Hi, all maintainers,
Kindly ping for review.
Thanks,
Jie Hai
On 2024/12/4 10:06, Jie Hai wrote:
The application send packets only when the buffer is full, or the
buffer is empty and the packets to be sent extends TX_PKT_BURST.
The change of MAX_PKT_BURST make TX buffer size and TX_PKT_BURST
in
1. Use rte_net_get_ptype() to parse packets instead.
2. Support TSO for packets with ipv6 extension header.
Signed-off-by: Jie Hai
---
app/test-pmd/csumonly.c | 522 ++--
1 file changed, 128 insertions(+), 394 deletions(-)
diff --git a/app/test-pmd/csumonly.c
Add packet types parse for vxlan/vxlan-gpe/gtp/geneve packets.
Signed-off-by: Jie Hai
---
lib/net/rte_net.c | 110 ++
1 file changed, 102 insertions(+), 8 deletions(-)
diff --git a/lib/net/rte_net.c b/lib/net/rte_net.c
index d680accc1631..45b63eafc7c6
This patch support parse ipv6 extension header, and
support TSO for ipv6tcp packets with extension header.
Jie Hai (2):
net: add ptype parse for tunnel packets
app/test-pmd: use ptype API parse packets
app/test-pmd/csumonly.c | 522 ++--
lib/net/rte_net.c
The memset was always doing 0 bytes since size computed later.
Link: https://pvs-studio.com/en/blog/posts/cpp/1179/
Fixes: 3a80d7fb2ecd ("crypto/qat: support SHA3 plain hash")
Cc: sta...@dpdk.org
Signed-off-by: Stephen Hemminger
Acked-by: Bruce Richardson
---
drivers/crypto/qat/qat_sym_sessio
Script that converts memset before free into rte_memset_sensitive
and memset before rte_free into rte_free_sensitive
Signed-off-by: Stephen Hemminger
---
devtools/cocci/memset_free.cocci | 9 +
1 file changed, 9 insertions(+)
create mode 100644 devtools/cocci/memset_free.cocci
diff --g
Calling memset before rte_free not necessary, and could be
removed by the compiler. In this case, the data is not security
sensitive so the memset can be removed. Some security scanning
tools will flag this.
Signed-off-by: Stephen Hemminger
Acked-by: Bruce Richardson
---
drivers/compress/octeon
Regular memset maybe removed by compiler if done before a free
function. Use new rte_free_sensitive instead.
Signed-off-by: Stephen Hemminger
Acked-by: Bruce Richardson
---
drivers/crypto/qat/qat_asym.c | 5 +
1 file changed, 1 insertion(+), 4 deletions(-)
diff --git a/drivers/crypto/qat/q
Calling memset before free() has no effect and will be flagged
by security parsing tools as a potential bug. None of these data
structures have sensitive information.
Signed-off-by: Stephen Hemminger
---
drivers/net/ntnic/nthw/core/nthw_hif.c| 5 +
drivers/net/ntnic/nthw/core
Need to check the result of malloc() before calling memset.
This is only place in this driver that forgot, other code
does check.
Fixes: 0d9bca480e26 ("net/ntnic: add FPGA modules for initialization")
cc: sta...@dpdk.org
Signed-off-by: Stephen Hemminger
---
drivers/net/ntnic/nthw/nthw_rac.c | 4
Since tmp is not used later in the function, this memset
is unnecessary. Even though this is harmless,
it causes tools that look for security issues
around memset to flag this a bug.
Signed-off-by: Stephen Hemminger
Acked-by: Bruce Richardson
---
app/test/test_cmdline_cirbuf.c | 2 --
1 file ch
Doing memset before free maybe removed by compiler, and
is flagged by security scanning tools as potential problem.
In this case the memset is unnecessary.
Signed-off-by: Stephen Hemminger
Acked-by: Chengwen Feng
---
drivers/bus/uacce/uacce.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/d
When memset() is used before a release function such as free,
the compiler if allowed to optimize the memset away under
the as-if rules. This is normally ok, but in certain cases such
as passwords or security keys it is problematic.
Introduce a DPDK wrapper which is equivalent to the
C23 memset_ex
Although internally rte_free does poison the buffer in most
cases, it is useful to have function that explicitly does
this to avoid any security issues.
Name of new API is chosen to be similar to Linux kernel
kfree_sensitive() to make porting drivers easier.
Signed-off-by: Stephen Hemminger
---
This series handles memset related bugs indentified by PVS Studio.
