public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
* [PATCH] pretty-print: Fix up ptrdiff handling for %tu, %to, %tx
@ 2024-02-13  8:21 Jakub Jelinek
  2024-02-13 16:11 ` Jeff Law
  0 siblings, 1 reply; 2+ messages in thread
From: Jakub Jelinek @ 2024-02-13  8:21 UTC (permalink / raw)
  To: Richard Biener, Joseph S. Myers; +Cc: gcc-patches

Hi!

This is IMHO more of a theoretical case, I believe my current code
doesn't handle %tu or %to or %tx correctly if ptrdiff_t is not one of
int, long int or long long int.  For size_t and %zd or %zi I'm
using va_arg (whatever, ssize_t) and hope that ssize_t is the signed
type corresponding to size_t which C99 talks about.
For ptrdiff_t there is no type for unsigned type corresponding to
ptrdiff_t and I'm not aware of a portable way on the host to get
such a type (the fmt tests use hacks like
#define signed /* Type might or might not have explicit 'signed'.  */
typedef unsigned __PTRDIFF_TYPE__ unsigned_ptrdiff_t;
#undef signed
but that only works with compilers which predefine __PTRDIFF_TYPE__),
std::make_unsigned<ptrdiff_t> I believe only works reliably if
ptrdiff_t is one of char, signed char, short, int, long or long long,
but won't work e.g. for __int20__ or whatever other weird ptrdiff_t
the host might have.

The following patch assumes host is two's complement (I think we
rely on it pretty much everywhere anyway) and prints unsigned type
corresponding to ptrdiff_t as unsigned long long printing of
ptrdiff_t value masked with 2ULL * PTRDIFF_MAX + 1.  So the only
actual limitation is that it doesn't support ptrdiff_t wider than
long long int.

Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?

2024-02-13  Jakub Jelinek  <jakub@redhat.com>

	* pretty-print.cc (PTRDIFF_MAX): Define if not yet defined.
	(pp_integer_with_precision): For unsigned ptrdiff_t printing
	with u, o or x print ptrdiff_t argument converted to
	unsigned long long and masked with 2ULL * PTRDIFF_MAX + 1.

	* error.cc (error_print): For u printing of ptrdiff_t,
	print ptrdiff_t argument converted to unsigned long long and
	masked with 2ULL * PTRDIFF_MAX + 1.

--- gcc/pretty-print.cc.jj	2024-02-12 12:44:09.344335964 +0100
+++ gcc/pretty-print.cc	2024-02-12 19:09:27.594483176 +0100
@@ -752,6 +752,9 @@ output_buffer::~output_buffer ()
   obstack_free (&formatted_obstack, NULL);
 }
 
+#ifndef PTRDIFF_MAX
+#define PTRDIFF_MAX INTTYPE_MAXIMUM (ptrdiff_t)
+#endif
 
 /* Format an integer given by va_arg (ARG, type-specifier T) where
    type-specifier is a precision modifier as indicated by PREC.  F is
@@ -783,7 +786,15 @@ output_buffer::~output_buffer ()
         break;                                               \
                                                              \
       case 4:                                                \
-        if (sizeof (ptrdiff_t) <= sizeof (int))              \
+        if (T (-1) >= T (0))                                 \
+          {                                                  \
+            unsigned long long a = va_arg (ARG, ptrdiff_t);  \
+            unsigned long long m = PTRDIFF_MAX;              \
+            m = 2 * m + 1;                                   \
+            pp_scalar (PP, "%" HOST_LONG_LONG_FORMAT F,      \
+		       a & m);                               \
+          }                                                  \
+        else if (sizeof (ptrdiff_t) <= sizeof (int))         \
           pp_scalar (PP, "%" F,                              \
                      (int) va_arg (ARG, ptrdiff_t));         \
         else if (sizeof (ptrdiff_t) <= sizeof (long))        \
--- gcc/fortran/error.cc.jj	2024-02-12 12:44:09.343335978 +0100
+++ gcc/fortran/error.cc	2024-02-12 19:11:23.564886530 +0100
@@ -886,13 +886,14 @@ error_print (const char *type, const cha
 	  format++;
 	  if (*format == 'u')
 	    {
-	      ptrdiff_t ptrdiffval = spec[n++].u.ptrdiffval;
-	      if (sizeof (ptrdiff_t) == sizeof (int))
-		error_uinteger ((unsigned) ptrdiffval);
-	      else if (sizeof (ptrdiff_t) == sizeof (long))
-		error_uinteger ((unsigned long) ptrdiffval);
-	      else
-		error_uinteger (ptrdiffval);
+	      unsigned long long a = spec[n++].u.ptrdiffval, m;
+#ifdef PTRDIFF_MAX
+	      m = PTRDIFF_MAX;
+#else
+	      m = INTTYPE_MAXIMUM (ptrdiff_t);
+#endif
+	      m = 2 * m + 1;  
+	      error_uinteger (a & m);
 	    }
 	  else
 	    error_integer (spec[n++].u.ptrdiffval);

