Re: firefox 52 - crashes at startup or later due to oom

[Sebastien Marie]

On Mon, Mar 13, 2017 at 11:19:09PM +0100, Peter N. M. Hansteen wrote:
> 
> On 03/13/17 08:01, Landry Breuil wrote:
> >> Alternatively, I would be glad if someone could test the diff below: I
> >> run with it, but I don't use javascript enough to be sure the allocation
> >> isn't too low.
> >>
> >> It makes the allocation to be 128 Mo instead of 1 Go on 64bits platform
> >> (it is the same value than for 32 bits platform in fact).
> > 
> > Yet, firefox is unusable on 32-bit hardware. Try *running* it on an atom
> > netbook with 1Gb RAM... so taking the same value as 32 bits platforms
> > isnt an argument :)
> > 
> >>From what i read in the various related bugs, the size was increased to
> > fix memory allocation crashes when running WASM (the new 'assembly in
> > javascript' standard) and i think to increase randomization of memory
> > addresses allocated. *shrug*
> > 
> > For us, i think coming back to 640M is the "safest" solution.
> 
> I've been seeing the same symptoms (firefox-52 crashes with various out
> of memory errors after a few minutes).
> 
> What is the most useful way to help debug this? Tweaking staff's
> stacksize-cur upwards, or some other value?
> 

the root cause of the segfault is known: firefox allocate a big chunk of
1Go for javascript, and still allocate some memory for others parts of
firefox. so the program will hit RLIMIT_DATA very quickly.

so you had to adjust the datastack-cur limit in your login-class (staff
or default depending the user).

I experimented several things:
  - make firefox allocate only 128 Mo for javascript. it works for me,
but I am not a big user of javascript.

  - make firefox allocate only 640 Mo for javascript (as landry@
proposed). it should make firefox usable in staff class, but it is
still too much for default class (reminder: the 640 Mo is for
javascript only, the application still do malloc() for others parts
of firefox).

Personally, what bother me is the fact that datastack limit is the same
for all processes, without convenient way (for my knowledge) to run some
programs with alternate class (and bigger datastack).

Else, I would have a "default" login-class with something like 256 Mo of
datastack, and running firefox or libreoffice with alternate login-class
with more datastack (like 2 Go).

I only know su(1) to be able to change the login-class, and the
parameter is reserved to root (and it is normal). So if I want to use it
I have to do something like:

semarie@foo$ doas su -c bigmem semarie -c "firefox"

And I am not very confident for the use of doas(1) and su(1) in the same
command-line...


Alternatively, having a small datastack-cur and a bigger datastack-max
is a possibility, and setting datastack-cur to datastack-max before
running firefox (ulimit -d `ulimit -dH`).


If people have others suggestions, I would be glad to hear them :)

-- 
Sebastien Marie

CVS: cvs.openbsd.org: ports

[Jeremie Courreges-Anglas]

CVSROOT:/cvs
Module name:ports
Changes by: j...@cvs.openbsd.org2017/03/13 20:57:27

Modified files:
graphics/libraw: Makefile 

Log message:
Use lang/gcc for atomic builtins on arm

CVS: cvs.openbsd.org: ports

[Jeremie Courreges-Anglas]

CVSROOT:/cvs
Module name:ports
Changes by: j...@cvs.openbsd.org2017/03/13 20:43:18

Added files:
misc/zzuf/patches: patch-configure 

Log message:
Honour CFLAGS

CVS: cvs.openbsd.org: ports

[Jeremie Courreges-Anglas]

CVSROOT:/cvs
Module name:ports
Changes by: j...@cvs.openbsd.org2017/03/13 20:36:36

Modified files:
misc/zzuf  : Makefile 

Log message:
Use lang/gcc on arm

Re: Update to weechat-1.7

[Daniel Jakots]

On Sun, 12 Mar 2017 20:15:42 -0400, Daniel Jakots 
wrote:

> Hi,
> 
> Here's a diff to update to latest weechat release.
> 
> Changelog: https://weechat.org/files/changelog/ChangeLog-1.7.html
> 
> Comments? OK?

Updated diff following a review from abieber@

There's still:
weechat-ruby-1.7(net/weechat,-ruby):
Extra:  c.89 gmp.10 m.10 pthread.23

but I have the same for textproc/redland-bindings:

ruby-librdf-1.0.17.1p2(textproc/redland-bindings,-ruby):
Extra:  c.89 gmp.10 m.10 pthread.23
Scanning: ok
*** Error 1 in /usr/ports/textproc/redland-bindings

So I'm not really sure where the problem lies

Index: Makefile
===
RCS file: /cvs/ports/net/weechat/Makefile,v
retrieving revision 1.26
diff -u -p -r1.26 Makefile
--- Makefile6 Sep 2016 15:37:53 -   1.26
+++ Makefile14 Mar 2017 01:43:18 -
@@ -6,7 +6,7 @@ COMMENT-python= Python bindings for weec
 COMMENT-ruby=  Ruby bindings for weechat
 COMMENT-tcl=   Tcl bindings for weechat
 
-V= 1.5
+V= 1.7
 DISTNAME=  weechat-${V}
 
 PKGNAME-main=  weechat-${V}
@@ -15,8 +15,6 @@ PKGNAME-python=   weechat-python-${V}
 PKGNAME-ruby=  weechat-ruby-${V}
 PKGNAME-tcl=   weechat-tcl-${V}
 
-REVISION-ruby =0
-
 CATEGORIES=net
 
 HOMEPAGE=  http://www.weechat.org/
@@ -30,7 +28,7 @@ MASTER_SITES= ${HOMEPAGE}/files/src/
 
 MULTI_PACKAGES=-main -lua -python -ruby -tcl
 
-WANTLIB-main=  c curl m ncursesw perl pthread util \
+WANTLIB-main=  c curl m ncursesw perl pthread \
gcrypt gnutls gpg-error z ${MODGETTEXT_WANTLIB}
 
 WANTLIB-lua=   ${MODLUA_WANTLIB} m
@@ -88,7 +86,7 @@ pre-configure:
 
 post-install:
$(INSTALL_DATA_DIR) ${PREFIX}/share/doc/weechat
-   $(INSTALL_DATA) $(WRKSRC)/doc/en/weechat_*.en.asciidoc \
+   $(INSTALL_DATA) $(WRKSRC)/doc/en/weechat_*.en.adoc \
${PREFIX}/share/doc/weechat/
 
 NO_TEST=   Yes
Index: distinfo
===
RCS file: /cvs/ports/net/weechat/distinfo,v
retrieving revision 1.11
diff -u -p -r1.11 distinfo
--- distinfo11 May 2016 00:31:47 -  1.11
+++ distinfo14 Mar 2017 01:43:18 -
@@ -1,2 +1,2 @@
-SHA256 (weechat-1.5.tar.gz) = MXRVhVaiCuj57jq79mt9QrZX0zcDIlVVAacH4znhB3E=
-SIZE (weechat-1.5.tar.gz) = 3582515
+SHA256 (weechat-1.7.tar.gz) = 12GqUBQcvV8+N1Zlx7hO5dpFFu5GeDFjO2itvDdNQus=
+SIZE (weechat-1.7.tar.gz) = 3650453
Index: patches/patch-doc_CMakeLists_txt
===
RCS file: /cvs/ports/net/weechat/patches/patch-doc_CMakeLists_txt,v
retrieving revision 1.6
diff -u -p -r1.6 patch-doc_CMakeLists_txt
--- patches/patch-doc_CMakeLists_txt11 Jan 2016 15:11:31 -  1.6
+++ patches/patch-doc_CMakeLists_txt14 Mar 2017 01:43:18 -
@@ -1,25 +1,28 @@
 doc/CMakeLists.txt.origSun Jan 10 00:17:50 2016
-+++ doc/CMakeLists.txt Sun Jan 10 06:09:03 2016
-@@ -45,14 +45,14 @@ if(ENABLE_MAN OR ENABLE_DOC)
- set(ASCIIDOC_ARGS_DEV${ASCIIDOC_ARGS} ${CSS_TOC2}   )
+$OpenBSD$
+--- doc/CMakeLists.txt.origSun Jan 15 01:41:25 2017
 doc/CMakeLists.txt Sun Mar 12 20:36:40 2017
+@@ -26,15 +26,15 @@ if(ENABLE_MAN OR ENABLE_DOC)
  
