[PATCH] c-family: char8_t and aliasing in C vs C++ [PR111884]

[PATCH] c-family: char8_t and aliasing in C vs C++ [PR111884]

[Marek Polacek]

Bootstrapped/regtested on x86_64-pc-linux-gnu, ok for trunk?

-- >8 --
In the PR, Joseph says that in C char8_t is not a distinct type.  So
we should behave as if it can alias anything, like ordinary char.
In C, unsigned_char_type_node == char8_type_node, so with this patch
we return 0 instead of -1.  And the following comment says:

  /* The C standard guarantees that any object may be accessed via an
     lvalue that has narrow character type (except char8_t).  */
  if (t == char_type_node
      || t == signed_char_type_node
      || t == unsigned_char_type_node)
    return 0;

Which appears to be wrong, so I'm adjusting that as well.

	PR c/111884

gcc/c-family/ChangeLog:

	* c-common.cc (c_common_get_alias_set): Return -1 for char8_t only
	in C++.

gcc/testsuite/ChangeLog:

	* c-c++-common/alias-1.c: New test.
---
 gcc/c-family/c-common.cc             |  7 ++++---
 gcc/testsuite/c-c++-common/alias-1.c | 23 +++++++++++++++++++++++
 2 files changed, 27 insertions(+), 3 deletions(-)
 create mode 100644 gcc/testsuite/c-c++-common/alias-1.c

diff --git a/gcc/c-family/c-common.cc b/gcc/c-family/c-common.cc
index f044db5b797..0efdc677217 100644
--- a/gcc/c-family/c-common.cc
+++ b/gcc/c-family/c-common.cc
@@ -3828,12 +3828,13 @@ c_common_get_alias_set (tree t)
   if (!TYPE_P (t))
     return -1;
 
-  /* Unlike char, char8_t doesn't alias. */
-  if (flag_char8_t && t == char8_type_node)
+  /* Unlike char, char8_t doesn't alias in C++.  (In C, char8_t is not
+     a distinct type.)  */
+  if (flag_char8_t && t == char8_type_node && c_dialect_cxx ())
     return -1;
 
   /* The C standard guarantees that any object may be accessed via an
-     lvalue that has narrow character type (except char8_t).  */
+     lvalue that has narrow character type.  */
   if (t == char_type_node
       || t == signed_char_type_node
       || t == unsigned_char_type_node)
diff --git a/gcc/testsuite/c-c++-common/alias-1.c b/gcc/testsuite/c-c++-common/alias-1.c
new file mode 100644
index 00000000000..d72fec47f76
--- /dev/null
+++ b/gcc/testsuite/c-c++-common/alias-1.c
@@ -0,0 +1,23 @@
+/* PR c/111884 */
+/* { dg-do compile } */
+/* { dg-options "-O2 -Wall" } */
+/* { dg-additional-options "-std=c++20" { target c++ } } */
+/* { dg-additional-options "-std=c2x" { target c } } */
+
+int f(int i)
+{
+    int f = 1;
+    return i[(unsigned char *)&f];
+}
+
+int g(int i)
+{
+    int f = 1;
+    return i[(signed char *)&f];
+}
+
+int h(int i)
+{
+    int f = 1;
+    return i[(char *)&f];
+}

base-commit: eb15fad3190a8b33e3e451b964ff1ecf08bbb113
-- 
2.41.0

Re: [PATCH] c-family: char8_t and aliasing in C vs C++ [PR111884]

[Jason Merrill]

On 10/20/23 12:31, Marek Polacek wrote:
> Bootstrapped/regtested on x86_64-pc-linux-gnu, ok for trunk?

OK.

> -- >8 --
> In the PR, Joseph says that in C char8_t is not a distinct type.  So
> we should behave as if it can alias anything, like ordinary char.
> In C, unsigned_char_type_node == char8_type_node, so with this patch
> we return 0 instead of -1.  And the following comment says:
> 
>    /* The C standard guarantees that any object may be accessed via an
>       lvalue that has narrow character type (except char8_t).  */
>    if (t == char_type_node
>        || t == signed_char_type_node
>        || t == unsigned_char_type_node)
>      return 0;
> 
> Which appears to be wrong, so I'm adjusting that as well.
> 
> 	PR c/111884
> 
> gcc/c-family/ChangeLog:
> 
> 	* c-common.cc (c_common_get_alias_set): Return -1 for char8_t only
> 	in C++.
> 
> gcc/testsuite/ChangeLog:
> 
> 	* c-c++-common/alias-1.c: New test.
> ---
>   gcc/c-family/c-common.cc             |  7 ++++---
>   gcc/testsuite/c-c++-common/alias-1.c | 23 +++++++++++++++++++++++
>   2 files changed, 27 insertions(+), 3 deletions(-)
>   create mode 100644 gcc/testsuite/c-c++-common/alias-1.c
> 
> diff --git a/gcc/c-family/c-common.cc b/gcc/c-family/c-common.cc
> index f044db5b797..0efdc677217 100644
> --- a/gcc/c-family/c-common.cc
> +++ b/gcc/c-family/c-common.cc
> @@ -3828,12 +3828,13 @@ c_common_get_alias_set (tree t)
>     if (!TYPE_P (t))
>       return -1;
>   
> -  /* Unlike char, char8_t doesn't alias. */
> -  if (flag_char8_t && t == char8_type_node)
> +  /* Unlike char, char8_t doesn't alias in C++.  (In C, char8_t is not
> +     a distinct type.)  */
> +  if (flag_char8_t && t == char8_type_node && c_dialect_cxx ())
>       return -1;
>   
>     /* The C standard guarantees that any object may be accessed via an
> -     lvalue that has narrow character type (except char8_t).  */
> +     lvalue that has narrow character type.  */
>     if (t == char_type_node
>         || t == signed_char_type_node
>         || t == unsigned_char_type_node)
> diff --git a/gcc/testsuite/c-c++-common/alias-1.c b/gcc/testsuite/c-c++-common/alias-1.c
> new file mode 100644
> index 00000000000..d72fec47f76
> --- /dev/null
> +++ b/gcc/testsuite/c-c++-common/alias-1.c
> @@ -0,0 +1,23 @@
> +/* PR c/111884 */
> +/* { dg-do compile } */
> +/* { dg-options "-O2 -Wall" } */
> +/* { dg-additional-options "-std=c++20" { target c++ } } */
> +/* { dg-additional-options "-std=c2x" { target c } } */
> +
> +int f(int i)
> +{
> +    int f = 1;
> +    return i[(unsigned char *)&f];
> +}
> +
> +int g(int i)
> +{
> +    int f = 1;
> +    return i[(signed char *)&f];
> +}
> +
> +int h(int i)
> +{
> +    int f = 1;
> +    return i[(char *)&f];
> +}
> 
> base-commit: eb15fad3190a8b33e3e451b964ff1ecf08bbb113