 8
98 timings 13610 144 24 29 0 120 3
99endmode
100
101mode "1024x768-72"
102 geometry 1024 768 2048 4096 8
103 timings 13232 144 24 29 0 120 3
104endmode
105
106mode "1024x768-75"
107 geometry 1024 768 2048 4096 8
108 timings 12703 144 24 29 0 120 3
109endmode
110
111mode "1024x768-80"
112 geometry 1024 768 2048 4096 8
113 timings 11910 144 24 29 0 120 3
114endmode
115
116mode "1024x768-85"
117 geometry 1024 768 2048 4096 8
118 timings 11209 144 24 29 0 120 3
119endmode
120
121mode "1280x1024-50"
122 geometry 1280 1024 2048 4096 8
123 timings 11114 232 16 39 0 160 3
124endmode
125
126mode "1280x1024-60"
127 geometry 1280 1024 2048 4096 8
128 timings 9262 232 16 39 0 160 3
129endmode
130
131mode "1280x1024-70"
132 geometry 1280 1024 2048 4096 8
133 timings 7939 232 16 39 0 160 3
134endmode
135
136mode "1280x1024-72"
137 geometry 1280 1024 2048 4096 8
138 timings 7719 232 16 39 0 160 3
139endmode
140
141mode "1280x1024-75"
142 geometry 1280 1024 2048 4096 8
143 timings 7410 232 16 39 0 160 3
144endmode
145
146mode "1280x1024-80"
147 geometry 1280 1024 2048 4096 8
148 timings 6946 232 16 39 0 160 3
149endmode
150
151mode "1280x1024-85"
152 geometry 1280 1024 2048 4096 8
153 timings 6538 232 16 39 0 160 3
154endmode
diff --git a/Documentation/fb/cyblafb/performance b/Documentation/fb/cyblafb/performance
deleted file mode 100644
index 8d15d5dfc6b3..000000000000
--- a/Documentation/fb/cyblafb/performance
+++ /dev/null
@@ -1,79 +0,0 @@
1Speed
2=====
3
4CyBlaFB is much faster than tridentfb and vesafb. Compare the performance data
5for mode 1280x1024-[8,16,32]@61 Hz.
6
7Test 1: Cat a file with 2000 lines of 0 characters.
8Test 2: Cat a file with 2000 lines of 80 characters.
9Test 3: Cat a file with 2000 lines of 160 characters.
10
11All values show system time use in seconds, kernel 2.6.12 was used for
12the measurements. 2.6.13 is a bit slower, 2.6.14 hopefully will include a
13patch that speeds up kernel bitblitting a lot ( > 20%).
14
15+-----------+-----------------------------------------------------+
16| | not accelerated |
17| TRIDENTFB +-----------------+-----------------+-----------------+
18| of 2.6.12 | 8 bpp | 16 bpp | 32 bpp |
19| | noypan | ypan | noypan | ypan | noypan | ypan |
20+-----------+--------+--------+--------+--------+--------+--------+
21| Test 1 | 4.31 | 4.33 | 6.05 | 12.81 | ---- | ---- |
22| Test 2 | 67.94 | 5.44 | 123.16 | 14.79 | ---- | ---- |
23| Test 3 | 131.36 | 6.55 | 240.12 | 16.76 | ---- | ---- |
24+-----------+--------+--------+--------+--------+--------+--------+
25| Comments | | | completely bro- |
26| | | | ken, monitor |
27| | | | switches off |
28+-----------+-----------------+-----------------+-----------------+
29
30
31+-----------+-----------------------------------------------------+
32| | accelerated |
33| TRIDENTFB +-----------------+-----------------+-----------------+
34| of 2.6.12 | 8 bpp | 16 bpp | 32 bpp |
35| | noypan | ypan | noypan | ypan | noypan | ypan |
36+-----------+--------+--------+--------+--------+--------+--------+
37| Test 1 | ---- | ---- | 20.62 | 1.22 | ---- | ---- |
38| Test 2 | ---- | ---- | 22.61 | 3.19 | ---- | ---- |
39| Test 3 | ---- | ---- | 24.59 | 5.16 | ---- | ---- |
40+-----------+--------+--------+--------+--------+--------+--------+
41| Comments | broken, writing | broken, ok only | completely bro- |
42| | to wrong places | if bgcolor is | ken, monitor |
43| | on screen + bug | black, bug in | switches off |
44| | in fillrect() | fillrect() | |
45+-----------+-----------------+-----------------+-----------------+
46
47
