boringssl/crypto/x509/x509_lu.c

727 lines
20 KiB
C
Raw Normal View History

/* crypto/x509/x509_lu.c */
/* Copyright (C) 1995-1998 Eric Young (eay@cryptsoft.com)
* All rights reserved.
*
* This package is an SSL implementation written
* by Eric Young (eay@cryptsoft.com).
* The implementation was written so as to conform with Netscapes SSL.
*
* This library is free for commercial and non-commercial use as long as
* the following conditions are aheared to. The following conditions
* apply to all code found in this distribution, be it the RC4, RSA,
* lhash, DES, etc., code; not just the SSL code. The SSL documentation
* included with this distribution is covered by the same copyright terms
* except that the holder is Tim Hudson (tjh@cryptsoft.com).
*
* Copyright remains Eric Young's, and as such any Copyright notices in
* the code are not to be removed.
* If this package is used in a product, Eric Young should be given attribution
* as the author of the parts of the library used.
* This can be in the form of a textual message at program startup or
* in documentation (online or textual) provided with the package.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* "This product includes cryptographic software written by
* Eric Young (eay@cryptsoft.com)"
* The word 'cryptographic' can be left out if the rouines from the library
* being used are not cryptographic related :-).
* 4. If you include any Windows specific code (or a derivative thereof) from
* the apps directory (application code) you must include an acknowledgement:
* "This product includes software written by Tim Hudson (tjh@cryptsoft.com)"
*
* THIS SOFTWARE IS PROVIDED BY ERIC YOUNG ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* The licence and distribution terms for any publically available version or
* derivative of this code cannot be changed. i.e. this code cannot simply be
* copied and put under another distribution licence
* [including the GNU Public Licence.] */
#include <string.h>
#include <openssl/err.h>
#include <openssl/lhash.h>
#include <openssl/mem.h>
#include <openssl/thread.h>
#include <openssl/x509.h>
#include <openssl/x509v3.h>
#include "../internal.h"
X509_LOOKUP *X509_LOOKUP_new(X509_LOOKUP_METHOD *method)
{
X509_LOOKUP *ret;
ret = (X509_LOOKUP *)OPENSSL_malloc(sizeof(X509_LOOKUP));
if (ret == NULL)
return NULL;
ret->init = 0;
ret->skip = 0;
ret->method = method;
ret->method_data = NULL;
ret->store_ctx = NULL;
if ((method->new_item != NULL) && !method->new_item(ret)) {
OPENSSL_free(ret);
return NULL;
}
return ret;
}
void X509_LOOKUP_free(X509_LOOKUP *ctx)
{
if (ctx == NULL)
return;
if ((ctx->method != NULL) && (ctx->method->free != NULL))
(*ctx->method->free) (ctx);
OPENSSL_free(ctx);
}
int X509_LOOKUP_init(X509_LOOKUP *ctx)
{
if (ctx->method == NULL)
return 0;
if (ctx->method->init != NULL)
return ctx->method->init(ctx);
else
return 1;
}
int X509_LOOKUP_shutdown(X509_LOOKUP *ctx)
{
if (ctx->method == NULL)
return 0;
if (ctx->method->shutdown != NULL)
return ctx->method->shutdown(ctx);
else
return 1;
}
int X509_LOOKUP_ctrl(X509_LOOKUP *ctx, int cmd, const char *argc, long argl,
char **ret)
{
if (ctx->method == NULL)
return -1;
if (ctx->method->ctrl != NULL)
return ctx->method->ctrl(ctx, cmd, argc, argl, ret);
else
return 1;
}
int X509_LOOKUP_by_subject(X509_LOOKUP *ctx, int type, X509_NAME *name,
X509_OBJECT *ret)
{
if ((ctx->method == NULL) || (ctx->method->get_by_subject == NULL))
Unwind X509_LU_RETRY and fix a lot of type confusion. (This change will be sent upstream. Since the legacy X.509 stack is just kept around for compatibility, if they decide to fix it in a different way, we may wish to revert this and apply their fix.) Dating back to SSLeay, X509_LOOKUP_METHOD had this X509_LU_RETRY machinery. But it's not documented and it appears to have never worked. Problems with the existing logic: - X509_LU_* is not sure whether it is a type enum (to be passed into X509_LOOKUP_by_*) or a return enum (to be retained by those same functions). - X509_LOOKUP_by_* is not sure whether it returns 0/1 or an X509_LU_* value. Looking at the functions themselves, one might think it's the latter, but for X509_LOOKUP_by_subject returning both 0 and X509_LU_FAIL. But looking at the call sites, some expect 0/1 (such as X509_STORE_get1_certs) while others expect an X509_LU_* enum (such as X509_STORE_CTX_get1_issuer). It is very fortunate that FAIL happens to be 0 and X509 happens to be 1. These functions primarily call to X509_LOOKUP_METHOD hooks. Looking through OpenSSL itself and code checked into Google, I found no evidence that any hooks have been implemented except for get_by_subject in by_dir.c. We take that one as definitive and observe it believes it returns 0/1. Notably, it returns 1 on success even if asked for a type other than X509_LU_X509. (X509_LU_X509 = 1. Others are different.) I found another piece of third-party software which corroborates this worldview. - X509_STORE_get_by_subject's handling of X509_LU_RETRY (it's the j < 0 check) is broken. It saves j into vs->current_method where it probably meant to save i. (This bug has existed since SSLeay.) It also returns j (supposedly X509_LU_RETRY) while all callers of X509_STORE_get_by_subject expect it to return 0/1 by checking with ! instead of <= 0. (Note that all other codepaths return 0 and 1 so this function did not actually believe it returned X509_LU_* most of the time.) This, in turn, gives us a free of uninitialized pointers in X509_STORE_get1_certs and other functions which expect that *ret is filled in if X509_STORE_get_by_subject returns success. GCC 4.9 with optimizations from the Android NDK noticed this, which trigged this saga. (It's only reachable if any X509_LOOKUP_METHOD returned X509_LU_RETRY.) - Although the code which expects X509_STORE_get_by_subject return 0/1 does not date to SSLeay, the X509_STORE_get_by_subject call in X509_STORE_CTX_get1_issuer *does* (though, at the time, it was inline in X509_verify_cert. That code believes X509_STORE_get_by_subject returns an X509_LU_* enum, but it doesn't work either! It believes *ret is filled in on X509_LU_RETRY, thus freeing another uninitialized pointer (GCC noticed this too). Since this "retry" code has clearly never worked, from SSLeay onwards, unwind it completely rather than attempt to fix it. No X509_LOOKUP_METHOD can possibly have depended on it. Matching all non-broken codepaths X509_LOOKUP_by_* now returns 0/1 and X509_STORE_get_by_subject returns 0/1. X509_LU_* is purely a type enum with X509_LU_{REJECT,FAIL} being legacy constants to keep old code compiling. (Upstream is recommended to remove those values altogether for 1.1.0.) On the off chance any get_by_* X509_LOOKUP_METHOD implementations did not return 0/1 (I have found no evidence anywhere of this, and I believe it wouldn't have worked anyway), the X509_LOOKUP_by_* wrapper functions will coerce the return values back to 0/1 before passing up to the callers which want 0/1. This both avoids the error-prone -1/0/1 calling convention and, more importantly, avoids problems with third-party callers which expect a X509_LU_* return code. 0/1 collide with FAIL/X509 while -1 will collide with RETRY and might confuse things. Change-Id: I98ecf6fa7342866b9124dc6f0b422cb9ce4a1ae7 Reviewed-on: https://boringssl-review.googlesource.com/8303 Reviewed-by: Adam Langley <agl@google.com>
2016-06-15 23:41:51 +01:00
return 0;
if (ctx->skip)
return 0;
