The new macro __rte_noreturn, for compiler hinting,
is now used where appropriate for consistency.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
The new macro __rte_cold, for compiler hinting,
is now used where appropriate for consistency.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: David Christensen <drc@linux.vnet.ibm.com>
The new macro __rte_used, forcing symbol to be generated,
is now used where appropriate for consistency.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
There is a common macro __rte_unused, avoiding warnings,
which is now used where appropriate for consistency.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
There is a macro __rte_noinline, preventing function to be inlined,
which is now used where appropriate for consistency.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
There is a macro __rte_always_inline, forcing functions to be inlined,
which is now used where appropriate for consistency.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
There is a common macro __rte_packed for packing structs,
which is now used where appropriate for consistency.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
There is a common macro __rte_aligned for alignment,
which is now used where appropriate for consistency.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: David Christensen <drc@linux.vnet.ibm.com>
rte_ipsec has a dependency on rte_hash
So we need the librte_hash to be compiled before librte_ipsec.
Add the DEPDIRs to make sure this.
Fixes: 3feb23609c ("ipsec: add SAD create/destroy implementation")
Cc: stable@dpdk.org
Reported-by: Raslan Darawsheh <rasland@mellanox.com>
Suggested-by: Ferruh Yigit <ferruh.yigit@intel.com>
Signed-off-by: Akhil Goyal <akhil.goyal@nxp.com>
Currently, in the case to use bitmap as resource allocator, after
bitmap creation, all the bitmap bits should be set to indicate the
bit available. Every time when allocate one bit, search for the set
bits and clear it to make it in use.
Add a new rte_bitmap_init_with_all_set() function to have a quick
fill up the bitmap bits.
Comparing with the case create the bitmap as empty and set the bitmap
one by one, the new function costs less cycles.
Signed-off-by: Suanming Mou <suanmingm@mellanox.com>
Acked-by: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
Remove setting ALLOW_EXPERIMENTAL_API individually for each Makefile and
meson.build. Instead, enable ALLOW_EXPERIMENTAL_API flag across app, lib
and drivers.
This changes reduces the clutter across the project while still
maintaining the functionality of ALLOW_EXPERIMENTAL_API i.e. warning
external applications about experimental API usage.
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Signed-off-by: David Marchand <david.marchand@redhat.com>
This is to support cases when the input data for
decoding a code block is larger than 64kB and would
not fit as a contiguous block of data into one
mbuf. In that case the length from the operation
supersedes the mbuf default structure.
Signed-off-by: Nicolas Chautru <nicolas.chautru@intel.com>
Acked-by: Dave Burley <dave.burley@accelercomm.com>
Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
This exposes the HARQ buffer size at the device driver level
instead of using the capability of a specific operation.
This is currently not yet used by a device until
future commit.
Signed-off-by: Nicolas Chautru <nicolas.chautru@intel.com>
Acked-by: Dave Burley <dave.burley@accelercomm.com>
Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
Adding capability flag for device variants when HARQ buffer
may or may not include the filler bits.
Minor cosmetic changes in same file.
Signed-off-by: Nicolas Chautru <nicolas.chautru@intel.com>
Acked-by: Dave Burley <dave.burley@accelercomm.com>
Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
This commit removes unused prototype of the function pointer
compressdev_queue_pair_count_t.
Signed-off-by: Artur Trybula <arturx.trybula@intel.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
This commit removes unused function pointer (queue_pair_count)
from struct rte_cryptodev_ops. Related functions removed as well.
Signed-off-by: Artur Trybula <arturx.trybula@intel.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Acked-by: Fiona Trahe <fiona.trahe@intel.com>
Acked-by: Anoob Joseph <anoobj@marvell.com>
This patch adds range-checking of the device id passed from
the user app code. It prevents out-of-range array accesses
which in some situations resulted in an
application crash (segfault).
Fixes: 3dd4435cf4 ("cryptodev: fix checks related to device id")
Cc: stable@dpdk.org
Signed-off-by: Adam Dybkowski <adamx.dybkowski@intel.com>
Acked-by: Fiona Trahe <fiona.trahe@intel.com>
Acked-by: Akhil Goyal <akhil.goyal@nxp.com>
This patch allows C++ programs to use librte_fib library routines.
