public inbox for binutils@sourceware.org
 help / color / mirror / Atom feed
From: Andrew Burgess <andrew.burgess@embecosm.com>
To: gdb-patches@sourceware.org, binutils@sourceware.org
Cc: Fredrik Hederstierna <fredrik@hederstierna.com>,
	Andrew Burgess <andrew.burgess@embecosm.com>
Subject: [PATCHv2 9/9] gdb/arm: add support for bare-metal core dumps
Date: Wed, 20 Jan 2021 20:23:15 +0000	[thread overview]
Message-ID: <1d767ed944205b58d1df09dd1aa218a6fabcc7f5.1611172468.git.andrew.burgess@embecosm.com> (raw)
In-Reply-To: <cover.1611172468.git.andrew.burgess@embecosm.com>

This commit is just a trimmed down version of Fredrik's patch (link
below), updated to fit in with my new common core dumping code.

As a significant piece of work Fredrik would need a copyright
assignment in place before this could be merged.  I don't know if that
is the case or not.

As with the RISC-V patch (#5) the documentation of the core file
format is given in the commit message, but it really needs a "better"
home.  Unlike RISC-V, I don't know where that home is for ARM.

I would prefer if most feedback on the documentation was given on the
RISC-V patch (#5 in this series), unless you're pointing out an error
that is specific just to this patch.

I have checked that this code builds, but otherwise I've done no
testing on this at all.  I assume Fredrik gave this some testing
before posting, but I have mucked about with things, so I'm sure any
mistakes are mine.

Thanks,
Andrew


---

This commit adds support for bare metal core dumps on the ARM target,
and is based off of this patch submitted to the mailing list:

  https://sourceware.org/pipermail/gdb-patches/2020-October/172845.html

Compared to the version linked above this version is updated to take
account of recent changes to the core dump infrastructure in GDB,
there is now more shared infrastructure for core dumping within GDB,
and also some common bare metal core dumping infrastructure.  As a
result this patch is smaller than the original proposed patch.

Further, the original patch included some unrelated changes to the
simulator that have been removed from this version.

I have written a ChangeLog entry as the original patch was missing
one.

I have done absolutely no testing of this patch.  It is based on the
original submitted patch, which I assume was tested, but after my
modifications things might have been broken.

The cores dump format is based around generating an ELF containing
sections for the writable regions of memory that a user could be
using.  Which regions are dumped rely on GDB's existing common core
dumping code, GDB will attempt to figure out the stack and heap as
well as copying out writable data sections as identified by the
original ELF.

Register information is added to the core dump using notes, just as it
is for Linux of FreeBSD core dumps.  The note types used consist of
the 2 basic types you would expect in a OS based core dump,
NT_PRPSINFO, NT_PRSTATUS, along with the architecture specific
NT_ARM_VFP note.

The data layouts for each note type are described blow, in all case,
all padding fields should be set to zero.

Note NT_PRPSINFO is optional.  Its data layout is:

  struct prpsinfo_t
  {
    uint8_t padding[28];
    char fname[16];
    char psargs[80];
  }

Field 'fname' - null terminated string consisting of the basename of
    (up to the fist 15 characters of) the executable.  Any additional
    space should be set to zero.  If there's no executable name then
    this field can be set to all zero.

Field 'psargs' - a null terminated string up to 80 characters in
    length.  Any additional space should be filled with zero.  This
    field contains the full executable path and any arguments passed
    to the executable.  If there's nothing sensible to write in this
    field then fill it with zero.

Note NT_PRSTATUS is required, its data layout is:

  struct prstatus_t
  {
    uint8_t padding_1[12];
    uint16_t sig;
    uint8_t padding_2[10];
    uint32_t thread_id;
    uint8_t padding_3[44];
    uint32_t gregs[18];
  }

Field 'sig' - the signal that stopped this thread.  Its implementation
    defined what this field actually means.  Within GDB this will be
    the signal number that the remote target reports as the stop
    reason for this thread.

Field 'thread_is' - the thread id for this thread, its implementation
    defined what this field actually means.  Within GDB this will be
    thread thread-id that is assigned to each remote thread.

