[PATCH] gcc5: Add PR65779 patch to fix powerpc compile issues

[PATCH] gcc5: Add PR65779 patch to fix powerpc compile issues

[Richard Purdie]

This fixes compile issues on powerpc with gcc 5 which show up with
errors like:

| make[2]: Entering directory '/media/build1/poky/build/tmp/work/ppc7400-poky-linux/xprop/1_1.2.2-r0/build'
| powerpc-poky-linux-gcc  -m32 -mhard-float -mcpu=7400 --sysroot=/media/build1/poky/build/tmp/sysroots/qemuppc -Wall -Wpointer-arith -Wmissing-declarations -Wformat=2 -Wstrict-prototypes -Wmissing-prototypes -Wnested-externs -Wbad-function-cast -Wold-style-definition -Wdeclaration-after-statement -Wunused -Wuninitialized -Wshadow -Wmissing-noreturn -Wmissing-format-attribute -Wredundant-decls -Wlogical-op -Werror=implicit -Werror=nonnull -Werror=init-self -Werror=main -Werror=missing-braces -Werror=sequence-point -Werror=return-type -Werror=trigraphs -Werror=array-bounds -Werror=write-strings -Werror=address -Werror=int-to-pointer-cast -Werror=pointer-to-int-cast -fno-strict-aliasing  -O2 -pipe -g -feliminate-unused-debug-types  -Wl,-O1 -Wl,--hash-style=gnu -Wl,--as-needed -o xprop dsimple.o clientwin.o xprop.o -lX11
| /media/build1/poky/build/tmp/sysroots/qemuppc/usr/lib/../lib/libX11.so: undefined reference to `.LCL2'
| collect2: error: ld returned 1 exit status

Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

diff --git a/meta/recipes-devtools/gcc/gcc-5.1.inc b/meta/recipes-devtools/gcc/gcc-5.1.inc
index db4c795..305736b 100644
--- a/meta/recipes-devtools/gcc/gcc-5.1.inc
+++ b/meta/recipes-devtools/gcc/gcc-5.1.inc
@@ -68,6 +68,7 @@ SRC_URI = "\
            file://0034-Don-t-search-host-directory-during-relink-if-inst_pr.patch \
            file://0035-Dont-link-the-plugins-with-libgomp-explicitly.patch \
            file://0036-Use-SYSTEMLIBS_DIR-replacement-instead-of-hardcoding.patch \
+           file://0037-pr65779.patch \
           "
 