- add_subdirectory(en)
+ set(ASCIIDOCTOR_ARGS -a experimental -a icons=font -a 
revnumber="${VERSION}" -a sectanchors -a source-highlighter=prettify)
+ 
+-add_subdirectory(cs)
+-add_subdirectory(de)
+-add_subdirectory(en)
+-add_subdirectory(es)
 -add_subdirectory(fr)
 -add_subdirectory(it)
--add_subdirectory(de)
+-add_subdirectory(ja)
 -add_subdirectory(pl)
--add_subdirectory(es)
 -add_subdirectory(ru)
--add_subdirectory(ja)
--add_subdirectory(cs)
-+#add_subdirectory(fr)
-+#add_subdirectory(it)
++#add_subdirectory(cs)
 +#add_subdirectory(de)
-+#add_subdirectory(pl)
++#add_subdirectory(en)
 +#add_subdirectory(es)
-+#add_subdirectory(ru)
++#add_subdirectory(fr)
++#add_subdirectory(it)
 +#add_subdirectory(ja)
-+#add_subdirectory(cs)
++#add_subdirectory(pl)
++##add_subdirectory(ru)
  
endif()
  
Index: patches/patch-src_gui_curses_CMakeLists_txt
===
RCS file: /cvs/ports/net/weechat/patches/patch-src_gui_curses_CMakeLists_txt,v
retrieving revision 1.5
diff -u -p -r1.5 patch-src_gui_curses_CMakeLists_txt
--- patches/patch-src_gui_curses_CMakeLists_txt 5 Oct 2015 11:11:47 -   
1.5
+++ patches/patch-src_gui_curses_CMakeLists_txt 14 Mar 2017 01:43:18 -
@@ -2,9 +2,9 @@ $OpenBSD: patch-src_gui_curses_CMakeList
 
 Fix cyclical dependency error
 
 src/gui/curses/CMakeLists.txt.orig Mon Oct  5 13:04:20 2015
-+++ src/gui/curses/CMakeLists.txt  Mon Oct  5 13:05:21 2015
-@@ -95,10 +95,10 @@ add_custom_command(
+--- src/gui/curses/CMakeLists.txt.orig Sun Jan 15 01:41:25 2017
 

CVS: cvs.openbsd.org: ports

[Jeremie Courreges-Anglas]

CVSROOT:/cvs
Module name:ports
Changes by: j...@cvs.openbsd.org2017/03/13 19:35:52

Modified files:
net/ucspi-tools: Makefile 
Added files:
net/ucspi-tools/patches: patch-config_mk 

Log message:
Honour CC/CFLAGS

CVS: cvs.openbsd.org: ports

[Daniel Jakots]

CVSROOT:/cvs
Module name:ports
Changes by: d...@cvs.openbsd.org2017/03/13 19:01:38

Modified files:
security/py-crypto: Tag: OPENBSD_6_0 Makefile 
Added files:
security/py-crypto/patches: Tag: OPENBSD_6_0 
patch-lib_Crypto_SelfTest_Cipher_common_py 
patch-src_block_template_c 

Log message:
Backport patches adapted from upstream by Sevan Janiyan for CVE-2013-7459

ok jca@

CVS: cvs.openbsd.org: ports

[Daniel Jakots]

CVSROOT:/cvs
Module name:ports
Changes by: d...@cvs.openbsd.org2017/03/13 18:56:48

Modified files:
security/py-crypto: Makefile 
Added files:
security/py-crypto/patches: 
patch-lib_Crypto_SelfTest_Cipher_common_py 
patch-src_block_template_c 

Log message:
Backport patches adapted from upstream by Sevan Janiyan for CVE-2013-7459

ok jca@

CVS: cvs.openbsd.org: ports

[Daniel Jakots]

CVSROOT:/cvs
Module name:ports
Changes by: d...@cvs.openbsd.org2017/03/13 18:43:28

Modified files:
security/py-crypto/patches: patch-lib_Crypto_Hash___init___py 

patch-lib_Crypto_SelfTest_Cipher___init___py 
patch-lib_Crypto_SelfTest_Hash___init__py 

Log message:
Regen patches

prodded by jca@

CVS: cvs.openbsd.org: ports

[Giovanni Bechis]

CVSROOT:/cvs
Module name:ports
Changes by: giova...@cvs.openbsd.org2017/03/13 16:25:14

Modified files:
net/unison : Makefile 
Added files:
net/unison/patches: patch-bytearray_stubs_c 
Removed files:
net/unison/patches: patch-patch-bytearray_stubs_c 

Log message:
patch the correct file
spotted and fixed by Caspar Shutijs

Re: firefox 52 - crashes at startup or later due to oom

[Peter N. M. Hansteen]

On 03/13/17 08:01, Landry Breuil wrote:
>> Alternatively, I would be glad if someone could test the diff below: I
>> run with it, but I don't use javascript enough to be sure the allocation
>> isn't too low.
>>
>> It makes the allocation to be 128 Mo instead of 1 Go on 64bits platform
>> (it is the same value than for 32 bits platform in fact).
> 
> Yet, firefox is unusable on 32-bit hardware. Try *running* it on an atom
> netbook with 1Gb RAM... so taking the same value as 32 bits platforms
> isnt an argument :)
> 
>>From what i read in the various related bugs, the size was increased to
> fix memory allocation crashes when running WASM (the new 'assembly in
> javascript' standard) and i think to increase randomization of memory
> addresses allocated. *shrug*
> 
> For us, i think coming back to 640M is the "safest" solution.

I've been seeing the same symptoms (firefox-52 crashes with various out
of memory errors after a few minutes).

What is the most useful way to help debug this? Tweaking staff's
stacksize-cur upwards, or some other value?

I did some limited experiments tonight, but I'm afraid I started too
late to get as far as seeing any useful conclusions.

But I can try again tomorrow afternoon CET (possibly with new snapshot
and packages).

All the best,
Peter

-- 
Peter N. M. Hansteen, member of the first RFC 1149 implementation team
http://bsdly.blogspot.com/ http://www.bsdly.net/ http://www.nuug.no/
"Remember to set the evil bit on all malicious network traffic"
delilah spamd[29949]: 85.152.224.147: disconnected after 42673 seconds.

Re: UPDATE: texlive 2016

[Edd Barrett]

Hi Nils,

Thanks for testing.

On Mon, Mar 13, 2017 at 08:06:06AM +0100, Nils Reuße wrote:
> The update went well for me, but i had to use an obj-directory outside
> of /usr (the build takes quite a bit of disk space), and had to
> deinstall texlive 2015 before installing the new packages with `make
> install-all`.

Alright, so I think you are telling me you were unable to test the
update process, only the install-from-scratch process.

> Regarding texlive, there are some problems with lualatex and some of
> the packages that ship with texlive 2016.
>...
> Most of the fallout should have been fixed upstream, but i guess there
> is no way to include updated packages from CTAN in the texlive
> package?

Hrm, that's unfortunate. If you can find a minimal patch to fix this I'd
happily include it. Ideally this would not change the packing lists
though. Or is it more involved?

> Anyway, lualatex is still considered beta and version 1.0 will be
> released with texlive 2017, and i guess that things will just work by
> then -- so maybe it's ok to ignore this issue for now or just point to
> the workaround.

If it helps, I plan to turn around texlive 2017 a bit swifter than I
have 2016.

> Downloading newer packages to a local texmf should also work.

So we have a couple of workarounds at least.

