dpdk.git
6 years agoeal: bring forward init of interrupt handling
Jianfeng Tan [Tue, 26 Jun 2018 10:53:16 +0000 (11:53 +0100)]
eal: bring forward init of interrupt handling

Next commit will make asynchronous IPC requests rely on alarm API,
which in turn relies on interrupts to work. Therefore, move the EAL
interrupt initialization before IPC initialization to avoid breaking
IPC in the next commit.

Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com>
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agoeal/bsd: support alarm API
Anatoly Burakov [Tue, 26 Jun 2018 10:53:15 +0000 (11:53 +0100)]
eal/bsd: support alarm API

Implement EAL alarm API support for FreeBSD. The implementation
is largely identical to that of Linux version, with one key
difference.

The alarm API is a little Linux-centric in that it is expecting
the alarm API to manage alarm timeouts without involvement of the
interrupt thread. This works on Linux because in Linux, there's
timerfd API which allows waiting for timer events on an fd.

On FreeBSD, however, there are no timerfd's, and timer events are
set up directly in kevent. There is no way to pass information from
the alarm API to the interrupt thread, so we also add a little
back-channel magic to get soonest alarm timeout from the alarm API.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agoeal/bsd: add interrupt thread
Anatoly Burakov [Tue, 26 Jun 2018 10:53:14 +0000 (11:53 +0100)]
eal/bsd: add interrupt thread

Add interrupt thread to FreeBSD. It is largely a copy-paste from
Linuxapp interrupt thread, except for a few key differences:

* Use kevent instead of epoll
* Do not recreate the event queue on adding/removing interrupt
  sources, add/remove them to/from the queue on the fly instead
* No support for UIO/VFIO handles

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agoeal/linux: use libc malloc in interrupt handling
Jianfeng Tan [Tue, 26 Jun 2018 10:53:13 +0000 (11:53 +0100)]
eal/linux: use libc malloc in interrupt handling

IPC uses interrupts API internally, and memory subsystem uses IPC.
Therefore, IPC should not use rte_malloc to avoid circular dependency.
Switch to using regular glibc malloc in interrupts API.

Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com>
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agoeal/linux: use libc malloc in alarm
Jianfeng Tan [Tue, 26 Jun 2018 10:53:12 +0000 (11:53 +0100)]
eal/linux: use libc malloc in alarm

Alarm API is going to be used by IPC internally. However, because
memory subsystem depends on IPC, alarm API cannot use rte_malloc as
it creates a circular dependency.

To avoid such chicken and egg problem, we change to use glibc malloc
in the alarm API.

Signed-off-by: Jianfeng Tan <jianfeng.tan@intel.com>
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agovfio: fix uninitialized variable
Anatoly Burakov [Fri, 1 Jun 2018 09:08:12 +0000 (10:08 +0100)]
vfio: fix uninitialized variable

Some static analyzers complain about it, even though
value is never used if not initialized. To avoid additional
false positives about a potential null-pointer dereferences,
also add a null-check.

Bugzilla ID: 58
Fixes: ea2dc1066870 ("vfio: add multi container support")
Cc: stable@dpdk.org
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agoeal/linux: fix uninitialized value
Anatoly Burakov [Fri, 1 Jun 2018 09:08:11 +0000 (10:08 +0100)]
eal/linux: fix uninitialized value

The value is not used, but some static analyzers may give out a
warning. Fix it by assigning default value of zero.

Bugzilla ID: 58
Fixes: cdc242f260e7 ("eal/linux: support running as unprivileged user")
Cc: stable@dpdk.org
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agoeal/linux: fix invalid syntax in interrupts
Anatoly Burakov [Fri, 1 Jun 2018 09:08:10 +0000 (10:08 +0100)]
eal/linux: fix invalid syntax in interrupts

Parentheses were missing. It worked because macro is enclosed in
parentheses, so syntax was valid after macro expansion.

Bugzilla ID: 58
Fixes: 0a45657a6794 ("pci: rework interrupt handling")
Cc: stable@dpdk.org
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agoeal: add option to limit memory allocation on sockets
Anatoly Burakov [Thu, 31 May 2018 17:35:33 +0000 (18:35 +0100)]
eal: add option to limit memory allocation on sockets

Previously, it was possible to limit maximum amount of memory
allowed for allocation by creating validator callbacks. Although a
powerful tool, it's a bit of a hassle and requires modifying the
application for it to work with DPDK example applications.

Fix this by adding a new parameter "--socket-limit", with syntax
similar to "--socket-mem", which would set per-socket memory
allocation limits, and set up a default validator callback to deny
all allocations above the limit.

This option is incompatible with legacy mode, as validator callbacks
are not supported there.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agomemzone: improve zero-length reserve
Anatoly Burakov [Thu, 31 May 2018 09:51:01 +0000 (10:51 +0100)]
memzone: improve zero-length reserve

Currently, reserving zero-length memzones is done by looking at
malloc statistics, and reserving biggest sized element found in those
statistics. This has two issues.

First, there is a race condition. The heap is unlocked between the
time we check stats, and the time we reserve malloc element for memzone.
This may lead to inability to reserve the memzone we wanted to reserve,
because another allocation might have taken place and biggest sized
element may no longer be available.

Second, the size returned by malloc statistics does not include any
alignment information, which is worked around by being conservative and
subtracting alignment length from the final result. This leads to
fragmentation and reserving memzones that could have been bigger but
aren't.

Fix all of this by using earlier-introduced operation to reserve
biggest possible malloc element. This, however, comes with a trade-off,
because we can only lock one heap at a time. So, if we check the first
available heap and find *any* element at all, that element will be
considered "the biggest", even though other heaps might have bigger
elements. We cannot know what other heaps have before we try and
allocate it, and it is not a good idea to lock all of the heaps at
the same time, so, we will just document this limitation and
encourage users to reserve memzones with socket id properly set.

