Re: v2.19.47 on Mac x86 (Jacques Menu Muzhic)

2016-10-03 Thread Trevor Bača 
On Sun, Sep 25, 2016 at 1:53 PM, Masamichi Hosoda 
wrote:

> >>>   https://bugs.freedesktop.org/show_bug.cgi?id=97546
> >>>
> >>> The proposed fix hasn't been applied yet to the fontconfig git
> >>> repository – maybe we should downgrade GUB's fontconfig version to
> >>> 2.12.0 until this is fixed.
> >>
> >> I think that there is another possible solution.  To use the
> >> proposed patch in GUB.
> >
> > Certainly.  However, I can't estimate whether there aren't side
> > effects...  On the other hand, given that the patch is from the
> > fontconfig maintainer, there are high chances that it is the right
> > one :-)
>
> I've created pull request for GUB.
> https://github.com/gperciva/gub/pull/29
>
> The patch is only applied to building binary for mac.


This is really great to see.

And accords exactly with the test results I added to the original thread
showing that 2.19.47 (and .48) are, in fact, definitely rebuilding one of
the font caches at every run of LilyPond.

Thanks very much for tracking this down!





-- 
Trevor Bača
www.trevorbaca.com
soundcloud.com/trevorbaca
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: v2.19.47 on Mac x86 (Jacques Menu Muzhic)

2016-09-25 Thread Masamichi Hosoda 
>>>   https://bugs.freedesktop.org/show_bug.cgi?id=97546
>>> 
>>> The proposed fix hasn't been applied yet to the fontconfig git
>>> repository – maybe we should downgrade GUB's fontconfig version to
>>> 2.12.0 until this is fixed.
>> 
>> I think that there is another possible solution.  To use the
>> proposed patch in GUB.
> 
> Certainly.  However, I can't estimate whether there aren't side
> effects...  On the other hand, given that the patch is from the
> fontconfig maintainer, there are high chances that it is the right
> one :-)

I've created pull request for GUB.
https://github.com/gperciva/gub/pull/29

The patch is only applied to building binary for mac.
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: v2.19.47 on Mac x86 (Jacques Menu Muzhic)

2016-09-24 Thread Werner LEMBERG 
>>   https://bugs.freedesktop.org/show_bug.cgi?id=97546
>> 
>> The proposed fix hasn't been applied yet to the fontconfig git
>> repository – maybe we should downgrade GUB's fontconfig version to
>> 2.12.0 until this is fixed.
> 
> I think that there is another possible solution.  To use the
> proposed patch in GUB.

Certainly.  However, I can't estimate whether there aren't side
effects...  On the other hand, given that the patch is from the
fontconfig maintainer, there are high chances that it is the right
one :-)


Werner
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: v2.19.47 on Mac x86 (Jacques Menu Muzhic)

2016-09-24 Thread Masamichi Hosoda 
>> Most likely cause of the issue will be the 2.12.1 change that
>> improved cache validation logic (and for yet unknown reasons always
>> invalidates the cache of the Mac OS System fonts)
>> 
>> https://www.freedesktop.org/software/fontconfig/release/ChangeLog-2.12.1
> 
> Aah, this rings a bell.
> 
>   https://bugs.freedesktop.org/show_bug.cgi?id=97546
> 
> The proposed fix hasn't been applied yet to the fontconfig git
> repository – maybe we should downgrade GUB's fontconfig version to
> 2.12.0 until this is fixed.

I think that there is another possible solution.
To use the proposed patch in GUB.
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: v2.19.47 on Mac x86 (Jacques Menu Muzhic)

2016-09-24 Thread Werner LEMBERG 
> Most likely cause of the issue will be the 2.12.1 change that
> improved cache validation logic (and for yet unknown reasons always
> invalidates the cache of the Mac OS System fonts)
> 
> https://www.freedesktop.org/software/fontconfig/release/ChangeLog-2.12.1

Aah, this rings a bell.

  https://bugs.freedesktop.org/show_bug.cgi?id=97546

The proposed fix hasn't been applied yet to the fontconfig git
repository – maybe we should downgrade GUB's fontconfig version to
2.12.0 until this is fixed.


Werner
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: v2.19.47 on Mac x86 (Jacques Menu Muzhic)

2016-09-24 Thread Hans Aikema 
Discovered already that indeed a newer version of font-config is packaged with 
2.19.47 (2.12.1 versus 2.11.95).

Most likely cause of the issue will be the 2.12.1 change that improved cache 
validation logic (and for yet unknown reasons always invalidates the cache of 
the Mac OS System fonts)

