[RFA/commit] (Ada) Add ravenscar tasking support on AArch64

[RFA/commit] (Ada) Add ravenscar tasking support on AArch64

[Joel Brobecker]

Hello,

This patch adds support for debugging Ravenscar tasks, similar to what
is done for ppc and sparc.

gdb/ChangeLog:

        * aarch64-ravenscar-thread.h, aarch64-ravenscar-thread.c:
        New files.
        * aarch64-tdep.c: #include "aarch64-ravenscar-thread.h".
        (aarch64_gdbarch_init): Add call to register_aarch64_ravenscar_ops.
        * Makefile.in (ALL_64_TARGET_OBS): Add aarch64-ravenscar-thread.o.
        (HFILES_NO_SRCDIR): Add aarch64-ravenscar-thread.h.
        (ALLDEPFILES): Add aarch64-ravenscar-thread.c.
        * configure.tgt (cpu_obs) [aarch64*-*-*]: Add ravenscar-thread.o
        and aarch64-ravenscar-thread.o.

Tested on aarch64-elf using AdaCore's testsuite.
OK to apply?

Thank you!
-- 
Joel

---
 gdb/Makefile.in                |   3 +
 gdb/aarch64-ravenscar-thread.c | 213 +++++++++++++++++++++++++++++++++++++++++
 gdb/aarch64-ravenscar-thread.h |  27 ++++++
 gdb/aarch64-tdep.c             |   3 +
 gdb/configure.tgt              |   3 +-
 5 files changed, 248 insertions(+), 1 deletion(-)
 create mode 100644 gdb/aarch64-ravenscar-thread.c
 create mode 100644 gdb/aarch64-ravenscar-thread.h

diff --git a/gdb/Makefile.in b/gdb/Makefile.in
index c76a4e4..c1ad8e4 100644
--- a/gdb/Makefile.in
+++ b/gdb/Makefile.in
@@ -627,6 +627,7 @@ ALL_64_TARGET_OBS = \
 	aarch64-fbsd-tdep.o \
 	aarch64-linux-tdep.o \
 	aarch64-newlib-tdep.o \
+	aarch64-ravenscar-thread.o \
 	aarch64-tdep.o \
 	alpha-bsd-tdep.o \
 	alpha-linux-tdep.o \
@@ -1165,6 +1166,7 @@ SFILES = \
 # right, it is probably easiest just to list .h files here directly.
 