Also, fixup unit tests to account for the new behavior.

Fixes: fafcc11985a2 ("mem: rework memzone to be allocated by malloc")

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agomalloc: allow reserving biggest element
Anatoly Burakov [Thu, 31 May 2018 09:51:00 +0000 (10:51 +0100)]
malloc: allow reserving biggest element

Add an internal-only function to allocate biggest element from
the heap. Nominally, it supports SOCKET_ID_ANY as its socket
argument, but it's essentially useless because other sockets
will only be allocated from if the entire heap on current or
specified socket is busy.

Still, asking to reserve a biggest element will allow fixing
race condition in memzone reserve that has been there for a
long time.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Remy Horton <remy.horton@intel.com>
6 years agomalloc: add finding biggest free IOVA-contiguous element
Anatoly Burakov [Thu, 31 May 2018 09:50:59 +0000 (10:50 +0100)]
malloc: add finding biggest free IOVA-contiguous element

Adding internal-only function to find biggest free IOVA-contiguous
malloc element. This is not exposed to external API.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Remy Horton <remy.horton@intel.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
6 years agomalloc: fix pad erasing
Anatoly Burakov [Thu, 31 May 2018 17:05:40 +0000 (18:05 +0100)]
malloc: fix pad erasing

Previously, when joining adjacent free elements, we were erasing
trailer and header, but did not erase the padding. Fix this by
accounting for padding on erase, and do not erase padding twice
by adjusting data pointer and data len to not include padding.

Fixes: bb372060dad4 ("malloc: make heap a doubly-linked list")
Cc: stable@dpdk.org
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agomem: provide thread-unsafe memseg list walk variant
Anatoly Burakov [Tue, 12 Jun 2018 09:46:16 +0000 (10:46 +0100)]
mem: provide thread-unsafe memseg list walk variant

Sometimes, user code needs to walk memseg list while being inside
a memory-related callback. Rather than making everyone copy around
the same iteration code and depending on DPDK internals, provide an
official way to do memseg_list_walk() inside callbacks.

Also, remove existing reimplementation from memalloc code and use
the new API instead.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agomem: provide thread-unsafe memseg walk variant
Anatoly Burakov [Tue, 12 Jun 2018 09:46:15 +0000 (10:46 +0100)]
mem: provide thread-unsafe memseg walk variant

Sometimes, user code needs to walk memseg list while being inside
a memory-related callback. Rather than making everyone copy around
the same iteration code and depending on DPDK internals, provide an
official way to do memseg_walk() inside callbacks.

Also, remove existing reimplementation from sPAPR VFIO code and use
the new API instead.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agomem: provide thread-unsafe contig walk variant
Anatoly Burakov [Tue, 12 Jun 2018 09:46:14 +0000 (10:46 +0100)]
mem: provide thread-unsafe contig walk variant

Sometimes, user code needs to walk memseg list while being inside
a memory-related callback. Rather than making everyone copy around
the same iteration code and depending on DPDK internals, provide an
official way to do memseg_contig_walk() inside callbacks.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agomem: mark pages as freeable on exit
Anatoly Burakov [Thu, 31 May 2018 16:11:47 +0000 (17:11 +0100)]
mem: mark pages as freeable on exit

When rte_eal_cleanup() is called, it is expected that DPDK will be able to
release all of its memory back to the system. However, if pages are marked
as unfreeable, the pages will not be released back. Fix this to mark all
pages as freeable on calling rte_eal_cleanup(), but only do it for primary
process, as secondaries can come and go.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agomem: allocate in reverse to reduce fragmentation
Anatoly Burakov [Mon, 11 Jun 2018 20:55:42 +0000 (21:55 +0100)]
mem: allocate in reverse to reduce fragmentation

Currently, all hugepages are allocated from lower VA address to
higher VA address, while malloc heap allocates from higher VA
address to lower VA address. This results in heap fragmentation
over time due to multiple reserves leaving small space below the
allocated elements.

Fix this by allocating VA memory from the top, thereby reducing
fragmentation and lowering overall memory usage.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agotest/fbarray: add autotests
Anatoly Burakov [Mon, 11 Jun 2018 20:55:41 +0000 (21:55 +0100)]
test/fbarray: add autotests

Introduce a suite of autotests to cover functionality of fbarray.
This will check for invalid parameters, check API return values and
errno codes, and will also do some basic functionality checks on the
indexing code.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agofbarray: add reverse finding of contiguous
Anatoly Burakov [Mon, 11 Jun 2018 20:55:40 +0000 (21:55 +0100)]
fbarray: add reverse finding of contiguous

Add a function to return starting point of current contiguous
block, going backwards. All semantics are kept the same as the
existing function, with the only difference being that given the
same input, results will be returned in reverse order.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agofbarray: add reverse finding of chunk
Anatoly Burakov [Mon, 11 Jun 2018 20:55:39 +0000 (21:55 +0100)]
fbarray: add reverse finding of chunk

Add a function to look for N used/free slots, but going backwards
instead of forwards. All semantics are kept similar to the existing
function, with the difference being that given the same input, the
same results will be returned in reverse order.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agofbarray: add reverse finding
Anatoly Burakov [Mon, 11 Jun 2018 20:55:38 +0000 (21:55 +0100)]
fbarray: add reverse finding

