qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Peter Maydell <peter.maydell@linaro.org>
To: qemu-devel@nongnu.org
Cc: patches@linaro.org, Richard Henderson <rth@twiddle.net>,
	Riku Voipio <riku.voipio@iki.fi>
Subject: [Qemu-devel] [PATCH v2] linux-user: provide frame information in x86-64 safe_syscall
Date: Mon,  6 Jun 2016 19:56:54 +0100	[thread overview]
Message-ID: <1465239414-4932-1-git-send-email-peter.maydell@linaro.org> (raw)

Use cfi directives in the x86-64 safe_syscall to allow gdb to get
backtraces right from within it. (In particular this will be
quite a common situation if the user interrupts QEMU while it's
in a blocked safe-syscall: at the point of the syscall insn RBP
is in use for something else, and so gdb can't find the frame then
without assistance.)

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
---
v1->v2 changes: minor tweaks as requested by rth

I was planning to send this in a series with some other safe-syscall code
for the other architectures, but I haven't got round to those yet, so
might as well send out the fixed version of this by itself.

 linux-user/host/x86_64/safe-syscall.inc.S | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/linux-user/host/x86_64/safe-syscall.inc.S b/linux-user/host/x86_64/safe-syscall.inc.S
index dde434c..e09368d 100644
--- a/linux-user/host/x86_64/safe-syscall.inc.S
+++ b/linux-user/host/x86_64/safe-syscall.inc.S
@@ -24,6 +24,7 @@
          * -1-and-errno-set convention is done by the calling wrapper.
          */
 safe_syscall_base:
+        .cfi_startproc
         /* This saves a frame pointer and aligns the stack for the syscall.
          * (It's unclear if the syscall ABI has the same stack alignment
          * requirements as the userspace function call ABI, but better safe than
@@ -31,6 +32,8 @@ safe_syscall_base:
          * does not list any ABI differences regarding stack alignment.)
          */
         push    %rbp
+        .cfi_adjust_cfa_offset 8
+        .cfi_rel_offset rbp, 0
 
         /* The syscall calling convention isn't the same as the
          * C one:
@@ -70,12 +73,19 @@ safe_syscall_start:
 safe_syscall_end:
         /* code path for having successfully executed the syscall */
         pop     %rbp
+        .cfi_remember_state
+        .cfi_def_cfa_offset 8
+        .cfi_restore rbp
         ret
 
 return_ERESTARTSYS:
         /* code path when we didn't execute the syscall */
+        .cfi_restore_state
         mov     $-TARGET_ERESTARTSYS, %rax
         pop     %rbp
+        .cfi_def_cfa_offset 8
+        .cfi_restore rbp
         ret
+        .cfi_endproc
 
         .size   safe_syscall_base, .-safe_syscall_base
-- 
1.9.1

             reply	other threads:[~2016-06-06 18:57 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-06-06 18:56 Peter Maydell [this message]
2016-06-06 21:03 ` [Qemu-devel] [PATCH v2] linux-user: provide frame information in x86-64 safe_syscall Richard Henderson

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=1465239414-4932-1-git-send-email-peter.maydell@linaro.org \
    --to=peter.maydell@linaro.org \
    --cc=patches@linaro.org \
    --cc=qemu-devel@nongnu.org \
    --cc=riku.voipio@iki.fi \
    --cc=rth@twiddle.net \
    /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 NNTP newsgroup(s).