kris/boringssl
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
eb43eca5a8
boringssl/crypto/abi_self_test.cc

524 lines
22 KiB
C++
Raw Normal View History

Add an ABI testing framework. Dear reader, I must apologize in advance. This CL contains the following: - A new 256-line perlasm file with non-trivial perl bits and a dual-ABI variadic function caller. - C preprocessor gymnastics, with variadic macros and fun facts about __VA_ARGS__'s behavior on empty argument lists. - C++ template gymnastics, including variadic arguments, template specialization, std::enable_if, and machinery to control template argument deduction. Enjoy. This tests that our assembly functions correctly honor platform ABI conventions. Right now this only tests callee-saved registers, but it should be extendable to SEH/CFI unwind testing with single-step debugging APIs. Register-checking does not involve anything funny and should be compatible with SDE. (The future unwind testing is unlikely to be compatible.) This CL adds support for x86_64 SysV and Win64 ABIs. ARM, AArch64, and x86 can be added in the future. The testing is injected in two places. First, all the assembly tests in p256-x86_64-test.cc are now instrumented. This is the intended workflow and should capture all registers. However, we currently do not unit-test our assembly much directly. We should do that as follow-up work[0] but, in the meantime, I've also wrapped all of the GTest main function in an ABI test. This is imperfect as ABI failures may be masked by other stack frames, but it costs nothing[1] and is pretty reliable at catching Win64 xmm register failures. [0] An alternate strategy would be, in debug builds, unconditionally instrument every assembly call in libcrypto. But the CHECK_ABI macro would be difficult to replicate in pure C, and unwind testing may be too invasive for this. Still, something to consider when we C++ libcrypto. [1] When single-stepped unwind testing exists, it won't cost nothing. The gtest_main.cc call will turn unwind testing off. Change-Id: I6643b26445891fd46abfacac52bc024024c8d7f6 Reviewed-on: https://boringssl-review.googlesource.com/c/33764 Reviewed-by: Adam Langley <agl@google.com> Reviewed-by: Adam Langley <alangley@gmail.com> Commit-Queue: David Benjamin <davidben@google.com>
2018-12-16 00:58:43 +00:00
/* Copyright (c) 2018, Google Inc.
*
* Permission to use, copy, modify, and/or distribute this software for any
* purpose with or without fee is hereby granted, provided that the above
* copyright notice and this permission notice appear in all copies.
*
* THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
* WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
* MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY
* SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION
* OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN
* CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE. */
#include <gtest/gtest.h>
#include <gtest/gtest-spi.h>
#include <openssl/rand.h>
#include "test/abi_test.h"
Add a CFI tester to CHECK_ABI. This uses the x86 trap flag and libunwind to test CFI works at each instruction. For now, it just uses the system one out of pkg-config and disables unwind tests if unavailable. We'll probably want to stick a copy into //third_party and perhaps try the LLVM one later. This tester caught two bugs in P-256 CFI annotations already: I47b5f9798b3bcee1748e537b21c173d312a14b42 and I9f576d868850312d6c14d1386f8fbfa85021b347 An earlier design used PTRACE_SINGLESTEP with libunwind's remote unwinding features. ptrace is a mess around stop signals (see group-stop discussion in ptrace(2)) and this is 10x faster, so I went with it. The question of which is more future-proof is complex: - There are two libunwinds with the same API, https://www.nongnu.org/libunwind/ and LLVM's. This currently uses the system nongnu.org for convenience. In future, LLVM's should be easier to bundle (less complex build) and appears to even support Windows, but I haven't tested this. Moreover, setting the trap flag keeps the test single-process, which is less complex on Windows. That suggests the trap flag design and switching to LLVM later. However... - Not all architectures have a trap flag settable by userspace. As far as I can tell, ARMv8's PSTATE.SS can only be set from the kernel. If we stick with nongnu.org libunwind, we can use PTRACE_SINGLESTEP and remote unwinding. Or we implement it for LLVM. Another thought is for the ptracer to bounce SIGTRAP back into the process, to share the local unwinding code. - ARMv7 has no trap flag at all and PTRACE_SINGLESTEP fails. Debuggers single-step by injecting breakpoints instead. However, ARMv8's trap flag seems to work in both AArch32 and AArch64 modes, so we may be able to condition it on a 64-bit kernel. Sadly, neither strategy works with Intel SDE. Adding flags to cpucap vectors as we do with ARM would help, but it would not emulate CPUs newer than the host CPU. For now, I've just had SDE tests disable these. Annoyingly, CMake does not allow object libraries to have dependencies, so make test_support a proper static library. Rename the target to test_support_lib to avoid https://gitlab.kitware.com/cmake/cmake/issues/17785 Update-Note: This adds a new optional test dependency, but it's disabled by default (define BORINGSSL_HAVE_LIBUNWIND), so consumers do not need to do anything. We'll probably want to adjust this in the future. Bug: 181 Change-Id: I817263d7907aff0904a9cee83f8b26747262cc0c Reviewed-on: https://boringssl-review.googlesource.com/c/33966 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2018-12-21 23:58:36 +00:00
static bool test_function_ok;
static int TestFunction(int a1, int a2, int a3, int a4, int a5, int a6, int a7,
Implement ABI testing for aarch64. This caught a bug in bn_mul_mont. Tested manually on iOS and Android. Change-Id: I1819fcd9ad34dbe3ba92bba952507d86dd12185a Reviewed-on: https://boringssl-review.googlesource.com/c/34805 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2019-02-02 00:25:07 +00:00
int a8) {
Add a CFI tester to CHECK_ABI. This uses the x86 trap flag and libunwind to test CFI works at each instruction. For now, it just uses the system one out of pkg-config and disables unwind tests if unavailable. We'll probably want to stick a copy into //third_party and perhaps try the LLVM one later. This tester caught two bugs in P-256 CFI annotations already: I47b5f9798b3bcee1748e537b21c173d312a14b42 and I9f576d868850312d6c14d1386f8fbfa85021b347 An earlier design used PTRACE_SINGLESTEP with libunwind's remote unwinding features. ptrace is a mess around stop signals (see group-stop discussion in ptrace(2)) and this is 10x faster, so I went with it. The question of which is more future-proof is complex: - There are two libunwinds with the same API, https://www.nongnu.org/libunwind/ and LLVM's. This currently uses the system nongnu.org for convenience. In future, LLVM's should be easier to bundle (less complex build) and appears to even support Windows, but I haven't tested this. Moreover, setting the trap flag keeps the test single-process, which is less complex on Windows. That suggests the trap flag design and switching to LLVM later. However... - Not all architectures have a trap flag settable by userspace. As far as I can tell, ARMv8's PSTATE.SS can only be set from the kernel. If we stick with nongnu.org libunwind, we can use PTRACE_SINGLESTEP and remote unwinding. Or we implement it for LLVM. Another thought is for the ptracer to bounce SIGTRAP back into the process, to share the local unwinding code. - ARMv7 has no trap flag at all and PTRACE_SINGLESTEP fails. Debuggers single-step by injecting breakpoints instead. However, ARMv8's trap flag seems to work in both AArch32 and AArch64 modes, so we may be able to condition it on a 64-bit kernel. Sadly, neither strategy works with Intel SDE. Adding flags to cpucap vectors as we do with ARM would help, but it would not emulate CPUs newer than the host CPU. For now, I've just had SDE tests disable these. Annoyingly, CMake does not allow object libraries to have dependencies, so make test_support a proper static library. Rename the target to test_support_lib to avoid https://gitlab.kitware.com/cmake/cmake/issues/17785 Update-Note: This adds a new optional test dependency, but it's disabled by default (define BORINGSSL_HAVE_LIBUNWIND), so consumers do not need to do anything. We'll probably want to adjust this in the future. Bug: 181 Change-Id: I817263d7907aff0904a9cee83f8b26747262cc0c Reviewed-on: https://boringssl-review.googlesource.com/c/33966 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2018-12-21 23:58:36 +00:00
test_function_ok = a1 == 1 || a2 == 2 || a3 == 3 || a4 == 4 || a5 == 5 ||
Implement ABI testing for aarch64. This caught a bug in bn_mul_mont. Tested manually on iOS and Android. Change-Id: I1819fcd9ad34dbe3ba92bba952507d86dd12185a Reviewed-on: https://boringssl-review.googlesource.com/c/34805 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2019-02-02 00:25:07 +00:00
a6 == 6 || a7 == 7 || a8 == 8;
Add a CFI tester to CHECK_ABI. This uses the x86 trap flag and libunwind to test CFI works at each instruction. For now, it just uses the system one out of pkg-config and disables unwind tests if unavailable. We'll probably want to stick a copy into //third_party and perhaps try the LLVM one later. This tester caught two bugs in P-256 CFI annotations already: I47b5f9798b3bcee1748e537b21c173d312a14b42 and I9f576d868850312d6c14d1386f8fbfa85021b347 An earlier design used PTRACE_SINGLESTEP with libunwind's remote unwinding features. ptrace is a mess around stop signals (see group-stop discussion in ptrace(2)) and this is 10x faster, so I went with it. The question of which is more future-proof is complex: - There are two libunwinds with the same API, https://www.nongnu.org/libunwind/ and LLVM's. This currently uses the system nongnu.org for convenience. In future, LLVM's should be easier to bundle (less complex build) and appears to even support Windows, but I haven't tested this. Moreover, setting the trap flag keeps the test single-process, which is less complex on Windows. That suggests the trap flag design and switching to LLVM later. However... - Not all architectures have a trap flag settable by userspace. As far as I can tell, ARMv8's PSTATE.SS can only be set from the kernel. If we stick with nongnu.org libunwind, we can use PTRACE_SINGLESTEP and remote unwinding. Or we implement it for LLVM. Another thought is for the ptracer to bounce SIGTRAP back into the process, to share the local unwinding code. - ARMv7 has no trap flag at all and PTRACE_SINGLESTEP fails. Debuggers single-step by injecting breakpoints instead. However, ARMv8's trap flag seems to work in both AArch32 and AArch64 modes, so we may be able to condition it on a 64-bit kernel. Sadly, neither strategy works with Intel SDE. Adding flags to cpucap vectors as we do with ARM would help, but it would not emulate CPUs newer than the host CPU. For now, I've just had SDE tests disable these. Annoyingly, CMake does not allow object libraries to have dependencies, so make test_support a proper static library. Rename the target to test_support_lib to avoid https://gitlab.kitware.com/cmake/cmake/issues/17785 Update-Note: This adds a new optional test dependency, but it's disabled by default (define BORINGSSL_HAVE_LIBUNWIND), so consumers do not need to do anything. We'll probably want to adjust this in the future. Bug: 181 Change-Id: I817263d7907aff0904a9cee83f8b26747262cc0c Reviewed-on: https://boringssl-review.googlesource.com/c/33966 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2018-12-21 23:58:36 +00:00
return 42;