Unwind X509_LU_RETRY and fix a lot of type confusion. (This change will be sent upstream. Since the legacy X.509 stack is just kept around for compatibility, if they decide to fix it in a different way, we may wish to revert this and apply their fix.) Dating back to SSLeay, X509_LOOKUP_METHOD had this X509_LU_RETRY machinery. But it's not documented and it appears to have never worked. Problems with the existing logic: - X509_LU_* is not sure whether it is a type enum (to be passed into X509_LOOKUP_by_*) or a return enum (to be retained by those same functions). - X509_LOOKUP_by_* is not sure whether it returns 0/1 or an X509_LU_* value. Looking at the functions themselves, one might think it's the latter, but for X509_LOOKUP_by_subject returning both 0 and X509_LU_FAIL. But looking at the call sites, some expect 0/1 (such as X509_STORE_get1_certs) while others expect an X509_LU_* enum (such as X509_STORE_CTX_get1_issuer). It is very fortunate that FAIL happens to be 0 and X509 happens to be 1. These functions primarily call to X509_LOOKUP_METHOD hooks. Looking through OpenSSL itself and code checked into Google, I found no evidence that any hooks have been implemented except for get_by_subject in by_dir.c. We take that one as definitive and observe it believes it returns 0/1. Notably, it returns 1 on success even if asked for a type other than X509_LU_X509. (X509_LU_X509 = 1. Others are different.) I found another piece of third-party software which corroborates this worldview. - X509_STORE_get_by_subject's handling of X509_LU_RETRY (it's the j < 0 check) is broken. It saves j into vs->current_method where it probably meant to save i. (This bug has existed since SSLeay.) It also returns j (supposedly X509_LU_RETRY) while all callers of X509_STORE_get_by_subject expect it to return 0/1 by checking with ! instead of <= 0. (Note that all other codepaths return 0 and 1 so this function did not actually believe it returned X509_LU_* most of the time.) This, in turn, gives us a free of uninitialized pointers in X509_STORE_get1_certs and other functions which expect that *ret is filled in if X509_STORE_get_by_subject returns success. GCC 4.9 with optimizations from the Android NDK noticed this, which trigged this saga. (It's only reachable if any X509_LOOKUP_METHOD returned X509_LU_RETRY.) - Although the code which expects X509_STORE_get_by_subject return 0/1 does not date to SSLeay, the X509_STORE_get_by_subject call in X509_STORE_CTX_get1_issuer *does* (though, at the time, it was inline in X509_verify_cert. That code believes X509_STORE_get_by_subject returns an X509_LU_* enum, but it doesn't work either! It believes *ret is filled in on X509_LU_RETRY, thus freeing another uninitialized pointer (GCC noticed this too). Since this "retry" code has clearly never worked, from SSLeay onwards, unwind it completely rather than attempt to fix it. No X509_LOOKUP_METHOD can possibly have depended on it. Matching all non-broken codepaths X509_LOOKUP_by_* now returns 0/1 and X509_STORE_get_by_subject returns 0/1. X509_LU_* is purely a type enum with X509_LU_{REJECT,FAIL} being legacy constants to keep old code compiling. (Upstream is recommended to remove those values altogether for 1.1.0.) On the off chance any get_by_* X509_LOOKUP_METHOD implementations did not return 0/1 (I have found no evidence anywhere of this, and I believe it wouldn't have worked anyway), the X509_LOOKUP_by_* wrapper functions will coerce the return values back to 0/1 before passing up to the callers which want 0/1. This both avoids the error-prone -1/0/1 calling convention and, more importantly, avoids problems with third-party callers which expect a X509_LU_* return code. 0/1 collide with FAIL/X509 while -1 will collide with RETRY and might confuse things. Change-Id: I98ecf6fa7342866b9124dc6f0b422cb9ce4a1ae7 Reviewed-on: https://boringssl-review.googlesource.com/8303 Reviewed-by: Adam Langley <agl@google.com>
2016-06-15 23:41:51 +01:00
return ctx->method->get_by_subject(ctx, type, name, ret) > 0;
}
int X509_LOOKUP_by_issuer_serial(X509_LOOKUP *ctx, int type, X509_NAME *name,
ASN1_INTEGER *serial, X509_OBJECT *ret)
{
if ((ctx->method == NULL) || (ctx->method->get_by_issuer_serial == NULL))
Unwind X509_LU_RETRY and fix a lot of type confusion. (This change will be sent upstream. Since the legacy X.509 stack is just kept around for compatibility, if they decide to fix it in a different way, we may wish to revert this and apply their fix.) Dating back to SSLeay, X509_LOOKUP_METHOD had this X509_LU_RETRY machinery. But it's not documented and it appears to have never worked. Problems with the existing logic: - X509_LU_* is not sure whether it is a type enum (to be passed into X509_LOOKUP_by_*) or a return enum (to be retained by those same functions). - X509_LOOKUP_by_* is not sure whether it returns 0/1 or an X509_LU_* value. Looking at the functions themselves, one might think it's the latter, but for X509_LOOKUP_by_subject returning both 0 and X509_LU_FAIL. But looking at the call sites, some expect 0/1 (such as X509_STORE_get1_certs) while others expect an X509_LU_* enum (such as X509_STORE_CTX_get1_issuer). It is very fortunate that FAIL happens to be 0 and X509 happens to be 1. These functions primarily call to X509_LOOKUP_METHOD hooks. Looking through OpenSSL itself and code checked into Google, I found no evidence that any hooks have been implemented except for get_by_subject in by_dir.c. We take that one as definitive and observe it believes it returns 0/1. Notably, it returns 1 on success even if asked for a type other than X509_LU_X509. (X509_LU_X509 = 1. Others are different.) I found another piece of third-party software which corroborates this worldview. - X509_STORE_get_by_subject's handling of X509_LU_RETRY (it's the j < 0 check) is broken. It saves j into vs->current_method where it probably meant to save i. (This bug has existed since SSLeay.) It also returns j (supposedly X509_LU_RETRY) while all callers of X509_STORE_get_by_subject expect it to return 0/1 by checking with ! instead of <= 0. (Note that all other codepaths return 0 and 1 so this function did not actually believe it returned X509_LU_* most of the time.) This, in turn, gives us a free of uninitialized pointers in X509_STORE_get1_certs and other functions which expect that *ret is filled in if X509_STORE_get_by_subject returns success. GCC 4.9 with optimizations from the Android NDK noticed this, which trigged this saga. (It's only reachable if any X509_LOOKUP_METHOD returned X509_LU_RETRY.) - Although the code which expects X509_STORE_get_by_subject return 0/1 does not date to SSLeay, the X509_STORE_get_by_subject call in X509_STORE_CTX_get1_issuer *does* (though, at the time, it was inline in X509_verify_cert. That code believes X509_STORE_get_by_subject returns an X509_LU_* enum, but it doesn't work either! It believes *ret is filled in on X509_LU_RETRY, thus freeing another uninitialized pointer (GCC noticed this too). Since this "retry" code has clearly never worked, from SSLeay onwards, unwind it completely rather than attempt to fix it. No X509_LOOKUP_METHOD can possibly have depended on it. Matching all non-broken codepaths X509_LOOKUP_by_* now returns 0/1 and X509_STORE_get_by_subject returns 0/1. X509_LU_* is purely a type enum with X509_LU_{REJECT,FAIL} being legacy constants to keep old code compiling. (Upstream is recommended to remove those values altogether for 1.1.0.) On the off chance any get_by_* X509_LOOKUP_METHOD implementations did not return 0/1 (I have found no evidence anywhere of this, and I believe it wouldn't have worked anyway), the X509_LOOKUP_by_* wrapper functions will coerce the return values back to 0/1 before passing up to the callers which want 0/1. This both avoids the error-prone -1/0/1 calling convention and, more importantly, avoids problems with third-party callers which expect a X509_LU_* return code. 0/1 collide with FAIL/X509 while -1 will collide with RETRY and might confuse things. Change-Id: I98ecf6fa7342866b9124dc6f0b422cb9ce4a1ae7 Reviewed-on: https://boringssl-review.googlesource.com/8303 Reviewed-by: Adam Langley <agl@google.com>
