kajgan / fbterm Goto Github PK

AC_ARG_ENABLE(gpm, 
              AC_HELP_STRING([--disable-gpm], [disable gpm mouse support [[default=auto]]]), 
              [GPM="$enalbeval"],
              [GPM=auto])

AC_ARG_ENABLE(vesa, 
              AC_HELP_STRING([--disable-vesa], [disable VESA video card support [[default=auto]]]), 
              [VESA="$enalbeval"],
              [VESA=auto])

AC_ARG_ENABLE(epoll, 
              AC_HELP_STRING([--disable-epoll], [do not use epoll system call [[default=auto]]]), 
              [EPOLL="$enalbeval"],
              [EPOLL=auto])

AC_ARG_ENABLE(signalfd,
              AC_HELP_STRING([--disable-signalfd], [do not use signalfd system call [[default=auto]]]),
              [SIGNALFD="$enalbeval"],
              [SIGNALFD=auto])

Hi，我在打包 fbterm 的时候发现无法禁用 vesa 或者 gpm 支持，
以上代码是从 fbterm 1.7 的 configure.ac 中截取的， enalbeval 
应该是enableval 的笔误吧？

in src/lib/vterm.cpp

        if (mode_flags.insert_mode) {
                changed_line(cursor_y, cursor_x, width - 1);

                u16 step = dw ? 2 : 1;
#if 0
                for (u16 i = cursor_x; i < width - step; i++) {
                        text[yp + i + step] = text[yp + i];
                        attrs[yp + i + step] = attrs[yp + i];
                }
#else
                /* BUG FIXED */
                for (s32 i = width - step; i >= cursor_x; i--) {
                        text[yp + i + step] = text[yp + i];
                        attrs[yp + i + step] = attrs[yp + i];
                }
#endif
        } else {
                changed_line(cursor_y, cursor_x, cursor_x + (dw ? 1 : 0));
        }

environment: debian testing linux 2.6.32-trunk-686 fbterm 1.6 gpm 1.20.4

i have two types of consoles - fbcon from the kernel and the fbterm. fbcon is
the default text console, and fbterm graphical console. with gpm i'm able to
select text with mouse and paste between consoles. it works fine if the
copy-paste is between two fbcons or fbterms, but if i select some text in an
fbcon i can't paste it to an fbterm, and vice-versa.

What steps will reproduce the problem?
1. Launch a program other than the shell with fbterm (like fbterm -- screen)
2. The program will be launched using TERM=linux

What is the expected output? What do you see instead?
TERM should be set to fbterm.

What version of the product are you using? On what operating system?
1.7

Please provide any additional information below.
Line 83 of /src/lib/shell.cpp
-- setenv("TERM","linux",1)
++setenv("TERM","fbterm",1)

As a workaround, you can launch the "shell" program using /usr/bin/env 
TERM=fbterm program.

我在gentoo下安装fbterm失败：

make[3]: Leaving directory
`/var/tmp/portage/app-i18n/fbterm-1.3/work/fbterm-1.3/src/lib'
make[2]: Leaving directory
`/var/tmp/portage/app-i18n/fbterm-1.3/work/fbterm-1.3/src/lib'
make[2]: Entering directory
`/var/tmp/portage/app-i18n/fbterm-1.3/work/fbterm-1.3/src'
make[3]: Entering directory
`/var/tmp/portage/app-i18n/fbterm-1.3/work/fbterm-1.3/src'
test -z "/usr/bin" || /bin/mkdir -p
"/var/tmp/portage/app-i18n/fbterm-1.3/image//usr/bin"
make[3]: Nothing to be done for `install-data-am'.
  /usr/bin/install -c 'fbterm'
'/var/tmp/portage/app-i18n/fbterm-1.3/image//usr/bin/fbterm'
make  install-exec-hook
make[4]: Entering directory
`/var/tmp/portage/app-i18n/fbterm-1.3/work/fbterm-1.3/src'
setcap 'cap_sys_tty_config+ep'
/var/tmp/portage/app-i18n/fbterm-1.3/image//usr/bin/fbterm
Failed to set capabilities on file
`/var/tmp/portage/app-i18n/fbterm-1.3/image//usr/bin/fbterm' (Operation not
supported)
usage: setcap [-q] [-v] (-r|-|<caps>) <filename> [ ... (-r|-|<capsN>)
<filenameN> ]