48+-----------+-----------------------------------------------------+
49| | not accelerated |
50| VESAFB +-----------------+-----------------+-----------------+
51| of 2.6.12 | 8 bpp | 16 bpp | 32 bpp |
52| | noypan | ypan | noypan | ypan | noypan | ypan |
53+-----------+--------+--------+--------+--------+--------+--------+
54| Test 1 | 4.26 | 3.76 | 5.99 | 7.23 | ---- | ---- |
55| Test 2 | 65.65 | 4.89 | 120.88 | 9.08 | ---- | ---- |
56| Test 3 | 126.91 | 5.94 | 235.77 | 11.03 | ---- | ---- |
57+-----------+--------+--------+--------+--------+--------+--------+
58| Comments | vga=0x307 | vga=0x31a | vga=0x31b not |
59| | fh=80kHz | fh=80kHz | supported by |
60| | fv=75kHz | fv=75kHz | video BIOS and |
61| | | | hardware |
62+-----------+-----------------+-----------------+-----------------+
63
64
65+-----------+-----------------------------------------------------+
66| | accelerated |
67| CYBLAFB +-----------------+-----------------+-----------------+
68| | 8 bpp | 16 bpp | 32 bpp |
69| | noypan | ypan | noypan | ypan | noypan | ypan |
70+-----------+--------+--------+--------+--------+--------+--------+
71| Test 1 | 8.02 | 0.23 | 19.04 | 0.61 | 57.12 | 2.74 |
72| Test 2 | 8.38 | 0.55 | 19.39 | 0.92 | 57.54 | 3.13 |
73| Test 3 | 8.73 | 0.86 | 19.74 | 1.24 | 57.95 | 3.51 |
74+-----------+--------+--------+--------+--------+--------+--------+
75| Comments | | | |
76| | | | |
77| | | | |
78| | | | |
79+-----------+-----------------+-----------------+-----------------+
diff --git a/Documentation/fb/cyblafb/todo b/Documentation/fb/cyblafb/todo
deleted file mode 100644
index c5f6d0eae545..000000000000
--- a/Documentation/fb/cyblafb/todo
+++ /dev/null
@@ -1,31 +0,0 @@
1TODO / Missing features
2=======================
3
4Verify LCD stuff "stretch" and "center" options are
5 completely untested ... this code needs to be
6 verified. As I don't have access to such
7 hardware, please contact me if you are
8 willing run some tests.
9
10Interlaced video modes The reason that interleaved
11 modes are disabled is that I do not know
12 the meaning of the vertical interlace
13 parameter. Also the datasheet mentions a
14 bit d8 of a horizontal interlace parameter,
15 but nowhere the lower 8 bits. Please help
16 if you can.
17
18low-res double scan modes Who needs it?
19
20accelerated color blitting Who needs it? The console driver does use color
21 blitting for nothing but drawing the penguine,
22 everything else is done using color expanding
23 blitting of 1bpp character bitmaps.
24
25ioctls Who needs it?
26
27TV-out Will be done later. Use "vga= " at boot time
28 to set a suitable video mode.
29
30??? Feel free to contact me if you have any
31 feature requests
diff --git a/Documentation/fb/cyblafb/usage b/Documentation/fb/cyblafb/usage
deleted file mode 100644
index a39bb3d402a2..000000000000
--- a/Documentation/fb/cyblafb/usage
+++ /dev/null
@@ -1,217 +0,0 @@
1CyBlaFB is a framebuffer driver for the Cyberblade/i1 graphics core integrated
2into the VIA Apollo PLE133 (aka vt8601) south bridge. It is developed and
3tested using a VIA EPIA 5000 board.
4
5Cyblafb - compiled into the kernel or as a module?
6==================================================
7
8You might compile cyblafb either as a module or compile it permanently into the
9kernel.
10
11Unless you have a real reason to do so you should not compile both vesafb and
12cyblafb permanently into the kernel. It's possible and it helps during the
13developement cycle, but it's useless and will at least block some otherwise
14usefull memory for ordinary users.
