From: Yao Qi <qiyaoltc@gmail.com>
To: Antoine Tremblay <antoine.tremblay@ericsson.com>
Cc: <gdb-patches@sourceware.org>
Subject: Re: [PATCH v6 5/7] Support software single step on ARM in GDBServer.
Date: Mon, 07 Dec 2015 15:18:00 -0000 [thread overview]
Message-ID: <8637ves2xp.fsf@gmail.com> (raw)
In-Reply-To: <1449254773-19019-6-git-send-email-antoine.tremblay@ericsson.com> (Antoine Tremblay's message of "Fri, 4 Dec 2015 13:46:11 -0500")
Antoine Tremblay <antoine.tremblay@ericsson.com> writes:
> gdb/ChangeLog:
> * Makefile.in: Add arm-get-next-pcs.c/o, arm-linux.c/o.
> to all arm targets.
Please describe how Makefile.in is changed, as I suggested in previous patch.
> * aarch32-linux-nat.c: Include arch/arm-get-next-pcs.h.
> * arch/arm-get-next-pcs.c: New file.
> * arch/arm-get-next-pcs.h: New file.
> * arch/arm-linux.h: New file.
> * arch/arm-linux.c: New file.
> * arm-linux-nat.c: Include arch/arm.h, arch/arm-get-next-pcs.h
> arch/arm-linux.h.
> * arm-linux-tdep.c: Include arch/arm.h,arch/arm-get-next-pcs.h
> arch/arm-linux.h.
> (arm_linux_get_next_pcs_ops): New struct.
> (ARM_SIGCONTEXT_R0, ARM_UCONTEXT_SIGCONTEXT,
> ARM_OLD_RT_SIGFRAME_SIGINFO, ARM_OLD_RT_SIGFRAME_UCONTEXT,
> ARM_NEW_RT_SIGFRAME_UCONTEXT, ARM_NEW_SIGFRAME_MAGIC): Move stack
> layout defines to arch/arm-linux.h.
> (arm_linux_sigreturn_next_pc_offset): Move to arch/arm-linux.c.
> (arm_linux_software_single_step): Adjust for
> arm_get_next_pcs implementation.
> * arm-tdep.c: Include arch/arm-get-next-pcs.h.
> (arm_get_next_pcs_ops): New struct.
> (submask): Move macro to arm-get-next-pcs.h.
> (bit): Likewise.
> (bits): Likewise.
> (sbits): Likewise.
> (BranchDest): Likewise.
> (thumb_instruction_changes_pc): Move to arm-get-next-pcs.c
> (thumb2_instruction_changes_pc): Likewise.
> (arm_instruction_changes_pc): Likewise.
I feel they should be moved to arch/arm.c, because they are not about
getting the next pcs.
> (shifted_reg_val): Likewise.
> (thumb_advance_itstate): Likewise.
and this one.
> (thumb_get_next_pc_raw): Likewise.
> (arm_get_next_pc_raw): Likewise.
> (arm_get_next_pc): Likewise.
> (thumb_deal_with_atomic_sequence_raw): Likewise.
> (arm_deal_with_atomic_sequence_raw): Likewise.
> (arm_deal_with_atomic_sequence): Likewise.
> (arm_get_next_pcs_read_memory_unsigned_integer): New function.
> (arm_get_next_pcs_addr_bits_remove): Likewise.
> (arm_get_next_pcs_syscall_next_pc): Likewise.
> (arm_get_next_pcs_is_thumb): Likewise.
> (arm_software_single_step): Adjust for arm_get_next_pcs
> implementation.
> * arm-tdep.h: (arm_get_next_pc): Remove declaration.
> (arm_get_next_pcs_read_memory_unsigned_integer):
> New declaration.
> (arm_get_next_pcs_addr_bits_remove): Likewise.
> (arm_get_next_pcs_syscall_next_pc): Likewise.
> (arm_software_single_step): Likewise.
> (arm_deal_with_atomic_sequence: Remove declaration.
> (arm_get_next_pcs_is_thumb): New delcaration.
> * arm-wince-tdep.c: Include arch/arm-get-next-pcs.h.
> * armbsd-tdep.c: Likewise.
> * armnbsd-tdep.c: Likewise.
