putty/marshal.h

324 строки
13 KiB
C
Исходник Обычный вид История

New centralised binary-data marshalling system. I've finally got tired of all the code throughout PuTTY that repeats the same logic about how to format the SSH binary primitives like uint32, string, mpint. We've got reasonably organised code in ssh.c that appends things like that to 'struct Packet'; something similar in sftp.c which repeats a lot of the work; utility functions in various places to format an mpint to feed to one or another hash function; and no end of totally ad-hoc stuff in functions like public key blob formatters which actually have to _count up_ the size of data painstakingly, then malloc exactly that much and mess about with PUT_32BIT. It's time to bring all of that into one place, and stop repeating myself in error-prone ways everywhere. The new marshal.h defines a system in which I centralise all the actual marshalling functions, and then layer a touch of C macro trickery on top to allow me to (look as if I) pass a wide range of different types to those functions, as long as the target type has been set up in the right way to have a write() function. This commit adds the new header and source file, and sets up some general centralised types (strbuf and the various hash-function contexts like SHA_State), but doesn't use the new calls for anything yet. (I've also renamed some internal functions in import.c which were using the same names that I've just defined macros over. That won't last long - those functions are going to go away soon, so the changed names are strictly temporary.)
2018-05-24 11:17:13 +03:00
#ifndef PUTTY_MARSHAL_H
#define PUTTY_MARSHAL_H
#include "defs.h"
#include <stdio.h>
New centralised binary-data marshalling system. I've finally got tired of all the code throughout PuTTY that repeats the same logic about how to format the SSH binary primitives like uint32, string, mpint. We've got reasonably organised code in ssh.c that appends things like that to 'struct Packet'; something similar in sftp.c which repeats a lot of the work; utility functions in various places to format an mpint to feed to one or another hash function; and no end of totally ad-hoc stuff in functions like public key blob formatters which actually have to _count up_ the size of data painstakingly, then malloc exactly that much and mess about with PUT_32BIT. It's time to bring all of that into one place, and stop repeating myself in error-prone ways everywhere. The new marshal.h defines a system in which I centralise all the actual marshalling functions, and then layer a touch of C macro trickery on top to allow me to (look as if I) pass a wide range of different types to those functions, as long as the target type has been set up in the right way to have a write() function. This commit adds the new header and source file, and sets up some general centralised types (strbuf and the various hash-function contexts like SHA_State), but doesn't use the new calls for anything yet. (I've also renamed some internal functions in import.c which were using the same names that I've just defined macros over. That won't last long - those functions are going to go away soon, so the changed names are strictly temporary.)
2018-05-24 11:17:13 +03:00
/*
* A sort of 'abstract base class' or 'interface' or 'trait' which is
* the common feature of all types that want to accept data formatted
* using the SSH binary conventions of uint32, string, mpint etc.
*/
struct BinarySink {
void (*write)(BinarySink *sink, const void *data, size_t len);
BinarySink *binarysink_;
};
/*
* To define a structure type as a valid target for binary formatted
* data, put 'BinarySink_IMPLEMENTATION' in its declaration, and when
* an instance is set up, use 'BinarySink_INIT' to initialise the
* 'base class' state, providing a function pointer to be the
* implementation of the write() call above.
*/
#define BinarySink_IMPLEMENTATION BinarySink binarysink_[1]
#define BinarySink_INIT(obj, writefn) \
((obj)->binarysink_->write = (writefn), \
(obj)->binarysink_->binarysink_ = (obj)->binarysink_)
/*
* To define a larger structure type as a valid BinarySink in such a
* way that it will delegate the write method to some other object,
* put 'BinarySink_DELEGATE_IMPLEMENTATION' in its declaration, and
* when an instance is set up, use 'BinarySink_DELEGATE_INIT' to point
* at the object it wants to delegate to.
*
* In such a delegated structure, you might sometimes want to have the
* delegation stop being valid (e.g. it might be delegating to an
* object that only sometimes exists). You can null out the delegate
* pointer using BinarySink_DELEGATE_CLEAR.