2016-06-15 23:41:51 +01:00
return 0;
return ctx->method->get_by_issuer_serial(ctx, type, name, serial, ret) > 0;
}
int X509_LOOKUP_by_fingerprint(X509_LOOKUP *ctx, int type,
unsigned char *bytes, int len,
X509_OBJECT *ret)
{
if ((ctx->method == NULL) || (ctx->method->get_by_fingerprint == NULL))
Unwind X509_LU_RETRY and fix a lot of type confusion. (This change will be sent upstream. Since the legacy X.509 stack is just kept around for compatibility, if they decide to fix it in a different way, we may wish to revert this and apply their fix.) Dating back to SSLeay, X509_LOOKUP_METHOD had this X509_LU_RETRY machinery. But it's not documented and it appears to have never worked. Problems with the existing logic: - X509_LU_* is not sure whether it is a type enum (to be passed into X509_LOOKUP_by_*) or a return enum (to be retained by those same functions). - X509_LOOKUP_by_* is not sure whether it returns 0/1 or an X509_LU_* value. Looking at the functions themselves, one might think it's the latter, but for X509_LOOKUP_by_subject returning both 0 and X509_LU_FAIL. But looking at the call sites, some expect 0/1 (such as X509_STORE_get1_certs) while others expect an X509_LU_* enum (such as X509_STORE_CTX_get1_issuer). It is very fortunate that FAIL happens to be 0 and X509 happens to be 1. These functions primarily call to X509_LOOKUP_METHOD hooks. Looking through OpenSSL itself and code checked into Google, I found no evidence that any hooks have been implemented except for get_by_subject in by_dir.c. We take that one as definitive and observe it believes it returns 0/1. Notably, it returns 1 on success even if asked for a type other than X509_LU_X509. (X509_LU_X509 = 1. Others are different.) I found another piece of third-party software which corroborates this worldview. - X509_STORE_get_by_subject's handling of X509_LU_RETRY (it's the j < 0 check) is broken. It saves j into vs->current_method where it probably meant to save i. (This bug has existed since SSLeay.) It also returns j (supposedly X509_LU_RETRY) while all callers of X509_STORE_get_by_subject expect it to return 0/1 by checking with ! instead of <= 0. (Note that all other codepaths return 0 and 1 so this function did not actually believe it returned X509_LU_* most of the time.) This, in turn, gives us a free of uninitialized pointers in X509_STORE_get1_certs and other functions which expect that *ret is filled in if X509_STORE_get_by_subject returns success. GCC 4.9 with optimizations from the Android NDK noticed this, which trigged this saga. (It's only reachable if any X509_LOOKUP_METHOD returned X509_LU_RETRY.) - Although the code which expects X509_STORE_get_by_subject return 0/1 does not date to SSLeay, the X509_STORE_get_by_subject call in X509_STORE_CTX_get1_issuer *does* (though, at the time, it was inline in X509_verify_cert. That code believes X509_STORE_get_by_subject returns an X509_LU_* enum, but it doesn't work either! It believes *ret is filled in on X509_LU_RETRY, thus freeing another uninitialized pointer (GCC noticed this too). Since this "retry" code has clearly never worked, from SSLeay onwards, unwind it completely rather than attempt to fix it. No X509_LOOKUP_METHOD can possibly have depended on it. Matching all non-broken codepaths X509_LOOKUP_by_* now returns 0/1 and X509_STORE_get_by_subject returns 0/1. X509_LU_* is purely a type enum with X509_LU_{REJECT,FAIL} being legacy constants to keep old code compiling. (Upstream is recommended to remove those values altogether for 1.1.0.) On the off chance any get_by_* X509_LOOKUP_METHOD implementations did not return 0/1 (I have found no evidence anywhere of this, and I believe it wouldn't have worked anyway), the X509_LOOKUP_by_* wrapper functions will coerce the return values back to 0/1 before passing up to the callers which want 0/1. This both avoids the error-prone -1/0/1 calling convention and, more importantly, avoids problems with third-party callers which expect a X509_LU_* return code. 0/1 collide with FAIL/X509 while -1 will collide with RETRY and might confuse things. Change-Id: I98ecf6fa7342866b9124dc6f0b422cb9ce4a1ae7 Reviewed-on: https://boringssl-review.googlesource.com/8303 Reviewed-by: Adam Langley <agl@google.com>
2016-06-15 23:41:51 +01:00
return 0;
return ctx->method->get_by_fingerprint(ctx, type, bytes, len, ret) > 0;
}
int X509_LOOKUP_by_alias(X509_LOOKUP *ctx, int type, char *str, int len,
X509_OBJECT *ret)
{
if ((ctx->method == NULL) || (ctx->method->get_by_alias == NULL))
Unwind X509_LU_RETRY and fix a lot of type confusion. (This change will be sent upstream. Since the legacy X.509 stack is just kept around for compatibility, if they decide to fix it in a different way, we may wish to revert this and apply their fix.) Dating back to SSLeay, X509_LOOKUP_METHOD had this X509_LU_RETRY machinery. But it's not documented and it appears to have never worked. Problems with the existing logic: - X509_LU_* is not sure whether it is a type enum (to be passed into X509_LOOKUP_by_*) or a return enum (to be retained by those same functions). - X509_LOOKUP_by_* is not sure whether it returns 0/1 or an X509_LU_* value. Looking at the functions themselves, one might think it's the latter, but for X509_LOOKUP_by_subject returning both 0 and X509_LU_FAIL. But looking at the call sites, some expect 0/1 (such as X509_STORE_get1_certs) while others expect an X509_LU_* enum (such as X509_STORE_CTX_get1_issuer). It is very fortunate that FAIL happens to be 0 and X509 happens to be 1. These functions primarily call to X509_LOOKUP_METHOD hooks. Looking through OpenSSL itself and code checked into Google, I found no evidence that any hooks have been implemented except for get_by_subject in by_dir.c. We take that one as definitive and observe it believes it returns 0/1. Notably, it returns 1 on success even if asked for a type other than X509_LU_X509. (X509_LU_X509 = 1. Others are different.) I found another piece of third-party software which corroborates this worldview. - X509_STORE_get_by_subject's handling of X509_LU_RETRY (it's the j < 0 check) is broken. It saves j into vs->current_method where it probably meant to save i. (This bug has existed since SSLeay.) It also returns j (supposedly X509_LU_RETRY) while all callers of X509_STORE_get_by_subject expect it to return 0/1 by checking with ! instead of <= 0. (Note that all other codepaths return 0 and 1 so this function did not actually believe it returned X509_LU_* most of the time.) This, in turn, gives us a free of uninitialized pointers in X509_STORE_get1_certs and other functions which expect that *ret is filled in if X509_STORE_get_by_subject returns success. GCC 4.9 with optimizations from the Android NDK noticed this, which trigged this saga. (It's only reachable if any X509_LOOKUP_METHOD returned X509_LU_RETRY.) - Although the code which expects X509_STORE_get_by_subject return 0/1 does not date to SSLeay, the X509_STORE_get_by_subject call in X509_STORE_CTX_get1_issuer *does* (though, at the time, it was inline in X509_verify_cert. That code believes X509_STORE_get_by_subject returns an X509_LU_* enum, but it doesn't work either! It believes *ret is filled in on X509_LU_RETRY, thus freeing another uninitialized pointer (GCC noticed this too). Since this "retry" code has clearly never worked, from SSLeay onwards, unwind it completely rather than attempt to fix it. No X509_LOOKUP_METHOD can possibly have depended on it. Matching all non-broken codepaths X509_LOOKUP_by_* now returns 0/1 and X509_STORE_get_by_subject returns 0/1. X509_LU_* is purely a type enum with X509_LU_{REJECT,FAIL} being legacy constants to keep old code compiling. (Upstream is recommended to remove those values altogether for 1.1.0.) On the off chance any get_by_* X509_LOOKUP_METHOD implementations did not return 0/1 (I have found no evidence anywhere of this, and I believe it wouldn't have worked anyway), the X509_LOOKUP_by_* wrapper functions will coerce the return values back to 0/1 before passing up to the callers which want 0/1. This both avoids the error-prone -1/0/1 calling convention and, more importantly, avoids problems with third-party callers which expect a X509_LU_* return code. 0/1 collide with FAIL/X509 while -1 will collide with RETRY and might confuse things. Change-Id: I98ecf6fa7342866b9124dc6f0b422cb9ce4a1ae7 Reviewed-on: https://boringssl-review.googlesource.com/8303 Reviewed-by: Adam Langley <agl@google.com>
