tun/tun.patch

Mon, 17 Sep 2012 19:10:10 +0200

author
Michael Schloh von Bennewitz <michael@schloh.com>
date
Mon, 17 Sep 2012 19:10:10 +0200
changeset 689
9fe04d4d4e5a
parent 75
b3a835aa617c
permissions
-rw-r--r--

Update to new version of vendor software although Oracle fails to deliver.
More specifically, newer db(3) patch revisions exist but Oracle has
removed them from the canonical download server URI for Berkely DB.

michael@667 1 Index: solaris/configure
michael@667 2 --- solaris/configure.orig 2000-05-04 21:24:53.000000000 +0200
michael@667 3 +++ solaris/configure 2012-09-07 01:25:49.313354333 +0200
michael@667 4 @@ -769,6 +769,7 @@
michael@667 5 s%@oldincludedir@%$oldincludedir%g
michael@667 6 s%@infodir@%$infodir%g
michael@667 7 s%@mandir@%$mandir%g
michael@667 8 +s%@CC@%$CC%g
michael@667 9 s%@INSTALL_PROGRAM@%$INSTALL_PROGRAM%g
michael@667 10 s%@INSTALL_SCRIPT@%$INSTALL_SCRIPT%g
michael@667 11 s%@INSTALL_DATA@%$INSTALL_DATA%g
michael@75 12 Index: linux/2.2/tun.c
michael@75 13 --- linux/2.2/tun.c.orig 2006-10-10 14:45:00.338589000 +0200
michael@75 14 +++ linux/2.2/tun.c 2006-10-10 14:44:05.695404000 +0200
michael@75 15 @@ -178,10 +178,22 @@
michael@75 16
michael@75 17 DBG( KERN_INFO "%s: tun_chr_poll\n", tun->name);
michael@75 18
michael@75 19 + /* Data written to the /dev/tunX device is immediately placed into a socket buffer, making it
michael@75 20 + * available to networking code at the tunX interface. Writes never block.
michael@75 21 + * Likewise, data flows from the network stack, through the tunX interface and into the /dev/tun* device,
michael@75 22 + * where it is queued, making it available for read().
michael@75 23 + * Thus the character device /dev/tunX is:
michael@75 24 + * - readable if data was "transmitted" to the tunX interface and is now queued at the /dev/tunX device.
michael@75 25 + * - always writable.
michael@75 26 + * Everything written here is equally true of taps.
michael@75 27 + * The author made a mistake when implementing this routine; he forgot that the device is always writable.
michael@75 28 + * -jeff stearns 22-Dec-2005
michael@75 29 + */
michael@75 30 +
michael@75 31 poll_wait(file, &tun->read_wait, wait);
michael@75 32
michael@75 33 if( skb_queue_len(&tun->txq) )
michael@75 34 - return POLLIN | POLLRDNORM;
michael@75 35 + return POLLIN | POLLRDNORM | POLLOUT | POLLWRNORM;
michael@75 36
michael@75 37 return POLLOUT | POLLWRNORM;
michael@75 38 }
michael@75 39 Index: linux/2.4/tun.c
michael@75 40 --- linux/2.4/tun.c.orig 2006-10-10 14:41:57.910408000 +0200
michael@75 41 +++ linux/2.4/tun.c 2006-10-10 14:43:40.067700000 +0200
michael@75 42 @@ -176,9 +176,21 @@
michael@75 43 DBG(KERN_INFO "%s: tun_chr_poll\n", tun->name);
michael@75 44
michael@75 45 poll_wait(file, &tun->read_wait, wait);
michael@75 46 +
michael@75 47 + /* Data written to the /dev/tunX device is immediately placed into a socket buffer, making it
michael@75 48 + * available to networking code at the tunX interface. Writes never block.
michael@75 49 + * Likewise, data flows from the network stack, through the tunX interface and into the /dev/tun* device,
michael@75 50 + * where it is queued, making it available for read().
michael@75 51 + * Thus the character device /dev/tunX is:
michael@75 52 + * - readable if data was "transmitted" to the tunX interface and is now queued at the /dev/tunX device.
michael@75 53 + * - always writable.
michael@75 54 + * Everything written here is equally true of taps.
michael@75 55 + * The author made a mistake when implementing this routine; he forgot that the device is always writable.
michael@75 56 + * -jeff stearns 22-Dec-2005
michael@75 57 + */
michael@75 58
michael@75 59 if (skb_queue_len(&tun->txq))
michael@75 60 - return POLLIN | POLLRDNORM;
michael@75 61 + return POLLIN | POLLRDNORM | POLLOUT | POLLWRNORM;
michael@75 62
michael@75 63 return POLLOUT | POLLWRNORM;
michael@75 64 }

mercurial