https://www.freedesktop.org/software/fontconfig/release/ChangeLog-2.12.1


> On 24 Sep 2016, at 12:55, Hans Aikema <hans.aik...@aikebah.net> wrote:
> 
> Looked a bit furher into this, and seems to be an issue of the fontconfig 
> library. Was that upgraded between the binaries of 2.19.46 and 2.19.47?
> 
> Seems that in Lilypond 2.19.47 fontconfig is no longer having a cache/dir 
> checksum match for the system font library and thus rescans it on each run to 
> build the cache. On 2.19.46 it still shows a match.
> 
> Lilypond 2.19.46:
> 
> GNU LilyPond 2.19.46
> FC_DEBUG=16
> FcCacheTimeValid dir 
> "/Users/aikebah/Downloads/LilyPond2.19.46.app/Contents/Resources/share/lilypond/current/fonts/otf"
>  cache checksum 1469570231 dir checksum 1469570231
> FcCacheTimeValid dir "/Library/Fonts" cache checksum 1458750870 dir checksum 
> 1458750870
> FcCacheTimeValid dir "/System/Library/Fonts" cache checksum 1458750878 dir 
> checksum 1458750878
> Verwerken van 'test.ly'
> Ontleden...
> Vertolken van muziek...
> Voorbewerken van grafische objecten...
> Zoeken naar het ideale aantal pagina's...Fontconfig warning: ignoring UTF-8: 
> not a valid region tag
> 
> Muziek passend maken voor 1 pagina...
> Tekenen van systemen...
> Opmaakuitvoer naar 
> '/var/folders/w1/2ppjzq0j4cj2wv3f06_rd6twgn/T//lilypond-GNcNkQ'...
> Converteren naar 'test.pdf'...
> Verwijderen van 
> '/var/folders/w1/2ppjzq0j4cj2wv3f06_rd6twgn/T//lilypond-GNcNkQ'...
> Gelukt: compilatie is met succes voltooid
> 
> 
> 
> 
> Lilypond 2.19.47:
> 
> GNU LilyPond 2.19.47
> FC_DEBUG=16
> FcCacheTimeValid dir 
> "/Users/aikebah/Downloads/LilyPond2.19.47.app/Contents/Resources/share/lilypond/current/fonts/otf"
>  cache checksum 1472556057 dir checksum 1472556057
> FcCacheTimeValid dir "/Library/Fonts" cache checksum 1458750870 dir checksum 
> 1458750870
> 
> charsets 281 -> 80 leaves 30419 -> 1687
> FcDirCacheWriteDir dir "/System/Library/Fonts" file 
> "/Users/aikebah/.lilypond-fonts.cache-2//b0a71e6bf6a8a1a908413a823d76e21f-i686-apple-darwin8.cache-7"
> Verwerken van 'test.ly'
> Ontleden...
> Vertolken van muziek...
> Voorbewerken van grafische objecten...
> Zoeken naar het ideale aantal pagina's...Fontconfig warning: ignoring UTF-8: 
> not a valid region tag
> 
> Muziek passend maken voor 1 pagina...
> Tekenen van systemen...
> Opmaakuitvoer naar 
> '/var/folders/w1/2ppjzq0j4cj2wv3f06_rd6twgn/T//lilypond-JVrgSZ'...
> Converteren naar 'test.pdf'...
> Verwijderen van 
> '/var/folders/w1/2ppjzq0j4cj2wv3f06_rd6twgn/T//lilypond-JVrgSZ'...
> Gelukt: compilatie is met succes voltooid
> 
> 
> 
>> On 05 Sep 2016, at 00:20, Hans Aikema <hans.aik...@aikebah.net> wrote:
>> 
>> On 04 Sep 2016, at 19:42, Cynthia Karl <pck...@mac.com> wrote:
>>> 
>>> 
>>>> Message: 5
>>>> Date: Sun, 4 Sep 2016 17:41:42 +0200
>>>> From: Jacques Menu Muzhic <imj-muz...@bluewin.ch>
>>>> To: Andrew Bernard <andrew.bern...@gmail.com>
>>>> Cc: Jacques Menu Muzhic <imj-muz...@bluewin.ch>,   lilypond-user
>>>><lilypond-user@gnu.org>
>>>> Subject: Re: v2.19.47 on Mac x86
>>>> I run El Capitan 10.11.6:
>>>> 
>>>> menu@macbookprojm:~/Documents/LaTeX/PartitionsLilypond > uname -a
>>>> Darwin macbookprojm 15.6.0 Darwin Kernel Version 15.6.0: Mon Aug 29 
>>>> 20:21:34 PDT 2016; root:xnu-3248.60.11~1/RELEASE_X86_64 x86_64
>>>> 
>>>> and I get:
>>>> 
>>>> menu@macbookprojm:~ > sudo dtruss lilypond --version
>>> 
>>> 
>>> I run El Capitan 10.11.6 and get the exact same output for “uname -a”.
>>> 
>>> I wanted to see what the difference was between v2.19.46 and v2.19.47, so I 
>>> ran them both on the following file:
>>> 
>>> bash-3.2$ cat 1note.ly
>>> \version "2.19.46"
>>> { c4 }
>>> 
>>> <….>
>> 
>>> I then ran dtruss -c on both versions to see what the difference in system 
>>> calls was.
>>> 
>>> The following table shows the number of system calls which have a Count > 
>>> 100 for the v2.19.47 version and the corresponding count for the v2.19.46 
>>&

Re: v2.19.47 on Mac x86 (Jacques Menu Muzhic)

2016-09-24 Thread Hans Aikema 
Looked a bit furher into this, and seems to be an issue of the fontconfig 
library. Was that upgraded between the binaries of 2.19.46 and 2.19.47?

Seems that in Lilypond 2.19.47 fontconfig is no longer having a cache/dir 
checksum match for the system font library and thus rescans it on each run to 
build the cache. On 2.19.46 it still shows a match.

Lilypond 2.19.46:

GNU LilyPond 2.19.46
FC_DEBUG=16
FcCacheTimeValid dir 
"/Users/aikebah/Downloads/LilyPond2.19.46.app/Contents/Resources/share/lilypond/current/fonts/otf"
 cache checksum 1469570231 dir checksum 1469570231
FcCacheTimeValid dir "/Library/Fonts" cache checksum 1458750870 dir checksum 
1458750870
FcCacheTimeValid dir "/System/Library/Fonts" cache checksum 1458750878 dir 
checksum 1458750878
Verwerken van 'test.ly'
Ontleden...
Vertolken van muziek...
Voorbewerken van grafische objecten...
Zoeken naar het ideale aantal pagina's...Fontconfig warning: ignoring UTF-8: 
not a valid region tag

Muziek passend maken voor 1 pagina...
Tekenen van systemen...
Opmaakuitvoer naar 
'/var/folders/w1/2ppjzq0j4cj2wv3f06_rd6twgn/T//lilypond-GNcNkQ'...
Converteren naar 'test.pdf'...
Verwijderen van 
'/var/folders/w1/2ppjzq0j4cj2wv3f06_rd6twgn/T//lilypond-GNcNkQ'...
Gelukt: compilatie is met succes voltooid




Lilypond 2.19.47:

