From: "Michael S. Tsirkin" <mst@redhat.com>
To: qemu-devel@nongnu.org
Cc: "Peter Maydell" <peter.maydell@linaro.org>,
"Carlos López" <clopez@suse.de>
Subject: [PULL 72/73] virtio: fix reachable assertion due to stale value of cached region size
Date: Tue, 7 Mar 2023 20:14:25 -0500 [thread overview]
Message-ID: <bbc1c327d7974261c61566cdb950cc5fa0196b41.1678237635.git.mst@redhat.com> (raw)
In-Reply-To: <cover.1678237635.git.mst@redhat.com>
From: Carlos López <clopez@suse.de>
In virtqueue_{split,packed}_get_avail_bytes() descriptors are read
in a loop via MemoryRegionCache regions and calls to
vring_{split,packed}_desc_read() - these take a region cache and the
index of the descriptor to be read.
For direct descriptors we use a cache provided by the caller, whose
size matches that of the virtqueue vring. We limit the number of
descriptors we can read by the size of that vring:
max = vq->vring.num;
...
MemoryRegionCache *desc_cache = &caches->desc;
For indirect descriptors, we initialize a new cache and limit the
number of descriptors by the size of the intermediate descriptor:
len = address_space_cache_init(&indirect_desc_cache,
vdev->dma_as,
desc.addr, desc.len, false);
desc_cache = &indirect_desc_cache;
...
max = desc.len / sizeof(VRingDesc);
However, the first initialization of `max` is done outside the loop
where we process guest descriptors, while the second one is done
inside. This means that a sequence of an indirect descriptor followed
by a direct one will leave a stale value in `max`. If the second
descriptor's `next` field is smaller than the stale value, but
greater than the size of the virtqueue ring (and thus the cached
region), a failed assertion will be triggered in
address_space_read_cached() down the call chain.
Fix this by initializing `max` inside the loop in both functions.
Fixes: 9796d0ac8fb0 ("virtio: use address_space_map/unmap to access descriptors")
Signed-off-by: Carlos López <clopez@suse.de>
Message-Id: <20230302100358.3613-1-clopez@suse.de>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
---
hw/virtio/virtio.c | 11 +++++------
1 file changed, 5 insertions(+), 6 deletions(-)
diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c
index f35178f5fc..98c4819fcc 100644
--- a/hw/virtio/virtio.c
+++ b/hw/virtio/virtio.c
@@ -1069,7 +1069,7 @@ static void virtqueue_split_get_avail_bytes(VirtQueue *vq,
VRingMemoryRegionCaches *caches)
{
VirtIODevice *vdev = vq->vdev;
- unsigned int max, idx;
+ unsigned int idx;
unsigned int total_bufs, in_total, out_total;
MemoryRegionCache indirect_desc_cache = MEMORY_REGION_CACHE_INVALID;
int64_t len = 0;
@@ -1078,13 +1078,12 @@ static void virtqueue_split_get_avail_bytes(VirtQueue *vq,
idx = vq->last_avail_idx;
total_bufs = in_total = out_total = 0;
- max = vq->vring.num;
-
while ((rc = virtqueue_num_heads(vq, idx)) > 0) {
MemoryRegionCache *desc_cache = &caches->desc;
unsigned int num_bufs;
VRingDesc desc;
unsigned int i;
+ unsigned int max = vq->vring.num;
num_bufs = total_bufs;
@@ -1206,7 +1205,7 @@ static void virtqueue_packed_get_avail_bytes(VirtQueue *vq,
VRingMemoryRegionCaches *caches)
{
VirtIODevice *vdev = vq->vdev;
- unsigned int max, idx;
+ unsigned int idx;
unsigned int total_bufs, in_total, out_total;
MemoryRegionCache *desc_cache;
MemoryRegionCache indirect_desc_cache = MEMORY_REGION_CACHE_INVALID;
@@ -1218,14 +1217,14 @@ static void virtqueue_packed_get_avail_bytes(VirtQueue *vq,
wrap_counter = vq->last_avail_wrap_counter;
total_bufs = in_total = out_total = 0;
- max = vq->vring.num;
-
for (;;) {
unsigned int num_bufs = total_bufs;
unsigned int i = idx;
int rc;
+ unsigned int max = vq->vring.num;
desc_cache = &caches->desc;
+
vring_packed_desc_read(vdev, &desc, desc_cache, idx, true);
if (!is_desc_avail(desc.flags, wrap_counter)) {
break;
--
MST
next prev parent reply other threads:[~2023-03-08 1:20 UTC|newest]
Thread overview: 99+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-03-08 1:10 [PULL 00/73] virtio,pc,pci: features, fixes Michael S. Tsirkin
2023-03-08 1:10 ` [PULL 01/73] cryptodev: Introduce cryptodev.json Michael S. Tsirkin
2023-03-08 1:10 ` [PULL 02/73] cryptodev: Remove 'name' & 'model' fields Michael S. Tsirkin
2023-03-08 1:10 ` [PULL 03/73] cryptodev: Introduce cryptodev alg type in QAPI Michael S. Tsirkin
2023-03-08 1:11 ` [PULL 04/73] cryptodev: Introduce server " Michael S. Tsirkin
2023-03-08 1:11 ` [PULL 05/73] cryptodev: Introduce 'query-cryptodev' QMP command Michael S. Tsirkin
2023-03-08 1:11 ` [PULL 06/73] cryptodev-builtin: Detect akcipher capability Michael S. Tsirkin
2023-03-08 1:11 ` [PULL 07/73] hmp: add cryptodev info command Michael S. Tsirkin
2023-03-08 1:11 ` [PULL 08/73] cryptodev: Use CryptoDevBackendOpInfo for operation Michael S. Tsirkin
2023-03-08 1:11 ` [PULL 09/73] cryptodev: Account statistics Michael S. Tsirkin
2023-03-08 1:11 ` [PULL 10/73] cryptodev: support QoS Michael S. Tsirkin
2023-03-08 1:11 ` [PULL 11/73] cryptodev: Support query-stats QMP command Michael S. Tsirkin
2023-05-02 17:03 ` Peter Maydell
2023-05-03 4:19 ` zhenwei pi
2023-03-08 1:11 ` [PULL 12/73] MAINTAINERS: add myself as the maintainer for cryptodev Michael S. Tsirkin
2023-03-08 1:11 ` [PULL 13/73] vdpa net: move iova tree creation from init to start Michael S. Tsirkin
2023-03-08 1:11 ` [PULL 14/73] vdpa: Remember last call fd set Michael S. Tsirkin
2023-03-08 1:11 ` [PULL 15/73] vdpa: Negotiate _F_SUSPEND feature Michael S. Tsirkin
2023-03-08 1:11 ` [PULL 16/73] vdpa: rewind at get_base, not set_base Michael S. Tsirkin
2023-03-08 1:11 ` [PULL 17/73] vdpa: add vhost_vdpa->suspended parameter Michael S. Tsirkin
2023-03-08 1:11 ` [PULL 18/73] vdpa: add vhost_vdpa_suspend Michael S. Tsirkin
2023-03-08 1:11 ` [PULL 19/73] vdpa: move vhost reset after get vring base Michael S. Tsirkin
2023-03-08 1:11 ` [PULL 20/73] vdpa: add vdpa net migration state notifier Michael S. Tsirkin
2023-03-08 1:11 ` [PULL 21/73] vdpa: disable RAM block discard only for the first device Michael S. Tsirkin
2023-03-08 1:11 ` [PULL 22/73] vdpa net: block migration if the device has CVQ Michael S. Tsirkin
2023-03-08 1:11 ` [PULL 23/73] vdpa: block migration if device has unsupported features Michael S. Tsirkin
2023-03-08 1:12 ` [PULL 24/73] vdpa: block migration if SVQ does not admit a feature Michael S. Tsirkin
2023-03-08 1:12 ` [PULL 25/73] vdpa net: allow VHOST_F_LOG_ALL Michael S. Tsirkin
