qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Darren Kenny <darren.kenny@oracle.com>
To: "Bulekov, Alexander" <alxndr@bu.edu>
Cc: Laurent Vivier <lvivier@redhat.com>,
	Thomas Huth <thuth@redhat.com>,
	"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>,
	"bsd@redhat.com" <bsd@redhat.com>,
	"stefanha@redhat.com" <stefanha@redhat.com>,
	"pbonzini@redhat.com" <pbonzini@redhat.com>
Subject: Re: [PATCH v8 19/21] fuzz: add virtio-net fuzz target
Date: Wed, 5 Feb 2020 13:57:40 +0000	[thread overview]
Message-ID: <20200205135740.f6o22uy5ss7bzoj2@starbug-mbp> (raw)
In-Reply-To: <20200129053357.27454-20-alxndr@bu.edu>

On Wed, Jan 29, 2020 at 05:34:27AM +0000, Bulekov, Alexander wrote:
>The virtio-net fuzz target feeds inputs to all three virtio-net
>virtqueues, and uses forking to avoid leaking state between fuzz runs.
>
>Signed-off-by: Alexander Bulekov <alxndr@bu.edu>
>---
> tests/qtest/fuzz/Makefile.include  |   1 +
> tests/qtest/fuzz/virtio_net_fuzz.c | 195 +++++++++++++++++++++++++++++
> 2 files changed, 196 insertions(+)
> create mode 100644 tests/qtest/fuzz/virtio_net_fuzz.c
>
>diff --git a/tests/qtest/fuzz/Makefile.include b/tests/qtest/fuzz/Makefile.include
>index 38b8cdd9f1..77385777ef 100644
>--- a/tests/qtest/fuzz/Makefile.include
>+++ b/tests/qtest/fuzz/Makefile.include
>@@ -8,6 +8,7 @@ fuzz-obj-y += tests/qtest/fuzz/qos_fuzz.o
>
> # Targets
> fuzz-obj-y += tests/qtest/fuzz/i440fx_fuzz.o
>+fuzz-obj-y += tests/qtest/fuzz/virtio_net_fuzz.o
>
> FUZZ_CFLAGS += -I$(SRC_PATH)/tests -I$(SRC_PATH)/tests/qtest
>
>diff --git a/tests/qtest/fuzz/virtio_net_fuzz.c b/tests/qtest/fuzz/virtio_net_fuzz.c
>new file mode 100644
>index 0000000000..e4e3e50865
>--- /dev/null
>+++ b/tests/qtest/fuzz/virtio_net_fuzz.c
>@@ -0,0 +1,195 @@
>+/*
>+ * virtio-net Fuzzing Target
>+ *
>+ * Copyright Red Hat Inc., 2019
>+ *
>+ * Authors:
>+ *  Alexander Bulekov   <alxndr@bu.edu>
>+ *
>+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
>+ * See the COPYING file in the top-level directory.
>+ */
>+
>+#include "qemu/osdep.h"
>+
>+#include "standard-headers/linux/virtio_config.h"
>+#include "tests/qtest/libqtest.h"
>+#include "tests/qtest/libqos/virtio-net.h"
>+#include "fuzz.h"
>+#include "fork_fuzz.h"
>+#include "qos_fuzz.h"
>+
>+
>+#define QVIRTIO_NET_TIMEOUT_US (30 * 1000 * 1000)
>+#define QVIRTIO_RX_VQ 0
>+#define QVIRTIO_TX_VQ 1
>+#define QVIRTIO_CTRL_VQ 2
>+
>+static int sockfds[2];
>+static bool sockfds_initialized;
>+
>+static void virtio_net_fuzz_multi(QTestState *s,
>+        const unsigned char *Data, size_t Size, bool check_used)
>+{
>+    typedef struct vq_action {
>+        uint8_t queue;
>+        uint8_t length;
>+        uint8_t write;
>+        uint8_t next;
>+        uint8_t rx;
>+    } vq_action;
>+
>+    uint32_t free_head = 0;
>+
>+    QGuestAllocator *t_alloc = fuzz_qos_alloc;
>+
>+    QVirtioNet *net_if = fuzz_qos_obj;
>+    QVirtioDevice *dev = net_if->vdev;
>+    QVirtQueue *q;
>+    vq_action vqa;
>+    while (Size >= sizeof(vqa)) {
>+        memcpy(&vqa, Data, sizeof(vqa));
>+        Data += sizeof(vqa);
>+        Size -= sizeof(vqa);
>+
>+        q = net_if->queues[vqa.queue % 3];
>+
>+        vqa.length = vqa.length >= Size ? Size :  vqa.length;
>+
>+        /*
>+         * Only attempt to write incoming packets, when using the socket
>+         * backend. Otherwise, always place the input on a virtqueue.
>+         */
>+        if (vqa.rx && sockfds_initialized) {
>+            write(sockfds[0], Data, vqa.length);
>+        } else {
>+            vqa.rx = 0;
>+            uint64_t req_addr = guest_alloc(t_alloc, vqa.length);
>+            /*
>+             * If checking used ring, ensure that the fuzzer doesn't trigger
>+             * trivial asserion failure on zero-zied buffer
>+             */
>+            qtest_memwrite(s, req_addr, Data, vqa.length);
>+
>+
>+            free_head = qvirtqueue_add(s, q, req_addr, vqa.length,
>+                    vqa.write, vqa.next);
>+            qvirtqueue_add(s, q, req_addr, vqa.length, vqa.write , vqa.next);
>+            qvirtqueue_kick(s, dev, q, free_head);
>+        }
>+
>+        /*
>+         * normally, we could just use qvirtio_wait_used_elem, but since we
>+         * must manually run the main-loop for all the bhs to run, we use
>+         * this hack with flush_events(), to run the main_loop
>+         */
>+        gint64 start_time = g_get_monotonic_time();

