Olivier Matz [Wed, 18 May 2016 11:04:26 +0000 (13:04 +0200)]
mempool: rename object constructor typedef
This commit renames mempool_obj_ctor_t as mempool_obj_cb_t.
In next commits, we will add the ability to populate the
mempool and iterate through objects using the same function.
We will use the same callback type for that. As the callback is
not a constructor anymore, rename it into rte_mempool_obj_cb_t.
The rte_mempool_obj_iter_t that was used to iterate over objects
will be removed in next commits.
No functional change.
In this commit, the API is preserved through a compat typedef.
Olivier Matz [Wed, 18 May 2016 11:04:25 +0000 (13:04 +0200)]
mempool: use sizeof to get size of header and trailer
Since commits d2e0ca22f and 97e7e685b the headers and trailers
of the mempool are defined as a structure. We can get their
size using a sizeof instead of doing a calculation that will
become wrong at the first structure update.
Olivier Matz [Wed, 18 May 2016 11:04:24 +0000 (13:04 +0200)]
mempool: uninline function to check cookies
There's no reason to keep this function inlined. Move it to
rte_mempool.c. We need to export the function for when compiling
with shared libraries + debug. We also need to keep the macro,
because we don't want to call an empty function when debug is
disabled.
Olivier Matz [Wed, 18 May 2016 11:04:23 +0000 (13:04 +0200)]
mempool: rename element size variables
This commit replaces elt_size by total_elt_size when appropriate.
In some mempool functions, we use the size of the elements as arguments
or variables. There is a confusion between the size including or not
including the header and trailer.
To avoid this confusion:
- update the API documentation
- rename the variables and argument names as "elt_size" when the size
does not include the header and trailer, or else as "total_elt_size".
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked by: Keith Wiles <keith.wiles@intel.com>
Olivier Matz [Wed, 18 May 2016 11:04:22 +0000 (13:04 +0200)]
mempool: rework comments and style
No functional change, just fix some comments and styling issues.
Also avoid to duplicate comments between rte_mempool_create()
and rte_mempool_xmem_create().
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Acked by: Keith Wiles <keith.wiles@intel.com>
Hiroyuki Mikita [Wed, 18 May 2016 14:41:03 +0000 (23:41 +0900)]
mbuf: decrease reference counter when detaching
The rte_pktmbuf_detach() function should decrease refcnt on a direct
buffer as stated in doc/guides/prog_guide/mbuf_lib.rst:
"whenever the indirect buffer is detached, the reference counter on the
direct buffer is decremented."
tools: allow binding to other network class devices
dpdk_nic_bind will only handle Ethernet devices, but Mellanox ConnectX-3 Pro,
for example, is a Network class device, but not an Ethernet one. Even though
this allows other devices in the list, like Wireless devices, this should not be
a problem.
Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@redhat.com>
Maxime Leroy [Mon, 9 May 2016 16:13:36 +0000 (18:13 +0200)]
log: fix level/type retrieving from a standard thread
The functions rte_log_cur_msg_loglevel() and rte_log_cur_msg_logtype()
return the current log level/type for the message being processed. They
are used when implementing a user-defined logging stream.
The current log levels and types were stored in a table indexed by the
lcore_id, only returning a valid value for dataplane threads. Setting
and getting these values in a non dataplane thread was ignored, using
the global value instead.
To fix this issue, a per-thread variable could be used (with
RTE_DEFINE_PER_LCORE), allowing any pthread to set and retrieve its
current log level or type.
Signed-off-by: Maxime Leroy <maxime.leroy@6wind.com> Signed-off-by: Olivier Matz <olivier.matz@6wind.com> Acked-by: David Marchand <david.marchand@6wind.com>
Hemant Agrawal [Wed, 11 May 2016 13:47:59 +0000 (19:17 +0530)]
mk: introduce NXP dpaa2 architecture based on armv8-a
This patch introduces dpaa2 machine target to address difference
in cpu parameter, number of core to 8 and no numa support
w.r.t default armv8-a machine
Keith Wiles [Thu, 14 Apr 2016 09:42:36 +0000 (11:42 +0200)]
mempool: reduce structure size if no cache needed
The rte_mempool structure is changed, which will cause an ABI change
for this structure. Providing backward compat is not reasonable
here as this structure is used in multiple defines/inlines.
