mirrors
/
SDL
mirror of https://github.com/libsdl-org/SDL
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
SDL/test/testplatform.c

495 lines
23 KiB
C
Raw Normal View History

Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
/*
Updated copyright for 2024
2024-01-02 00:15:26 +03:00
Copyright (C) 1997-2024 Sam Lantinga <slouken@libsdl.org>
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
This software is provided 'as-is', without any express or implied
Fixed accidental changing of All to Allu in previous commit (thanks @zturtleman!)
2024-02-04 10:46:22 +03:00
warranty. In no event will the authors be held liable for any damages
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
arising from the use of this software.
Permission is granted to anyone to use this software for any purpose,
including commercial applications, and to alter it and redistribute it
freely.
*/
Switch header convention from `#include "SDL.h"` to `#include <SDL3/SDLh>` I ran this script in the include directory: ```sh sed -i '' -e 's,#include "\(SDL.*\)",#include <SDL3/\1>,' *.h ``` I ran this script in the src directory: ```sh for i in ../include/SDL3/SDL*.h do hdr=$(basename $i) if [ x"$(echo $hdr | egrep 'SDL_main|SDL_name|SDL_test|SDL_syswm|SDL_opengl|SDL_egl|SDL_vulkan')" != x ]; then find . -type f -exec sed -i '' -e 's,#include "\('$hdr'\)",#include <SDL3/\1>,' {} \; else find . -type f -exec sed -i '' -e '/#include "'$hdr'"/d' {} \; fi done ``` Fixes https://github.com/libsdl-org/SDL/issues/6575
2022-11-27 07:43:38 +03:00
#include <SDL3/SDL.h>
Add SDL_main.h includes to tests .. including cmake/test/*.c, whatever that does (this is needed now because SDL.h doesn't include SDL_main.h anymore)
2022-12-15 07:58:20 +03:00
#include <SDL3/SDL_main.h>
testprograms: parse arguments using SDLTest_CommonState
2023-03-17 02:25:39 +03:00
#include <SDL3/SDL_test.h>
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
/*
* Watcom C flags these as Warning 201: "Unreachable code" if you just
* compare them directly, so we push it through a function to keep the
* compiler quiet. --ryan.
*/
static int
badsize(size_t sizeoftype, size_t hardcodetype)
{
return sizeoftype != hardcodetype;
}
testplatform.c: move static asserts out of TestTypes().
2022-03-17 14:56:50 +03:00
SDL_COMPILE_TIME_ASSERT(SDL_MAX_SINT8, SDL_MAX_SINT8 == 127);
SDL_COMPILE_TIME_ASSERT(SDL_MIN_SINT8, SDL_MIN_SINT8 == -128);
SDL_COMPILE_TIME_ASSERT(SDL_MAX_UINT8, SDL_MAX_UINT8 == 255);
SDL_COMPILE_TIME_ASSERT(SDL_MIN_UINT8, SDL_MIN_UINT8 == 0);
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
testplatform.c: move static asserts out of TestTypes().
2022-03-17 14:56:50 +03:00
SDL_COMPILE_TIME_ASSERT(SDL_MAX_SINT16, SDL_MAX_SINT16 == 32767);
SDL_COMPILE_TIME_ASSERT(SDL_MIN_SINT16, SDL_MIN_SINT16 == -32768);
SDL_COMPILE_TIME_ASSERT(SDL_MAX_UINT16, SDL_MAX_UINT16 == 65535);
SDL_COMPILE_TIME_ASSERT(SDL_MIN_UINT16, SDL_MIN_UINT16 == 0);
Added min/max macros for the sized SDL datatypes
2017-10-17 00:39:56 +03:00
testplatform.c: move static asserts out of TestTypes().
2022-03-17 14:56:50 +03:00
SDL_COMPILE_TIME_ASSERT(SDL_MAX_SINT32, SDL_MAX_SINT32 == 2147483647);
SDL_COMPILE_TIME_ASSERT(SDL_MIN_SINT32, SDL_MIN_SINT32 == ~0x7fffffff); /* Instead of -2147483648, which is treated as unsigned by some compilers */
SDL_COMPILE_TIME_ASSERT(SDL_MAX_UINT32, SDL_MAX_UINT32 == 4294967295u);
SDL_COMPILE_TIME_ASSERT(SDL_MIN_UINT32, SDL_MIN_UINT32 == 0);
Added min/max macros for the sized SDL datatypes
2017-10-17 00:39:56 +03:00
System dependent 64 bit integer suffixes Added macros SDL_SINT64_C() and SDL_UINT64_C(). Integer suffixes of SDL_MAX_SINT64, SDL_MIN_SINT64, SDL_MAX_UINT64 and SDL_MIN_UINT64 are now system dependent.
2024-05-05 01:59:38 +03:00
SDL_COMPILE_TIME_ASSERT(SDL_MAX_SINT64, SDL_MAX_SINT64 == INT64_C(9223372036854775807));
SDL_COMPILE_TIME_ASSERT(SDL_MIN_SINT64, SDL_MIN_SINT64 == ~INT64_C(0x7fffffffffffffff)); /* Instead of -9223372036854775808, which is treated as unsigned by compilers */
SDL_COMPILE_TIME_ASSERT(SDL_MAX_UINT64, SDL_MAX_UINT64 == UINT64_C(18446744073709551615));
testplatform.c: move static asserts out of TestTypes().
2022-03-17 14:56:50 +03:00
SDL_COMPILE_TIME_ASSERT(SDL_MIN_UINT64, SDL_MIN_UINT64 == 0);
Added min/max macros for the sized SDL datatypes
2017-10-17 00:39:56 +03:00
Removed SDL_bool in favor of plain bool We require stdbool.h in the build environment, so we might as well use the plain bool type. If your environment doesn't have stdbool.h, this simple replacement will suffice: typedef signed char bool;
2024-09-18 17:52:28 +03:00
static int TestTypes(bool verbose)
testplatform.c: move static asserts out of TestTypes().
2022-03-17 14:56:50 +03:00
{
int error = 0;
Added min/max macros for the sized SDL datatypes
2017-10-17 00:39:56 +03:00
testplatform: make sure bool is 1 byte
2024-09-18 19:34:09 +03:00
if (badsize(sizeof(bool), 1)) {
if (verbose) {
SDL_Log("sizeof(bool) != 1, instead = %u\n", (unsigned int)sizeof(bool));
}
++error;
}
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
if (badsize(sizeof(Uint8), 1)) {
Cleanup add brace (#6545) * Add braces after if conditions * More add braces after if conditions * Add braces after while() conditions * Fix compilation because of macro being modified * Add braces to for loop * Add braces after if/goto * Move comments up * Remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements after merge * Fix inconsistent patterns are xxx == NULL vs !xxx * More "{}" for "if() break;" and "if() continue;" * More "{}" after if() short statement * More "{}" after "if () return;" statement * More fix inconsistent patterns are xxx == NULL vs !xxx * Revert some modificaion on SDL_RLEaccel.c * SDL_RLEaccel: no short statement * Cleanup 'if' where the bracket is in a new line * Cleanup 'while' where the bracket is in a new line * Cleanup 'for' where the bracket is in a new line * Cleanup 'else' where the bracket is in a new line
2022-11-27 19:38:43 +03:00
if (verbose) {
SDL_Log("sizeof(Uint8) != 1, instead = %u\n", (unsigned int)sizeof(Uint8));
}
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
++error;
}
if (badsize(sizeof(Uint16), 2)) {
Cleanup add brace (#6545) * Add braces after if conditions * More add braces after if conditions * Add braces after while() conditions * Fix compilation because of macro being modified * Add braces to for loop * Add braces after if/goto * Move comments up * Remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements after merge * Fix inconsistent patterns are xxx == NULL vs !xxx * More "{}" for "if() break;" and "if() continue;" * More "{}" after if() short statement * More "{}" after "if () return;" statement * More fix inconsistent patterns are xxx == NULL vs !xxx * Revert some modificaion on SDL_RLEaccel.c * SDL_RLEaccel: no short statement * Cleanup 'if' where the bracket is in a new line * Cleanup 'while' where the bracket is in a new line * Cleanup 'for' where the bracket is in a new line * Cleanup 'else' where the bracket is in a new line
2022-11-27 19:38:43 +03:00
if (verbose) {
SDL_Log("sizeof(Uint16) != 2, instead = %u\n", (unsigned int)sizeof(Uint16));
}
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
++error;
}
if (badsize(sizeof(Uint32), 4)) {
Cleanup add brace (#6545) * Add braces after if conditions * More add braces after if conditions * Add braces after while() conditions * Fix compilation because of macro being modified * Add braces to for loop * Add braces after if/goto * Move comments up * Remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements after merge * Fix inconsistent patterns are xxx == NULL vs !xxx * More "{}" for "if() break;" and "if() continue;" * More "{}" after if() short statement * More "{}" after "if () return;" statement * More fix inconsistent patterns are xxx == NULL vs !xxx * Revert some modificaion on SDL_RLEaccel.c * SDL_RLEaccel: no short statement * Cleanup 'if' where the bracket is in a new line * Cleanup 'while' where the bracket is in a new line * Cleanup 'for' where the bracket is in a new line * Cleanup 'else' where the bracket is in a new line
2022-11-27 19:38:43 +03:00
if (verbose) {
SDL_Log("sizeof(Uint32) != 4, instead = %u\n", (unsigned int)sizeof(Uint32));
}
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
++error;
}
if (badsize(sizeof(Uint64), 8)) {
Cleanup add brace (#6545) * Add braces after if conditions * More add braces after if conditions * Add braces after while() conditions * Fix compilation because of macro being modified * Add braces to for loop * Add braces after if/goto * Move comments up * Remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements after merge * Fix inconsistent patterns are xxx == NULL vs !xxx * More "{}" for "if() break;" and "if() continue;" * More "{}" after if() short statement * More "{}" after "if () return;" statement * More fix inconsistent patterns are xxx == NULL vs !xxx * Revert some modificaion on SDL_RLEaccel.c * SDL_RLEaccel: no short statement * Cleanup 'if' where the bracket is in a new line * Cleanup 'while' where the bracket is in a new line * Cleanup 'for' where the bracket is in a new line * Cleanup 'else' where the bracket is in a new line
2022-11-27 19:38:43 +03:00
if (verbose) {
SDL_Log("sizeof(Uint64) != 8, instead = %u\n", (unsigned int)sizeof(Uint64));
}
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
++error;
}
Cleanup add brace (#6545) * Add braces after if conditions * More add braces after if conditions * Add braces after while() conditions * Fix compilation because of macro being modified * Add braces to for loop * Add braces after if/goto * Move comments up * Remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements after merge * Fix inconsistent patterns are xxx == NULL vs !xxx * More "{}" for "if() break;" and "if() continue;" * More "{}" after if() short statement * More "{}" after "if () return;" statement * More fix inconsistent patterns are xxx == NULL vs !xxx * Revert some modificaion on SDL_RLEaccel.c * SDL_RLEaccel: no short statement * Cleanup 'if' where the bracket is in a new line * Cleanup 'while' where the bracket is in a new line * Cleanup 'for' where the bracket is in a new line * Cleanup 'else' where the bracket is in a new line
2022-11-27 19:38:43 +03:00
if (verbose && !error) {
Fixed accidental changing of All to Allu in previous commit (thanks @zturtleman!)
