https://github.com/cyndyishida updated
https://github.com/llvm/llvm-project/pull/94508
>From c547d990aca29ecfe6f51d37c5c3f8712dfc5e44 Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Fri, 10 May 2024 09:19:22 -0700
Subject: [PATCH 1/2] [InstallAPI] Pick up input headers by directory traversal
@@ -97,6 +97,14 @@ class HeaderFile {
Other.Excluded, Other.Extra,
Other.Umbrella);
}
+
+ bool operator<(const HeaderFile &Other) const {
cyndyishida wrote:
I could infer, b
https://github.com/cyndyishida updated
https://github.com/llvm/llvm-project/pull/94508
>From 95cc0c9a2b135706e80a5e73ef5e4257aa89984c Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Fri, 10 May 2024 09:19:22 -0700
Subject: [PATCH] [InstallAPI] Pick up input headers by directory traversal
Mat
cyndyishida wrote:
ping
https://github.com/llvm/llvm-project/pull/94508
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/cyndyishida closed
https://github.com/llvm/llvm-project/pull/94508
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/cyndyishida created
https://github.com/llvm/llvm-project/pull/88750
Apple's ld supports alias_lists, described as
```
-alias_list filename
The specified filename contains a list of aliases. The symbol name
and its alias are on one
line, separate
https://github.com/cyndyishida updated
https://github.com/llvm/llvm-project/pull/88750
>From d7a4e72f80a9dcd18770dec1e06fd54518b60233 Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Thu, 11 Apr 2024 13:13:45 -0700
Subject: [PATCH] [InstallAPI] Add support for aliased exports
Apple's ld suppo
https://github.com/cyndyishida updated
https://github.com/llvm/llvm-project/pull/88750
>From b2b54ad8ccf930a19327563a64c52c9f4013103f Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Thu, 11 Apr 2024 13:13:45 -0700
Subject: [PATCH] [InstallAPI] Add support for aliased exports
Apple's ld suppo
@@ -973,5 +978,24 @@ bool DylibVerifier::verifyBinaryAttrs(const
ArrayRef ProvidedTargets,
return true;
}
+std::unique_ptr DylibVerifier::getExports() {
+ for (const auto &[Alias, Base] : Aliases) {
cyndyishida wrote:
`getExports()` only gets called once
https://github.com/cyndyishida updated
https://github.com/llvm/llvm-project/pull/88750
>From b2b54ad8ccf930a19327563a64c52c9f4013103f Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Thu, 11 Apr 2024 13:13:45 -0700
Subject: [PATCH 1/2] [InstallAPI] Add support for aliased exports
Apple's ld s
https://github.com/cyndyishida updated
https://github.com/llvm/llvm-project/pull/88750
>From b2b54ad8ccf930a19327563a64c52c9f4013103f Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Thu, 11 Apr 2024 13:13:45 -0700
Subject: [PATCH 1/3] [InstallAPI] Add support for aliased exports
Apple's ld s
cyndyishida wrote:
ping, is there anything else needed to land this patch?
https://github.com/llvm/llvm-project/pull/88750
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/cyndyishida updated
https://github.com/llvm/llvm-project/pull/88750
>From b2b54ad8ccf930a19327563a64c52c9f4013103f Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Thu, 11 Apr 2024 13:13:45 -0700
Subject: [PATCH 1/4] [InstallAPI] Add support for aliased exports
Apple's ld s
@@ -259,7 +259,10 @@ bool Options::processLinkerOptions(InputArgList &Args) {
if (auto *Arg = Args.getLastArg(drv::OPT_umbrella))
LinkerOpts.ParentUmbrella = Arg->getValue();
- LinkerOpts.IsDylib = Args.hasArg(drv::OPT_dynamiclib);
cyndyishida wrote:
n
https://github.com/cyndyishida closed
https://github.com/llvm/llvm-project/pull/88750
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/cyndyishida created
https://github.com/llvm/llvm-project/pull/89270
None
>From efefd04b1545b62f3a22bc0ff53bfb0a2a512b1f Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Thu, 18 Apr 2024 10:24:35 -0700
Subject: [PATCH] [InstallAPI][Tests] Update tests to be resilient to reve
cyndyishida wrote:
> It looks like this is causing failures on the reverse-iteration buildbot
> (https://lab.llvm.org/buildbot/#/builders/54/builds/9683)
Thanks for the heads up, should be fixed by
https://github.com/llvm/llvm-project/pull/89270
https://github.com/llvm/llvm-project/pull/87674
cyndyishida wrote:
> How hard would it be to make the output here have a consistent order? I mean,
> emitting diagnostics in a consistent order isn't nearly as important as
> emitting consistent binaries, but it's still nice to have if we can get it
> easily.
