Patchwork [RFC,6/6] Linux kernel thread awareness AArch64 target support

login
register
mail settings
Submitter Omair Javaid
Date Jan. 29, 2019, 5:03 a.m.
Message ID <1548738199-9403-7-git-send-email-omair.javaid@linaro.org>
Download mbox | patch
Permalink /patch/31240/
State New
Headers show

Comments

Omair Javaid - Jan. 29, 2019, 5:03 a.m.
This patch adds support for linux kernel thread aware debugging on
AArch64 targets. It implements aarch64_linux_kernel_ops derived class to
provide AArch64 targets specific functionality to linux_kernel_ops class.

gdb/ChangeLog:

	* Makefile.in (ALL_TARGET_OBS): Add aarch64-lk-tdep.o.
	(ALLDEPFILES): Add aarch64-lk-tdep.c.
	* configure.tgt (aarch64*-*-linux*): Add aarch64-lk-tdep.o and lk_tobjs.
	* aarch64-lk-tdep.c: New file.
---
 gdb/Makefile.in       |   2 +
 gdb/aarch64-lk-tdep.c | 135 ++++++++++++++++++++++++++++++++++++++++++++++++++
 gdb/configure.tgt     |   2 +-
 3 files changed, 138 insertions(+), 1 deletion(-)
 create mode 100644 gdb/aarch64-lk-tdep.c

Patch

diff --git a/gdb/Makefile.in b/gdb/Makefile.in
index f6dbfdd..e0f4ff7 100644
--- a/gdb/Makefile.in
+++ b/gdb/Makefile.in
@@ -633,6 +633,7 @@  TARGET_OBS = @TARGET_OBS@
 ALL_64_TARGET_OBS = \
 	aarch64-fbsd-tdep.o \
 	aarch64-linux-tdep.o \
+	aarch64-lk-tdep.o \
 	aarch64-newlib-tdep.o \
 	aarch64-ravenscar-thread.o \
 	aarch64-tdep.o \
@@ -2161,6 +2162,7 @@  ALLDEPFILES = \
 	aarch64-fbsd-tdep.c \
 	aarch64-linux-nat.c \
 	aarch64-linux-tdep.c \
+	aarch64-lk-tdep.c \
 	aarch64-newlib-tdep.c \
 	aarch64-ravenscar-thread.c \
 	aarch64-tdep.c \
diff --git a/gdb/aarch64-lk-tdep.c b/gdb/aarch64-lk-tdep.c
new file mode 100644
index 0000000..f380e6e
--- /dev/null
+++ b/gdb/aarch64-lk-tdep.c
@@ -0,0 +1,135 @@ 
+/* Target-dependent code for linux-kernel target on AArch64 architecture.
+   Copyright (C) 2019 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 "osabi.h"
+#include "regset.h"
+#include "gdbcore.h"
+
+#include "arch/aarch64.h"
+#include "lk-low.h"
+
+/* Define register map and set.  */
+
+/* From linux kernel source arch/arm64/include/asm/processor.h
+   struct cpu_context.  */
+
+static const struct regcache_map_entry aarch64_gregmap_lk[] =
+  {
+    { 11, AARCH64_X0_REGNUM + 19, 8 },
+    { 1, AARCH64_SP_REGNUM, 8 },
+    { 1, AARCH64_PC_REGNUM, 8 },
+    { 0 }
+  };
+
+const struct regset aarch64_gregset_lk =
+  {
+    aarch64_gregmap_lk,
+    regcache_supply_regset,
+    regcache_collect_regset
+  };
+
+/* AArch64 specific implementation for linux_kernel_ops.  */
+
+class aarch64_linux_kernel_ops : public linux_kernel_ops
+{
+public:
+  aarch64_linux_kernel_ops ()
+  {}
+
+  void get_registers (CORE_ADDR task, struct regcache *regcache,
+                      int regnum) override;
+protected:
+  void arch_read_symbols () override;
+};
+
+/* Implement linux_kernel_ops->arch_read_symbols virtual method.  */
+
+void
+aarch64_linux_kernel_ops::arch_read_symbols ()
+{
+  declare_field ("task_struct->thread", LK_CONFIG_ALWAYS);
+  declare_field ("thread_struct->cpu_context", LK_CONFIG_ALWAYS);
+}
+
+/* Implement linux_kernel_ops->get_registers virtual method.  */
+
+void
+aarch64_linux_kernel_ops::get_registers (CORE_ADDR task,
+                                         struct regcache *regcache, int regnum)
+{
+  CORE_ADDR cpu_context;
+  gdb_byte buf[104];
+  size_t size;
+
+  /* Calculate cpu_context address from task_struct address.  */
+  cpu_context = task + offset ("task_struct->thread")
+                + offset ("thread_struct->cpu_context");
+
+  size = FIELD_SIZE (field ("thread_struct->cpu_context"));
+
+  gdb_assert (size <= sizeof (buf));
+
+  read_memory (cpu_context, buf, size);
+
+  regcache->supply_regset (&aarch64_gregset_lk, -1, buf, size);
+}
+
+/* Implement gdbarch get_new_lk_ops hook.  */
+
+static linux_kernel_ops *
+aarch64_get_new_lk_ops (struct gdbarch *gdbarch)
+{
+  return new aarch64_linux_kernel_ops;
+}
+
+/* Initialize Linux kernel specific gdbarch hooks.  */
+
+static void
+aarch64_lk_init_abi (struct gdbarch_info info, struct gdbarch *gdbarch)
+{
+  /* Support linux kernel debugging.  */
+  set_gdbarch_get_new_lk_ops (gdbarch, aarch64_get_new_lk_ops);
+}
+
+/* From the ELF-headers point of view Linux kernel- and user-space binaries
+   are the same.  So we have to rely on some heuristics to distinguish them.
+
+   For the heuristic we check for __ksymtab section in the vmlinux and
+   modules .ko files.  */
+
+static enum gdb_osabi
+aarch64_lk_osabi_sniffer (bfd *abfd)
+{
+  if (bfd_get_section_by_name (abfd, "__ksymtab"))
+    return GDB_OSABI_LINUX_KERNEL;
+
+  return GDB_OSABI_UNKNOWN;
+}
+
+void
+_initialize_aarch64_lk_tdep (void)
+{
+  /* Hook us into the OSABI mechanism.  */
+  gdbarch_register_osabi (bfd_arch_aarch64, 0, GDB_OSABI_LINUX_KERNEL,
+                          aarch64_lk_init_abi);
+
+  /* Add OSABI sniffer.  */
+  gdbarch_register_osabi_sniffer (bfd_arch_aarch64, bfd_target_elf_flavour,
+                                  aarch64_lk_osabi_sniffer);
+}
diff --git a/gdb/configure.tgt b/gdb/configure.tgt
index fe5a653..d45deef 100644
--- a/gdb/configure.tgt
+++ b/gdb/configure.tgt
@@ -128,7 +128,7 @@  aarch64*-*-linux*)
 			arch/arm.o arch/arm-linux.o arch/arm-get-next-pcs.o \
 			arm-tdep.o arm-linux-tdep.o \
 			glibc-tdep.o linux-tdep.o solib-svr4.o \
-			symfile-mem.o linux-record.o"
+			symfile-mem.o linux-record.o aarch64-lk-tdep.o ${lk_tobjs}"
 	build_gdbserver=yes
 	;;