Add an ABI testing framework. Dear reader, I must apologize in advance. This CL contains the following: - A new 256-line perlasm file with non-trivial perl bits and a dual-ABI variadic function caller. - C preprocessor gymnastics, with variadic macros and fun facts about __VA_ARGS__'s behavior on empty argument lists. - C++ template gymnastics, including variadic arguments, template specialization, std::enable_if, and machinery to control template argument deduction. Enjoy. This tests that our assembly functions correctly honor platform ABI conventions. Right now this only tests callee-saved registers, but it should be extendable to SEH/CFI unwind testing with single-step debugging APIs. Register-checking does not involve anything funny and should be compatible with SDE. (The future unwind testing is unlikely to be compatible.) This CL adds support for x86_64 SysV and Win64 ABIs. ARM, AArch64, and x86 can be added in the future. The testing is injected in two places. First, all the assembly tests in p256-x86_64-test.cc are now instrumented. This is the intended workflow and should capture all registers. However, we currently do not unit-test our assembly much directly. We should do that as follow-up work[0] but, in the meantime, I've also wrapped all of the GTest main function in an ABI test. This is imperfect as ABI failures may be masked by other stack frames, but it costs nothing[1] and is pretty reliable at catching Win64 xmm register failures. [0] An alternate strategy would be, in debug builds, unconditionally instrument every assembly call in libcrypto. But the CHECK_ABI macro would be difficult to replicate in pure C, and unwind testing may be too invasive for this. Still, something to consider when we C++ libcrypto. [1] When single-stepped unwind testing exists, it won't cost nothing. The gtest_main.cc call will turn unwind testing off. Change-Id: I6643b26445891fd46abfacac52bc024024c8d7f6 Reviewed-on: https://boringssl-review.googlesource.com/c/33764 Reviewed-by: Adam Langley <agl@google.com> Reviewed-by: Adam Langley <alangley@gmail.com> Commit-Queue: David Benjamin <davidben@google.com>
2018-12-16 00:58:43 +00:00
}
TEST(ABITest, SanityCheck) {
Avoid unwind tests on libc functions. When built under UBSan, it gets confused inside a PLT stub. Change-Id: Ib082ecc076ba2111337ff5921e465e4beb99aab5 Reviewed-on: https://boringssl-review.googlesource.com/c/34448 Reviewed-by: Adam Langley <agl@google.com>
2019-01-22 03:42:25 +00:00
EXPECT_NE(0, CHECK_ABI_NO_UNWIND(strcmp, "hello", "world"));
Add an ABI testing framework. Dear reader, I must apologize in advance. This CL contains the following: - A new 256-line perlasm file with non-trivial perl bits and a dual-ABI variadic function caller. - C preprocessor gymnastics, with variadic macros and fun facts about __VA_ARGS__'s behavior on empty argument lists. - C++ template gymnastics, including variadic arguments, template specialization, std::enable_if, and machinery to control template argument deduction. Enjoy. This tests that our assembly functions correctly honor platform ABI conventions. Right now this only tests callee-saved registers, but it should be extendable to SEH/CFI unwind testing with single-step debugging APIs. Register-checking does not involve anything funny and should be compatible with SDE. (The future unwind testing is unlikely to be compatible.) This CL adds support for x86_64 SysV and Win64 ABIs. ARM, AArch64, and x86 can be added in the future. The testing is injected in two places. First, all the assembly tests in p256-x86_64-test.cc are now instrumented. This is the intended workflow and should capture all registers. However, we currently do not unit-test our assembly much directly. We should do that as follow-up work[0] but, in the meantime, I've also wrapped all of the GTest main function in an ABI test. This is imperfect as ABI failures may be masked by other stack frames, but it costs nothing[1] and is pretty reliable at catching Win64 xmm register failures. [0] An alternate strategy would be, in debug builds, unconditionally instrument every assembly call in libcrypto. But the CHECK_ABI macro would be difficult to replicate in pure C, and unwind testing may be too invasive for this. Still, something to consider when we C++ libcrypto. [1] When single-stepped unwind testing exists, it won't cost nothing. The gtest_main.cc call will turn unwind testing off. Change-Id: I6643b26445891fd46abfacac52bc024024c8d7f6 Reviewed-on: https://boringssl-review.googlesource.com/c/33764 Reviewed-by: Adam Langley <agl@google.com> Reviewed-by: Adam Langley <alangley@gmail.com> Commit-Queue: David Benjamin <davidben@google.com>
2018-12-16 00:58:43 +00:00
Add a CFI tester to CHECK_ABI. This uses the x86 trap flag and libunwind to test CFI works at each instruction. For now, it just uses the system one out of pkg-config and disables unwind tests if unavailable. We'll probably want to stick a copy into //third_party and perhaps try the LLVM one later. This tester caught two bugs in P-256 CFI annotations already: I47b5f9798b3bcee1748e537b21c173d312a14b42 and I9f576d868850312d6c14d1386f8fbfa85021b347 An earlier design used PTRACE_SINGLESTEP with libunwind's remote unwinding features. ptrace is a mess around stop signals (see group-stop discussion in ptrace(2)) and this is 10x faster, so I went with it. The question of which is more future-proof is complex: - There are two libunwinds with the same API, https://www.nongnu.org/libunwind/ and LLVM's. This currently uses the system nongnu.org for convenience. In future, LLVM's should be easier to bundle (less complex build) and appears to even support Windows, but I haven't tested this. Moreover, setting the trap flag keeps the test single-process, which is less complex on Windows. That suggests the trap flag design and switching to LLVM later. However... - Not all architectures have a trap flag settable by userspace. As far as I can tell, ARMv8's PSTATE.SS can only be set from the kernel. If we stick with nongnu.org libunwind, we can use PTRACE_SINGLESTEP and remote unwinding. Or we implement it for LLVM. Another thought is for the ptracer to bounce SIGTRAP back into the process, to share the local unwinding code. - ARMv7 has no trap flag at all and PTRACE_SINGLESTEP fails. Debuggers single-step by injecting breakpoints instead. However, ARMv8's trap flag seems to work in both AArch32 and AArch64 modes, so we may be able to condition it on a 64-bit kernel. Sadly, neither strategy works with Intel SDE. Adding flags to cpucap vectors as we do with ARM would help, but it would not emulate CPUs newer than the host CPU. For now, I've just had SDE tests disable these. Annoyingly, CMake does not allow object libraries to have dependencies, so make test_support a proper static library. Rename the target to test_support_lib to avoid https://gitlab.kitware.com/cmake/cmake/issues/17785 Update-Note: This adds a new optional test dependency, but it's disabled by default (define BORINGSSL_HAVE_LIBUNWIND), so consumers do not need to do anything. We'll probably want to adjust this in the future. Bug: 181 Change-Id: I817263d7907aff0904a9cee83f8b26747262cc0c Reviewed-on: https://boringssl-review.googlesource.com/c/33966 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2018-12-21 23:58:36 +00:00
test_function_ok = false;
Implement ABI testing for aarch64. This caught a bug in bn_mul_mont. Tested manually on iOS and Android. Change-Id: I1819fcd9ad34dbe3ba92bba952507d86dd12185a Reviewed-on: https://boringssl-review.googlesource.com/c/34805 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2019-02-02 00:25:07 +00:00
EXPECT_EQ(42, CHECK_ABI_SEH(TestFunction, 1, 2, 3, 4, 5, 6, 7, 8));
Add a CFI tester to CHECK_ABI. This uses the x86 trap flag and libunwind to test CFI works at each instruction. For now, it just uses the system one out of pkg-config and disables unwind tests if unavailable. We'll probably want to stick a copy into //third_party and perhaps try the LLVM one later. This tester caught two bugs in P-256 CFI annotations already: I47b5f9798b3bcee1748e537b21c173d312a14b42 and I9f576d868850312d6c14d1386f8fbfa85021b347 An earlier design used PTRACE_SINGLESTEP with libunwind's remote unwinding features. ptrace is a mess around stop signals (see group-stop discussion in ptrace(2)) and this is 10x faster, so I went with it. The question of which is more future-proof is complex: - There are two libunwinds with the same API, https://www.nongnu.org/libunwind/ and LLVM's. This currently uses the system nongnu.org for convenience. In future, LLVM's should be easier to bundle (less complex build) and appears to even support Windows, but I haven't tested this. Moreover, setting the trap flag keeps the test single-process, which is less complex on Windows. That suggests the trap flag design and switching to LLVM later. However... - Not all architectures have a trap flag settable by userspace. As far as I can tell, ARMv8's PSTATE.SS can only be set from the kernel. If we stick with nongnu.org libunwind, we can use PTRACE_SINGLESTEP and remote unwinding. Or we implement it for LLVM. Another thought is for the ptracer to bounce SIGTRAP back into the process, to share the local unwinding code. - ARMv7 has no trap flag at all and PTRACE_SINGLESTEP fails. Debuggers single-step by injecting breakpoints instead. However, ARMv8's trap flag seems to work in both AArch32 and AArch64 modes, so we may be able to condition it on a 64-bit kernel. Sadly, neither strategy works with Intel SDE. Adding flags to cpucap vectors as we do with ARM would help, but it would not emulate CPUs newer than the host CPU. For now, I've just had SDE tests disable these. Annoyingly, CMake does not allow object libraries to have dependencies, so make test_support a proper static library. Rename the target to test_support_lib to avoid https://gitlab.kitware.com/cmake/cmake/issues/17785 Update-Note: This adds a new optional test dependency, but it's disabled by default (define BORINGSSL_HAVE_LIBUNWIND), so consumers do not need to do anything. We'll probably want to adjust this in the future. Bug: 181 Change-Id: I817263d7907aff0904a9cee83f8b26747262cc0c Reviewed-on: https://boringssl-review.googlesource.com/c/33966 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2018-12-21 23:58:36 +00:00
EXPECT_TRUE(test_function_ok);
Add an ABI testing framework. Dear reader, I must apologize in advance. This CL contains the following: - A new 256-line perlasm file with non-trivial perl bits and a dual-ABI variadic function caller. - C preprocessor gymnastics, with variadic macros and fun facts about __VA_ARGS__'s behavior on empty argument lists. - C++ template gymnastics, including variadic arguments, template specialization, std::enable_if, and machinery to control template argument deduction. Enjoy. This tests that our assembly functions correctly honor platform ABI conventions. Right now this only tests callee-saved registers, but it should be extendable to SEH/CFI unwind testing with single-step debugging APIs. Register-checking does not involve anything funny and should be compatible with SDE. (The future unwind testing is unlikely to be compatible.) This CL adds support for x86_64 SysV and Win64 ABIs. ARM, AArch64, and x86 can be added in the future. The testing is injected in two places. First, all the assembly tests in p256-x86_64-test.cc are now instrumented. This is the intended workflow and should capture all registers. However, we currently do not unit-test our assembly much directly. We should do that as follow-up work[0] but, in the meantime, I've also wrapped all of the GTest main function in an ABI test. This is imperfect as ABI failures may be masked by other stack frames, but it costs nothing[1] and is pretty reliable at catching Win64 xmm register failures. [0] An alternate strategy would be, in debug builds, unconditionally instrument every assembly call in libcrypto. But the CHECK_ABI macro would be difficult to replicate in pure C, and unwind testing may be too invasive for this. Still, something to consider when we C++ libcrypto. [1] When single-stepped unwind testing exists, it won't cost nothing. The gtest_main.cc call will turn unwind testing off. Change-Id: I6643b26445891fd46abfacac52bc024024c8d7f6 Reviewed-on: https://boringssl-review.googlesource.com/c/33764 Reviewed-by: Adam Langley <agl@google.com> Reviewed-by: Adam Langley <alangley@gmail.com> Commit-Queue: David Benjamin <davidben@google.com>
2018-12-16 00:58:43 +00:00
#if defined(SUPPORTS_ABI_TEST)
abi_test::internal::CallerState state;
RAND_bytes(reinterpret_cast<uint8_t *>(&state), sizeof(state));
Avoid unwind tests on libc functions. When built under UBSan, it gets confused inside a PLT stub. Change-Id: Ib082ecc076ba2111337ff5921e465e4beb99aab5 Reviewed-on: https://boringssl-review.googlesource.com/c/34448 Reviewed-by: Adam Langley <agl@google.com>
2019-01-22 03:42:25 +00:00
crypto_word_t argv[] = {
Implement ABI testing for aarch64. This caught a bug in bn_mul_mont. Tested manually on iOS and Android. Change-Id: I1819fcd9ad34dbe3ba92bba952507d86dd12185a Reviewed-on: https://boringssl-review.googlesource.com/c/34805 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2019-02-02 00:25:07 +00:00
