diff mbox

[PATCHv3] tests/tcg: fix build

Message ID 1350370438-32610-1-git-send-email-catalinp@google.com
State New
Headers show

Commit Message

Catalin Patulea Oct. 16, 2012, 6:53 a.m. UTC 
Sorry this got abandoned.. here is a fresh (and slightly leaner) patch on top of master. Can anyone take a look?

---
This broke when the tests were moved from tests/ to tests/tcg/.

On x86_64 host/i386-linux-user non-kvm guest, test-i386 and test-mmap are broken, but at least they build.

To build/run the tests:
$ cd $BUILD_PATH/tests/tcg
$ SRC_PATH=path/to/qemu make <target>
---
 tests/tcg/Makefile     |   16 ++++++++++------
 tests/tcg/linux-test.c |    2 ++
 tests/tcg/test-i386.c  |    3 ++-
 tests/tcg/test_path.c  |   13 +++++++------
 4 files changed, 21 insertions(+), 13 deletions(-)

Comments

Peter Maydell Oct. 16, 2012, 12:50 p.m. UTC | #1 
On 16 October 2012 07:53, Catalin Patulea <catalinp@google.com> wrote:
> Sorry this got abandoned.. here is a fresh (and slightly leaner) patch on top of master. Can anyone take a look?
>
> ---
> This broke when the tests were moved from tests/ to tests/tcg/.
>
> On x86_64 host/i386-linux-user non-kvm guest, test-i386 and test-mmap are broken, but at least they build.
>
> To build/run the tests:
> $ cd $BUILD_PATH/tests/tcg
> $ SRC_PATH=path/to/qemu make <target>

> @@ -86,7 +90,7 @@ testthread: testthread.c
>  # i386/x86_64 emulation test (test various opcodes) */
>  test-i386: test-i386.c test-i386-code16.S test-i386-vm86.S \
>             test-i386.h test-i386-shift.h test-i386-muldiv.h
> -       $(CC_I386) $(CFLAGS) $(LDFLAGS) -o $@ \
> +       $(CC_I386) $(QEMU_INCLUDES) $(CFLAGS) $(LDFLAGS) -o $@ \
>                $(<D)/test-i386.c $(<D)/test-i386-code16.S $(<D)/test-i386-vm86.S -lm

The compile line for the test-x86_64 target also needs
QEMU_INCLUDES added to it.

I can't work out how to get a working i386 compiler on my 64-bit Ubuntu system
but I think that's an Ubuntu issue :-) Tests run ok on a 32 bit native box
(there are some failures, as you say, but I'm unsurprised given how long it's
been since we ran them last.)

-- PMM
Catalin Patulea Oct. 16, 2012, 6:09 p.m. UTC | #2 
On Tue, Oct 16, 2012 at 8:50 AM, Peter Maydell <peter.maydell@linaro.org> wrote:
> The compile line for the test-x86_64 target also needs
> QEMU_INCLUDES added to it.
Done.

> I can't work out how to get a working i386 compiler on my 64-bit Ubuntu system
> but I think that's an Ubuntu issue :-) Tests run ok on a 32 bit native box
> (there are some failures, as you say, but I'm unsurprised given how long it's
> been since we ran them last.)
I'm running x86_64 precise with CC_I386="gcc -m32" (autodetected by
configure) - does that not work for you? Is there anything wrong with
building 32-bit binaries this way?
Peter Maydell Oct. 16, 2012, 6:47 p.m. UTC | #3 
On 16 October 2012 19:09, Catalin Patulea <catalinp@google.com> wrote:
> On Tue, Oct 16, 2012 at 8:50 AM, Peter Maydell <peter.maydell@linaro.org> wrote:
>> I can't work out how to get a working i386 compiler on my 64-bit Ubuntu system
>> but I think that's an Ubuntu issue :-) Tests run ok on a 32 bit native box
>> (there are some failures, as you say, but I'm unsurprised given how long it's
>> been since we ran them last.)
> I'm running x86_64 precise with CC_I386="gcc -m32" (autodetected by
> configure) - does that not work for you? Is there anything wrong with
> building 32-bit binaries this way?

I think that should work but my setup is broken somehow: it can't
find the 32 bit libgcc (and in passing finds and discards the 64 bit
version):

cam-vm-266:precise:tcg$ make linux-test
ccache gcc -m32 -Wall -O2 -g -fno-strict-aliasing  -o linux-test
linux-test.c -lm
/usr/bin/ld: skipping incompatible
/usr/lib/gcc/x86_64-linux-gnu/4.6/libgcc.a when searching for -lgcc
/usr/bin/ld: cannot find -lgcc
/usr/bin/ld: skipping incompatible
/usr/lib/gcc/x86_64-linux-gnu/4.6/libgcc_s.so when searching for
-lgcc_s
/usr/bin/ld: cannot find -lgcc_s
collect2: ld returned 1 exit status
make: *** [linux-test] Error 1

I have a feeling that precise's mostly-working multiarch support
may be the issue...