Allow mempool cache support to be dynamic depending on if the
mempool being created needs cache support. Saves about 1.5M of
memory used by the rte_mempool structure.
Allocating small mempools which do not require cache can consume
larges amounts of memory if you have a number of these mempools.
Signed-off-by: Keith Wiles <keith.wiles@intel.com> Acked-by: Olivier Matz <olivier.matz@6wind.com>
Olivier Matz [Wed, 11 May 2016 14:43:46 +0000 (16:43 +0200)]
mbuf: add raw allocation function
Many drivers provide their own implementation of rte_mbuf_raw_alloc(),
duplicating the code. Introduce a new public function in rte_mbuf to
allocate a raw mbuf (uninitialized).
Olivier Matz [Fri, 13 May 2016 13:27:59 +0000 (15:27 +0200)]
doc: move release notes instructions as comments
We don't want to have this instructions in the generated docs, so use
comments. It's also less confusing for people adding entries in the
documentation.
Signed-off-by: Olivier Matz <olivier.matz@6wind.com> Acked-by: John McNamara <john.mcnamara@intel.com>
Infinite loop: The loop does not have a normal termination condition,
so will continue until an abnormal condition arises.
In rte_netmap_poll: Infinite loop with unsatisfiable exit condition.
overrun-local: Overrunning array tokens of 8 8-byte elements
at element index 4294967294 (byte offset 34359738352)
using index i (which evaluates to 4294967294).
Daniel Mrzyglod [Wed, 20 Apr 2016 14:39:30 +0000 (16:39 +0200)]
power: fix error messages
Function strerror(errno) has built strings only for non-negative errno values.
for negative values of errno it describe error as "Unknown error -errno"
to be more descriptive i put string "channel not found" taken from header.
The negative argument will be interpreted as a very large unsigned value.
Coverity issue: 13266
Coverity issue: 13269 Fixes: 445c6528b55f ("power: common interface for guest and host") Signed-off-by: Daniel Mrzyglod <danielx.t.mrzyglod@intel.com>
Jan Viktorin [Mon, 9 May 2016 21:34:16 +0000 (23:34 +0200)]
config: disable qede for ARMv7
The qede PMD driver is failing when building for ARMv7:
drivers/net/qede/base/ecore_dev.c:1150:6:
error: variable ‘prs_reg’ set but not used
[...]
drivers/net/qede/base/ecore_dev.c:2492:13:
error: variable ‘p_phys’ set but not used
[...]
drivers/net/qede/base/ecore_dev.c:2532:39:
error: variable ‘pbl_size’ set but not used
Fixes: 3eae93a9bfd5 ("qede: enable PMD build") Signed-off-by: Jan Viktorin <viktorin@rehivetech.com>
Panu Matilainen [Tue, 10 May 2016 10:01:25 +0000 (13:01 +0300)]
qede: fix build with gcc 6
With gcc >= 6.0, qede base driver fails to build with:
drivers/net/qede/base/ecore_cxt.c: In function 'ecore_cdu_init_common':
cc1: error: left shift of negative value [-Werror=shift-negative-value]
Since the base drivers are untouchable, work around by disabling
the warning.
Fixes: ec94dbc57362 ("qede: add base driver") Signed-off-by: Panu Matilainen <pmatilai@redhat.com>
Daniel Mrzyglod [Tue, 10 May 2016 16:11:18 +0000 (18:11 +0200)]
vhost: fix name not null terminated
Fix issue reported by Coverity.
Coverity ID 124556
If the buffer is treated as a null terminated string in later operations,
a buffer overflow or over-read may occur.