 Note <filename> must be a regular (non-symlink) file.
make[4]: *** [install-exec-hook] Error 1
make[4]: Leaving directory
`/var/tmp/portage/app-i18n/fbterm-1.3/work/fbterm-1.3/src'
make[3]: *** [install-exec-am] Error 2

内核版本：
Linux polaris 2.6.27-gentoo-r5 #1 SMP PREEMPT Sat Dec 20 14:49:27 CST 2008
i686 Genuine Intel(R) CPU T2300 @ 1.66GHz GenuineIntel GNU/Linux

Hi,

*What steps will reproduce the problem?
1. Run fbterm with locale LANG=ja_JP.UTF-8

*What is the expected output? What do you see instead?
-Glyphs like '□', '◆', ..., known as 'ambiguous width characters', are 
drawn as single width.
-They should be double width in that locale.
-Other terminal apps have special option, like mlterm's '--ac', but fbterm 
doesn't.
-It is a highly necessary functionality for Japanese people.

*What version of the product are you using? On what operating system?
-Ubuntu 10.04

*Please provide any additional information below.
-I wrote a patch for new option '--ambiguous-wide'.
-It treats ambiguous width characters as wide width.
-See attached file.

I apologize if we already have another way to correct the problem.
Thank you.

tabata

What steps will reproduce the problem?
1. download fbterm-1.7.0.tar.gz and build as usual ( also make imexample )
2. cd top_srcdir of fbterm ( into fbterm-1.7/ )
3. ./src/fbterm -i ./im/imexample

What is the expected output? What do you see instead?

Two unknown characters ( seems to be [0x9d,0xb9] ) will be dumped to tty. 

Please use labels and text to provide additional information.

http://code.google.com/p/ucimf/issues/detail?id=77
http://code.google.com/p/ucimf/issues/detail?id=79

What steps will reproduce the problem?
1. start fbterm
2. start an application using a framebuffer inside the fbterm

What is the expected output? What do you see instead?

One would expect the application to be able to draw on the fb, just like when 
it's started from the plain Linux console. Instead it notices it's in a pts and 
thus doesn't dare opening the fb.

What version of the product are you using? On what operating system?

1.6, Debian

Please provide any additional information below.

To make the application dare to open the fb, fbterm should define FRAMEBUFFER 
to the device that it has itself opened, e.g. /dev/fb0

What steps will reproduce the problem?
1. ./configure
2. echo $?
1

What is the expected output? What do you see instead?
0

What version of the product are you using? On what operating system?
1.6 on GNU/Linux (Slackware 13.0)

Please provide any additional information below.
The './configure' step does not show any errors but just return '1'. And
then 'make' successes. Fbterm runs well.
Version 1.5 has no problem like this. "configure" returns '0'.

The attachment is 'config.log'. What makes 'configure' return '1' not '0'?

fbterm cannot use 'fbi'[1] to view images under framebuffer console.
When run 'fbi' under fbterm, it will complain as below:

    ioctl VT_GETSTATE: 无效的参数 (not a linux console?)

(Thanks to lurz's feedback)


[1]. http://linux.bytesex.org/fbida/

Attempted to convert consolefonts (psf) using psftools 
http://www.seasip.demon.co.uk/Unix/PSF/
to bdf format and installed in fontconfig dir.  Unfortunately did not work with 
fbterm.   

發現 Ctrl-Space 其實會影響到一些編輯器程式。 ( 
已確定有影響的有 vim, nano
)。目前在
jfbterm+ucimf 
上並沒有產生這個問題，猜想會不會是沒有將用過的 
Ctrl-Space 丟掉?

我測試過的版本有 fbterm-1.3 及 fbterm-1.4，使用 gentoo 安裝

我的主機資訊如下:
Linux localhost 2.6.26-gentoo-r1 #7 SMP Tue Mar 10 12:50:20 CST 2009 i686
Intel(R) Core(TM)2 CPU T5500 @ 1.66GHz GenuineIntel GNU/Linux

All the special keys are not recognized. I don't know why, I added the 
user into group fbdev, fbterm launches.

Addtionally, with Shift-PageUp, I find the real tty was flipped up, but 
the fbterm does nothing..

With root, all works..

I basically would want to use my fbcondecor background with fbterm, but the 
thing is I'd need to resize the text area because I need the output to start at 
128px from the top.