2024-02-04 10:46:22 +03:00
SDL_Log("All data types are the expected size.\n");
Cleanup add brace (#6545) * Add braces after if conditions * More add braces after if conditions * Add braces after while() conditions * Fix compilation because of macro being modified * Add braces to for loop * Add braces after if/goto * Move comments up * Remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements after merge * Fix inconsistent patterns are xxx == NULL vs !xxx * More "{}" for "if() break;" and "if() continue;" * More "{}" after if() short statement * More "{}" after "if () return;" statement * More fix inconsistent patterns are xxx == NULL vs !xxx * Revert some modificaion on SDL_RLEaccel.c * SDL_RLEaccel: no short statement * Cleanup 'if' where the bracket is in a new line * Cleanup 'while' where the bracket is in a new line * Cleanup 'for' where the bracket is in a new line * Cleanup 'else' where the bracket is in a new line
2022-11-27 19:38:43 +03:00
}
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
Cleanup add brace (#6545) * Add braces after if conditions * More add braces after if conditions * Add braces after while() conditions * Fix compilation because of macro being modified * Add braces to for loop * Add braces after if/goto * Move comments up * Remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements after merge * Fix inconsistent patterns are xxx == NULL vs !xxx * More "{}" for "if() break;" and "if() continue;" * More "{}" after if() short statement * More "{}" after "if () return;" statement * More fix inconsistent patterns are xxx == NULL vs !xxx * Revert some modificaion on SDL_RLEaccel.c * SDL_RLEaccel: no short statement * Cleanup 'if' where the bracket is in a new line * Cleanup 'while' where the bracket is in a new line * Cleanup 'for' where the bracket is in a new line * Cleanup 'else' where the bracket is in a new line
2022-11-27 19:38:43 +03:00
return error ? 1 : 0;
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
}
Removed SDL_bool in favor of plain bool We require stdbool.h in the build environment, so we might as well use the plain bool type. If your environment doesn't have stdbool.h, this simple replacement will suffice: typedef signed char bool;
2024-09-18 17:52:28 +03:00
static int TestEndian(bool verbose)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
{
int error = 0;
Uint16 value = 0x1234;
int real_byteorder;
Add SDL_FLOATWORDORDER for older ARM toolchains
2022-05-04 01:21:10 +03:00
int real_floatwordorder = 0;
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
Uint16 value16 = 0xCDAB;
Uint16 swapped16 = 0xABCD;
Uint32 value32 = 0xEFBEADDE;
Uint32 swapped32 = 0xDEADBEEF;
Uint64 value64, swapped64;
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
union
{
double d;
Uint32 ui32[2];
Add SDL_FLOATWORDORDER for older ARM toolchains
2022-05-04 01:21:10 +03:00
} value_double;
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
value64 = 0xEFBEADDE;
value64 <<= 32;
value64 |= 0xCDAB3412;
swapped64 = 0x1234ABCD;
swapped64 <<= 32;
swapped64 |= 0xDEADBEEF;
Add SDL_FLOATWORDORDER for older ARM toolchains
2022-05-04 01:21:10 +03:00
value_double.d = 3.141593;
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
if (verbose) {
SDL_Log("Detected a %s endian machine.\n",
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
(SDL_BYTEORDER == SDL_LIL_ENDIAN) ? "little" : "big");
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
}
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
if ((*((char *)&value) >> 4) == 0x1) {
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
real_byteorder = SDL_BIG_ENDIAN;
} else {
real_byteorder = SDL_LIL_ENDIAN;
}
if (real_byteorder != SDL_BYTEORDER) {
if (verbose) {
SDL_Log("Actually a %s endian machine!\n",
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
(real_byteorder == SDL_LIL_ENDIAN) ? "little" : "big");
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
}
++error;
}
Add SDL_FLOATWORDORDER for older ARM toolchains
2022-05-04 01:21:10 +03:00
if (verbose) {
SDL_Log("Detected a %s endian float word order machine.\n",
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
(SDL_FLOATWORDORDER == SDL_LIL_ENDIAN) ? "little" : "big");
Add SDL_FLOATWORDORDER for older ARM toolchains
2022-05-04 01:21:10 +03:00
}
if (value_double.ui32[0] == 0x82c2bd7f && value_double.ui32[1] == 0x400921fb) {
real_floatwordorder = SDL_LIL_ENDIAN;
} else if (value_double.ui32[0] == 0x400921fb && value_double.ui32[1] == 0x82c2bd7f) {
real_floatwordorder = SDL_BIG_ENDIAN;
}
if (real_floatwordorder != SDL_FLOATWORDORDER) {
if (verbose) {
SDL_Log("Actually a %s endian float word order machine!\n",
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
(real_floatwordorder == SDL_LIL_ENDIAN) ? "little" : (real_floatwordorder == SDL_BIG_ENDIAN) ? "big"
: "unknown");
Add SDL_FLOATWORDORDER for older ARM toolchains
2022-05-04 01:21:10 +03:00
}
++error;
}
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
if (verbose) {
SDL_Log("Value 16 = 0x%X, swapped = 0x%X\n", value16,
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
SDL_Swap16(value16));
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
}
if (SDL_Swap16(value16) != swapped16) {
if (verbose) {
SDL_Log("16 bit value swapped incorrectly!\n");
}
++error;
}
if (verbose) {
N3DS: Fix `-Wformat` warnings in tests. All warnings were about invalid specifiers. Since U/Sint32 is a long, using `%d` emits a -Wformat warning.
