Age | Commit message (Collapse) | Author | Files | Lines |
|
With heavy traffic, tx code path may crash due to memory corruption
Thread 5 "vpp_wk_2" received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7fff3995c700 (LWP 2505)]
0x00007ffff73675e8 in vhost_user_if_input (vm=0x7fffb5f5bf9c,
vum=0x7ffff7882a40 <vhost_user_main>, vui=0x7fffb65570c4, qid=0,
node=0x7fffb6577dac, mode=VNET_HW_INTERFACE_RX_MODE_POLLING)
at /home/sluong/vpp-master/vpp/build-data/../src/vnet/devices/virtio/vhost-user.c:1610
1610 bi_current = (vum->cpus[thread_index].rx_buffers)
[vum->cpus[thread_index].rx_buffers_len];
(gdb) p vum->cpus[thread_index].rx_buffers_len
$2 = 793212607
(gdb)
Apparently, some code accidentally wrote the bad value in rx_buffers_len.
rx_buffers_len should never be greater than 1024 since that is how many buffers
we request each time.
After debugging many hours, I discovered that the memory corruption happens
in the tx code path right here on line 2176.
{
vhost_copy_t *cpy = &vum->cpus[thread_index].copy[copy_len];
copy_len++;
cpy->len = bytes_left;
cpy->len = (cpy->len > buffer_len) ? buffer_len : cpy->len;
cpy->dst = buffer_map_addr;
cpy->src = (uword) vlib_buffer_get_current (current_b0) +
current_b0->current_length - bytes_left;
(gdb) p cpy
$3 = (vhost_copy_t *) 0x7fffb554077c
(gdb) p copy_len
$4 = 1025
(gdb) p &vum->cpus[3].rx_buffers_len
$8 = (u32 *) 0x7fffb5540784
copy_len is picking up the index entry 1024 before it was incremented. copy array has only
1024 members (0 - 1023 are valid).
The assignment here in cpy surely causes memory corruption. It is only discovered later
when the memory location that it corrupted is used.
The condition for the crash is to transmit jumbo frames under heavy volume. Since ring
size is 1024, with one packet taking up one index for frame size (less 2048), it does
not cause overflow. With jumbo frames, it requires multiple indices for one packet,
it can cause the overflow under heavy traffic.
The fix is to do copy out when we have 1000 entries in the array to avoid
overflow.
Change-Id: Iefbc739b8e80470f1cf13123113f8331ffcd0eb2
Signed-off-by: Steven <sluong@cisco.com>
(cherry picked from commit aa5df48cb233b377b5910694e2440a16e5973864)
|
|
A bug was reported where a jumbo packet would stay in vhost
queue forever or until a large enough number of other packets
arrived in the queue too.
This is due to a bug in vhost input node buffer allocation.
The fix is to make sure that vhost always allocates at least
enough buffers for one single big packet. '40' is used to
account for 65kB frames.
Change-Id: I1d293028854165083e30cd798fab9d4140230b78
Signed-off-by: Pierre Pfister <ppfister@cisco.com>
|
|
On host interface if a VLAN tagged packet is received, linux kernel removes
the VLAN header from packet byte stream and adds metadata in tpacket2_hdr.
This patch explicitely checks for the presense of VLAN metadata and adds it
in VPP packet.
Change-Id: I0ba35c1e98dbc008ce18d032f22f2717d610c1aa
Signed-off-by: Akshaya N <akshaya@rtbrick.com>
|
|
When changing the admin state of a vhost-user interface, do not put it
in link-up mode if the interface is not actually ready.
Change-Id: Idbc631a7126efa79d199909f9e7656d21bd412ca
Signed-off-by: Yoann Desmouceaux <ydesmouc@cisco.com>
|
|
Change-Id: Iff33694fc42cc3bcc73cf1372339053a6365039c
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
This will allow us to use this code in client libraries without vlib.
Change-Id: I8557b752496841ba588aa36b6082cbe2cd1867fe
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: Id79d2c2be7a98e15416a537c890a8f2dd6d4464d
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
When I type in set interface rx-mode with a nonexistent queue, I got a crash with the following
traceback. It looks like the vm is NULL when vlib_node_get_runtime is called.
DBGvpp# sh int rx
Thread 0 (vpp_main):
node dpdk-input:
TenGigabitEthernet5/0/0 queue 0 (polling)
TenGigabitEthernet5/0/1 queue 0 (polling)
TenGigabitEthernet7/0/0 queue 0 (polling)
TenGigabitEthernet7/0/1 queue 0 (polling)
node vhost-user-input:
VirtualEthernet0/0/2 queue 0 (adaptive)
DBGvpp# set interface rx-mode VirtualEthernet0/0/2 queue 1 polling
Thread 1 "vpp_main" received signal SIGSEGV, Segmentation fault.
