Commit Graph

1112 Commits

Author SHA1 Message Date
Christoph Lohmann
980991fa6e Fix the new -e handling. An empty cmd has to work for backwards compatibility. 2015-05-10 15:19:48 +02:00
suigin
ae1923d275 Clean up xdraws and optimize glyph drawing with non-unit kerning values
I have another patch here for review that optimizes the performance of
glyph drawing, primarily when using non-unit kerning values, and fixes a
few other minor issues. It's dependent on the earlier patch from me that
stores unicode codepoints in a Rune type, typedef'd to uint_least32_t.

This patch is a pretty big change to xdraws so your scrutiny is
appreciated.

First, some performance numbers. I used Yu-Jie Lin termfps.sh shell
script to benchmark before and after, and you can find it in the
attachments. On my Kaveri A10 7850k machine, I get the following
results:

Before Patch
============

1) Font: "Liberation Mono:pixelsize=12:antialias=false:autohint=false"
   cwscale: 1.0, chscale: 1.0
   For 273x83 100 frames.
   Elapsed time :     1.553
   Frames/second:    64.352
   Chars /second: 1,458,159

2) Font: "Inconsolata:pixelsize=14:antialias=true:autohint=true"
   cwscale: 1.001, chscale: 1.001
   For 239x73 100 frames.
   Elapsed time :   159.286
   Frames/second:     0.627
   Chars /second:    10,953

After Patch
===========

3) Font: "Liberation Mono:pixelsize=12:antialias=false:autohint=false"
   cwscale: 1.0, chscale: 1.0
   For 273x83 100 frames.
   Elapsed time :     1.544
   Frames/second:    64.728
   Chars /second: 1,466,690

4) Font: "Inconsolata:pixelsize=14:antialias=true:autohint=true"
   cwscale: 1.001, chscale: 1.001
   For 239x73 100 frames.
   Elapsed time :     1.955
   Frames/second:    51.146
   Chars /second:   892,361

As you can see, while the improvements for fonts with unit-kerning is
marginal, there's a huge ~81x performance increase with the patch when
using kerning values other than 1.0.

So what does the patch do?

The `xdraws' function would render each glyph one at a time if non-unit
kerning values were configured, and this was the primary cause of the
slow down. Xft provides a handful of functions which allow you to render
multiple characters or glyphs at time, each with a unique <x,y> position,
so it was simply a matter of massaging the data into a format that would
allow us to use one of these functions.

I've split `xdraws' up into two functions. In the first pass with
`xmakeglyphfontspecs' it will iterate over all of the glyphs in a given
row and it will build up an array of corresponding XftGlyphFontSpec
records. Much of the old logic for resolving fonts for glyphs using Xft
and fontconfig went into this function.

The second pass is done with `xrenderglyphfontspecs' which contains the
old logic for determining colors, clearing the background, and finally
rendering the array of XftGlyphFontSpec records.

There's a couple of other things that have been improved by this patch.
For instance, the UTF-32 codepoints in the Line's were being re-encoded
back into UTF-8 strings to be passed to `xdraws' which in turn would then
decode back to UTF-32 to verify that the Font contained a matching glyph
for the code point. Next, the UTF-8 string was being passed to
`XftDrawStringUtf8' which internally mallocs a scratch buffer and decodes
back to UTF-32 and does the lookup of the glyphs all over again.

This patch gets rid of all of this redundant round-trip encoding and
decoding of characters to be rendered and only looks up the glyph index
once (per font) during the font resolution phase. So this is probably
what's responsible for the marginal improvements seen when kerning values
are kept to 1.0.

I imagine there are other performance improvements here too, not seen in
the above benchmarks, if the user has lots of non-ASCII code plane characters
on the screen, or several different fonts are being utilized during
screen redraw.

Anyway, if you see any problems, please let me know and I can fix them.
2015-05-07 12:03:44 +02:00
suigin
38af006b5e Changed type for UTF-32 codepoints from long to uint_least32_t 2015-05-06 08:15:41 +02:00
noname
c990abfedf Fix empty selection highlighting bug.
When user clicks LMB, one character is selected, but will not be copied
to selection until the user moves cursor a bit. Therefore, the character
should not be highlighted as selected yet.

Before the patch, the trick was not to mark line as dirty to avoid
highlighting it. However, if user has already selected something and
clicks in line that contains selection, selclear sets the line as dirty
and one character is highlighted when it should not.