What steps will reproduce the problem?
1.Set fbterm to use the Terminus font (and remove all other fonts if installed)
2.Start fbterm and use any application that draws lines, e.g. elinks

What is the expected output? What do you see instead?
Expected: lines drawn in menus, dialog boxes, etc.
See: Lines not drawn.

What version of the product are you using? On what operating system?
fbterm 1.7, compiled from Arch User Repository(AUR), on Arch Linux.
Terminus font installed from Arch repos.

Please provide any additional information below.
If other fonts are installed, e.g. WenQuanYi Bitmap Song, lines are drawn in 
that font instead, which are not aligned properly and looks ugly.

Hi,

This issue is forward from debian/BTS #590831
  http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=590831

-----
From: "Trent W. Buck" <[email protected]>
To: Debian Bug Tracking System <[email protected]>
Subject: wish for --font-height=16, fix 1px padding of unifont
Date: Fri, 30 Jul 2010 00:02:14 +1000
Package: fbterm
Version: 1.6-1
Severity: wishlist

I want to replace jfbterm (using PCF unifont) with fbterm (using TTF
unifont).  Unfortunately, it seems to force 1px padding between cells:

    $ fbterm -vs16 -nunifont -- login -f twb
    [screen] driver: inteldrmfb, mode: 1024x600-32bpp, scrolling: redraw
    [font] width: 9px, height: 17px, ordered list: unifont, DejaVu Sans, DejaVu Sans Mono, DejaVu Serif
    [term] size: 113x35, default codec: UTF-8

In jfbterm I get the same size unifont glyphs, but they're packed a
little more tightly (and look better):

    # jfbterm -e login -f twb
    $ echo $COLUMNS'x'$LINES
    128x37

I can "fix" the width with --font-width=8, but I can't force the
height to 16px instead of 17px.  I also tried -s15, -s15.9, -s16pt,
-s16px, -nunifont-16 and -nunifont:pixelSize=16 -- none of which
worked.

I *think* a cheap way to get what I want is to simply patch in a
corresponding --font-height=16 option that sets mHeight.

PS: I'd actually prefer DejaVu + Dotum over unifont, but fbterm
doesn't do antialiasing, so to me ALL outline fonts look crap in it.

-----

I tested attached patch. Work fine for me.
I applied this patch on Debian.
Could you apply?

Best regards,
  Nobuhiro

A feature request.

1. use fb in true rgb way if available (do not use the 256-color palette to
convert 8bpp color to 32bpp), then render text to screen with its
antialiasing, hinting feature;

2. use bold font (bitmap or vector) when text is bold (<esc>[1m); if bold
font not availble, try to embolden it or use regular one.

What version of the product are you using? On what operating system?
fbterm-1.5, Debian GNU/Linux testing (squeeze)

Fedora 11中启动fbterm 
1.5后ctrl-alt-Fn组合键无效，不能切换回X中。

It's used to toggle input method.
Maybe it's better to set this shortcut in .fbtermrc.

A feature request.

I use ibus-fbterm as input method for fbterm, but somethimes ibus-fbterm 
freezes or crashes when I've already typed some words. I don't want to 
restart fbterm then redo the typing of many words. Is there a way to kill
and restart the input method or can you implement this?

ncurses line drawing with ACS_HLINE etc (functions hline(),
box() shows up as inverted questionmarks. Example programs
ncmpcpp/centerim/... (and as well a small minimal hack I
made myself).

TERM is "fbterm" (no difference with "linux"), utf8 locale.
Is this some problem with the terminfo/termcap perhaps?

This is fbterm 1.6 on Ubuntu 10.4. Font is DejaVu Sans Mono
(does it matter?)

Is it possible for fbterm to use an image as background. Then, I will dump
fbcondecor in my customized kernel.

我是第一次安装FBTERM，有几个问题：
1.~/.fbtermrc里的字体配置怎么做？我怎么才能让fbterm找到我的�
��体？
我的情况是只指定名称和同时指定名称和字体文件的路径都��
�行。
fbterm和X同样使用freetype2来设定字体，可是X的配置文件里有FON
TPATH参数来让X找到字
体，而fbterm里是什么参数在起这样的作用。
每次我启动fbterm时总是提示init font error!

This is compile time bug, to version 1.6.

1. Change to the top source directory
2. run `./configure ; echo $?`

The result will return 1 when 'libx86.h' existed.

