Hi，Adrien Mazarguil
There is a bug present with 18.05-rci when I test the feature "Move RSS to
rte_flow" in i40e NIC
The test steps are as below:
./usertools/dpdk-devbind.py -b igb_uio 05:00.0 05:00.1
./x86_64-native-linuxapp-gcc/app/testpmd -c 0x1fffe -n 4 -- -i --nb-cores=8
--rxq=8 --txq=8 --
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Marvin Liu
> Sent: Saturday, April 28, 2018 5:54 PM
> To: dev@dpdk.org
> Cc: Liu, Yong
> Subject: [dpdk-dev] [PATCH] mk: using initial-exec model for thread local
> variable
>
> When building share library, thre
On 4/28/2018 9:24 AM, Stephen Hemminger wrote:
On Fri, 27 Apr 2018 21:52:26 +0200
Thomas Monjalon wrote:
27/04/2018 19:45, Shreyansh Jain:
From: Stephen Hemminger [mailto:step...@networkplumber.org]
Shreyansh Jain wrote:
From: Jianfeng Tan
Below commit introduced pthread barrier for syn
On 4/28/2018 9:21 AM, Stephen Hemminger wrote:
On Fri, 27 Apr 2018 21:52:26 +0200
Thomas Monjalon wrote:
27/04/2018 19:45, Shreyansh Jain:
From: Stephen Hemminger [mailto:step...@networkplumber.org]
Shreyansh Jain wrote:
From: Jianfeng Tan
Below commit introduced pthread barrier for syn
Hi，Adrien Mazarguil
We have just use new RC.1 code on the feature of flow RSS API, but we
find some abnormal phenomenon.
After that I check code again, I find that it is introduced in this patch:
SHA-1: ac8d22de2394e03ba4a77d8fd24381147aafb1d3
* ethdev: flatten RSS configuration in flow
From: Tonghao Zhang
ixgbe doesn't write the EITR counter, disable it now.
Signed-off-by: Tonghao Zhang
---
drivers/net/ixgbe/ixgbe_ethdev.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/drivers/net/ixgbe/ixgbe_ethdev.c
index fcc10f6..0
From: Tonghao Zhang
If dpdk APPs call the rte_eth_dev_rx_intr_enable or
rte_eth_dev_rx_intr_disable frequently, and ixgbe vf will read
the IXGBE_VTEIMS register everytime. The patch saves the IXGBE_VTEIMS
to mask to avoid read frequently.
Signed-off-by: Tonghao Zhang
Acked-by: Beilei Xing
Acke
From: Tonghao Zhang
Set EITR interval as default. This patch can improve the
performance when we enable the rx-interrupt to process the
packets because we hope rx-interrupt reduce CPU. For example,
the 200us value of EITR makes the performance better with
the low CPU. The default value of ITR is
From: Tonghao Zhang
The ixgbe PF and VF use the same default value
for EITR defined in header file.
Signed-off-by: Tonghao Zhang
---
drivers/net/ixgbe/ixgbe_ethdev.c | 5 +
drivers/net/ixgbe/ixgbe_rxtx.c | 3 ++-
2 files changed, 3 insertions(+), 5 deletions(-)
diff --git a/drivers/net/
From: Tonghao Zhang
v4: for reducing config options, remove the config option and use the
default value defined in header file. Remove the patch:
http://dpdk.org/dev/patchwork/patch/36433
rebase the net-next.
v3: configure eitr via configuration file and remove the patch:
http:/
When building share library, thread-local storage model will be changed
to global-dynamic. It will cost additional protect for read thread local
variable. By now only lcore id is this kind of varaible and not need to
dynamic share with other threads. So make TLS model back to initial-exec
like stat
On Fri, 27 Apr 2018 21:52:26 +0200
Thomas Monjalon wrote:
> 27/04/2018 19:45, Shreyansh Jain:
> > From: Stephen Hemminger [mailto:step...@networkplumber.org]
> > > Shreyansh Jain wrote:
> > > > From: Jianfeng Tan
> > > > > Below commit introduced pthread barrier for synchronization.