1, 2, 3, 4, 5, 6, 7, 8,
Add an ABI testing framework. Dear reader, I must apologize in advance. This CL contains the following: - A new 256-line perlasm file with non-trivial perl bits and a dual-ABI variadic function caller. - C preprocessor gymnastics, with variadic macros and fun facts about __VA_ARGS__'s behavior on empty argument lists. - C++ template gymnastics, including variadic arguments, template specialization, std::enable_if, and machinery to control template argument deduction. Enjoy. This tests that our assembly functions correctly honor platform ABI conventions. Right now this only tests callee-saved registers, but it should be extendable to SEH/CFI unwind testing with single-step debugging APIs. Register-checking does not involve anything funny and should be compatible with SDE. (The future unwind testing is unlikely to be compatible.) This CL adds support for x86_64 SysV and Win64 ABIs. ARM, AArch64, and x86 can be added in the future. The testing is injected in two places. First, all the assembly tests in p256-x86_64-test.cc are now instrumented. This is the intended workflow and should capture all registers. However, we currently do not unit-test our assembly much directly. We should do that as follow-up work[0] but, in the meantime, I've also wrapped all of the GTest main function in an ABI test. This is imperfect as ABI failures may be masked by other stack frames, but it costs nothing[1] and is pretty reliable at catching Win64 xmm register failures. [0] An alternate strategy would be, in debug builds, unconditionally instrument every assembly call in libcrypto. But the CHECK_ABI macro would be difficult to replicate in pure C, and unwind testing may be too invasive for this. Still, something to consider when we C++ libcrypto. [1] When single-stepped unwind testing exists, it won't cost nothing. The gtest_main.cc call will turn unwind testing off. Change-Id: I6643b26445891fd46abfacac52bc024024c8d7f6 Reviewed-on: https://boringssl-review.googlesource.com/c/33764 Reviewed-by: Adam Langley <agl@google.com> Reviewed-by: Adam Langley <alangley@gmail.com> Commit-Queue: David Benjamin <davidben@google.com>
2018-12-16 00:58:43 +00:00
};
Implement unwind testing for Windows. Unfortunately, due to most OpenSSL assembly using custom exception handlers to unwind, most of our assembly doesn't work with non-destructive unwind. For now, CHECK_ABI behaves like CHECK_ABI_NO_UNWIND on Windows, and CHECK_ABI_SEH will test unwinding on both platforms. The tests do, however, work with the unwind-code-based assembly we recently added, as well as the clmul-based GHASH which is also code-based. Remove the ad-hoc SEH tests which intentionally hit memory access exceptions, now that we can test unwind directly. Now that we can test it, the next step is to implement SEH directives in perlasm so writing these unwind codes is less of a chore. Bug: 259 Change-Id: I23a57a22c5dc9fa4513f575f18192335779678a5 Reviewed-on: https://boringssl-review.googlesource.com/c/34784 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2019-02-03 03:29:58 +00:00
CHECK_ABI_SEH(abi_test_trampoline,
Implement ABI testing for aarch64. This caught a bug in bn_mul_mont. Tested manually on iOS and Android. Change-Id: I1819fcd9ad34dbe3ba92bba952507d86dd12185a Reviewed-on: https://boringssl-review.googlesource.com/c/34805 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2019-02-02 00:25:07 +00:00
reinterpret_cast<crypto_word_t>(TestFunction), &state, argv, 8,
Implement unwind testing for Windows. Unfortunately, due to most OpenSSL assembly using custom exception handlers to unwind, most of our assembly doesn't work with non-destructive unwind. For now, CHECK_ABI behaves like CHECK_ABI_NO_UNWIND on Windows, and CHECK_ABI_SEH will test unwinding on both platforms. The tests do, however, work with the unwind-code-based assembly we recently added, as well as the clmul-based GHASH which is also code-based. Remove the ad-hoc SEH tests which intentionally hit memory access exceptions, now that we can test unwind directly. Now that we can test it, the next step is to implement SEH directives in perlasm so writing these unwind codes is less of a chore. Bug: 259 Change-Id: I23a57a22c5dc9fa4513f575f18192335779678a5 Reviewed-on: https://boringssl-review.googlesource.com/c/34784 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2019-02-03 03:29:58 +00:00
0 /* no breakpoint */);
Add a CFI tester to CHECK_ABI. This uses the x86 trap flag and libunwind to test CFI works at each instruction. For now, it just uses the system one out of pkg-config and disables unwind tests if unavailable. We'll probably want to stick a copy into //third_party and perhaps try the LLVM one later. This tester caught two bugs in P-256 CFI annotations already: I47b5f9798b3bcee1748e537b21c173d312a14b42 and I9f576d868850312d6c14d1386f8fbfa85021b347 An earlier design used PTRACE_SINGLESTEP with libunwind's remote unwinding features. ptrace is a mess around stop signals (see group-stop discussion in ptrace(2)) and this is 10x faster, so I went with it. The question of which is more future-proof is complex: - There are two libunwinds with the same API, https://www.nongnu.org/libunwind/ and LLVM's. This currently uses the system nongnu.org for convenience. In future, LLVM's should be easier to bundle (less complex build) and appears to even support Windows, but I haven't tested this. Moreover, setting the trap flag keeps the test single-process, which is less complex on Windows. That suggests the trap flag design and switching to LLVM later. However... - Not all architectures have a trap flag settable by userspace. As far as I can tell, ARMv8's PSTATE.SS can only be set from the kernel. If we stick with nongnu.org libunwind, we can use PTRACE_SINGLESTEP and remote unwinding. Or we implement it for LLVM. Another thought is for the ptracer to bounce SIGTRAP back into the process, to share the local unwinding code. - ARMv7 has no trap flag at all and PTRACE_SINGLESTEP fails. Debuggers single-step by injecting breakpoints instead. However, ARMv8's trap flag seems to work in both AArch32 and AArch64 modes, so we may be able to condition it on a 64-bit kernel. Sadly, neither strategy works with Intel SDE. Adding flags to cpucap vectors as we do with ARM would help, but it would not emulate CPUs newer than the host CPU. For now, I've just had SDE tests disable these. Annoyingly, CMake does not allow object libraries to have dependencies, so make test_support a proper static library. Rename the target to test_support_lib to avoid https://gitlab.kitware.com/cmake/cmake/issues/17785 Update-Note: This adds a new optional test dependency, but it's disabled by default (define BORINGSSL_HAVE_LIBUNWIND), so consumers do not need to do anything. We'll probably want to adjust this in the future. Bug: 181 Change-Id: I817263d7907aff0904a9cee83f8b26747262cc0c Reviewed-on: https://boringssl-review.googlesource.com/c/33966 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2018-12-21 23:58:36 +00:00
Add ABI testing for 32-bit x86. This is much less interesting (stack-based parameters, Windows and SysV match, no SEH concerns as far as I can tell) than x86_64, but it was easy to do and I'm more familiar with x86 than ARM, so it made a better second architecture to make sure all the architecture ifdefs worked out. Also fix a bug in the x86_64 direction flag code. It was shifting in the wrong direction, making give 0 or 1<<20 rather than 0 or 1. (Happily, x86_64 appears to be unique in having vastly different calling conventions between OSs. x86 is the same between SysV and Windows, and ARM had the good sense to specify a (mostly) common set of rules.) Since a lot of the assembly functions use the same names and the tests were written generically, merely dropping in a trampoline and CallerState implementation gives us a bunch of ABI tests for free. Change-Id: I15408c18d43e88cfa1c5c0634a8b268a150ed961 Reviewed-on: https://boringssl-review.googlesource.com/c/34624 Reviewed-by: Adam Langley <agl@google.com> Commit-Queue: David Benjamin <davidben@google.com>
2019-01-26 17:29:08 +00:00
#if defined(OPENSSL_X86_64)
Add a CFI tester to CHECK_ABI. This uses the x86 trap flag and libunwind to test CFI works at each instruction. For now, it just uses the system one out of pkg-config and disables unwind tests if unavailable. We'll probably want to stick a copy into //third_party and perhaps try the LLVM one later. This tester caught two bugs in P-256 CFI annotations already: I47b5f9798b3bcee1748e537b21c173d312a14b42 and I9f576d868850312d6c14d1386f8fbfa85021b347 An earlier design used PTRACE_SINGLESTEP with libunwind's remote unwinding features. ptrace is a mess around stop signals (see group-stop discussion in ptrace(2)) and this is 10x faster, so I went with it. The question of which is more future-proof is complex: - There are two libunwinds with the same API, https://www.nongnu.org/libunwind/ and LLVM's. This currently uses the system nongnu.org for convenience. In future, LLVM's should be easier to bundle (less complex build) and appears to even support Windows, but I haven't tested this. Moreover, setting the trap flag keeps the test single-process, which is less complex on Windows. That suggests the trap flag design and switching to LLVM later. However... - Not all architectures have a trap flag settable by userspace. As far as I can tell, ARMv8's PSTATE.SS can only be set from the kernel. If we stick with nongnu.org libunwind, we can use PTRACE_SINGLESTEP and remote unwinding. Or we implement it for LLVM. Another thought is for the ptracer to bounce SIGTRAP back into the process, to share the local unwinding code. - ARMv7 has no trap flag at all and PTRACE_SINGLESTEP fails. Debuggers single-step by injecting breakpoints instead. However, ARMv8's trap flag seems to work in both AArch32 and AArch64 modes, so we may be able to condition it on a 64-bit kernel. Sadly, neither strategy works with Intel SDE. Adding flags to cpucap vectors as we do with ARM would help, but it would not emulate CPUs newer than the host CPU. For now, I've just had SDE tests disable these. Annoyingly, CMake does not allow object libraries to have dependencies, so make test_support a proper static library. Rename the target to test_support_lib to avoid https://gitlab.kitware.com/cmake/cmake/issues/17785 Update-Note: This adds a new optional test dependency, but it's disabled by default (define BORINGSSL_HAVE_LIBUNWIND), so consumers do not need to do anything. We'll probably want to adjust this in the future. Bug: 181 Change-Id: I817263d7907aff0904a9cee83f8b26747262cc0c Reviewed-on: https://boringssl-review.googlesource.com/c/33966 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2018-12-21 23:58:36 +00:00
if (abi_test::UnwindTestsEnabled()) {
Implement unwind testing for Windows. Unfortunately, due to most OpenSSL assembly using custom exception handlers to unwind, most of our assembly doesn't work with non-destructive unwind. For now, CHECK_ABI behaves like CHECK_ABI_NO_UNWIND on Windows, and CHECK_ABI_SEH will test unwinding on both platforms. The tests do, however, work with the unwind-code-based assembly we recently added, as well as the clmul-based GHASH which is also code-based. Remove the ad-hoc SEH tests which intentionally hit memory access exceptions, now that we can test unwind directly. Now that we can test it, the next step is to implement SEH directives in perlasm so writing these unwind codes is less of a chore. Bug: 259 Change-Id: I23a57a22c5dc9fa4513f575f18192335779678a5 Reviewed-on: https://boringssl-review.googlesource.com/c/34784 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2019-02-03 03:29:58 +00:00
EXPECT_NONFATAL_FAILURE(CHECK_ABI_SEH(abi_test_bad_unwind_wrong_register),
Use Windows symbol APIs in the unwind tester. This should make things a bit easier to debug. Update-Note: Test binaries on Windows now link to dbghelp. Bug: 259 Change-Id: I9da1fc89d429080c5250238e4341445922b1dd8e Reviewed-on: https://boringssl-review.googlesource.com/c/34868 Reviewed-by: Adam Langley <agl@google.com> Commit-Queue: David Benjamin <davidben@google.com>
2019-02-10 07:04:01 +00:00
"was not recovered");
