maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/clang/test/Modules/diag-flags.cpp

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

57 lines
4.9 KiB
C++
Raw Normal View History

Fix initial diagnostic state setup for an explicit module with no diagnostic pragmas. If a file has no diagnostic pragmas, we build its diagnostic state lazily, but in this case we never set up the root state to be the diagnostic state in which the module was originally built, so the diagnostic flags for files in the module with no diagnostic pragmas were incorrectly based on the user of the module rather than the diagnostic state when the module was built. llvm-svn: 301846
2017-05-02 06:10:47 +08:00
// RUN: rm -rf %t
//
[modules] Round-trip -Werror flag through explicit module build. The intent for an explicit module build is that the diagnostics produced within the module are those that were configured when the module was built, not those that are enabled within a user of the module. This includes diagnostics that don't actually show up until the module is used (for instance, diagnostics produced during template instantiation and weird cases like -Wpadded). We serialized and restored the diagnostic state for individual warning groups, but previously did not track the state for flags like -Werror and -Weverything, which are implemented as separate bits rather than as part of the diagnostics mapping information. llvm-svn: 301992
2017-05-03 08:28:49 +08:00
// For an implicit module, all that matters are the warning flags in the user.
// RUN: %clang_cc1 -triple %itanium_abi_triple -fmodules -fimplicit-module-maps -emit-module -fmodules-cache-path=%t -fmodule-name=diag_flags -x c++ %S/Inputs/module.map -fmodules-ts
// RUN: %clang_cc1 -triple %itanium_abi_triple -fmodules -fimplicit-module-maps -verify -fmodules-cache-path=%t -I %S/Inputs %s -fmodules-ts
[modules] Fix incorrect diagnostic mapping computation when a module changes diagnostic settings using _Pragma within a macro. The AST writer had previously been assuming that all diagnostic state transitions would occur within a FileID corresponding to a file. When a diagnostic state change occured within a macro, it was unable to form a location for that state change and would instead corrupt the diagnostic state of the "root" node (and thus that of the main compilation). Also introduce a "#pragma clang __debug diag_mapping" debugging utility that I added to track this issue down. llvm-svn: 324695
2018-02-09 09:15:13 +08:00
// RUN: %clang_cc1 -triple %itanium_abi_triple -fmodules -fimplicit-module-maps -verify -fmodules-cache-path=%t -I %S/Inputs %s -fmodules-ts -DWARNING -DLOCAL_WARNING -Wpadded
// RUN: %clang_cc1 -triple %itanium_abi_triple -fmodules -fimplicit-module-maps -verify -fmodules-cache-path=%t -I %S/Inputs %s -fmodules-ts -DERROR -DLOCAL_ERROR -Wpadded -Werror
// RUN: %clang_cc1 -triple %itanium_abi_triple -fmodules -fimplicit-module-maps -verify -fmodules-cache-path=%t -I %S/Inputs %s -fmodules-ts -DERROR -DLOCAL_ERROR -Werror=padded
Fix initial diagnostic state setup for an explicit module with no diagnostic pragmas. If a file has no diagnostic pragmas, we build its diagnostic state lazily, but in this case we never set up the root state to be the diagnostic state in which the module was originally built, so the diagnostic flags for files in the module with no diagnostic pragmas were incorrectly based on the user of the module rather than the diagnostic state when the module was built. llvm-svn: 301846
2017-05-02 06:10:47 +08:00
//
[modules] Round-trip -Werror flag through explicit module build. The intent for an explicit module build is that the diagnostics produced within the module are those that were configured when the module was built, not those that are enabled within a user of the module. This includes diagnostics that don't actually show up until the module is used (for instance, diagnostics produced during template instantiation and weird cases like -Wpadded). We serialized and restored the diagnostic state for individual warning groups, but previously did not track the state for flags like -Werror and -Weverything, which are implemented as separate bits rather than as part of the diagnostics mapping information. llvm-svn: 301992
2017-05-03 08:28:49 +08:00
// For an explicit module, all that matters are the warning flags in the module build.
