Commit graph

25 commits

Author SHA1 Message Date
Jerry Yu
0a6272d6c9 revert padlock from aesni module
Signed-off-by: Jerry Yu <jerry.h.yu@arm.com>
2023-08-18 17:35:59 +08:00
Jerry Yu
35b59d7805 exclude arm64ec mode for aesni
AESNI does not work correctly for msvc arm64ec

Signed-off-by: Jerry Yu <jerry.h.yu@arm.com>
2023-08-17 10:34:15 +08:00
Jerry Yu
3ce0398d1d Add compiler cflags error message
Signed-off-by: Jerry Yu <jerry.h.yu@arm.com>
2023-08-16 17:22:18 +08:00
Jerry Yu
c628486cd9 enable runtime detection when padlock enabled and plain c disabled
Signed-off-by: Jerry Yu <jerry.h.yu@arm.com>
2023-08-16 16:14:02 +08:00
Jerry Yu
e62ff09569 Restore aesni for i386
intrinsic code can be work on i386 also

Signed-off-by: Jerry Yu <jerry.h.yu@arm.com>
2023-08-16 14:33:16 +08:00
Jerry Yu
240bb11171 Add gnu check for aseni assembly code
Signed-off-by: Jerry Yu <jerry.h.yu@arm.com>
2023-08-11 10:45:35 +08:00
Jerry Yu
8189f32945 improve aesni check for x86_64
`MBEDTLS_AESNI_C` does not depends on `MBEDTLS_HAVE_ASM`
when intrinsic is available.

And compiler relative checks only work on x86_64, it should
be only checked on x86_64.

Signed-off-by: Jerry Yu <jerry.h.yu@arm.com>
2023-08-10 15:17:45 +08:00
Jerry Yu
9c0b7d13bf Remove unnecessary name check tag
Signed-off-by: Jerry Yu <jerry.h.yu@arm.com>
2023-08-04 17:25:59 +08:00
Jerry Yu
02b1519ab6 move accelerator checks to aes.c
Origin position is always validate due to conflict
between the guards in `aes.c` and module undef check

Signed-off-by: Jerry Yu <jerry.h.yu@arm.com>
2023-08-02 17:44:02 +08:00
Jerry Yu
3660623e59 Rename plain c option and update comments
Signed-off-by: Jerry Yu <jerry.h.yu@arm.com>
2023-08-02 17:44:01 +08:00
Jerry Yu
8840a8c574 fix wrong checks
Signed-off-by: Jerry Yu <jerry.h.yu@arm.com>
2023-08-02 17:44:00 +08:00
Jerry Yu
3fcf2b5053 Rename HAS_NO_PLAIN_C to DONT_USE_SOFTWARE_CRYPTO
Signed-off-by: Jerry Yu <jerry.h.yu@arm.com>
2023-08-02 17:44:00 +08:00
Jerry Yu
4d030f3acd Add check for no aes implementation provided
Signed-off-by: Jerry Yu <jerry.h.yu@arm.com>
2023-08-02 17:43:59 +08:00
Jerry Yu
315fd30201 Rename plain c disable option
Signed-off-by: Jerry Yu <jerry.h.yu@arm.com>
2023-08-02 17:43:59 +08:00
Jerry Yu
0d4f4e5b01 Add option to disable built-in aes implementation.
For time being, there are only two aes implementations for known
architectures. I define runtime detection function as const when
built-in was disabled. In this case, compiler will remove dead
built-in code.

Signed-off-by: Jerry Yu <jerry.h.yu@arm.com>
2023-08-02 17:43:54 +08:00
Dave Rodgman
96a9e6a9dd Address test review comments
Signed-off-by: Dave Rodgman <dave.rodgman@arm.com>
2023-06-16 20:18:36 +01:00
Dave Rodgman
e07c670e47 Allow all.sh to override intrinsics vs asm selection
Signed-off-by: Dave Rodgman <dave.rodgman@arm.com>
2023-06-16 13:21:28 +01:00
Dave Rodgman
6365a681c8 Prefer intrinsics over asm for AES-NI
Signed-off-by: Dave Rodgman <dave.rodgman@arm.com>
2023-05-22 11:14:36 +01:00
Gilles Peskine
36b9e47eed Fix preprocessor conditional
This was intended as an if-else-if chain. Make it so.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2023-03-17 17:31:18 +01:00
Gilles Peskine
30e9f2a293 Finish sentence in comment
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2023-03-17 17:29:58 +01:00
Gilles Peskine
9c682e724a AESNI: Overhaul implementation selection
Have clearly separated code to:
* determine whether the assembly-based implementation is available;
* determine whether the intrinsics-based implementation is available;
* select one of the available implementations if any.

Now MBEDTLS_AESNI_HAVE_CODE can be the single interface for aes.c and
aesni.c to determine which AESNI is built.

Change the implementation selection: now, if both implementations are
available, always prefer assembly. Before, the intrinsics were used if
available. This preference is to minimize disruption, and will likely
be revised in a later minor release.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2023-03-16 17:21:33 +01:00
Gilles Peskine
9af58cd7f8 New preprocessor symbol indicating that AESNI support is present
The configuration symbol MBEDTLS_AESNI_C requests AESNI support, but it is
ignored if the platform doesn't have AESNI. This allows keeping
MBEDTLS_AESNI_C enabled (as it is in the default build) when building for
platforms other than x86_64, or when MBEDTLS_HAVE_ASM is disabled.

To facilitate maintenance, always use the symbol MBEDTLS_AESNI_HAVE_CODE to
answer the question "can I call mbedtls_aesni_xxx functions?", rather than
repeating the check `defined(MBEDTLS_AESNI_C) && ...`.

Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2023-03-15 19:38:37 +01:00
Gilles Peskine
449bd8303e Switch to the new code style
Signed-off-by: Gilles Peskine <Gilles.Peskine@arm.com>
2023-01-11 14:50:10 +01:00
Bence Szépkúti
c662b36af2 Replace all inclusions of config.h
Also remove preprocessor logic for MBEDTLS_CONFIG_FILE, since
build_info.h alreadyy handles it.

This commit was generated using the following script:

# ========================
#!/bin/sh
git ls-files | grep -v '^include/mbedtls/build_info\.h$' | xargs sed -b -E -i '
/^#if !?defined\(MBEDTLS_CONFIG_FILE\)/i#include "mbedtls/build_info.h"
//,/^#endif/d
'
# ========================

Signed-off-by: Bence Szépkúti <bence.szepkuti@arm.com>
2021-06-28 09:24:07 +01:00
Chris Jones
187782f41e Move aesni.h to library
`aesni.h` is an internal header and is moved accordingly.

Also removes some references to internal headers in scripts with
only public headers.

Signed-off-by: Chris Jones <christopher.jones@arm.com>
2021-03-10 12:52:37 +00:00
Renamed from include/mbedtls/aesni.h (Browse further)