2022-10-10 08:31:14 +03:00
SDL_Log("Value 32 = 0x%" SDL_PRIX32 ", swapped = 0x%" SDL_PRIX32 "\n",
value32,
SDL_Swap32(value32));
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
}
if (SDL_Swap32(value32) != swapped32) {
if (verbose) {
SDL_Log("32 bit value swapped incorrectly!\n");
}
++error;
}
if (verbose) {
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
SDL_Log("Value 64 = 0x%" SDL_PRIX64 ", swapped = 0x%" SDL_PRIX64 "\n", value64,
SDL_Swap64(value64));
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
}
if (SDL_Swap64(value64) != swapped64) {
if (verbose) {
SDL_Log("64 bit value swapped incorrectly!\n");
}
++error;
}
Cleanup add brace (#6545) * Add braces after if conditions * More add braces after if conditions * Add braces after while() conditions * Fix compilation because of macro being modified * Add braces to for loop * Add braces after if/goto * Move comments up * Remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements after merge * Fix inconsistent patterns are xxx == NULL vs !xxx * More "{}" for "if() break;" and "if() continue;" * More "{}" after if() short statement * More "{}" after "if () return;" statement * More fix inconsistent patterns are xxx == NULL vs !xxx * Revert some modificaion on SDL_RLEaccel.c * SDL_RLEaccel: no short statement * Cleanup 'if' where the bracket is in a new line * Cleanup 'while' where the bracket is in a new line * Cleanup 'for' where the bracket is in a new line * Cleanup 'else' where the bracket is in a new line
2022-11-27 19:38:43 +03:00
return error ? 1 : 0;
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
}
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
static int TST_allmul(void *a, void *b, int arg, void *result, void *expected)
Fixed bug 3468 - _allshr in SDL_stdlib.c is not working properly Mark Pizzolato On Windows with Visual Studio, when building SDL as a static library using the x86 (32bit) mode, several intrinsic operations are implemented in code in SDL_stdlib.c. One of these, _allshr() is not properly implemented and fails for some input. As a result, some operations on 64bit data elements (long long) don't always work. I classified this bug as a blocker since things absolutely don't work when the affected code is invoked. The affected code is only invoked when SDL is compiled in x86 mode on Visual Studio when building a SDL as a static library. This build environment isn't common, and hence the bug hasn't been noticed previously. I reopened #2537 and mentioned this problem and provided a fix. That fix is provided again here along with test code which could be added to some of the SDL test code. This test code verifies that the x86 intrinsic routines produce the same results as the native x64 instructions which these routines emulate under the Microsoft compiler. The point of the tests is to make sure that Visual Studio x86 code produces the same results as Visual Studio x64 code. Some of the arguments (or boundary conditions) may produce different results on other compiler environments, so the tests really shouldn't be run on all compilers. The test driver only actually exercised code when the compiler defines _MSC_VER, so the driver can generically be invoked without issue.
2016-11-06 21:01:08 +03:00
{
(*(long long *)result) = ((*(long long *)a) * (*(long long *)b));
return (*(long long *)result) == (*(long long *)expected);
}
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
static int TST_alldiv(void *a, void *b, int arg, void *result, void *expected)
Fixed bug 3468 - _allshr in SDL_stdlib.c is not working properly Mark Pizzolato On Windows with Visual Studio, when building SDL as a static library using the x86 (32bit) mode, several intrinsic operations are implemented in code in SDL_stdlib.c. One of these, _allshr() is not properly implemented and fails for some input. As a result, some operations on 64bit data elements (long long) don't always work. I classified this bug as a blocker since things absolutely don't work when the affected code is invoked. The affected code is only invoked when SDL is compiled in x86 mode on Visual Studio when building a SDL as a static library. This build environment isn't common, and hence the bug hasn't been noticed previously. I reopened #2537 and mentioned this problem and provided a fix. That fix is provided again here along with test code which could be added to some of the SDL test code. This test code verifies that the x86 intrinsic routines produce the same results as the native x64 instructions which these routines emulate under the Microsoft compiler. The point of the tests is to make sure that Visual Studio x86 code produces the same results as Visual Studio x64 code. Some of the arguments (or boundary conditions) may produce different results on other compiler environments, so the tests really shouldn't be run on all compilers. The test driver only actually exercised code when the compiler defines _MSC_VER, so the driver can generically be invoked without issue.
2016-11-06 21:01:08 +03:00
{
(*(long long *)result) = ((*(long long *)a) / (*(long long *)b));
return (*(long long *)result) == (*(long long *)expected);
}
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
static int TST_allrem(void *a, void *b, int arg, void *result, void *expected)
Fixed bug 3468 - _allshr in SDL_stdlib.c is not working properly Mark Pizzolato On Windows with Visual Studio, when building SDL as a static library using the x86 (32bit) mode, several intrinsic operations are implemented in code in SDL_stdlib.c. One of these, _allshr() is not properly implemented and fails for some input. As a result, some operations on 64bit data elements (long long) don't always work. I classified this bug as a blocker since things absolutely don't work when the affected code is invoked. The affected code is only invoked when SDL is compiled in x86 mode on Visual Studio when building a SDL as a static library. This build environment isn't common, and hence the bug hasn't been noticed previously. I reopened #2537 and mentioned this problem and provided a fix. That fix is provided again here along with test code which could be added to some of the SDL test code. This test code verifies that the x86 intrinsic routines produce the same results as the native x64 instructions which these routines emulate under the Microsoft compiler. The point of the tests is to make sure that Visual Studio x86 code produces the same results as Visual Studio x64 code. Some of the arguments (or boundary conditions) may produce different results on other compiler environments, so the tests really shouldn't be run on all compilers. The test driver only actually exercised code when the compiler defines _MSC_VER, so the driver can generically be invoked without issue.
2016-11-06 21:01:08 +03:00
{
(*(long long *)result) = ((*(long long *)a) % (*(long long *)b));
return (*(long long *)result) == (*(long long *)expected);
}
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
static int TST_ualldiv(void *a, void *b, int arg, void *result, void *expected)
Fixed bug 3468 - _allshr in SDL_stdlib.c is not working properly Mark Pizzolato On Windows with Visual Studio, when building SDL as a static library using the x86 (32bit) mode, several intrinsic operations are implemented in code in SDL_stdlib.c. One of these, _allshr() is not properly implemented and fails for some input. As a result, some operations on 64bit data elements (long long) don't always work. I classified this bug as a blocker since things absolutely don't work when the affected code is invoked. The affected code is only invoked when SDL is compiled in x86 mode on Visual Studio when building a SDL as a static library. This build environment isn't common, and hence the bug hasn't been noticed previously. I reopened #2537 and mentioned this problem and provided a fix. That fix is provided again here along with test code which could be added to some of the SDL test code. This test code verifies that the x86 intrinsic routines produce the same results as the native x64 instructions which these routines emulate under the Microsoft compiler. The point of the tests is to make sure that Visual Studio x86 code produces the same results as Visual Studio x64 code. Some of the arguments (or boundary conditions) may produce different results on other compiler environments, so the tests really shouldn't be run on all compilers. The test driver only actually exercised code when the compiler defines _MSC_VER, so the driver can generically be invoked without issue.
2016-11-06 21:01:08 +03:00
{
(*(unsigned long long *)result) = ((*(unsigned long long *)a) / (*(unsigned long long *)b));
return (*(unsigned long long *)result) == (*(unsigned long long *)expected);
}
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
static int TST_uallrem(void *a, void *b, int arg, void *result, void *expected)
Fixed bug 3468 - _allshr in SDL_stdlib.c is not working properly Mark Pizzolato On Windows with Visual Studio, when building SDL as a static library using the x86 (32bit) mode, several intrinsic operations are implemented in code in SDL_stdlib.c. One of these, _allshr() is not properly implemented and fails for some input. As a result, some operations on 64bit data elements (long long) don't always work. I classified this bug as a blocker since things absolutely don't work when the affected code is invoked. The affected code is only invoked when SDL is compiled in x86 mode on Visual Studio when building a SDL as a static library. This build environment isn't common, and hence the bug hasn't been noticed previously. I reopened #2537 and mentioned this problem and provided a fix. That fix is provided again here along with test code which could be added to some of the SDL test code. This test code verifies that the x86 intrinsic routines produce the same results as the native x64 instructions which these routines emulate under the Microsoft compiler. The point of the tests is to make sure that Visual Studio x86 code produces the same results as Visual Studio x64 code. Some of the arguments (or boundary conditions) may produce different results on other compiler environments, so the tests really shouldn't be run on all compilers. The test driver only actually exercised code when the compiler defines _MSC_VER, so the driver can generically be invoked without issue.
2016-11-06 21:01:08 +03:00
{
(*(unsigned long long *)result) = ((*(unsigned long long *)a) % (*(unsigned long long *)b));
return (*(unsigned long long *)result) == (*(unsigned long long *)expected);
}
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
static int TST_allshl(void *a, void *b, int arg, void *result, void *expected)
Fixed bug 3468 - _allshr in SDL_stdlib.c is not working properly Mark Pizzolato On Windows with Visual Studio, when building SDL as a static library using the x86 (32bit) mode, several intrinsic operations are implemented in code in SDL_stdlib.c. One of these, _allshr() is not properly implemented and fails for some input. As a result, some operations on 64bit data elements (long long) don't always work. I classified this bug as a blocker since things absolutely don't work when the affected code is invoked. The affected code is only invoked when SDL is compiled in x86 mode on Visual Studio when building a SDL as a static library. This build environment isn't common, and hence the bug hasn't been noticed previously. I reopened #2537 and mentioned this problem and provided a fix. That fix is provided again here along with test code which could be added to some of the SDL test code. This test code verifies that the x86 intrinsic routines produce the same results as the native x64 instructions which these routines emulate under the Microsoft compiler. The point of the tests is to make sure that Visual Studio x86 code produces the same results as Visual Studio x64 code. Some of the arguments (or boundary conditions) may produce different results on other compiler environments, so the tests really shouldn't be run on all compilers. The test driver only actually exercised code when the compiler defines _MSC_VER, so the driver can generically be invoked without issue.
