"mbedtls_ssl_handshake_params: reorder fields to save code size" moved this
filed earlier along with byte-sized fields that should be in the 128-element
access window on Arm Thumb. This took away precious room in the 128-byte
window. Move it back further out.
Results (same architecture, config-suite-b.h + MBEDTLS_ECDH_LEGACY_CONTEXT +
MBEDTLS_ECP_RESTARTABLE):
library/ssl_cli.o: 2860 -> 2816 (diff: 44)
library/ssl_msg.o: 3080 -> 3076 (diff: 4)
library/ssl_srv.o: 3340 -> 3300 (diff: 40)
library/ssl_tls.o: 6546 -> 6478 (diff: 68)
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
The current testing of the PSA configuration is
based on test code located in the library itself.
Remove this code as we are moving to using a
test library instead.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Rename test driver entry points to
libtestdriver1_<name of the Mbed TLS entry point>.
This aligns with the renaming of all Mbed TLS APIs
for the test driver library (that will be put in place
in the following commits) to avoid name conflicts
when linking it with the Mbed TLS library.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
In preparation of the driver test entry points to be
provided by a test driver library, move their prototypes
to tests directory.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
In psa_asymmetric_encrypt/decrypt(), always return
PSA_ERROR_INVALID_ARGUMENT if the key is a PSA key
and the algorithm is not a PSA algorithm we know
about, whether RSA is supported or not.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Define test driver entry points that provide an alternative
to Mbed TLS driver entry points only when the PSA configuration
is used. Their purpose is only to test the PSA configuration
thus there is no good reason to use them out of this scope.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
The test entry points defined in psa_crypto_hash.c
and psa_crypto_mac.c are supposed to be exact
clones of the Mbed TLS driver entry points. Thus
the operation type should be the Mbed TLS operation
type not a test one. There was no compilation error
as the hash and cipher operation test types are
currently equal to the Mbed TLS ones.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Align RSA/ECP sign/verify hash dispatch with the
corresponding code of the library. The library
code was modified recently but not the test code
one and these modifications ease the following work.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Reorder fields mbedtls_ssl_handshake_params in order to save code on Arm
Thumb builds. The general idea is to put often-used fields in the direct
access window of 128 elements from the beginning of the structure.
The reordering is a human selection based on a report of field offset and
use counts, and informed by measuring the code size with various
arrangements. Some notes:
* This is the same reordering as the corresponding commit in #5189 for 2.2x.
* I moved most byte-sized fields at the beginning where they're sure to be
in the direct access window.
* I moved buffering earlier because it can be around the threshold depending
on the configuration, and it's accessed in a lot of places.
* I moved several fields, including update_checksum and friends, early so
that they're guaranteed to be in the early access window.
Results (arm-none-eabi-gcc 7.3.1, build_arm_none_eabi_gcc_m0plus build):
library/ssl_cli.o: 19763 -> 19687 (diff: 76)
library/ssl_msg.o: 24874 -> 24834 (diff: 40)
library/ssl_srv.o: 20754 -> 20562 (diff: 192)
library/ssl_tls.o: 21003 -> 20907 (diff: 96)
library/ssl_tls13_client.o: 7284 -> 7272 (diff: 12)
library/ssl_tls13_generic.o: 4749 -> 4721 (diff: 28)
library/ssl_tls13_keys.o: 5133 -> 5077 (diff: 56)
Results (same architecture, config-suite-b.h + MBEDTLS_ECDH_LEGACY_CONTEXT +
MBEDTLS_ECP_RESTARTABLE):
library/ssl_cli.o: 3000 -> 2936 (diff: 64)
library/ssl_msg.o: 3084 -> 3080 (diff: 4)
library/ssl_srv.o: 3428 -> 3400 (diff: 28)
library/ssl_tls.o: 6754 -> 6730 (diff: 24)
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Replace bitfields mbedtls_ssl_handshake_params by bytes. This saves some
code size, and since the bitfields weren't group, this doesn't increase the
RAM usage.
Replace several ints that only store values in the range 0..255 by uint8_t.
This can increase or decrease the code size depending on the architecture
and on how the field is used. I chose changes that save code size on Arm
Thumb builds and will save more after field reordering.
Leave the bitfields in struct mbedtls_ssl_hs_buffer alone: replacing them by
uint8_t slightly increases the code size.
