[PATCHv2 9/9] gdb/arm: add support for bare-metal core dumps
Luis Machado
luis.machado@linaro.org
Mon Feb 1 14:51:13 GMT 2021
Andrew,
Thanks a lot for going out of your way to come up with this patch and
documentation changes. It is great that this can be reused for other
architectures and it definitely saves developers some time. Much
appreciated.
With the changes below, I think it will be pretty simple to support ARM
bare metal cores files (it might even work as-is).
There are some nits in the patch, most of which I pointed out in the
RISCV bare metal core file documentation patch. But other than those, I
think this looks good.
Fredrik, would you mind taking a look at Andrew's changes for ARM bare
metal core file support and rebase your changes on top of his RISCV
series? I think that's the way forward.
You will also need a copyright assignment before your changes can be
accepted upstream, as this work is not considered a trivial change.
Finally, it would be nice to have some testing to make sure this is
working as expected. Do you have a test setup that you can use to
validate things?
Patch nits below...
On 1/20/21 5:23 PM, Andrew Burgess wrote:
> 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
cores -> core
> 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,
blow -> below
case -> cases?
> 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
Its -> It's
> 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
its -> it's
> 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.
2020~2021
> +
> + 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. */
CSPR -> CPSR
> +#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
>
More information about the Binutils
mailing list