WARNING - OLD ARCHIVES

This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
   
 
 
Xen 
 
Home Products Support Community News
 
   
 

xen-devel

Re: [Xen-devel] Re: [PATCH] libxl: check for early failures of qemu-dm (

To: Ian Jackson <Ian.Jackson@xxxxxxxxxxxxx>
Subject: Re: [Xen-devel] Re: [PATCH] libxl: check for early failures of qemu-dm (v2)
From: Vincent Hanquez <vincent.hanquez@xxxxxxxxxxxxx>
Date: Mon, 16 Nov 2009 14:52:56 +0000
Cc: "xen-devel@xxxxxxxxxxxxxxxxxxx" <xen-devel@xxxxxxxxxxxxxxxxxxx>
Delivery-date: Mon, 16 Nov 2009 06:46:56 -0800
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
In-reply-to: <19201.16668.970391.316245@xxxxxxxxxxxxxxxxxxxxxxxx>
List-help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-id: Xen developer discussion <xen-devel.lists.xensource.com>
List-post: <mailto:xen-devel@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
References: <19201.16668.970391.316245@xxxxxxxxxxxxxxxxxxxxxxxx>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: Thunderbird 2.0.0.23 (X11/20091001)
Ian Jackson wrote:
This patch makes xl create check whether qemu-dm has started
correctly, and causes it to fail immediately with appropriate errors
if not.  There are other bugfixes too.

More specifically:

 * libxl_create_device_model forks twice rather than once so that the
   process which calls libxl does not end up being the actual parent
   of qemu.  That avoids the need for the qemu-dm process to be reaped
   at some indefinite time in the future.

 * The first fork generates an intermediate process which is
   responsible for writing the qemu-dm pid to xenstore and then merely
   waits to collect and report on qemu-dm's exit status during
   startup.  New arguments to libxl_create_device_model allow the
   preservation of its pid so that a later call can check whether the
   startup is successful.
Did you have the qemu-dm ready patch in qemu ?
 * xl.c's create_domain checks for errors in its libxl calls.

Consequential changes:

 * libxl_wait_for_device_model now takes a callback function parameter
   which is called repeatedly in the loop iteration and allows the
   caller to abort the wait.

 * libxl_exec no longer calls fork; there is a new libxl_fork.
this is not libxl goal to provide wrapper for syscalls.
the exec should do the double fork+exec itself, no need to have a fork function.
 * osdep.[ch] new use #define _GNU_SOURCE.  The provided asprintf
   declarations are suppressed when not needed (currently, always).

 * There is a hook to override waitpid, which will be necessary for
   some callers.
why ?

Remaining problems and other issues I noticed or we found:

 * The error handling is rather inconsistent still and lacking in
   places.
* xl_logv is declared but not defined.
 * _GNU_SOURCE should be used throughout.  The asprintf implementation
   should be disabled in favour of the system one.
osdeps was just suppose to be available for netbsd. not sure why the contrary happens on christoph's patch.

 * XL_LOG_ERROR_ERRNO needs to actually print the errno value.
there shouldn't be a LOG_ERROR_ERRNO value. what you want is the different log function that will embedd an errno automatically in the fmt, not a different logging level. not sure what you meant later by "remove new error level, should be in future patch".

 * destroy_device_model can kill random dom0 processes (!)
* struct libxl_ctx should be defined in libxl_internal.h.
no, otherwise the init ctx need to allocate itself the memory of the context, instead of having
the caller "allocate it" itself on the fct stack.
Signed-off-by: Ian Jackson <ian.jackson@xxxxxxxxxxxxx>

(Changes since v1:
 * Remove new error log level; should be in a future patch
 * Properly fixed the asprintf problem
 * Indentation no uses literal tabs

--
Vincent

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel