Commit graph

7617 commits

Author SHA1 Message Date
Dawid Drozd 428cc52a73 Allow to forward declare of public structs #1215
Thanks to forward declare we can declare `struct` in our header file instead making #include
2018-07-24 10:02:47 +02:00
Simon Butcher 4072aaea73 Merge remote-tracking branch 'public/pr/1875' into development 2018-07-24 08:41:07 +01:00
Simon Butcher a7b35434e6 Merge remote-tracking branch 'public/pr/1871' into development 2018-07-24 08:26:13 +01:00
Simon Butcher 6965f771de Fix use of Knowledge Base in documentation
Made the use of 'Knowledge Base' consistent in the CONTRIBUTING.md and
README.md files.
2018-07-23 23:59:24 +01:00
Brian J Murray 4736e96568 add myself to changelog 2018-07-23 10:34:47 -07:00
Brian J Murray ca2ea4e217 Fix issue if salt = NULL and salt_len !=0 in mbedtls_hkdf_extract() 2018-07-23 10:34:47 -07:00
Simon Butcher 5a5d03f025 Add test guidelines to CONTRIBUTING.md 2018-07-23 16:16:04 +01:00
Simon Butcher 10f9663fcb Added clarification to CONTRIBUTING.md
Added rationale as to when API changes are permitted, and clarified why we
try to preserve the API/ABI.
2018-07-23 16:10:36 +01:00
Simon Butcher 5cf4d0694f Refine the definition of net_sockets.h
Clarified the purport of net_socket.h to make its purpose clearer.
2018-07-23 14:39:45 +01:00
Simon Butcher cce68bec1d Add a macro to define the memory size in ssl_server2.c
When MBEDTLS_MEMORY_BUFFER_ALLOC_C was defined, the sample ssl_server2.c was
using its own memory buffer for memory allocated by the library. The memory
used wasn't obvious, so this adds a macro for the memory buffer allocated to
make the allocated memory size more obvious and hence easier to configure.
2018-07-23 14:26:09 +01:00
Simon Butcher e459f07b55 Correct logic to exclude i386 inline assenbly when -O0
The i386 MPI inline assembly code was being incorrectly included when
all compiler optimisation was disabled.
2018-07-23 13:31:32 +01:00
Simon Butcher 8e6a22ac3d Add additional i386 tests to all.sh
Added an additional i386 test to all.sh, to allow one test with -O0 which
compiles out inline assembly, and one to test with -01 which includes the inline
assembly.
2018-07-23 13:31:19 +01:00
Simon Butcher 7a6da6ed5e Expand i386 all.sh tests to full config ASan builds
The i386 test builds were only building the default configuration and had
no address sanitisation. This commit expands the test configuration to the full
configuration in all.sh and builds with ASan for when the test suites are
executed.
2018-07-22 18:01:19 +01:00
Simon Butcher ccbd46435f Increase the memory buffer size for ssl_server2.c
Newer features in the library have increased the overall RAM usage of the
library, when all features are enabled. ssl_server2.c, with all features enabled
was running out of memory for the ssl-opt.sh test 'Authentication: client
max_int chain, server required'.