> > >
On Fri, 27 Apr 2018 21:52:26 +0200
Thomas Monjalon wrote:
> 27/04/2018 19:45, Shreyansh Jain:
> > From: Stephen Hemminger [mailto:step...@networkplumber.org]
> > > Shreyansh Jain wrote:
> > > > From: Jianfeng Tan
> > > > > Below commit introduced pthread barrier for synchronization.
> > >
A new DPDK release candidate is ready for testing:
https://dpdk.org/browse/dpdk/tag/?id=v18.05-rc1
1138 patches: more than 18.02-rc2 which was really huge!
This is one of the explanations for the big delay in this release.
The release notes so far:
http://dpdk.org/doc/guides/rel_n
This adds general compression drivers feature guide as well as the ISA-L PMD
documentation and guide.
Signed-off-by: Lee Daly
---
.gitignore | 1 +
MAINTAINERS| 12
devtools/test-build.sh
Adds decompression functionality, similar to compression, this sets internal
ISA-L structures, provides input & output mbuf addresses, executes
decompression, which ISA-L calls inflate and finally error checks.
Signed-off-by: Lee Daly
---
drivers/compress/isal/isal_compress_pmd.c | 55 ++
Adds compression functionality, this sets internal ISA-L structures, provides
input & output mbuf addresses, executes compression, which ISA-L calls deflate,
and finally error checks.
Signed-off-by: Lee Daly
---
drivers/compress/isal/isal_compress_pmd.c | 93 ++-
driv
Add functions for statistic retrieval and resetting.
Signed-off-by: Lee Daly
---
drivers/compress/isal/isal_compress_pmd_ops.c | 33 +--
1 file changed, 31 insertions(+), 2 deletions(-)
diff --git a/drivers/compress/isal/isal_compress_pmd_ops.c
b/drivers/compress/isal/i
This patchs adds support for the compressdev enqueue_burst and
dequeue_burst API operations.
Signed-off-by: Lee Daly
---
drivers/compress/isal/isal_compress_pmd.c | 64 +++
drivers/compress/isal/isal_compress_pmd_ops.c | 2 +
2 files changed, 66 insertions(+)
diff -
This adds queue pair operations such as setup and release.
Signed-off-by: Lee Daly
---
drivers/compress/isal/isal_compress_pmd_ops.c | 111 +-
drivers/compress/isal/isal_compress_pmd_private.h | 2 +
2 files changed, 111 insertions(+), 2 deletions(-)
diff --git a/drive
This patch adds device initialization functionality such as probe and create,
also adding
deinitialize functionality.
Dynamic logging component also added.
Signed-off-by: Lee Daly
---
drivers/compress/isal/Makefile| 1 +
drivers/compress/isal/isal_compress_pmd.c | 7
This patchset contains the first compression PMD written under the DPDK
compression API, compressdev. The ISA-L compression driver utilizes Intel's
ISA-L compression library. It therefore has dependencies on both compressdev
and the ISA-L library, v2.22.0.
V2:
- Changes to keep in compliance wit
This patch creates, configures and frees the private xform, taking the
applications xform and using it to populate the PMDs own private xform
with the information which will be required for the compress/decompress
functions. This information includes the level, algorithm,
type of huffman code, typ
Add basic device poll mode driver operations to give ability to start, stop,
close device etc.
Signed-off-by: Lee Daly
---
drivers/compress/isal/isal_compress_pmd_ops.c | 63 +--
drivers/compress/isal/isal_compress_pmd_private.h | 12 +
2 files changed, 70 insertions(
Adding basic skeleton of the ISA-L compression driver. No compression
functionality, but lays the foundation for operations in the rest of the
patchset. The ISA-L compression driver utilizes Intel's ISA-L compression
library and compressdev API. It therefore has dependencies on both compressdev
and
27/04/2018 15:14, Burakov, Anatoly:
> On 27-Apr-18 1:59 PM, Reshma Pattan wrote:
> > EAL option -m is supported in FreeBSD,
> > so move it under supported heading from non
> > supported heading.
