[systemd-devel] [PATCH] core/namespace: Protect /usr instead of /home with ProtectSystem=yes

2015-05-31 Thread Jason Pleau
A small typo in ee818b8 caused /home to be put in read-only instead of /usr when ProtectSystem was enabled (ie: not set to no). --- src/core/namespace.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/src/core/namespace.c b/src/core/namespace.c index 7d0b7e7..01a817b 100644

Re: [systemd-devel] [PATCH] core/namespace: Protect /usr instead of /home with ProtectSystem=yes

2015-05-31 Thread Martin Pitt
Hey Jason, Jason Pleau [2015-05-31 12:51 -0400]: A small typo in ee818b8 caused /home to be put in read-only instead of /usr when ProtectSystem was enabled (ie: not set to no). Argh, thanks for spotting and the patch! Applied. Martin -- Martin Pitt|

Re: [systemd-devel] udev --daemon is broken again

2015-05-31 Thread Abdó Roig-Maranges
Andrei Borzenkov writes: Was not it fixed by 693d371d30fee1da58365121801445b404416ada? No. The first time it broke was due to the udev manager wanting to be used by a single process, and was fixed in 040e689654ef08c63. Then it broke again by the commit you point 693d371d30fee1da58, due to

Re: [systemd-devel] udev --daemon is broken again

2015-05-31 Thread Tom Gundersen
On Sun, May 31, 2015 at 8:45 PM, Abdó Roig-Maranges abdo.r...@gmail.com wrote: Andrei Borzenkov writes: Was not it fixed by 693d371d30fee1da58365121801445b404416ada? No. The first time it broke was due to the udev manager wanting to be used by a single process, and was fixed in

Re: [systemd-devel] [RFC] load-fragment: use unquote_first_word in config_parse_exec

2015-05-31 Thread Filipe Brandenburger
Hi, On Sun, May 31, 2015 at 12:06 AM, Andrei Borzenkov arvidj...@gmail.com wrote: В Sat, 30 May 2015 23:29:29 -0700 Filipe Brandenburger filbran...@google.com пишет: - Handling a \; is ugly, it looks like a hack... unquote_first_word is not equipped to recognize that sequence, so I had to

Re: [systemd-devel] udev --daemon is broken again

2015-05-31 Thread Andrei Borzenkov
В Sun, 31 May 2015 15:47:25 +0200 Abdó Roig-Maranges abdo.r...@gmail.com пишет: Hi, Since the move of udev to sd-event, udevd --daemon is broken again, and breaks the initramfs on arch linux, for example. It turns out that sd_event_add_signal (in manager_listen) complains because the

Re: [systemd-devel] [PATCH v3 3/3] udev: input_id - use ABS_MT_SLOT{-1} to exclude non touch screen devices

2015-05-31 Thread Peter Hutterer
On Fri, May 29, 2015 at 02:54:54PM +0200, Andreas Pokorny wrote: Require touch screens to have a ABS_MT_SLOT axis while exclude devices that overlap with the MT range of axes. --- src/udev/udev-builtin-input_id.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git

Re: [systemd-devel] udev --daemon is broken again

2015-05-31 Thread Tom Gundersen
On Sun, May 31, 2015 at 9:45 PM, Tom Gundersen t...@jklm.no wrote: On Sun, May 31, 2015 at 8:45 PM, Abdó Roig-Maranges abdo.r...@gmail.com wrote: Andrei Borzenkov writes: Was not it fixed by 693d371d30fee1da58365121801445b404416ada? No. The first time it broke was due to the udev

[systemd-devel] bcache broke in udev 220

2015-05-31 Thread David Mohr
Hi, udev 220 arrived in Debian sid and this update broke bcache support [1]: no /dev/disk/by-uuid/ symlink is being created for the filesystem on top of /dev/bcache*. That's because 60-persistent-storage now uses a whitelist instead of blacklist. The simple attached patch fixes this, could

Re: [systemd-devel] [PATCH v2] udev: input_id: fix detection of various touchscreens

2015-05-31 Thread Peter Hutterer
On Fri, May 29, 2015 at 12:32:44PM +0200, Andreas Pokorny wrote: Hi, On Fri, May 29, 2015 at 2:07 AM, Peter Hutterer peter.hutte...@who-t.net wrote: On Thu, May 28, 2015 at 02:47:35PM +0200, Andreas Pokorny wrote: There are touch screens that do not provide 'BTN_TOUCH' and hence no

[systemd-devel] [PATCH] util: Refactor common cunescape block in unquote_first_word

2015-05-31 Thread Filipe Brandenburger
Get rid of code duplication here. --- Some alternatives I considered: 1) Keep a separate bool escape and set it when we see a backslash, still keeping state set to VALUE or SINGLE_QUOTE or DOUBLE_QUOTE. 2) Create the enum so that ESCAPE has a bit value like 0x100 or similar, so that we

[systemd-devel] [RFC] load-fragment: use unquote_first_word in config_parse_exec

2015-05-31 Thread Filipe Brandenburger
This is an attempt to convert it from using FOREACH_WORD_QUOTED into a loop using unquote_first_word repeatedly. Additionally, we're looping through the arguments only once and using _cleanup_ functions to manage objects owned by this function. Some notes: - There is some difference in how

Re: [systemd-devel] [RFC] load-fragment: use unquote_first_word in config_parse_exec

2015-05-31 Thread Andrei Borzenkov
В Sat, 30 May 2015 23:29:29 -0700 Filipe Brandenburger filbran...@google.com пишет: - Handling a \; is ugly, it looks like a hack... unquote_first_word is not equipped to recognize that sequence, so I had to move it outside unquote_first_word looking for those two characters followed by

Re: [systemd-devel] [PATCH] util: Refactor common cunescape block in unquote_first_word

2015-05-31 Thread systemd github import bot
Patchset imported to github. Pull request: https://github.com/systemd-devs/systemd/compare/master...systemd-mailing-devs:1433053753-2598-1-git-send-email-filbranden%40google.com -- Generated by https://github.com/haraldh/mail2git ___ systemd-devel

[systemd-devel] udev --daemon is broken again

2015-05-31 Thread Abdó Roig-Maranges
Hi, Since the move of udev to sd-event, udevd --daemon is broken again, and breaks the initramfs on arch linux, for example. It turns out that sd_event_add_signal (in manager_listen) complains because the sd_event event loop has moved through a fork. I assume the intention in sd-event is not

Re: [systemd-devel] [systemd-commits] Makefile.am

2015-05-31 Thread Kay Sievers
On Fri, May 29, 2015 at 6:53 PM, Lennart Poettering lenn...@poettering.net wrote: On Fri, 29.05.15 09:45, Filipe Brandenburger (filbran...@google.com) wrote: Hi, I was thinking about this one recently... I really think the correct solution is for man/custom-entities.ent to be generated by