As and when I make this SSH server into a test suite, I'm not going to
want to wait for a gratuitous RSA key generation in every test run. So
now you can provide one in advance.
It has to be in SSH-1 format, because that's the format for which I
happen to already have internal API routines that return an RSAKey
instead of an opaque ssh_key. But since you also have to store it
without a passphrase, that doesn't really matter anyway.
This checks that the public Diffie-Hellman value sent by the server is
not an obviously silly one like 1 or -1 (mod p). We already had the
validation function, and were using it in standard DH key exchange,
but the parallel code in the GSSAPI case missed it out.
I completely forgot to check that the server had actually sent a key
of at least MINKLEN bits, as RFC 4432 clearly says that it MUST.
Without this restriction, not only can a server trick the client into
using a shared secret with inadequate entropy, but it can send a key
so short that the client attempts to generate a secret integer of
negative length, with integer-overflowing results.
The _nm strategy is slower, so I don't want to just change everything
over no matter what its contents. In this pass I've tried to catch
everything that holds the _really_ sensitive things like passwords,
private keys and session keys.
The local put_mp_*_from_string functions in import.c now take ptrlen
(which simplifies essentially all their call sites); so does the local
function logwrite() in logging.c, and so does ssh2_fingerprint_blob.
Similarly to the 'AES (unaccelerated)' naming scheme I added in the
AES rewrite, the hash functions that have multiple implementations now
each come with an annotation saying which one they are.
This was more tricky for hashes than for ciphers, because the
annotation for a hash has to be a separate string literal from the
base text name, so that it can propagate into the name field for each
HMAC wrapper without looking silly.
This is in preparation for a PRNG revamp which will want to have a
well defined boundary for any given request-for-randomness, so that it
can destroy the evidence afterwards. So no more looping round calling
random_byte() and then stopping when we feel like it: now you say up
front how many random bytes you want, and call random_read() which
gives you that many in one go.
Most of the call sites that had to be fixed are fairly mechanical, and
quite a few ended up more concise afterwards. A few became more
cumbersome, such as mp_random_bits, in which the new API doesn't let
me load the random bytes directly into the target integer without
triggering undefined behaviour, so instead I have to allocate a
separate temporary buffer.
The _most_ interesting call site was in the PKCS#1 v1.5 padding code
in sshrsa.c (used in SSH-1), in which you need a stream of _nonzero_
random bytes. The previous code just looped on random_byte, retrying
if it got a zero. Now I'm doing a much more interesting thing with an
mpint, essentially scaling a binary fraction repeatedly to extract a
number in the range [0,255) and then adding 1 to it.
ssh_rsakex_encrypt took an input (pointer, length) pair, which I've
replaced with a ptrlen; it also took an _output_ (pointer, length)
pair, and then re-computed the right length internally and enforced by
assertion that the one passed in matched it. Now it just returns a
strbuf of whatever length it computed, which saves the caller having
to compute the length at all.
Also, both ssh_rsakex_encrypt and ssh_rsakex_decrypt took their
arguments in a weird order; I think it looks more sensible to put the
RSA key first rather than last, so now they both have the common order
(key, hash, input data).
The abstract method ssh_key_sign(), and the concrete functions
ssh_rsakex_newkey() and rsa_ssh1_public_blob_len(), now each take a
ptrlen argument in place of a separate pointer and length pair.
Partly that's because I'm generally preferring ptrlens these days and
it keeps argument lists short and tidy-looking, but mostly it's
because it will make those functions easier to wrap in my upcoming
test system.
The old 'Bignum' data type is gone completely, and so is sshbn.c. In
its place is a new thing called 'mp_int', handled by an entirely new
library module mpint.c, with API differences both large and small.
The main aim of this change is that the new library should be free of
timing- and cache-related side channels. I've written the code so that
it _should_ - assuming I haven't made any mistakes - do all of its
work without either control flow or memory addressing depending on the
data words of the input numbers. (Though, being an _arbitrary_
precision library, it does have to at least depend on the sizes of the
numbers - but there's a 'formal' size that can vary separately from
the actual magnitude of the represented integer, so if you want to
keep it secret that your number is actually small, it should work fine
to have a very long mp_int and just happen to store 23 in it.) So I've
done all my conditionalisation by means of computing both answers and
doing bit-masking to swap the right one into place, and all loops over
the words of an mp_int go up to the formal size rather than the actual
size.
I haven't actually tested the constant-time property in any rigorous
way yet (I'm still considering the best way to do it). But this code
is surely at the very least a big improvement on the old version, even
if I later find a few more things to fix.
I've also completely rewritten the low-level elliptic curve arithmetic
from sshecc.c; the new ecc.c is closer to being an adjunct of mpint.c
than it is to the SSH end of the code. The new elliptic curve code
keeps all coordinates in Montgomery-multiplication transformed form to
speed up all the multiplications mod the same prime, and only converts
them back when you ask for the affine coordinates. Also, I adopted
extended coordinates for the Edwards curve implementation.
sshecc.c has also had a near-total rewrite in the course of switching
it over to the new system. While I was there, I've separated ECDSA and
EdDSA more completely - they now have separate vtables, instead of a
single vtable in which nearly every function had a big if statement in
it - and also made the externally exposed types for an ECDSA key and
an ECDH context different.
