Signed-off-by: David Marchand <david.marchand@6wind.com>
[Thomas: move inclusion used by ixgbe bypass]
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
This fixes cases in KNI where kernel allocation function return value
is needlessly casted.
Detected with coccinelle:
lib/librte_eal/linuxapp/kni/ethtool/igb/igb_main.c:3181:25-28:
WARNING: casting value returned by memory allocation function to (u32 *) is useless.
lib/librte_eal/linuxapp/kni/kni_vhost.c:690:9-28:
WARNING: casting value returned by memory allocation function to (struct rte_kni_fifo *) is useless.
lib/librte_eal/linuxapp/kni/kni_vhost.c:684:13-27:
WARNING: casting value returned by memory allocation function to (struct sk_buff *) is useless
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
nic_uio requires the pci ids to be present in rte_pci_dev_ids.h in order to
bind the devices to nic_uio. However, it's better to remove this whitelist
of pci ids, and instead rely on hw.nic_uio.bdfs kenv parameter to allow
binding any device to nic_uio.
Suggested-by: David Marchand <david.marchand@6wind.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
Signed-off-by: Kumar Sanghvi <kumaras@chelsio.com>
Acked-by: David Marchand <david.marchand@6wind.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
This problem was discovered when passing invalid PCI id to the
blacklist API in devargs.
Any failures in rte_devargs_add would cause a core dump because
it would call rte_log() before the the EAL log environment was
initialized. Rather than try and log just remove the messages
and leave it up to the caller to check the return value.
Most of the other failure possibilities are when malloc() fails, and if
that happens any logging that used malloc() would also fail.
This failure was not caught by the standalone tests to devargs
because the tests are run after calling rte_eal_init (which is not
how devargs is intended to be used).
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Change the log level of startup messages. Anything that is
just normal activity (like getting virtual areas) is changed
to debug level. Anything that is a failure should be NOTICE
or ERR severity.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
The read for events in the interrupt thread may get interrupted
by signals from application. Avoid generating stray log message.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
There are close and detach functions in ethdev.
To keep a consistent naming, PCI functions called by ethdev detach
must be named "detach" instead of "close".
Fix also comments which mix close and uninit names.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
SLES 12 has kernel 3.12, which original does not have skb_set_hash,
but SuSE has added that function to the kernel integrated on it.
Therefore, the function is not declared when compiling on this OS.
Reported-by: Sotiris Salloumis <sotiris.salloumis@ericsson.com>
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Implement rte_memzone_free which, as its name implies, would free a
memzone.
Currently memzone are tracked in an array and cannot be free.
To be able to reuse the same array to track memzones, we have to
change how we keep track of reserved memzones.
With this patch, any memzone with addr NULL is not used, so we also need
to change how we look for the next memzone entry free.
Add new unit test for rte_memzone_free API.
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
In the current memory hierarchy, memsegs are groups of physically
contiguous hugepages, memzones are slices of memsegs and malloc further
slices memzones into smaller memory chunks.
This patch modifies malloc so it partitions memsegs instead of memzones.
Thus memzones would call malloc internally for memory allocation while
maintaining its ABI.
During initialization malloc sets all available memory as part of the heaps.
CONFIG_RTE_MALLOC_MEMZONE_SIZE was used to specify the default memory
block size to expand the heap. The option is not used/relevant anymore,
so we remove it.
Remove free_memseg field from internal mem config structure as it is
not used anymore.
Also remove code in ivshmem that was setting up free_memseg on init.
It would be possible to free memzones and therefore any other structure
based on memzones, ie. mempools
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
Move malloc inside eal and create a new section in MAINTAINERS file for
Memory Allocation in EAL.
Create a dummy malloc library to avoid breaking applications that have
librte_malloc in their DT_NEEDED entries.
This is the first step towards using malloc to allocate memory directly
from memsegs. Thus, memzones would allocate memory through malloc,
allowing to free memzones.
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
In order to unify the packet type, the field of 'packet_type' in
'struct rte_mbuf' needs to be extended from 16 to 32 bits.
Accordingly, some fields in 'struct rte_mbuf' are re-organized to support
this change for Vector PMD.
As 'struct rte_kni_mbuf' for KNI should be right mapped to
'struct rte_mbuf', it should be modified accordingly.
In ixgbe PMD driver, corresponding changes are added for the mbuf changes,
especially the bit masks of packet type for 'ol_flags' are replaced by
unified packet type. In addition, more packet types (UDP, TCP and SCTP)
are supported in vectorized ixgbe PMD.
To avoid breaking ABI compatibility, all the changes would be enabled by
RTE_NEXT_ABI.
Note that around 2% performance drop (64B) was observed of doing 4 ports
(1 port per 82599 card) IO forwarding on the same SNB core.
Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Signed-off-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
After code rework from bellow commit, logic expects hugepage_sz field to
always be set (ie. not zero value).
When using --no-huge, this field was left unset defaulting to zero.
Set hugepage_sz to RTE_PGSIZE_4K when using --no-huge.
Fixes: b3dfffd962 ("mem: allow multiple page sizes to be requested")
Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
Commit 7c5d0cc915 added const in prefetch functions for
X86 and PPC. This patch does the same for Tile arch.
Signed-off-by: Zhigang Lu <zlu@ezchip.com>
When using vfio, the probe fails for BAR > 0 after the
commit-id 90a1633b2 (eal/linux: allow to map BARs with MSI-X tables).
While debugging further, found that the BAR region offset and size read from
vfio are u64, but are assigned to uint32_t variables. This results in the u64
value getting truncated to 0 and passing wrong offset and size to mmap for
subsequent BAR regions.
The fix is to use unsigned long for the offset and size.