-- PMM
Catalin Patulea Oct. 16, 2012, 7:34 p.m. UTC | #4 
On Tue, Oct 16, 2012 at 2:47 PM, Peter Maydell <peter.maydell@linaro.org> wrote:
> I have a feeling that precise's mostly-working multiarch support
> may be the issue...
Seems I've got libgcc1:i386 and lib32gcc1 which provide libgcc_s.so*
in various forms. Not entirely sure what the 'right' setup is.
diff mbox

Patch

diff --git a/tests/tcg/Makefile b/tests/tcg/Makefile
index 15e36a2..9ff47b8 100644
--- a/tests/tcg/Makefile
+++ b/tests/tcg/Makefile
@@ -1,13 +1,13 @@ 
--include ../config-host.mak
+-include ../../config-host.mak
 -include $(SRC_PATH)/rules.mak
 
-$(call set-vpath, $(SRC_PATH)/tests)
+$(call set-vpath, $(SRC_PATH)/tests/tcg)
 
-QEMU=../i386-linux-user/qemu-i386
-QEMU_X86_64=../x86_64-linux-user/qemu-x86_64
+QEMU=../../i386-linux-user/qemu-i386
+QEMU_X86_64=../../x86_64-linux-user/qemu-x86_64
 CC_X86_64=$(CC_I386) -m64
 
-QEMU_INCLUDES += -I..
+QEMU_INCLUDES += -I../..
 CFLAGS=-Wall -O2 -g -fno-strict-aliasing
 #CFLAGS+=-msse2
 LDFLAGS=
@@ -36,6 +36,7 @@  TESTS += $(I386_TESTS)
 endif
 
 all: $(patsubst %,run-%,$(TESTS))
+test: all
 
 # rules to run tests
 
@@ -74,7 +75,10 @@  run-test_path: test_path
 # rules to compile tests
 
 test_path: test_path.o
+	$(CC_I386) $(LDFLAGS) -o $@ $^ $(LIBS)
+
 test_path.o: test_path.c
+	$(CC_I386) $(QEMU_INCLUDES) $(GLIB_CFLAGS) $(CFLAGS) -c -o $@ $^
 
 hello-i386: hello-i386.c
 	$(CC_I386) -nostdlib $(CFLAGS) -static $(LDFLAGS) -o $@ $<
@@ -86,7 +90,7 @@  testthread: testthread.c
 # i386/x86_64 emulation test (test various opcodes) */
 test-i386: test-i386.c test-i386-code16.S test-i386-vm86.S \
            test-i386.h test-i386-shift.h test-i386-muldiv.h
-	$(CC_I386) $(CFLAGS) $(LDFLAGS) -o $@ \
+	$(CC_I386) $(QEMU_INCLUDES) $(CFLAGS) $(LDFLAGS) -o $@ \
               $(<D)/test-i386.c $(<D)/test-i386-code16.S $(<D)/test-i386-vm86.S -lm
 
 test-x86_64: test-i386.c \
diff --git a/tests/tcg/linux-test.c b/tests/tcg/linux-test.c
index 2e4a746..83cb32d 100644
--- a/tests/tcg/linux-test.c
+++ b/tests/tcg/linux-test.c
@@ -16,6 +16,7 @@ 
  *  You should have received a copy of the GNU General Public License
  *  along with this program; if not, see <http://www.gnu.org/licenses/>.
  */
+#define _GNU_SOURCE
 #include <stdarg.h>
 #include <stdlib.h>
 #include <stdio.h>
@@ -38,6 +39,7 @@ 
 #include <dirent.h>
 #include <setjmp.h>
 #include <sys/shm.h>
+#include <sched.h>
 
 #define TESTPATH "/tmp/linux-test.tmp"
 #define TESTPORT 7654
diff --git a/tests/tcg/test-i386.c b/tests/tcg/test-i386.c
index 8e64bba..64d929e 100644
--- a/tests/tcg/test-i386.c
+++ b/tests/tcg/test-i386.c
@@ -17,6 +17,7 @@ 
  *  along with this program; if not, see <http://www.gnu.org/licenses/>.
  */
 #define _GNU_SOURCE
+#include "compiler.h"
 #include <stdlib.h>
 #include <stdio.h>
 #include <string.h>
@@ -1827,7 +1828,7 @@  void test_exceptions(void)
     printf("lock nop exception:\n");
     if (setjmp(jmp_env) == 0) {
         /* now execute an invalid instruction */
-        asm volatile("lock nop");
+        asm volatile(".byte 0xf0, 0x90"); /* lock nop */
     }
 
     printf("INT exception:\n");
diff --git a/tests/tcg/test_path.c b/tests/tcg/test_path.c
index 7265a94..a064eea 100644
--- a/tests/tcg/test_path.c
+++ b/tests/tcg/test_path.c
@@ -1,11 +1,12 @@ 
 /* Test path override code */
-#include "../config-host.h"
-#include "../qemu-malloc.c"
-#include "../cutils.c"
-#include "../path.c"
-#include "../trace.c"
+#define _GNU_SOURCE
+#include "config-host.h"
+#include "iov.c"
+#include "cutils.c"
+#include "path.c"
+#include "trace.c"
 #ifdef CONFIG_TRACE_SIMPLE
-#include "../simpletrace.c"
+#include "../trace/simple.c"
 #endif
 
 #include <stdarg.h>