Add function to look up used/free indexes starting from specified
index, but going backwards instead of forward. Semantics are kept
similar to the existing function, except for the fact that, given
the same input, the results returned will be in reverse order.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agofbarray: reduce duplication in element finding
Anatoly Burakov [Mon, 11 Jun 2018 20:55:37 +0000 (21:55 +0100)]
fbarray: reduce duplication in element finding

Just code move to put all checks and calls in one place.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agofbarray: reduce duplication in chunk finding
Anatoly Burakov [Mon, 11 Jun 2018 20:55:36 +0000 (21:55 +0100)]
fbarray: reduce duplication in chunk finding

Mostly code move, aside from more quick checks done to avoid
doing computations in obviously hopeless cases.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agofbarray: reduce duplication in contiguous finding
Anatoly Burakov [Mon, 11 Jun 2018 20:55:35 +0000 (21:55 +0100)]
fbarray: reduce duplication in contiguous finding

Mostly a code move, to have all code related to find_contig in
one place. This slightly changes the API in that previously,
calling find_contig_free() on a full fbarray would've been
an error, but equivalent call to find_contig_used() on an empty
array does not return an error, leading to an inconsistency in
the API.

The decision was made to not treat this condition as an error,
because it is equivalent to calling find_contig() on an index
that just happens to be used/free, which is not an error and
will return 0.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agofbarray: fix errno values returned from functions
Anatoly Burakov [Mon, 11 Jun 2018 20:55:34 +0000 (21:55 +0100)]
fbarray: fix errno values returned from functions

Errno values are supposed to be positive, yet they were negative.

This changes API, so not backporting.

Fixes: c44d09811b40 ("eal: add shared indexed file-backed array")

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agomem: make segment preallocation OS-specific
Anatoly Burakov [Thu, 28 Jun 2018 11:41:49 +0000 (12:41 +0100)]
mem: make segment preallocation OS-specific

In the perfect world, it wouldn't matter how much memory was
preallocated because most of it was always going to be private
anonymous zero-page mappings for the duration of the program.
However, in practice, due to peculiarities of FreeBSD, we need
to additionally limit memory allocation there. This patch moves
the segment preallocation to EAL private functions that will be
implemented by an OS-specific EAL rather than being in the common
memory-related code.

Since there is no support for growing/shrinking memory use at
runtime on FreeBSD anyway, this does not inhibit any functionality
but makes core dumps faster even on default settings.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agoeal/bsd: concatenate adjacent memory segments
Anatoly Burakov [Thu, 28 Jun 2018 11:41:48 +0000 (12:41 +0100)]
eal/bsd: concatenate adjacent memory segments

Previously, memory allocator always left holes between mapped
contigmem segments, even if they were IOVA-contiguous. Fix this
by remembering last IOVA address and memseg index, and checking
against those when mapping new contigmem segments.

Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agoeal/bsd: fix memory segment index display
Anatoly Burakov [Thu, 28 Jun 2018 11:41:47 +0000 (12:41 +0100)]
eal/bsd: fix memory segment index display

Segment index was set to 0 at start but was never incremented.
This has no consequences other than displayed number of segments
allocated at initialization. Fix this by incrementing it after
displaying.

Fixes: 66cc45e293ed ("mem: replace memseg with memseg lists")
Cc: stable@dpdk.org
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agoeal: fix return codes on control thread failure
Dariusz Stojaczyk [Tue, 10 Jul 2018 10:44:47 +0000 (12:44 +0200)]
eal: fix return codes on control thread failure

This function returned positive error numbers instead
of negative ones as desbribed in the doc. What's worse,
multiple of its callers only check for (rc < 0) to detect
failure.

It was incorrectly assumed that pthread_create
and pthread_setaffinity_np return negative errnos. They
always returns positive ones, so this patch negates their
return values before returning.

Fixes: 9e5afc72c909 ("eal: add function to create control threads")
Cc: stable@dpdk.org
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
Reviewed-by: Olivier Matz <olivier.matz@6wind.com>
6 years agoeal: fix return codes on thread naming failure
Dariusz Stojaczyk [Tue, 10 Jul 2018 10:44:46 +0000 (12:44 +0200)]
eal: fix return codes on thread naming failure

The doc says this function returns negative errno
on error, but it currently returns either -1 or
positive errno.

It was incorrectly assumed that pthread_setname_np()
returns negative error numbers. It always returns
positive ones, so this patch negates its return value
before returning.

Fixes: 3901ed99c2f8 ("eal: fix thread naming on FreeBSD")
Cc: stable@dpdk.org
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
Reviewed-by: Olivier Matz <olivier.matz@6wind.com>
6 years agoeal: ignore failure of naming a control thread
Dariusz Stojaczyk [Tue, 10 Jul 2018 10:44:45 +0000 (12:44 +0200)]
eal: ignore failure of naming a control thread

The error is not fatal and we can physically continue
creating the thread. It simply won't have a name.

If rte_thread_setname() fails, we will just print
a debug log now. EAL does the same for lcore threads.

Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
Reviewed-by: Olivier Matz <olivier.matz@6wind.com>
6 years agomem: do not use --base-virtaddr in secondary processes
Dariusz Stojaczyk [Mon, 18 Jun 2018 19:53:09 +0000 (21:53 +0200)]
mem: do not use --base-virtaddr in secondary processes

Since secondary process' address space is highly dictated
by the primary process' mappings, it doesn't make much
sense to use base-virtaddr for secondary processes.