 #S = "${TMPDIR}/work-shared/gcc-${PV}-${PR}/gcc-${SNAP}"
diff --git a/meta/recipes-devtools/gcc/gcc-5.1/0037-pr65779.patch b/meta/recipes-devtools/gcc/gcc-5.1/0037-pr65779.patch
new file mode 100644
index 0000000..1424673
--- /dev/null
+++ b/meta/recipes-devtools/gcc/gcc-5.1/0037-pr65779.patch
@@ -0,0 +1,173 @@
+List-Id: <gcc-patches.gcc.gnu.org>
+List-Archive: <http://gcc.gnu.org/ml/gcc-patches/>
+List-Post: <mailto:gcc-patches at gcc dot gnu dot org>
+List-Help: <mailto:gcc-patches-help at gcc dot gnu dot org>
+Date: Mon, 20 Apr 2015 12:40:49 +0930
+From: Alan Modra <amodra at gmail dot com>
+To: gcc-patches at gcc dot gnu dot org
+Subject: [Patch] pr65779 - [5/6 Regression] undefined local symbol on powerpc
+
+This patch removes bogus debug info left around by shrink-wrapping,
+which on some powerpc targets with just the right register allocation
+led to assembly errors.
+
+Bootstrapped and regression tested powerpc64-linux and x86_64-linux.
+
+https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65779
+
+gcc/
+	PR debug/65779
+	* shrink-wrap.c (insn_uses_reg): New function.
+	(move_insn_for_shrink_wrap): Remove debug insns using regs set
+	by the moved insn.
+gcc/testsuite/
+	* gcc.dg/pr65779.c: New.
+	
+Upstream-Status: Pending (from mailing list, not merged yet)
+
+Index: a/gcc/shrink-wrap.c
+===================================================================
+--- a/gcc/shrink-wrap.c.orig
++++ b/gcc/shrink-wrap.c
+@@ -182,6 +182,21 @@ live_edge_for_reg (basic_block bb, int r
+   return live_edge;
+ }
+ 
++static bool
++insn_uses_reg (rtx_insn *insn, unsigned int regno, unsigned int end_regno)
++{
++  df_ref use;
++
++  FOR_EACH_INSN_USE (use, insn)
++    {
++      rtx reg = DF_REF_REG (use);
++
++      if (REG_P (reg) && REGNO (reg) >= regno && REGNO (reg) < end_regno)
++	return true;
++    }
++  return false;
++}
++
+ /* Try to move INSN from BB to a successor.  Return true on success.
+    USES and DEFS are the set of registers that are used and defined
+    after INSN in BB.  SPLIT_P indicates whether a live edge from BB
+@@ -340,10 +355,15 @@ move_insn_for_shrink_wrap (basic_block b
+       *split_p = true;
+     }
+ 
++  vec<basic_block> live_bbs;
++  if (MAY_HAVE_DEBUG_INSNS)
++    live_bbs.create (5);
+   /* At this point we are committed to moving INSN, but let's try to
+      move it as far as we can.  */
+   do
+     {
++      if (MAY_HAVE_DEBUG_INSNS)
++	live_bbs.safe_push (bb);
+       live_out = df_get_live_out (bb);
+       live_in = df_get_live_in (next_block);
+       bb = next_block;
+@@ -426,6 +446,34 @@ move_insn_for_shrink_wrap (basic_block b
+ 	SET_REGNO_REG_SET (bb_uses, i);
+     }
+ 
++  /* Remove debug insns using regs set by the insn we are moving.  */
++  if (MAY_HAVE_DEBUG_INSNS)
++    {
++      while (!live_bbs.is_empty ())
++	{
++	  rtx_insn *dinsn;
++	  basic_block tmp_bb = live_bbs.pop ();
++
++	  FOR_BB_INSNS_REVERSE (tmp_bb, dinsn)
++	    {
++	      if (dinsn == insn)
++		break;
++	      if (DEBUG_INSN_P (dinsn)
++		  && insn_uses_reg (dinsn, dregno, end_dregno))
++		{
++		  if (*split_p)
++		    /* If split, then we will be moving insn into a
++		       newly created block immediately after the entry
++		       block.  Move the debug info there too.  */
++		    emit_debug_insn_after (PATTERN (dinsn), bb_note (bb));
++		  delete_insn (dinsn);
++		  break;
++		}
++	    }
++	}
++      live_bbs.release ();
++    }
++
+   emit_insn_after (PATTERN (insn), bb_note (bb));
+   delete_insn (insn);
+   return true;
+Index: b/gcc/testsuite/gcc.dg/pr65779.c
+===================================================================
+--- /dev/null
++++ b/gcc/testsuite/gcc.dg/pr65779.c
+@@ -0,0 +1,64 @@
++/* { dg-do run } */
++/* { dg-options "-O2 -g" } */
++/* { dg-additional-options "-mrelocatable" { target powerpc-*-rtems* } } */
++
++unsigned long __attribute__ ((noinline))
++adler32 (unsigned long adler, unsigned char *buf, unsigned int len)
++{
++  unsigned long s1 = adler & 0xffff;
++  unsigned long s2 = (adler >> 16) & 0xffff;
++  int k;
++
++  if (buf == 0)
++    return 1L;
++
++  while (len > 0)
++    {
++      k = len < 5552 ? len : 5552;
++      len -= k;
++      while (k >= 16)
++	{
++	  s1 += *buf++; s2 += s1;
++	  s1 += *buf++; s2 += s1;
++	  s1 += *buf++; s2 += s1;
++	  s1 += *buf++; s2 += s1;
++	  s1 += *buf++; s2 += s1;
++	  s1 += *buf++; s2 += s1;
++	  s1 += *buf++; s2 += s1;
++	  s1 += *buf++; s2 += s1;
++	  s1 += *buf++; s2 += s1;
++	  s1 += *buf++; s2 += s1;
++	  s1 += *buf++; s2 += s1;
++	  s1 += *buf++; s2 += s1;
++	  s1 += *buf++; s2 += s1;
++	  s1 += *buf++; s2 += s1;
++	  s1 += *buf++; s2 += s1;
++	  s1 += *buf++; s2 += s1;
++	  k -= 16;
++	}
++      if (k != 0)
++	do
++	  {
++	    s1 += *buf++; s2 += s1;
++	  } while (--k);
++      s1 &= 0xffffffffUL;
++      s2 &= 0xffffffffUL;
++      s1 %= 65521L;
++      s2 %= 65521L;
++    }
++  return (s2 << 16) | s1;
++}
++
++unsigned char buf[] = { 0, 1, 2, 3, 4, 5, 6, 7,
++			8, 9, 10, 11, 12, 13, 14, 15,
++			0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87,
++			0x88, 0x89, 0x8a, 0x8b, 0x8c, 0x8d, 0x8e, 0x8f,
++			0x55, 0xaa };
++int
++main ()
++{
++  unsigned long x = adler32 (0, buf, sizeof buf);
++  if (x != 0x640409efUL)
++    __builtin_abort ();
++  return 0;
++}

Re: [PATCH] gcc5: Add PR65779 patch to fix powerpc compile issues

[akuster808]

Richard,

There is a Yocto bug opened for this #7721, please add the reference.

thanks for the patch.

