[kvm-unit-tests PATCH 0/8] arm: fix building by adding a feature

[kvm-unit-tests PATCH 0/8] arm: fix building by adding a feature

[Andrew Jones]

The main motivator of this series is to get the arm build fixed.
522667917 "lib: backtrace printing" added a common backtrace()
function, but we missed that arm's __builtin_return_address only
accepts the input 0, and thus the function, which calls it with
all values 0 to 20, can't compile on arm. To fix backtrace() for
arm, the easiest thing to do is to override it with a backtrace
that works. So this series makes some cleanups to stack tracing
support and adds the support for ARM.


Andrew Jones (8):
  x86: change exit to abort again
  x86: trivial: there's no dump_stack.o
  pretty_print_stacks: keep the 'STACK:' line
  pretty_print_stacks: use elf file for the kernel
  pretty_print_stacks: addr2line may need a cross prefix
  stack: share api prototypes
  stack: copy common asm/stack.h bits to all arches
  arm: stack: add dump_stack support

 arm/Makefile.arm               |  4 ++++
 configure                      |  2 ++
 lib/arm/asm/stack.h            | 11 +++++++++++
 lib/arm/processor.c            |  1 +
 lib/arm/stack.c                | 41 +++++++++++++++++++++++++++++++++++++++++
 lib/arm64/asm/stack.h          |  8 ++++++++
 lib/powerpc/asm/stack.h        |  8 ++++++++
 lib/ppc64/asm/stack.h          |  8 ++++++++
 lib/stack.h                    |  9 +++++----
 lib/x86/asm/stack.h            |  3 ---
 lib/x86/desc.c                 |  2 +-
 scripts/pretty_print_stacks.py | 14 +++++++++++---
 x86/Makefile.common            |  2 +-
 13 files changed, 101 insertions(+), 12 deletions(-)
 create mode 100644 lib/arm/stack.c

-- 
2.4.11

[kvm-unit-tests PATCH 1/8] x86: change exit to abort again

[Andrew Jones]

This is needed again because ed0a50e5 "x86: lib: debug dump
on unhandled exceptions" must have missed it in a rebase.

Signed-off-by: Andrew Jones <drjones@redhat.com>
---
 lib/x86/desc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/lib/x86/desc.c b/lib/x86/desc.c
index 75765ef5d7804..402204ddcac44 100644
--- a/lib/x86/desc.c
+++ b/lib/x86/desc.c
@@ -96,7 +96,7 @@ static void unhandled_exception(struct ex_regs *regs, bool cpu)
 #endif
 	);
 	dump_frame_stack((void*) regs->rip, (void*) regs->rbp);
-	exit(7);
+	abort();
 }
 
 static void check_exception_table(struct ex_regs *regs)
-- 
2.4.11

Re: [kvm-unit-tests PATCH 1/8] x86: change exit to abort again

[Peter Feiner]

On Tue, Apr 19, 2016 at 7:28 AM, Andrew Jones <drjones@redhat.com> wrote:
> This is needed again because ed0a50e5 "x86: lib: debug dump
> on unhandled exceptions" must have missed it in a rebase.
>
> Signed-off-by: Andrew Jones <drjones@redhat.com>
Reviewed-by: Peter Feiner <pfeiner@google.com>

[kvm-unit-tests PATCH 2/8] x86: trivial: there's no dump_stack.o

[Andrew Jones]

Signed-off-by: Andrew Jones <drjones@redhat.com>
---
 x86/Makefile.common | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/x86/Makefile.common b/x86/Makefile.common
index ca8036722a535..298e5f721ff15 100644
--- a/x86/Makefile.common
+++ b/x86/Makefile.common
@@ -20,7 +20,7 @@ $(libcflat): CFLAGS += -ffreestanding -I lib
 CFLAGS += -m$(bits)
 CFLAGS += -O1
 
-# dump_stack.o relies on frame pointers.
+# stack.o relies on frame pointers.
 KEEP_FRAME_POINTER := y
 
 libgcc := $(shell $(CC) -m$(bits) --print-libgcc-file-name)
-- 
2.4.11

Re: [kvm-unit-tests PATCH 2/8] x86: trivial: there's no dump_stack.o

[Peter Feiner]

On Tue, Apr 19, 2016 at 7:28 AM, Andrew Jones <drjones@redhat.com> wrote:
> Signed-off-by: Andrew Jones <drjones@redhat.com>
Reviewed-by: Peter Feiner <pfeiner@google.com>

[kvm-unit-tests PATCH 3/8] pretty_print_stacks: keep the 'STACK:' line

[Andrew Jones]

It's sometimes useful to have the addresses for searching in a
disassembly. Anyway, there's no reason to drop it.