0x00007ffff6d4e0bc in vlib_node_get_runtime (vm=0x0, node_index=125)
at /home/sluong/vpp/build-data/../src/vlib/node_funcs.h:92
92 vlib_node_t *n = vec_elt (nm->nodes, node_index);
(gdb) where
at /home/sluong/vpp/build-data/../src/vlib/node_funcs.h:92
at /home/sluong/vpp/build-data/../src/vlib/node_funcs.h:112
vnm=0x6f0fa0 <vnet_main>, hw_if_index=7, queue_id=1, mode=0x7fffb62099e8)
at /home/sluong/vpp/build-data/../src/vnet/devices/devices.c:307
hw_if_index=7, queue_id=1, mode=VNET_HW_INTERFACE_RX_MODE_POLLING)
at /home/sluong/vpp/build-data/../src/vnet/interface_cli.c:1192
vm=0x7ffff7b9d440 <vlib_global_main>, input=0x7fffb6209ef0,
cmd=0x7fffb61d5d14)
at /home/sluong/vpp/build-data/../src/vnet/interface_cli.c:1288
vm=0x7ffff7b9d440 <vlib_global_main>,
cm=0x7ffff7b9d630 <vlib_global_main+496>, input=0x7fffb6209ef0,
parent_command_index=18)
at /home/sluong/vpp/build-data/../src/vlib/cli.c:588
vm=0x7ffff7b9d440 <vlib_global_main>,
cm=0x7ffff7b9d630 <vlib_global_main+496>, input=0x7fffb6209ef0,
parent_command_index=12)
The fix is to add a check for vec_len(hw->input_node_thread_index_by_queue)
and vec_len (hw->rx_mode_by_queue) to reject the command if the queue_id is
out of bound. While at it, I notice inputting queue_id=-1 is being interpreted
as all queues. An easy fix is to not overload the queue_id variable with -1 to
mean something else.
Change-Id: Id70ec3e7d06ccc67635e6d28ef53420bdac4a988
Signed-off-by: Steven <sluong@cisco.com>
|
|
According to the spec, supporting interrupt mode from the driver is optional,
not a must. When interrupt mode is configured on the interface, we should
check to make sure that the driver didn't opt out for the kickfd support and
reject the configuration if it did.
Change-Id: I7d3dbaddde65458e1a6a802754a3768ae8685a0e
Signed-off-by: Steven <sluong@cisco.com>
|
|
In the data path, we grab qsz from vhost_user_vring_t to compute
qsz_mask and store it in a stack variable to use on many occasions.
We never use qsz for any meaningful purpose. It is more useful to
cache qsz_mask in vhost_user_vring_t to avoid the needless computation
in the data path.
Change-Id: Idf4d94a9754d5c75c899f1f4f59602275b9904a6
Signed-off-by: Steven <sluong@cisco.com>
|
|
create/delete/modify vhost_user APIs no longer support the operation mode
(polling/interrupt/adaptive). They are now done via the generic interface.
Change-Id: I9e9bd503f9b56c953ecd2b271b3e2007da20c72a
Signed-off-by: Steven <sluong@cisco.com>
|
|
Change-Id: I0124fa264f7f390fc7cd9722da59be03116831c5
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
Change-Id: I06a10a4291e61aec3f1396d2514ed6fe3901897a
Signed-off-by: Neale Ranns <neale.ranns@cisco.com>
Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
|
|
Change-Id: Idb97e573961b3bc2acdeef77582314590795f8c3
Signed-off-by: Pavel Kotucek <pkotucek@cisco.com>
|
|
(VPP-916)
The syntax for debug vhost-user is
debug vhost-user <on | off>
However, currently the code does not reject the invalid command such as below
debug vhost-user
debug vhost-user on blah
debug vhost-user off blah
The fix is to enforece the correct syntax and reject the command when invalid
option is entered.
Change-Id: I1a04ae8ddb6dd299aa6d15b043362964e685ddde
Signed-off-by: Steven <sluong@cisco.com>
(cherry picked from commit 6a4de2764d9e6cadf36af824dddb3f33c2d6dc7e)
|
|
A call to 'clib_smp_swap (&((dq)->interrupt_pending), 0)' was creating
a SIGBUS.
Instead of making dq->interrupt_pending aligned on 64bits, we reduce the size
from uword (u64) to u32, as the number of pending interrupts will never
go above max of u32.