This patch replaces dirty trick with explicit check for sel.mode inside
selected().
2015-05-04 12:06:43 +02:00
noname
3cb7f27afe Fix indentation. 2015-05-04 12:00:10 +02:00
noname
1811b6030c Add enumeration for sel.mode
This patch also prevents sel.mode from increasing beyond 2. It is almost
impossible, but sel.mode may overflow if mouse is moved around for too
long while selecting.
2015-05-04 11:57:17 +02:00
noname
22571ea4e8 selnormalize: make special case explicit
Special case is when regular selection spans multiple lines.
Otherwise, just sort sel.ob.x and sel.ob.y.
2015-05-04 11:47:53 +02:00
noname
8751809aff selsnap: simplify SNAP_LINE case
Also make sure y never exceeds term.row-1 even if ATTR_WRAP is set for
some reason.
2015-05-04 11:41:55 +02:00
noname
765bb0fd14 Remove first argument of selsnap. 2015-05-04 11:16:08 +02:00
Jochen Sprickerhof
07ce96a3a0 Fix sigchld
Only wait for termination of the shell.
2015-05-04 10:57:40 +02:00
mvdan@mvdan.cc
190b94c7a2 len assignment is never used
Signed-off-by: Daniel Martí <mvdan@mvdan.cc>
2015-04-27 10:10:30 +02:00
mvdan@mvdan.cc
4f21c41a1c Clarify calculation precedence for '&' and '?'
Signed-off-by: Daniel Martí <mvdan@mvdan.cc>
2015-04-27 10:09:49 +02:00
mvdan@mvdan.cc
3a5053f6c1 Use %u for uint
Signed-off-by: Daniel Martí <mvdan@mvdan.cc>
2015-04-27 10:09:15 +02:00
suigin
7ab6c92e18 Optimize memory footprint of line buffers 2015-04-27 10:05:14 +02:00
noname@inventati.org
0622ad9bad Make tputc, tsetchar and techo accept unicode 2015-04-27 09:50:40 +02:00
noname@inventati.org
21f765426c Change internal character representation. 2015-04-27 09:50:01 +02:00
noname@inventati.org
753fe862b1 Remove last parameter of utf8encode
This parameter was always UTF_SIZ, so it is better remove it and
use directly UTF_SIZ in it.
2015-04-27 08:59:45 +02:00
noname@inventati.org
61c35cd246 Use utf8len instead of utf8decode. 2015-04-27 08:58:37 +02:00
alp@alexpilon.ca
742a41d655 Make build shut up about system() without return value check.
st.c:1321:2: warning: ignoring return value of function declared with warn_unused_result attribute [-Wunused-result]
            system(cmd);
            ^~~~~~ ~~~

Debatable whether an error here should case exit(EXIT_FAILURE). Just
preserving the existing behaviour for now.
2015-04-27 08:57:51 +02:00
Roberto E. Vargas Caballero
f36dd277a2 Merge remote-tracking branch 'origin/master' 2015-04-23 17:59:39 +02:00
Roberto E. Vargas Caballero
0d7448dabc Fix segmentation fault in strhandle()
We cannot pass strescseq.args[0] to atoi when nargs is zero,
because in this case it will be null.
2015-04-23 17:59:10 +02:00
Roberto E. Vargas Caballero
b17aa18f7c Uses a &[] pointer loop instead of + pointer loop 2015-04-23 17:59:10 +02:00
noname@inventati.org
736685d641 Do not set terminal title based on stty arguments. 2015-04-23 17:59:10 +02:00
noname@inventati.org
89807ed453 Move common code to xloadcolor 2015-04-21 09:48:33 +02:00
noname@inventati.org
84c756b97e Use LEN(dc.col) instead of LEN(colorname).
LEN(colorname) may be below 256 for some configurations.
2015-04-21 08:13:46 +02:00
noname@inventati.org
2fdcc5e5f6 Remove WIN_REDRAW flag.
WIN_REDRAW flag was not used since introduction of Xdbe
in commit 94771d0588
2015-04-21 07:49:07 +02:00
noname@inventati.org
ab69ea89b7 Place memset arguments in the correct order. 2015-04-20 09:29:24 +02:00
noname@inventati.org
5528280fae Remove explicit 'return' from 'void' functions. 2015-04-20 09:28:14 +02:00
noname@inventati.org
6dc2b546ec Increment accuaracy in drawtime calculation
This way is a bit more accurate.
2015-04-20 09:20:35 +02:00
noname@inventati.org
c27c731b9f Monotonic clock cannot jump backwards.
The check was introduced back when st used gettimeofday.
The condition is also modified to increment the accuaracy of the
calculation.
2015-04-20 09:18:40 +02:00
noname@inventati.org
6ee56d6590 Place tlinelen type on separate line. 2015-04-20 08:30:49 +02:00
Roberto E. Vargas Caballero
215bdb2da3 Add tty line support
Not always is desirable to create a pseudo terminal, and some times
we want to open a terminal emulator over a tty line. With this new
patch is possible to do someting like:

	$ st -l /dev/ttyS0 115200

Without this option was needed to launch another terminal emulator
over st (for example minicom, picocom, cu, ...).
2015-04-15 10:52:44 +02:00
sin
56abffb4b6 Fix memmove() invocation with src/dst being NULL
This fixes a segmentation fault on some systems.
2015-04-15 10:51:00 +02:00
noname@inventati.org
aff35af275 Use as command arguments the remaining parameters
This change allows execute st as 'st mutt' while it keeps the
compability with xterm and urxt.
2015-04-14 09:55:15 +02:00
Roberto E. Vargas Caballero
83e73c6242 Merge branch 'master' of ssh://suckless.org/gitrepos/st 2015-04-14 09:47:19 +02:00
Markus Wichmann
42fa1f5ce4 Implement most ICCCM rules for selection handling.
ICCCM mandates the use of real timestamps to interact with the
selection, to rule out race conditions if the clients are run at
different speeds. I have implemented the low hanging fruit, putting the
timestamps into text selection. Also, ICCCM mandates a check for whether
XSetSelectionOwner() worked. Not sure my version is correct, though.
2015-04-13 22:18:45 +02:00
Roberto E. Vargas Caballero
23ed12857f Merge branch 'master' of ssh://suckless.org/gitrepos/st 2015-04-13 22:01:40 +02:00
noname@inventati.org
9eb70a2d3e Do not use tmoveto in tputtab.
tmoveto resets CURSOR_WRAPNEXT.

Simple testcase:

for i in $(seq 1 200); do
	printf '\t.';
	usleep 100000;
	printf '\t@';
	usleep 100000;
done

In st executing this script causes @ and . to overwrite each other in
the last column.
2015-04-13 21:33:36 +02:00
noname@inventati.org
9619760e12 tresize: remove unnecessary if 2015-04-13 17:12:49 +02:00
noname@inventati.org
a6af2cc469 Remove old TODO entry.
It probably refers to
http://lists.suckless.org/dev/1211/13427.html
and does not seem like a bug in st.
2015-04-13 17:06:11 +02:00
noname@inventati.org
b94ad75e5d Remove 'titles' variable.
We do not free it until exit anyway.
2015-04-13 17:00:08 +02:00
noname@inventati.org
e6dd0f825d Remove useless if in tstrsequence. 2015-04-13 15:17:24 +02:00
noname@inventati.org
b0310fba5d Simplify tmoveto.
LIMIT returns value. This fact is already used in x2col and y2row.
2015-04-13 15:15:47 +02:00
noname@inventati.org
9d1495f9ee Fix typo.
It seems that LICENSE files are more common than LICENCE files.
At least this patch makes spelling consistent.
2015-04-13 14:26:01 +02:00
noname@inventati.org
c569e3146e Remove 'slide' variable in tresize. 2015-04-13 10:18:08 +02:00
noname@inventati.org
39ae1a4de5 Move tresize comments around. 2015-04-13 09:26:03 +02:00
noname@inventati.org
6352502d64 tresize: move for loop outside if
There is no need to check that slide > 0 before executing loop.
If slide <= 0, loop stops immediately.
2015-04-13 09:21:03 +02:00
noname@inventati.org
b9390a5496 Simplify loop condition. 2015-04-13 09:20:00 +02:00
noname@inventati.org
d2937b05ae Remove unnecessary XFilterEvent call.
XFilterEvent usually filters KeyPress events according to input method.
At this point the window is not mapped. The only events that we process
are ConfigureNotify and MapNotify. They should not be filtered by input
method.
2015-04-13 09:17:06 +02:00
noname@inventati.org
d3e0f3444b Use do..while in window mapping loop. 2015-04-13 09:15:42 +02:00