I don't think it's too hard in ge
https://github.com/cyndyishida closed
https://github.com/llvm/llvm-project/pull/89270
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/cyndyishida created
https://github.com/llvm/llvm-project/pull/89483
Matches up with the clang tools
>From 76ece379c8a8325b0f866a89e1de523cbeaa0c25 Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Fri, 19 Apr 2024 19:34:14 -0700
Subject: [PATCH] [clang][darwin][cmake] Disabl
https://github.com/cyndyishida edited
https://github.com/llvm/llvm-project/pull/89483
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/cyndyishida closed
https://github.com/llvm/llvm-project/pull/89483
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
cyndyishida wrote:
It appears your PR branch is based on a stale version of the target `llvm/main`
branch, resulting in more commits than what this PR intended for. Please merge
in the target branch and resolve any conflicts.
https://github.com/llvm/llvm-project/pull/89480
___
https://github.com/cyndyishida closed
https://github.com/llvm/llvm-project/pull/99520
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/cyndyishida approved this pull request.
https://github.com/llvm/llvm-project/pull/100126
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/cyndyishida created
https://github.com/llvm/llvm-project/pull/100636
None
>From f96d9b79d7efb744182a5016295e6dd305564968 Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Thu, 25 Jul 2024 12:15:45 -0700
Subject: [PATCH] [InstallAPI] Fix early return for scanning sub director
@@ -0,0 +1,300 @@
+//===- DirectoryScanner.cpp
---===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM
Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Ap
https://github.com/cyndyishida closed
https://github.com/llvm/llvm-project/pull/100636
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
cyndyishida wrote:
Thank you @zeroomega ! I am also running on arm64-mac and that test is passing.
What is your `cmake` invocation? I'd like to dig into the underlying cause of
non-stable ordering. It looks like the darwin-aarch64 is passing the test as
well https://lab.llvm.org/buildbot/#/bui
https://github.com/cyndyishida created
https://github.com/llvm/llvm-project/pull/99520
Previously source input like `#import ` resulted in infinite calls append the
same token into `CurDirTokens`. This patch now ignores those directive lines if
they won't actually end up being compiled. (e.g.
Author: Cyndy Ishida
Date: 2024-07-18T10:12:36-07:00
New Revision: 1c55586e9a475a09b7d769e7fc9a254e7150c972
URL:
https://github.com/llvm/llvm-project/commit/1c55586e9a475a09b7d769e7fc9a254e7150c972
DIFF:
https://github.com/llvm/llvm-project/commit/1c55586e9a475a09b7d769e7fc9a254e7150c972.diff
https://github.com/cyndyishida approved this pull request.
Thank you for looking into & resolving this again!
https://github.com/llvm/llvm-project/pull/99399
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailma
@@ -544,7 +544,7 @@ Scanner::lexIncludeFilename(const char *&First, const char
*const End) {
void Scanner::lexPPDirectiveBody(const char *&First, const char *const End) {
while (true) {
const dependency_directives_scan::Token &Tok = lexToken(First, End);
-if (Tok.is(
https://github.com/cyndyishida updated
https://github.com/llvm/llvm-project/pull/99520
>From bcddefdce00a1e15f29181bc92eab86098a8b328 Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Thu, 18 Jul 2024 08:56:24 -0700
Subject: [PATCH 1/2] [clang-scan-deps] Ignore import/include directives with
m
cyndyishida wrote:
> This is similar feedback that I gave for #97654, could you test the change
> with a unit test in `DependencyDirectivesScannerTest.cpp` instead? It's much
> simpler to set up and execution-wise it's orders of magnitude more
> lightweight than a whole new lit test.