The root cause is that Gcc and other compilers are free to
optimize away memset called before free.
v6 - rename rte_memset_sensative to rte_memzero_explicit to more
closely match kernel usage.
use rte_free_sensistive in c
Just doing memset() on keys is not enough, compiler can optimize
it away. Use new rte_memzero_explicit() and rte_free_sensitive().
Signed-off-by: Stephen Hemminger
Acked-by: Bruce Richardson
---
drivers/crypto/qat/qat_sym_session.c | 33 ++--
1 file changed, 17 insertion
All messages that because of an error should be at log level
NOTICE or above. Do not break log messages across lines.
Reword message for the case where getting info about a port
fails and it is skipped.
Signed-off-by: Stephen Hemminger
Acked-by: Tyler Retzlaff
---
lib/latencystats/rte_latencyst
The TSC frequency is not always an integral number of nanoseconds.
For examples, cycles per nanoseconds on my test system is 2.8.
Fix by using floating point where needed and calculate value once.
Signed-off-by: Stephen Hemminger
---
lib/latencystats/rte_latencystats.c | 17 ++---
1
No need for loop and memset() here. Just use structure
initialization. Remove unnecessary initializations.
Look at the number of samples, min, max and average
numbers to make sure they look correct.
Forward for longer to make sure multiple samples are taken.
Signed-off-by: Stephen Hemminger
---
Keeping track of the number of samples is useful when doing
debug and testing to make sure that samples are actually happening.
Signed-off-by: Stephen Hemminger
---
lib/latencystats/rte_latencystats.c | 20
1 file changed, 12 insertions(+), 8 deletions(-)
diff --git a/lib/l
ISO C does not allow casting function pointer to void *.
Resolve by enforcing the reserved argument.
The user_cb argument for rte_latencystats_init() was not
implemented, and had to be NULL anyway.
The log type is local to this function and therefore
can be local to this file.
Signed-off-by: Step
Include what is used here.
Signed-off-by: Stephen Hemminger
---
lib/latencystats/rte_latencystats.c | 19 ++-
1 file changed, 14 insertions(+), 5 deletions(-)
diff --git a/lib/latencystats/rte_latencystats.c
b/lib/latencystats/rte_latencystats.c
index 82c310da92..b990e5760f 100
The cycle counts do not need to be stored as floating point.
Instead keep track of latency in cycles, and convert to
nanoseconds when read.
Use scaled math with exponential Weighted Moving Average weight
of .25 to avoid use of floating point for that.
The average latency took too long to "warm up
Latencystats uses unnecessary floating point and is inaccurate if the number
of cycles per ns is not an integral value. Cleanup the code and add
more tests as well.
v7 - rebase and drop already merged VLA patch
Stephen Hemminger (7):
latencystats: handle fractional cycles per ns
latencystats:
On Thu, 13 Feb 2025 14:41:16 +0800
Junlong Wang wrote:
> V1:
> - updated net zxdh driver.
> optimize init and some ops.
> provided csum/lro/tso 、extend stats、fw_version、module_info 、meter, etc.
>
> Junlong Wang (16):
> net/zxdh: optimize np dtb channel initialization
> net/zxdh: op
From: Long Li
The current code has the wrong assumption that a PCI device can have only
one Ethernet device. This is not correct as a PCI device can be multi
functional and have multiple Ethernet devices.
Fix this by scanning all the devices under a PCI device.
Fixes: a2a23a794b3a ("net/netvsc:
From: Long Li
There is no recovery code path if netvsc failed to switch data path.
Log an error in this case for troubleshooting.
Signed-off-by: Long Li
---
drivers/net/netvsc/hn_vf.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/drivers/net/netvsc/hn_vf.c b/drivers
From: Long Li
An RTE device can have multiple Ethernet devices. On hot plug events, it
can't be removed until all its Ethernet devices have been removed.
Fixes: a2a23a794b3a ("net/netvsc: support VF device hot add/remove")
Cc: sta...@dpdk.org
Signed-off-by: Long Li
---
drivers/net/netvsc/hn_vf
From: Long Li
On some systems, one PCI device may have muiltple NET devices. This
patchset fixes issues on those systems when such PCI device is doing hot
add/remove.
Change in v2: added cover letter, drop the 4th patch that will be implemented
in EAL instead for device args caching
Long Li (
On Wed, 12 Feb 2025 23:32:21 +0800
Wenbo Cao wrote:
> Add basic PMD library and doc build infrastructure
> Update maintainers file to claim responsibility.