-- 
Best Regards
Edd Barrett

http://www.theunixzoo.co.uk

Re: [PATCH] emulators/virtualjaguar (Honour CFLAGS and CXXFLAGS)

[Jeremie Courreges-Anglas]

Frederic Cambus  writes:

> Hi ports@,
>
> Here is a diff to make emulators/virtualjaguar honour CFLAGS and
> CXXFLAGS everywhere.
>
> Comments? OK?

ok jca@

-- 
jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF  DDCC 0DFA 74AE 1524 E7EE

CVS: cvs.openbsd.org: ports

[Jeremie Courreges-Anglas]

CVSROOT:/cvs
Module name:ports
Changes by: j...@cvs.openbsd.org2017/03/13 15:32:51

Modified files:
www/ruby-passenger: Makefile 

Log message:
Mark as BROKEN on arm, undefined reference to __sync_val_compare_and_swap_4

The __sync_val_compare_and_swap_4 reference comes from the internal
libuv copy.  Using lang/gcc doesn't seem to help.

CVS: cvs.openbsd.org: ports

[Joerg Jung]

CVSROOT:/cvs
Module name:ports
Changes by: j...@cvs.openbsd.org2017/03/13 14:35:47

Modified files:
mail/opensmtpd-extras: Makefile distinfo 

Log message:
update to newer (bugfix) snapshot from today

CVS: cvs.openbsd.org: ports

[Alexander Bluhm]

CVSROOT:/cvs
Module name:ports
Changes by: bl...@cvs.openbsd.org   2017/03/13 13:39:11

Modified files:
devel/p5-ExtUtils-Helpers: Makefile distinfo 

Log message:
update p5-ExtUtils-Helpers to 0.026
from maintainer Sergey Bronnikov

Re: Firefox and Chrome and linkedin account create/recaptcha

[Landry Breuil]

On Mon, Mar 13, 2017 at 12:07:29PM +, Stuart Henderson wrote:
> A desktop user reported "aw snap"s and hangs in Chromium (51.0, on a
> 6.0-stable system) in linkedin, mostly during searches. I'm trying to
> create an account there to check if it also happens on -current too
> but haven't been able to:
> 
> - If I try to create an account with Chromium, I enter Name/Email/Password,
> click Join Now, the button greys out but it just spins. Not sure if it's
> related but when I select the password box, I get this on console:
> 
> [35311:112030272:0313/115918.940732:ERROR:KeyboardEventManager.cpp(424)] Not 
> implemented reached in static bool 
> blink::KeyboardEventManager::currentCapsLockState()
> 
> - If I try to create an account with Firefox, I enter Name/Email/Password,
> click Join Now, I get a recaptcha "I am not a robot" page, but clicking on
> it displays the captcha again.

I've had some funky stuff with the captcha on no-ip.com when i need to
renew a stupid free dynamic domain at their site every 30 days, but most
of the times i just need to click the cehckbox without even having to do
the challenge and it proceeds.

[PATCH] emulators/virtualjaguar (Honour CFLAGS and CXXFLAGS)

[Frederic Cambus]

Hi ports@,

Here is a diff to make emulators/virtualjaguar honour CFLAGS and
CXXFLAGS everywhere.

Comments? OK?

Index: Makefile
===
RCS file: /cvs/ports/emulators/virtualjaguar/Makefile,v
retrieving revision 1.4
diff -u -p -r1.4 Makefile
--- Makefile4 Mar 2017 16:16:32 -   1.4
+++ Makefile13 Mar 2017 16:57:20 -
@@ -34,6 +34,8 @@ MAKE_FLAGS =  CC="${CC}" \
LINK="${CXX}" \
V=1
 
+MAKE_ENV = CXXFLAGS="${CXXFLAGS}"
+
 NO_TEST =  Yes
 
 pre-configure:
Index: patches/patch-Makefile
===
RCS file: patches/patch-Makefile
diff -N patches/patch-Makefile
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-Makefile  13 Mar 2017 16:57:20 -
@@ -0,0 +1,14 @@
+$OpenBSD$
+--- Makefile.orig  Fri Oct 10 15:54:48 2014
 Makefile   Mon Mar 13 17:25:21 2017
+@@ -24,8 +24,8 @@ QMAKE_EXTRA := -spec macx-g++
+ endif
+ 
+ # Set basic flags, these can be overridden from the environment
+-CFLAGS = -O2
+-CXXFLAGS = -O2
++CFLAGS ?= -O2
++CXXFLAGS ?= -O2
+ 
+ # Add CPPFLAGS
+ CFLAGS += $(CPPFLAGS)
Index: patches/patch-virtualjaguar_pro
===
RCS file: patches/patch-virtualjaguar_pro
diff -N patches/patch-virtualjaguar_pro
--- /dev/null   1 Jan 1970 00:00:00 -
+++ patches/patch-virtualjaguar_pro 13 Mar 2017 16:57:20 -
@@ -0,0 +1,14 @@
+$OpenBSD$
+--- virtualjaguar.pro.orig Fri Oct 10 15:54:48 2014
 virtualjaguar.pro  Mon Mar 13 17:25:21 2017
+@@ -47,8 +47,8 @@ macx  { ICON = res/vj-icon.icns }
+ # NOTE: May have to put -Wall back in, but only on non-release cycles. It can
+ #   cause problems if you're not careful. (Can do this via command line in
+ #   qmake)
+-QMAKE_CFLAGS += `$(CROSS)sdl-config --cflags`
+-QMAKE_CXXFLAGS += `$(CROSS)sdl-config --cflags`
++QMAKE_CFLAGS = `$(CROSS)sdl-config --cflags`
++QMAKE_CXXFLAGS = `$(CROSS)sdl-config --cflags`
+ 
+ # Need to add libcdio stuffola (checking/including)...
+ 

CVS: cvs.openbsd.org: ports

[Alexander Bluhm]

CVSROOT:/cvs
Module name:ports
Changes by: bl...@cvs.openbsd.org   2017/03/13 10:26:44

Modified files:
devel/p5-Class-Method-Modifiers: Makefile distinfo 

Log message:
update p5-Class-Method-Modifiers to 2.12
from maintainer Sergey Bronnikov

[UPDATE] devel/p5-ExtUtils-Helpers

[Sergey Bronnikov]

It's a minor update (0.022 -> 0.026).
Tested on amd64.
diff --git a/devel/p5-ExtUtils-Helpers/Makefile 
b/devel/p5-ExtUtils-Helpers/Makefile
index 0e5d02e..16d29b3 100644
--- a/devel/p5-ExtUtils-Helpers/Makefile
+++ b/devel/p5-ExtUtils-Helpers/Makefile
@@ -4,7 +4,7 @@ COMMENT =   various portability utilities for module 
builders
 
 MODULES =  cpan
 PKG_ARCH = *
-DISTNAME = ExtUtils-Helpers-0.022
+DISTNAME = ExtUtils-Helpers-0.026
 CATEGORIES =   devel
 
 MAINTAINER =Sergey Bronnikov 
diff --git a/devel/p5-ExtUtils-Helpers/distinfo 
b/devel/p5-ExtUtils-Helpers/distinfo
index 112a42d..09db563 100644
--- a/devel/p5-ExtUtils-Helpers/distinfo
+++ b/devel/p5-ExtUtils-Helpers/distinfo
@@ -1,2 +1,2 @@
-SHA256 (ExtUtils-Helpers-0.022.tar.gz) = 
0/jPcA+zQUyhJgCJdVy/ZAQUVeS3RBEGd7G6W7mjqpU=
-SIZE (ExtUtils-Helpers-0.022.tar.gz) = 17078
+SHA256 (ExtUtils-Helpers-0.026.tar.gz) = 
3pAbZ5CkVXz07JCBSeA1eDsSW/EV65ZA/rG8HCTDNBY=
+SIZE (ExtUtils-Helpers-0.026.tar.gz) = 15099

Re: [new] www/nextcloud

[Johan Huldtgren]

On 3/13/17 10:58, Gonzalo L. Rodriguez wrote:
> On [27/02/17] [07:44P], Johan Huldtgren wrote:
> ; On 2/26/17 18:35, Gonzalo L. Rodriguez wrote:
> ; >
> ; > I do some tweaks on your port, can you try if this works for you?
> ; 
> ; looks good to me, packages and installs fine on my test system.
> ; 
> ; .jh
> ; 
> 
> New version attached, any interested on this in?