>
> For
cyndyishida wrote:
> > I switched to the unit test, though I didn't figure out how to capture the
> > same sort of round-trip testing as with the lit test.
>
> Are you not seeing the same "infinite calls append the same token into
> CurDirTokens" issue via the unit test?
Oh yeah, that part is
cyndyishida wrote:
> Overall I really like the direction of this. I'm curious if any of the
> maintainers from Apple have thoughts since they're the primary users of
> availability annotations.
While I like the approach of aligning availability parameters closer to
`llvm::Triple`, I am concer
https://github.com/cyndyishida approved this pull request.
https://github.com/llvm/llvm-project/pull/104653
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -1592,6 +1592,12 @@ succeeds but Clang emits a warning specifying that the
function is deprecated.
Finally, if Clang is instructed to compile code for macOS 10.7, the call
fails because ``f()`` is no longer available.
+Clang is instructed to compile code for a particular O
@@ -1592,6 +1592,12 @@ succeeds but Clang emits a warning specifying that the
function is deprecated.
Finally, if Clang is instructed to compile code for macOS 10.7, the call
fails because ``f()`` is no longer available.
+Clang is instructed to compile code for a particular O
@@ -1637,31 +1637,42 @@ specifies availability for the current target platform,
the availability
attributes are ignored. Supported platforms are:
``ios``
+``ios_app_extension``
cyndyishida wrote:
I do think it's still useful to define what those platforms' v
https://github.com/cyndyishida created
https://github.com/llvm/llvm-project/pull/98171
`$ld$previous` symbols need to be exported for them to be seen by clients. TAPI
cannot omit them in tbd files, so account for this in installapi verification
when handling reexport verification.
Reviewed in
https://github.com/cyndyishida updated
https://github.com/llvm/llvm-project/pull/98171
>From 03ee88e0b208d42a675dbbcf89ace869d193023b Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Tue, 9 Jul 2024 08:22:53 -0700
Subject: [PATCH] [InstallAPI] Don't look for linker directive symbols in
reexpo
https://github.com/cyndyishida closed
https://github.com/llvm/llvm-project/pull/98171
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/cyndyishida created
https://github.com/llvm/llvm-project/pull/90664
Add validation in the FileList reader to check that the headers exist and use
similar diagnostics in Options.cpp
>From 42ad6921fdf179647f6529ddeb7d060e93e0f52f Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Da
https://github.com/cyndyishida closed
https://github.com/llvm/llvm-project/pull/90664
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/cyndyishida created
https://github.com/llvm/llvm-project/pull/90686
Projects like libc use mutually exclusive macros to compile files multiple
times and then merge the result into the final library. For installapi to
accept these, we'd need to parse the same declarations in
https://github.com/cyndyishida updated
https://github.com/llvm/llvm-project/pull/90686
>From c4fec501607059b1030b7c53794ad271194d Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Thu, 18 Apr 2024 07:28:44 -0700
Subject: [PATCH 1/2] [InstallAPI] Support mutually exclusive parse options
Pro
https://github.com/cyndyishida updated
https://github.com/llvm/llvm-project/pull/90686
>From faabe7752dbb596f13e0e676b5c8db9fb8ad9258 Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Thu, 18 Apr 2024 07:28:44 -0700
Subject: [PATCH] [InstallAPI] Support mutually exclusive parse options
Project
cyndyishida wrote:
ping
https://github.com/llvm/llvm-project/pull/90686
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/cyndyishida closed
https://github.com/llvm/llvm-project/pull/90686
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/cyndyishida created
https://github.com/llvm/llvm-project/pull/91770
None
>From 480f1a6dd0c78c8018ac08259b4d3cba64c25165 Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Thu, 2 May 2024 19:53:07 -0700
Subject: [PATCH] [InstallAPI] add JSON option to pass X arguments
---
..
