Sorry, wrong patch.

In D63678#1554508, @rnk wrote:

I would almost prefer to XFAIL this on Solaris, I feel like the intent is clearer with echo.

Rebased an slightly fixed version. Without it, man profiling tests on Solaris FAIL like

Command Output (stderr):
--
Undefined                       first referenced
 symbol                             in file
__llvm_profile_register_names_function /tmp/lit_tmp_Nqu4eh/infinite_loop-9dc638.o
__llvm_profile_register_function    /tmp/lit_tmp_Nqu4eh/infinite_loop-9dc638.o

I don't think we need to worry about older versions of Solaris ld that don't support
__start_SECNAME and stop_SECNAME labels any longer: that feature is in
Solaris 11.4 and we already use Solaris 11.4-only features like ld -z gnu-version-script-compat
and fully working .preinit_array suport in compiler-rt.

Further testing in the exact buildbot user environment revealed that I'd forgotten one
crucial option: clang calls the linker with Solaris ld-only options. If GNU ld happens to
be earlier in PATH at link time, this breaks the build.

Ping^2. It's been two weeks now for a trivial patch fixing the build...

In D62944#1542217, @efriedma wrote:

For format-strings.c, I'm not really happy suggesting #if defined(__sun) && !defined(__LP64__), but I don't think the alternative is better. We could restrict the test so it doesn't run using a Solaris target triple, but we actually want coverage here: the difference in wint_t affects the semantics of "%lc", so we want some coverage of that path.

Adapt Sema/format-strings.c.

Ping? This has been a week, too. According to gcc/config headers, there are quite a number of targets with long int for wchar_t/wint_t,
so there should be some generic way to handle the Clang :: Sema/format-strings.c failure. Suggestions?

Ping? It's been a week and the patch is close to obvious.

In D60046#1458399, @mgorny wrote:

Thanks, looks good. Do you have commit access or do you need me to commit it for you?

In D60046#1456034, @mgorny wrote:

Just add it to the regex above.

Skip SPARC like other targets.

Ping? Who's an appropriate reviewer here? CODE_OWNERS.TXT doesn't list anyone.

Ping. Another Solaris-specific patch which might be considered obvious?

Ping. Given that this patch is Solaris-only, I guess it could go in as obvious?

In D60043#1451326, @jdenny wrote:

FileCheck.h needs those other non-LLVM headers. This header is different.

Move <cmath> include to FileCheck.cpp.

In D60043#1449930, @jdenny wrote:

Revised patch.

In D60043#1449881, @jdenny wrote:

In D55596#1334312, @krytarowski wrote:

@ro could you please test this patch for Solaris? I need to know whether it works with TLS destructor on SunOS.

In D54099#1293960, @krytarowski wrote:

There is also sparc and sparc64 support in sanitizers on NetBSD... but I cannot say for now how well it works. For sure we run UBSan and ASan for both of them.

Refreshed.

In D54030#1286927, @marxin wrote:

I just built GCC on gcc202 machine of Compile farm. Even though the target is not supported, the libsanitizer runtime library is still built in GCC.

Overall, this looks good to me, with the nits noted.

In D46006#1078139, @krytarowski wrote:

What are the SunOS results? For NetBSD ASan broke badly and results for other sanitizers (UBSan, TSan, MSan) are deteriorated...

In D46006#1076822, @krytarowski wrote:

Can you run tests on SunOS? check-asan, check-ubsan etc.

ping^5

ping^3

TBH, I've always disliked it that way: my preference would be to go for the SANITIZE_<PLATFORM> forms as far as possible.