2023-03-08 1:12 ` [PULL 26/73] vdpa: return VHOST_F_LOG_ALL in vhost-vdpa devices Michael S. Tsirkin
2023-03-08 1:12 ` [PULL 27/73] Revert "tests/qtest: Check for devices in bios-tables-test" Michael S. Tsirkin
2023-03-08 1:12 ` [PULL 28/73] tests: acpi: whitelist new q35.noacpihp test and pc.hpbrroot Michael S. Tsirkin
2023-03-08 1:12 ` [PULL 29/73] tests: acpi: add test_acpi_q35_tcg_no_acpi_hotplug test and extend test_acpi_piix4_no_acpi_pci_hotplug Michael S. Tsirkin
2023-03-08 1:12 ` [PULL 30/73] tests: acpi: update expected blobs Michael S. Tsirkin
2023-03-13 10:57 ` Philippe Mathieu-Daudé
2023-03-13 12:59 ` Michael S. Tsirkin
2023-03-08 1:12 ` [PULL 31/73] tests: acpi: whitelist q35/DSDT.multi-bridge before extending testcase Michael S. Tsirkin
2023-03-08 1:12 ` [PULL 32/73] tests: acpi: extend multi-bridge case with case 'root-port,id=HOHP,hotplug=off root-port,bus=NOHP' Michael S. Tsirkin
2023-03-08 1:12 ` [PULL 33/73] x86: pcihp: fix missing PCNT callchain when intermediate root-port has 'hotplug=off' set Michael S. Tsirkin
2023-03-08 1:12 ` [PULL 34/73] tests: acpi: whitelist pc/DSDT.hpbrroot and pc/DSDT.hpbridge tests Michael S. Tsirkin
2023-03-08 1:12 ` [PULL 35/73] x86: pcihp: fix missing bridge AML when intermediate root-port has 'hotplug=off' set Michael S. Tsirkin
2023-03-08 1:12 ` [PULL 36/73] tests: acpi: update expected blobs Michael S. Tsirkin
2023-03-08 1:12 ` [PULL 37/73] pcihp: piix4: do not redirect hotplug controller to piix4 when ACPI hotplug is disabled Michael S. Tsirkin
2023-03-08 1:12 ` [PULL 38/73] pci: fix 'hotplugglable' property behavior Michael S. Tsirkin
2023-03-08 1:12 ` [PULL 39/73] tests: acpi: whitelist DSDT blobs before isolating PCI _DSM func 0 prolog Michael S. Tsirkin
2023-03-08 1:12 ` [PULL 40/73] pcihp: move PCI _DSM function 0 prolog into separate function Michael S. Tsirkin
2023-03-08 1:12 ` [PULL 41/73] tests: acpi: update expected blobs Michael S. Tsirkin
2023-03-08 1:12 ` [PULL 42/73] tests: acpi: whitelist DSDT before adding EDSM method Michael S. Tsirkin
2023-03-08 1:12 ` [PULL 43/73] acpi: pci: add EDSM method to DSDT Michael S. Tsirkin
2023-03-08 1:13 ` [PULL 44/73] tests: acpi: update expected blobs Michael S. Tsirkin
2023-03-08 1:13 ` [PULL 45/73] tests: acpi: whitelist DSDT before adding device with acpi-index to testcases Michael S. Tsirkin
2023-03-08 1:13 ` [PULL 46/73] tests: acpi: add device with acpi-index on non-hotpluggble bus Michael S. Tsirkin
2023-03-08 1:13 ` [PULL 47/73] acpi: pci: support acpi-index for non-hotpluggable devices Michael S. Tsirkin
2023-03-08 1:13 ` [PULL 48/73] tests: acpi: update expected blobs Michael S. Tsirkin
2023-03-08 1:13 ` [PULL 49/73] tests: acpi: whitelist DSDT before exposing non zero functions Michael S. Tsirkin
2023-03-08 1:13 ` [PULL 50/73] acpi: pci: describe all functions on populated slots Michael S. Tsirkin
2023-03-08 1:13 ` [PULL 51/73] tests: acpi: update expected blobs Michael S. Tsirkin
2023-03-08 1:13 ` [PULL 52/73] tests: acpi: whitelist DSDT before adding non-0 function device with acpi-index to testcases Michael S. Tsirkin
2023-03-08 1:13 ` [PULL 53/73] tests: acpi: add non zero function device with acpi-index on non-hotpluggble bus Michael S. Tsirkin
