> > Add a comment for outer-ip option in csum command.
> >
> > Set outer-ip option only when the packet is a IPv4 packet.
> >
> > Signed-off-by: Jijiang Liu
> Acked-by: Marvin Liu
Applied, thanks
2015-06-10 17:13, Mcnamara, John:
> > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Bernard Iremonger
> > Sent: Wednesday, June 10, 2015 4:13 PM
> > To: dev at dpdk.org
> > Subject: [dpdk-dev] [PATCH 0/2] doc: kni command line fixes
> >
> > Several kni command lines did not work.
> > The
2015-07-23 14:18, Liang, Cunming:
> Hi Thomas and all,
>
> This patch set postponed from v2.0, and widely reviewed during this release
> cycle.
> The packet I/O interrupt framework is the prerequisite of all PMDs to support
> packet I/O intr.
> There's no significant change since the last three
2015-07-20 11:02, Cunming Liang:
> To make bsd compiling happy with new intr changes.
>
> Signed-off-by: Cunming Liang
The dummy BSD functions must be added along the Linux ones to avoid
breaking BSD build in the middle of the series.
2015-07-27 16:54, Bernard Iremonger:
> Bernard Iremonger (4):
> bonding: add support for PCI Port Hotplug
> test-pmd: modified testpmd for link_bonding
> bonding: free queue memory in close function
> testpmd_app_ug: add example of re-attaching bonded port
Applied with last mintue fixes
2015-07-27 16:54, Bernard Iremonger:
> uint8_t need_reconfig_queues; /**< need reconfiguring
> queues or not */
> uint8_t rss_flag; /**< enable rss or not */
> - uint8_t dcb_flag; /**< enable dcb */
> + uint8_t
2015-07-27 16:54, Bernard Iremonger:
> --- a/drivers/net/bonding/rte_eth_bond_version.map
> +++ b/drivers/net/bonding/rte_eth_bond_version.map
> @@ -5,6 +5,7 @@ DPDK_2.0 {
> rte_eth_bond_8023ad_setup;
> rte_eth_bond_active_slaves_get;
> rte_eth_bond_create;
> + rte_eth
2015-07-20 11:02, Cunming Liang:
> +static int turn_on_intr(struct lcore_conf *qconf)
> +{
> + int i;
> + struct lcore_rx_queue *rx_queue;
> + uint8_t port_id, queue_id;
> +
> + for (i = 0; i < qconf->n_rx_queue; ++i) {
> + rx_queue = &(qconf->rx_queue_list[i])
From: Robert Sanford
Eliminate problematic race condition in rte_timer_manage() that can
lead to corruption of per-lcore pending-lists (implemented as
skip-lists). The race condition occurs when rte_timer_manage() expires
multiple timers on lcore A, while lcore B simultaneously invokes
rte_timer_
From: Robert Sanford
Add new timer-manage race-condition test: We wrote a test to confirm
our suspicion that we could crash rte_timer_manage() under the right
circumstances. We repeatedly set several timers to expire at roughly
the same time on the master core. The master lcore just delays and ru
From: Robert Sanford
Fix app/test timer stress test 2: Sometimes this test fails and
seg-faults because the slave lcores get out of phase with the master.
The master uses a single int, 'ready', to synchronize multiple slave
lcores through multiple phases of the test.
To resolve, we construct sim
From: Robert Sanford
This patchset fixes a bug in timer stress test 2, adds a new stress test
to expose a race condition bug in API rte_timer_manage(), and then fixes
the rte_timer_manage() bug.
Description of rte_timer_manage() race condition bug: Through code
inspection, we notice a potential
2015-07-27 15:46, Sanford, Robert:
> Hi Thomas,
>
> > Please, could you re-send this serie after having added the description
> >of
> > each patch in the commit messages?
>
> Yes, I will move the paragraphs that begin with "Patch n" from patch 0 to
> their respective patches.
>
> > It seems you
Signed-off-by: Bernard Iremonger
---
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 13 +
1 file changed, 13 insertions(+)
diff --git a/doc/guides/testpmd_app_ug/testpmd_funcs.rst
b/doc/guides/testpmd_app_ug/testpmd_funcs.rst
index 4652962..5272bf4 100644
--- a/doc/guides/testpmd_app
add function bond_ethdev_free_queues() and call from the bond_ethdev_close()
function.
Signed-off-by: Bernard Iremonger
---
drivers/net/bonding/rte_eth_bond_pmd.c | 21 -
drivers/net/bonding/rte_eth_bond_private.h | 2 +-
2 files changed, 21 insertions(+), 2 deletions(-
When the bonded port is started it also starts the slave port,
but the slave port status is not set. A slave_flag has been
added to struct rte_port to resolve this issue.