Signed-off-by: Andrew Jones <drjones@redhat.com>
---
 scripts/pretty_print_stacks.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/scripts/pretty_print_stacks.py b/scripts/pretty_print_stacks.py
index ee5a52edf07cd..11090deb122b1 100755
--- a/scripts/pretty_print_stacks.py
+++ b/scripts/pretty_print_stacks.py
@@ -67,8 +67,9 @@ def main():
             if line == '':
                 break
 
+            puts(line)
+
             if not line.strip().startswith('STACK:'):
-                puts(line)
                 continue
 
             try:
-- 
2.4.11

Re: [kvm-unit-tests PATCH 3/8] pretty_print_stacks: keep the 'STACK:' line

[Peter Feiner]

On Tue, Apr 19, 2016 at 7:28 AM, Andrew Jones <drjones@redhat.com> wrote:
> It's sometimes useful to have the addresses for searching in a
> disassembly. Anyway, there's no reason to drop it.
>
> Signed-off-by: Andrew Jones <drjones@redhat.com>
Reviewed-by: Peter Feiner <pfeiner@google.com>

[kvm-unit-tests PATCH 4/8] pretty_print_stacks: use elf file for the kernel

[Andrew Jones]

x86 doesn't seem to care, but other architectures do.

Signed-off-by: Andrew Jones <drjones@redhat.com>
---
 scripts/pretty_print_stacks.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/scripts/pretty_print_stacks.py b/scripts/pretty_print_stacks.py
index 11090deb122b1..db2f145323c45 100755
--- a/scripts/pretty_print_stacks.py
+++ b/scripts/pretty_print_stacks.py
@@ -58,7 +58,7 @@ def main():
         sys.stderr.write('usage: %s <kernel>\n' % sys.argv[0])
         sys.exit(1)
 
-    binary = sys.argv[1]
+    binary = sys.argv[1].replace("flat", "elf")
 
     try:
         while True:
-- 
2.4.11

Re: [kvm-unit-tests PATCH 4/8] pretty_print_stacks: use elf file for the kernel

[Andrew Jones]

On Tue, Apr 19, 2016 at 04:28:06PM +0200, Andrew Jones wrote:
> x86 doesn't seem to care, but other architectures do.
> 
> Signed-off-by: Andrew Jones <drjones@redhat.com>
> ---
>  scripts/pretty_print_stacks.py | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/scripts/pretty_print_stacks.py b/scripts/pretty_print_stacks.py
> index 11090deb122b1..db2f145323c45 100755
> --- a/scripts/pretty_print_stacks.py
> +++ b/scripts/pretty_print_stacks.py
> @@ -58,7 +58,7 @@ def main():
>          sys.stderr.write('usage: %s <kernel>\n' % sys.argv[0])
>          sys.exit(1)
>  
> -    binary = sys.argv[1]
> +    binary = sys.argv[1].replace("flat", "elf")

grr... I should have used (".flat", ".elf") since the string 'flat'
could easily show up in other parts of a test name.

>  
>      try:
>          while True:
> -- 
> 2.4.11
> 

Re: [kvm-unit-tests PATCH 4/8] pretty_print_stacks: use elf file for the kernel

[Peter Feiner]

On Tue, Apr 19, 2016 at 7:53 AM, Andrew Jones <drjones@redhat.com> wrote:
> On Tue, Apr 19, 2016 at 04:28:06PM +0200, Andrew Jones wrote:
>> x86 doesn't seem to care, but other architectures do.
>>
>> Signed-off-by: Andrew Jones <drjones@redhat.com>
With "." fix,
Reviewed-by: Peter Feiner <pfeiner@google.com>

[kvm-unit-tests PATCH 5/8] pretty_print_stacks: addr2line may need a cross prefix

[Andrew Jones]

Signed-off-by: Andrew Jones <drjones@redhat.com>
---
 configure                      | 2 ++
 scripts/pretty_print_stacks.py | 9 ++++++++-
 2 files changed, 10 insertions(+), 1 deletion(-)

diff --git a/configure b/configure
index ba6c55b3bc5b3..ca0ef83b942e5 100755
--- a/configure
+++ b/configure
@@ -7,6 +7,7 @@ ld=ld
 objcopy=objcopy
 objdump=objdump
 ar=ar
+addr2line=addr2line
 arch=`uname -m | sed -e 's/i.86/i386/;s/arm.*/arm/;s/ppc64.*/ppc64/'`
 host=$arch
 cross_prefix=
@@ -154,6 +155,7 @@ LD=$cross_prefix$ld
 OBJCOPY=$cross_prefix$objcopy
 OBJDUMP=$cross_prefix$objdump
 AR=$cross_prefix$ar
+ADDR2LINE=$cross_prefix$addr2line
 API=$api
 TEST_DIR=$testdir
 FIRMWARE=$firmware
diff --git a/scripts/pretty_print_stacks.py b/scripts/pretty_print_stacks.py
index db2f145323c45..336fc681e45b2 100755
--- a/scripts/pretty_print_stacks.py
+++ b/scripts/pretty_print_stacks.py
@@ -5,6 +5,8 @@ import subprocess
 import sys
 import traceback
 