*/
#define BinarySink_DELEGATE_IMPLEMENTATION BinarySink *binarysink_
#define BinarySink_DELEGATE_INIT(obj, othersink) \
((obj)->binarysink_ = BinarySink_UPCAST(othersink))
#define BinarySink_DELEGATE_CLEAR(obj) ((obj)->binarysink_ = NULL)
New centralised binary-data marshalling system. I've finally got tired of all the code throughout PuTTY that repeats the same logic about how to format the SSH binary primitives like uint32, string, mpint. We've got reasonably organised code in ssh.c that appends things like that to 'struct Packet'; something similar in sftp.c which repeats a lot of the work; utility functions in various places to format an mpint to feed to one or another hash function; and no end of totally ad-hoc stuff in functions like public key blob formatters which actually have to _count up_ the size of data painstakingly, then malloc exactly that much and mess about with PUT_32BIT. It's time to bring all of that into one place, and stop repeating myself in error-prone ways everywhere. The new marshal.h defines a system in which I centralise all the actual marshalling functions, and then layer a touch of C macro trickery on top to allow me to (look as if I) pass a wide range of different types to those functions, as long as the target type has been set up in the right way to have a write() function. This commit adds the new header and source file, and sets up some general centralised types (strbuf and the various hash-function contexts like SHA_State), but doesn't use the new calls for anything yet. (I've also renamed some internal functions in import.c which were using the same names that I've just defined macros over. That won't last long - those functions are going to go away soon, so the changed names are strictly temporary.)
2018-05-24 11:17:13 +03:00
/*
* The implementing type's write function will want to downcast its
* 'BinarySink *' parameter back to the more specific type. Also,
* sometimes you'll want to upcast a pointer to a particular
* implementing type into an abstract 'BinarySink *' to pass to
* generic subroutines not defined in this file. These macros do that
* job.
*
* Importantly, BinarySink_UPCAST can also be applied to a BinarySink
* * itself (and leaves it unchanged). That's achieved by a small
* piece of C trickery: implementing structures and the BinarySink
* structure itself both contain a field called binarysink_, but in
* implementing objects it's a BinarySink[1] whereas in the abstract
* type it's a 'BinarySink *' pointing back to the same structure,
* meaning that you can say 'foo->binarysink_' in either case and get
* a pointer type by different methods.
*/
#define BinarySink_DOWNCAST(object, type) \
TYPECHECK((object) == ((type *)0)->binarysink_, \
((type *)(((char *)(object)) - offsetof(type, binarysink_))))
#define BinarySink_UPCAST(object) \
TYPECHECK((object)->binarysink_ == (BinarySink *)0, \
(object)->binarysink_)
/*
* If you structure-copy an object that's implementing BinarySink,
* then that tricky self-pointer in its trait subobject will point to
* the wrong place. You could call BinarySink_INIT again, but this
* macro is terser and does all that's needed to fix up the copied
* object.
*/
#define BinarySink_COPIED(obj) \
((obj)->binarysink_->binarysink_ = (obj)->binarysink_)
/*
* The put_* macros are the main client to this system. Any structure
* which implements the BinarySink 'trait' is valid for use as the
* first parameter of any of these put_* macros.
*/
/* Basic big-endian integer types. */
New centralised binary-data marshalling system. I've finally got tired of all the code throughout PuTTY that repeats the same logic about how to format the SSH binary primitives like uint32, string, mpint. We've got reasonably organised code in ssh.c that appends things like that to 'struct Packet'; something similar in sftp.c which repeats a lot of the work; utility functions in various places to format an mpint to feed to one or another hash function; and no end of totally ad-hoc stuff in functions like public key blob formatters which actually have to _count up_ the size of data painstakingly, then malloc exactly that much and mess about with PUT_32BIT. It's time to bring all of that into one place, and stop repeating myself in error-prone ways everywhere. The new marshal.h defines a system in which I centralise all the actual marshalling functions, and then layer a touch of C macro trickery on top to allow me to (look as if I) pass a wide range of different types to those functions, as long as the target type has been set up in the right way to have a write() function. This commit adds the new header and source file, and sets up some general centralised types (strbuf and the various hash-function contexts like SHA_State), but doesn't use the new calls for anything yet. (I've also renamed some internal functions in import.c which were using the same names that I've just defined macros over. That won't last long - those functions are going to go away soon, so the changed names are strictly temporary.)
