[PATCH] D54438: [analyzer] Reimplement dependencies between checkers

Szelethus added a comment. In D54438#1378425 , @Szelethus wrote: > In D54438#1375858 , > @george.karpenkov wrote: > > > After this landed, options for RetainCountChecker stopped working - e.g. I > > can't use

[PATCH] D54438: [analyzer] Reimplement dependencies between checkers

Szelethus added a comment. In D54438#1375858 , @george.karpenkov wrote: > After this landed, options for RetainCountChecker stopped working - e.g. I > can't use `osx.cocoa.RetainCount:blah=X`. > Do you know why is this the case / how to fix it?

[PATCH] D54438: [analyzer] Reimplement dependencies between checkers

george.karpenkov added a comment. After this landed, options for RetainCountChecker stopped working - e.g. I can't use `osx.cocoa.RetainCount:blah=X`. Do you know why is this the case / how to fix it? Repository: rL LLVM CHANGES SINCE LAST ACTION https://reviews.llvm.org/D54438/new/

[PATCH] D54438: [analyzer] Reimplement dependencies between checkers

uabelho added inline comments. Comment at: cfe/trunk/include/clang/StaticAnalyzer/Checkers/Checkers.td:230 def NullReturnedFromNonnullChecker : Checker<"NullReturnedFromNonnull">, HelpText<"Warns when a null pointer is returned from a function that has "

[PATCH] D54438: [analyzer] Reimplement dependencies between checkers

Szelethus marked an inline comment as done. Szelethus added inline comments. Comment at: cfe/trunk/include/clang/StaticAnalyzer/Checkers/Checkers.td:230 def NullReturnedFromNonnullChecker : Checker<"NullReturnedFromNonnull">, HelpText<"Warns when a null pointer is returned

[PATCH] D54438: [analyzer] Reimplement dependencies between checkers

uabelho added inline comments. Comment at: cfe/trunk/include/clang/StaticAnalyzer/Checkers/Checkers.td:230 def NullReturnedFromNonnullChecker : Checker<"NullReturnedFromNonnull">, HelpText<"Warns when a null pointer is returned from a function that has "

[PATCH] D54438: [analyzer] Reimplement dependencies between checkers

Szelethus added a comment. In D54438#1372324 , @NoQ wrote: > *gets hyped for the upcoming patchlanding party* Oh yeah, super excited about this! It was a blast! > In D54438#1329425 , @Szelethus wrote: > >>

[PATCH] D54438: [analyzer] Reimplement dependencies between checkers

NoQ accepted this revision. NoQ added a comment. *gets hyped for the upcoming patchlanding party* In D54438#1329425 , @Szelethus wrote: > Hmmm, I left plenty of room for improvement in the tblgen generation, we > could generate compile-time errors on

[PATCH] D54438: [analyzer] Reimplement dependencies between checkers

Szelethus updated this revision to Diff 182726. Szelethus added a comment. - Rebase - Resolve the issue mentioned above by not enabling checkers that has any of their dependencies explicitly disabled - Introduce `osx.RetainCountBase` to "solve" the issue mentioned in D55400#1364683

[PATCH] D54438: [analyzer] Reimplement dependencies between checkers

Szelethus added a comment. > I still expect some skeletons to fall out of the closet This patch doesn't handle -analyzer-disable-checker, which leads to assertation failures in later pathes. Since the way which checker should/shouldnt be enabled is implemented rather poorly, I'll probably try

[PATCH] D54438: [analyzer] Reimplement dependencies between checkers

Szelethus added a comment. Hmmm, I left plenty of room for improvement in the tblgen generation, we could generate compile-time errors on cycles within the dependency graph, try to include the generated file only once, but these clearly are very low-prio issues, so I'll do it later. I'll have

[PATCH] D54438: [analyzer] Reimplement dependencies between checkers

Szelethus added a comment. I can't fix these right away, but I don't want myself to forget it before commiting. Pay no attention. Comment at: include/clang/StaticAnalyzer/Checkers/CheckerBase.td:51-56 +/// relies on information MallocBase gathers. +/// Example: +/// def

[PATCH] D54438: [analyzer] Reimplement dependencies between checkers

NoQ added a comment. Also you might need to rebase on top of D55400 . CHANGES SINCE LAST ACTION https://reviews.llvm.org/D54438/new/ https://reviews.llvm.org/D54438 ___ cfe-commits mailing list

[PATCH] D54438: [analyzer] Reimplement dependencies between checkers

NoQ added a comment. Generally, because the primary user of checker names is clang-tidy, the correct name to display is the name of the specific checker that the user needs to disable in order to get rid of the warning. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D54438/new/

[PATCH] D54438: [analyzer] Reimplement dependencies between checkers

NoQ accepted this revision. NoQ added a comment. This revision is now accepted and ready to land. I'm seeing no problems with this patch and i'm happy that we're replacing hacks with well-defined patterns :) In D54438#1322970 , @Szelethus wrote: > -

[PATCH] D54438: [analyzer] Reimplement dependencies between checkers

Szelethus added a comment. In D54438#1315953 , @Szelethus wrote: > - ❌ Move `CheckerManager::registerChecker` out of the registry functions. > - ❌ Since `shouldRegister##CHECKERNAME` is a thing, we can move everything > to the checker's constructor,

[PATCH] D54438: [analyzer] Reimplement dependencies between checkers

Szelethus requested review of this revision. Szelethus added a comment. Please take a second look, this is a pretty important to get right. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D54438/new/ https://reviews.llvm.org/D54438 ___

[PATCH] D54438: [analyzer] Reimplement dependencies between checkers

Szelethus updated this revision to Diff 177189. Szelethus added a comment. - Register the checker after it's dependencies (accidentally left this change in the next branch). This implies that `unix.MallocChecker:Optimistic` is no longer a thing, and is now called

[PATCH] D54438: [analyzer] Reimplement dependencies between checkers

Szelethus updated this revision to Diff 177182. Szelethus retitled this revision from "[analyzer][WIP] Reimplement dependencies between checkers" to "[analyzer] Reimplement dependencies between checkers". Szelethus edited the summary of this revision. Szelethus added reviewers: MTC,