+config = {}
+
 # Subvert output buffering.
 def puts(string):
     sys.stdout.write(string)
@@ -25,7 +27,7 @@ def pretty_print_stack(binary, line):
     # Output like this:
     #        0x004002be: start64 at path/to/kvm-unit-tests/x86/cstart64.S:208
     #         (inlined by) test_ept_violation at path/to/kvm-unit-tests/x86/vmx_tests.c:1719 (discriminator 1)
-    cmd = ['addr2line', '-e', binary, '-i', '-f', '--pretty', '--address']
+    cmd = [config["ADDR2LINE"], '-e', binary, '-i', '-f', '--pretty', '--address']
     cmd.extend(addrs)
 
     p = subprocess.Popen(cmd, stdout=subprocess.PIPE)
@@ -60,6 +62,11 @@ def main():
 
     binary = sys.argv[1].replace("flat", "elf")
 
+    with open("config.mak") as config_file:
+        for line in config_file:
+            name, val = line.partition("=")[::2]
+            config[name.strip()] = val.strip()
+
     try:
         while True:
             # Subvert input buffering.
-- 
2.4.11

Re: [kvm-unit-tests PATCH 5/8] pretty_print_stacks: addr2line may need a cross prefix

[Peter Feiner]

On Tue, Apr 19, 2016 at 7:28 AM, Andrew Jones <drjones@redhat.com> wrote:
> Signed-off-by: Andrew Jones <drjones@redhat.com>
> ---
>  configure                      | 2 ++
>  scripts/pretty_print_stacks.py | 9 ++++++++-
>  2 files changed, 10 insertions(+), 1 deletion(-)
>
> diff --git a/configure b/configure
> index ba6c55b3bc5b3..ca0ef83b942e5 100755
> --- a/configure
> +++ b/configure
> @@ -7,6 +7,7 @@ ld=ld
>  objcopy=objcopy
>  objdump=objdump
>  ar=ar
> +addr2line=addr2line
>  arch=`uname -m | sed -e 's/i.86/i386/;s/arm.*/arm/;s/ppc64.*/ppc64/'`
>  host=$arch
>  cross_prefix=
> @@ -154,6 +155,7 @@ LD=$cross_prefix$ld
>  OBJCOPY=$cross_prefix$objcopy
>  OBJDUMP=$cross_prefix$objdump
>  AR=$cross_prefix$ar
> +ADDR2LINE=$cross_prefix$addr2line
>  API=$api
>  TEST_DIR=$testdir
>  FIRMWARE=$firmware
> diff --git a/scripts/pretty_print_stacks.py b/scripts/pretty_print_stacks.py
> index db2f145323c45..336fc681e45b2 100755
> --- a/scripts/pretty_print_stacks.py
> +++ b/scripts/pretty_print_stacks.py
> @@ -5,6 +5,8 @@ import subprocess
>  import sys
>  import traceback
>
> +config = {}
> +
>  # Subvert output buffering.
>  def puts(string):
>      sys.stdout.write(string)
> @@ -25,7 +27,7 @@ def pretty_print_stack(binary, line):
>      # Output like this:
>      #        0x004002be: start64 at path/to/kvm-unit-tests/x86/cstart64.S:208
>      #         (inlined by) test_ept_violation at path/to/kvm-unit-tests/x86/vmx_tests.c:1719 (discriminator 1)
> -    cmd = ['addr2line', '-e', binary, '-i', '-f', '--pretty', '--address']
> +    cmd = [config["ADDR2LINE"], '-e', binary, '-i', '-f', '--pretty', '--address']

Perhaps config.get('ADDR2LINE', 'addr2line') so my old config still works?

>      cmd.extend(addrs)
>
>      p = subprocess.Popen(cmd, stdout=subprocess.PIPE)
> @@ -60,6 +62,11 @@ def main():
>
>      binary = sys.argv[1].replace("flat", "elf")
>
> +    with open("config.mak") as config_file:
> +        for line in config_file:
> +            name, val = line.partition("=")[::2]

Very elegant trick here! I always do name, _, val = line.partition("=")!

