[dpdk-dev] [PATCH 1/2] test/ipsec: fix return values

[Gagandeep Singh]

During SA creation, if the required algorithm is not supported,
drivers can return ENOTSUP. But in most of the IPsec test cases,
if the SA creation does not success, it just returns
TEST_FAILED.

This patch fixes this issue by returning the actual return values
from the driver to the application, so that it can make decisions
whether the test case is passed, failed or unsupported.

Fixes: 05fe65eb66b2 ("test/ipsec: introduce functional test")
Cc: konstantin.anan...@intel.com

Signed-off-by: Gagandeep Singh 
---
 app/test/test_ipsec.c | 32 
 1 file changed, 16 insertions(+), 16 deletions(-)

diff --git a/app/test/test_ipsec.c b/app/test/test_ipsec.c
index 9ad07a1..d18220a 100644
--- a/app/test/test_ipsec.c
+++ b/app/test/test_ipsec.c
@@ -744,7 +744,7 @@ create_sa(enum rte_security_session_action_type action_type,
ut->ss[j].type = action_type;
rc = create_session(ut, &ts->qp_conf, ts->valid_dev, j);
if (rc != 0)
-   return TEST_FAILED;
+   return rc;
 
rc = rte_ipsec_sa_init(ut->ss[j].sa, &ut->sa_prm, sz);
rc = (rc > 0 && (uint32_t)rc <= sz) ? 0 : -EINVAL;
@@ -1247,7 +1247,7 @@ test_ipsec_crypto_inb_burst_null_null(int i)
test_cfg[i].replay_win_sz, test_cfg[i].flags, 0);
if (rc != 0) {
RTE_LOG(ERR, USER1, "create_sa failed, cfg %d\n", i);
-   return TEST_FAILED;
+   return rc;
}
 
/* Generate test mbuf data */
@@ -1349,7 +1349,7 @@ test_ipsec_crypto_outb_burst_null_null(int i)
test_cfg[i].replay_win_sz, test_cfg[i].flags, 0);
if (rc != 0) {
RTE_LOG(ERR, USER1, "create_sa failed, cfg %d\n", i);
-   return TEST_FAILED;
+   return rc;
}
 
/* Generate input mbuf data */
@@ -1458,7 +1458,7 @@ test_ipsec_inline_crypto_inb_burst_null_null(int i)
test_cfg[i].replay_win_sz, test_cfg[i].flags, 0);
if (rc != 0) {
RTE_LOG(ERR, USER1, "create_sa failed, cfg %d\n", i);
-   return TEST_FAILED;
+   return rc;
}
 
/* Generate inbound mbuf data */
@@ -1536,7 +1536,7 @@ test_ipsec_inline_proto_inb_burst_null_null(int i)
test_cfg[i].replay_win_sz, test_cfg[i].flags, 0);
if (rc != 0) {
RTE_LOG(ERR, USER1, "create_sa failed, cfg %d\n", i);
-   return TEST_FAILED;
+   return rc;
}
 
/* Generate inbound mbuf data */
@@ -1644,7 +1644,7 @@ test_ipsec_inline_crypto_outb_burst_null_null(int i)
test_cfg[i].replay_win_sz, test_cfg[i].flags, 0);
if (rc != 0) {
RTE_LOG(ERR, USER1, "create_sa failed, cfg %d\n", i);
-   return TEST_FAILED;
+   return rc;
}
 
/* Generate test mbuf data */
@@ -1722,7 +1722,7 @@ test_ipsec_inline_proto_outb_burst_null_null(int i)
test_cfg[i].replay_win_sz, test_cfg[i].flags, 0);
if (rc != 0) {
RTE_LOG(ERR, USER1, "create_sa failed, cfg %d\n", i);
-   return TEST_FAILED;
+   return rc;
}
 
/* Generate test mbuf data */
@@ -1798,7 +1798,7 @@ test_ipsec_lksd_proto_inb_burst_null_null(int i)
test_cfg[i].replay_win_sz, test_cfg[i].flags, 0);
if (rc != 0) {
RTE_LOG(ERR, USER1, "create_sa failed, cfg %d\n", i);
-   return TEST_FAILED;
+   return rc;
}
 
/* Generate test mbuf data */
@@ -1911,7 +1911,7 @@ test_ipsec_replay_inb_inside_null_null(int i)
test_cfg[i].replay_win_sz, test_cfg[i].flags, 0);
if (rc != 0) {
RTE_LOG(ERR, USER1, "create_sa failed, cfg %d\n", i);
-   return TEST_FAILED;
+   return rc;
}
 
/* Generate inbound mbuf data */
@@ -2004,7 +2004,7 @@ test_ipsec_replay_inb_outside_null_null(int i)
test_cfg[i].replay_win_sz, test_cfg[i].flags, 0);
if (rc != 0) {
RTE_LOG(ERR, USER1, "create_sa failed, cfg %d\n", i);
-   return TEST_FAILED;
+   return rc;
}
 
/* Generate test mbuf data */
@@ -2104,7 +2104,7 @@ test_ipsec_replay_inb_repeat_null_null(int i)
test_cfg[i].replay_win_sz, test_cfg[i].flags, 0);
if (rc != 0) {
RTE_LOG(ERR, USER1, "create_sa failed, cfg %d\n", i);
-   return TEST_FAILED;
+   return rc;
}
 
/* Generate test mbuf data */
@@ -2205,7 +2205,7 @@ test_ipsec_replay_inb_inside_burst_null_null(int i)
test_cfg[i].replay_win_sz, test_cfg[i].flags, 0);
if (rc != 0) {
RTE_LOG(ERR, USER1, "create_sa failed, cfg %d\n", i);
-   return TEST_FAILED;
+   return rc;
}
 

[dpdk-dev] [PATCH 2/2] crypto/dpaa2_sec: fix if check on memory allocation

[Gagandeep Singh]

When key length is 0, zmalloc will return NULL pointor
and in that case it should not return NOMEM.
So in this patch, adding a check on key length.

Fixes: 8d1f3a5d751b ("crypto/dpaa2_sec: support crypto operation")
Cc: akhil.go...@nxp.com
---
 drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c 
b/drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c
index cab79db..05b194c 100644
--- a/drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c
+++ b/drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c
@@ -1842,7 +1842,7 @@ dpaa2_sec_cipher_init(struct rte_cryptodev *dev,
session->ctxt_type = DPAA2_SEC_CIPHER;
session->cipher_key.data = rte_zmalloc(NULL, xform->cipher.key.length,
RTE_CACHE_LINE_SIZE);
-   if (session->cipher_key.data == NULL) {
+   if (session->cipher_key.data == NULL && xform->cipher.key.length > 0) {
DPAA2_SEC_ERR("No Memory for cipher key");
rte_free(priv);
return -ENOMEM;
-- 
2.7.4

Re: [dpdk-dev] [PATCH 3/4] net/mlx5: fix secondary process attach port Tx queue

[Raslan Darawsheh]

Hi,

> -Original Message-
> From: Suanming Mou 
> Sent: Sunday, January 24, 2021 1:02 PM
> To: Slava Ovsiienko ; Matan Azrad
> 
> Cc: Raslan Darawsheh ; dev@dpdk.org;
> sta...@dpdk.org
> Subject: [PATCH 3/4] net/mlx5: fix secondary process attach port Tx queue
> 
> Currently, the secondary process port UAR register mapping used by Tx
> queue is done during port initializing.
> 
> Unluckily, in port hot-plug case, the secondary process was requested
> to initialize the port when primary process did not complete the
> device configuration and the port Tx queue number is not configured
> yet. Hence, the secondary process getS the zero Tx queue number during
> probing, causing the UAR registers not be mapped in the correct
> fashion.
> 
> This commit checks the configured number of Tx queues in secondary
> process when the port start is requested. In case the Tx queue
> number mismatch found the UAR mapping is reinitialized accordingly.
> 
> Fixes: 2aac5b5d119f ("net/mlx5: sync stop/start with secondary process")
> cc: sta...@dpdk.org
> 
> Signed-off-by: Suanming Mou 
> Acked-by: Viacheslav Ovsiienko 
> ---
>  drivers/net/mlx5/linux/mlx5_mp_os.c | 19 +++
>  drivers/net/mlx5/mlx5.c |  2 +-
>  drivers/net/mlx5/mlx5.h |  6 +-
>  3 files changed, 25 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/net/mlx5/linux/mlx5_mp_os.c
> b/drivers/net/mlx5/linux/mlx5_mp_os.c
> index 08ade75..95372e2 100644
> --- a/drivers/net/mlx5/linux/mlx5_mp_os.c
> +++ b/drivers/net/mlx5/linux/mlx5_mp_os.c
[...]
> 
> -/* Per-process private structure. */
> +/*
> + * Per-process private structure.
> + * Caution, secondary pocess may rebuid the struct during port start.
> + */
Typo in
pocess-> process
rebuid-> rebuild
will fix during integration.
[...]
> 1.8.3.1

Kindest regards
Raslan Darawsheh

Re: [dpdk-dev] [PATCH 1/4] net/mlx5: fix invalid multi-process ID

[Raslan Darawsheh]

Hi,
> -Original Message-
> From: Suanming Mou 
> Sent: Sunday, January 24, 2021 1:02 PM
> To: Slava Ovsiienko ; Matan Azrad
> 
> Cc: Raslan Darawsheh ; dev@dpdk.org
> Subject: [PATCH 1/4] net/mlx5: fix invalid multi-process ID
> 
> The device port_id is used for inter-process communication and must
> be the same both for primary and secondary process
> 
> This IPC port_id was configured with the invalid temporary value in
> port spawn routine. This temporary value was used by the function
> rte_eth_dev_get_port_by_name() to check whether the port exists.
> 
> This commit corrects the mp port_id with rte_eth_dev port_id.
> 
> Fixes: 2eb4d0107acc ("net/mlx5: refactor PCI probing on Linux")
> 
Missing Cc: sta...@dpdk.org
Will add during integration.
> Signed-off-by: Suanming Mou 
> Acked-by: Viacheslav Ovsiienko 
> ---


Kindest regards,
Raslan Darawsheh

[dpdk-dev] [PATCH v4 0/2] fix 'max-pkt-len' errors

[Steve Yang]

Here fixed 3 issues for 'max-pkt-len':
1. When cmdline option '--max-pkt-len' set the value less then
   '1500 + overhead', the app/testpmd will force to resize the 'max-pkt-len'
   to '1500 + overhead'. However, the user really want to configure
   'max-pkt-len' to a specified value (< 1500 + overhead);

2. If the large value of '--max-pkt-len' gave (e.g.: 8000), and user want to
   reset the value to a small one (e.g.: 1400), it will became invalid due to
   JUMBO_FRAME offload state doesn't change before port started;

3. When rx/tx queue offloads capabilities aren't specified, the rx/tx queue
   setup will be failed once the port offloads changed.

---
v4:
 * combined testpmd patches;
 * updated the commit log for patch 2;
v3:
 * rebased code to latest;
 * splited to 3 commits;

v2:
 * moved the update logic to 'rxtx_port_config';
 * added the 'tx_conf' part;
 * optimized the 'default' assignment;
---

Steve Yang (2):
  ethdev: fix MTU doesn't update when jumbo frame disabled
  app/testpmd: fix max-pkt-len option invalid

 app/test-pmd/cmdline.c |  1 +
 app/test-pmd/parameters.c  |  1 +
 app/test-pmd/testpmd.c | 63 +-
 app/test-pmd/testpmd.h |  2 ++
 lib/librte_ethdev/rte_ethdev.c |  8 ++---
 5 files changed, 54 insertions(+), 21 deletions(-)

-- 
2.17.1

[dpdk-dev] [PATCH v4 1/2] ethdev: fix MTU doesn't update when jumbo frame disabled

[Steve Yang]

The MTU value should be updated to 'max_rx_pkt_len - overhead'
no matter if the JUMBO FRAME offload enabled. If not update this MTU,
use will get the wrong MTU info via some command.
E.g.: 'show port info all' in testpmd tool.

Actually, the 'max_rx_pkt_len' has been used for other purposes in many
places now, even though the 'max_rx_pkt_len' is expected 'Only used if
JUMBO_FRAME enabled'.

For examples,
'max_rx_pkt_len' perhaps can be used as the 'rx_ctx.rxmax' in i40e.

Fixes: bf0f90d92d30 ("ethdev: fix max Rx packet length check")

Signed-off-by: Steve Yang 
---
 lib/librte_ethdev/rte_ethdev.c | 8 
 1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/lib/librte_ethdev/rte_ethdev.c b/lib/librte_ethdev/rte_ethdev.c
index daf5f24f7e..42857e3b67 100644
--- a/lib/librte_ethdev/rte_ethdev.c
+++ b/lib/librte_ethdev/rte_ethdev.c
@@ -1421,10 +1421,6 @@ rte_eth_dev_configure(uint16_t port_id, uint16_t 
nb_rx_q, uint16_t nb_tx_q,
ret = -EINVAL;
goto rollback;
}
-
-   /* Scale the MTU size to adapt max_rx_pkt_len */
-   dev->data->mtu = dev->data->dev_conf.rxmode.max_rx_pkt_len -
-   overhead_len;
} else {
uint16_t pktlen = dev_conf->rxmode.max_rx_pkt_len;
if (pktlen < RTE_ETHER_MIN_MTU + overhead_len ||
@@ -1434,6 +1430,10 @@ rte_eth_dev_configure(uint16_t port_id, uint16_t 
nb_rx_q, uint16_t nb_tx_q,
RTE_ETHER_MTU + overhead_len;
}
 
+   /* Scale the MTU size to adapt max_rx_pkt_len */
+   dev->data->mtu = dev->data->dev_conf.rxmode.max_rx_pkt_len -
+   overhead_len;
+
/*
 * If LRO is enabled, check that the maximum aggregated packet
 * size is supported by the configured device.
-- 
2.17.1

[dpdk-dev] [PATCH v4 2/2] app/testpmd: fix max-pkt-len option invalid

[Steve Yang]

Moved the setting of 'DEV_RX_OFFLOAD_JUMBO_FRAME' from
'cmd_config_max_pkt_len_parsed()' to 'init_config()' caused fail the case
where 'max_rx_pkt_len' is changed from the command line via
"port config all max-pkt-len".

The 'init_config()' function is only called when testpmd is started,
but the DEV_RX_OFFLOAD_JUMBO_FRAME setting needs to be recomputed whenever
'max_rx_pkt_len' changes.

Define the 'update_jumbo_frame_offload()' function for both 'init_config()'
and 'cmd_config_max_pkt_len_parsed()', and detect if 'max_rx_pkt_len'
should be update to 1500 + overhead as default configuration. At the same
time, the offloads of rx queue also should update the value once port
offloads changed (e.g.: disabled JUMBO_FRAME offload via changed
max-pkt-len, reproduce steps [1]), otherwise, it would cause unexpected
result.

[1]
--
./x86_64-native-linuxapp-gcc/app/dpdk-testpmd  -c 0xf -n 4  -- -i
--max-pkt-len=9000 --tx-offloads=0x8000 --rxq=4 --txq=4 --disable-rss
testpmd>  set verbose 3
testpmd>  port stop all
testpmd>  port config all max-pkt-len 1518  port start all

// Got fail error info without this patch
Configuring Port 0 (socket 1)
Ethdev port_id=0 rx_queue_id=0, new added offloads 0x800 must be
within per-queue offload capabilities 0x0 in rte_eth_rx_queue_setup()
Fail to configure port 0 rx queues //<-- Fail error info;
--

Fixes: 761c4d6690 ("app/testpmd: fix max Rx packet length for VLAN packets")

Signed-off-by: Steve Yang 
---
 app/test-pmd/cmdline.c|  1 +
 app/test-pmd/parameters.c |  1 +
 app/test-pmd/testpmd.c| 63 ---
 app/test-pmd/testpmd.h|  2 ++
 4 files changed, 50 insertions(+), 17 deletions(-)

diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
index 89034c8b72..8b6b7b6206 100644
--- a/app/test-pmd/cmdline.c
+++ b/app/test-pmd/cmdline.c
@@ -1901,6 +1901,7 @@ cmd_config_max_pkt_len_parsed(void *parsed_result,
printf("Unknown parameter\n");
return;
}
+   update_jumbo_frame_offload(pid, false);
}
 
init_port_config();
diff --git a/app/test-pmd/parameters.c b/app/test-pmd/parameters.c
index df5eb10d84..1c63156ddd 100644
--- a/app/test-pmd/parameters.c
+++ b/app/test-pmd/parameters.c
@@ -833,6 +833,7 @@ launch_args_parse(int argc, char** argv)
 "total-num-mbufs should be > 
1024\n");
}
if (!strcmp(lgopts[opt_idx].name, "max-pkt-len")) {
+   default_max_pktlen = false;
n = atoi(optarg);
if (n >= RTE_ETHER_MIN_LEN)
rx_mode.max_rx_pkt_len = (uint32_t) n;
diff --git a/app/test-pmd/testpmd.c b/app/test-pmd/testpmd.c
index c256e719ae..f22d2be46d 100644
--- a/app/test-pmd/testpmd.c
+++ b/app/test-pmd/testpmd.c
@@ -531,6 +531,8 @@ uint16_t gso_max_segment_size = RTE_ETHER_MAX_LEN - 
RTE_ETHER_CRC_LEN;
 /* Holds the registered mbuf dynamic flags names. */
 char dynf_names[64][RTE_MBUF_DYN_NAMESIZE];
 
+bool default_max_pktlen = true;
+
 /*
  * Helper function to check if socket is already discovered.
  * If yes, return positive value. If not, return zero.
@@ -1410,7 +1412,6 @@ init_config(void)
struct rte_gro_param gro_param;
uint32_t gso_types;
uint16_t data_size;
-   uint16_t eth_overhead;
bool warning = 0;
int k;
int ret;
@@ -1447,22 +1448,7 @@ init_config(void)
rte_exit(EXIT_FAILURE,
 "rte_eth_dev_info_get() failed\n");
 
-   /* Update the max_rx_pkt_len to have MTU as RTE_ETHER_MTU */
-   if (port->dev_info.max_mtu != UINT16_MAX &&
-   port->dev_info.max_rx_pktlen > port->dev_info.max_mtu)
-   eth_overhead = port->dev_info.max_rx_pktlen -
-   port->dev_info.max_mtu;
-   else
-   eth_overhead =
-   RTE_ETHER_HDR_LEN + RTE_ETHER_CRC_LEN;
-
-   if (port->dev_conf.rxmode.max_rx_pkt_len <=
-   (uint32_t)(RTE_ETHER_MTU + eth_overhead))
-   port->dev_conf.rxmode.max_rx_pkt_len =
-   RTE_ETHER_MTU + eth_overhead;
-   else
-   port->dev_conf.rxmode.offloads |=
-   DEV_RX_OFFLOAD_JUMBO_FRAME;
+   update_jumbo_frame_offload(pid, default_max_pktlen);
 
if (!(port->dev_info.tx_offload_capa &
  DEV_TX_OFFLOAD_MBUF_FAST_FREE))
@@ -3358,6 +3344,49 @@ rxtx_port_config(struct rte_port *port)
}
 }
 
+void
+update_jumbo_frame_offload(portid_t po

Re: [dpdk-dev] [PATCH 1/1] ethdev: fix handling of close failure

[Andrew Rybchenko]

On 1/22/21 8:58 PM, Thomas Monjalon wrote:
> If a failure happens when closing a port,
> it was unnecessarily failing again in the function eth_err(),
> because of a check against HW removal cause.
> Indeed there is a big chance the port is released at this point.
> Given the port is in the middle (or at the end) of a close process,
> checking the error cause by accessing the port is a non-sense.
> The error check is replaced by a simple return in the close function.
> 
> Bugzilla ID: 624
> Fixes: 8a5a0aad5d3e ("ethdev: allow close function to return an error")
> Cc: sta...@dpdk.org
> 
> Reported-by: Anatoly Burakov 
> Signed-off-by: Thomas Monjalon 

Acked-by: Andrew Rybchenko 

Re: [dpdk-dev] [PATCH] examples/pipeline: fix VXLAN script permission

[David Marchand]

On Fri, Jan 22, 2021 at 1:19 PM Dumitrescu, Cristian
 wrote:
> > This python script provides a shebang that only makes sense if the
> > script has the executable bit set.
> >
> > Fixes: fde7a772701a ("examples/pipeline: add VXLAN encapsulation
> > example")
> > Cc: sta...@dpdk.org
> >
> > Reported-by: Timothy Redaelli 
> > Signed-off-by: David Marchand 
> Acked-by: Cristian Dumitrescu 

Applied.


-- 
David Marchand

Re: [dpdk-dev] [PATCH v8 2/3] build: use Python pmdinfogen

[Kinsella, Ray]

On 23/01/2021 11:38, Thomas Monjalon wrote:
> 22/01/2021 23:24, Dmitry Kozlyuk:
>> On Fri, 22 Jan 2021 21:57:15 +0100, Thomas Monjalon wrote:
>>> 22/01/2021 21:31, Dmitry Kozlyuk:
 On Wed, 20 Jan 2021 11:24:21 +0100, Thomas Monjalon wrote:  
> 20/01/2021 08:23, Dmitry Kozlyuk:  
>> On Wed, 20 Jan 2021 01:05:59 +0100, Thomas Monjalon wrote:
>>> This is now the right timeframe to introduce this change
>>> with the new Python module dependency.
>>> Unfortunately, the ABI check is returning an issue:
>>>
>>> 'const char mlx5_common_pci_pmd_info[62]' was changed
>>> to 'const char mlx5_common_pci_pmd_info[60]' at rte_common_mlx5.pmd.c   
>>>  
>>
>> Will investigate and fix ASAP.  

 Now that I think of it: strings like this change every time new PCI IDs are
 added to a PMD, but AFAIK adding PCI IDs is not considered an ABI breakage,
 is it? One example is 28c9a7d7b48e ("net/mlx5: add ConnectX-6 Lx device 
 ID")
 added 2020-07-08, i.e. clearly outside of ABI change window.  
>>>
>>> You're right.
>>>
 "xxx_pmd_info" changes are due to JSON formatting (new is more canonical),
 which can be worked around easily, if the above is wrong.  
>>>
>>> If the new format is better, please keep it.
>>> What we need is an exception for the pmdinfo symbols
>>> in the file devtools/libabigail.abignore.
>>> You can probably use a regex for these symbols.
>>
>> This would allow real breakages to pass ABI check, abidiff doesn't analyze
>> variable content and it's not easy to compare. Maybe later a script can be
>> added that checks lines with RTE_DEVICE_IN in patches. There are at most 32 
>> of
>> 5494 relevant commits between 19.11 and 20.11, though.
>>
>> To verify there are no meaningful changes I ensured empty diff between
>> results of the following command for "main" and the branch:
>>
>>  find build/drivers -name '*.so' -exec usertools/dpdk-pmdinfo.py
> 
> For now we cannot do such check as part of the ABI checker.
> And we cannot merge this patch if the ABI check fails.
> I think the only solution is to allow any change in the pmdinfo variables.
> 

So my 2c on this is that this is an acceptable work-around for the v21 (DPDK 
v20.11) ABI.
However we are going to end up carrying this rule in libabigail.ignore 
indefinitely.