Previously C++ programs were not able to link this library and programs
would fail to get linked. With this patch compilation and linking
will be successful.
Fixes: 39e9272484 ("fib: add FIB library")
Fixes: 40d41a8a7b ("fib: support IPv6")
Cc: stable@dpdk.org
Signed-off-by: Asim Jamshed <asim.jamshed@gmail.com>
Signed-off-by: Saikrishna Edupuganti <saikrishna.edupuganti@intel.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>
Acked-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>
None of the public functions modify IPv6 address passed. So their
parameters are made const - with the exception of bulk functions.
This exception is due to compatibility problems - some compilers report
problems with const-casting of array of arrays.
Previously only lookup and add were updated to have addresses passed as
const so I'm adding this fixline.
Fixes: d82927d2f8 ("lpm6: make IPv6 address immutable")
Cc: stable@dpdk.org
Signed-off-by: Andrzej Ostruszka <aostruszka@marvell.com>
Acked-by: Vladimir Medvedkin <vladimir.medvedkin@intel.com>
Clean up indent and line ordering in Makefile and meson.build
for consistency in linux/ and freebsd/ directories.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: David Marchand <david.marchand@redhat.com>
Since the kernel modules are moved to kernel/ directory,
there is no need anymore for the sub-directory eal/ in
linux/, freebsd/ and windows/.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: David Marchand <david.marchand@redhat.com>
The EAL API (with doxygen documentation) is moved from
common/include/ to include/, which makes more clear that
it is the global API for all environments and architectures.
Note that the arch-specific and OS-specific include files are not
in this global include directory, but include/generic/ should
cover the doxygen documentation for them.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: David Marchand <david.marchand@redhat.com>
The variable common_sources allows to change the list of common files
from an arch-specific or OS-specific meson file.
For consistency and least surprise principle, this variable is removed.
Now the list of common sources is defined only in common/meson.build.
As a consequence, the temporary list of files which are currently supported
on Windows, is moved from windows/eal/meson.build.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: David Marchand <david.marchand@redhat.com>
The arch-specific directories arm, ppc and x86 in common/include/arch/
are moved as include/ sub-directories of respective arch directories:
- arm/include/
- ppc/include/
- x86/include/
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: Gavin Hu <gavin.hu@arm.com>
Reviewed-by: David Christensen <drc@linux.vnet.ibm.com>
Acked-by: David Marchand <david.marchand@redhat.com>
The arch-specific directories arm, ppc and x86 in common/arch/
are moved at the same level as the OS-specific directories.
It makes more clear that EAL is covering a matrix combining OS and arch.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: David Christensen <drc@linux.vnet.ibm.com>
Acked-by: David Marchand <david.marchand@redhat.com>
The directories ppc_64 are renamed as ppc in
- config/
- lib/librte_eal/common/arch/
- lib/librte_eal/common/include/arch/
The EAL directories are not really renamed, but symbolically linked,
because they will be moved with their new name in the next commits.
If ppc_32 needs to be supported, it can be in the same directory.
The arch directories arm and x86 are already covering both 32 and 64-bit
sub-architectures.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Reviewed-by: David Christensen <drc@linux.vnet.ibm.com>
Acked-by: David Marchand <david.marchand@redhat.com>
When moving files to the directory kernel/,
the file BSDmakefile.meson was left in eal/.
Also the intermediate makefiles in linux/ and freebsd/ became useless.
Fixes: acaa9ee991 ("move kernel modules directories")
Cc: stable@dpdk.org
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: David Marchand <david.marchand@redhat.com>
Introduction of pinned external buffers doubled memory loads in the
rte_pktmbuf_prefree_seg() function. Analysis of the generated assembly
code shows unnecessary load of the pool field of the rte_mbuf structure.