> +            config[name.strip()] = val.strip()
> +
>      try:
>          while True:
>              # Subvert input buffering.
> --
> 2.4.11
>

Otherwise, reviewed-by: Peter Feiner <pfeiner@google.com>

Re: [kvm-unit-tests PATCH 5/8] pretty_print_stacks: addr2line may need a cross prefix

[Andrew Jones]

On Tue, Apr 19, 2016 at 08:33:26AM -0700, Peter Feiner wrote:
> On Tue, Apr 19, 2016 at 7:28 AM, Andrew Jones <drjones@redhat.com> wrote:
> > Signed-off-by: Andrew Jones <drjones@redhat.com>
> > ---
> >  configure                      | 2 ++
> >  scripts/pretty_print_stacks.py | 9 ++++++++-
> >  2 files changed, 10 insertions(+), 1 deletion(-)
> >
> > diff --git a/configure b/configure
> > index ba6c55b3bc5b3..ca0ef83b942e5 100755
> > --- a/configure
> > +++ b/configure
> > @@ -7,6 +7,7 @@ ld=ld
> >  objcopy=objcopy
> >  objdump=objdump
> >  ar=ar
> > +addr2line=addr2line
> >  arch=`uname -m | sed -e 's/i.86/i386/;s/arm.*/arm/;s/ppc64.*/ppc64/'`
> >  host=$arch
> >  cross_prefix=
> > @@ -154,6 +155,7 @@ LD=$cross_prefix$ld
> >  OBJCOPY=$cross_prefix$objcopy
> >  OBJDUMP=$cross_prefix$objdump
> >  AR=$cross_prefix$ar
> > +ADDR2LINE=$cross_prefix$addr2line
> >  API=$api
> >  TEST_DIR=$testdir
> >  FIRMWARE=$firmware
> > diff --git a/scripts/pretty_print_stacks.py b/scripts/pretty_print_stacks.py
> > index db2f145323c45..336fc681e45b2 100755
> > --- a/scripts/pretty_print_stacks.py
> > +++ b/scripts/pretty_print_stacks.py
> > @@ -5,6 +5,8 @@ import subprocess
> >  import sys
> >  import traceback
> >
> > +config = {}
> > +
> >  # Subvert output buffering.
> >  def puts(string):
> >      sys.stdout.write(string)
> > @@ -25,7 +27,7 @@ def pretty_print_stack(binary, line):
> >      # Output like this:
> >      #        0x004002be: start64 at path/to/kvm-unit-tests/x86/cstart64.S:208
> >      #         (inlined by) test_ept_violation at path/to/kvm-unit-tests/x86/vmx_tests.c:1719 (discriminator 1)
> > -    cmd = ['addr2line', '-e', binary, '-i', '-f', '--pretty', '--address']
> > +    cmd = [config["ADDR2LINE"], '-e', binary, '-i', '-f', '--pretty', '--address']
> 
> Perhaps config.get('ADDR2LINE', 'addr2line') so my old config still works?

OK, will do in v2.

> 
> >      cmd.extend(addrs)
> >
> >      p = subprocess.Popen(cmd, stdout=subprocess.PIPE)
> > @@ -60,6 +62,11 @@ def main():
> >
> >      binary = sys.argv[1].replace("flat", "elf")
> >
> > +    with open("config.mak") as config_file:
> > +        for line in config_file:
> > +            name, val = line.partition("=")[::2]
> 
> Very elegant trick here! I always do name, _, val = line.partition("=")!
> 
> > +            config[name.strip()] = val.strip()
> > +
> >      try:
> >          while True:
> >              # Subvert input buffering.
> > --
> > 2.4.11
> >
> 
> Otherwise, reviewed-by: Peter Feiner <pfeiner@google.com>
> --
> To unsubscribe from this list: send the line "unsubscribe kvm" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

[kvm-unit-tests PATCH 6/8] stack: share api prototypes

[Andrew Jones]

Signed-off-by: Andrew Jones <drjones@redhat.com>
---
 lib/stack.h         | 9 +++++----
 lib/x86/asm/stack.h | 3 ---
 2 files changed, 5 insertions(+), 7 deletions(-)

diff --git a/lib/stack.h b/lib/stack.h
index bb6b9aa70876d..cfc66f442f997 100644
--- a/lib/stack.h
+++ b/lib/stack.h
@@ -4,7 +4,10 @@
 #include <libcflat.h>
 #include <asm/stack.h>
 
-#ifndef HAVE_ARCH_BACKTRACE_FRAME
+#ifdef HAVE_ARCH_BACKTRACE_FRAME
+extern int backtrace_frame(const void *frame, const void **return_addrs,
+			   int max_depth);
+#else
 static inline int
 backtrace_frame(const void *frame __unused, const void **return_addrs __unused,
 		int max_depth __unused)
@@ -13,8 +16,6 @@ backtrace_frame(const void *frame __unused, const void **return_addrs __unused,
 }
 #endif
 