This is based on patch by Alejandro Lucero <alejandro.lucero@netronome.com>
posted at below:
http://dpdk.org/ml/archives/dev/2015-June/020201.html
and updated with diff from below to fix 32-bit compilation:
http://dpdk.org/ml/archives/dev/2015-July/020963.html
Fixes: 90a1633b23 ("eal/linux: allow to map BARs with MSI-X tables")
Signed-off-by: Alejandro Lucero <alejandro.lucero@netronome.com>
Signed-off-by: Rahul Lakkireddy <rahul.lakkireddy@chelsio.com>
Signed-off-by: Kumar Sanghvi <kumaras@chelsio.com>
The patch fixes vfio initialization issue introduced by below patch.
Root cause is that VFIO_PRESENT is inaccessible in eal common level.
To fix it, remove pci_map/unmap_device from common code, then implement
in linux and bsd code.
Fixes: 35b3313e32 ("pci: merge mapping functions for linux and bsd")
Reported-by: Michael Qiu <michael.qiu@intel.com>
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
This commit adds support for the TILE-Gx platform, as well as the TILE
CPU architecture. This architecture port is fairly simple due to its
reliance on generics for most arch stuff.
Signed-off-by: Cyril Chemparathy <cchemparathy@ezchip.com>
Signed-off-by: Zhigang Lu <zlu@ezchip.com>
This patch adds a few new page sizes that are supported on the TILE-Gx
and TILE-Mx platforms.
Signed-off-by: Cyril Chemparathy <cchemparathy@ezchip.com>
Signed-off-by: Zhigang Lu <zlu@ezchip.com>
This patch extends the memzone allocator to remove the restriction
that prevented callers from specifying multiple page sizes in the
flags argument.
In doing so, we also sanitize the free segment matching logic to get
rid of architecture specific disjunctions (2MB vs 1GB on x86, and 16MB
vs 16GB on PPC), thereby allowing for a broader range of hugepages on
architectures that support it.
Signed-off-by: Cyril Chemparathy <cchemparathy@ezchip.com>
Signed-off-by: Zhigang Lu <zlu@ezchip.com>
The definitions of rte_memzone_reserve_aligned() and
rte_memzone_reserve_bounded() were identical with the exception of the
bound argument passed into rte_memzone_reserve_thread_safe().
This patch removes this replication of code by unifying it into
rte_memzone_reserve_thread_safe(), which is then called by all three
variants of rte_memzone_reserve().
Signed-off-by: Cyril Chemparathy <cchemparathy@ezchip.com>
Signed-off-by: Zhigang Lu <zlu@ezchip.com>
When RTE_COMPILE_TIME_CPUFLAGS is empty, the rte_cpu_check_supported()
code breaks with a "comparison is always false due to limited range of
data type". This is because the compile_time_flags[] array is empty.
Assigning the array dimension to a local variable apparently solves this.
Signed-off-by: Cyril Chemparathy <cchemparathy@ezchip.com>
Signed-off-by: Zhigang Lu <zlu@ezchip.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Add support for 82583V (E1000) PCI device id.
Signed-off-by: James Davidson <james.davidson@brocade.com>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
Newer kernels make this unreadable for security reasons for non-roots.
Running the application will then fill the logs with
rte_mem_virt2phy: cannot open /proc/self/pagemap
messages.
However, there are cases when DPDK is and should be run as non-root,
without the need for virtual-to-physical address translations: a
typical example is when working with PCAP input/output. This patch
adds a start-time check for /proc/self/pagemap readability, and
directly returns an error code from rte_mem_virt2phy().
This way, there is only a one-time warning at startup instead of
constant warnings all the time.
Signed-off-by: Simon Kagstrom <simon.kagstrom@netinsight.net>
Signed-off-by: Johan Faltstrom <johan.faltstrom@netinsight.net>
A missing port from memcpy_toiovecend to copy_to_iter
is showed when vHost HDR is enabled. DPDK would not build.
This patch add this validation to build with kernel > 3.19.
Fixes: 45e63ba8db ("kni: fix vhost build with kernels 3.19 and 4.0")
Linux: ba7438aed924 ("vhost: don't bother copying iovecs in handle_rx(), kill memcpy_toiovecend()")
Signed-off-by: Miguel Bernal Marin <miguel.bernal.marin@linux.intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Parameters from sendmsg and recvmsg has been changed in 4.1 kernel.
The function pointers belong to proto_ops structure were updated removing
the struct kiocb parameter.
Linux: 1b784140474e ("net: Remove iocb argument from sendmsg and recvmsg")
Signed-off-by: Miguel Bernal Marin <miguel.bernal.marin@linux.intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
rebuild member was removed from headers_ops in kernel release
4.1. Therefore kni module compilation breaks.
This patch add the properly checks to fix it.
Linux: d476059e77d1 ("net: Kill dev_rebuild_header")
Signed-off-by: Miguel Bernal Marin <miguel.bernal.marin@linux.intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
ndo_bridge_getlink has changed in kernel release 4.1. It
adds new parameter which breaks compilation.
This patch add the properly checks to fix it.
Linux: 46c264daaaa5 ("bridge/nl: remove wrong use of NLM_F_MULTI")
Signed-off-by: Miguel Bernal Marin <miguel.bernal.marin@linux.intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Using IBM advance toolchain on Ubuntu 14.04 (package 8.0-3), gcc is complaining
about out of bound accesses.