Would it make sense to just fix the size of _pmd_info to some reasonably large 
value - 
say 128 bytes, to allow us to drop the rule in the DPDK 21.11 v22 release?

Ray K

Re: [dpdk-dev] [dpdk-stable] [PATCH v11 1/4] raw/ifpga: add fpga rsu function

[Ferruh Yigit]

On 1/22/2021 2:18 AM, Huang, Wei wrote:





-Original Message-
From: Ferruh Yigit 
Sent: Friday, January 22, 2021 00:30
To: Huang, Wei ; dev@dpdk.org; Xu, Rosen ; 
Zhang, Qi Z 
Cc: sta...@dpdk.org; Zhang, Tianfei ; Ray Kinsella 

Subject: Re: [dpdk-stable] [PATCH v11 1/4] raw/ifpga: add fpga rsu function

On 1/21/2021 6:03 AM, Wei Huang wrote:

RSU (Remote System Update) depends on secure manager which may be
different on various implementations, so a new secure manager device
is implemented for adapting such difference.
There are three major functions added:
1. ifpga_rawdev_update_flash() updates flash with specific image file.
2. ifpga_rawdev_stop_flash_update() aborts flash update process.
3. ifpga_rawdev_reload() reloads FPGA from updated flash.

Signed-off-by: Wei Huang 
Acked-by: Tianfei Zhang 
Acked-by: Rosen Xu 


<...>


@@ -76,4 +76,9 @@ int
   ifpga_unregister_msix_irq(enum ifpga_irq_type type,
   int vec_start, rte_intr_callback_fn handler, void *arg);

+int ifpga_rawdev_update_flash(struct rte_rawdev *dev, const char *image,
+uint64_t *status);
+int ifpga_rawdev_stop_flash_update(struct rte_rawdev *dev, int
+force); int ifpga_rawdev_reload(struct rte_rawdev *dev, int type, int
+page);
+
   #endif /* _IFPGA_RAWDEV_H_ */



Hi Wei,

Please help me understand the rawdev, who should be calling the above newly 
added functions?


>
> Hi Ferruh,
>
> Cyborg is an OpenStack project that aims to provide a general purpose 
management framework for acceleration resources (i.e. various types of 
accelerators such as GPU, FPGA, NP, ODP, DPDK/SPDK and so on).

>
> To update the FPGA flash is one of requirements from Cyborg. Originally there 
are no such interfaces, so I added them.

>
> These interfaces use rte_rawdev to identify which FPGA to access, they will 
be called in opae_update_flash(),

> opae_cancel_flash_update() and opae_reboot_device() in ifpga_opae_api.c .
>
> These opae_xxx function use PCI address to identify FPGA instead of 
rte_rawdev, so the caller has no need to know the existence of rte_rawdev.
> In fact, Cyborg is Python application, these opae_xxx functions will be 
eventually wrapped in a Python module for Cyborg to call.

>

Thanks for clarification, I see what you are doing, but still I think these APIs 
are not belong to a driver, they look like more application level.


@Wei, @Rosen, what do you think to keep only generic raw/ipfga APIs in the 
driver and move all 'ifpga_opae_api.c/h' to the sample application?


In that case raw/ifpga APIs still can get the port_id as parameter, and the 
'opae' layer in the sample can do the conversion?

Re: [dpdk-dev] [PATCH v8 2/3] build: use Python pmdinfogen

[Kinsella, Ray]

On 25/01/2021 09:25, Kinsella, Ray wrote:
> 
> 
> On 23/01/2021 11:38, Thomas Monjalon wrote:
>> 22/01/2021 23:24, Dmitry Kozlyuk:
>>> On Fri, 22 Jan 2021 21:57:15 +0100, Thomas Monjalon wrote:
 22/01/2021 21:31, Dmitry Kozlyuk:
> On Wed, 20 Jan 2021 11:24:21 +0100, Thomas Monjalon wrote:  
>> 20/01/2021 08:23, Dmitry Kozlyuk:  
>>> On Wed, 20 Jan 2021 01:05:59 +0100, Thomas Monjalon wrote:
 This is now the right timeframe to introduce this change
 with the new Python module dependency.
 Unfortunately, the ABI check is returning an issue:

 'const char mlx5_common_pci_pmd_info[62]' was changed
 to 'const char mlx5_common_pci_pmd_info[60]' at rte_common_mlx5.pmd.c  
   
>>>
>>> Will investigate and fix ASAP.  
>
> Now that I think of it: strings like this change every time new PCI IDs 
> are
> added to a PMD, but AFAIK adding PCI IDs is not considered an ABI 
> breakage,
> is it? One example is 28c9a7d7b48e ("net/mlx5: add ConnectX-6 Lx device 
> ID")
> added 2020-07-08, i.e. clearly outside of ABI change window.  

 You're right.

> "xxx_pmd_info" changes are due to JSON formatting (new is more canonical),
> which can be worked around easily, if the above is wrong.  

 If the new format is better, please keep it.
 What we need is an exception for the pmdinfo symbols
 in the file devtools/libabigail.abignore.
 You can probably use a regex for these symbols.
>>>
>>> This would allow real breakages to pass ABI check, abidiff doesn't analyze
>>> variable content and it's not easy to compare. Maybe later a script can be
>>> added that checks lines with RTE_DEVICE_IN in patches. There are at most 32 
>>> of
>>> 5494 relevant commits between 19.11 and 20.11, though.
>>>
>>> To verify there are no meaningful changes I ensured empty diff between
>>> results of the following command for "main" and the branch:
>>>
>>> find build/drivers -name '*.so' -exec usertools/dpdk-pmdinfo.py
>>
>> For now we cannot do such check as part of the ABI checker.
>> And we cannot merge this patch if the ABI check fails.
>> I think the only solution is to allow any change in the pmdinfo variables.
>>
> 
> So my 2c on this is that this is an acceptable work-around for the v21 (DPDK 
> v20.11) ABI.
> However we are going to end up carrying this rule in libabigail.ignore 
> indefinitely.
> 
> Would it make sense to just fix the size of _pmd_info to some reasonably 
> large value - 
> say 128 bytes, to allow us to drop the rule in the DPDK 21.11 v22 release?
> 
> Ray K


Another point is - shouldn't _pmd_info probably live in "INTERNAL" is anycase?

Re: [dpdk-dev] [PATCH v8 2/3] build: use Python pmdinfogen

[Dmitry Kozlyuk]

On Mon, 25 Jan 2021 09:25:51 +, Kinsella, Ray wrote:
> On 23/01/2021 11:38, Thomas Monjalon wrote:
> > 22/01/2021 23:24, Dmitry Kozlyuk:  
> >> On Fri, 22 Jan 2021 21:57:15 +0100, Thomas Monjalon wrote:  
> >>> 22/01/2021 21:31, Dmitry Kozlyuk:  
>  On Wed, 20 Jan 2021 11:24:21 +0100, Thomas Monjalon wrote:
> > 20/01/2021 08:23, Dmitry Kozlyuk:
> >> On Wed, 20 Jan 2021 01:05:59 +0100, Thomas Monjalon wrote:  
> >>> This is now the right timeframe to introduce this change
> >>> with the new Python module dependency.
> >>> Unfortunately, the ABI check is returning an issue:
> >>>
> >>> 'const char mlx5_common_pci_pmd_info[62]' was changed
> >>> to 'const char mlx5_common_pci_pmd_info[60]' at rte_common_mlx5.pmd.c 
> >>>  
> >>
> >> Will investigate and fix ASAP.
> 
>  Now that I think of it: strings like this change every time new PCI IDs 
>  are
>  added to a PMD, but AFAIK adding PCI IDs is not considered an ABI 
>  breakage,
>  is it? One example is 28c9a7d7b48e ("net/mlx5: add ConnectX-6 Lx device 
>  ID")
>  added 2020-07-08, i.e. clearly outside of ABI change window.
> >>>
> >>> You're right.
> >>>  
>  "xxx_pmd_info" changes are due to JSON formatting (new is more 
>  canonical),
>  which can be worked around easily, if the above is wrong.
> >>>
> >>> If the new format is better, please keep it.
> >>> What we need is an exception for the pmdinfo symbols
> >>> in the file devtools/libabigail.abignore.
> >>> You can probably use a regex for these symbols.  
> >>
> >> This would allow real breakages to pass ABI check, abidiff doesn't analyze
> >> variable content and it's not easy to compare. Maybe later a script can be
> >> added that checks lines with RTE_DEVICE_IN in patches. There are at most 
> >> 32 of
> >> 5494 relevant commits between 19.11 and 20.11, though.
> >>
> >> To verify there are no meaningful changes I ensured empty diff between
> >> results of the following command for "main" and the branch:
> >>
> >>find build/drivers -name '*.so' -exec usertools/dpdk-pmdinfo.py  
> > 
> > For now we cannot do such check as part of the ABI checker.
> > And we cannot merge this patch if the ABI check fails.
> > I think the only solution is to allow any change in the pmdinfo variables.
> >   
> 
> So my 2c on this is that this is an acceptable work-around for the v21 (DPDK 
> v20.11) ABI.
> However we are going to end up carrying this rule in libabigail.ignore 
> indefinitely.
> 
> Would it make sense to just fix the size of _pmd_info to some reasonably 
> large value - 
> say 128 bytes, to allow us to drop the rule in the DPDK 21.11 v22 release?

I don't think so. This is a JSON *string to be parsed;* considering its size
as part of application *binary* interface is wrong in the first place. As for
content, checking that no PCI IDs are removed is out of scope for libabigail
anyway. Technically we could fix _pmd_info size, but this still allows
breaking changes to pass the check with no benefit.

Re: [dpdk-dev] [PATCH v11 3/4] raw/ifpga: add OPAE API for OpenStack Cyborg

[Ferruh Yigit]

On 1/22/2021 2:43 AM, Huang, Wei wrote:




-Original Message-
From: Ferruh Yigit 
Sent: Friday, January 22, 2021 00:30
To: Huang, Wei ; dev@dpdk.org; Xu, Rosen ; 
Zhang, Qi Z 
Cc: sta...@dpdk.org; Zhang, Tianfei 
Subject: Re: [dpdk-dev] [PATCH v11 3/4] raw/ifpga: add OPAE API for OpenStack 
Cyborg

On 1/21/2021 6:03 AM, Wei Huang wrote:

Cyborg is an OpenStack project that aims to provide a general purpose
management framework for acceleration resources (i.e. various types of
accelerators such as GPU, FPGA, NP, ODP, DPDK/SPDK and so on).
It needs some OPAE type APIs to manage PACs (Programmable Acceleration
Card) with Intel FPGA. Below major functions are added to meets Cyborg
requirements.
1. opae_init() set up OPAE environment.
2. opae_cleanup() clean up OPAE environment.
3. opae_enumerate() searches PAC with specific FPGA.
4. opae_get_property() gets properties of FPGA.
5. opae_partial_reconfigure() perform partial configuration on FPGA.
6. opae_get_image_info() gets information of image file.
7. opae_update_flash() updates FPGA flash with specific image file.
8. opae_cancel_flash_update() cancel process of FPGA flash update.
9. opae_probe_device() manually probe specific FPGA with ifpga driver.
10. opae_remove_device() manually remove specific FPGA from ifpga driver.
11. opae_bind_driver() binds specific FPGA with specified kernel driver.
12. opae_unbind_driver() unbinds specific FPGA from kernel driver.
13. opae_reboot_device() reboots specific FPGA (do reconfiguration).

Signed-off-by: Wei Huang 
Acked-by: Tianfei Zhang 
Acked-by: Rosen Xu 


<...>


+
+RTE_INIT(init_api_env)
+{
+eal_inited = 0;
+opae_log_level = OPAE_LOG_ERR;
+opae_log_file = NULL;
+ifpga_rawdev_logtype = 0;
+
+opae_log_info("API environment is initialized\n"); }
+
+RTE_FINI(clean_api_env)
+{
+if (opae_log_file) {
+fclose(opae_log_file);
+opae_log_file = NULL;
+}
+opae_log_info("API environment is cleaned\n"); }
+


There are called in constructor and destructor, which means all DPDK 
applications have this PMD compiled will run above and print some log, is this 
really what we want here?
Is there a way to limit the initialize to the PMD context, like 'probe()' etc?



Hi Ferruh,

These log information will not be output for the default log level is 
OPAE_LOG_ERR, they are output only when log level is set to OPAE_LOG_INFO. I 
can remove them if you think it's necessary.



Although they don't output because of default log level, I think it is wrong 
that they run for _all_ dpdk applications, I still think it should run for the 
scope of the driver, not for all dpdk applications.



All these opae_xxx functions will check the input PCI address is valid or not, 
and the rawdev PMD will also filter input PCI device ID. I think it's enough to 
prevent these function to be misused.

Thanks,
Wei

Re: [dpdk-dev] [PATCH 0/4] net/mlx: fix secondary process bugs

[Raslan Darawsheh]

Hi,

> -Original Message-
> From: Suanming Mou 
> Sent: Sunday, January 24, 2021 1:02 PM
> To: Slava Ovsiienko ; Matan Azrad
> 
> Cc: Raslan Darawsheh ; dev@dpdk.org
> Subject: [PATCH 0/4] net/mlx: fix secondary process bugs
> 
> This patch series fix several secondary process bugs.
> 
> Suanming Mou (4):
>   net/mlx5: fix invalid multi-process ID
>   net/mlx5: fix secondary process port detach crash
>   net/mlx5: fix secondary process attach port Tx queue
>   net/mlx4: fix secondary process attach port Tx queue
> 
>  drivers/net/mlx4/mlx4.c | 10 +-
>  drivers/net/mlx4/mlx4.h |  4 
>  drivers/net/mlx4/mlx4_mp.c  | 24 
>  drivers/net/mlx4/mlx4_rxtx.h|  1 +
>  drivers/net/mlx4/mlx4_txq.c | 28 
>  drivers/net/mlx5/linux/mlx5_mp_os.c | 19 +++
>  drivers/net/mlx5/linux/mlx5_os.c|  4 ++--
>  drivers/net/mlx5/mlx5.c |  8 
>  drivers/net/mlx5/mlx5.h |  6 +-
>  drivers/net/mlx5/mlx5_txq.c | 21 +
>  10 files changed, 105 insertions(+), 20 deletions(-)
> 
> --
> 1.8.3.1

Series applied to next-net-mlx with small modifications mentioned on each patch 
separately. 

Kindest regards,
Raslan Darawsheh

Re: [dpdk-dev] [PATCH v8 2/3] build: use Python pmdinfogen

[Kinsella, Ray]

On 25/01/2021 10:05, Dmitry Kozlyuk wrote:
> On Mon, 25 Jan 2021 09:25:51 +, Kinsella, Ray wrote:
>> On 23/01/2021 11:38, Thomas Monjalon wrote:
>>> 22/01/2021 23:24, Dmitry Kozlyuk:  
 On Fri, 22 Jan 2021 21:57:15 +0100, Thomas Monjalon wrote:  
> 22/01/2021 21:31, Dmitry Kozlyuk:  
>> On Wed, 20 Jan 2021 11:24:21 +0100, Thomas Monjalon wrote:
>>> 20/01/2021 08:23, Dmitry Kozlyuk:
 On Wed, 20 Jan 2021 01:05:59 +0100, Thomas Monjalon wrote:  
> This is now the right timeframe to introduce this change
> with the new Python module dependency.
> Unfortunately, the ABI check is returning an issue:
>
> 'const char mlx5_common_pci_pmd_info[62]' was changed
> to 'const char mlx5_common_pci_pmd_info[60]' at rte_common_mlx5.pmd.c 
>  

 Will investigate and fix ASAP.
>>
>> Now that I think of it: strings like this change every time new PCI IDs 
>> are
>> added to a PMD, but AFAIK adding PCI IDs is not considered an ABI 
>> breakage,
>> is it? One example is 28c9a7d7b48e ("net/mlx5: add ConnectX-6 Lx device 
>> ID")
>> added 2020-07-08, i.e. clearly outside of ABI change window.
>
> You're right.
>  
>> "xxx_pmd_info" changes are due to JSON formatting (new is more 
>> canonical),
>> which can be worked around easily, if the above is wrong.
>
> If the new format is better, please keep it.
> What we need is an exception for the pmdinfo symbols
> in the file devtools/libabigail.abignore.
> You can probably use a regex for these symbols.  

 This would allow real breakages to pass ABI check, abidiff doesn't analyze
 variable content and it's not easy to compare. Maybe later a script can be
 added that checks lines with RTE_DEVICE_IN in patches. There are at most 
 32 of
 5494 relevant commits between 19.11 and 20.11, though.

 To verify there are no meaningful changes I ensured empty diff between
 results of the following command for "main" and the branch:

find build/drivers -name '*.so' -exec usertools/dpdk-pmdinfo.py  
>>>
>>> For now we cannot do such check as part of the ABI checker.
>>> And we cannot merge this patch if the ABI check fails.
>>> I think the only solution is to allow any change in the pmdinfo variables.
>>>   
>>
>> So my 2c on this is that this is an acceptable work-around for the v21 (DPDK 
>> v20.11) ABI.
>> However we are going to end up carrying this rule in libabigail.ignore 
>> indefinitely.
>>
>> Would it make sense to just fix the size of _pmd_info to some reasonably 
>> large value - 
>> say 128 bytes, to allow us to drop the rule in the DPDK 21.11 v22 release?
> 
> I don't think so. This is a JSON *string to be parsed;* considering its size
> as part of application *binary* interface is wrong in the first place.

Right - then is belongs in INTERNAL, I would say. 

> As for
> content, checking that no PCI IDs are removed is out of scope for libabigail
> anyway. 

Lets be clear PCI IDs - are _nothing_ to do with ABI.

> Technically we could fix _pmd_info size, but this still allows
> breaking changes to pass the check with no benefit.

ABI changes or other, please explain?

Re: [dpdk-dev] [PATCH v10 2/3] build: use Python pmdinfogen

[Thomas Monjalon]

24/01/2021 21:51, Dmitry Kozlyuk:
> --- a/devtools/libabigail.abignore
> +++ b/devtools/libabigail.abignore
> +; Ignore all changes in generated PMD information strings.
> +[suppress_variable]
> +name_regex = _pmd_info$

I will fix this rule as "name_regexp" so it works.

Re: [dpdk-dev] [PATCH v1 3/3] ring: rename and refactor ring library

[Ananyev, Konstantin]

Hi Feifei,

> > > For legacy modes, rename ring_generic/c11 to ring_generic/c11_pvt.
> > > Furthermore, add new file ring_elem_pvt.h which includes
> > > ring_do_eq/deq and ring element copy/delete APIs.
> > >
> > > For other modes, rename xx_c11_mem to xx_elem_pvt. Move all private
> > > APIs into these new header files.
> > >
> > > Suggested-by: Honnappa Nagarahalli 
> > > Signed-off-by: Feifei Wang 
> > > Reviewed-by: Honnappa Nagarahalli 
> > > Reviewed-by: Ruifeng Wang 
> > > ---
> > >  lib/librte_ring/meson.build   |  15 +-
> > >  .../{rte_ring_c11_mem.h => ring_c11_pvt.h}|   9 +-
> > >  lib/librte_ring/ring_elem_pvt.h   | 385 ++
> > >  ...{rte_ring_generic.h => ring_generic_pvt.h} |   6 +-
> > >  ...ring_hts_c11_mem.h => ring_hts_elem_pvt.h} |  88 +++-
> > > ...ng_peek_c11_mem.h => ring_peek_elem_pvt.h} |  75 +++-
> > > ...ring_rts_c11_mem.h => ring_rts_elem_pvt.h} |  88 +++-
> > >  lib/librte_ring/rte_ring_elem.h   | 374 +
> > >  lib/librte_ring/rte_ring_hts.h|  84 +---
> > >  lib/librte_ring/rte_ring_peek.h   |  71 +---
> > >  lib/librte_ring/rte_ring_peek_zc.h|   2 +-
> > >  lib/librte_ring/rte_ring_rts.h|  84 +---
> > >  12 files changed, 646 insertions(+), 635 deletions(-)  rename
> > > lib/librte_ring/{rte_ring_c11_mem.h => ring_c11_pvt.h} (96%)  create
> > > mode 100644 lib/librte_ring/ring_elem_pvt.h  rename
> > > lib/librte_ring/{rte_ring_generic.h => ring_generic_pvt.h} (98%)
> > > rename lib/librte_ring/{rte_ring_hts_c11_mem.h => ring_hts_elem_pvt.h}
> > > (60%)  rename lib/librte_ring/{rte_ring_peek_c11_mem.h =>
> > > ring_peek_elem_pvt.h} (62%)  rename
> > > lib/librte_ring/{rte_ring_rts_c11_mem.h => ring_rts_elem_pvt.h} (62%)
> > >
> >
> > Sorry, but I don't understand the purpose of that patch.
> > As I remember by DPDK naming convention all installable headers should
> > have 'rte_' prefix. Same for public defines (RTE_).
> > Why to abandon it here?
> 
> This change refers to you and Honnappa's discussion about ring:
> https://mails.dpdk.org/archives/dev/2020-May/166803.html
> 
> And it is to separate the external APIs that can be called by users,
> from internal APIs that cannot be called by users.
> The internal functions are included in the files with xx_pvt.h.

