bug#25474: 26.0.50; Buffer content invisible

bug#25474: 26.0.50; Buffer content invisible

[Ivor Durham]

[-- Attachment #1.1: Type: text/plain, Size: 3776 bytes --]

--text follows this line--

 

After a fresh "git clone; autoconfig.sh; autoconfig.sh git; configure; make"
sequence I have an Emacs that displays a menu bar and right-hand scroll bar
but the actual buffer content is invisible. (See attached screenshot after
opening my .bashrc file.) If I use the "-nw" option to avoid the window
system, the buffer content is visible. I'm connecting to my Linux system via
SecureCRT from my Windows 10 laptop which has VcXsrv running locally.

 

The officially installed Emacs 25.1.1 displays the buffer content correctly
which this new 26.0.50 build does not with the same Windows client
connection.

 

Where should I look for more information on this problem. Thanks in advance.

 

 

In GNU Emacs 26.0.50.1 (x86_64-unknown-linux-gnu, GTK+ Version 2.24.31)

of 2017-01-18 built on clowder

Repository revision: dbb29d7eb428dd53617d31a9cc159d889deb1e8e

System Description:     Fedora release 25 (Twenty Five)

 

Recent messages:

Loading paren...done

For information about GNU Emacs and the GNU system, type C-h C-a.

 

Configured features:

XPM JPEG TIFF GIF PNG SOUND GSETTINGS NOTIFY GNUTLS LIBXML2 FREETYPE XFT

ZLIB TOOLKIT_SCROLL_BARS GTK2 X11

 

Important settings:

  value of $LANG: en_US.UTF-8

  locale-coding-system: utf-8-unix

 

Major mode: Lisp Interaction

 

Minor modes in effect:

  gpm-mouse-mode: t

  show-paren-mode: t

  tooltip-mode: t

  global-eldoc-mode: t

  electric-indent-mode: t

  mouse-wheel-mode: t

  tool-bar-mode: t

  menu-bar-mode: t

  file-name-shadow-mode: t

  auto-composition-mode: t

  auto-encryption-mode: t

  auto-compression-mode: t

  size-indication-mode: t

  line-number-mode: t

  transient-mark-mode: t

 

Load-path shadows:

None found.

 

Features:

(shadow sort mail-extr emacsbug message subr-x puny seq dired

dired-loaddefs format-spec rfc822 mml mml-sec password-cache epa epg

epg-config gnus-util rmail rmail-loaddefs mm-decode mm-bodies mm-encode

mail-parse rfc2231 mailabbrev gmm-utils mailheader sendmail rfc2047

rfc2045 ietf-drums mm-util mail-prsvr mail-utils t-mouse term/linux

php-mode advice derived rx edmacro kmacro speedbar sb-image ezimage

dframe etags xref cl-seq project eieio byte-opt bytecomp byte-compile

cl-extra help-mode cconv eieio-core cl-macs eieio-loaddefs flymake

compile comint ansi-color ring add-log cc-langs cc-mode cc-fonts

easymenu cc-guess cc-menus cc-cmds cc-styles cc-align cc-engine cc-vars

cc-defs cl gv cl-loaddefs pcase cl-lib time-date paren cus-start

cus-load mule-util tooltip eldoc electric uniquify ediff-hook vc-hooks

lisp-float-type mwheel term/x-win x-win term/common-win x-dnd tool-bar

dnd fontset image regexp-opt fringe tabulated-list replace newcomment

text-mode elisp-mode lisp-mode prog-mode register page menu-bar

rfn-eshadow isearch timer select scroll-bar mouse jit-lock font-lock

syntax facemenu font-core term/tty-colors frame cl-generic cham georgian

utf-8-lang misc-lang vietnamese tibetan thai tai-viet lao korean

japanese eucjp-ms cp51932 hebrew greek romanian slovak czech european

ethiopic indian cyrillic chinese composite charscript case-table

epa-hook jka-cmpr-hook help simple abbrev obarray minibuffer

cl-preloaded nadvice loaddefs button faces cus-face macroexp files

text-properties overlay sha1 md5 base64 format env code-pages mule

custom widget hashtable-print-readable backquote inotify dynamic-setting

system-font-setting font-render-setting move-toolbar gtk x-toolkit x

multi-tty make-network-process emacs)

