openssl/crypto/evp/keymgmt_meth.c

338 lines
11 KiB
C
Raw Normal View History

/*
* Copyright 2019 The OpenSSL Project Authors. All Rights Reserved.
*
* Licensed under the Apache License 2.0 (the "License"). You may not use
* this file except in compliance with the License. You can obtain a copy
* in the file LICENSE in the source distribution or at
* https://www.openssl.org/source/license.html
*/
#include <openssl/crypto.h>
#include <openssl/core_numbers.h>
#include <openssl/evp.h>
#include <openssl/err.h>
#include "internal/provider.h"
#include "internal/refcount.h"
#include "crypto/evp.h"
#include "evp_local.h"
static void *keymgmt_new(void)
{
EVP_KEYMGMT *keymgmt = NULL;
if ((keymgmt = OPENSSL_zalloc(sizeof(*keymgmt))) == NULL
|| (keymgmt->lock = CRYPTO_THREAD_lock_new()) == NULL) {
EVP_KEYMGMT_free(keymgmt);
EVPerr(0, ERR_R_MALLOC_FAILURE);
return NULL;
}
keymgmt->refcnt = 1;
return keymgmt;
}
static void *keymgmt_from_dispatch(int name_id,
const OSSL_DISPATCH *fns,
OSSL_PROVIDER *prov)
{
EVP_KEYMGMT *keymgmt = NULL;
int setparamfncnt = 0, getparamfncnt = 0, importfncnt = 0, exportfncnt = 0;
if ((keymgmt = keymgmt_new()) == NULL) {
EVP_KEYMGMT_free(keymgmt);
return NULL;
}
keymgmt->name_id = name_id;
for (; fns->function_id != 0; fns++) {
switch (fns->function_id) {
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
case OSSL_FUNC_KEYMGMT_NEW:
if (keymgmt->new == NULL)
keymgmt->new = OSSL_get_OP_keymgmt_new(fns);
break;
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
case OSSL_FUNC_KEYMGMT_FREE:
if (keymgmt->free == NULL)
keymgmt->free = OSSL_get_OP_keymgmt_free(fns);
break;
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
case OSSL_FUNC_KEYMGMT_GET_PARAMS:
if (keymgmt->get_params == NULL) {
getparamfncnt++;
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
keymgmt->get_params = OSSL_get_OP_keymgmt_get_params(fns);
}
break;
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
case OSSL_FUNC_KEYMGMT_GETTABLE_PARAMS:
if (keymgmt->gettable_params == NULL) {
getparamfncnt++;
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
keymgmt->gettable_params =
OSSL_get_OP_keymgmt_gettable_params(fns);
}
break;
case OSSL_FUNC_KEYMGMT_SET_PARAMS:
if (keymgmt->set_params == NULL) {
setparamfncnt++;
keymgmt->set_params = OSSL_get_OP_keymgmt_set_params(fns);
}
break;
case OSSL_FUNC_KEYMGMT_SETTABLE_PARAMS:
if (keymgmt->settable_params == NULL) {
setparamfncnt++;
keymgmt->settable_params =
OSSL_get_OP_keymgmt_settable_params(fns);
}
break;
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
case OSSL_FUNC_KEYMGMT_QUERY_OPERATION_NAME:
if (keymgmt->query_operation_name == NULL)
keymgmt->query_operation_name =
OSSL_get_OP_keymgmt_query_operation_name(fns);
break;
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
case OSSL_FUNC_KEYMGMT_HAS:
if (keymgmt->has == NULL)
keymgmt->has = OSSL_get_OP_keymgmt_has(fns);
break;
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
case OSSL_FUNC_KEYMGMT_VALIDATE:
if (keymgmt->validate == NULL)
keymgmt->validate = OSSL_get_OP_keymgmt_validate(fns);
break;
case OSSL_FUNC_KEYMGMT_MATCH:
if (keymgmt->match == NULL)
keymgmt->match = OSSL_get_OP_keymgmt_match(fns);
break;
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
case OSSL_FUNC_KEYMGMT_IMPORT:
if (keymgmt->import == NULL) {
importfncnt++;
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
keymgmt->import = OSSL_get_OP_keymgmt_import(fns);
}
break;
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
case OSSL_FUNC_KEYMGMT_IMPORT_TYPES:
if (keymgmt->import_types == NULL) {
importfncnt++;
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
keymgmt->import_types = OSSL_get_OP_keymgmt_import_types(fns);
}
break;
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
case OSSL_FUNC_KEYMGMT_EXPORT:
if (keymgmt->export == NULL) {
exportfncnt++;
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
keymgmt->export = OSSL_get_OP_keymgmt_export(fns);
}
break;
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
case OSSL_FUNC_KEYMGMT_EXPORT_TYPES:
if (keymgmt->export_types == NULL) {
exportfncnt++;
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
keymgmt->export_types = OSSL_get_OP_keymgmt_export_types(fns);
}
break;
}
}
/*
* Try to check that the method is sensible.