-#ifndef HAVE_ARCH_BACKTRACE
-int backtrace(const void **return_addrs, int max_depth);
-#endif
+extern int backtrace(const void **return_addrs, int max_depth);
 
 #endif
diff --git a/lib/x86/asm/stack.h b/lib/x86/asm/stack.h
index fc4766d37b172..b14e2c0fa012e 100644
--- a/lib/x86/asm/stack.h
+++ b/lib/x86/asm/stack.h
@@ -6,9 +6,6 @@
 #endif
 
 #define HAVE_ARCH_BACKTRACE_FRAME
-int backtrace_frame(const void *frame, const void **return_addrs, int max_depth);
-
 #define HAVE_ARCH_BACKTRACE
-int backtrace(const void **return_addrs, int max_depth);
 
 #endif
-- 
2.4.11

Re: [kvm-unit-tests PATCH 6/8] stack: share api prototypes

[Peter Feiner]

On Tue, Apr 19, 2016 at 7:28 AM, Andrew Jones <drjones@redhat.com> wrote:
> Signed-off-by: Andrew Jones <drjones@redhat.com>
Reviewed-by: Peter Feiner <pfeiner@google.com>

[kvm-unit-tests PATCH 7/8] stack: copy common asm/stack.h bits to all arches

[Andrew Jones]

Signed-off-by: Andrew Jones <drjones@redhat.com>
---
 lib/arm/asm/stack.h     | 8 ++++++++
 lib/arm64/asm/stack.h   | 8 ++++++++
 lib/powerpc/asm/stack.h | 8 ++++++++
 lib/ppc64/asm/stack.h   | 8 ++++++++
 4 files changed, 32 insertions(+)

diff --git a/lib/arm/asm/stack.h b/lib/arm/asm/stack.h
index e69de29bb2d1d..fe97ac5d4b7d9 100644
--- a/lib/arm/asm/stack.h
+++ b/lib/arm/asm/stack.h
@@ -0,0 +1,8 @@
+#ifndef _ASMARM_STACK_H_
+#define _ASMARM_STACK_H_
+
+#ifndef _STACK_H_
+#error Do not directly include <asm/stack.h>. Just use <stack.h>.
+#endif
+
+#endif
diff --git a/lib/arm64/asm/stack.h b/lib/arm64/asm/stack.h
index e69de29bb2d1d..d0006242fdaaf 100644
--- a/lib/arm64/asm/stack.h
+++ b/lib/arm64/asm/stack.h
@@ -0,0 +1,8 @@
+#ifndef _ASMARM64_STACK_H_
+#define _ASMARM64_STACK_H_
+
+#ifndef _STACK_H_
+#error Do not directly include <asm/stack.h>. Just use <stack.h>.
+#endif
+
+#endif
diff --git a/lib/powerpc/asm/stack.h b/lib/powerpc/asm/stack.h
index e69de29bb2d1d..e1c46ee092a10 100644
--- a/lib/powerpc/asm/stack.h
+++ b/lib/powerpc/asm/stack.h
@@ -0,0 +1,8 @@
+#ifndef _ASMPOWERPC_STACK_H_
+#define _ASMPOWERPC_STACK_H_
+
+#ifndef _STACK_H_
+#error Do not directly include <asm/stack.h>. Just use <stack.h>.
+#endif
+
+#endif
diff --git a/lib/ppc64/asm/stack.h b/lib/ppc64/asm/stack.h
index e69de29bb2d1d..9734bbb8f7bd9 100644
--- a/lib/ppc64/asm/stack.h
+++ b/lib/ppc64/asm/stack.h
@@ -0,0 +1,8 @@
+#ifndef _ASMPPC64_STACK_H_
+#define _ASMPPC64_STACK_H_
+
+#ifndef _STACK_H_
+#error Do not directly include <asm/stack.h>. Just use <stack.h>.
+#endif
+
+#endif
-- 
2.4.11

Re: [kvm-unit-tests PATCH 7/8] stack: copy common asm/stack.h bits to all arches

[Peter Feiner]

On Tue, Apr 19, 2016 at 7:28 AM, Andrew Jones <drjones@redhat.com> wrote:
> Signed-off-by: Andrew Jones <drjones@redhat.com>
Reviewed-by: Peter Feiner <pfeiner@google.com>

[kvm-unit-tests PATCH 8/8] arm: stack: add dump_stack support

[Andrew Jones]

This actually fixes the arm build too, as arm's __builtin_return_address
doesn't allow any input other than zero, and thus the common backtrace()
wasn't compiling.