I have been running 11.0.2 in production for the past week or so. Works
well for me.

.jh

CVS: cvs.openbsd.org: ports

[Frederic Cambus]

CVSROOT:/cvs
Module name:ports
Changes by: fcam...@cvs.openbsd.org 2017/03/13 09:38:24

Modified files:
devel/uthash   : Makefile distinfo 
devel/uthash/patches: patch-tests_Makefile 
devel/uthash/pkg: PLIST 

Log message:
Update uthash to 2.0.1.

Notable changes:

- Upstream moved to GitHub
- HTML documentation is not bundled in release tarballs anymore, and
must be generated with asciidoc, so only ship plain text documentation
to avoid adding a build dependency

OK jca@

CVS: cvs.openbsd.org: ports

[Antoine Jacoutot]

CVSROOT:/cvs
Module name:ports
Changes by: ajacou...@cvs.openbsd.org   2017/03/13 09:28:33

Modified files:
mail/evolution-ews: Makefile distinfo 

Log message:
Update to evolution-ews-3.22.6.

CVS: cvs.openbsd.org: ports

[Antoine Jacoutot]

CVSROOT:/cvs
Module name:ports
Changes by: ajacou...@cvs.openbsd.org   2017/03/13 09:28:18

Modified files:
mail/evolution : Makefile distinfo 

Log message:
Update to evolution-3.22.6.

CVS: cvs.openbsd.org: ports

[Antoine Jacoutot]

CVSROOT:/cvs
Module name:ports
Changes by: ajacou...@cvs.openbsd.org   2017/03/13 09:27:55

Modified files:
databases/evolution-data-server: Makefile distinfo 

Log message:
Update to evolution-data-server-3.22.6.

CVS: cvs.openbsd.org: ports

[Alexander Bluhm]

CVSROOT:/cvs
Module name:ports
Changes by: bl...@cvs.openbsd.org   2017/03/13 09:13:54

Modified files:
converters/p5-Cpanel-JSON-XS: Makefile distinfo 

Log message:
update p5-Cpanel-JSON-XS to 3.0230

CVS: cvs.openbsd.org: ports

[Alexander Bluhm]

CVSROOT:/cvs
Module name:ports
Changes by: bl...@cvs.openbsd.org   2017/03/13 09:02:48

Modified files:
security   : Makefile 

Log message:
+botan2

CVS: cvs.openbsd.org: ports

[Alexander Bluhm]

CVSROOT:/cvs
Module name:ports
Changes by: bl...@cvs.openbsd.org   2017/03/13 09:00:39

Log message:
import botan2 2.0.1
input and OK jca@

Comment:
crypto and TLS for C++11

Description:
Botan (Japanese for peony) is a cryptography library written in
C++11 and released under the permissive Simplified BSD license.
Botan's goal is to be the best option for cryptography in new C++
code by offering the tools necessary to implement a range of practical
systems, such as TLS/DTLS, PKIX certificate handling, PKCS#11 and
TPM hardware support, password hashing, and post quantum crypto
schemes.  In addition to the C++, botan has a C89 API specifically
designed to be easy to call from other languages.  A Python binding
using ctypes calling the C89 API is included.

Status:

Vendor Tag: bluhm
Release Tags:   bluhm_20170313

N ports/security/botan2/Makefile
N ports/security/botan2/distinfo
N ports/security/botan2/pkg/DESCR
N ports/security/botan2/pkg/PLIST
N ports/security/botan2/patches/patch-src_tests_test_name_constraint_cpp
N ports/security/botan2/patches/patch-src_build-data_os_openbsd_txt
N ports/security/botan2/patches/patch-src_scripts_install_py
N ports/security/botan2/patches/patch-configure_py
N ports/security/botan2/patches/patch-src_build-data_cc_gcc_txt

No conflicts created by this import

CVS: cvs.openbsd.org: ports

[Antoine Jacoutot]

CVSROOT:/cvs
Module name:ports
Changes by: ajacou...@cvs.openbsd.org   2017/03/13 07:26:52

Modified files:
devel/meson: Makefile distinfo 

Log message:
Update to meson-0.39.0.

Re: openbsd ports of perl modules

[Marc Espie]

On Mon, Mar 13, 2017 at 04:00:06PM +0300, Sergey Bronnikov wrote:
> Hello,
> 
> some time ago I made an OpenBSD port for Tapper testing system.
> This port depended from huge amount of Perl modules and I made port
> for each of them. Right now Tapper development is discontinued
> and I removed Tapper from the openbsd-wip repository, but all dependencies
> are kept in the repo.
> 
> The question to community is: do we need these modules in the official ports 
> tree?
> Is anyone ready to import all these ports or part of them to CVS?

Err, at least some of these are in base perl.  Are you talking about
a more recent version ? that will often be a nightmare.

At least those:
> p5-Data-Dumper
> p5-ExtUtils-MakeMaker
> p5-Getopt-Long
> p5-Test-Simple
> p5-Text-Abbrev
> p5-Text-Balanced

openbsd ports of perl modules

[Sergey Bronnikov]

Hello,

some time ago I made an OpenBSD port for Tapper testing system.
This port depended from huge amount of Perl modules and I made port
for each of them. Right now Tapper development is discontinued
and I removed Tapper from the openbsd-wip repository, but all dependencies
are kept in the repo.

The question to community is: do we need these modules in the official ports 
tree?
Is anyone ready to import all these ports or part of them to CVS?

p5-DBIx-Class-InflateColumn-Object-Enum
p5-App-Daemon
p5-App-Rad
p5-BSD-Sysctl
p5-Benchmark-Perl-Formance
p5-Bencode
p5-CHI
p5-Compress-PPMd
p5-Convert-Bencode
p5-Convert-Bencode_XS
p5-Daemon-Daemonize
p5-Data-Clone
p5-Data-DPath
p5-Data-Dumper
p5-Data-Rmap
p5-Data-Serializer
p5-Data-Taxi
p5-Data-YAML
p5-DateTime-Format-DateParse
p5-Debug-ShowStuff
p5-Devel-Backtrace
p5-Devel-Platform-Info
p5-ExtUtils-MakeMaker
p5-File-Pid
p5-Format-Human-Bytes
p5-Getopt-Long
p5-HTML-FromANSI
p5-Hash-MoreUtils
p5-Kwalify
p5-Language-Expr
p5-Lexical-SealRequireHints
p5-Math-BigInt-Pari
p5-MooX-Types-MooseLike-Numeric
p5-MooseX-ClassAttribute
p5-MooseX-Singleton
p5-Object-Enum
p5-Regexp-Grammars
p5-Scalar-List-Utils
p5-Set-Intersection
p5-Sys-Info-Base
p5-Sys-Info-Driver-BSD
p5-Sys-Info
p5-Sysadm-Install
p5-TAP-DOM
p5-Term-VT102-Boundless
p5-Term-VT102
p5-Test-Fixture-DBIC-Schema
p5-Test-Mock-LWP
p5-Test-Nginx
p5-Test-Simple
p5-Test-Sys-Info
p5-Test-TAPv13
p5-Unix-Processors
p5-bareword-filehandles
p5-experimental
p5-multidimensional
p5--Crypt-DH
p5-Digest-JHash
p5-Sys-Syslog
p5-String-Diff
p5-String-Escape
p5-String-Util
p5-Text-Abbrev
p5-Text-Balanced
p5-Text-CSV-Slurp
p5-Catalyst-View-HTML-Mason
p5-TAP-Formatter-HTML
p5-Test-WWW-Selenium

Re: Firefox and Chrome and linkedin account create/recaptcha

[Marc Espie]

On Mon, Mar 13, 2017 at 12:07:29PM +, Stuart Henderson wrote:
> There's also an annoying UK webshop (broadbandbuyer.co.uk) that also tries
> to force a recaptcha if I use it from OpenBSD (I guess they're doing some
> user-agent check or something), I get a similar thing where the recaptcha
> just gets displayed again when I click on it.