Change-Id: Ifa5a6d3b7adee222329a671be01305cf50853b33
Signed-off-by: Christophe Fontaine <christophe.fontaine@enea.com>
|
|
If interface is down and queues are not configured then we are not able
to change rx-mode. This change introducess default mode which is stored
per interface and applied if driver wants.
Change-Id: I70149c21c1530eafc148d5e4aa03fbee53dec62f
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I25ccf8260dbe7e1550aee3904a688fc135ce1f03
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Add runtime debug vhost-user on | off CLI to facilitate troubleshooting.
This feature is needed to avoid recompiling the code to debug vhost issues.
The debugging messages should not be on the data path to avoid performance
hit.
Change-Id: I4c40f65dbb222557cba3fb8706fa3b7b62eec95f
Signed-off-by: Steven <sluong@cisco.com>
|
|
Fix coverity error associated with fd.
Change-Id: I0648aebaf356308bc03cc7217922479bfc4e22f7
Signed-off-by: Ray Kinsella <ray.kinsella@intel.com>
|
|
and add adaptive mode support to receive queue
- Migrate vhost to use device infra which does the interface/queue to worker
thread assignment.
- Retire vhost thread CLI and corresponding code which assigns interface/queue
to worker thread. set interface placement should be used instead to customize
the interface/queue to worker thread assignment.
- Retire vhost interrupt/polling option when creating vhost-user interface.
Instead, set interface rx-mode should be used.
- Add code in vnet_device_input_unassign_thread to change the node state
to interrupt if the last polling interface has left the worker thread for the
device of the corresponding interface/queue.
- Add adaptive mode support. The node state is set to interrupt initially.
When the scheduler detects a burst of traffic, it switches the input node to
polling. Then we inform the device that we don't need interrupt notification.
When the traffic subsides, the scheduler switches the input node back to
interrupt. Then we immediately tell the driver that we want interrupt
notification again.
- Remove some duplicate code in vlib/main.c
Change-Id: Id19bb1b9e50e6521c6464f470f5825c26924d3a8
Signed-off-by: Steven <sluong@cisco.com>
|
|
When processing a vring descriptor which is outside of mmap, we disable
the interface and spit a message to shut/no shut the interface. This is
not practical as application using vhost cannot constantly checking the
logs and do the recovery. The proposed fix is to log an error, like
other errors that we encounter.
The other bug is buffer leak in the function rewind. At the end of the
while loop when b_current != b_head, we still have to give back 1 more
buffer or add 1 to rx_buffers_len.
Change-Id: I68c0b24f070e644cd8878f42272a7b518f14393f
Signed-off-by: Steven <sluong@cisco.com>
|
|
Added support to the interfaces mac address.
Resolved an fd leak when the interface is a bridge.
Change-Id: I6608c51b11a50bd0ae4aabe0dc5788c4301b5a1e
Signed-off-by: Ray Kinsella <ray.kinsella@intel.com>
|
|
When the descriptor is chained via multiple parts, vhost is supposed
to reassemble the different parts to form a packet prior to passing
the packet to the next input node. However, bad packet was seen, having
bad ethertype, source, and destination mac addresses.
The problem was due to the destination pointer not being incremented as
each chain is processed. THe result was the first chain is copied to the
beginning of the buffer, the next chain is copied, then the last chain
is also copied to the beginning of the buffer. As a result, the ethertype,
source and destination mac, etc, are being overwritten by the very last
chain of the descriptor.
Change-Id: I78f9a91de68c85574047912576dcc311d7597e21
Signed-off-by: Steven <sluong@cisco.com>
|
|
Added support to the af_packet device to change the MTU size.
Change-Id: I9c9e1e17323721f3efccf70a10b753e12eef94d5
Signed-off-by: Ray Kinsella <ray.kinsella@intel.com>
|
|
- When the interface rx-mode is changed via CLI, the corresponding device
may want to know about it and to reset the driver. This patch is to add
the callback.
- In the function vnet_hw_interface_set_rx_mode, it appears it is missing
a line
hw->rx_mode_by_queue[queue_id] = mode
because the function is checking if the new mode is the same as
hw->rx_mode_by_queue which is initialized to POLLING. So if the function is
called to change the mode to interrupt, it just returns without doing
anything. This is the check that I am talking about in the same function.
if (hw->rx_mode_by_queue[queue_id] == mode)
return 0;
Change-Id: Iaca2651c43e0ae3fda6fd8dc128e247b0851cc65
Signed-off-by: Steven <sluong@cisco.com>
|
|
Change-Id: I9664214652229b663c3e3ba7406b4ede96bfb123
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: Ib506c3e9d66170f29e3266ad6dc4d32b829befba
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Setting the interface state in VPP on an af_packet device, was not being
reflected on the host. This implied the user had to set the device state
in VPP and then on the host, in order to put the interface into an 'up'
state. This changes makes the device state consisent in VPP and the host.