GNU LilyPond 2.19.47
FC_DEBUG=16
FcCacheTimeValid dir 
"/Users/aikebah/Downloads/LilyPond2.19.47.app/Contents/Resources/share/lilypond/current/fonts/otf"
 cache checksum 1472556057 dir checksum 1472556057
FcCacheTimeValid dir "/Library/Fonts" cache checksum 1458750870 dir checksum 
1458750870

charsets 281 -> 80 leaves 30419 -> 1687
FcDirCacheWriteDir dir "/System/Library/Fonts" file 
"/Users/aikebah/.lilypond-fonts.cache-2//b0a71e6bf6a8a1a908413a823d76e21f-i686-apple-darwin8.cache-7"
Verwerken van 'test.ly'
Ontleden...
Vertolken van muziek...
Voorbewerken van grafische objecten...
Zoeken naar het ideale aantal pagina's...Fontconfig warning: ignoring UTF-8: 
not a valid region tag

Muziek passend maken voor 1 pagina...
Tekenen van systemen...
Opmaakuitvoer naar 
'/var/folders/w1/2ppjzq0j4cj2wv3f06_rd6twgn/T//lilypond-JVrgSZ'...
Converteren naar 'test.pdf'...
Verwijderen van 
'/var/folders/w1/2ppjzq0j4cj2wv3f06_rd6twgn/T//lilypond-JVrgSZ'...
Gelukt: compilatie is met succes voltooid



> On 05 Sep 2016, at 00:20, Hans Aikema <hans.aik...@aikebah.net> wrote:
> 
> On 04 Sep 2016, at 19:42, Cynthia Karl <pck...@mac.com> wrote:
>> 
>> 
>>> Message: 5
>>> Date: Sun, 4 Sep 2016 17:41:42 +0200
>>> From: Jacques Menu Muzhic <imj-muz...@bluewin.ch>
>>> To: Andrew Bernard <andrew.bern...@gmail.com>
>>> Cc: Jacques Menu Muzhic <imj-muz...@bluewin.ch>,lilypond-user
>>> <lilypond-user@gnu.org>
>>> Subject: Re: v2.19.47 on Mac x86
>>> I run El Capitan 10.11.6:
>>> 
>>> menu@macbookprojm:~/Documents/LaTeX/PartitionsLilypond > uname -a
>>> Darwin macbookprojm 15.6.0 Darwin Kernel Version 15.6.0: Mon Aug 29 
>>> 20:21:34 PDT 2016; root:xnu-3248.60.11~1/RELEASE_X86_64 x86_64
>>> 
>>> and I get:
>>> 
>>> menu@macbookprojm:~ > sudo dtruss lilypond --version
>> 
>> 
>> I run El Capitan 10.11.6 and get the exact same output for “uname -a”.
>> 
>> I wanted to see what the difference was between v2.19.46 and v2.19.47, so I 
>> ran them both on the following file:
>> 
>> bash-3.2$ cat 1note.ly
>> \version "2.19.46"
>> { c4 }
>> 
>> <….>
> 
>> I then ran dtruss -c on both versions to see what the difference in system 
>> calls was.
>> 
>> The following table shows the number of system calls which have a Count > 
>> 100 for the v2.19.47 version and the corresponding count for the v2.19.46 
>> version:
>> 
>> CALL COUNT LP46  COUNT LP47
>> ……   …
>> getattrlist  112 128
>> stat 178 171
>> stat64   207 207
>> sigaltstack  222 228
>> sigprocmask  263 269
>> select_nocancel  320 323
>> lseek  57  123013
>> read_nocancel341   125474
>> 
>> I then did a count of the number of lseeks on file descriptors <= 13 (at 
>> first glance there are no file descriptors greater th

Re: v2.19.47 on Mac x86

2016-09-05 Thread Hans Aikema 
> Op 5 sep. 2016 om 08:31 heeft Jacques Menu Muzhic  het 
> volgende geschreven:
> 
> Hello Andrew,
> 
> I hadn’t heard of rootless protection and csrutil prior to this post and yes, 
> I have Xcode 7.3.1 installed.
> 
> JM
> 
> 
>> Le 5 sept. 2016 à 02:33, Andrew Bernard  a écrit :
>> 
>> Hi Jacques,
>> 
>> This is puzzling. Have you explicitly turned off the rootless protection? If 
>> not, have you got XCode installed? The question is, why does your system 
>> work, rather than why does mine not work, in fact! :=) There are hundreds of 
>> threads about this rootless problem on the internet. I am not a Mac 
>> developer, so I don’t fully comprehend it, and I am not prepared to 
>> sacrifice my Mac by adjusting the csrutl settings in recovery mode to turn 
>> it all off.
>> 
>> Shoud this thread move to the devel list?
>> 
>> Andrew
>> 
>> 
>> 
>>> On 5 September 2016 at 1:41:44 AM, Jacques Menu Muzhic 
>>> (imj-muz...@bluewin.ch) wrote:
>>> 
>>> Hello Andrew,
>>> 
>>> I run El Capitan 10.11.6:
>>> 
>>> menu@macbookprojm:~/Documents/LaTeX/PartitionsLilypond > uname -a
>>> Darwin macbookprojm 15.6.0 Darwin Kernel Version 15.6.0: Mon Aug 29 
>>> 20:21:34 PDT 2016; root:xnu-3248.60.11~1/RELEASE_X86_64 x86_64
>>> 
>>> and I get:
>>> 
>>> menu@macbookprojm:~ > sudo dtruss lilypond --version
>>> 
>>> ...
>>> …
>>> 
>>> Your configuration seems to have some other problem as far as I can guess.
> 
My system is also a vanilla Mac OS X with XCode installed (no explicit security 
disabling; also hadn't heard about the rootless and csrutil until this thread 
drew my attention) . 
So Andrew, if you don't have XCode installed it might be the missing piece of 
the puzzle. dtruss and dtrace worked fine on my system using sudo

Hans Aikema

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: v2.19.47 on Mac x86

2016-09-05 Thread Jacques Menu Muzhic 
Hello Andrew,

I hadn’t heard of rootless protection and csrutil prior to this post and yes, I 
have Xcode 7.3.1 installed.