Not so far.
User agents checks are easily bypassed with extensions.

I have "User agent switcher for chrome" and I would suggest to try something
like it to get that specific concern out of the way.

Firefox and Chrome and linkedin account create/recaptcha

[Stuart Henderson]

A desktop user reported "aw snap"s and hangs in Chromium (51.0, on a
6.0-stable system) in linkedin, mostly during searches. I'm trying to
create an account there to check if it also happens on -current too
but haven't been able to:

- If I try to create an account with Chromium, I enter Name/Email/Password,
click Join Now, the button greys out but it just spins. Not sure if it's
related but when I select the password box, I get this on console:

[35311:112030272:0313/115918.940732:ERROR:KeyboardEventManager.cpp(424)] Not 
implemented reached in static bool 
blink::KeyboardEventManager::currentCapsLockState()

- If I try to create an account with Firefox, I enter Name/Email/Password,
click Join Now, I get a recaptcha "I am not a robot" page, but clicking on
it displays the captcha again.

There's also an annoying UK webshop (broadbandbuyer.co.uk) that also tries
to force a recaptcha if I use it from OpenBSD (I guess they're doing some
user-agent check or something), I get a similar thing where the recaptcha
just gets displayed again when I click on it.

Any ideas anyone?  Do others see problems with recaptcha on OpenBSD?

qt-creator / x11/qt5/qtdeclarative loop [Re: qt creator Abort trap (core dumped)]

[Stuart Henderson]

Does anyone have ideas about this? I tried changing the deps to
x11/qt5/qtdeclarative,no_examples,no_qch,no_html,bootstrap but that
didn't help.


On 2017/02/18 13:25, Stuart Henderson wrote:
> Moved from misc.
> 
> On 2017-02-17, Caspar Schutijser  wrote:
> > Does anybody now why qt-creator 4.1.0 doesn't show up on the mirrors?
> > Does it fail to build on the ports machines? On my machine, I built the
> > package successfully a couple of times.
> 
> Looks like there's a problem with the bootstrap bits for qt5/qtdeclarative 
> that
> are used to break a dependency loop.
> 
> 55437@1487279019: T: x11/qt5/qtdeclarative
> 55437@1487279019: B: 
> x11/qt5/qtdeclarative,-examples:qt/qtdeclarative-opensource-src-5.6.2.tar.gz
> 55437@1487279226: !: devel/qt-creator because of 
> x11/qt5/qtdeclarative,-docindex
> 55437@1487279372: !: meta/qt5,-qch because of x11/qt5/qtdeclarative,-qch
> 55437@1487279372: !: meta/qt5,-html because of x11/qt5/qtdeclarative,-html
> 55437@1487279798: T: x11/qt5/qtdeclarative,,-main
> 55437@1487279798: T: x11/qt5/qtdeclarative,,-qch
> 55437@1487279798: T: x11/qt5/qtdeclarative,,-examples
> 55437@1487279798: T: x11/qt5/qtdeclarative,,-docindex
> 55437@1487279798: T: x11/qt5/qtdeclarative,,-html
> 55437@1487279835: !: x11/qt5/qtdeclarative,-html Ignored as FLAVOR contains 
> no_html
> 55437@1487279835: T: x11/qt5/qtdeclarative,-main
> 55437@1487279835: !: x11/qt5/qtdeclarative,-examples Ignored as FLAVOR 
> contains no_examples
> 55437@1487279836: T: x11/qt5/qtdeclarative
> 55437@1487279836: !: x11/qt5/qtdeclarative,-docindex vanishes from 
> BUILD_PACKAGES
> 55437@1487279836: !: x11/qt5/qtdeclarative,-qch Ignored as FLAVOR contains 
> no_qch
> 55437@1487279836: T: x11/qt5/qtdeclarative,,-docindex
> 55437@1487296882: Q: x11/qt5/qtdeclarative,-main
> 55437@1487296882: Q: x11/qt5/qtdeclarative
> 55437@1487297476: J: x11/qt5/qtdeclarative localhost
> 55437@1487298797: B: x11/qt5/qtdeclarative
> 55437@1487298797: B: x11/qt5/qtdeclarative,-main
> 55437@1487298830: I: x11/qt5/qtdeclarative,-main # qtdeclarative-5.6.2
> 55437@1487298830: I: x11/qt5/qtdeclarative # qtdeclarative-5.6.2
> 55437@1487298830: B: x11/qt5/qtdeclarative,,-main
> 55437@1487298887: I: x11/qt5/qtdeclarative,,-main # qtdeclarative-5.6.2
> 55437@1487300466: Q: x11/qt5/qtdeclarative,,-docindex
> 55437@1487300466: Q: x11/qt5/qtdeclarative,,-qch
> 55437@1487300466: Q: x11/qt5/qtdeclarative,,-examples
> 55437@1487300466: Q: x11/qt5/qtdeclarative,,-html
> 55437@1487300627: J: x11/qt5/qtdeclarative,,-docindex localhost
> 55437@1487301963: B: x11/qt5/qtdeclarative,,-docindex
> 55437@1487301963: B: x11/qt5/qtdeclarative,
> 55437@1487301963: B: x11/qt5/qtdeclarative,,-html
> 55437@1487301963: B: x11/qt5/qtdeclarative,,-qch
> 55437@1487301963: B: x11/qt5/qtdeclarative,,-examples
> 55437@1487301963: B: x11/qt5/qtdeclarative,,-main
> 55437@1487301994: I: x11/qt5/qtdeclarative,,-docindex # 
> qtdeclarative-docindex-5.6.2
> 55437@1487301994: I: x11/qt5/qtdeclarative,,-main # qtdeclarative-5.6.2
> 55437@1487301994: I: x11/qt5/qtdeclarative,,-html # qtdeclarative-html-5.6.2p0
> 55437@1487301994: I: x11/qt5/qtdeclarative, # qtdeclarative-5.6.2
> 55437@1487301994: I: x11/qt5/qtdeclarative,,-qch # qtdeclarative-qch-5.6.2p0
> 55437@1487305507: I: x11/qt5/qtdeclarative,,-examples # 
> qtdeclarative-examples-5.6.2p0
> 93565@1487335269: B: x11/qt5/qtdeclarative
> 93565@1487335269: B: 
> x11/qt5/qtdeclarative,-html:qt/qtdeclarative-opensource-src-5.6.2.tar.gz
> 93565@1487335283: I: x11/qt5/qtdeclarative # qtdeclarative-5.6.2
> 93565@1487335431: !: devel/qt-creator because of 
> x11/qt5/qtdeclarative,-docindex
> 93565@1487335531: !: meta/qt5,-qch because of x11/qt5/qtdeclarative,-qch
> 93565@1487335531: !: meta/qt5,-html because of x11/qt5/qtdeclarative,-html
> 93565@1487335531: !: meta/qt5,-examples because of 
> x11/qt5/qtdeclarative,-examples
> 93565@1487335784: B: x11/qt5/qtdeclarative,,-examples
> 93565@1487335784: B: x11/qt5/qtdeclarative,,-main
> 93565@1487335784: B: x11/qt5/qtdeclarative,,-html
> 93565@1487335784: B: x11/qt5/qtdeclarative,,-docindex
> 93565@1487335784: B: x11/qt5/qtdeclarative,,-qch
> 93565@1487335808: B: x11/qt5/qtdeclarative
> 93565@1487335808: Q: x11/qt5/qtdeclarative
> 93565@1487335808: B: x11/qt5/qtdeclarative,-main
> 93565@1487335808: Q: x11/qt5/qtdeclarative,-main
> 93565@1487335808: !: x11/qt5/qtdeclarative,-docindex vanishes from 
> BUILD_PACKAGES
> 93565@1487335808: !: x11/qt5/qtdeclarative,-html Ignored as FLAVOR contains 
> no_html
> 93565@1487335809: !: x11/qt5/qtdeclarative,-examples Ignored as FLAVOR 
> contains no_examples
> 93565@1487335809: !: x11/qt5/qtdeclarative,-qch Ignored as FLAVOR contains 
> no_qch
> 93565@1487335818: I: x11/qt5/qtdeclarative,-main # qtdeclarative-5.6.2
> 93565@1487335818: I: x11/qt5/qtdeclarative # qtdeclarative-5.6.2
> 93565@1487335818: I: x11/qt5/qtdeclarative,,-main # qtdeclarative-5.6.2
> 93565@1487335818: I: x11/qt5/qtdeclarative,,-examples # 
> 