NIT: It's a little unclear using a for(;;) when in reality, for most
use-cases, other than virtio_net_fork_fuzz_check_used() it will
never really loop around.

Maybe it would be clearer if the parts that never repeat were
outside the loop? e.g.

    /* Run the main loop */
    qtest_clock_step(s, 100);
    flush_events(s);

    /* Poll the used vring only if we added to the TX or CTRL vq */
    while (check_used && !vqa.rx && q != net_if->queues[QVIRTIO_RX_VQ]) {
        uint32_t got_desc_idx;
        /* Input led to a virtio_error */
        if (dev->bus->get_status(dev) & VIRTIO_CONFIG_S_NEEDS_RESET) {
            break;
        }
        if (dev->bus->get_queue_isr_status(dev, q) &&
                qvirtqueue_get_buf(s, q, &got_desc_idx, NULL)) {
            g_assert_cmpint(got_desc_idx, ==, free_head);
            break;
        }
        g_assert(g_get_monotonic_time() - start_time
                 <= QVIRTIO_NET_TIMEOUT_US);

        /* Run the main loop */
        qtest_clock_step(s, 100);
        flush_events(s);
    }

Unless I'm reading it incorrectly that should do the same thing.

Thanks,

Darren.

>+        for (;;) {
>+            /* Run the main loop */
>+            qtest_clock_step(s, 100);
>+            flush_events(s);
>+            /* Poll the used vring only if we added to the TX or CTRL vq */
>+            if (check_used && !vqa.rx && q != net_if->queues[QVIRTIO_RX_VQ]) {
>+                uint32_t got_desc_idx;
>+                /* Input led to a virtio_error */
>+                if (dev->bus->get_status(dev) & VIRTIO_CONFIG_S_NEEDS_RESET) {
>+                    break;
>+                }
>+                if (dev->bus->get_queue_isr_status(dev, q) &&
>+                        qvirtqueue_get_buf(s, q, &got_desc_idx, NULL)) {
>+                    g_assert_cmpint(got_desc_idx, ==, free_head);
>+                    break;
>+                }
>+                g_assert(g_get_monotonic_time() - start_time
>+                         <= QVIRTIO_NET_TIMEOUT_US);
>+            } else {
>+                break;
>+            }
>+        }
>+        Data += vqa.length;
>+        Size -= vqa.length;
>+    }
>+}
>+
>+static void virtio_net_fork_fuzz(QTestState *s,
>+        const unsigned char *Data, size_t Size)
>+{
>+    if (fork() == 0) {
>+        virtio_net_fuzz_multi(s, Data, Size, false);
>+        flush_events(s);
>+        _Exit(0);
>+    } else {
>+        wait(NULL);
>+    }
>+}
>+
>+static void virtio_net_fork_fuzz_check_used(QTestState *s,
>+        const unsigned char *Data, size_t Size)