 

Memory information:

((conses 16 149429 9392)

(symbols 48 26038 1)

(miscs 40 41 120)

(strings 32 33443 4625)

(string-bytes 1 1117918)

(vectors 16 18190)

(vector-slots 8 484789 5293)

(floats 8 249 245)

(intervals 56 155 0)

(buffers 976 12))


[-- Attachment #1.2: Type: text/html, Size: 8967 bytes --]

[-- Attachment #2: emacs-blank-buffer.JPG --]
[-- Type: image/jpeg, Size: 33750 bytes --]

bug#25474: 26.0.50; Buffer content invisible

[Eli Zaretskii]

> From: "Ivor Durham" <ivor.durham@ivor.cc>
> Date: Wed, 18 Jan 2017 01:50:45 -0800
> 
> After a fresh "git clone; autoconfig.sh; autoconfig.sh git; configure; make" sequence I have an Emacs that
> displays a menu bar and right-hand scroll bar but the actual buffer content is invisible. (See attached
> screenshot after opening my .bashrc file.) If I use the “-nw” option to avoid the window system, the buffer
> content is visible. I’m connecting to my Linux system via SecureCRT from my Windows 10 laptop which has
> VcXsrv running locally.
> 
> The officially installed Emacs 25.1.1 displays the buffer content correctly which this new 26.0.50 build does
> not with the same Windows client connection.

Does Emacs respond to input?  For example, if you move the mouse
pointer to hover above one of the tool-bar buttons and wait for a
second, does a tooltip pop up describing the button?  And if you click
on a menu-bar item, do you see a menu drop down?

If the above produce no response, can you see if Emacs is consuming
CPU cycles?

Thanks.

bug#25474: 26.0.50; Buffer content invisible

[Ivor Durham]

Clicking on menu items produces the expected drop-down. Hovering over icons produces the expected tooltip. Selecting Help/About produces a blank buffer. Selecting Help/GNU launches the browser as expected. So it appears to be only the buffer content display that is not working.

-----Original Message-----
From: Eli Zaretskii [mailto:eliz@gnu.org] 
Sent: Wednesday, January 18, 2017 7:33 AM
To: Ivor Durham
Cc: 25474@debbugs.gnu.org
Subject: Re: bug#25474: 26.0.50; Buffer content invisible

> From: "Ivor Durham" <ivor.durham@ivor.cc>
> Date: Wed, 18 Jan 2017 01:50:45 -0800
> 
> After a fresh "git clone; autoconfig.sh; autoconfig.sh git; configure; 
> make" sequence I have an Emacs that displays a menu bar and right-hand 
> scroll bar but the actual buffer content is invisible. (See attached 
> screenshot after opening my .bashrc file.) If I use the “-nw” option 
> to avoid the window system, the buffer content is visible. I’m connecting to my Linux system via SecureCRT from my Windows 10 laptop which has VcXsrv running locally.
> 
> The officially installed Emacs 25.1.1 displays the buffer content 
> correctly which this new 26.0.50 build does not with the same Windows client connection.

Does Emacs respond to input?  For example, if you move the mouse pointer to hover above one of the tool-bar buttons and wait for a second, does a tooltip pop up describing the button?  And if you click on a menu-bar item, do you see a menu drop down?

If the above produce no response, can you see if Emacs is consuming CPU cycles?

Thanks.

bug#25474: 26.0.50; Buffer content invisible

[Eli Zaretskii]

> From: "Ivor Durham" <ivor.durham@ivor.cc>
> Cc: <25474@debbugs.gnu.org>
> Date: Wed, 18 Jan 2017 08:04:53 -0800
> 
> Clicking on menu items produces the expected drop-down. Hovering over icons produces the expected tooltip. Selecting Help/About produces a blank buffer. Selecting Help/GNU launches the browser as expected. So it appears to be only the buffer content display that is not working.