CVS: cvs.openbsd.org: ports

[Stuart Henderson]

CVSROOT:/cvs
Module name:ports
Changes by: st...@cvs.openbsd.org   2017/03/13 04:57:00

Modified files:
databases/mariadb: Makefile distinfo 
databases/mariadb/pkg: PLIST-server PLIST-tests 

Log message:
update to MariaDB 10.0.30, from Brad. Fixes CVE-2017-3302 and CVE-2017-3313.

Re: rust-doc removal ?

[Laurence Tratt]

On Sun, Mar 12, 2017 at 06:58:50PM +0100, Sebastien Marie wrote:

Hello Sebastien,

> I would like to know if there are any user of rust-doc.

Yes, I use it a lot. It's particularly useful to have the API docs when I'm
offline (or with limited bandwidth). But if I'm the only user then I don't
think it's worth the maintenance overhead!


Laurie
-- 
Personal http://tratt.net/laurie/
Software Development Teamhttp://soft-dev.org/
   https://github.com/ltratt  http://twitter.com/laurencetratt

CVS: cvs.openbsd.org: ports

[Stuart Henderson]

CVSROOT:/cvs
Module name:ports
Changes by: st...@cvs.openbsd.org   2017/03/13 03:42:57

Modified files:
sysutils/bacula: Makefile distinfo 
Added files:
sysutils/bacula/patches: patch-src_filed_xacl_c 

Log message:
update to bacula-7.4.6

CVS: cvs.openbsd.org: ports

[Antoine Jacoutot]

CVSROOT:/cvs
Module name:ports
Changes by: ajacou...@cvs.openbsd.org   2017/03/13 03:11:59

Modified files:
x11/gnome/documents: Makefile distinfo 
x11/gnome/documents/pkg: PLIST 

Log message:
Update to gnome-documents-3.22.1.

CVS: cvs.openbsd.org: ports

[Antoine Jacoutot]

CVSROOT:/cvs
Module name:ports
Changes by: ajacou...@cvs.openbsd.org   2017/03/13 02:52:09

Modified files:
devel/harfbuzz : Makefile distinfo 

Log message:
Update to harfbuzz-1.4.5.

CVS: cvs.openbsd.org: ports

[Benoit Lecocq]

CVSROOT:/cvs
Module name:ports
Changes by: ben...@cvs.openbsd.org  2017/03/13 02:47:27

Modified files:
net/tintin++   : Makefile distinfo 

Log message:
Update to tintin-2.01.2.

from Ted Roby (Maintainer)

Re: Fwd: [elinks-dev] [Bug][Security] elinks doesn't verify server certificate

[Stuart Henderson]

On 2017/03/12 17:57, Edd Barrett wrote:
> On Fri, Mar 10, 2017 at 10:54:29AM +, Edd Barrett wrote:
> > +# Elinks does not check SSL certificates properly!
> > +# Disable SSL support to protect our users.
> > +CONFIGURE_ARGS +=  --without-gnutls \
> > +   --without-openssl
> 
> Any comments on this? Kill SSL support or kill elinks?

Is this patch any good for us?

http://lists.linuxfromscratch.org/pipermail/elinks-dev/2015-June/002099.html

CVS: cvs.openbsd.org: ports

[Antoine Jacoutot]

CVSROOT:/cvs
Module name:ports
Changes by: ajacou...@cvs.openbsd.org   2017/03/13 02:42:29

Modified files:
graphics/gegl03: Makefile distinfo 

Log message:
Update to gegl03-0.3.14.

CVS: cvs.openbsd.org: ports

[Antoine Jacoutot]

CVSROOT:/cvs
Module name:ports
Changes by: ajacou...@cvs.openbsd.org   2017/03/13 02:42:14

Modified files:
graphics/babl  : Makefile distinfo 
graphics/babl/pkg: PLIST 

Log message:
Update to babl-0.1.24.

CVS: cvs.openbsd.org: ports

[Benoit Lecocq]

CVSROOT:/cvs
Module name:ports
Changes by: ben...@cvs.openbsd.org  2017/03/13 02:39:13

Modified files:
net/lftp   : Makefile distinfo 
net/lftp/patches: patch-configure 

Log message:
Update to lftp-4.7.7.

from Rafael Sadowski (Maintainer)

simple update for net/tintin++

[Ted Roby]

This is a simple version update for net/tintin++.
Compiled and tested on amd64.


Index: Makefile
===
RCS file: /cvs/ports/net/tintin++/Makefile,v
retrieving revision 1.19
diff -u -p -u -r1.19 Makefile
--- Makefile4 Jan 2014 07:51:53 -   1.19
+++ Makefile13 Mar 2017 07:20:49 -
@@ -1,7 +1,7 @@
 # $OpenBSD: Makefile,v 1.19 2014/01/04 07:51:53 bcallah Exp $ 
 
 COMMENT=   client program for playing muds with advanced features
-DISTNAME=   tintin-2.01.0
+DISTNAME=   tintin-2.01.2
 CATEGORIES= net games
 
 HOMEPAGE=   http://tintin.sourceforge.net/
Index: distinfo
===
RCS file: /cvs/ports/net/tintin++/distinfo,v
retrieving revision 1.4
diff -u -p -u -r1.4 distinfo
--- distinfo4 Jan 2014 07:51:53 -   1.4
+++ distinfo13 Mar 2017 07:20:49 -
@@ -1,2 +1,2 @@
-SHA256 (tintin-2.01.0.tar.gz) = 4ONUY6l+5bM+8LKbLFf6gnbE52MoyxnJim6pLGA6nHY=
-SIZE (tintin-2.01.0.tar.gz) = 293651
+SHA256 (tintin-2.01.2.tar.gz) = AeEePN7Uj/aGsuoW52es8fa16jJlUez/CRVS6J9KA44=
+SIZE (tintin-2.01.2.tar.gz) = 274380

Re: UPDATE: texlive 2016

[Nils Reuße]

> Edd Barrett  hat am 9. März 2017 um 23:07
> geschrieben:
> 
> 
> Hi,
> 
> Here's an update to tex live 2016.
> 
> Can people start playing with this and reporting problems when building
> their docs.
> 
> The pregen tarball is here for now:
> http://theunixzoo.co.uk/random/texmf-var-pregen-2016.tar.xz
> 
> Stuart, would you mind hosting this file please?
> 
> The plist make target needs the new texscythe I posted the other day.
> 
> I've run this through a partial bulk with all the build dependants. No
> issues, but I've not yet tested it thoroughly on my own stuff.
> 
> I only really use pdflatex, so if you are a user of one of the other
> typesetters included, please step up to test.
> 
> $ wc -l ~/tmp/tl16.diff
>21361 /home/edd/tmp/tl16.diff
> 
> :P
> 
> Thanks
> 
> -- 
> Best Regards
> Edd Barrett
> 
> http://www.theunixzoo.co.uk

Hi Ed,

thanks for your hard work.

The update went well for me, but i had to use an obj-directory outside of /usr 
(the build takes quite a bit of disk space), and had to deinstall texlive 2015 
before installing the new packages with `make install-all`.