I guess the problem comes from the last line of configure.ac

129 [[ "$VESA_WARN" = "yes" ]] && AC_MSG_WARN([libx86.h doesn't exist! VESA
support will be disabled!    ])

fbterm does not work on radeonfb unfortunately.
The text is garbled (truncated) and unreadable both in
normal and portrait mode. I've tried different fonts, 
but this doesn't help.

The fbterm on 32bpp radeonfb looks like in the attched screenshot.

If launch fbterm from current virtual console, everything is fine.

If launch fbterm from non-current virtual console，the display get mess.
1. Login into the virtual console /dev/tty5.
2. Run "sleep 3; fbterm -i fbterm_ucimf".
3. Press ALT+F2, switch to the virtual console /dev/tty2.
4. Wait a second, and the display get mess.
5. Press ALT+F5 and ALT+F2, and the display back to normal.


I try to run fbterm using /sbin/init, and find this bug.
0. Patch fbterm. (issuse 15 fbterm-1.4_run_command.patch
)
1. Change one line from /etc/inittab
"5:2345:respawn:/sbin/agetty -I '\033(K' tty5 38400"
to
"5:2345:respawn:/sbin/agetty -I '\033(K' tty5 38400 -n -l /bin/login.sh".
2. Create the file /bin/login.sh
cat > /bin/login.sh << eof
#!/bin/sh
export LANG=zh_CN.UTF-8
export LC_ALL=zh_CN.UTF-8
exec /usr/bin/fbterm -i /usr/bin/fbterm_ucimf /bin/login
eof
chmod 700 /bin/login.sh 
3. Run "init q".
4. Press ALT+F2.
5. Wait a while, and the display get mess.
6. Press ALT+F5 and ALT+F2, and the display back to normal.
After that, we can find a config file /.fbtermrc. It will be nice using a
system level config file /etc/fbterm.conf.

http://groups.google.com/group/ucimf/browse_thread/thread/baa5fd4dffdcb27d

fbterm is really nice. Thank you for starting this project. I do enjoy it.
Especially when "jfbterm" is not maintained or developed. fbterm is just
the one I need.
I'd like to suggest that in the future fbterm can support xfonts-unifont
(/usr/share/fonts/X11/misc/unifont.pcf.gz). The reasons I'd like to use
unifont.pcf in fbterm are:
1. Only one font file can meet almost all languages.
2. Because no ttf files are required, X-related programs won't be installed
due to dependence. 

Therefore we can use pure console environment. This is especially nice for
some light-weight environment.

Thanks again.

A feature request. 

Now with the <ESC>]P sequence I can set the 16-color palette to my favorite
Tango scheme. But the color settings for background and foreground in
~/.fbtermrc are limited to be in the color palette, then I can't use
lightgrey (0x07) for foreground text and can't set background to a color
not defined in the palette. 

I think the following features should be added:
1. background/foreground color can be set in rgb value;
2. color palette can be configured in ~/.fbtermrc

> What version of the product are you using? On what operating system?
fbterm-1.5, Debian GNU/Linux 5.0.1 (lenny)

1. uninitialized variable in get_mode()

if 0
        u32 xres_tmp, yres_tmp, bpp_tmp;
#else
        /* BUG FIXED */
        u32 xres_tmp = 0, yres_tmp = 0, bpp_tmp = 0;
#endif

2.  'unsupported' mode in get_mode()

#if 0
                if (minfo->memory_model != VBE_MODEL_PACKED
                        && minfo->memory_model != VBE_MODEL_RGB) continue;
#else
                /* BUG FIXED */
                switch (minfo->bits_per_pixel) {
                case 8:
                        if (minfo->memory_model != VBE_MODEL_PACKED) 
continue;
                        break;
                case 15:
                case 16:
                case 32:
                        if (minfo->memory_model != VBE_MODEL_RGB) continue;
                        break;
                default:
                        continue;
                }
#endif

Hi,

I've rotate my screen to portrait for document displaying and showing more 
lines of code. Framebuffer does support the screen rotation by adding 
"fbcon=rotate:3" in kernel booting parameter, I wonder whether it is 
possible for fbterm to support screen rotation.

What steps will reproduce the problem?
1. debian stable (currently etch)
2. normally install (configure/make/make install)
3. run fbterm

What version of the product are you using? On what operating system?
fbterm-1.1 on debian stable

Please provide any additional information below.
I have to say everything is ok, just one thing is borther me.
The gaps between two words are very large, I think there are
so many spaces between the words. I've changed several fonts,
but all efforts are useless.

