For simple TX the virtio-net header must be zeroed, but it was using memory
that had been initialized with indirect descriptor tables. This resulted in
"unsupported gso type" errors from librte_vhost.
We can use the same memory for every descriptor to save cachelines in the
vswitch.
Fixes:
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Tomasz Kulasek
> Sent: Monday, April 04, 2016 3:45 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH] examples/l3fwd: fix segfault with gcc 5.x
>
> It seems that with gcc >5.x and -O2/-O3 optimization
> -Original Message-
> From: Kulasek, TomaszX
> Sent: Monday, April 04, 2016 5:20 PM
> To: Ananyev, Konstantin
> Cc: dev at dpdk.org
> Subject: RE: [dpdk-dev] [PATCH] examples/l3fwd: fix segfault with gcc 5.x
>
> Hi Konstantin,
>
> > -Original Message-
> > From: Ananyev,
On Tue, 5 Apr 2016 00:57:16 +
"Lu, Wenzhuo" wrote:
> Hi Stephen,
>
> > -Original Message-
> > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Stephen Hemminger
> > Sent: Tuesday, April 5, 2016 12:14 AM
> > To: Zhang, Helin; Ananyev, Konstantin
> > Cc: dev at dpdk.org;
Pass the same flags as in rte_mempool_create().
Signed-off-by: Lazaros Koromilas
---
app/test/test_mempool.c | 18 +--
app/test/test_mempool_perf.c | 4 +--
lib/librte_mempool/rte_mempool.h | 66 +---
3 files changed, 45 insertions(+),
The mempool cache is only available to EAL threads as a per-lcore
resource. Change this so that the user can create and provide their own
cache on mempool get and put operations. This works with non-EAL threads
too. This commit introduces the new API calls:
rte_mempool_cache_create(size,
Since commit a88ba49e51, values larger than 4 are allowed,
the autotests need to be updated accordingly.
Fixes: a88ba49e51 ("config: fix CPU and memory parameters on IBM POWER8")
Signed-off-by: Olivier Matz
---
app/test/test_eal_flags.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
renamed rte_cryptodev_sym_session.type -> dev_type
(as it's not a session type, but a device type)
renamed rte_crypto_sym_op.type -> sess_type
(as it's not an op type, but a session type)
Signed-off-by: Fiona Trahe
---
drivers/crypto/aesni_gcm/aesni_gcm_pmd.c | 7 ---
Hi Piotr,
On 04/04/2016 04:11 PM, Azarewicz, PiotrX T wrote:
>> Using token_len + 1 as the buffer size in the snprintf looks a bit
>> dangerous, as
>> it won't protect from overflows.
>>
>> See the following example:
>
> > That's why snprintf() should still use STR_TOKEN_SIZE.
>>
> Okay, I
Coverity reports an issue in ethdev:
*** CID 124562: Null pointer dereferences (FORWARD_NULL)
/lib/librte_ether/rte_ethdev.c: 1518 in rte_eth_xstats_get()
1512
1513 /* global stats */
1514 for (i = 0; i < RTE_NB_STATS; i++) {
1515 stats_ptr =
> >> > Announce the ABI breakage due to addition of external mempool
> >> > manager functionality which requires changes to rte_mempool
> >> > structure.
> >> >
> >> > Signed-off-by: David Hunt
> >>
> >> Acked-by: Olivier Matz
> >>
> >Acked-by: Bruce Richardson
>
> Acked-by: John McNamara
It seems that with gcc >5.x and -O2/-O3 optimization breaks packet grouping
algorithm.
When last packet pointer "lp" and "pnum->u64" buffer points the same
memory buffer, high optimization can cause unpredictable results. It seems
that assignment of precalculated group sizes may interfere with
2016-03-17 10:05, Olivier Matz:
> Add a deprecation notice for coming changes in mempool for 16.07.
[...]
> +* librte_mempool: new fixes and features will be added in 16.07:
> + allocation of large mempool in several virtual memory chunks, new API
> + to populate a mempool, new API to free a
> >> Deprecation notice for 16.04 for changes to occur in
> >> release 16.07 for rte_mempool memory reduction.
> >>
> >> Signed-off-by: Keith Wiles
> > Acked-by: Olivier Matz
> Acked-by: David Hunt
> Acked-by: John McNamara
Applied, thanks.
Hi Konstantin,
> -Original Message-
> From: Ananyev, Konstantin
> Sent: Monday, April 4, 2016 17:35
> To: Kulasek, TomaszX
> Cc: dev at dpdk.org
> Subject: RE: [dpdk-dev] [PATCH] examples/l3fwd: fix segfault with gcc 5.x
>
> Hi Tomasz,
>
> > -Original Message-
> > From: dev
2016-04-01 14:20, Keith Wiles:
> The new compiler option -Og causes a few warning on variables
> being used before being set warnings.
Sometimes the compiler is wrong. It seems this option makes it
even wronger. Why not use -Wno-error with -Og?
More details below:
>
On Mon, Apr 4, 2016 at 1:05 PM, Yuanhan Liu
wrote:
> On Mon, Apr 04, 2016 at 03:13:37PM +0200, Thomas Monjalon wrote:
> > Huawei, Yuanhan, any comment?
> >
> > 2016-03-31 13:01, Rich Lane:
> > > vq->vq_ring.desc[i + mid_idx].next = i;
> > >
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Ananyev,
> Konstantin
> Sent: Monday, April 04, 2016 4:35 PM
> To: Kulasek, TomaszX
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH] examples/l3fwd: fix segfault with gcc 5.x
>
> Hi Tomasz,
>
> >
Hi Tomasz,
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Tomasz Kulasek
> Sent: Monday, April 04, 2016 3:45 PM
> To: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH] examples/l3fwd: fix segfault with gcc 5.x
>
> It seems that with gcc >5.x and -O2/-O3
Huawei, Yuanhan, any comment?