Author: Cyndy Ishida
Date: 2024-05-10T17:17:38-07:00
New Revision: 504cf554639360525c3f746e7296a242350b2af9
URL:
https://github.com/llvm/llvm-project/commit/504cf554639360525c3f746e7296a242350b2af9
DIFF:
https://github.com/llvm/llvm-project/commit/504cf554639360525c3f746e7296a242350b2af9.diff
https://github.com/cyndyishida updated
https://github.com/llvm/llvm-project/pull/91770
>From 291412a203ea60465d4ecae9317f3490c59bfb50 Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Thu, 2 May 2024 19:53:07 -0700
Subject: [PATCH 1/2] [InstallAPI] add JSON option to pass X arguments
---
.../
https://github.com/cyndyishida updated
https://github.com/llvm/llvm-project/pull/91770
>From 291412a203ea60465d4ecae9317f3490c59bfb50 Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Thu, 2 May 2024 19:53:07 -0700
Subject: [PATCH 1/3] [InstallAPI] add JSON option to pass X arguments
---
.../
https://github.com/cyndyishida closed
https://github.com/llvm/llvm-project/pull/91770
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/cyndyishida approved this pull request.
the InstallAPI changes LGTM, thanks!
https://github.com/llvm/llvm-project/pull/93190
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe
https://github.com/cyndyishida approved this pull request.
LGTM
https://github.com/llvm/llvm-project/pull/102239
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Author: Cyndy Ishida
Date: 2021-04-05T10:24:42-07:00
New Revision: 0116d04d04f20e9ae62ba847075840c3cb298080
URL:
https://github.com/llvm/llvm-project/commit/0116d04d04f20e9ae62ba847075840c3cb298080
DIFF:
https://github.com/llvm/llvm-project/commit/0116d04d04f20e9ae62ba847075840c3cb298080.diff
Author: Cyndy Ishida
Date: 2021-10-15T09:12:31-07:00
New Revision: 395e1fe3057482576eb0bc38d1079fcc1ff29193
URL:
https://github.com/llvm/llvm-project/commit/395e1fe3057482576eb0bc38d1079fcc1ff29193
DIFF:
https://github.com/llvm/llvm-project/commit/395e1fe3057482576eb0bc38d1079fcc1ff29193.diff
@@ -1637,31 +1637,42 @@ specifies availability for the current target platform,
the availability
attributes are ignored. Supported platforms are:
``ios``
+``ios_app_extension``
cyndyishida wrote:
The description only describes the platform. Can you also incl
@@ -1637,31 +1637,42 @@ specifies availability for the current target platform,
the availability
attributes are ignored. Supported platforms are:
``ios``
+``ios_app_extension``
Apple's iOS operating system. The minimum deployment target is specified
as part of the ``-ta
@@ -1637,31 +1637,42 @@ specifies availability for the current target platform,
the availability
attributes are ignored. Supported platforms are:
``ios``
+``ios_app_extension``
Apple's iOS operating system. The minimum deployment target is specified
as part of the ``-ta
@@ -2953,7 +2953,15 @@ static bool sdkSupportsBuiltinModules(
case Darwin::MacOS:
return SDKVersion >= VersionTuple(15U);
cyndyishida wrote:
The extra test coverage is more for future-proofing. There is a decent amount
of `TargetVariantTriple` logic happ
https://github.com/cyndyishida reopened
https://github.com/llvm/llvm-project/pull/115048
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
cyndyishida wrote:
> `darwin::Linker::ConstructJob`
https://github.com/llvm/llvm-project/pull/115048
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
cyndyishida wrote:
> Do we also need to update `darwin::Linker::ConstructJob(...)`?