Field 'gregs' - holds the general purpose registers $a1 through to $pc
    at indices 0 to 15.  At index 16 the program status register.
    Index 17 should be set to zero.

Note NT_ARM_VFP is optional, its data layout is:

  armvfp_t
  {
    uint64_t regs[32];
    uint32_t fpscr;
  }

Field 'regs' - holds the 32 d-registers 0 to 31 in order.

Field 'fpscr' - holds the fpscr register.

The rules for ordering the notes is the same as for Linux.  The
NT_PRSTATUS note must come before any other notes about additional
register sets.  And for multi-threaded targets all registers for a
single thread should be grouped together.  This is because only
NT_PRSTATUS includes a thread-id, all additional register notes after
a NT_PRSTATUS are assumed to belong to the same thread until a
different NT_PRSTATUS is seen.

gdb/ChangeLog:

	PR gdb/14383
	* Makefile.in (ALL_TARGET_OBS): Add arm-none-tdep.o.
	(ALLDEPFILES): Add arm-none-tdep.c
	* arm-none-tdep.c: New file.
	* configure.tgt (arm*-*-*): Add arm-none-tdep.o to cpu_obs.
---
 gdb/ChangeLog       |   9 ++
 gdb/Makefile.in     |   2 +
 gdb/arm-none-tdep.c | 208 ++++++++++++++++++++++++++++++++++++++++++++
 gdb/configure.tgt   |   3 +-
 4 files changed, 221 insertions(+), 1 deletion(-)
 create mode 100644 gdb/arm-none-tdep.c