 HFILES_NO_SRCDIR = \
+	aarch64-ravenscar-thread.h \
 	aarch64-tdep.h \
 	ada-lang.h \
 	addrmap.h \
@@ -2169,6 +2171,7 @@ ALLDEPFILES = \
 	aarch64-linux-nat.c \
 	aarch64-linux-tdep.c \
 	aarch64-newlib-tdep.c \
+	aarch64-ravenscar-thread.c \
 	aarch64-tdep.c \
 	aix-thread.c \
 	alpha-bsd-nat.c \
diff --git a/gdb/aarch64-ravenscar-thread.c b/gdb/aarch64-ravenscar-thread.c
new file mode 100644
index 0000000..88629e4
--- /dev/null
+++ b/gdb/aarch64-ravenscar-thread.c
@@ -0,0 +1,213 @@
+/* Ravenscar Aarch64 target support.
+
+   Copyright (C) 2017-2018 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 "gdbcore.h"
+#include "regcache.h"
+#include "aarch64-tdep.h"
+#include "inferior.h"
+#include "ravenscar-thread.h"
+#include "aarch64-ravenscar-thread.h"
+
+#define NO_OFFSET -1
+
+/* See aarch64-tdep.h for register numbers.  */
+
+static const int aarch64_context_offsets[] =
+{
+  /* X0 - X28 */
+  NO_OFFSET, NO_OFFSET, NO_OFFSET, NO_OFFSET,
+  NO_OFFSET, NO_OFFSET, NO_OFFSET, NO_OFFSET,
+  NO_OFFSET, NO_OFFSET, NO_OFFSET, NO_OFFSET,
+  NO_OFFSET, NO_OFFSET, NO_OFFSET, NO_OFFSET,
+  NO_OFFSET, NO_OFFSET, NO_OFFSET, 0,
+  8,         16,        24,        32,
+  40,        48,        56,        64,
+  72,
+
+  /* FP, LR, SP, PC, CPSR */
+  /* Note that as task switch is synchronous, PC is in fact the LR here */
+  80,        88,        96,        88,
+  NO_OFFSET,
+
+  /* Q0 - Q31 */
+  112,       128,       144,       160,
+  176,       192,       208,       224,
+  240,       256,       272,       288,
+  304,       320,       336,       352,
+  368,       384,       400,       416,
+  432,       448,       464,       480,
+  496,       512,       528,       544,
+  560,       576,       592,       608,
+
+  /* FPSR, FPCR */
+  104,       108,
+
+  /* FPU Saved field */
+  624
+};
+
+/* The register layout info.  */
+
+struct ravenscar_reg_info
+{
+  /* A table providing the offset relative to the context structure
+     where each register is saved.  */
+  const int *context_offsets;
+
+  /* The number of elements in the context_offsets table above.  */
+  int context_offsets_size;
+};
+
+/* supply register REGNUM, which has been saved on REGISTER_ADDR, to the
+   regcache.  */
+
+static void
+supply_register_at_address (struct regcache *regcache, int regnum,
+                            CORE_ADDR register_addr)
+{
+  struct gdbarch *gdbarch = regcache->arch ();
+  int buf_size = register_size (gdbarch, regnum);
+  gdb_byte *buf;
+
+  buf = (gdb_byte *) alloca (buf_size);
+  read_memory (register_addr, buf, buf_size);
+  regcache->raw_supply (regnum, buf);
+}
+
+/* Return true if, for a non-running thread, REGNUM has been saved on the
+   Thread_Descriptor.  */
+
+static int
+register_in_thread_descriptor_p (const struct ravenscar_reg_info *reg_info,
+				 int regnum)
+{
+  /* Check FPU registers */
+  return (regnum < reg_info->context_offsets_size
+	  && reg_info->context_offsets[regnum] != NO_OFFSET);
+}
+
+/* to_fetch_registers when inferior_ptid is different from the running
+   thread.  */
+
+static void
+aarch64_ravenscar_generic_fetch_registers
+  (const struct ravenscar_reg_info *reg_info,
+   struct regcache *regcache, int regnum)
+{
+  struct gdbarch *gdbarch = regcache->arch ();
+  const int num_regs = gdbarch_num_regs (gdbarch);
+  int current_regnum;
+  CORE_ADDR current_address;
+  CORE_ADDR thread_descriptor_address;
+
+  /* The tid is the thread_id field, which is a pointer to the thread.  */
+  thread_descriptor_address = (CORE_ADDR) inferior_ptid.tid ();
+
+  /* Read registers.  */
+  for (current_regnum = 0; current_regnum < num_regs; current_regnum++)
+    {
+      if (register_in_thread_descriptor_p (reg_info, current_regnum))
+        {
+          current_address = thread_descriptor_address
+            + reg_info->context_offsets[current_regnum];
+          supply_register_at_address (regcache, current_regnum,
+                                      current_address);
+        }
+    }
+}
+
+/* to_prepare_to_store when inferior_ptid is different from the running
+   thread.  */
+
+static void
+aarch64_ravenscar_generic_prepare_to_store (struct regcache *regcache)
+{
+  /* Nothing to do.  */
+}
+
+/* to_store_registers when inferior_ptid is different from the running
+   thread.  */
+
+static void
+aarch64_ravenscar_generic_store_registers
+  (const struct ravenscar_reg_info *reg_info,
+   struct regcache *regcache, int regnum)
+{
+  struct gdbarch *gdbarch = regcache->arch ();
+  int buf_size = register_size (gdbarch, regnum);
+  gdb_byte buf[buf_size];
+  ULONGEST register_address;
+
+  if (register_in_thread_descriptor_p (reg_info, regnum))
+    register_address
+      = inferior_ptid.tid () + reg_info->context_offsets [regnum];
+  else
+    return;
+
+  regcache->raw_collect (regnum, buf);
+  write_memory (register_address,
+                buf,
+                buf_size);
+}
+
+/* The ravenscar_reg_info for most Aarch64 targets.  */
+
+static const struct ravenscar_reg_info aarch64_reg_info =
+{
+  aarch64_context_offsets,
+  ARRAY_SIZE (aarch64_context_offsets),
+};
+
+/* Implement the to_fetch_registers ravenscar_arch_ops method
+   for most Aarch64 targets.  */
+
+static void
+aarch64_ravenscar_fetch_registers (struct regcache *regcache, int regnum)
+{
+  aarch64_ravenscar_generic_fetch_registers
+    (&aarch64_reg_info, regcache, regnum);
+}
+
+/* Implement the to_store_registers ravenscar_arch_ops method
+   for most Aarch64 targets.  */
+
+static void
+aarch64_ravenscar_store_registers (struct regcache *regcache, int regnum)
+{
+  aarch64_ravenscar_generic_store_registers
+    (&aarch64_reg_info, regcache, regnum);
+}
+
+/* The ravenscar_arch_ops vector for most Aarch64 targets.  */
+
+static struct ravenscar_arch_ops aarch64_ravenscar_ops =
+{
+  aarch64_ravenscar_fetch_registers,
+  aarch64_ravenscar_store_registers,
+  aarch64_ravenscar_generic_prepare_to_store
+};
+
+/* Register aarch64_ravenscar_ops in GDBARCH.  */
+
+void
+register_aarch64_ravenscar_ops (struct gdbarch *gdbarch)
+{
+  set_gdbarch_ravenscar_ops (gdbarch, &aarch64_ravenscar_ops);
+}
diff --git a/gdb/aarch64-ravenscar-thread.h b/gdb/aarch64-ravenscar-thread.h
new file mode 100644
index 0000000..e900233
--- /dev/null
+++ b/gdb/aarch64-ravenscar-thread.h
@@ -0,0 +1,27 @@
+/* Ravenscar Aarch64 target support.
+ 
+   Copyright (C) 2017-2018 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 AARCH64_RAVENSCAR_THREAD_H
+#define AARCH64_RAVENSCAR_THREAD_H
+
+struct gdbarch;
+
+extern void register_aarch64_ravenscar_ops (struct gdbarch *gdbarch);
+
+#endif
diff --git a/gdb/aarch64-tdep.c b/gdb/aarch64-tdep.c
index d2e6ac6..02d1f0f 100644
--- a/gdb/aarch64-tdep.c
+++ b/gdb/aarch64-tdep.c
@@ -46,6 +46,7 @@
 #include "selftest.h"
 