CC eal_hugepage_info.o
lib/librte_eal/linuxapp/eal/eal_hugepage_info.c:
In function ‘eal_hugepage_info_init’:
lib/librte_eal/linuxapp/eal/eal_hugepage_info.c:350:35:
error: array subscript is above array bounds [-Werror=array-bounds]
internal_config.hugepage_info[j].hugepage_sz)
^
lib/librte_eal/linuxapp/eal/eal_hugepage_info.c:350:35:
error: array subscript is above array bounds [-Werror=array-bounds]
lib/librte_eal/linuxapp/eal/eal_hugepage_info.c:349:37:
error: array subscript is above array bounds [-Werror=array-bounds]
if (internal_config.hugepage_info[j-1].hugepage_sz <
^
lib/librte_eal/linuxapp/eal/eal_hugepage_info.c:350:35:
error: array subscript is above array bounds [-Werror=array-bounds]
internal_config.hugepage_info[j].hugepage_sz)
Looking at the code, these warnings are invalid from my pov and they disappeared
when upgrading the toolchain to new version (8.0-4).
However, the code was buggy (sorting code is wrong), so fix this by using qsort
and adding a check on num_sizes to avoid potential out of bound accesses.
Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
With this, we should be checkpatch compliant.
Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
Replace this while loop with a for loop and simplify error handling.
Indent is broken on purpose, fixed in next commit.
Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
Rather than cast the huge pages number returned by get_num_hugepages, rework
this function so that it returns 0 when something goes wrong.
And no need for casts in log.
Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
The code in eal_hugepage_info.c is not reachable by secondary processes.
Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
Added code that saves the pointers to the detached devices, during
driver loading, and during driver unloading, go over the list,
and re-attach them by calling device_probe_and_attach
on each device.
Signed-off-by: Raz Amir <razamir22@gmail.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
This patch removes CONFIG_RTE_LIBRTE_EAL_HOTPLUG option, and enables it
as default in both Linux and BSD.
Also, to support port hotplug, rte_eal_pci_scan() and below missing
symbols should be exported to ethdev library.
- rte_eal_parse_devargs_str()
- rte_eal_pci_close_one()
- rte_eal_pci_probe_one()
- rte_eal_pci_scan()
- rte_eal_vdev_init()
- rte_eal_vdev_uninit()
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
This patch consolidates below functions, and implements these in common
eal code.
- rte_eal_pci_probe_one_driver()
- rte_eal_pci_close_one_driver()
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
Acked-by: David Marchand <david.marchand@6wind.com>
The patch consolidates below functions, and implemented in common
eal code.
- pci_map_device()
- pci_unmap_device()
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
The patch consolidates below functions, and implemented in common
eal code.
- pci_map_resource()
- pci_unmap_resource()
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
This patch consolidates below structures, and defines them in common code.
- struct pci_map
- struct mapped_pci_resources
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
This patch adds a new function called pci_uio_map_resource_by_index().
The function hides how to map uio resource in linuxapp and bsdapp.
With the function, pci_uio_map_resource() will be more abstracted.
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
Acked-by: David Marchand <david.marchand@6wind.com>
This patch adds new functions called pci_uio_alloc_resource() and
pci_uio_free_resource().
The functions hides how to prepare or free uio resource in linuxapp
and bsdapp. With the function, pci_uio_map_resource() will be more
abstracted.
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
Acked-by: David Marchand <david.marchand@6wind.com>
This patch fixes below.
- bsdapp
- Use map_id in pci_uio_map_resource().
- Fix interface of pci_map_resource().
- Move path variable of mapped_pci_resource structure to pci_map.
- linuxapp
- Remove redundant error message of linuxapp.
'pci_uio_map_resource()' is implemented in both linuxapp and bsdapp,
but interface is different. The patch fixes the function of bsdapp
to do same as linuxapp. After applying it, file descriptor should be
opened and closed out of pci_map_resource().
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
To merge pci code of linuxapp and bsdapp, this patch changes names
like below.
- uio_map to pci_map
- uio_resource to mapped_pci_resource
- uio_res_list to mapped_pci_res_list
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
This patch fixes following memory leaks.
- When open() is failed, uio_res and fds won't be freed in
pci_uio_map_resource().
- When pci_map_resource() is failed but path is allocated correctly,
path and fds won't be freed in pci_uio_map_recource().
Also, some mapped resources should be freed.
- When pci_uio_unmap() is called, path should be freed.
Also, fixes below.
- When pci_map_resource() is failed, mapaddr will be MAP_FAILED.
In this case, pci_map_addr should not be incremented in
pci_uio_map_resource().
- To shrink code, move close().
- Remove fail variable.
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
When pci_uio_unmap_resource() is called, a file descriptor that is used
for uio configuration should be closed.
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
This patch fixes coding style of below files in linuxapp and bsdapp.
- eal_pci.c
- eal_pci_uio.c
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: David Marchand <david.marchand@6wind.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>
rte_prefetchX functions included volatile void *p as parameter,
but the function does not modify it,
so it should include the const keyword.
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Since the commit below includes rte_cpuflags.h in rte_spinlock.h,
compilation of the mlx4 driver fails when CONFIG_RTE_LIBRTE_MLX4_DEBUG=y.
This mode adds -pedantic to the compiler's command line for mlx4, which
complains about the static definition of an empty cpu_feature_table[] in
common rte_cpuflags.h, then about its redefinition as a larger array in
arch-specific rte_cpuflags.h.
While DPDK does not officially support -pedantic internally, external
applications may enable it and include rte_spinlock.h from the public API.
Instead of removing -pedantic from mlx4, this commit fixes rte_cpuflags.h.
Fixes: ba7468997e ("spinlock: add HTM lock elision for x86")
Signed-off-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
Acked-by: David Marchand <david.marchand@6wind.com>
On machines that are strict on pointer alignment, current code breaks
on GCC's -Wcast-align checks on casts from narrower to wider types.
This patch introduces new unaligned_uint(16|32|64)_t types, which
correctly retain alignment in such cases. Strict alignment
architectures will need to define CONFIG_RTE_ARCH_STRICT_ALIGN in
order to effect these new types.
Signed-off-by: Cyril Chemparathy <cchemparathy@ezchip.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Fixes: 88701645c9 ("eal: move interrupt type out of igb_uio")
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
The RTE_LOG(DEBUG, ...) messages in rte_eal_cpu_init() are printed
even when the log level on the command line was set to INFO or lower.