// RUN: %clang_cc1 -triple %itanium_abi_triple -fmodules -fimplicit-module-maps -emit-module -fmodule-name=diag_flags -x c++ %S/Inputs/module.map -fmodules-ts -o %t/nodiag.pcm
[modules] Fix incorrect diagnostic mapping computation when a module changes diagnostic settings using _Pragma within a macro. The AST writer had previously been assuming that all diagnostic state transitions would occur within a FileID corresponding to a file. When a diagnostic state change occured within a macro, it was unable to form a location for that state change and would instead corrupt the diagnostic state of the "root" node (and thus that of the main compilation). Also introduce a "#pragma clang __debug diag_mapping" debugging utility that I added to track this issue down. llvm-svn: 324695
2018-02-09 09:15:13 +08:00
// RUN: %clang_cc1 -triple %itanium_abi_triple -fmodules -fimplicit-module-maps -verify -fmodules-cache-path=%t -I %S/Inputs %s -fmodules-ts -fmodule-file=%t/nodiag.pcm -Wpadded -DLOCAL_WARNING
// RUN: %clang_cc1 -triple %itanium_abi_triple -fmodules -fimplicit-module-maps -verify -fmodules-cache-path=%t -I %S/Inputs %s -fmodules-ts -fmodule-file=%t/nodiag.pcm -Werror -Wpadded -DLOCAL_ERROR
[modules] Round-trip -Werror flag through explicit module build. The intent for an explicit module build is that the diagnostics produced within the module are those that were configured when the module was built, not those that are enabled within a user of the module. This includes diagnostics that don't actually show up until the module is used (for instance, diagnostics produced during template instantiation and weird cases like -Wpadded). We serialized and restored the diagnostic state for individual warning groups, but previously did not track the state for flags like -Werror and -Weverything, which are implemented as separate bits rather than as part of the diagnostics mapping information. llvm-svn: 301992
2017-05-03 08:28:49 +08:00
//
// RUN: %clang_cc1 -triple %itanium_abi_triple -fmodules -fimplicit-module-maps -emit-module -fmodule-name=diag_flags -x c++ %S/Inputs/module.map -fmodules-ts -o %t/warning.pcm -Wpadded
// RUN: %clang_cc1 -triple %itanium_abi_triple -fmodules -fimplicit-module-maps -verify -fmodules-cache-path=%t -I %S/Inputs %s -fmodules-ts -DWARNING -fmodule-file=%t/warning.pcm
[modules] Fix incorrect diagnostic mapping computation when a module changes diagnostic settings using _Pragma within a macro. The AST writer had previously been assuming that all diagnostic state transitions would occur within a FileID corresponding to a file. When a diagnostic state change occured within a macro, it was unable to form a location for that state change and would instead corrupt the diagnostic state of the "root" node (and thus that of the main compilation). Also introduce a "#pragma clang __debug diag_mapping" debugging utility that I added to track this issue down. llvm-svn: 324695
2018-02-09 09:15:13 +08:00
// RUN: %clang_cc1 -triple %itanium_abi_triple -fmodules -fimplicit-module-maps -verify -fmodules-cache-path=%t -I %S/Inputs %s -fmodules-ts -DWARNING -fmodule-file=%t/warning.pcm -Werror=padded -DLOCAL_ERROR
[modules] Round-trip -Werror flag through explicit module build. The intent for an explicit module build is that the diagnostics produced within the module are those that were configured when the module was built, not those that are enabled within a user of the module. This includes diagnostics that don't actually show up until the module is used (for instance, diagnostics produced during template instantiation and weird cases like -Wpadded). We serialized and restored the diagnostic state for individual warning groups, but previously did not track the state for flags like -Werror and -Weverything, which are implemented as separate bits rather than as part of the diagnostics mapping information. llvm-svn: 301992
2017-05-03 08:28:49 +08:00
// RUN: %clang_cc1 -triple %itanium_abi_triple -fmodules -fimplicit-module-maps -verify -fmodules-cache-path=%t -I %S/Inputs %s -fmodules-ts -DWARNING -fmodule-file=%t/warning.pcm -Werror