 #include "aarch64-tdep.h"
+#include "aarch64-ravenscar-thread.h"
 
 #include "elf-bfd.h"
 #include "elf/aarch64.h"
@@ -3175,6 +3176,8 @@ aarch64_gdbarch_init (struct gdbarch_info info, struct gdbarch_list *arches)
 		  value_of_aarch64_user_reg,
 		  &aarch64_register_aliases[i].regnum);
 
+  register_aarch64_ravenscar_ops (gdbarch);
+
   return gdbarch;
 }
 
diff --git a/gdb/configure.tgt b/gdb/configure.tgt
index 6d1a4df..ebe7dff 100644
--- a/gdb/configure.tgt
+++ b/gdb/configure.tgt
@@ -48,7 +48,8 @@ amd64_tobjs="amd64-tdep.o arch/amd64.o ${x86_tobjs}"
 
 case "${targ}" in
 aarch64*-*-*)
-	cpu_obs="aarch64-tdep.o arch/aarch64-insn.o arch/aarch64.o";;
+	cpu_obs="aarch64-tdep.o arch/aarch64-insn.o arch/aarch64.o \
+		 ravenscar-thread.o aarch64-ravenscar-thread.o";;
 
 alpha*-*-*)
 	# Target: Alpha
-- 
2.1.4

Re: [RFA/commit] (Ada) Add ravenscar tasking support on AArch64

[Pedro Alves]

On 09/08/2018 10:58 PM, Joel Brobecker wrote:
> Hello,
> 
> This patch adds support for debugging Ravenscar tasks, similar to what
> is done for ppc and sparc.
> 
> gdb/ChangeLog:
> 
>         * aarch64-ravenscar-thread.h, aarch64-ravenscar-thread.c:
>         New files.
>         * aarch64-tdep.c: #include "aarch64-ravenscar-thread.h".
>         (aarch64_gdbarch_init): Add call to register_aarch64_ravenscar_ops.
>         * Makefile.in (ALL_64_TARGET_OBS): Add aarch64-ravenscar-thread.o.
>         (HFILES_NO_SRCDIR): Add aarch64-ravenscar-thread.h.
>         (ALLDEPFILES): Add aarch64-ravenscar-thread.c.
>         * configure.tgt (cpu_obs) [aarch64*-*-*]: Add ravenscar-thread.o
>         and aarch64-ravenscar-thread.o.
> 
> Tested on aarch64-elf using AdaCore's testsuite.
> OK to apply?

