public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: "H.J. Lu" <hjl.tools@gmail.com>
To: gcc-patches@gcc.gnu.org, Uros Bizjak <ubizjak@gmail.com>,
	Jan Hubicka <hubicka@ucw.cz>
Cc: 刘袋鼠 <crazylht@gmail.com>
Subject: V2 [PATCH] i386: Insert ENDBR for NOTE_INSN_DELETED_LABEL only if needed
Date: Sat, 16 Feb 2019 15:02:00 -0000	[thread overview]
Message-ID: <20190216150211.GA13329@gmail.com> (raw)
In-Reply-To: <20190215041332.24720-1-hjl.tools@gmail.com>

On Thu, Feb 14, 2019 at 08:13:32PM -0800, H.J. Lu wrote:
> NOTE_INSN_DELETED_LABEL is used to mark what used to be a 'code_label',
> but was not used for other purposes than taking its address and was
> transformed to mark that no code jumps to it.  NOTE_INSN_DELETED_LABEL
> is generated only in 3 places:
> 
> 1. When delete_insn sees an unused label which is an explicit label in
> the input source code or its address is taken, it turns the label into
> a NOTE_INSN_DELETED_LABEL note.
> 2. When rtl_tidy_fallthru_edge deletes a tablejump, it turns the
> tablejump into a NOTE_INSN_DELETED_LABEL note.
> 3. ix86_init_large_pic_reg creats a NOTE_INSN_DELETED_LABEL note, .L2,
> to initialize large model PIC register:
> 
> L2:
> 	movabsq	$_GLOBAL_OFFSET_TABLE_-.L2, %r11
> 	leaq	.L2(%rip), %rax
> 	movabsq	$val@GOT, %rdx
> 	addq	%r11, %rax
> 
> Among of them, ENDBR is needed only when the label address is taken.
> rest_of_insert_endbranch has
> 
>           if ((LABEL_P (insn) && LABEL_PRESERVE_P (insn))
>               || (NOTE_P (insn)
>                   && NOTE_KIND (insn) == NOTE_INSN_DELETED_LABEL))
>             /* TODO.  Check /s bit also.  */
>             {
>               cet_eb = gen_nop_endbr ();
>               emit_insn_after (cet_eb, insn);
>               continue;
>             }
> 
> For NOTE_INSN_DELETED_LABEL, we should check if forced_labels to see
> if its address is taken.  Also ix86_init_large_pic_reg shouldn't set
> LABEL_PRESERVE_P (in_struct) since NOTE_INSN_DELETED_LABEL is suffcient
> to keep the label.
> 
> gcc/
> 
> 	PR target/89355
> 	* config/i386/i386.c (rest_of_insert_endbranch): Check
> 	forced_labels to see if the address of NOTE_INSN_DELETED_LABEL
> 	is taken.
> 	(ix86_init_large_pic_reg): Don't set LABEL_PRESERVE_P.
> 

Here is the updated patch.  We should check LABEL_PRESERVE_P on
NOTE_INSN_DELETED_LABEL to see if its address is taken.

OK for trunk?

Thanks.

H.J.
---
NOTE_INSN_DELETED_LABEL is used to mark what used to be a 'code_label',
but was not used for other purposes than taking its address and was
transformed to mark that no code jumps to it.  Since LABEL_PRESERVE_P is
true only if the label address was taken, check LABEL_PRESERVE_P before
inserting ENDBR.

2019-02-15  H.J. Lu  <hongjiu.lu@intel.com>
	    Hongtao Liu  <hongtao.liu@intel.com>

gcc/

	PR target/89355
	* config/i386/i386.c (rest_of_insert_endbranch): LABEL_PRESERVE_P
	to see if the address of NOTE_INSN_DELETED_LABEL is taken.
	(ix86_init_large_pic_reg): Don't set LABEL_PRESERVE_P.

gcc/testsuite/

	PR target/89355
	* gcc.target/i386/cet-label-3.c: New test.
	* gcc.target/i386/cet-label-4.c: Likewise.
	* gcc.target/i386/cet-label-5.c: Likewise.
---
 gcc/config/i386/i386.c                      |  9 ++++----
 gcc/testsuite/gcc.target/i386/cet-label-3.c | 23 +++++++++++++++++++++
 gcc/testsuite/gcc.target/i386/cet-label-4.c | 12 +++++++++++
 gcc/testsuite/gcc.target/i386/cet-label-5.c | 13 ++++++++++++
 4 files changed, 52 insertions(+), 5 deletions(-)
 create mode 100644 gcc/testsuite/gcc.target/i386/cet-label-3.c
 create mode 100644 gcc/testsuite/gcc.target/i386/cet-label-4.c
 create mode 100644 gcc/testsuite/gcc.target/i386/cet-label-5.c

diff --git a/gcc/config/i386/i386.c b/gcc/config/i386/i386.c
index 609273e4fc4..acdc789b834 100644
--- a/gcc/config/i386/i386.c
+++ b/gcc/config/i386/i386.c
@@ -2734,10 +2734,10 @@ rest_of_insert_endbranch (void)
 	      continue;
 	    }
 