2018-05-24 11:17:13 +03:00
#define put_byte(bs, val) \
BinarySink_put_byte(BinarySink_UPCAST(bs), val)
#define put_uint16(bs, val) \
BinarySink_put_uint16(BinarySink_UPCAST(bs), val)
#define put_uint32(bs, val) \
BinarySink_put_uint32(BinarySink_UPCAST(bs), val)
#define put_uint64(bs, val) \
BinarySink_put_uint64(BinarySink_UPCAST(bs), val)
/* SSH booleans, encoded as a single byte storing either 0 or 1. */
#define put_bool(bs, val) \
BinarySink_put_bool(BinarySink_UPCAST(bs), val)
/* SSH strings, with a leading uint32 length field. 'stringz' is a
* convenience function that takes an ordinary C zero-terminated
* string as input. 'stringsb' takes a strbuf * as input, and
* finalises it as a side effect (handy for multi-level marshalling in
* which you use these same functions to format an inner blob of data
* that then gets wrapped into a string container in an outer one). */
#define put_string(bs, val, len) \
BinarySink_put_string(BinarySink_UPCAST(bs),val,len)
#define put_stringpl(bs, ptrlen) \
BinarySink_put_stringpl(BinarySink_UPCAST(bs),ptrlen)
New centralised binary-data marshalling system. I've finally got tired of all the code throughout PuTTY that repeats the same logic about how to format the SSH binary primitives like uint32, string, mpint. We've got reasonably organised code in ssh.c that appends things like that to 'struct Packet'; something similar in sftp.c which repeats a lot of the work; utility functions in various places to format an mpint to feed to one or another hash function; and no end of totally ad-hoc stuff in functions like public key blob formatters which actually have to _count up_ the size of data painstakingly, then malloc exactly that much and mess about with PUT_32BIT. It's time to bring all of that into one place, and stop repeating myself in error-prone ways everywhere. The new marshal.h defines a system in which I centralise all the actual marshalling functions, and then layer a touch of C macro trickery on top to allow me to (look as if I) pass a wide range of different types to those functions, as long as the target type has been set up in the right way to have a write() function. This commit adds the new header and source file, and sets up some general centralised types (strbuf and the various hash-function contexts like SHA_State), but doesn't use the new calls for anything yet. (I've also renamed some internal functions in import.c which were using the same names that I've just defined macros over. That won't last long - those functions are going to go away soon, so the changed names are strictly temporary.)
2018-05-24 11:17:13 +03:00
#define put_stringz(bs, val) \
BinarySink_put_stringz(BinarySink_UPCAST(bs), val)
#define put_stringsb(bs, val) \
BinarySink_put_stringsb(BinarySink_UPCAST(bs), val)
/* Other string outputs: 'asciz' emits the string data directly into
* the output including the terminating \0, and 'pstring' emits the
* string in Pascal style with a leading _one_-byte length field.
* pstring can fail if the string is too long. */
#define put_asciz(bs, val) \
BinarySink_put_asciz(BinarySink_UPCAST(bs), val)
#define put_pstring(bs, val) \
BinarySink_put_pstring(BinarySink_UPCAST(bs), val)
/* Multiprecision integers, in both the SSH-1 and SSH-2 formats. */
#define put_mp_ssh1(bs, val) \
BinarySink_put_mp_ssh1(BinarySink_UPCAST(bs), val)
#define put_mp_ssh2(bs, val) \
BinarySink_put_mp_ssh2(BinarySink_UPCAST(bs), val)
/* Padding with a specified byte. */
#define put_padding(bs, len, padbyte) \
BinarySink_put_padding(BinarySink_UPCAST(bs), len, padbyte)
New centralised binary-data marshalling system. I've finally got tired of all the code throughout PuTTY that repeats the same logic about how to format the SSH binary primitives like uint32, string, mpint. We've got reasonably organised code in ssh.c that appends things like that to 'struct Packet'; something similar in sftp.c which repeats a lot of the work; utility functions in various places to format an mpint to feed to one or another hash function; and no end of totally ad-hoc stuff in functions like public key blob formatters which actually have to _count up_ the size of data painstakingly, then malloc exactly that much and mess about with PUT_32BIT. It's time to bring all of that into one place, and stop repeating myself in error-prone ways everywhere. The new marshal.h defines a system in which I centralise all the actual marshalling functions, and then layer a touch of C macro trickery on top to allow me to (look as if I) pass a wide range of different types to those functions, as long as the target type has been set up in the right way to have a write() function. This commit adds the new header and source file, and sets up some general centralised types (strbuf and the various hash-function contexts like SHA_State), but doesn't use the new calls for anything yet. (I've also renamed some internal functions in import.c which were using the same names that I've just defined macros over. That won't last long - those functions are going to go away soon, so the changed names are strictly temporary.)