Signed-off-by: Andrew Jones <drjones@redhat.com>
---
 arm/Makefile.arm    |  4 ++++
 lib/arm/asm/stack.h |  3 +++
 lib/arm/processor.c |  1 +
 lib/arm/stack.c     | 41 +++++++++++++++++++++++++++++++++++++++++
 4 files changed, 49 insertions(+)
 create mode 100644 lib/arm/stack.c

diff --git a/arm/Makefile.arm b/arm/Makefile.arm
index 946422872532d..92f375700c45e 100644
--- a/arm/Makefile.arm
+++ b/arm/Makefile.arm
@@ -8,12 +8,16 @@ ldarch = elf32-littlearm
 kernel_offset = 0x10000
 machine = -marm
 
+# stack.o relies on frame pointers.
+KEEP_FRAME_POINTER := y
+
 CFLAGS += $(machine)
 CFLAGS += -mcpu=$(PROCESSOR)
 
 cstart.o = $(TEST_DIR)/cstart.o
 cflatobjs += lib/arm/spinlock.o
 cflatobjs += lib/arm/processor.o
+cflatobjs += lib/arm/stack.o
 
 # arm specific tests
 tests =
diff --git a/lib/arm/asm/stack.h b/lib/arm/asm/stack.h
index fe97ac5d4b7d9..ebcedc2ef938c 100644
--- a/lib/arm/asm/stack.h
+++ b/lib/arm/asm/stack.h
@@ -5,4 +5,7 @@
 #error Do not directly include <asm/stack.h>. Just use <stack.h>.
 #endif
 
+#define HAVE_ARCH_BACKTRACE_FRAME
+#define HAVE_ARCH_BACKTRACE
+
 #endif
diff --git a/lib/arm/processor.c b/lib/arm/processor.c
index 1cef46ab28647..54fdb87ef0196 100644
--- a/lib/arm/processor.c
+++ b/lib/arm/processor.c
@@ -108,6 +108,7 @@ void do_handle_exception(enum vector v, struct pt_regs *regs)
 		asm volatile("mrc p15, 0, %0, c5, c0, 1": "=r" (fsr));
 		printf("IFAR: %08lx    IFSR: %08lx\n", far, fsr);
 	}
+	dump_frame_stack((void *)regs->ARM_pc, (void *)regs->ARM_fp);
 	abort();
 }
 
diff --git a/lib/arm/stack.c b/lib/arm/stack.c
new file mode 100644
index 0000000000000..f58b731f277bf
--- /dev/null
+++ b/lib/arm/stack.c
@@ -0,0 +1,41 @@
+/*
+ * backtrace support (this is a modified lib/x86/stack.c)
+ *
+ * Copyright (C) 2016, Red Hat Inc, Andrew Jones <drjones@redhat.com>
+ *
+ * This work is licensed under the terms of the GNU LGPL, version 2.
+ */
+#include <libcflat.h>
+#include <stack.h>
+
+int backtrace_frame(const void *frame, const void **return_addrs,
+		    int max_depth)
+{
+	static int walking;
+	int depth = 0;
+	const unsigned long *fp = (unsigned long *)frame;
+
+	if (walking) {
+		printf("RECURSIVE STACK WALK!!!\n");
+		return 0;
+	}
+	walking = 1;
+
+	for (depth = 0; depth < max_depth; depth++) {
+		if (!fp)
+			break;
+		return_addrs[depth] = (void *)fp[0];
+		if (return_addrs[depth] == 0)
+			break;
+		fp = (unsigned long *)fp[-1];
+	}
+
+	walking = 0;
+	return depth;
+}
+
+int backtrace(const void **return_addrs, int max_depth)
+{
+	return backtrace_frame(__builtin_frame_address(0),
+			       return_addrs, max_depth);
+}
-- 
2.4.11

Re: [kvm-unit-tests PATCH 8/8] arm: stack: add dump_stack support

[Peter Feiner]