regards,
Armin

On 05/11/2015 09:54 AM, Richard Purdie wrote:
> This fixes compile issues on powerpc with gcc 5 which show up with
> errors like:
>
> | make[2]: Entering directory '/media/build1/poky/build/tmp/work/ppc7400-poky-linux/xprop/1_1.2.2-r0/build'
> | powerpc-poky-linux-gcc  -m32 -mhard-float -mcpu=7400 --sysroot=/media/build1/poky/build/tmp/sysroots/qemuppc -Wall -Wpointer-arith -Wmissing-declarations -Wformat=2 -Wstrict-prototypes -Wmissing-prototypes -Wnested-externs -Wbad-function-cast -Wold-style-definition -Wdeclaration-after-statement -Wunused -Wuninitialized -Wshadow -Wmissing-noreturn -Wmissing-format-attribute -Wredundant-decls -Wlogical-op -Werror=implicit -Werror=nonnull -Werror=init-self -Werror=main -Werror=missing-braces -Werror=sequence-point -Werror=return-type -Werror=trigraphs -Werror=array-bounds -Werror=write-strings -Werror=address -Werror=int-to-pointer-cast -Werror=pointer-to-int-cast -fno-strict-aliasing  -O2 -pipe -g -feliminate-unused-debug-types  -Wl,-O1 -Wl,--hash-style=gnu -Wl,--as-needed -o xprop dsimple.o clientwin.o xprop.o -lX11
> | /media/build1/poky/build/tmp/sysroots/qemuppc/usr/lib/../lib/libX11.so: undefined reference to `.LCL2'
> | collect2: error: ld returned 1 exit status
>
> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
>
> diff --git a/meta/recipes-devtools/gcc/gcc-5.1.inc b/meta/recipes-devtools/gcc/gcc-5.1.inc
> index db4c795..305736b 100644
> --- a/meta/recipes-devtools/gcc/gcc-5.1.inc
> +++ b/meta/recipes-devtools/gcc/gcc-5.1.inc
> @@ -68,6 +68,7 @@ SRC_URI = "\
>              file://0034-Don-t-search-host-directory-during-relink-if-inst_pr.patch \
>              file://0035-Dont-link-the-plugins-with-libgomp-explicitly.patch \
>              file://0036-Use-SYSTEMLIBS_DIR-replacement-instead-of-hardcoding.patch \
> +           file://0037-pr65779.patch \
>             "
>
>   #S = "${TMPDIR}/work-shared/gcc-${PV}-${PR}/gcc-${SNAP}"
> diff --git a/meta/recipes-devtools/gcc/gcc-5.1/0037-pr65779.patch b/meta/recipes-devtools/gcc/gcc-5.1/0037-pr65779.patch
> new file mode 100644
> index 0000000..1424673
> --- /dev/null
> +++ b/meta/recipes-devtools/gcc/gcc-5.1/0037-pr65779.patch
> @@ -0,0 +1,173 @@
> +List-Id: <gcc-patches.gcc.gnu.org>
> +List-Archive: <http://gcc.gnu.org/ml/gcc-patches/>
> +List-Post: <mailto:gcc-patches at gcc dot gnu dot org>
> +List-Help: <mailto:gcc-patches-help at gcc dot gnu dot org>
> +Date: Mon, 20 Apr 2015 12:40:49 +0930
> +From: Alan Modra <amodra at gmail dot com>
> +To: gcc-patches at gcc dot gnu dot org
> +Subject: [Patch] pr65779 - [5/6 Regression] undefined local symbol on powerpc
> +
> +This patch removes bogus debug info left around by shrink-wrapping,
> +which on some powerpc targets with just the right register allocation
> +led to assembly errors.
> +
> +Bootstrapped and regression tested powerpc64-linux and x86_64-linux.
> +
> +https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65779
> +
> +gcc/
> +	PR debug/65779
> +	* shrink-wrap.c (insn_uses_reg): New function.
> +	(move_insn_for_shrink_wrap): Remove debug insns using regs set
> +	by the moved insn.
> +gcc/testsuite/
> +	* gcc.dg/pr65779.c: New.
> +	
> +Upstream-Status: Pending (from mailing list, not merged yet)
> +
> +Index: a/gcc/shrink-wrap.c
> +===================================================================
> +--- a/gcc/shrink-wrap.c.orig
> ++++ b/gcc/shrink-wrap.c
> +@@ -182,6 +182,21 @@ live_edge_for_reg (basic_block bb, int r
> +   return live_edge;
> + }
> +
> ++static bool
> ++insn_uses_reg (rtx_insn *insn, unsigned int regno, unsigned int end_regno)
> ++{
> ++  df_ref use;
> ++
> ++  FOR_EACH_INSN_USE (use, insn)
> ++    {
> ++      rtx reg = DF_REF_REG (use);
> ++
> ++      if (REG_P (reg) && REGNO (reg) >= regno && REGNO (reg) < end_regno)
> ++	return true;
> ++    }
> ++  return false;
> ++}
> ++
> + /* Try to move INSN from BB to a successor.  Return true on success.
> +    USES and DEFS are the set of registers that are used and defined
> +    after INSN in BB.  SPLIT_P indicates whether a live edge from BB
> +@@ -340,10 +355,15 @@ move_insn_for_shrink_wrap (basic_block b
> +       *split_p = true;
> +     }
> +
> ++  vec<basic_block> live_bbs;
> ++  if (MAY_HAVE_DEBUG_INSNS)
> ++    live_bbs.create (5);
> +   /* At this point we are committed to moving INSN, but let's try to
> +      move it as far as we can.  */
> +   do
> +     {
> ++      if (MAY_HAVE_DEBUG_INSNS)
> ++	live_bbs.safe_push (bb);
> +       live_out = df_get_live_out (bb);
> +       live_in = df_get_live_in (next_block);
> +       bb = next_block;
> +@@ -426,6 +446,34 @@ move_insn_for_shrink_wrap (basic_block b
> + 	SET_REGNO_REG_SET (bb_uses, i);
> +     }
> +
> ++  /* Remove debug insns using regs set by the insn we are moving.  */
> ++  if (MAY_HAVE_DEBUG_INSNS)
> ++    {
> ++      while (!live_bbs.is_empty ())
> ++	{
> ++	  rtx_insn *dinsn;
> ++	  basic_block tmp_bb = live_bbs.pop ();
> ++
> ++	  FOR_BB_INSNS_REVERSE (tmp_bb, dinsn)
> ++	    {
> ++	      if (dinsn == insn)
> ++		break;
> ++	      if (DEBUG_INSN_P (dinsn)
> ++		  && insn_uses_reg (dinsn, dregno, end_dregno))
> ++		{
> ++		  if (*split_p)
> ++		    /* If split, then we will be moving insn into a
> ++		       newly created block immediately after the entry
> ++		       block.  Move the debug info there too.  */
> ++		    emit_debug_insn_after (PATTERN (dinsn), bb_note (bb));
> ++		  delete_insn (dinsn);
> ++		  break;
> ++		}
> ++	    }
> ++	}
> ++      live_bbs.release ();
> ++    }
> ++
> +   emit_insn_after (PATTERN (insn), bb_note (bb));
> +   delete_insn (insn);
> +   return true;
> +Index: b/gcc/testsuite/gcc.dg/pr65779.c
> +===================================================================
> +--- /dev/null
> ++++ b/gcc/testsuite/gcc.dg/pr65779.c
> +@@ -0,0 +1,64 @@
> ++/* { dg-do run } */
> ++/* { dg-options "-O2 -g" } */
> ++/* { dg-additional-options "-mrelocatable" { target powerpc-*-rtems* } } */
> ++
> ++unsigned long __attribute__ ((noinline))
> ++adler32 (unsigned long adler, unsigned char *buf, unsigned int len)
> ++{
> ++  unsigned long s1 = adler & 0xffff;
> ++  unsigned long s2 = (adler >> 16) & 0xffff;
> ++  int k;
> ++
> ++  if (buf == 0)
> ++    return 1L;
> ++
> ++  while (len > 0)
> ++    {
> ++      k = len < 5552 ? len : 5552;
> ++      len -= k;
> ++      while (k >= 16)
> ++	{
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  k -= 16;
> ++	}
> ++      if (k != 0)
> ++	do
> ++	  {
> ++	    s1 += *buf++; s2 += s1;
> ++	  } while (--k);
> ++      s1 &= 0xffffffffUL;
> ++      s2 &= 0xffffffffUL;
> ++      s1 %= 65521L;
> ++      s2 %= 65521L;
> ++    }
> ++  return (s2 << 16) | s1;
> ++}
> ++
> ++unsigned char buf[] = { 0, 1, 2, 3, 4, 5, 6, 7,
> ++			8, 9, 10, 11, 12, 13, 14, 15,
> ++			0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87,
> ++			0x88, 0x89, 0x8a, 0x8b, 0x8c, 0x8d, 0x8e, 0x8f,
> ++			0x55, 0xaa };
> ++int
> ++main ()
> ++{
> ++  unsigned long x = adler32 (0, buf, sizeof buf);
> ++  if (x != 0x640409efUL)
> ++    __builtin_abort ();
> ++  return 0;
> ++}
>
>