This commit increases the memory buffer allocation for ssl_server2.c to allow
the test to work with all features enabled.
2018-07-22 14:43:39 +01:00
Andres Amaya Garcia 4c761fab7f Add test for empty app data records to ssl-opt.sh 2018-07-20 23:09:29 +01:00
Andres Amaya Garcia 81f0633c16 Add ChangeLog entry for empty app data fix 2018-07-20 23:09:29 +01:00
Andres Amaya Garcia ce6fbac247 Fix ssl_client2 to send 0-length app data 2018-07-20 23:09:29 +01:00
Angus Gratton 1a7a17e548 Check for invalid short Alert messages
(Short Change Cipher Spec & Handshake messages are already checked for.)
2018-07-20 23:09:29 +01:00
Angus Gratton 34817929ea TLSv1.2: Treat zero-length fragments as invalid, unless they are application data
TLS v1.2 explicitly disallows other kinds of zero length fragments (earlier standards
don't mention zero-length fragments at all).
2018-07-20 23:09:29 +01:00
Angus Gratton b512bc1d29 CBC mode: Allow zero-length message fragments (100% padding)
Fixes https://github.com/ARMmbed/mbedtls/issues/1632
2018-07-20 23:09:29 +01:00
Simon Butcher 7870ec12c4 Add API/ABI section to CONTRIBUTING.md
Added a section on API/ABI compatibility on the development branch to the
CONTRIBUTING.md guidelines. Also added to the testing section, refined the LTS
section and changed some formatting for consistency.
2018-07-20 19:39:10 +01:00
Jaeden Amero 7c1258dc78 all.sh: Return error on keep-going failure
When calling all.sh from a script and using "--keep-going", errors were
sometimes missed due to all.sh always returning 0 "success" return code.
Return 1 if there is any failure encountered during a "keep-going" run.
2018-07-20 17:14:21 +01:00
Simon Butcher 7d728bd70e Merge remote-tracking branch 'public/pr/1454' into development 2018-07-20 14:33:44 +01:00
Simon Butcher 922bd1efb2 Merge remote-tracking branch 'public/pr/1752' into development 2018-07-20 14:33:18 +01:00
Simon Butcher 862e703d51 Merge remote-tracking branch 'public/pr/921' into development 2018-07-20 14:30:50 +01:00
Simon Butcher be685ba337 Merge remote-tracking branch 'public/pr/1836' into development 2018-07-19 20:01:23 +01:00
Simon Butcher 4f37bcabf9 Fix ChangeLog entry for issue #1663
The ChangeLog entry was under the wrong version, and under Changes, not
BugFixes.
2018-07-19 19:52:32 +01:00
Simon Butcher df15356259 Merge remote-tracking branch 'public/pr/1663' into development 2018-07-19 19:48:10 +01:00
Simon Butcher fad547072a Merge remote-tracking branch 'public/pr/532' into development 2018-07-19 16:15:51 +01:00
Simon Butcher a72098b4d6 Merge remote-tracking branch 'public/pr/1778' into development 2018-07-19 16:10:38 +01:00
Simon Butcher ca33cafab0 Add definition of purpose and use of networking module
The purpose of the networking module can sometimes be misunderstood. This adds
a definition and explanation of what the networking module is and what it can be
used for.
2018-07-18 17:52:14 +01:00
Ron Eldor 00115034ea Repharse comments
Rephrase comments to clarify them.
2018-07-17 11:21:50 +03:00
k-stachowiak 6ca436a457 Update change log 2018-07-16 12:20:10 +02:00
Manuel Pégourié-Gonnard 830ce11eba Clarify attack conditions in the ChangeLog.
Referring to the previous entry could imply that the current one was limited
to SHA-384 too, which it isn't.
2018-07-11 18:27:08 +02:00
k-stachowiak 21feae58cb Update change log 2018-07-11 17:34:55 +02:00
k-stachowiak a47911cb70 Fix memory leak in ssl_setup 2018-07-11 17:26:07 +02:00
Dawid Drozd 0e2c07e83e
Remove unnecessary mark as unused #1098
`ret` is used always at line 1305 in statement:
`if( ( ret = pk_parse_key_pkcs8_unencrypted_der( pk, key, keylen ) ) == 0 )`
2018-07-11 15:16:53 +02:00
Manuel Pégourié-Gonnard 8744a02357 Clarify a few comments
The "+" sign could be misinterpreted as addition.
2018-07-11 12:30:40 +02:00
Simon Butcher 4b9a3addb6 Disable use of the i386 assembly for option -O0
We don't compile in the assembly code if compiler optimisations are disabled as
the number of registers used in the assembly code doesn't work with the -O0
option. Also anyone select -O0 probably doesn't want to compile in the assembly
code anyway.
2018-07-10 20:18:29 +01:00
Simon Butcher 00af447ba8 Add ChangeLog entry for PR #536 2018-07-10 15:35:43 +01:00
Simon Butcher e7aeef09ee Merge remote-tracking branch 'public/pr/536' into development 2018-07-10 15:24:26 +01:00
Simon Butcher 32b074720e Merge remote-tracking branch 'public/pr/1737' into development 2018-07-10 14:57:50 +01:00
Simon Butcher 727325baa7 Merge remote-tracking branch 'public/pr/1788' into development 2018-07-10 14:54:52 +01:00
Simon Butcher 24c0dbf71e Merge remote-tracking branch 'public/pr/1821' into development 2018-07-10 14:34:01 +01:00
Simon Butcher cdbb2f2168 Merge remote-tracking branch 'public/pr/1563' into development 2018-07-10 12:49:26 +01:00
Simon Butcher 231d7e5669 Add ChangeLog entry for PR #1567.
ChangeLog entry for platform support for the Haiku OS. PR #1567.
2018-07-10 11:56:19 +01:00
Simon Butcher 6331cb0607 Fix some whitespace issues in ChangeLog and CMakeLists.txt
Stray tab in library/CMakeLists.txt and incorrect formatting in ChangeLog.
2018-07-10 11:48:42 +01:00
Simon Butcher d21bd31759 Merge remote-tracking branch 'public/pr/1567' into development 2018-07-10 11:43:06 +01:00
Manuel Pégourié-Gonnard 6a25cfae2a Avoid debug message that might leak length
The length to the debug message could conceivably leak through the time it
takes to print it, and that length would in turn reveal whether padding was
correct or not.
2018-07-10 11:15:36 +02:00
k-stachowiak a5fbfd7cd8 Enable snprintf on FreeBSD 2018-07-08 13:22:11 +01:00