2016-11-06 21:01:08 +03:00
{
(*(long long *)result) = (*(long long *)a) << arg;
return (*(long long *)result) == (*(long long *)expected);
}
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
static int TST_aullshl(void *a, void *b, int arg, void *result, void *expected)
Fixed bug 3468 - _allshr in SDL_stdlib.c is not working properly Mark Pizzolato On Windows with Visual Studio, when building SDL as a static library using the x86 (32bit) mode, several intrinsic operations are implemented in code in SDL_stdlib.c. One of these, _allshr() is not properly implemented and fails for some input. As a result, some operations on 64bit data elements (long long) don't always work. I classified this bug as a blocker since things absolutely don't work when the affected code is invoked. The affected code is only invoked when SDL is compiled in x86 mode on Visual Studio when building a SDL as a static library. This build environment isn't common, and hence the bug hasn't been noticed previously. I reopened #2537 and mentioned this problem and provided a fix. That fix is provided again here along with test code which could be added to some of the SDL test code. This test code verifies that the x86 intrinsic routines produce the same results as the native x64 instructions which these routines emulate under the Microsoft compiler. The point of the tests is to make sure that Visual Studio x86 code produces the same results as Visual Studio x64 code. Some of the arguments (or boundary conditions) may produce different results on other compiler environments, so the tests really shouldn't be run on all compilers. The test driver only actually exercised code when the compiler defines _MSC_VER, so the driver can generically be invoked without issue.
2016-11-06 21:01:08 +03:00
{
(*(unsigned long long *)result) = (*(unsigned long long *)a) << arg;
return (*(unsigned long long *)result) == (*(unsigned long long *)expected);
}
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
static int TST_allshr(void *a, void *b, int arg, void *result, void *expected)
Fixed bug 3468 - _allshr in SDL_stdlib.c is not working properly Mark Pizzolato On Windows with Visual Studio, when building SDL as a static library using the x86 (32bit) mode, several intrinsic operations are implemented in code in SDL_stdlib.c. One of these, _allshr() is not properly implemented and fails for some input. As a result, some operations on 64bit data elements (long long) don't always work. I classified this bug as a blocker since things absolutely don't work when the affected code is invoked. The affected code is only invoked when SDL is compiled in x86 mode on Visual Studio when building a SDL as a static library. This build environment isn't common, and hence the bug hasn't been noticed previously. I reopened #2537 and mentioned this problem and provided a fix. That fix is provided again here along with test code which could be added to some of the SDL test code. This test code verifies that the x86 intrinsic routines produce the same results as the native x64 instructions which these routines emulate under the Microsoft compiler. The point of the tests is to make sure that Visual Studio x86 code produces the same results as Visual Studio x64 code. Some of the arguments (or boundary conditions) may produce different results on other compiler environments, so the tests really shouldn't be run on all compilers. The test driver only actually exercised code when the compiler defines _MSC_VER, so the driver can generically be invoked without issue.
2016-11-06 21:01:08 +03:00
{
(*(long long *)result) = (*(long long *)a) >> arg;
return (*(long long *)result) == (*(long long *)expected);
}
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
static int TST_aullshr(void *a, void *b, int arg, void *result, void *expected)
Fixed bug 3468 - _allshr in SDL_stdlib.c is not working properly Mark Pizzolato On Windows with Visual Studio, when building SDL as a static library using the x86 (32bit) mode, several intrinsic operations are implemented in code in SDL_stdlib.c. One of these, _allshr() is not properly implemented and fails for some input. As a result, some operations on 64bit data elements (long long) don't always work. I classified this bug as a blocker since things absolutely don't work when the affected code is invoked. The affected code is only invoked when SDL is compiled in x86 mode on Visual Studio when building a SDL as a static library. This build environment isn't common, and hence the bug hasn't been noticed previously. I reopened #2537 and mentioned this problem and provided a fix. That fix is provided again here along with test code which could be added to some of the SDL test code. This test code verifies that the x86 intrinsic routines produce the same results as the native x64 instructions which these routines emulate under the Microsoft compiler. The point of the tests is to make sure that Visual Studio x86 code produces the same results as Visual Studio x64 code. Some of the arguments (or boundary conditions) may produce different results on other compiler environments, so the tests really shouldn't be run on all compilers. The test driver only actually exercised code when the compiler defines _MSC_VER, so the driver can generically be invoked without issue.
2016-11-06 21:01:08 +03:00
{
(*(unsigned long long *)result) = (*(unsigned long long *)a) >> arg;
return (*(unsigned long long *)result) == (*(unsigned long long *)expected);
}
typedef int (*LL_Intrinsic)(void *a, void *b, int arg, void *result, void *expected);
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
typedef struct
{
Fixed bug 3468 - _allshr in SDL_stdlib.c is not working properly Mark Pizzolato On Windows with Visual Studio, when building SDL as a static library using the x86 (32bit) mode, several intrinsic operations are implemented in code in SDL_stdlib.c. One of these, _allshr() is not properly implemented and fails for some input. As a result, some operations on 64bit data elements (long long) don't always work. I classified this bug as a blocker since things absolutely don't work when the affected code is invoked. The affected code is only invoked when SDL is compiled in x86 mode on Visual Studio when building a SDL as a static library. This build environment isn't common, and hence the bug hasn't been noticed previously. I reopened #2537 and mentioned this problem and provided a fix. That fix is provided again here along with test code which could be added to some of the SDL test code. This test code verifies that the x86 intrinsic routines produce the same results as the native x64 instructions which these routines emulate under the Microsoft compiler. The point of the tests is to make sure that Visual Studio x86 code produces the same results as Visual Studio x64 code. Some of the arguments (or boundary conditions) may produce different results on other compiler environments, so the tests really shouldn't be run on all compilers. The test driver only actually exercised code when the compiler defines _MSC_VER, so the driver can generically be invoked without issue.