Even these xx_pvt.h files contain internal API and not supposed to be included 
directly,
they have to be installed into public dir ( 
> Best Regards
> Feifei
> >
> > > diff --git a/lib/librte_ring/meson.build b/lib/librte_ring/meson.build
> > > index 36fdcb6a5..98eac5810 100644
> > > --- a/lib/librte_ring/meson.build
> > > +++ b/lib/librte_ring/meson.build
> > > @@ -2,15 +2,16 @@
> > >  # Copyright(c) 2017 Intel Corporation
> > >
> > >  sources = files('rte_ring.c')
> > > -headers = files('rte_ring.h',
> > > +headers = files('ring_c11_pvt.h',
> > > + 'ring_elem_pvt.h',
> > > + 'ring_generic_pvt.h',
> > > + 'ring_hts_elem_pvt.h',
> > > + 'ring_peek_elem_pvt.h',
> > > + 'ring_rts_elem_pvt.h',
> > > + 'rte_ring.h',
> > >   'rte_ring_core.h',
> > >   'rte_ring_elem.h',
> > > - 'rte_ring_c11_mem.h',
> > > - 'rte_ring_generic.h',
> > >   'rte_ring_hts.h',
> > > - 'rte_ring_hts_c11_mem.h',
> > >   'rte_ring_peek.h',
> > > - 'rte_ring_peek_c11_mem.h',
> > >   'rte_ring_peek_zc.h',
> > > - 'rte_ring_rts.h',
> > > - 'rte_ring_rts_c11_mem.h')
> > > + 'rte_ring_rts.h')
> > > diff --git a/lib/librte_ring/rte_ring_c11_mem.h
> > > b/lib/librte_ring/ring_c11_pvt.h similarity index 96% rename from
> > > lib/librte_ring/rte_ring_c11_mem.h
> > > rename to lib/librte_ring/ring_c11_pvt.h index 7f5eba262..9f2f5318f
> > > 100644
> > > --- a/lib/librte_ring/rte_ring_c11_mem.h
> > > +++ b/lib/librte_ring/ring_c11_pvt.h
> > > @@ -7,8 +7,8 @@
> > >   * Used as BSD-3 Licensed with permission from Kip Macy.
> > >   */
> > >
> > > -#ifndef _RTE_RING_C11_MEM_H_
> > > -#define _RTE_RING_C11_MEM_H_
> > > +#ifndef _RING_C11_PVT_H_
> > > +#define _RING_C11_PVT_H_
> > >
> > >  static __rte_always_inline void
> > >  __rte_ring_update_tail(struct rte_ring_headtail *ht, uint32_t
> > > old_val, @@ -69,9 +69,6 @@ __rte_ring_move_prod_head(struct rte_ring
> > *r, unsigned int is_sp,
> > >   /* Ensure the head is read before tail */
> > >   __atomic_thread_fence(__ATOMIC_ACQUIRE);
> > >
> > > - /* load-acquire synchronize with store-release of ht->tail
> > > -  * in update_tail.
> > > -  */
> > >   cons_tail = __atomic_load_n(&r->cons.tail,
> > >   __ATOMIC_ACQUIRE);
> > >
> > > @@ -178,4 +175,4 @@ __rte_ring_move_cons_head(struct rte_ring *r, int
> > is_sc,
> > >   return n;
> > >  }
> > >
> > > -#endif /* _RTE_RING_C11_MEM_H_ */
> > > +#endif /* _RING_C11_PVT_H_ */
> > > diff --git a/lib/librte_ring/ring_elem_p

Re: [dpdk-dev] [PATCH v1 3/3] ring: rename and refactor ring library

[Ananyev, Konstantin]

> 
> Hi Feifei,
> 
> > > > For legacy modes, rename ring_generic/c11 to ring_generic/c11_pvt.
> > > > Furthermore, add new file ring_elem_pvt.h which includes
> > > > ring_do_eq/deq and ring element copy/delete APIs.
> > > >
> > > > For other modes, rename xx_c11_mem to xx_elem_pvt. Move all private
> > > > APIs into these new header files.
> > > >
> > > > Suggested-by: Honnappa Nagarahalli 
> > > > Signed-off-by: Feifei Wang 
> > > > Reviewed-by: Honnappa Nagarahalli 
> > > > Reviewed-by: Ruifeng Wang 
> > > > ---
> > > >  lib/librte_ring/meson.build   |  15 +-
> > > >  .../{rte_ring_c11_mem.h => ring_c11_pvt.h}|   9 +-
> > > >  lib/librte_ring/ring_elem_pvt.h   | 385 ++
> > > >  ...{rte_ring_generic.h => ring_generic_pvt.h} |   6 +-
> > > >  ...ring_hts_c11_mem.h => ring_hts_elem_pvt.h} |  88 +++-
> > > > ...ng_peek_c11_mem.h => ring_peek_elem_pvt.h} |  75 +++-
> > > > ...ring_rts_c11_mem.h => ring_rts_elem_pvt.h} |  88 +++-
> > > >  lib/librte_ring/rte_ring_elem.h   | 374 +
> > > >  lib/librte_ring/rte_ring_hts.h|  84 +---
> > > >  lib/librte_ring/rte_ring_peek.h   |  71 +---
> > > >  lib/librte_ring/rte_ring_peek_zc.h|   2 +-
> > > >  lib/librte_ring/rte_ring_rts.h|  84 +---
> > > >  12 files changed, 646 insertions(+), 635 deletions(-)  rename
> > > > lib/librte_ring/{rte_ring_c11_mem.h => ring_c11_pvt.h} (96%)  create
> > > > mode 100644 lib/librte_ring/ring_elem_pvt.h  rename
> > > > lib/librte_ring/{rte_ring_generic.h => ring_generic_pvt.h} (98%)
> > > > rename lib/librte_ring/{rte_ring_hts_c11_mem.h => ring_hts_elem_pvt.h}
> > > > (60%)  rename lib/librte_ring/{rte_ring_peek_c11_mem.h =>
> > > > ring_peek_elem_pvt.h} (62%)  rename
> > > > lib/librte_ring/{rte_ring_rts_c11_mem.h => ring_rts_elem_pvt.h} (62%)
> > > >
> > >
> > > Sorry, but I don't understand the purpose of that patch.
> > > As I remember by DPDK naming convention all installable headers should
> > > have 'rte_' prefix. Same for public defines (RTE_).
> > > Why to abandon it here?
> >
> > This change refers to you and Honnappa's discussion about ring:
> > https://mails.dpdk.org/archives/dev/2020-May/166803.html
> >
> > And it is to separate the external APIs that can be called by users,
> > from internal APIs that cannot be called by users.
> > The internal functions are included in the files with xx_pvt.h.
> 
> Even these xx_pvt.h files contain internal API and not supposed to be 
> included directly,
> they have to be installed into public dir ( So I think we need to keep DPDK public prefixes for them (and inside them):
> rte_xx_pvt.h (_RTE_).

As another thought - it is probably good to introduce new generic naming 
convention
for such sort of files (internal ones that still have to be exposed).
Might be '_rte_' prefix or so...
Though I suppose it would require broader discussion first.

> 
> Konstantin
> 
> >
> > Best Regards
> > Feifei
> > >
> > > > diff --git a/lib/librte_ring/meson.build b/lib/librte_ring/meson.build
> > > > index 36fdcb6a5..98eac5810 100644
> > > > --- a/lib/librte_ring/meson.build
> > > > +++ b/lib/librte_ring/meson.build
> > > > @@ -2,15 +2,16 @@
> > > >  # Copyright(c) 2017 Intel Corporation
> > > >
> > > >  sources = files('rte_ring.c')
> > > > -headers = files('rte_ring.h',
> > > > +headers = files('ring_c11_pvt.h',
> > > > +   'ring_elem_pvt.h',
> > > > +   'ring_generic_pvt.h',
> > > > +   'ring_hts_elem_pvt.h',
> > > > +   'ring_peek_elem_pvt.h',
> > > > +   'ring_rts_elem_pvt.h',
> > > > +   'rte_ring.h',
> > > > 'rte_ring_core.h',
> > > > 'rte_ring_elem.h',
> > > > -   'rte_ring_c11_mem.h',
> > > > -   'rte_ring_generic.h',
> > > > 'rte_ring_hts.h',
> > > > -   'rte_ring_hts_c11_mem.h',
> > > > 'rte_ring_peek.h',
> > > > -   'rte_ring_peek_c11_mem.h',
> > > > 'rte_ring_peek_zc.h',
> > > > -   'rte_ring_rts.h',
> > > > -   'rte_ring_rts_c11_mem.h')
> > > > +   'rte_ring_rts.h')
> > > > diff --git a/lib/librte_ring/rte_ring_c11_mem.h
> > > > b/lib/librte_ring/ring_c11_pvt.h similarity index 96% rename from
> > > > lib/librte_ring/rte_ring_c11_mem.h
> > > > rename to lib/librte_ring/ring_c11_pvt.h index 7f5eba262..9f2f5318f
> > > > 100644
> > > > --- a/lib/librte_ring/rte_ring_c11_mem.h
> > > > +++ b/lib/librte_ring/ring_c11_pvt.h
> > > > @@ -7,8 +7,8 @@
> > > >   * Used as BSD-3 Licensed with permission from Kip Macy.
> > > >   */
> > > >
> > > > -#ifndef _RTE_RING_C11_MEM_H_
> > > > -#define _RTE_RING_C11_MEM_H_
> > > > +#ifndef _RING_C11_PVT_H_
> > > > +#define _RING_C11_PVT_H_
> > > >
> > > >  static __rte_always_inline void
> > > >  __rte_ring_update_tail(struct rte_ring_headtail *ht, uint32_t
> > > > old_val, @@ -69,9 +69,6 @@ __rte_

Re: [dpdk-dev] [PATCH v8 2/3] build: use Python pmdinfogen

[David Marchand]

On Mon, Jan 25, 2021 at 11:01 AM Kinsella, Ray  wrote:
>
>
>
> On 25/01/2021 09:25, Kinsella, Ray wrote:
> >
> >
> > On 23/01/2021 11:38, Thomas Monjalon wrote:
> >> 22/01/2021 23:24, Dmitry Kozlyuk:
> >>> On Fri, 22 Jan 2021 21:57:15 +0100, Thomas Monjalon wrote:
>  22/01/2021 21:31, Dmitry Kozlyuk:
> > On Wed, 20 Jan 2021 11:24:21 +0100, Thomas Monjalon wrote:
> >> 20/01/2021 08:23, Dmitry Kozlyuk:
> >>> On Wed, 20 Jan 2021 01:05:59 +0100, Thomas Monjalon wrote:
>  This is now the right timeframe to introduce this change
>  with the new Python module dependency.
>  Unfortunately, the ABI check is returning an issue:
> 
>  'const char mlx5_common_pci_pmd_info[62]' was changed
>  to 'const char mlx5_common_pci_pmd_info[60]' at rte_common_mlx5.pmd.c
> >>>
> >>> Will investigate and fix ASAP.
> >
> > Now that I think of it: strings like this change every time new PCI IDs 
> > are
> > added to a PMD, but AFAIK adding PCI IDs is not considered an ABI 
> > breakage,
> > is it? One example is 28c9a7d7b48e ("net/mlx5: add ConnectX-6 Lx device 
> > ID")
> > added 2020-07-08, i.e. clearly outside of ABI change window.
> 
>  You're right.
> 
> > "xxx_pmd_info" changes are due to JSON formatting (new is more 
> > canonical),
> > which can be worked around easily, if the above is wrong.
> 
>  If the new format is better, please keep it.
>  What we need is an exception for the pmdinfo symbols
>  in the file devtools/libabigail.abignore.
>  You can probably use a regex for these symbols.
> >>>
> >>> This would allow real breakages to pass ABI check, abidiff doesn't analyze
> >>> variable content and it's not easy to compare. Maybe later a script can be
> >>> added that checks lines with RTE_DEVICE_IN in patches. There are at most 
> >>> 32 of
> >>> 5494 relevant commits between 19.11 and 20.11, though.
> >>>
> >>> To verify there are no meaningful changes I ensured empty diff between
> >>> results of the following command for "main" and the branch:
> >>>
> >>> find build/drivers -name '*.so' -exec usertools/dpdk-pmdinfo.py
> >>
> >> For now we cannot do such check as part of the ABI checker.
> >> And we cannot merge this patch if the ABI check fails.
> >> I think the only solution is to allow any change in the pmdinfo variables.
> >>
> >
> > So my 2c on this is that this is an acceptable work-around for the v21 
> > (DPDK v20.11) ABI.
> > However we are going to end up carrying this rule in libabigail.ignore 
> > indefinitely.
> >
> > Would it make sense to just fix the size of _pmd_info to some reasonably 
> > large value -
> > say 128 bytes, to allow us to drop the rule in the DPDK 21.11 v22 release?
> >
> > Ray K
>
>
> Another point is - shouldn't _pmd_info probably live in "INTERNAL" is anycase?

The symbol itself can be hidden from the ABeyes.
It is only a placeholder for the PMD_INFO_STRING= string used by
usertools/dpdk-pmdinfo.py and maybe some other parsing tool.

I guess a static symbol would be enough:

diff --git a/buildtools/pmdinfogen/pmdinfogen.c
b/buildtools/pmdinfogen/pmdinfogen.c
index a68d1ea999..14bf7d9f42 100644
--- a/buildtools/pmdinfogen/pmdinfogen.c
+++ b/buildtools/pmdinfogen/pmdinfogen.c
@@ -393,7 +393,7 @@ static void output_pmd_info_string(struct elf_info
*info, char *outfile)
drv = info->drivers;

while (drv) {
-   fprintf(ofd, "const char %s_pmd_info[] __attribute__((used)) = "
+   fprintf(ofd, "static const char %s_pmd_info[]
__attribute__((used)) = "
"\"PMD_INFO_STRING= {",
drv->name);
fprintf(ofd, "\\\"name\\\" : \\\"%s\\\", ", drv->name);


We will need an exception for the v21 ABI though.


-- 
David Marchand

Re: [dpdk-dev] [PATCH v8 2/3] build: use Python pmdinfogen

[Dmitry Kozlyuk]

On Mon, 25 Jan 2021 10:11:07 +, Kinsella, Ray wrote:
> On 25/01/2021 10:05, Dmitry Kozlyuk wrote:
> > On Mon, 25 Jan 2021 09:25:51 +, Kinsella, Ray wrote:  
> >> On 23/01/2021 11:38, Thomas Monjalon wrote:  
> >>> 22/01/2021 23:24, Dmitry Kozlyuk:
>  On Fri, 22 Jan 2021 21:57:15 +0100, Thomas Monjalon wrote:
> > 22/01/2021 21:31, Dmitry Kozlyuk:
> >> On Wed, 20 Jan 2021 11:24:21 +0100, Thomas Monjalon wrote:  
> >>> 20/01/2021 08:23, Dmitry Kozlyuk:  
>  On Wed, 20 Jan 2021 01:05:59 +0100, Thomas Monjalon wrote:
> > This is now the right timeframe to introduce this change
> > with the new Python module dependency.
> > Unfortunately, the ABI check is returning an issue:
> >
> > 'const char mlx5_common_pci_pmd_info[62]' was changed
> > to 'const char mlx5_common_pci_pmd_info[60]' at 
> > rte_common_mlx5.pmd.c
> 
>  Will investigate and fix ASAP.  
> >>
> >> Now that I think of it: strings like this change every time new PCI 
> >> IDs are
> >> added to a PMD, but AFAIK adding PCI IDs is not considered an ABI 
> >> breakage,
> >> is it? One example is 28c9a7d7b48e ("net/mlx5: add ConnectX-6 Lx 
> >> device ID")
> >> added 2020-07-08, i.e. clearly outside of ABI change window.  
> >
> > You're right.
> >
> >> "xxx_pmd_info" changes are due to JSON formatting (new is more 
> >> canonical),
> >> which can be worked around easily, if the above is wrong.  
> >
> > If the new format is better, please keep it.
> > What we need is an exception for the pmdinfo symbols
> > in the file devtools/libabigail.abignore.
> > You can probably use a regex for these symbols.
> 
>  This would allow real breakages to pass ABI check, abidiff doesn't 
>  analyze
>  variable content and it's not easy to compare. Maybe later a script can 
>  be
>  added that checks lines with RTE_DEVICE_IN in patches. There are at most 
>  32 of
>  5494 relevant commits between 19.11 and 20.11, though.
> 
>  To verify there are no meaningful changes I ensured empty diff between
>  results of the following command for "main" and the branch:
> 
>   find build/drivers -name '*.so' -exec usertools/dpdk-pmdinfo.py
> >>>
> >>> For now we cannot do such check as part of the ABI checker.
> >>> And we cannot merge this patch if the ABI check fails.
> >>> I think the only solution is to allow any change in the pmdinfo variables.
> >>> 
> >>
> >> So my 2c on this is that this is an acceptable work-around for the v21 
> >> (DPDK v20.11) ABI.
> >> However we are going to end up carrying this rule in libabigail.ignore 
> >> indefinitely.
> >>
> >> Would it make sense to just fix the size of _pmd_info to some reasonably 
> >> large value - 
> >> say 128 bytes, to allow us to drop the rule in the DPDK 21.11 v22 release? 
> >>  
> > 
> > I don't think so. This is a JSON *string to be parsed;* considering its size
> > as part of application *binary* interface is wrong in the first place.  
> 
> Right - then is belongs in INTERNAL, I would say. 
>
> > As for
> > content, checking that no PCI IDs are removed is out of scope for libabigail
> > anyway.   
> 
> Lets be clear PCI IDs - are _nothing_ to do with ABI.

Technically, yes, but they're referred to in abi_policy.rst, because DPDK
behavior depends on them. Same issue as with as return values: no formats
change, yet compatibility is broken.

> > Technically we could fix _pmd_info size, but this still allows
> > breaking changes to pass the check with no benefit.  
> 
> ABI changes or other, please explain?

Behavioral changes via PCI ID removal, see above.

Re: [dpdk-dev] [dpdk-stable] [PATCH v11 1/4] raw/ifpga: add fpga rsu function

[Xu, Rosen]

Hi,

> -Original Message-
> From: Ferruh Yigit 
> Sent: Monday, January 25, 2021 18:01
> To: Huang, Wei ; dev@dpdk.org; Xu, Rosen
> ; Zhang, Qi Z 
> Cc: sta...@dpdk.org; Zhang, Tianfei ; Ray Kinsella
> 
> Subject: Re: [dpdk-stable] [PATCH v11 1/4] raw/ifpga: add fpga rsu function
> 
> On 1/22/2021 2:18 AM, Huang, Wei wrote:
> 
> 
> 
> >>
> >> -Original Message-
> >> From: Ferruh Yigit 
> >> Sent: Friday, January 22, 2021 00:30
> >> To: Huang, Wei ; dev@dpdk.org; Xu, Rosen
> >> ; Zhang, Qi Z 
> >> Cc: sta...@dpdk.org; Zhang, Tianfei ; Ray
> >> Kinsella 
> >> Subject: Re: [dpdk-stable] [PATCH v11 1/4] raw/ifpga: add fpga rsu
> >> function
> >>
> >> On 1/21/2021 6:03 AM, Wei Huang wrote:
> >>> RSU (Remote System Update) depends on secure manager which may be
> >>> different on various implementations, so a new secure manager device
> >>> is implemented for adapting such difference.
> >>> There are three major functions added:
> >>> 1. ifpga_rawdev_update_flash() updates flash with specific image file.
> >>> 2. ifpga_rawdev_stop_flash_update() aborts flash update process.
> >>> 3. ifpga_rawdev_reload() reloads FPGA from updated flash.
> >>>
> >>> Signed-off-by: Wei Huang 
> >>> Acked-by: Tianfei Zhang 
> >>> Acked-by: Rosen Xu 
> >>
> >> <...>
> >>
> >>> @@ -76,4 +76,9 @@ int
> >>>ifpga_unregister_msix_irq(enum ifpga_irq_type type,
> >>>int vec_start, rte_intr_callback_fn handler, void *arg);
> >>>
> >>> +int ifpga_rawdev_update_flash(struct rte_rawdev *dev, const char
> >>> +*image, uint64_t *status); int
> >>> +ifpga_rawdev_stop_flash_update(struct rte_rawdev *dev, int force);
> >>> +int ifpga_rawdev_reload(struct rte_rawdev *dev, int type, int
> >>> +page);
> >>> +
> >>>#endif /* _IFPGA_RAWDEV_H_ */
> >>>
> >>
> >> Hi Wei,
> >>
> >> Please help me understand the rawdev, who should be calling the above
> newly added functions?
> >>
>  >
>  > Hi Ferruh,
>  >
>  > Cyborg is an OpenStack project that aims to provide a general purpose
> management framework for acceleration resources (i.e. various types of
> accelerators such as GPU, FPGA, NP, ODP, DPDK/SPDK and so on).
>  >
>  > To update the FPGA flash is one of requirements from Cyborg. Originally
> there are no such interfaces, so I added them.
>  >
>  > These interfaces use rte_rawdev to identify which FPGA to access, they will
> be called in opae_update_flash(),  > opae_cancel_flash_update() and
> opae_reboot_device() in ifpga_opae_api.c .
>  >
>  > These opae_xxx function use PCI address to identify FPGA instead of
> rte_rawdev, so the caller has no need to know the existence of rte_rawdev.
>  > In fact, Cyborg is Python application, these opae_xxx functions will be
> eventually wrapped in a Python module for Cyborg to call.
>  >
> 
> Thanks for clarification, I see what you are doing, but still I think these 
> APIs
> are not belong to a driver, they look like more application level.
> 
> @Wei, @Rosen, what do you think to keep only generic raw/ipfga APIs in the
> driver and move all 'ifpga_opae_api.c/h' to the sample application?

@Huang, Wei, could you classify which functions are only used in sample 
application?

> In that case raw/ifpga APIs still can get the port_id as parameter, and the
> 'opae' layer in the sample can do the conversion?

Re: [dpdk-dev] [PATCH v8 2/3] build: use Python pmdinfogen

[Kinsella, Ray]

On 25/01/2021 10:29, David Marchand wrote:
> On Mon, Jan 25, 2021 at 11:01 AM Kinsella, Ray  wrote:
>>
>>
>>
>> On 25/01/2021 09:25, Kinsella, Ray wrote:
>>>
>>>
>>> On 23/01/2021 11:38, Thomas Monjalon wrote:
 22/01/2021 23:24, Dmitry Kozlyuk:
> On Fri, 22 Jan 2021 21:57:15 +0100, Thomas Monjalon wrote:
>> 22/01/2021 21:31, Dmitry Kozlyuk:
>>> On Wed, 20 Jan 2021 11:24:21 +0100, Thomas Monjalon wrote:
 20/01/2021 08:23, Dmitry Kozlyuk:
> On Wed, 20 Jan 2021 01:05:59 +0100, Thomas Monjalon wrote:
>> This is now the right timeframe to introduce this change
>> with the new Python module dependency.
>> Unfortunately, the ABI check is returning an issue:
>>
>> 'const char mlx5_common_pci_pmd_info[62]' was changed
>> to 'const char mlx5_common_pci_pmd_info[60]' at rte_common_mlx5.pmd.c
>
> Will investigate and fix ASAP.
>>>
>>> Now that I think of it: strings like this change every time new PCI IDs 
>>> are
>>> added to a PMD, but AFAIK adding PCI IDs is not considered an ABI 
>>> breakage,
>>> is it? One example is 28c9a7d7b48e ("net/mlx5: add ConnectX-6 Lx device 
>>> ID")
>>> added 2020-07-08, i.e. clearly outside of ABI change window.
>>
>> You're right.
>>
>>> "xxx_pmd_info" changes are due to JSON formatting (new is more 
>>> canonical),
>>> which can be worked around easily, if the above is wrong.
>>
>> If the new format is better, please keep it.
>> What we need is an exception for the pmdinfo symbols
>> in the file devtools/libabigail.abignore.
>> You can probably use a regex for these symbols.
>
> This would allow real breakages to pass ABI check, abidiff doesn't analyze
> variable content and it's not easy to compare. Maybe later a script can be
> added that checks lines with RTE_DEVICE_IN in patches. There are at most 
> 32 of
> 5494 relevant commits between 19.11 and 20.11, though.
>
> To verify there are no meaningful changes I ensured empty diff between
> results of the following command for "main" and the branch:
>
> find build/drivers -name '*.so' -exec usertools/dpdk-pmdinfo.py

 For now we cannot do such check as part of the ABI checker.
 And we cannot merge this patch if the ABI check fails.
 I think the only solution is to allow any change in the pmdinfo variables.