2016-06-15 23:41:51 +01:00
return 0;
return ctx->method->get_by_alias(ctx, type, str, len, ret) > 0;
}
static int x509_object_cmp(const X509_OBJECT **a, const X509_OBJECT **b)
{
int ret;
ret = ((*a)->type - (*b)->type);
if (ret)
return ret;
switch ((*a)->type) {
case X509_LU_X509:
ret = X509_subject_name_cmp((*a)->data.x509, (*b)->data.x509);
break;
case X509_LU_CRL:
ret = X509_CRL_cmp((*a)->data.crl, (*b)->data.crl);
break;
default:
/* abort(); */
return 0;
}
return ret;
}
X509_STORE *X509_STORE_new(void)
{
X509_STORE *ret;
if ((ret = (X509_STORE *)OPENSSL_malloc(sizeof(X509_STORE))) == NULL)
return NULL;
OPENSSL_memset(ret, 0, sizeof(*ret));
CRYPTO_MUTEX_init(&ret->objs_lock);
ret->objs = sk_X509_OBJECT_new(x509_object_cmp);
if (ret->objs == NULL)
goto err;
ret->cache = 1;
ret->get_cert_methods = sk_X509_LOOKUP_new_null();
if (ret->get_cert_methods == NULL)
goto err;
ret->param = X509_VERIFY_PARAM_new();
if (ret->param == NULL)
goto err;
ret->references = 1;
return ret;
err:
if (ret) {
CRYPTO_MUTEX_cleanup(&ret->objs_lock);
if (ret->param)
X509_VERIFY_PARAM_free(ret->param);
if (ret->get_cert_methods)
sk_X509_LOOKUP_free(ret->get_cert_methods);
if (ret->objs)
sk_X509_OBJECT_free(ret->objs);
OPENSSL_free(ret);
}
return NULL;
}
int X509_STORE_up_ref(X509_STORE *store)
{
CRYPTO_refcount_inc(&store->references);
return 1;
}
static void cleanup(X509_OBJECT *a)
{
if (a == NULL) {
return;
}
if (a->type == X509_LU_X509) {
X509_free(a->data.x509);
} else if (a->type == X509_LU_CRL) {
X509_CRL_free(a->data.crl);
} else {
/* abort(); */
}
OPENSSL_free(a);
}
void X509_STORE_free(X509_STORE *vfy)
{
size_t j;
STACK_OF(X509_LOOKUP) *sk;
X509_LOOKUP *lu;
if (vfy == NULL)
return;
if (!CRYPTO_refcount_dec_and_test_zero(&vfy->references)) {
return;
}
CRYPTO_MUTEX_cleanup(&vfy->objs_lock);
sk = vfy->get_cert_methods;
for (j = 0; j < sk_X509_LOOKUP_num(sk); j++) {
lu = sk_X509_LOOKUP_value(sk, j);
X509_LOOKUP_shutdown(lu);
X509_LOOKUP_free(lu);
}
sk_X509_LOOKUP_free(sk);
sk_X509_OBJECT_pop_free(vfy->objs, cleanup);
if (vfy->param)
X509_VERIFY_PARAM_free(vfy->param);
OPENSSL_free(vfy);
}
X509_LOOKUP *X509_STORE_add_lookup(X509_STORE *v, X509_LOOKUP_METHOD *m)
{
size_t i;
STACK_OF(X509_LOOKUP) *sk;
X509_LOOKUP *lu;
sk = v->get_cert_methods;
for (i = 0; i < sk_X509_LOOKUP_num(sk); i++) {
lu = sk_X509_LOOKUP_value(sk, i);
if (m == lu->method) {
return lu;
}
}
/* a new one */
lu = X509_LOOKUP_new(m);
if (lu == NULL)
return NULL;
else {
lu->store_ctx = v;
if (sk_X509_LOOKUP_push(v->get_cert_methods, lu))
return lu;
else {
X509_LOOKUP_free(lu);
return NULL;
}
}
}
int X509_STORE_get_by_subject(X509_STORE_CTX *vs, int type, X509_NAME *name,
X509_OBJECT *ret)
{
X509_STORE *ctx = vs->ctx;
X509_LOOKUP *lu;
X509_OBJECT stmp, *tmp;
Unwind X509_LU_RETRY and fix a lot of type confusion. (This change will be sent upstream. Since the legacy X.509 stack is just kept around for compatibility, if they decide to fix it in a different way, we may wish to revert this and apply their fix.) Dating back to SSLeay, X509_LOOKUP_METHOD had this X509_LU_RETRY machinery. But it's not documented and it appears to have never worked. Problems with the existing logic: - X509_LU_* is not sure whether it is a type enum (to be passed into X509_LOOKUP_by_*) or a return enum (to be retained by those same functions). - X509_LOOKUP_by_* is not sure whether it returns 0/1 or an X509_LU_* value. Looking at the functions themselves, one might think it's the latter, but for X509_LOOKUP_by_subject returning both 0 and X509_LU_FAIL. But looking at the call sites, some expect 0/1 (such as X509_STORE_get1_certs) while others expect an X509_LU_* enum (such as X509_STORE_CTX_get1_issuer). It is very fortunate that FAIL happens to be 0 and X509 happens to be 1. These functions primarily call to X509_LOOKUP_METHOD hooks. Looking through OpenSSL itself and code checked into Google, I found no evidence that any hooks have been implemented except for get_by_subject in by_dir.c. We take that one as definitive and observe it believes it returns 0/1. Notably, it returns 1 on success even if asked for a type other than X509_LU_X509. (X509_LU_X509 = 1. Others are different.) I found another piece of third-party software which corroborates this worldview. - X509_STORE_get_by_subject's handling of X509_LU_RETRY (it's the j < 0 check) is broken. It saves j into vs->current_method where it probably meant to save i. (This bug has existed since SSLeay.) It also returns j (supposedly X509_LU_RETRY) while all callers of X509_STORE_get_by_subject expect it to return 0/1 by checking with ! instead of <= 0. (Note that all other codepaths return 0 and 1 so this function did not actually believe it returned X509_LU_* most of the time.) This, in turn, gives us a free of uninitialized pointers in X509_STORE_get1_certs and other functions which expect that *ret is filled in if X509_STORE_get_by_subject returns success. GCC 4.9 with optimizations from the Android NDK noticed this, which trigged this saga. (It's only reachable if any X509_LOOKUP_METHOD returned X509_LU_RETRY.) - Although the code which expects X509_STORE_get_by_subject return 0/1 does not date to SSLeay, the X509_STORE_get_by_subject call in X509_STORE_CTX_get1_issuer *does* (though, at the time, it was inline in X509_verify_cert. That code believes X509_STORE_get_by_subject returns an X509_LU_* enum, but it doesn't work either! It believes *ret is filled in on X509_LU_RETRY, thus freeing another uninitialized pointer (GCC noticed this too). Since this "retry" code has clearly never worked, from SSLeay onwards, unwind it completely rather than attempt to fix it. No X509_LOOKUP_METHOD can possibly have depended on it. Matching all non-broken codepaths X509_LOOKUP_by_* now returns 0/1 and X509_STORE_get_by_subject returns 0/1. X509_LU_* is purely a type enum with X509_LU_{REJECT,FAIL} being legacy constants to keep old code compiling. (Upstream is recommended to remove those values altogether for 1.1.0.) On the off chance any get_by_* X509_LOOKUP_METHOD implementations did not return 0/1 (I have found no evidence anywhere of this, and I believe it wouldn't have worked anyway), the X509_LOOKUP_by_* wrapper functions will coerce the return values back to 0/1 before passing up to the callers which want 0/1. This both avoids the error-prone -1/0/1 calling convention and, more importantly, avoids problems with third-party callers which expect a X509_LU_* return code. 0/1 collide with FAIL/X509 while -1 will collide with RETRY and might confuse things. Change-Id: I98ecf6fa7342866b9124dc6f0b422cb9ce4a1ae7 Reviewed-on: https://boringssl-review.googlesource.com/8303 Reviewed-by: Adam Langley <agl@google.com>
2016-06-15 23:41:51 +01:00
int i;
CRYPTO_MUTEX_lock_write(&ctx->objs_lock);
tmp = X509_OBJECT_retrieve_by_subject(ctx->objs, type, name);
CRYPTO_MUTEX_unlock_write(&ctx->objs_lock);