Re: [PATCH] gcc5: Add PR65779 patch to fix powerpc compile issues

[akuster808]

On 05/11/2015 09:54 AM, Richard Purdie wrote:
> This fixes compile issues on powerpc with gcc 5 which show up with
> errors like:
>
> | make[2]: Entering directory '/media/build1/poky/build/tmp/work/ppc7400-poky-linux/xprop/1_1.2.2-r0/build'
> | powerpc-poky-linux-gcc  -m32 -mhard-float -mcpu=7400 --sysroot=/media/build1/poky/build/tmp/sysroots/qemuppc -Wall -Wpointer-arith -Wmissing-declarations -Wformat=2 -Wstrict-prototypes -Wmissing-prototypes -Wnested-externs -Wbad-function-cast -Wold-style-definition -Wdeclaration-after-statement -Wunused -Wuninitialized -Wshadow -Wmissing-noreturn -Wmissing-format-attribute -Wredundant-decls -Wlogical-op -Werror=implicit -Werror=nonnull -Werror=init-self -Werror=main -Werror=missing-braces -Werror=sequence-point -Werror=return-type -Werror=trigraphs -Werror=array-bounds -Werror=write-strings -Werror=address -Werror=int-to-pointer-cast -Werror=pointer-to-int-cast -fno-strict-aliasing  -O2 -pipe -g -feliminate-unused-debug-types  -Wl,-O1 -Wl,--hash-style=gnu -Wl,--as-needed -o xprop dsimple.o clientwin.o xprop.o -lX11
> | /media/build1/poky/build/tmp/sysroots/qemuppc/usr/lib/../lib/libX11.so: undefined reference to `.LCL2'
> | collect2: error: ld returned 1 exit status
>
> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>

Acked-by: Armin Kuster <Akuster808@gmail.com>


this fixes all but one package(lzop) failure on qemuppc base yocto world 
build.