JM


> Le 5 sept. 2016 à 02:33, Andrew Bernard  a écrit :
> 
> Hi Jacques,
> 
> This is puzzling. Have you explicitly turned off the rootless protection? If 
> not, have you got XCode installed? The question is, why does your system 
> work, rather than why does mine not work, in fact! :=) There are hundreds of 
> threads about this rootless problem on the internet. I am not a Mac 
> developer, so I don’t fully comprehend it, and I am not prepared to sacrifice 
> my Mac by adjusting the csrutl settings in recovery mode to turn it all off.
> 
> Shoud this thread move to the devel list?
> 
> Andrew
> 
> 
> 
> On 5 September 2016 at 1:41:44 AM, Jacques Menu Muzhic 
> (imj-muz...@bluewin.ch) wrote:
> 
>> Hello Andrew,
>> 
>> I run El Capitan 10.11.6:
>> 
>> menu@macbookprojm:~/Documents/LaTeX/PartitionsLilypond > uname -a
>> Darwin macbookprojm 15.6.0 Darwin Kernel Version 15.6.0: Mon Aug 29 20:21:34 
>> PDT 2016; root:xnu-3248.60.11~1/RELEASE_X86_64 x86_64
>> 
>> and I get:
>> 
>> menu@macbookprojm:~ > sudo dtruss lilypond --version
>> 
>> ...
>> …
>> 
>> Your configuration seems to have some other problem as far as I can guess.
>> 
>> 
> 


___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: v2.19.47 on Mac x86

2016-09-04 Thread Andrew Bernard 
Hi Jacques,

This is puzzling. Have you explicitly turned off the rootless protection?
If not, have you got XCode installed? The question is, why does your system
work, rather than why does mine not work, in fact! :=) There are hundreds
of threads about this rootless problem on the internet. I am not a Mac
developer, so I don’t fully comprehend it, and I am not prepared to
sacrifice my Mac by adjusting the csrutl settings in recovery mode to turn
it all off.

Shoud this thread move to the devel list?

Andrew



On 5 September 2016 at 1:41:44 AM, Jacques Menu Muzhic (
imj-muz...@bluewin.ch) wrote:

Hello Andrew,

I run El Capitan 10.11.6:

menu@macbookprojm:~/Documents/LaTeX/PartitionsLilypond > uname -a
Darwin macbookprojm 15.6.0 Darwin Kernel Version 15.6.0: Mon Aug 29
20:21:34 PDT 2016; root:xnu-3248.60.11~1/RELEASE_X86_64 x86_64

and I get:

menu@macbookprojm:~ > sudo dtruss lilypond --version

...
…

Your configuration seems to have some other problem as far as I can guess.
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: v2.19.47 on Mac x86 (Jacques Menu Muzhic)

2016-09-04 Thread Hans Aikema 
On 04 Sep 2016, at 19:42, Cynthia Karl <pck...@mac.com> wrote:
> 
> 
>> Message: 5
>> Date: Sun, 4 Sep 2016 17:41:42 +0200
>> From: Jacques Menu Muzhic <imj-muz...@bluewin.ch>
>> To: Andrew Bernard <andrew.bern...@gmail.com>
>> Cc: Jacques Menu Muzhic <imj-muz...@bluewin.ch>, lilypond-user
>>  <lilypond-user@gnu.org>
>> Subject: Re: v2.19.47 on Mac x86
>> I run El Capitan 10.11.6:
>> 
>> menu@macbookprojm:~/Documents/LaTeX/PartitionsLilypond > uname -a
>> Darwin macbookprojm 15.6.0 Darwin Kernel Version 15.6.0: Mon Aug 29 20:21:34 
>> PDT 2016; root:xnu-3248.60.11~1/RELEASE_X86_64 x86_64
>> 
>> and I get:
>> 
>> menu@macbookprojm:~ > sudo dtruss lilypond --version
> 
> 
> I run El Capitan 10.11.6 and get the exact same output for “uname -a”.
> 
> I wanted to see what the difference was between v2.19.46 and v2.19.47, so I 
> ran them both on the following file:
> 
> bash-3.2$ cat 1note.ly
> \version "2.19.46"
> { c4 }
> 
> <….>

> I then ran dtruss -c on both versions to see what the difference in system 
> calls was.
> 
> The following table shows the number of system calls which have a Count > 100 
> for the v2.19.47 version and the corresponding count for the v2.19.46 version:
> 
> CALL  COUNT LP46  COUNT LP47
> … …   …
> getattrlist   112 128
> stat  178 171
> stat64207 207
> sigaltstack   222 228
> sigprocmask   263 269
> select_nocancel   320 323
> lseek   57  123013
> read_nocancel 341   125474
> 
> I then did a count of the number of lseeks on file descriptors <= 13 (at 
> first glance there are no file descriptors greater than 12:
> 
> lseek(0xfiledes   v46 v47
> 
> lseek(0x0  1   23
> lseek(0x1  1 1
> lseek(0x2  1 1
> lseek(0x3  2 2
> lseek(0x4  0 0
> lseek(0x5  0 0
> lseek(0x6  2 2
> lseek(0x7 35   35
> lseek(0x8   8 122969
> lseek(0x9   3   3
> lseek(0xA   11
> lseek(0xB   33
> lseek(0xC   00
> 
> Hopefully, this information might be of use to someone trying to figure this 
> problem out.
> 
> The most salient questions would be:  what is filedes 8 used for?  And why is 
> this problem only occurring on Macs?
> 
> 
> 
> 
> 
> ___
> lilypond-user mailing list
> lilypond-user@gnu.org
> https://lists.gnu.org/mailman/listinfo/lilypond-user