2016-03-31 13:01, Rich Lane:
> vq->vq_ring.desc[i + mid_idx].next = i;
> vq->vq_ring.desc[i + mid_idx].addr =
> - vq->virtio_net_hdr_mem +
> -
2016-04-04 10:46, Daniel Mrzyglod:
> This series of patches is to solve errors for Intel C Compiler
>
> Daniel Mrzyglod (2):
> examples/l2fwd-crypto: fix for icc
> examples: fix build errors for icc
Applied with precision about icc 15.0.1, thanks
On 01/04/16 12:23, Pablo de Lara wrote:
> A new process to request the libsso library required by the SNOW3G PMD
> has been put in place, through a website, replacing the previous email method.
> This commit updates the SNOW3G documentation, to reflect this change.
>
> Since the library does not
Fixes: 94b0ad8e0aa5 ("null_crypto: add driver for null crypto operations")
Signed-off-by: Pablo de Lara
---
doc/guides/cryptodevs/null.rst | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/doc/guides/cryptodevs/null.rst b/doc/guides/cryptodevs/null.rst
index 3ea4d21..6408391
This patchset fixes some minor typos in the Crypto PMD documentation
and add an extra section to help user to initialize the devices.
The patchset depends on:
"doc: update the snow3g library information"
(http://dpdk.org/dev/patchwork/patch/11889/)
Pablo de Lara (3):
doc: fix typo in Crypto
Hi Olivier,
> -Original Message-
> From: Olivier Matz [mailto:olivier.matz at 6wind.com]
> Sent: Monday, April 4, 2016 10:01 AM
> To: Azarewicz, PiotrX T
> Cc: dev at dpdk.org
> Subject: Re: [PATCH v1 1/1] cmdline: add any multi string mode to token
> string
>
> Hi Piotr,
>
> This is
error: loops in this subroutine are not good vectorization candidates
(try compiling with O3 and/or IPO).
this error occurs in icc 15.0.1
Solution to disable this diagnostic message
https://software.intel.com/en-us/forums/intel-c-compiler/topic/537688
Fixes: d299106e8e31
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Friday, April 01, 2016 8:43 PM
> To: De Lara Guarch, Pablo
> Cc: dev at dpdk.org
> Subject: [PATCH] scripts: build with libsso
>
> Signed-off-by: Thomas Monjalon
Acked-by: Pablo de Lara
Hello experts,
We are currently running the latest DPDK driver on a Linux VM on KVM host.
We are using Intel XL710 NIC (40G mode) in SRIOV mode on the VM.
The PF side is the standard i40e Linux Kernel driver.
What is the maximum number of HW RSS queues that can be allocated for the
i40e poll
error: loops in this subroutine are not good vectorization candidates
(try compiling with O3 and/or IPO).
Solution to disable this diagnostic message
https://software.intel.com/en-us/forums/intel-c-compiler/topic/537688
Fixes: d299106e8e31 ("examples/ipsec-secgw: add IPsec sample application")
This series of patches is to solve errors for Intel C Compiler
Daniel Mrzyglod (2):
examples/l2fwd-crypto: fix for icc
examples: fix build errors for icc
examples/ipsec-secgw/Makefile | 4
examples/l2fwd-crypto/main.c | 4 ++--
examples/vmdq_dcb/Makefile| 4 +++-
3 files changed,
> -Original Message-
> From: Mrzyglod, DanielX T
> Sent: Monday, April 04, 2016 10:38 AM
> To: De Lara Guarch, Pablo; dev at dpdk.org
> Cc: Wu, Jingjing
> Subject: RE: [PATCH 2/2] examples: fix build errors for icc
>
> >From: De Lara Guarch, Pablo
> >Sent: Monday, April 04, 2016 11:34
Hi Piotr,
This is globally ok for me. Please see a comment below.
On 04/01/2016 01:36 PM, Piotr Azarewicz wrote:
> @@ -162,12 +174,15 @@ cmdline_parse_string(cmdline_parse_token_hdr_t *tk,
> const char *buf, void *res,
> }
>
> if (res) {
> - /* we are sure that
>From: De Lara Guarch, Pablo
>Sent: Monday, April 04, 2016 11:34 AM
>To: Mrzyglod, DanielX T ; dev at dpdk.org
>Cc: Wu, Jingjing
>Subject: RE: [PATCH 2/2] examples: fix build errors for icc
>
>Hi Daniel,
>
>> -Original Message-
>> From: Mrzyglod, DanielX T
>> Sent: Monday, April 04, 2016
> -Original Message-
> From: Mrzyglod, DanielX T
> Sent: Monday, April 04, 2016 9:47 AM
> To: dev at dpdk.org
> Cc: Wu, Jingjing; De Lara Guarch, Pablo
> Subject: [PATCH 1/2] examples/l2fwd-crypto: fix for icc
>
> Fix for compilation errors for icc:
> error #188: enumerated type mixed
Hi Daniel,
> -Original Message-
> From: Mrzyglod, DanielX T
> Sent: Monday, April 04, 2016 9:47 AM
> To: dev at dpdk.org
> Cc: Wu, Jingjing; De Lara Guarch, Pablo
> Subject: [PATCH 2/2] examples: fix build errors for icc
>
> error: loops in this subroutine are not good vectorization
This driver was one of the originals and has lots of little
whitespace issues.
PS: I know Intel doesn't like whitespace changes, there is never a
good time to do this, but no resuliting binary changes and it is
unlikely that more changes to this driver will occur this late in
release cycle.
36 matches
Mail list logo