In vhost_set_ifname: The string buffer may not have a null terminator if
the source string's length is equal to the buffer size
Fixes: 54292e9520e0 ("vhost: support ifname for vhost-user") Signed-off-by: Daniel Mrzyglod <danielx.t.mrzyglod@intel.com> Acked-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Yuanhan Liu [Mon, 2 May 2016 21:23:48 +0000 (14:23 -0700)]
examples/vhost: fix mbuf allocation failure
It has always been a mystery (at least to me before) that how many
mbuf is enough while creating an mbuf pool. While current macro
NUM_MBUFS_PER_PORT gives your some insights, it's not that accurate:
it doesn't consider the case we may receive a big packet, say 64K
when TSO is enabled.
We actually have tried to fix it once before, with commit 5499c1fc9baa
("examples/vhost: fix mbuf allocation"), but it just workarounded it
by enlarging it a bit so that the case described in the commit log
by passes. So, while trying to fix it ultimately, I'm thinking how
big is big enough, and what are the factors need consider to figure
out a proper value.
Therefore, here you are. I introduced a helper function to create
the mbuf pool, and do the "how many mbufs are needed" calculation
there. Also, I put detailed comments how that comes, to serve as
the guidelines.
Yuanhan Liu [Mon, 2 May 2016 21:23:47 +0000 (14:23 -0700)]
examples/vhost: handle broadcast packet
Every time I do a VM2VM iperf test with vhost example, I have to set
the arp table manually, as vhost-switch just ignores the broadcast
packet, leaving the ARP request not served.
Here we do a transmit a broadcast packet (such as ARP request) to
every vhost device, as well as the physical port, to fix above
arp table issue.
Signed-off-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Yuanhan Liu [Mon, 2 May 2016 21:23:44 +0000 (14:23 -0700)]
examples/vhost: remove unused macro and struct
Interestingly, DESC_PER_CACHELINE has never been used since the
introduction of vhost example. Remove it.
vlan_ethhdr struct and VLAN_ETH_HLEN macro reference had been removed
by commit 4d50b6acbd95 ("examples/vhost: adapt Tx routing to lib"), but
had forgot to remove the definition.
Fixes: 4d50b6acbd95 ("examples/vhost: adapt Tx routing to lib") Signed-off-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Yuanhan Liu [Mon, 2 May 2016 21:23:43 +0000 (14:23 -0700)]
examples/vhost: remove the non-working zero copy code
It's reported that it's has not been working for a long while. And due
to it's complex, it's better to redesign it than to fix it to make it
work again.
Signed-off-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Jianfeng Tan [Fri, 29 Apr 2016 00:48:46 +0000 (00:48 +0000)]
virtio: fix memory leak of virtqueue memzones
When virtio was proposed in DPDK, there is no API to free memzones.
But this has changed since rte_memzone_free() has been implemented by
commit ff909fe21f0a ("mem: introduce memzone freeing").
This patch is to make sure memzones in struct virtqueue, like mz and
virtio_net_hdr_mz, are freed when queue is released or setup fails.
Fixes: c1f86306a026 ("virtio: add new driver") Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com> Acked-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Jianfeng Tan [Mon, 9 May 2016 16:35:57 +0000 (09:35 -0700)]
virtio: fix overwritten driver flags
The "drv_flags" is set with device as the input, which means different
device (say, modern vs legacy) could end up with a different value. And
the fact that "drv_flags" is shared by all devices means that every time
we add a new device, it simply overwrites the value configured from the
last device.
Therefore, when two virtio devices have different flags, it may lead to
wrong result, such as virtio would set irq config when it's not supported.
Making the flag per device (using "dev->data->dev_flags") could let us
have different value for each device, which would avoid the above issue.
Fixes: da978dfdc43 ("virtio: use port IO to get PCI resource") Reported-by: David Marchand <david.marchand@6wind.com> Suggested-by: David Marchand <david.marchand@6wind.com> Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com> Acked-by: David Marchand <david.marchand@6wind.com> Acked-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Avail ring is updated by the frontend and consumed by the backend.
There are frequent core to core cache transfers for the avail ring.
This optmization avoids avail ring entry index update if the entry
already holds the same value.
As DPDK virtio PMD implements FIFO free descriptor list (also for
performance reason of CACHE), in which descriptors are allocated
from the head and freed to the tail, with this patch in most cases
avail ring will remain the same, then it would be valid in both caches
of frontend and backend.