Implement unwind testing for Windows. Unfortunately, due to most OpenSSL assembly using custom exception handlers to unwind, most of our assembly doesn't work with non-destructive unwind. For now, CHECK_ABI behaves like CHECK_ABI_NO_UNWIND on Windows, and CHECK_ABI_SEH will test unwinding on both platforms. The tests do, however, work with the unwind-code-based assembly we recently added, as well as the clmul-based GHASH which is also code-based. Remove the ad-hoc SEH tests which intentionally hit memory access exceptions, now that we can test unwind directly. Now that we can test it, the next step is to implement SEH directives in perlasm so writing these unwind codes is less of a chore. Bug: 259 Change-Id: I23a57a22c5dc9fa4513f575f18192335779678a5 Reviewed-on: https://boringssl-review.googlesource.com/c/34784 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2019-02-03 03:29:58 +00:00
EXPECT_NONFATAL_FAILURE(CHECK_ABI_SEH(abi_test_bad_unwind_temporary),
Use Windows symbol APIs in the unwind tester. This should make things a bit easier to debug. Update-Note: Test binaries on Windows now link to dbghelp. Bug: 259 Change-Id: I9da1fc89d429080c5250238e4341445922b1dd8e Reviewed-on: https://boringssl-review.googlesource.com/c/34868 Reviewed-by: Adam Langley <agl@google.com> Commit-Queue: David Benjamin <davidben@google.com>
2019-02-10 07:04:01 +00:00
"was not recovered");
Add a CFI tester to CHECK_ABI. This uses the x86 trap flag and libunwind to test CFI works at each instruction. For now, it just uses the system one out of pkg-config and disables unwind tests if unavailable. We'll probably want to stick a copy into //third_party and perhaps try the LLVM one later. This tester caught two bugs in P-256 CFI annotations already: I47b5f9798b3bcee1748e537b21c173d312a14b42 and I9f576d868850312d6c14d1386f8fbfa85021b347 An earlier design used PTRACE_SINGLESTEP with libunwind's remote unwinding features. ptrace is a mess around stop signals (see group-stop discussion in ptrace(2)) and this is 10x faster, so I went with it. The question of which is more future-proof is complex: - There are two libunwinds with the same API, https://www.nongnu.org/libunwind/ and LLVM's. This currently uses the system nongnu.org for convenience. In future, LLVM's should be easier to bundle (less complex build) and appears to even support Windows, but I haven't tested this. Moreover, setting the trap flag keeps the test single-process, which is less complex on Windows. That suggests the trap flag design and switching to LLVM later. However... - Not all architectures have a trap flag settable by userspace. As far as I can tell, ARMv8's PSTATE.SS can only be set from the kernel. If we stick with nongnu.org libunwind, we can use PTRACE_SINGLESTEP and remote unwinding. Or we implement it for LLVM. Another thought is for the ptracer to bounce SIGTRAP back into the process, to share the local unwinding code. - ARMv7 has no trap flag at all and PTRACE_SINGLESTEP fails. Debuggers single-step by injecting breakpoints instead. However, ARMv8's trap flag seems to work in both AArch32 and AArch64 modes, so we may be able to condition it on a 64-bit kernel. Sadly, neither strategy works with Intel SDE. Adding flags to cpucap vectors as we do with ARM would help, but it would not emulate CPUs newer than the host CPU. For now, I've just had SDE tests disable these. Annoyingly, CMake does not allow object libraries to have dependencies, so make test_support a proper static library. Rename the target to test_support_lib to avoid https://gitlab.kitware.com/cmake/cmake/issues/17785 Update-Note: This adds a new optional test dependency, but it's disabled by default (define BORINGSSL_HAVE_LIBUNWIND), so consumers do not need to do anything. We'll probably want to adjust this in the future. Bug: 181 Change-Id: I817263d7907aff0904a9cee83f8b26747262cc0c Reviewed-on: https://boringssl-review.googlesource.com/c/33966 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2018-12-21 23:58:36 +00:00
CHECK_ABI_NO_UNWIND(abi_test_bad_unwind_wrong_register);
CHECK_ABI_NO_UNWIND(abi_test_bad_unwind_temporary);
Implement unwind testing for Windows. Unfortunately, due to most OpenSSL assembly using custom exception handlers to unwind, most of our assembly doesn't work with non-destructive unwind. For now, CHECK_ABI behaves like CHECK_ABI_NO_UNWIND on Windows, and CHECK_ABI_SEH will test unwinding on both platforms. The tests do, however, work with the unwind-code-based assembly we recently added, as well as the clmul-based GHASH which is also code-based. Remove the ad-hoc SEH tests which intentionally hit memory access exceptions, now that we can test unwind directly. Now that we can test it, the next step is to implement SEH directives in perlasm so writing these unwind codes is less of a chore. Bug: 259 Change-Id: I23a57a22c5dc9fa4513f575f18192335779678a5 Reviewed-on: https://boringssl-review.googlesource.com/c/34784 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2019-02-03 03:29:58 +00:00
#if defined(OPENSSL_WINDOWS)
// The invalid epilog makes Windows believe the epilog starts later than it
// actually does. As a result, immediately after the popq, it does not
// realize the stack has been unwound and repeats the work.
EXPECT_NONFATAL_FAILURE(CHECK_ABI_SEH(abi_test_bad_unwind_epilog),
"unwound past starting frame");
CHECK_ABI_NO_UNWIND(abi_test_bad_unwind_epilog);
#endif // OPENSSL_WINDOWS
Add a CFI tester to CHECK_ABI. This uses the x86 trap flag and libunwind to test CFI works at each instruction. For now, it just uses the system one out of pkg-config and disables unwind tests if unavailable. We'll probably want to stick a copy into //third_party and perhaps try the LLVM one later. This tester caught two bugs in P-256 CFI annotations already: I47b5f9798b3bcee1748e537b21c173d312a14b42 and I9f576d868850312d6c14d1386f8fbfa85021b347 An earlier design used PTRACE_SINGLESTEP with libunwind's remote unwinding features. ptrace is a mess around stop signals (see group-stop discussion in ptrace(2)) and this is 10x faster, so I went with it. The question of which is more future-proof is complex: - There are two libunwinds with the same API, https://www.nongnu.org/libunwind/ and LLVM's. This currently uses the system nongnu.org for convenience. In future, LLVM's should be easier to bundle (less complex build) and appears to even support Windows, but I haven't tested this. Moreover, setting the trap flag keeps the test single-process, which is less complex on Windows. That suggests the trap flag design and switching to LLVM later. However... - Not all architectures have a trap flag settable by userspace. As far as I can tell, ARMv8's PSTATE.SS can only be set from the kernel. If we stick with nongnu.org libunwind, we can use PTRACE_SINGLESTEP and remote unwinding. Or we implement it for LLVM. Another thought is for the ptracer to bounce SIGTRAP back into the process, to share the local unwinding code. - ARMv7 has no trap flag at all and PTRACE_SINGLESTEP fails. Debuggers single-step by injecting breakpoints instead. However, ARMv8's trap flag seems to work in both AArch32 and AArch64 modes, so we may be able to condition it on a 64-bit kernel. Sadly, neither strategy works with Intel SDE. Adding flags to cpucap vectors as we do with ARM would help, but it would not emulate CPUs newer than the host CPU. For now, I've just had SDE tests disable these. Annoyingly, CMake does not allow object libraries to have dependencies, so make test_support a proper static library. Rename the target to test_support_lib to avoid https://gitlab.kitware.com/cmake/cmake/issues/17785 Update-Note: This adds a new optional test dependency, but it's disabled by default (define BORINGSSL_HAVE_LIBUNWIND), so consumers do not need to do anything. We'll probably want to adjust this in the future. Bug: 181 Change-Id: I817263d7907aff0904a9cee83f8b26747262cc0c Reviewed-on: https://boringssl-review.googlesource.com/c/33966 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2018-12-21 23:58:36 +00:00
}
Add ABI testing for 32-bit x86. This is much less interesting (stack-based parameters, Windows and SysV match, no SEH concerns as far as I can tell) than x86_64, but it was easy to do and I'm more familiar with x86 than ARM, so it made a better second architecture to make sure all the architecture ifdefs worked out. Also fix a bug in the x86_64 direction flag code. It was shifting in the wrong direction, making give 0 or 1<<20 rather than 0 or 1. (Happily, x86_64 appears to be unique in having vastly different calling conventions between OSs. x86 is the same between SysV and Windows, and ARM had the good sense to specify a (mostly) common set of rules.) Since a lot of the assembly functions use the same names and the tests were written generically, merely dropping in a trampoline and CallerState implementation gives us a bunch of ABI tests for free. Change-Id: I15408c18d43e88cfa1c5c0634a8b268a150ed961 Reviewed-on: https://boringssl-review.googlesource.com/c/34624 Reviewed-by: Adam Langley <agl@google.com> Commit-Queue: David Benjamin <davidben@google.com>
2019-01-26 17:29:08 +00:00
#endif // OPENSSL_X86_64
Add an ABI testing framework. Dear reader, I must apologize in advance. This CL contains the following: - A new 256-line perlasm file with non-trivial perl bits and a dual-ABI variadic function caller. - C preprocessor gymnastics, with variadic macros and fun facts about __VA_ARGS__'s behavior on empty argument lists. - C++ template gymnastics, including variadic arguments, template specialization, std::enable_if, and machinery to control template argument deduction. Enjoy. This tests that our assembly functions correctly honor platform ABI conventions. Right now this only tests callee-saved registers, but it should be extendable to SEH/CFI unwind testing with single-step debugging APIs. Register-checking does not involve anything funny and should be compatible with SDE. (The future unwind testing is unlikely to be compatible.) This CL adds support for x86_64 SysV and Win64 ABIs. ARM, AArch64, and x86 can be added in the future. The testing is injected in two places. First, all the assembly tests in p256-x86_64-test.cc are now instrumented. This is the intended workflow and should capture all registers. However, we currently do not unit-test our assembly much directly. We should do that as follow-up work[0] but, in the meantime, I've also wrapped all of the GTest main function in an ABI test. This is imperfect as ABI failures may be masked by other stack frames, but it costs nothing[1] and is pretty reliable at catching Win64 xmm register failures. [0] An alternate strategy would be, in debug builds, unconditionally instrument every assembly call in libcrypto. But the CHECK_ABI macro would be difficult to replicate in pure C, and unwind testing may be too invasive for this. Still, something to consider when we C++ libcrypto. [1] When single-stepped unwind testing exists, it won't cost nothing. The gtest_main.cc call will turn unwind testing off. Change-Id: I6643b26445891fd46abfacac52bc024024c8d7f6 Reviewed-on: https://boringssl-review.googlesource.com/c/33764 Reviewed-by: Adam Langley <agl@google.com> Reviewed-by: Adam Langley <alangley@gmail.com> Commit-Queue: David Benjamin <davidben@google.com>
2018-12-16 00:58:43 +00:00
#endif // SUPPORTS_ABI_TEST
}
#if defined(OPENSSL_X86_64) && defined(SUPPORTS_ABI_TEST)
extern "C" {
void abi_test_clobber_rax(void);
void abi_test_clobber_rbx(void);
void abi_test_clobber_rcx(void);
void abi_test_clobber_rdx(void);
void abi_test_clobber_rsi(void);
void abi_test_clobber_rdi(void);
void abi_test_clobber_rbp(void);
void abi_test_clobber_r8(void);
void abi_test_clobber_r9(void);
void abi_test_clobber_r10(void);
void abi_test_clobber_r11(void);
void abi_test_clobber_r12(void);
void abi_test_clobber_r13(void);
void abi_test_clobber_r14(void);
void abi_test_clobber_r15(void);
void abi_test_clobber_xmm0(void);
void abi_test_clobber_xmm1(void);
void abi_test_clobber_xmm2(void);
void abi_test_clobber_xmm3(void);
void abi_test_clobber_xmm4(void);
void abi_test_clobber_xmm5(void);
void abi_test_clobber_xmm6(void);
void abi_test_clobber_xmm7(void);
void abi_test_clobber_xmm8(void);
void abi_test_clobber_xmm9(void);
void abi_test_clobber_xmm10(void);
void abi_test_clobber_xmm11(void);
void abi_test_clobber_xmm12(void);
void abi_test_clobber_xmm13(void);
void abi_test_clobber_xmm14(void);
void abi_test_clobber_xmm15(void);
} // extern "C"
TEST(ABITest, X86_64) {
// abi_test_trampoline hides unsaved registers from the caller, so we can
// safely call the abi_test_clobber_* functions below.
