[Sema] Fix for PR50741

[chrish_ericsson_atx]

Summary

Fixed crash when doing pointer math on a void pointer.

Also, reworked test to use -verify rather than FileCheck.

[chrish_ericsson_atx]

Updated summary.

[abhinavgaba]

@erichkeane, can you please take a look?

[erichkeane]

Two small changes to the test that you can do as a part of committing. Thanks!

clang/test/Sema/unbounded-array-bounds.c

17–19

Rather than actual line numbers, do @-1 here instead (you can do relative values). That way all of these are easy to read when going back through these.

80

As this warning is unrelated here, I'd suggest either casting this entire thing to (void) to suppress this, or just doing -Wno-unused on the command line flags.

[chrish_ericsson_atx]

Addressed review feedback

[erichkeane]

Yep, that works, still approved.

[chrish_ericsson_atx]

Thanks for the hints, @erichkeane ! Especially thanks for pointing out the root cause early on.

[chrish_ericsson_atx]

Delivered as rGfc6ec9b98cf9

[abhinavgaba]

Thanks for fixing this, @chrish_ericsson_atx!

clang/lib/Sema/SemaChecking.cpp

void Sema::CheckArrayAccess(const Expr *BaseExpr, const Expr *IndexExpr,
   if (IsUnboundedArray) {
     if (index.isUnsigned() || !index.isNegative()) {
       const auto &ASTC = getASTContext();
       unsigned AddrBits =
           ASTC.getTargetInfo().getPointerWidth(ASTC.getTargetAddressSpace(
               EffectiveType->getCanonicalTypeInternal()));
       if (index.getBitWidth() < AddrBits)
         index = index.zext(AddrBits);
-      CharUnits ElemCharUnits = ASTC.getTypeSizeInChars(EffectiveType);
-      llvm::APInt ElemBytes(index.getBitWidth(), ElemCharUnits.getQuantity());
+      Optional<CharUnits> ElemCharUnits =
+          ASTC.getTypeSizeInCharsIfKnown(EffectiveType);
+      // PR50741 - If EffectiveType has unknown size (e.g., if it's a void
+      // pointer) bounds-checking isn't meaningful.
+      if (!ElemCharUnits)
+        return;
+      llvm::APInt ElemBytes(index.getBitWidth(), ElemCharUnits->getQuantity());
       // If index has more active bits than address space, we already know
       // we have a bounds violation to warn about.  Otherwise, compute
       // address of (index + 1)th element, and warn about bounds violation
       // only if that address exceeds address space.
       if (index.getActiveBits() <= AddrBits) {
         bool Overflow;
         llvm::APInt Product(index);
         Product += 1;
if (index.isUnsigned() || !index.isNegative()) {
           ASE ? diag::warn_array_index_exceeds_max_addressable_bounds
               : diag::warn_ptr_arith_exceeds_max_addressable_bounds;
 
       // Diag message shows element size in bits and in "bytes" (platform-
       // dependent CharUnits)
       DiagRuntimeBehavior(BaseExpr->getBeginLoc(), BaseExpr,
                           PDiag(DiagID)
                               << toString(index, 10, true) << AddrBits
-                              << (unsigned)ASTC.toBits(ElemCharUnits)
+                              << (unsigned)ASTC.toBits(*ElemCharUnits)
                               << toString(ElemBytes, 10, false)
                               << toString(MaxElems, 10, false)
                               << (unsigned)MaxElems.getLimitedValue(~0U)
                               << IndexExpr->getSourceRange());
 
       if (!ND) {
         // Try harder to find a NamedDecl to point at in the note.
         while (const auto *ASE = dyn_cast<ArraySubscriptExpr>(BaseExpr))

clang/test/Sema/unbounded-array-bounds.c

-// RUN: %clang_cc1 -triple x86_64-pc-linux-gnu -fsyntax-only %s 2>&1 | FileCheck --check-prefix=CHECK-X86-ADDR64 %s  \
-// RUN:              --implicit-check-not 'past the last possible element'
-// RUN: %clang_cc1 -triple i386-pc-linux-gnu   -fsyntax-only %s 2>&1 | FileCheck --check-prefix=CHECK-I386-ADDR32 %s \
-// RUN:              --implicit-check-not 'past the last possible element'
-// RUN: %clang_cc1 -triple avr-pc-linux-gnu    -fsyntax-only %s 2>&1 | FileCheck --check-prefix=CHECK-AVR-ADDR16 %s  \
-// RUN:              --implicit-check-not 'past the last possible element'
+// RUN: %clang_cc1 -triple x86_64-pc-linux-gnu -fsyntax-only -Wno-unused -verify=addr64,expected %s
+// RUN: %clang_cc1 -triple i386-pc-linux-gnu   -fsyntax-only -Wno-unused -verify=addr32,expected %s
+// RUN: %clang_cc1 -triple avr-pc-linux-gnu    -fsyntax-only -Wno-unused -verify=addr16,expected %s
 
 struct S {
   long long a;
   char b;
   long long c;
   short d;
 };
 