if (tmp == NULL || type == X509_LU_CRL) {
Unwind X509_LU_RETRY and fix a lot of type confusion. (This change will be sent upstream. Since the legacy X.509 stack is just kept around for compatibility, if they decide to fix it in a different way, we may wish to revert this and apply their fix.) Dating back to SSLeay, X509_LOOKUP_METHOD had this X509_LU_RETRY machinery. But it's not documented and it appears to have never worked. Problems with the existing logic: - X509_LU_* is not sure whether it is a type enum (to be passed into X509_LOOKUP_by_*) or a return enum (to be retained by those same functions). - X509_LOOKUP_by_* is not sure whether it returns 0/1 or an X509_LU_* value. Looking at the functions themselves, one might think it's the latter, but for X509_LOOKUP_by_subject returning both 0 and X509_LU_FAIL. But looking at the call sites, some expect 0/1 (such as X509_STORE_get1_certs) while others expect an X509_LU_* enum (such as X509_STORE_CTX_get1_issuer). It is very fortunate that FAIL happens to be 0 and X509 happens to be 1. These functions primarily call to X509_LOOKUP_METHOD hooks. Looking through OpenSSL itself and code checked into Google, I found no evidence that any hooks have been implemented except for get_by_subject in by_dir.c. We take that one as definitive and observe it believes it returns 0/1. Notably, it returns 1 on success even if asked for a type other than X509_LU_X509. (X509_LU_X509 = 1. Others are different.) I found another piece of third-party software which corroborates this worldview. - X509_STORE_get_by_subject's handling of X509_LU_RETRY (it's the j < 0 check) is broken. It saves j into vs->current_method where it probably meant to save i. (This bug has existed since SSLeay.) It also returns j (supposedly X509_LU_RETRY) while all callers of X509_STORE_get_by_subject expect it to return 0/1 by checking with ! instead of <= 0. (Note that all other codepaths return 0 and 1 so this function did not actually believe it returned X509_LU_* most of the time.) This, in turn, gives us a free of uninitialized pointers in X509_STORE_get1_certs and other functions which expect that *ret is filled in if X509_STORE_get_by_subject returns success. GCC 4.9 with optimizations from the Android NDK noticed this, which trigged this saga. (It's only reachable if any X509_LOOKUP_METHOD returned X509_LU_RETRY.) - Although the code which expects X509_STORE_get_by_subject return 0/1 does not date to SSLeay, the X509_STORE_get_by_subject call in X509_STORE_CTX_get1_issuer *does* (though, at the time, it was inline in X509_verify_cert. That code believes X509_STORE_get_by_subject returns an X509_LU_* enum, but it doesn't work either! It believes *ret is filled in on X509_LU_RETRY, thus freeing another uninitialized pointer (GCC noticed this too). Since this "retry" code has clearly never worked, from SSLeay onwards, unwind it completely rather than attempt to fix it. No X509_LOOKUP_METHOD can possibly have depended on it. Matching all non-broken codepaths X509_LOOKUP_by_* now returns 0/1 and X509_STORE_get_by_subject returns 0/1. X509_LU_* is purely a type enum with X509_LU_{REJECT,FAIL} being legacy constants to keep old code compiling. (Upstream is recommended to remove those values altogether for 1.1.0.) On the off chance any get_by_* X509_LOOKUP_METHOD implementations did not return 0/1 (I have found no evidence anywhere of this, and I believe it wouldn't have worked anyway), the X509_LOOKUP_by_* wrapper functions will coerce the return values back to 0/1 before passing up to the callers which want 0/1. This both avoids the error-prone -1/0/1 calling convention and, more importantly, avoids problems with third-party callers which expect a X509_LU_* return code. 0/1 collide with FAIL/X509 while -1 will collide with RETRY and might confuse things. Change-Id: I98ecf6fa7342866b9124dc6f0b422cb9ce4a1ae7 Reviewed-on: https://boringssl-review.googlesource.com/8303 Reviewed-by: Adam Langley <agl@google.com>
2016-06-15 23:41:51 +01:00
for (i = 0; i < (int)sk_X509_LOOKUP_num(ctx->get_cert_methods); i++) {
lu = sk_X509_LOOKUP_value(ctx->get_cert_methods, i);
Unwind X509_LU_RETRY and fix a lot of type confusion. (This change will be sent upstream. Since the legacy X.509 stack is just kept around for compatibility, if they decide to fix it in a different way, we may wish to revert this and apply their fix.) Dating back to SSLeay, X509_LOOKUP_METHOD had this X509_LU_RETRY machinery. But it's not documented and it appears to have never worked. Problems with the existing logic: - X509_LU_* is not sure whether it is a type enum (to be passed into X509_LOOKUP_by_*) or a return enum (to be retained by those same functions). - X509_LOOKUP_by_* is not sure whether it returns 0/1 or an X509_LU_* value. Looking at the functions themselves, one might think it's the latter, but for X509_LOOKUP_by_subject returning both 0 and X509_LU_FAIL. But looking at the call sites, some expect 0/1 (such as X509_STORE_get1_certs) while others expect an X509_LU_* enum (such as X509_STORE_CTX_get1_issuer). It is very fortunate that FAIL happens to be 0 and X509 happens to be 1. These functions primarily call to X509_LOOKUP_METHOD hooks. Looking through OpenSSL itself and code checked into Google, I found no evidence that any hooks have been implemented except for get_by_subject in by_dir.c. We take that one as definitive and observe it believes it returns 0/1. Notably, it returns 1 on success even if asked for a type other than X509_LU_X509. (X509_LU_X509 = 1. Others are different.) I found another piece of third-party software which corroborates this worldview. - X509_STORE_get_by_subject's handling of X509_LU_RETRY (it's the j < 0 check) is broken. It saves j into vs->current_method where it probably meant to save i. (This bug has existed since SSLeay.) It also returns j (supposedly X509_LU_RETRY) while all callers of X509_STORE_get_by_subject expect it to return 0/1 by checking with ! instead of <= 0. (Note that all other codepaths return 0 and 1 so this function did not actually believe it returned X509_LU_* most of the time.) This, in turn, gives us a free of uninitialized pointers in X509_STORE_get1_certs and other functions which expect that *ret is filled in if X509_STORE_get_by_subject returns success. GCC 4.9 with optimizations from the Android NDK noticed this, which trigged this saga. (It's only reachable if any X509_LOOKUP_METHOD returned X509_LU_RETRY.) - Although the code which expects X509_STORE_get_by_subject return 0/1 does not date to SSLeay, the X509_STORE_get_by_subject call in X509_STORE_CTX_get1_issuer *does* (though, at the time, it was inline in X509_verify_cert. That code believes X509_STORE_get_by_subject returns an X509_LU_* enum, but it doesn't work either! It believes *ret is filled in on X509_LU_RETRY, thus freeing another uninitialized pointer (GCC noticed this too). Since this "retry" code has clearly never worked, from SSLeay onwards, unwind it completely rather than attempt to fix it. No X509_LOOKUP_METHOD can possibly have depended on it. Matching all non-broken codepaths X509_LOOKUP_by_* now returns 0/1 and X509_STORE_get_by_subject returns 0/1. X509_LU_* is purely a type enum with X509_LU_{REJECT,FAIL} being legacy constants to keep old code compiling. (Upstream is recommended to remove those values altogether for 1.1.0.) On the off chance any get_by_* X509_LOOKUP_METHOD implementations did not return 0/1 (I have found no evidence anywhere of this, and I believe it wouldn't have worked anyway), the X509_LOOKUP_by_* wrapper functions will coerce the return values back to 0/1 before passing up to the callers which want 0/1. This both avoids the error-prone -1/0/1 calling convention and, more importantly, avoids problems with third-party callers which expect a X509_LU_* return code. 0/1 collide with FAIL/X509 while -1 will collide with RETRY and might confuse things. Change-Id: I98ecf6fa7342866b9124dc6f0b422cb9ce4a1ae7 Reviewed-on: https://boringssl-review.googlesource.com/8303 Reviewed-by: Adam Langley <agl@google.com>