Another suggestion, I'd like to see a chinese input method 
in fbterm, maybe you will add a input method, such as Wubi,
by lib unicon-im or something else.

the 256-color mode (by setting TERM=fbterm) is rather useful.
Wonder if you would add an option to set TERM=fbterm ?
Like this:

$ fbterm
 -> TERM=linux

$ fbterm --256-color
 -> TERM=fbterm

What steps will reproduce the problem?
1. install fbterm w3m w3m-img
2. w3m www.google.cn
3. the google picture cannot be displayed

What is the expected output? What do you see instead?
w3m-img can display pictures in a normal console when framebuffer is enabled.
if i start fbterm, it cannot, it should display the pictures.
if i use jfbterm, it can.

What version of the product are you using? On what operating system?
the issue exists in fbterm 1.4.2 and 1.5. i use Ubuntu 8.04.2.

Please provide any additional information below.
if i start fbterm and then jfbterm, w3m-img can display the picture, so i 
assume this is a fbterm problem.
fbterm is the best framebuffer terminal i've ever used. i really don't want to 
install jfbterm cause it cannot exit in 
Ubuntu and this bug is still not fixed. it looks like that no one is now 
actively maintaining jfbterm.
I also compile and tested the newest 1.5, the issue still exists.
please kindly fix this issue. thanks a lot in advance.

目标：字符界面下，使用中文用户名（非ascii）登录系统。

这要求把 /etc/inittab 的一行如
6:2345:respawn:/sbin/agetty -I '\033(K' tty6 38400
替换为
6:2345:respawn:/sbin/agetty -I '\033(K' tty6 38400 -l /usr/bin/env -
LANG=zh_CN.UTF-8 /usr/bin/fbterm -i /usr/bin/fbterm_ucimf /bin/login

简单的说，需要
LANG=zh_CN.UTF-8 fbterm -i fbterm_ucimf vi /somefile
直接在中文环境下运行 vi 编辑 /somefile 退出时回到原环境。

256 colours support like xterm or konsole have whould be nice.

The manpage contains a typo (-font-size vs. --font-size). The patch fixes
the problem.

debian testing with fbterm 1.6 using font=WenQuanYi Bitmap Song (xfonts-wqy 
0.9.9-3.2)

Glyphs like ♦ are dual-width in wenquanyi bitmap song, but 
incorrectly determined as singal-width and truncated to the left half.

The spacing property of the font should be used in determining this.

What steps will reproduce the problem?
1.  Set TERM to fbterm
2.  start emacs
3.  Evakuate (list-colors-display) in emacs

What is the expected output? What do you see instead?

The expected output is a list of 256 colors. This works when runing emacs in 
xterm, with TERM=xterm-256color, but does not work in fbterm, with TERM=fbterm.

Instead, I see a list of only 8 colors.

What version of the product are you using? On what operating system?

fbterm version 1.7, in Arch linux.

Please provide any additional information below.

When TERM=fbterm, 'tput colors' reports that there are 256 colors. But emacs 
can only see 8 colors.

I currently use the Linux system console in 800x600x32 mode, and have 
console-setup configured to use the "Fixed" font (size 16). I was considering 
switching to fbterm so I can use 256 colors on the console, but I can't seem to 
get it to use the console fonts in /usr/share/consolefonts. Is this possible?

System is amd64 Debian Squeeze, linux 2.6.32-5, fbterm 1.6-2.

(I would have asked this on a mailing list, but I can't find one. Apologies if 
there is one and I missed it.)

I am new to Linux so please do not flame me for asking a stupid question.

I have compiled fbterm 1.6 on Debian Lenny. After adding the kernel option 
vga=791 to activate the framebuffer device, fbterm works.

However, when I am in fbterm and hot-plug an USB device, the kernel messages 
(like "[123.456789] usb 1-2.1: New USB device found, blah blah blah") are 
printed directly on top of what is displayed in fbterm. These are in the font 
used by the native console and at the cursor location of the native console 
before launching fbterm. I believe the system console is still drawing on the 
framebuffer although fbterm is open.

Is there a way to:
1. display these messages in fbterm, by either somehow directing them to fbterm 
or having fbterm intercept and display these messages;
2. or make fbterm stay "on top of" these messages and suppress them?