A minor new feature: since the new arithmetic code includes a modular
square root function, we can now support the compressed point
representation for the NIST curves. We seem to have been getting along
fine without that so far, but it seemed a shame not to put it in,
since it was suddenly easy.
In sshrsa.c, one major change is that I've removed the RSA blinding
step in rsa_privkey_op, in which we randomise the ciphertext before
doing the decryption. The purpose of that was to avoid timing leaks
giving away the plaintext - but the new arithmetic code should take
that in its stride in the course of also being careful enough to avoid
leaking the _private key_, which RSA blinding had no way to do
anything about in any case.
Apart from those specific points, most of the rest of the changes are
more or less mechanical, just changing type names and translating code
into the new API.
In the new modular SSH architecture, ssh2transport.c delegates the
actual KEX packet exchange to ssh2kex_coroutine, which has different
implementations for client and server. The KEX code actually in
ssh2transport.c consists of looping on the coroutine until it zeroes
out its state field in the ssh2transport state.
But if something goes wrong enough during KEX that we call
ssh_proto_error or any other fatal connection-terminating function,
then when we return to ssh2transport.c, the ssh2transport state won't
even exist for it to check that flag. Address Sanitiser pointed that
out to me recently, so here's a fix in which we set an 'aborted' flag
to tell the caller that its state has already been freed.
In the past, I've had a lot of macros which you call with double
parentheses, along the lines of debug(("format string", params)), so
that the inner parens protect the commas and permit the macro to treat
the whole printf-style argument list as one macro argument.
That's all very well, but it's a bit inconvenient (it doesn't leave
you any way to implement such a macro by prepending another argument
to the list), and now this code base's rules allow C99isms, I can
switch all those macros to using a single pair of parens, using the
C99 ability to say '...' in the parameter list of the #define and get
at the corresponding suffix of the arguments as __VA_ARGS__.
So I'm doing it. I've made the following printf-style macros variadic:
bpp_logevent, ppl_logevent, ppl_printf and debug.
While I'm here, I've also fixed up a collection of conditioned-out
calls to debug() in the Windows front end which were clearly expecting
a macro with a different calling syntax, because they had an integer
parameter first. If I ever have a need to condition those back in,
they should actually work now.
The flag 'cross_certifying' in the SSH-2 transport layer state is now
a pointer to the host key algorithm we expect to be certifying,
instead of a plain bool. That lets me check by assertion that it's
what we expected it to be after all the complicated key exchange has
happened.
(I have no reason to think this _will_ go wrong. When we cross-
certify, the desired algorithm should be the only one we put into our
KEXINIT host key algorithm list, so it should also be the only one we
can come out of the far end of KEXINIT having selected. But if
anything ever does go wrong with my KEXINIT handling then I'd prefer
an assertion failure to silently certifying the wrong key, and also,
this makes it clearer to static analysers - and perhaps also humans
reading the code - what we expect the situation to be.)
The event log messages generated during DH key exchange now include both the
modulus size and hash algorithm used as well as whether the DH parameters
are from one of the standardized groups or were supplied by the server
during Group Exchange.
My normal habit these days, in new code, is to treat int and bool as
_almost_ completely separate types. I'm still willing to use C's
implicit test for zero on an integer (e.g. 'if (!blob.len)' is fine,
no need to spell it out as blob.len != 0), but generally, if a
variable is going to be conceptually a boolean, I like to declare it
bool and assign to it using 'true' or 'false' rather than 0 or 1.
PuTTY is an exception, because it predates the C99 bool, and I've
stuck to its existing coding style even when adding new code to it.
But it's been annoying me more and more, so now that I've decided C99
bool is an acceptable thing to require from our toolchain in the first
place, here's a quite thorough trawl through the source doing
'boolification'. Many variables and function parameters are now typed
as bool rather than int; many assignments of 0 or 1 to those variables
are now spelled 'true' or 'false'.
I managed this thorough conversion with the help of a custom clang
plugin that I wrote to trawl the AST and apply heuristics to point out
where things might want changing. So I've even managed to do a decent
job on parts of the code I haven't looked at in years!
To make the plugin's work easier, I pushed platform front ends
generally in the direction of using standard 'bool' in preference to
platform-specific boolean types like Windows BOOL or GTK's gboolean;
I've left the platform booleans in places they _have_ to be for the
platform APIs to work right, but variables only used by my own code
have been converted wherever I found them.