2016-06-15 23:41:51 +01:00
if (X509_LOOKUP_by_subject(lu, type, name, &stmp)) {
tmp = &stmp;
break;
}
}
if (tmp == NULL)
return 0;
}
/*
* if (ret->data.ptr != NULL) X509_OBJECT_free_contents(ret);
*/
ret->type = tmp->type;
ret->data.ptr = tmp->data.ptr;
X509_OBJECT_up_ref_count(ret);
return 1;
}
int X509_STORE_add_cert(X509_STORE *ctx, X509 *x)
{
X509_OBJECT *obj;
int ret = 1;
if (x == NULL)
return 0;
obj = (X509_OBJECT *)OPENSSL_malloc(sizeof(X509_OBJECT));
if (obj == NULL) {
OPENSSL_PUT_ERROR(X509, ERR_R_MALLOC_FAILURE);
return 0;
}
obj->type = X509_LU_X509;
obj->data.x509 = x;
CRYPTO_MUTEX_lock_write(&ctx->objs_lock);
X509_OBJECT_up_ref_count(obj);
if (X509_OBJECT_retrieve_match(ctx->objs, obj)) {
X509_OBJECT_free_contents(obj);
OPENSSL_free(obj);
OPENSSL_PUT_ERROR(X509, X509_R_CERT_ALREADY_IN_HASH_TABLE);
ret = 0;
} else if (!sk_X509_OBJECT_push(ctx->objs, obj)) {
X509_OBJECT_free_contents(obj);
OPENSSL_free(obj);
OPENSSL_PUT_ERROR(X509, ERR_R_MALLOC_FAILURE);
ret = 0;
}
CRYPTO_MUTEX_unlock_write(&ctx->objs_lock);
return ret;
}
int X509_STORE_add_crl(X509_STORE *ctx, X509_CRL *x)
{
X509_OBJECT *obj;
int ret = 1;
if (x == NULL)
return 0;
obj = (X509_OBJECT *)OPENSSL_malloc(sizeof(X509_OBJECT));
if (obj == NULL) {
OPENSSL_PUT_ERROR(X509, ERR_R_MALLOC_FAILURE);
return 0;
}
obj->type = X509_LU_CRL;
obj->data.crl = x;
CRYPTO_MUTEX_lock_write(&ctx->objs_lock);
X509_OBJECT_up_ref_count(obj);
if (X509_OBJECT_retrieve_match(ctx->objs, obj)) {
X509_OBJECT_free_contents(obj);
OPENSSL_free(obj);
OPENSSL_PUT_ERROR(X509, X509_R_CERT_ALREADY_IN_HASH_TABLE);
ret = 0;
} else if (!sk_X509_OBJECT_push(ctx->objs, obj)) {
X509_OBJECT_free_contents(obj);
OPENSSL_free(obj);
OPENSSL_PUT_ERROR(X509, ERR_R_MALLOC_FAILURE);
ret = 0;
}
CRYPTO_MUTEX_unlock_write(&ctx->objs_lock);
return ret;
}
void X509_STORE_set0_additional_untrusted(X509_STORE *ctx,
STACK_OF(X509) *untrusted) {
ctx->additional_untrusted = untrusted;
}
int X509_OBJECT_up_ref_count(X509_OBJECT *a)
{
switch (a->type) {
case X509_LU_X509:
X509_up_ref(a->data.x509);
break;
case X509_LU_CRL:
X509_CRL_up_ref(a->data.crl);
break;
}
return 1;
}
void X509_OBJECT_free_contents(X509_OBJECT *a)
{
switch (a->type) {
case X509_LU_X509:
X509_free(a->data.x509);
break;
case X509_LU_CRL:
X509_CRL_free(a->data.crl);
break;
}
}
Switch OPENSSL_VERSION_NUMBER to 1.1.0. Although we are derived from 1.0.2, we mimic 1.1.0 in some ways around our FOO_up_ref functions and opaque libssl types. This causes some difficulties when porting third-party code as any OPENSSL_VERSION_NUMBER checks for 1.1.0 APIs we have will be wrong. Moreover, adding accessors without changing OPENSSL_VERSION_NUMBER can break external projects. It is common to implement a compatibility version of an accessor under #ifdef as a static function. This then conflicts with our headers if we, unlike OpenSSL 1.0.2, have this function. This change switches OPENSSL_VERSION_NUMBER to 1.1.0 and atomically adds enough accessors for software with 1.1.0 support already. The hope is this will unblock hiding SSL_CTX and SSL_SESSION, which will be especially useful with C++-ficiation. The cost is we will hit some growing pains as more 1.1.0 consumers enter the ecosystem and we converge on the right set of APIs to import from upstream. It does not remove any 1.0.2 APIs, so we will not require that all projects support 1.1.0. The exception is APIs which changed in 1.1.0 but did not change the function signature. Those are breaking changes. Specifically: - SSL_CTX_sess_set_get_cb is now const-correct. - X509_get0_signature is now const-correct. For C++ consumers only, this change temporarily includes an overload hack for SSL_CTX_sess_set_get_cb that keeps the old callback working. This is a workaround for Node not yet supporting OpenSSL 1.1.0. The version number is set at (the as yet unreleased) 1.1.0g to denote that this change includes https://github.com/openssl/openssl/pull/4384. Bug: 91 Change-Id: I5eeb27448a6db4c25c244afac37f9604d9608a76 Reviewed-on: https://boringssl-review.googlesource.com/10340 Commit-Queue: David Benjamin <davidben@google.com> CQ-Verified: CQ bot account: commit-bot@chromium.org <commit-bot@chromium.org> Reviewed-by: Adam Langley <agl@google.com>
2016-08-12 19:48:19 +01:00
int X509_OBJECT_get_type(const X509_OBJECT *a)
{
return a->type;
}
X509 *X509_OBJECT_get0_X509(const X509_OBJECT *a)
{
if (a == NULL || a->type != X509_LU_X509) {
return NULL;
}
return a->data.x509;
}
static int x509_object_idx_cnt(STACK_OF(X509_OBJECT) *h, int type,
X509_NAME *name, int *pnmatch)
{
X509_OBJECT stmp;
X509 x509_s;
X509_CINF cinf_s;
X509_CRL crl_s;
X509_CRL_INFO crl_info_s;
stmp.type = type;
switch (type) {
case X509_LU_X509:
stmp.data.x509 = &x509_s;
x509_s.cert_info = &cinf_s;
cinf_s.subject = name;
break;
case X509_LU_CRL:
stmp.data.crl = &crl_s;
crl_s.crl = &crl_info_s;
crl_info_s.issuer = name;
break;
default:
/* abort(); */
return -1;
}
size_t idx;
if (!sk_X509_OBJECT_find(h, &idx, &stmp))
return -1;
if (pnmatch != NULL) {
int tidx;
const X509_OBJECT *tobj, *pstmp;
*pnmatch = 1;
pstmp = &stmp;
for (tidx = idx + 1; tidx < (int)sk_X509_OBJECT_num(h); tidx++) {
tobj = sk_X509_OBJECT_value(h, tidx);
if (x509_object_cmp(&tobj, &pstmp))
break;
(*pnmatch)++;
}
}
return idx;
}
int X509_OBJECT_idx_by_subject(STACK_OF(X509_OBJECT) *h, int type,
X509_NAME *name)
{
return x509_object_idx_cnt(h, type, name, NULL);
}
X509_OBJECT *X509_OBJECT_retrieve_by_subject(STACK_OF(X509_OBJECT) *h,
int type, X509_NAME *name)
{
int idx;
idx = X509_OBJECT_idx_by_subject(h, type, name);
if (idx == -1)
return NULL;
return sk_X509_OBJECT_value(h, idx);
}
Switch OPENSSL_VERSION_NUMBER to 1.1.0. Although we are derived from 1.0.2, we mimic 1.1.0 in some ways around our FOO_up_ref functions and opaque libssl types. This causes some difficulties when porting third-party code as any OPENSSL_VERSION_NUMBER checks for 1.1.0 APIs we have will be wrong. Moreover, adding accessors without changing OPENSSL_VERSION_NUMBER can break external projects. It is common to implement a compatibility version of an accessor under #ifdef as a static function. This then conflicts with our headers if we, unlike OpenSSL 1.0.2, have this function. This change switches OPENSSL_VERSION_NUMBER to 1.1.0 and atomically adds enough accessors for software with 1.1.0 support already. The hope is this will unblock hiding SSL_CTX and SSL_SESSION, which will be especially useful with C++-ficiation. The cost is we will hit some growing pains as more 1.1.0 consumers enter the ecosystem and we converge on the right set of APIs to import from upstream. It does not remove any 1.0.2 APIs, so we will not require that all projects support 1.1.0. The exception is APIs which changed in 1.1.0 but did not change the function signature. Those are breaking changes. Specifically: - SSL_CTX_sess_set_get_cb is now const-correct. - X509_get0_signature is now const-correct. For C++ consumers only, this change temporarily includes an overload hack for SSL_CTX_sess_set_get_cb that keeps the old callback working. This is a workaround for Node not yet supporting OpenSSL 1.1.0. The version number is set at (the as yet unreleased) 1.1.0g to denote that this change includes https://github.com/openssl/openssl/pull/4384. Bug: 91 Change-Id: I5eeb27448a6db4c25c244afac37f9604d9608a76 Reviewed-on: https://boringssl-review.googlesource.com/10340 Commit-Queue: David Benjamin <davidben@google.com> CQ-Verified: CQ bot account: commit-bot@chromium.org <commit-bot@chromium.org> Reviewed-by: Adam Langley <agl@google.com>