>
> Signed-off-by: Wenbo Cao
> Reviewed-by: Thomas Monjalon
> ---
> .mailmap | 1 +
> MAINTAINERS
On Wed, 12 Feb 2025 23:32:20 +0800
Wenbo Cao wrote:
> For This patchset just to support the basic chip init work
> and user can just found the eth_dev, but can't control more.
> For Now just support 2*10g nic,the chip can support
> 2*10g,4*10g,4*1g,8*1g,8*10g.
> The Feature rx side can support rx
On Thu, 13 Feb 2025 14:41:16 +0800
Junlong Wang wrote:
> V1:
> - updated net zxdh driver.
> optimize init and some ops.
> provided csum/lro/tso 、extend stats、fw_version、module_info 、meter, etc.
>
> Junlong Wang (16):
> net/zxdh: optimize np dtb channel initialization
> net/zxdh: op
On Thu, 13 Feb 2025 12:17:51 +0100
Ariel Otilibili wrote:
> Hello Maryam, hello Stephen;
>
> On Fri, Feb 7, 2025 at 1:33 PM Ariel Otilibili
> wrote:
>
> >
> >
> > On Fri, Feb 7, 2025 at 12:13 PM Maryam Tahhan wrote:
> >
> >>
> >> On 07/02/2025 10:45, Ariel Otilibili wrote:
> >>
> >> Both le
Add method cleanup_sut to framework that removes a
DPDK source from a SUT node at the conclusion of a testrun.
This will only run when the DPDK source is being copied
from the DTS engine node during the testrun
(when remote=false in the configuration).
Signed-off-by: Dean Marx
---
dts/framework/
> -Original Message-
> From: Stephen Hemminger [mailto:step...@networkplumber.org]
> Sent: Saturday, December 21, 2024 1:11 AM
> To: Wangyunjian(wangyunjian,TongTu)
> Cc: dev@dpdk.org; maxime.coque...@redhat.com; chen...@nvidia.com;
> Lilijun (Jerry) ; xiawei (H) ;
> wangzengyuan ; sta..
Hello Maryam, hello Stephen;
On Fri, Feb 7, 2025 at 1:33 PM Ariel Otilibili
wrote:
>
>
> On Fri, Feb 7, 2025 at 12:13 PM Maryam Tahhan wrote:
>
>>
>> On 07/02/2025 10:45, Ariel Otilibili wrote:
>>
>> Both legs of the loop share the same logic: the common parts are about
>> reserving and filling
Hi Thomas,
On Wed, Feb 12, 2025 at 3:00 PM Thomas Monjalon wrote:
> Applied, thanks.
>
Thanks for the feedback. Have a good day.
Ariel
Hi Konstantin,
On Fri, Feb 7, 2025 at 3:31 PM Konstantin Ananyev <
konstantin.anan...@huawei.com> wrote:
> Personally, I don't see much profit in such refactoring.
>
Then, please drop this patch. I gave room for other people to react, none
came up.
Thanks for your feedback,
Ariel
>
> > 2.30.2
> From: Thomas Monjalon [mailto:tho...@monjalon.net]
> Sent: Thursday, 13 February 2025 09.34
>
> 13/02/2025 04:09, Varghese, Vipin:
> > [AMD Official Use Only - AMD Internal Distribution Only]
> >
> > Adding Thomas and Ajit to the loop.
> >
> > Hi Ajit, we have been using the patch series for ide
13/02/2025 04:09, Varghese, Vipin:
> [AMD Official Use Only - AMD Internal Distribution Only]
>
> Adding Thomas and Ajit to the loop.
>
> Hi Ajit, we have been using the patch series for identifying the topology and
> getting l3 cache id for populating the steering tag for Device Specific Model
From: Tamar Mashiah
Add DROP action to multi destination actions, set NOP
destination type when DROP is specified.
cc: sta...@dpdk.org
Signed-off-by: Tamar Mashiah
---
drivers/net/mlx5/hws/mlx5dr_action.c | 4
drivers/net/mlx5/hws/mlx5dr_cmd.c| 4
2 files changed, 8 insertions(+
From: Tamar Mashiah
Add DROP destination to mirror clone actions.
cc: sta...@dpdk.org
Signed-off-by: Tamar Mashiah
---
.mailmap| 1 +
drivers/common/mlx5/mlx5_prm.h | 1 +
drivers/net/mlx5/mlx5_flow_hw.c | 5 +
3 files changed, 7 insertions(+)
diff --git a/.mailm
42 matches
Mail list logo