>
> diff --git a/meta/recipes-devtools/gcc/gcc-5.1.inc b/meta/recipes-devtools/gcc/gcc-5.1.inc
> index db4c795..305736b 100644
> --- a/meta/recipes-devtools/gcc/gcc-5.1.inc
> +++ b/meta/recipes-devtools/gcc/gcc-5.1.inc
> @@ -68,6 +68,7 @@ SRC_URI = "\
>              file://0034-Don-t-search-host-directory-during-relink-if-inst_pr.patch \
>              file://0035-Dont-link-the-plugins-with-libgomp-explicitly.patch \
>              file://0036-Use-SYSTEMLIBS_DIR-replacement-instead-of-hardcoding.patch \
> +           file://0037-pr65779.patch \
>             "
>
>   #S = "${TMPDIR}/work-shared/gcc-${PV}-${PR}/gcc-${SNAP}"
> diff --git a/meta/recipes-devtools/gcc/gcc-5.1/0037-pr65779.patch b/meta/recipes-devtools/gcc/gcc-5.1/0037-pr65779.patch
> new file mode 100644
> index 0000000..1424673
> --- /dev/null
> +++ b/meta/recipes-devtools/gcc/gcc-5.1/0037-pr65779.patch
> @@ -0,0 +1,173 @@
> +List-Id: <gcc-patches.gcc.gnu.org>
> +List-Archive: <http://gcc.gnu.org/ml/gcc-patches/>
> +List-Post: <mailto:gcc-patches at gcc dot gnu dot org>
> +List-Help: <mailto:gcc-patches-help at gcc dot gnu dot org>
> +Date: Mon, 20 Apr 2015 12:40:49 +0930
> +From: Alan Modra <amodra at gmail dot com>
> +To: gcc-patches at gcc dot gnu dot org
> +Subject: [Patch] pr65779 - [5/6 Regression] undefined local symbol on powerpc
> +
> +This patch removes bogus debug info left around by shrink-wrapping,
> +which on some powerpc targets with just the right register allocation
> +led to assembly errors.
> +
> +Bootstrapped and regression tested powerpc64-linux and x86_64-linux.
> +
> +https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65779
> +
> +gcc/
> +	PR debug/65779
> +	* shrink-wrap.c (insn_uses_reg): New function.
> +	(move_insn_for_shrink_wrap): Remove debug insns using regs set
> +	by the moved insn.
> +gcc/testsuite/
> +	* gcc.dg/pr65779.c: New.
> +	
> +Upstream-Status: Pending (from mailing list, not merged yet)
> +
> +Index: a/gcc/shrink-wrap.c
> +===================================================================
> +--- a/gcc/shrink-wrap.c.orig
> ++++ b/gcc/shrink-wrap.c
> +@@ -182,6 +182,21 @@ live_edge_for_reg (basic_block bb, int r
> +   return live_edge;
> + }
> +
> ++static bool
> ++insn_uses_reg (rtx_insn *insn, unsigned int regno, unsigned int end_regno)
> ++{
> ++  df_ref use;
> ++
> ++  FOR_EACH_INSN_USE (use, insn)
> ++    {
> ++      rtx reg = DF_REF_REG (use);
> ++
> ++      if (REG_P (reg) && REGNO (reg) >= regno && REGNO (reg) < end_regno)
> ++	return true;
> ++    }
> ++  return false;
> ++}
> ++
> + /* Try to move INSN from BB to a successor.  Return true on success.
> +    USES and DEFS are the set of registers that are used and defined
> +    after INSN in BB.  SPLIT_P indicates whether a live edge from BB
> +@@ -340,10 +355,15 @@ move_insn_for_shrink_wrap (basic_block b
> +       *split_p = true;
> +     }
> +
> ++  vec<basic_block> live_bbs;
> ++  if (MAY_HAVE_DEBUG_INSNS)
> ++    live_bbs.create (5);
> +   /* At this point we are committed to moving INSN, but let's try to
> +      move it as far as we can.  */
> +   do
> +     {
> ++      if (MAY_HAVE_DEBUG_INSNS)
> ++	live_bbs.safe_push (bb);
> +       live_out = df_get_live_out (bb);
> +       live_in = df_get_live_in (next_block);
> +       bb = next_block;
> +@@ -426,6 +446,34 @@ move_insn_for_shrink_wrap (basic_block b
> + 	SET_REGNO_REG_SET (bb_uses, i);
> +     }
> +
> ++  /* Remove debug insns using regs set by the insn we are moving.  */
> ++  if (MAY_HAVE_DEBUG_INSNS)
> ++    {
> ++      while (!live_bbs.is_empty ())
> ++	{
> ++	  rtx_insn *dinsn;
> ++	  basic_block tmp_bb = live_bbs.pop ();
> ++
> ++	  FOR_BB_INSNS_REVERSE (tmp_bb, dinsn)
> ++	    {
> ++	      if (dinsn == insn)
> ++		break;
> ++	      if (DEBUG_INSN_P (dinsn)
> ++		  && insn_uses_reg (dinsn, dregno, end_dregno))
> ++		{
> ++		  if (*split_p)
> ++		    /* If split, then we will be moving insn into a
> ++		       newly created block immediately after the entry
> ++		       block.  Move the debug info there too.  */
> ++		    emit_debug_insn_after (PATTERN (dinsn), bb_note (bb));
> ++		  delete_insn (dinsn);
> ++		  break;
> ++		}
> ++	    }
> ++	}
> ++      live_bbs.release ();
> ++    }
> ++
> +   emit_insn_after (PATTERN (insn), bb_note (bb));
> +   delete_insn (insn);
> +   return true;
> +Index: b/gcc/testsuite/gcc.dg/pr65779.c
> +===================================================================
> +--- /dev/null
> ++++ b/gcc/testsuite/gcc.dg/pr65779.c
> +@@ -0,0 +1,64 @@
> ++/* { dg-do run } */
> ++/* { dg-options "-O2 -g" } */
> ++/* { dg-additional-options "-mrelocatable" { target powerpc-*-rtems* } } */
> ++
> ++unsigned long __attribute__ ((noinline))
> ++adler32 (unsigned long adler, unsigned char *buf, unsigned int len)
> ++{
> ++  unsigned long s1 = adler & 0xffff;
> ++  unsigned long s2 = (adler >> 16) & 0xffff;
> ++  int k;
> ++
> ++  if (buf == 0)
> ++    return 1L;
> ++
> ++  while (len > 0)
> ++    {
> ++      k = len < 5552 ? len : 5552;
> ++      len -= k;
> ++      while (k >= 16)
> ++	{
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  s1 += *buf++; s2 += s1;
> ++	  k -= 16;
> ++	}
> ++      if (k != 0)
> ++	do
> ++	  {
> ++	    s1 += *buf++; s2 += s1;
> ++	  } while (--k);
> ++      s1 &= 0xffffffffUL;
> ++      s2 &= 0xffffffffUL;
> ++      s1 %= 65521L;
> ++      s2 %= 65521L;
> ++    }
> ++  return (s2 << 16) | s1;
> ++}
> ++
> ++unsigned char buf[] = { 0, 1, 2, 3, 4, 5, 6, 7,
> ++			8, 9, 10, 11, 12, 13, 14, 15,
> ++			0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87,
> ++			0x88, 0x89, 0x8a, 0x8b, 0x8c, 0x8d, 0x8e, 0x8f,
> ++			0x55, 0xaa };
> ++int
> ++main ()
> ++{
> ++  unsigned long x = adler32 (0, buf, sizeof buf);
> ++  if (x != 0x640409efUL)
> ++    __builtin_abort ();
> ++  return 0;
> ++}
>
>

Re: [PATCH] gcc5: Add PR65779 patch to fix powerpc compile issues

[Khem Raj]