Signed-off-by: Bernard Iremonger
---
app/test-pmd/cmdline.c | 4 +++-
app/test-pmd/testpmd.c | 22 --
ap
This patch depends on the Port Hotplug Framework.
It implements the rte_dev_uninit_t() function for the link bonding pmd.
Signed-off-by: Bernard Iremonger
---
drivers/net/bonding/rte_eth_bond.h | 13 -
drivers/net/bonding/rte_eth_bond_api.c | 80 ++--
Changes in V7:
Rebase.
Move call of bond_ethdev_free_queues() from bond_ethdev_stop() to
bond_ethdev_close().
Add patch 4, doc patch for testpmd_app_ug, adds bonding example.
Changes in V6:
Rebase.
add rte_eth_bond_free() to rte_eth_bond_version.map
Changes in V5:
Rebase.
Refactor initialisation
2015-07-23 17:12, Pablo de Lara:
> With the new hash implementation, the minimum lookup burst size
> to get good performance is 8, since its internal pipeline
> consists of 4 stages of 2 entries each, so to avoid
> duplication, burst size should be 8 or more entries.
>
> Signed-off-by: Pablo de La
To keep consistency with another repo, I add macro definitions about
FM10K_IS_VALID_ETHER_ADD in fm10k_type.h, though they have already
been defined in fm10k_osdep.h.
Signed-off-by: Wang Xiao W
---
drivers/net/fm10k/base/fm10k_type.h | 17 +++--
1 file changed, 15 insertions(+), 2 de
When a VF issues an LPORT_STATE request to enable a port which is
already enabled, the PF will first disable the VF. Then it is supposed
to re-enable the VF again with new settings. This is primarily done in
order to ensure that the switch management software properly clears the
previous VF setting
The reference to err_no was left around after an old re-factor. We never
use this value again, and the macros called on the function appear to
have no relevant side effect I could see. Discovered via cppcheck
fm10k_mbx.c:1312: (style) Variable 'err_no' is assigned a value that is never
used.
This
The SYSTIME_CFG.Adjust field has a Direction bit to indicate whether the
adjustment is positive or negative. However, we incorrectly read the
documentation and the direction bit should be set 1 when positive, not
when negative.
Signed-off-by: Wang Xiao W
---
drivers/net/fm10k/base/fm10k_pf.c |
This patch adds #pragma pack(push, 1) around some structures which are passed
via TLV messages. These structures must not be left unpacked as GCC and
other compilers are wont to do. Otherwise, we get invalid message
responses from the Switch Manager software since it sends 20 bytes and
we expect 24
During initialization, the VF counts its rings by walking the TQDLOC
registers. This only works if the TQMAP/RQMAP registers are set to map
the out-of-bound rings to the first one, so the VF driver can detect when
it has run out of queues cleanly. Update the PF to reset the empty
TQMAP/RQMAP regist
VFs were being improperly added to the switch's multicast group. The
error stems from the fact that incorrect arguments were passed to the
"update_mc_addr" function. It would seem to be a copy paste error since
the parameters are similar to the "update_uc_addr" function.
Signed-off-by: Wang Xiao W
To prevent divide-by-zero issues, ensure that we always have an ITR
scale. Default to Gen3 scaling if we don't know the speed. Also ensure
the VF checks the register value and ensures we use Gen3 if we are
provided a zero value.
Signed-off-by: Wang Xiao W
---
drivers/net/fm10k/base/fm10k_pf.c |
In a recent change, the ITR scale was given to the VF via TDLEN
registers at driver initialization. This allows the VF to determine what
the ITR scale for PCIe link speed should be. However, a VF unload
followed by a reload incorrectly left this value as 0. Thus, the driver
reloads and uses 0 as th
When we call update_max_size, it does not drop all oversized messages.
This is due to the difficulty in performing this operation, since it is
a FIFO which makes updating anything other than head or tail very
difficult. To fix this, modify validate_msg_size to ensure that we error
out later when tr
After shutting down the mailbox by force, we then go about resetting max
size to 0, and clearing all messages in the FIFO. However, we should
just reset the head pointer so that the FIFO will become empty, rather than
changing the max size to 0. This helps prevent increment in tx_dropped
counter du
Red Rock Canyon's interrupt throttle timers are based on the PCIe link
speed. Because of this, the value being programmed into the ITR
registers must be scaled.