diff --git a/gdb/Makefile.in b/gdb/Makefile.in
index 5f88b6a78cf..c40dd7d4688 100644
--- a/gdb/Makefile.in
+++ b/gdb/Makefile.in
@@ -726,6 +726,7 @@ ALL_TARGET_OBS = \
 	arm-fbsd-tdep.o \
 	arm-linux-tdep.o \
 	arm-netbsd-tdep.o \
+	arm-none-tdep.o \
 	arm-obsd-tdep.o \
 	arm-pikeos-tdep.o \
 	arm-symbian-tdep.o \
@@ -2167,6 +2168,7 @@ ALLDEPFILES = \
 	arm-linux-tdep.c \
 	arm-netbsd-nat.c \
 	arm-netbsd-tdep.c \
+	arm-none-tdep.c \
 	arm-obsd-tdep.c \
 	arm-symbian-tdep.c \
 	arm-tdep.c \
diff --git a/gdb/arm-none-tdep.c b/gdb/arm-none-tdep.c
new file mode 100644
index 00000000000..199ac652129
--- /dev/null
+++ b/gdb/arm-none-tdep.c
@@ -0,0 +1,208 @@
+/* none on ARM target support.
+
+   Copyright (C) 2020 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/>.  */
+
+#include "defs.h"
+#include "arm-tdep.h"
+#include "arch-utils.h"
+#include "regcache.h"
+#include "elf-bfd.h"
+#include "regset.h"
+#include "none-tdep.h"
+#include "user-regs.h"
+
+/* Core file and register set support.  */
+#define ARM_NONE_SIZEOF_GREGSET (18 * ARM_INT_REGISTER_SIZE)
+
+/* Support VFP register format.  */
+#define ARM_NONE_SIZEOF_VFP (32 * 8 + 4)
+
+/* The index to access CSPR in user_regs as defined in GLIBC.  */
+#define ARM_NONE_CPSR_GREGNUM 16
+
+/* Supply register REGNUM from buffer GREGS_BUF (length LEN bytes) into
+   REGCACHE.  If REGNUM is -1 then supply all registers.  The set of
+   registers that this function will supply is limited to the general
+   purpose registers.
+
+   The layout of the registers here is based on the ARM GNU/Linux
+   layout.  */
+
+static void
+arm_none_supply_gregset (const struct regset *regset,
+			 struct regcache *regcache,
+			 int regnum, const void *gregs_buf, size_t len)
+{
+  struct gdbarch *gdbarch = regcache->arch ();
+  enum bfd_endian byte_order = gdbarch_byte_order (gdbarch);
+  const gdb_byte *gregs = (const gdb_byte *) gregs_buf;
+
+  for (int regno = ARM_A1_REGNUM; regno < ARM_PC_REGNUM; regno++)
+    if (regnum == -1 || regnum == regno)
+      regcache->raw_supply (regno, gregs + ARM_INT_REGISTER_SIZE * regno);
+
+  if (regnum == ARM_PS_REGNUM || regnum == -1)
+    {
+      if (arm_apcs_32)
+	regcache->raw_supply (ARM_PS_REGNUM,
+			      gregs + ARM_INT_REGISTER_SIZE
+			      * ARM_NONE_CPSR_GREGNUM);
+      else
+	regcache->raw_supply (ARM_PS_REGNUM,
+			     gregs + ARM_INT_REGISTER_SIZE * ARM_PC_REGNUM);
+    }
+
+  if (regnum == ARM_PC_REGNUM || regnum == -1)
+    {
+      gdb_byte pc_buf[ARM_INT_REGISTER_SIZE];
+
+      CORE_ADDR reg_pc
+	= extract_unsigned_integer (gregs + ARM_INT_REGISTER_SIZE
+				    * ARM_PC_REGNUM,
+				    ARM_INT_REGISTER_SIZE, byte_order);
+      reg_pc = gdbarch_addr_bits_remove (gdbarch, reg_pc);
+      store_unsigned_integer (pc_buf, ARM_INT_REGISTER_SIZE, byte_order,
+			      reg_pc);
+      regcache->raw_supply (ARM_PC_REGNUM, pc_buf);
+    }
+}
+
+/* Collect register REGNUM from REGCACHE and place it into buffer GREGS_BUF
+   (length LEN bytes).  If REGNUM is -1 then collect all registers.  The
+   set of registers that this function will collect is limited to the
+   general purpose registers.
+
+   The layout of the registers here is based on the ARM GNU/Linux
+   layout.  */
+
+static void
+arm_none_collect_gregset (const struct regset *regset,
+			  const struct regcache *regcache,
+			  int regnum, void *gregs_buf, size_t len)
+{
+  gdb_byte *gregs = (gdb_byte *) gregs_buf;
+