>>>
>>> So my 2c on this is that this is an acceptable work-around for the v21 
>>> (DPDK v20.11) ABI.
>>> However we are going to end up carrying this rule in libabigail.ignore 
>>> indefinitely.
>>>
>>> Would it make sense to just fix the size of _pmd_info to some reasonably 
>>> large value -
>>> say 128 bytes, to allow us to drop the rule in the DPDK 21.11 v22 release?
>>>
>>> Ray K
>>
>>
>> Another point is - shouldn't _pmd_info probably live in "INTERNAL" is 
>> anycase?
> 
> The symbol itself can be hidden from the ABeyes.
> It is only a placeholder for the PMD_INFO_STRING= string used by
> usertools/dpdk-pmdinfo.py and maybe some other parsing tool.
> 
> I guess a static symbol would be enough:
> 
> diff --git a/buildtools/pmdinfogen/pmdinfogen.c
> b/buildtools/pmdinfogen/pmdinfogen.c
> index a68d1ea999..14bf7d9f42 100644
> --- a/buildtools/pmdinfogen/pmdinfogen.c
> +++ b/buildtools/pmdinfogen/pmdinfogen.c
> @@ -393,7 +393,7 @@ static void output_pmd_info_string(struct elf_info
> *info, char *outfile)
> drv = info->drivers;
> 
> while (drv) {
> -   fprintf(ofd, "const char %s_pmd_info[] __attribute__((used)) 
> = "
> +   fprintf(ofd, "static const char %s_pmd_info[]
> __attribute__((used)) = "
> "\"PMD_INFO_STRING= {",
> drv->name);
> fprintf(ofd, "\\\"name\\\" : \\\"%s\\\", ", drv->name);
> 
> 
> We will need an exception for the v21 ABI though.
> 

Good suggestion +1

Re: [dpdk-dev] [PATCH 1/1] ethdev: fix handling of close failure

[Burakov, Anatoly]

On 22-Jan-21 5:58 PM, Thomas Monjalon wrote:

If a failure happens when closing a port,
it was unnecessarily failing again in the function eth_err(),
because of a check against HW removal cause.
Indeed there is a big chance the port is released at this point.
Given the port is in the middle (or at the end) of a close process,
checking the error cause by accessing the port is a non-sense.
The error check is replaced by a simple return in the close function.

Bugzilla ID: 624
Fixes: 8a5a0aad5d3e ("ethdev: allow close function to return an error")
Cc: sta...@dpdk.org

Reported-by: Anatoly Burakov 
Signed-off-by: Thomas Monjalon 
---


Tested-by: Anatoly Burakov 

--
Thanks,
Anatoly

Re: [dpdk-dev] [PATCH v2 01/44] bus/vdev: add helper to get vdev from eth dev

[Maxime Coquelin]

On 1/20/21 1:56 AM, Thomas Monjalon wrote:
> 19/01/2021 22:24, Maxime Coquelin:
>> This patch adds an helper macro to get the rte_vdev_device
>> pointer from a rte_eth_dev pointer.
>>
>> This is similar to RTE_ETH_DEV_TO_PCI().
> [...]
>> --- a/drivers/bus/vdev/rte_bus_vdev.h
>> +++ b/drivers/bus/vdev/rte_bus_vdev.h
>> @@ -34,6 +34,8 @@ struct rte_vdev_device {
>> +#define RTE_ETH_DEV_TO_VDEV(eth_dev)
>> RTE_DEV_TO_VDEV((eth_dev)->device)
> 
> Why these macros (vdev and PCI) are not in ethdev?
> 
> 

I think because that would put a dependency on vdev & pci bus drivers in
ethdev library.

Indeed, RTE_DEV_TO_VDEV needs to know about rte_vdev_device layout:

#define RTE_DEV_TO_VDEV(ptr) \
container_of(ptr, struct rte_vdev_device, device)

Regards,
Maxime

Re: [dpdk-dev] [PATCH v8 2/3] build: use Python pmdinfogen

[Thomas Monjalon]

25/01/2021 11:46, Kinsella, Ray:
> On 25/01/2021 10:29, David Marchand wrote:
> > The symbol itself can be hidden from the ABeyes.
> > It is only a placeholder for the PMD_INFO_STRING= string used by
> > usertools/dpdk-pmdinfo.py and maybe some other parsing tool.
> > 
> > I guess a static symbol would be enough:
> > 
> > diff --git a/buildtools/pmdinfogen/pmdinfogen.c
> > b/buildtools/pmdinfogen/pmdinfogen.c
> > index a68d1ea999..14bf7d9f42 100644
> > --- a/buildtools/pmdinfogen/pmdinfogen.c
> > +++ b/buildtools/pmdinfogen/pmdinfogen.c
> > @@ -393,7 +393,7 @@ static void output_pmd_info_string(struct elf_info
> > *info, char *outfile)
> > drv = info->drivers;
> > 
> > while (drv) {
> > -   fprintf(ofd, "const char %s_pmd_info[] 
> > __attribute__((used)) = "
> > +   fprintf(ofd, "static const char %s_pmd_info[]
> > __attribute__((used)) = "
> > "\"PMD_INFO_STRING= {",
> > drv->name);
> > fprintf(ofd, "\\\"name\\\" : \\\"%s\\\", ", drv->name);
> > 
> > 
> > We will need an exception for the v21 ABI though.
> > 
> 
> Good suggestion +1

Yes +1 for adding static on *_pmd_info

Re: [dpdk-dev] [PATCH v2 01/44] bus/vdev: add helper to get vdev from eth dev

[Thomas Monjalon]

25/01/2021 11:53, Maxime Coquelin:
> 
> On 1/20/21 1:56 AM, Thomas Monjalon wrote:
> > 19/01/2021 22:24, Maxime Coquelin:
> >> This patch adds an helper macro to get the rte_vdev_device
> >> pointer from a rte_eth_dev pointer.
> >>
> >> This is similar to RTE_ETH_DEV_TO_PCI().
> > [...]
> >> --- a/drivers/bus/vdev/rte_bus_vdev.h
> >> +++ b/drivers/bus/vdev/rte_bus_vdev.h
> >> @@ -34,6 +34,8 @@ struct rte_vdev_device {
> >> +#define RTE_ETH_DEV_TO_VDEV(eth_dev)  
> >> RTE_DEV_TO_VDEV((eth_dev)->device)
> > 
> > Why these macros (vdev and PCI) are not in ethdev?
> > 
> > 
> 
> I think because that would put a dependency on vdev & pci bus drivers in
> ethdev library.
> 
> Indeed, RTE_DEV_TO_VDEV needs to know about rte_vdev_device layout:
> 
> #define RTE_DEV_TO_VDEV(ptr) \
>   container_of(ptr, struct rte_vdev_device, device)

OK

Re: [dpdk-dev] [PATCH v4 2/5] examples/multi_process: cleanup bus objects while terminating app

[oulijun]

在 2020/10/18 17:25, David Marchand 写道:

On Thu, Oct 8, 2020 at 5:31 PM  wrote:


From: Rohit Raj 

Certain bus objects may need to be closed and re-acquired
while terminating and rerunning the client application.
Hence a signal handler is required to catch the termination
of the App and hence closing the bus objects.

This patch adds the missing signal handler in the client
app and closes the Bus objects in both client and server
applications when the signal Handler is called.

Signed-off-by: Rohit Raj 
---
  .../multi_process/client_server_mp/mp_client/client.c | 11 +++
  .../multi_process/client_server_mp/mp_server/main.c   |  4 +++-
  2 files changed, 14 insertions(+), 1 deletion(-)

diff --git a/examples/multi_process/client_server_mp/mp_client/client.c 
b/examples/multi_process/client_server_mp/mp_client/client.c
index 361d90b54..c37516b4c 100644
--- a/examples/multi_process/client_server_mp/mp_client/client.c
+++ b/examples/multi_process/client_server_mp/mp_client/client.c
@@ -11,6 +11,7 @@
  #include 
  #include 
  #include 
+#include 

  #include 
  #include 
@@ -196,6 +197,14 @@ handle_packet(struct rte_mbuf *buf)

  }

+static void
+signal_handler(int signal)
+{
+   if (signal == SIGINT)
+   rte_eal_cleanup();
+   exit(0);
+}


Calling rte_eal_cleanup from a signal handler is a bad idea.
In most cases, you are racing with other threads still using DPDK resources.
https://git.dpdk.org/dpdk/commit?id=2c434431f4
https://git.dpdk.org/dpdk/commit?id=613ce6691c

This might not be a problem in this multi_process example, but let's
keep a consistent way across all examples.
Thanks.

Hi，
  I want to know why you don't think he's a problem. recently, we use 
the patch

https://patchwork.dpdk.org/patch/86988/
startup with multiprocess. Use the tester to send traffic and kill the 
slave process repeatedly.
The coredump exception occurs. According to my analysis, the cause is 
that resources are not released after the slave process is killed.


Thanks
Lijun Ou

[dpdk-dev] [PATCH v1] net/iavf: adjust the VLAN failure handling

[Haiyue Wang]

Change the fatal returning to print the error message only, so that the
VF device can continue to be configured.

Fixes: 1c301e8c3cff ("net/iavf: support new VLAN capabilities")

Signed-off-by: Haiyue Wang 
---
 drivers/net/iavf/iavf_ethdev.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/drivers/net/iavf/iavf_ethdev.c b/drivers/net/iavf/iavf_ethdev.c
index 8e741031ec..af655084d2 100644
--- a/drivers/net/iavf/iavf_ethdev.c
+++ b/drivers/net/iavf/iavf_ethdev.c
@@ -429,10 +429,8 @@ iavf_dev_configure(struct rte_eth_dev *dev)
}
 
ret = iavf_dev_init_vlan(dev);
-   if (ret) {
+   if (ret)
PMD_DRV_LOG(ERR, "configure VLAN failed: %d", ret);
-   return -1;
-   }
 
if (vf->vf_res->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_RSS_PF) {
if (iavf_init_rss(ad) != 0) {
-- 
2.30.0

Re: [dpdk-dev] [PATCH v4 02/22] app/testpmd: fix max rx packet length for VLAN packets

[Ferruh Yigit]

On 1/21/2021 3:27 PM, Lance Richardson wrote:

On Mon, Jan 18, 2021 at 2:08 AM Steve Yang  wrote:


When the max rx packet length is smaller than the sum of mtu size and
ether overhead size, it should be enlarged, otherwise the VLAN packets
will be dropped.

Removed the rx_offloads assignment for jumbo frame during command line
parsing, and set the correct jumbo frame flag if MTU size is larger than
the default value 'RTE_ETHER_MTU' within 'init_config()'.

Fixes: 384161e00627 ("app/testpmd: adjust on the fly VLAN configuration")
Fixes: 35b2d13fd6fd ("net: add rte prefix to ether defines")
Fixes: ce17eddefc20 ("ethdev: introduce Rx queue offloads API")
Fixes: 150c9ac2df13 ("app/testpmd: update Rx offload after setting MTU")

Cc: Wenzhuo Lu 
Cc: Beilei Xing 
Cc: Bernard Iremonger 

Signed-off-by: Steve Yang 
---
  app/test-pmd/cmdline.c|  6 --
  app/test-pmd/config.c |  2 +-
  app/test-pmd/parameters.c |  7 ++-
  app/test-pmd/testpmd.c| 18 ++
  4 files changed, 21 insertions(+), 12 deletions(-)

diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
index 2ccbaa039e..65042fcff5 100644
--- a/app/test-pmd/cmdline.c
+++ b/app/test-pmd/cmdline.c
@@ -1886,7 +1886,6 @@ cmd_config_max_pkt_len_parsed(void *parsed_result,

 RTE_ETH_FOREACH_DEV(pid) {
 struct rte_port *port = &ports[pid];
-   uint64_t rx_offloads = port->dev_conf.rxmode.offloads;

 if (!strcmp(res->name, "max-pkt-len")) {
 if (res->value < RTE_ETHER_MIN_LEN) {
@@ -1898,11 +1897,6 @@ cmd_config_max_pkt_len_parsed(void *parsed_result,
 return;

 port->dev_conf.rxmode.max_rx_pkt_len = res->value;
-   if (res->value > RTE_ETHER_MAX_LEN)
-   rx_offloads |= DEV_RX_OFFLOAD_JUMBO_FRAME;
-   else
-   rx_offloads &= ~DEV_RX_OFFLOAD_JUMBO_FRAME;
-   port->dev_conf.rxmode.offloads = rx_offloads;
 } else {
 printf("Unknown parameter\n");
 return;
diff --git a/app/test-pmd/config.c b/app/test-pmd/config.c
index 3f6c8642b1..1195f054f3 100644
--- a/app/test-pmd/config.c
+++ b/app/test-pmd/config.c
@@ -1434,7 +1434,7 @@ port_mtu_set(portid_t port_id, uint16_t mtu)
  * device supports jumbo frame.
  */
 eth_overhead = dev_info.max_rx_pktlen - dev_info.max_mtu;
-   if (mtu > RTE_ETHER_MAX_LEN - eth_overhead) {
+   if (mtu > RTE_ETHER_MTU) {
 rte_port->dev_conf.rxmode.offloads |=
 DEV_RX_OFFLOAD_JUMBO_FRAME;
 rte_port->dev_conf.rxmode.max_rx_pkt_len =
diff --git a/app/test-pmd/parameters.c b/app/test-pmd/parameters.c
index 414a0068fb..df5eb10d84 100644
--- a/app/test-pmd/parameters.c
+++ b/app/test-pmd/parameters.c
@@ -834,12 +834,9 @@ launch_args_parse(int argc, char** argv)
 }
 if (!strcmp(lgopts[opt_idx].name, "max-pkt-len")) {
 n = atoi(optarg);
-   if (n >= RTE_ETHER_MIN_LEN) {
+   if (n >= RTE_ETHER_MIN_LEN)
 rx_mode.max_rx_pkt_len = (uint32_t) n;
-   if (n > RTE_ETHER_MAX_LEN)
-   rx_offloads |=
-   
DEV_RX_OFFLOAD_JUMBO_FRAME;
-   } else
+   else
 rte_exit(EXIT_FAILURE,
  "Invalid max-pkt-len=%d - should be 
> %d\n",
  n, RTE_ETHER_MIN_LEN);
diff --git a/app/test-pmd/testpmd.c b/app/test-pmd/testpmd.c
index 2b60f6c5d3..c256e719ae 100644
--- a/app/test-pmd/testpmd.c
+++ b/app/test-pmd/testpmd.c
@@ -1410,6 +1410,7 @@ init_config(void)
 struct rte_gro_param gro_param;
 uint32_t gso_types;
 uint16_t data_size;
+   uint16_t eth_overhead;
 bool warning = 0;
 int k;
 int ret;
@@ -1446,6 +1447,23 @@ init_config(void)
 rte_exit(EXIT_FAILURE,
  "rte_eth_dev_info_get() failed\n");

+   /* Update the max_rx_pkt_len to have MTU as RTE_ETHER_MTU */
+   if (port->dev_info.max_mtu != UINT16_MAX &&
+   port->dev_info.max_rx_pktlen > port->dev_info.max_mtu)
+   eth_overhead = port->dev_info.max_rx_pktlen -
+   port->dev_info.max_mtu;
+   else
+   eth_overhead =
+   RTE_ETHER_HDR_LEN + RTE_ETHER_CRC_LEN;
+
+   if (

Re: [dpdk-dev] [PATCH v3 01/22] ethdev: fix MTU size exceeds max rx packet length

[Ferruh Yigit]

On 1/19/2021 8:46 AM, oulijun wrote:



在 2021/1/18 18:42, Ferruh Yigit 写道:

On 1/15/2021 10:44 AM, oulijun wrote:

Hi Steve
This is a very good job! But I have some question and suggestions.
Please check it.

在 2021/1/14 17:45, Steve Yang 写道:

Ethdev is using default Ethernet overhead to decide if provided
'max_rx_pkt_len' value is bigger than max (non jumbo) MTU value,
and limits it to MAX if it is.

Since the application/driver used Ethernet overhead is different than
the ethdev one, check result is wrong.

If the driver is using Ethernet overhead bigger than the default one,
the provided 'max_rx_pkt_len' is trimmed down, and in the driver when
correct Ethernet overhead is used to convert back, the resulting MTU is
less than the intended one, causing some packets to be dropped.

Like,
app -> max_rx_pkt_len = 1500/*mtu*/ + 22/*overhead*/ = 1522
ethdev  -> 1522 > 1518/*MAX*/; max_rx_pkt_len = 1518
driver  -> MTU = 1518 - 22 = 1496
Packets with size 1497-1500 are dropped although intention is to be able
to send/receive them.

The fix is to make ethdev use the correct Ethernet overhead for port,
instead of default one.

Fixes: 59d0ecdbf0e1 ("ethdev: MTU accessors")

Signed-off-by: Steve Yang 
---
  lib/librte_ethdev/rte_ethdev.c | 27 ---
  1 file changed, 24 insertions(+), 3 deletions(-)

diff --git a/lib/librte_ethdev/rte_ethdev.c b/lib/librte_ethdev/rte_ethdev.c
index 17ddacc78d..19ca4c4512 100644
--- a/lib/librte_ethdev/rte_ethdev.c
+++ b/lib/librte_ethdev/rte_ethdev.c
@@ -1292,8 +1292,10 @@ rte_eth_dev_configure(uint16_t port_id, uint16_t 
nb_rx_q, uint16_t nb_tx_q,

  struct rte_eth_dev *dev;
  struct rte_eth_dev_info dev_info;
  struct rte_eth_conf orig_conf;
+    uint16_t overhead_len;
  int diag;
  int ret;
+    uint16_t old_mtu;
  RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -ENODEV);
@@ -1319,10 +1321,20 @@ rte_eth_dev_configure(uint16_t port_id, uint16_t 
nb_rx_q, uint16_t nb_tx_q,

  memcpy(&dev->data->dev_conf, dev_conf,
 sizeof(dev->data->dev_conf));
+    /* Backup mtu for rollback */
+    old_mtu = dev->data->mtu;
+
  ret = rte_eth_dev_info_get(port_id, &dev_info);
  if (ret != 0)
  goto rollback;
+    /* Get the real Ethernet overhead length */
+    if (dev_info.max_mtu != UINT16_MAX &&
+    dev_info.max_rx_pktlen > dev_info.max_mtu)
+    overhead_len = dev_info.max_rx_pktlen - dev_info.max_mtu;
+    else
+    overhead_len = RTE_ETHER_HDR_LEN + RTE_ETHER_CRC_LEN;
+
The Ethernet frame header length supported by each NIC may be different, 
which cause the difference of allowed packets size and drop you say.
The diference of Ethernet frame header length have an impact for the maximum 
Ethernet frame length, which is a boundary value

of enabling jumbo frame through the ' max_rx_pkt_len '.
However, we need to do the same thing you do above to get the overhead_len 
every time, which will
cause a lot of duplicate code in the framework and app. For examples, parsing 
and processing for '--max-pkt-len=xxx' parameter,
  and " cmd_config_max_pkt_len_parsed " in testpmd, and here modifying here 
dev_configure API.

It's a little redundant and troublesome.

Maybe, it is necessary for driver to directly report the supported maximum 
Ethernet frame length by rte_dev_info_get API.

As following:
struct rte_eth_dev_info {
 
 /**
  * The maximum Ethernet frame length supported by each
  * driver varies with the Ethernet header length.
  */
 uint16_t eth_max_len;
 
}

int
rte_eth_dev_info_get(uint16_t port_id, struct rte_eth_dev_info *dev_info)
{
 xxx
 dev_info->min_mtu = RTE_ETHER_MIN_MTU;
 dev_info->max_mtu = UINT16_MAX;
 dev_info->eth_max_len = RTE_ETHER_MAX_LEN;
 xxx
}

And then:
xxx_devv_infos_get(struct rte_eth_dev *eth_dev, struct rte_eth_dev_info *info)
{
 xxx
 info->eth_max_len = xxx_ETHER_MAX _LEN;
 xxx
}
What do you think?


Hi Lijun,

The 'max_mtu' has been added in the past to solve exact same problem, perhaps 
it would be better to add something like 'eth_max_len' at that time.


But as Steve mentioned we are planning to switch to more MTU based 
configuration, which should remove the need of the field you mentioned.



  /* If number of queues specified by application for both Rx and Tx is
   * zero, use driver preferred values. This cannot be done individually
   * as it is valid for either Tx or Rx (but not both) to be zero.
@@ -1410,11 +1422,18 @@ rte_eth_dev_configure(uint16_t port_id, uint16_t 
nb_rx_q, uint16_t nb_tx_q,

  goto rollback;
  }
  } else {
-    if (dev_conf->rxmode.max_rx_pkt_len < RTE_ETHER_MIN_LEN ||
-    dev_conf->rxmode.max_rx_pkt_len > RTE_ETHER_MAX_LEN)
+    uint16_t pktlen = dev_conf->rxmode.max_rx_pkt_len;
+    if (pktlen < RTE_ETHER_MIN_MTU + overhead_len ||
+    pktlen > RTE_ETHER_MTU + overhead_len)
  /* Use default value */

Re: [dpdk-dev] [PATCH v2 19/44] net/virtio: move features definition to generic header

[Maxime Coquelin]

On 1/21/21 7:47 AM, Xia, Chenbo wrote:
> Hi Maxime,
> 
>> -Original Message-
>> From: Maxime Coquelin 
>> Sent: Wednesday, January 20, 2021 5:25 AM
>> To: dev@dpdk.org; Xia, Chenbo ; olivier.m...@6wind.com;
>> amore...@redhat.com; david.march...@redhat.com
>> Cc: Maxime Coquelin 
>> Subject: [PATCH v2 19/44] net/virtio: move features definition to generic
>> header
>>
>> This patch moves all the Virtio definition to the generic
>> header. It also renames some helpers to no more reference
>> PCI.
>>
>> Signed-off-by: Maxime Coquelin 
>> ---
>>  drivers/net/virtio/meson.build|   3 +-
>>  drivers/net/virtio/virtio.c   |  22 
>>  drivers/net/virtio/virtio.h   |  94 +++
>>  drivers/net/virtio/virtio_ethdev.c| 114 +-
>>  drivers/net/virtio/virtio_pci.c   |  21 +---
>>  drivers/net/virtio/virtio_pci.h   |  90 --
>>  drivers/net/virtio/virtio_ring.h  |   2 +-
>>  drivers/net/virtio/virtio_rxtx.c  |  38 +++---
>>  drivers/net/virtio/virtio_rxtx_packed.h   |   6 +-
>>  .../net/virtio/virtio_user/vhost_kernel_tap.c |   2 +-
>>  drivers/net/virtio/virtio_user_ethdev.c   |   6 +-
>>  drivers/net/virtio/virtqueue.c|   4 +-
>>  drivers/net/virtio/virtqueue.h|   8 +-
>>  13 files changed, 211 insertions(+), 199 deletions(-)
>>  create mode 100644 drivers/net/virtio/virtio.c
>>
>> diff --git a/drivers/net/virtio/meson.build b/drivers/net/virtio/meson.build
>> index f2873d6180..d595cfdcab 100644
>> --- a/drivers/net/virtio/meson.build
>> +++ b/drivers/net/virtio/meson.build
> 
> 
> 
>>
>> -if (vtpci_with_feature(hw, VIRTIO_F_RING_PACKED)) {
>> +if (virtio_with_feature(hw, VIRTIO_F_RING_PACKED)) {
> 
> As discussed before, let's also replace here with virtio_with_packed_queue 😊

Indeed, I missed that one.
It is fixed now and will be in v3.