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
* At least one constructor and the destructor are MANDATORY
* The functions 'has' is MANDATORY
* It makes no sense being able to free stuff if you can't create it.
* It makes no sense providing OSSL_PARAM descriptors for import and
* export if you can't import or export.
*/
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
if (keymgmt->free == NULL
|| keymgmt->new == NULL
|| keymgmt->has == NULL
|| (getparamfncnt != 0 && getparamfncnt != 2)
|| (setparamfncnt != 0 && setparamfncnt != 2)
|| (importfncnt != 0 && importfncnt != 2)
|| (exportfncnt != 0 && exportfncnt != 2)) {
EVP_KEYMGMT_free(keymgmt);
EVPerr(0, EVP_R_INVALID_PROVIDER_FUNCTIONS);
return NULL;
}
keymgmt->prov = prov;
if (prov != NULL)
ossl_provider_up_ref(prov);
return keymgmt;
}
EVP_KEYMGMT *evp_keymgmt_fetch_by_number(OPENSSL_CTX *ctx, int name_id,
const char *properties)
{
return evp_generic_fetch_by_number(ctx,
OSSL_OP_KEYMGMT, name_id, properties,
keymgmt_from_dispatch,
(int (*)(void *))EVP_KEYMGMT_up_ref,
(void (*)(void *))EVP_KEYMGMT_free);
}
EVP_KEYMGMT *EVP_KEYMGMT_fetch(OPENSSL_CTX *ctx, const char *algorithm,
const char *properties)
{
return evp_generic_fetch(ctx, OSSL_OP_KEYMGMT, algorithm, properties,
keymgmt_from_dispatch,
(int (*)(void *))EVP_KEYMGMT_up_ref,
(void (*)(void *))EVP_KEYMGMT_free);
}
int EVP_KEYMGMT_up_ref(EVP_KEYMGMT *keymgmt)
{
int ref = 0;
CRYPTO_UP_REF(&keymgmt->refcnt, &ref, keymgmt->lock);
return 1;
}
void EVP_KEYMGMT_free(EVP_KEYMGMT *keymgmt)
{
int ref = 0;
if (keymgmt == NULL)
return;
CRYPTO_DOWN_REF(&keymgmt->refcnt, &ref, keymgmt->lock);
if (ref > 0)
return;
ossl_provider_free(keymgmt->prov);
CRYPTO_THREAD_lock_free(keymgmt->lock);
OPENSSL_free(keymgmt);
}
const OSSL_PROVIDER *EVP_KEYMGMT_provider(const EVP_KEYMGMT *keymgmt)
{
return keymgmt->prov;
}
int EVP_KEYMGMT_number(const EVP_KEYMGMT *keymgmt)
{
return keymgmt->name_id;
}
int EVP_KEYMGMT_is_a(const EVP_KEYMGMT *keymgmt, const char *name)
{
return evp_is_a(keymgmt->prov, keymgmt->name_id, NULL, name);
}
void EVP_KEYMGMT_do_all_provided(OPENSSL_CTX *libctx,
void (*fn)(EVP_KEYMGMT *keymgmt, void *arg),
void *arg)
{
evp_generic_do_all(libctx, OSSL_OP_KEYMGMT,
(void (*)(void *, void *))fn, arg,
keymgmt_from_dispatch,
(void (*)(void *))EVP_KEYMGMT_free);
}
void EVP_KEYMGMT_names_do_all(const EVP_KEYMGMT *keymgmt,
void (*fn)(const char *name, void *data),
void *data)
{
if (keymgmt->prov != NULL)
evp_names_do_all(keymgmt->prov, keymgmt->name_id, fn, data);
}
/*
* Internal API that interfaces with the method function pointers
*/
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
void *evp_keymgmt_newdata(const EVP_KEYMGMT *keymgmt)
{
void *provctx = ossl_provider_ctx(EVP_KEYMGMT_provider(keymgmt));
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
/*
* TODO(3.0) 'new' is currently mandatory on its own, but when new
* constructors appear, it won't be quite as mandatory, so we have
* a check for future cases.