>+{
>+    if (fork() == 0) {
>+        virtio_net_fuzz_multi(s, Data, Size, true);
>+        flush_events(s);
>+        _Exit(0);
>+    } else {
>+        wait(NULL);
>+    }
>+}
>+
>+static void virtio_net_pre_fuzz(QTestState *s)
>+{
>+    qos_init_path(s);
>+    counter_shm_init();
>+}
>+
>+static void *virtio_net_test_setup_socket(GString *cmd_line, void *arg)
>+{
>+    int ret = socketpair(PF_UNIX, SOCK_STREAM, 0, sockfds);
>+    g_assert_cmpint(ret, !=, -1);
>+    fcntl(sockfds[0], F_SETFL, O_NONBLOCK);
>+    sockfds_initialized = true;
>+    g_string_append_printf(cmd_line, " -netdev socket,fd=%d,id=hs0 ",
>+                           sockfds[1]);
>+    return arg;
>+}
>+
>+static void *virtio_net_test_setup_user(GString *cmd_line, void *arg)
>+{
>+    g_string_append_printf(cmd_line, " -netdev user,id=hs0 ");
>+    return arg;
>+}
>+
>+static void register_virtio_net_fuzz_targets(void)
>+{
>+    fuzz_add_qos_target(&(FuzzTarget){
>+            .name = "virtio-net-socket",
>+            .description = "Fuzz the virtio-net virtual queues. Fuzz incoming "
>+            "traffic using the socket backend",
>+            .pre_fuzz = &virtio_net_pre_fuzz,
>+            .fuzz = virtio_net_fork_fuzz,},
>+            "virtio-net",
>+            &(QOSGraphTestOptions){.before = virtio_net_test_setup_socket}
>+            );
>+
>+    fuzz_add_qos_target(&(FuzzTarget){
>+            .name = "virtio-net-socket-check-used",
>+            .description = "Fuzz the virtio-net virtual queues. Wait for the "
>+            "descriptors to be used. Timeout may indicate improperly handled "
>+            "input",
>+            .pre_fuzz = &virtio_net_pre_fuzz,
>+            .fuzz = virtio_net_fork_fuzz_check_used,},
>+            "virtio-net",
>+            &(QOSGraphTestOptions){.before = virtio_net_test_setup_socket}
>+            );
>+    fuzz_add_qos_target(&(FuzzTarget){
>+            .name = "virtio-net-slirp",
>+            .description = "Fuzz the virtio-net virtual queues with the slirp "
>+            " backend. Warning: May result in network traffic emitted from the "
>+            " process. Run in an isolated network environment.",

NIT: s/" /"/    (fix double-spaces)

>+            .pre_fuzz = &virtio_net_pre_fuzz,
>+            .fuzz = virtio_net_fork_fuzz,},
>+            "virtio-net",
>+            &(QOSGraphTestOptions){.before = virtio_net_test_setup_user}
>+            );
>+}
>+
>+fuzz_target_init(register_virtio_net_fuzz_targets);
>-- 
>2.23.0
>
>


  parent reply	other threads:[~2020-02-05 13:58 UTC|newest]