This patch is intended to fix PCI resource mapping
in secondary processes using the same base-virtaddr
as their primary processes. PCI uses the end of the hugepage
memory area to map all resources. [pci_find_max_end_va()]
It works for primary processes, but can't be mapped 1:1
by secondary ones, as the same addresses are currently always
occupied by shadow memseg lists, which were created with
eal_get_virtual_area(NULL, ...).

```
PRIMARY PROCESS
0x6e00e00000    388K rw-s- fbarray_memseg-2048k-1-3
0x6e01000000 16777216K r----   [ anon ]
0x7201000000     16K rw-s- resource0

SECONDARY PROCESS
0x6e00e00000    388K rw-s- fbarray_memseg-2048k-1-3
0x6e01000000 16777216K r----   [ anon ]
0x7201000000      4K rw-s- fbarray_memseg-1048576k-0-0_203213
```

Fixes: 524e43c2ad9a ("mem: prepare memseg lists for multiprocess sync")
Cc: stable@dpdk.org
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agomem: fix alignment requested with --base-virtaddr
Dariusz Stojaczyk [Fri, 15 Jun 2018 15:13:04 +0000 (17:13 +0200)]
mem: fix alignment requested with --base-virtaddr

Whenever a calculated base-virtaddr offset had to be
manually aligned to requested page_sz, we did not take
account of that alignment in incrementing the base-virtaddr
offset further. The next requested virtual area could print
a warning "hint [...] not respected!" and let the system
pick an address instead. As a result, this breaks secondary
process support on many system configurations.

Fixes: b7cc54187ea4 ("mem: move virtual area function in common directory")
Cc: stable@dpdk.org
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agomem: fix alignment of requested virtual areas
Dariusz Stojaczyk [Wed, 13 Jun 2018 19:08:14 +0000 (21:08 +0200)]
mem: fix alignment of requested virtual areas

Although the alignment mechanism works as intended, the
`no_align` bool flag was set incorrectly. We were aligning
buffers that didn't need extra alignment, and weren't
aligning ones that really needed it.

Fixes: b7cc54187ea4 ("mem: move virtual area function in common directory")
Cc: stable@dpdk.org
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agomem: avoid crash on memseg query with invalid address
Dariusz Stojaczyk [Mon, 4 Jun 2018 05:33:41 +0000 (07:33 +0200)]
mem: avoid crash on memseg query with invalid address

When trying to use it with an address that's not
managed by DPDK it would segfault due to a missing
check. The doc says this function returns either
a pointer or NULL, so let it do so.

Fixes: 66cc45e293ed ("mem: replace memseg with memseg lists")
Cc: stable@dpdk.org
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agomem: do not unmap overlapping region on mmap failure
Dariusz Stojaczyk [Fri, 1 Jun 2018 12:59:20 +0000 (14:59 +0200)]
mem: do not unmap overlapping region on mmap failure

This isn't documented in the manuals, but a failed
mmap(..., MAP_FIXED) may still unmap overlapping
regions. In such case, we need to remap these regions
back into our address space to ensure mem contiguity.
We do it unconditionally now on mmap failure just to
be safe.

Verified on Linux 4.9.0-4-amd64. I was getting
ENOMEM when trying to map hugetlbfs with no space
left, and the previous anonymous mapping was still
being removed.

Fixes: 582bed1e1d1d ("mem: support mapping hugepages at runtime")
Cc: stable@dpdk.org
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agomem: do not leave unmapped holes in EAL memory area
Dariusz Stojaczyk [Fri, 1 Jun 2018 12:59:19 +0000 (14:59 +0200)]
mem: do not leave unmapped holes in EAL memory area

EAL reserves a huge area in virtual address space
to provide virtual address contiguity for e.g.
future memory extensions (memory hotplug). During
memory hotplug, if the hugepage mmap succeeds but
doesn't suffice EAL's requiriments, the EAL would
unmap this mapping straight away, leaving a hole in
its virtual memory area and making it available
to everyone. As EAL still thinks it owns the entire
region, it may try to mmap it later with MAP_FIXED,
possibly overriding a user's mapping that was made
in the meantime.

This patch ensures each hole is mapped back by EAL,
so that it won't be available to anyone else.

Fixes: 582bed1e1d1d ("mem: support mapping hugepages at runtime")
Cc: stable@dpdk.org
Signed-off-by: Dariusz Stojaczyk <dariuszx.stojaczyk@intel.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
6 years agoethdev: fix flow expansion matching types
Nelio Laranjeiro [Wed, 11 Jul 2018 06:49:35 +0000 (08:49 +0200)]
ethdev: fix flow expansion matching types

Node RSS types are generally covering more RSS kind than the user is
requesting, it should accept to expand even if only a single bit is
remains after masking.  Setting the correct RSS kind for the rule
remains the driver job.

Fixes: 4ed05fcd441b ("ethdev: add flow API to expand RSS flows")

Signed-off-by: Nelio Laranjeiro <nelio.laranjeiro@6wind.com>
Acked-by: Adrien Mazarguil <adrien.mazarguil@6wind.com>
6 years agoapp/testpmd: distribute queues to cores
Xueming Li [Sat, 26 May 2018 15:15:20 +0000 (23:15 +0800)]
app/testpmd: distribute queues to cores

Current topology distribute forwarding streams to lcores by port, this
make unbalanced loading when port number larger than 2:
lcore 0: P0Q0->P1Q0, P0Q1->P1Q1
locre 1: P1Q0->P0Q0, P1Q1->P0Q1
If only one port has traffic, only one locre get fully loaded and the
other one get no forwarding. Performance is bad as only one core doing
forwarding in such case.

This patch distributes forwarding streams by queue, try to get streams
of each port handled by different lcore:
lcore 0: P0Q0->P1Q0, P1Q0->P1Q0
locre 1: P0Q1->P0Q1, P1Q1->P0Q1

Signed-off-by: Xueming Li <xuemingl@mellanox.com>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
6 years agohash: add API to query the key count
Yipeng Wang [Tue, 10 Jul 2018 17:00:01 +0000 (10:00 -0700)]
hash: add API to query the key count

Add a new function, rte_hash_count, to return the number of keys that
are currently stored in the hash table. Corresponding test functions are
added into hash_test and hash_multiwriter test.