Suggested-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Huawei Xie <huawei.xie@intel.com> Acked-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Currently, vhost PMD doesn't have linkage for librte_vhost, even though
it depends on librte_vhost APIs. This causes a linkage error if below
conditions are fulfilled.
- DPDK libraries are compiled as shared libraries.
- DPDK application doesn't link librte_vhost.
- Above application tries to link vhost PMD using '-d' DPDK option.
The patch adds linkage for librte_vhost to vhost PMD not to cause an
above error.
Fixes: ee584e9710b9 ("vhost: add driver on top of the library") Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp> Acked-by: Panu Matilainen <pmatilai@redhat.com> Acked-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Jianfeng Tan [Mon, 25 Apr 2016 02:37:45 +0000 (02:37 +0000)]
virtio: fix segfault on Tx desc flags setup
After the do-while loop, idx could be VQ_RING_DESC_CHAIN_END (32768)
when it's the last vring desc buf we can get. Therefore, following
expresssion could lead to a segfault error, as it tries to access
beyond the desc memory boundary.
start_dp[idx].flags &= ~VRING_DESC_F_NEXT;
This bug could be reproduced easily with "set fwd txonly" in the
guest PMD, where the dequeue on host is slower than the guest Tx,
that running out of free desc buf is pretty easy.
The fix is straightforward and easy, just remove it, as we have
already set desc flags properly inside the do-while loop.
Fixes: dd856dfcb9e ("virtio: use any layout on Tx")
[Yuanhan Liu: commit log reword] Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com> Acked-by: Huawei Xie <huawei.xie@intel.com> Acked-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Jianfeng Tan [Tue, 19 Apr 2016 05:22:37 +0000 (05:22 +0000)]
virtio: fix newline under debug mode
Issue: output of appliations and debug info of DPDK may be mixed up
in same line when enabling below debug options of virtio:
CONFIG_RTE_LIBRTE_VIRTIO_DEBUG_INIT
CONFIG_RTE_LIBRTE_VIRTIO_DEBUG_TX
CONFIG_RTE_LIBRTE_VIRTIO_DEBUG_DRIVER
This patch adds "\n" in the tail of definitions like PMD_RX_LOG,
PMD_TX_LOG, and PMD_DRV_LOG, and removes some "\n" when using these
macros.
Fixes: c1f86306a026 ("virtio: add new driver") Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com> Acked-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Jingjing Wu [Wed, 4 May 2016 05:42:25 +0000 (13:42 +0800)]
i40e: fix VLAN stripping from inner header
Previously, for tunnel packets, such as VXLAN/NVGRE, the vlan
tags of the inner header will be stripped without putting vlan
info to descriptor, what is not expected behaviour.
This patch fixes it by changing hardware configuration to leave
the inner packet alone.
Fixes: 4861cde46116 ("i40e: new poll mode driver") Fixes: a778a1fa2e4e ("i40e: set up and initialize flow director") Signed-off-by: Jingjing Wu <jingjing.wu@intel.com>
John Daley [Wed, 27 Apr 2016 02:51:56 +0000 (19:51 -0700)]
enic: fix offset for Rx mbuf data
The code to provide mbufs for RX used m->data_off instead of
RTE_PKTMBUF_HEADROOM as the position inside the mbuf for the data to be
written. As the mbuf is uninitialised, this could potentially cause Rx
data to be placed at the wrong address in the mbuf - or even outside it.
Fixes: 947d860c821f ("enic: improve Rx performance") Signed-off-by: John Daley <johndale@cisco.com>
nfp: serialize access for hardware reconfiguration
Some apps calling some functions from different threads at the
same time could lead to reconfig problems. Reconfig mechanism is
based on a hardware queue where incrementing a counter signals the
firmware to do the reconfig. If there is a second increment before the
first one has been processed the firmware will stop and a device
reset is necessary.