Thread overview: 55+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-01-29  5:34 [PATCH v8 00/21] Add virtual device fuzzing support Bulekov, Alexander
2020-01-29  5:34 ` [PATCH v8 01/21] softmmu: split off vl.c:main() into main.c Bulekov, Alexander
2020-01-30 14:39   ` Stefan Hajnoczi
2020-01-30 15:06   ` Alex Bennée
2020-01-30 17:44     ` Alexander Bulekov
2020-01-30 18:41       ` Alex Bennée
2020-02-05 14:02   ` Darren Kenny
2020-01-29  5:34 ` [PATCH v8 02/21] module: check module wasn't already initialized Bulekov, Alexander
2020-01-29  5:34 ` [PATCH v8 03/21] fuzz: add FUZZ_TARGET module type Bulekov, Alexander
2020-02-05 11:28   ` Darren Kenny
2020-01-29  5:34 ` [PATCH v8 04/21] qtest: add qtest_server_send abstraction Bulekov, Alexander
2020-02-05 11:30   ` Darren Kenny
2020-01-29  5:34 ` [PATCH v8 05/21] libqtest: add a layer of abstraction to send/recv Bulekov, Alexander
2020-02-05 11:32   ` Darren Kenny
2020-01-29  5:34 ` [PATCH v8 06/21] libqtest: make bufwrite rely on the TransportOps Bulekov, Alexander
2020-02-05 11:36   ` Darren Kenny
2020-01-29  5:34 ` [PATCH v8 07/21] qtest: add in-process incoming command handler Bulekov, Alexander
2020-02-05 11:39   ` Darren Kenny
2020-01-29  5:34 ` [PATCH v8 08/21] libqos: rename i2c_send and i2c_recv Bulekov, Alexander
2020-02-05 11:40   ` Darren Kenny
2020-01-29  5:34 ` [PATCH v8 09/21] libqos: split qos-test and libqos makefile vars Bulekov, Alexander
2020-01-29  5:34 ` [PATCH v8 10/21] libqos: move useful qos-test funcs to qos_external Bulekov, Alexander
2020-01-29 10:03   ` Paolo Bonzini
2020-02-05 13:34   ` Darren Kenny
2020-01-29  5:34 ` [PATCH v8 11/21] fuzz: add fuzzer skeleton Bulekov, Alexander
2020-02-05 11:55   ` Darren Kenny
2020-01-29  5:34 ` [PATCH v8 12/21] exec: keep ram block across fork when using qtest Bulekov, Alexander
2020-02-05 13:00   ` Darren Kenny
2020-01-29  5:34 ` [PATCH v8 13/21] main: keep rcu_atfork callback enabled for qtest Bulekov, Alexander
2020-01-30 14:42   ` Stefan Hajnoczi
2020-01-30 17:24     ` Paolo Bonzini
2020-01-30 17:42       ` Alexander Bulekov
2020-01-30 18:14         ` Paolo Bonzini
2020-02-05 13:58   ` Darren Kenny
2020-06-18  7:34   ` Thomas Huth
2020-06-18 15:08     ` Alexander Bulekov
2020-01-29  5:34 ` [PATCH v8 14/21] fuzz: support for fork-based fuzzing Bulekov, Alexander
2020-01-29  5:34 ` [PATCH v8 16/21] fuzz: add target/fuzz makefile rules Bulekov, Alexander
2020-01-29  5:34 ` [PATCH v8 15/21] fuzz: add support for qos-assisted fuzz targets Bulekov, Alexander
2020-02-05 13:18   ` Darren Kenny
2020-02-11 16:19     ` Alexander Bulekov
2020-01-29  5:34 ` [PATCH v8 17/21] fuzz: add configure flag --enable-fuzzing Bulekov, Alexander
2020-02-05 13:19   ` Darren Kenny
2020-01-29  5:34 ` [PATCH v8 18/21] fuzz: add i440fx fuzz targets Bulekov, Alexander
2020-02-05 13:26   ` Darren Kenny
2020-02-06 14:47     ` Alexander Bulekov
2020-01-29  5:34 ` [PATCH v8 19/21] fuzz: add virtio-net fuzz target Bulekov, Alexander
2020-01-30 14:43   ` Stefan Hajnoczi
2020-02-05 13:57   ` Darren Kenny [this message]
2020-02-06 14:51     ` Alexander Bulekov
2020-01-29  5:34 ` [PATCH v8 20/21] fuzz: add virtio-scsi " Bulekov, Alexander
2020-02-05 13:31   ` Darren Kenny
2020-01-29  5:34 ` [PATCH v8 21/21] fuzz: add documentation to docs/devel/ Bulekov, Alexander
2020-02-05 13:33   ` Darren Kenny
2020-01-30 14:44 ` [PATCH v8 00/21] Add virtual device fuzzing support Stefan Hajnoczi

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=20200205135740.f6o22uy5ss7bzoj2@starbug-mbp \
    --to=darren.kenny@oracle.com \
    --cc=alxndr@bu.edu \
    --cc=bsd@redhat.com \
    --cc=lvivier@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@redhat.com \
    --cc=thuth@redhat.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).