IMO this deserves a NEWS entry.

Thanks,
Pedro Alves

[RFA/NEWS] document support for Ravenscar task switching on aarch64-elf (was: "Re: [RFA/commit] (Ada) Add ravenscar tasking support on AArch64")

[Joel Brobecker]

[-- Attachment #1: Type: text/plain, Size: 939 bytes --]

Hi Pedro,

> > gdb/ChangeLog:
> > 
> >         * aarch64-ravenscar-thread.h, aarch64-ravenscar-thread.c:
> >         New files.
> >         * aarch64-tdep.c: #include "aarch64-ravenscar-thread.h".
> >         (aarch64_gdbarch_init): Add call to register_aarch64_ravenscar_ops.
> >         * Makefile.in (ALL_64_TARGET_OBS): Add aarch64-ravenscar-thread.o.
> >         (HFILES_NO_SRCDIR): Add aarch64-ravenscar-thread.h.
> >         (ALLDEPFILES): Add aarch64-ravenscar-thread.c.
> >         * configure.tgt (cpu_obs) [aarch64*-*-*]: Add ravenscar-thread.o
> >         and aarch64-ravenscar-thread.o.
> > 
> > Tested on aarch64-elf using AdaCore's testsuite.
> > OK to apply?
> 
> IMO this deserves a NEWS entry.

Yes, indeed. Thanks for catching that.

Attached is a patch that does just that. OK to apply, Eli?

gdb/ChangeLog:

        * gdb/NEWS: Add entry documenting Ravenscar tasking support
        on AArch64 ELF.

Thanks,
-- 
Joel

[-- Attachment #2: 0001-gdb-NEWS-Document-support-for-Ravenscar-tasking-on-A.patch --]
[-- Type: text/x-diff, Size: 1010 bytes --]

From 740b34f0ec95c2bcac833b05bd545bb416d1ae34 Mon Sep 17 00:00:00 2001
From: Joel Brobecker <brobecker@adacore.com>
Date: Tue, 11 Sep 2018 15:38:19 +0200
Subject: [PATCH] gdb/NEWS: Document support for Ravenscar tasking on AArch64
 ELF

gdb/ChangeLog:

        * gdb/NEWS: Add entry documenting Ravenscar tasking support
        on AArch64 ELF.
---
 gdb/NEWS | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/gdb/NEWS b/gdb/NEWS
index 75436b0fc32..cba67c5a4db 100644
--- a/gdb/NEWS
+++ b/gdb/NEWS
@@ -18,6 +18,11 @@
 * DWARF index cache: GDB can now automatically save indices of DWARF
   symbols on disk to speed up further loading of the same binaries.
 
+* Ada task switching is now supported on aarch64-elf targets when
+  debugging a program using the Ravenscar Profile.  For more information,
+  see the "Tasking Support when using the Ravenscar Profile" section
+  in the GDB user manual.
+
 * GDB in batch mode now exits with status 1 if the last command to be
   executed failed.
 
-- 
2.17.1

Re: [RFA/commit] (Ada) Add ravenscar tasking support on AArch64

[Joel Brobecker]

[-- Attachment #1: Type: text/plain, Size: 1755 bytes --]

> > > gdb/ChangeLog:
> > > 
> > >         * aarch64-ravenscar-thread.h, aarch64-ravenscar-thread.c:
> > >         New files.
> > >         * aarch64-tdep.c: #include "aarch64-ravenscar-thread.h".
> > >         (aarch64_gdbarch_init): Add call to register_aarch64_ravenscar_ops.
> > >         * Makefile.in (ALL_64_TARGET_OBS): Add aarch64-ravenscar-thread.o.
> > >         (HFILES_NO_SRCDIR): Add aarch64-ravenscar-thread.h.
> > >         (ALLDEPFILES): Add aarch64-ravenscar-thread.c.
> > >         * configure.tgt (cpu_obs) [aarch64*-*-*]: Add ravenscar-thread.o
> > >         and aarch64-ravenscar-thread.o.
> > > 
> > > Tested on aarch64-elf using AdaCore's testsuite.
> > > OK to apply?
> > 
> > IMO this deserves a NEWS entry.
> 
> Yes, indeed. Thanks for catching that.
> Attached is a patch that does just that. OK to apply, Eli?