On Mon, May 11, 2015 at 4:20 PM, akuster808 <akuster808@gmail.com> wrote:
>
>
> On 05/11/2015 09:54 AM, Richard Purdie wrote:
>>
>> This fixes compile issues on powerpc with gcc 5 which show up with
>> errors like:
>>
>> | make[2]: Entering directory
>> '/media/build1/poky/build/tmp/work/ppc7400-poky-linux/xprop/1_1.2.2-r0/build'
>> | powerpc-poky-linux-gcc  -m32 -mhard-float -mcpu=7400
>> --sysroot=/media/build1/poky/build/tmp/sysroots/qemuppc -Wall
>> -Wpointer-arith -Wmissing-declarations -Wformat=2 -Wstrict-prototypes
>> -Wmissing-prototypes -Wnested-externs -Wbad-function-cast
>> -Wold-style-definition -Wdeclaration-after-statement -Wunused
>> -Wuninitialized -Wshadow -Wmissing-noreturn -Wmissing-format-attribute
>> -Wredundant-decls -Wlogical-op -Werror=implicit -Werror=nonnull
>> -Werror=init-self -Werror=main -Werror=missing-braces -Werror=sequence-point
>> -Werror=return-type -Werror=trigraphs -Werror=array-bounds
>> -Werror=write-strings -Werror=address -Werror=int-to-pointer-cast
>> -Werror=pointer-to-int-cast -fno-strict-aliasing  -O2 -pipe -g
>> -feliminate-unused-debug-types  -Wl,-O1 -Wl,--hash-style=gnu -Wl,--as-needed
>> -o xprop dsimple.o clientwin.o xprop.o -lX11
>> | /media/build1/poky/build/tmp/sysroots/qemuppc/usr/lib/../lib/libX11.so:
>> undefined reference to `.LCL2'
>> | collect2: error: ld returned 1 exit status
>>
>> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
>
>
> Acked-by: Armin Kuster <Akuster808@gmail.com>
>
>
> this fixes all but one package(lzop) failure on qemuppc base yocto world
> build.
>

I am fine witj this patch for now, since we are mostly build testing,
we have to keep in mind
its a codegen bug and can bite us at runtime. I am just hoping for an
update to this patch and that will eliminate
that doubt.