//
// RUN: %clang_cc1 -triple %itanium_abi_triple -fmodules -fimplicit-module-maps -emit-module -fmodule-name=diag_flags -x c++ %S/Inputs/module.map -fmodules-ts -o %t/werror-no-error.pcm -Werror -Wpadded -Wno-error=padded
// RUN: %clang_cc1 -triple %itanium_abi_triple -fmodules -fimplicit-module-maps -verify -fmodules-cache-path=%t -I %S/Inputs %s -fmodules-ts -DWARNING -fmodule-file=%t/werror-no-error.pcm
// RUN: %clang_cc1 -triple %itanium_abi_triple -fmodules -fimplicit-module-maps -verify -fmodules-cache-path=%t -I %S/Inputs %s -fmodules-ts -DWARNING -fmodule-file=%t/werror-no-error.pcm -Wno-padded
[modules] Fix incorrect diagnostic mapping computation when a module changes diagnostic settings using _Pragma within a macro. The AST writer had previously been assuming that all diagnostic state transitions would occur within a FileID corresponding to a file. When a diagnostic state change occured within a macro, it was unable to form a location for that state change and would instead corrupt the diagnostic state of the "root" node (and thus that of the main compilation). Also introduce a "#pragma clang __debug diag_mapping" debugging utility that I added to track this issue down. llvm-svn: 324695
2018-02-09 09:15:13 +08:00
// RUN: %clang_cc1 -triple %itanium_abi_triple -fmodules -fimplicit-module-maps -verify -fmodules-cache-path=%t -I %S/Inputs %s -fmodules-ts -DWARNING -fmodule-file=%t/werror-no-error.pcm -Werror=padded -DLOCAL_ERROR
[modules] Round-trip -Werror flag through explicit module build. The intent for an explicit module build is that the diagnostics produced within the module are those that were configured when the module was built, not those that are enabled within a user of the module. This includes diagnostics that don't actually show up until the module is used (for instance, diagnostics produced during template instantiation and weird cases like -Wpadded). We serialized and restored the diagnostic state for individual warning groups, but previously did not track the state for flags like -Werror and -Weverything, which are implemented as separate bits rather than as part of the diagnostics mapping information. llvm-svn: 301992
2017-05-03 08:28:49 +08:00
//
// RUN: %clang_cc1 -triple %itanium_abi_triple -fmodules -fimplicit-module-maps -emit-module -fmodule-name=diag_flags -x c++ %S/Inputs/module.map -fmodules-ts -o %t/error.pcm -Werror=padded
// RUN: %clang_cc1 -triple %itanium_abi_triple -fmodules -fimplicit-module-maps -verify -fmodules-cache-path=%t -I %S/Inputs %s -fmodules-ts -DERROR -fmodule-file=%t/error.pcm -Wno-padded
// RUN: %clang_cc1 -triple %itanium_abi_triple -fmodules -fimplicit-module-maps -verify -fmodules-cache-path=%t -I %S/Inputs %s -fmodules-ts -DERROR -fmodule-file=%t/error.pcm -Wno-error=padded
//
// RUN: %clang_cc1 -triple %itanium_abi_triple -fmodules -fimplicit-module-maps -emit-module -fmodule-name=diag_flags -x c++ %S/Inputs/module.map -fmodules-ts -o %t/werror.pcm -Werror -Wpadded
// RUN: %clang_cc1 -triple %itanium_abi_triple -fmodules -fimplicit-module-maps -verify -fmodules-cache-path=%t -I %S/Inputs %s -fmodules-ts -DERROR -fmodule-file=%t/werror.pcm -Wno-error
// RUN: %clang_cc1 -triple %itanium_abi_triple -fmodules -fimplicit-module-maps -verify -fmodules-cache-path=%t -I %S/Inputs %s -fmodules-ts -DERROR -fmodule-file=%t/werror.pcm -Wno-padded
Fix initial diagnostic state setup for an explicit module with no diagnostic pragmas. If a file has no diagnostic pragmas, we build its diagnostic state lazily, but in this case we never set up the root state to be the diagnostic state in which the module was originally built, so the diagnostic flags for files in the module with no diagnostic pragmas were incorrectly based on the user of the module rather than the diagnostic state when the module was built. llvm-svn: 301846
2017-05-02 06:10:47 +08:00
import diag_flags;
// Diagnostic flags from the module user make no difference to diagnostics
// emitted within the module when using an explicitly-loaded module.