2016-11-06 21:01:08 +03:00
const char *operation;
LL_Intrinsic routine;
unsigned long long a, b;
int arg;
unsigned long long expected_result;
} LL_Test;
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
static LL_Test LL_Tests[] = {
Fixed warning C4245: 'initializing': conversion from '__int64' to 'unsigned __int64', signed/unsigned mismatch
2024-02-03 22:27:35 +03:00
/* UNDEFINED {"_allshl", &TST_allshl, 0xFFFFFFFFFFFFFFFFllu, 0llu, 65, 0x0000000000000000ll}, */
{ "_allshl", &TST_allshl, 0xFFFFFFFFFFFFFFFFllu, 0llu, 1, 0xFFFFFFFFFFFFFFFEllu },
{ "_allshl", &TST_allshl, 0xFFFFFFFFFFFFFFFFllu, 0llu, 32, 0xFFFFFFFF00000000llu },
{ "_allshl", &TST_allshl, 0xFFFFFFFFFFFFFFFFllu, 0llu, 33, 0xFFFFFFFE00000000llu },
{ "_allshl", &TST_allshl, 0xFFFFFFFFFFFFFFFFllu, 0llu, 0, 0xFFFFFFFFFFFFFFFFllu },
{ "_allshr", &TST_allshr, 0xAAAAAAAA55555555llu, 0llu, 63, 0xFFFFFFFFFFFFFFFFllu },
/* UNDEFINED {"_allshr", &TST_allshr, 0xFFFFFFFFFFFFFFFFllu, 0llu, 65, 0xFFFFFFFFFFFFFFFFll}, */
{ "_allshr", &TST_allshr, 0xFFFFFFFFFFFFFFFFllu, 0llu, 1, 0xFFFFFFFFFFFFFFFFllu },
{ "_allshr", &TST_allshr, 0xFFFFFFFFFFFFFFFFllu, 0llu, 32, 0xFFFFFFFFFFFFFFFFllu },
{ "_allshr", &TST_allshr, 0xFFFFFFFFFFFFFFFFllu, 0llu, 33, 0xFFFFFFFFFFFFFFFFllu },
{ "_allshr", &TST_allshr, 0xFFFFFFFFFFFFFFFFllu, 0llu, 0, 0xFFFFFFFFFFFFFFFFllu },
/* UNDEFINED {"_allshr", &TST_allshr, 0x5F5F5F5F5F5F5F5Fllu, 0llu, 65, 0x0000000000000000ll}, */
{ "_allshr", &TST_allshr, 0x5F5F5F5F5F5F5F5Fllu, 0llu, 1, 0x2FAFAFAFAFAFAFAFllu },
{ "_allshr", &TST_allshr, 0x5F5F5F5F5F5F5F5Fllu, 0llu, 32, 0x000000005F5F5F5Fllu },
{ "_allshr", &TST_allshr, 0x5F5F5F5F5F5F5F5Fllu, 0llu, 33, 0x000000002FAFAFAFllu },
/* UNDEFINED {"_aullshl", &TST_aullshl, 0xFFFFFFFFFFFFFFFFllu, 0llu, 65, 0x0000000000000000ll}, */
{ "_aullshl", &TST_aullshl, 0xFFFFFFFFFFFFFFFFllu, 0llu, 1, 0xFFFFFFFFFFFFFFFEllu },
{ "_aullshl", &TST_aullshl, 0xFFFFFFFFFFFFFFFFllu, 0llu, 32, 0xFFFFFFFF00000000llu },
{ "_aullshl", &TST_aullshl, 0xFFFFFFFFFFFFFFFFllu, 0llu, 33, 0xFFFFFFFE00000000llu },
{ "_aullshl", &TST_aullshl, 0xFFFFFFFFFFFFFFFFllu, 0llu, 0, 0xFFFFFFFFFFFFFFFFllu },
/* UNDEFINED {"_aullshr", &TST_aullshr, 0xFFFFFFFFFFFFFFFFllu, 0llu, 65, 0x0000000000000000ll}, */
{ "_aullshr", &TST_aullshr, 0xFFFFFFFFFFFFFFFFllu, 0llu, 1, 0x7FFFFFFFFFFFFFFFllu },
{ "_aullshr", &TST_aullshr, 0xFFFFFFFFFFFFFFFFllu, 0llu, 32, 0x00000000FFFFFFFFllu },
{ "_aullshr", &TST_aullshr, 0xFFFFFFFFFFFFFFFFllu, 0llu, 33, 0x000000007FFFFFFFllu },
{ "_aullshr", &TST_aullshr, 0xFFFFFFFFFFFFFFFFllu, 0llu, 0, 0xFFFFFFFFFFFFFFFFllu },
{ "_allmul", &TST_allmul, 0xFFFFFFFFFFFFFFFFllu, 0x0000000000000000llu, 0, 0x0000000000000000llu },
{ "_allmul", &TST_allmul, 0x0000000000000000llu, 0xFFFFFFFFFFFFFFFFllu, 0, 0x0000000000000000llu },
{ "_allmul", &TST_allmul, 0x000000000FFFFFFFllu, 0x0000000000000001llu, 0, 0x000000000FFFFFFFllu },
{ "_allmul", &TST_allmul, 0x0000000000000001llu, 0x000000000FFFFFFFllu, 0, 0x000000000FFFFFFFllu },
{ "_allmul", &TST_allmul, 0x000000000FFFFFFFllu, 0x0000000000000010llu, 0, 0x00000000FFFFFFF0llu },
{ "_allmul", &TST_allmul, 0x0000000000000010llu, 0x000000000FFFFFFFllu, 0, 0x00000000FFFFFFF0llu },
{ "_allmul", &TST_allmul, 0x000000000FFFFFFFllu, 0x0000000000000100llu, 0, 0x0000000FFFFFFF00llu },
{ "_allmul", &TST_allmul, 0x0000000000000100llu, 0x000000000FFFFFFFllu, 0, 0x0000000FFFFFFF00llu },
{ "_allmul", &TST_allmul, 0x000000000FFFFFFFllu, 0x0000000010000000llu, 0, 0x00FFFFFFF0000000llu },
{ "_allmul", &TST_allmul, 0x0000000010000000llu, 0x000000000FFFFFFFllu, 0, 0x00FFFFFFF0000000llu },
{ "_allmul", &TST_allmul, 0x000000000FFFFFFFllu, 0x0000000080000000llu, 0, 0x07FFFFFF80000000llu },
{ "_allmul", &TST_allmul, 0x0000000080000000llu, 0x000000000FFFFFFFllu, 0, 0x07FFFFFF80000000llu },
{ "_allmul", &TST_allmul, 0xFFFFFFFFFFFFFFFEllu, 0x0000000080000000llu, 0, 0xFFFFFFFF00000000llu },
{ "_allmul", &TST_allmul, 0x0000000080000000llu, 0xFFFFFFFFFFFFFFFEllu, 0, 0xFFFFFFFF00000000llu },
{ "_allmul", &TST_allmul, 0xFFFFFFFFFFFFFFFEllu, 0x0000000080000008llu, 0, 0xFFFFFFFEFFFFFFF0llu },
{ "_allmul", &TST_allmul, 0x0000000080000008llu, 0xFFFFFFFFFFFFFFFEllu, 0, 0xFFFFFFFEFFFFFFF0llu },
{ "_allmul", &TST_allmul, 0x00000000FFFFFFFFllu, 0x00000000FFFFFFFFllu, 0, 0xFFFFFFFE00000001llu },
{ "_alldiv", &TST_alldiv, 0x0000000000000000llu, 0x0000000000000001llu, 0, 0x0000000000000000llu },
{ "_alldiv", &TST_alldiv, 0x0000000000000000llu, 0xFFFFFFFFFFFFFFFFllu, 0, 0x0000000000000000llu },
{ "_alldiv", &TST_alldiv, 0x0000000000000001llu, 0xFFFFFFFFFFFFFFFFllu, 0, 0xFFFFFFFFFFFFFFFFllu },
{ "_alldiv", &TST_alldiv, 0xFFFFFFFFFFFFFFFFllu, 0x0000000000000001llu, 0, 0xFFFFFFFFFFFFFFFFllu },
{ "_alldiv", &TST_alldiv, 0x0000000000000001llu, 0xFFFFFFFFFFFFFFFFllu, 0, 0xFFFFFFFFFFFFFFFFllu },
{ "_alldiv", &TST_alldiv, 0x0000000000000001llu, 0x0000000000000001llu, 0, 0x0000000000000001llu },
{ "_alldiv", &TST_alldiv, 0xFFFFFFFFFFFFFFFFllu, 0xFFFFFFFFFFFFFFFFllu, 0, 0x0000000000000001llu },
{ "_alldiv", &TST_alldiv, 0x000000000FFFFFFFllu, 0x0000000000000001llu, 0, 0x000000000FFFFFFFllu },
{ "_alldiv", &TST_alldiv, 0x0000000FFFFFFFFFllu, 0x0000000000000010llu, 0, 0x00000000FFFFFFFFllu },
{ "_alldiv", &TST_alldiv, 0x0000000000000100llu, 0x000000000FFFFFFFllu, 0, 0x0000000000000000llu },
{ "_alldiv", &TST_alldiv, 0x00FFFFFFF0000000llu, 0x0000000010000000llu, 0, 0x000000000FFFFFFFllu },
{ "_alldiv", &TST_alldiv, 0x07FFFFFF80000000llu, 0x0000000080000000llu, 0, 0x000000000FFFFFFFllu },
{ "_alldiv", &TST_alldiv, 0xFFFFFFFFFFFFFFFEllu, 0x0000000080000000llu, 0, 0x0000000000000000llu },
{ "_alldiv", &TST_alldiv, 0xFFFFFFFEFFFFFFF0llu, 0xFFFFFFFFFFFFFFFEllu, 0, 0x0000000080000008llu },
{ "_alldiv", &TST_alldiv, 0x7FFFFFFEFFFFFFF0llu, 0xFFFFFFFFFFFFFFFEllu, 0, 0xC000000080000008llu },
{ "_alldiv", &TST_alldiv, 0x7FFFFFFEFFFFFFF0llu, 0x0000FFFFFFFFFFFEllu, 0, 0x0000000000007FFFllu },
{ "_alldiv", &TST_alldiv, 0x7FFFFFFEFFFFFFF0llu, 0x7FFFFFFEFFFFFFF0llu, 0, 0x0000000000000001llu },
{ "_allrem", &TST_allrem, 0x0000000000000000llu, 0x0000000000000001llu, 0, 0x0000000000000000llu },
{ "_allrem", &TST_allrem, 0x0000000000000000llu, 0xFFFFFFFFFFFFFFFFllu, 0, 0x0000000000000000llu },