>
>
>>
>> diff --git a/meta/recipes-devtools/gcc/gcc-5.1.inc
>> b/meta/recipes-devtools/gcc/gcc-5.1.inc
>> index db4c795..305736b 100644
>> --- a/meta/recipes-devtools/gcc/gcc-5.1.inc
>> +++ b/meta/recipes-devtools/gcc/gcc-5.1.inc
>> @@ -68,6 +68,7 @@ SRC_URI = "\
>>
>> file://0034-Don-t-search-host-directory-during-relink-if-inst_pr.patch \
>>
>> file://0035-Dont-link-the-plugins-with-libgomp-explicitly.patch \
>>
>> file://0036-Use-SYSTEMLIBS_DIR-replacement-instead-of-hardcoding.patch \
>> +           file://0037-pr65779.patch \
>>             "
>>
>>   #S = "${TMPDIR}/work-shared/gcc-${PV}-${PR}/gcc-${SNAP}"
>> diff --git a/meta/recipes-devtools/gcc/gcc-5.1/0037-pr65779.patch
>> b/meta/recipes-devtools/gcc/gcc-5.1/0037-pr65779.patch
>> new file mode 100644
>> index 0000000..1424673
>> --- /dev/null
>> +++ b/meta/recipes-devtools/gcc/gcc-5.1/0037-pr65779.patch
>> @@ -0,0 +1,173 @@
>> +List-Id: <gcc-patches.gcc.gnu.org>
>> +List-Archive: <http://gcc.gnu.org/ml/gcc-patches/>
>> +List-Post: <mailto:gcc-patches at gcc dot gnu dot org>
>> +List-Help: <mailto:gcc-patches-help at gcc dot gnu dot org>
>> +Date: Mon, 20 Apr 2015 12:40:49 +0930
>> +From: Alan Modra <amodra at gmail dot com>
>> +To: gcc-patches at gcc dot gnu dot org
>> +Subject: [Patch] pr65779 - [5/6 Regression] undefined local symbol on
>> powerpc
>> +
>> +This patch removes bogus debug info left around by shrink-wrapping,
>> +which on some powerpc targets with just the right register allocation
>> +led to assembly errors.
>> +
>> +Bootstrapped and regression tested powerpc64-linux and x86_64-linux.
>> +
>> +https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65779
>> +
>> +gcc/
>> +       PR debug/65779
>> +       * shrink-wrap.c (insn_uses_reg): New function.
>> +       (move_insn_for_shrink_wrap): Remove debug insns using regs set
>> +       by the moved insn.
>> +gcc/testsuite/
>> +       * gcc.dg/pr65779.c: New.
>> +
>> +Upstream-Status: Pending (from mailing list, not merged yet)
>> +
>> +Index: a/gcc/shrink-wrap.c
>> +===================================================================
>> +--- a/gcc/shrink-wrap.c.orig
>> ++++ b/gcc/shrink-wrap.c
>> +@@ -182,6 +182,21 @@ live_edge_for_reg (basic_block bb, int r
>> +   return live_edge;
>> + }
>> +
>> ++static bool
>> ++insn_uses_reg (rtx_insn *insn, unsigned int regno, unsigned int
>> end_regno)
>> ++{
>> ++  df_ref use;
>> ++
>> ++  FOR_EACH_INSN_USE (use, insn)
>> ++    {
>> ++      rtx reg = DF_REF_REG (use);
>> ++
>> ++      if (REG_P (reg) && REGNO (reg) >= regno && REGNO (reg) <
>> end_regno)
>> ++      return true;
>> ++    }
>> ++  return false;
>> ++}
>> ++
>> + /* Try to move INSN from BB to a successor.  Return true on success.
>> +    USES and DEFS are the set of registers that are used and defined
>> +    after INSN in BB.  SPLIT_P indicates whether a live edge from BB
>> +@@ -340,10 +355,15 @@ move_insn_for_shrink_wrap (basic_block b
>> +       *split_p = true;
>> +     }
>> +
>> ++  vec<basic_block> live_bbs;
>> ++  if (MAY_HAVE_DEBUG_INSNS)
>> ++    live_bbs.create (5);
>> +   /* At this point we are committed to moving INSN, but let's try to
>> +      move it as far as we can.  */
>> +   do
>> +     {
>> ++      if (MAY_HAVE_DEBUG_INSNS)
>> ++      live_bbs.safe_push (bb);
>> +       live_out = df_get_live_out (bb);
>> +       live_in = df_get_live_in (next_block);
>> +       bb = next_block;
>> +@@ -426,6 +446,34 @@ move_insn_for_shrink_wrap (basic_block b
>> +       SET_REGNO_REG_SET (bb_uses, i);
>> +     }
>> +
>> ++  /* Remove debug insns using regs set by the insn we are moving.  */
>> ++  if (MAY_HAVE_DEBUG_INSNS)
>> ++    {
>> ++      while (!live_bbs.is_empty ())
>> ++      {
>> ++        rtx_insn *dinsn;
>> ++        basic_block tmp_bb = live_bbs.pop ();
>> ++
>> ++        FOR_BB_INSNS_REVERSE (tmp_bb, dinsn)
>> ++          {
>> ++            if (dinsn == insn)
>> ++              break;
>> ++            if (DEBUG_INSN_P (dinsn)
>> ++                && insn_uses_reg (dinsn, dregno, end_dregno))
>> ++              {
>> ++                if (*split_p)
>> ++                  /* If split, then we will be moving insn into a
>> ++                     newly created block immediately after the entry
>> ++                     block.  Move the debug info there too.  */
>> ++                  emit_debug_insn_after (PATTERN (dinsn), bb_note (bb));
>> ++                delete_insn (dinsn);
>> ++                break;
>> ++              }
>> ++          }
>> ++      }
>> ++      live_bbs.release ();
>> ++    }
>> ++
>> +   emit_insn_after (PATTERN (insn), bb_note (bb));
>> +   delete_insn (insn);
>> +   return true;
>> +Index: b/gcc/testsuite/gcc.dg/pr65779.c
>> +===================================================================
>> +--- /dev/null
>> ++++ b/gcc/testsuite/gcc.dg/pr65779.c
>> +@@ -0,0 +1,64 @@
>> ++/* { dg-do run } */
>> ++/* { dg-options "-O2 -g" } */
>> ++/* { dg-additional-options "-mrelocatable" { target powerpc-*-rtems* } }
>> */
>> ++
>> ++unsigned long __attribute__ ((noinline))
>> ++adler32 (unsigned long adler, unsigned char *buf, unsigned int len)
>> ++{
>> ++  unsigned long s1 = adler & 0xffff;
>> ++  unsigned long s2 = (adler >> 16) & 0xffff;
>> ++  int k;
>> ++
>> ++  if (buf == 0)
>> ++    return 1L;
>> ++
>> ++  while (len > 0)
>> ++    {
>> ++      k = len < 5552 ? len : 5552;
>> ++      len -= k;
>> ++      while (k >= 16)
>> ++      {
>> ++        s1 += *buf++; s2 += s1;
>> ++        s1 += *buf++; s2 += s1;
>> ++        s1 += *buf++; s2 += s1;
>> ++        s1 += *buf++; s2 += s1;
>> ++        s1 += *buf++; s2 += s1;
>> ++        s1 += *buf++; s2 += s1;
>> ++        s1 += *buf++; s2 += s1;
>> ++        s1 += *buf++; s2 += s1;
>> ++        s1 += *buf++; s2 += s1;
>> ++        s1 += *buf++; s2 += s1;
>> ++        s1 += *buf++; s2 += s1;
>> ++        s1 += *buf++; s2 += s1;
>> ++        s1 += *buf++; s2 += s1;
>> ++        s1 += *buf++; s2 += s1;
>> ++        s1 += *buf++; s2 += s1;
>> ++        s1 += *buf++; s2 += s1;
>> ++        k -= 16;
>> ++      }
>> ++      if (k != 0)
>> ++      do
>> ++        {
>> ++          s1 += *buf++; s2 += s1;
>> ++        } while (--k);
>> ++      s1 &= 0xffffffffUL;
>> ++      s2 &= 0xffffffffUL;
>> ++      s1 %= 65521L;
>> ++      s2 %= 65521L;
>> ++    }
>> ++  return (s2 << 16) | s1;
>> ++}
>> ++
>> ++unsigned char buf[] = { 0, 1, 2, 3, 4, 5, 6, 7,
>> ++                      8, 9, 10, 11, 12, 13, 14, 15,
>> ++                      0x80, 0x81, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87,
>> ++                      0x88, 0x89, 0x8a, 0x8b, 0x8c, 0x8d, 0x8e, 0x8f,
>> ++                      0x55, 0xaa };
>> ++int
>> ++main ()
>> ++{
>> ++  unsigned long x = adler32 (0, buf, sizeof buf);
>> ++  if (x != 0x640409efUL)
>> ++    __builtin_abort ();
>> ++  return 0;
>> ++}
>>
>>
>

Re: [PATCH] gcc5: Add PR65779 patch to fix powerpc compile issues

[Richard Purdie]