2018-05-24 11:17:13 +03:00
/* Fallback: just emit raw data bytes, using a syntax that matches the
* rest of these macros. */
#define put_data(bs, val, len) \
BinarySink_put_data(BinarySink_UPCAST(bs), val, len)
#define put_datapl(bs, pl) \
BinarySink_put_datapl(BinarySink_UPCAST(bs), pl)
New centralised binary-data marshalling system. I've finally got tired of all the code throughout PuTTY that repeats the same logic about how to format the SSH binary primitives like uint32, string, mpint. We've got reasonably organised code in ssh.c that appends things like that to 'struct Packet'; something similar in sftp.c which repeats a lot of the work; utility functions in various places to format an mpint to feed to one or another hash function; and no end of totally ad-hoc stuff in functions like public key blob formatters which actually have to _count up_ the size of data painstakingly, then malloc exactly that much and mess about with PUT_32BIT. It's time to bring all of that into one place, and stop repeating myself in error-prone ways everywhere. The new marshal.h defines a system in which I centralise all the actual marshalling functions, and then layer a touch of C macro trickery on top to allow me to (look as if I) pass a wide range of different types to those functions, as long as the target type has been set up in the right way to have a write() function. This commit adds the new header and source file, and sets up some general centralised types (strbuf and the various hash-function contexts like SHA_State), but doesn't use the new calls for anything yet. (I've also renamed some internal functions in import.c which were using the same names that I've just defined macros over. That won't last long - those functions are going to go away soon, so the changed names are strictly temporary.)
2018-05-24 11:17:13 +03:00
/*
* The underlying real C functions that implement most of those
* macros. Generally you won't want to call these directly, because
* they have such cumbersome names; you call the wrapper macros above
* instead.
*
* A few functions whose wrapper macros are defined above are actually
* declared in other headers, so as to guarantee that the
* declaration(s) of their other parameter type(s) are in scope.
*/
void BinarySink_put_data(BinarySink *, const void *data, size_t len);
void BinarySink_put_datapl(BinarySink *, ptrlen);
void BinarySink_put_padding(BinarySink *, size_t len, unsigned char padbyte);
New centralised binary-data marshalling system. I've finally got tired of all the code throughout PuTTY that repeats the same logic about how to format the SSH binary primitives like uint32, string, mpint. We've got reasonably organised code in ssh.c that appends things like that to 'struct Packet'; something similar in sftp.c which repeats a lot of the work; utility functions in various places to format an mpint to feed to one or another hash function; and no end of totally ad-hoc stuff in functions like public key blob formatters which actually have to _count up_ the size of data painstakingly, then malloc exactly that much and mess about with PUT_32BIT. It's time to bring all of that into one place, and stop repeating myself in error-prone ways everywhere. The new marshal.h defines a system in which I centralise all the actual marshalling functions, and then layer a touch of C macro trickery on top to allow me to (look as if I) pass a wide range of different types to those functions, as long as the target type has been set up in the right way to have a write() function. This commit adds the new header and source file, and sets up some general centralised types (strbuf and the various hash-function contexts like SHA_State), but doesn't use the new calls for anything yet. (I've also renamed some internal functions in import.c which were using the same names that I've just defined macros over. That won't last long - those functions are going to go away soon, so the changed names are strictly temporary.)
2018-05-24 11:17:13 +03:00
void BinarySink_put_byte(BinarySink *, unsigned char);
Convert a lot of 'int' variables to 'bool'. 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'!
2018-11-02 22:23:19 +03:00
void BinarySink_put_bool(BinarySink *, bool);
New centralised binary-data marshalling system. I've finally got tired of all the code throughout PuTTY that repeats the same logic about how to format the SSH binary primitives like uint32, string, mpint. We've got reasonably organised code in ssh.c that appends things like that to 'struct Packet'; something similar in sftp.c which repeats a lot of the work; utility functions in various places to format an mpint to feed to one or another hash function; and no end of totally ad-hoc stuff in functions like public key blob formatters which actually have to _count up_ the size of data painstakingly, then malloc exactly that much and mess about with PUT_32BIT. It's time to bring all of that into one place, and stop repeating myself in error-prone ways everywhere. The new marshal.h defines a system in which I centralise all the actual marshalling functions, and then layer a touch of C macro trickery on top to allow me to (look as if I) pass a wide range of different types to those functions, as long as the target type has been set up in the right way to have a write() function. This commit adds the new header and source file, and sets up some general centralised types (strbuf and the various hash-function contexts like SHA_State), but doesn't use the new calls for anything yet. (I've also renamed some internal functions in import.c which were using the same names that I've just defined macros over. That won't last long - those functions are going to go away soon, so the changed names are strictly temporary.)