2016-08-12 19:48:19 +01:00
STACK_OF(X509_OBJECT) *X509_STORE_get0_objects(X509_STORE *st)
{
return st->objs;
}
STACK_OF (X509) * X509_STORE_get1_certs(X509_STORE_CTX *ctx, X509_NAME *nm)
{
int i, idx, cnt;
STACK_OF(X509) *sk;
X509 *x;
X509_OBJECT *obj;
sk = sk_X509_new_null();
if (sk == NULL)
return NULL;
CRYPTO_MUTEX_lock_write(&ctx->ctx->objs_lock);
idx = x509_object_idx_cnt(ctx->ctx->objs, X509_LU_X509, nm, &cnt);
if (idx < 0) {
/*
* Nothing found in cache: do lookup to possibly add new objects to
* cache
*/
X509_OBJECT xobj;
CRYPTO_MUTEX_unlock_write(&ctx->ctx->objs_lock);
if (!X509_STORE_get_by_subject(ctx, X509_LU_X509, nm, &xobj)) {
sk_X509_free(sk);
return NULL;
}
X509_OBJECT_free_contents(&xobj);
CRYPTO_MUTEX_lock_write(&ctx->ctx->objs_lock);
idx = x509_object_idx_cnt(ctx->ctx->objs, X509_LU_X509, nm, &cnt);
if (idx < 0) {
CRYPTO_MUTEX_unlock_write(&ctx->ctx->objs_lock);
sk_X509_free(sk);
return NULL;
}
}
for (i = 0; i < cnt; i++, idx++) {
obj = sk_X509_OBJECT_value(ctx->ctx->objs, idx);
x = obj->data.x509;
if (!sk_X509_push(sk, x)) {
CRYPTO_MUTEX_unlock_write(&ctx->ctx->objs_lock);
sk_X509_pop_free(sk, X509_free);
return NULL;
}
X509_up_ref(x);
}
CRYPTO_MUTEX_unlock_write(&ctx->ctx->objs_lock);
return sk;
}
STACK_OF (X509_CRL) * X509_STORE_get1_crls(X509_STORE_CTX *ctx, X509_NAME *nm)
{
int i, idx, cnt;
STACK_OF(X509_CRL) *sk;
X509_CRL *x;
X509_OBJECT *obj, xobj;
sk = sk_X509_CRL_new_null();
if (sk == NULL)
return NULL;
/* Always do lookup to possibly add new CRLs to cache. */
if (!X509_STORE_get_by_subject(ctx, X509_LU_CRL, nm, &xobj)) {
sk_X509_CRL_free(sk);
return NULL;
}
X509_OBJECT_free_contents(&xobj);
CRYPTO_MUTEX_lock_write(&ctx->ctx->objs_lock);
idx = x509_object_idx_cnt(ctx->ctx->objs, X509_LU_CRL, nm, &cnt);
if (idx < 0) {
CRYPTO_MUTEX_unlock_write(&ctx->ctx->objs_lock);
sk_X509_CRL_free(sk);
return NULL;
}
for (i = 0; i < cnt; i++, idx++) {
obj = sk_X509_OBJECT_value(ctx->ctx->objs, idx);
x = obj->data.crl;
X509_CRL_up_ref(x);
if (!sk_X509_CRL_push(sk, x)) {
CRYPTO_MUTEX_unlock_write(&ctx->ctx->objs_lock);
X509_CRL_free(x);
sk_X509_CRL_pop_free(sk, X509_CRL_free);
return NULL;
}
}
CRYPTO_MUTEX_unlock_write(&ctx->ctx->objs_lock);
return sk;
}
X509_OBJECT *X509_OBJECT_retrieve_match(STACK_OF(X509_OBJECT) *h,
X509_OBJECT *x)
{
size_t idx, i;
X509_OBJECT *obj;
if (!sk_X509_OBJECT_find(h, &idx, x)) {
return NULL;
}
if ((x->type != X509_LU_X509) && (x->type != X509_LU_CRL))
return sk_X509_OBJECT_value(h, idx);
for (i = idx; i < sk_X509_OBJECT_num(h); i++) {
obj = sk_X509_OBJECT_value(h, i);
if (x509_object_cmp
((const X509_OBJECT **)&obj, (const X509_OBJECT **)&x))
return NULL;
if (x->type == X509_LU_X509) {
if (!X509_cmp(obj->data.x509, x->data.x509))
return obj;
} else if (x->type == X509_LU_CRL) {
if (!X509_CRL_match(obj->data.crl, x->data.crl))
return obj;
} else
return obj;
}
return NULL;
}
/*
* Try to get issuer certificate from store. Due to limitations of the API
* this can only retrieve a single certificate matching a given subject name.
* However it will fill the cache with all matching certificates, so we can
* examine the cache for all matches. Return values are: 1 lookup
* successful. 0 certificate not found. -1 some other error.
*/
int X509_STORE_CTX_get1_issuer(X509 **issuer, X509_STORE_CTX *ctx, X509 *x)
{
X509_NAME *xn;
X509_OBJECT obj, *pobj;
Unwind X509_LU_RETRY and fix a lot of type confusion. (This change will be sent upstream. Since the legacy X.509 stack is just kept around for compatibility, if they decide to fix it in a different way, we may wish to revert this and apply their fix.) Dating back to SSLeay, X509_LOOKUP_METHOD had this X509_LU_RETRY machinery. But it's not documented and it appears to have never worked. Problems with the existing logic: - X509_LU_* is not sure whether it is a type enum (to be passed into X509_LOOKUP_by_*) or a return enum (to be retained by those same functions). - X509_LOOKUP_by_* is not sure whether it returns 0/1 or an X509_LU_* value. Looking at the functions themselves, one might think it's the latter, but for X509_LOOKUP_by_subject returning both 0 and X509_LU_FAIL. But looking at the call sites, some expect 0/1 (such as X509_STORE_get1_certs) while others expect an X509_LU_* enum (such as X509_STORE_CTX_get1_issuer). It is very fortunate that FAIL happens to be 0 and X509 happens to be 1. These functions primarily call to X509_LOOKUP_METHOD hooks. Looking through OpenSSL itself and code checked into Google, I found no evidence that any hooks have been implemented except for get_by_subject in by_dir.c. We take that one as definitive and observe it believes it returns 0/1. Notably, it returns 1 on success even if asked for a type other than X509_LU_X509. (X509_LU_X509 = 1. Others are different.) I found another piece of third-party software which corroborates this worldview. - X509_STORE_get_by_subject's handling of X509_LU_RETRY (it's the j < 0 check) is broken. It saves j into vs->current_method where it probably meant to save i. (This bug has existed since SSLeay.) It also returns j (supposedly X509_LU_RETRY) while all callers of X509_STORE_get_by_subject expect it to return 0/1 by checking with ! instead of <= 0. (Note that all other codepaths return 0 and 1 so this function did not actually believe it returned X509_LU_* most of the time.) This, in turn, gives us a free of uninitialized pointers in X509_STORE_get1_certs and other functions which expect that *ret is filled in if X509_STORE_get_by_subject returns success. GCC 4.9 with optimizations from the Android NDK noticed this, which trigged this saga. (It's only reachable if any X509_LOOKUP_METHOD returned X509_LU_RETRY.) - Although the code which expects X509_STORE_get_by_subject return 0/1 does not date to SSLeay, the X509_STORE_get_by_subject call in X509_STORE_CTX_get1_issuer *does* (though, at the time, it was inline in X509_verify_cert. That code believes X509_STORE_get_by_subject returns an X509_LU_* enum, but it doesn't work either! It believes *ret is filled in on X509_LU_RETRY, thus freeing another uninitialized pointer (GCC noticed this too). Since this "retry" code has clearly never worked, from SSLeay onwards, unwind it completely rather than attempt to fix it. No X509_LOOKUP_METHOD can possibly have depended on it. Matching all non-broken codepaths X509_LOOKUP_by_* now returns 0/1 and X509_STORE_get_by_subject returns 0/1. X509_LU_* is purely a type enum with X509_LU_{REJECT,FAIL} being legacy constants to keep old code compiling. (Upstream is recommended to remove those values altogether for 1.1.0.) On the off chance any get_by_* X509_LOOKUP_METHOD implementations did not return 0/1 (I have found no evidence anywhere of this, and I believe it wouldn't have worked anyway), the X509_LOOKUP_by_* wrapper functions will coerce the return values back to 0/1 before passing up to the callers which want 0/1. This both avoids the error-prone -1/0/1 calling convention and, more importantly, avoids problems with third-party callers which expect a X509_LU_* return code. 0/1 collide with FAIL/X509 while -1 will collide with RETRY and might confuse things. Change-Id: I98ecf6fa7342866b9124dc6f0b422cb9ce4a1ae7 Reviewed-on: https://boringssl-review.googlesource.com/8303 Reviewed-by: Adam Langley <agl@google.com>
2016-06-15 23:41:51 +01:00
int idx, ret;
size_t i;
xn = X509_get_issuer_name(x);