-struct S s[];
+struct S s[]; // expected-warning {{tentative array definition}} expected-note {{declared here}} addr16-note {{declared here}}
 
 void f1() {
   ++s[3].a;
   ++s[7073650413200313099].b;
-  // CHECK-X86-ADDR64:  :[[@LINE-1]]:5: warning: array index 7073650413200313099 refers past the last possible element for an array in 64-bit address space containing 256-bit (32-byte) elements (max possible 576460752303423488 elements)
-  // CHECK-I386-ADDR32: :[[@LINE-2]]:5: warning: {{.*}} past the last possible element {{.*}} in 32-bit {{.*}} (max possible 178956970 elements)
-  // CHECK-AVR-ADDR16:  :[[@LINE-3]]:5: warning: {{.*}} past the last possible element {{.*}} in 16-bit {{.*}} (max possible 3276 elements)
+  // addr16-warning@-1 {{array index 7073650413200313099 refers past the last possible element for an array in 16-bit address space containing 160-bit (20-byte) elements (max possible 3276 elements)}}
+  // addr32-warning@-2 {{array index 7073650413200313099 refers past the last possible element for an array in 32-bit address space containing 192-bit (24-byte) elements (max possible 178956970 elements)}}
+  // addr64-warning@-3 {{array index 7073650413200313099 refers past the last possible element for an array in 64-bit address space containing 256-bit (32-byte) elements (max possible 576460752303423488 elements)}}
   ++s[7073650].c;
-  // CHECK-AVR-ADDR16:  :[[@LINE-1]]:5: warning: {{.*}} past the last possible element {{.*}} in 16-bit {{.*}} (max possible 3276 elements)
+  // addr16-warning@-1 {{array index 7073650 refers past the last possible element for an array in 16-bit address space containing 160-bit (20-byte) elements (max possible 3276 elements)}}
 }
 
-long long ll[];
+long long ll[]; // expected-warning {{tentative array definition}} expected-note {{declared here}} addr16-note {{declared here}} addr32-note {{declared here}}
 
 void f2() {
   ++ll[3];
   ++ll[2705843009213693952];
-  // CHECK-X86-ADDR64:  :[[@LINE-1]]:5: warning: {{.*}} past the last possible element {{.*}} in 64-bit {{.*}} (max possible 2305843009213693952 elements)
-  // CHECK-I386-ADDR32: :[[@LINE-2]]:5: warning: {{.*}} past the last possible element {{.*}} in 32-bit {{.*}} (max possible 536870912 elements)
-  // CHECK-AVR-ADDR16:  :[[@LINE-3]]:5: warning: {{.*}} past the last possible element {{.*}} in 16-bit {{.*}} (max possible 8192 elements)
+  // addr16-warning@-1 {{array index 2705843009213693952 refers past the last possible element for an array in 16-bit address space containing 64-bit (8-byte) elements (max possible 8192 elements)}}
+  // addr32-warning@-2 {{array index 2705843009213693952 refers past the last possible element for an array in 32-bit address space containing 64-bit (8-byte) elements (max possible 536870912 elements)}}
+  // addr64-warning@-3 {{array index 2705843009213693952 refers past the last possible element for an array in 64-bit address space containing 64-bit (8-byte) elements (max possible 2305843009213693952 elements)}}
   ++ll[847073650];
-  // CHECK-I386-ADDR32: :[[@LINE-1]]:5: warning: {{.*}} past the last possible element {{.*}} in 32-bit {{.*}} (max possible 536870912 elements)
-  // CHECK-AVR-ADDR16:  :[[@LINE-2]]:5: warning: {{.*}} past the last possible element {{.*}} in 16-bit {{.*}} (max possible 8192 elements)
+  // addr16-warning@-1 {{array index 847073650 refers past the last possible element for an array in 16-bit address space containing 64-bit (8-byte) elements (max possible 8192 elements)}}
+  // addr32-warning@-2 {{array index 847073650 refers past the last possible element for an array in 32-bit address space containing 64-bit (8-byte) elements (max possible 536870912 elements)}}
 }
 