Regarding texlive, there are some problems with lualatex and some of the 
packages that ship with texlive 2016, e.g. typearea, which is needed by all 
KOMA classes like scrartcl.  The package has a bug in version 3.20 [1], which 
breaks compilation (lualatex only) and got fixed upstream with version 3.21.  I 
found similar problems with the datetime [2] and standalone [3] packages, and 
there are probably more.  A workaround exists, namely including the luatex85 
package before these packages [4].  Most of the fallout should have been fixed 
upstream, but i guess there is no way to include updated packages from CTAN in 
the texlive package?

Anyway, lualatex is still considered beta and version 1.0 will be released with 
texlive 2017, and i guess that things will just work by then -- so maybe it's 
ok to ignore this issue for now or just point to the workaround.  Downloading 
newer packages to a local texmf should also work.

Nils

[1] https://www.tug.org/pipermail/tex-live/2016-June/038675.html
[2] 
http://tex.stackexchange.com/questions/315198/datetime-package-does-not-work-with-lualatex-in-tl-2016
[3] 
http://tex.stackexchange.com/questions/315025/lualatex-texlive-2016-standalone-undefined-control-sequence
[4] http://ctan.org/pkg/luatex85

Re: firefox 52 - crashes at startup or later due to oom

[Landry Breuil]

On Mon, Mar 13, 2017 at 06:34:12AM +0100, Sebastien Marie wrote:
> On Sun, Mar 12, 2017 at 10:00:55PM +0100, frantisek holop wrote:
> > it seems that i can 100% crash firefox by trying to use
> > the web.whatsapp.com client for varying lengths of
> > time.  this seems to be a regression, is anyone else
> > seeing this?
> > 
> 
> with firefox-52.0, the way memory allocation is done for javascript has
> changed:
> 
>   - Bug 1334933 - Allocate executable pages from a pre-reserved range.
> https://hg.mozilla.org/releases/mozilla-release/rev/6b35bbf96b67
> 
> and in the same release, the size of initial malloc has been raised from
> 640Mo to 1Go:
> 
>   - Bug 1337561 - Baldr: call largeAllocationCallback and retry if executable 
> allocation fails. r=jandem, a=jcristau
> https://hg.mozilla.org/releases/mozilla-release/rev/65bb26d07408
> 
> 
> It makes firefox to malloc a chunk of 1Go at startup.
> 
> As you crash after the start, I assume you are running in "staff" class
> or have modified login.conf datastack-cur value in "default" class.
> 
> The current way to deal with that is to raise your datastack-cur to make
> firefox to fit in...
> 
> 
> Alternatively, I would be glad if someone could test the diff below: I
> run with it, but I don't use javascript enough to be sure the allocation
> isn't too low.
> 
> It makes the allocation to be 128 Mo instead of 1 Go on 64bits platform
> (it is the same value than for 32 bits platform in fact).

Yet, firefox is unusable on 32-bit hardware. Try *running* it on an atom
netbook with 1Gb RAM... so taking the same value as 32 bits platforms
isnt an argument :)

>From what i read in the various related bugs, the size was increased to
fix memory allocation crashes when running WASM (the new 'assembly in
javascript' standard) and i think to increase randomization of memory
addresses allocated. *shrug*

For us, i think coming back to 640M is the "safest" solution.

Re: firefox 52 - crashes at startup or later due to oom

[Landry Breuil]

On Mon, Mar 13, 2017 at 05:40:25AM +, Tinker wrote:
> On 2017-03-13 05:34, Sebastien Marie wrote:
> ..
> > Note that the value is the total allocated bytes for javascript. It
> > isn't a initial value, and it shouldn't grow (if I correctly understand
> > the code). So I expect 128 Mo to be a bit "restrictive".
> ..
> > +-static const size_t MaxCodeBytesPerProcess = 1 * 1024 * 1024 * 1024;
> > ++static const size_t MaxCodeBytesPerProcess = 128 * 1024 * 1024;
> 
> So Firefox has a blanked fixed-heap size model for all of its deployments??
> 
> If so, that is like, ridiculously archaic, what were they thinking. I wonder
> if it can be made to run in a dynamic heap model. I can't see why Wikipedia
> and alert("Hello world"); would take up any much memory at all, like, more
> than the binary and dep libraries + 1-100MB allocations.

Talking about it here won't help. Take that upstream.

Re: new: security/botan2

[Jeremie Courreges-Anglas]

Alexander Bluhm  writes:

> Hi
>
> I would like to add a new port for the crypto library botan 2.  We
> already have the botan 1.10 in ports, but they are not API compatible.
> So I think it it best to put it under security/botan2.  Note that
> all internal paths do not conflict with the existing security/botan.
>
> Comment:
> crypto and TLS for C++11
>
> Description:
> Botan (Japanese for peony) is a cryptography library written in
> C++11 and released under the permissive Simplified BSD license.
> Botan's goal is to be the best option for cryptography in new C++
> code by offering the tools necessary to implement a range of practical
> systems, such as TLS/DTLS, PKIX certificate handling, PKCS#11 and
> TPM hardware support, password hashing, and post quantum crypto
> schemes.  In addition to the C++, botan has a C89 API specifically
> designed to be easy to call from other languages.  A Python binding
> using ctypes calling the C89 API is included.
>
> I have added one patch from upstream that makes the tests pass.
>
> The other patches convert the library naming to our schema.  It
> seems that upstream is respecting ABI compatibility the same way
> we do, so I user their version numbering, which is 0.0 currently.
> If a future version is incompatible, I can still add a patch to
> override it.
>
> The python binding is included by default.  It is only one file and
> I don't think the effort of a py-subpackage is worth it.

Same here regarding a py- subpackage, but then should botan2 depend on
python at runtime?  (MODPY_RUNDEP defaults to Yes)

> ok?

Looks good to me, however the build system hardcodes eg++ and -O3.  The
updated tarball below forces the build system to respect CXX/CXXFLAGS.



botan2.tgz
Description: Binary data

-- 
jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF  DDCC 0DFA 74AE 1524 E7EE

Re: [UPDATE] devel/uthash to 2.0.1

[Jeremie Courreges-Anglas]

Frederic Cambus  writes:

> Hi ports@,
>
> Here is a diff to update devel/uthash to 2.0.1.