For the PF, this is as simple as reading the PCIe link speed and storing
the result. However, in the case of SR-IOV, the VF's interrupt t
The header comment included a miscopy of a C-code line, and also
mis-used Rx FIFO when it clearly meant Tx FIFO.
Signed-off-by: Wang Xiao W
---
drivers/net/fm10k/base/fm10k_mbx.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/drivers/net/fm10k/base/fm10k_mbx.c
b/drivers/n
Set PF queues used for VMDq to unlimited bandwidth when virtualization
resources are assigned.
Signed-off-by: Wang Xiao W
---
drivers/net/fm10k/base/fm10k_pf.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/net/fm10k/base/fm10k_pf.c
b/drivers/net/fm10k/base/fm10k_
This patch resolves a bug in Linux where we called the
request_tx_timestamp_mode function that is undefined for VF. Implement a
no-op function that simply ensures that the mode is NONE, otherwise it
would fail with ERR_PARAM.
Signed-off-by: Wang Xiao W
---
drivers/net/fm10k/base/fm10k_vf.c | 23
We need a handler function to be able to listen for Tx timestamp mode
responses. Without this, core driver code for PTP can't determine if the
Timestamp mode request was successful. This was overlooked in the
previous commit.
Signed-off-by: Wang Xiao W
---
drivers/net/fm10k/base/fm10k_pf.c | 6
A quick review of this long pending patch would be great.
Thanks
2015-07-16 14:45, xuelin.shi at freescale.com:
> From: Xuelin Shi
>
> 1. cpu use data owned by ixgbe must use rte_le_to_cpu_xx(...)
> 2. cpu fill data to ixgbe must use rte_cpu_to_le_xx(...)
> 3. checking pci status with converted
From: "Chen Jing D(Mark)"
fm10k has the capability to do checksum offload in TX side. This
change will expose the capability to application in infos_get
function.
Signed-off-by: Chen Jing D(Mark)
---
drivers/net/fm10k/fm10k_ethdev.c |6 +-
1 files changed, 5 insertions(+), 1 deletions(
> As Steve pointed out, "ixgbe: fix release queue mbufs" is not complete.
> As at queue stop we don't reset vector related rx queue fields to their
> initial values.
>
> Reported-by: Cunming Liang
> Signed-off-by: Konstantin Ananyev
Applied, thanks
> > This patch fixes parsing value of core variable in pipeline config.
> > Before not every combination of cores (c), sockets (s) and
> > hyperthreading (h) was parsed correctly.
> >
> > Signed-off-by: Maciej Gajdzica
>
> Acked-by: Cristian Dumitrescu
Applied, thanks
Hi Thomas,
> Please, could you re-send this serie after having added the description
>of
> each patch in the commit messages?
Yes, I will move the paragraphs that begin with "Patch n" from patch 0 to
their respective patches.
> It seems you fix 2 bugs in the first patch. It may be clearer to spl
2015-07-27 09:42, Iremonger, Bernard:
> Seeing a build error on Linux with build of bnx2x enabled
>
> == Build drivers/net/bnx2x
> CC bnx2x.o
> /root/dpdk_sforge_2/drivers/net/bnx2x/bnx2x.c:50:18: fatal error: zlib.h: No
> such file or directory
> #include
bnx2x requires zlib to be installed
2015-07-03 14:58, Jijiang Liu:
> Add a VXLAN sample guide in the sample_app_ug directory.
>
> It includes:
>
> - Add the overlay networking picture with svg format.
>
> - Add the TEP termination framework picture with svg format.
>
> - Add the tep_termination.rst file
>
> -
2015-07-27 13:16, Jan Viktorin:
> I am building in a slightly non-standard way, because I tried to integrate
> DPDK into Buildroot system.
> The problem is with kzalloc and kfree which are undefined in the igb_uio
> driver. I do not use that
> driver, however, it has failed to build. See below:
>
Hi Tony,
Sorry for the late reply, I was on vacation.
I'll prepare patch soon.
Regards,
Vladimir
2015-07-22 10:55 GMT+03:00 Tony Lu :
> Hi, Vladimir
>
> When compiling thash for no-X86 arches, it fails with the following errors.
> I wonder if
> it is possible to make the thash library arch-inde
> fm10k has the capability to do checksum offload in TX side. This
> change will expose the capability to application in infos_get
> function.
>
> Signed-off-by: Chen Jing D(Mark)
Applied, thanks
Those of you who have looked at the website for the event
(https://dpdksummit.com/us/en/userspace2015) will have seen that we're planning
to have 3 different types of presentations:
Standard presentation sessions:
- Scheduled for both Thursday and Friday mornings.