Björn Töpel [Fri, 22 Apr 2016 05:39:22 +0000 (07:39 +0200)]
i40evf: report error for unsupported CRC stripping config
On hosts running a non-DPDK PF driver, the VF has no means of changing
the HW CRC strip setting for a RX queue. It's implicitly enabled.
This patch checks if the host is running a non-DPDK PF kernel driver,
and returns an error, if HW CRC stripping was not requested in the port
configuration.
Signed-off-by: Björn Töpel <bjorn.topel@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com>
Daniel Mrzyglod [Mon, 18 Apr 2016 17:13:23 +0000 (19:13 +0200)]
i40e: remove unneeded NULL check
In i40evf_config_vlan_pvid the check for NULL for the dev value is
unnecessary, since this value is passed in from the ethdev API which
will ensure that a valid rte_eth_dev structure is provided.
Furthermore, all code paths leading to this function already use the
dev value.
Issue identified by Coverity.
Coverity ID 13302:
There may be a null pointer dereference, or else the comparison against
null is unnecessary.
In i40evf_config_vlan_pvid: All paths that lead to this null pointer
comparison already dereference the pointer earlier
Fixes: 2b12431b5369 ("i40e: add vlan stripping and insertion to VF") Signed-off-by: Daniel Mrzyglod <danielx.t.mrzyglod@intel.com> Acked-by: Helin Zhang <helin.zhang@intel.com> Acked-by: Jingjing Wu <jingjing.wu@intel.com>
Bruce Richardson [Thu, 14 Apr 2016 16:02:37 +0000 (17:02 +0100)]
i40e: simplify SSE packet length extraction code
In Table 8-16 of the "Intel® Ethernet Controller XL710 Datasheet" it is
stated that when the whole packet is written to a single buffer, the
header length field in the descriptor will be 0. This means that when
extracting the packet/data_len field from the descriptor in the driver
we do not need to mask out the extra header-length bits.
Inside the vector driver, this reduces the need to pull all four pktlen
fields into a single register to work on. Instead of a shift and mask,
we now need to only do a shift. Therefore, we can work on each descriptor
independently, processing each using one shift intrinsic and a blend.
This change makes the code shorter and easier to read, so we can pull it
into the main descriptor processing loop instead of needing its own
function. This in turn makes the descriptor processing in the loop as a
whole slightly easier to read as it's more linear.
In terms of performance, in testing this change shows little effect, with
single-core perf tests showing a very slight improvement.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Zhe Tao <zhe.tao@intel.com>
Bruce Richardson [Thu, 14 Apr 2016 16:02:36 +0000 (17:02 +0100)]
i40e: improve performance of vector PMD
An analysis of the i40e code using Intel® VTune™ Amplifier 2016 showed
that the code was unexpectedly causing stalls due to "Loads blocked by
Store Forwards". This can occur when a load from memory has to wait
due to the prior store being to the same address, but being of a smaller
size i.e. the stored value cannot be directly returned to the loader.
[See ref: https://software.intel.com/en-us/node/544454]
These stalls are due to the way in which the data_len values are handled
in the driver. The lengths are extracted using vector operations, but those
16-bit lengths are then assigned using scalar operations i.e. 16-bit
stores.
These regular 16-bit stores actually have two effects in the code:
* they cause the "Loads blocked by Store Forwards" issues reported
* they also cause the previous loads in the RX function to actually be a
load followed by a store to an address on the stack, because the 16-bit
assignment can't be done to an xmm register.
By converting the 16-bit store operations into a sequence of SSE blend
operations, we can ensure that the descriptor loads only occur once, and
avoid both the additional stores and loads from the stack, as well as the
stalls due to the blocked loads.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Zhe Tao <zhe.tao@intel.com>
Bruce Richardson [Thu, 14 Apr 2016 16:02:35 +0000 (17:02 +0100)]
i40e: require SSE4.1 support for vector driver
Later commits to improve the driver will make use of the SSE4.1
_mm_blend_epi16 intrinsic, so:
* set the compilation level to always have SSE4.1 support,
* and add in a runtime check for SSE4.1 as part of the condition checks
for vector driver selection.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com> Acked-by: Zhe Tao <zhe.tao@intel.com>
The physical link is handled by the management Firmware.
