mirror of
https://git.tartarus.org/simon/putty.git
synced 2025-04-10 15:48:06 -05:00
dfdb73e103
sshaes.c is more or less completely changed by this commit. Firstly, I've changed the top-level structure. In the old structure, there were three levels of indirection controlling what an encryption function would actually do: first the ssh2_cipher vtable, then a subsidiary set of function pointers within that to select the software or hardware implementation, and then inside the main encryption function, a switch on the key length to jump into the right place in the unrolled loop of cipher rounds. That was all a bit untidy. So now _all_ of that is done by means of just one selection system, namely the ssh2_cipher vtable. The software and hardware implementations of a given SSH cipher each have their own separate vtable, e.g. ssh2_aes256_sdctr_sw and ssh2_aes256_sdctr_hw; this allows them to have their own completely different state structures too, and not have to try to coexist awkwardly in the same universal AESContext with workaround code to align things correctly. The old implementation-agnostic vtables like ssh2_aes256_sdctr still exist, but now they're mostly empty, containing only the constructor function, which will decide whether AES-NI is currently available and then choose one of the other _real_ vtables to instantiate. As well as the cleaner data representation, this also means the vtables can have different description strings, which means the Event Log will indicate which AES implementation is actually in use; it means the SW and HW vtables are available for testcrypt to use (although actually using them is left for the next commit); and in principle it would also make it easy to support a user override for the automatic SW/HW selection (in case anyone turns out to want one). The AES-NI implementation has been reorganised to fit into the new framework. One thing I've done is to de-optimise the key expansion: instead of having a separate blazingly fast loop-unrolled key setup function for each key length, there's now just one, which uses AES intrinsics for the actual transformations of individual key words, but wraps them in a common loop structure for all the key lengths which has a clear correspondence to the cipher spec. (Sorry to throw away your work there, Pavel, but this isn't an application where key setup really _needs_ to be hugely fast, and I decided I prefer a version I can understand and debug.) The software AES implementation is also completely replaced with one that uses a bit-sliced representation, i.e. the cipher state is split across eight integers in such a way that each logical byte of the state occupies a single bit in each of those integers. The S-box lookup is done by a long string of AND and XOR operations on the eight bits (removing the potential cache side channel from a lookup table), and this representation allows 64 S-box lookups to be done in parallel simply by extending those AND/XOR operations to be bitwise ones on a whole word. So now we can perform four AES encryptions or decryptions in parallel, at least when the cipher mode permits it (which SDCTR and CBC decryption both do). The result is slower than the old implementation, but (a) not by as much as you might think - those parallel S-boxes are surprisingly competitive with 64 separate table lookups; (b) the compensation is that now it should run in constant time with no data-dependent control flow or memory addressing; and (c) in any case the really fast hardware implementation will supersede it for most users.
This is the README for the source archive of PuTTY, a free Windows
and Unix Telnet and SSH client.
If you want to rebuild PuTTY from source, we provide a variety of
Makefiles and equivalents. (If you have fetched the source from
Git, you'll have to generate the Makefiles yourself -- see
below.)
There are various compile-time directives that you can use to
disable or modify certain features; it may be necessary to do this
in some environments. They are documented in `Recipe', and in
comments in many of the generated Makefiles.
For building on Windows:
- windows/Makefile.vc is for command-line builds on MS Visual C++
systems. Change into the `windows' subdirectory and type `nmake
-f Makefile.vc' to build all the PuTTY binaries.
As of 2017, we successfully compile PuTTY with both Visual Studio
7 (2003) and Visual Studio 14 (2015), so our guess is that it will
probably build with versions in between those as well.
(The binaries from Visual Studio 14 are only compatible with
Windows XP and up. Binaries from Visual Studio 7 ought to work
with anything from Windows 95 onward.)
- Inside the windows/MSVC subdirectory are MS Visual Studio project
files for doing GUI-based builds of the various PuTTY utilities.
These have been tested on Visual Studio 7 and 10.
You should be able to build each PuTTY utility by loading the
corresponding .dsp file in Visual Studio. For example,
MSVC/putty/putty.dsp builds PuTTY itself, MSVC/plink/plink.dsp
builds Plink, and so on.
- windows/Makefile.mgw is for MinGW / Cygwin installations. Type
`make -f Makefile.mgw' while in the `windows' subdirectory to
build all the PuTTY binaries.