15
16Selecting Modes
17===============
18
19 Startup Mode
20 ============
21
22 First of all, you might use the "vga=???" boot parameter as it is
23 documented in vesafb.txt and svga.txt. Cyblafb will detect the video
24 mode selected and will use the geometry and timings found by
25 inspecting the hardware registers.
26
27 video=cyblafb vga=0x317
28
29 Alternatively you might use a combination of the mode, ref and bpp
30 parameters. If you compiled the driver into the kernel, add something
31 like this to the kernel command line:
32
33 video=cyblafb:1280x1024,bpp=16,ref=50 ...
34
35 If you compiled the driver as a module, the same mode would be
36 selected by the following command:
37
38 modprobe cyblafb mode=1280x1024 bpp=16 ref=50 ...
39
40 None of the modes possible to select as startup modes are affected by
41 the problems described at the end of the next subsection.
42
43 For all startup modes cyblafb chooses a virtual x resolution of 2048,
44 the only exception is mode 1280x1024 in combination with 32 bpp. This
45 allows ywrap scrolling for all those modes if rotation is 0 or 2, and
46 also fast scrolling if rotation is 1 or 3. The default virtual y reso-
47 lution is 4096 for bpp == 8, 2048 for bpp==16 and 1024 for bpp == 32,
48 again with the only exception of 1280x1024 at 32 bpp.
49
50 Please do set your video memory size to 8 Mb in the Bios setup. Other
51 values will work, but performace is decreased for a lot of modes.
52
53 Mode changes using fbset
54 ========================
55
56 You might use fbset to change the video mode, see "man fbset". Cyblafb
57 generally does assume that you know what you are doing. But it does
58 some checks, especially those that are needed to prevent you from
59 damaging your hardware.
60
61 - only 8, 16, 24 and 32 bpp video modes are accepted
62 - interlaced video modes are not accepted
63 - double scan video modes are not accepted
64 - if a flat panel is found, cyblafb does not allow you
65 to program a resolution higher than the physical
66 resolution of the flat panel monitor
67 - cyblafb does not allow vclk to exceed 230 MHz. As 32 bpp
68 and (currently) 24 bit modes use a doubled vclk internally,
69 the dotclock limit as seen by fbset is 115 MHz for those
70 modes and 230 MHz for 8 and 16 bpp modes.
71 - cyblafb will allow you to select very high resolutions as
72 long as the hardware can be programmed to these modes. The
73 documented limit 1600x1200 is not enforced, but don't expect
74 perfect signal quality.
75
76 Any request that violates the rules given above will be either changed
77 to something the hardware supports or an error value will be returned.
78
79 If you program a virtual y resolution higher than the hardware limit,
80 cyblafb will silently decrease that value to the highest possible
81 value. The same is true for a virtual x resolution that is not
82 supported by the hardware. Cyblafb tries to adapt vyres first because
83 vxres decides if ywrap scrolling is possible or not.
84
85 Attempts to disable acceleration are ignored, I believe that this is
86 safe.
87
88 Some video modes that should work do not work as expected. If you use
89 the standard fb.modes, fbset 640x480-60 will program that mode, but
90 you will see a vertical area, about two characters wide, with only
91 much darker characters than the other characters on the screen.
92 Cyblafb does allow that mode to be set, as it does not violate the
93 official specifications. It would need a lot of code to reliably sort
94 out all invalid modes, playing around with the margin values will
95 give a valid mode quickly. And if cyblafb would detect such an invalid
96 mode, should it silently alter the requested values or should it
97 report an error? Both options have some pros and cons. As stated
98 above, none of the startup modes are affected, and if you set
99 verbosity to 1 or higher, cyblafb will print the fbset command that
100 would be needed to program that mode using fbset.
101
102
103Other Parameters
104================
105
106
107crt don't autodetect, assume monitor connected to
108 standard VGA connector
109
110fp don't autodetect, assume flat panel display
111 connected to flat panel monitor interface
112
113nativex inform driver about native x resolution of
114 flat panel monitor connected to special
115 interface (should be autodetected)
116
117stretch stretch image to adapt low resolution modes to
118 higer resolutions of flat panel monitors
119 connected to special interface
120
121center center image to adapt low resolution modes to
122 higer resolutions of flat panel monitors
123 connected to special interface
124
125memsize use if autodetected memsize is wrong ...