The problem is the rte_eal_cpu_init() routine was called before
the command line args are scanned. Setting --log-level=7 now
correctly does not print the messages from the rte_eal_cpu_init() routine.
Signed-off-by: Keith Wiles <keith.wiles@intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
In rte_log_dump_history() the log_history list is reinitialized without
resetting the log_history_size. In the next call to rte_log_add_in_history()
the log_history_size > RTE_LOG_HISTORY and the code unconditionally tries
to remove the first entry:
Program received signal SIGSEGV, Segmentation fault.
rte_log_add_in_history (
buf=buf@entry=0x7f02035cd000 "[snip]\n", size=size@entry=86)
at /usr/src/packages/BUILD/lib/librte_eal/common/eal_common_log.c:122
Signed-off-by: Jan Blunck <jblunck@infradead.org>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
This patch adds methods that use hardware memory transactions (HTM) on
fast-path for rwlock (a.k.a. lock elision). Here the methods are implemented
for x86 using Restricted Transactional Memory instructions (Intel(r)
Transactional Synchronization Extensions). The implementation fall-backs to
the normal rwlock if HTM is not available or memory transactions fail. This is
not a replacement for all rwlock usages since not all critical sections
protected by locks are friendly to HTM. For example, an attempt to perform
a HW I/O operation inside a hardware memory transaction always aborts
the transaction since the CPU is not able to roll-back should the transaction
fail. Therefore, hardware transactional locks are not advised to be used around
rte_eth_rx_burst() and rte_eth_tx_burst() calls.
Signed-off-by: Roman Dementiev <roman.dementiev@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
This patch adds methods that use hardware memory transactions (HTM) on fast-path
for spinlocks (a.k.a. lock elision). Here the methods are implemented for x86
using Restricted Transactional Memory instructions (Intel(r) Transactional
Synchronization Extensions). The implementation fall-backs to the normal
spinlock if HTM is not available or memory transactions fail. This is not
a replacement for all spinlock usages since not all critical sections protected
by spinlocks are friendly to HTM. For example, an attempt to perform a HW I/O
operation inside a hardware memory transaction always aborts the transaction
since the CPU is not able to roll-back should the transaction fail.
Therefore, hardware transactional locks are not advised to be used around
rte_eth_rx_burst() and rte_eth_tx_burst() calls.
Signed-off-by: Roman Dementiev <roman.dementiev@intel.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
In containers like docker, current->pid returns current process's global
PID instead of its own PID under containers's PID namespace, and
get_net_ns_by_pid() suppose to accept a virtual PID under its own
namespace, so we should use task_pid_vnr(current) to get current process's
virtual PID instead of current->pid.
Signed-off-by: Wenfeng Liu <liuwf@arraynetworks.com.cn>
Acked-by: Helin Zhang <helin.zhang@intel.com>
We did some (very basic) tests with IGMP, which involves adding
multicast addresses to ETH interfaces. This is done via the ip tool,
an example can be found on e.g.,
http://superuser.com/questions/324824/linux-built-in-or-open-source-program-to-join-multicast-group
and this will fail on KNI interfaces because of an unimplemented ioctl
SIOCADDMULTI. The patch simply adds an empty callback for set_rx_mode
(typically used for setting up hardware) so that the ioctl succeeds.
This is the same thing as the Linux tap interface does.
Signed-off-by: Simon Kagstrom <simon.kagstrom@netinsight.net>
Signed-off-by: Johan Faltstrom <johan.faltstrom@netinsight.net>
Reviewed-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Loop processing packets dequeued from rx_q was using the number of
packets requested, not how many it actually received.
Variable rename to make code a little more clear
Signed-off-by: Jay Rolette <rolette@infiniteio.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
No reason to check out many entries are in kni->rx_q prior to
actually pulling them from the fifo. You can't dequeue more than
are there anyway. Max entries to dequeue is either the max batch
size or however much space is available on kni->free_q (lesser of the two).
Signed-off-by: Jay Rolette <rolette@infiniteio.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Do not need the 'safe' version of list_for_each_entry() if you are
not deleting from the list as you iterate over it.
Signed-off-by: Jay Rolette <rolette@infiniteio.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Implement .ndo_change_carrier to enable
DPDK applications to propagate link state changes to
kni virtual interfaces through sysfs
Signed-off-by: Vijayakumar Muthuvel Manickam <mmvijay@gmail.com>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Needed to run as non-root but with higher memory allocations, and
removes a constraint on no-huge mode being limited to 64M. A usage
example is if running with file input with the pcap PMD, which can be
done as non-root after this patch via e.g.,
./test-dpdk --no-huge -m 1024 -l 0,1 -n3
--vdev 'eth_pcap0,rx_pcap=eth-rx.pcap,tx_pcap=eth-tx.pcap'
Signed-off-by: Simon Kagstrom <simon.kagstrom@netinsight.net>
Signed-off-by: Johan Faltstrom <johan.faltstrom@netinsight.net>
Acked-by: David Marchand <david.marchand@6wind.com>
Ran this code base through a script which:
- removes trailing whitespace
- removes space before tabs
- removes blank lines at end of file
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Helin Zhang <helin.zhang@intel.com>
Using the "physical_package_id" as a fallback for determining the
numa node of a core tends to be unreliable. Fix this by using a
detection routine which reads the numa information from
/sys/devices/system/node and just returns a numa node of 0 on
failure.