On Tue, Apr 19, 2016 at 7:28 AM, Andrew Jones <drjones@redhat.com> wrote:
> This actually fixes the arm build too, as arm's __builtin_return_address
> doesn't allow any input other than zero, and thus the common backtrace()
> wasn't compiling.
>
> Signed-off-by: Andrew Jones <drjones@redhat.com>
> ---
>  arm/Makefile.arm    |  4 ++++
>  lib/arm/asm/stack.h |  3 +++
>  lib/arm/processor.c |  1 +
>  lib/arm/stack.c     | 41 +++++++++++++++++++++++++++++++++++++++++
>  4 files changed, 49 insertions(+)
>  create mode 100644 lib/arm/stack.c
>
> diff --git a/arm/Makefile.arm b/arm/Makefile.arm
> index 946422872532d..92f375700c45e 100644
> --- a/arm/Makefile.arm
> +++ b/arm/Makefile.arm
> @@ -8,12 +8,16 @@ ldarch = elf32-littlearm
>  kernel_offset = 0x10000
>  machine = -marm
>
> +# stack.o relies on frame pointers.
> +KEEP_FRAME_POINTER := y
> +
>  CFLAGS += $(machine)
>  CFLAGS += -mcpu=$(PROCESSOR)
>
>  cstart.o = $(TEST_DIR)/cstart.o
>  cflatobjs += lib/arm/spinlock.o
>  cflatobjs += lib/arm/processor.o
> +cflatobjs += lib/arm/stack.o
>
>  # arm specific tests
>  tests =
> diff --git a/lib/arm/asm/stack.h b/lib/arm/asm/stack.h
> index fe97ac5d4b7d9..ebcedc2ef938c 100644
> --- a/lib/arm/asm/stack.h
> +++ b/lib/arm/asm/stack.h
> @@ -5,4 +5,7 @@
>  #error Do not directly include <asm/stack.h>. Just use <stack.h>.
>  #endif
>
> +#define HAVE_ARCH_BACKTRACE_FRAME
> +#define HAVE_ARCH_BACKTRACE
> +
>  #endif
> diff --git a/lib/arm/processor.c b/lib/arm/processor.c
> index 1cef46ab28647..54fdb87ef0196 100644
> --- a/lib/arm/processor.c
> +++ b/lib/arm/processor.c
> @@ -108,6 +108,7 @@ void do_handle_exception(enum vector v, struct pt_regs *regs)
>                 asm volatile("mrc p15, 0, %0, c5, c0, 1": "=r" (fsr));
>                 printf("IFAR: %08lx    IFSR: %08lx\n", far, fsr);
>         }
> +       dump_frame_stack((void *)regs->ARM_pc, (void *)regs->ARM_fp);
>         abort();
>  }
>
> diff --git a/lib/arm/stack.c b/lib/arm/stack.c
> new file mode 100644
> index 0000000000000..f58b731f277bf
> --- /dev/null
> +++ b/lib/arm/stack.c
> @@ -0,0 +1,41 @@
> +/*
> + * backtrace support (this is a modified lib/x86/stack.c)
> + *
> + * Copyright (C) 2016, Red Hat Inc, Andrew Jones <drjones@redhat.com>
> + *
> + * This work is licensed under the terms of the GNU LGPL, version 2.
> + */
> +#include <libcflat.h>
> +#include <stack.h>
> +
> +int backtrace_frame(const void *frame, const void **return_addrs,
> +                   int max_depth)
> +{
> +       static int walking;
> +       int depth = 0;

Nit: unnecessary initialization.

> +       const unsigned long *fp = (unsigned long *)frame;
> +
> +       if (walking) {
> +               printf("RECURSIVE STACK WALK!!!\n");
> +               return 0;
> +       }
> +       walking = 1;
> +
> +       for (depth = 0; depth < max_depth; depth++) {
> +               if (!fp)

Interesting difference in termination case here from x86. Instead of
checking for a null return address, you check for a null frame
address. Any particular reason why?

> +                       break;
> +               return_addrs[depth] = (void *)fp[0];
> +               if (return_addrs[depth] == 0)
> +                       break;
> +               fp = (unsigned long *)fp[-1];
> +       }
> +
> +       walking = 0;
> +       return depth;
> +}
> +
> +int backtrace(const void **return_addrs, int max_depth)
> +{
> +       return backtrace_frame(__builtin_frame_address(0),
> +                              return_addrs, max_depth);
> +}
> --
> 2.4.11
>

Nice fix!

Re: [kvm-unit-tests PATCH 8/8] arm: stack: add dump_stack support

[Andrew Jones]

