This is the mail archive of the gdb-patches@sourceware.org mailing list for the GDB project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[PATCH 06/11] Little-endian fixes: 128-bit DFP parameters / registers


Hello,

the powerpc64le-linux ABI specifies that when a 128-bit DFP value is
passed in a pair of floating-point registers, the first register holds
the most-significant part of the value.  This is as opposed to the
usual rule on little-endian systems, where the first register would
hold the least-significant part.

This affects two places in GDB, the read/write routines for the
128-bit DFP pseudo-registers, and the function call / return
sequence.  For the former, current code already distinguishes
between big- and little-endian targets, but gets the latter
wrong.  This is presumably because *GCC* also got it wrong,
and GDB matches the old GCC behavior.  But GCC is now fixed:
http://gcc.gnu.org/ml/gcc-patches/2013-11/msg02145.html
so GDB needs to be fixed too.  (Old code shouldn't really be
an issue since there is no code "out there" so far that uses
dfp128 on little-endian ...)

Tested on powerpc64-linux and powerpc64le-linux.

Bye,
Ulrich



ChangeLog:

	* ppc-sysv-tdep.c (ppc64_sysv_abi_push_freg): Use correct order
	within a register pair holding a DFP 128-bit value on little-endian.
	(ppc64_sysv_abi_return_value_base): Likewise.
	* rs6000-tdep.c (dfp_pseudo_register_read): Likewise.
	(dfp_pseudo_register_write): Likewise.

testsuite/ChangeLog:

	* gdb.arch/powerpc-d128-regs.exp: Enable on powerpc64*-*.


Index: binutils-gdb/gdb/ppc-sysv-tdep.c
===================================================================
--- binutils-gdb.orig/gdb/ppc-sysv-tdep.c
+++ binutils-gdb/gdb/ppc-sysv-tdep.c
@@ -1267,8 +1267,11 @@ ppc64_sysv_abi_push_freg (struct gdbarch
       if (argpos->regcache && argpos->freg <= 12)
 	{
 	  int regnum = tdep->ppc_fp0_regnum + argpos->freg;
-	  regcache_cooked_write (argpos->regcache, regnum, val);
-	  regcache_cooked_write (argpos->regcache, regnum + 1, val + 8);
+	  int lopart = gdbarch_byte_order (gdbarch) == BFD_ENDIAN_BIG ? 8 : 0;
+	  int hipart = gdbarch_byte_order (gdbarch) == BFD_ENDIAN_BIG ? 0 : 8;
+
+	  regcache_cooked_write (argpos->regcache, regnum, val + hipart);
+	  regcache_cooked_write (argpos->regcache, regnum + 1, val + lopart);
 	}
 
       argpos->freg += 2;
@@ -1676,16 +1679,18 @@ ppc64_sysv_abi_return_value_base (struct
       && TYPE_CODE (valtype) == TYPE_CODE_DECFLOAT)
     {
       int regnum = tdep->ppc_fp0_regnum + 2 + 2 * index;
+      int lopart = gdbarch_byte_order (gdbarch) == BFD_ENDIAN_BIG ? 8 : 0;
+      int hipart = gdbarch_byte_order (gdbarch) == BFD_ENDIAN_BIG ? 0 : 8;
 
       if (writebuf != NULL)
 	{
-	  regcache_cooked_write (regcache, regnum, writebuf);
-	  regcache_cooked_write (regcache, regnum + 1, writebuf + 8);
+	  regcache_cooked_write (regcache, regnum, writebuf + hipart);
+	  regcache_cooked_write (regcache, regnum + 1, writebuf + lopart);
 	}
       if (readbuf != NULL)
 	{
-	  regcache_cooked_read (regcache, regnum, readbuf);
-	  regcache_cooked_read (regcache, regnum + 1, readbuf + 8);
+	  regcache_cooked_read (regcache, regnum, readbuf + hipart);
+	  regcache_cooked_read (regcache, regnum + 1, readbuf + lopart);
 	}
       return 1;
     }
Index: binutils-gdb/gdb/rs6000-tdep.c
===================================================================
--- binutils-gdb.orig/gdb/rs6000-tdep.c
+++ binutils-gdb/gdb/rs6000-tdep.c
@@ -2672,10 +2672,10 @@ dfp_pseudo_register_read (struct gdbarch
   else
     {
       status = regcache_raw_read (regcache, tdep->ppc_fp0_regnum +
-				  2 * reg_index + 1, buffer + 8);
+				  2 * reg_index + 1, buffer);
       if (status == REG_VALID)
 	status = regcache_raw_read (regcache, tdep->ppc_fp0_regnum +
-				    2 * reg_index, buffer);
+				    2 * reg_index, buffer + 8);
     }
 
   return status;
@@ -2701,9 +2701,9 @@ dfp_pseudo_register_write (struct gdbarc
   else
     {
       regcache_raw_write (regcache, tdep->ppc_fp0_regnum +
-			  2 * reg_index + 1, buffer + 8);
+			  2 * reg_index + 1, buffer);
       regcache_raw_write (regcache, tdep->ppc_fp0_regnum +
-			  2 * reg_index, buffer);
+			  2 * reg_index, buffer + 8);
     }
 }
 
Index: binutils-gdb/gdb/testsuite/gdb.arch/powerpc-d128-regs.exp
===================================================================
--- binutils-gdb.orig/gdb/testsuite/gdb.arch/powerpc-d128-regs.exp
+++ binutils-gdb/gdb/testsuite/gdb.arch/powerpc-d128-regs.exp
@@ -20,7 +20,7 @@
 
 # Testcase for ppc decimal128 pseudo-registers.
 
-if ![istarget "powerpc64-*"] then {
+if ![istarget "powerpc64*-*"] then {
     verbose "Skipping powerpc Decimal128 pseudo-registers testcase."
     return
 }
-- 
  Dr. Ulrich Weigand
  GNU/Linux compilers and toolchain
  Ulrich.Weigand@de.ibm.com


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]