Additionally, although this is not a question regarding fbterm, I would be very 
happy if someone could tell me how to suppress or redirect these kernel 
messages in the native console, like sending all these messages to tty2 when I 
work on tty1 and such.

Thanks in advance.

I'm filing this issue on behalf of a friend :)

What steps will reproduce the problem?
Check the file configure.ac. All "enalbeval" should be changed to
"enableval", and the file configure should thus be regenerated.

What is the expected output? What do you see instead?
Variables should be able to be set when I pass --enable-...; but it is
impossible to do so with this typo present.

What version of the product are you using? On what operating system?
1.5 and 1.6 on GNU/Linux

Not sure if this is a vim or fbterm issue.  In fbterm and bash, the cursor can 
change using echo -e '\033[?2c' and echo -e '\033[7c'.  But within vim, even by 
using the t_SI and t_EI, the cursor appearance cannot change.  It would be an 
enhancement to able to do this, and even more if the color of the cursor could 
change. (For switching vim modes).

What steps will reproduce the problem?
1. using fbterm-1.0.tar.gz
2. ran "./configure"
3. then "make"

What is the expected output? What do you see instead?

make[3]: Entering directory `/usr/local/src/fbterm-1.0/src'
g++ -DHAVE_CONFIG_H -I. -I..     -g -O2 -fno-exceptions -Ilib
-I/usr/include/freetype2     -MT screen.o -MD -MP -MF .deps/screen.Tpo -c
-o screen.o screen.cpp
screen.cpp: In static member function ?static Screen*
Screen::createInstance()?:
screen.cpp:83: error: 'O_CLOEXEC' was not declared in this scope
make[3]: *** [screen.o] Error 1
make[3]: Leaving directory `/usr/local/src/fbterm-1.0/src'
make[2]: *** [all-recursive] Error 1
make[2]: Leaving directory `/usr/local/src/fbterm-1.0/src'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/usr/local/src/fbterm-1.0'
make: *** [all] Error 2


What version of the product are you using? On what operating system?

> cat /etc/issue

Welcome to openSUSE 10.2 (X86-64) - Kernel 2.6.24.2


Please provide any additional information below.

I made the change from pendsig() to pendsigs() as in the previous issue
report suggested and I got by that problem but now this one occurs.

Please add build support for android devices
and some improvements for input special symbols without hardware keyboard.

in src/screen.cpp

#ifdef ENABLE_VESA
        s8 buf[16];
        Config::instance()->getOption("vesa-mode", buf, sizeof(buf));
        if (!strcmp(buf, "list")) {
                VesaDev::printModes();
                return 0;
        }
        u32 mode = 0;
        Config::instance()->getOption("vesa-mode", mode);

        if (!mode) pScreen = FbDev::initFbDev();
        if (!pScreen) pScreen = VesaDev::initVesaDev(mode);

        if (!pScreen) return 0;
#else
        pScreen = FbDev::initFbDev();
        /* BUG FIXED */
        if (!pScreen) return 0;
#endif

What steps will reproduce the problem?
1. using fbterm-1.0 source
2. ran "./configure"
3. then "make"

What is the expected output? What do you see instead?