126 should never be necessary
127
128nopcirr disable PCI read retry
129nopciwr disable PCI write retry
130nopcirb disable PCI read bursts
131nopciwb disable PCI write bursts
132
133bpp bpp for specified modes
134 valid values: 8 || 16 || 24 || 32
135
136ref refresh rate for specified mode
137 valid values: 50 <= ref <= 85
138
139mode 640x480 or 800x600 or 1024x768 or 1280x1024
140 if not specified, the startup mode will be detected
141 and used, so you might also use the vga=??? parameter
142 described in vesafb.txt. If you do not specify a mode,
143 bpp and ref parameters are ignored.
144
145verbosity 0 is the default, increase to at least 2 for every
146 bug report!
147
148Development hints
149=================
150
151It's much faster do compile a module and to load the new version after
152unloading the old module than to compile a new kernel and to reboot. So if you
153try to work on cyblafb, it might be a good idea to use cyblafb as a module.
154In real life, fast often means dangerous, and that's also the case here. If
155you introduce a serious bug when cyblafb is compiled into the kernel, the
156kernel will lock or oops with a high probability before the file system is
157mounted, and the danger for your data is low. If you load a broken own version
158of cyblafb on a running system, the danger for the integrity of the file
159system is much higher as you might need a hard reset afterwards. Decide
160yourself.
161
162Module unloading, the vfb method
163================================
164
165If you want to unload/reload cyblafb using the virtual framebuffer, you need
166to enable vfb support in the kernel first. After that, load the modules as
167shown below:
168
169 modprobe vfb vfb_enable=1
170 modprobe fbcon
171 modprobe cyblafb
172 fbset -fb /dev/fb1 1280x1024-60 -vyres 2662
173 con2fb /dev/fb1 /dev/tty1
174 ...
175
176If you now made some changes to cyblafb and want to reload it, you might do it
177as show below:
178
179 con2fb /dev/fb0 /dev/tty1
180 ...
181 rmmod cyblafb
182 modprobe cyblafb
183 con2fb /dev/fb1 /dev/tty1
184 ...
185
186Of course, you might choose another mode, and most certainly you also want to
187map some other /dev/tty* to the real framebuffer device. You might also choose
188to compile fbcon as a kernel module or place it permanently in the kernel.
189
190I do not know of any way to unload fbcon, and fbcon will prevent the
191framebuffer device loaded first from unloading. [If there is a way, then
192please add a description here!]
193
194Module unloading, the vesafb method
195===================================
196
197Configure the kernel:
198
199 <*> Support for frame buffer devices
200 [*] VESA VGA graphics support
201 <M> Cyberblade/i1 support
202
203Add e.g. "video=vesafb:ypan vga=0x307" to the kernel parameters. The ypan
204parameter is important, choose any vga parameter you like as long as it is
205a graphics mode.
206
207After booting, load cyblafb without any mode and bpp parameter and assign
208cyblafb to individual ttys using con2fb, e.g.:
209
210 modprobe cyblafb
211 con2fb /dev/fb1 /dev/tty1
212
213Unloading cyblafb works without problems after you assign vesafb to all
214ttys again, e.g.:
215
216 con2fb /dev/fb0 /dev/tty1
217 rmmod cyblafb
diff --git a/Documentation/fb/cyblafb/whatsnew b/Documentation/fb/cyblafb/whatsnew
deleted file mode 100644
index 76c07a26e044..000000000000
--- a/Documentation/fb/cyblafb/whatsnew
+++ /dev/null
@@ -1,29 +0,0 @@
10.62
2====
3
4 - the vesafb parameter has been removed as I decided to allow the
5 feature without any special parameter.
6
7 - Cyblafb does not use the vga style of panning any longer, now the
8 "right view" register in the graphics engine IO space is used. Without
9 that change it was impossible to use all available memory, and without
10 access to all available memory it is impossible to ywrap.
11
12 - The imageblit function now uses hardware acceleration for all font
13 widths. Hardware blitting across pixel column 2048 is broken in the
14 cyberblade/i1 graphics core, but we work around that hardware bug.