On Tue, Apr 19, 2016 at 08:41:02AM -0700, Peter Feiner wrote:
> On Tue, Apr 19, 2016 at 7:28 AM, Andrew Jones <drjones@redhat.com> wrote:
> > This actually fixes the arm build too, as arm's __builtin_return_address
> > doesn't allow any input other than zero, and thus the common backtrace()
> > wasn't compiling.
> >
> > Signed-off-by: Andrew Jones <drjones@redhat.com>
> > ---
> >  arm/Makefile.arm    |  4 ++++
> >  lib/arm/asm/stack.h |  3 +++
> >  lib/arm/processor.c |  1 +
> >  lib/arm/stack.c     | 41 +++++++++++++++++++++++++++++++++++++++++
> >  4 files changed, 49 insertions(+)
> >  create mode 100644 lib/arm/stack.c
> >
> > diff --git a/arm/Makefile.arm b/arm/Makefile.arm
> > index 946422872532d..92f375700c45e 100644
> > --- a/arm/Makefile.arm
> > +++ b/arm/Makefile.arm
> > @@ -8,12 +8,16 @@ ldarch = elf32-littlearm
> >  kernel_offset = 0x10000
> >  machine = -marm
> >
> > +# stack.o relies on frame pointers.
> > +KEEP_FRAME_POINTER := y
> > +
> >  CFLAGS += $(machine)
> >  CFLAGS += -mcpu=$(PROCESSOR)
> >
> >  cstart.o = $(TEST_DIR)/cstart.o
> >  cflatobjs += lib/arm/spinlock.o
> >  cflatobjs += lib/arm/processor.o
> > +cflatobjs += lib/arm/stack.o
> >
> >  # arm specific tests
> >  tests =
> > diff --git a/lib/arm/asm/stack.h b/lib/arm/asm/stack.h
> > index fe97ac5d4b7d9..ebcedc2ef938c 100644
> > --- a/lib/arm/asm/stack.h
> > +++ b/lib/arm/asm/stack.h
> > @@ -5,4 +5,7 @@
> >  #error Do not directly include <asm/stack.h>. Just use <stack.h>.
> >  #endif
> >
> > +#define HAVE_ARCH_BACKTRACE_FRAME
> > +#define HAVE_ARCH_BACKTRACE
> > +
> >  #endif
> > diff --git a/lib/arm/processor.c b/lib/arm/processor.c
> > index 1cef46ab28647..54fdb87ef0196 100644
> > --- a/lib/arm/processor.c
> > +++ b/lib/arm/processor.c
> > @@ -108,6 +108,7 @@ void do_handle_exception(enum vector v, struct pt_regs *regs)
> >                 asm volatile("mrc p15, 0, %0, c5, c0, 1": "=r" (fsr));
> >                 printf("IFAR: %08lx    IFSR: %08lx\n", far, fsr);
> >         }
> > +       dump_frame_stack((void *)regs->ARM_pc, (void *)regs->ARM_fp);
> >         abort();
> >  }
> >
> > diff --git a/lib/arm/stack.c b/lib/arm/stack.c
> > new file mode 100644
> > index 0000000000000..f58b731f277bf
> > --- /dev/null
> > +++ b/lib/arm/stack.c
> > @@ -0,0 +1,41 @@
> > +/*
> > + * backtrace support (this is a modified lib/x86/stack.c)
> > + *
> > + * Copyright (C) 2016, Red Hat Inc, Andrew Jones <drjones@redhat.com>
> > + *
> > + * This work is licensed under the terms of the GNU LGPL, version 2.
> > + */
> > +#include <libcflat.h>
> > +#include <stack.h>
> > +
> > +int backtrace_frame(const void *frame, const void **return_addrs,
> > +                   int max_depth)
> > +{
> > +       static int walking;
> > +       int depth = 0;
> 
> Nit: unnecessary initialization.

/me catches the nit and flings it over to lib/x86/stack.c :-)

I'll fix for v2.

> 
> > +       const unsigned long *fp = (unsigned long *)frame;
> > +
> > +       if (walking) {
> > +               printf("RECURSIVE STACK WALK!!!\n");
> > +               return 0;
> > +       }
> > +       walking = 1;
> > +
> > +       for (depth = 0; depth < max_depth; depth++) {
> > +               if (!fp)
> 
> Interesting difference in termination case here from x86. Instead of
> checking for a null return address, you check for a null frame
> address. Any particular reason why?

When I configured a unit test that had more than one vcpu, then
if I dereferenced a null frame, address zero, I actually got
non-zero junk. Then fp was getting updated to fffffffc, and when
I dereferenced that (an unmapped address) it led to an exception.
So I just threw this test in to make sure that can't happen. It'll
also keep me safe in case regs->ARM_fp is ever null in
do_handle_exception.

> 
> > +                       break;
> > +               return_addrs[depth] = (void *)fp[0];
> > +               if (return_addrs[depth] == 0)
> > +                       break;
> > +               fp = (unsigned long *)fp[-1];
> > +       }
> > +
> > +       walking = 0;
> > +       return depth;
> > +}
> > +
> > +int backtrace(const void **return_addrs, int max_depth)
> > +{
> > +       return backtrace_frame(__builtin_frame_address(0),
> > +                              return_addrs, max_depth);
> > +}
> > --
> > 2.4.11
> >
> 
> Nice fix!
> --
> To unsubscribe from this list: send the line "unsubscribe kvm" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html