Thanks,
Maxime

> Thanks
> Chenbo

Re: [dpdk-dev] [dpdk-stable] [PATCH 1/1] ethdev: fix handling of close failure

[Thomas Monjalon]

25/01/2021 10:13, Andrew Rybchenko:
> On 1/22/21 8:58 PM, Thomas Monjalon wrote:
> > If a failure happens when closing a port,
> > it was unnecessarily failing again in the function eth_err(),
> > because of a check against HW removal cause.
> > Indeed there is a big chance the port is released at this point.
> > Given the port is in the middle (or at the end) of a close process,
> > checking the error cause by accessing the port is a non-sense.
> > The error check is replaced by a simple return in the close function.
> > 
> > Bugzilla ID: 624
> > Fixes: 8a5a0aad5d3e ("ethdev: allow close function to return an error")
> > Cc: sta...@dpdk.org
> > 
> > Reported-by: Anatoly Burakov 
> > Signed-off-by: Thomas Monjalon 
> 
> Acked-by: Andrew Rybchenko 

Applied

Re: [dpdk-dev] [PATCH v3 1/3] ethdev: fix MTU doesn't update when jumbo frame disabled

[Ferruh Yigit]

On 1/25/2021 9:49 AM, Yang, SteveX wrote:

Hi Huisong,

Thanks for your review.

The validity of the pair  should be 
checked from application layer (e.g.: testpmd),


and the RTE layer should keep open enough to adapt the high-layer requirement.

I’m not sure if exists some applications/NICs that treat ‘packet size < 1500’ as 
JUMBO_FRAME. If so, that also can work as expect with current code.


@Yigit, Ferruh , please correct me if something 
understand wrong.




Hi Huisong,

Agree that there is a grey area in the API, the question is if 'JUMBO_FRAME' is 
set, can application set the 'max_rx_pkt_len' less than "RTE_ETHER_MTU + 
overhead_len". Lijun (cc'ed) has the same concern.


The API documentation, and checks in the 'rte_eth_dev_configure()' enables 
setting this for a long time, I am reluctant to add this limitation now.
Although agree that application should set 'JUMBO_FRAME' properly based on 
requested 'MTU' value.



BTW, there perhaps are some confused problems about jumbo frame and 
max_rx_pkt_len, and Ferruh has scheduled to re-factor this part at release 21.11.


If you’re interesting about it, please refer to following link: [RFC,v2] doc: 
announce max Rx packet len field deprecation - Patchwork (dpdk.org) 



Thanks & Regards,

Steve Yang.

*From:* Huisong Li 
*Sent:* Monday, January 25, 2021 3:12 PM
*To:* Yang, SteveX ; dev@dpdk.org
*Cc:* Lu, Wenzhuo ; Li, Xiaoyun ; 
Iremonger, Bernard ; tho...@monjalon.net; Yigit, 
Ferruh ; andrew.rybche...@oktetlabs.ru; Yang, Qiming 
; ouli...@huawei.com; huangda...@huawei.com
*Subject:* Re: [dpdk-dev] [PATCH v3 1/3] ethdev: fix MTU doesn't update when 
jumbo frame disabled


Hi Steve,

In the current modification, the MTU is updated based on 'max_rx_pkt_len' 
regardless of whether jumbo frame is enabled.


Now, MTU is correct when jumbo frmae is disabled. However, when jumbo frame is 
enabled, the MTU value may be inconsistent with


the definition of the enabled jumbo frame. Like:

1/ DEV_RX_OFFLOAD_JUMBO_FRAME is set;

2/ max_rx_pkt_len = 1200

3/ dev->data->mtu = 1200 - overhead_len(18) = 1182

In rte_eth_dev_configure API, the check for 'max_rx_pkt_len' is as follows:

if (dev_conf->rxmode.offloads & DEV_RX_OFFLOAD_JUMBO_FRAME) {  //jumbo frame 
enabled
         if (dev_conf->rxmode.max_rx_pkt_len > dev_info.max_rx_pktlen) {
             
             goto rollback;
         } else if (*dev_conf->rxmode.max_rx_pkt_len < RTE_ETHER_MIN_LEN*) {
             
             goto rollback;
         }
} else { //jumbo frame disabled

         if (pktlen < RTE_ETHER_MIN_MTU + overhead_len ||
     pktlen > RTE_ETHER_MTU + overhead_len)
     /* Use default value */
     dev->data->dev_conf.rxmode.max_rx_pkt_len =
     RTE_ETHER_MTU + overhead_len;

}

Since the applicatin sets DEV_RX_OFFLOAD_JUMBO_FRAME to enable jumbo frame, and 
the framework API needs to update


the MTU based on 'max_rx_pkt_len', but the framework API uses 
*RTE_ETHER_MIN_LEN(64)* to verify the boundary value of


'max_rx_pkt_len', instead of "RTE_ETHER_MTU + overhead_len".  As far as I know, 
if the applicatin sets DEV_RX_OFFLOAD_JUMBO_FRAME


and 'max_rx_pkt_len' is 1200, the framework API or driver should return a 
failure. As mentioned in this patch set, the jumbo frame


offload is set only when 'max_rx_pkt_len' requested is greater than 
"RTE_ETHER_MTU + eth_overhead" in testpmd.


I really don't understand it.  How do you understand this behavior?

Thanks.

在 2021/1/22 17:01, Steve Yang 写道:

The MTU value should be updated to 'max_rx_pkt_len - overhead'

no matter if the JUMBO FRAME offload enabled. If not update this MTU,

use will get the wrong MTU info via some command.

E.g.: 'show port info all' in testpmd tool.

Actually, the 'max_rx_pkt_len' has been used for other purposes in many

places now, even though the 'max_rx_pkt_len' is expected 'Only used if

JUMBO_FRAME enabled'.

For examples,

'max_rx_pkt_len' perhaps can be used as the 'rx_ctx.rxmax' in i40e.

Fixes: bf0f90d92d30 ("ethdev: fix max Rx packet length check")

Signed-off-by: Steve Yang  


---

  lib/librte_ethdev/rte_ethdev.c | 8 

  1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/lib/librte_ethdev/rte_ethdev.c b/lib/librte_ethdev/rte_ethdev.c

index daf5f24f7e..42857e3b67 100644

--- a/lib/librte_ethdev/rte_ethdev.c

+++ b/lib/librte_ethdev/rte_ethdev.c

@@ -1421,10 +1421,6 @@ rte_eth_dev_configure(uint16_t port_id, uint16_t 
nb_rx_q, uint16_t nb_tx_q,

     ret = -EINVAL;

     goto rollback;

     }

-

-   /* Scale the MTU size to adapt max_rx_pkt_len */

-   dev->data->mtu = dev->data->dev_conf.rxmode.max_rx_

Re: [dpdk-dev] [dpdk-stable] [PATCH 1/1] ethdev: fix handling of close failure

[Thomas Monjalon]

25/01/2021 13:37, Thomas Monjalon:
> 25/01/2021 10:13, Andrew Rybchenko:
> > On 1/22/21 8:58 PM, Thomas Monjalon wrote:
> > > If a failure happens when closing a port,
> > > it was unnecessarily failing again in the function eth_err(),
> > > because of a check against HW removal cause.
> > > Indeed there is a big chance the port is released at this point.
> > > Given the port is in the middle (or at the end) of a close process,
> > > checking the error cause by accessing the port is a non-sense.
> > > The error check is replaced by a simple return in the close function.
> > > 
> > > Bugzilla ID: 624
> > > Fixes: 8a5a0aad5d3e ("ethdev: allow close function to return an error")
> > > Cc: sta...@dpdk.org
> > > 
> > > Reported-by: Anatoly Burakov 
> > > Signed-off-by: Thomas Monjalon 
> > 
> > Acked-by: Andrew Rybchenko 
> 
> Applied

Sorry please ignore this wrong message, patch not applied.
(will be considered by Ferruh)

Re: [dpdk-dev] [PATCH v2 09/44] net/virtio: move MSIX detection to PCI ethdev

[Maxime Coquelin]

On 1/21/21 8:12 AM, Xia, Chenbo wrote:
> Hi Maxime,
> 
>> -Original Message-
>> From: Maxime Coquelin 
>> Sent: Wednesday, January 20, 2021 5:25 AM
>> To: dev@dpdk.org; Xia, Chenbo ; olivier.m...@6wind.com;
>> amore...@redhat.com; david.march...@redhat.com
>> Cc: Maxime Coquelin 
>> Subject: [PATCH v2 09/44] net/virtio: move MSIX detection to PCI ethdev
>>
>> This patch introduces a new callback to notify the bus
>> driver some interrupt related operation was done.
>>
>> This is used by Virtio PCI driver to check msix status.
>>
>> Signed-off-by: Maxime Coquelin 
>> ---
>>  drivers/net/virtio/virtio_ethdev.c |  12 +--
>>  drivers/net/virtio/virtio_pci.c| 120 ++---
>>  drivers/net/virtio/virtio_pci.h|   6 +-
>>  drivers/net/virtio/virtio_pci_ethdev.c |   2 +
>>  4 files changed, 82 insertions(+), 58 deletions(-)
>>
>> diff --git a/drivers/net/virtio/virtio_ethdev.c
>> b/drivers/net/virtio/virtio_ethdev.c
>> index a3e81f336d..13d5a76376 100644
>> --- a/drivers/net/virtio/virtio_ethdev.c
>> +++ b/drivers/net/virtio/virtio_ethdev.c
>> @@ -1287,8 +1287,8 @@ virtio_intr_unmask(struct rte_eth_dev *dev)
>>  if (rte_intr_ack(dev->intr_handle) < 0)
>>  return -1;
>>
>> -if (hw->bus_type == VIRTIO_BUS_PCI_LEGACY || hw->bus_type ==
>> VIRTIO_BUS_PCI_MODERN)
>> -hw->use_msix = vtpci_msix_detect(RTE_ETH_DEV_TO_PCI(dev));
>> +if (VTPCI_OPS(hw)->intr_event)
> 
> Emmm.. why is callback called intr_event? The callback detects/checks msix 
> and lsc.
> IMHO, maybe intr_detect? Or another similar verb you like 😊

I called it intr_event because it could be used for other purpose than
intr detection. But this is the only use for now so let's immplement
your suggestion and we can revisit later if needed.

>> +VTPCI_OPS(hw)->intr_event(hw);
>>
>>  return 0;
> 
> 
> 
>>
>> +msix_detect:
>> +hw->use_msix = vtpci_msix_detect(dev);
>> +
> 
> Just use use the callback introduced? Should also work later when intr_lsc is 
> added.

I didn't do it given that it was in the same file and that the callback
was to report an event happened. With renaming the callback as you
suggest, it makes sense to use it.

I will fix in v3.

Thanks,
Maxime

> Thanks,
> Chenbo
> 

Re: [dpdk-dev] [PATCH v4 1/2] ethdev: fix MTU doesn't update when jumbo frame disabled

[Ferruh Yigit]

On 1/25/2021 8:32 AM, Steve Yang wrote:

The MTU value should be updated to 'max_rx_pkt_len - overhead'
no matter if the JUMBO FRAME offload enabled. If not update this MTU,
use will get the wrong MTU info via some command.
E.g.: 'show port info all' in testpmd tool.

Actually, the 'max_rx_pkt_len' has been used for other purposes in many
places now, even though the 'max_rx_pkt_len' is expected 'Only used if
JUMBO_FRAME enabled'.

For examples,
'max_rx_pkt_len' perhaps can be used as the 'rx_ctx.rxmax' in i40e.

Fixes: bf0f90d92d30 ("ethdev: fix max Rx packet length check")

Signed-off-by: Steve Yang 
---
  lib/librte_ethdev/rte_ethdev.c | 8 
  1 file changed, 4 insertions(+), 4 deletions(-)

diff --git a/lib/librte_ethdev/rte_ethdev.c b/lib/librte_ethdev/rte_ethdev.c
index daf5f24f7e..42857e3b67 100644
--- a/lib/librte_ethdev/rte_ethdev.c
+++ b/lib/librte_ethdev/rte_ethdev.c
@@ -1421,10 +1421,6 @@ rte_eth_dev_configure(uint16_t port_id, uint16_t 
nb_rx_q, uint16_t nb_tx_q,
ret = -EINVAL;
goto rollback;
}
-
-   /* Scale the MTU size to adapt max_rx_pkt_len */
-   dev->data->mtu = dev->data->dev_conf.rxmode.max_rx_pkt_len -
-   overhead_len;
} else {
uint16_t pktlen = dev_conf->rxmode.max_rx_pkt_len;
if (pktlen < RTE_ETHER_MIN_MTU + overhead_len ||
@@ -1434,6 +1430,10 @@ rte_eth_dev_configure(uint16_t port_id, uint16_t 
nb_rx_q, uint16_t nb_tx_q,
RTE_ETHER_MTU + overhead_len;
}
  
+	/* Scale the MTU size to adapt max_rx_pkt_len */

+   dev->data->mtu = dev->data->dev_conf.rxmode.max_rx_pkt_len -
+   overhead_len;
+
/*
 * If LRO is enabled, check that the maximum aggregated packet
 * size is supported by the configured device.



This will cause 'max_rx_pkt_len' taken into account even 'JUMBO_FRAME' is not 
set, this is against the API documentation.


PMD can ignore the 'max_rx_pkt_len' when 'JUMBO_FRAME' is not set, so updating 
the 'max_rx_pkt_len' is relatively harmless, but updating 'MTU' can have affect.


Instead what do you think explicitly call 'rte_eth_dev_set_mtu()' in the testpmd 
when the MTU of the port needs to be changed because of "port config all 
max-pkt-len" command?

Re: [dpdk-dev] [PATCH] net/iavf: fix symmetric rule creating

[Zhang, Qi Z]

> -Original Message-
> From: Xing, Beilei 
> Sent: Friday, January 22, 2021 12:05 PM
> To: Ding, Xuan ; Zhang, Qi Z ;
> Wu, Jingjing 
> Cc: dev@dpdk.org; sta...@dpdk.org
> Subject: RE: [PATCH] net/iavf: fix symmetric rule creating
> 
> 
> 
> > -Original Message-
> > From: Ding, Xuan 
> > Sent: Friday, January 22, 2021 11:19 AM
> > To: Zhang, Qi Z ; Wu, Jingjing
> > ; Xing, Beilei 
> > Cc: dev@dpdk.org; Ding, Xuan ; sta...@dpdk.org
> > Subject: [PATCH] net/iavf: fix symmetric rule creating
> >
> > Only allow to create symmetric rule for L3/L4.
> >
> > Fixes: 91f27b2e39ab("net/iavf: refactor RSS")
> > Cc: sta...@dpdk.org
> >
> > Signed-off-by: Xuan Ding 
> Acked-by: Beilei Xing 

Applied to dpdk-next-net-intel.

Thanks
Qi

Re: [dpdk-dev] [PATCH] net/iavf: fix port VLAN cfg err for AVF with SVM

[Zhang, Qi Z]

> -Original Message-
> From: Wang, Haiyue 
> Sent: Friday, January 22, 2021 2:12 PM
> To: Guo, Junfeng ; Zhang, Qi Z
> ; Wu, Jingjing ; Xing, Beilei
> 
> Cc: dev@dpdk.org; Zhang, Yuying 
> Subject: RE: [PATCH] net/iavf: fix port VLAN cfg err for AVF with SVM
> 
> > -Original Message-
> > From: Guo, Junfeng 
> > Sent: Friday, January 22, 2021 21:36
> > To: Zhang, Qi Z ; Wu, Jingjing
> > ; Xing, Beilei 
> > Cc: dev@dpdk.org; Wang, Haiyue ; Zhang, Yuying
> > ; Guo, Junfeng 
> > Subject: [PATCH] net/iavf: fix port VLAN cfg err for AVF with SVM
> >
> > For AVF with single vlan mode (SVM), port vlan stripping config has
> > already been disabled by PF. In this scenario, the error of -ENOTSUP
> > can be ignored.
> >
> > Fixes: 1c301e8c3cff ("net/iavf: support new VLAN capabilities")
> > Signed-off-by: Junfeng Guo 
> > ---
> >  drivers/net/iavf/iavf_ethdev.c | 3 +++
> >  1 file changed, 3 insertions(+)
> 
> Good catch, I missed this scenario handling. ;-)
> 
> Acked-by: Haiyue Wang 

Applied to dpdk-next-net-intel.

Thanks
Qi

Re: [dpdk-dev] [PATCH v10 0/3] pmdinfogen: rewrite in Python

[Thomas Monjalon]

> Dmitry Kozlyuk (3):
>   pmdinfogen: add Python implementation
>   build: use Python pmdinfogen
>   pmdinfogen: remove C implementation

Applied, thanks.

Reminder: it is adding a NEW DEPENDENCY on pyelftools,
as highlighted already in CI and maintainers meetings:
http://mails.dpdk.org/archives/ci/2021-January/000939.html
http://mails.dpdk.org/archives/dev/2021-January/197814.html

If not already done, please UPDATE your environment! Thanks

[dpdk-dev] [PATCH v3 1/4] eal: add missing include to mcslock

[Bruce Richardson]

Include 'rte_branch_prediction.h' to get the likely/unlikely macro
definitions.

Fixes: 2173fb61 ("mcslock: add MCS queued lock implementation")
Cc: sta...@dpdk.org

Signed-off-by: Bruce Richardson 
---
 lib/librte_eal/include/generic/rte_mcslock.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/lib/librte_eal/include/generic/rte_mcslock.h 
b/lib/librte_eal/include/generic/rte_mcslock.h
index d370bef17a..9f323bd2a2 100644
--- a/lib/librte_eal/include/generic/rte_mcslock.h
+++ b/lib/librte_eal/include/generic/rte_mcslock.h
@@ -22,6 +22,7 @@
 #include 
 #include 
 #include 
+#include 
 
 /**
  * The rte_mcslock_t type.
-- 
2.27.0

[dpdk-dev] [PATCH v3 2/4] build: separate out headers for include checking

[Bruce Richardson]

For some libraries, there may be some header files which are not for direct
inclusion, but rather are to be included via other header files. To allow
later checking of these files for missing includes, we separate out the
indirect include files from the direct ones.

Signed-off-by: Bruce Richardson 
---
 doc/guides/contributing/coding_style.rst | 12 
 lib/librte_eal/include/meson.build   |  2 +-
 lib/librte_eal/x86/include/meson.build   | 14 +-
 lib/librte_ethdev/meson.build|  4 ++--
 lib/librte_hash/meson.build  |  4 ++--
 lib/librte_ipsec/meson.build |  3 ++-
 lib/librte_lpm/meson.build   |  2 +-
 lib/librte_regexdev/meson.build  |  2 +-
 lib/librte_ring/meson.build  |  4 +++-
 lib/librte_stack/meson.build |  4 +++-
 lib/librte_table/meson.build |  7 +++
 lib/meson.build  |  3 +++
 meson.build  |  1 +
 meson_options.txt|  2 ++
 14 files changed, 45 insertions(+), 19 deletions(-)

diff --git a/doc/guides/contributing/coding_style.rst 
b/doc/guides/contributing/coding_style.rst
index bb3f3efcbc..dba4145228 100644
--- a/doc/guides/contributing/coding_style.rst
+++ b/doc/guides/contributing/coding_style.rst
@@ -891,6 +891,18 @@ headers
installed to $PREFIX/include when ``ninja install`` is run. As with
source files, these should be specified using the meson ``files()``
function.
+   When ``check_includes`` build option is set to ``true``, each header 
file
+   has additional checks performed on it, for example to ensure that it is
+   not missing any include statements for dependent headers.
+   For header files which are public, but only included indirectly in
+   applications, these checks can be skipped by using the 
``headers_no_chkincs``
+   variable rather than ``headers``.
+
+headers_no_chkincs
+   **Default Value = []**.
+   As with ``headers`` option above, except that the files are not checked
+   for all needed include files as part of a DPDK build when
+   ``check_includes`` is set to ``true``.
 
 includes:
**Default Value = []**.
diff --git a/lib/librte_eal/include/meson.build 
b/lib/librte_eal/include/meson.build
index 0dea342e1d..449740e510 100644
--- a/lib/librte_eal/include/meson.build
+++ b/lib/librte_eal/include/meson.build
@@ -16,7 +16,6 @@ headers += files(
'rte_dev.h',
'rte_devargs.h',
'rte_eal.h',
-   'rte_eal_interrupts.h',
'rte_eal_memconfig.h',
'rte_eal_trace.h',
'rte_errno.h',
@@ -49,6 +48,7 @@ headers += files(
'rte_version.h',
'rte_vfio.h',
 )
+headers_no_chkincs += files('rte_eal_interrupts.h')
 
 # special case install the generic headers, since they go in a subdir
 generic_headers = files(
diff --git a/lib/librte_eal/x86/include/meson.build 
b/lib/librte_eal/x86/include/meson.build
index 549cc21a42..835ea22947 100644
--- a/lib/librte_eal/x86/include/meson.build
+++ b/lib/librte_eal/x86/include/meson.build
@@ -2,11 +2,7 @@
 # Copyright(c) 2017 Intel Corporation
 
 arch_headers = files(
-   'rte_atomic_32.h',
-   'rte_atomic_64.h',
'rte_atomic.h',
-   'rte_byteorder_32.h',
-   'rte_byteorder_64.h',
'rte_byteorder.h',
'rte_cpuflags.h',
'rte_cycles.h',
@@ -22,4 +18,12 @@ arch_headers = files(
'rte_ticketlock.h',
'rte_vect.h',
 )
-install_headers(arch_headers, subdir: get_option('include_subdir_arch'))
+arch_headers_no_chkincs = files(
+   'rte_atomic_32.h',
+   'rte_atomic_64.h',
+   'rte_byteorder_32.h',
+   'rte_byteorder_64.h',
+)
+install_headers(arch_headers + arch_headers_no_chkincs,
+   subdir: get_option('include_subdir_arch'))
+dpdk_chkinc_headers += arch_headers
diff --git a/lib/librte_ethdev/meson.build b/lib/librte_ethdev/meson.build
index e4b610246f..ab84869ea8 100644
--- a/lib/librte_ethdev/meson.build
+++ b/lib/librte_ethdev/meson.build
@@ -12,12 +12,10 @@ sources = files('ethdev_private.c',
 
 headers = files('rte_ethdev.h',
'rte_ethdev_driver.h',
-   'rte_ethdev_core.h',
'rte_ethdev_pci.h',
'rte_ethdev_trace.h',
'rte_ethdev_trace_fp.h',
'rte_ethdev_vdev.h',
-   'rte_eth_ctrl.h',
'rte_dev_info.h',
'rte_flow.h',
'rte_flow_driver.h',
@@ -25,5 +23,7 @@ headers = files('rte_ethdev.h',
'rte_mtr_driver.h',
'rte_tm.h',
'rte_tm_driver.h')
+headers_no_chkincs += files('rte_eth_ctrl.h',
+   'rte_ethdev_core.h')
 
 deps += ['net', 'kvargs', 'meter', 'telemetry']
diff --git a/lib/librte_hash/meson.build b/lib/librte_hash/meson.build
index 0977a63fd2..b3ebc8b078 100644
--- a/lib/librte_hash/meson.build
+++ b/lib/librte_hash/meson.build
@@ -1,12 +1,12 @@
 # SPDX-License-Identifier: BSD-3-Clause
 # Copyright(c) 2017 Intel Corporation
 
-headers = files('rte_

[dpdk-dev] [PATCH v3 0/4] add checking of header includes

[Bruce Richardson]

As a general principle, each header file should include any other
headers it needs to provide data type definitions or macros. For
example, any header using the uintX_t types in structures or function
prototypes should include "stdint.h" to provide those type definitions.