What about the cursor -- do you see it?  If you click the mouse on the
text area, does it appear and/or move to the place where you clicked?

Also, does invoking Emacs like below make the text visible?

  emacs -Q -bg blue -fg read

How about this:

  emacs -Q -fn 8x12

bug#25474: 26.0.50; Buffer content invisible

[Ivor Durham]

The cursor does not appear. If I click in the buffer area the cut and copy
icons highlight while the mouse is down.

"emacs -Q -bg blue -fg red" produces a blue background for the buffer but no
text appears in red when I type.

"emacs -Q -fn 8x12" reports "Font '8x12' is not defined". However, I checked
my .Xdefaults and it specifies "emacs*font" as "9x15". Using "9x15" instead
of "8x12" the window opens with the blank buffer area.

-----Original Message-----
From: Eli Zaretskii [mailto:eliz@gnu.org] 
Sent: Wednesday, January 18, 2017 8:50 AM
To: Ivor Durham
Cc: 25474@debbugs.gnu.org
Subject: Re: bug#25474: 26.0.50; Buffer content invisible

> From: "Ivor Durham" <ivor.durham@ivor.cc>
> Cc: <25474@debbugs.gnu.org>
> Date: Wed, 18 Jan 2017 08:04:53 -0800
> 
> Clicking on menu items produces the expected drop-down. Hovering over
icons produces the expected tooltip. Selecting Help/About produces a blank
buffer. Selecting Help/GNU launches the browser as expected. So it appears
to be only the buffer content display that is not working.

What about the cursor -- do you see it?  If you click the mouse on the text
area, does it appear and/or move to the place where you clicked?

Also, does invoking Emacs like below make the text visible?

  emacs -Q -bg blue -fg read

How about this:

  emacs -Q -fn 8x12

bug#25474: 26.0.50; Buffer content invisible

[Eli Zaretskii]

> From: "Ivor Durham" <ivor.durham@ivor.cc>
> Cc: <25474@debbugs.gnu.org>
> Date: Wed, 18 Jan 2017 09:24:47 -0800
> 
> The cursor does not appear. If I click in the buffer area the cut and copy
> icons highlight while the mouse is down.
> 
> "emacs -Q -bg blue -fg red" produces a blue background for the buffer but no
> text appears in red when I type.
> 
> "emacs -Q -fn 8x12" reports "Font '8x12' is not defined". However, I checked
> my .Xdefaults and it specifies "emacs*font" as "9x15". Using "9x15" instead
> of "8x12" the window opens with the blank buffer area.

Can you reconfigure with --enable-checking='yes,glyphs', rebuild, and
then invoke "M-x dump-glyph-matrix RET" after starting Emacs from the
shell command line?  Please tell what does that show in the shell
window.

Thanks.

bug#25474: 26.0.50; Buffer content invisible

[Ivor Durham]

Here is the output from M-X dump-glyph-matrix after invoking emacs -Q:

PT = 146, BEGV = 1. ZV = 146
Cursor x = 0, y = 54, hpos = 0, vpos = 3
=============================================
  0: (1) ''
  0: (1) ';; This buffer is for text that is not saved, and for Lisp
evaluation.[\n]'
  0: (1) ''
  1: (1) ''
  1: (1) ';; To create a file, visit it with C-x C-f and enter text in its
buffer.[\n]'
  1: (1) ''
  2: (1) ''
  2: (1) '[\n]'
  2: (1) ''
  3: (1) ''
  3: (1) '[\n]'
  3: (1) ''
  4: (1) ''
  4: (1) '[\n]'
  4: (1) ''
  5: (1) ''
  5: (1) '[\n]'
  5: (1) ''
  6: (1) ''
  6: (1) '[\n]'
  6: (1) ''
  7: (1) ''
  7: (1) '[\n]'
  7: (1) ''
  8: (1) ''
  8: (1) '[\n]'
  8: (1) ''
  9: (1) ''
  9: (1) '[\n]'
  9: (1) ''
 10: (1) ''
 10: (1) '[\n]'
 10: (1) ''
 11: (1) ''
 11: (1) '[\n]'
 11: (1) ''
 12: (1) ''
 12: (1) '[\n]'
 12: (1) ''
 13: (1) ''
 13: (1) '[\n]'
 13: (1) ''
 14: (1) ''
 14: (1) '[\n]'
 14: (1) ''
 15: (1) ''
 15: (1) '[\n]'
 15: (1) ''
 16: (1) ''
 16: (1) '[\n]'
 16: (1) ''
 17: (1) ''
 17: (1) '[\n]'
 17: (1) ''
 18: (1) ''
 18: (1) '[\n]'
 18: (1) ''
 19: (1) ''
 19: (1) '[\n]'
 19: (1) ''
 20: (1) ''
 20: (1) '[\n]'
 20: (1) ''
 21: (1) ''
 21: (1) '[\n]'
 21: (1) ''
 22: (1) ''
 22: (1) '[\n]'
 22: (1) ''
 23: (1) ''
 23: (1) '[\n]'
 23: (1) ''
 24: (1) ''
 24: (1) '[\n]'
 24: (1) ''
 25: (1) ''
 25: (1) '[\n]'
 25: (1) ''
 26: (1) ''
 26: (1) '[\n]'
 26: (1) ''
 27: (1) ''
 27: (1) '[\n]'
 27: (1) ''
 28: (1) ''
 28: (1) '[\n]'
 28: (1) ''
 29: (1) ''
 29: (1) '[\n]'
 29: (1) ''
 30: (1) ''
 30: (1) '[\n]'
 30: (1) ''
 31: (1) ''
 31: (1) '[\n]'
 31: (1) ''
 32: (1) ''
 32: (1) '[\n]'
 32: (1) ''
 33: (1) ''
 33: (1) '[\n]'
 33: (1) ''
 34: (0) ''
 34: (0) ''
 34: (0) ''
 35: (0) ''
 35: (0) ''
 35: (0) ''
 36: (0) ''
 36: (0) ''
 36: (0) ''
 37: (1) ''
 37: (1) ' U:---  *scratch*      All L4     (Lisp Interaction)
'
 37: (1) ''

-----Original Message-----
From: Eli Zaretskii [mailto:eliz@gnu.org] 
Sent: Wednesday, January 18, 2017 10:30 AM
To: Ivor Durham
Cc: 25474@debbugs.gnu.org
Subject: Re: bug#25474: 26.0.50; Buffer content invisible

> From: "Ivor Durham" <ivor.durham@ivor.cc>
> Cc: <25474@debbugs.gnu.org>
> Date: Wed, 18 Jan 2017 09:24:47 -0800
> 
> The cursor does not appear. If I click in the buffer area the cut and 
> copy icons highlight while the mouse is down.
> 
> "emacs -Q -bg blue -fg red" produces a blue background for the buffer 
> but no text appears in red when I type.
> 
> "emacs -Q -fn 8x12" reports "Font '8x12' is not defined". However, I 
> checked my .Xdefaults and it specifies "emacs*font" as "9x15". Using 
> "9x15" instead of "8x12" the window opens with the blank buffer area.

Can you reconfigure with --enable-checking='yes,glyphs', rebuild, and then
invoke "M-x dump-glyph-matrix RET" after starting Emacs from the shell
command line?  Please tell what does that show in the shell window.

Thanks.

bug#25474: 26.0.50; Buffer content invisible

[Eli Zaretskii]

> From: "Ivor Durham" <ivor.durham@ivor.cc>
> Cc: <25474@debbugs.gnu.org>
> Date: Wed, 18 Jan 2017 13:27:21 -0800
> 
> Here is the output from M-X dump-glyph-matrix after invoking emacs -Q:

Thanks.  This output means that the Emacs display engine produced the
required display of the initial frame, and therefore the prime suspect
is the interaction with X.

Can you try running in X synchronous mode?  Like this:

  emacs -Q -xrm "emacs.synchronous: true"

bug#25474: 26.0.50; Buffer content invisible

[Ivor Durham]

emacs -Q -xrm "emacs.synchronous: true" -- still produces a blank buffer
area.