2018-05-24 11:17:13 +03:00
void BinarySink_put_uint16(BinarySink *, unsigned long);
void BinarySink_put_uint32(BinarySink *, unsigned long);
void BinarySink_put_uint64(BinarySink *, uint64_t);
New centralised binary-data marshalling system. I've finally got tired of all the code throughout PuTTY that repeats the same logic about how to format the SSH binary primitives like uint32, string, mpint. We've got reasonably organised code in ssh.c that appends things like that to 'struct Packet'; something similar in sftp.c which repeats a lot of the work; utility functions in various places to format an mpint to feed to one or another hash function; and no end of totally ad-hoc stuff in functions like public key blob formatters which actually have to _count up_ the size of data painstakingly, then malloc exactly that much and mess about with PUT_32BIT. It's time to bring all of that into one place, and stop repeating myself in error-prone ways everywhere. The new marshal.h defines a system in which I centralise all the actual marshalling functions, and then layer a touch of C macro trickery on top to allow me to (look as if I) pass a wide range of different types to those functions, as long as the target type has been set up in the right way to have a write() function. This commit adds the new header and source file, and sets up some general centralised types (strbuf and the various hash-function contexts like SHA_State), but doesn't use the new calls for anything yet. (I've also renamed some internal functions in import.c which were using the same names that I've just defined macros over. That won't last long - those functions are going to go away soon, so the changed names are strictly temporary.)
2018-05-24 11:17:13 +03:00
void BinarySink_put_string(BinarySink *, const void *data, size_t len);
void BinarySink_put_stringpl(BinarySink *, ptrlen);
New centralised binary-data marshalling system. I've finally got tired of all the code throughout PuTTY that repeats the same logic about how to format the SSH binary primitives like uint32, string, mpint. We've got reasonably organised code in ssh.c that appends things like that to 'struct Packet'; something similar in sftp.c which repeats a lot of the work; utility functions in various places to format an mpint to feed to one or another hash function; and no end of totally ad-hoc stuff in functions like public key blob formatters which actually have to _count up_ the size of data painstakingly, then malloc exactly that much and mess about with PUT_32BIT. It's time to bring all of that into one place, and stop repeating myself in error-prone ways everywhere. The new marshal.h defines a system in which I centralise all the actual marshalling functions, and then layer a touch of C macro trickery on top to allow me to (look as if I) pass a wide range of different types to those functions, as long as the target type has been set up in the right way to have a write() function. This commit adds the new header and source file, and sets up some general centralised types (strbuf and the various hash-function contexts like SHA_State), but doesn't use the new calls for anything yet. (I've also renamed some internal functions in import.c which were using the same names that I've just defined macros over. That won't last long - those functions are going to go away soon, so the changed names are strictly temporary.)
2018-05-24 11:17:13 +03:00
void BinarySink_put_stringz(BinarySink *, const char *str);
struct strbuf;
void BinarySink_put_stringsb(BinarySink *, struct strbuf *);
void BinarySink_put_asciz(BinarySink *, const char *str);
Convert a lot of 'int' variables to 'bool'. 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'!
2018-11-02 22:23:19 +03:00
bool BinarySink_put_pstring(BinarySink *, const char *str);
Complete rewrite of PuTTY's bignum library. 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.
