63 Commits

Author SHA1 Message Date
Marcelo Roberto Jimenez
28799f26c9 [pupnp-devel] Type mixup on x86_64 causes UPNP_E_OUTOF_MEMORY
Submitted By: Glen Masgai
after an UpnpSendActionAsync() for example, i get UPNP_E_OUTOF_MEMORY
in the callback using 1.4.4 on a x86_64 system. This happens in
http_MakeMessage(), which in some cases get called with wrong types
(int instead of size_t) in combination with format "b" and "Q".
The attached patch should fix this.


git-svn-id: https://pupnp.svn.sourceforge.net/svnroot/pupnp/trunk@166 119443c7-1b9e-41f8-b6fc-b9c35fce742c
2007-04-28 18:42:14 +00:00
Marcelo Roberto Jimenez
948c959be5 Case insensitive comparison in raw_find_str() (httpparser.c) as
suggested by Craig Nelson in SF Tracker [ 1689382 ] DLINK DIR-625 patch.


git-svn-id: https://pupnp.svn.sourceforge.net/svnroot/pupnp/trunk@158 119443c7-1b9e-41f8-b6fc-b9c35fce742c
2007-04-19 13:05:33 +00:00
Marcelo Roberto Jimenez
286a45d4ad * SF Tracker [ 1652827 ] UpnpRegisterRootDevice returned -104
Submitted By: Michael Andersen - miwer

	Issue was found to be related to sizeof (size_t) != sizeof (int)
	on AMD64 systems. Emil Ljungdahl's AMD64 patch has been applied along
	with some other fixes. Original user report follows:

	When I run upnpd I get the above mentioned error (UPNP_E_OUTOF_MEMORY).
	I've tried with 1.4.1 and 1.4.2-RC3, it's the same. I don't understand why,
	because I have plenty of RAM, and I even tried closing some applications,
	but it didn't help.

	$ upnpd eth1 br0

	The following is logged in the /var/log/messages:
	Feb 6 01:33:47 server upnpd[6933]: Error registering the root device with
	descDocUrl: http://192.168.0.1:49152/gatedesc.xml
	Feb 6 01:33:47 server upnpd[6933]: UpnpRegisterRootDevice returned -104


	I tried enabling debugging and it looks like it cannot allocate memory
	through the membuffer_append function. It's wierd because it's only a few
	bytes.

	Please note, that I enabled some extra debugging lines that were commented,
	in order to get more information. See attached files.



git-svn-id: https://pupnp.svn.sourceforge.net/svnroot/pupnp/trunk@154 119443c7-1b9e-41f8-b6fc-b9c35fce742c
2007-04-17 03:55:20 +00:00
Marcelo Roberto Jimenez
bdda54a66b Fix for 64 bit machines.
git-svn-id: https://pupnp.svn.sourceforge.net/svnroot/pupnp/trunk@152 119443c7-1b9e-41f8-b6fc-b9c35fce742c
2007-03-13 17:46:26 +00:00
Oxy
c3060e0d0b typedef modifications for use with Borland Builder
git-svn-id: https://pupnp.svn.sourceforge.net/svnroot/pupnp/trunk@141 119443c7-1b9e-41f8-b6fc-b9c35fce742c
2007-03-03 14:36:58 +00:00
Oxy
2ac3819f8c typedefs corrected, made MS VC++ specific to fix problems of that compiler
git-svn-id: https://pupnp.svn.sourceforge.net/svnroot/pupnp/trunk@140 119443c7-1b9e-41f8-b6fc-b9c35fce742c
2007-03-02 13:53:08 +00:00
Oxy
0d451c9c8e (long long) cast replaced by a generic data type ulong64 that can be defined separately for every used system/compiler (did not work with VC++)
git-svn-id: https://pupnp.svn.sourceforge.net/svnroot/pupnp/trunk@139 119443c7-1b9e-41f8-b6fc-b9c35fce742c
2007-03-01 10:08:04 +00:00
Marcelo Roberto Jimenez
bbd16d75ed Not a comment.
git-svn-id: https://pupnp.svn.sourceforge.net/svnroot/pupnp/trunk@138 119443c7-1b9e-41f8-b6fc-b9c35fce742c
2007-02-17 02:18:43 +00:00
Marcelo Roberto Jimenez
08834bf0f0 32/64 bits portability issues on *printf.
Use %zd for size_t, and cast to (long long) for off_t.


git-svn-id: https://pupnp.svn.sourceforge.net/svnroot/pupnp/trunk@129 119443c7-1b9e-41f8-b6fc-b9c35fce742c
2007-02-09 19:54:22 +00:00
Marcelo Roberto Jimenez
f64c539395 SF Tracker [ 1634922 ] Support for large files (>= 2 GiB), part 2
Submitted By: Jonathan - no_dice
Summary: This patch hopefully fixes the remaining types and related
code to enable files >= 2 GiB to be streamed. Jonathan claims to have
tested this with a patched version of ushare-0.9.8 and a D-Link DSM-520.



git-svn-id: https://pupnp.svn.sourceforge.net/svnroot/pupnp/trunk@121 119443c7-1b9e-41f8-b6fc-b9c35fce742c
2007-01-23 13:22:52 +00:00
Marcelo Roberto Jimenez
1d9e8712b7 Thorough revision of every call of http_MakeMessage() due to a bug introduced
in rev.79: largefile patch added.

http_MakeMessage() has a worst than brain damaged "printf" like interface.
In rev.79, the "N" format parameter must be an off_t. Every call of this
function with an "N" format parameter and an int passed on the stack 
would fail terribly.

Not every place I touched had a bug, but to review the format and keep my
sanity, I had to format it properly.



git-svn-id: https://pupnp.svn.sourceforge.net/svnroot/pupnp/trunk@96 119443c7-1b9e-41f8-b6fc-b9c35fce742c
2006-12-23 18:38:00 +00:00
Oxy
1716c7a10d largefile patch added
git-svn-id: https://pupnp.svn.sourceforge.net/svnroot/pupnp/trunk@79 119443c7-1b9e-41f8-b6fc-b9c35fce742c
2006-08-17 05:17:50 +00:00
Marcelo Roberto Jimenez
89e7a40fcc Removing unnecessary additional directory level.
git-svn-id: https://pupnp.svn.sourceforge.net/svnroot/pupnp/trunk@29 119443c7-1b9e-41f8-b6fc-b9c35fce742c
2006-07-04 02:44:17 +00:00