[modules] Round-trip -Werror flag through explicit module build. The intent for an explicit module build is that the diagnostics produced within the module are those that were configured when the module was built, not those that are enabled within a user of the module. This includes diagnostics that don't actually show up until the module is used (for instance, diagnostics produced during template instantiation and weird cases like -Wpadded). We serialized and restored the diagnostic state for individual warning groups, but previously did not track the state for flags like -Werror and -Weverything, which are implemented as separate bits rather than as part of the diagnostics mapping information. llvm-svn: 301992
2017-05-03 08:28:49 +08:00
#if ERROR
[modules] Fix incorrect diagnostic mapping computation when a module changes diagnostic settings using _Pragma within a macro. The AST writer had previously been assuming that all diagnostic state transitions would occur within a FileID corresponding to a file. When a diagnostic state change occured within a macro, it was unable to form a location for that state change and would instead corrupt the diagnostic state of the "root" node (and thus that of the main compilation). Also introduce a "#pragma clang __debug diag_mapping" debugging utility that I added to track this issue down. llvm-svn: 324695
2018-02-09 09:15:13 +08:00
// expected-error@diag_flags.h:4 {{padding struct}}
[modules] Round-trip -Werror flag through explicit module build. The intent for an explicit module build is that the diagnostics produced within the module are those that were configured when the module was built, not those that are enabled within a user of the module. This includes diagnostics that don't actually show up until the module is used (for instance, diagnostics produced during template instantiation and weird cases like -Wpadded). We serialized and restored the diagnostic state for individual warning groups, but previously did not track the state for flags like -Werror and -Weverything, which are implemented as separate bits rather than as part of the diagnostics mapping information. llvm-svn: 301992
2017-05-03 08:28:49 +08:00
#elif WARNING
[modules] Fix incorrect diagnostic mapping computation when a module changes diagnostic settings using _Pragma within a macro. The AST writer had previously been assuming that all diagnostic state transitions would occur within a FileID corresponding to a file. When a diagnostic state change occured within a macro, it was unable to form a location for that state change and would instead corrupt the diagnostic state of the "root" node (and thus that of the main compilation). Also introduce a "#pragma clang __debug diag_mapping" debugging utility that I added to track this issue down. llvm-svn: 324695
2018-02-09 09:15:13 +08:00
// expected-warning@diag_flags.h:4 {{padding struct}}
Fix initial diagnostic state setup for an explicit module with no diagnostic pragmas. If a file has no diagnostic pragmas, we build its diagnostic state lazily, but in this case we never set up the root state to be the diagnostic state in which the module was originally built, so the diagnostic flags for files in the module with no diagnostic pragmas were incorrectly based on the user of the module rather than the diagnostic state when the module was built. llvm-svn: 301846
2017-05-02 06:10:47 +08:00
#endif
Fix this test to use a construct that actually forces struct layout to happen when testing -Wpadded. llvm-svn: 306349
2017-06-27 08:22:07 +08:00
int arr[sizeof(Padded)];
[modules] Fix incorrect diagnostic mapping computation when a module changes diagnostic settings using _Pragma within a macro. The AST writer had previously been assuming that all diagnostic state transitions would occur within a FileID corresponding to a file. When a diagnostic state change occured within a macro, it was unable to form a location for that state change and would instead corrupt the diagnostic state of the "root" node (and thus that of the main compilation). Also introduce a "#pragma clang __debug diag_mapping" debugging utility that I added to track this issue down. llvm-svn: 324695
2018-02-09 09:15:13 +08:00
// Diagnostic flags from the module make no difference to diagnostics emitted
// in the module user.
#if LOCAL_ERROR
// expected-error@+4 {{padding struct}}
#elif LOCAL_WARNING
// expected-warning@+2 {{padding struct}}
#endif
struct Padded2 { char x; int y; };
int arr2[sizeof(Padded2)];
#if !ERROR && !WARNING && !LOCAL_ERROR && !LOCAL_WARNING
// expected-no-diagnostics
#endif