> >
> > Signed-off-by: Reshma Pattan
> > ---
> > v2: remove recommendation of socket-mem option.
> > -
13/04/2018 10:14, Pablo de Lara:
> Commits for bbdev and security libraries are merged
> into the Next Crypto subtree.
>
> Signed-off-by: Pablo de Lara
Acked-by: Thomas Monjalon
Applied, thanks
> > Signed-off-by: Akhil Goyal
>
> Acked-by: Pablo de Lara
>
> Thanks for volunteering!
Acked-by: Thomas Monjalon
and approved by the techboard.
Applied, thanks
23/04/2018 14:29, Anatoly Burakov:
> Claim maintainership of all areas of EAL memory init, including
> OS-specific parts of it.
>
> Also, claim maintainership of fbarray, since although it's not
> related to memory allocation, it is heavily used by it and its
> primary purpose is to serve memory a
> Anatoly Burakov (2):
> mem: add memalloc init stage
> mem: revert to using flock() and add per-segment lockfiles
Applied, thanks
> > For non-legacy mode, we are preallocating space for hugepages, so we know
> > in advance which pages we will be able to allocate, and which we won't.
> > However, the init procedure was using hugepage counts gathered from sysfs
> > and paid no attention to hugepage sizes that were actually avai
26/04/2018 10:10, Burakov, Anatoly:
> On 26-Apr-18 9:06 AM, Jianfeng Tan wrote:
> > resize_hugefile() returns either 0 (which indicates success) or -1
> > (which indicates failure). We failed to check the success as we
> > use --single-file-segments option.
> >
> > Fixes: 2a04139f66b4 ("eal: add s
25/04/2018 12:15, Anatoly Burakov:
> This patchset fixes a few Coverity issues in malloc
> introduced by recent DPDK memory hotplug patchset.
>
> Coverity issues fixed:
> - 272566 - possible null dereference
> - 272574 - use value before verification
>
> The following coverity issues were not fix
> Anatoly Burakov (9):
> mem: use strlcpy instead of snprintf
> mem: fix resource leak
> mem: fix potential double close
> mem: fix potential resource leak
> mem: fix potential resource leak
> mem: fix comparing pointer to value
> mem: fix potential bad unmap
> mem: fix statement ha
Hi Declan,
Thank you (and DPDK) for driving this proposal, much appreciated. Upon
quick review of the proposal one thing stood out.
In rte_flow.h the new tunnel action,
RLTE_FLOW_ACTION_VXLAN_ENCAP/DECAP, similar with NVGRE_ENCAP/DECAP, is
defined as:
/**
* Decapsulate outer most VXLAN tunnel fr
https://dpdk.org/tracker/show_bug.cgi?id=29
Bug ID: 29
Summary: pktgen hangs when it tries to send packets through
libvirt driver, works for all other drivers
Product: DPDK
Version: 18.02
Hardware: x86
OS: L
27/04/2018 19:45, Shreyansh Jain:
> From: Stephen Hemminger [mailto:step...@networkplumber.org]
> > Shreyansh Jain wrote:
> > > From: Jianfeng Tan
> > > > Below commit introduced pthread barrier for synchronization.
> > > > But two IPC threads block on the barrier, and never wake up.
> > > >
> > >
> Shreyansh Jain (3):
> crypto/dpaa_sec: remove ctx based offset for PA-VA conversion
> bus/fslmc: optimize physical to virtual address searching
> bus/dpaa: optimize physical to virtual address searching
Applied with fixes for:
- 32-bit compilation
- symbols export for share
27/04/2018 19:20, Shreyansh Jain:
> --- a/drivers/bus/dpaa/rte_dpaa_bus.h
> +++ b/drivers/bus/dpaa/rte_dpaa_bus.h
> @@ -95,9 +95,34 @@ struct dpaa_portal {
> uint64_t tid;/**< Parent Thread id for this portal */
> };
>
> -/* TODO - this is costly, need to write a fast coversion routine */
On Fri, Apr 27, 2018 at 04:55:20PM +0100, Ferruh Yigit wrote:
> On 4/27/2018 4:42 PM, Neil Horman wrote:
> > On Fri, Apr 27, 2018 at 04:17:55PM +0100, Ferruh Yigit wrote:
> >> Remove version tag from experimental block in linker version scripts
> >> (.map files).
