Move interceptors for libresolv functions to MSan

The functions are not relevant for most sanitizers and only required for
MSan to see which regions have been written to. This eliminates a link
dependency for all other sanitizers and fixes #59007: while `-lresolv`
had been added for the static runtime in 6dce56b, it wasn't added
to the shared runtimes.

Instead of just moving the interceptors, we adapt them to MSan
conventions:
* We don't skip intercepting when `msan_init_is_running` is true, but
  directly call ENSURE_MSAN_INITED() like most other interceptors. It
  seems unlikely that these functions are called during initialization.
* We don't unpoison `errno`, because none of the functions is specified
  to use it.

Author: aaronpuchert

clang/lib/Driver/ToolChains/CommonArgs.cpp

@@ -1410,7 +1410,7 @@ void tools::linkSanitizerRuntimeDeps(const ToolChain &TC,
   // libresolv.a, even if exists, is an empty archive to satisfy POSIX -lresolv
   // requirement.
   if (TC.getTriple().isOSLinux() && !TC.getTriple().isAndroid() &&
-      !TC.getTriple().isMusl())
+      !TC.getTriple().isMusl() && TC.getSanitizerArgs(Args).needsMsanRt())
     CmdArgs.push_back("-lresolv");
 }
 

compiler-rt/lib/msan/msan_interceptors.cpp

@@ -1358,6 +1358,79 @@ INTERCEPTOR(int, forkpty, int *aparent, char *name, const void *termp,
 #define MSAN_MAYBE_INTERCEPT_FORKPTY
 #endif
 
+#if SANITIZER_LINUX && !SANITIZER_ANDROID
+INTERCEPTOR(int, __b64_ntop, unsigned char const *src, SIZE_T srclength,
+            char *target, SIZE_T targsize) {
+  ENSURE_MSAN_INITED();
+  CHECK_UNPOISONED(src, srclength);
+  InterceptorScope interceptor_scope;
+  int res = REAL(__b64_ntop)(src, srclength, target, targsize);
+  if (res >= 0)
+    __msan_unpoison(target, res + 1);
+  return res;
+}
+INTERCEPTOR(int, __b64_pton, char const *src, char *target, SIZE_T targsize) {
+  ENSURE_MSAN_INITED();
+  CHECK_UNPOISONED(src, internal_strlen(src) + 1);
+  InterceptorScope interceptor_scope;
+  int res = REAL(__b64_pton)(src, target, targsize);
+  if (res >= 0)
+    __msan_unpoison(target, res);
+  return res;
+}
+#  define MSAN_MAYBE_INTERCEPT___B64_TO \
+    MSAN_INTERCEPT_FUNC(__b64_ntop);    \
+    COMMON_INTERCEPT_FUNCTION(__b64_pton);
+#else
+#  define MSAN_MAYBE_INTERCEPT___B64_TO
+#endif
+
+#if SANITIZER_LINUX && !SANITIZER_ANDROID
+#  if __GLIBC_PREREQ(2, 34)
+// Changed with https://sourceware.org/git/?p=glibc.git;h=640bbdf
+#    define DN_COMP_INTERCEPTOR_NAME dn_comp
+#    define DN_EXPAND_INTERCEPTOR_NAME dn_expand
+#  else
+#    define DN_COMP_INTERCEPTOR_NAME __dn_comp
+#    define DN_EXPAND_INTERCEPTOR_NAME __dn_expand
+#  endif
+INTERCEPTOR(int, DN_COMP_INTERCEPTOR_NAME, unsigned char *exp_dn,
+            unsigned char *comp_dn, int length, unsigned char **dnptrs,
+            unsigned char **lastdnptr) {
+  ENSURE_MSAN_INITED();
+  InterceptorScope interceptor_scope;
+  int res = REAL(DN_COMP_INTERCEPTOR_NAME)(exp_dn, comp_dn, length, dnptrs,
+                                           lastdnptr);
+  if (res >= 0) {
+    __msan_unpoison(comp_dn, res);
+    if (dnptrs && lastdnptr) {
+      unsigned char **p = dnptrs;
+      for (; p != lastdnptr && *p; ++p);
+      if (p != lastdnptr)
+        ++p;
+      __msan_unpoison(dnptrs, (p - dnptrs) * sizeof(*p));
+    }
+  }
+  return res;
+}
+INTERCEPTOR(int, DN_EXPAND_INTERCEPTOR_NAME, unsigned char const *base,
+            unsigned char const *end, unsigned char const *src, char *dest,
+            int space) {
+  ENSURE_MSAN_INITED();
+  // TODO: add read check if __dn_comp intercept added
+  InterceptorScope interceptor_scope;
+  int res = REAL(DN_EXPAND_INTERCEPTOR_NAME)(base, end, src, dest, space);
+  if (res >= 0)
+    __msan_unpoison(dest, internal_strlen(dest) + 1);
+  return res;
+}
+#  define MSAN_MAYBE_INTERCEPT_DN_COMP_EXPAND      \
+    MSAN_INTERCEPT_FUNC(DN_COMP_INTERCEPTOR_NAME); \
+    MSAN_INTERCEPT_FUNC(DN_EXPAND_INTERCEPTOR_NAME);
+#else
+#  define MSAN_MAYBE_INTERCEPT_DN_COMP_EXPAND
+#endif
+
 struct MSanInterceptorContext {
   bool in_interceptor_scope;
 };
@@ -1916,6 +1989,9 @@ void InitializeInterceptors() {
   MSAN_MAYBE_INTERCEPT_OPENPTY;
   MSAN_MAYBE_INTERCEPT_FORKPTY;
 