{ "_allrem", &TST_allrem, 0x0000000000000001llu, 0xFFFFFFFFFFFFFFFFllu, 0, 0x0000000000000000llu },
{ "_allrem", &TST_allrem, 0xFFFFFFFFFFFFFFFFllu, 0x0000000000000001llu, 0, 0x0000000000000000llu },
{ "_allrem", &TST_allrem, 0x0000000000000001llu, 0xFFFFFFFFFFFFFFFFllu, 0, 0x0000000000000000llu },
{ "_allrem", &TST_allrem, 0x0000000000000001llu, 0x0000000000000001llu, 0, 0x0000000000000000llu },
{ "_allrem", &TST_allrem, 0xFFFFFFFFFFFFFFFFllu, 0xFFFFFFFFFFFFFFFFllu, 0, 0x0000000000000000llu },
{ "_allrem", &TST_allrem, 0x000000000FFFFFFFllu, 0x0000000000000001llu, 0, 0x0000000000000000llu },
{ "_allrem", &TST_allrem, 0x0000000FFFFFFFFFllu, 0x0000000000000010llu, 0, 0x000000000000000Fllu },
{ "_allrem", &TST_allrem, 0x0000000000000100llu, 0x000000000FFFFFFFllu, 0, 0x0000000000000100llu },
{ "_allrem", &TST_allrem, 0x00FFFFFFF0000000llu, 0x0000000010000000llu, 0, 0x0000000000000000llu },
{ "_allrem", &TST_allrem, 0x07FFFFFF80000000llu, 0x0000000080000000llu, 0, 0x0000000000000000llu },
{ "_allrem", &TST_allrem, 0xFFFFFFFFFFFFFFFEllu, 0x0000000080000000llu, 0, 0xFFFFFFFFFFFFFFFEllu },
{ "_allrem", &TST_allrem, 0xFFFFFFFEFFFFFFF0llu, 0xFFFFFFFFFFFFFFFEllu, 0, 0x0000000000000000llu },
{ "_allrem", &TST_allrem, 0x7FFFFFFEFFFFFFF0llu, 0xFFFFFFFFFFFFFFFEllu, 0, 0x0000000000000000llu },
{ "_allrem", &TST_allrem, 0x7FFFFFFEFFFFFFF0llu, 0x0000FFFFFFFFFFFEllu, 0, 0x0000FFFF0000FFEEllu },
{ "_allrem", &TST_allrem, 0x7FFFFFFEFFFFFFF0llu, 0x7FFFFFFEFFFFFFF0llu, 0, 0x0000000000000000llu },
{ "_ualldiv", &TST_ualldiv, 0x0000000000000000llu, 0x0000000000000001llu, 0, 0x0000000000000000llu },
{ "_ualldiv", &TST_ualldiv, 0x0000000000000000llu, 0xFFFFFFFFFFFFFFFFllu, 0, 0x0000000000000000llu },
{ "_ualldiv", &TST_ualldiv, 0x0000000000000001llu, 0xFFFFFFFFFFFFFFFFllu, 0, 0x0000000000000000llu },
{ "_ualldiv", &TST_ualldiv, 0xFFFFFFFFFFFFFFFFllu, 0x0000000000000001llu, 0, 0xFFFFFFFFFFFFFFFFllu },
{ "_ualldiv", &TST_ualldiv, 0x0000000000000001llu, 0xFFFFFFFFFFFFFFFFllu, 0, 0x0000000000000000llu },
{ "_ualldiv", &TST_ualldiv, 0x0000000000000001llu, 0x0000000000000001llu, 0, 0x0000000000000001llu },
{ "_ualldiv", &TST_ualldiv, 0xFFFFFFFFFFFFFFFFllu, 0xFFFFFFFFFFFFFFFFllu, 0, 0x0000000000000001llu },
{ "_ualldiv", &TST_ualldiv, 0x000000000FFFFFFFllu, 0x0000000000000001llu, 0, 0x000000000FFFFFFFllu },
{ "_ualldiv", &TST_ualldiv, 0x0000000FFFFFFFFFllu, 0x0000000000000010llu, 0, 0x00000000FFFFFFFFllu },
{ "_ualldiv", &TST_ualldiv, 0x0000000000000100llu, 0x000000000FFFFFFFllu, 0, 0x0000000000000000llu },
{ "_ualldiv", &TST_ualldiv, 0x00FFFFFFF0000000llu, 0x0000000010000000llu, 0, 0x000000000FFFFFFFllu },
{ "_ualldiv", &TST_ualldiv, 0x07FFFFFF80000000llu, 0x0000000080000000llu, 0, 0x000000000FFFFFFFllu },
{ "_ualldiv", &TST_ualldiv, 0xFFFFFFFFFFFFFFFEllu, 0x0000000080000000llu, 0, 0x00000001FFFFFFFFllu },
{ "_ualldiv", &TST_ualldiv, 0xFFFFFFFEFFFFFFF0llu, 0xFFFFFFFFFFFFFFFEllu, 0, 0x0000000000000000llu },
{ "_ualldiv", &TST_ualldiv, 0x7FFFFFFEFFFFFFF0llu, 0xFFFFFFFFFFFFFFFEllu, 0, 0x0000000000000000llu },
{ "_ualldiv", &TST_ualldiv, 0x7FFFFFFEFFFFFFF0llu, 0x0000FFFFFFFFFFFEllu, 0, 0x0000000000007FFFllu },
{ "_ualldiv", &TST_ualldiv, 0x7FFFFFFEFFFFFFF0llu, 0x7FFFFFFEFFFFFFF0llu, 0, 0x0000000000000001llu },
{ "_uallrem", &TST_uallrem, 0x0000000000000000llu, 0x0000000000000001llu, 0, 0x0000000000000000llu },
{ "_uallrem", &TST_uallrem, 0x0000000000000000llu, 0xFFFFFFFFFFFFFFFFllu, 0, 0x0000000000000000llu },
{ "_uallrem", &TST_uallrem, 0x0000000000000001llu, 0xFFFFFFFFFFFFFFFFllu, 0, 0x0000000000000001llu },
{ "_uallrem", &TST_uallrem, 0xFFFFFFFFFFFFFFFFllu, 0x0000000000000001llu, 0, 0x0000000000000000llu },
{ "_uallrem", &TST_uallrem, 0x0000000000000001llu, 0xFFFFFFFFFFFFFFFFllu, 0, 0x0000000000000001llu },
{ "_uallrem", &TST_uallrem, 0x0000000000000001llu, 0x0000000000000001llu, 0, 0x0000000000000000llu },
{ "_uallrem", &TST_uallrem, 0xFFFFFFFFFFFFFFFFllu, 0xFFFFFFFFFFFFFFFFllu, 0, 0x0000000000000000llu },
{ "_uallrem", &TST_uallrem, 0x000000000FFFFFFFllu, 0x0000000000000001llu, 0, 0x0000000000000000llu },
{ "_uallrem", &TST_uallrem, 0x0000000FFFFFFFFFllu, 0x0000000000000010llu, 0, 0x000000000000000Fllu },
{ "_uallrem", &TST_uallrem, 0x0000000000000100llu, 0x000000000FFFFFFFllu, 0, 0x0000000000000100llu },
{ "_uallrem", &TST_uallrem, 0x00FFFFFFF0000000llu, 0x0000000010000000llu, 0, 0x0000000000000000llu },
{ "_uallrem", &TST_uallrem, 0x07FFFFFF80000000llu, 0x0000000080000000llu, 0, 0x0000000000000000llu },
{ "_uallrem", &TST_uallrem, 0xFFFFFFFFFFFFFFFEllu, 0x0000000080000000llu, 0, 0x000000007FFFFFFEllu },
{ "_uallrem", &TST_uallrem, 0xFFFFFFFEFFFFFFF0llu, 0xFFFFFFFFFFFFFFFEllu, 0, 0xFFFFFFFEFFFFFFF0llu },
{ "_uallrem", &TST_uallrem, 0x7FFFFFFEFFFFFFF0llu, 0xFFFFFFFFFFFFFFFEllu, 0, 0x7FFFFFFEFFFFFFF0llu },
{ "_uallrem", &TST_uallrem, 0x7FFFFFFEFFFFFFF0llu, 0x0000FFFFFFFFFFFEllu, 0, 0x0000FFFF0000FFEEllu },
{ "_uallrem", &TST_uallrem, 0x7FFFFFFEFFFFFFF0llu, 0x7FFFFFFEFFFFFFF0llu, 0, 0x0000000000000000llu },
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
More fix warnings about static function and prototype
2023-03-08 18:12:45 +03:00
{ NULL, NULL, 0, 0, 0, 0 }
Fixed bug 3468 - _allshr in SDL_stdlib.c is not working properly Mark Pizzolato On Windows with Visual Studio, when building SDL as a static library using the x86 (32bit) mode, several intrinsic operations are implemented in code in SDL_stdlib.c. One of these, _allshr() is not properly implemented and fails for some input. As a result, some operations on 64bit data elements (long long) don't always work. I classified this bug as a blocker since things absolutely don't work when the affected code is invoked. The affected code is only invoked when SDL is compiled in x86 mode on Visual Studio when building a SDL as a static library. This build environment isn't common, and hence the bug hasn't been noticed previously. I reopened #2537 and mentioned this problem and provided a fix. That fix is provided again here along with test code which could be added to some of the SDL test code. This test code verifies that the x86 intrinsic routines produce the same results as the native x64 instructions which these routines emulate under the Microsoft compiler. The point of the tests is to make sure that Visual Studio x86 code produces the same results as Visual Studio x64 code. Some of the arguments (or boundary conditions) may produce different results on other compiler environments, so the tests really shouldn't be run on all compilers. The test driver only actually exercised code when the compiler defines _MSC_VER, so the driver can generically be invoked without issue.