Reported-by: Wang Sheng-Hui <shhuiw@gmail.com>
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Stephen Hemminger <stephen@networkplumber.org>
On Fedora 22, with GCC 5.1, errors are reported due to array accesses
being potentially out of bounds. This commit fixes this by ensuring the
bounds check in the loop takes account of the array size.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
rte_pci.h depends upon stdio.h for the definition of the FILE type. Add
in #include <stdio.h> to the file to satisfy this dependency in cases
where the including C file does not already include stdio.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Marc Sune <marc.sune@bisdn.de>
Move xenvirt PMD to drivers/net directory
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Move ring PMD to drivers directory
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Move pcap pmd to drivers/net directory
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
move af_packet pmd to drivers/net directory
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: John McNamara <john.mcnamara@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
The introduction of uio_pci_generic broke interrupt handling with
igb_uio. The igb_uio device uses the kernel read/write method to
enable disable IRQ's; the uio_pci_generic has to use PCI intx
config read/write to enable disable interrupts.
Since igb_uio uses MSI-X the PCI intx config read/write won't
work.
Fixes: c112df6875 ("eal/linux: toggle interrupt for uio_pci_generic")
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Set internal event file descriptor to be non-block and not
inherited across exec. This prevents accidental hangs and
passing in another thread.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
Compilation on FreeBSD with clang was broken, giving the error message:
lib/librte_eal/bsdapp/eal/eal_pci.c:438:16: fatal error: assigning to
'struct rte_pci_id *' from 'const struct rte_pci_id *' discards qualifiers
[-Wincompatible-pointer-types-discards-qualifiers]
for (id_table = dr->id_table ; id_table->vendor_id != 0; id_table++) {
^ ~~~~~~~~~~~~
This patch fixes the issue by adding "const" to the type of id_table.
Fixes: 6065355a03 ("pci: make device id tables const")
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
This patch adds support for enic in the nic_uio driver so that enic
could be used on FreeBSD.
Signed-off-by: Sujith Sankar <ssujith@cisco.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
Due to commit c0371da6 in kernel 3.19, which removed msg_iov
and msg_iovlen from struct msghdr, DPDK would not build.
Also, functions memcpy_toiovecend and memcpy_fromiovecend
were removed in commits ba7438ae and 57dd8a07, being substituted by
copy_from_iter and copy_to_iter.
This patch makes use of struct iov_iter, which has references
to msg_iov and msg_iovln, and makes use of copy_from_iter
and copy_to_iter.
Reported-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Due to API changes in function pointer ndo_bridge_setlink
(commit ad41faa8) and the rename of functions vlan_tx_*
(commit df8a39de) in kernel 4.0, DPDK would not build.
This patch adds the properly checks to fix the compilation.
Reported-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Closing /dev/io fd causes SIGBUS in inb/outb instructions
as the process loses the IOPL privileges once the fd is closed:
(gdb) bt
0 0x0000000000492f2c in outb (port=49170, data=0 '\000')
at /usr/include/machine/cpufunc.h:244
1 0x0000000000492f7a in outb_p (data=0 '\000', port=49170)
at /dpdk/dpdk-2.0.0/lib/librte_pmd_virtio/virtio_pci.h:211
2 0x000000000049328d in vtpci_set_status (hw=0x80331f380, status=0 '\000')
at /dpdk/dpdk-2.0.0/lib/librte_pmd_virtio/virtio_pci.c:130
3 0x00000000004931fe in vtpci_reset (hw=0x80331f380)
at /dpdk/dpdk-2.0.0/lib/librte_pmd_virtio/virtio_pci.c:108
4 0x00000000004a175e in eth_virtio_dev_init (eth_dev=0x831b80 <rte_eth_devices>)
at /dpdk/dpdk-2.0.0/lib/librte_pmd_virtio/virtio_ethdev.c:1150
5 0x0000000000462c09 in rte_eth_dev_init (pci_drv=0x79d1a0 <rte_virtio_pmd>,
pci_dev=0x802417560) at /dpdk/dpdk-2.0.0/lib/librte_ether/rte_ethdev.c:326
6 0x000000000046f03f in rte_eal_pci_probe_one_driver (dr=0x79d1a0 <rte_virtio_pmd>,
dev=0x802417560) at /dpdk/dpdk-2.0.0/lib/librte_eal/bsdapp/eal/eal_pci.c:487
7 0x0000000000475b06 in pci_probe_all_drivers (dev=0x802417560)
at /dpdk/dpdk-2.0.0/lib/librte_eal/common/eal_common_pci.c:116
8 0x0000000000475bb9 in rte_eal_pci_probe ()
at /dpdk/dpdk-2.0.0/lib/librte_eal/common/eal_common_pci.c:246
9 0x000000000046cd63 in rte_eal_init (argc=5, argv=0x7fffffffeaf0)
at /dpdk/dpdk-2.0.0/lib/librte_eal/bsdapp/eal/eal.c:554
10 0x0000000000404544 in main ()
Fixes: 8a312224bc ("eal/bsd: fix fd leak")
Signed-off-by: Raz Amir <razamir22@gmail.com>
Reviewed-by: Bruce Richardson <bruce.richardson@intel.com>
According to the api, rte_log() / rte_vlog() are supposed to check the log level
and type but they were not doing so. This check was only done in the RTE_LOG
macro while this macro is only there to remove log messages at build time.
rte_log() always calls rte_vlog(), so do the check in rte_vlog() only.
Signed-off-by: Jean Dao <jean.dao@6wind.com>
Signed-off-by: David Marchand <david.marchand@6wind.com>
The old device id: IXGBE_DEV_ID_X550EM_X is split into 2 new device id:
IXGBE_DEV_ID_X550EM_X_10G_T and IXGBE_DEV_ID_X550EM_X_1G_T
Signed-off-by: Changchun Ouyang <changchun.ouyang@intel.com>
Acked-by: Jijiang Liu <jijiang.liu@intel.com>
probe and close both don't modify the rte_pci_addr structure
that is passed.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
The PCI device id table is immutable and should be made const
in all drivers. The pseudo drivers can initialize their local
copy as necessary.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Kernel driver (kdrv) seems easier to understand than
passthrough driver (pt_driver). It's also more generic
as a PMD could run on top of any PCI kernel driver if
it would offer such support.
