[PATCH] kdump: x86_64 save cpu registers upon crash

- Saving the cpu registers of all cpus before booting in to the crash
  kernel.

- crash_setup_regs will save the registers of the cpu on which panic has
  occured.  One of the concerns ppc64 folks raised is that after capturing the
  register states, one should not pop the current call frame and push new one.
   Hence it has been inlined.  More call frames later get pushed on to stack
  (machine_crash_shutdown() and machine_kexec()), but one will not want to
  backtrace those.

- Not very sure about the CFI annotations.  With this patch I am getting
  decent backtrace with gdb.  Assuming, compiler has generated enough
  debugging information for crash_kexec().  Coding crash_setup_regs() in pure
  assembly makes it tricky because then it can not be inlined and we don't
  want to return back after capturing register states we don't want to pop
  this call frame.

- Saving the non-panicing cpus registers will be done in the NMI handler
  while shooting down them in machine_crash_shutdown.

- Introducing CRASH_DUMP option in Kconfig for x86_64.

Signed-off-by: Murali M Chakravarthy <muralim@in.ibm.com>
Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com>
Cc: Andi Kleen <ak@muc.de>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

2 files changed, 77 insertions, 0 deletions

diff --git a/arch/x86_64/Kconfig b/arch/x86_64/Kconfig
index 4f3e925962c..dd2d116b9ab 100644
--- a/arch/x86_64/Kconfig
+++ b/arch/x86_64/Kconfig
@@ -427,6 +427,13 @@ config KEXEC
 	  support.  As of this writing the exact hardware interface is
 	  strongly in flux, so no good recommendation can be made.
 
+config CRASH_DUMP
+	bool "kernel crash dumps (EXPERIMENTAL)"
+	depends on EMBEDDED
+	depends on EXPERIMENTAL
+	help
+		Generate crash dump after being started by kexec.
+
 config SECCOMP
 	bool "Enable seccomp to safely compute untrusted bytecode"
 	depends on PROC_FS
diff --git a/arch/x86_64/kernel/crash.c b/arch/x86_64/kernel/crash.c
index abc601f36b6..4e6c3b729e3 100644
--- a/arch/x86_64/kernel/crash.c
+++ b/arch/x86_64/kernel/crash.c
@@ -11,9 +11,12 @@
 #include <linux/types.h>
 #include <linux/kernel.h>
 #include <linux/smp.h>
+#include <linux/irq.h>
 #include <linux/reboot.h>
 #include <linux/kexec.h>
 #include <linux/delay.h>
+#include <linux/elf.h>
+#include <linux/elfcore.h>
 
 #include <asm/processor.h>
 #include <asm/hardirq.h>
@@ -24,6 +27,71 @@
 /* This keeps a track of which one is crashing cpu. */
 static int crashing_cpu;
 
+static u32 *append_elf_note(u32 *buf, char *name, unsigned type,
+						void *data, size_t data_len)
+{
+	struct elf_note note;
+
+	note.n_namesz = strlen(name) + 1;
+	note.n_descsz = data_len;
+	note.n_type   = type;
+	memcpy(buf, &note, sizeof(note));
+	buf += (sizeof(note) +3)/4;
+	memcpy(buf, name, note.n_namesz);
+	buf += (note.n_namesz + 3)/4;
+	memcpy(buf, data, note.n_descsz);
+	buf += (note.n_descsz + 3)/4;
+
+	return buf;
+}
+
+static void final_note(u32 *buf)
+{
+	struct elf_note note;
+
+	note.n_namesz = 0;
+	note.n_descsz = 0;
+	note.n_type   = 0;
+	memcpy(buf, &note, sizeof(note));
+}
+
+static void crash_save_this_cpu(struct pt_regs *regs, int cpu)
+{
+	struct elf_prstatus prstatus;
+	u32 *buf;
+
+	if ((cpu < 0) || (cpu >= NR_CPUS))
+		return;
+
+	/* Using ELF notes here is opportunistic.
+	 * I need a well defined structure format
+	 * for the data I pass, and I need tags
+	 * on the data to indicate what information I have
+	 * squirrelled away.  ELF notes happen to provide
+	 * all of that that no need to invent something new.
+	 */
+
+	buf = (u32*)per_cpu_ptr(crash_notes, cpu);
+
+	if (!buf)
+		return;
+
+	memset(&prstatus, 0, sizeof(prstatus));
+	prstatus.pr_pid = current->pid;
+	elf_core_copy_regs(&prstatus.pr_reg, regs);
+	buf = append_elf_note(buf, "CORE", NT_PRSTATUS, &prstatus,
+					sizeof(prstatus));
+	final_note(buf);
+}
+
+static void crash_save_self(struct pt_regs *regs)
+{
+	int cpu;
+
+	cpu = smp_processor_id();
+	crash_save_this_cpu(regs, cpu);
+}
+
 #ifdef CONFIG_SMP
 static atomic_t waiting_for_crash_ipi;
 
@@ -38,6 +106,7 @@ static int crash_nmi_callback(struct pt_regs *regs, int cpu)
 		return 1;
 	local_irq_disable();
 
+	crash_save_this_cpu(regs, cpu);
 	disable_local_APIC();
 	atomic_dec(&waiting_for_crash_ipi);
 	/* Assume hlt works */
@@ -113,4 +182,5 @@ void machine_crash_shutdown(struct pt_regs *regs)
 	disable_IO_APIC();
 #endif
 
+	crash_save_self(regs);
 }