I did a dtrace on 2.19.47, using the default introductionary song that Lilypond 
uses as a starting sample (when starting the Lilypond app)

#!/usr/sbin/dtrace -s
syscall::lseek:entry /execname == "lilypond"/ { printf("lseek for %s ", 
execname); }
syscall::open*:entry /execname == "lilypond"/ { printf("%s %s", execname, 
copyinstr(arg0)); }


Searching within the created log-file of all dtrace hits for "open", so that 
only the open* entries are found shows that apparently the lseek-explosion 
occurs while lilypond is scanning the Mac OS X Fonts (this was not the first 
run of 2.19.47, so the initial font-scanning of the first run can not be blamed 
(unless the changes from 2.19.46 to 2.19.47 broke the font-caching system for 
Mac OS X of course)):

CC-ing lilypond-devel as this may provide useful information for them too


  0151   open:entry lilypond 
/Users/aikebah/Downloads/LilyPond2.19.47.app/Contents/Resources/bin/../lib/libintl.8.dylib
<..some additional open/open_nocancel calls left out..>
  0937  open_nocancel:entry lilypond 
/Users/aikebah/Downloads/LilyPond2.19.47.app/Contents/Resources/share/lilypond/current/fonts/otf
> LINE 283 0937  open_nocancel:entry lilypond 
/Users/aikebah/.lilypond-fonts.cache-2//33296e741d33efa1650f388486e232c6-i686-apple-darwin8.cache-7
> LINE 284 0937  open_nocancel:entry lilypond /Library/Fonts
> LINE 285 0937  open_nocancel:entry lilypond 
/

Fwd: Re: v2.19.47 on Mac x86

2016-09-04 Thread Patrick Karl 




 Forwarded Message 
Subject:Re: v2.19.47 on Mac x86
Date:   Sun, 4 Sep 2016 15:16:03 -0500
From:   Patrick Karl <jpk...@gmail.com>
To: Jacques Menu Muzhic <imj-muz...@bluewin.ch>





Message: 5
Date: Sun, 4 Sep 2016 17:41:42 +0200
From: Jacques Menu Muzhic <imj-muz...@bluewin.ch>
To: Andrew Bernard <andrew.bern...@gmail.com>
Cc: Jacques Menu Muzhic <imj-muz...@bluewin.ch>,  lilypond-user
<lilypond-user@gnu.org>
Subject: Re: v2.19.47 on Mac x86
I run El Capitan 10.11.6:

menu@macbookprojm:~/Documents/LaTeX/PartitionsLilypond > uname -a
Darwin macbookprojm 15.6.0 Darwin Kernel Version 15.6.0: Mon Aug 29
20:21:34 PDT 2016; root:xnu-3248.60.11~1/RELEASE_X86_64 x86_64

and I get:

menu@macbookprojm:~ > sudo dtruss lilypond --version


I run El Capitan 10.11.6 and get the exact same output for “uname -a”.

I wanted to see what the difference was between v2.19.46 and v2.19.47,
so I ran them both on the following file:

bash-3.2$ cat 1note.ly
\version "2.19.46"
{ c4 }

bash-3.2$ time lilypond 1note.ly
GNU LilyPond 2.19.46
Processing `1note.ly'
Parsing...
Interpreting music...
Preprocessing graphical objects...
Finding the ideal number of pages...
Fitting music on 1 page...
Drawing systems...
Layout output to `/tmp/lilypond-22EfXs'...
Converting to `1note.pdf'...
Deleting `/tmp/lilypond-22EfXs'...
Success: compilation successfully completed

real0m0.820s
user0m0.638s
sys0m0.093s


bash-3.2$ LP47=$l47
bash-3.2$ echo $LP47
/Users/pat/Downloads/LilyPond.app/Contents/Resources/bin/lilypond
bash-3.2$ time $LP47 1note.ly
GNU LilyPond 2.19.47
Processing `1note.ly'
Parsing...
Interpreting music...
Preprocessing graphical objects...
Finding the ideal number of pages...
Fitting music on 1 page...
Drawing systems...
Layout output to `/tmp/lilypond-B2ws6y'...
Converting to `1note.pdf'...
Deleting `/tmp/lilypond-B2ws6y'...
Success: compilation successfully completed

real1m2.364s
user0m46.848s
sys0m15.145s

I then ran dtruss -c on both versions to see what the difference in
system calls was.

The following table shows the number of system calls which have a Count

100 for the v2.19.47 version and the corresponding count for the

v2.19.46 version:

CALLCOUNT LP46COUNT LP47
………
getattrlist112128
stat178171
stat64207207
sigaltstack222228
sigprocmask263269
select_nocancel320323
lseek  57  123013
read_nocancel341  125474

I then did a count of the number of lseeks on file descriptors <= 13 (at
first glance there are no file descriptors greater than 12:

lseek(0xfiledesv46v47

lseek(0x0  1  23
lseek(0x1  11
lseek(0x2  11
lseek(0x3  22
lseek(0x4  00
lseek(0x5  0   0
lseek(0x6  22
lseek(0x7 35  35
lseek(0x8   8 122969
lseek(0x9   3   3
lseek(0xA   11
lseek(0xB   33
lseek(0xC   00

Hopefully, this information might be of use to someone trying to figure
this problem out.

The most salient questions would be:  what is filedes 8 used for? And
why is this problem only occurring on Macs?



___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: v2.19.47 on Mac x86

2016-09-04 Thread Jacques Menu Muzhic 
Hello Andrew,

I run El Capitan 10.11.6:

menu@macbookprojm:~/Documents/LaTeX/PartitionsLilypond > uname -a
Darwin macbookprojm 15.6.0 Darwin Kernel Version 15.6.0: Mon Aug 29 20:21:34 
PDT 2016; root:xnu-3248.60.11~1/RELEASE_X86_64 x86_64

and I get:

menu@macbookprojm:~ > sudo dtruss lilypond --version

GNU LilyPond 2.19.44

Copyright (c) 1996--2015 by
  Han-Wen Nienhuys 
  Jan Nieuwenhuizen 
  and others.