MinGW and friends can lag behind other toolchains in their support
for the Windows API. Compile-time levers are provided to exclude
some features; the defaults are set appropriately for the
'mingw-w64' cross-compiler provided with Ubuntu 14.04. If you are
using an older toolchain, you may need to exclude more features;
alternatively, you may find that upgrading to a recent version of
the 'w32api' package helps.
- windows/Makefile.lcc is for lcc-win32. Type `make -f
Makefile.lcc' while in the `windows' subdirectory. (You will
probably need to specify COMPAT=-DNO_MULTIMON.)
- Inside the windows/DEVCPP subdirectory are Dev-C++ project
files for doing GUI-based builds of the various PuTTY utilities.
The PuTTY team actively use Makefile.vc (with VC7/10) and Makefile.mgw
(with mingw32), so we'll probably notice problems with those
toolchains fairly quickly. Please report any problems with the other
toolchains mentioned above.
For building on Unix:
- unix/configure is for Unix and GTK. If you don't have GTK, you
should still be able to build the command-line utilities (PSCP,
PSFTP, Plink, PuTTYgen) using this script. To use it, change into
the `unix' subdirectory, run `./configure' and then `make'. Or you
can do the same in the top-level directory (we provide a little
wrapper that invokes configure one level down), which is more like
a normal Unix source archive but doesn't do so well at keeping the
per-platform stuff in each platform's subdirectory; it's up to you.
- unix/Makefile.gtk and unix/Makefile.ux are for non-autoconfigured
builds. These makefiles expect you to change into the `unix'
subdirectory, then run `make -f Makefile.gtk' or `make -f
Makefile.ux' respectively. Makefile.gtk builds all the programs but
relies on Gtk, whereas Makefile.ux builds only the command-line
utilities and has no Gtk dependence.
- For the graphical utilities, any of Gtk+-1.2, Gtk+-2.0, and Gtk+-3.0
should be supported. If you have more than one installed, you can
manually specify which one you want by giving the option
'--with-gtk=N' to the configure script where N is 1, 2, or 3.
(The default is the newest available, of course.) In the absence
of any Gtk version, the configure script will automatically
construct a Makefile which builds only the command-line utilities;
you can manually create this condition by giving configure the
option '--without-gtk'.
- pterm would like to be setuid or setgid, as appropriate, to permit
it to write records of user logins to /var/run/utmp and
/var/log/wtmp. (Of course it will not use this privilege for
anything else, and in particular it will drop all privileges before
starting up complex subsystems like GTK.) By default the makefile
will not attempt to add privileges to the pterm executable at 'make
install' time, but you can ask it to do so by running configure
with the option '--enable-setuid=USER' or '--enable-setgid=GROUP'.
- The Unix Makefiles have an `install' target. Note that by default
it tries to install `man' pages; if you have fetched the source via
Git then you will need to have built these using Halibut
first - see below.
- It's also possible to build the Windows version of PuTTY to run
on Unix by using Winelib. To do this, change to the `windows'
directory and run `make -f Makefile.mgw CC=winegcc RC=wrc'.
All of the Makefiles are generated automatically from the file
`Recipe' by the Perl script `mkfiles.pl' (except for the Unix one,
which is generated by the `configure' script; mkfiles.pl only
generates the input to automake). Additions and corrections to Recipe,
mkfiles.pl and/or configure.ac are much more useful than additions and
corrections to the actual Makefiles, Makefile.am or Makefile.in.
The Unix `configure' script and its various requirements are generated
by the shell script `mkauto.sh', which requires GNU Autoconf, GNU
Automake, and Gtk; if you've got the source from Git rather
than using one of our source snapshots, you'll need to run this
yourself. The input file to Automake is generated by mkfiles.pl along
with all the rest of the makefiles, so you will need to run mkfiles.pl
and then mkauto.sh.
Documentation (in various formats including Windows Help and Unix
`man' pages) is built from the Halibut (`.but') files in the `doc'
subdirectory using `doc/Makefile'. If you aren't using one of our
source snapshots, you'll need to do this yourself. Halibut can be
found at <https://www.chiark.greenend.org.uk/~sgtatham/halibut/>.
The PuTTY home web site is
https://www.chiark.greenend.org.uk/~sgtatham/putty/
If you want to send bug reports or feature requests, please read the
Feedback section of the web site before doing so. Sending one-line
reports saying `it doesn't work' will waste your time as much as
ours.
See the file LICENCE for the licence conditions.