> >>
> >> That label is not used by
27/04/2018 20:49, Thomas Monjalon:
> 27/04/2018 19:20, Shreyansh Jain:
> > --- a/drivers/bus/fslmc/portal/dpaa2_hw_pvt.h
> > +++ b/drivers/bus/fslmc/portal/dpaa2_hw_pvt.h
> > @@ -254,15 +254,38 @@ enum qbman_fd_format {
> > */
> > #define DPAA2_EQ_RESP_ALWAYS 1
> >
> > +/* Variou
27/04/2018 19:20, Shreyansh Jain:
> --- a/drivers/bus/fslmc/portal/dpaa2_hw_pvt.h
> +++ b/drivers/bus/fslmc/portal/dpaa2_hw_pvt.h
> @@ -254,15 +254,38 @@ enum qbman_fd_format {
> */
> #define DPAA2_EQ_RESP_ALWAYS 1
>
> +/* Various structures representing contiguous memory maps */
> +st
27/04/2018 19:22, Yongseok Koh:
> This patch introduces a new way of attaching an external buffer to a mbuf.
>
> Attaching an external buffer is quite similar to mbuf indirection in
> replacing buffer addresses and length of a mbuf, but a few differences:
> - When an indirect mbuf is attached, r
Hi,
2018-04-27 16:51 GMT+03:00 Bruce Richardson :
> On Fri, Apr 27, 2018 at 02:49:27PM +0100, Bruce Richardson wrote:
> > When we check for libbsd we were just checking for the library itself.
> > However, on some distros, e.g. CentOS, libbsd will be found even though
> > it's not usable by DPDK
> -Original Message-
> From: Stephen Hemminger [mailto:step...@networkplumber.org]
> Sent: Friday, April 27, 2018 11:10 PM
> To: Shreyansh Jain
> Cc: Jianfeng Tan ; dev@dpdk.org;
> tho...@monjalon.net; Olivier Matz ; Anatoly
> Burakov
> Subject: Re: [dpdk-dev] [PATCH] eal: fix threads blo
27/04/2018 16:21, Maxime Coquelin:
>
> On 04/27/2018 04:06 PM, Fan Zhang wrote:
> > Fixes: 3bb595ecd682 ("vhost/crypto: add request handler")
> >
> > This patch fix final condition check while moving virtqueue
> > descriptors.
> >
> > Signed-off-by: Fan Zhang
> > ---
> > lib/librte_vhost/vhos
On Fri, 27 Apr 2018 17:36:56 +
Shreyansh Jain wrote:
> > -Original Message-
> > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Jianfeng Tan
> > Sent: Friday, April 27, 2018 10:12 PM
> > To: dev@dpdk.org
> > Cc: tho...@monjalon.net; Jianfeng Tan ; Olivier
> > Matz ; Anatoly Burak
27/04/2018 16:19, Maxime Coquelin:
> Hi Fan,
>
> On 04/27/2018 03:52 PM, Fan Zhang wrote:
> > Fixes: 473387aaae4a ("vhost/crypto: move to safe GPA translation API")
> >
> > This patch fixes the missing head descriptor correction for
> > indirect descriptors.
> >
> > Signed-off-by: Fan Zhang
> >
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Jianfeng Tan
> Sent: Friday, April 27, 2018 10:12 PM
> To: dev@dpdk.org
> Cc: tho...@monjalon.net; Jianfeng Tan ; Olivier
> Matz ; Anatoly Burakov
>
> Subject: [dpdk-dev] [PATCH] eal: fix threads block on barrier
>
This patch is to accommodate an experimental feature of mbuf - external
buffer attachment. If mbuf is attached to an external buffer, its ol_flags
will have EXT_ATTACHED_MBUF set. Without enabling/using the feature,
everything remains same.
If PMD delivers Rx packets with non-direct mbuf, ol_flags
This patch introduces a new way of attaching an external buffer to a mbuf.