Actually - I thought that the patch for which the NEWS entry was
already in, but I immediately realized that it was not. So better
yet, here is a patch that merges the two:

gdb/ChangeLog:

        * aarch64-ravenscar-thread.h, aarch64-ravenscar-thread.c:
        New files.
        * aarch64-tdep.c: #include "aarch64-ravenscar-thread.h".
        (aarch64_gdbarch_init): Add call to register_aarch64_ravenscar_ops.
        * Makefile.in (ALL_64_TARGET_OBS): Add aarch64-ravenscar-thread.o.
        (HFILES_NO_SRCDIR): Add aarch64-ravenscar-thread.h.
        (ALLDEPFILES): Add aarch64-ravenscar-thread.c.
        * configure.tgt (cpu_obs) [aarch64*-*-*]: Add ravenscar-thread.o
        and aarch64-ravenscar-thread.o.
        * gdb/NEWS: Add entry documenting Ravenscar tasking support
        on AArch64 ELF.

No change compared to the last version, except for the addition of
the NEWS entry.

Thank you!
-- 
Joel

[-- Attachment #2: 0001-Ada-Add-ravenscar-tasking-support-on-AArch64.patch --]
[-- Type: text/x-diff, Size: 11382 bytes --]

From ce6f16cdf46a266db68aa4610fa5a1bef6323f23 Mon Sep 17 00:00:00 2001
From: Joel Brobecker <brobecker@adacore.com>
Date: Fri, 7 Sep 2018 17:53:37 -0400
Subject: [PATCH] (Ada) Add ravenscar tasking support on AArch64

This patch adds support for debugging Ravenscar tasks, similar to what
is done for ppc and sparc.

gdb/ChangeLog:

        * aarch64-ravenscar-thread.h, aarch64-ravenscar-thread.c:
        New files.
        * aarch64-tdep.c: #include "aarch64-ravenscar-thread.h".
        (aarch64_gdbarch_init): Add call to register_aarch64_ravenscar_ops.
        * Makefile.in (ALL_64_TARGET_OBS): Add aarch64-ravenscar-thread.o.
        (HFILES_NO_SRCDIR): Add aarch64-ravenscar-thread.h.
        (ALLDEPFILES): Add aarch64-ravenscar-thread.c.
        * configure.tgt (cpu_obs) [aarch64*-*-*]: Add ravenscar-thread.o
        and aarch64-ravenscar-thread.o.
        * gdb/NEWS: Add entry documenting Ravenscar tasking support
        on AArch64 ELF.
---
 gdb/Makefile.in                |   3 +
 gdb/NEWS                       |   5 +
 gdb/aarch64-ravenscar-thread.c | 213 +++++++++++++++++++++++++++++++++++++++++
 gdb/aarch64-ravenscar-thread.h |  27 ++++++
 gdb/aarch64-tdep.c             |   3 +
 gdb/configure.tgt              |   3 +-
 6 files changed, 253 insertions(+), 1 deletion(-)
 create mode 100644 gdb/aarch64-ravenscar-thread.c
 create mode 100644 gdb/aarch64-ravenscar-thread.h

diff --git a/gdb/Makefile.in b/gdb/Makefile.in
index d49f3ee..e985dfc 100644
--- a/gdb/Makefile.in
+++ b/gdb/Makefile.in
@@ -627,6 +627,7 @@ ALL_64_TARGET_OBS = \
 	aarch64-fbsd-tdep.o \
 	aarch64-linux-tdep.o \
 	aarch64-newlib-tdep.o \
+	aarch64-ravenscar-thread.o \
 	aarch64-tdep.o \
 	alpha-bsd-tdep.o \
 	alpha-linux-tdep.o \
@@ -1165,6 +1166,7 @@ SFILES = \
 # right, it is probably easiest just to list .h files here directly.
 