Here is the snippet of the assembly for "if (!RTE_MBUF_DIRECT(m))":
Before the change the code was:
movq 0x18(%rbx), %rax // load the ol_flags field
test %r13, %rax // check if ol_flags equals to 0x60...0
jz 0x9a8718 <Block 2> // jump out to "if (m->next != NULL)"
After the change the code became:
movq 0x18(%rbx), %rax // load ol_flags
test %r14, %rax // check if ol_flags equals to 0x60...0
jnz 0x9bea38 <Block 2> // jump in to "if (!RTE_MBUF_HAS_EXTBUF(m)"
movq 0x48(%rbx), %rax // load the pool field
jmp 0x9bea78 <Block 7> // jump out to "if (m->next != NULL)"
Look like this absolutely unneeded memory load of the pool field is an
optimization for the external buffer case in GCC (4.8.5), since Clang
generates the same assembly for both before and after the change versions.
Plus, GCC favors the external buffer case over the simple case.
This assembly code layout causes the performance degradation because the
rte_pktmbuf_prefree_seg() function is a part of a very hot path.
Workaround this compilation issue by moving the check for pinned buffer
apart from the check for external buffer and restore the initial code
flow that favors the direct mbuf case over the external one.
Fixes: 6ef1107ad4 ("mbuf: detach mbuf with pinned external buffer")
Cc: stable@dpdk.org
Signed-off-by: Alexander Kozyrev <akozyrev@mellanox.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@mellanox.com>
Acked-by: Olivier Matz <olivier.matz@6wind.com>
When the input string is "key=[", the ending '\0' is replaced
by a ',', leading to a heap buffer overflow.
Check the content of ctx1 to avoid this problem.
Fixes: cc0579f233 ("kvargs: support list value")
Cc: stable@dpdk.org
Signed-off-by: Yunjian Wang <wangyunjian@huawei.com>
Signed-off-by: Olivier Matz <olivier.matz@6wind.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>
This is a helper function in case components would like to do more work
than just logging a message based on log level, like for example
collecting some stats if the log type is DEBUG etc..
A few existing relevant usage converted to this new API.
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
Acked-by: Hyong Youb Kim <hyonkim@cisco.com>
Acked-by: Hemant Agrawal <hemant.agrawal@nxp.com>
Reviewed-by: Andrzej Ostruszka <aostruszka@marvell.com>
Acked-by: David Marchand <david.marchand@redhat.com>
When --no-huge mode is used, the memory is currently allocated with
mmap(NULL, ...). This is fine in most cases, but can fail in cases
where DPDK is run on a machine with an IOMMU that is of more limited
address width than that of a VA, because we're not specifying the
address hint for mmap() call.
Fix it by preallocating VA space before mapping it.
Cc: stable@dpdk.org
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Tested-by: David Marchand <david.marchand@redhat.com>
Tested-by: Jun W Zhou <junx.w.zhou@intel.com>
Currently, when we are creating DMA mappings for memory that's
either external or is backed by hugepages in IOVA as PA mode, we
assume that each page is necessarily discontiguous. This may not
actually be the case, especially for external memory, where the
user is able to create their own IOVA table and make it
contiguous. This is a problem because VFIO has a limited number
of DMA mappings, and it does not appear to concatenate them and
treats each mapping as separate, even when they cover adjacent
areas.
Fix this so that we always map contiguous memory in a single
chunk, as opposed to mapping each segment separately.
Signed-off-by: Anatoly Burakov <anatoly.burakov@intel.com>
Acked-by: Ray Kinsella <ray.kinsella@intel.com>
The source callback list grows infinitely when more than alarm
is queued.
This fix recognizes that an alarm interrupt in FreeBSD should never
have more than one callback on its list, so if
rte_intr_callback_register() is called with an interrupt handle type
of RTE_INTR_HANDLE_ALARM, so if such an interrupt type already has a
non-empty list, then a new callback is not created, but the kevent
timer is restarted properly.
Fixes: 23150bd8d8 ("eal/bsd: add interrupt thread")
Cc: stable@dpdk.org
Signed-off-by: Mit Matelske <mit@pt.net>
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
_cpuset TLS variable stores the CPU affinity of eal thread.
Populate the _cpuset TLS variable for control thread to
1) Make rte_thread_get_affinity() and eal_thread_dump_affinity
functional with control thread.
2) Quick access to cpu affinity.
Signed-off-by: Jerin Jacob <jerinj@marvell.com>
Reviewed-by: David Marchand <david.marchand@redhat.com>
The 32-bit arm machine doesn't support unaligned memory access. It
will cause a bus error on aarch32 with the custom element size ring.