This patch lays the infrastructure for interrupt/attention handling in
the driver, as link change notifications arrive via async interrupts,
as well as the handling of such notifications. It adds async event
notification handler interfaces to the PMD.
This patch adds following SRIOV features to qede PMD:
- VF configuration
- VF intialization/de-initialization
- VF PF communications channel
- statistics capture and query
The Qlogic Everest Driver for Ethernet(QEDE) Poll Mode Driver(PMD) is
the DPDK specific module for QLogic FastLinQ QL4xxxx 25G/40G CNA family
of adapters as well as their virtual functions (VF) in SR-IOV context.
This patch adds QEDE PMD, which interacts with base driver and
initialises the HW.
This patch content also includes:
- eth_dev_ops callbacks
- Rx/Tx support for the driver
- link default configuration
- change link property
- link up/down/update notifications
- vlan offload and filtering capability
- device/function/port statistics
- qede nic guide and updated overview.rst
Note that the follow on commits contain the code for the features mentioned
in documents but not implemented in this patch.
The base driver is the backend module for the QLogic FastLinQ QL4xxxx
25G/40G CNA family of adapters as well as their virtual functions (VF)
in SR-IOV context.
The purpose of the base module is to:
- provide all the common code that will be shared between the various
drivers that would be used with said line of products. Flows such as
chip initialization and de-initialization fall under this category.
- abstract the protocol-specific HW & FW components, allowing the
protocol drivers to have clean APIs, which are detached in its
slowpath configuration from the actual Hardware Software Interface(HSI).
This patch adds a base module without any protocol-specific bits.
I.e., this adds a basic implementation that almost entirely falls under
the first category.
Tomasz Kulasek [Fri, 22 Apr 2016 15:35:57 +0000 (17:35 +0200)]
ixgbe: fix bit shift overflow in VMDQ pool setup
Fix issue reported by Coverity.
Coverity ID 13193: Bad bit shift operation (BAD_SHIFT)
large_shift: In expression 1 << pool, left shifting by more than 31 bits
has undefined behavior. The shift amount, pool, is at least 32.
This patch is a rework of register addr selection logic and mask
computation to made it more readable and avoid bit overflow when 32 bit
value is shifted over its size for pool > 31.
Fixes: fe3a45fd4104 ("ixgbe: add VMDq support") Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com> Acked-by: Bruce Richardson <bruce.richardson@intel.com>
The statistics queried by calling rte_eth_stats_get are zero when
the API is first called on the port. The root cause is because the
offset_loaded flag is not set correctly after device start.
This patch fixes this issue by resetting statistics at initialization
time. The resetting process will set offset_loaded flag.
Michael Frasca [Mon, 18 Apr 2016 12:51:52 +0000 (08:51 -0400)]
fm10k: fix packet type for multi-segment packets
When building a chain of mbufs for a multi-segment packet, the
packet_type field resides at the end of the chain. It should be
copied forward to the head of the list.
Also, uses RTE_LIBRTE_FM10K_RX_OLFLAGS_ENABLE to guard packet-type
computation. The mbuf fields are not copied when this define is not set.
Piotr Azarewicz [Thu, 14 Apr 2016 09:59:30 +0000 (11:59 +0200)]
ixgbe: fix bit masking in queue stop
The masking for the RX/TX enable bit was incorrect in the rx and tx
queue stop functions. Instead of using "& MASK" it used "| MASK" which
would always return true. This error was found by converity scan.
CID 13215 : Wrong operator used (CONSTANT_EXPRESSION_RESULT)
operator_confusion: txdctl | 33554432 is always 1/true regardless of the
values of its operand. This occurs as the logical second operand of
'&&'.
CID 13216 : Wrong operator used (CONSTANT_EXPRESSION_RESULT)
operator_confusion: rxdctl | 33554432 is always 1/true regardless of the
values of its operand. This occurs as the logical second operand of
'&&'.