-----Original Message-----
From: Eli Zaretskii [mailto:eliz@gnu.org] 
Sent: Wednesday, January 18, 2017 7:34 PM
To: Ivor Durham
Cc: 25474@debbugs.gnu.org
Subject: Re: bug#25474: 26.0.50; Buffer content invisible

> From: "Ivor Durham" <ivor.durham@ivor.cc>
> Cc: <25474@debbugs.gnu.org>
> Date: Wed, 18 Jan 2017 13:27:21 -0800
> 
> Here is the output from M-X dump-glyph-matrix after invoking emacs -Q:

Thanks.  This output means that the Emacs display engine produced the
required display of the initial frame, and therefore the prime suspect is
the interaction with X.

Can you try running in X synchronous mode?  Like this:

  emacs -Q -xrm "emacs.synchronous: true"

bug#25474: 26.0.50; Buffer content invisible

[Eli Zaretskii]

> From: "Ivor Durham" <ivor.durham@ivor.cc>
> Cc: <25474@debbugs.gnu.org>
> Date: Wed, 18 Jan 2017 21:03:02 -0800
> 
> emacs -Q -xrm "emacs.synchronous: true" -- still produces a blank buffer
> area.

Thanks.  One possible suspect could be the double-buffering feature we
have on master.  Do you have HAVE_XDBE defined in src/config.h?

bug#25474: 26.0.50; Buffer content invisible

[Ivor Durham]

"#define HAVE_XDBE 1" is in src/config.h

-----Original Message-----
From: Eli Zaretskii [mailto:eliz@gnu.org] 
Sent: Thursday, January 19, 2017 7:51 AM
To: Ivor Durham
Cc: 25474@debbugs.gnu.org
Subject: Re: bug#25474: 26.0.50; Buffer content invisible

> From: "Ivor Durham" <ivor.durham@ivor.cc>
> Cc: <25474@debbugs.gnu.org>
> Date: Wed, 18 Jan 2017 21:03:02 -0800
> 
> emacs -Q -xrm "emacs.synchronous: true" -- still produces a blank 
> buffer area.

Thanks.  One possible suspect could be the double-buffering feature we have
on master.  Do you have HAVE_XDBE defined in src/config.h?

bug#25474: 26.0.50; Buffer content invisible

[Eli Zaretskii]

> From: "Ivor Durham" <ivor.durham@ivor.cc>
> Cc: <25474@debbugs.gnu.org>
> Date: Thu, 19 Jan 2017 07:57:52 -0800
> 
> "#define HAVE_XDBE 1" is in src/config.h

Can you replace it with

  #undef HAVE_XDBE

then edit src/Makefile to make the XDBE-related variables empty, like
this:

  XDBE_LIBS =
  XDBE_CFLAGS =

and then rebuild Emacs and see if the problem persists.

Alternatively, temporarily rename the header X11/extensions/Xdbe.h in
your include directory to something else, then re-run 'configure' and
rebuild.

bug#25474: 26.0.50; Buffer content invisible

[Ivor Durham]

Disabling XDBE fixes the problem. Thank you.

-----Original Message-----
From: Eli Zaretskii [mailto:eliz@gnu.org] 
Sent: Thursday, January 19, 2017 8:18 AM
To: Ivor Durham
Cc: 25474@debbugs.gnu.org
Subject: Re: bug#25474: 26.0.50; Buffer content invisible

> From: "Ivor Durham" <ivor.durham@ivor.cc>
> Cc: <25474@debbugs.gnu.org>
> Date: Thu, 19 Jan 2017 07:57:52 -0800
> 
> "#define HAVE_XDBE 1" is in src/config.h

Can you replace it with

  #undef HAVE_XDBE

then edit src/Makefile to make the XDBE-related variables empty, like
this:

  XDBE_LIBS =
  XDBE_CFLAGS =

and then rebuild Emacs and see if the problem persists.