On Mon, 2015-05-11 at 17:24 -0700, Khem Raj wrote:
> On Mon, May 11, 2015 at 4:20 PM, akuster808 <akuster808@gmail.com> wrote:
> >
> >
> > On 05/11/2015 09:54 AM, Richard Purdie wrote:
> >>
> >> This fixes compile issues on powerpc with gcc 5 which show up with
> >> errors like:
> >>
> >> | make[2]: Entering directory
> >> '/media/build1/poky/build/tmp/work/ppc7400-poky-linux/xprop/1_1.2.2-r0/build'
> >> | powerpc-poky-linux-gcc  -m32 -mhard-float -mcpu=7400
> >> --sysroot=/media/build1/poky/build/tmp/sysroots/qemuppc -Wall
> >> -Wpointer-arith -Wmissing-declarations -Wformat=2 -Wstrict-prototypes
> >> -Wmissing-prototypes -Wnested-externs -Wbad-function-cast
> >> -Wold-style-definition -Wdeclaration-after-statement -Wunused
> >> -Wuninitialized -Wshadow -Wmissing-noreturn -Wmissing-format-attribute
> >> -Wredundant-decls -Wlogical-op -Werror=implicit -Werror=nonnull
> >> -Werror=init-self -Werror=main -Werror=missing-braces -Werror=sequence-point
> >> -Werror=return-type -Werror=trigraphs -Werror=array-bounds
> >> -Werror=write-strings -Werror=address -Werror=int-to-pointer-cast
> >> -Werror=pointer-to-int-cast -fno-strict-aliasing  -O2 -pipe -g
> >> -feliminate-unused-debug-types  -Wl,-O1 -Wl,--hash-style=gnu -Wl,--as-needed
> >> -o xprop dsimple.o clientwin.o xprop.o -lX11
> >> | /media/build1/poky/build/tmp/sysroots/qemuppc/usr/lib/../lib/libX11.so:
> >> undefined reference to `.LCL2'
> >> | collect2: error: ld returned 1 exit status
> >>
> >> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
> >
> >
> > Acked-by: Armin Kuster <Akuster808@gmail.com>
> >
> >
> > this fixes all but one package(lzop) failure on qemuppc base yocto world
> > build.
> >
> 
> I am fine witj this patch for now, since we are mostly build testing,
> we have to keep in mind
> its a codegen bug and can bite us at runtime. I am just hoping for an
> update to this patch and that will eliminate
> that doubt.

Agreed, an update upstream would obviously help. FWIW the automated
runtime tests did look reasonable on the autobuilder apart from:

core-image-sato-sdk has a problem with the C++ toolchain not finding
limits (all arches).

qemuarm is not booting (minimal, sato or sato-sdk).

However these issues were present before the patch so I don't think its
related.

Since the failures list is better with the patch I'll merge it until
something better comes along.

Cheers,

Richard

Re: [PATCH] gcc5: Add PR65779 patch to fix powerpc compile issues

[Khem Raj]

[-- Attachment #1: Type: text/plain, Size: 2754 bytes --]


> On May 12, 2015, at 12:28 AM, Richard Purdie <richard.purdie@linuxfoundation.org> wrote:
> 
> On Mon, 2015-05-11 at 17:24 -0700, Khem Raj wrote:
>> On Mon, May 11, 2015 at 4:20 PM, akuster808 <akuster808@gmail.com> wrote:
>>> 
>>> 
>>> On 05/11/2015 09:54 AM, Richard Purdie wrote:
>>>> 
>>>> This fixes compile issues on powerpc with gcc 5 which show up with
>>>> errors like:
>>>> 
>>>> | make[2]: Entering directory
>>>> '/media/build1/poky/build/tmp/work/ppc7400-poky-linux/xprop/1_1.2.2-r0/build'
>>>> | powerpc-poky-linux-gcc  -m32 -mhard-float -mcpu=7400
>>>> --sysroot=/media/build1/poky/build/tmp/sysroots/qemuppc -Wall
>>>> -Wpointer-arith -Wmissing-declarations -Wformat=2 -Wstrict-prototypes
>>>> -Wmissing-prototypes -Wnested-externs -Wbad-function-cast
>>>> -Wold-style-definition -Wdeclaration-after-statement -Wunused
>>>> -Wuninitialized -Wshadow -Wmissing-noreturn -Wmissing-format-attribute
>>>> -Wredundant-decls -Wlogical-op -Werror=implicit -Werror=nonnull
>>>> -Werror=init-self -Werror=main -Werror=missing-braces -Werror=sequence-point
>>>> -Werror=return-type -Werror=trigraphs -Werror=array-bounds
>>>> -Werror=write-strings -Werror=address -Werror=int-to-pointer-cast
>>>> -Werror=pointer-to-int-cast -fno-strict-aliasing  -O2 -pipe -g
>>>> -feliminate-unused-debug-types  -Wl,-O1 -Wl,--hash-style=gnu -Wl,--as-needed
>>>> -o xprop dsimple.o clientwin.o xprop.o -lX11
>>>> | /media/build1/poky/build/tmp/sysroots/qemuppc/usr/lib/../lib/libX11.so:
>>>> undefined reference to `.LCL2'
>>>> | collect2: error: ld returned 1 exit status
>>>> 
>>>> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
>>> 
>>> 
>>> Acked-by: Armin Kuster <Akuster808@gmail.com>
>>> 
>>> 
>>> this fixes all but one package(lzop) failure on qemuppc base yocto world
>>> build.
>>> 
>> 
>> I am fine witj this patch for now, since we are mostly build testing,
>> we have to keep in mind
>> its a codegen bug and can bite us at runtime. I am just hoping for an
>> update to this patch and that will eliminate
>> that doubt.
> 
> Agreed, an update upstream would obviously help. FWIW the automated
> runtime tests did look reasonable on the autobuilder apart from:
> 
> core-image-sato-sdk has a problem with the C++ toolchain not finding
> limits (all arches).
> 

hmm thats interesting. Can you point to an error for more details ?

> qemuarm is not booting (minimal, sato or sato-sdk).

yes I have had problems with linux-yocto on emulator too.

> 
> However these issues were present before the patch so I don't think its
> related.

right

> 
> Since the failures list is better with the patch I'll merge it until
> something better comes along.

thats OK


[-- Attachment #2: Message signed with OpenPGP using GPGMail --]
[-- Type: application/pgp-signature, Size: 211 bytes --]