+  for (int regno = ARM_A1_REGNUM; regno < ARM_PC_REGNUM; regno++)
+    if (regnum == -1 || regnum == regno)
+      regcache->raw_collect (regno,
+			     gregs + ARM_INT_REGISTER_SIZE * regno);
+
+  if (regnum == ARM_PS_REGNUM || regnum == -1)
+    {
+      if (arm_apcs_32)
+	regcache->raw_collect (ARM_PS_REGNUM,
+			       gregs + ARM_INT_REGISTER_SIZE
+			       * ARM_NONE_CPSR_GREGNUM);
+      else
+	regcache->raw_collect (ARM_PS_REGNUM,
+			       gregs + ARM_INT_REGISTER_SIZE * ARM_PC_REGNUM);
+    }
+
+  if (regnum == ARM_PC_REGNUM || regnum == -1)
+    regcache->raw_collect (ARM_PC_REGNUM,
+			   gregs + ARM_INT_REGISTER_SIZE * ARM_PC_REGNUM);
+}
+
+/* Supply VFP registers from REGS_BUF into REGCACHE.  */
+
+static void
+arm_none_supply_vfp (const struct regset *regset,
+		     struct regcache *regcache,
+		     int regnum, const void *regs_buf, size_t len)
+{
+  const gdb_byte *regs = (const gdb_byte *) regs_buf;
+
+  if (regnum == ARM_FPSCR_REGNUM || regnum == -1)
+    regcache->raw_supply (ARM_FPSCR_REGNUM, regs + 32 * 8);
+
+  for (int regno = ARM_D0_REGNUM; regno <= ARM_D31_REGNUM; regno++)
+    if (regnum == -1 || regnum == regno)
+      regcache->raw_supply (regno, regs + (regno - ARM_D0_REGNUM) * 8);
+}
+
+/* Collect VFP registers from REGCACHE into REGS_BUF.  */
+
+static void
+arm_none_collect_vfp (const struct regset *regset,
+		      const struct regcache *regcache,
+		      int regnum, void *regs_buf, size_t len)
+{
+  gdb_byte *regs = (gdb_byte *) regs_buf;
+
+  if (regnum == ARM_FPSCR_REGNUM || regnum == -1)
+    regcache->raw_collect (ARM_FPSCR_REGNUM, regs + 32 * 8);
+
+  for (int regno = ARM_D0_REGNUM; regno <= ARM_D31_REGNUM; regno++)
+    if (regnum == -1 || regnum == regno)
+      regcache->raw_collect (regno, regs + (regno - ARM_D0_REGNUM) * 8);
+}
+
+/* The general purpose register set.  */
+
+static const struct regset arm_none_gregset =
+  {
+    nullptr, arm_none_supply_gregset, arm_none_collect_gregset
+  };
+
+/* The VFP register set.  */
+
+static const struct regset arm_none_vfpregset =
+  {
+    nullptr, arm_none_supply_vfp, arm_none_collect_vfp
+  };
+
+/* Iterate over core file register note sections.  */
+
+static void
+arm_none_iterate_over_regset_sections (struct gdbarch *gdbarch,
+				       iterate_over_regset_sections_cb *cb,
+				       void *cb_data,
+				       const struct regcache *regcache)
+{
+  struct gdbarch_tdep *tdep = gdbarch_tdep (gdbarch);
+
+  cb (".reg", ARM_NONE_SIZEOF_GREGSET, ARM_NONE_SIZEOF_GREGSET,
+      &arm_none_gregset, nullptr, cb_data);
+
+  if (tdep->vfp_register_count > 0)
+    cb (".reg-arm-vfp", ARM_NONE_SIZEOF_VFP, ARM_NONE_SIZEOF_VFP,
+	&arm_none_vfpregset, "VFP floating-point", cb_data);
+}
+
+/* Initialize ARM bare-metal ABI info.  */
+
+static void
+arm_none_init_abi (struct gdbarch_info info, struct gdbarch *gdbarch)
+{
+  none_init_abi (gdbarch);
+
+  /* Iterate over registers for reading and writing bare metal ARM core
+     files.  */
+  set_gdbarch_iterate_over_regset_sections
+    (gdbarch, arm_none_iterate_over_regset_sections);
+}
+
+/* Initialize ARM bare-metal target support.  */
+
+void _initialize_arm_none_tdep ();
+void
+_initialize_arm_none_tdep ()
+{
+  gdbarch_register_osabi (bfd_arch_arm, 0, GDB_OSABI_NONE,
+			  arm_none_init_abi);
+}
diff --git a/gdb/configure.tgt b/gdb/configure.tgt
index ad88ddd9302..4f3b9df90d7 100644
--- a/gdb/configure.tgt
+++ b/gdb/configure.tgt
@@ -65,7 +65,8 @@ arc*-*-*)
 