Thread 1 "test" received signal SIGBUS, Bus error.
__rte_ring_enqueue_elems_64 (n=1, obj_table=0xf5edfe41, prod_head=0, \
r=0xf5edfb80) at /build/dpdk/build/include/rte_ring_elem.h:177
177 ring[idx++] = obj[i++];
Fixes: cc4b218790 ("ring: support configurable element size")
Signed-off-by: Phil Yang <phil.yang@arm.com>
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Minor spelling errors found by aspell and codespell
Fixes: 1daa338058 ("ethdev: validate offloads set by PMD")
Fixes: 81f9db8ecc ("ethdev: add vlan offload support")
Fixes: c8231c63dd ("ethdev: insert Rx callback as head of list")
Cc: stable@dpdk.org
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>
This patch adds the new flow item RTE_FLOW_ITEM_TYPE_PFCP to flow API to
match a PFCP header.
Add sample PFCP rules for testpmd guide. Since Session Endpoint
Identifier (SEID) only will be present in PFCP Session header and PFCP
Session headers shall be identified when the S field is equal to 1, when
create rules for PFCP Session header with certain SEID the S field need
be set 1.
Signed-off-by: Xiao Zhang <xiao.zhang@intel.com>
Acked-by: Ori Kam <orika@mellanox.com>
MinGW-w64 linker does not mimic MS linker options, so the build system
must differentiate between linkers on Windows. Use GNU linker options
with GCC and MS linker options with Clang.
MinGW-w64 by default uses MSVCRT stdio, which does not comply to ANSI,
most notably its formatting and string handling functions. MinGW-w64
support for the Universal CRT (UCRT) is ongoing, but the toolchain
provides its own standard-complying implementation of stdio. The latter
is used in the patch to support formatting in DPDK.
Signed-off-by: Dmitry Kozlyuk <dmitry.kozliuk@gmail.com>
Tested-by: William Tu <u9012063@gmail.com>
Tested-by: Pallavi Kadam <pallavi.kadam@intel.com>
Disable bundled getopt implementation and GNU extensions shim when
building in GNU environment.
Signed-off-by: Dmitry Kozlyuk <dmitry.kozliuk@gmail.com>
Tested-by: William Tu <u9012063@gmail.com>
Tested-by: Pallavi Kadam <pallavi.kadam@intel.com>
Acked-by: Pallavi Kadam <pallavi.kadam@intel.com>
Mixed case in Windows header names causes errors when cross-compiling
from Linux with case-sensitive filesystem using MinGW, because MinGW
distribution provides all platform SDK headers in lowercase. The change
does not affect Windows native builds on case-insensitive filesystems
(NTFS default).
Reported-by: Narcisa Vasile <navasile@microsoft.com>
Reported-by: William Tu <u9012063@gmail.com>
Signed-off-by: Dmitry Kozlyuk <dmitry.kozliuk@gmail.com>
Tested-by: William Tu <u9012063@gmail.com>
Tested-by: Pallavi Kadam <pallavi.kadam@intel.com>
Acked-by: Pallavi Kadam <pallavi.kadam@intel.com>
When using __attribute__((format(...)) on functions, GCC on Windows
assumes MS-specific format string by default, even if the underlying
stdio implementation is ANSI-compliant (either MS Unicersal CRT
or MinGW implementation). Wrap attribute into a macro that forces
GNU-specific format string when using GCC.
Use this new attribute for logging and panic messages in EAL
and for output strings in cmdline library.
Signed-off-by: Dmitry Kozlyuk <dmitry.kozliuk@gmail.com>
Tested-by: William Tu <u9012063@gmail.com>
Tested-by: Pallavi Kadam <pallavi.kadam@intel.com>
RTE_TOOLCHAIN_* macros are defining the target compiler.
Some API may be used in a host application like pmdinfogen.
That's why the the public headers should check the effective compiler
in use instead of the target compiler.
Detecting the compiler with macros is easy, except for __GNUC__
which is defined in all compilers supporting some GNU compatibility.
