qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Alex Bennée" <alex.bennee@linaro.org>
To: qemu-devel@nongnu.org
Cc: "Zhao Liu" <zhao1.liu@intel.com>,
	"Eduardo Habkost" <eduardo@habkost.net>,
	"Pierrick Bouvier" <pierrick.bouvier@linaro.org>,
	"Peter Maydell" <peter.maydell@linaro.org>,
	"Yanan Wang" <wangyanan55@huawei.com>,
	"John Snow" <jsnow@redhat.com>,
	"Philippe Mathieu-Daudé" <philmd@linaro.org>,
	devel@lists.libvirt.org, "Marcelo Tosatti" <mtosatti@redhat.com>,
	"Riku Voipio" <riku.voipio@iki.fi>,
	"Wainer dos Santos Moschetta" <wainersm@redhat.com>,
	"Marcel Apfelbaum" <marcel.apfelbaum@gmail.com>,
	"Paolo Bonzini" <pbonzini@redhat.com>,
	"Mahmoud Mandour" <ma.mandourr@gmail.com>,
	"Laurent Vivier" <laurent@vivier.eu>,
	"Jiaxun Yang" <jiaxun.yang@flygoat.com>,
	"Edgar E. Iglesias" <edgar.iglesias@gmail.com>,
	"Richard Henderson" <richard.henderson@linaro.org>,
	"Alexandre Iooss" <erdnaxe@crans.org>,
	qemu-arm@nongnu.org, "Daniel P. Berrangé" <berrange@redhat.com>,
	"Marc-André Lureau" <marcandre.lureau@redhat.com>,
	"Cleber Rosa" <crosa@redhat.com>,
	kvm@vger.kernel.org, "Beraldo Leal" <bleal@redhat.com>,
	"Thomas Huth" <thuth@redhat.com>,
	"Alex Bennée" <alex.bennee@linaro.org>,
	"Ilya Leoshkevich" <iii@linux.ibm.com>
Subject: [PATCH v3 07/18] tests/tcg/x86_64: Add cross-modifying code test
Date: Wed, 23 Oct 2024 12:33:55 +0100	[thread overview]
Message-ID: <20241023113406.1284676-8-alex.bennee@linaro.org> (raw)
In-Reply-To: <20241023113406.1284676-1-alex.bennee@linaro.org>

From: Ilya Leoshkevich <iii@linux.ibm.com>

commit f025692c992c ("accel/tcg: Clear PAGE_WRITE before translation")
fixed cross-modifying code handling, but did not add a test. The
changed code was further improved recently [1], and I was not sure
whether these modifications were safe (spoiler: they were fine).

Add a test to make sure there are no regressions.

[1] https://lists.gnu.org/archive/html/qemu-devel/2022-09/msg00034.html

Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Message-Id: <20241022105614.839199-8-alex.bennee@linaro.org>
Message-Id: <20241001150617.9977-1-iii@linux.ibm.com>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
---
 tests/tcg/x86_64/cross-modifying-code.c | 80 +++++++++++++++++++++++++
 tests/tcg/x86_64/Makefile.target        |  4 ++
 2 files changed, 84 insertions(+)
 create mode 100644 tests/tcg/x86_64/cross-modifying-code.c

diff --git a/tests/tcg/x86_64/cross-modifying-code.c b/tests/tcg/x86_64/cross-modifying-code.c
new file mode 100644
index 0000000000..2704df6061
--- /dev/null
+++ b/tests/tcg/x86_64/cross-modifying-code.c
@@ -0,0 +1,80 @@
+/*
+ * Test patching code, running in one thread, from another thread.
+ *
+ * Intel SDM calls this "cross-modifying code" and recommends a special
+ * sequence, which requires both threads to cooperate.
+ *
+ * Linux kernel uses a different sequence that does not require cooperation and
+ * involves patching the first byte with int3.
+ *
+ * Finally, there is user-mode software out there that simply uses atomics, and
+ * that seems to be good enough in practice. Test that QEMU has no problems
+ * with this as well.
+ */
+
+#include <assert.h>
+#include <pthread.h>
+#include <stdbool.h>
+#include <stdlib.h>
+
+void add1_or_nop(long *x);
+asm(".pushsection .rwx,\"awx\",@progbits\n"
+    ".globl add1_or_nop\n"
+    /* addq $0x1,(%rdi) */
+    "add1_or_nop: .byte 0x48, 0x83, 0x07, 0x01\n"
+    "ret\n"
+    ".popsection\n");
+
+#define THREAD_WAIT 0
+#define THREAD_PATCH 1
+#define THREAD_STOP 2
+
+static void *thread_func(void *arg)
+{
+    int val = 0x0026748d; /* nop */
+
+    while (true) {
+        switch (__atomic_load_n((int *)arg, __ATOMIC_SEQ_CST)) {
+        case THREAD_WAIT:
+            break;
+        case THREAD_PATCH:
+            val = __atomic_exchange_n((int *)&add1_or_nop, val,
+                                      __ATOMIC_SEQ_CST);
+            break;
+        case THREAD_STOP:
+            return NULL;
+        default:
+            assert(false);
+            __builtin_unreachable();
+        }
+    }
+}
+
+#define INITIAL 42
+#define COUNT 1000000
+
+int main(void)
+{
+    int command = THREAD_WAIT;
+    pthread_t thread;
+    long x = 0;
+    int err;
+    int i;
+
+    err = pthread_create(&thread, NULL, &thread_func, &command);
+    assert(err == 0);
+
+    __atomic_store_n(&command, THREAD_PATCH, __ATOMIC_SEQ_CST);
+    for (i = 0; i < COUNT; i++) {
+        add1_or_nop(&x);
+    }
+    __atomic_store_n(&command, THREAD_STOP, __ATOMIC_SEQ_CST);
+
+    err = pthread_join(thread, NULL);
+    assert(err == 0);
+
+    assert(x >= INITIAL);
+    assert(x <= INITIAL + COUNT);
+
+    return EXIT_SUCCESS;
+}
diff --git a/tests/tcg/x86_64/Makefile.target b/tests/tcg/x86_64/Makefile.target
index 783ab5b21a..d6dff559c7 100644
--- a/tests/tcg/x86_64/Makefile.target
+++ b/tests/tcg/x86_64/Makefile.target
@@ -17,6 +17,7 @@ X86_64_TESTS += cmpxchg
 X86_64_TESTS += adox
 X86_64_TESTS += test-1648
 X86_64_TESTS += test-2175
+X86_64_TESTS += cross-modifying-code
 TESTS=$(MULTIARCH_TESTS) $(X86_64_TESTS) test-x86_64
 else
 TESTS=$(MULTIARCH_TESTS)
@@ -27,6 +28,9 @@ adox: CFLAGS=-O2
 run-test-i386-ssse3: QEMU_OPTS += -cpu max
 run-plugin-test-i386-ssse3-%: QEMU_OPTS += -cpu max
 