2023-03-08 1:13 ` [PULL 54/73] tests: acpi: update expected blobs Michael S. Tsirkin
2023-03-08 1:13 ` [PULL 55/73] pci: move acpi-index uniqueness check to generic PCI device code Michael S. Tsirkin
2023-03-08 1:13 ` [PULL 56/73] acpi: pci: drop BSEL usage when deciding that device isn't hotpluggable Michael S. Tsirkin
2023-03-08 1:13 ` [PULL 57/73] acpi: pci: move BSEL into build_append_pcihp_slots() Michael S. Tsirkin
2023-03-08 1:13 ` [PULL 58/73] acpi: pci: move out ACPI PCI hotplug generator from generic slot generator build_append_pci_bus_devices() Michael S. Tsirkin
2023-03-08 1:13 ` [PULL 59/73] pcihp: move fields enabling hotplug into AcpiPciHpState Michael S. Tsirkin
2023-03-08 1:13 ` [PULL 60/73] pcihp: add ACPI PCI hotplug specific is_hotpluggable_bus() callback Michael S. Tsirkin
2023-03-08 1:13 ` [PULL 61/73] hw/pci/aer: Implement PCI_ERR_UNCOR_MASK register Michael S. Tsirkin
2023-04-26 0:42 ` Peter Xu
2023-04-26 6:12 ` Michael S. Tsirkin
2023-04-26 7:19 ` Juan Quintela
2023-05-03 0:32 ` Leonardo Brás
2023-05-03 4:08 ` Michael S. Tsirkin
2023-05-03 9:31 ` Jonathan Cameron via
2023-03-08 1:13 ` [PULL 62/73] hw/pci/aer: Add missing routing for AER errors Michael S. Tsirkin
2023-03-08 1:13 ` [PULL 63/73] hw/pci-bridge/cxl_root_port: Wire up AER Michael S. Tsirkin
2023-03-08 1:14 ` [PULL 64/73] hw/pci-bridge/cxl_root_port: Wire up MSI Michael S. Tsirkin
2023-03-08 1:14 ` [PULL 65/73] hw/mem/cxl-type3: Add AER extended capability Michael S. Tsirkin
2023-03-08 1:14 ` [PULL 66/73] hw/cxl: Fix endian issues in CXL RAS capability defaults / masks Michael S. Tsirkin
2023-03-08 1:14 ` [PULL 67/73] hw/pci/aer: Make PCIE AER error injection facility available for other emulation to use Michael S. Tsirkin
2023-03-08 1:14 ` [PULL 68/73] hw/mem/cxl_type3: Add CXL RAS Error Injection Support Michael S. Tsirkin
2023-03-08 1:14 ` [PULL 69/73] hw/pci: Add pcie_count_ds_port() and pcie_find_port_first() helpers Michael S. Tsirkin
2023-03-08 1:14 ` [PULL 70/73] hw/pxb-cxl: Support passthrough HDM Decoders unless overridden Michael S. Tsirkin
2023-04-11 10:26 ` Peter Maydell
2023-04-17 11:22 ` Thomas Huth
2023-04-17 11:29 ` Michael S. Tsirkin
2023-04-17 13:04 ` Thomas Huth
2023-04-19 13:43 ` Jonathan Cameron via
2023-04-19 13:57 ` Jonathan Cameron via
2023-04-19 14:49 ` Jonathan Cameron via
2023-04-19 16:25 ` Peter Maydell
2023-04-19 17:18 ` Jonathan Cameron via
2023-03-08 1:14 ` [PULL 71/73] hw/virtio/vhost-user: avoid using unitialized errp Michael S. Tsirkin
2023-03-08 1:14 ` Michael S. Tsirkin [this message]
2023-03-08 1:14 ` [PULL 73/73] virtio: refresh vring region cache after updating a virtqueue size Michael S. Tsirkin
2023-03-09 14:48 ` Michael S. Tsirkin
2023-03-09 14:47 ` [PULL 00/73] virtio,pc,pci: features, fixes Michael S. Tsirkin
2023-03-10 17:32 ` Peter Maydell
2023-03-10 22:20 ` Philippe Mathieu-Daudé
2023-03-11 19:22 ` Michael S. Tsirkin
2023-03-13 8:03 ` Philippe Mathieu-Daudé
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=bbc1c327d7974261c61566cdb950cc5fa0196b41.1678237635.git.mst@redhat.com \
--to=mst@redhat.com \
--cc=clopez@suse.de \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
/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).