abi_test::internal::CallerState state;
RAND_bytes(reinterpret_cast<uint8_t *>(&state), sizeof(state));
Add a CFI tester to CHECK_ABI. This uses the x86 trap flag and libunwind to test CFI works at each instruction. For now, it just uses the system one out of pkg-config and disables unwind tests if unavailable. We'll probably want to stick a copy into //third_party and perhaps try the LLVM one later. This tester caught two bugs in P-256 CFI annotations already: I47b5f9798b3bcee1748e537b21c173d312a14b42 and I9f576d868850312d6c14d1386f8fbfa85021b347 An earlier design used PTRACE_SINGLESTEP with libunwind's remote unwinding features. ptrace is a mess around stop signals (see group-stop discussion in ptrace(2)) and this is 10x faster, so I went with it. The question of which is more future-proof is complex: - There are two libunwinds with the same API, https://www.nongnu.org/libunwind/ and LLVM's. This currently uses the system nongnu.org for convenience. In future, LLVM's should be easier to bundle (less complex build) and appears to even support Windows, but I haven't tested this. Moreover, setting the trap flag keeps the test single-process, which is less complex on Windows. That suggests the trap flag design and switching to LLVM later. However... - Not all architectures have a trap flag settable by userspace. As far as I can tell, ARMv8's PSTATE.SS can only be set from the kernel. If we stick with nongnu.org libunwind, we can use PTRACE_SINGLESTEP and remote unwinding. Or we implement it for LLVM. Another thought is for the ptracer to bounce SIGTRAP back into the process, to share the local unwinding code. - ARMv7 has no trap flag at all and PTRACE_SINGLESTEP fails. Debuggers single-step by injecting breakpoints instead. However, ARMv8's trap flag seems to work in both AArch32 and AArch64 modes, so we may be able to condition it on a 64-bit kernel. Sadly, neither strategy works with Intel SDE. Adding flags to cpucap vectors as we do with ARM would help, but it would not emulate CPUs newer than the host CPU. For now, I've just had SDE tests disable these. Annoyingly, CMake does not allow object libraries to have dependencies, so make test_support a proper static library. Rename the target to test_support_lib to avoid https://gitlab.kitware.com/cmake/cmake/issues/17785 Update-Note: This adds a new optional test dependency, but it's disabled by default (define BORINGSSL_HAVE_LIBUNWIND), so consumers do not need to do anything. We'll probably want to adjust this in the future. Bug: 181 Change-Id: I817263d7907aff0904a9cee83f8b26747262cc0c Reviewed-on: https://boringssl-review.googlesource.com/c/33966 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2018-12-21 23:58:36 +00:00
CHECK_ABI_NO_UNWIND(abi_test_trampoline,
reinterpret_cast<crypto_word_t>(abi_test_clobber_rbx),
&state, nullptr, 0, 0 /* no breakpoint */);
CHECK_ABI_NO_UNWIND(abi_test_clobber_rax);
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_rbx),
"rbx was not restored after return");
CHECK_ABI_NO_UNWIND(abi_test_clobber_rcx);
CHECK_ABI_NO_UNWIND(abi_test_clobber_rdx);
Add an ABI testing framework. Dear reader, I must apologize in advance. This CL contains the following: - A new 256-line perlasm file with non-trivial perl bits and a dual-ABI variadic function caller. - C preprocessor gymnastics, with variadic macros and fun facts about __VA_ARGS__'s behavior on empty argument lists. - C++ template gymnastics, including variadic arguments, template specialization, std::enable_if, and machinery to control template argument deduction. Enjoy. This tests that our assembly functions correctly honor platform ABI conventions. Right now this only tests callee-saved registers, but it should be extendable to SEH/CFI unwind testing with single-step debugging APIs. Register-checking does not involve anything funny and should be compatible with SDE. (The future unwind testing is unlikely to be compatible.) This CL adds support for x86_64 SysV and Win64 ABIs. ARM, AArch64, and x86 can be added in the future. The testing is injected in two places. First, all the assembly tests in p256-x86_64-test.cc are now instrumented. This is the intended workflow and should capture all registers. However, we currently do not unit-test our assembly much directly. We should do that as follow-up work[0] but, in the meantime, I've also wrapped all of the GTest main function in an ABI test. This is imperfect as ABI failures may be masked by other stack frames, but it costs nothing[1] and is pretty reliable at catching Win64 xmm register failures. [0] An alternate strategy would be, in debug builds, unconditionally instrument every assembly call in libcrypto. But the CHECK_ABI macro would be difficult to replicate in pure C, and unwind testing may be too invasive for this. Still, something to consider when we C++ libcrypto. [1] When single-stepped unwind testing exists, it won't cost nothing. The gtest_main.cc call will turn unwind testing off. Change-Id: I6643b26445891fd46abfacac52bc024024c8d7f6 Reviewed-on: https://boringssl-review.googlesource.com/c/33764 Reviewed-by: Adam Langley <agl@google.com> Reviewed-by: Adam Langley <alangley@gmail.com> Commit-Queue: David Benjamin <davidben@google.com>
2018-12-16 00:58:43 +00:00
#if defined(OPENSSL_WINDOWS)
Add a CFI tester to CHECK_ABI. This uses the x86 trap flag and libunwind to test CFI works at each instruction. For now, it just uses the system one out of pkg-config and disables unwind tests if unavailable. We'll probably want to stick a copy into //third_party and perhaps try the LLVM one later. This tester caught two bugs in P-256 CFI annotations already: I47b5f9798b3bcee1748e537b21c173d312a14b42 and I9f576d868850312d6c14d1386f8fbfa85021b347 An earlier design used PTRACE_SINGLESTEP with libunwind's remote unwinding features. ptrace is a mess around stop signals (see group-stop discussion in ptrace(2)) and this is 10x faster, so I went with it. The question of which is more future-proof is complex: - There are two libunwinds with the same API, https://www.nongnu.org/libunwind/ and LLVM's. This currently uses the system nongnu.org for convenience. In future, LLVM's should be easier to bundle (less complex build) and appears to even support Windows, but I haven't tested this. Moreover, setting the trap flag keeps the test single-process, which is less complex on Windows. That suggests the trap flag design and switching to LLVM later. However... - Not all architectures have a trap flag settable by userspace. As far as I can tell, ARMv8's PSTATE.SS can only be set from the kernel. If we stick with nongnu.org libunwind, we can use PTRACE_SINGLESTEP and remote unwinding. Or we implement it for LLVM. Another thought is for the ptracer to bounce SIGTRAP back into the process, to share the local unwinding code. - ARMv7 has no trap flag at all and PTRACE_SINGLESTEP fails. Debuggers single-step by injecting breakpoints instead. However, ARMv8's trap flag seems to work in both AArch32 and AArch64 modes, so we may be able to condition it on a 64-bit kernel. Sadly, neither strategy works with Intel SDE. Adding flags to cpucap vectors as we do with ARM would help, but it would not emulate CPUs newer than the host CPU. For now, I've just had SDE tests disable these. Annoyingly, CMake does not allow object libraries to have dependencies, so make test_support a proper static library. Rename the target to test_support_lib to avoid https://gitlab.kitware.com/cmake/cmake/issues/17785 Update-Note: This adds a new optional test dependency, but it's disabled by default (define BORINGSSL_HAVE_LIBUNWIND), so consumers do not need to do anything. We'll probably want to adjust this in the future. Bug: 181 Change-Id: I817263d7907aff0904a9cee83f8b26747262cc0c Reviewed-on: https://boringssl-review.googlesource.com/c/33966 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2018-12-21 23:58:36 +00:00
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_rdi),
"rdi was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_rsi),
"rsi was not restored after return");
Add an ABI testing framework. Dear reader, I must apologize in advance. This CL contains the following: - A new 256-line perlasm file with non-trivial perl bits and a dual-ABI variadic function caller. - C preprocessor gymnastics, with variadic macros and fun facts about __VA_ARGS__'s behavior on empty argument lists. - C++ template gymnastics, including variadic arguments, template specialization, std::enable_if, and machinery to control template argument deduction. Enjoy. This tests that our assembly functions correctly honor platform ABI conventions. Right now this only tests callee-saved registers, but it should be extendable to SEH/CFI unwind testing with single-step debugging APIs. Register-checking does not involve anything funny and should be compatible with SDE. (The future unwind testing is unlikely to be compatible.) This CL adds support for x86_64 SysV and Win64 ABIs. ARM, AArch64, and x86 can be added in the future. The testing is injected in two places. First, all the assembly tests in p256-x86_64-test.cc are now instrumented. This is the intended workflow and should capture all registers. However, we currently do not unit-test our assembly much directly. We should do that as follow-up work[0] but, in the meantime, I've also wrapped all of the GTest main function in an ABI test. This is imperfect as ABI failures may be masked by other stack frames, but it costs nothing[1] and is pretty reliable at catching Win64 xmm register failures. [0] An alternate strategy would be, in debug builds, unconditionally instrument every assembly call in libcrypto. But the CHECK_ABI macro would be difficult to replicate in pure C, and unwind testing may be too invasive for this. Still, something to consider when we C++ libcrypto. [1] When single-stepped unwind testing exists, it won't cost nothing. The gtest_main.cc call will turn unwind testing off. Change-Id: I6643b26445891fd46abfacac52bc024024c8d7f6 Reviewed-on: https://boringssl-review.googlesource.com/c/33764 Reviewed-by: Adam Langley <agl@google.com> Reviewed-by: Adam Langley <alangley@gmail.com> Commit-Queue: David Benjamin <davidben@google.com>
2018-12-16 00:58:43 +00:00
#else
Add a CFI tester to CHECK_ABI. This uses the x86 trap flag and libunwind to test CFI works at each instruction. For now, it just uses the system one out of pkg-config and disables unwind tests if unavailable. We'll probably want to stick a copy into //third_party and perhaps try the LLVM one later. This tester caught two bugs in P-256 CFI annotations already: I47b5f9798b3bcee1748e537b21c173d312a14b42 and I9f576d868850312d6c14d1386f8fbfa85021b347 An earlier design used PTRACE_SINGLESTEP with libunwind's remote unwinding features. ptrace is a mess around stop signals (see group-stop discussion in ptrace(2)) and this is 10x faster, so I went with it. The question of which is more future-proof is complex: - There are two libunwinds with the same API, https://www.nongnu.org/libunwind/ and LLVM's. This currently uses the system nongnu.org for convenience. In future, LLVM's should be easier to bundle (less complex build) and appears to even support Windows, but I haven't tested this. Moreover, setting the trap flag keeps the test single-process, which is less complex on Windows. That suggests the trap flag design and switching to LLVM later. However... - Not all architectures have a trap flag settable by userspace. As far as I can tell, ARMv8's PSTATE.SS can only be set from the kernel. If we stick with nongnu.org libunwind, we can use PTRACE_SINGLESTEP and remote unwinding. Or we implement it for LLVM. Another thought is for the ptracer to bounce SIGTRAP back into the process, to share the local unwinding code. - ARMv7 has no trap flag at all and PTRACE_SINGLESTEP fails. Debuggers single-step by injecting breakpoints instead. However, ARMv8's trap flag seems to work in both AArch32 and AArch64 modes, so we may be able to condition it on a 64-bit kernel. Sadly, neither strategy works with Intel SDE. Adding flags to cpucap vectors as we do with ARM would help, but it would not emulate CPUs newer than the host CPU. For now, I've just had SDE tests disable these. Annoyingly, CMake does not allow object libraries to have dependencies, so make test_support a proper static library. Rename the target to test_support_lib to avoid https://gitlab.kitware.com/cmake/cmake/issues/17785 Update-Note: This adds a new optional test dependency, but it's disabled by default (define BORINGSSL_HAVE_LIBUNWIND), so consumers do not need to do anything. We'll probably want to adjust this in the future. Bug: 181 Change-Id: I817263d7907aff0904a9cee83f8b26747262cc0c Reviewed-on: https://boringssl-review.googlesource.com/c/33966 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2018-12-21 23:58:36 +00:00