 HFILES_NO_SRCDIR = \
+	aarch64-ravenscar-thread.h \
 	aarch64-tdep.h \
 	ada-lang.h \
 	addrmap.h \
@@ -2169,6 +2171,7 @@ ALLDEPFILES = \
 	aarch64-linux-nat.c \
 	aarch64-linux-tdep.c \
 	aarch64-newlib-tdep.c \
+	aarch64-ravenscar-thread.c \
 	aarch64-tdep.c \
 	aix-thread.c \
 	alpha-bsd-nat.c \
diff --git a/gdb/NEWS b/gdb/NEWS
index 75436b0..cba67c5 100644
--- a/gdb/NEWS
+++ b/gdb/NEWS
@@ -18,6 +18,11 @@
 * DWARF index cache: GDB can now automatically save indices of DWARF
   symbols on disk to speed up further loading of the same binaries.
 
+* Ada task switching is now supported on aarch64-elf targets when
+  debugging a program using the Ravenscar Profile.  For more information,
+  see the "Tasking Support when using the Ravenscar Profile" section
+  in the GDB user manual.
+
 * GDB in batch mode now exits with status 1 if the last command to be
   executed failed.
 
diff --git a/gdb/aarch64-ravenscar-thread.c b/gdb/aarch64-ravenscar-thread.c
new file mode 100644
index 0000000..88629e4
--- /dev/null
+++ b/gdb/aarch64-ravenscar-thread.c
@@ -0,0 +1,213 @@
+/* Ravenscar Aarch64 target support.
+
+   Copyright (C) 2017-2018 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 "gdbcore.h"
+#include "regcache.h"
+#include "aarch64-tdep.h"
+#include "inferior.h"
+#include "ravenscar-thread.h"
+#include "aarch64-ravenscar-thread.h"
+
+#define NO_OFFSET -1
+
+/* See aarch64-tdep.h for register numbers.  */
+
+static const int aarch64_context_offsets[] =
+{
+  /* X0 - X28 */
+  NO_OFFSET, NO_OFFSET, NO_OFFSET, NO_OFFSET,
+  NO_OFFSET, NO_OFFSET, NO_OFFSET, NO_OFFSET,
+  NO_OFFSET, NO_OFFSET, NO_OFFSET, NO_OFFSET,
+  NO_OFFSET, NO_OFFSET, NO_OFFSET, NO_OFFSET,
+  NO_OFFSET, NO_OFFSET, NO_OFFSET, 0,
+  8,         16,        24,        32,
+  40,        48,        56,        64,
+  72,
+
+  /* FP, LR, SP, PC, CPSR */
+  /* Note that as task switch is synchronous, PC is in fact the LR here */
+  80,        88,        96,        88,
+  NO_OFFSET,
+
+  /* Q0 - Q31 */
+  112,       128,       144,       160,
+  176,       192,       208,       224,
+  240,       256,       272,       288,
+  304,       320,       336,       352,
+  368,       384,       400,       416,
+  432,       448,       464,       480,
+  496,       512,       528,       544,
+  560,       576,       592,       608,
+
+  /* FPSR, FPCR */
+  104,       108,
+
+  /* FPU Saved field */
+  624
+};
+
+/* The register layout info.  */
+
+struct ravenscar_reg_info
+{
+  /* A table providing the offset relative to the context structure
+     where each register is saved.  */
+  const int *context_offsets;
+
+  /* The number of elements in the context_offsets table above.  */
+  int context_offsets_size;
+};
+
+/* supply register REGNUM, which has been saved on REGISTER_ADDR, to the
+   regcache.  */
+
+static void
+supply_register_at_address (struct regcache *regcache, int regnum,
+                            CORE_ADDR register_addr)
+{
+  struct gdbarch *gdbarch = regcache->arch ();
+  int buf_size = register_size (gdbarch, regnum);
+  gdb_byte *buf;
+
+  buf = (gdb_byte *) alloca (buf_size);
+  read_memory (register_addr, buf, buf_size);
+  regcache->raw_supply (regnum, buf);
+}
+
+/* Return true if, for a non-running thread, REGNUM has been saved on the
+   Thread_Descriptor.  */
+
+static int
+register_in_thread_descriptor_p (const struct ravenscar_reg_info *reg_info,
+				 int regnum)
+{
+  /* Check FPU registers */
+  return (regnum < reg_info->context_offsets_size
+	  && reg_info->context_offsets[regnum] != NO_OFFSET);
+}
+
+/* to_fetch_registers when inferior_ptid is different from the running
+   thread.  */
+
+static void