Attaching an external buffer is quite similar to mbuf indirection in
replacing buffer addresses and length of a mbuf, but a few differences:
- When an indirect mbuf is attached, refcnt of the direct mbuf would be
2 as
On Fri, Apr 27, 2018, 19:40 Thomas Monjalon wrote:
> 27/04/2018 18:31, Arnon Warshavsky:
> > On Fri, Apr 27, 2018, 17:22 Thomas Monjalon wrote:
> > > What is the status of this patchset?
> > > It seems not ready for RC1 (today).
> > > I don't want to push it in RC2 because it changes too many th
On 18-04-27 09:49 AM, Burakov, Anatoly wrote:
On 27-Apr-18 5:30 PM, Scott Branden wrote:
Hi Anatoly,
We'd appreciate your input so we can come to a solution of supporting
the necessary memory allocations?
Hi Scott,
I'm currently starting to work on a prototype that will be at least
RFC
If user has specified a flag to unmap the area right after mapping it,
we were passing an already-unmapped pointer to RTE_LOG. This is not an
issue since RTE_LOG doesn't actually dereference the pointer, but fix
it anyway by moving call to RTE_LOG to before unmap.
Coverity issue: 272584
Fixes: b7
Previously, if mmap failed to map page address at requested
address, we were attempting to unmap the wrong address. Fix it
by unmapping our actual mapped address, and jump further to
avoid unmapping memory that is not allocated.
Coverity issue: 272602
Fixes: 582bed1e1d1d ("mem: support mapping hu
Coverity reports these lines as having no effect. Technically, we do
want for those lines to have no effect, however they would've likely
been optimized out. Add volatile qualifiers to ensure the code has
effects.
Coverity issue: 272608
Fixes: 582bed1e1d1d ("mem: support mapping hugepages at runt
Previous code had an old rebase leftover from the time when
oldpolicy was an actual int, instead of a pointer. Fix it to
do comparison with dereferencing the pointer.
Coverity issue: 272589
Fixes: 582bed1e1d1d ("mem: support mapping hugepages at runtime")
Cc: anatoly.bura...@intel.com
Signed-off
Normally, tailq entry should have a valid fd by the time we attempt
to map the segment. However, in case it doesn't, we're leaking fd,
so fix it.
Coverity issue: 272570
Fixes: 2a04139f66b4 ("eal: add single file segments option")
Cc: anatoly.bura...@intel.com
Signed-off-by: Anatoly Burakov
---
Coverity issue: 272601
Fixes: 66cc45e293ed ("mem: replace memseg with memseg lists")
Cc: anatoly.bura...@intel.com
Signed-off-by: Anatoly Burakov
Acked-by: Bruce Richardson
---
lib/librte_eal/linuxapp/eal/eal_memory.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/lib/librte_eal/linuxapp/
We close fd if we managed to find it in the list of allocated
segment lists (which should always be the case under normal
conditions), but if we didn't, the fd was leaking. Close it if
we couldn't find it in the segment list. This is not an issue
as if the segment is zero length, we're getting rid
We were closing descriptor before checking if mapping has
failed, but if it did, we did a second close afterwards. Fix
it by moving closing descriptor to after we've done all error
checks.
Coverity issue: 272560
Fixes: 2a04139f66b4 ("eal: add single file segments option")
Cc: anatoly.bura...@inte
This patchset fixes a host of coverity issues in memory subsystem
introduced with recent DPDK memory hotplug patchset.
Coverity issues fixed:
- 272601 - leaking fd
- 272560 - double close fd
- 272568 - leaking fd
- 272570 - leaking fd
- 272589 - dereference before null check
- 272602 - freeing wro
Signed-off-by: Anatoly Burakov
Acked-by: Bruce Richardson
---
Notes:
One other instance of using snprintf is left alone as
it is expected to be addressed by another patch [1].
[1] http://dpdk.org/dev/patchwork/patch/38301/
lib/librte_eal/common/eal_common_memalloc.c | 5 +++--
With Hotplugging memory support, the order of memseg has been changed
from physically contiguous to virtual contiguous. DPAA bus and drivers
depend on PA to VA address conversion for I/O.