- Opportunity to present on a t
On Mon, Jul 27, 2015 at 2:09 PM, Stephen Hemminger <
stephen at networkplumber.org> wrote:
> On Mon, 27 Jul 2015 13:40:08 -0700
> Ravi Kerur wrote:
>
> > On Sun, Jul 26, 2015 at 2:54 PM, Thomas Monjalon <
> thomas.monjalon at 6wind.com>
> > wrote:
> >
> > > Hi Ravi,
> > > It seems to be a nice im
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Monday, July 27, 2015 3:18 PM
> To: Ananyev, Konstantin
> Cc: xuelin.shi at freescale.com; dev at dpdk.org
> Subject: Re: [PATCH v4] enforce rules of the cpu and ixgbe exchange data.
>
> A quick re
2015-07-27 13:17, Sergio Gonzalez Monroy:
> Fix error where TSC freq is 0.
>
> The logical OR operator evaluates to 1 if any of its operands is
> different than 0.
>
> Error showed later while initializing PMD:
> EAL: TSC frequency is ~0 KHz
>
> PMD: eth_ixgbe_dev_init(): Hardware Initialization
As Steve pointed out, "ixgbe: fix release queue mbufs" is not complete.
As at queue stop we don't reset vector related rx queue fields to their
initial values.
Reported-by: Cunming Liang
Signed-off-by: Konstantin Ananyev
---
drivers/net/ixgbe/ixgbe_rxtx.c | 5 +
1 file changed, 5 insertions
On Mon, 27 Jul 2015 13:40:08 -0700
Ravi Kerur wrote:
> On Sun, Jul 26, 2015 at 2:54 PM, Thomas Monjalon 6wind.com>
> wrote:
>
> > Hi Ravi,
> > It seems to be a nice improvement but it needs some cleanup.
> >
> > Checkpatch returns some errors.
> >
> > 2015-04-22 14:06, Ravi Kerur:
> > > use pth
On Sun, Jul 26, 2015 at 2:54 PM, Thomas Monjalon
wrote:
> Hi Ravi,
> It seems to be a nice improvement but it needs some cleanup.
>
> Checkpatch returns some errors.
>
> 2015-04-22 14:06, Ravi Kerur:
> > use pthread_setname_np and pthread_set_name_np for Linux and
> > FreeBSD respectively.
> > Re
Fix error where TSC freq is 0.
The logical OR operator evaluates to 1 if any of its operands is
different than 0.
Error showed later while initializing PMD:
EAL: TSC frequency is ~0 KHz
PMD: eth_ixgbe_dev_init(): Hardware Initialization Failure: -30
EAL: Error - exiting with code: 1
Cause: Req
Hello Thomas,
I am building in a slightly non-standard way, because I tried to integrate DPDK
into Buildroot system.
The problem is with kzalloc and kfree which are undefined in the igb_uio
driver. I do not use that
driver, however, it has failed to build. See below:
/usr/bin/make -j5 -C
/home
> -Original Message-
> From: Gonzalez Monroy, Sergio
> Sent: Monday, July 27, 2015 1:07 PM
> To: Mcnamara, John
> Subject: [PATCH] eal: fix set_tsc_freq
>
> The logical OR operator evaluates to 1 if any of its operands is different
> than 0.
>
> Fixes: 040cf8a41187 ("eal: deduplicate time
Hi Konstantin,
Thanks! Another question I would have: why does _recv_raw_pkts_vec()
insist on (nb_pkts > RTE_IXGBE_VPMD_RX_BURST)? Looking at the code it
should be able to return packets when nb_pkts >=
RTE_IXGBE_DESCS_PER_LOOP. The split_flags check in
ixgbe_recv_scattered_pkts_vec() would be
2015-07-27 08:31, Iremonger, Bernard:
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > 2015-07-21 16:16, Bernard Iremonger:
> > > add function bond_ethdev_free_queues() and call from the
> > bond_ethdev_stop() function.
> >
> > Other drivers free their queues when closing.
> > Why
Hi Stephen,
> -Original Message-
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Stephen
> Hemminger
> Sent: Monday, July 20, 2015 5:33 PM
> To: harish.patil at qlogic.com
> Cc: dev at dpdk.org
> Subject: [dpdk-dev] [PATCH v5 4/4] bnx2x: enable PMD build
>
> This is build infras
> -Original Message-
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> Sent: Monday, July 27, 2015 3:48 AM
> To: Iremonger, Bernard
> Cc: dev at dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v6 3/3] bonding: free queue memory in stop
> function
>
> 2015-07-21 16:16, Bernard Irem
2015-07-07 14:09, Bernard Iremonger:
> This patch depends on the Port Hotplug Framework.