Results (arm-none-eabi-gcc 7.3.1, build_arm_none_eabi_gcc_m0plus build):
library/ssl_cli.o: 19759 -> 19763 (diff: -4)
library/ssl_srv.o: 20790 -> 20754 (diff: 36)
library/ssl_tls13_keys.o: 5153 -> 5133 (diff: 20)
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Don't use the output buffer in psa_aead_generate_nonce()
to pass the generated nonce to the driver as a local
attacker could potentially control it.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Don't use the output buffer in psa_cipher_encrypt()
to pass the generated IV to the driver as local
attacker could potentially control it.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Align the execution of cipher one-shot APIs with
that of cipher multi-part APIs: always exit
through the exit-labelled section.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Don't use the output buffer in psa_cipher_generate_iv()
to pass the generated IV to the driver as local
attacker could potentially control it.
Signed-off-by: Ronald Cron <ronald.cron@arm.com>
Move fields around to have fewer accesses outside the 128-element Thumb
direct access window.
Make the same change as in 2.27+, for the same small benefit.
Results (arm-none-eabi-gcc 7.3.1, build_arm_none_eabi_gcc_m0plus build):
library/psa_crypto.o: 16434 -> 16414 (diff: 20)
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Remove support signature PKCS1 v1.5 in CertificateVerify.
Remove useless server states in test script
Signed-off-by: XiaokangQian <xiaokang.qian@arm.com>
Change client test code to support rsa pss signatures
Add test cases for rsa pss signature in ssl-opt.sh
Signed-off-by: XiaokangQian <xiaokang.qian@arm.com>
psa_key_derivation_output_key: prioritize BAD_STATE over NOT_PERMITTED
If psa_key_derivation_output_key() is called on an operation which hasn't been
set up or which has been aborted, return PSA_ERROR_BAD_STATE. Only return
PSA_ERROR_NOT_PERMITTED if the operation state is ok for
psa_key_derivation_input_bytes() or psa_key_derivation_output_bytes() but not
ok to output a key.
Ideally psa_key_derivation_output_key() would return PSA_ERROR_NOT_PERMITTED
only when psa_key_derivation_output_bytes() is possible, but this is clumsier
to implement.
Signed-off-by: Dave Rodgman <dave.rodgman@arm.com>
Previously passing a NULL or zero length password into either
mbedtls_pkcs12_pbe() or mbedtls_pkcs12_derive() could cause an infinate
loop, and it was also possible to pass a NULL password, with a non-zero
length, which would cause memory corruption.
I have fixed these errors, and improved the documentation to reflect the
changes and further explain what is expected of the inputs.
Signed-off-by: Paul Elliott <paul.elliott@arm.com>
Remove client certificate verify in tests.
Change the layout of structure to fix abi_api check issues.
Add comments of Finished.
Align with the coding styles.
Signed-off-by: XiaokangQian <xiaokang.qian@arm.com>
This is a workaround for an issue with mkstemp() in older MinGW releases that
causes simultaneous creation of .a files in the same directory to fail.
Fixes#5146
Signed-off-by: Tom Cosgrove <tom.cosgrove@arm.com>
Change debug messag for server finished.
Change name of generate_application_keys.
Remove the client vertificate tests from ssl-opt.sh.
Add test strings for server finished in ssl-opt.sh.
Signed-off-by: XiaokangQian <xiaokang.qian@arm.com>
Move set_state function into client
Add back export_key callback function in generate
application keys
Signed-off-by: XiaokangQian <xiaokang.qian@arm.com>
Fix the variable not inialized issue, remove the client
certificate related code, remove early data related code.
Signed-off-by: XiaokangQian <xiaokang.qian@arm.com>
The current definition of PSA_ALG_IS_HASH_AND_SIGN includes
PSA_ALG_RSA_PKCS1V15_SIGN_RAW and PSA_ALG_ECDSA_ANY, which don't strictly
follow the hash-and-sign paradigm: the algorithm does not encode a hash
algorithm that is applied prior to the signature step. The definition in
fact encompasses what can be used with psa_sign_hash/psa_verify_hash, so
it's the correct definition for PSA_ALG_IS_SIGN_HASH. Therefore this commit
moves definition of PSA_ALG_IS_HASH_AND_SIGN to PSA_ALG_IS_SIGN_HASH, and
replace the definition of PSA_ALG_IS_HASH_AND_SIGN by a correct one (based
on PSA_ALG_IS_SIGN_HASH, excluding the algorithms where the pre-signature
step isn't to apply the hash encoded in the algorithm).
In the definition of PSA_ALG_SIGN_GET_HASH, keep the condition for a nonzero
output to be PSA_ALG_IS_HASH_AND_SIGN.
Everywhere else in the code base (definition of PSA_ALG_IS_SIGN_MESSAGE, and
every use of PSA_ALG_IS_HASH_AND_SIGN outside of crypto_values.h), we meant
PSA_ALG_IS_SIGN_HASH where we wrote PSA_ALG_IS_HASH_AND_SIGN, so do a
global replacement.
```
git grep -l IS_HASH_AND_SIGN ':!include/psa/crypto_values.h' | xargs perl -i -pe 's/ALG_IS_HASH_AND_SIGN/ALG_IS_SIGN_HASH/g'
```
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
The makefile build specifies -L. -lmbedx509 -lmbedcrypto flags first,
and only then object files referencing symbols from those libraries.
In this order the linker will not add the linked libraries to the
DT_NEEDED section because they are not referenced yet (at least that
happens for me on ubuntu 20.04 with the default gnu compiler tools).
By first specifying the object files and then the linked libraries, we
do end up with libmbedx509 and libmbedcrypto in the DT_NEEDED sections.
This way running dlopen(...) on libmedtls.so just works.
Note that the CMake build does this by default.
Signed-off-by: Harmen Stoppels <harmenstoppels@gmail.com>
The refactoring is needed for the group api to work
properly. Code is modified to use mbedtls_get_supported_groups
instead of direct access so that both deprecated and new api
are useable.
Signed-off-by: Brett Warren <brett.warren@arm.com>
mbedtls_ssl_conf_groups allows supported groups for key
sharing to be configured via their IANA NamedGroup ID.
This is added in anticipation of PQC and Hybrid key
sharing algorithms being integrated into Mbed TLS.
mbedtls_ssl_conf_curves is deprecated in favor of
mbedtls_ssl_conf_groups. handshake_init has been
modified to translate and copy curves configured
via conf_curves into a heap allocatied array of
NamedGroup IDs. This allows the refactoring of code
interacting with conf_curve related variables (such
as curve_list) to use NamedGroup IDs while retaining
the deprecated API.
Signed-off-by: Brett Warren <brett.warren@arm.com>
To test c <= high, instead of testing the sign of (high + 1) - c, negate the
sign of high - c (as we're doing for c - low). This is a little easier to
read and shaves 2 instructions off the arm thumb build with
arm-none-eabi-gcc 7.3.1.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
n was used for two different purposes. Give it a different name the second
time. This does not seem to change the generated code when compiling with
optimization for size or performance.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Instead of doing constant-flow table lookup, which requires 64 memory loads
for each lookup into a 64-entry table, do a range-based calculation, which
requires more CPU instructions per range but there are only 5 ranges.
I expect a significant performance gain (although smaller than for decoding
since the encoding table is half the size), but I haven't measured. Code
size is slightly smaller.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Document what each local variable does when it isn't obvious from the name.
Don't reuse a variable for different purposes.
This commit has very little impact on the generated code (same code size on
a sample Thumb build), although it does fix a theoretical bug that 2^32
spaces inside a line would be ignored instead of treated as an error.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Instead of doing constant-flow table lookup, which requires 128 memory loads
for each lookup into a 128-entry table, do a range-based calculation, which
requires more CPU instructions per range but there are only 5 ranges.
Experimentally, this is ~12x faster on my PC (based on
programs/x509/load_roots). The code is slightly smaller, too.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
Base64 decoding uses equality comparison tests for characters that don't
leak information about the content of the data other than its length, such
as whitespace. Do this with '=' as well, since it only reveals information
about the length. This way the table lookup can focus on character validity
and decoding value.
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
When the option is On, CMake will have rules to generate the generated
files using scripts etc. When the option is Off, CMake will assume the
files are available from the source tree; in that mode, it won't require
any extra tools (Perl for example) compared to when we committed the
files to git.
The intention is that users will never need to adjust this option:
- in the development branch (and features branches etc.) the option is
always On (development mode);
- in released tarballs, which include the generated files, we'll switch
the option to Off (release mode) in the same commit that re-adds the
generated files.
Signed-off-by: Manuel Pégourié-Gonnard <manuel.pegourie-gonnard@arm.com>