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Michael Qiu <michael.qiu@intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
eal options OPT_CREATE_UIO_DEV does not need argument so set it to zero.
It needs to reset create_uio_dev explicitly.
Fixes: f7f97c1604 ("pci: add option --create-uio-dev to run without hotplug")
Signed-off-by: Haifeng Gao <gaohaifeng.gao@huawei.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Due to API changes in functions ndo_dflt_bridge_getlink
(commit 2c3c031c) and ndo_fdb_add (commit f6f6424b)
in kernel 3.19, DPDK would not build.
This patch solves the problem, by checking the kernel version
and adding the necessary new parameters.
Mind that function igb_ndo_fdb_add does not need the extra parameter
if USE_CONST_DEV_UC_CHAR is not set, since that macro is only defined
when kernel is greater or equal than 3.7
Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
In Suse11 SP3, there'll be errors for not found sse3 functions.
rte_memcpy.h: In function ‘rte_memcpy’:
rte_memcpy.h:625: error: implicit declaration of function ‘_mm_alignr_epi8’
rte_memcpy.h:625: error: nested extern declaration of ‘_mm_alignr_epi8’
rte_memcpy.h:625: error: incompatible type for argument 2 of ‘_mm_storeu_si128’
These functions defined in tmmintrin.h and should be included in.
Fixes: 9144d6bcde ("eal/x86: optimize memcpy for SSE and AVX")
Signed-off-by: Marvin Liu <yong.liu@intel.com>
Function pread need macro _XOPEN_SOURCE be defined.
Add _GNU_SOURCE will fix this issue.
error: implicit declaration of function ‘pread’
Fixes: 4a499c6495 ("eal/linux: enable uio_pci_generic support")
Signed-off-by: Marvin Liu <yong.liu@intel.com>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
When including rte_version.h without string.h, there is a compilation error:
include/rte_version.h: error: implicit declaration of function ‘strlen’
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Fix a warning when the rte_common.h header is included in a compilation
using -Wbad-function-cast, such as in Open vSwitch where the
following warning is emitted repeatedly:
../rte_common.h: In function 'rte_is_aligned':
../rte_common.h:184:9: warning: cast from function call of
type 'uintptr_t' to non-matching type 'void *' [-Wbad-function-cast]
This change fixes the issue in rte_common.h by using the RTE_ALIGN_FLOOR
macro to get the aligned floor value with generic type casting.
Also removed the rte_align_floor_int() function and replaced it with
the RTE_PTR_ALIGN_FLOOR() macro.
Signed-off-by: John McNamara <john.mcnamara@intel.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
x86intrin.h cannot be directly included as it is not
always available.
rte_common_vect.h handles compiler differences.
Suggested-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Commit a2348166ea ("tailq: move to dynamic tailq") introduced a bug in
uio/vfio resources list init.
These resources list were pointed at through a pointer initialised only once but
too early in the eal init (before tailqs init).
Fix this by "resolving" this pointer when used (which is well after tailqs
init).
Fixes: a2348166ea ("tailq: move to dynamic tailq")
Reported-by: Marvin Liu <yong.liu@intel.com>
Reported-by: Tetsuya Mukawa <mukawa@igel.co.jp>
Signed-off-by: David Marchand <david.marchand@6wind.com>
Tested-by: John McNamara <john.mcnamara@intel.com>
Tested-by: Tetsuya Mukawa <mukawa@igel.co.jp>
The contigmem module was using an "int" type for specifying the
size of blocks of memory to be reserved. A 2GB block was therefore
overflowing the signed 32-bit value, making 1GB the largest block
size that could be reserved as a single unit.
The fix is to change the type used for the buffer/block size to
an "int64_t" value.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
There is no remaining reference to E_RTE_NO_TAILQ.
Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
No static entry remaining, the rte_tailq api is for "internal use" only, get rid
of the static slots.
Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
This register system makes it possible to reserve a tailq for the dpdk
libraries.
The "dynamic" tailqs are right after the "static" tailqs in shared mem.
Primary process is responsible for writing the tailq names, so that secondary
processes can find them.
This is a temp commit, "static" tailqs are removed after conversion of all
users in next commits.
Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
A lot of places just protect against concurrent access and I can not see the
gain of having those macros.
Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
The "reserve" macros and functions do not check if the requested entry is free.
They do nothing more than the lookup function (which itself "creates" entries
...).
The rte_tailq api is marked as "internal use" in documentation and these macros
are only used in test application, so just get rid of them.
Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
No need to cast everywhere, define a common macro for this, plus it can be used
in future commits.
Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
There is no reason why we should use the "reserve" tailq api, since the pci
entry is already statically reserved.
Signed-off-by: David Marchand <david.marchand@6wind.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
include/rte_cpuflags.h:154:2: error: redeclaration of enumerator ‘REG_EAX’
In i686, from REG_EAX to REG_EDX are all defined in
/usr/include/sys/ucontext.h
Rename to RTE_REG_EAX to avoid this issue.
Signed-off-by: Michael Qiu <michael.qiu@intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
When compiling C++11-code or above (--std=c++11), the build fails with
lots of
rte_eth_ctrl.h:517:3: note: in expansion of macro RTE_ALIGN
(RTE_ALIGN(RTE_ETH_FLOW_MAX, UINT32_BIT)/UINT32_BIT)
^
When reading the GCC info pages, I get the feeling that __typeof__ is
a better choice, and that indeed works when including the headers in
C++ files (--std=c++11).