In practice, while many, but not all, headers in DPDK did include all
necessary headers, it was never actually checked that each header could
be included in a C file and compiled without having any compiler errors
about missing definitions.  The script "check-includes.sh" could be used
for this job, but it was not called out in the documentation, so many
contributors may not have been aware of it's existance. It also was
difficult to run from a source-code directory, as the script did not
automatically allow finding of headers from one DPDK library directory
to another [this was probably based on running it on a build created by
the "make" build system, where all headers were in a single directory].
To attempt to have a build-system integrated replacement, this patchset
adds a "chkincs" app in the buildtools directory to verify this on an
ongoing basis.

This chkincs app does nothing when run, and is not installed as part of
a DPDK "ninja install", it's for build-time checking only. Its source
code consists of one C file per public DPDK header, where that C file
contains nothing except an include for that header.  Therefore, if any
header is added to the lib folder which fails to compile when included
alone, the build of chkincs will fail with a suitable error message.
Since this compile checking is not needed on most builds of DPDK, the
building of chkincs is disabled by default, but can be enabled by the
"test_includes" meson option. To catch errors with patch submissions,
the final patch of this series enables it for a single build in
test-meson-builds script.

Future work could involve doing similar checks on headers for C++
compatibility, which was something done by the check-includes.sh script
but which is missing here..

V3:
* Shrunk patchset as most header fixes already applied
* Moved chkincs from "apps" to the "buildtools" directory, which is a
  better location for something not for installation for end-user use.
* Added patch to drop check-includes script.

V2:
* Add maintainers file entry for new app
* Drop patch for c11 ring header
* Use build variable "headers_no_chkincs" for tracking exceptions

Bruce Richardson (4):
  eal: add missing include to mcslock
  build: separate out headers for include checking
  buildtools/chkincs: add app to verify header includes
  devtools: remove check-includes script

 MAINTAINERS  |   5 +-
 buildtools/chkincs/gen_c_file_for_header.py  |  12 +
 buildtools/chkincs/main.c|   4 +
 buildtools/chkincs/meson.build   |  40 +++
 devtools/check-includes.sh   | 259 ---
 devtools/test-meson-builds.sh|   2 +-
 doc/guides/contributing/coding_style.rst |  12 +
 lib/librte_eal/include/generic/rte_mcslock.h |   1 +
 lib/librte_eal/include/meson.build   |   2 +-
 lib/librte_eal/x86/include/meson.build   |  14 +-
 lib/librte_ethdev/meson.build|   4 +-
 lib/librte_hash/meson.build  |   4 +-
 lib/librte_ipsec/meson.build |   3 +-
 lib/librte_lpm/meson.build   |   2 +-
 lib/librte_regexdev/meson.build  |   2 +-
 lib/librte_ring/meson.build  |   4 +-
 lib/librte_stack/meson.build |   4 +-
 lib/librte_table/meson.build |   7 +-
 lib/meson.build  |   3 +
 meson.build  |   6 +
 meson_options.txt|   2 +
 21 files changed, 112 insertions(+), 280 deletions(-)
 create mode 100755 buildtools/chkincs/gen_c_file_for_header.py
 create mode 100644 buildtools/chkincs/main.c
 create mode 100644 buildtools/chkincs/meson.build
 delete mode 100755 devtools/check-includes.sh

--
2.27.0

[dpdk-dev] [PATCH v3 3/4] buildtools/chkincs: add app to verify header includes

[Bruce Richardson]

To verify that all DPDK headers are ok for inclusion directly in a C file,
and are not missing any other pre-requisite headers, we can auto-generate
for each header an empty C file that includes that header. Compiling these
files will throw errors if any header has unmet dependencies.

To ensure ongoing compliance, we enable this build test as part of the
default x86 build in "test-meson-builds.sh".

Signed-off-by: Bruce Richardson 
---
 MAINTAINERS |  4 +++
 buildtools/chkincs/gen_c_file_for_header.py | 12 +++
 buildtools/chkincs/main.c   |  4 +++
 buildtools/chkincs/meson.build  | 40 +
 devtools/test-meson-builds.sh   |  2 +-
 meson.build |  5 +++
 6 files changed, 66 insertions(+), 1 deletion(-)
 create mode 100755 buildtools/chkincs/gen_c_file_for_header.py
 create mode 100644 buildtools/chkincs/main.c
 create mode 100644 buildtools/chkincs/meson.build

diff --git a/MAINTAINERS b/MAINTAINERS
index aa973a3960..6df494e367 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -1560,6 +1560,10 @@ F: app/test/test_resource.c
 F: app/test/virtual_pmd.c
 F: app/test/virtual_pmd.h
 
+Header build sanity checking
+M: Bruce Richardson 
+F: buildtools/chkincs/
+
 Sample packet helper functions for unit test
 M: Reshma Pattan 
 F: app/test/sample_packet_forward.c
diff --git a/buildtools/chkincs/gen_c_file_for_header.py 
b/buildtools/chkincs/gen_c_file_for_header.py
new file mode 100755
index 00..ed46948aea
--- /dev/null
+++ b/buildtools/chkincs/gen_c_file_for_header.py
@@ -0,0 +1,12 @@
+#! /usr/bin/env python3
+# SPDX-License-Identifier: BSD-3-Clause
+# Copyright(c) 2021 Intel Corporation
+
+from sys import argv
+from os.path import abspath
+
+(h_file, c_file) = argv[1:]
+
+contents = '#include "' + abspath(h_file) + '"'
+with open(c_file, 'w') as cf:
+cf.write(contents)
diff --git a/buildtools/chkincs/main.c b/buildtools/chkincs/main.c
new file mode 100644
index 00..d25bb8852a
--- /dev/null
+++ b/buildtools/chkincs/main.c
@@ -0,0 +1,4 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2021 Intel Corporation
+ */
+int main(void) { return 0; }
diff --git a/buildtools/chkincs/meson.build b/buildtools/chkincs/meson.build
new file mode 100644
index 00..5dc43283e0
--- /dev/null
+++ b/buildtools/chkincs/meson.build
@@ -0,0 +1,40 @@
+# SPDX-License-Identifier: BSD-3-Clause
+# Copyright(c) 2021 Intel Corporation
+
+if not get_option('check_includes')
+   build = false
+   subdir_done()
+endif
+
+if is_windows
+   # for windows, the shebang line in the script won't work.
+   error('option "check_includes" is not supported on windows')
+endif
+
+gen_c_file_for_header = find_program('gen_c_file_for_header.py')
+gen_c_files = generator(gen_c_file_for_header,
+   output: '@BASENAME@.c',
+   arguments: ['@INPUT@', '@OUTPUT@'])
+
+cflags = machine_args
+cflags += '-Wno-unused-function' # needed if we include generic headers
+cflags += '-DALLOW_EXPERIMENTAL_API'
+
+# some ethdev headers depend on bus headers
+includes = include_directories('../../drivers/bus/pci',
+   '../../drivers/bus/vdev')
+
+sources = files('main.c')
+sources += gen_c_files.process(dpdk_chkinc_headers)
+
+deps = []
+foreach l:enabled_libs
+   deps += get_variable('static_rte_' + l)
+endforeach
+
+executable('chkincs', sources,
+   c_args: cflags,
+   include_directories: includes,
+   dependencies: deps,
+   link_whole: dpdk_static_libraries + dpdk_drivers,
+   install: false)
diff --git a/devtools/test-meson-builds.sh b/devtools/test-meson-builds.sh
index efa91e0e40..07b5e6aeca 100755
--- a/devtools/test-meson-builds.sh
+++ b/devtools/test-meson-builds.sh
@@ -227,7 +227,7 @@ default_machine='nehalem'
 if ! check_cc_flags "-march=$default_machine" ; then
default_machine='corei7'
 fi
-build build-x86-default cc skipABI \
+build build-x86-default cc skipABI -Dcheck_includes=true\
-Dlibdir=lib -Dmachine=$default_machine $use_shared
 
 # 32-bit with default compiler
diff --git a/meson.build b/meson.build
index 7c9a8fce5f..9aa7882911 100644
--- a/meson.build
+++ b/meson.build
@@ -69,6 +69,11 @@ if get_option('enable_kmods')
subdir('kernel')
 endif
 
+# check header includes if requested
+if get_option('check_includes')
+   subdir('buildtools/chkincs')
+endif
+
 # write the build config
 build_cfg = 'rte_build_config.h'
 configure_file(output: build_cfg,
-- 
2.27.0

[dpdk-dev] [PATCH v3 4/4] devtools: remove check-includes script

[Bruce Richardson]

The check-includes script allowed checking header files in a given
directory to ensure that each header compiled alone without requiring
any other header inclusions.

With header checking now being done by the chkincs app in the build
system this script can be removed.

Signed-off-by: Bruce Richardson 
---
 MAINTAINERS|   1 -
 devtools/check-includes.sh | 259 -
 2 files changed, 260 deletions(-)
 delete mode 100755 devtools/check-includes.sh

diff --git a/MAINTAINERS b/MAINTAINERS
index 6df494e367..30721f8561 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -81,7 +81,6 @@ F: devtools/check-dup-includes.sh
 F: devtools/check-maintainers.sh
 F: devtools/check-forbidden-tokens.awk
 F: devtools/check-git-log.sh
-F: devtools/check-includes.sh
 F: devtools/check-spdx-tag.sh
 F: devtools/check-symbol-maps.sh
 F: devtools/checkpatches.sh
diff --git a/devtools/check-includes.sh b/devtools/check-includes.sh
deleted file mode 100755
index 940cf0eb2a..00
--- a/devtools/check-includes.sh
+++ /dev/null
@@ -1,259 +0,0 @@
-#!/bin/sh -e
-# SPDX-License-Identifier: BSD-3-Clause
-# Copyright 2016 6WIND S.A.
-
-# This script checks that header files in a given directory do not miss
-# dependencies when included on their own, do not conflict and accept being
-# compiled with the strictest possible flags.
-#
-# Files are looked up in the directory provided as the first argument,
-# otherwise build/include by default.
-#
-# Recognized environment variables:
-#
-# VERBOSE=1 is the same as -v.
-#
-# QUIET=1 is the same as -q.
-#
-# SUMMARY=1 is the same as -s.
-#
-# CC, CPPFLAGS, CFLAGS, CXX, CXXFLAGS are taken into account.
-#
-# PEDANTIC_CFLAGS, PEDANTIC_CXXFLAGS and PEDANTIC_CPPFLAGS provide strict
-# C/C++ compilation flags.
-#
-# IGNORE contains a list of globbing patterns matching files (relative to the
-# include directory) to avoid. It is set by default to known DPDK headers
-# which must not be included on their own.
-#
-# IGNORE_CXX provides additional files for C++.
-
-while getopts hqvs arg; do
-   case $arg in
-   h)
-   cat < /dev/null
-
-[ "$VERBOSE" = 1 ] &&
-output ()
-{
-   local CCV
-   local CXXV
-
-   shift
-   CCV=$CC
-   CXXV=$CXX
-   CC="echo $CC" CXX="echo $CXX" "$@"
-   CC=$CCV
-   CXX=$CXXV
-
-   "$@"
-} ||
-output ()
-{
-
-   printf '  %s\n' "$1"
-   shift
-   "$@"
-}
-
-trap 'rm -f "$temp_cc" "$temp_cxx"' EXIT
-
-compile_cc ()
-{
-   ${CC} -I"$include_dir" \
-   ${PEDANTIC_CPPFLAGS} ${CPPFLAGS} \
-   ${PEDANTIC_CFLAGS} ${CFLAGS} \
-   -c -o /dev/null "${temp_cc}"
-}
-
-compile_cxx ()
-{
-   ${CXX} -I"$include_dir" \
-   ${PEDANTIC_CPPFLAGS} ${CPPFLAGS} \
-   ${PEDANTIC_CXXFLAGS} ${CXXFLAGS} \
-   -c -o /dev/null "${temp_cxx}"
-}
-
-ignore ()
-{
-   file="$1"
-   shift
-   while [ $# -ne 0 ]; do
-   case "$file" in
-   $1)
-   return 0
-   ;;
-   esac
-   shift
-   done
-   return 1
-}
-
-# Check C/C++ compilation for each header file.
-
-while read -r path
-do
-   file=${path#$include_dir}
-   file=${file##/}
-   if ignore "$file" $IGNORE; then
-   output "SKIP $file" :
-   continue
-   fi
-   if printf "\
-#include <%s>
-
-int main(void)
-{
-   return 0;
-}
-" "$file" > "$temp_cc" &&
-   output "CC $file" compile_cc
-   then
-   pass_cc="$pass_cc $file"
-   else
-   failures_cc=$(($failures_cc + 1))
-   fi
-   if ignore "$file" $IGNORE_CXX; then
-   output "SKIP CXX $file" :
-   continue
-   fi
-   if printf "\
-#include <%s>
-
-int main()
-{
-}
-" "$file" > "$temp_cxx" &&
-   output "CXX $file" compile_cxx
-   then
-   pass_cxx="$pass_cxx $file"
-   else
-   failures_cxx=$(($failures_cxx + 1))
-   fi
-done < "$temp_cc" &&
-for file in $pass_cc; do
-   printf "\
-#include <%s>
-" "$file" >> $temp_cc
-done
-if printf "\
-int main(void)
-{
-   return 0;
-}
-" >> "$temp_cc" &&
-   output "CC (all includes that did not fail)" compile_cc
-then
-   :
-else
-   failures_cc=$(($failures_cc + 1))
-fi
-
-# Check C++ compilation with all includes.
-
-: > "$temp_cxx" &&
-for file in $pass_cxx; do
-   printf "\
-#include <%s>
-" "$file" >> $temp_cxx
-done
-if printf "\
-int main()
-{
-}
-" >> "$temp_cxx" &&
-   output "CXX (all includes that did not fail)" compile_cxx
-then
-   :
-else
-   failures_cxx=$(($failures_cxx + 1))
-fi
-
-# Report results.
-
-if [ "$SUMMARY" = 1 ]; then
-   printf "\
-Summary:
- %u failure(s) for C using '%s'.
- %u failure(s) for C++ using '%s'.
-" $failures_cc "$CC" $failures_cxx "$CXX" 1>&2
-fi
-
-# Exit with nonzero status if there are failures.
-
-[ $failures_c

Re: [dpdk-dev] [PATCH v3 0/3] AVX512 vPMD on i40e

[David Marchand]

On Thu, Jan 21, 2021 at 6:02 AM Kadam, Pallavi  wrote:
> On 1/20/2021 11:21 PM, Ferruh Yigit wrote:
>
> And for the mingw, I have same result with Ali, I can reproduce with (Fedora 
> MinGW 9.2.1-6.fc32).
>
> But different from the CI, I am getting the error [1] for all following files:
> rte_random.c
> i40e_rxtx_vec_sse.c
> i40e_rxtx_vec_avx512.c
> i40e_rxtx_vec_avx2.c
> rte_ethdev.c
>
> [1] Error: invalid register for .seh_savexmm
>
>
> There is a stackoverflow entry for it:
> https://stackoverflow.com/questions/43152633/invalid-register-for-seh-savexmm-in-cygwin
>
> If I use '-fno-asynchronous-unwind-tables' as suggested there, the build 
> works fine.
>
> So the problem may not be just 'i40e_rxtx_vec_avx512.c'.
>
>
> If I change the machine type from 'native' to 'corei7' [2], the build error 
> reduced to only 'i40e_rxtx_vec_avx512.c', so the problem seems happens when 
> avx512 is supported by CPU, in this case compiler seems has a defect.
> And since for 'i40e_rxtx_vec_avx512.c' the '-march=skylake-avx512' explicitly 
> set can cause the problem seen in all machines.
>
> [2]
>  diff --git a/config/x86/cross-mingw b/config/x86/cross-mingw
>  index 4c15a7fa2e..7cee238add 100644
>  --- a/config/x86/cross-mingw
>  +++ b/config/x86/cross-mingw
>  @@ -9,5 +9,5 @@ pkgconfig = 'x86_64-w64-mingw32-pkg-config'
>   [host_machine]
>   system = 'windows'
>   cpu_family = 'x86_64'
>  -cpu = 'native'
>  +cpu = 'corei7'
>   endian = 'little'
>
> @Ranjit, @Pallavi,
> Are you building using mingw, and if so are you observing same problem?
>

Thanks Ferruh.


> We usually build using Clang. However, we verify with mingw as well before 
> submitting the patch.
> As mentioned in the patch [1] comments, we replaced #include x86intrin.h with 
>  in the file i40e_rxtx_vec_avx2.c
> And this helped fixing an error related to conflicting types for 
> '__m_prefethw' with Clang on Windows.
> I was able to build this patch using Clang as well as mingw.
> [1] http://patches.dpdk.org/patch/84770/
>
> I verified patch fix submitted by Rong, Leyi, it builds successfully with 
> Clang.
> However, I am getting same error "Error: invalid register for .seh_savexmm" 
> using mingw for 'i40e_rxtx_vec_avx512.c' file.

The patch I sent https://patchwork.dpdk.org/patch/86999/ is not enough.
I have neither time nor environment to find a fix.

For now I simply stopped checking mingw builds.

Will it get fixed?


-- 
David Marchand

Re: [dpdk-dev] [PATCH v4 2/2] app/testpmd: fix max-pkt-len option invalid

[Ferruh Yigit]

On 1/25/2021 8:32 AM, Steve Yang wrote:

Moved the setting of 'DEV_RX_OFFLOAD_JUMBO_FRAME' from
'cmd_config_max_pkt_len_parsed()' to 'init_config()' caused fail the case
where 'max_rx_pkt_len' is changed from the command line via
"port config all max-pkt-len".

The 'init_config()' function is only called when testpmd is started,
but the DEV_RX_OFFLOAD_JUMBO_FRAME setting needs to be recomputed whenever
'max_rx_pkt_len' changes.

Define the 'update_jumbo_frame_offload()' function for both 'init_config()'
and 'cmd_config_max_pkt_len_parsed()', and detect if 'max_rx_pkt_len'
should be update to 1500 + overhead as default configuration. At the same
time, the offloads of rx queue also should update the value once port
offloads changed (e.g.: disabled JUMBO_FRAME offload via changed
max-pkt-len, reproduce steps [1]), otherwise, it would cause unexpected
result.

[1]
--
./x86_64-native-linuxapp-gcc/app/dpdk-testpmd  -c 0xf -n 4  -- -i
 --max-pkt-len=9000 --tx-offloads=0x8000 --rxq=4 --txq=4 --disable-rss
testpmd>  set verbose 3
testpmd>  port stop all
testpmd>  port config all max-pkt-len 1518  port start all

// Got fail error info without this patch
Configuring Port 0 (socket 1)
Ethdev port_id=0 rx_queue_id=0, new added offloads 0x800 must be
within per-queue offload capabilities 0x0 in rte_eth_rx_queue_setup()
Fail to configure port 0 rx queues //<-- Fail error info;
--

Fixes: 761c4d6690 ("app/testpmd: fix max Rx packet length for VLAN packets")

Signed-off-by: Steve Yang 
---
  app/test-pmd/cmdline.c|  1 +
  app/test-pmd/parameters.c |  1 +
  app/test-pmd/testpmd.c| 63 ---
  app/test-pmd/testpmd.h|  2 ++
  4 files changed, 50 insertions(+), 17 deletions(-)

diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
index 89034c8b72..8b6b7b6206 100644
--- a/app/test-pmd/cmdline.c
+++ b/app/test-pmd/cmdline.c
@@ -1901,6 +1901,7 @@ cmd_config_max_pkt_len_parsed(void *parsed_result,
printf("Unknown parameter\n");
return;
}
+   update_jumbo_frame_offload(pid, false);
}
  
  	init_port_config();

diff --git a/app/test-pmd/parameters.c b/app/test-pmd/parameters.c
index df5eb10d84..1c63156ddd 100644
--- a/app/test-pmd/parameters.c
+++ b/app/test-pmd/parameters.c
@@ -833,6 +833,7 @@ launch_args_parse(int argc, char** argv)
 "total-num-mbufs should be > 
1024\n");
}
if (!strcmp(lgopts[opt_idx].name, "max-pkt-len")) {
+   default_max_pktlen = false;
n = atoi(optarg);
if (n >= RTE_ETHER_MIN_LEN)
rx_mode.max_rx_pkt_len = (uint32_t) n;
diff --git a/app/test-pmd/testpmd.c b/app/test-pmd/testpmd.c
index c256e719ae..f22d2be46d 100644
--- a/app/test-pmd/testpmd.c
+++ b/app/test-pmd/testpmd.c
@@ -531,6 +531,8 @@ uint16_t gso_max_segment_size = RTE_ETHER_MAX_LEN - 
RTE_ETHER_CRC_LEN;
  /* Holds the registered mbuf dynamic flags names. */
  char dynf_names[64][RTE_MBUF_DYN_NAMESIZE];
  
+bool default_max_pktlen = true;

+
  /*
   * Helper function to check if socket is already discovered.
   * If yes, return positive value. If not, return zero.
@@ -1410,7 +1412,6 @@ init_config(void)
struct rte_gro_param gro_param;
uint32_t gso_types;
uint16_t data_size;
-   uint16_t eth_overhead;
bool warning = 0;
int k;
int ret;
@@ -1447,22 +1448,7 @@ init_config(void)
rte_exit(EXIT_FAILURE,
 "rte_eth_dev_info_get() failed\n");
  
-		/* Update the max_rx_pkt_len to have MTU as RTE_ETHER_MTU */

-   if (port->dev_info.max_mtu != UINT16_MAX &&
-   port->dev_info.max_rx_pktlen > port->dev_info.max_mtu)
-   eth_overhead = port->dev_info.max_rx_pktlen -
-   port->dev_info.max_mtu;
-   else
-   eth_overhead =
-   RTE_ETHER_HDR_LEN + RTE_ETHER_CRC_LEN;
-
-   if (port->dev_conf.rxmode.max_rx_pkt_len <=
-   (uint32_t)(RTE_ETHER_MTU + eth_overhead))
-   port->dev_conf.rxmode.max_rx_pkt_len =
-   RTE_ETHER_MTU + eth_overhead;
-   else
-   port->dev_conf.rxmode.offloads |=
-   DEV_RX_OFFLOAD_JUMBO_FRAME;
+   update_jumbo_frame_offload(pid, default_max_pktlen);
  
  		if (!(port->dev_info.tx_offload_capa &

  DEV_TX_OFFLOAD_MBUF_FAST_FREE))
@@ -3358,6 +3344,49 @@ rxtx_port_config(struct rte_port *port)
}
  }
  
+void

+upda

Re: [dpdk-dev] [PATCH v2 36/44] net/virtio: move Vhost-user reqs to Vhost-user backend

[Maxime Coquelin]

On 1/21/21 9:56 AM, Xia, Chenbo wrote:
> Hi Maxime,
> 
>> -Original Message-
>> From: Maxime Coquelin 
>> Sent: Wednesday, January 20, 2021 5:25 AM
>> To: dev@dpdk.org; Xia, Chenbo ; olivier.m...@6wind.com;
>> amore...@redhat.com; david.march...@redhat.com
>> Cc: Maxime Coquelin 
>> Subject: [PATCH v2 36/44] net/virtio: move Vhost-user reqs to Vhost-user
> 
> Better use full name 'requests' if the title's not too long?