I thought about it, but clang already doesn't for today's default paths so I
decided against it. It seems only pass extra paths to the linker for
platform-specific directories like `DriverKit`. Xcode's `ld` pas
https://github.com/cyndyishida closed
https://github.com/llvm/llvm-project/pull/115048
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/cyndyishida created
https://github.com/llvm/llvm-project/pull/115048
* Have clang always append & pass `System/Library/SubFrameworks` when
determining default sdk search paths.
* Teach `clang-installapi` to traverse there for framework input.
* Teach `llvm-readtapi` that the
https://github.com/cyndyishida updated
https://github.com/llvm/llvm-project/pull/115048
>From 9359df8624a75b7c0c361fdfcd9fc9102824dc68 Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Tue, 5 Nov 2024 10:48:09 -0800
Subject: [PATCH 1/2] [Clang][Darwin] Introduce `SubFrameworks` as a SDK
defaul
cyndyishida wrote:
```
> git checkout -f baecb5c50287efebf2482739bfbe320c8147b7ff
error: unable to unlink old 'clang/lib/AST/Decl.cpp': Invalid argument
error: unable to unlink old 'clang/lib/Serialization/ASTReaderDecl.cpp':
Invalid argument
```
Seems like an unrelated issue is happening on the
https://github.com/cyndyishida updated
https://github.com/llvm/llvm-project/pull/115048
>From dd22f1347530d8f866822074ec9b89cb8ee56867 Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Tue, 5 Nov 2024 10:48:09 -0800
Subject: [PATCH 1/2] [Clang][Darwin] Introduce `SubFrameworks` as a SDK
defaul
https://github.com/cyndyishida updated
https://github.com/llvm/llvm-project/pull/115048
>From 9359df8624a75b7c0c361fdfcd9fc9102824dc68 Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Tue, 5 Nov 2024 10:48:09 -0800
Subject: [PATCH 1/3] [Clang][Darwin] Introduce `SubFrameworks` as a SDK
defaul
https://github.com/cyndyishida approved this pull request.
I initially touched this code because I needed to be able to map back input
headers from a destination location (DSTROOT) to where they were resolved with
headermaps (SRCROOT) for `tapi`/`clang-installapi`. But looking at the code it
o
https://github.com/cyndyishida updated
https://github.com/llvm/llvm-project/pull/115048
>From 9359df8624a75b7c0c361fdfcd9fc9102824dc68 Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Tue, 5 Nov 2024 10:48:09 -0800
Subject: [PATCH 1/4] [Clang][Darwin] Introduce `SubFrameworks` as a SDK
defaul
@@ -347,6 +347,7 @@ void InitHeaderSearch::AddDefaultIncludePaths(
} else {
AddPath("/System/Library/Frameworks", System, true);
AddPath("/Library/Frameworks", System, true);
+AddPath("/System/Library/SubFrameworks", System, true);
@@ -255,16 +255,21 @@ static void stubifyDirectory(const StringRef InputPath,
Context &Ctx) {
if (EC)
reportError(IT->path() + ": " + EC.message());
-// Skip header directories (include/Headers/PrivateHeaders) and module
cyndyishida wrote:
My s
cyndyishida wrote:
ping
https://github.com/llvm/llvm-project/pull/115048
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/cyndyishida edited
https://github.com/llvm/llvm-project/pull/115048
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/cyndyishida created
https://github.com/llvm/llvm-project/pull/110724
None
>From b78a9a6649adaa1b0f7252733995ece7d1cebee0 Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Mon, 30 Sep 2024 17:07:06 -0700
Subject: [PATCH] [clang] Remove 'PCH' from more diagnostics that were wr
https://github.com/cyndyishida closed
https://github.com/llvm/llvm-project/pull/110724
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/cyndyishida updated
https://github.com/llvm/llvm-project/pull/115048
>From 9359df8624a75b7c0c361fdfcd9fc9102824dc68 Mon Sep 17 00:00:00 2001
From: Cyndy Ishida
Date: Tue, 5 Nov 2024 10:48:09 -0800
Subject: [PATCH 1/2] [Clang][Darwin] Introduce `SubFrameworks` as a SDK
defaul
cyndyishida wrote:
> > However, notably, even on Darwin, if you pass _both_ flags, you still get
> > the intended behavior of `-isysroot` affecting headers, and `--sysroot`
> > affecting libraries. Your change would break that.