This patch creates a list of blocks requested to be pinned to the
DPAA mempool. For searching physical address
With Hotplugging memory support, the order of memseg has been changed
from physically contiguous to virtual contiguous. FSLMC bus and dpaa2
drivers depend on PA to VA address conversion when in Physical
addressing mode.
This patch creates a list of blocks requested to be pinned to the
DPAA2 mempoo
Crypto requires physical to virtual address conversion for
descriptors. Prior to memory hotplugging this was based on memseg
iteration assuming memsegs are all physical contiguous and using
cached start address fast calculations can be done. This
assumption now stands invalid with memory hotpluggin
Once the hotplugging (6b42f7563) patchset were merged, DPAA2 Physical
Addressing mode and DPAA observed drastic performance drop (~95%)
This was because of an inherent assumption while doing some memory
translation that memsegs would be physically contiguous
This series attempts to add a workaroun
On Fri, 27 Apr 2018 16:41:42 +
Jianfeng Tan wrote:
> Below commit introduced pthread barrier for synchronization.
> But two IPC threads block on the barrier, and never wake up.
>
> (gdb) bt
> #0 futex_wait (private=0, expected=0, futex_word=0x7fffcff4)
> at ../sysdeps/unix/sys
On 27-Apr-18 4:15 PM, Bruce Richardson wrote:
On Wed, Apr 25, 2018 at 10:56:41AM +0100, Anatoly Burakov wrote:
We were closing descriptor before checking if mapping has
failed, but if it did, we did a second close afterwards. Fix
it by moving closing descriptor to after we check if mmap has
succ
On 27-Apr-18 5:30 PM, Scott Branden wrote:
Hi Anatoly,
We'd appreciate your input so we can come to a solution of supporting
the necessary memory allocations?
Hi Scott,
I'm currently starting to work on a prototype that will be at least
RFC'd (if not v1'd) during 18.08 timeframe. Basicall
On 27-Apr-18 5:17 PM, Tan, Jianfeng wrote:
On 4/27/2018 11:46 PM, Tan, Jianfeng wrote:
Hi Olivier,
After this patch, I find the two IPC threads block at
pthread_barrier_wait(), and never wake up. Please refer below for more
information. The system is Ubuntu 16.04.
On 4/24/2018 10:46 PM, O
27/04/2018 18:41, Jianfeng Tan:
> Below commit introduced pthread barrier for synchronization.
> But two IPC threads block on the barrier, and never wake up.
>
> (gdb) bt
> #0 futex_wait (private=0, expected=0, futex_word=0x7fffcff4)
> at ../sysdeps/unix/sysv/linux/futex-internal.h:
On 27-Apr-18 5:27 PM, Bruce Richardson wrote:
On Fri, Apr 27, 2018 at 04:55:51PM +0100, Burakov, Anatoly wrote:
On 27-Apr-18 4:21 PM, Bruce Richardson wrote:
On Wed, Apr 25, 2018 at 10:56:43AM +0100, Anatoly Burakov wrote:
Normally, tailq entry should have a valid fd by the time we attempt
to
27/04/2018 18:31, Arnon Warshavsky:
> On Fri, Apr 27, 2018, 17:22 Thomas Monjalon wrote:
> > What is the status of this patchset?
> > It seems not ready for RC1 (today).
> > I don't want to push it in RC2 because it changes too many things.
> > So it's today or wait for 18.08. What do you think?
>
Below commit introduced pthread barrier for synchronization.
But two IPC threads block on the barrier, and never wake up.
