public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: John Baldwin <jhb@FreeBSD.org>
To: gdb-patches@sourceware.org
Subject: [PATCH v2 00/12] FreeBSD/aarch64 hardware watchpoint support
Date: Wed, 16 Mar 2022 13:19:11 -0700	[thread overview]
Message-ID: <20220316201923.89694-1-jhb@FreeBSD.org> (raw)

Changes since V1:

- The unordered_map<>'s in x86-nat.c and aarch64-nat.c both now store
  objects directly rather than pointers to objects.

- Trimmed "Contributed by" notices from new files.

- I have compiled and (very lightly) tested this on Linux Aarch64.
  By light testing I mean that I ran a test program with a harware
  breakpoint set on main and it stopped correctly.  I haven't run a
  full test suite as my Aarch64 test box is a lowly Raspberry Pi
  for which such a run would take a fairly long time.

I still have some open questions about Patch 6 from the first
version:

Patch 6 has an open question about how best to handle having a
platform-specific hook for when debug registers have been changed.
Right now we require the platform to supply the function that
nat/aarch64-hw-point.c calls.  I did not choose to create an
equivalent to x86_dr_low, but perhaps that sort of structure, or at
least a function pointer should be used instead?

There is also some messiness around the Linux-specific
kernel_supports_any_contiguous_range workaround in patch 6.

OTOH, some of the FreeBSD/x86 cleanups in the first half of the series
(such as adding x86-fbsd-nat.*) might be nice to reuse in my XSAVE
series, so if that half of the series is ok (first 5 patches), it
might be nice to push that in sooner.

John Baldwin (12):
  Remove USE_SIGTRAP_SIGINFO condition for FreeBSD/x86 debug regs
    support.
  x86-nat: Use an unordered_map to store per-pid debug reg state.
  x86-nat: Add x86_lookup_debug_reg_state.
  Add an x86_fbsd_nat_target mixin class for FreeBSD x86 native targets.
  fbsd-nat: Add a low_new_fork virtual method.
  x86-fbsd-nat: Copy debug register state on fork.
  nat: Split out platform-independent aarch64 debug register support.
  aarch64: Add an aarch64_nat_target mixin class.
  fbsd-nat: Add helper routine to fetch siginfo_t for a ptid.
  fbsd-nat: Add a low_delete_thread virtual method.
  fbsd-nat: Add a low_prepare_to_resume virtual method.
  Add support for hardware breakpoints/watchpoints on FreeBSD/Aarch64.

 gdb/NEWS                         |   2 +
 gdb/aarch64-fbsd-nat.c           | 260 ++++++++++++-
 gdb/aarch64-linux-nat.c          | 352 +----------------
 gdb/aarch64-nat.c                | 302 +++++++++++++++
 gdb/aarch64-nat.h                | 109 ++++++
 gdb/amd64-fbsd-nat.c             |  20 +-
 gdb/configure.nat                |  12 +-
 gdb/fbsd-nat.c                   |  28 +-
 gdb/fbsd-nat.h                   |  18 +
 gdb/i386-fbsd-nat.c              |  20 +-
 gdb/nat/aarch64-hw-point.c       | 624 +++++++++++++++++++++++++++++++
 gdb/nat/aarch64-hw-point.h       | 126 +++++++
 gdb/nat/aarch64-linux-hw-point.c | 605 +-----------------------------
 gdb/nat/aarch64-linux-hw-point.h | 105 +-----
 gdb/nat/aarch64-linux.c          |   4 +-
 gdb/x86-fbsd-nat.c               |  45 +++
 gdb/x86-fbsd-nat.h               |  36 ++
 gdb/x86-nat.c                    |  92 +----
 gdb/x86-nat.h                    |   5 +
 gdbserver/configure.srv          |   1 +
 gdbserver/linux-aarch64-low.cc   |  13 +-
 21 files changed, 1612 insertions(+), 1167 deletions(-)
 create mode 100644 gdb/aarch64-nat.c
 create mode 100644 gdb/aarch64-nat.h
 create mode 100644 gdb/nat/aarch64-hw-point.c
 create mode 100644 gdb/nat/aarch64-hw-point.h
 create mode 100644 gdb/x86-fbsd-nat.c
 create mode 100644 gdb/x86-fbsd-nat.h

-- 
2.34.1


             reply	other threads:[~2022-03-16 20:19 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-03-16 20:19 John Baldwin [this message]
2022-03-16 20:19 ` [PATCH v2 01/12] Remove USE_SIGTRAP_SIGINFO condition for FreeBSD/x86 debug regs support John Baldwin
2022-03-16 20:19 ` [PATCH v2 02/12] x86-nat: Use an unordered_map to store per-pid debug reg state John Baldwin
2022-03-21 18:07   ` Pedro Alves
2022-03-16 20:19 ` [PATCH v2 03/12] x86-nat: Add x86_lookup_debug_reg_state John Baldwin
2022-03-21 18:10   ` Pedro Alves
2022-03-16 20:19 ` [PATCH v2 04/12] Add an x86_fbsd_nat_target mixin class for FreeBSD x86 native targets John Baldwin
2022-03-16 20:19 ` [PATCH v2 05/12] fbsd-nat: Add a low_new_fork virtual method John Baldwin
2022-03-16 20:19 ` [PATCH v2 06/12] x86-fbsd-nat: Copy debug register state on fork John Baldwin
2022-03-16 20:19 ` [PATCH v2 07/12] nat: Split out platform-independent aarch64 debug register support John Baldwin
2022-03-17 15:37   ` Luis Machado
2022-03-16 20:19 ` [PATCH v2 08/12] aarch64: Add an aarch64_nat_target mixin class John Baldwin
2022-03-17 15:35   ` Luis Machado
2022-03-16 20:19 ` [PATCH v2 09/12] fbsd-nat: Add helper routine to fetch siginfo_t for a ptid John Baldwin
2022-03-16 20:19 ` [PATCH v2 10/12] fbsd-nat: Add a low_delete_thread virtual method John Baldwin
2022-03-16 20:19 ` [PATCH v2 11/12] fbsd-nat: Add a low_prepare_to_resume " John Baldwin
2022-03-16 20:19 ` [PATCH v2 12/12] Add support for hardware breakpoints/watchpoints on FreeBSD/Aarch64 John Baldwin
2022-03-30 15:23 ` [PATCH v2 00/12] FreeBSD/aarch64 hardware watchpoint support Luis Machado
2022-03-30 15:31   ` Luis Machado

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=20220316201923.89694-1-jhb@FreeBSD.org \
    --to=jhb@freebsd.org \
    --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).