Alternatively, temporarily rename the header X11/extensions/Xdbe.h in your
include directory to something else, then re-run 'configure' and rebuild.

bug#25474: 26.0.50; Buffer content invisible

[Eli Zaretskii]

> From: "Ivor Durham" <ivor.durham@ivor.cc>
> Cc: <25474@debbugs.gnu.org>
> Date: Thu, 19 Jan 2017 09:05:19 -0800
> 
> Disabling XDBE fixes the problem. Thank you.

Thanks for testing.

Daniel, could you please look into this?  If this is something
specific to this particular X server, and it cannot be fixed/worked
around in Emacs sources, perhaps we need a run-time option to disable
double-buffering for a certain display?

bug#25474: 26.0.50; Buffer content invisible

[Ivor Durham]

My environment in which this problem occurs:

I am using SecureCRT 8.0.3 on a Dell laptop running Windows 10 to connect
via ssh to a Fedora 25 Linux system. The laptop is running VcXsrv 1.15.2.0
as its X server, which is where the blank buffer content is displayed.

-----Original Message-----
From: Eli Zaretskii [mailto:eliz@gnu.org] 
Sent: Thursday, January 19, 2017 11:51 PM
To: Ivor Durham; Daniel Colascione
Cc: 25474@debbugs.gnu.org
Subject: Re: bug#25474: 26.0.50; Buffer content invisible

> From: "Ivor Durham" <ivor.durham@ivor.cc>
> Cc: <25474@debbugs.gnu.org>
> Date: Thu, 19 Jan 2017 09:05:19 -0800
> 
> Disabling XDBE fixes the problem. Thank you.

Thanks for testing.

Daniel, could you please look into this?  If this is something specific to
this particular X server, and it cannot be fixed/worked around in Emacs
sources, perhaps we need a run-time option to disable double-buffering for a
certain display?

bug#25474: 26.0.50; Buffer content invisible

[Daniel Colascione]

On Fri, Jan 20 2017, Eli Zaretskii wrote:
>> From: "Ivor Durham" <ivor.durham@ivor.cc>
>> Cc: <25474@debbugs.gnu.org>
>> Date: Thu, 19 Jan 2017 09:05:19 -0800
>> 
>> Disabling XDBE fixes the problem. Thank you.
>
> Thanks for testing.
>
> Daniel, could you please look into this?  If this is something
> specific to this particular X server, and it cannot be fixed/worked
> around in Emacs sources, perhaps we need a run-time option to disable
> double-buffering for a certain display?

I'll take a look. It may be worthwhile to just ditch XDBE and use a
pixmap instead --- that'd complicate the code in Emacs and potentially
(but not necessarily) reduce performance, but it would at least trigger
fewer X server bugs.

In the meantime, you can disable double buffering using the
inhibit-double-buffering frame parameter, e.g.,

  (modify-frame-parameters nil '((inhibit-double-buffering . nil)))

bug#25474: 26.0.50; Buffer content invisible

[Eli Zaretskii]

> From: Daniel Colascione <dancol@dancol.org>
> Cc: "Ivor Durham" <ivor.durham@ivor.cc>,  25474@debbugs.gnu.org
> Date: Fri, 20 Jan 2017 01:07:29 -0800
> 
> In the meantime, you can disable double buffering using the
> inhibit-double-buffering frame parameter, e.g.,
> 
>   (modify-frame-parameters nil '((inhibit-double-buffering . nil)))

Thanks, I forgot we had this option already.

So, Ivor, you can keep your "normal" build, without the HAVE_XDBE
tweaking ion src/config.h, and instead disable the feature (e.g., in
your ~/.emacs) as Daniel points out above.

bug#25474: 26.0.50; Buffer content invisible

[Ivor Durham]

I did a "make clean; ./configure; make" to re-generate src/config.h and
src/Makefile and confirmed the XDBE options were back. Then confirmed that
the blank buffer problem occurred again. Then I appended:

(modify-frame-parameters nil '((inhibit-double-buffering . nil)))

to ~/.emacs as prescribed. I launched the re-built src/emacs again with no
command-line arguments and the blank buffer area problem still occurred. So
the .emacs addition does not appear to be equivalent to the src/config.h &
src/Makefile changes.