> * armobsd-tdep.c: Likewise.
Why do we need to include arm-get-next-pcs.h in these *-tdep.c files?
> * common/gdb_vecs.h: Add CORE_ADDR vector definition.
> * configure.tgt: Add arm-get-next-pcs.o, arm-linux.o to all
> relevent arm targets.
> * symtab.h: Move CORE_ADDR vector definition to gdb_vecs.h.
>
> gdb/gdbserver/ChangeLog:
> * Makefile.in : Add arm-get-next-pcs.c/o, arm-linux.c/o.
Please describe what are changed.
> * configure.srv: Add arm-get-next-pcs.o, arm-linux.o
> to needed arm targets.
> * linux-arm-low.c: Include arch/arm-linux.h,
> aarch/arm-get-next-pcs.h.
> (get_next_pcs_ops): New struct.
> (get_next_pcs_addr_bits_remove): New function.
> (get_next_pcs_is_thumb): New function.
> (get_next_pcs_read_memory_unsigned_integer): Likewise.
> (arm_sigreturn_next_pc): Likewise.
> (get_next_pcs_syscall_next_pc): Likewise.
> (arm_gdbserver_get_next_pcs): Likewise.
> (struct linux_target_ops) <arm_gdbserver_get_next_pcs>:
> Initialize.
> * linux-low.h: Move CORE_ADDR vector definition to gdb_vecs.h.
> * server.h: Include gdb_vecs.h.
> ---
> gdb/Makefile.in | 16 +-
> gdb/aarch32-linux-nat.c | 1 +
> gdb/arch/arm-get-next-pcs.c | 1216 +++++++++++++++++++++++++++++++++
> gdb/arch/arm-get-next-pcs.h | 95 +++
> gdb/arch/arm-linux.c | 59 ++
> gdb/arch/arm-linux.h | 73 ++
> gdb/arm-linux-nat.c | 2 +
> gdb/arm-linux-tdep.c | 136 ++--
> gdb/arm-tdep.c | 1495 +++++------------------------------------
> gdb/arm-tdep.h | 18 +-
> gdb/arm-wince-tdep.c | 1 +
> gdb/armbsd-tdep.c | 1 +
> gdb/armnbsd-tdep.c | 1 +
> gdb/armobsd-tdep.c | 1 +
> gdb/common/gdb_vecs.h | 2 +
> gdb/configure.tgt | 20 +-
> gdb/gdbserver/Makefile.in | 9 +-
> gdb/gdbserver/configure.srv | 2 +
> gdb/gdbserver/linux-arm-low.c | 144 +++-
> gdb/gdbserver/linux-low.h | 2 -
> gdb/gdbserver/server.h | 1 +
> gdb/symtab.h | 2 -
> 22 files changed, 1871 insertions(+), 1426 deletions(-)
> create mode 100644 gdb/arch/arm-get-next-pcs.c
> create mode 100644 gdb/arch/arm-get-next-pcs.h
> create mode 100644 gdb/arch/arm-linux.c
> create mode 100644 gdb/arch/arm-linux.h
>
> +
> +/* Checks for an atomic sequence of instructions beginning with a LDREX{,B,H,D}
> + instruction and ending with a STREX{,B,H,D} instruction. If such a sequence
> + is found, attempt to step through it. The end of the sequence address is
> + added to the next_pcs list. */
> +
> +static int
> +thumb_deal_with_atomic_sequence_raw (struct arm_get_next_pcs *self,
> + CORE_ADDR pc,
> + VEC (CORE_ADDR) **next_pcs)
We can return VEC (CORE_ADDR) * here, if it is NULL, the sequence is not
found, otherwise, the end of the sequence addresses are returned in the vector.
> +
> +/* Checks for an atomic sequence of instructions beginning with a LDREX{,B,H,D}
> + instruction and ending with a STREX{,B,H,D} instruction. If such a sequence
> + is found, attempt to step through it. The end of the sequence address is
> + added to the next_pcs list. */
> +
> +static int
> +arm_deal_with_atomic_sequence_raw (struct arm_get_next_pcs *self,
> + CORE_ADDR pc,
> + VEC (CORE_ADDR) **next_pcs)
Likewise.