(gdb) bt
#0 futex_wait (private=0, expected=0, futex_word=0x7fffcff4)
at ../sysdeps/unix/sysv/linux/futex-internal.h:61
#1 futex_wait_simple (private=0, expec
On Thu, Apr 19, 2018 at 05:35:25PM +0100, Anatoly Burakov wrote:
> When we ask to reserve virtual areas, we usually include
> alignment in the mapping size, and that memory ends up
> being wasted. Wasting a gigabyte of VA space while trying to
> reserve one gigabyte is pretty expensive on 32-bit, s
On Thu, Apr 19, 2018 at 05:35:25PM +0100, Anatoly Burakov wrote:
> When we ask to reserve virtual areas, we usually include
> alignment in the mapping size, and that memory ends up
> being wasted. Wasting a gigabyte of VA space while trying to
> reserve one gigabyte is pretty expensive on 32-bit, s
On Fri, Apr 27, 2018, 17:22 Thomas Monjalon wrote:
> 26/04/2018 08:20, Arnon Warshavsky:
> > The purpose of this patch series is to cleanup the library code
> > from paths that end up aborting the process,
> > and move to checking error values, in order to allow the running process
> > perform an
Hi Anatoly,
We'd appreciate your input so we can come to a solution of supporting the
necessary memory allocations?
On 23 April 2018 at 02:23, Srinath Mannam
wrote:
> Hi Anatoly,
>
> Our requirement is, that separate memory segment (speed memory window)
> need to be allocated outside huge pages
On 4/26/2018 11:40 AM, Declan Doherty wrote:
> Introduces a new port attribute to ethdev port's which denotes the
> switch domain a port belongs to. By default all port's switch
> identifiers are set to RTE_ETH_DEV_SWITCH_DOMAIN_ID_INVALID. Ports
> which supported the concept of switch domains can
On Fri, Apr 27, 2018 at 04:55:51PM +0100, Burakov, Anatoly wrote:
> On 27-Apr-18 4:21 PM, Bruce Richardson wrote:
> > On Wed, Apr 25, 2018 at 10:56:43AM +0100, Anatoly Burakov wrote:
> > > Normally, tailq entry should have a valid fd by the time we attempt
> > > to map the segment. However, in case
On 4/27/2018 11:46 PM, Tan, Jianfeng wrote:
Hi Olivier,
After this patch, I find the two IPC threads block at
pthread_barrier_wait(), and never wake up. Please refer below for more
information. The system is Ubuntu 16.04.
On 4/24/2018 10:46 PM, Olivier Matz wrote:
To avoid code duplicatio
With Hotplugging memory support, the order of memseg has been changed
from physically contiguous to virtual contiguous. DPAA bus and drivers
depend on PA to VA address conversion for I/O.
This patch creates a list of blocks requested to be pinned to the
DPAA mempool. For searching physical address
Crypto requires physical to virtual address conversion for
descriptors. Prior to memory hotplugging this was based on memseg
iteration assuming memsegs are all physical contiguous and using
cached start address fast calculations can be done. This
assumption now stands invalid with memory hotpluggin
With Hotplugging memory support, the order of memseg has been changed
from physically contiguous to virtual contiguous. FSLMC bus and dpaa2
drivers depend on PA to VA address conversion when in Physical
addressing mode.
This patch creates a list of blocks requested to be pinned to the
DPAA2 mempoo
Once the hotplugging (6b42f7563) patchset were merged, DPAA2 Physical
Addressing mode and DPAA observed drastic performance drop (~95%)
This was because of an inherent assumption while doing some memory
translation that memsegs would be physically contiguous
This series attempts to add a workaroun
On 27-Apr-18 4:57 PM, Van Haaren, Harry wrote:
From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Anatoly Burakov
Sent: Wednesday, April 25, 2018 11:16 AM
To: dev@dpdk.org
Cc: tho...@monjalon.net; Burakov, Anatoly
Subject: [dpdk-dev] [PATCH v2 3/3] malloc: fix potential dereferencing of NULL
p
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Anatoly Burakov
> Sent: Wednesday, April 25, 2018 11:16 AM
> To: dev@dpdk.org
> Cc: tho...@monjalon.net; Burakov, Anatoly
> Subject: [dpdk-dev] [PATCH v2 3/3] malloc: fix potential dereferencing of NULL
> pointer
>
> Previous code checked for
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Anatoly Burakov
> Sent: Wednesday, April 25, 2018 11:16 AM
> To: dev@dpdk.org
> Cc: tho...@monjalon.net; Burakov, Anatoly
> Subject: [dpdk-dev] [PATCH v2 2/3] malloc: fix potential out-of-bounds array
> access
>
> Technically, while the point
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Anatoly Burakov
> Sent: Wednesday, April 25, 2018 11:16 AM
> To: dev@dpdk.org
> Cc: tho...@monjalon.net
> Subject: [dpdk-dev] [PATCH v2 1/3] malloc: replace snprintf with strlcpy
>
> Signed-off-by: Anatoly Burakov
Acked-by: Harry van Haaren
On 27-Apr-18 4:21 PM, Bruce Richardson wrote:
On Wed, Apr 25, 2018 at 10:56:43AM +0100, Anatoly Burakov wrote:
Normally, tailq entry should have a valid fd by the time we attempt
to map the segment. However, in case it doesn't, we're leaking fd,
so fix it.