Change-Id: I6dc6aee79503e04576683db937b861337a2b375b
Signed-off-by: Ray Kinsella <ray.kinsella@intel.com>
|
|
When creating or modifying a vhost interface, verify if the path name
already existed and reject the command.
Change-Id: I8b2d33b77c847f774492874f7d194fa72c488479
Signed-off-by: Steven <sluong@cisco.com>
|
|
I had a bug where a requested size of 1G was resulting in
an aligned size of '1G + 2M', resulting in an OOM error.
Previous code was adding one huge page size
when memory is already aligned.
Change-Id: Idd3aa0e9b893fb3efccba6ae1c7161e26d3f9456
Signed-off-by: Pierre Pfister <ppfister@cisco.com>
|
|
Patch 6347 removed the socket file when the interface is deleted and
when VPP process is exitting. The CLI for deleting the interface has
builtin vlib_worker_thread_barrier_sync to prevent the worker
threads from running. Unfortunately, the CLI quit does not have the
builtin vlib_worker_thread_barrier_sync. As a result, it may cause the
worker thread to crash.
The fix is to add the vlib_worker_thread_barrier_sync in vhost_user_exit.
Change-Id: I1eff81170e131098f1799662f0ab48d6fca3def7
Signed-off-by: Steven <sluong@cisco.com>
|
|
Change-Id: I72298aaae7d172082ece3a8edea4217c11b28d79
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
- Unlink the file created for the socket when deleting vhost interface
if we are the server mode.
- Remove all vhost interfaces when VPP process is exitting.
Change-Id: Id9b676cd027bbd67b473bbd01901d1ecc4d8e6cb
Signed-off-by: Steven <sluong@cisco.com>
|
|
The command set interface placement is supposed to remove the existing
interface placement and add a new interface placement based on the given
options. My quick test for the CLI shows that the old interface placement
continues to exist. But the new interface placement is also added.
The bug exists in vnet_device_input_unassign_thread which checks
the old thread index is the same as the passed thread index and skips the
deletion if they are the same.
The fix is to remove the check which is not supposed to be there.
Change-Id: Ib055721fad47513949a03b3cb6dc292bd19fd1e8
Signed-off-by: Steven <sluong@cisco.com>
|
|
- Add cpu index to the vring structure for quick lookup
- Reduce the code that needs to be protected by
vlib_worker_thread_barrier_sync
- Set minimum timer no less than 1 ms
Change-Id: Iafef4bf6879a8efb350abf4e0f517e38f7ff7a8b
Signed-off-by: Steven <sluong@cisco.com>
|
|
Change-Id: I23b588eb56a3f5690158449a1f9bc8053cd3d251
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
1 - interface-DPO
Used in the Data-plane to change a packet's input interface
2 - MPLS multicast FIB entry
Same as a unicast entry but it links to a replicate not a load-balance DPO
3 - Multicast MPLS tunnel
Update MPLS tunnels to use a FIB path-list to describe the endpoint[s]. Use the path-list to generate the forwarding chain (DPOs) to link to .
4 - Resolve a path via a local label (of an mLDP LSP)
For IP multicast entries to use an LSP in the replication list, we need to decribe the 'resolve-via-label' where the label is that of a multicast LSP.
5 - MPLS disposition path sets RPF-ID
For a interface-less LSP (i.e. mLDP not RSVP-TE) at the tail of the LSP we still need to perform an RPF check. An MPLS disposition DPO performs the MPLS pop validation checks and sets the RPF-ID in the packet.
6 - RPF check with per-entry RPF-ID
An RPF-ID is used instead of a real interface SW if index in the case the IP traffic arrives from an LSP that does not have an associated interface.
Change-Id: Ib92e177be919147bafeb599729abf3d1abc2f4b3
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
This patch deprecates stack-based thread identification,
Also removes requirement that thread stacks are adjacent.
Finally, possibly annoying for some folks, it renames
all occurences of cpu_index and cpu_number with thread
index. Using word "cpu" is misleading here as thread can
be migrated ti different CPU, and also it is not related
to linux cpu index.
Change-Id: I68cdaf661e701d2336fc953dcb9978d10a70f7c1
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Change-Id: I86089e9bb604adfc260a111685001be1c897ce53
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
(VPP-668)
When the number of open files is reached in the system, vhost may
encounter a failure in socket call and return from vhost-user-process.
