The file's uint32_t alignment should be more restrictive than the
uint16_t alignment, so this SHOULD be safe ... and reduces build noise.
Co-authored-by: Robert Lipe <robertlipe@gpsbabel.org>
The ipv6cp-accept-local option was supposed to enable it, but it is
already enabled by default, with no way to disable it.
For coherency with IPv4 and IPX, this disables ipv6cp-accept-local by
default, and the option can be used to enable it.
This also enables it automatically when the local id is not
specified, in coherency with IPv4 and IPX, and as the documentation was
saying.
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
As seen in `inode_del` (and `inode_alloc`), a free (`NOT_INIT`) node can be found anywhere in `g_vfs_dev_nodes`; it’s a “sparse” list.
So when checking for free nodes, the iterating index variable (`e`) shall be used, not the counter (`count`). The code got that wrong.
For instance, if the first node is not available (type is not `VFS_TYPE_NOT_INIT`), then the function previously returned 0, no matter what, and was iterating the node list in vain.
This adds some basic checks to the subroutines of eap_input to check
that we have requested or agreed to doing EAP authentication before
doing any processing on the received packet. The motivation is to
make it harder for a malicious peer to disrupt the operation of pppd
by sending unsolicited EAP packets. Note that eap_success() already
has a check that the EAP client state is reasonable, and does nothing
(apart from possibly printing a debug message) if not.
Signed-off-by: Paul Mackerras <paulus@ozlabs.org>