CHECK_ABI_NO_UNWIND(abi_test_clobber_rdi);
CHECK_ABI_NO_UNWIND(abi_test_clobber_rsi);
Add an ABI testing framework. Dear reader, I must apologize in advance. This CL contains the following: - A new 256-line perlasm file with non-trivial perl bits and a dual-ABI variadic function caller. - C preprocessor gymnastics, with variadic macros and fun facts about __VA_ARGS__'s behavior on empty argument lists. - C++ template gymnastics, including variadic arguments, template specialization, std::enable_if, and machinery to control template argument deduction. Enjoy. This tests that our assembly functions correctly honor platform ABI conventions. Right now this only tests callee-saved registers, but it should be extendable to SEH/CFI unwind testing with single-step debugging APIs. Register-checking does not involve anything funny and should be compatible with SDE. (The future unwind testing is unlikely to be compatible.) This CL adds support for x86_64 SysV and Win64 ABIs. ARM, AArch64, and x86 can be added in the future. The testing is injected in two places. First, all the assembly tests in p256-x86_64-test.cc are now instrumented. This is the intended workflow and should capture all registers. However, we currently do not unit-test our assembly much directly. We should do that as follow-up work[0] but, in the meantime, I've also wrapped all of the GTest main function in an ABI test. This is imperfect as ABI failures may be masked by other stack frames, but it costs nothing[1] and is pretty reliable at catching Win64 xmm register failures. [0] An alternate strategy would be, in debug builds, unconditionally instrument every assembly call in libcrypto. But the CHECK_ABI macro would be difficult to replicate in pure C, and unwind testing may be too invasive for this. Still, something to consider when we C++ libcrypto. [1] When single-stepped unwind testing exists, it won't cost nothing. The gtest_main.cc call will turn unwind testing off. Change-Id: I6643b26445891fd46abfacac52bc024024c8d7f6 Reviewed-on: https://boringssl-review.googlesource.com/c/33764 Reviewed-by: Adam Langley <agl@google.com> Reviewed-by: Adam Langley <alangley@gmail.com> Commit-Queue: David Benjamin <davidben@google.com>
2018-12-16 00:58:43 +00:00
#endif
Add a CFI tester to CHECK_ABI. This uses the x86 trap flag and libunwind to test CFI works at each instruction. For now, it just uses the system one out of pkg-config and disables unwind tests if unavailable. We'll probably want to stick a copy into //third_party and perhaps try the LLVM one later. This tester caught two bugs in P-256 CFI annotations already: I47b5f9798b3bcee1748e537b21c173d312a14b42 and I9f576d868850312d6c14d1386f8fbfa85021b347 An earlier design used PTRACE_SINGLESTEP with libunwind's remote unwinding features. ptrace is a mess around stop signals (see group-stop discussion in ptrace(2)) and this is 10x faster, so I went with it. The question of which is more future-proof is complex: - There are two libunwinds with the same API, https://www.nongnu.org/libunwind/ and LLVM's. This currently uses the system nongnu.org for convenience. In future, LLVM's should be easier to bundle (less complex build) and appears to even support Windows, but I haven't tested this. Moreover, setting the trap flag keeps the test single-process, which is less complex on Windows. That suggests the trap flag design and switching to LLVM later. However... - Not all architectures have a trap flag settable by userspace. As far as I can tell, ARMv8's PSTATE.SS can only be set from the kernel. If we stick with nongnu.org libunwind, we can use PTRACE_SINGLESTEP and remote unwinding. Or we implement it for LLVM. Another thought is for the ptracer to bounce SIGTRAP back into the process, to share the local unwinding code. - ARMv7 has no trap flag at all and PTRACE_SINGLESTEP fails. Debuggers single-step by injecting breakpoints instead. However, ARMv8's trap flag seems to work in both AArch32 and AArch64 modes, so we may be able to condition it on a 64-bit kernel. Sadly, neither strategy works with Intel SDE. Adding flags to cpucap vectors as we do with ARM would help, but it would not emulate CPUs newer than the host CPU. For now, I've just had SDE tests disable these. Annoyingly, CMake does not allow object libraries to have dependencies, so make test_support a proper static library. Rename the target to test_support_lib to avoid https://gitlab.kitware.com/cmake/cmake/issues/17785 Update-Note: This adds a new optional test dependency, but it's disabled by default (define BORINGSSL_HAVE_LIBUNWIND), so consumers do not need to do anything. We'll probably want to adjust this in the future. Bug: 181 Change-Id: I817263d7907aff0904a9cee83f8b26747262cc0c Reviewed-on: https://boringssl-review.googlesource.com/c/33966 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2018-12-21 23:58:36 +00:00
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_rbp),
"rbp was not restored after return");
CHECK_ABI_NO_UNWIND(abi_test_clobber_r8);
CHECK_ABI_NO_UNWIND(abi_test_clobber_r9);
CHECK_ABI_NO_UNWIND(abi_test_clobber_r10);
CHECK_ABI_NO_UNWIND(abi_test_clobber_r11);
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_r12),
"r12 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_r13),
"r13 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_r14),
"r14 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_r15),
"r15 was not restored after return");
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm0);
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm1);
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm2);
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm3);
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm4);
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm5);
Add an ABI testing framework. Dear reader, I must apologize in advance. This CL contains the following: - A new 256-line perlasm file with non-trivial perl bits and a dual-ABI variadic function caller. - C preprocessor gymnastics, with variadic macros and fun facts about __VA_ARGS__'s behavior on empty argument lists. - C++ template gymnastics, including variadic arguments, template specialization, std::enable_if, and machinery to control template argument deduction. Enjoy. This tests that our assembly functions correctly honor platform ABI conventions. Right now this only tests callee-saved registers, but it should be extendable to SEH/CFI unwind testing with single-step debugging APIs. Register-checking does not involve anything funny and should be compatible with SDE. (The future unwind testing is unlikely to be compatible.) This CL adds support for x86_64 SysV and Win64 ABIs. ARM, AArch64, and x86 can be added in the future. The testing is injected in two places. First, all the assembly tests in p256-x86_64-test.cc are now instrumented. This is the intended workflow and should capture all registers. However, we currently do not unit-test our assembly much directly. We should do that as follow-up work[0] but, in the meantime, I've also wrapped all of the GTest main function in an ABI test. This is imperfect as ABI failures may be masked by other stack frames, but it costs nothing[1] and is pretty reliable at catching Win64 xmm register failures. [0] An alternate strategy would be, in debug builds, unconditionally instrument every assembly call in libcrypto. But the CHECK_ABI macro would be difficult to replicate in pure C, and unwind testing may be too invasive for this. Still, something to consider when we C++ libcrypto. [1] When single-stepped unwind testing exists, it won't cost nothing. The gtest_main.cc call will turn unwind testing off. Change-Id: I6643b26445891fd46abfacac52bc024024c8d7f6 Reviewed-on: https://boringssl-review.googlesource.com/c/33764 Reviewed-by: Adam Langley <agl@google.com> Reviewed-by: Adam Langley <alangley@gmail.com> Commit-Queue: David Benjamin <davidben@google.com>
2018-12-16 00:58:43 +00:00
#if defined(OPENSSL_WINDOWS)
Add a CFI tester to CHECK_ABI. This uses the x86 trap flag and libunwind to test CFI works at each instruction. For now, it just uses the system one out of pkg-config and disables unwind tests if unavailable. We'll probably want to stick a copy into //third_party and perhaps try the LLVM one later. This tester caught two bugs in P-256 CFI annotations already: I47b5f9798b3bcee1748e537b21c173d312a14b42 and I9f576d868850312d6c14d1386f8fbfa85021b347 An earlier design used PTRACE_SINGLESTEP with libunwind's remote unwinding features. ptrace is a mess around stop signals (see group-stop discussion in ptrace(2)) and this is 10x faster, so I went with it. The question of which is more future-proof is complex: - There are two libunwinds with the same API, https://www.nongnu.org/libunwind/ and LLVM's. This currently uses the system nongnu.org for convenience. In future, LLVM's should be easier to bundle (less complex build) and appears to even support Windows, but I haven't tested this. Moreover, setting the trap flag keeps the test single-process, which is less complex on Windows. That suggests the trap flag design and switching to LLVM later. However... - Not all architectures have a trap flag settable by userspace. As far as I can tell, ARMv8's PSTATE.SS can only be set from the kernel. If we stick with nongnu.org libunwind, we can use PTRACE_SINGLESTEP and remote unwinding. Or we implement it for LLVM. Another thought is for the ptracer to bounce SIGTRAP back into the process, to share the local unwinding code. - ARMv7 has no trap flag at all and PTRACE_SINGLESTEP fails. Debuggers single-step by injecting breakpoints instead. However, ARMv8's trap flag seems to work in both AArch32 and AArch64 modes, so we may be able to condition it on a 64-bit kernel. Sadly, neither strategy works with Intel SDE. Adding flags to cpucap vectors as we do with ARM would help, but it would not emulate CPUs newer than the host CPU. For now, I've just had SDE tests disable these. Annoyingly, CMake does not allow object libraries to have dependencies, so make test_support a proper static library. Rename the target to test_support_lib to avoid https://gitlab.kitware.com/cmake/cmake/issues/17785 Update-Note: This adds a new optional test dependency, but it's disabled by default (define BORINGSSL_HAVE_LIBUNWIND), so consumers do not need to do anything. We'll probably want to adjust this in the future. Bug: 181 Change-Id: I817263d7907aff0904a9cee83f8b26747262cc0c Reviewed-on: https://boringssl-review.googlesource.com/c/33966 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2018-12-21 23:58:36 +00:00
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm6),
"xmm6 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm7),
"xmm7 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm8),
"xmm8 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm9),
"xmm9 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm10),
"xmm10 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm11),
"xmm11 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm12),
"xmm12 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm13),
"xmm13 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm14),
"xmm14 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm15),
"xmm15 was not restored after return");