Signed-off-by: Simon Kagstrom <simon.kagstrom@netinsight.net>
Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
./i686-native-linuxapp-gcc/include/rte_memcpy.h:592:23: error:
cast from pointer to integer of different size
[-Werror=pointer-to-int-cast]
dstofss = 16 - (int)((long long)(void *)dst & 0x0F) + 16;
Type 'long long' is 64-bit in i686 platform while 'void *'
is 32-bit.
Signed-off-by: Michael Qiu <michael.qiu@intel.com>
Signed-off-by: Zhihong Wang <zhihong.wang@intel.com>
The path variable is set via snprintf, and does not need to
memset before that.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
The mempool header forces error on -Wcast-qual:
error: cast discards ‘const’ qualifier from pointer target type
Let's fix it by removing const qualifier of pci driver from commit
5e9f6d1340 ("pci: reference driver structure for each device")
It's needed because the driver flags are changed depending on using uio or not.
Actually these driver flags should be directly attached to each device.
Fixes: da978dfdc4 ("virtio: use port IO to get PCI resource")
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: Changchun Ouyang <changchun.ouyang@intel.com>
Compilation fails in some distributions because of missing unistd.h
needed for pread/pwrite (seen with Suse):
lib/librte_eal/linuxapp/eal/eal_pci_uio.c:62:2:
error: implicit declaration of function ‘pread’
Fixes: 4a499c6495 ("eal/linux: enable uio_pci_generic support")
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: David Marchand <david.marchand@6wind.com>
On failure devargs->args should not be accessed if devargs is NULL.
Fixes: c07691ae10 ("devargs: remove limit on parameters length")
Signed-off-by: Pawel Wodkowski <pawelx.wodkowski@intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
Theres no need to export this variable. Its set and queried from an API call
that doesn't exist in the hot path. Instead just export the rte_socket_id
symbol and make the variable private to protect it from type changes. We should
do this with the other exported variables too, but I think its too late in the
release cycle to do that.
tested using distributor_autotest (which uses rte_socket_id), successfully.
Only tested on linux, as I don't currently have a bsd system spun up, but the
changes are symmetric, and should be fine
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Acked-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
These functions are used for attaching or detaching a port.
When rte_eth_dev_attach() is called, the function tries to realize the
device name as pci address. If this is done successfully,
rte_eth_dev_attach() will attach physical device port. If not, attaches
virtual devive port.
When rte_eth_dev_detach() is called, the function gets the device type
of this port to know whether the port is come from physical or virtual.
And then specific detaching function will be called.
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
To remove assumption, do like followings.
This patch adds "RTE_PCI_DRV_DETACHABLE" to drv_flags of rte_pci_driver
structure. The flags indicate the driver can detach devices at runtime.
Also, remove assumption that port will not be detached.
To remove the assumption.
- Add 'attached' member to rte_eth_dev structure.
This member is used for indicating the port is attached, or not.
DEV_ATTACHED indicates a port is attached.
DEV_DETACHED indicates a port is detached.
- Add rte_eth_dev_allocate_new_port().
This function is used for allocating new port.
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
Acked-by: Thomas Monjalon <thomas.monjalon@6wind.com>
The patch adds following functions.
- rte_eal_vdev_init();
- rte_eal_vdev_uninit();
- rte_eal_parse_devargs_str().
These functions are used for driver initialization and finalization.
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
- Add pci_close_all_drivers()
The function tries to find a driver for the specified device, and
then close the driver.
- Add rte_eal_pci_probe_one() and rte_eal_pci_close_one()
The functions are used for probe and close a device.
First the function tries to find a device that has the specified
PCI address. Then, probe or close the device.
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
The patch adds functions for unmapping igb_uio resources. The patch is only
for Linux and igb_uio environment. VFIO and BSD are not supported.
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
This patch replaces pci_addr_comparison() and memcmp() of pci addresses by
rte_eal_compare_pci_addr().
To compare PCI addresses, rte_eal_compare_pci_addr() doesn't use memcmp().
This is because sizeof(struct rte_pci_addr) returns 6, but actually
this structure is like below.
struct rte_pci_addr {
uint16_t domain; /**< Device domain */
uint8_t bus; /**< Device bus */
uint8_t devid; /**< Device ID */
uint8_t function; /**< Device function. */
};
If the structure is dynamically allocated in a function without bzero,
last 1 byte may have value. As a result, memcmp may not work.
To avoid such a case, rte_eal_compare_pci_addr() compare following values.
dev_addr = (addr->domain << 24) | (addr->bus << 16) |
(addr->devid << 8) | addr->function;
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
With the driver type flag in struct rte_pci_dev, we do not need
to always map uio devices with vfio related function when
vfio enabled.
Signed-off-by: Michael Qiu <michael.qiu@intel.com>
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>
Currently, dpdk has no ability to know which type of driver(
vfio-pci/igb_uio/uio_pci_generic) the device used. It only can
check whether vfio is enabled or not statically.
It really useful to have the flag, because different type need to
handle differently in runtime. For example, pci memory map,
pot hotplug, and so on.
This patch add a flag field for pci device to solve above issue.
Signed-off-by: Michael Qiu <michael.qiu@intel.com>
Signed-off-by: Tetsuya Mukawa <mukawa@igel.co.jp>
Acked-by: Bernard Iremonger <bernard.iremonger@intel.com>
Main code changes:
1. Differentiate architectural features based on CPU flags
a. Implement separated move functions for SSE/AVX/AVX2 to make full utilization of cache bandwidth
b. Implement separated copy flow specifically optimized for target architecture
2. Rewrite the memcpy function "rte_memcpy"
a. Add store aligning
b. Add load aligning based on architectural features
c. Put block copy loop into inline move functions for better control of instruction order
d. Eliminate unnecessary MOVs
3. Rewrite the inline move functions
a. Add move functions for unaligned load cases
b. Change instruction order in copy loops for better pipeline utilization
c. Use intrinsics instead of assembly code
4. Remove slow glibc call for constant copies
Test report: http://dpdk.org/ml/archives/dev/2015-January/011848.html
Signed-off-by: Zhihong Wang <zhihong.wang@intel.com>
Tested-by: Jingguo Fu <jingguox.fu@intel.com>
Reviewed-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Acked-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
As per_lcore__socket_id and rte_sys_gettid are missing in version map,
it causes compiling error when CONFIG_RTE_BUILD_SHARED_LIB is enabled.