Coverity issue: 13215
Coverity issue: 13216 Fixes: 029fd06d40fa ("ixgbe: queue start and stop") Signed-off-by: Piotr Azarewicz <piotrx.t.azarewicz@intel.com> Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Run ixgbe driver through checkpatch and fix the issues highlighted
Fix line spacing, some bad indentation, and in a couple
of cases use short circuit (already there) return to lessen indentation.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org> Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Applied with four additional fixes for issues highlighted by checkpatch Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Jan Viktorin [Fri, 15 Apr 2016 22:33:06 +0000 (00:33 +0200)]
mk: do not enforce any specific ARM ABI
The dpdk build system passes -mfloat-abi=softfp, which makes the build fail
when the selected ABI is EABIhf. The dpdk build system should not make
assumptions on the selected ARM ABI.
Reported-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com> Signed-off-by: Jan Viktorin <viktorin@rehivetech.com>
Thomas Monjalon [Fri, 22 Apr 2016 12:21:26 +0000 (14:21 +0200)]
eal: add assert macro for debug
The macro RTE_VERIFY always checks a condition.
It is optimized with "unlikely" hint.
While this macro is well suited for test applications, it is preferred
in libraries and examples to enable such check in debug mode.
That's why the macro RTE_ASSERT is introduced to call RTE_VERIFY only
if built with debug logs enabled.
A lot of assert macros were duplicated and enabled with a specific flag.
Removing these #ifdef allows to test these code branches more easily
and avoid dead code pitfalls.
The ENA_ASSERT is kept (in debug mode only) because it has more
parameters to log.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Thomas Monjalon [Fri, 22 Apr 2016 10:52:20 +0000 (12:52 +0200)]
examples: remove useless debug flags
The debug logs must be enabled at compile-time and run-time.
There are also some internal flags in some examples to enable the debug
logs of the applications. They are now enabled in debug configs and
can be disabled thanks to the more generic logtype mechanism:
rte_set_log_type(RTE_LOGTYPE_USER1, 0);
Removing these #ifdef allows to test these code branches more easily
and avoid dead code pitfalls.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com> Acked-by: Yuanhan Liu <yuanhan.liu@linux.intel.com>
Thomas Monjalon [Fri, 22 Apr 2016 10:25:59 +0000 (12:25 +0200)]
log: increase default level to info
The default was to compile every logs (including debug) and set
the default level to debug.
As some debug logs may hurt performance, a notice is added and the
default level is now info.
In order to enable debug logs, they must be compiled with
RTE_LOG_LEVEL=RTE_LOG_DEBUG and enabled at runtime with --log-level=8.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Thomas Monjalon [Fri, 22 Apr 2016 13:03:02 +0000 (15:03 +0200)]
eal: increase log level of some messages
Some relevant EAL logs are raised from debug to info level in order to
show the available number of cores and the detected devices.
The driver ids and name are logged only if the device is not blacklisted.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Back then when we fixed the missing free lpm I was to quickly to say yes
if it applies not only to the lpm6 but also to all of the lpm code.
It turned out to not apply to all of them. In rte_lpm_create_v20 there
is an unexpected fused allocation:
mem_size = sizeof(*lpm) + (sizeof(lpm->rules_tbl[0]) * max_rules);
[...]
lpm = (struct rte_lpm_v20 *)rte_zmalloc_socket(mem_name,mem_size,
RTE_CACHE_LINE_SIZE, socket_id);
That causes lpm->rules_tbl not to have an own struct malloc_elem that
can be derived via RTE_PTR_SUB(data, MALLOC_ELEM_HEADER_LEN) in
malloc_elem_from_data.
Due to that the rte_lpm_free_v20 accidentially misderives the elem and
assumes it is ELEM_FREE triggering in malloc_elem_free
if (!malloc_elem_cookies_ok(elem) || elem->state !=
return -1;
While it seems counter-intuitive the way to properly remove rules_tbl in
the old fused allocation style of rte_lpm_free_v20 is to not remove it.
The newer rte_lpm_free_v1604 is safe because in rte_lpm_create_v1604
rules_tbl is a separate allocation.
If any fragment hole is found in ipv4_frag_reassemble() and
ipv6_frag_reassemble(), whole ip_frag_pkt mbufs are moved to death-row.