15
16 - modes with vxres != xres are supported now.
17
18 - ywrap scrolling is supported now and the default. This is a big
19 performance gain.
20
21 - default video modes use vyres > yres and vxres > xres to allow
22 almost optimal scrolling speed for normal and rotated screens
23
24 - some features mainly usefull for debugging the upper layers of the
25 framebuffer system have been added, have a look at the code
26
27 - fixed: Oops after unloading cyblafb when reading /proc/io*
28
29 - we work around some bugs of the higher framebuffer layers.
diff --git a/Documentation/fb/cyblafb/whycyblafb b/Documentation/fb/cyblafb/whycyblafb
deleted file mode 100644
index a123bc11e698..000000000000
--- a/Documentation/fb/cyblafb/whycyblafb
+++ /dev/null
@@ -1,85 +0,0 @@
1I tried the following framebuffer drivers:
2
3 - TRIDENTFB is full of bugs. Acceleration is broken for Blade3D
4 graphics cores like the cyberblade/i1. It claims to support a great
5 number of devices, but documentation for most of these devices is
6 unfortunately not available. There is _no_ reason to use tridentfb
7 for cyberblade/i1 + CRT users. VESAFB is faster, and the one
8 advantage, mode switching, is broken in tridentfb.
9
10 - VESAFB is used by many distributions as a standard. Vesafb does
11 not support mode switching. VESAFB is a bit faster than the working
12 configurations of TRIDENTFB, but it is still too slow, even if you
13 use ypan.
14
15 - EPIAFB (you'll find it on sourceforge) supports the Cyberblade/i1
16 graphics core, but it still has serious bugs and developement seems
17 to have stopped. This is the one driver with TV-out support. If you
18 do need this feature, try epiafb.
19
20None of these drivers was a real option for me.
21
22I believe that is unreasonable to change code that announces to support 20
23devices if I only have more or less sufficient documentation for exactly one
24of these. The risk of breaking device foo while fixing device bar is too high.
25
26So I decided to start CyBlaFB as a stripped down tridentfb.
27
28All code specific to other Trident chips has been removed. After that there
29were a lot of cosmetic changes to increase the readability of the code. All
30register names were changed to those mnemonics used in the datasheet. Function
31and macro names were changed if they hindered easy understanding of the code.
32
33After that I debugged the code and implemented some new features. I'll try to
34give a little summary of the main changes:
35
36 - calculation of vertical and horizontal timings was fixed
37
38 - video signal quality has been improved dramatically
39
40 - acceleration:
41
42 - fillrect and copyarea were fixed and reenabled
43
44 - color expanding imageblit was newly implemented, color
45 imageblit (only used to draw the penguine) still uses the
46 generic code.
47
48 - init of the acceleration engine was improved and moved to a
49 place where it really works ...
50
51 - sync function has a timeout now and tries to reset and
52 reinit the accel engine if necessary
53
54 - fewer slow copyarea calls when doing ypan scrolling by using
55 undocumented bit d21 of screen start address stored in
56 CR2B[5]. BIOS does use it also, so this should be safe.
57
58 - cyblafb rejects any attempt to set modes that would cause vclk
59 values above reasonable 230 MHz. 32bit modes use a clock
60 multiplicator of 2, so fbset does show the correct values for
61 pixclock but not for vclk in this case. The fbset limit is 115 MHz
62 for 32 bpp modes.
63
64 - cyblafb rejects modes known to be broken or unimplemented (all
65 interlaced modes, all doublescan modes for now)
66
67 - cyblafb now works independant of the video mode in effect at startup
68 time (tridentfb does not init all needed registers to reasonable
69 values)
70
71 - switching between video modes does work reliably now
72
73 - the first video mode now is the one selected on startup using the
74 vga=???? mechanism or any of
75 - 640x480, 800x600, 1024x768, 1280x1024
76 - 8, 16, 24 or 32 bpp
77 - refresh between 50 Hz and 85 Hz, 1 Hz steps (1280x1024-32
78 is limited to 63Hz)
79
80 - pci retry and pci burst mode are settable (try to disable if you
81 experience latency problems)
82
83 - built as a module cyblafb might be unloaded and reloaded using
84 the vfb module and con2vt or might be used together with vesafb
85