	Jakub


^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: [PATCH] pretty-print: Fix up ptrdiff handling for %tu, %to, %tx
  2024-02-13  8:21 [PATCH] pretty-print: Fix up ptrdiff handling for %tu, %to, %tx Jakub Jelinek
@ 2024-02-13 16:11 ` Jeff Law
  0 siblings, 0 replies; 2+ messages in thread
From: Jeff Law @ 2024-02-13 16:11 UTC (permalink / raw)
  To: Jakub Jelinek, Richard Biener, Joseph S. Myers; +Cc: gcc-patches



On 2/13/24 01:21, Jakub Jelinek wrote:
> Hi!
> 
> This is IMHO more of a theoretical case, I believe my current code
> doesn't handle %tu or %to or %tx correctly if ptrdiff_t is not one of
> int, long int or long long int.  For size_t and %zd or %zi I'm
> using va_arg (whatever, ssize_t) and hope that ssize_t is the signed
> type corresponding to size_t which C99 talks about.
> For ptrdiff_t there is no type for unsigned type corresponding to
> ptrdiff_t and I'm not aware of a portable way on the host to get
> such a type (the fmt tests use hacks like
> #define signed /* Type might or might not have explicit 'signed'.  */
> typedef unsigned __PTRDIFF_TYPE__ unsigned_ptrdiff_t;
> #undef signed
> but that only works with compilers which predefine __PTRDIFF_TYPE__),
> std::make_unsigned<ptrdiff_t> I believe only works reliably if
> ptrdiff_t is one of char, signed char, short, int, long or long long,
> but won't work e.g. for __int20__ or whatever other weird ptrdiff_t
> the host might have.
> 
> The following patch assumes host is two's complement (I think we
> rely on it pretty much everywhere anyway) and prints unsigned type
> corresponding to ptrdiff_t as unsigned long long printing of
> ptrdiff_t value masked with 2ULL * PTRDIFF_MAX + 1.  So the only
> actual limitation is that it doesn't support ptrdiff_t wider than
> long long int.
> 
> Bootstrapped/regtested on x86_64-linux and i686-linux, ok for trunk?
> 
> 2024-02-13  Jakub Jelinek  <jakub@redhat.com>
> 
> 	* pretty-print.cc (PTRDIFF_MAX): Define if not yet defined.
> 	(pp_integer_with_precision): For unsigned ptrdiff_t printing
> 	with u, o or x print ptrdiff_t argument converted to
> 	unsigned long long and masked with 2ULL * PTRDIFF_MAX + 1.
> 
> 	* error.cc (error_print): For u printing of ptrdiff_t,
> 	print ptrdiff_t argument converted to unsigned long long and
> 	masked with 2ULL * PTRDIFF_MAX + 1.
OK
jeff

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2024-02-13 16:11 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2024-02-13  8:21 [PATCH] pretty-print: Fix up ptrdiff handling for %tu, %to, %tx Jakub Jelinek
2024-02-13 16:11 ` Jeff Law

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).