-----Original Message-----
From: Eli Zaretskii [mailto:eliz@gnu.org] 
Sent: Friday, January 20, 2017 2:30 AM
To: Daniel Colascione
Cc: ivor.durham@ivor.cc; 25474@debbugs.gnu.org
Subject: Re: bug#25474: 26.0.50; Buffer content invisible

> From: Daniel Colascione <dancol@dancol.org>
> Cc: "Ivor Durham" <ivor.durham@ivor.cc>,  25474@debbugs.gnu.org
> Date: Fri, 20 Jan 2017 01:07:29 -0800
> 
> In the meantime, you can disable double buffering using the 
> inhibit-double-buffering frame parameter, e.g.,
> 
>   (modify-frame-parameters nil '((inhibit-double-buffering . nil)))

Thanks, I forgot we had this option already.

So, Ivor, you can keep your "normal" build, without the HAVE_XDBE tweaking
ion src/config.h, and instead disable the feature (e.g., in your ~/.emacs)
as Daniel points out above.

bug#25474: 26.0.50; Buffer content invisible

[martin rudalics]

 > In the meantime, you can disable double buffering using the
 > inhibit-double-buffering frame parameter, e.g.,
 >
 >    (modify-frame-parameters nil '((inhibit-double-buffering . nil)))

I suppose Daniel meant

(modify-frame-parameters nil '((inhibit-double-buffering . t)))

martin

bug#25474: 26.0.50; Buffer content invisible

[Daniel Colascione]

On 01/20/2017 09:41 AM, martin rudalics wrote:
>> In the meantime, you can disable double buffering using the
>> inhibit-double-buffering frame parameter, e.g.,
>>
>>    (modify-frame-parameters nil '((inhibit-double-buffering . nil)))
>
> I suppose Daniel meant
>
> (modify-frame-parameters nil '((inhibit-double-buffering . t)))

Err, yes.

bug#25474: 26.0.50; Buffer content invisible

[Ivor Durham]

t instead of nil does correct the blank buffer problem. I was a little surprised by nil, but deferred to my betters. 

(I'm leaving the build with the config.h/Makefile patches installed to avoid updating multiple .emacs files though!)

-----Original Message-----
From: Daniel Colascione [mailto:dancol@dancol.org] 
Sent: Friday, January 20, 2017 9:43 AM
To: martin rudalics; Eli Zaretskii
Cc: 25474@debbugs.gnu.org; Ivor Durham
Subject: Re: bug#25474: 26.0.50; Buffer content invisible

On 01/20/2017 09:41 AM, martin rudalics wrote:
>> In the meantime, you can disable double buffering using the 
>> inhibit-double-buffering frame parameter, e.g.,
>>
>>    (modify-frame-parameters nil '((inhibit-double-buffering . nil)))
>
> I suppose Daniel meant
>
> (modify-frame-parameters nil '((inhibit-double-buffering . t)))

Err, yes.

bug#25474: 26.0.50; Buffer content invisible

[Eli Zaretskii]

> From: "Ivor Durham" <ivor.durham@ivor.cc>
> Cc: <25474@debbugs.gnu.org>
> Date: Fri, 20 Jan 2017 12:29:17 -0800
> 
> (I'm leaving the build with the config.h/Makefile patches installed to avoid updating multiple .emacs files though!)

You should be able to make that setting conditional on the X display,
I think.  Martin, am I right?

bug#25474: 26.0.50; Buffer content invisible

[martin rudalics]

 >> (I'm leaving the build with the config.h/Makefile patches installed to avoid updating multiple .emacs files though!)
 >
 > You should be able to make that setting conditional on the X display,
 > I think.  Martin, am I right?