It is improved by defining RTE_CC_CLANG, RTE_CC_ICC and RTE_CC_GCC.
The extra macro RTE_CC_IS_GNU is defined to 0 or 1 in GCC case,
so it can be used simply with #if.
Signed-off-by: Thomas Monjalon <thomas@monjalon.net>
Acked-by: Dmitry Kozlyuk <dmitry.kozliuk@gmail.com>
The include file rte_debug.h is pulled into x86 builds through the
following callchain:
lib/librte_eal/common/include/generic/rte_cycles.h:17,
lib/librte_eal/common/include/arch/x86/rte_cycles.h:13,
lib/librte_eal/common/include/arch/x86/rte_spinlock.h:18,
lib/librte_eal/common/include/arch/x86/rte_rwlock.h:13,
The PPC specific version of the same callchain does not include
rte_debug.h. Add rte_debug.h back into the C files that require it.
Fixes: 0dcba52562 ("pci: remove unneeded includes in public header file")
Cc: stable@dpdk.org
Signed-off-by: David Christensen <drc@linux.vnet.ibm.com>
Tested-by: David Wilder <dwilder@us.ibm.com>
Acked-by: David Marchand <david.marchand@redhat.com>
When removing the extra headers from rte_pci.h stdlib should have been
removed instead of stdio, since off_t is missing for BSD builds when just
including stdlib.h
Fixes: 0dcba52562 ("pci: remove unneeded includes in public header file")
Cc: stable@dpdk.org
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
The rte_pci.h file includes more header files than are actually needed,
which means that all users of it also include those headers. This patch
removes the unneeded headers - adding them elsewhere where other components
were requiring them but not including them directly.
Fixes: c752998b5e ("pci: introduce library and driver")
Cc: stable@dpdk.org
Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
The ep struct is used to track what type of stats are required by the
client. For PORT_STATS type, it contains the lists of port and metric
ids to query, and the number of ids in each list.
The ep struct has values set (num of port and metric ids) when a request
for port stats values by name is received. However, after this value
assignment, the struct is reset to all 0 values, meaning the number of
port and metric ids required now both show as 0, and the client will not
receive the requested data in response. To fix this issue, the memset
call is now moved above the ep struct value assignment.
Fixes: 4080e46c80 ("telemetry: support global metrics")
Cc: stable@dpdk.org
Signed-off-by: Ciara Power <ciara.power@intel.com>
Reviewed-by: Kevin Laatz <kevin.laatz@intel.com>
This commit releases all service cores from their role,
returning them to ROLE_RTE on rte_service_finalize().
This fixes an issue relating to the service cores causing
a race-condition on rte_eal_cleanup(), where the service core
could still be executing while the main thread has already
free-d the service memory, leading to a segfault.
Fixes: da23f0aa87 ("service: fix memory leak with new function")
Cc: stable@dpdk.org
Reported-by: David Marchand <david.marchand@redhat.com>
Reported-by: Aaron Conole <aconole@redhat.com>
Signed-off-by: David Marchand <david.marchand@redhat.com>
Signed-off-by: Harry van Haaren <harry.van.haaren@intel.com>
Acked-by: Aaron Conole <aconole@redhat.com>
When the memory allocator reserves virtual addresses, it still does not
know what they will be used for.
Besides, huge areas are reserved for memory hotplug in multiprocess
setups. But most of the pages are unused in the whole life of the
processes.
Change protection mode to PROT_NONE when only reserving VA.
The memory allocator already switches to the right mode when making use
of it.
It also has the nice effect of getting those pages skipped by the kernel
when calling mlockall() or when a coredump gets generated.
Cc: stable@dpdk.org
Suggested-by: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: David Marchand <david.marchand@redhat.com>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Acked-by: Aaron Conole <aconole@redhat.com>
Acked-by: Anatoly Burakov <anatoly.burakov@intel.com>
In order to get more accurate the cntvct_el0 reading,
SW must invoke isb.
Fixes: ccad39ea07 ("eal/arm: add cpu cycle operations for ARMv8")
Cc: stable@dpdk.org
Reviewed-by: Gavin Hu <gavin.hu@arm.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
Signed-off-by: Haifeng Lin <haifeng.lin@huawei.com>
