public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Luis Machado <luis.machado@arm.com>
To: Yichao Yu <yyc1992@gmail.com>, gdb-patches@sourceware.org
Subject: Re: [PATCH v3] Return the regnum for PC (32) on aarch64
Date: Wed, 18 May 2022 10:53:55 +0100	[thread overview]
Message-ID: <8d311bd0-d814-934f-8639-8841a10c803d@arm.com> (raw)
In-Reply-To: <CAMvDr+QzsPQyChNormEVa2T+deVNJ8HYtoas1O-qhujhnPbVvA@mail.gmail.com>

On 5/18/22 02:03, Yichao Yu wrote:
> Sorry for the duplicated post.... I think somehow the university
> network prevented the email from going out earlier today...

No problem. I've been having some inbound e-mail issues as well.

> 
> On Tue, May 17, 2022 at 8:45 PM Yichao Yu <yyc1992@gmail.com> wrote:
>>
>> This will allow the unwind info to explicitly specify a different value
>> for the return address from the link register.
>> Such usage, although uncommon, is valid and useful for signal frames.
>> It is also supported by aadwarf64 from ARM (Note 9 in [1]).
>>
>> Ref https://sourceware.org/pipermail/gdb/2022-May/050091.html
>>
>> [1] https://github.com/ARM-software/abi-aa/blob/2022Q1/aadwarf64/aadwarf64.rst#dwarf-register-names
>> ---
>>   gdb/aarch64-tdep.c                           |  3 +
>>   gdb/aarch64-tdep.h                           |  1 +
>>   gdb/testsuite/gdb.arch/aarch64-unwind-pc.S   | 48 ++++++++++++++
>>   gdb/testsuite/gdb.arch/aarch64-unwind-pc.exp | 70 ++++++++++++++++++++
>>   4 files changed, 122 insertions(+)
>>   create mode 100644 gdb/testsuite/gdb.arch/aarch64-unwind-pc.S
>>   create mode 100644 gdb/testsuite/gdb.arch/aarch64-unwind-pc.exp
>>
>> diff --git a/gdb/aarch64-tdep.c b/gdb/aarch64-tdep.c
>> index 9d06ebfe27c..f9eb455232f 100644
>> --- a/gdb/aarch64-tdep.c
>> +++ b/gdb/aarch64-tdep.c
>> @@ -2219,6 +2219,9 @@ aarch64_dwarf_reg_to_regnum (struct gdbarch *gdbarch, int reg)
>>     if (reg == AARCH64_DWARF_SP)
>>       return AARCH64_SP_REGNUM;
>>
>> +  if (reg == AARCH64_DWARF_PC)
>> +    return AARCH64_PC_REGNUM;
>> +
>>     if (reg >= AARCH64_DWARF_V0 && reg <= AARCH64_DWARF_V0 + 31)
>>       return AARCH64_V0_REGNUM + reg - AARCH64_DWARF_V0;
>>
>> diff --git a/gdb/aarch64-tdep.h b/gdb/aarch64-tdep.h
>> index e4cdebb6311..aa1bedbdac6 100644
>> --- a/gdb/aarch64-tdep.h
>> +++ b/gdb/aarch64-tdep.h
>> @@ -34,6 +34,7 @@ struct regset;
>>   /* AArch64 Dwarf register numbering.  */
>>   #define AARCH64_DWARF_X0   0
>>   #define AARCH64_DWARF_SP  31
>> +#define AARCH64_DWARF_PC  32
>>   #define AARCH64_DWARF_PAUTH_RA_STATE  34
>>   #define AARCH64_DWARF_PAUTH_DMASK  35
>>   #define AARCH64_DWARF_PAUTH_CMASK  36
>> diff --git a/gdb/testsuite/gdb.arch/aarch64-unwind-pc.S b/gdb/testsuite/gdb.arch/aarch64-unwind-pc.S
>> new file mode 100644
>> index 00000000000..6cc4f80e349
>> --- /dev/null
>> +++ b/gdb/testsuite/gdb.arch/aarch64-unwind-pc.S
>> @@ -0,0 +1,48 @@
>> +/* Copyright 2022 Free Software Foundation, Inc.
>> +
>> +   This file is part of GDB.
>> +
>> +   This program is free software; you can redistribute it and/or modify
>> +   it under the terms of the GNU General Public License as published by
>> +   the Free Software Foundation; either version 3 of the License, or
>> +   (at your option) any later version.
>> +
>> +   This program is distributed in the hope that it will be useful,
>> +   but WITHOUT ANY WARRANTY; without even the implied warranty of
>> +   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
>> +   GNU General Public License for more details.
>> +
>> +   You should have received a copy of the GNU General Public License
>> +   along with this program.  If not, see <http://www.gnu.org/licenses/>.  */
>> +
>> +        .text
>> +        .globl main
>> +        .type main,#function
>> +main:
>> +        .cfi_startproc
>> +        stp     x29, x30, [sp, -16]!
>> +        .cfi_def_cfa sp, 16
>> +        .cfi_offset x29, 0
>> +        .cfi_offset x30, 8
>> +        bl test_func
>> +        ldp     x29, x30, [sp], 16
>> +        .cfi_restore x29
>> +        .cfi_restore x30
>> +        .cfi_def_cfa sp, 0
>> +        mov     x0, 0
>> +        ret
>> +        .cfi_endproc
>> +
>> +        .globl test_func
>> +test_func:
>> +        .cfi_startproc
>> +        // Unwind x30 to a different value
>> +        // CFA_val_expression x30 const2u 0x1234
>> +        .cfi_escape 0x16, 30, 0x03, 0x0a, 0x34, 0x12
>> +        // CFA_val_expression pc breg30 0
>> +        .cfi_escape 0x16, 32, 0x02, 0x8e, 0x00
>> +        mov     x0, x30
>> +        .cfi_register 32, x0
>> +        mov     x30, 0x1234
>> +        ret     x0
>> +        .cfi_endproc
>> diff --git a/gdb/testsuite/gdb.arch/aarch64-unwind-pc.exp b/gdb/testsuite/gdb.arch/aarch64-unwind-pc.exp
>> new file mode 100644
>> index 00000000000..105b9a9cc3a
>> --- /dev/null
>> +++ b/gdb/testsuite/gdb.arch/aarch64-unwind-pc.exp
>> @@ -0,0 +1,70 @@
>> +# Copyright 2022 Free Software Foundation, Inc.
>> +
>> +# This program is free software; you can redistribute it and/or modify
>> +# it under the terms of the GNU General Public License as published by
>> +# the Free Software Foundation; either version 3 of the License, or
>> +# (at your option) any later version.
>> +#
>> +# This program is distributed in the hope that it will be useful,
>> +# but WITHOUT ANY WARRANTY; without even the implied warranty of
>> +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
>> +# GNU General Public License for more details.
>> +#
>> +# You should have received a copy of the GNU General Public License
>> +# along with this program.  If not, see <http://www.gnu.org/licenses/>.
>> +
>> +# This file is part of the gdb testsuite.
>> +
>> +# Test explicitly unwinding the PC DWARF register on aarch64
>> +
>> +if {![is_aarch64_target]} then {
>> +    verbose "Skipping ${gdb_test_file_name}."
>> +    return
>> +}
>> +
>> +standard_testfile .S
>> +
>> +if { [prepare_for_testing "failed to prepare" $testfile $srcfile] } {
>> +    return -1
>> +}
>> +
>> +if ![runto_main] {
>> +    return -1
>> +}
>> +
>> +proc test_reg_vals {} {
>> +    gdb_test "p \$pc - &main" "= 8" "p \$pc"
>> +    gdb_test "p/x \$x30" "= 0x1234" "p \$x30"
>> +}
>> +
>> +proc test_unwind_pc { inst } {
>> +    gdb_test "si" "$inst" "single step"
>> +    gdb_test "backtrace" \
>> +       ".*#1.*in main ().*" \
>> +       "backtrace"
>> +    gdb_test "up" "in main ().*" "parent frame"
>> +    test_reg_vals
>> +}
>> +
>> +# Ready to enter the function
>> +gdb_test "si" "bl test_func" "call site"
>> +# Step through the 3 instructions in the function to make sure that
>> +# we have the same unwind info throughout.
>> +with_test_prefix "1st stepi" {
>> +    test_unwind_pc "mov     x0, x30"
>> +}
>> +with_test_prefix "2nd stepi" {
>> +    test_unwind_pc "mov     x30, 0x1234"
>> +}
>> +with_test_prefix "3rd stepi" {
>> +    test_unwind_pc "ret     x0"
>> +}
>> +# Check again after we returned
>> +with_test_prefix "final" {
>> +    # Check that we've stepped out (si prints out the new function name)
>> +    gdb_test "si" ".*main *().*" "single step out"
>> +    gdb_test "backtrace" \
>> +       "#0\[\t \]+main ().*" \
>> +       "backtrace"
>> +    test_reg_vals
>> +}
>> --
>> 2.36.1
>>