I'm not sure what "avoid updating multiple .emacs files" means.  Ivor
can customize the frame parameter in ‘default-frame-alist’ in which case
it applies to all frames that "understand" the parameter - in the
particular case to all frames of a >= 26.1 Emacs build.  It can also be
added manually to this list based on the ‘window-system’ used (`x' in
the case at hand) preferably via ‘customize-set-variable’ in which case
it will apply to X frames only.

martin

bug#25474: 26.0.50; Buffer content invisible

[Eli Zaretskii]

> Date: Sat, 21 Jan 2017 14:20:17 +0100
> From: martin rudalics <rudalics@gmx.at>
> CC: dancol@dancol.org, 25474@debbugs.gnu.org
> 
>  > You should be able to make that setting conditional on the X display,
>  > I think.  Martin, am I right?
> 
> I'm not sure what "avoid updating multiple .emacs files" means.  Ivor
> can customize the frame parameter in ‘default-frame-alist’ in which case
> it applies to all frames that "understand" the parameter - in the
> particular case to all frames of a >= 26.1 Emacs build.  It can also be
> added manually to this list based on the ‘window-system’ used (`x' in
> the case at hand) preferably via ‘customize-set-variable’ in which case
> it will apply to X frames only.

Ivor's problem only happens when he uses Emacs via a certain X server
that runs on Windows.  I thought there was a way of specifying
parameters default-frame-alist in a way that would cause some
parameter to have a special value in frames on specific X display.

bug#25474: 26.0.50; Buffer content invisible

[Ivor Durham]

Sorry for the imprecision: What I meant by "multiple .emacs files" is that I use multiple accounts on the same Linux system for different consulting clients so I would have to make sure I copied the patched .emacs to all accounts (or link to a public version). Not a big deal but easier to fix in one place.

I also use the Linux system both remotely from my laptop and locally from the keyboard so I have two different X implementations to contend with, the one on the Fedora Linux system itself and the VcXsrv on my laptop.

-----Original Message-----
From: martin rudalics [mailto:rudalics@gmx.at] 
Sent: Saturday, January 21, 2017 5:20 AM
To: Eli Zaretskii; Ivor Durham
Cc: dancol@dancol.org; 25474@debbugs.gnu.org
Subject: Re: bug#25474: 26.0.50; Buffer content invisible

 >> (I'm leaving the build with the config.h/Makefile patches installed to avoid updating multiple .emacs files though!)  >  > You should be able to make that setting conditional on the X display,  > I think.  Martin, am I right?

I'm not sure what "avoid updating multiple .emacs files" means.  Ivor can customize the frame parameter in ‘default-frame-alist’ in which case it applies to all frames that "understand" the parameter - in the particular case to all frames of a >= 26.1 Emacs build.  It can also be added manually to this list based on the ‘window-system’ used (`x' in the case at hand) preferably via ‘customize-set-variable’ in which case it will apply to X frames only.

martin

bug#25474: 26.0.50; Buffer content invisible

[martin rudalics]

 > Ivor's problem only happens when he uses Emacs via a certain X server
 > that runs on Windows.  I thought there was a way of specifying
 > parameters default-frame-alist in a way that would cause some
 > parameter to have a special value in frames on specific X display.

You probably mean ‘window-system-default-frame-alist’.  That's a
separate variable.  I suppose the most simple solution is to provide the
parameter as additional argument of ‘make-frame’.

martin

bug#25474: 26.0.50; Buffer content invisible

[Eli Zaretskii]

> From: Daniel Colascione <dancol@dancol.org>
> Date: Fri, 20 Jan 2017 09:42:31 -0800
> Cc: 25474@debbugs.gnu.org, Ivor Durham <ivor.durham@ivor.cc>
> 
> On 01/20/2017 09:41 AM, martin rudalics wrote:
> >> In the meantime, you can disable double buffering using the
> >> inhibit-double-buffering frame parameter, e.g.,
> >>
> >>    (modify-frame-parameters nil '((inhibit-double-buffering . nil)))
> >
> > I suppose Daniel meant
> >
> > (modify-frame-parameters nil '((inhibit-double-buffering . t)))
> 
> Err, yes.

Daniel, any chance you could look into fixing this problem without
disabling double-buffering?  This bug is currently one of those which
block the release of Emacs 26.1, so it would be good to have it solved
in some way.

Thanks.