Add an ABI testing framework. Dear reader, I must apologize in advance. This CL contains the following: - A new 256-line perlasm file with non-trivial perl bits and a dual-ABI variadic function caller. - C preprocessor gymnastics, with variadic macros and fun facts about __VA_ARGS__'s behavior on empty argument lists. - C++ template gymnastics, including variadic arguments, template specialization, std::enable_if, and machinery to control template argument deduction. Enjoy. This tests that our assembly functions correctly honor platform ABI conventions. Right now this only tests callee-saved registers, but it should be extendable to SEH/CFI unwind testing with single-step debugging APIs. Register-checking does not involve anything funny and should be compatible with SDE. (The future unwind testing is unlikely to be compatible.) This CL adds support for x86_64 SysV and Win64 ABIs. ARM, AArch64, and x86 can be added in the future. The testing is injected in two places. First, all the assembly tests in p256-x86_64-test.cc are now instrumented. This is the intended workflow and should capture all registers. However, we currently do not unit-test our assembly much directly. We should do that as follow-up work[0] but, in the meantime, I've also wrapped all of the GTest main function in an ABI test. This is imperfect as ABI failures may be masked by other stack frames, but it costs nothing[1] and is pretty reliable at catching Win64 xmm register failures. [0] An alternate strategy would be, in debug builds, unconditionally instrument every assembly call in libcrypto. But the CHECK_ABI macro would be difficult to replicate in pure C, and unwind testing may be too invasive for this. Still, something to consider when we C++ libcrypto. [1] When single-stepped unwind testing exists, it won't cost nothing. The gtest_main.cc call will turn unwind testing off. Change-Id: I6643b26445891fd46abfacac52bc024024c8d7f6 Reviewed-on: https://boringssl-review.googlesource.com/c/33764 Reviewed-by: Adam Langley <agl@google.com> Reviewed-by: Adam Langley <alangley@gmail.com> Commit-Queue: David Benjamin <davidben@google.com>
2018-12-16 00:58:43 +00:00
#else
Add a CFI tester to CHECK_ABI. This uses the x86 trap flag and libunwind to test CFI works at each instruction. For now, it just uses the system one out of pkg-config and disables unwind tests if unavailable. We'll probably want to stick a copy into //third_party and perhaps try the LLVM one later. This tester caught two bugs in P-256 CFI annotations already: I47b5f9798b3bcee1748e537b21c173d312a14b42 and I9f576d868850312d6c14d1386f8fbfa85021b347 An earlier design used PTRACE_SINGLESTEP with libunwind's remote unwinding features. ptrace is a mess around stop signals (see group-stop discussion in ptrace(2)) and this is 10x faster, so I went with it. The question of which is more future-proof is complex: - There are two libunwinds with the same API, https://www.nongnu.org/libunwind/ and LLVM's. This currently uses the system nongnu.org for convenience. In future, LLVM's should be easier to bundle (less complex build) and appears to even support Windows, but I haven't tested this. Moreover, setting the trap flag keeps the test single-process, which is less complex on Windows. That suggests the trap flag design and switching to LLVM later. However... - Not all architectures have a trap flag settable by userspace. As far as I can tell, ARMv8's PSTATE.SS can only be set from the kernel. If we stick with nongnu.org libunwind, we can use PTRACE_SINGLESTEP and remote unwinding. Or we implement it for LLVM. Another thought is for the ptracer to bounce SIGTRAP back into the process, to share the local unwinding code. - ARMv7 has no trap flag at all and PTRACE_SINGLESTEP fails. Debuggers single-step by injecting breakpoints instead. However, ARMv8's trap flag seems to work in both AArch32 and AArch64 modes, so we may be able to condition it on a 64-bit kernel. Sadly, neither strategy works with Intel SDE. Adding flags to cpucap vectors as we do with ARM would help, but it would not emulate CPUs newer than the host CPU. For now, I've just had SDE tests disable these. Annoyingly, CMake does not allow object libraries to have dependencies, so make test_support a proper static library. Rename the target to test_support_lib to avoid https://gitlab.kitware.com/cmake/cmake/issues/17785 Update-Note: This adds a new optional test dependency, but it's disabled by default (define BORINGSSL_HAVE_LIBUNWIND), so consumers do not need to do anything. We'll probably want to adjust this in the future. Bug: 181 Change-Id: I817263d7907aff0904a9cee83f8b26747262cc0c Reviewed-on: https://boringssl-review.googlesource.com/c/33966 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2018-12-21 23:58:36 +00:00
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm6);
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm7);
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm8);
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm9);
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm10);
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm11);
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm12);
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm13);
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm14);
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm15);
Add an ABI testing framework. Dear reader, I must apologize in advance. This CL contains the following: - A new 256-line perlasm file with non-trivial perl bits and a dual-ABI variadic function caller. - C preprocessor gymnastics, with variadic macros and fun facts about __VA_ARGS__'s behavior on empty argument lists. - C++ template gymnastics, including variadic arguments, template specialization, std::enable_if, and machinery to control template argument deduction. Enjoy. This tests that our assembly functions correctly honor platform ABI conventions. Right now this only tests callee-saved registers, but it should be extendable to SEH/CFI unwind testing with single-step debugging APIs. Register-checking does not involve anything funny and should be compatible with SDE. (The future unwind testing is unlikely to be compatible.) This CL adds support for x86_64 SysV and Win64 ABIs. ARM, AArch64, and x86 can be added in the future. The testing is injected in two places. First, all the assembly tests in p256-x86_64-test.cc are now instrumented. This is the intended workflow and should capture all registers. However, we currently do not unit-test our assembly much directly. We should do that as follow-up work[0] but, in the meantime, I've also wrapped all of the GTest main function in an ABI test. This is imperfect as ABI failures may be masked by other stack frames, but it costs nothing[1] and is pretty reliable at catching Win64 xmm register failures. [0] An alternate strategy would be, in debug builds, unconditionally instrument every assembly call in libcrypto. But the CHECK_ABI macro would be difficult to replicate in pure C, and unwind testing may be too invasive for this. Still, something to consider when we C++ libcrypto. [1] When single-stepped unwind testing exists, it won't cost nothing. The gtest_main.cc call will turn unwind testing off. Change-Id: I6643b26445891fd46abfacac52bc024024c8d7f6 Reviewed-on: https://boringssl-review.googlesource.com/c/33764 Reviewed-by: Adam Langley <agl@google.com> Reviewed-by: Adam Langley <alangley@gmail.com> Commit-Queue: David Benjamin <davidben@google.com>
2018-12-16 00:58:43 +00:00
#endif
Add direction flag checking to CHECK_ABI. Linux and Windows ABIs both require that the direction flag be cleared on function exit, so that functions can rely on it being cleared on entry. (Some OpenSSL assembly preserves it, which is stronger, but we only require what is specified by the ABI so CHECK_ABI works with C compiler output.) Change-Id: I1a320aed4371176b4b44fe672f1a90167b84160f Reviewed-on: https://boringssl-review.googlesource.com/c/34187 Commit-Queue: Adam Langley <agl@google.com> Reviewed-by: Adam Langley <agl@google.com>
2019-01-07 02:05:52 +00:00
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_set_direction_flag),
"Direction flag set after return");
EXPECT_EQ(0, abi_test_get_and_clear_direction_flag())
<< "CHECK_ABI did not insulate the caller from direction flag errors";
Add an ABI testing framework. Dear reader, I must apologize in advance. This CL contains the following: - A new 256-line perlasm file with non-trivial perl bits and a dual-ABI variadic function caller. - C preprocessor gymnastics, with variadic macros and fun facts about __VA_ARGS__'s behavior on empty argument lists. - C++ template gymnastics, including variadic arguments, template specialization, std::enable_if, and machinery to control template argument deduction. Enjoy. This tests that our assembly functions correctly honor platform ABI conventions. Right now this only tests callee-saved registers, but it should be extendable to SEH/CFI unwind testing with single-step debugging APIs. Register-checking does not involve anything funny and should be compatible with SDE. (The future unwind testing is unlikely to be compatible.) This CL adds support for x86_64 SysV and Win64 ABIs. ARM, AArch64, and x86 can be added in the future. The testing is injected in two places. First, all the assembly tests in p256-x86_64-test.cc are now instrumented. This is the intended workflow and should capture all registers. However, we currently do not unit-test our assembly much directly. We should do that as follow-up work[0] but, in the meantime, I've also wrapped all of the GTest main function in an ABI test. This is imperfect as ABI failures may be masked by other stack frames, but it costs nothing[1] and is pretty reliable at catching Win64 xmm register failures. [0] An alternate strategy would be, in debug builds, unconditionally instrument every assembly call in libcrypto. But the CHECK_ABI macro would be difficult to replicate in pure C, and unwind testing may be too invasive for this. Still, something to consider when we C++ libcrypto. [1] When single-stepped unwind testing exists, it won't cost nothing. The gtest_main.cc call will turn unwind testing off. Change-Id: I6643b26445891fd46abfacac52bc024024c8d7f6 Reviewed-on: https://boringssl-review.googlesource.com/c/33764 Reviewed-by: Adam Langley <agl@google.com> Reviewed-by: Adam Langley <alangley@gmail.com> Commit-Queue: David Benjamin <davidben@google.com>
2018-12-16 00:58:43 +00:00
}
#endif // OPENSSL_X86_64 && SUPPORTS_ABI_TEST
Add ABI testing for 32-bit x86. This is much less interesting (stack-based parameters, Windows and SysV match, no SEH concerns as far as I can tell) than x86_64, but it was easy to do and I'm more familiar with x86 than ARM, so it made a better second architecture to make sure all the architecture ifdefs worked out. Also fix a bug in the x86_64 direction flag code. It was shifting in the wrong direction, making give 0 or 1<<20 rather than 0 or 1. (Happily, x86_64 appears to be unique in having vastly different calling conventions between OSs. x86 is the same between SysV and Windows, and ARM had the good sense to specify a (mostly) common set of rules.) Since a lot of the assembly functions use the same names and the tests were written generically, merely dropping in a trampoline and CallerState implementation gives us a bunch of ABI tests for free. Change-Id: I15408c18d43e88cfa1c5c0634a8b268a150ed961 Reviewed-on: https://boringssl-review.googlesource.com/c/34624 Reviewed-by: Adam Langley <agl@google.com> Commit-Queue: David Benjamin <davidben@google.com>
2019-01-26 17:29:08 +00:00
#if defined(OPENSSL_X86) && defined(SUPPORTS_ABI_TEST)
extern "C" {
void abi_test_clobber_eax(void);
void abi_test_clobber_ebx(void);
void abi_test_clobber_ecx(void);
void abi_test_clobber_edx(void);
void abi_test_clobber_esi(void);
void abi_test_clobber_edi(void);
void abi_test_clobber_ebp(void);
void abi_test_clobber_xmm0(void);
void abi_test_clobber_xmm1(void);
void abi_test_clobber_xmm2(void);
void abi_test_clobber_xmm3(void);
void abi_test_clobber_xmm4(void);
void abi_test_clobber_xmm5(void);
void abi_test_clobber_xmm6(void);
void abi_test_clobber_xmm7(void);
} // extern "C"
TEST(ABITest, X86) {
// abi_test_trampoline hides unsaved registers from the caller, so we can
// safely call the abi_test_clobber_* functions below.