*/
if (keymgmt->new == NULL)
return NULL;
return keymgmt->new(provctx);
}
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
void evp_keymgmt_freedata(const EVP_KEYMGMT *keymgmt, void *keydata)
{
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
/* This is mandatory, no need to check for its presence */
keymgmt->free(keydata);
}
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
int evp_keymgmt_get_params(const EVP_KEYMGMT *keymgmt, void *keydata,
OSSL_PARAM params[])
{
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
if (keymgmt->get_params == NULL)
return 1;
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
return keymgmt->get_params(keydata, params);
}
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
const OSSL_PARAM *evp_keymgmt_gettable_params(const EVP_KEYMGMT *keymgmt)
{
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
if (keymgmt->gettable_params == NULL)
return NULL;
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
return keymgmt->gettable_params();
}
int evp_keymgmt_set_params(const EVP_KEYMGMT *keymgmt, void *keydata,
const OSSL_PARAM params[])
{
if (keymgmt->set_params == NULL)
return 1;
return keymgmt->set_params(keydata, params);
}
const OSSL_PARAM *evp_keymgmt_settable_params(const EVP_KEYMGMT *keymgmt)
{
if (keymgmt->settable_params == NULL)
return NULL;
return keymgmt->settable_params();
}
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
int evp_keymgmt_has(const EVP_KEYMGMT *keymgmt, void *keydata, int selection)
{
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
/* This is mandatory, no need to check for its presence */
return keymgmt->has(keydata, selection);
}
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
int evp_keymgmt_validate(const EVP_KEYMGMT *keymgmt, void *keydata,
int selection)
{
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
/* We assume valid if the implementation doesn't have a function */
if (keymgmt->validate == NULL)
return 1;
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
return keymgmt->validate(keydata, selection);
}
int evp_keymgmt_match(const EVP_KEYMGMT *keymgmt,
const void *keydata1, const void *keydata2,
int selection)
{
/* We assume no match if the implementation doesn't have a function */
if (keymgmt->match == NULL)
return 0;
return keymgmt->match(keydata1, keydata2, selection);
}
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
int evp_keymgmt_import(const EVP_KEYMGMT *keymgmt, void *keydata,
int selection, const OSSL_PARAM params[])
{
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
if (keymgmt->import == NULL)
return 0;
return keymgmt->import(keydata, selection, params);
}
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
const OSSL_PARAM *evp_keymgmt_import_types(const EVP_KEYMGMT *keymgmt,
int selection)
{
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
if (keymgmt->import_types == NULL)
return NULL;
return keymgmt->import_types(selection);
}
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
int evp_keymgmt_export(const EVP_KEYMGMT *keymgmt, void *keydata,
int selection, OSSL_CALLBACK *param_cb, void *cbarg)
{
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
if (keymgmt->export == NULL)
return 0;
return keymgmt->export(keydata, selection, param_cb, cbarg);
}
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
const OSSL_PARAM *evp_keymgmt_export_types(const EVP_KEYMGMT *keymgmt,
int selection)
{
Redesign the KEYMGMT libcrypto <-> provider interface - the basics The KEYMGMT libcrypto <-> provider interface currently makes a few assumptions: 1. provider side domain parameters and key data isn't mutable. In other words, as soon as a key has been created in any (loaded, imported data, ...), it's set in stone. 2. provider side domain parameters can be strictly separated from the key data. This does work for the most part, but there are places where that's a bit too rigid for the functionality that the EVP_PKEY API delivers. Key data needs to be mutable to allow the flexibility that functions like EVP_PKEY_copy_parameters promise, as well as to provide the combinations of data that an EVP_PKEY is generally assumed to be able to hold: - domain parameters only - public key only - public key + private key - domain parameters + public key - domain parameters + public key + private key To remedy all this, we: 1. let go of the distinction between domain parameters and key material proper in the libcrypto <-> provider interface. As a consequence, functions that still need it gain a selection argument, which is a set of bits that indicate what parts of the key object are to be considered in a specific call. This allows a reduction of very similar functions into one. 2. Rework the libcrypto <-> provider interface so provider side key objects are created and destructed with a separate function, and get their data filled and extracted in through import and export. (future work will see other key object constructors and other functions to fill them with data) Fixes #10979 squash! Redesign the KEYMGMT libcrypto <-> provider interface - the basics Remedy 1 needs a rewrite: Reviewed-by: Matt Caswell <matt@openssl.org> Reviewed-by: Shane Lontis <shane.lontis@oracle.com> Reviewed-by: Paul Dale <paul.dale@oracle.com> (Merged from https://github.com/openssl/openssl/pull/11006)
2020-02-03 01:56:07 +08:00
if (keymgmt->export_types == NULL)
return NULL;
return keymgmt->export_types(selection);
}