Any mbufs already chained to another mbuf are freed multiple times as
there are still in ip_frag_pkt array.
Signed-off-by: Chaeyong Chong <cychong@gmail.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Igor Ryzhov [Thu, 21 Apr 2016 16:27:30 +0000 (19:27 +0300)]
kni: keep ethernet MAC address when interface goes up
Currently every time a KNI interface goes up, its ethernet address
is reassigned.
After this patch ethernet address is assigned only once,
at initialization time.
Tomasz Kulasek [Mon, 11 Apr 2016 16:03:48 +0000 (18:03 +0200)]
app/testpmd: fix string overrun in engines listing
CID 13307 (#1 of 1): Copy into fixed size buffer (STRING_OVERFLOW)
fixed_size_dest: You might overrun the 128 byte fixed-size string fwd_modes
by copying fwd_eng->fwd_mode_name without checking the length.
Fixes: 769ce6b17835 ("app/testpmd: list forwarding engines") Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com>
Tomasz Kulasek [Tue, 26 Apr 2016 15:47:56 +0000 (17:47 +0200)]
examples/performance-thread: fix crash with gcc 5
It seems that with gcc >5.x and -O2/-O3 optimization breaks packet
grouping algorithm in l3fwd-thread application causing segfault.
When last packet pointer "lp" and "pnum->u64" buffer points the same
memory buffer, high optimization can cause unpredictable results.
It seems that assignment of precalculated group sizes may interfere
with initialization of new group size when lp points value inside
current group and didn't should be changed.
With gcc >5.x and optimization we cannot be sure which assignment will be
done first, so the group size can be counted incorrectly causing segfault.
This patch eliminates intersection of assignment of initial group size
(lp[0] = 1) and precalculated group sizes when gptbl[v].idx < 4.
The same patch was applied for original l3fwd (af1694d94).
Fixes: d48415e1fee3 ("examples/performance-thread: add l3fwd-thread app") Signed-off-by: Tomasz Kulasek <tomaszx.kulasek@intel.com>
Daniel Mrzyglod [Fri, 22 Apr 2016 16:00:50 +0000 (18:00 +0200)]
acl: fix division by float zero in test app
Fix issue reported by Coverity.
Coverity ID 13240
This could cause an immediate crash or incorrect computation.
In search_ip5tuples: An expression which may be zero is used
as a divisor in floating-point arithmetic.
divide_by_zero: In expression (long double)tm / pkt,
division by expression pkt which may be zero has undefined behavior.
Fixes: 26c057ab6c45 ("acl: new test-acl application") Signed-off-by: Daniel Mrzyglod <danielx.t.mrzyglod@intel.com> Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Thomas Monjalon [Tue, 19 Apr 2016 20:38:31 +0000 (22:38 +0200)]
eal: remove useless internal function from memcpy headers
The function rte_memcpy_func() is used in ARM and PPC implementations
of rte_memcpy().
There are some useless copies in Tile and some ARM branches.
It was also declared without doxygen comment in the generic header.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com> Acked-by: David Marchand <david.marchand@6wind.com>
John McNamara [Mon, 11 Apr 2016 22:27:25 +0000 (23:27 +0100)]
doc: improve nics features matrix
Changed symbol on NIC overview table from X to Y to help
clarify the indicated features are supported. The X caused
confusion for some readers.
Also, added * character to indicate partially supported
features. This can be used in the future to direct the reader
to more specific details in the individual NIC guides.
Signed-off-by: John McNamara <john.mcnamara@intel.com>
Thomas Monjalon [Tue, 19 Apr 2016 19:49:03 +0000 (21:49 +0200)]
lib: remove poisoned flags
Some flags were poisoned after having been removed from EAL and mbuf
in releases 1.8 (b10eef348d, 62814bc2e9) and 2.0 (4769bc5a27cc).
After several releases, they have probably disappeared from all
applications going to upgrade to DPDK 16.07.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com> Acked-by: David Marchand <david.marchand@6wind.com>
git.droids-corp.org / dpdk.git / log