I ran ./configure, then make:
>> make
make  all-recursive
make[1]: Entering directory `/disk0/home/xtof/download/fbterm-1.0'
Making all in src
make[2]: Entering directory `/disk0/home/xtof/download/fbterm-1.0/src'
Making all in lib
make[3]: Entering directory `/disk0/home/xtof/download/fbterm-1.0/src/lib'
g++ -DHAVE_CONFIG_H -I. -I../..     -g -O2 -fno-exceptions -Ilib
-I/usr/include/freetype2     -MT io.o -MD -MP -MF .deps/io.Tpo -c -o io.o
io.cpp
mv -f .deps/io.Tpo .deps/io.Po
g++ -DHAVE_CONFIG_H -I. -I../..     -g -O2 -fno-exceptions -Ilib
-I/usr/include/freetype2     -MT shell.o -MD -MP -MF .deps/shell.Tpo -c -o
shell.o shell.cpp
mv -f .deps/shell.Tpo .deps/shell.Po
g++ -DHAVE_CONFIG_H -I. -I../..     -g -O2 -fno-exceptions -Ilib
-I/usr/include/freetype2     -MT vterm_action.o -MD -MP -MF
.deps/vterm_action.Tpo -c -o vterm_action.o vterm_action.cpp
mv -f .deps/vterm_action.Tpo .deps/vterm_action.Po
g++ -DHAVE_CONFIG_H -I. -I../..     -g -O2 -fno-exceptions -Ilib
-I/usr/include/freetype2     -MT vterm.o -MD -MP -MF .deps/vterm.Tpo -c -o
vterm.o vterm.cpp
mv -f .deps/vterm.Tpo .deps/vterm.Po
g++ -DHAVE_CONFIG_H -I. -I../..     -g -O2 -fno-exceptions -Ilib
-I/usr/include/freetype2     -MT vterm_states.o -MD -MP -MF
.deps/vterm_states.Tpo -c -o vterm_states.o vterm_states.cpp
mv -f .deps/vterm_states.Tpo .deps/vterm_states.Po
rm -f libshell.a
ar cru libshell.a io.o shell.o vterm_action.o vterm.o vterm_states.o 
ranlib libshell.a
make[3]: Leaving directory `/disk0/home/xtof/download/fbterm-1.0/src/lib'
make[3]: Entering directory `/disk0/home/xtof/download/fbterm-1.0/src'
g++ -DHAVE_CONFIG_H -I. -I..     -g -O2 -fno-exceptions -Ilib
-I/usr/include/freetype2     -MT fbconfig.o -MD -MP -MF .deps/fbconfig.Tpo
-c -o fbconfig.o fbconfig.cpp
mv -f .deps/fbconfig.Tpo .deps/fbconfig.Po
g++ -DHAVE_CONFIG_H -I. -I..     -g -O2 -fno-exceptions -Ilib
-I/usr/include/freetype2     -MT fbio.o -MD -MP -MF .deps/fbio.Tpo -c -o
fbio.o fbio.cpp
mv -f .deps/fbio.Tpo .deps/fbio.Po
g++ -DHAVE_CONFIG_H -I. -I..     -g -O2 -fno-exceptions -Ilib
-I/usr/include/freetype2     -MT fbshell.o -MD -MP -MF .deps/fbshell.Tpo -c
-o fbshell.o fbshell.cpp
mv -f .deps/fbshell.Tpo .deps/fbshell.Po
g++ -DHAVE_CONFIG_H -I. -I..     -g -O2 -fno-exceptions -Ilib
-I/usr/include/freetype2     -MT fbterm.o -MD -MP -MF .deps/fbterm.Tpo -c
-o fbterm.o fbterm.cpp
fbterm.cpp: In member function 'virtual void SignalIo::readyRead(s8*, u32)':
fbterm.cpp:59: error: 'struct signalfd_siginfo' has no member named 'ssi_signo'
make[3]: *** [fbterm.o] Error 1
make[3]: Leaving directory `/disk0/home/xtof/download/fbterm-1.0/src'
make[2]: *** [all-recursive] Error 1
make[2]: Leaving directory `/disk0/home/xtof/download/fbterm-1.0/src'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/disk0/home/xtof/download/fbterm-1.0'
make: *** [all] Error 2

What version of the product are you using? On what operating system?
fbterm-1.0 source on gentoo amd64.

Please provide any additional information below.
default/linux/amd64/2008.0/desktop, gcc-4.1.2, glibc-2.6.1-r0,
2.6.25-gentoo-r6 x86_64

What steps will reproduce the problem?
1. using fbterm-1.0.tar.gz
2. ran "./configure"
3. then "make"

What is the expected output? What do you see instead?

make[3]: Entering directory `/usr/local/src/fbterm-1.0/src'
g++ -DHAVE_CONFIG_H -I. -I..     -g -O2 -fno-exceptions -Ilib
-I/usr/include/freetype2     -MT fbterm.o -MD -MP -MF .deps/fbterm.Tpo -c
-o fbterm.o fbterm.cpp
fbterm.cpp: In function ‘void pollSignal()’:
fbterm.cpp:73: error: ‘pendsig’ was not declared in this scope
make[3]: *** [fbterm.o] Error 1
make[3]: Leaving directory `/usr/local/src/fbterm-1.0/src'
make[2]: *** [all-recursive] Error 1
make[2]: Leaving directory `/usr/local/src/fbterm-1.0/src'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/usr/local/src/fbterm-1.0'
make: *** [all] Error 2


What version of the product are you using? On what operating system?

