Re: [blfs-dev] Seeking advice for testing seamonkey and thunderbird
On 11/02/2018 17:25, Ken Moffat wrote: On Sun, Feb 11, 2018 at 05:04:50PM +0100, Pierre Labastie wrote: On 11/02/2018 16:24, Bruce Dubbs wrote: Pierre Labastie wrote: Hi, For both seamonkey and thunderbird, system cairo is currently disabled. But it seems recent versions of firefox can handle system cairo. So I'd like to enable system cairo also for new seamonkey (based on firefox 52) and thunderbird. But how can I make sure that it works? Pointer to the mailing list maybe, or direct advice. Thanks for your answers. From memory (it was maybe 2 years ago), firefox crashed frequently - if I opened multiple youtube tabs (hopefully paused, or with sound off!) and then tried moving to new positions. I think there were also other crashes blamed on system cairo. But firefox is more robust these days. Unfortunately, ff59-beta does not compile if system cairo is enabled (one or more of the cargo packages seem to need an older version). As of -O2, our firefox config file has it too, and says optimizing for size is broken in gcc 7, whatever that means... BTW, I have removed the extra CFLAGS, they seem to not be needed anymore. I think it was gcc-6, or perhaps 6.1, where we had to add those. Without them, the firefox UI was severely broken (I think Bruce reported inaccessible menu items). I think I tested without that in gcc-7, but I don't remember the result. ĸen Things are not getting well: I was looking for a translation from French to English, so I opened the Collins French-English dictionary and entered my query. Then the result page began to display, but seamonkey crashed (segmentation fault)! Tried again, same. At the time, my compilation options were no --enable-optimize switch (means the optimization switch for gcc is -Os), and system cairo. I first rebuilt with --enable-optimize=-02, same crash. Then I rebuilt disabling system cairo, and... still the same result. Segfault when displaying a Collins dictionary result... Here is the faulty link: https://www.collinsdictionary.com/dictionary/french-english The page is displayed for less than a second, then the crash happens... Note that I have watched youtube videos for a few hours without any crash! Switching back and forth between different tabs playing different video. So I must be missing something, but what? Pierre -- http://lists.linuxfromscratch.org/listinfo/blfs-dev FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
[blfs-dev] Archived openssl
Guys, I just archived openssl (1.1.x) in BLFS. I had to touch about 65 packages. When updating a page, please review the dependencies to see that they make sense. In some cases we had openssl or x or y. If the dependency was required or recommended, I moved x and y to optional. -- Bruce -- http://lists.linuxfromscratch.org/listinfo/blfs-dev FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: [blfs-dev] Seeking advice for testing seamonkey and thunderbird
On Sun, Feb 11, 2018 at 05:04:50PM +0100, Pierre Labastie wrote: > On 11/02/2018 16:24, Bruce Dubbs wrote: > > Pierre Labastie wrote: > > > Hi, > > > > > > For both seamonkey and thunderbird, system cairo is currently > > > disabled. But it > > > seems recent versions of firefox can handle system cairo. So I'd like to > > > enable system cairo also for new seamonkey (based on firefox 52) and > > > thunderbird. But how can I make sure that it works? Pointer to the > > > mailing > > > list maybe, or direct advice. Thanks for your answers. From memory (it was maybe 2 years ago), firefox crashed frequently - if I opened multiple youtube tabs (hopefully paused, or with sound off!) and then tried moving to new positions. I think there were also other crashes blamed on system cairo. But firefox is more robust these days. Unfortunately, ff59-beta does not compile if system cairo is enabled (one or more of the cargo packages seem to need an older version). > As of -O2, our firefox config file has it too, and says optimizing for size > is broken in gcc 7, whatever that means... BTW, I have removed the extra > CFLAGS, they seem to not be needed anymore. > I think it was gcc-6, or perhaps 6.1, where we had to add those. Without them, the firefox UI was severely broken (I think Bruce reported inaccessible menu items). I think I tested without that in gcc-7, but I don't remember the result. ĸen -- Truth, in front of her huge walk-in wardrobe, selected black leather boots with stiletto heels for such a barefaced truth. - Unseen Academicals -- http://lists.linuxfromscratch.org/listinfo/blfs-dev FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: [blfs-dev] Seeking advice for testing seamonkey and thunderbird