abi_test::internal::CallerState state;
RAND_bytes(reinterpret_cast<uint8_t *>(&state), sizeof(state));
CHECK_ABI_NO_UNWIND(abi_test_trampoline,
reinterpret_cast<crypto_word_t>(abi_test_clobber_ebx),
&state, nullptr, 0, 0 /* no breakpoint */);
CHECK_ABI_NO_UNWIND(abi_test_clobber_eax);
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_ebx),
"ebx was not restored after return");
CHECK_ABI_NO_UNWIND(abi_test_clobber_ecx);
CHECK_ABI_NO_UNWIND(abi_test_clobber_edx);
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_edi),
"edi was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_esi),
"esi was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_ebp),
"ebp was not restored after return");
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm0);
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm1);
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm2);
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm3);
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm4);
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm5);
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm6);
CHECK_ABI_NO_UNWIND(abi_test_clobber_xmm7);
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_set_direction_flag),
"Direction flag set after return");
EXPECT_EQ(0, abi_test_get_and_clear_direction_flag())
<< "CHECK_ABI did not insulate the caller from direction flag errors";
}
#endif // OPENSSL_X86 && SUPPORTS_ABI_TEST
Implement ABI testing for ARM. Update-Note: There's some chance this'll break iOS since I was unable to test it there. The iPad I have to test on is too new to run 32-bit code at all. Change-Id: I6593f91b67a5e8a82828237d3b69ed948b07922d Reviewed-on: https://boringssl-review.googlesource.com/c/34725 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2019-01-07 20:29:48 +00:00
#if defined(OPENSSL_ARM) && defined(SUPPORTS_ABI_TEST)
extern "C" {
void abi_test_clobber_r0(void);
void abi_test_clobber_r1(void);
void abi_test_clobber_r2(void);
void abi_test_clobber_r3(void);
void abi_test_clobber_r4(void);
void abi_test_clobber_r5(void);
void abi_test_clobber_r6(void);
void abi_test_clobber_r7(void);
void abi_test_clobber_r8(void);
void abi_test_clobber_r9(void);
void abi_test_clobber_r10(void);
void abi_test_clobber_r11(void);
void abi_test_clobber_r12(void);
// r13, r14, and r15, are sp, lr, and pc, respectively.
void abi_test_clobber_d0(void);
void abi_test_clobber_d1(void);
void abi_test_clobber_d2(void);
void abi_test_clobber_d3(void);
void abi_test_clobber_d4(void);
void abi_test_clobber_d5(void);
void abi_test_clobber_d6(void);
void abi_test_clobber_d7(void);
void abi_test_clobber_d8(void);
void abi_test_clobber_d9(void);
void abi_test_clobber_d10(void);
void abi_test_clobber_d11(void);
void abi_test_clobber_d12(void);
void abi_test_clobber_d13(void);
void abi_test_clobber_d14(void);
void abi_test_clobber_d15(void);
} // extern "C"
TEST(ABITest, ARM) {
// abi_test_trampoline hides unsaved registers from the caller, so we can
// safely call the abi_test_clobber_* functions below.
abi_test::internal::CallerState state;
RAND_bytes(reinterpret_cast<uint8_t *>(&state), sizeof(state));
CHECK_ABI_NO_UNWIND(abi_test_trampoline,
reinterpret_cast<crypto_word_t>(abi_test_clobber_r4),
&state, nullptr, 0, 0 /* no breakpoint */);
CHECK_ABI_NO_UNWIND(abi_test_clobber_r0);
CHECK_ABI_NO_UNWIND(abi_test_clobber_r1);
CHECK_ABI_NO_UNWIND(abi_test_clobber_r2);
CHECK_ABI_NO_UNWIND(abi_test_clobber_r3);
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_r4),
"r4 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_r5),
"r5 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_r6),
"r6 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_r7),
"r7 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_r8),
"r8 was not restored after return");
#if defined(OPENSSL_APPLE)
CHECK_ABI_NO_UNWIND(abi_test_clobber_r9);
#else
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_r9),
"r9 was not restored after return");
#endif
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_r10),
"r10 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_r11),
"r11 was not restored after return");
CHECK_ABI_NO_UNWIND(abi_test_clobber_r12);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d0);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d1);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d2);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d3);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d4);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d5);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d6);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d7);
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_d8),
"d8 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_d9),
"d9 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_d10),
"d10 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_d11),
"d11 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_d12),
"d12 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_d13),
"d13 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_d14),
"d14 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_d15),
"d15 was not restored after return");
}
#endif // OPENSSL_ARM && SUPPORTS_ABI_TEST
Implement ABI testing for aarch64. This caught a bug in bn_mul_mont. Tested manually on iOS and Android. Change-Id: I1819fcd9ad34dbe3ba92bba952507d86dd12185a Reviewed-on: https://boringssl-review.googlesource.com/c/34805 Commit-Queue: David Benjamin <davidben@google.com> Reviewed-by: Adam Langley <agl@google.com>
2019-02-02 00:25:07 +00:00
#if defined(OPENSSL_AARCH64) && defined(SUPPORTS_ABI_TEST)
extern "C" {
void abi_test_clobber_x0(void);
void abi_test_clobber_x1(void);
void abi_test_clobber_x2(void);
void abi_test_clobber_x3(void);
void abi_test_clobber_x4(void);
void abi_test_clobber_x5(void);
void abi_test_clobber_x6(void);
void abi_test_clobber_x7(void);
void abi_test_clobber_x8(void);
void abi_test_clobber_x9(void);
void abi_test_clobber_x10(void);
void abi_test_clobber_x11(void);
void abi_test_clobber_x12(void);
void abi_test_clobber_x13(void);
void abi_test_clobber_x14(void);
void abi_test_clobber_x15(void);
void abi_test_clobber_x16(void);
void abi_test_clobber_x17(void);
// x18 is the platform register and off limits.
void abi_test_clobber_x19(void);
void abi_test_clobber_x20(void);
void abi_test_clobber_x21(void);
void abi_test_clobber_x22(void);
void abi_test_clobber_x23(void);
void abi_test_clobber_x24(void);
void abi_test_clobber_x25(void);
void abi_test_clobber_x26(void);
void abi_test_clobber_x27(void);
void abi_test_clobber_x28(void);
void abi_test_clobber_x29(void);
void abi_test_clobber_d0(void);
void abi_test_clobber_d1(void);
void abi_test_clobber_d2(void);
void abi_test_clobber_d3(void);
void abi_test_clobber_d4(void);
void abi_test_clobber_d5(void);
void abi_test_clobber_d6(void);
void abi_test_clobber_d7(void);
void abi_test_clobber_d8(void);
void abi_test_clobber_d9(void);
void abi_test_clobber_d10(void);
void abi_test_clobber_d11(void);
void abi_test_clobber_d12(void);
void abi_test_clobber_d13(void);
void abi_test_clobber_d14(void);
void abi_test_clobber_d15(void);
void abi_test_clobber_d16(void);
void abi_test_clobber_d17(void);
void abi_test_clobber_d18(void);
void abi_test_clobber_d19(void);
void abi_test_clobber_d20(void);
void abi_test_clobber_d21(void);
void abi_test_clobber_d22(void);
void abi_test_clobber_d23(void);
void abi_test_clobber_d24(void);
void abi_test_clobber_d25(void);
void abi_test_clobber_d26(void);
void abi_test_clobber_d27(void);
void abi_test_clobber_d28(void);
void abi_test_clobber_d29(void);
void abi_test_clobber_d30(void);
void abi_test_clobber_d31(void);
void abi_test_clobber_v8_upper(void);
void abi_test_clobber_v9_upper(void);
void abi_test_clobber_v10_upper(void);
void abi_test_clobber_v11_upper(void);
void abi_test_clobber_v12_upper(void);
void abi_test_clobber_v13_upper(void);
void abi_test_clobber_v14_upper(void);
void abi_test_clobber_v15_upper(void);
} // extern "C"
TEST(ABITest, AArch64) {
// abi_test_trampoline hides unsaved registers from the caller, so we can
// safely call the abi_test_clobber_* functions below.
abi_test::internal::CallerState state;
RAND_bytes(reinterpret_cast<uint8_t *>(&state), sizeof(state));
CHECK_ABI_NO_UNWIND(abi_test_trampoline,
reinterpret_cast<crypto_word_t>(abi_test_clobber_x19),
&state, nullptr, 0, 0 /* no breakpoint */);
CHECK_ABI_NO_UNWIND(abi_test_clobber_x0);
CHECK_ABI_NO_UNWIND(abi_test_clobber_x1);
CHECK_ABI_NO_UNWIND(abi_test_clobber_x2);
CHECK_ABI_NO_UNWIND(abi_test_clobber_x3);
CHECK_ABI_NO_UNWIND(abi_test_clobber_x4);
CHECK_ABI_NO_UNWIND(abi_test_clobber_x5);
CHECK_ABI_NO_UNWIND(abi_test_clobber_x6);
CHECK_ABI_NO_UNWIND(abi_test_clobber_x7);
CHECK_ABI_NO_UNWIND(abi_test_clobber_x8);
CHECK_ABI_NO_UNWIND(abi_test_clobber_x9);
CHECK_ABI_NO_UNWIND(abi_test_clobber_x10);
CHECK_ABI_NO_UNWIND(abi_test_clobber_x11);
CHECK_ABI_NO_UNWIND(abi_test_clobber_x12);
CHECK_ABI_NO_UNWIND(abi_test_clobber_x13);
CHECK_ABI_NO_UNWIND(abi_test_clobber_x14);
CHECK_ABI_NO_UNWIND(abi_test_clobber_x15);
CHECK_ABI_NO_UNWIND(abi_test_clobber_x16);
CHECK_ABI_NO_UNWIND(abi_test_clobber_x17);
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_x19),
"x19 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_x20),
"x20 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_x21),
"x21 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_x22),
"x22 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_x23),
"x23 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_x24),
"x24 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_x25),
"x25 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_x26),
"x26 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_x27),
"x27 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_x28),
"x28 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_x29),
"x29 was not restored after return");
CHECK_ABI_NO_UNWIND(abi_test_clobber_d0);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d1);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d2);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d3);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d4);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d5);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d6);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d7);
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_d8),
"d8 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_d9),
"d9 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_d10),
"d10 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_d11),
"d11 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_d12),
"d12 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_d13),
"d13 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_d14),
"d14 was not restored after return");
EXPECT_NONFATAL_FAILURE(CHECK_ABI_NO_UNWIND(abi_test_clobber_d15),
"d15 was not restored after return");
CHECK_ABI_NO_UNWIND(abi_test_clobber_d16);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d18);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d19);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d20);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d21);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d22);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d23);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d24);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d25);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d26);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d27);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d28);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d29);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d30);
CHECK_ABI_NO_UNWIND(abi_test_clobber_d31);
// The lower halves of v8-v15 (accessed as d8-d15) must be preserved, but not
// the upper halves.
CHECK_ABI_NO_UNWIND(abi_test_clobber_v8_upper);
CHECK_ABI_NO_UNWIND(abi_test_clobber_v9_upper);
CHECK_ABI_NO_UNWIND(abi_test_clobber_v10_upper);
CHECK_ABI_NO_UNWIND(abi_test_clobber_v11_upper);
CHECK_ABI_NO_UNWIND(abi_test_clobber_v12_upper);
CHECK_ABI_NO_UNWIND(abi_test_clobber_v13_upper);
CHECK_ABI_NO_UNWIND(abi_test_clobber_v14_upper);
CHECK_ABI_NO_UNWIND(abi_test_clobber_v15_upper);
}
#endif // OPENSSL_AARCH64 && SUPPORTS_ABI_TEST
Reference in New Issue Copy Permalink