-void f3(struct S p[]) {
+void f3(struct S p[]) { // expected-note {{declared here}} addr16-note {{declared here}}
   ++p[3].a;
   ++p[7073650413200313099].b;
-  // CHECK-X86-ADDR64:  :[[@LINE-1]]:5: warning: {{.*}} past the last possible element {{.*}} in 64-bit {{.*}} (max possible 576460752303423488 elements)
-  // CHECK-I386-ADDR32: :[[@LINE-2]]:5: warning: {{.*}} past the last possible element {{.*}} in 32-bit {{.*}} (max possible 178956970 elements)
-  // CHECK-AVR-ADDR16:  :[[@LINE-3]]:5: warning: {{.*}} past the last possible element {{.*}} in 16-bit {{.*}} (max possible 3276 elements)
+  // addr16-warning@-1 {{array index 7073650413200313099 refers past the last possible element for an array in 16-bit address space containing 160-bit (20-byte) elements (max possible 3276 elements)}}
+  // addr32-warning@-2 {{array index 7073650413200313099 refers past the last possible element for an array in 32-bit address space containing 192-bit (24-byte) elements (max possible 178956970 elements)}}
+  // addr64-warning@-3 {{array index 7073650413200313099 refers past the last possible element for an array in 64-bit address space containing 256-bit (32-byte) elements (max possible 576460752303423488 elements)}}
   ++p[7073650].c;
-  // CHECK-AVR-ADDR16:  :[[@LINE-1]]:5: warning: {{.*}} past the last possible element {{.*}} in 16-bit {{.*}} (max possible 3276 elements)
+  // addr16-warning@-1 {{array index 7073650 refers past the last possible element for an array in 16-bit address space containing 160-bit (20-byte) elements (max possible 3276 elements)}}
 }
 
-void f4(struct S *p) {
+void f4(struct S *p) { // expected-note {{declared here}} addr16-note {{declared here}}
   p += 3;
   p += 7073650413200313099;
-  // CHECK-X86-ADDR64:  :[[@LINE-1]]:3: warning: the pointer incremented by 7073650413200313099 refers past the last possible element for an array in 64-bit address space containing 256-bit (32-byte) elements (max possible 576460752303423488 elements)
-  // CHECK-I386-ADDR32: :[[@LINE-2]]:3: warning: {{.*}} past the last possible element {{.*}} in 32-bit {{.*}} (max possible 178956970 elements)
-  // CHECK-AVR-ADDR16:  :[[@LINE-3]]:3: warning: {{.*}} past the last possible element {{.*}} in 16-bit {{.*}} (max possible 3276 elements)
+  // addr16-warning@-1 {{the pointer incremented by 7073650413200313099 refers past the last possible element for an array in 16-bit address space containing 160-bit (20-byte) elements (max possible 3276 elements)}}
+  // addr32-warning@-2 {{the pointer incremented by 7073650413200313099 refers past the last possible element for an array in 32-bit address space containing 192-bit (24-byte) elements (max possible 178956970 elements)}}
+  // addr64-warning@-3 {{the pointer incremented by 7073650413200313099 refers past the last possible element for an array in 64-bit address space containing 256-bit (32-byte) elements (max possible 576460752303423488 elements)}}
   p += 7073650;
-  // CHECK-AVR-ADDR16:  :[[@LINE-1]]:3: warning: {{.*}} past the last possible element {{.*}} in 16-bit {{.*}} (max possible 3276 elements)
+  // addr16-warning@-1 {{the pointer incremented by 7073650 refers past the last possible element for an array in 16-bit address space containing 160-bit (20-byte) elements (max possible 3276 elements)}}
 }
 
 struct BQ {
   struct S bigblock[3276];
 };
 
-struct BQ bq[];
+struct BQ bq[]; // expected-warning {{tentative array definition}} addr16-note {{declared here}}
 
 void f5() {
   ++bq[0].bigblock[0].a;
   ++bq[1].bigblock[0].a;
-  // CHECK-AVR-ADDR16:  :[[@LINE-1]]:5: warning: {{.*}} past the last possible element {{.*}} in 16-bit {{.*}} (max possible 1 element)
+  // addr16-warning@-1 {{array index 1 refers past the last possible element for an array in 16-bit address space containing 524160-bit (65520-byte) elements (max possible 1 element)}}
 }
 
 void f6() {
   int ints[] = {1, 3, 5, 7, 8, 6, 4, 5, 9};
   int const n_ints = sizeof(ints) / sizeof(int);
   unsigned long long const N = 3;
 
   int *middle = &ints[0] + n_ints / 2;
   // Should NOT produce a warning.
   *(middle + 5 - N) = 22;
 }
+
+void pr50741() {
+  (void *)0 + 0xdead000000000000UL;
+  // no array-bounds warning, and no crash
+}