Fixes: ef76436c68 ("eal: get unique thread id")
Fixes: 9e29251b2a ("eal: thread affinity API")
Reported-by: Tetsuya Mukawa <mukawa@igel.co.jp>
Signed-off-by: Cunming Liang <cunming.liang@intel.com>
Tested-by: Tetsuya Mukawa <mukawa@igel.co.jp>
Acked-by: John McNamara <john.mcnamara@intel.com>
In pci_uio_map_resource we check that we are in a primary process
before calling pci_uio_set_bus_master. However, there is already
an earlier check which means that we are always in a primary instance
at this point in the code, so the check can be removed.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
Rather than scanning the resource file in sysfs a second time, we
can pull the information on physical addresses of BARs from the
pci resource information already present in the dev structure.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
Instead of distinguishing the BAR mappings via offset within a single
file, originally /dev/uioX, switch to mapping each individual bar via
the appropriately numbered resourceX file.
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: David Marchand <david.marchand@6wind.com>
Following commit c07691ae10, an implicit change has been done in the
devargs API.
This triggers problem in virtual pmds that did not check for parameters
validity as it was implicitely valid.
Fix this by restoring the empty argument as "" and add a note in the api.
Restore associated tests.
Fixes: c07691ae10 ("devargs: remove limit on parameters length")
Reported-by: Tetsuya Mukawa <mukawa@igel.co.jp>
Signed-off-by: David Marchand <david.marchand@6wind.com>
Tested-by: Tetsuya Mukawa <mukawa@igel.co.jp>
In non-EAL thread, lcore_id always be LCORE_ID_ANY.
It can't be used as unique id for recursive spinlock.
Then use rte_gettid() to replace it.
Signed-off-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
For those non-EAL thread, *_lcore_id* is invalid and probably larger than RTE_MAX_LCORE.
The patch adds the check and allows only EAL thread using EAL per thread log level and log type.
Others shares the global log level.
Signed-off-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Set _lcore_id and _socket_id to (-1) by default.
For those non EAL thread, _lcore_id shall always be LCORE_ID_ANY.
The libraries using _lcore_id as index need to take care.
_socket_id always be SOCKET_ID_ANY until the thread changes the affinity
by rte_thread_set_affinity().
Signed-off-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
EAL threads use assigned cpuset to set core affinity during startup.
It keeps 1:1 mapping, if no '--lcores' option is used.
Signed-off-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
1. add two TLS *_socket_id* and *_cpuset*
2. add one internal API, eal_cpu_socket_id/eal_thread_dump_affinity
3. add two public API, rte_thread_set/get_affinity
4. update EAL version map for EAL public API
The API works for both EAL thread and non EAL thread.
When calling rte_thread_set_affinity, the *_socket_id* and
*_cpuset* of calling thread will be updated if the thread
successfully set the cpu affinity.
Signed-off-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
The rte_gettid() wraps the linux and freebsd syscall gettid().
It provides a persistent unique thread id for the calling thread.
It will save the unique id in TLS on the first time.
Signed-off-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
It defines eal_cpu_socket_id() which exposing the origin private cpu_socket_id().
The function is only used inside EAL. It returns socket_id of the specified cpu_id.
Signed-off-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
The problem is that strnlen() here may return invalid value with 32bit icc.
(actually it returns it’s second parameter,e.g: sysconf(_SC_ARG_MAX)).
It starts to manifest hwen max_len parameter is > 2M and using icc –m32 –O2 (or above).
Suggested-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Signed-off-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
It supports one new eal long option '--lcores' for EAL thread cpuset assignment.
The format pattern:
--lcores='<lcores[@cpus]>[<,lcores[@cpus]>...]'
lcores, cpus could be a single digit/range or a group.
'(' and ')' are necessary if it's a group.
If not supply '@cpus', the value of cpus uses the same as lcores.
e.g. '1,2@(5-7),(3-5)@(0,2),(0,6),7-8' means starting 9 EAL thread as below
lcore 0 runs on cpuset 0x41 (cpu 0,6)
lcore 1 runs on cpuset 0x2 (cpu 1)
lcore 2 runs on cpuset 0xe0 (cpu 5,6,7)
lcore 3,4,5 runs on cpuset 0x5 (cpu 0,2)
lcore 6 runs on cpuset 0x41 (cpu 0,6)
lcore 7 runs on cpuset 0x80 (cpu 7)
lcore 8 runs on cpuset 0x100 (cpu 8)
Test report: http://dpdk.org/ml/archives/dev/2015-February/013383.html
Signed-off-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Tested-by: Qun Wan <qun.wan@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
The patch adds 'cpuset' into per-lcore configure 'lcore_config[]',
as the lcore no longer always 1:1 pinning with physical cpu.
The lcore now stands for a EAL thread rather than a logical cpu.
It doesn't change the default behavior of 1:1 mapping, but allows to
affinity the EAL thread to multiple cpus.
Signed-off-by: Cunming Liang <cunming.liang@intel.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
Help is printed with -h or --help.
Help is also printed for an unknown option.
This was broken since the rework of options.
Fixes: 489a9d6c9f ("merge bsd and linux common options parsing")
Signed-off-by: Thomas Monjalon <thomas.monjalon@6wind.com>
Acked-by: David Marchand <david.marchand@6wind.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