Signed-off-by: Yipeng Wang <yipeng1.wang@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
6 years agotest/hash: add test for read/write concurrency
Yipeng Wang [Tue, 10 Jul 2018 17:00:00 +0000 (10:00 -0700)]
test/hash: add test for read/write concurrency

This commits add a new test case for testing read/write concurrency.

Signed-off-by: Yipeng Wang <yipeng1.wang@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
6 years agotest/hash: update for read/write concurrency perf
Yipeng Wang [Tue, 10 Jul 2018 16:59:59 +0000 (09:59 -0700)]
test/hash: update for read/write concurrency perf

New code is added to support read-write concurrency for
rte_hash. Due to the newly added code in critial path,
the perf test is modified to show any performance impact.
It is still a single-thread test.

Signed-off-by: Yipeng Wang <yipeng1.wang@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
6 years agohash: support read/write concurrency
Yipeng Wang [Tue, 10 Jul 2018 16:59:58 +0000 (09:59 -0700)]
hash: support read/write concurrency

The existing implementation of librte_hash does not support read-write
concurrency. This commit implements read-write safety using rte_rwlock
and rte_rwlock TM version if hardware transactional memory is available.

Both multi-writer and read-write concurrency is protected by rte_rwlock
now. The x86 specific header file is removed since the x86 specific RTM
function is not called directly by rte hash now.

Signed-off-by: Yipeng Wang <yipeng1.wang@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
6 years agohash: move duplicated code into functions
Yipeng Wang [Tue, 10 Jul 2018 16:59:57 +0000 (09:59 -0700)]
hash: move duplicated code into functions

This commit refactors the hash table lookup/add/del code
to remove some code duplication. Processing on primary bucket can
also apply to secondary bucket with same code.

Signed-off-by: Yipeng Wang <yipeng1.wang@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
6 years agohash: fix key slot size accuracy
Yipeng Wang [Tue, 10 Jul 2018 16:59:56 +0000 (09:59 -0700)]
hash: fix key slot size accuracy

This commit calculates the needed key slot size more
accurately. The previous local cache fix requires
the free slot ring to be larger than actually needed.
The calculation of the value is inaccurate.

Fixes: 5915699153d7 ("hash: fix scaling by reducing contention")
Cc: stable@dpdk.org
Signed-off-by: Yipeng Wang <yipeng1.wang@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
6 years agohash: fix a multi-writer race condition
Yipeng Wang [Tue, 10 Jul 2018 16:59:55 +0000 (09:59 -0700)]
hash: fix a multi-writer race condition

Current multi-writer implementation uses Intel TSX to
protect the cuckoo path moving but not the cuckoo
path searching. After searching, we need to verify again if
the same empty slot still exists at the beginning of the TSX
region. Otherwise another writer could occupy the empty slot
before the TSX region. Current code does not verify.

Fixes: be856325cba3 ("hash: add scalable multi-writer insertion with Intel TSX")
Cc: stable@dpdk.org
Signed-off-by: Yipeng Wang <yipeng1.wang@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
6 years agohash: fix multiwriter lock memory allocation
Yipeng Wang [Tue, 10 Jul 2018 16:59:54 +0000 (09:59 -0700)]
hash: fix multiwriter lock memory allocation

When malloc for multiwriter_lock, the align should be
RTE_CACHE_LINE_SIZE rather than LCORE_CACHE_SIZE.

Also there should be check to verify the success of
rte_malloc.

Fixes: be856325cba3 ("hash: add scalable multi-writer insertion with Intel TSX")
Cc: stable@dpdk.org
Signed-off-by: Yipeng Wang <yipeng1.wang@intel.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
6 years agomempool/octeontx: fix pool to aura mapping
Pavan Nikhilesh [Tue, 3 Jul 2018 04:50:41 +0000 (10:20 +0530)]
mempool/octeontx: fix pool to aura mapping

HW needs each pool to be mapped to an aura set of 16 auras.
Previously, pool to aura mapping was considered to be 1:1.

Fixes: 02fd6c744350 ("mempool/octeontx: support allocation")
Cc: stable@dpdk.org
Signed-off-by: Pavan Nikhilesh <pbhagavatula@caviumnetworks.com>
Acked-by: Santosh Shukla <santosh.shukla@caviumnetworks.com>
Acked-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
6 years agoexamples/l3fwd-power: add high/regular perf cores options
Radu Nicolau [Mon, 11 Jun 2018 10:03:23 +0000 (11:03 +0100)]
examples/l3fwd-power: add high/regular perf cores options

Added high/regular performance core pinning configuration options
that can be used in place of the existing 'config' option.

'--high-perf-cores CORELIST' option allow the user to specify a
high performance cores list; if this option is not used and the
'perf-config' option is used, the application will query the
system using the rte_power library in order to get a list of
available high performance cores. The cores that are considered
high performance are the cores that have turbo enabled.

'--perf-config (port,queue,hi_perf,lcore_index)'
option is similar to the existing config option, the cores are specified
as indices for bins containing high or regular performance cores.