Fixed, that's not too long with requests.

> Same thing could be done for 'net/virtio: make Vhost-user req sender 
> consistent'

Ditto

> With this fixed:
> 
> Reviewed-by: Chenbo Xia 
> 

Thanks,
Maxime

>> backend
>>
>> Now that we have a proper isolation of the backends,
>> we can move Vhost-user requests declaration to the
>> Vhost-user backend file.
>>
>> Signed-off-by: Maxime Coquelin 
>> ---
>>  drivers/net/virtio/virtio_user/vhost.h  | 25 -
>>  drivers/net/virtio/virtio_user/vhost_user.c | 25 +
>>  2 files changed, 25 insertions(+), 25 deletions(-)
>>
>> diff --git a/drivers/net/virtio/virtio_user/vhost.h
>> b/drivers/net/virtio/virtio_user/vhost.h
>> index 6294b8afee..2aa6b2cb70 100644
>> --- a/drivers/net/virtio/virtio_user/vhost.h
>> +++ b/drivers/net/virtio/virtio_user/vhost.h
>> @@ -63,31 +63,6 @@ struct vhost_vring_addr {
>>  #define VHOST_USER_PROTOCOL_F_STATUS 16
>>  #endif
>>
>> -enum vhost_user_request {
>> -VHOST_USER_NONE = 0,
>> -VHOST_USER_GET_FEATURES = 1,
>> -VHOST_USER_SET_FEATURES = 2,
>> -VHOST_USER_SET_OWNER = 3,
>> -VHOST_USER_RESET_OWNER = 4,
>> -VHOST_USER_SET_MEM_TABLE = 5,
>> -VHOST_USER_SET_LOG_BASE = 6,
>> -VHOST_USER_SET_LOG_FD = 7,
>> -VHOST_USER_SET_VRING_NUM = 8,
>> -VHOST_USER_SET_VRING_ADDR = 9,
>> -VHOST_USER_SET_VRING_BASE = 10,
>> -VHOST_USER_GET_VRING_BASE = 11,
>> -VHOST_USER_SET_VRING_KICK = 12,
>> -VHOST_USER_SET_VRING_CALL = 13,
>> -VHOST_USER_SET_VRING_ERR = 14,
>> -VHOST_USER_GET_PROTOCOL_FEATURES = 15,
>> -VHOST_USER_SET_PROTOCOL_FEATURES = 16,
>> -VHOST_USER_GET_QUEUE_NUM = 17,
>> -VHOST_USER_SET_VRING_ENABLE = 18,
>> -VHOST_USER_SET_STATUS = 39,
>> -VHOST_USER_GET_STATUS = 40,
>> -VHOST_USER_MAX
>> -};
>> -
>>  #ifndef VHOST_BACKEND_F_IOTLB_MSG_V2
>>  #define VHOST_BACKEND_F_IOTLB_MSG_V2 1
>>  #endif
>> diff --git a/drivers/net/virtio/virtio_user/vhost_user.c
>> b/drivers/net/virtio/virtio_user/vhost_user.c
>> index d46e25b64b..fb6fcc82c9 100644
>> --- a/drivers/net/virtio/virtio_user/vhost_user.c
>> +++ b/drivers/net/virtio/virtio_user/vhost_user.c
>> @@ -27,6 +27,31 @@ struct vhost_memory {
>>  struct vhost_memory_region regions[VHOST_MEMORY_MAX_NREGIONS];
>>  };
>>
>> +enum vhost_user_request {
>> +VHOST_USER_NONE = 0,
>> +VHOST_USER_GET_FEATURES = 1,
>> +VHOST_USER_SET_FEATURES = 2,
>> +VHOST_USER_SET_OWNER = 3,
>> +VHOST_USER_RESET_OWNER = 4,
>> +VHOST_USER_SET_MEM_TABLE = 5,
>> +VHOST_USER_SET_LOG_BASE = 6,
>> +VHOST_USER_SET_LOG_FD = 7,
>> +VHOST_USER_SET_VRING_NUM = 8,
>> +VHOST_USER_SET_VRING_ADDR = 9,
>> +VHOST_USER_SET_VRING_BASE = 10,
>> +VHOST_USER_GET_VRING_BASE = 11,
>> +VHOST_USER_SET_VRING_KICK = 12,
>> +VHOST_USER_SET_VRING_CALL = 13,
>> +VHOST_USER_SET_VRING_ERR = 14,
>> +VHOST_USER_GET_PROTOCOL_FEATURES = 15,
>> +VHOST_USER_SET_PROTOCOL_FEATURES = 16,
>> +VHOST_USER_GET_QUEUE_NUM = 17,
>> +VHOST_USER_SET_VRING_ENABLE = 18,
>> +VHOST_USER_SET_STATUS = 39,
>> +VHOST_USER_GET_STATUS = 40,
>> +VHOST_USER_MAX
>> +};
>> +
>>  struct vhost_user_msg {
>>  enum vhost_user_request request;
>>
>> --
>> 2.29.2
> 

Re: [dpdk-dev] [PATCH v2 01/44] bus/vdev: add helper to get vdev from eth dev

[Maxime Coquelin]

On 1/21/21 9:58 AM, Xia, Chenbo wrote:
> Hi Maxime,
> 
>> -Original Message-
>> From: Maxime Coquelin 
>> Sent: Wednesday, January 20, 2021 5:24 AM
>> To: dev@dpdk.org; Xia, Chenbo ; olivier.m...@6wind.com;
>> amore...@redhat.com; david.march...@redhat.com
>> Cc: Maxime Coquelin 
>> Subject: [PATCH v2 01/44] bus/vdev: add helper to get vdev from eth dev
> 
> Sorry that I missed the title in v1..
> 
> Better use 'ethdev' or 'eth device'here?

Changed to ethedv.

Thanks,
Maxime

> Thanks,
> Chenbo
> 
>>
>> This patch adds an helper macro to get the rte_vdev_device
>> pointer from a rte_eth_dev pointer.
>>
>> This is similar to RTE_ETH_DEV_TO_PCI().
>>
>> Signed-off-by: Maxime Coquelin 
>> Reviewed-by: Chenbo Xia 
>> Reviewed-by: David Marchand 
>> ---
>>  drivers/bus/vdev/rte_bus_vdev.h | 2 ++
>>  1 file changed, 2 insertions(+)
>>
>> diff --git a/drivers/bus/vdev/rte_bus_vdev.h 
>> b/drivers/bus/vdev/rte_bus_vdev.h
>> index d14eeb41b0..f99a41f825 100644
>> --- a/drivers/bus/vdev/rte_bus_vdev.h
>> +++ b/drivers/bus/vdev/rte_bus_vdev.h
>> @@ -34,6 +34,8 @@ struct rte_vdev_device {
>>  #define RTE_DEV_TO_VDEV_CONST(ptr) \
>>  container_of(ptr, const struct rte_vdev_device, device)
>>
>> +#define RTE_ETH_DEV_TO_VDEV(eth_dev)
>> RTE_DEV_TO_VDEV((eth_dev)->device)
>> +
>>  static inline const char *
>>  rte_vdev_device_name(const struct rte_vdev_device *dev)
>>  {
>> --
>> 2.29.2
> 

[dpdk-dev] [PATCH v2 0/3] support SSL/TLS way of cipher-auth operations

[Tejasree Kondoj]

This series adds SSL/TLS way of cipher-auth operations support i.e. auth
generation followed by encryption and decryption followed by auth verify in
OCTEON TX, OCTEON TX2 PMDs and sample unit test application.
It also adds feature flag for SSL/TLS order in cryptodev library so that
SSL test cases are skipped if the device doesn't support the feature.

v2:
* Added feature flag for SSL/TLS order in cryptodev library
* Moved SSL/TLS test cases to proper test sub-suite

Tejasree Kondoj (3):
  cryptodev: add feature flag for SSL/TLS order
  common/cpt: add support for SSL/TLS way of cipher-auth operations
  test/crypto: add support for SSL/TLS way of cipher-auth operations

 app/test/test_cryptodev_aes_test_vectors.h| 589 ++
 app/test/test_cryptodev_blockcipher.c | 107 +++-
 app/test/test_cryptodev_blockcipher.h |  10 +
 doc/guides/cryptodevs/features/default.ini|   1 +
 doc/guides/cryptodevs/features/octeontx.ini   |   1 +
 doc/guides/cryptodevs/features/octeontx2.ini  |   1 +
 doc/guides/rel_notes/release_21_02.rst|  18 +
 drivers/common/cpt/cpt_mcode_defines.h|   7 +-
 drivers/common/cpt/cpt_ucode.h|  42 +-
 drivers/crypto/octeontx/otx_cryptodev_ops.c   |  11 +-
 drivers/crypto/octeontx2/otx2_cryptodev.c |   3 +-
 drivers/crypto/octeontx2/otx2_cryptodev_ops.c |   8 +-
 lib/librte_cryptodev/rte_cryptodev.h  |   2 +
 13 files changed, 774 insertions(+), 26 deletions(-)

-- 
2.27.0

[dpdk-dev] [PATCH v2 1/3] cryptodev: add feature flag for SSL/TLS order

[Tejasree Kondoj]

Added device feature flag for SSL/TLS way of cipher-auth
operations support i.e. auth generation followed by encryption
and decryption followed by auth verify so that SSL/TLS tests
are skipped if the device doesn't support the feature.

Signed-off-by: Tejasree Kondoj 
---
 doc/guides/cryptodevs/features/default.ini | 1 +
 doc/guides/rel_notes/release_21_02.rst | 7 +++
 lib/librte_cryptodev/rte_cryptodev.h   | 2 ++
 3 files changed, 10 insertions(+)

diff --git a/doc/guides/cryptodevs/features/default.ini 
b/doc/guides/cryptodevs/features/default.ini
index 17b177fc45..cd5f75d8f6 100644
--- a/doc/guides/cryptodevs/features/default.ini
+++ b/doc/guides/cryptodevs/features/default.ini
@@ -31,6 +31,7 @@ CPU crypto =
 Symmetric sessionless  =
 Non-Byte aligned data  =
 Sym raw data path API  =
+SSL way of chaining=
 
 ;
 ; Supported crypto algorithms of a default crypto driver.
diff --git a/doc/guides/rel_notes/release_21_02.rst 
b/doc/guides/rel_notes/release_21_02.rst
index e72a582b1b..45854d5d33 100644
--- a/doc/guides/rel_notes/release_21_02.rst
+++ b/doc/guides/rel_notes/release_21_02.rst
@@ -82,6 +82,13 @@ New Features
   enable applications to add/remove user callbacks which gets called
   for every enqueue/dequeue operation.
 
+* **Added feature flag in cryptodev library.**
+
+  Added device feature flag for SSL/TLS way of cipher-auth operations support
+  i.e. auth generation followed by encryption and decryption followed by
+  auth verify in cryptodev library so that SSL/TLS tests are skipped if the
+  device doesn't support the feature.
+
 * **Updated the OCTEON TX2 crypto PMD.**
 
   * Updated the OCTEON TX2 crypto PMD lookaside protocol offload for IPsec with
diff --git a/lib/librte_cryptodev/rte_cryptodev.h 
b/lib/librte_cryptodev/rte_cryptodev.h
index ae34f33f69..6ee3f33c06 100644
--- a/lib/librte_cryptodev/rte_cryptodev.h
+++ b/lib/librte_cryptodev/rte_cryptodev.h
@@ -461,6 +461,8 @@ rte_cryptodev_asym_get_xform_enum(enum 
rte_crypto_asym_xform_type *xform_enum,
 /**< Support operations on data which is not byte aligned */
 #define RTE_CRYPTODEV_FF_SYM_RAW_DP(1ULL << 24)
 /**< Support accelerator specific symmetric raw data-path APIs */
+#define RTE_CRYPTODEV_FF_SSL_ORDER (1ULL << 25)
+/**< Support SSL order of cipher-auth xforms chaining */
 
 /**
  * Get the name of a crypto device feature flag
-- 
2.27.0

[dpdk-dev] [PATCH v2 2/3] common/cpt: add support for SSL/TLS way of cipher-auth operations

[Tejasree Kondoj]

Adding support for SSL/TLS way of cipher-auth operations order
- auth generation followed by encryption
- decryption followed by auth verify

Signed-off-by: Tejasree Kondoj 
---
 doc/guides/cryptodevs/features/octeontx.ini   |  1 +
 doc/guides/cryptodevs/features/octeontx2.ini  |  1 +
 doc/guides/rel_notes/release_21_02.rst|  5 +++
 drivers/common/cpt/cpt_mcode_defines.h|  7 +++-
 drivers/common/cpt/cpt_ucode.h| 42 +++
 drivers/crypto/octeontx/otx_cryptodev_ops.c   | 11 +++--
 drivers/crypto/octeontx2/otx2_cryptodev.c |  3 +-
 drivers/crypto/octeontx2/otx2_cryptodev_ops.c |  8 +++-
 8 files changed, 64 insertions(+), 14 deletions(-)

diff --git a/doc/guides/cryptodevs/features/octeontx.ini 
b/doc/guides/cryptodevs/features/octeontx.ini
index 10d94e3f7b..4b93368d88 100644
--- a/doc/guides/cryptodevs/features/octeontx.ini
+++ b/doc/guides/cryptodevs/features/octeontx.ini
@@ -14,6 +14,7 @@ OOP SGL In SGL Out = Y
 OOP LB  In LB  Out = Y
 RSA PRIV OP KEY QT = Y
 Symmetric sessionless  = Y
+SSL way of chaining= Y
 
 ;
 ; Supported crypto algorithms of 'octeontx' crypto driver.
diff --git a/doc/guides/cryptodevs/features/octeontx2.ini 
b/doc/guides/cryptodevs/features/octeontx2.ini
index b0d50ce984..7769b7da03 100644
--- a/doc/guides/cryptodevs/features/octeontx2.ini
+++ b/doc/guides/cryptodevs/features/octeontx2.ini
@@ -15,6 +15,7 @@ OOP SGL In SGL Out = Y
 OOP LB  In LB  Out = Y
 RSA PRIV OP KEY QT = Y
 Symmetric sessionless  = Y
+SSL way of chaining= Y
 
 ;
 ; Supported crypto algorithms of 'octeontx2' crypto driver.
diff --git a/doc/guides/rel_notes/release_21_02.rst 
b/doc/guides/rel_notes/release_21_02.rst
index 45854d5d33..b8c9cb7825 100644
--- a/doc/guides/rel_notes/release_21_02.rst
+++ b/doc/guides/rel_notes/release_21_02.rst
@@ -98,6 +98,11 @@ New Features
 PMD lookaside protocol offload for IPsec.
   * Added support for aes-cbc sha256-128-hmac cipher combination in OCTEON TX2
 crypto PMD lookaside protocol offload for IPsec.
+  * Added SSL/TLS way of cipher-auth operations support in OCTEON TX2 crypto 
PMD.
+
+* **Updated the OCTEON TX crypto PMD.**
+
+  * Added SSL/TLS way of cipher-auth operations support in OCTEON TX crypto 
PMD.
 
 
 Removed Items
diff --git a/drivers/common/cpt/cpt_mcode_defines.h 
b/drivers/common/cpt/cpt_mcode_defines.h
index 56a745f419..624bdcf3cf 100644
--- a/drivers/common/cpt/cpt_mcode_defines.h
+++ b/drivers/common/cpt/cpt_mcode_defines.h
@@ -20,6 +20,9 @@
 #define CPT_MAJOR_OP_ZUC_SNOW3G0x37
 #define CPT_MAJOR_OP_KASUMI0x38
 #define CPT_MAJOR_OP_MISC  0x01
+#define CPT_HMAC_FIRST_BIT_POS 0x4
+#define CPT_FC_MINOR_OP_ENCRYPT0x0
+#define CPT_FC_MINOR_OP_DECRYPT0x1
 
 /* AE opcodes */
 #define CPT_MAJOR_OP_MODEX 0x03
@@ -314,8 +317,10 @@ struct cpt_ctx {
uint64_t hmac   :1;
uint64_t zsk_flags  :3;
uint64_t k_ecb  :1;
+   uint64_t auth_enc   :1;
+   uint64_t dec_auth   :1;
uint64_t snow3g :2;
-   uint64_t rsvd   :21;
+   uint64_t rsvd   :19;
/* Below fields are accessed by hardware */
union {
mc_fc_context_t fctx;
diff --git a/drivers/common/cpt/cpt_ucode.h b/drivers/common/cpt/cpt_ucode.h
index 0536620710..ee6d49aae7 100644
--- a/drivers/common/cpt/cpt_ucode.h
+++ b/drivers/common/cpt/cpt_ucode.h
@@ -752,7 +752,9 @@ cpt_enc_hmac_prep(uint32_t flags,
 
/* Encryption */
vq_cmd_w0.s.opcode.major = CPT_MAJOR_OP_FC;
-   vq_cmd_w0.s.opcode.minor = 0;
+   vq_cmd_w0.s.opcode.minor = CPT_FC_MINOR_OP_ENCRYPT;
+   vq_cmd_w0.s.opcode.minor |= (cpt_ctx->auth_enc <<
+   CPT_HMAC_FIRST_BIT_POS);
 
if (hash_type == GMAC_TYPE) {
encr_offset = 0;
@@ -779,6 +781,9 @@ cpt_enc_hmac_prep(uint32_t flags,
outputlen = enc_dlen + mac_len;
}
 
+   if (cpt_ctx->auth_enc != 0)
+   outputlen = enc_dlen;
+
/* GP op header */
vq_cmd_w0.s.param1 = encr_data_len;
vq_cmd_w0.s.param2 = auth_data_len;
@@ -1112,7 +1117,9 @@ cpt_dec_hmac_prep(uint32_t flags,
 
/* Decryption */
vq_cmd_w0.s.opcode.major = CPT_MAJOR_OP_FC;
-   vq_cmd_w0.s.opcode.minor = 1;
+   vq_cmd_w0.s.opcode.minor = CPT_FC_MINOR_OP_DECRYPT;
+   vq_cmd_w0.s.opcode.minor |= (cpt_ctx->dec_auth <<
+   CPT_HMAC_FIRST_BIT_POS);
 
if (hash_type == GMAC_TYPE) {
encr_offset = 0;
@@ -1130,6 +1137,9 @@ cpt_dec_hmac_prep(uint32_t flags,
outputlen = enc_dlen;
}
 
+   if (cpt_ctx->dec_auth != 0)
+   outputlen = inputlen = enc_dlen;
+
vq_cmd_w0.s.param1 = encr_data_len;
vq_cmd_w0.s.param2 = auth_data_len;
 
@@ -2566,6 +2576,7 @@ fill_sess_cipher(struct rte_crypto_sym_xform *xform,
 struct cp

[dpdk-dev] [PATCH v2 3/3] test/crypto: add support for SSL/TLS way of cipher-auth operations

[Tejasree Kondoj]

Adding support for SSL/TLS way of cipher-auth operations order
- auth generation followed by encryption
- decryption followed by auth verify

Signed-off-by: Tejasree Kondoj 
---
 app/test/test_cryptodev_aes_test_vectors.h | 589 +
 app/test/test_cryptodev_blockcipher.c  | 107 +++-
 app/test/test_cryptodev_blockcipher.h  |  10 +
 doc/guides/rel_notes/release_21_02.rst |   6 +
 4 files changed, 700 insertions(+), 12 deletions(-)

diff --git a/app/test/test_cryptodev_aes_test_vectors.h 
b/app/test/test_cryptodev_aes_test_vectors.h
index c192d75a7e..3c3e3159c4 100644
--- a/app/test/test_cryptodev_aes_test_vectors.h
+++ b/app/test/test_cryptodev_aes_test_vectors.h
@@ -1093,6 +1093,172 @@ static const uint8_t ciphertext512_aes128cbc_aad[] = {
0x73, 0x65, 0x72, 0x73, 0x2C, 0x20, 0x73, 0x75
 };
 