Coverity issue: 272570
Fixes: 2a04139
On 4/27/2018 4:42 PM, Neil Horman wrote:
> On Fri, Apr 27, 2018 at 04:17:55PM +0100, Ferruh Yigit wrote:
>> Remove version tag from experimental block in linker version scripts
>> (.map files).
>>
>> That label is not used by linker and information only. It is useful
>> for version blocks but not u
On Fri, Apr 27, 2018 at 04:49:03PM +0100, Burakov, Anatoly wrote:
> On 27-Apr-18 4:21 PM, Bruce Richardson wrote:
> > On Wed, Apr 25, 2018 at 10:56:43AM +0100, Anatoly Burakov wrote:
> > > Normally, tailq entry should have a valid fd by the time we attempt
> > > to map the segment. However, in case
On 27-Apr-18 4:21 PM, Bruce Richardson wrote:
On Wed, Apr 25, 2018 at 10:56:43AM +0100, Anatoly Burakov wrote:
Normally, tailq entry should have a valid fd by the time we attempt
to map the segment. However, in case it doesn't, we're leaking fd,
so fix it.
Coverity issue: 272570
Fixes: 2a04139
On Fri, Apr 27, 2018 at 04:39:34PM +0100, Burakov, Anatoly wrote:
> On 27-Apr-18 4:18 PM, Bruce Richardson wrote:
> > On Wed, Apr 25, 2018 at 10:56:42AM +0100, Anatoly Burakov wrote:
> > > We close fd if we managed to find it in the list of allocated
> > > segment lists (which should always be the
Hi Olivier,
After this patch, I find the two IPC threads block at
pthread_barrier_wait(), and never wake up. Please refer below for more
information. The system is Ubuntu 16.04.
On 4/24/2018 10:46 PM, Olivier Matz wrote:
To avoid code duplication, add a parameter to rte_ctrl_thread_create()
On Wed, Apr 25, 2018 at 10:56:47AM +0100, Anatoly Burakov wrote:
> If user has specified a flag to unmap the area right after mapping it,
> we were passing an already-unmapped pointer to RTE_LOG. This is not an
> issue since RTE_LOG doesn't actually dereference the pointer, but fix
> it anyway by m
On Fri, Apr 27, 2018 at 04:17:55PM +0100, Ferruh Yigit wrote:
> Remove version tag from experimental block in linker version scripts
> (.map files).
>
> That label is not used by linker and information only. It is useful
> for version blocks but not useful for experimental block but confusing.
> R
On 27-Apr-18 4:18 PM, Bruce Richardson wrote:
On Wed, Apr 25, 2018 at 10:56:42AM +0100, Anatoly Burakov wrote:
We close fd if we managed to find it in the list of allocated
segment lists (which should always be the case under normal
conditions), but if we didn't, the fd was leaking. Close it if
On Wed, Apr 25, 2018 at 10:56:46AM +0100, Anatoly Burakov wrote:
> Coverity reports these lines as having no effect. Technically, we do
> want for those lines to have no effect, however they would've likely
> been optimized out. Add volatile qualifiers to ensure the code has
> effects.
>
> Coverit
1 - 100 of 202 matches
Mail list logo