No consumer so I see no reason to hold this back.  ok jca@

> Notable changes:
>
> - Upstream moved to GitHub
> - HTML documentation is not bundled in release tarballs anymore, and
>   must be generated with asciidoc, so only ship plain text documentation
>   to avoid adding a build dependency
>
> Index: Makefile
> ===
> RCS file: /cvs/ports/devel/uthash/Makefile,v
> retrieving revision 1.13
> diff -u -p -r1.13 Makefile
> --- Makefile  27 Mar 2014 08:47:56 -  1.13
> +++ Makefile  1 Mar 2017 21:03:57 -
> @@ -2,29 +2,25 @@
>  
>  COMMENT= C macros to add, find and delete items from a hash
>  
> -DISTNAME=uthash-1.9.7
> -CATEGORIES=  devel
> +GH_ACCOUNT=  troydhanson
> +GH_PROJECT=  uthash
> +GH_TAGNAME=  v2.0.1
>  
> -HOMEPAGE=http://uthash.sf.net/
> +CATEGORIES=  devel
>  
>  # BSD
>  PERMIT_PACKAGE_CDROM=Yes
>  
> -MASTER_SITES=${MASTER_SITE_SOURCEFORGE:=uthash/}
> -EXTRACT_SUFX=.tar.bz2
> -
>  NO_BUILD=Yes
>  PKG_ARCH=*
>  
>  PORTHOME=${WRKDIR}
> -TEST_TARGET= run_tests
> +TEST_TARGET= run_tests
>  
>  do-install:
>   ${INSTALL_DATA} ${WRKSRC}/src/*.h ${PREFIX}/include
> - ${INSTALL_DATA_DIR} ${PREFIX}/share/doc/uthash/img
> - ${INSTALL_DATA} ${WRKSRC}/doc/html/*.{css,html} \
> - ${PREFIX}/share/doc/uthash
> - ${INSTALL_DATA} ${WRKSRC}/doc/html/img/* ${PREFIX}/share/doc/uthash/img
> + ${INSTALL_DATA_DIR} ${PREFIX}/share/doc/uthash
> + ${INSTALL_DATA} ${WRKSRC}/doc/u*.txt ${PREFIX}/share/doc/uthash
>  
>  do-test:
>   @cd ${WRKSRC}/tests/ && exec ${SETENV} ${MAKE_ENV} \
> Index: distinfo
> ===
> RCS file: /cvs/ports/devel/uthash/distinfo,v
> retrieving revision 1.10
> diff -u -p -r1.10 distinfo
> --- distinfo  13 Oct 2012 10:40:38 -  1.10
> +++ distinfo  1 Mar 2017 21:03:57 -
> @@ -1,2 +1,2 @@
> -SHA256 (uthash-1.9.7.tar.bz2) = lW9cmXmDScQTJ1/kyf8SjXLigGVdrb5DZfjp+9qROT8=
> -SIZE (uthash-1.9.7.tar.bz2) = 373059
> +SHA256 (uthash-2.0.1.tar.gz) = YTuV/MNot9AVrS0IAjEydwEvUMSsKQw9/BQtQuvqMzc=
> +SIZE (uthash-2.0.1.tar.gz) = 189051
> Index: patches/patch-tests_Makefile
> ===
> RCS file: /cvs/ports/devel/uthash/patches/patch-tests_Makefile,v
> retrieving revision 1.7
> diff -u -p -r1.7 patch-tests_Makefile
> --- patches/patch-tests_Makefile  19 Apr 2012 06:39:42 -  1.7
> +++ patches/patch-tests_Makefile  1 Mar 2017 21:03:57 -
> @@ -2,9 +2,9 @@ $OpenBSD: patch-tests_Makefile,v 1.7 201
>  
>  - Remove GNU make crap that never applied to us anyway.
>  
>  tests/Makefile.orig  Thu Apr 19 08:38:39 2012
> -+++ tests/Makefile   Thu Apr 19 08:39:02 2012
> -@@ -19,48 +19,10 @@ CFLAGS += -Wall 
> +--- tests/Makefile.orig  Tue Jul  5 23:05:33 2016
>  tests/Makefile   Wed Mar  1 21:55:35 2017
> +@@ -24,48 +24,10 @@ CFLAGS += -Wall 
>   #CFLAGS += -std=c89
>   CFLAGS += ${EXTRA_CFLAGS}
>   
> @@ -13,7 +13,7 @@ $OpenBSD: patch-tests_Makefile,v 1.7 201
>  -endif
>  -
>  -ifeq ($(HASH_PEDANTIC),1)
> --CFLAGS += -pedantic 
> +-CFLAGS += -pedantic
>  -endif
>  -
>   TEST_TARGET=run_tests
> Index: pkg/PLIST
> ===
> RCS file: /cvs/ports/devel/uthash/pkg/PLIST,v
> retrieving revision 1.4
> diff -u -p -r1.4 PLIST
> --- pkg/PLIST 3 Nov 2010 16:37:47 -   1.4
> +++ pkg/PLIST 1 Mar 2017 21:03:57 -
> @@ -1,26 +1,16 @@
> -@comment $OpenBSD: PLIST,v 1.4 2010/11/03 16:37:47 jasper Exp $
> +@comment $OpenBSD$
> +include/libut.h
> +include/ringbuf.h
>  include/utarray.h
>  include/uthash.h
>  include/utlist.h
> +include/utmm.h
> +include/utringbuffer.h
>  include/utstring.h
> +include/utvector.h
>  share/doc/uthash/
> -share/doc/uthash/ChangeLog.html
> -share/doc/uthash/img/
> -share/doc/uthash/img/banner.png
> -share/doc/uthash/img/banner.svg
> -share/doc/uthash/img/grad_blue.png
> -share/doc/uthash/img/grad_blue.svg
> -share/doc/uthash/img/rss.png
> -share/doc/uthash/img/uthash-mini.png
> -share/doc/uthash/img/uthash-mini.svg
> -share/doc/uthash/img/uthash.png
> -share/doc/uthash/index.html
> -share/doc/uthash/license.html
> -share/doc/uthash/styles.css
> -share/doc/uthash/tdh-quirks.css
> -share/doc/uthash/tdh.css
> -share/doc/uthash/toc.css
> -share/doc/uthash/userguide.html
> -share/doc/uthash/utarray.html
> -share/doc/uthash/utlist.html
> -share/doc/uthash/utstring.html
> +share/doc/uthash/userguide.txt
> +share/doc/uthash/utarray.txt
> +share/doc/uthash/utlist.txt
> +share/doc/uthash/utringbuffer.txt
> +share/doc/uthash/utstring.txt
>

-- 
jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF  DDCC 0DFA 

Re: UPDATE: security/botan

[Jeremie Courreges-Anglas]

Alexander Bluhm  writes:

> On Fri, Mar 10, 2017 at 09:38:20PM +0100, Rafael Sadowski wrote:
>> please find below a simple diff to update botan to the last stable
>> version.
>
> I have already commited this, but I was too fast.  Several new
> functions show up in the library, so I think the library version
> has to be bumped.

When comparing the output of

  nm -g libbotan-1.10.so | awk '{ print $3 }' | sort -u

I do not find additional symbols.

--- a   Mon Mar 13 07:02:10 2017
+++ b   Mon Mar 13 07:02:21 2017
@@ -3269,7 +3269,7 @@
 _ZTIN5Botan5EMSA4E
 _ZTIN5Botan5PBKDFE
 _ZTIN5Botan5TigerE
-_ZTIN5Botan64_GLOBAL__N_._src_alloc_alloc_mmap_mmap_mem.cpp__05C3E6FB20MemoryMapping_FailedE
+_ZTIN5Botan64_GLOBAL__N_._src_alloc_alloc_mmap_mmap_mem.cpp__C578C44620MemoryMapping_FailedE
 _ZTIN5Botan6BigInt12DivideByZeroE
 _ZTIN5Botan6CTR_BEE
 _ZTIN5Botan6Comb4PE
@@ -3556,7 +3556,7 @@
 _ZTSN5Botan5EMSA4E
 _ZTSN5Botan5PBKDFE
 _ZTSN5Botan5TigerE
-_ZTSN5Botan64_GLOBAL__N_._src_alloc_alloc_mmap_mmap_mem.cpp__05C3E6FB20MemoryMapping_FailedE
+_ZTSN5Botan64_GLOBAL__N_._src_alloc_alloc_mmap_mmap_mem.cpp__C578C44620MemoryMapping_FailedE
 _ZTSN5Botan6BigInt12DivideByZeroE
 _ZTSN5Botan6CTR_BEE
 _ZTSN5Botan6Comb4PE

(The mismatch for 'Botan::(anonymous namespace)::MemoryMapping_Failed'
above appears to be normal and unrelated to the update.)

> ok?

I do not think it is needed.

> bluhm
>
> Index: Makefile
> ===
> RCS file: /cvs/ports/security/botan/Makefile,v
> retrieving revision 1.18
> diff -u -p -r1.18 Makefile
> --- Makefile  13 Mar 2017 01:31:20 -  1.18
> +++ Makefile  13 Mar 2017 01:46:23 -
> @@ -6,11 +6,12 @@ COMMENT=portable, easy to use, and effi
>  
>  DISTNAME=Botan-1.10.15
>  PKGNAME= ${DISTNAME:L}
> +REVISION =   0
>  CATEGORIES=  security
>  MASTER_SITES=${HOMEPAGE}releases/
>  EXTRACT_SUFX=.tgz
>  
> -SHARED_LIBS= botan-1.10  1.0
> +SHARED_LIBS= botan-1.10  1.1
>  
>  HOMEPAGE=http://botan.randombit.net/
>  
>

-- 
jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF  DDCC 0DFA 74AE 1524 E7EE