Example:

l3fwd-power -l 6,7 -- -p 0xff \
--high-perf-cores 6 --perf-config="(0,0,0,0),(1,0,1,0)"

cores 6 and 7 are used, core 6 is specified as a high performance core.
port 0 queue 0 will use a regular performance core, index 0 (core 7)
port 1 queue 0 will use a high performance core, index 0 (core 6)

Signed-off-by: Radu Nicolau <radu.nicolau@intel.com>
Acked-by: David Hunt <david.hunt@intel.com>
6 years agotest/power: add unit test for get capabilities API
Radu Nicolau [Mon, 11 Jun 2018 10:03:22 +0000 (11:03 +0100)]
test/power: add unit test for get capabilities API

Signed-off-by: Radu Nicolau <radu.nicolau@intel.com>
Acked-by: David Hunt <david.hunt@intel.com>
6 years agopower: add get capabilities API
Radu Nicolau [Mon, 11 Jun 2018 10:03:21 +0000 (11:03 +0100)]
power: add get capabilities API

New API added, rte_power_get_capabilities(), that allows the
application to query the power and performance capabilities
of the CPU cores.

Signed-off-by: Radu Nicolau <radu.nicolau@intel.com>
Acked-by: David Hunt <david.hunt@intel.com>
6 years agoexamples/ip_pipeline: remove commands restriction
Jasvinder Singh [Tue, 10 Jul 2018 20:43:39 +0000 (21:43 +0100)]
examples/ip_pipeline: remove commands restriction

Currently, some CLI commands (for examples- add or delete pipeline
table entries, add meter profile etc.) fails to execute when
application pipeline threads are not running. Therefore,
command for enabling pipeline on the thread is required to be
executed first or specified in the script file before any of
such commands.

This patch removes above restriction and adds support for
executing all CLI commands regardless of the pipeline thread state.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agoapp/testpmd: rework softnic forward mode
Jasvinder Singh [Fri, 6 Jul 2018 17:21:16 +0000 (18:21 +0100)]
app/testpmd: rework softnic forward mode

Modied the testpmd softnic forwarding mode as per the
changes in softnic PMD.

To run testpmd application with softnic fwd mode, following
command is used;

$ ./testpmd -c 0xc -n 4 --vdev 'net_softnic0,firmware=script.cli'
  -- -i --forward-mode=softnic

Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
Signed-off-by: Reshma Pattan <reshma.pattan@intel.com>
6 years agonet/softnic: add firmware script
Jasvinder Singh [Fri, 6 Jul 2018 17:21:15 +0000 (18:21 +0100)]
net/softnic: add firmware script

Add default firmware script for softnic.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agonet/softnic: implement start and stop
Jasvinder Singh [Fri, 6 Jul 2018 17:21:14 +0000 (18:21 +0100)]
net/softnic: implement start and stop

Implements softnic start and stop function.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agonet/softnic: setup Rx/Tx queues
Jasvinder Singh [Fri, 6 Jul 2018 17:21:13 +0000 (18:21 +0100)]
net/softnic: setup Rx/Tx queues

Implements softnic receive and transmit queues setup using swq object.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agonet/softnic: add command for TTL action
Jasvinder Singh [Fri, 6 Jul 2018 17:21:12 +0000 (18:21 +0100)]
net/softnic: add command for TTL action

Add cli commands for ttl action in softnic pipeline objects.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agonet/softnic: add command for meter action
Jasvinder Singh [Fri, 6 Jul 2018 17:21:11 +0000 (18:21 +0100)]
net/softnic: add command for meter action

Add cli commands for meter action in softnic pipeline objects.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agonet/softnic: add command to read stats
Jasvinder Singh [Fri, 6 Jul 2018 17:21:10 +0000 (18:21 +0100)]
net/softnic: add command to read stats

Add cli commands to read port and table stats of
softnic pipeline objects.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agonet/softnic: add command for pipeline table entries
Jasvinder Singh [Fri, 6 Jul 2018 17:21:09 +0000 (18:21 +0100)]
net/softnic: add command for pipeline table entries

Add cli commands for table entries in softnic pipeline objects.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agonet/softnic: add command to enable/disable pipeline
Jasvinder Singh [Fri, 6 Jul 2018 17:21:08 +0000 (18:21 +0100)]
net/softnic: add command to enable/disable pipeline

Add cli commands to enable and disable pipelines on specific threads in
softnic.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agonet/softnic: add command to create objects
Jasvinder Singh [Fri, 8 Jun 2018 12:41:48 +0000 (13:41 +0100)]
net/softnic: add command to create objects

Add cli commands to create softnic objects such as mempool, swq,
pipeline, etc.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agonet/softnic: add connection agent
Jasvinder Singh [Fri, 6 Jul 2018 17:21:06 +0000 (18:21 +0100)]
net/softnic: add connection agent

Add connection agent to enable connectivity with external agent
(e.g. telnet, netcat, Python script, etc).

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agonet/softnic: add command interface
Jasvinder Singh [Fri, 8 Jun 2018 12:41:46 +0000 (13:41 +0100)]
net/softnic: add command interface

Add interface for softnic cli commands.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agonet/softnic: add run API
Cristian Dumitrescu [Wed, 11 Jul 2018 12:43:09 +0000 (13:43 +0100)]
net/softnic: add run API

Implements softnic API function to run pipeline objects.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agonet/softnic: add thread
Jasvinder Singh [Fri, 6 Jul 2018 17:21:03 +0000 (18:21 +0100)]
net/softnic: add thread

Add thread data structure and init function to run softnic pipelines
objects.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agonet/softnic: add pipeline object
Jasvinder Singh [Fri, 6 Jul 2018 17:21:02 +0000 (18:21 +0100)]
net/softnic: add pipeline object

Add pipeline object implementation to the softnic.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agonet/softnic: add table action profile
Jasvinder Singh [Fri, 6 Jul 2018 17:21:01 +0000 (18:21 +0100)]
net/softnic: add table action profile