> It implements the rte_dev_uninit_t() function for the ring pmd.
>
> Signed-off-by: Bernard Iremonger
> Acked-by: Bruce Richardson
Applied.
The EAL must reworked without fake PCI before any other change to
2015-07-21 16:16, Bernard Iremonger:
> add function bond_ethdev_free_queues() and call from the bond_ethdev_stop()
> function.
Other drivers free their queues when closing.
Why is it done in stop() for bonding?
> >Stephen Hemminger (4):
> > eal: provide functions to access PCI config
> > bnx2x: driver core
> > bnx2x: driver support routines
> > bnx2x: enable PMD build
>
> Acked-by: Harish Patil
This driver is too big, its style do not comply with the DPDK coding style,
there is no git history (drop
2015-07-27 02:56, Thomas Monjalon:
> v9 was a subset of previous deduplications by Ravi Kerur.
> This v10 address the comments I've done on v9.
>
> Ravi Kerur (3):
> eal: deduplicate lcore initialization
> eal: deduplicate timer functions
> eal: deduplicate memory initialization
Applied sho
From: Ravi Kerur
Move common functions from BSD/Linux to eal_common_memory.c file.
BSD uses contigmem kernel module and Linux uses /proc/self/pagemap file.
Signed-off-by: Ravi Kerur
Signed-off-by: Thomas Monjalon
---
lib/librte_eal/bsdapp/eal/eal_memory.c| 39 ++-
lib/librte_e
From: Ravi Kerur
Move common functions from BSD/Linux to eal_common_timer.c.
BSD uses sysctl and Linux uses CLOCK_MONOTIC_RAW to calibrate TSC.
HPET is specific to Linux and not integrated in the common init.
Signed-off-by: Ravi Kerur
Signed-off-by: Thomas Monjalon
---
lib/librte_eal/bsdapp/e
From: Ravi Kerur
Implement cpu_detected() for BSD.
Move common function in eal_lcore.c to eal_common_lcore.c file.
Signed-off-by: Ravi Kerur
---
lib/librte_eal/bsdapp/eal/Makefile | 1 +
lib/librte_eal/bsdapp/eal/eal_lcore.c | 72 +-
.../eal/ea
v9 was a subset of previous deduplications by Ravi Kerur.
This v10 address the comments I've done on v9.
Ravi Kerur (3):
eal: deduplicate lcore initialization
eal: deduplicate timer functions
eal: deduplicate memory initialization
lib/librte_eal/bsdapp/eal/Makefile | 2 +
2015-07-25 12:36, Ravi Kerur:
> --- a/lib/librte_eal/bsdapp/eal/eal_memory.c
> +++ b/lib/librte_eal/bsdapp/eal/eal_memory.c
[...]
> +/*
> + * Wrapper function to initialize contigmem.
> + */
> +int
> +rte_eal_hugepage_init(void)
> +{
> + return rte_eal_contigmem_init();
> +}
The useless wrappe
2015-07-25 12:36, Ravi Kerur
> static int
> set_tsc_freq_from_sysctl(void)
> {
> size_t sz;
> int tmp;
> + uint64_t tsc_hz;
>
> sz = sizeof(tmp);
> tmp = 0;
> @@ -94,42 +77,23 @@ set_tsc_freq_from_sysctl(void)
> else if (tmp != 1)
> RTE_LOG(WARNI
2015-07-25 12:36, Ravi Kerur:
> +static void
> +set_tsc_freq_fallback(void)
> +{
> + RTE_LOG(WARNING, EAL, "WARNING: clock_gettime cannot use "
> + "CLOCK_MONOTONIC_RAW and HPET is not available"
Not generic enough for BSD.
We can say "TSC frequency estimated roughly"
> +
2015-07-24 19:11, Jan Viktorin:
> The missing header prevents to build with linux v3.18.
Please could you give more details of why build is broken?
What is the error message?
I don't remember having seen such issue.
Thanks
2015-07-23 09:15, David Marchand:
> Here is some trivial cleanup for eal.
>
> Nothing really astounding but this will ease an unfinished patchset I have
> been working on for some time (getting rid of any driver specific stuff in
> eal).
>
> David Marchand (2):
> eal: remove dependency on drive
2015-07-23 09:15, David Marchand:
> --- a/app/test-pmd/cmdline.c
> +++ b/app/test-pmd/cmdline.c
> @@ -84,7 +84,6 @@
> #include
> #include
> #include
> -#include
This one is "needed" to check bypass support.
The bypass support should be managed inside the ixgbe driver, and the
specific code
73 matches
Mail list logo