The return terminates all attempts of incoming socket connections
in the future, even if the condition is reconciled.
The fix is to not return from vhost-user-process, record the error in
the interface, spit out the error, and retry the connection every 3 seconds.
Change-Id: I806baedf13e8c9b73e7c7820c094240f39949950
Signed-off-by: Steven <sluong@cisco.com>
|
|
when walking vlib_buffer_t next_buffer chain on transmit.
On buffer allocation:
- next_buffer is not and may contain a stale invalid value that
should be ignored if not overwritten by a valid value.
- VLIB_BUFFER_NEXT_PRESENT flag is cleared and only set
if a valid value is written to next_buffer.
Change-Id: I9b0ccdc54f4f7456f8328ce7c4a0d52d0fba8caa
Signed-off-by: Jim Gibson <gibson+fdio@cisco.com>
|
|
To line up with "show interface placement," recently added. Otherwise,
"show int" refers only to "show interface placement," which tends to
annoy the cash customers...
Change-Id: Iea9e3681aeb051e2b0e1ecbf06706d98af9a3abf
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
vhost currently supports only polling mode. This patch is to add
interrupt mode. When the interface is configured for interrupt
mode, our input node does not get called unless there is a packet
in the vring.
If a particular CPU has one interface configured for polling mode
and another in interrupt, the input node is set to polling for
that CPU.
This diffs also includes two crashes in vlib's dispatch_node. One is
included in https://gerrit.fd.io/r/#/c/5516. The other crash is in
the ASSERT. The ASSERT can become true when the caller of
dispatch_node is in a loop. The first call converted the node
to polling. The second call thereafter will hit the ASSERT.
Change-Id: If17b6d48b20d7d8605c6a161459828637173cd32
Signed-off-by: Steven <sluong@cisco.com>
|
|
This change adds two new debug CLI command:
- "show interface placmenet" to display which
thread (main or worker) is responsible for processing
interface rx queue
vpp# show interface placement
Thread 0 (vpp_main):
node af-packet-input:
host-vpp1 queue 0
Thread 1 (vpp_wk_0):
node af-packet-input:
host-virbr0 queue 0
Thread 2 (vpp_wk_1):
node af-packet-input:
host-vpp2 queue 0
host-lxcbr0 queue 0
- "set interface placmenet" to assign thread (main or worker)
which process specific interface rx queue
vpp# set interface placement host-vpp1 queue 0 main
Change-Id: Id4dd00cf2b05e10fae2125ac7cb4411b446c5e9c
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Drop comes before lookup when enabled. is_first_or_last is not required when setting a feature, the anchor is added in find_config_with_features().
Don't make the PG interfaces automatically L3 enabled, this way we can have tests that check the L3 protocol disbaled behaviour.
Change-Id: Icef22a920b27ff9cec6ab2da6b05f05c532cb60f
Signed-off-by: Neale Ranns <nranns@cisco.com>
|
|
When the VM is migrated, the driver sends VHOST_USER_VRING_GET_BASE
message to the device to get the vring offset. The device is
supposed to shut down the vring, and return the current vring offset.
What the code did was to shutdown the vring, initialize the vring,
and return 0 to the driver.
The fix is to first store last_avail_idx in the message and then close
the vring.
Change-Id: I432e9f50f36d89fe53a45e050edcf5e1218caf7a
Signed-off-by: Steven <sluong@cisco.com>
|
|
vpp_lite platform is not needed anymore as same efect can be
achieved with following startup.conf config:
plugins {
plugin dpdk_plugin.so { disable }
}
Change-Id: I690ea8ceb1c6e1fe32e01e7da54e9958019a93bf
Signed-off-by: Damjan Marion <damarion@cisco.com>
|
|
Clean up spurious binary API client link dependency on libvlib.so,
which managed to hide behind vlib_mains == 0 checks reached by
VLIB_xxx_FUNCTION macros.
Change-Id: I5df1f8ab07dca1944250e643ccf06e60a8462325
Signed-off-by: Dave Barach <dave@barachs.net>
|
|
Add operation_mode for create_vhost_user_if, modify_vhost_user_if,
and sw_interface_vhost_user_details. Only polling mode is supported
for these APIs. Other mode is rejected and gets
VNET_API_ERROR_UNIMPLEMENTED error.
Change-Id: I0596f4e2c087aa2b6f78eb3e0b63910b1859641e
Signed-off-by: Steven <sluong@cisco.com>
|