Unwind X509_LU_RETRY and fix a lot of type confusion. (This change will be sent upstream. Since the legacy X.509 stack is just kept around for compatibility, if they decide to fix it in a different way, we may wish to revert this and apply their fix.) Dating back to SSLeay, X509_LOOKUP_METHOD had this X509_LU_RETRY machinery. But it's not documented and it appears to have never worked. Problems with the existing logic: - X509_LU_* is not sure whether it is a type enum (to be passed into X509_LOOKUP_by_*) or a return enum (to be retained by those same functions). - X509_LOOKUP_by_* is not sure whether it returns 0/1 or an X509_LU_* value. Looking at the functions themselves, one might think it's the latter, but for X509_LOOKUP_by_subject returning both 0 and X509_LU_FAIL. But looking at the call sites, some expect 0/1 (such as X509_STORE_get1_certs) while others expect an X509_LU_* enum (such as X509_STORE_CTX_get1_issuer). It is very fortunate that FAIL happens to be 0 and X509 happens to be 1. These functions primarily call to X509_LOOKUP_METHOD hooks. Looking through OpenSSL itself and code checked into Google, I found no evidence that any hooks have been implemented except for get_by_subject in by_dir.c. We take that one as definitive and observe it believes it returns 0/1. Notably, it returns 1 on success even if asked for a type other than X509_LU_X509. (X509_LU_X509 = 1. Others are different.) I found another piece of third-party software which corroborates this worldview. - X509_STORE_get_by_subject's handling of X509_LU_RETRY (it's the j < 0 check) is broken. It saves j into vs->current_method where it probably meant to save i. (This bug has existed since SSLeay.) It also returns j (supposedly X509_LU_RETRY) while all callers of X509_STORE_get_by_subject expect it to return 0/1 by checking with ! instead of <= 0. (Note that all other codepaths return 0 and 1 so this function did not actually believe it returned X509_LU_* most of the time.) This, in turn, gives us a free of uninitialized pointers in X509_STORE_get1_certs and other functions which expect that *ret is filled in if X509_STORE_get_by_subject returns success. GCC 4.9 with optimizations from the Android NDK noticed this, which trigged this saga. (It's only reachable if any X509_LOOKUP_METHOD returned X509_LU_RETRY.) - Although the code which expects X509_STORE_get_by_subject return 0/1 does not date to SSLeay, the X509_STORE_get_by_subject call in X509_STORE_CTX_get1_issuer *does* (though, at the time, it was inline in X509_verify_cert. That code believes X509_STORE_get_by_subject returns an X509_LU_* enum, but it doesn't work either! It believes *ret is filled in on X509_LU_RETRY, thus freeing another uninitialized pointer (GCC noticed this too). Since this "retry" code has clearly never worked, from SSLeay onwards, unwind it completely rather than attempt to fix it. No X509_LOOKUP_METHOD can possibly have depended on it. Matching all non-broken codepaths X509_LOOKUP_by_* now returns 0/1 and X509_STORE_get_by_subject returns 0/1. X509_LU_* is purely a type enum with X509_LU_{REJECT,FAIL} being legacy constants to keep old code compiling. (Upstream is recommended to remove those values altogether for 1.1.0.) On the off chance any get_by_* X509_LOOKUP_METHOD implementations did not return 0/1 (I have found no evidence anywhere of this, and I believe it wouldn't have worked anyway), the X509_LOOKUP_by_* wrapper functions will coerce the return values back to 0/1 before passing up to the callers which want 0/1. This both avoids the error-prone -1/0/1 calling convention and, more importantly, avoids problems with third-party callers which expect a X509_LU_* return code. 0/1 collide with FAIL/X509 while -1 will collide with RETRY and might confuse things. Change-Id: I98ecf6fa7342866b9124dc6f0b422cb9ce4a1ae7 Reviewed-on: https://boringssl-review.googlesource.com/8303 Reviewed-by: Adam Langley <agl@google.com>
2016-06-15 23:41:51 +01:00
if (!X509_STORE_get_by_subject(ctx, X509_LU_X509, xn, &obj))
return 0;
/* If certificate matches all OK */
if (ctx->check_issued(ctx, x, obj.data.x509)) {
*issuer = obj.data.x509;
return 1;
}
X509_OBJECT_free_contents(&obj);
/* Else find index of first cert accepted by 'check_issued' */
ret = 0;
CRYPTO_MUTEX_lock_write(&ctx->ctx->objs_lock);
idx = X509_OBJECT_idx_by_subject(ctx->ctx->objs, X509_LU_X509, xn);
if (idx != -1) { /* should be true as we've had at least one
* match */
/* Look through all matching certs for suitable issuer */
for (i = idx; i < sk_X509_OBJECT_num(ctx->ctx->objs); i++) {
pobj = sk_X509_OBJECT_value(ctx->ctx->objs, i);
/* See if we've run past the matches */
if (pobj->type != X509_LU_X509)
break;
if (X509_NAME_cmp(xn, X509_get_subject_name(pobj->data.x509)))
break;
if (ctx->check_issued(ctx, x, pobj->data.x509)) {
*issuer = pobj->data.x509;
X509_OBJECT_up_ref_count(pobj);
ret = 1;
break;
}
}
}
CRYPTO_MUTEX_unlock_write(&ctx->ctx->objs_lock);
return ret;
}
int X509_STORE_set_flags(X509_STORE *ctx, unsigned long flags)
{
return X509_VERIFY_PARAM_set_flags(ctx->param, flags);
}
int X509_STORE_set_depth(X509_STORE *ctx, int depth)
{
X509_VERIFY_PARAM_set_depth(ctx->param, depth);
return 1;
}
int X509_STORE_set_purpose(X509_STORE *ctx, int purpose)
{
return X509_VERIFY_PARAM_set_purpose(ctx->param, purpose);
}
int X509_STORE_set_trust(X509_STORE *ctx, int trust)
{
return X509_VERIFY_PARAM_set_trust(ctx->param, trust);
}
int X509_STORE_set1_param(X509_STORE *ctx, X509_VERIFY_PARAM *param)
{
return X509_VERIFY_PARAM_set1(ctx->param, param);
}
Switch OPENSSL_VERSION_NUMBER to 1.1.0. Although we are derived from 1.0.2, we mimic 1.1.0 in some ways around our FOO_up_ref functions and opaque libssl types. This causes some difficulties when porting third-party code as any OPENSSL_VERSION_NUMBER checks for 1.1.0 APIs we have will be wrong. Moreover, adding accessors without changing OPENSSL_VERSION_NUMBER can break external projects. It is common to implement a compatibility version of an accessor under #ifdef as a static function. This then conflicts with our headers if we, unlike OpenSSL 1.0.2, have this function. This change switches OPENSSL_VERSION_NUMBER to 1.1.0 and atomically adds enough accessors for software with 1.1.0 support already. The hope is this will unblock hiding SSL_CTX and SSL_SESSION, which will be especially useful with C++-ficiation. The cost is we will hit some growing pains as more 1.1.0 consumers enter the ecosystem and we converge on the right set of APIs to import from upstream. It does not remove any 1.0.2 APIs, so we will not require that all projects support 1.1.0. The exception is APIs which changed in 1.1.0 but did not change the function signature. Those are breaking changes. Specifically: - SSL_CTX_sess_set_get_cb is now const-correct. - X509_get0_signature is now const-correct. For C++ consumers only, this change temporarily includes an overload hack for SSL_CTX_sess_set_get_cb that keeps the old callback working. This is a workaround for Node not yet supporting OpenSSL 1.1.0. The version number is set at (the as yet unreleased) 1.1.0g to denote that this change includes https://github.com/openssl/openssl/pull/4384. Bug: 91 Change-Id: I5eeb27448a6db4c25c244afac37f9604d9608a76 Reviewed-on: https://boringssl-review.googlesource.com/10340 Commit-Queue: David Benjamin <davidben@google.com> CQ-Verified: CQ bot account: commit-bot@chromium.org <commit-bot@chromium.org> Reviewed-by: Adam Langley <agl@google.com>
2016-08-12 19:48:19 +01:00
X509_VERIFY_PARAM *X509_STORE_get0_param(X509_STORE *ctx)
{
return ctx->param;
}
void X509_STORE_set_verify_cb(X509_STORE *ctx,
int (*verify_cb) (int, X509_STORE_CTX *))
{
ctx->verify_cb = verify_cb;
}
void X509_STORE_set_lookup_crls_cb(X509_STORE *ctx,
STACK_OF (X509_CRL) *
(*cb) (X509_STORE_CTX *ctx, X509_NAME *nm))
{
ctx->lookup_crls = cb;
}
X509_STORE *X509_STORE_CTX_get0_store(X509_STORE_CTX *ctx)
{
return ctx->ctx;
}