[PULL] revert arm backtrace fix, introduce client id tracking

2010-10-21 Thread Rami Ylimäki

 Hi Keith,

The following changes applied on top of current master branch 
(1a0d9324b3d9fd93e685066e0e5cea0611878c0d)


Revert "Set DamageSetReportAfterOp to true for the damage 
extension" (#30260)


are available at:

git://gitorious.org/rjy-fdo/xserver.git

Rami Ylimäki (3):
  Revert "os: Prevent backtrace from being stopped in noreturn
functions."
  dix: Add facilities for client ID tracking.
  Xext: Use general OS functions to determine client command string in
SELinux.

 Xext/xselinux_hooks.c|   22 +--
 configure.ac |   28 ++--
 dix/Makefile.am  |1 +
 dix/client.c |  345 
++

 dix/main.c   |3 +
 hw/xfree86/loader/sdksyms.sh |1 +
 include/Makefile.am  |1 +
 include/client.h |   60 
 include/dix-config.h.in  |3 +
 include/os.h |3 +
 os/Makefile.am   |   17 +--
 os/access.c  |   76 +
 12 files changed, 517 insertions(+), 43 deletions(-)
 create mode 100644 dix/client.c
 create mode 100644 include/client.h

___
xorg-devel@lists.x.org: X.Org development
Archives: http://lists.x.org/archives/xorg-devel
Info: http://lists.x.org/mailman/listinfo/xorg-devel


Re: [PULL] revert arm backtrace fix, introduce client id tracking

2010-11-10 Thread Keith Packard
On Thu, 21 Oct 2010 12:47:04 +0300, Rami Ylimäki  wrote:

>dix: Add facilities for client ID tracking.

Seems like just sticking new fields in the client and having the OS
initialize them when the client connects would make most of this code
unnecessary.

-- 
keith.pack...@intel.com


pgpmFPIHvZoAy.pgp
Description: PGP signature
___
xorg-devel@lists.x.org: X.Org development
Archives: http://lists.x.org/archives/xorg-devel
Info: http://lists.x.org/mailman/listinfo/xorg-devel

Re: [PULL] revert arm backtrace fix, introduce client id tracking

2010-11-10 Thread Pat Kane
On Wed, Nov 10, 2010 at 4:54 PM, Keith Packard  wrote:
> Seems like just sticking new fields in the client and having the OS
> initialize them when the client connects would make most of this code
> unnecessary.

Yeah, I have a local hack that uses the 3 pad bytes in xConnClientPrefix
to pass client ID (PID) to the X server.

Pat
---
___
xorg-devel@lists.x.org: X.Org development
Archives: http://lists.x.org/archives/xorg-devel
Info: http://lists.x.org/mailman/listinfo/xorg-devel


Re: [PULL] revert arm backtrace fix, introduce client id tracking

2010-11-10 Thread Keith Packard
On Wed, 10 Nov 2010 17:22:37 -0600, Pat Kane  wrote:
> On Wed, Nov 10, 2010 at 4:54 PM, Keith Packard  wrote:
> > Seems like just sticking new fields in the client and having the OS
> > initialize them when the client connects would make most of this code
> > unnecessary.
> 
> Yeah, I have a local hack that uses the 3 pad bytes in xConnClientPrefix
> to pass client ID (PID) to the X server.

The code which computes uid/gid/pid from the connection seems
like the right idea, I'd just rather see that all done neatly in the os
layer without resorting to devPrivates in the client structure.

-- 
keith.pack...@intel.com


pgpmoUhzGCjss.pgp
Description: PGP signature
___
xorg-devel@lists.x.org: X.Org development
Archives: http://lists.x.org/archives/xorg-devel
Info: http://lists.x.org/mailman/listinfo/xorg-devel