 arm*-*-*)
 	cpu_obs="aarch32-tdep.o arch/aarch32.o arch/arm.o \
-		 arch/arm-get-next-pcs.o arm-tdep.o";;
+		 arch/arm-get-next-pcs.o arm-tdep.o arm-none-tdep.o"
+	;;
 
 hppa*-*-*)
 	# Target: HP PA-RISC
-- 
2.25.4


  parent reply	other threads:[~2021-01-20 20:23 UTC|newest]

Thread overview: 55+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-01-20 20:23 [PATCHv2 0/9] Bare-metal core dumps for RISC-V Andrew Burgess
2021-01-20 20:23 ` [PATCHv2 1/9] gdb: unify parts of the Linux and FreeBSD core dumping code Andrew Burgess
2021-01-22 12:01   ` Strasuns, Mihails
2021-01-22 18:50   ` Tom Tromey
2021-02-01 11:56   ` Andrew Burgess
2021-02-09 21:52     ` Andrew Burgess
2021-01-20 20:23 ` [PATCHv2 2/9] bfd/binutils: support for gdb target descriptions in the core file Andrew Burgess
2021-01-22 10:47   ` Strasuns, Mihails
2021-01-22 19:30     ` Andrew Burgess
2021-01-25 10:11       ` Strasuns, Mihails
2021-01-25 11:20         ` Andrew Burgess
2021-02-01 12:05   ` PING: " Andrew Burgess
2021-02-01 15:10     ` Strasuns, Mihails
2021-02-01 13:29   ` Luis Machado
2021-02-10 20:45   ` Jim Wilson
2021-01-20 20:23 ` [PATCHv2 3/9] gdb: write target description into " Andrew Burgess
2021-01-22 19:15   ` Tom Tromey
2021-02-01 13:37   ` Luis Machado
2021-01-20 20:23 ` [PATCHv2 4/9] bfd/riscv: prepare to handle bare metal core dump creation Andrew Burgess
2021-02-01 12:03   ` PING: " Andrew Burgess
2021-02-01 13:48   ` Luis Machado
2021-02-01 14:44     ` Andrew Burgess
2021-02-10 20:57   ` Jim Wilson
2021-01-20 20:23 ` [PATCHv2 5/9] gdb/riscv: introduce bare metal core dump support Andrew Burgess
2021-02-01 14:05   ` Luis Machado
2021-02-03  3:04     ` Palmer Dabbelt
2021-01-20 20:23 ` [PATCHv2 6/9] bfd/binutils: add support for RISC-V CSRs in core files Andrew Burgess
2021-02-01 12:00   ` Andrew Burgess
2021-02-01 14:08     ` Luis Machado
2021-02-10 21:00     ` Jim Wilson
2021-01-20 20:23 ` [PATCHv2 7/9] gdb/riscv: make riscv target description names global Andrew Burgess
2021-02-01 14:22   ` Luis Machado
2021-01-20 20:23 ` [PATCHv2 8/9] gdb/riscv: write CSRs into baremetal core dumps Andrew Burgess
2021-02-01 14:33   ` Luis Machado
2021-01-20 20:23 ` Andrew Burgess [this message]
2021-02-01 14:51   ` [PATCHv2 9/9] gdb/arm: add support for bare-metal " Luis Machado
2021-01-22 19:28 ` [PATCHv2 0/9] Bare-metal core dumps for RISC-V Tom Tromey
2021-02-07 21:48 ` PING: " Andrew Burgess
2021-02-10 21:01   ` Jim Wilson
2021-02-15 11:14     ` Andrew Burgess
2021-02-15 17:29 ` [PATCHv3 " Andrew Burgess
2021-02-15 17:29   ` [PATCHv3 1/9] gdb: unify parts of the Linux and FreeBSD core dumping code Andrew Burgess
2021-02-15 22:56     ` Lancelot SIX
2021-02-16 16:55       ` Andrew Burgess
2021-02-15 17:29   ` [PATCHv3 2/9] bfd/binutils: support for gdb target descriptions in the core file Andrew Burgess
2021-02-15 17:29   ` [PATCHv3 3/9] gdb: write target description into " Andrew Burgess
2021-02-15 17:29   ` [PATCHv3 4/9] bfd/riscv: prepare to handle bare metal core dump creation Andrew Burgess
2021-02-15 17:29   ` [PATCHv3 5/9] gdb/riscv: introduce bare metal core dump support Andrew Burgess
2021-02-15 17:29   ` [PATCHv3 6/9] bfd/binutils: add support for RISC-V CSRs in core files Andrew Burgess
2021-02-15 17:29   ` [PATCHv3 7/9] gdb/riscv: make riscv target description names global Andrew Burgess
2021-02-15 17:29   ` [PATCHv3 8/9] gdb/riscv: write CSRs into baremetal core dumps Andrew Burgess
2021-02-15 17:29   ` [PATCHv3 9/9] gdb/arm: add support for bare-metal " Andrew Burgess
2021-03-01 10:32   ` [PATCHv3 0/9] Bare-metal core dumps for RISC-V Andrew Burgess
2021-03-01 14:45     ` Nick Clifton
2021-03-05 17:35     ` Andrew Burgess

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=1d767ed944205b58d1df09dd1aa218a6fabcc7f5.1611172468.git.andrew.burgess@embecosm.com \
    --to=andrew.burgess@embecosm.com \
    --cc=binutils@sourceware.org \
    --cc=fredrik@hederstierna.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).