+  MSAN_MAYBE_INTERCEPT___B64_TO;
+  MSAN_MAYBE_INTERCEPT_DN_COMP_EXPAND;
+
   inited = 1;
 }
 } // namespace __msan

compiler-rt/lib/sanitizer_common/sanitizer_common_interceptors.inc

@@ -2538,82 +2538,6 @@ INTERCEPTOR(int, glob64, const char *pattern, int flags,
 #define INIT_GLOB64
 #endif  // SANITIZER_INTERCEPT_GLOB64
 
-#if SANITIZER_INTERCEPT___B64_TO
-INTERCEPTOR(int, __b64_ntop, unsigned char const *src, SIZE_T srclength,
-            char *target, SIZE_T targsize) {
-  void *ctx;
-  COMMON_INTERCEPTOR_ENTER(ctx, __b64_ntop, src, srclength, target, targsize);
-  COMMON_INTERCEPTOR_READ_RANGE(ctx, src, srclength);
-  int res = REAL(__b64_ntop)(src, srclength, target, targsize);
-  if (res >= 0)
-    COMMON_INTERCEPTOR_WRITE_RANGE(ctx, target, res + 1);
-  return res;
-}
-INTERCEPTOR(int, __b64_pton, char const *src, char *target, SIZE_T targsize) {
-  void *ctx;
-  COMMON_INTERCEPTOR_ENTER(ctx, __b64_pton, src, target, targsize);
-  COMMON_INTERCEPTOR_READ_RANGE(ctx, src, internal_strlen(src) + 1);
-  int res = REAL(__b64_pton)(src, target, targsize);
-  if (res >= 0)
-    COMMON_INTERCEPTOR_WRITE_RANGE(ctx, target, res);
-  return res;
-}
-#define INIT___B64_TO                      \
-    COMMON_INTERCEPT_FUNCTION(__b64_ntop); \
-    COMMON_INTERCEPT_FUNCTION(__b64_pton);
-#else  // SANITIZER_INTERCEPT___B64_TO
-#define INIT___B64_TO
-#endif  // SANITIZER_INTERCEPT___B64_TO
-
-#if SANITIZER_INTERCEPT_DN_COMP_EXPAND
-#  if __GLIBC_PREREQ(2, 34)
-// Changed with https://sourceware.org/git/?p=glibc.git;h=640bbdf
-#    define DN_COMP_INTERCEPTOR_NAME dn_comp
-#    define DN_EXPAND_INTERCEPTOR_NAME dn_expand
-#  else
-#    define DN_COMP_INTERCEPTOR_NAME __dn_comp
-#    define DN_EXPAND_INTERCEPTOR_NAME __dn_expand
-#  endif
-INTERCEPTOR(int, DN_COMP_INTERCEPTOR_NAME, unsigned char *exp_dn,
-            unsigned char *comp_dn, int length, unsigned char **dnptrs,
-            unsigned char **lastdnptr) {
-  void *ctx;
-  COMMON_INTERCEPTOR_ENTER(ctx, DN_COMP_INTERCEPTOR_NAME, exp_dn, comp_dn,
-                           length, dnptrs, lastdnptr);
-  int res = REAL(DN_COMP_INTERCEPTOR_NAME)(exp_dn, comp_dn, length, dnptrs,
-                                           lastdnptr);
-  if (res >= 0) {
-    COMMON_INTERCEPTOR_WRITE_RANGE(ctx, comp_dn, res);
-    if (dnptrs && lastdnptr) {
-      unsigned char **p = dnptrs;
-      for (; p != lastdnptr && *p; ++p)
-        ;
-      if (p != lastdnptr)
-        ++p;
-      COMMON_INTERCEPTOR_WRITE_RANGE(ctx, dnptrs, (p - dnptrs) * sizeof(*p));
-    }
-  }
-  return res;
-}
-INTERCEPTOR(int, DN_EXPAND_INTERCEPTOR_NAME, unsigned char const *base,
-            unsigned char const *end, unsigned char const *src, char *dest,
-            int space) {
-  void *ctx;
-  COMMON_INTERCEPTOR_ENTER(ctx, DN_EXPAND_INTERCEPTOR_NAME, base, end, src,
-                           dest, space);
-  // TODO: add read check if __dn_comp intercept added
-  int res = REAL(DN_EXPAND_INTERCEPTOR_NAME)(base, end, src, dest, space);
-  if (res >= 0)
-    COMMON_INTERCEPTOR_WRITE_RANGE(ctx, dest, internal_strlen(dest) + 1);
-  return res;
-}
-#  define INIT_DN_COMP_EXPAND                            \
-    COMMON_INTERCEPT_FUNCTION(DN_COMP_INTERCEPTOR_NAME); \
-    COMMON_INTERCEPT_FUNCTION(DN_EXPAND_INTERCEPTOR_NAME);
-#else  // SANITIZER_INTERCEPT_DN_COMP_EXPAND
-#  define INIT_DN_COMP_EXPAND
-#endif  // SANITIZER_INTERCEPT_DN_COMP_EXPAND
-
 #if SANITIZER_INTERCEPT_POSIX_SPAWN
 