2016-11-06 21:01:08 +03:00
};
Removed SDL_bool in favor of plain bool We require stdbool.h in the build environment, so we might as well use the plain bool type. If your environment doesn't have stdbool.h, this simple replacement will suffice: typedef signed char bool;
2024-09-18 17:52:28 +03:00
static int Test64Bit(bool verbose)
Fixed bug 3468 - _allshr in SDL_stdlib.c is not working properly Mark Pizzolato On Windows with Visual Studio, when building SDL as a static library using the x86 (32bit) mode, several intrinsic operations are implemented in code in SDL_stdlib.c. One of these, _allshr() is not properly implemented and fails for some input. As a result, some operations on 64bit data elements (long long) don't always work. I classified this bug as a blocker since things absolutely don't work when the affected code is invoked. The affected code is only invoked when SDL is compiled in x86 mode on Visual Studio when building a SDL as a static library. This build environment isn't common, and hence the bug hasn't been noticed previously. I reopened #2537 and mentioned this problem and provided a fix. That fix is provided again here along with test code which could be added to some of the SDL test code. This test code verifies that the x86 intrinsic routines produce the same results as the native x64 instructions which these routines emulate under the Microsoft compiler. The point of the tests is to make sure that Visual Studio x86 code produces the same results as Visual Studio x64 code. Some of the arguments (or boundary conditions) may produce different results on other compiler environments, so the tests really shouldn't be run on all compilers. The test driver only actually exercised code when the compiler defines _MSC_VER, so the driver can generically be invoked without issue.
2016-11-06 21:01:08 +03:00
{
LL_Test *t;
int failed = 0;
Pointer as bool (libsdl-org#7214)
2023-11-10 00:29:15 +03:00
for (t = LL_Tests; t->routine; t++) {
Fixed bug 3468 - _allshr in SDL_stdlib.c is not working properly Mark Pizzolato On Windows with Visual Studio, when building SDL as a static library using the x86 (32bit) mode, several intrinsic operations are implemented in code in SDL_stdlib.c. One of these, _allshr() is not properly implemented and fails for some input. As a result, some operations on 64bit data elements (long long) don't always work. I classified this bug as a blocker since things absolutely don't work when the affected code is invoked. The affected code is only invoked when SDL is compiled in x86 mode on Visual Studio when building a SDL as a static library. This build environment isn't common, and hence the bug hasn't been noticed previously. I reopened #2537 and mentioned this problem and provided a fix. That fix is provided again here along with test code which could be added to some of the SDL test code. This test code verifies that the x86 intrinsic routines produce the same results as the native x64 instructions which these routines emulate under the Microsoft compiler. The point of the tests is to make sure that Visual Studio x86 code produces the same results as Visual Studio x64 code. Some of the arguments (or boundary conditions) may produce different results on other compiler environments, so the tests really shouldn't be run on all compilers. The test driver only actually exercised code when the compiler defines _MSC_VER, so the driver can generically be invoked without issue.
2016-11-06 21:01:08 +03:00
unsigned long long result = 0;
unsigned int *al = (unsigned int *)&t->a;
unsigned int *bl = (unsigned int *)&t->b;
unsigned int *el = (unsigned int *)&t->expected_result;
unsigned int *rl = (unsigned int *)&result;
if (!t->routine(&t->a, &t->b, t->arg, &result, &t->expected_result)) {
Cleanup add brace (#6545) * Add braces after if conditions * More add braces after if conditions * Add braces after while() conditions * Fix compilation because of macro being modified * Add braces to for loop * Add braces after if/goto * Move comments up * Remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements after merge * Fix inconsistent patterns are xxx == NULL vs !xxx * More "{}" for "if() break;" and "if() continue;" * More "{}" after if() short statement * More "{}" after "if () return;" statement * More fix inconsistent patterns are xxx == NULL vs !xxx * Revert some modificaion on SDL_RLEaccel.c * SDL_RLEaccel: no short statement * Cleanup 'if' where the bracket is in a new line * Cleanup 'while' where the bracket is in a new line * Cleanup 'for' where the bracket is in a new line * Cleanup 'else' where the bracket is in a new line
2022-11-27 19:38:43 +03:00
if (verbose) {
SDL_Log("%s(0x%08X%08X, 0x%08X%08X, %3d, produced: 0x%08X%08X, expected: 0x%08X%08X\n", t->operation, al[1], al[0], bl[1], bl[0],
t->arg, rl[1], rl[0], el[1], el[0]);
}
Fixed bug 3468 - _allshr in SDL_stdlib.c is not working properly Mark Pizzolato On Windows with Visual Studio, when building SDL as a static library using the x86 (32bit) mode, several intrinsic operations are implemented in code in SDL_stdlib.c. One of these, _allshr() is not properly implemented and fails for some input. As a result, some operations on 64bit data elements (long long) don't always work. I classified this bug as a blocker since things absolutely don't work when the affected code is invoked. The affected code is only invoked when SDL is compiled in x86 mode on Visual Studio when building a SDL as a static library. This build environment isn't common, and hence the bug hasn't been noticed previously. I reopened #2537 and mentioned this problem and provided a fix. That fix is provided again here along with test code which could be added to some of the SDL test code. This test code verifies that the x86 intrinsic routines produce the same results as the native x64 instructions which these routines emulate under the Microsoft compiler. The point of the tests is to make sure that Visual Studio x86 code produces the same results as Visual Studio x64 code. Some of the arguments (or boundary conditions) may produce different results on other compiler environments, so the tests really shouldn't be run on all compilers. The test driver only actually exercised code when the compiler defines _MSC_VER, so the driver can generically be invoked without issue.
2016-11-06 21:01:08 +03:00
++failed;
}
}
Cleanup add brace (#6545) * Add braces after if conditions * More add braces after if conditions * Add braces after while() conditions * Fix compilation because of macro being modified * Add braces to for loop * Add braces after if/goto * Move comments up * Remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements after merge * Fix inconsistent patterns are xxx == NULL vs !xxx * More "{}" for "if() break;" and "if() continue;" * More "{}" after if() short statement * More "{}" after "if () return;" statement * More fix inconsistent patterns are xxx == NULL vs !xxx * Revert some modificaion on SDL_RLEaccel.c * SDL_RLEaccel: no short statement * Cleanup 'if' where the bracket is in a new line * Cleanup 'while' where the bracket is in a new line * Cleanup 'for' where the bracket is in a new line * Cleanup 'else' where the bracket is in a new line
2022-11-27 19:38:43 +03:00
if (verbose && (failed == 0)) {
Fixed accidental changing of All to Allu in previous commit (thanks @zturtleman!)
2024-02-04 10:46:22 +03:00
SDL_Log("All 64bit intrinsic tests passed\n");
Cleanup add brace (#6545) * Add braces after if conditions * More add braces after if conditions * Add braces after while() conditions * Fix compilation because of macro being modified * Add braces to for loop * Add braces after if/goto * Move comments up * Remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements after merge * Fix inconsistent patterns are xxx == NULL vs !xxx * More "{}" for "if() break;" and "if() continue;" * More "{}" after if() short statement * More "{}" after "if () return;" statement * More fix inconsistent patterns are xxx == NULL vs !xxx * Revert some modificaion on SDL_RLEaccel.c * SDL_RLEaccel: no short statement * Cleanup 'if' where the bracket is in a new line * Cleanup 'while' where the bracket is in a new line * Cleanup 'for' where the bracket is in a new line * Cleanup 'else' where the bracket is in a new line
2022-11-27 19:38:43 +03:00
}
return failed ? 1 : 0;
Fixed bug 3468 - _allshr in SDL_stdlib.c is not working properly Mark Pizzolato On Windows with Visual Studio, when building SDL as a static library using the x86 (32bit) mode, several intrinsic operations are implemented in code in SDL_stdlib.c. One of these, _allshr() is not properly implemented and fails for some input. As a result, some operations on 64bit data elements (long long) don't always work. I classified this bug as a blocker since things absolutely don't work when the affected code is invoked. The affected code is only invoked when SDL is compiled in x86 mode on Visual Studio when building a SDL as a static library. This build environment isn't common, and hence the bug hasn't been noticed previously. I reopened #2537 and mentioned this problem and provided a fix. That fix is provided again here along with test code which could be added to some of the SDL test code. This test code verifies that the x86 intrinsic routines produce the same results as the native x64 instructions which these routines emulate under the Microsoft compiler. The point of the tests is to make sure that Visual Studio x86 code produces the same results as Visual Studio x64 code. Some of the arguments (or boundary conditions) may produce different results on other compiler environments, so the tests really shouldn't be run on all compilers. The test driver only actually exercised code when the compiler defines _MSC_VER, so the driver can generically be invoked without issue.
2016-11-06 21:01:08 +03:00
}
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
Removed SDL_bool in favor of plain bool We require stdbool.h in the build environment, so we might as well use the plain bool type. If your environment doesn't have stdbool.h, this simple replacement will suffice: typedef signed char bool;
2024-09-18 17:52:28 +03:00
static int TestCPUInfo(bool verbose)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
{
if (verbose) {
cpuinfo: Rename SDL_GetCPUCount to SDL_GetNumLogicalCPUCores This was the only API that broke the "GetNumThings" convention used elsewhere, so renaming it helps with consistency. Adding "logical cores" to the name also makes it a bit more immediately obvious what the count actually represents.
