Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
From: "H.J. Lu" <hongjiu.lu@intel.com>
To: GDB <gdb-patches@sourceware.org>
Subject: Re: PATCH: Fix 32bit coredump read on Linux/AVX
Date: Tue, 20 Apr 2010 18:43:00 -0000	[thread overview]
Message-ID: <20100420184339.GA18641@intel.com> (raw)
In-Reply-To: <20100420032234.GA10544@intel.com>

On Mon, Apr 19, 2010 at 08:22:34PM -0700, H.J. Lu wrote:
> Hi,
> 
> This patch:
> 
> http://sourceware.org/ml/gdb-patches/2010-04/msg00276.html
> 
> breaks 32bit coredump read on Linux/AVX since we only dump .reg and
> .reg-xstate sections on AVX. But i386_linux_core_read_description
> checks .reg2 and .reg-xfp sections first. Since there are no
> .reg2 and .reg-xfp sections, NULL is returned and SSE target is used.
> This patch changes the section check order to .reg-xstate, .reg-xfp,
> .reg2.  OK to install?
> 
> Thanks.
> 

.reg2 section has x87 regiters on i386 and SSE registers on amd64.
Here is the updated patch to properly handle it.  OK to install?

Thanks.


H.J.
--
2010-04-20  H.J. Lu  <hongjiu.lu@intel.com>

	PR corefiles/11523
	* amd64-linux-tdep.c (amd64_linux_core_read_description): Check
	XCR0 first.

	* i386-linux-tdep.c (i386_linux_core_read_xcr0): Return 0 if
	there is no .reg-xstate section.
	(i386_linux_core_read_description): Check XCR0 first.

diff --git a/gdb/amd64-linux-tdep.c b/gdb/amd64-linux-tdep.c
index f249d5d..1f7a052 100644
--- a/gdb/amd64-linux-tdep.c
+++ b/gdb/amd64-linux-tdep.c
@@ -1269,18 +1269,23 @@ amd64_linux_core_read_description (struct gdbarch *gdbarch,
 				  struct target_ops *target,
 				  bfd *abfd)
 {
-  asection *section = bfd_get_section_by_name (abfd, ".reg2");
-  uint64_t xcr0;
-
-  if (section == NULL)
-    return NULL;
-
   /* Linux/x86-64.  */
-  xcr0 = i386_linux_core_read_xcr0 (gdbarch, target, abfd);
-  if ((xcr0 & I386_XSTATE_AVX_MASK) == I386_XSTATE_AVX_MASK)
-    return tdesc_amd64_avx_linux;
-  else
+  uint64_t xcr0 = i386_linux_core_read_xcr0 (gdbarch, target, abfd);
+  switch ((xcr0 & I386_XSTATE_AVX_MASK))
+    {
+    case I386_XSTATE_AVX_MASK:
+      return tdesc_amd64_avx_linux;
+      return tdesc_i386_avx_linux;
+    case I386_XSTATE_SSE_MASK:
+      return tdesc_amd64_linux;
+    default:
+      break;
+    }
+
+  if (bfd_get_section_by_name (abfd, ".reg2") != NULL)
     return tdesc_amd64_linux;
+  else
+    return NULL;
 }
 
 static void
diff --git a/gdb/i386-linux-tdep.c b/gdb/i386-linux-tdep.c
index 5952153..ee7c23f 100644
--- a/gdb/i386-linux-tdep.c
+++ b/gdb/i386-linux-tdep.c
@@ -611,7 +611,7 @@ i386_linux_core_read_xcr0 (struct gdbarch *gdbarch,
 	}
     }
   else
-    xcr0 = I386_XSTATE_SSE_MASK;
+    xcr0 = 0;
 
   return xcr0;
 }
@@ -623,22 +623,26 @@ i386_linux_core_read_description (struct gdbarch *gdbarch,
 				  struct target_ops *target,
 				  bfd *abfd)
 {
-  asection *section = bfd_get_section_by_name (abfd, ".reg2");
-  uint64_t xcr0;
-
-  if (section == NULL)
-    return NULL;
+  /* Linux/i386.  */
+  uint64_t xcr0 = i386_linux_core_read_xcr0 (gdbarch, target, abfd);
+  switch ((xcr0 & I386_XSTATE_AVX_MASK))
+    {
+    case I386_XSTATE_AVX_MASK:
+      return tdesc_i386_avx_linux;
+    case I386_XSTATE_SSE_MASK:
+      return tdesc_i386_linux;
+    case I386_XSTATE_X87_MASK:
+      return tdesc_i386_mmx_linux;
+    default:
+      break;
+    }
 
-  section = bfd_get_section_by_name (abfd, ".reg-xfp");
-  if (section == NULL)
+  if (bfd_get_section_by_name (abfd, ".reg-xfp") != NULL)
+    return tdesc_i386_linux;
+  else if (bfd_get_section_by_name (abfd, ".reg2") != NULL)
     return tdesc_i386_mmx_linux;
-
-  /* Linux/i386.  */
-  xcr0 = i386_linux_core_read_xcr0 (gdbarch, target, abfd);
-  if ((xcr0 & I386_XSTATE_AVX_MASK) == I386_XSTATE_AVX_MASK)
-    return tdesc_i386_avx_linux;
   else
-    return tdesc_i386_linux;
+    return NULL;
 }
 
 static void


  reply	other threads:[~2010-04-20 18:43 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-04-20  3:22 H.J. Lu
2010-04-20 18:43 ` H.J. Lu [this message]
2010-04-21 14:32   ` H.J. Lu
2010-04-21 19:38   ` Mark Kettenis
2010-04-21 20:18   ` H.J. Lu
2010-04-21 20:24     ` Mark Kettenis

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=20100420184339.GA18641@intel.com \
    --to=hongjiu.lu@intel.com \
    --cc=gdb-patches@sourceware.org \
    --cc=hjl.tools@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