PPP, move protocols initialization from ppp_clear to ppp_new

What protocols init functions are meant to is to be called once to set
the default configuration before user specific configuration is set.

Until now, we reset to the default configuration just before
reconnecting, thus without allowing any time frame to let users change
it. That was fine until one user asked to be able to do that.

This change move protocols init functions calls from ppp_clear to
ppp_new, meaning user configuration is not overwritten anymore.

PPP, CCP, move user configuration from ccp_init to ccp_resetci

ccp_init() is meant to be called once, providing default configuration
before user specific configuration is applied. In pppd user configuration
is set just after all protocol init functions are called, in lwIP we
use protocol reset functions to do that, which is more or less the same.

Therefore, be consistent and move CCP user configuration from ccp_init
to ccp_resetci.

PPP: Use LWIP_ARRAYSIZE at appropriate places

Use LWIP_ARRAYSIZE to simplify the code.

Signed-off-by: Axel Lin <axel.lin@ingics.com>
Reviewed-by: Sylvain Rochet <gradator@gradator.net>
[Sylvain Rochet <gradator@gradator.net>: added missing casts]
Signed-off-by: Sylvain Rochet <gradator@gradator.net>

Work on task #13907 PPP cleanups: Move PPP options from opt.h to dedicated ppp_opts.h file

PPP: more const and mixed u_char/char types fixes

PPP: fix constness in PPP related files when PRINTPKT_SUPPORT is enabled

PPP, removed unnecessary memset()

Everything is cleared in ppp_clear(), we don't need to clear all
structures twice.

PPP, MPPE, issue CCP reset request if synchronization is lost

If MPPE synchronization is lost, it is advised to send a CCP reset
request in order to recover to a clean state.

PPP, place print packet debug codenames in ROM

PPP, CCP, MPPE, handle CCP Reset-Request and Reset-Ack for MPPE

PPP, CCP, added receive and transmit chosen protocols in ccp_flags_set() (renamed ccp_set())

We need to know which methods were chosen when CCP is up, this used to be done
using ccp_test() which we are in the process of removing.

Using non-existing method 0 instead of -1 in CCP for unset method, allowing
type change from s16_t to u8_t for method.

PPP, CCP, added MPPE user configuration flags

PPP, MPPE, reworked MPPE debug, using ppp_pcb metadata

PPP, MSCHAP, CCP, MPPE: reworked key passing from CCP to MPPE

Removed mutiple copies of keys by pre-setting MPPE keys during MSCHAP
negotiation.

Improved MPPE init so we don't need to pass a buffer formatted in a
special way to MPPE, this is necessary for pppd to talk to the kernel,
we don't need that here.

PPP, CCP, MPPE: calling mppe_init() in ccp.c instead of using ccp_test()

We don't need this ccp_test() stuff which is used to test if kernel have
the ability to process MPPE packets. Reworked MPPE so we don't need it.

PPP, CCP, reworked ppp_pcb ccp_localstate and all_rejected fields

PPP, MPPE, moved MPPE global variables from MSCHAP to ppp_pcb

PPP, CCP, deflate, BSD compress, predictor 1 & 2 are now optional at compile time

PPP, CCP, compile-out a bit more MPPE code if MPPE is disabled

PPP, CCP, re-added missing ccp_test()

ccp_test() is not only used to test kernel support, but also to set MPPE keys,
we will change that further, but for now, re-add the necessary ccp_test()

PPP, CCP, added missing defines from linux/ppp-comp.h to ccp.h

PPP, CCP, moved ccp_test() from ccp_addci() to ccp_resetci()

We don't need to test our available compression protocols that much, we know
what we are going to be able to use in embedded. Even PPPd don't need that.

Checking only in ccp_resetci() is enough, this way we know our cilen size
before ccp_addci() is called. PPPd is designed that *_addci() can use less
bytes than it previously asked in *_cilen(), we prefer to avoid that here.

We actually don't need at all ccp_test() calls, we know what we are
able to support at compile time, we don't need to ask the kernel like
PPPd does. This is just the first step in ccp_test() removal.

PPP, CCP: Fix ccp_cilen returning size if both deflate_correct and deflate_draft are enabled

This fixes a bug returning 4 byte less than necessary for the addci buffer
if both deflate_correct and deflate_draft are enabled.