Add pipeline's table action profile implementation to the softnic.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agonet/softnic: add port action profile
Jasvinder Singh [Fri, 6 Jul 2018 17:21:00 +0000 (18:21 +0100)]
net/softnic: add port action profile

Add pipeline's port action profile implementation to the softnic.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agonet/softnic: add traffic manager object
Jasvinder Singh [Fri, 6 Jul 2018 17:20:59 +0000 (18:20 +0100)]
net/softnic: add traffic manager object

Add traffic manager(tmgr) object to the softnic.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agonet/softnic: add tap object
Jasvinder Singh [Fri, 6 Jul 2018 17:20:58 +0000 (18:20 +0100)]
net/softnic: add tap object

Add tap object implementation to the softnic.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agonet/softnic: add mempool object
Jasvinder Singh [Fri, 6 Jul 2018 17:20:57 +0000 (18:20 +0100)]
net/softnic: add mempool object

Add mempool object implementation to the softnic.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agonet/softnic: add link object
Jasvinder Singh [Fri, 6 Jul 2018 17:20:56 +0000 (18:20 +0100)]
net/softnic: add link object

Add link object implementation to the softnic.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agonet/softnic: add software queue object
Jasvinder Singh [Fri, 6 Jul 2018 17:20:55 +0000 (18:20 +0100)]
net/softnic: add software queue object

Add swq object implementation to the softnic.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agonet/softnic: restructure
Jasvinder Singh [Fri, 6 Jul 2018 17:20:54 +0000 (18:20 +0100)]
net/softnic: restructure

Rework the softnic implementation to have flexiblity in enabling
more features to its receive and transmit data path.

Signed-off-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Signed-off-by: Jasvinder Singh <jasvinder.singh@intel.com>
6 years agoexamples/ip_pipeline: add link command
Kevin Laatz [Fri, 22 Jun 2018 14:31:47 +0000 (15:31 +0100)]
examples/ip_pipeline: add link command

Add the functionality to track links in the application. This enables the
user to print the name, mac address and statistics for each link
in the application.

Signed-off-by: Kevin Laatz <kevin.laatz@intel.com>
Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
6 years agoraw/dpaa2_qdma: fix IOVA as VA flag
Hemant Agrawal [Thu, 21 Jun 2018 09:15:14 +0000 (14:45 +0530)]
raw/dpaa2_qdma: fix IOVA as VA flag

Fixes: b1ee472fed58 ("raw/dpaa2_qdma: introduce the DPAA2 QDMA driver")
Cc: stable@dpdk.org
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
6 years agobus/dpaa: support scatter/gather config
Hemant Agrawal [Fri, 6 Jul 2018 08:10:12 +0000 (13:40 +0530)]
bus/dpaa: support scatter/gather config

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
6 years agonet/dpaa: move the push queue set to global init
Hemant Agrawal [Fri, 6 Jul 2018 08:10:11 +0000 (13:40 +0530)]
net/dpaa: move the push queue set to global init

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
6 years agodrivers: support function name in NXP logs
Hemant Agrawal [Fri, 6 Jul 2018 08:10:10 +0000 (13:40 +0530)]
drivers: support function name in NXP logs

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
6 years agobus/fslmc: cleanup unnecessary global variables
Hemant Agrawal [Fri, 6 Jul 2018 08:10:09 +0000 (13:40 +0530)]
bus/fslmc: cleanup unnecessary global variables

Signed-off-by: Pavan Nikhilesh <pbhagavatula@caviumnetworks.com>
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
6 years agobus/dpaa: cleanup unnecessary global variables
Hemant Agrawal [Fri, 6 Jul 2018 08:10:08 +0000 (13:40 +0530)]
bus/dpaa: cleanup unnecessary global variables

Signed-off-by: Pavan Nikhilesh <pbhagavatula@caviumnetworks.com>
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
6 years agonet/dpaa: remove experimental tag from PMD API
Hemant Agrawal [Fri, 6 Jul 2018 08:10:07 +0000 (13:40 +0530)]
net/dpaa: remove experimental tag from PMD API

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
6 years agonet/dpaa: support default queue mode
Hemant Agrawal [Fri, 6 Jul 2018 08:10:06 +0000 (13:40 +0530)]
net/dpaa: support default queue mode

In case DPAA FMAN configuration tool (FMC) is not available,
the system can still work with default queue(1 queue per port).

This patch also fixes some logs related to FQ ids, which were
idetified while testing this support.

Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
6 years agobus/dpaa: make vdqcr configurable
Nipun Gupta [Fri, 6 Jul 2018 08:10:05 +0000 (13:40 +0530)]
bus/dpaa: make vdqcr configurable

This patch add support for configurable vdqcr exact flag.
This boost the performance, however this can give
side effects for some extra packet fetch. Which has been
taken care in the patch as well.

Signed-off-by: Nipun Gupta <nipun.gupta@nxp.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
6 years agobus/dpaa: add API to get MAC address
Akhil Goyal [Fri, 6 Jul 2018 08:10:04 +0000 (13:40 +0530)]
bus/dpaa: add API to get MAC address

Signed-off-by: Akhil Goyal <akhil.goyal@nxp.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
6 years agobus/dpaa: optimize fq callback routine
Hemant Agrawal [Fri, 6 Jul 2018 08:10:03 +0000 (13:40 +0530)]
bus/dpaa: optimize fq callback routine

Avoid array of fq as packets are dq only from a single q.

Signed-off-by: Sunil Kumar Kori <sunil.kori@nxp.com>
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
6 years agonet/dpaa2: fix prefetch Rx to honor number of packets
Hemant Agrawal [Fri, 6 Jul 2018 08:10:02 +0000 (13:40 +0530)]
net/dpaa2: fix prefetch Rx to honor number of packets

This patch fixes prefetch rx routine to
set the next prefetch request to the size of nb_pkts.
It assumes that next request would ideally be
of same size.