This program is free software.  It is covered by the GNU General Public
License and you are welcome to change it and/or distribute copies of it
under certain conditions.  Invoke as `lilypond --warranty' for more
information.

SYSCALL(args)= return
thread_selfid(0x0, 0x0, 0x0) = 124530 0
csops(0x0, 0x0, 0xBFFFE7E0)  = 0 0
issetugid(0x0, 0x0, 0xBFFFE7E0)  = 0 0
shared_region_check_np(0xBFFFC740, 0x0, 0xBFFFE7E0)  = 0 0
stat64("/usr/lib/dtrace/libdtrace_dyld.dylib\0", 0xBFFFDD78, 0xBFFFE7E0)
 = 0 0
open("/usr/lib/dtrace/libdtrace_dyld.dylib\0", 0x0, 0x0) = 3 0
pread(0x3, "\312\376\272\276\0", 0x1000, 0x0)= 4096 0
pread(0x3, "\316\372\355\376\a\0", 0x1000, 0x1000)   = 4096 0
fcntl(0x3, 0x61, 0xBFFFC250) = 0 0
mmap(0x0, 0x564, 0x5, 0x1, 0x3, 0x1000)  = 0x7A8000 0
munmap(0x7A8000, 0x564)  = 0 0
mmap(0x7A8000, 0x2000, 0x1, 0x12, 0x3, 0x1000)   = 0x7A8000 0
mmap(0x7AA000, 0x1000, 0x3, 0x12, 0x3, 0x3000)   = 0x7AA000 0
mmap(0x7AB000, 0x2F10, 0x1, 0x12, 0x3, 0x4000)   = 0x7AB000 0
close(0x3)   = 0 0
stat64("/usr/lib/dtrace/libdtrace_dyld.dylib\0", 0xBFFFE290, 0x1)   
 = 0 0
…

Your configuration seems to have some other problem as far as I can guess.

JM


> Le 4 sept. 2016 à 15:56, Andrew Bernard  a écrit :
> 
> Hi Jacques,
> 
> Sorry to be repetitive. What OS are you running?
> 
> On 10.11.6 El Capitan, this is the scenario:
> 
> sudo dtruss lilypond --version
> Password:
> 
> dtrace: failed to execute lilypond: dtrace cannot control executables signed 
> with restricted entitlements
> 
> It all about this hard to understand ‘rootless’ system that Apple have 
> introduced. It seems you have to adjust things with csrutil, but I cannot get 
> that to work.
> 
> Andrew
> 
> 
> 
> 
> On 4 September 2016 at 11:39:19 PM, Jacques Menu Muzhic 
> (imj-muz...@bluewin.ch ) wrote:
> 
>> Hello Andrew,
>> 
>> Here is what I got spying a running Frescobaldi instance when I closed an 
>> open file:
>> 
>> menu@macbookprojm:~/Documents/LaTeX/PartitionsLilypond > sudo dtruss -p 1273
>> 

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: v2.19.47 on Mac x86

2016-09-04 Thread Br. Samuel Springuel 
It is possible to disable SIP (System Integrity Protection, the official 
name for Mac's "rootless" system).  You can find instructions here:


https://developer.apple.com/library/mac/documentation/Security/Conceptual/System_Integrity_Protection_Guide/ConfiguringSystemIntegrityProtection/ConfiguringSystemIntegrityProtection.html

Obviously Apple recommends not doing this as a general principle, but 
disabling it should enable you to run the traces you want to run in the 
same way you'd do that on any other Unix machine.

--
✝
Br. Samuel, OSB
St. Anselm’s Abbey
Washington, DC
(R. Padraic Springuel)

PAX ☧ ΧΡΙΣΤΟΣ

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: v2.19.47 on Mac x86

2016-09-04 Thread Andrew Bernard 
Hi Jacques,

Sorry to be repetitive. What OS are you running?

On 10.11.6 El Capitan, this is the scenario:

sudo dtruss lilypond --version

Password:


dtrace: failed to execute lilypond: dtrace cannot control executables
signed with restricted entitlements

It all about this hard to understand ‘rootless’ system that Apple have
introduced. It seems you have to adjust things with csrutil, but I cannot
get that to work.

Andrew




On 4 September 2016 at 11:39:19 PM, Jacques Menu Muzhic (
imj-muz...@bluewin.ch) wrote:

Hello Andrew,

Here is what I got spying a running Frescobaldi instance when I closed an
open file:

menu@macbookprojm:~/Documents/LaTeX/PartitionsLilypond > sudo dtruss -p 1273
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: v2.19.47 on Mac x86

2016-09-04 Thread Jacques Menu Muzhic 
Hello Andrew,

Here is what I got spying a running Frescobaldi instance when I closed an open 
file:

menu@macbookprojm:~/Documents/LaTeX/PartitionsLilypond > sudo dtruss -p 1273

SYSCALL(args)= return
workq_kernreturn(0x20, 0x0, 0x1) = 0 0
thread_selfid(0x0, 0x0, 0x0) = 85229 0
bsdthread_ctl(0x100, 0x8FF, 0xE8CF)  = 0 0
workq_kernreturn(0x20, 0x0, 0x1) = 0 0
bsdthread_ctl(0x100, 0x8FF, 0x0) = 0 0
thread_selfid(0x0, 0x0, 0x0) = 85234 0
bsdthread_ctl(0x100, 0x10FF, 0xE8CF) = 0 0
bsdthread_ctl(0x100, 0x10FF, 0x0)= 0 0
bsdthread_ctl(0x100, 0x0, 0xE8D3)= 0 0
bsdthread_ctl(0x100, 0x0, 0x0)   = 0 0
bsdthread_ctl(0x100, 0x0, 0x10957)   = 0 0
bsdthread_ctl(0x100, 0x0, 0x0)   = 0 0
workq_kernreturn(0x20, 0x0, 0x1) = 0 0
workq_kernreturn(0x20, 0x0, 0x1) = 0 0
kevent_qos(0x6, 0x7FFF8EC7B1E8, 0x1) = 0 0
kevent_qos(0x6, 0x7FFF78D81B08, 0x1) = 1 0
kevent_qos(0x6, 0x7FFF8EC7B1E8, 0x1) = 0 0
kevent_qos(0x6, 0x7FFF8EC7B1E8, 0x1) = 0 0
kevent_qos(0x6, 0x7FFF78D81A20, 0x1) = 0 0
kevent_qos(0x6, 0x0, 0x0)= 1 0
kevent_qos(0x6, 0x0, 0x0)= 1 0
workq_kernreturn(0x20, 0x0, 0x1) = 0 0
bsdthread_ctl(0x100, 0x0, 0x12C87)   = 0 0
workq_kernreturn(0x20, 0x0, 0x1) = 0 0
bsdthread_ctl(0x100, 0x8FF, 0x12C87) = 0 0
kevent_qos(0x6, 0x7FFF8EC7B1E8, 0x1) = 0 0
kevent_qos(0x6, 0x0, 0x0)= 1 0
bsdthread_ctl(0x100, 0x8FF, 0x0) = 0 0
bsdthread_ctl(0x100, 0x8FF, 0x0) = 0 0
kevent_qos(0x6, 0x0, 0x0)= 1 0
workq_kernreturn(0x20, 0x0, 0x1) = 0 0
bsdthread_ctl(0x100, 0x0, 0x12C8B)   = 0 0
workq_kernreturn(0x20, 0x0, 0x1) = 0 0
kevent_qos(0x6, 0x7FFF8EC7B1E8, 0x1) = 0 0
bsdthread_ctl(0x100, 0x8FF, 0x12C8B) = 0 0
kevent_qos(0x6, 0x0, 0x0)= 1 0
bsdthread_ctl(0x100, 0x8FF, 0x0) = 0 0
bsdthread_ctl(0x100, 0x8FF, 0x0) = 0 0
madvise(0x105366000, 0x3, 0x9)   = 0 0
stat64("/var/folders/jc/xrpy67_x6_vcjfzpzds_9_6mgn/T/frescobaldi-WJRNtP/tmpf0bsGM/document.pdf\0",
 0x7FFF5FBF61E0, 0x9)  = 0 0
access("/var/folders/jc/xrpy67_x6_vcjfzpzds_9_6mgn/T/frescobaldi-WJRNtP/tmpf0bsGM/document.pdf\0",
 0x1, 0x9) = -1 Err#13
lstat64("/var/folders/jc/xrpy67_x6_vcjfzpzds_9_6mgn/T/frescobaldi-WJRNtP/tmpf0bsGM/document.pdf\0",
 0x7FFF5FBF61E0, 0x9) = 0 0
getattrlist("/var/folders/jc/xrpy67_x6_vcjfzpzds_9_6mgn/T/frescobaldi-WJRNtP/tmpf0bsGM/document.pdf\0",
 0x7FFF5FBF4A08, 0x7FFF5FBF4670)  = 0 0
geteuid(0x1006A2948, 0x7FFF5FBF4A08, 0x7FFF5FBF4670) = 501 0
geteuid(0x1006A2948, 0x7FFF5FBF4A08, 0x7FFF5FBF4670) = 501 0
geteuid(0x1006A2948, 0x7FFF5FBF4A08, 0x7FFF5FBF4670) = 501 0
fsgetpath(0x7FFF5FBF36E0, 0x400, 0x7FFF5FBF3268) = 99 0
getattrlist("/private/var/folders/jc/xrpy67_x6_vcjfzpzds_9_6mgn/T/frescobaldi-WJRNtP/tmpf0bsGM/document.pdf\0",
 0x7FFF5FBF3F00, 0x7FFF5FBF3B30)  = 0 0
…


> Le 4 sept. 2016 à 14:26, Andrew Bernard  a écrit :
> 
> Hello Jacques,
> 
> Have you been able to run these commands on El Capitan? That’s the issue. The 
> ‘rootless’ envrionemnt prevents root from havng superuser powers. [I do not 
> appreciate this restriction of normal UNIX capabilities.]
> 
> 
> Andrew
> 
> 
> 
> On 4 September 2016 at 5:14:02 PM, Jacques Menu Muzhic (imj-muz...@bluewin.ch 
> ) wrote:
> 
>> Hello Andrew,
>> 
>> I found information at :
>> 
>> https://8thlight.com/blog/colin-jones/2015/11/06/dtrace-even-better-than-strace-for-osx.html
>>  
>> 
>> 
>> The dtruce and dtrace options lists are shown below. They require sudo right.

___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: v2.19.47 on Mac x86

2016-09-04 Thread Andrew Bernard 
Hello Jacques,

Have you been able to run these commands on El Capitan? That’s the issue.
The ‘rootless’ envrionemnt prevents root from havng superuser powers. [I do
not appreciate this restriction of normal UNIX capabilities.]


Andrew



On 4 September 2016 at 5:14:02 PM, Jacques Menu Muzhic (
imj-muz...@bluewin.ch) wrote:

Hello Andrew,

I found information at :

https://8thlight.com/blog/colin-jones/2015/11/06/dtrace-even-better-than-strace-for-osx.html

The dtruce and dtrace options lists are shown below. They require sudo
right.
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

Re: v2.19.47 on Mac x86

2016-09-04 Thread Jacques Menu Muzhic 
Hello Andrew,

I found information at :


https://8thlight.com/blog/colin-jones/2015/11/06/dtrace-even-better-than-strace-for-osx.html
 


The dtruce and dtrace options lists are shown below. They require sudo right.