> +
> +/* See arm-get-next-pcs.h. */
> +
> +VEC (CORE_ADDR) *
> +arm_get_next_pcs (struct arm_get_next_pcs *self, CORE_ADDR pc)
> +{
> + VEC (CORE_ADDR) *next_pcs = NULL;
> +
> + if (self->ops->is_thumb (self))
> + {
> + if (!thumb_deal_with_atomic_sequence_raw (self, pc, &next_pcs))
> + return thumb_get_next_pcs_raw (self, pc, &next_pcs);
Why do we need to pass next_pcs to thumb_get_next_pcs_raw? Isn't
simpler to write it like this,
next_pcs = thumb_deal_with_atomic_sequence_raw (self, pc);
if (next_pcs == NULL)
next_pcs = thumb_get_next_pcs_raw (self, pc);
> + }
> + else
> + {
> + if (!arm_deal_with_atomic_sequence_raw (self, pc, &next_pcs))
> + return arm_get_next_pcs_raw (self, pc, &next_pcs);
> + }
> +
> + return next_pcs;
> +}
> diff --git a/gdb/arch/arm-get-next-pcs.h b/gdb/arch/arm-get-next-pcs.h
> new file mode 100644
> index 0000000..39921e0
> --- /dev/null
> +++ b/gdb/arch/arm-get-next-pcs.h
> @@ -0,0 +1,95 @@
> +/* Common code for ARM software single stepping support.
> +
> + Copyright (C) 1988-2015 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/>. */
> +
> +#ifndef ARM_GET_NEXT_PCS_H
> +#define ARM_GET_NEXT_PCS_H 1
> +
> +/* Support routines for instruction parsing. */
> +#define submask(x) ((1L << ((x) + 1)) - 1)
> +#define bits(obj,st,fn) (((obj) >> (st)) & submask ((fn) - (st)))
> +#define bit(obj,st) (((obj) >> (st)) & 1)
> +#define sbits(obj,st,fn) \
> + ((long) (bits(obj,st,fn) | ((long) bit(obj,fn) * ~ submask (fn - st))))
> +#define BranchDest(addr,instr) \
> + ((CORE_ADDR) (((unsigned long) (addr)) + 8 + (sbits (instr, 0, 23) << 2)))
> +
These macros should be moved to arch/arm.h.
--
Yao (齐尧)
next prev parent reply other threads:[~2015-12-07 15:18 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-12-04 18:46 [PATCH v6 0/6] Support software single step and conditional breakpoints " Antoine Tremblay
2015-12-04 18:46 ` [PATCH v6 2/7] Share some ARM target dependent code from GDB with GDBServer Antoine Tremblay
2015-12-04 18:46 ` [PATCH v6 4/7] Share regcache function regcache_raw_read_unsigned Antoine Tremblay
2015-12-07 14:40 ` Yao Qi
2015-12-07 14:49 ` Antoine Tremblay
2015-12-04 18:46 ` [PATCH v6 3/7] Refactor arm_software_single_step to use regcache Antoine Tremblay
2015-12-07 14:32 ` Yao Qi
2015-12-07 14:48 ` Antoine Tremblay
2015-12-07 15:21 ` Yao Qi
2015-12-07 17:37 ` Antoine Tremblay
2015-12-07 14:51 ` Yao Qi
2015-12-07 14:51 ` Antoine Tremblay
2015-12-04 18:46 ` [PATCH v6 1/7] Replace breakpoint_reinsert_addr by get_next_pcs operation in GDBServer Antoine Tremblay
2015-12-04 18:47 ` [PATCH v6 5/7] Support software single step on ARM " Antoine Tremblay
2015-12-07 15:18 ` Yao Qi [this message]
2015-12-07 15:58 ` Antoine Tremblay
2015-12-04 18:47 ` [PATCH v6 6/7] Enable software single stepping for while-stepping actions " Antoine Tremblay
2015-12-04 18:47 ` [PATCH v6 7/7] Enable conditional breakpoints for targets that support software single step " Antoine Tremblay
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=8637ves2xp.fsf@gmail.com \
--to=qiyaoltc@gmail.com \
--cc=antoine.tremblay@ericsson.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