Thanks. This LGTM for aarch64. Please push it.

  reply	other threads:[~2022-05-18  9:54 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-05-13 12:27 [PATCH] " Yichao Yu
2022-05-16  9:05 ` Luis Machado
2022-05-16 13:17   ` Yichao Yu
2022-05-16 13:25     ` [PATCH v2] " Yichao Yu
2022-05-17  6:42       ` Luis Machado
2022-05-17 14:41         ` [PATCH v3] " Yichao Yu
2022-05-17 17:34         ` [PATCH v2] " Yichao Yu
2022-05-17 17:36           ` [PATCH v3] " Yichao Yu
2022-05-18  1:03             ` Yichao Yu
2022-05-18  9:53               ` Luis Machado [this message]
2022-05-18 13:01                 ` Yichao Yu
2022-05-18 13:36                   ` Luis Machado
2022-05-18 14:01                     ` Yichao Yu
2022-05-18 14:03                       ` Luis Machado
2022-05-19  1:11                         ` Yichao Yu
2022-05-16 13:29     ` [PATCH] " Luis Machado
2022-05-16 14:10       ` Yichao Yu

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=8d311bd0-d814-934f-8639-8841a10c803d@arm.com \
    --to=luis.machado@arm.com \
    --cc=gdb-patches@sourceware.org \
    --cc=yyc1992@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).