PPP, MPPE, replaced #ifdef MPPE to #if MPPE_SUPPORT

PPP, CCP, various fix if MPPE is enabled

PPP, CCP, various compiler warnings fix

PPP, CCP, updated to our utils.c function name

PPP, CCP, updated ccp_printpkt() prototype to match other *_printpkt()

PPP, CCP, updated ANY_COMPRESS() to use ccp_options*

PPP, CCP, commented-in unused global variables used for options

PPP, CCP, reworked to our ppp_pcb* pointer and ppp_pcb->ccp* data field

PPP, CCP, moved CCP global variables to ppp_pcb (ppp.h)

PPP, CCP, updated FSM state with our FSM state define PPP_FSM_x

PPP, CCP, commented-in unused PPP config file parsing

PPP, CORE, Removed (*datainput) from struct protent if not used

Data input is only used by CCP and ECP, which are not supported at this time,
remove this entry from struct protent to save some flash.

PPP, CORE, Removed useless enabled_flag from struct protent

Our struct protent are const everywhere to save RAM, enable/disable
flag on a const struct is useless, saving some flash, removed.

PPP, fixed some compiler warnings

Using -Wall -pedantic -Wparentheses -Wsequence-point
-Wswitch-default -Wextra -Wundef -Wshadow -Wpointer-arith
-Wbad-function-cast -Wc++-compat -Wwrite-strings -Wold-style-definition
-Wmissing-prototypes -Wredundant-decls -Wnested-externs -Wno-address

PPP, adding const pragma to FSM callbacks, saving about 350 bytes in .data segment

PPP, adding const pragma to struct protent, saving about 200 bytes in .data segment

Moved PPP headers into include/netif/ppp/, fixing bug #37040.

started the unit to ppp_pcb replacement

cleaned depreacted __P() and __V() macros, removed the legacy varargs.h header

ppp_impl.h is back!

don't build any PPP file if PPP support is disabled

removed pppd.h and ppp.c, renamed pppmy.c to ppp.c and pppmy.h to ppp.h

added pppmy.h in all .c files to easily find conflicts

removed more protent fields only used by print packets functions

print packet functions are now optional

demand support is now a compile-time option

Obviously, it requires some wiring to know if there is new activity
on a not-yet established PPP interface with the default route
already set.

I don't think any lwIP user will ever need that, all should know
when to bring the link up and down.

Linux ioctl() gone, ECP support optional, CCP support optional

Removed all the used Linux ioctl(), sys_linux.c is entirely disabled.

Replaced unecessary functions to set up a PPP link to null fonctions,
however all stuff necessary to shutdown a PPP link should be done.

Some utils functions related to I/O files are disabled too.

ECP and CCP support only set PPP attributes into the PPP kernel support,
they are now compile-time options and are disabled by default and
obviously not supported (yet? :p)

replaced BCOPY to lwip-MEMCPY

more and more ppp options removal

disabled almost all PPP options strings and support as well as useless file-based auth code

Revert "Removed all stuff requiring encryption."

This reverts commit c268c5e07c046eb2cb8e5798a1f3eba7e0ad13c1.

Conflicts:

src/netif/ppp/auth.c
src/netif/ppp/ppp.c
src/netif/ppp/pppmy.c
src/netif/ppp/pppmy.h

Removed all stuff requiring encryption.

OpenBSD DES require more than 70 kB of static memory.

The GNU libcrypt DES, for information, require more than 131 kB,
so this is probably the case with all non memory optimised DES.

PolarSSL only required 132 bytes of stack with some kB of .rodata
precomputed tables :-)

I personally don't need MS CHAP v1 or MS CHAP v2, and that was not
supported in the previous PPP port, so there is no regression,
I feel comfortable about removing those hard to port stuff.

If someone want to do the MS CHAP port, he first have to find or
do a small memory footprint DES implementation.

Early development stage of an attempt to port PPPd 2.4.5 to lwIP.

PPPoE works, PPPoS code is not ported at all.

I am using the RP-PPPoE server to do my tests using the following
configuration:

$ cat /etc/ppp/pppoe-server-options
debug
login
lcp-echo-interval 10
lcp-echo-failure 10
ms-dns 192.168.4.130
ms-dns 192.168.4.231
netmask 255.255.255.0
defaultroute
noipdefault
usepeerdns

$ cat /etc/ppp/allip
192.168.4.1-200

$ pppoe-server -C isp -L 192.168.4.254 -p /etc/ppp/allip -I tap0

Plus the usual auth-lines in /etc/ppp/pap-secrets and
/etc/ppp/chap-secrets .

And the unix port minimal "echo" project slightly modified to use
the "tcpip" API, so with threads, which I am going to commit with
NO_SYS as a -Dmacro.

It still use some of the linux'ism, such as syslog() and crypt(),

I do not want to drop the syslog() supports at the moment, this is
pretty useful to debug, and we may just convert the way the syslog() is
done to provide a trace feature to our PPP users, as a compile-time
option.