-	  if ((LABEL_P (insn) && LABEL_PRESERVE_P (insn))
-	      || (NOTE_P (insn)
-		  && NOTE_KIND (insn) == NOTE_INSN_DELETED_LABEL))
-	    /* TODO.  Check /s bit also.  */
+	  if ((LABEL_P (insn)
+	       || (NOTE_P (insn)
+		   && NOTE_KIND (insn) == NOTE_INSN_DELETED_LABEL))
+	      && LABEL_PRESERVE_P (insn))
 	    {
 	      cet_eb = gen_nop_endbr ();
 	      emit_insn_after (cet_eb, insn);
@@ -6997,7 +6997,6 @@ ix86_init_large_pic_reg (unsigned int tmp_regno)
   gcc_assert (Pmode == DImode);
   label = gen_label_rtx ();
   emit_label (label);
-  LABEL_PRESERVE_P (label) = 1;
   tmp_reg = gen_rtx_REG (Pmode, tmp_regno);
   gcc_assert (REGNO (pic_offset_table_rtx) != tmp_regno);
   emit_insn (gen_set_rip_rex64 (pic_offset_table_rtx,
diff --git a/gcc/testsuite/gcc.target/i386/cet-label-3.c b/gcc/testsuite/gcc.target/i386/cet-label-3.c
new file mode 100644
index 00000000000..9f427a866f3
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/cet-label-3.c
@@ -0,0 +1,23 @@
+/* PR target/89355  */
+/* { dg-do compile } */
+/* { dg-options "-O2 -fcf-protection" } */
+/* { dg-final { scan-assembler-times "endbr32" 1 { target ia32 } } } */
+/* { dg-final { scan-assembler-times "endbr64" 1 { target { ! ia32 } } } } */
+int
+test (int* val)
+{
+  int status = 99;
+
+  if (!val)
+    {
+      status = 22;
+      goto end;
+    }
+
+  extern int x;
+  *val = x;
+
+  status = 0;
+end:
+  return status;
+}
diff --git a/gcc/testsuite/gcc.target/i386/cet-label-4.c b/gcc/testsuite/gcc.target/i386/cet-label-4.c
new file mode 100644
index 00000000000..d743d2bf202
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/cet-label-4.c
@@ -0,0 +1,12 @@
+/* PR target/89355  */
+/* { dg-do compile { target { fpic && lp64 } } } */
+/* { dg-options "-O2 -fcf-protection -fPIC -mcmodel=large" } */
+/* { dg-final { scan-assembler-times "endbr64" 1 } } */
+
+extern int val;
+
+int
+test (void)
+{
+  return val;
+}
diff --git a/gcc/testsuite/gcc.target/i386/cet-label-5.c b/gcc/testsuite/gcc.target/i386/cet-label-5.c
new file mode 100644
index 00000000000..4d5ca816598
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/cet-label-5.c
@@ -0,0 +1,13 @@
+/* PR target/89355  */
+/* { dg-do compile } */
+/* { dg-options "-O2 -fcf-protection -Wno-return-local-addr" } */
+/* { dg-final { scan-assembler-times "endbr32" 2 { target ia32 } } } */
+/* { dg-final { scan-assembler-times "endbr64" 2 { target { ! ia32 } } } } */
+
+void *
+func (void)
+{
+  return &&bar;
+bar:
+  return 0;
+}
-- 
2.20.1

  reply	other threads:[~2019-02-16 15:02 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-02-15  4:13 H.J. Lu
2019-02-16 15:02 ` H.J. Lu [this message]
2019-05-21 21:48   ` PING^1: V2 " H.J. Lu
2019-05-28  7:21     ` Uros Bizjak
2019-05-28  9:22   ` Jakub Jelinek
2019-05-28 15:16     ` H.J. Lu
2019-05-28 15:31       ` Jakub Jelinek
2019-05-28 15:34         ` H.J. Lu
2019-05-28 16:11           ` Jakub Jelinek
2019-05-28 17:37             ` Jeff Law
2019-05-31 17:39               ` [PATCH] Don't insert ENDBR after NOTE_INSN_DELETED_LABEL H.J. Lu
2019-05-31 18:03                 ` Jakub Jelinek
2019-05-31 18:26                   ` H.J. Lu

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20190216150211.GA13329@gmail.com \
    --to=hjl.tools@gmail.com \
    --cc=crazylht@gmail.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=hubicka@ucw.cz \
    --cc=ubizjak@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
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).