2018-12-31 16:53:41 +03:00
void BinarySink_put_mp_ssh1(BinarySink *bs, mp_int *x);
void BinarySink_put_mp_ssh2(BinarySink *bs, mp_int *x);
New centralised binary-data marshalling system. I've finally got tired of all the code throughout PuTTY that repeats the same logic about how to format the SSH binary primitives like uint32, string, mpint. We've got reasonably organised code in ssh.c that appends things like that to 'struct Packet'; something similar in sftp.c which repeats a lot of the work; utility functions in various places to format an mpint to feed to one or another hash function; and no end of totally ad-hoc stuff in functions like public key blob formatters which actually have to _count up_ the size of data painstakingly, then malloc exactly that much and mess about with PUT_32BIT. It's time to bring all of that into one place, and stop repeating myself in error-prone ways everywhere. The new marshal.h defines a system in which I centralise all the actual marshalling functions, and then layer a touch of C macro trickery on top to allow me to (look as if I) pass a wide range of different types to those functions, as long as the target type has been set up in the right way to have a write() function. This commit adds the new header and source file, and sets up some general centralised types (strbuf and the various hash-function contexts like SHA_State), but doesn't use the new calls for anything yet. (I've also renamed some internal functions in import.c which were using the same names that I've just defined macros over. That won't last long - those functions are going to go away soon, so the changed names are strictly temporary.)
2018-05-24 11:17:13 +03:00
/* ---------------------------------------------------------------------- */
/*
* A complementary trait structure for _un_-marshalling.
*
* This structure contains client-visible data fields rather than
* methods, because that seemed more useful than leaving it totally
* opaque. But it's still got the self-pointer system that will allow
* the set of get_* macros to target one of these itself or any other
* type that 'derives' from it. So, for example, an SSH packet
* structure can act as a BinarySource while also having additional
* fields like the packet type.
*/
typedef enum BinarySourceError {
BSE_NO_ERROR,
BSE_OUT_OF_DATA,
BSE_INVALID
} BinarySourceError;
struct BinarySource {
/*
* (data, len) is the data block being decoded. pos is the current
* position within the block.
*/
const void *data;
size_t pos, len;
/*
* 'err' indicates whether a decoding error has happened at any
* point. Once this has been set to something other than
* BSE_NO_ERROR, it shouldn't be changed by any unmarshalling
* function. So you can safely do a long sequence of get_foo()
* operations and then test err just once at the end, rather than
* having to conditionalise every single get.
*
* The unmarshalling functions should always return some value,
* even if a decoding error occurs. Generally on error they'll
* return zero (if numeric) or the empty string (if string-based),
* or some other appropriate default value for more complicated
* types.
*
* If the usual return value is dynamically allocated (e.g. a
Complete rewrite of PuTTY's bignum library. 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.
2018-12-31 16:53:41 +03:00
* bignum, or a normal C 'char *' string), then the error value is
* also dynamic in the same way. So you have to free exactly the
* same set of things whether or not there was a decoding error,
* which simplifies exit paths - for example, you could call a big
* pile of get_foo functions, then put the actual handling of the
* results under 'if (!get_err(src))', and then free everything
* outside that if.
*/
BinarySourceError err;
/*
* Self-pointer for the implicit derivation trick, same as
* BinarySink above.
*/
BinarySource *binarysource_;
};
/*
* Implementation macros, similar to BinarySink.
*/
#define BinarySource_IMPLEMENTATION BinarySource binarysource_[1]
static inline void BinarySource_INIT__(BinarySource *src, ptrlen data)
{
src->data = data.ptr;
src->len = data.len;
src->pos = 0;
src->err = BSE_NO_ERROR;
src->binarysource_ = src;
}
#define BinarySource_BARE_INIT_PL(obj, pl) \
TYPECHECK(&(obj)->binarysource_ == (BinarySource **)0, \
BinarySource_INIT__(obj, pl))
#define BinarySource_BARE_INIT(obj, data_, len_) \
BinarySource_BARE_INIT_PL(obj, make_ptrlen(data_, len_))
#define BinarySource_INIT_PL(obj, pl) \
TYPECHECK(&(obj)->binarysource_ == (BinarySource (*)[1])0, \
BinarySource_INIT__(BinarySource_UPCAST(obj), pl))
#define BinarySource_INIT(obj, data_, len_) \