>
> What's the use-case for this?
>
> Because for most users, you
cyndyishida wrote:
ping
https://github.com/llvm/llvm-project/pull/115048
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -346,6 +346,7 @@ void InitHeaderSearch::AddDefaultIncludePaths(
AddPath("/System/DriverKit/System/Library/Frameworks", System, true);
cyndyishida wrote:
I'd rather discuss DriverKit support separately, so I will merge this pr now.
> Also, why's there
https://github.com/cyndyishida edited
https://github.com/llvm/llvm-project/pull/115048
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/cyndyishida closed
https://github.com/llvm/llvm-project/pull/115048
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/cyndyishida edited
https://github.com/llvm/llvm-project/pull/118543
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
cyndyishida wrote:
> The main concern is it happens when setting `DEFAULT_SYSROOT`. If only
> `isysroot` is supported on macOS, does this mean `DEFAULT_SYSROOT` should be
> deprecated on Apple platforms?
I don't think it's deprecated in the sense we have plans to drop support for
it. The p
@@ -346,6 +346,7 @@ void InitHeaderSearch::AddDefaultIncludePaths(
AddPath("/System/DriverKit/System/Library/Frameworks", System, true);
cyndyishida wrote:
Followup in: https://github.com/llvm/llvm-project/pull/118543
https://github.com/llvm/llvm-projec
@@ -339,13 +340,11 @@ void InitHeaderSearch::AddDefaultIncludePaths(
if (triple.isOSDarwin()) {
cyndyishida wrote:
Looks like that was previously attempted by @ldionne in
https://github.com/llvm/llvm-project/pull/75841
The problem was appending `/Library/F
cyndyishida wrote:
> Sorry, which env var are you referring to?
I'm suggesting an explicit test case that checks that `DEFAULT_SYSROOT` is set
and behaves as expected. It's ignored if sysroot or isysroot is passed, it's
used when they're missing.
https://github.com/llvm/llvm-project/pull/1
https://github.com/cyndyishida approved this pull request.
https://github.com/llvm/llvm-project/pull/120149
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/cyndyishida approved this pull request.
https://github.com/llvm/llvm-project/pull/126007
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -2,6 +2,7 @@ set(LLVM_LINK_COMPONENTS
OrcShared
OrcTargetProcess
Support
+ Core
cyndyishida wrote:
Why did you need to add link dependencies?
https://github.com/llvm/llvm-project/pull/119473
___
cfe-commi
https://github.com/cyndyishida approved this pull request.
LGTM
https://github.com/llvm/llvm-project/pull/119473
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -339,13 +340,11 @@ void InitHeaderSearch::AddDefaultIncludePaths(
if (triple.isOSDarwin()) {
cyndyishida wrote:
I looked into this a bit more. IMO the only way to move all this code to the
driver is to either
1. changing search path order between Framew
@@ -0,0 +1,26 @@
+// UNSUPPORTED: system-windows
+// Windows is unsupported because we use the Unix path separator `/` in the
test.
+
+// Add default directories before running clang to check default
+// search paths.
+// RUN: rm -rf %t && mkdir -p %t
+// RUN: cp -R %S/Inputs/M
https://github.com/cyndyishida edited
https://github.com/llvm/llvm-project/pull/120149
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -2539,6 +2550,18 @@ void DarwinClang::AddClangSystemIncludeArgs(const
llvm::opt::ArgList &DriverArgs
llvm::sys::path::append(P, "usr", "include");
addExternCSystemInclude(DriverArgs, CC1Args, P.str());
}
+
+ // Add default framework search paths
+ auto addFramew
201 - 300 of 344 matches
Mail list logo
