public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Yao Qi <yao@codesourcery.com>
To: <gdb-patches@sourceware.org>
Subject: [PATCH] Detect 64-bit-ness in PowerPC Book III-E
Date: Thu, 11 Dec 2014 11:15:00 -0000	[thread overview]
Message-ID: <1418296515-8689-1-git-send-email-yao@codesourcery.com> (raw)

This patch is to teach GDBServer to detect 64-bit inferior correctly.
We find a problem that GDBServer is unable to detect on an e5500 core
processor.  Current GDBServer assumes that MSR is a 64-bit register,
but MSR is a 32-bit register in Book III-E.  This patch is to fix this
problem by checking both bits in MSR, in order to handle both Book
III-S and Book III-E.

Is it OK?

gdb/gdbserver:

2014-12-11  Yao Qi  <yao@codesourcery.com>

	* linux-ppc-low.c (ppc_arch_setup) [__powerpc64__]: Change
	variable msr to type 'unsigned long'.  Check bit 63 or bit 31
	is one.
---
 gdb/gdbserver/linux-ppc-low.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/gdb/gdbserver/linux-ppc-low.c b/gdb/gdbserver/linux-ppc-low.c
index 8fd4b38..0c0ced6 100644
--- a/gdb/gdbserver/linux-ppc-low.c
+++ b/gdb/gdbserver/linux-ppc-low.c
@@ -401,7 +401,7 @@ ppc_arch_setup (void)
 {
   const struct target_desc *tdesc;
 #ifdef __powerpc64__
-  long msr;
+  unsigned long msr;
   struct regcache *regcache;
 
   /* On a 64-bit host, assume 64-bit inferior process with no
@@ -411,13 +411,17 @@ ppc_arch_setup (void)
   current_process ()->tdesc = tdesc;
   ppc_hwcap = 0;
 
-  /* Only if the high bit of the MSR is set, we actually have
-     a 64-bit inferior.  */
+  /* We actually have a 64-bit inferior only if the certain bit of the
+     MSR is set.  The PowerISA Book III-S MSR is different from the
+     PowerISA Book III-E MSR.  The Book III-S MSR is 64 bits wide, and
+     its MSR[SF] is the bit 0 of a 64-bit value.  Book III-E MSR is 32
+     bits wide, and its MSR[CM] is the bit 0 of a 32-bit value.  We check
+     both here.  */
   regcache = new_register_cache (tdesc);
   fetch_inferior_registers (regcache, find_regno (tdesc, "msr"));
   collect_register_by_name (regcache, "msr", &msr);
   free_register_cache (regcache);
-  if (msr < 0)
+  if (msr & 0x8000000080000000)
     {
       ppc_get_hwcap (&ppc_hwcap);
       if (ppc_hwcap & PPC_FEATURE_CELL)
-- 
1.9.3

             reply	other threads:[~2014-12-11 11:15 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-12-11 11:15 Yao Qi [this message]
2014-12-11 15:31 ` Ulrich Weigand
2014-12-19  8:06   ` Yao Qi
2014-12-19 15:49     ` Ulrich Weigand
2015-01-11  8:27       ` [PATCH 0/2, V2] " Yao Qi
2015-01-11  8:27         ` [PATCH 1/2] Move some ppc macros to nat/ppc-linux.h Yao Qi
2015-01-13 16:57           ` Ulrich Weigand
2015-01-11  8:27         ` [PATCH 2/2] Detect 64-bit-ness in PowerPC Book III-E Yao Qi
2015-01-13 17:00           ` Ulrich Weigand
2015-01-14  2:54             ` Yao Qi
2015-01-14 12:47               ` Ulrich Weigand
2015-01-14 14:22                 ` Yao Qi

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=1418296515-8689-1-git-send-email-yao@codesourcery.com \
    --to=yao@codesourcery.com \
    --cc=gdb-patches@sourceware.org \
    /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).