On 11/02/2018 16:24, Bruce Dubbs wrote: Pierre Labastie wrote: Hi, For both seamonkey and thunderbird, system cairo is currently disabled. But it seems recent versions of firefox can handle system cairo. So I'd like to enable system cairo also for new seamonkey (based on firefox 52) and thunderbird. But how can I make sure that it works? Pointer to the mailing list maybe, or direct advice. Thanks for your answers. IIRC, we disabled cairo to be consistent with firefox. I do not recall the problematic pages, but I think it may have been menu appearance on all pages. Looking at my logs, the last time I enabled cairo in SM was seamonkey-2.35. I didn't build every version, but it was disabled in seamonkey-2.38. You might want to look at the optimize setting also. Do we still need -O2? Also, with only required and recommended dependencies (that is only alsa and no pulseaudio) of seamonkey, I can see html5 video (youtube), but I get no sound, although "speaker-test" and "aplay /usr/share/sounds/Front_Left.wav" both run OK. Maybe the config file is missing --enable-alsa... But maybe we should switch to require pulseaudio, as on the firefox page. Thoughts? I had pulse disabled in seamonkey-2.40, but enabled after that. There definately is an option --enable-alsa in SM, so you might want to try that. I added --enable-alsa, and I got sound. I think I'll try pulseaudio now. Large parts of the seamonkey code are based upon firefox-52.0.4-ESR, and Firefox upstream recommend pulseaudio. As of -O2, our firefox config file has it too, and says optimizing for size is broken in gcc 7, whatever that means... BTW, I have removed the extra CFLAGS, they seem to not be needed anymore. I haven't seen any crash yet, so I think system cairo is OK. Pierre -- http://lists.linuxfromscratch.org/listinfo/blfs-dev FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: [blfs-dev] Seeking advice for testing seamonkey and thunderbird
Pierre Labastie wrote: Hi, For both seamonkey and thunderbird, system cairo is currently disabled. But it seems recent versions of firefox can handle system cairo. So I'd like to enable system cairo also for new seamonkey (based on firefox 52) and thunderbird. But how can I make sure that it works? Pointer to the mailing list maybe, or direct advice. Thanks for your answers. IIRC, we disabled cairo to be consistent with firefox. I do not recall the problematic pages, but I think it may have been menu appearance on all pages. Looking at my logs, the last time I enabled cairo in SM was seamonkey-2.35. I didn't build every version, but it was disabled in seamonkey-2.38. You might want to look at the optimize setting also. Do we still need -O2? Also, with only required and recommended dependencies (that is only alsa and no pulseaudio) of seamonkey, I can see html5 video (youtube), but I get no sound, although "speaker-test" and "aplay /usr/share/sounds/Front_Left.wav" both run OK. Maybe the config file is missing --enable-alsa... But maybe we should switch to require pulseaudio, as on the firefox page. Thoughts? I had pulse disabled in seamonkey-2.40, but enabled after that. There definately is an option --enable-alsa in SM, so you might want to try that. -- Bruce -- http://lists.linuxfromscratch.org/listinfo/blfs-dev FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: [blfs-dev] math.h : No such file or directory (from gcc)
On 11/02/2018 11:58, Pierre Labastie wrote: On 10/02/2018 19:35, Alain Toussaint wrote: Hello, Someone have any clue how to solve this: In file included from ../../include/c99_math.h:194:0, from ../../src/gallium/auxiliary/util/u_math.h:44, from ../../src/mesa/main/macros.h:35, from glsl_types.cpp:25: /usr/include/c++/7.3.0/cmath:45:15: erreur fatale: math.h : Aucun fichier ou dossier de ce type #include_next ^~~~ Aucun fichier ou dossier de ce type == No such file or directory. When include_next does not work it may be that some headers have been "include-fixed" and should not have been. Try running (as root): --- /usr/libexec/gcc/x86_64-pc-linux-gnu/7.1.0/install-tools/mkheaders --- before anything else. See the Caution on the BLFS GCC page. Pierre Forgot to remove the SPAM header in the preceding, and also this should be updated to: /usr/libexec/gcc/x86_64-pc-linux-gnu/7.3.0/install-tools/mkheaders Pierre -- http://lists.linuxfromscratch.org/listinfo/blfs-dev FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page
Re: [blfs-dev] ***SPAM*** math.h : No such file or directory (from gcc)
On 10/02/2018 19:35, Alain Toussaint wrote: Hello, Someone have any clue how to solve this: In file included from ../../include/c99_math.h:194:0, from ../../src/gallium/auxiliary/util/u_math.h:44, from ../../src/mesa/main/macros.h:35, from glsl_types.cpp:25: /usr/include/c++/7.3.0/cmath:45:15: erreur fatale: math.h : Aucun fichier ou dossier de ce type #include_next ^~~~ Aucun fichier ou dossier de ce type == No such file or directory. When include_next does not work it may be that some headers have been "include-fixed" and should not have been. Try running (as root): --- /usr/libexec/gcc/x86_64-pc-linux-gnu/7.1.0/install-tools/mkheaders --- before anything else. See the Caution on the BLFS GCC page. Pierre -- http://lists.linuxfromscratch.org/listinfo/blfs-dev FAQ: http://www.linuxfromscratch.org/blfs/faq.html Unsubscribe: See the above information page