+cross-modifying-code: CFLAGS+=-pthread
+cross-modifying-code: LDFLAGS+=-pthread
+
 test-x86_64: LDFLAGS+=-lm -lc
 test-x86_64: test-i386.c test-i386.h test-i386-shift.h test-i386-muldiv.h
 	$(CC) $(CFLAGS) $< -o $@ $(LDFLAGS)
-- 
2.39.5



  parent reply	other threads:[~2024-10-23 11:40 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-10-23 11:33 [PATCH v3 00/18] maintainer updates (testing, gdbstub, plugins) pre-PR Alex Bennée
2024-10-23 11:33 ` [PATCH v3 01/18] tests/docker: Fix microblaze atomics Alex Bennée
2024-10-23 11:33 ` [PATCH v3 02/18] tests/docker: add NOFETCH env variable for testing Alex Bennée
2024-10-23 11:33 ` [PATCH v3 03/18] MAINTAINERS: mention my testing/next tree Alex Bennée
2024-10-23 11:33 ` [PATCH v3 04/18] meson: hide tsan related warnings Alex Bennée
2024-10-23 11:33 ` [PATCH v3 05/18] docs/devel: update tsan build documentation Alex Bennée
2024-10-23 11:33 ` [PATCH v3 06/18] scripts/ci: remove architecture checks for build-environment updates Alex Bennée
2024-10-23 11:33 ` Alex Bennée [this message]
2024-10-23 11:33 ` [PATCH v3 08/18] accel/tcg: add tracepoints for cpu_loop_exit_atomic Alex Bennée
2024-10-23 11:33 ` [PATCH v3 09/18] dockerfiles: fix default targets for debian-loongarch-cross Alex Bennée
2024-10-23 11:33 ` [PATCH v3 10/18] gitlab: make check-[dco|patch] a little more verbose Alex Bennée
2024-10-23 13:14   ` Daniel P. Berrangé
2024-10-23 11:33 ` [PATCH v3 11/18] MAINTAINERS: mention my gdbstub/next tree Alex Bennée
2024-10-23 11:34 ` [PATCH v3 12/18] config/targets: update aarch64_be-linux-user gdb XML list Alex Bennée
2024-10-23 11:34 ` [PATCH v3 13/18] tests/tcg: enable basic testing for aarch64_be-linux-user Alex Bennée
2024-10-23 11:34 ` [PATCH v3 14/18] tests/tcg/aarch64: Use raw strings for regexes in test-mte.py Alex Bennée
2024-10-23 11:34 ` [PATCH v3 15/18] testing: Enhance gdb probe script Alex Bennée
2024-10-23 11:34 ` [PATCH v3 16/18] MAINTAINERS: mention my plugins/next tree Alex Bennée
2024-10-23 11:34 ` [PATCH v3 17/18] contrib/plugins: remove Makefile for contrib/plugins Alex Bennée
2024-10-23 11:34 ` [PATCH v3 18/18] plugins: fix qemu_plugin_reset Alex Bennée

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20241023113406.1284676-8-alex.bennee@linaro.org \
    --to=alex.bennee@linaro.org \
    --cc=berrange@redhat.com \
    --cc=bleal@redhat.com \
    --cc=crosa@redhat.com \
    --cc=devel@lists.libvirt.org \
    --cc=edgar.iglesias@gmail.com \
    --cc=eduardo@habkost.net \
    --cc=erdnaxe@crans.org \
    --cc=iii@linux.ibm.com \
    --cc=jiaxun.yang@flygoat.com \
    --cc=jsnow@redhat.com \
    --cc=kvm@vger.kernel.org \
    --cc=laurent@vivier.eu \
    --cc=ma.mandourr@gmail.com \
    --cc=marcandre.lureau@redhat.com \
    --cc=marcel.apfelbaum@gmail.com \
    --cc=mtosatti@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=peter.maydell@linaro.org \
    --cc=philmd@linaro.org \
    --cc=pierrick.bouvier@linaro.org \
    --cc=qemu-arm@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=richard.henderson@linaro.org \
    --cc=riku.voipio@iki.fi \
    --cc=thuth@redhat.com \
    --cc=wainersm@redhat.com \
    --cc=wangyanan55@huawei.com \
    --cc=zhao1.liu@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).