+aarch64_ravenscar_generic_fetch_registers
+  (const struct ravenscar_reg_info *reg_info,
+   struct regcache *regcache, int regnum)
+{
+  struct gdbarch *gdbarch = regcache->arch ();
+  const int num_regs = gdbarch_num_regs (gdbarch);
+  int current_regnum;
+  CORE_ADDR current_address;
+  CORE_ADDR thread_descriptor_address;
+
+  /* The tid is the thread_id field, which is a pointer to the thread.  */
+  thread_descriptor_address = (CORE_ADDR) inferior_ptid.tid ();
+
+  /* Read registers.  */
+  for (current_regnum = 0; current_regnum < num_regs; current_regnum++)
+    {
+      if (register_in_thread_descriptor_p (reg_info, current_regnum))
+        {
+          current_address = thread_descriptor_address
+            + reg_info->context_offsets[current_regnum];
+          supply_register_at_address (regcache, current_regnum,
+                                      current_address);
+        }
+    }
+}
+
+/* to_prepare_to_store when inferior_ptid is different from the running
+   thread.  */
+
+static void
+aarch64_ravenscar_generic_prepare_to_store (struct regcache *regcache)
+{
+  /* Nothing to do.  */
+}
+
+/* to_store_registers when inferior_ptid is different from the running
+   thread.  */
+
+static void
+aarch64_ravenscar_generic_store_registers
+  (const struct ravenscar_reg_info *reg_info,
+   struct regcache *regcache, int regnum)
+{
+  struct gdbarch *gdbarch = regcache->arch ();
+  int buf_size = register_size (gdbarch, regnum);
+  gdb_byte buf[buf_size];
+  ULONGEST register_address;
+
+  if (register_in_thread_descriptor_p (reg_info, regnum))
+    register_address
+      = inferior_ptid.tid () + reg_info->context_offsets [regnum];
+  else
+    return;
+
+  regcache->raw_collect (regnum, buf);
+  write_memory (register_address,
+                buf,
+                buf_size);
+}
+
+/* The ravenscar_reg_info for most Aarch64 targets.  */
+
+static const struct ravenscar_reg_info aarch64_reg_info =
+{
+  aarch64_context_offsets,
+  ARRAY_SIZE (aarch64_context_offsets),
+};
+
+/* Implement the to_fetch_registers ravenscar_arch_ops method
+   for most Aarch64 targets.  */
+
+static void
+aarch64_ravenscar_fetch_registers (struct regcache *regcache, int regnum)
+{
+  aarch64_ravenscar_generic_fetch_registers
+    (&aarch64_reg_info, regcache, regnum);
+}
+
+/* Implement the to_store_registers ravenscar_arch_ops method
+   for most Aarch64 targets.  */
+
+static void
+aarch64_ravenscar_store_registers (struct regcache *regcache, int regnum)
+{
+  aarch64_ravenscar_generic_store_registers
+    (&aarch64_reg_info, regcache, regnum);
+}
+
+/* The ravenscar_arch_ops vector for most Aarch64 targets.  */
+
+static struct ravenscar_arch_ops aarch64_ravenscar_ops =
+{
+  aarch64_ravenscar_fetch_registers,
+  aarch64_ravenscar_store_registers,
+  aarch64_ravenscar_generic_prepare_to_store
+};
+
+/* Register aarch64_ravenscar_ops in GDBARCH.  */
+
+void
+register_aarch64_ravenscar_ops (struct gdbarch *gdbarch)
+{
+  set_gdbarch_ravenscar_ops (gdbarch, &aarch64_ravenscar_ops);
+}
diff --git a/gdb/aarch64-ravenscar-thread.h b/gdb/aarch64-ravenscar-thread.h
new file mode 100644
index 0000000..e900233
--- /dev/null
+++ b/gdb/aarch64-ravenscar-thread.h
@@ -0,0 +1,27 @@
+/* Ravenscar Aarch64 target support.
+ 
+   Copyright (C) 2017-2018 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 AARCH64_RAVENSCAR_THREAD_H
+#define AARCH64_RAVENSCAR_THREAD_H
+
+struct gdbarch;
+
+extern void register_aarch64_ravenscar_ops (struct gdbarch *gdbarch);
+
+#endif
diff --git a/gdb/aarch64-tdep.c b/gdb/aarch64-tdep.c
index d2e6ac6..02d1f0f 100644
--- a/gdb/aarch64-tdep.c
+++ b/gdb/aarch64-tdep.c
@@ -46,6 +46,7 @@
 #include "selftest.h"
 