2024-09-14 01:06:38 +03:00
SDL_Log("Number of logical CPU cores: %d\n", SDL_GetNumLogicalCPUCores());
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
SDL_Log("CPU cache line size: %d\n", SDL_GetCPUCacheLineSize());
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
SDL_Log("AltiVec %s\n", SDL_HasAltiVec() ? "detected" : "not detected");
SDL_Log("MMX %s\n", SDL_HasMMX() ? "detected" : "not detected");
SDL_Log("SSE %s\n", SDL_HasSSE() ? "detected" : "not detected");
SDL_Log("SSE2 %s\n", SDL_HasSSE2() ? "detected" : "not detected");
SDL_Log("SSE3 %s\n", SDL_HasSSE3() ? "detected" : "not detected");
SDL_Log("SSE4.1 %s\n", SDL_HasSSE41() ? "detected" : "not detected");
SDL_Log("SSE4.2 %s\n", SDL_HasSSE42() ? "detected" : "not detected");
SDL_Log("AVX %s\n", SDL_HasAVX() ? "detected" : "not detected");
SDL_Log("AVX2 %s\n", SDL_HasAVX2() ? "detected" : "not detected");
SDL_Log("AVX-512F %s\n", SDL_HasAVX512F() ? "detected" : "not detected");
SDL_Log("ARM SIMD %s\n", SDL_HasARMSIMD() ? "detected" : "not detected");
SDL_Log("NEON %s\n", SDL_HasNEON() ? "detected" : "not detected");
SDL_Log("LSX %s\n", SDL_HasLSX() ? "detected" : "not detected");
SDL_Log("LASX %s\n", SDL_HasLASX() ? "detected" : "not detected");
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
SDL_Log("System RAM %d MB\n", SDL_GetSystemRAM());
}
Cleanup add brace (#6545) * Add braces after if conditions * More add braces after if conditions * Add braces after while() conditions * Fix compilation because of macro being modified * Add braces to for loop * Add braces after if/goto * Move comments up * Remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements after merge * Fix inconsistent patterns are xxx == NULL vs !xxx * More "{}" for "if() break;" and "if() continue;" * More "{}" after if() short statement * More "{}" after "if () return;" statement * More fix inconsistent patterns are xxx == NULL vs !xxx * Revert some modificaion on SDL_RLEaccel.c * SDL_RLEaccel: no short statement * Cleanup 'if' where the bracket is in a new line * Cleanup 'while' where the bracket is in a new line * Cleanup 'for' where the bracket is in a new line * Cleanup 'else' where the bracket is in a new line
2022-11-27 19:38:43 +03:00
return 0;
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
}
Removed SDL_bool in favor of plain bool We require stdbool.h in the build environment, so we might as well use the plain bool type. If your environment doesn't have stdbool.h, this simple replacement will suffice: typedef signed char bool;
2024-09-18 17:52:28 +03:00
static int TestAssertions(bool verbose)
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
{
SDL_assert(1);
SDL_assert_release(1);
SDL_assert_paranoid(1);
SDL_assert(0 || 1);
SDL_assert_release(0 || 1);
SDL_assert_paranoid(0 || 1);
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
#if 0 /* enable this to test assertion failures. */
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
SDL_assert_release(1 == 2);
SDL_assert_release(5 < 4);
SDL_assert_release(0 && "This is a test");
#endif
{
const SDL_AssertData *item = SDL_GetAssertionReport();
while (item) {
SDL_Log("'%s', %s (%s:%d), triggered %u times, always ignore: %s.\n",
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
item->condition, item->function, item->filename,
item->linenum, item->trigger_count,
item->always_ignore ? "yes" : "no");
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
item = item->next;
}
}
Cleanup add brace (#6545) * Add braces after if conditions * More add braces after if conditions * Add braces after while() conditions * Fix compilation because of macro being modified * Add braces to for loop * Add braces after if/goto * Move comments up * Remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements * More remove extra () in the 'return ...;' statements after merge * Fix inconsistent patterns are xxx == NULL vs !xxx * More "{}" for "if() break;" and "if() continue;" * More "{}" after if() short statement * More "{}" after "if () return;" statement * More fix inconsistent patterns are xxx == NULL vs !xxx * Revert some modificaion on SDL_RLEaccel.c * SDL_RLEaccel: no short statement * Cleanup 'if' where the bracket is in a new line * Cleanup 'while' where the bracket is in a new line * Cleanup 'for' where the bracket is in a new line * Cleanup 'else' where the bracket is in a new line
2022-11-27 19:38:43 +03:00
return 0;
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
}
Update for SDL3 coding style (#6717) I updated .clang-format and ran clang-format 14 over the src and test directories to standardize the code base. In general I let clang-format have it's way, and added markup to prevent formatting of code that would break or be completely unreadable if formatted. The script I ran for the src directory is added as build-scripts/clang-format-src.sh This fixes: #6592 #6593 #6594
2022-11-30 23:51:59 +03:00
int main(int argc, char *argv[])
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
{
testprograms: parse arguments using SDLTest_CommonState
2023-03-17 02:25:39 +03:00
int i;
Removed SDL_bool in favor of plain bool We require stdbool.h in the build environment, so we might as well use the plain bool type. If your environment doesn't have stdbool.h, this simple replacement will suffice: typedef signed char bool;
2024-09-18 17:52:28 +03:00
bool verbose = true;
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
int status = 0;
testprograms: parse arguments using SDLTest_CommonState
2023-03-17 02:25:39 +03:00
SDLTest_CommonState *state;
/* Initialize test framework */
state = SDLTest_CommonCreateState(argv, 0);
Pointer as bool (libsdl-org#7214)
2023-11-10 00:29:15 +03:00
if (!state) {
testprograms: parse arguments using SDLTest_CommonState
2023-03-17 02:25:39 +03:00
return 1;
}
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
testprograms: parse arguments using SDLTest_CommonState
2023-03-17 02:25:39 +03:00
/* Parse commandline */
for (i = 1; i < argc;) {
int consumed;
consumed = SDLTest_CommonArg(state, i);
if (!consumed) {
if (SDL_strcmp(argv[i], "-q") == 0) {
Removed SDL_bool in favor of plain bool We require stdbool.h in the build environment, so we might as well use the plain bool type. If your environment doesn't have stdbool.h, this simple replacement will suffice: typedef signed char bool;
2024-09-18 17:52:28 +03:00
verbose = false;
testprograms: parse arguments using SDLTest_CommonState
2023-03-17 02:25:39 +03:00
consumed = 1;
}
}
if (consumed <= 0) {
static const char *options[] = { "[-q]", NULL };
SDLTest_CommonLogUsage(state, argv[0], options);
return 1;
}
i += consumed;
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
}
testprograms: parse arguments using SDLTest_CommonState
2023-03-17 02:25:39 +03:00
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
if (verbose) {
SDL_Log("This system is running %s\n", SDL_GetPlatform());
}
status += TestTypes(verbose);
status += TestEndian(verbose);
Fixed bug 3468 - _allshr in SDL_stdlib.c is not working properly Mark Pizzolato On Windows with Visual Studio, when building SDL as a static library using the x86 (32bit) mode, several intrinsic operations are implemented in code in SDL_stdlib.c. One of these, _allshr() is not properly implemented and fails for some input. As a result, some operations on 64bit data elements (long long) don't always work. I classified this bug as a blocker since things absolutely don't work when the affected code is invoked. The affected code is only invoked when SDL is compiled in x86 mode on Visual Studio when building a SDL as a static library. This build environment isn't common, and hence the bug hasn't been noticed previously. I reopened #2537 and mentioned this problem and provided a fix. That fix is provided again here along with test code which could be added to some of the SDL test code. This test code verifies that the x86 intrinsic routines produce the same results as the native x64 instructions which these routines emulate under the Microsoft compiler. The point of the tests is to make sure that Visual Studio x86 code produces the same results as Visual Studio x64 code. Some of the arguments (or boundary conditions) may produce different results on other compiler environments, so the tests really shouldn't be run on all compilers. The test driver only actually exercised code when the compiler defines _MSC_VER, so the driver can generically be invoked without issue.
2016-11-06 21:01:08 +03:00
status += Test64Bit(verbose);
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
status += TestCPUInfo(verbose);
status += TestAssertions(verbose);
Make sure we call SDL_Quit() before SDLTest_CommonDestroyState() Otherwise we'll report leaks that will be cleaned up in SDL_Quit()
2024-07-28 17:46:44 +03:00
SDL_Quit();
testprograms: parse arguments using SDLTest_CommonState
2023-03-17 02:25:39 +03:00
SDLTest_CommonDestroyState(state);
Fixed crash if initialization of EGL failed but was tried again later. The internal function SDL_EGL_LoadLibrary() did not delete and remove a mostly uninitialized data structure if loading the library first failed. A later try to use EGL then skipped initialization and assumed it was previously successful because the data structure now already existed. This led to at least one crash in the internal function SDL_EGL_ChooseConfig() because a NULL pointer was dereferenced to make a call to eglBindAPI().
2015-06-21 18:33:46 +03:00
return status;
}