Fixes: 4bc5ab88dbd6 ("net/dpaa2: fix Tx only mode")
Cc: stable@dpdk.org
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
6 years agonet/dpaa: fix queue error handling and logs
Hemant Agrawal [Fri, 6 Jul 2018 08:10:01 +0000 (13:40 +0530)]
net/dpaa: fix queue error handling and logs

Fixes: 5e7455931442 ("net/dpaa: support Rx queue configurations with eventdev")
Cc: stable@dpdk.org
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
6 years agobus/dpaa: fix buffer offset setting in FMAN
Hemant Agrawal [Fri, 6 Jul 2018 08:10:00 +0000 (13:40 +0530)]
bus/dpaa: fix buffer offset setting in FMAN

The buffer offset was incorrectly being set at 64,
thus not honoring the packet headroom.

Fixes: 6d6b4f49a155 ("bus/dpaa: add FMAN hardware operations")
Cc: stable@dpdk.org
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
6 years agobus/dpaa: fix SVR id fetch location
Hemant Agrawal [Fri, 6 Jul 2018 08:09:59 +0000 (13:39 +0530)]
bus/dpaa: fix SVR id fetch location

Otherwise the SVR may not be available for dpaa init.

Fixes: 3b59b73dea08 ("bus/dpaa: update platform SoC value register routines")
Cc: stable@dpdk.org
Signed-off-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
6 years agobus/dpaa: fix phandle support for Linux 4.16
Alok Makhariya [Fri, 6 Jul 2018 08:09:58 +0000 (13:39 +0530)]
bus/dpaa: fix phandle support for Linux 4.16

Fixes: 2183c6f69d7e ("bus/dpaa: add OF parser for device scanning")
Cc: stable@dpdk.org
Signed-off-by: Alok Makhariya <alok.makhariya@nxp.com>
Acked-by: Shreyansh Jain <shreyansh.jain@nxp.com>
6 years agomk: change TLS model for DPAA machine
Sachin Saxena [Wed, 4 Jul 2018 09:54:45 +0000 (15:24 +0530)]
mk: change TLS model for DPAA machine

Random corruptions observed on platfoms with using
the dpdk library in shared mode with VPP software (plugin).

using traditional TLS scheme resolved the issue.

Tested with VPP with DPDK as a plugin.

Signed-off-by: Sachin Saxena <sachin.saxena@nxp.com>
6 years agohash: validate hash bucket entries while compiling
Honnappa Nagarahalli [Thu, 31 May 2018 15:30:49 +0000 (10:30 -0500)]
hash: validate hash bucket entries while compiling

Validate RTE_HASH_BUCKET_ENTRIES during compilation instead of
run time.

Signed-off-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>
Acked-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
6 years agonet/kni: fix check for meson build
Bruce Richardson [Tue, 3 Jul 2018 10:31:19 +0000 (11:31 +0100)]
net/kni: fix check for meson build

The configuration value indicating that KNI was build was incorrect,
causing the driver to never be built.

Fixes: 3479586fe636 ("net/kni: add to meson build")

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
6 years agonet/avp: fix 32-bit meson builds
Bruce Richardson [Tue, 3 Jul 2018 10:31:18 +0000 (11:31 +0100)]
net/avp: fix 32-bit meson builds

When compiling with meson, extra warnings are enabled about casting from
integers to different size pointers. This triggers an error in AVP as the
addition of the offset to the pointer address causes the result to be a
64-bit integer which doesn't fit a 32-bit pointer. The fix here is to
explicitly indicate that the offset is of type "uintptr_t" which prevents
any promotion which would cause errors.

Fixes: c0ad584222b5 ("net/avp: add device initialization")

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Allain Legacy <allain.legacy@windriver.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
6 years agoexamples/kni: fix dependency check for meson build
Bruce Richardson [Tue, 3 Jul 2018 10:31:17 +0000 (11:31 +0100)]
examples/kni: fix dependency check for meson build

Rather than hard-coding the example app to be built only when a set of
conditions are met, we can simplify things by having the app built when
KNI library itself is available. That saves us duplicating the same set
of restrictions on both library and example app.

Fixes: 89f0711f9ddf ("examples: build some samples with meson")

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Acked-by: Ferruh Yigit <ferruh.yigit@intel.com>
6 years agobus/dpaa2: fix default IOVA setting for meson builds
Bruce Richardson [Tue, 3 Jul 2018 10:31:16 +0000 (11:31 +0100)]
bus/dpaa2: fix default IOVA setting for meson builds

By default, the DPAA2_USE_PHYS_IOVA setting with the make build-system
is "y", which is overridden to "n" in the config file specifically for
the arm64-dpaa2-linuxapp-gcc builds. The lack of this setting in meson
builds causes issues for 32-bit build, as the alternative code paths
have compiler warnings e.g. for i686 builds. Therefore we should align
the meson and make settings, setting the value to "true" by default and
overriding it to "false" for dpaa2-specific builds.

Fixes: 6ec78c2463ac ("build: add meson support for dpaaX platforms")

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
6 years agobuild: disable pointer to int warnings for 32-bit
Bruce Richardson [Tue, 3 Jul 2018 10:31:15 +0000 (11:31 +0100)]
build: disable pointer to int warnings for 32-bit

Converting a 32-bit pointer to a 64-bit integer is generally safe, but
generates a lot of warnings when compiling 32-bit code with meson. The
warnings are not flagged when using make, so just disable them for
32-bit meson builds.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>