In a few places there are int values that look very like booleans in
_most_ of the places they're used, but have a rarely-used third value,
or a distinction between different nonzero values that most users
don't care about. In these cases, I've _removed_ uses of 'true' and
'false' for the return values, to emphasise that there's something
more subtle going on than a simple boolean answer:
- the 'multisel' field in dialog.h's list box structure, for which
the GTK front end in particular recognises a difference between 1
and 2 but nearly everything else treats as boolean
- the 'urgent' parameter to plug_receive, where 1 vs 2 tells you
something about the specific location of the urgent pointer, but
most clients only care about 0 vs 'something nonzero'
- the return value of wc_match, where -1 indicates a syntax error in
the wildcard.
- the return values from SSH-1 RSA-key loading functions, which use
-1 for 'wrong passphrase' and 0 for all other failures (so any
caller which already knows it's not loading an _encrypted private_
key can treat them as boolean)
- term->esc_query, and the 'query' parameter in toggle_mode in
terminal.c, which _usually_ hold 0 for ESC[123h or 1 for ESC[?123h,
but can also hold -1 for some other intervening character that we
don't support.
In a few places there's an integer that I haven't turned into a bool
even though it really _can_ only take values 0 or 1 (and, as above,
tried to make the call sites consistent in not calling those values
true and false), on the grounds that I thought it would make it more
confusing to imply that the 0 value was in some sense 'negative' or
bad and the 1 positive or good:
- the return value of plug_accepting uses the POSIXish convention of
0=success and nonzero=error; I think if I made it bool then I'd
also want to reverse its sense, and that's a job for a separate
piece of work.
- the 'screen' parameter to lineptr() in terminal.c, where 0 and 1
represent the default and alternate screens. There's no obvious
reason why one of those should be considered 'true' or 'positive'
or 'success' - they're just indices - so I've left it as int.
ssh_scp_recv had particularly confusing semantics for its previous int
return value: its call sites used '<= 0' to check for error, but it
never actually returned a negative number, just 0 or 1. Now the
function and its call sites agree that it's a bool.
In a couple of places I've renamed variables called 'ret', because I
don't like that name any more - it's unclear whether it means the
return value (in preparation) for the _containing_ function or the
return value received from a subroutine call, and occasionally I've
accidentally used the same variable for both and introduced a bug. So
where one of those got in my way, I've renamed it to 'toret' or 'retd'
(the latter short for 'returned') in line with my usual modern
practice, but I haven't done a thorough job of finding all of them.
Finally, one amusing side effect of doing this is that I've had to
separate quite a few chained assignments. It used to be perfectly fine
to write 'a = b = c = TRUE' when a,b,c were int and TRUE was just a
the 'true' defined by stdbool.h, that idiom provokes a warning from
gcc: 'suggest parentheses around assignment used as truth value'!
This commit includes <stdbool.h> from defs.h and deletes my
traditional definitions of TRUE and FALSE, but other than that, it's a
100% mechanical search-and-replace transforming all uses of TRUE and
FALSE into the C99-standardised lowercase spellings.
No actual types are changed in this commit; that will come next. This
is just getting the noise out of the way, so that subsequent commits
can have a higher proportion of signal.
This is a major code reorganisation in preparation for making this
code base into one that can build an SSH server as well as a client.
(Mostly for purposes of using the server as a regression test suite
for the client, though I have some other possible uses in mind too.
However, it's currently no part of my plan to harden the server to the
point where it can sensibly be deployed in a hostile environment.)
In this preparatory commit, I've broken up the SSH-2 transport and
connection layers, and the SSH-1 connection layer, into multiple
source files, with each layer having its own header file containing
the shared type definitions. In each case, the new source file
contains code that's specific to the client side of the protocol, so
that a new file can be swapped in in its place when building the
server.
Mostly this is just a straightforward moving of code without changing
it very much, but there are a couple of actual changes in the process:
The parsing of SSH-2 global-request and channel open-messages is now
done by a new pair of functions in the client module. For channel
opens, I've invented a new union data type to be the return value from
that function, representing either failure (plus error message),
success (plus Channel instance to manage the new channel), or an
instruction to hand the channel over to a sharing downstream (plus a
pointer to the downstream in question).
Also, the tree234 of remote port forwardings in ssh2connection is now
initialised on first use by the client-specific code, so that's where
its compare function lives. The shared ssh2connection_free() still
takes responsibility for freeing it, but now has to check if it's
non-null first.
The outer shell of the ssh2_lportfwd_open method, for making a
local-to-remote port forwarding, is still centralised in
ssh2connection.c, but the part of it that actually constructs the
outgoing channel-open message has moved into the client code, because
that will have to change depending on whether the channel-open has to
have type direct-tcpip or forwarded-tcpip.
In the SSH-1 connection layer, half the filter_queue method has moved
out into the new client-specific code, but not all of it -
bidirectional channel maintenance messages are still handled
centrally. One exception is SSH_MSG_PORT_OPEN, which can be sent in
both directions, but with subtly different semantics - from server to
client, it's referring to a previously established remote forwarding
(and must be rejected if there isn't one that matches it), but from
client to server it's just a "direct-tcpip" request with no prior
context. So that one is in the client-specific module, and when I add
the server code it will have its own different handler.
Simon Tatham
Mark Tolley