 #include "aarch64-tdep.h"
+#include "aarch64-ravenscar-thread.h"
 
 #include "elf-bfd.h"
 #include "elf/aarch64.h"
@@ -3175,6 +3176,8 @@ aarch64_gdbarch_init (struct gdbarch_info info, struct gdbarch_list *arches)
 		  value_of_aarch64_user_reg,
 		  &aarch64_register_aliases[i].regnum);
 
+  register_aarch64_ravenscar_ops (gdbarch);
+
   return gdbarch;
 }
 
diff --git a/gdb/configure.tgt b/gdb/configure.tgt
index 6d1a4df..ebe7dff 100644
--- a/gdb/configure.tgt
+++ b/gdb/configure.tgt
@@ -48,7 +48,8 @@ amd64_tobjs="amd64-tdep.o arch/amd64.o ${x86_tobjs}"
 
 case "${targ}" in
 aarch64*-*-*)
-	cpu_obs="aarch64-tdep.o arch/aarch64-insn.o arch/aarch64.o";;
+	cpu_obs="aarch64-tdep.o arch/aarch64-insn.o arch/aarch64.o \
+		 ravenscar-thread.o aarch64-ravenscar-thread.o";;
 
 alpha*-*-*)
 	# Target: Alpha
-- 
2.1.4

Re: [RFA/NEWS] document support for Ravenscar task switching on aarch64-elf (was: "Re: [RFA/commit] (Ada) Add ravenscar tasking support on AArch64")

[Eli Zaretskii]

> Date: Tue, 11 Sep 2018 15:41:38 +0200
> From: Joel Brobecker <brobecker@adacore.com>
> Cc: gdb-patches@sourceware.org
> 
> > > gdb/ChangeLog:
> > > 
> > >         * aarch64-ravenscar-thread.h, aarch64-ravenscar-thread.c:
> > >         New files.
> > >         * aarch64-tdep.c: #include "aarch64-ravenscar-thread.h".
> > >         (aarch64_gdbarch_init): Add call to register_aarch64_ravenscar_ops.
> > >         * Makefile.in (ALL_64_TARGET_OBS): Add aarch64-ravenscar-thread.o.
> > >         (HFILES_NO_SRCDIR): Add aarch64-ravenscar-thread.h.
> > >         (ALLDEPFILES): Add aarch64-ravenscar-thread.c.
> > >         * configure.tgt (cpu_obs) [aarch64*-*-*]: Add ravenscar-thread.o
> > >         and aarch64-ravenscar-thread.o.
> > > 
> > > Tested on aarch64-elf using AdaCore's testsuite.
> > > OK to apply?
> > 
> > IMO this deserves a NEWS entry.
> 
> Yes, indeed. Thanks for catching that.
> 
> Attached is a patch that does just that. OK to apply, Eli?

Yes, thanks.

pushed: [RFA/commit] (Ada) Add ravenscar tasking support on AArch64

[Joel Brobecker]

Hello,

> gdb/ChangeLog:
> 
>         * aarch64-ravenscar-thread.h, aarch64-ravenscar-thread.c:
>         New files.
>         * aarch64-tdep.c: #include "aarch64-ravenscar-thread.h".
>         (aarch64_gdbarch_init): Add call to register_aarch64_ravenscar_ops.
>         * Makefile.in (ALL_64_TARGET_OBS): Add aarch64-ravenscar-thread.o.
>         (HFILES_NO_SRCDIR): Add aarch64-ravenscar-thread.h.
>         (ALLDEPFILES): Add aarch64-ravenscar-thread.c.
>         * configure.tgt (cpu_obs) [aarch64*-*-*]: Add ravenscar-thread.o
>         and aarch64-ravenscar-thread.o.
>         * gdb/NEWS: Add entry documenting Ravenscar tasking support
>         on AArch64 ELF.
> 
> No change compared to the last version, except for the addition of
> the NEWS entry.

For the record, I just pushed this patch to master.

Cheers,
-- 
Joel