> cat /etc/issue

Welcome to openSUSE 10.2 (X86-64) - Kernel \r (\l).


Please provide any additional information below.

Output from configure:

checking for a BSD-compatible install... /usr/bin/install -c
checking whether build environment is sane... yes
checking for a thread-safe mkdir -p... /bin/mkdir -p
checking for gawk... gawk
checking whether make sets $(MAKE)... yes
checking for g++... g++
checking for C++ compiler default output file name... a.out
checking whether the C++ compiler works... yes
checking whether we are cross compiling... no
checking for suffix of executables...
checking for suffix of object files... o
checking whether we are using the GNU C++ compiler... yes
checking whether g++ accepts -g... yes
checking for style of include used by make... GNU
checking dependency style of g++... gcc3
checking for gcc... gcc
checking whether we are using the GNU C compiler... yes
checking whether gcc accepts -g... yes
checking for gcc option to accept ISO C89... none needed
checking dependency style of gcc... gcc3
checking for ranlib... ranlib
checking for pkg-config... /usr/bin/pkg-config
checking pkg-config is at least version 0.9.0... yes
checking for forkpty in -lutil... yes
checking for freetype2... yes
checking for fontconfig... yes
checking how to run the C preprocessor... gcc -E
checking for grep that handles long lines and -e... /usr/bin/grep
checking for egrep... /usr/bin/grep -E
checking for ANSI C header files... yes
checking for sys/wait.h that is POSIX.1 compatible... yes
checking for sys/types.h... yes
checking for sys/stat.h... yes
checking for stdlib.h... yes
checking for string.h... yes
checking for memory.h... yes
checking for strings.h... yes
checking for inttypes.h... yes
checking for stdint.h... yes
checking for unistd.h... yes
checking fcntl.h usability... yes
checking fcntl.h presence... yes
checking for fcntl.h... yes
checking for stdlib.h... (cached) yes
checking for string.h... (cached) yes
checking sys/ioctl.h usability... yes
checking sys/ioctl.h presence... yes
checking for sys/ioctl.h... yes
checking sys/socket.h usability... yes
checking sys/socket.h presence... yes
checking for sys/socket.h... yes
checking sys/time.h usability... yes
checking sys/time.h presence... yes
checking for sys/time.h... yes
checking termios.h usability... yes
checking termios.h presence... yes
checking for termios.h... yes
checking for unistd.h... (cached) yes
checking dlfcn.h usability... yes
checking dlfcn.h presence... yes
checking for dlfcn.h... yes
checking linux/fb.h usability... yes
checking linux/fb.h presence... yes
checking for linux/fb.h... yes
checking linux/kd.h usability... yes
checking linux/kd.h presence... yes
checking for linux/kd.h... yes
checking linux/keyboard.h usability... yes
checking linux/keyboard.h presence... yes
checking for linux/keyboard.h... yes
checking gpm.h usability... yes
checking gpm.h presence... yes
checking for gpm.h... yes
checking for stdbool.h that conforms to C99... yes
checking for _Bool... yes
checking for an ANSI C-conforming const... yes
checking whether struct tm is in sys/time.h or time.h... time.h
checking whether gcc needs -traditional... no
checking for stdlib.h... (cached) yes
checking for unistd.h... (cached) yes
checking for getpagesize... yes
checking for working mmap... yes
checking return type of signal handlers... void
checking whether lstat dereferences a symlink specified with a trailing
slash... yes
checking whether stat accepts an empty string... no
checking for bzero... yes
checking for memmove... yes
checking for munmap... yes
checking for setenv... yes
checking for socket... yes
checking for strchr... yes
checking for strstr... yes
checking for strtol... yes
configure: creating ./config.status
config.status: creating Makefile
config.status: creating src/Makefile
config.status: creating src/lib/Makefile
config.status: creating config.h
config.status: executing depfiles commands

I know it might be inappropriate to ask, but is that possible for fbterm 
to support VGA mode, like zhcon do?

Sometimes we forgot to load frame buffer in boot time, but don't want to 
reboot, as some important services or processes are running. If fbterm can 
do that, it will make zhcon obsolete. :-)

Now we can run "fbterm -i fbterm_ucimf -- vi newfile"，
and exit to shell.

It's same with issue 13.
I'm sorry for put it here again in english.