 template <class RealSpawnPtr>
@@ -10346,8 +10270,6 @@ static void InitializeCommonInterceptors() {
   INIT_TIMESPEC_GET;
   INIT_GLOB;
   INIT_GLOB64;
-  INIT___B64_TO;
-  INIT_DN_COMP_EXPAND;
   INIT_POSIX_SPAWN;
   INIT_WAIT;
   INIT_WAIT4;

compiler-rt/lib/sanitizer_common/sanitizer_platform_interceptors.h

@@ -266,8 +266,6 @@ SANITIZER_WEAK_IMPORT void *aligned_alloc(__sanitizer::usize __alignment,
 #define SANITIZER_INTERCEPT_TIMESPEC_GET SI_LINUX
 #define SANITIZER_INTERCEPT_GLOB (SI_GLIBC || SI_SOLARIS)
 #define SANITIZER_INTERCEPT_GLOB64 SI_GLIBC
-#define SANITIZER_INTERCEPT___B64_TO SI_LINUX_NOT_ANDROID
-#define SANITIZER_INTERCEPT_DN_COMP_EXPAND SI_LINUX_NOT_ANDROID
 #define SANITIZER_INTERCEPT_POSIX_SPAWN SI_POSIX
 #define SANITIZER_INTERCEPT_WAIT SI_POSIX
 #define SANITIZER_INTERCEPT_INET SI_POSIX

compiler-rt/test/sanitizer_common/TestCases/Linux/dn_expand.cpp renamed to compiler-rt/test/msan/Linux/dn_expand.cpp

@@ -4,7 +4,7 @@
 #include <resolv.h>
 #include <string.h>
 
-#include "sanitizer_common/sanitizer_specific.h"
+#include <sanitizer/msan_interface.h>
 
 void testWrite() {
   char unsigned input[] = {0xff, 0xc5, 0xf7, 0xff, 0x00, 0x00, 0xff, 0x0a, 0x00,
@@ -18,7 +18,7 @@ void testWrite() {
 
   assert(res == 12);
   assert(strcmp(output, "google\\.com") == 0);
-  check_mem_is_good(output, strlen(output) + 1);
+  __msan_check_mem_is_initialized(output, strlen(output) + 1);
 }
 
 void testWriteZeroLength() {
@@ -33,7 +33,7 @@ void testWriteZeroLength() {
 
   assert(res == 1);
   assert(strcmp(output, "") == 0);
-  check_mem_is_good(output, strlen(output) + 1);
+  __msan_check_mem_is_initialized(output, strlen(output) + 1);
 }
 
 void testComp() {
@@ -49,17 +49,17 @@ void testComp() {
   char unsigned *n1 = mb;
   int res = dn_comp("llvm.org", mb, me - mb, pb, pe);
   assert(res == 10);
-  check_mem_is_good(mb, res);
+  __msan_check_mem_is_initialized(mb, res);
   // pb is [msg, llvm.org, nullptr]
-  check_mem_is_good(pb, sizeof(*pb) * 3);
+  __msan_check_mem_is_initialized(pb, sizeof(*pb) * 3);
   mb += res;
 
   char unsigned *n2 = mb;
   res = dn_comp("lab.llvm.org", mb, me - mb, pb, pe);
   assert(res == 6);
-  check_mem_is_good(mb, res);
+  __msan_check_mem_is_initialized(mb, res);
   // pb is [msg, llvm.org, lab.llvm.org, nullptr]
-  check_mem_is_good(pb, sizeof(*pb) * 4);
+  __msan_check_mem_is_initialized(pb, sizeof(*pb) * 4);
   mb += res;
 
   {
@@ -69,7 +69,7 @@ void testComp() {
     fprintf(stderr, "%d\n", res);
     assert(res == 10);
     assert(strcmp(output, "llvm.org") == 0);
-    check_mem_is_good(output, strlen(output) + 1);
+    __msan_check_mem_is_initialized(output, strlen(output) + 1);
   }
 
   {
@@ -78,7 +78,7 @@ void testComp() {
 
     assert(res == 6);
     assert(strcmp(output, "lab.llvm.org") == 0);
-    check_mem_is_good(output, strlen(output) + 1);
+    __msan_check_mem_is_initialized(output, strlen(output) + 1);
   }
 }