JM

--

menu@macbookprojm:~ > dtruss 
USAGE: dtruss [-acdefholLs] [-t syscall] { -p PID | -n name | command }

  -p PID  # examine this PID
  -n name # examine this process name
  -t syscall  # examine this syscall only
  -a  # print all details
  -c  # print syscall counts
  -d  # print relative times (us)
  -e  # print elapsed times (us)
  -f  # follow children
  -l  # force printing pid/lwpid
  -o  # print on cpu times
  -s  # print stack backtraces
  -L  # don't print pid/lwpid
  -b bufsize  # dynamic variable buf size
   eg,
   dtruss df -h   # run and examine "df -h"
   dtruss -p 1871 # examine PID 1871
   dtruss -n tar  # examine all processes called "tar"
   dtruss -f test.sh  # run test.sh and follow children
menu@macbookprojm:~ > dtrace

Usage: dtrace [-aACeFHlqSvVwZ] [-arch i386|x86_64] [-b bufsz] [-c cmd] [-D 
name[=def]]
[-I path] [-L path] [-o output] [-p pid] [-s script] [-U name]
[-x opt[=val]]

[-P provider [[ predicate ] action ]]
[-m [ provider: ] module [[ predicate ] action ]]
[-f [[ provider: ] module: ] func [[ predicate ] action ]]
[-n [[[ provider: ] module: ] func: ] name [[ predicate ] action ]]
[-i probe-id [[ predicate ] action ]] [ args ... ]

predicate -> '/' D-expression '/'
   action -> '{' D-statements '}'

-arch Generate programs and Mach-O files for the specified architecture

-a  claim anonymous tracing state
-A  generate plist(5) entries for anonymous tracing
-b  set trace buffer size
-c  run specified command and exit upon its completion
-C  run cpp(1) preprocessor on script files
-D  define symbol when invoking preprocessor
-e  exit after compiling request but prior to enabling probes
-f  enable or list probes matching the specified function name
-F  coalesce trace output by function
-h  generate a header file with definitions for static probes
-H  print included files when invoking preprocessor
-i  enable or list probes matching the specified probe id
-I  add include directory to preprocessor search path
-l  list probes matching specified criteria
-L  add library directory to library search path
-m  enable or list probes matching the specified module name
-n  enable or list probes matching the specified probe name
-o  set output file
-p  grab specified process-ID and cache its symbol tables
-P  enable or list probes matching the specified provider name
-q  set quiet mode (only output explicitly traced data)
-s  enable or list probes according to the specified D script
-S  print D compiler intermediate code
-U  undefine symbol when invoking preprocessor
-v  set verbose mode (report stability attributes, arguments)
-V  report DTrace API version
-w  permit destructive actions
-W  wait for specified process and exit upon its completion
-x  enable or modify compiler and tracing options
-Z  permit probe descriptions that match zero probes
menu@macbookprojm:~ > 


> Le 4 sept. 2016 à 03:04, Andrew Bernard  a écrit :
> 
> Hi Stan,
> 
> I have confirmed this and posted on the devel list. The suggestion was to run 
> a trace on it but on Mac El Capitan this is really hard to to the new 
> 'rootless' system where even root cannot trace things without going through 
> great complexity, which even as a UNIX developer myself I cannot figure out.
> 
> My request for Mac developers to give some advice on the devel list has so 
> far had nil response.
> 
> The time to compile a single note file is over forty seconds every time I 
> run. The 2.19.47 release is therefore unusable by any standard.
> 
> Previously, the leading edge releases have been very stable and always 
> anticipated. Recently, they have become problematic and buggy. I am not what 
> has led to this decline. Even Mr Kastrup himself has made a comment along the 
> same lines. In the past I have always recommended going with the latest dev 
> releases to people. This no longer seems like wise advice. Hopefully things 
> will revert to stability again in the future.
> 
> Are there any Mac developers on the user list here who know how to trace a 
> process on El Capitan?
> 
> Andrew
> 
> 
> On 4 September

Re: v2.19.47 on Mac x86

2016-09-03 Thread Andrew Bernard 
Hi Stan,

I have confirmed this and posted on the devel list. The suggestion was to
run a trace on it but on Mac El Capitan this is really hard to to the new
'rootless' system where even root cannot trace things without going through
great complexity, which even as a UNIX developer myself I cannot figure out.

My request for Mac developers to give some advice on the devel list has so
far had nil response.

The time to compile a single note file is over forty seconds every time I
run. The 2.19.47 release is therefore unusable by any standard.

Previously, the leading edge releases have been very stable and always
anticipated. Recently, they have become problematic and buggy. I am not
what has led to this decline. Even Mr Kastrup himself has made a comment
along the same lines. In the past I have always recommended going with the
latest dev releases to people. This no longer seems like wise advice.
Hopefully things will revert to stability again in the future.

Are there any Mac developers on the user list here who know how to trace a
process on El Capitan?

Andrew


On 4 September 2016 at 05:03, Stan Sanderson  wrote:

>
> As reported on the Developer list, wait times for compilation on my OS
> 10.11.6 machines are excessive with LilyPond 2.19.47.
>
> The wait time “feels” longer than the normal first run font cache build
> wait, but I have no objective data.
>
> I’ve reverted  to v 2.19.46.
> ___
> lilypond-user mailing list
> lilypond-user@gnu.org
> https://lists.gnu.org/mailman/listinfo/lilypond-user
>
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user

v2.19.47 on Mac x86

2016-09-03 Thread Stan Sanderson 
 
As reported on the Developer list, wait times for compilation on my OS 10.11.6 
machines are excessive with LilyPond 2.19.47. 

The wait time “feels” longer than the normal first run font cache build wait, 
but I have no objective data.

I’ve reverted  to v 2.19.46.  
___
lilypond-user mailing list
lilypond-user@gnu.org
https://lists.gnu.org/mailman/listinfo/lilypond-user