BinarySource_INIT_PL(obj, make_ptrlen(data_, len_))
#define BinarySource_DOWNCAST(object, type) \
TYPECHECK((object) == ((type *)0)->binarysource_, \
((type *)(((char *)(object)) - offsetof(type, binarysource_))))
#define BinarySource_UPCAST(object) \
TYPECHECK((object)->binarysource_ == (BinarySource *)0, \
(object)->binarysource_)
#define BinarySource_COPIED(obj) \
((obj)->binarysource_->binarysource_ = (obj)->binarysource_)
#define get_data(src, len) \
BinarySource_get_data(BinarySource_UPCAST(src), len)
#define get_byte(src) \
BinarySource_get_byte(BinarySource_UPCAST(src))
#define get_bool(src) \
BinarySource_get_bool(BinarySource_UPCAST(src))
#define get_uint16(src) \
BinarySource_get_uint16(BinarySource_UPCAST(src))
#define get_uint32(src) \
BinarySource_get_uint32(BinarySource_UPCAST(src))
#define get_uint64(src) \
BinarySource_get_uint64(BinarySource_UPCAST(src))
#define get_string(src) \
BinarySource_get_string(BinarySource_UPCAST(src))
#define get_asciz(src) \
BinarySource_get_asciz(BinarySource_UPCAST(src))
#define get_pstring(src) \
BinarySource_get_pstring(BinarySource_UPCAST(src))
#define get_mp_ssh1(src) \
BinarySource_get_mp_ssh1(BinarySource_UPCAST(src))
#define get_mp_ssh2(src) \
BinarySource_get_mp_ssh2(BinarySource_UPCAST(src))
#define get_rsa_ssh1_pub(src, rsa, order) \
BinarySource_get_rsa_ssh1_pub(BinarySource_UPCAST(src), rsa, order)
#define get_rsa_ssh1_priv(src, rsa) \
BinarySource_get_rsa_ssh1_priv(BinarySource_UPCAST(src), rsa)
#define get_err(src) (BinarySource_UPCAST(src)->err)
#define get_avail(src) (BinarySource_UPCAST(src)->len - \
BinarySource_UPCAST(src)->pos)
#define get_ptr(src) \
((const void *)( \
(const unsigned char *)(BinarySource_UPCAST(src)->data) + \
BinarySource_UPCAST(src)->pos))
ptrlen BinarySource_get_data(BinarySource *, size_t);
unsigned char BinarySource_get_byte(BinarySource *);
Convert a lot of 'int' variables to 'bool'. 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'!
2018-11-02 22:23:19 +03:00
bool BinarySource_get_bool(BinarySource *);
unsigned BinarySource_get_uint16(BinarySource *);
unsigned long BinarySource_get_uint32(BinarySource *);
uint64_t BinarySource_get_uint64(BinarySource *);
ptrlen BinarySource_get_string(BinarySource *);
const char *BinarySource_get_asciz(BinarySource *);
ptrlen BinarySource_get_pstring(BinarySource *);
Complete rewrite of PuTTY's bignum library. 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.
2018-12-31 16:53:41 +03:00
mp_int *BinarySource_get_mp_ssh1(BinarySource *src);
mp_int *BinarySource_get_mp_ssh2(BinarySource *src);
/*
* A couple of useful standard BinarySink implementations, which live
* as sensibly here as anywhere else: one that makes a BinarySink
* whose effect is to write to a stdio stream, and one whose effect is
* to append to a bufchain.
*/
struct stdio_sink {
FILE *fp;
BinarySink_IMPLEMENTATION;
};
struct bufchain_sink {
bufchain *ch;
BinarySink_IMPLEMENTATION;
};
void stdio_sink_init(stdio_sink *sink, FILE *fp);
void bufchain_sink_init(bufchain_sink *sink, bufchain *ch);
New centralised binary-data marshalling system. I've finally got tired of all the code throughout PuTTY that repeats the same logic about how to format the SSH binary primitives like uint32, string, mpint. We've got reasonably organised code in ssh.c that appends things like that to 'struct Packet'; something similar in sftp.c which repeats a lot of the work; utility functions in various places to format an mpint to feed to one or another hash function; and no end of totally ad-hoc stuff in functions like public key blob formatters which actually have to _count up_ the size of data painstakingly, then malloc exactly that much and mess about with PUT_32BIT. It's time to bring all of that into one place, and stop repeating myself in error-prone ways everywhere. The new marshal.h defines a system in which I centralise all the actual marshalling functions, and then layer a touch of C macro trickery on top to allow me to (look as if I) pass a wide range of different types to those functions, as long as the target type has been set up in the right way to have a write() function. This commit adds the new header and source file, and sets up some general centralised types (strbuf and the various hash-function contexts like SHA_State), but doesn't use the new calls for anything yet. (I've also renamed some internal functions in import.c which were using the same names that I've just defined macros over. That won't last long - those functions are going to go away soon, so the changed names are strictly temporary.)
2018-05-24 11:17:13 +03:00
#endif /* PUTTY_MARSHAL_H */