+static const uint8_t plaintext_aes_common_ssl[] = {
+   0x57, 0x68, 0x61, 0x74, 0x20, 0x61, 0x20, 0x6C,
+   0x6F, 0x75, 0x73, 0x79, 0x20, 0x65, 0x61, 0x72,
+   0x74, 0x68, 0x21, 0x20, 0x48, 0x65, 0x20, 0x77,
+   0x6F, 0x6E, 0x64, 0x65, 0x72, 0x65, 0x64, 0x20,
+   0x68, 0x6F, 0x77, 0x20, 0x6D, 0x61, 0x6E, 0x79,
+   0x20, 0x70, 0x65, 0x6F, 0x70, 0x6C, 0x65, 0x20,
+   0x77, 0x65, 0x72, 0x65, 0x20, 0x64, 0x65, 0x73,
+   0x74, 0x69, 0x74, 0x75, 0x74, 0x65, 0x20, 0x74,
+   0x68, 0x61, 0x74, 0x20, 0x73, 0x61, 0x6D, 0x65,
+   0x20, 0x6E, 0x69, 0x67, 0x68, 0x74, 0x20, 0x65,
+   0x76, 0x65, 0x6E, 0x20, 0x69, 0x6E, 0x20, 0x68,
+   0x69, 0x73, 0x20, 0x6F, 0x77, 0x6E, 0x20, 0x70,
+   0x72, 0x6F, 0x73, 0x70, 0x65, 0x72, 0x6F, 0x75,
+   0x73, 0x20, 0x63, 0x6F, 0x75, 0x6E, 0x74, 0x72,
+   0x79, 0x2C, 0x20, 0x68, 0x6F, 0x77, 0x20, 0x6D,
+   0x61, 0x6E, 0x79, 0x20, 0x68, 0x6F, 0x6D, 0x65,
+   0x73, 0x20, 0x77, 0x65, 0x72, 0x65, 0x20, 0x73,
+   0x68, 0x61, 0x6E, 0x74, 0x69, 0x65, 0x73, 0x2C,
+   0x20, 0x68, 0x6F, 0x77, 0x20, 0x6D, 0x61, 0x6E,
+   0x79, 0x20, 0x68, 0x75, 0x73, 0x62, 0x61, 0x6E,
+   0x64, 0x73, 0x20, 0x77, 0x65, 0x72, 0x65, 0x20,
+   0x64, 0x72, 0x75, 0x6E, 0x6B, 0x20, 0x61, 0x6E,
+   0x64, 0x20, 0x77, 0x69, 0x76, 0x65, 0x73, 0x20,
+   0x73, 0x6F, 0x63, 0x6B, 0x65, 0x64, 0x2C, 0x20,
+   0x61, 0x6E, 0x64, 0x20, 0x68, 0x6F, 0x77, 0x20,
+   0x6D, 0x61, 0x6E, 0x79, 0x20, 0x63, 0x68, 0x69,
+   0x6C, 0x64, 0x72, 0x65, 0x6E, 0x20, 0x77, 0x65,
+   0x72, 0x65, 0x20, 0x62, 0x75, 0x6C, 0x6C, 0x69,
+   0x65, 0x64, 0x2C, 0x20, 0x61, 0x62, 0x75, 0x73,
+   0x65, 0x64, 0x2C, 0x20, 0x6F, 0x72, 0x20, 0x61,
+   0x62, 0x61, 0x6E, 0x64, 0x6F, 0x6E, 0x65, 0x64,
+   0x2E, 0x20, 0x48, 0x6F, 0x77, 0x20, 0x6D, 0x61,
+   0x6E, 0x79, 0x20, 0x66, 0x61, 0x6D, 0x69, 0x6C,
+   0x69, 0x65, 0x73, 0x20, 0x68, 0x75, 0x6E, 0x67,
+   0x65, 0x72, 0x65, 0x64, 0x20, 0x66, 0x6F, 0x72,
+   0x20, 0x66, 0x6F, 0x6F, 0x64, 0x20, 0x74, 0x68,
+   0x65, 0x79, 0x20, 0x63, 0x6F, 0x75, 0x6C, 0x64,
+   0x20, 0x6E, 0x6F, 0x74, 0x20, 0x61, 0x66, 0x66,
+   0x6F, 0x72, 0x64, 0x20, 0x74, 0x6F, 0x20, 0x62,
+   0x75, 0x79, 0x3F, 0x20, 0x48, 0x6F, 0x77, 0x20,
+   0x6D, 0x61, 0x6E, 0x79, 0x20, 0x68, 0x65, 0x61,
+   0x72, 0x74, 0x73, 0x20, 0x77, 0x65, 0x72, 0x65,
+   0x20, 0x62, 0x72, 0x6F, 0x6B, 0x65, 0x6E, 0x3F,
+   0x20, 0x48, 0x6F, 0x77, 0x20, 0x6D, 0x61, 0x6E,
+   0x79, 0x20, 0x73, 0x75, 0x69, 0x63, 0x69, 0x64,
+   0x65, 0x73, 0x20, 0x77, 0x6F, 0x75, 0x6C, 0x64,
+   0x20, 0x74, 0x61, 0x6B, 0x65, 0x20, 0x70, 0x6C,
+   0x61, 0x63, 0x65, 0x20, 0x74, 0x68, 0x61, 0x74,
+   0x20, 0x73, 0x61, 0x6D, 0x65, 0x20, 0x6E, 0x69,
+   0x67, 0x68, 0x74, 0x2C, 0x20, 0x68, 0x6F, 0x77,
+   0x20, 0x6D, 0x61, 0x6E, 0x79, 0x20, 0x70, 0x65,
+   0x6F, 0x70, 0x6C, 0x65, 0x20, 0x77, 0x6F, 0x75,
+   0x6C, 0x64, 0x20, 0x67, 0x6F, 0x20, 0x69, 0x6E,
+   0x73, 0x61, 0x6E, 0x65, 0x3F, 0x20, 0x48, 0x6F,
+   0x77, 0x20, 0x6D, 0x61, 0x6E, 0x79, 0x20, 0x63,
+   0x6F, 0x63, 0x6B, 0x72, 0x6F, 0x61, 0x63, 0x68,
+   0x65, 0x73, 0x20, 0x61, 0x6E, 0x64, 0x20, 0x6C,
+   0x61, 0x6E, 0x64, 0x6C, 0x6F, 0x72, 0x64, 0x73,
+   0x20, 0x77, 0x6F, 0x75, 0x6C, 0x64, 0x20, 0x74,
+   0x72, 0x69, 0x75, 0x6D, 0x70, 0x68, 0x3F, 0x20,
+   0x48, 0x6F, 0x77, 0x20, 0x6D, 0x61, 0x6E, 0x79,
+   0x20, 0x77, 0x69, 0x6E, 0x6E, 0x65, 0x72, 0x73,
+   0x20, 0x77, 0x65, 0x72, 0x65, 0x20, 0x6c, 0x6f,
+   0x73, 0x65, 0x72, 0x73, 0x2c, 0x20, 0x73, 0x75,
+   /* mac */
+   0xC4, 0xB7, 0x0E, 0x6B, 0xDE, 0xD1, 0xE7, 0x77,
+   0x7E, 0x2E, 0x8F, 0xFC, 0x48, 0x39, 0x46, 0x17,
+   0x3F, 0x91, 0x64, 0x59
+};
+
+static const uint8_t ciphertext512_aes128cbc_ssl[] = {
+   0x8B, 0x4D, 0xDA, 0x1B, 0xCF, 0x04, 0xA0, 0x31,
+   0xB4, 0xBF, 0xBD, 0x68, 0x43, 0x20, 0x7E, 0x76,
+   0xB1, 0x96, 0x8B, 0xA2, 0x7C, 0xA2, 0x83, 0x9E,
+   0x39, 0x5A, 0x2F, 0x7E, 0x92, 0xB4, 0x48, 0x1A,
+   0x3F, 0x6B, 0x5D, 0xDF, 0x52, 0x85, 0x5F, 0x8E,
+   0x42

Re: [dpdk-dev] [PATCH v2 27/44] net/virtio: add Virtio-user features ops

[Maxime Coquelin]

On 1/22/21 8:25 AM, Xia, Chenbo wrote:
> Hi Maxime,
> 
>> -Original Message-
>> From: Maxime Coquelin 
>> Sent: Wednesday, January 20, 2021 5:25 AM
>> To: dev@dpdk.org; Xia, Chenbo ; olivier.m...@6wind.com;
>> amore...@redhat.com; david.march...@redhat.com
>> Cc: Maxime Coquelin 
>> Subject: [PATCH v2 27/44] net/virtio: add Virtio-user features ops
>>
>> This patch introduces new callbacks for getting
>> and setting Virtio features, and implements them
>> for the different backend types.
>>
>> Signed-off-by: Maxime Coquelin 
>> ---
>>  drivers/net/virtio/virtio_user/vhost.h|   2 +
>>  drivers/net/virtio/virtio_user/vhost_kernel.c | 150 +-
>>  .../net/virtio/virtio_user/vhost_kernel_tap.c |  23 +++
>>  .../net/virtio/virtio_user/vhost_kernel_tap.h |   1 +
>>  drivers/net/virtio/virtio_user/vhost_user.c   |  64 +++-
>>  drivers/net/virtio/virtio_user/vhost_vdpa.c   |  38 +++--
>>  .../net/virtio/virtio_user/virtio_user_dev.c  |   5 +-
>>  drivers/net/virtio/virtio_user_ethdev.c   |   3 +-
>>  8 files changed, 190 insertions(+), 96 deletions(-)
>>
>> diff --git a/drivers/net/virtio/virtio_user/vhost.h
>> b/drivers/net/virtio/virtio_user/vhost.h
>> index 5413ec6778..13a88c7671 100644
>> --- a/drivers/net/virtio/virtio_user/vhost.h
>> +++ b/drivers/net/virtio/virtio_user/vhost.h
>> @@ -110,6 +110,8 @@ struct virtio_user_dev;
> 
> 
> 
>> +
>> +ret = tap_support_features(&tap_features);
>> +if (ret < 0) {
>> +PMD_DRV_LOG(ERR, "Failed to get TAP features)");
> 
> Delete the ')' before " ?

Indeed, done.

>> +return -1;
>> +}
>> +
> 
> 
> 
>>
>> +static int
>> +vhost_user_get_features(struct virtio_user_dev *dev, uint64_t *features)
>> +{
>> +int ret;
>> +struct vhost_user_msg msg = {
>> +.request = VHOST_USER_GET_FEATURES,
>> +.flags = VHOST_USER_VERSION,
>> +};
>> +
>> +ret = vhost_user_write(dev->vhostfd, &msg, NULL, 0);
>> +if (ret < 0)
>> +goto err;
>> +
>> +ret = vhost_user_read(dev->vhostfd, &msg);
>> +if (ret < 0)
>> +goto err;
> 
> Better add error log both after write and read?

As you replied later, we already log in read & write functions, so
skipping this comment.

> Thanks!
> Chenbo
> 

Thanks,
Maxime

Re: [dpdk-dev] [PATCH v2 29/44] net/virtio: add Virtio-user memory tables ops

[Maxime Coquelin]

On 1/22/21 8:34 AM, Xia, Chenbo wrote:
> Hi Maxime,
> 
>> -Original Message-
>> From: Maxime Coquelin 
>> Sent: Wednesday, January 20, 2021 5:25 AM
>> To: dev@dpdk.org; Xia, Chenbo ; olivier.m...@6wind.com;
>> amore...@redhat.com; david.march...@redhat.com
>> Cc: Maxime Coquelin 
>> Subject: [PATCH v2 29/44] net/virtio: add Virtio-user memory tables ops
>>
>> This patch implements a dedicated callback for
>> preparing and sending memory table to the backends.
>>
>> Signed-off-by: Maxime Coquelin 
>> ---
>>  drivers/net/virtio/virtio_user/vhost.h|  1 +
>>  drivers/net/virtio/virtio_user/vhost_kernel.c | 60 ++---
>>  drivers/net/virtio/virtio_user/vhost_user.c   | 86 ++-
>>  drivers/net/virtio/virtio_user/vhost_vdpa.c   |  8 +-
>>  .../net/virtio/virtio_user/virtio_user_dev.c  |  4 +-
>>  5 files changed, 101 insertions(+), 58 deletions(-)
>>
...
>> +static int
>> +vhost_user_check_reply_ack(struct virtio_user_dev *dev, struct 
>> vhost_user_msg
>> *msg)
>> +{
>> +enum vhost_user_request req = msg->request;
>> +int ret;
>> +
>> +if (!(msg->flags & VHOST_USER_NEED_REPLY_MASK))
>> +return 0;
>> +
>> +ret = vhost_user_read(dev->vhostfd, msg);
>> +if (ret < 0) {
>> +PMD_DRV_LOG(ERR, "Failed to read reply-ack");
>> +return -1;
>> +}
>> +
>> +if (req != msg->request) {
>> +PMD_DRV_LOG(ERR, "Unexpected reply-ack request type (%d)", msg-
>>> request);
>> +return -1;
>> +}
>> +
>> +if (msg->size != sizeof(msg->payload.u64)) {
>> +PMD_DRV_LOG(ERR, "Unexpected reply-ack payload size (%d)", msg-
>>> size);
> 
> Better to use '%u'?

Done

>> +return -1;
>> +}
>> +
>> +if (msg->payload.u64) {
>> +PMD_DRV_LOG(ERR, "Slave replied NACK to request type %d", msg-
>>> request);
> 
> Replace '%d' with '(%d)' to align with the above style? :P

Added to make style consistent.

> Thanks,
> Chenbo
> 

Thanks,
Maxime

Re: [dpdk-dev] [PATCH v15 09/12] build: disable drivers in Arm builds

[Honnappa Nagarahalli]

> 
> 22/01/2021 10:07, Jerin Jacob:
> > On Fri, Jan 22, 2021 at 2:28 PM Thomas Monjalon 
> wrote:
> > > 22/01/2021 09:39, Juraj Linkeš:
> > > > > > > > disabled drivers, similarly how the command line option
> > > > > > > > works and remove unneeded driver options ported from the
> > > > > > > > old makefile system, since they don't work in the current Meson
> build system.
> > > > > > > > Add support for removing drivers for cross builds so that
> > > > > > > > we can disable them in cross files.
> > > > > > >
> > > > > > > Why disabling them?
> > > > > > > If a driver is not supported it should disable itseld in its 
> > > > > > > meson file.
> > > > > > >
> > > > > >
> > > > > > This is helpful when building for an SoC where we don't want
> > > > > > to build to build a driver, but the build machine actually supports
> the driver.
> > > > > > I believe in this case the meson build system would find the
> > > > > > dependencies and designate the driver to be build, but we
> > > > > > don't want to build
> > > > > the driver for that SoC.
> > > > > >
> > > > > > There may be other reasons as well - Honnappa or others from
> > > > > > the Arm community may shed more light on this.
> > > > > IMO, the assumption should be everything compiles on all the
> > > > > platforms. Hence, the disables should be applied to the
> > > > > platforms where the drivers do not compile.
> > >
> > > If a driver does not compile, it can disable itself.
> > > No need for a configuration.
> > >
> > > > Would it be okay to leave the disabled as they're in this commit and
> leave the updates to the plaform owners? Thomas, what do you think?
> > >
> > > I think this patch should not disable drivers but just add the infra to 
> > > do it.
> >
> > IMO, If the SOC has "fixed" set of dpdk devices, probably better to
> > have positive logic to enable only those in config file.
> > I think, that will be portable and useful.
> > IMO, We can have infrastructure code enable only specific drivers and
> > config owners can later enable the required set.
> 
> Yes you're right, enabling makes more sense than disabling for SoCs.
Every SoC also supports PCIe interfaces. This means, one could use them with a 
PCIe based NIC (we do use these interfaces internally at Arm, I am not sure 
from a deployment perspective).

If we use the enable logic, the list will be huge?

> 

Re: [dpdk-dev] [PATCH v2 41/44] net/virtio: move Vhost-kernel data to its backend

[Maxime Coquelin]

On 1/22/21 9:55 AM, Xia, Chenbo wrote:
> Hi Maxime,
> 
>> -Original Message-
>> From: Maxime Coquelin 
>> Sent: Wednesday, January 20, 2021 5:25 AM
>> To: dev@dpdk.org; Xia, Chenbo ; olivier.m...@6wind.com;
>> amore...@redhat.com; david.march...@redhat.com
>> Cc: Maxime Coquelin 
>> Subject: [PATCH v2 41/44] net/virtio: move Vhost-kernel data to its backend
>>
>> As done earlier for Vhost-user, this patch moves the
>> Vhost-Kernel specific data to its backend file.
>>
>> Signed-off-by: Maxime Coquelin 
>> ---
>>  drivers/net/virtio/virtio_user/vhost_kernel.c | 106 +++---
>>  .../net/virtio/virtio_user/virtio_user_dev.c  |  43 ++-
>>  .../net/virtio/virtio_user/virtio_user_dev.h  |   7 +-
>>  3 files changed, 98 insertions(+), 58 deletions(-)
>>
>> diff --git a/drivers/net/virtio/virtio_user/vhost_kernel.c
>> b/drivers/net/virtio/virtio_user/vhost_kernel.c
>> index aa1f9ece5e..bc4b3461e1 100644
>> --- a/drivers/net/virtio/virtio_user/vhost_kernel.c
>> +++ b/drivers/net/virtio/virtio_user/vhost_kernel.c
>> @@ -14,6 +14,11 @@
>>  #include "virtio_user_dev.h"
>>  #include "vhost_kernel_tap.h"
>>
>> +struct vhost_kernel_data {
>> +int *vhostfds;
>> +int *tapfds;
>> +};
>> +
>>  struct vhost_memory_kernel {
>>  uint32_t nregions;
>>  uint32_t padding;
>> @@ -96,7 +101,9 @@ vhost_kernel_ioctl(int fd, uint64_t request, void *arg)
>>  static int
>>  vhost_kernel_set_owner(struct virtio_user_dev *dev)
>>  {
>> -return vhost_kernel_ioctl(dev->vhostfds[0], VHOST_SET_OWNER, NULL);
>> +struct vhost_kernel_data *data = dev->backend_data;
>> +
>> +return vhost_kernel_ioctl(data->vhostfds[0], VHOST_SET_OWNER, NULL);
>>  }
>>
>>  static int
>> @@ -104,8 +111,9 @@ vhost_kernel_get_features(struct virtio_user_dev *dev,
>> uint64_t *features)
>>  {
>>  int ret;
>>  unsigned int tap_features;
>> +struct vhost_kernel_data *data = dev->backend_data;
>>
>> -ret = vhost_kernel_ioctl(dev->vhostfds[0], VHOST_GET_FEATURES, 
>> features);
>> +ret = vhost_kernel_ioctl(data->vhostfds[0], VHOST_GET_FEATURES,
>> features);
>>  if (ret < 0) {
>>  PMD_DRV_LOG(ERR, "Failed to get features");
>>  return -1;
>> @@ -138,6 +146,8 @@ vhost_kernel_get_features(struct virtio_user_dev *dev,
>> uint64_t *features)
>>  static int
>>  vhost_kernel_set_features(struct virtio_user_dev *dev, uint64_t features)
>>  {
>> +struct vhost_kernel_data *data = dev->backend_data;
>> +
>>  /* We don't need memory protection here */
>>  features &= ~(1ULL << VIRTIO_F_IOMMU_PLATFORM);
>>  /* VHOST kernel does not know about below flags */
>> @@ -145,7 +155,7 @@ vhost_kernel_set_features(struct virtio_user_dev *dev,
>> uint64_t features)
>>  features &= ~VHOST_KERNEL_HOST_OFFLOADS_MASK;
>>  features &= ~(1ULL << VIRTIO_NET_F_MQ);
>>
>> -return vhost_kernel_ioctl(dev->vhostfds[0], VHOST_SET_FEATURES,
>> &features);
>> +return vhost_kernel_ioctl(data->vhostfds[0], VHOST_SET_FEATURES,
>> &features);
>>  }
>>
>>  static int
>> @@ -185,6 +195,7 @@ add_memseg_list(const struct rte_memseg_list *msl, void
>> *arg)
>>  static int
>>  vhost_kernel_set_memory_table(struct virtio_user_dev *dev)
>>  {
>> +struct vhost_kernel_data *data = dev->backend_data;
>>  struct vhost_memory_kernel *vm;
>>  int ret;
>>
>> @@ -205,7 +216,7 @@ vhost_kernel_set_memory_table(struct virtio_user_dev 
>> *dev)
>>  if (ret < 0)
>>  goto err_free;
>>
>> -ret = vhost_kernel_ioctl(dev->vhostfds[0], VHOST_SET_MEM_TABLE, vm);
>> +ret = vhost_kernel_ioctl(data->vhostfds[0], VHOST_SET_MEM_TABLE, vm);
>>  if (ret < 0)
>>  goto err_free;
>>
>> @@ -224,9 +235,10 @@ vhost_kernel_set_vring(struct virtio_user_dev *dev,
>> uint64_t req, struct vhost_v
>>  {
>>  int ret, fd;
>>  unsigned int index = state->index;
>> +struct vhost_kernel_data *data = dev->backend_data;
>>
>>  /* Convert from queue index to queue-pair & offset */
>> -fd = dev->vhostfds[state->index / 2];
>> +fd = data->vhostfds[state->index / 2];
>>  state->index %= 2;
>>
>>  ret = vhost_kernel_ioctl(fd, req, state);
>> @@ -265,9 +277,10 @@ vhost_kernel_set_vring_file(struct virtio_user_dev *dev,
>> uint64_t req,
>>  {
>>  int ret, fd;
>>  unsigned int index = file->index;
>> +struct vhost_kernel_data *data = dev->backend_data;
>>
>>  /* Convert from queue index to queue-pair & offset */
>> -fd = dev->vhostfds[file->index / 2];
>> +fd = data->vhostfds[file->index / 2];
>>  file->index %= 2;
>>
>>  ret = vhost_kernel_ioctl(fd, req, file);
>> @@ -299,9 +312,10 @@ vhost_kernel_set_vring_addr(struct virtio_user_dev *dev,
>> struct vhost_vring_addr
>>  {
>>  int ret, fd;
>>  unsigned int index = addr->index;
>> +struct vhost_kernel_data *data = dev->backend_data;
>>
>>  /* Convert from queue index to queue-pair & offset */
>> -fd = dev->vhostfds[addr->index / 2];
>> +fd = data->vhostfds

Re: [dpdk-dev] [PATCH v2 42/44] net/virtio: move Vhost-vDPA data to its backend

[Maxime Coquelin]

On 1/22/21 10:06 AM, Xia, Chenbo wrote:
> Hi Maxime,
> 
>> -Original Message-
>> From: Maxime Coquelin 
>> Sent: Wednesday, January 20, 2021 5:25 AM
>> To: dev@dpdk.org; Xia, Chenbo ; olivier.m...@6wind.com;
>> amore...@redhat.com; david.march...@redhat.com
>> Cc: Maxime Coquelin 
>> Subject: [PATCH v2 42/44] net/virtio: move Vhost-vDPA data to its backend
>>
>> As done earlier for Vhost-user and Vhost-kernel, this
>> patch moves the Vhost-vDPA specific data to its backend
>> file.
>>
>> Signed-off-by: Maxime Coquelin 
>> ---
>>  drivers/net/virtio/virtio_user/vhost_vdpa.c   | 120 +-
>>  .../net/virtio/virtio_user/virtio_user_dev.h  |   4 -
>>  2 files changed, 89 insertions(+), 35 deletions(-)
>>
...
>>
>> diff --git a/drivers/net/virtio/virtio_user/virtio_user_dev.h
>> b/drivers/net/virtio/virtio_user/virtio_user_dev.h
>> index 36a9cadcad..9a6723da24 100644
>> --- a/drivers/net/virtio/virtio_user/virtio_user_dev.h
>> +++ b/drivers/net/virtio/virtio_user/virtio_user_dev.h
>> @@ -29,9 +29,6 @@ struct virtio_user_dev {
>>  enum virtio_user_backend_type backend_type;
>>  boolis_server;  /* server or client mode */
>>
>> -/* for vhost_vdpa backend */
>> -int vhostfd;
>> -
>>  /* for both vhost_user and vhost_kernel */
> 
> I remember you agreed that this comment is no longer needed?

Indeed, I missed to remove it. Done now.

> I also noticed that there're multiple calls of PMD_DRV_LOG that add
> '\n' in the end, which is necessary. But, let's clean it up later..

Yes, there may still be redundant new lines at some places, I agree we
fix that later.

> With above fixed:
> 
> Reviewed-by: Chenbo Xia 
> 


Thanks,
Maxime

Re: [dpdk-dev] [PATCH v2 43/44] net/virtio: improve Vhost-user error logging

[Maxime Coquelin]

On 1/22/21 10:11 AM, Xia, Chenbo wrote:
> Hi Maxime,
> 
>> -Original Message-
>> From: Maxime Coquelin 
>> Sent: Wednesday, January 20, 2021 5:25 AM
>> To: dev@dpdk.org; Xia, Chenbo ; olivier.m...@6wind.com;
>> amore...@redhat.com; david.march...@redhat.com
>> Cc: Maxime Coquelin 
>> Subject: [PATCH v2 43/44] net/virtio: improve Vhost-user error logging
>>
>> This patch improves error logging in vhost_user_read,
>> especially printing errno when recv() fails.
>>
>> Suggested-by: Adrian Moreno 
>> Signed-off-by: Maxime Coquelin 
>> ---
>>  drivers/net/virtio/virtio_user/vhost_user.c | 29 -
>>  1 file changed, 17 insertions(+), 12 deletions(-)
>>
>> diff --git a/drivers/net/virtio/virtio_user/vhost_user.c
>> b/drivers/net/virtio/virtio_user/vhost_user.c
>> index f046655af6..be91c99cea 100644
>> --- a/drivers/net/virtio/virtio_user/vhost_user.c
>> +++ b/drivers/net/virtio/virtio_user/vhost_user.c
>> @@ -148,38 +148,43 @@ vhost_user_read(int fd, struct vhost_user_msg *msg)
>>  int ret, sz_hdr = VHOST_USER_HDR_SIZE, sz_payload;
>>
>>  ret = recv(fd, (void *)msg, sz_hdr, 0);
>> -if (ret < sz_hdr) {
>> +if (ret < 0) {
>> +PMD_DRV_LOG(ERR, "Failed to recv msg header: %s", 
>> strerror(errno));
>> +return -1;
>> +} else if (ret < sz_hdr) {
>>  PMD_DRV_LOG(ERR, "Failed to recv msg hdr: %d instead of %d.",
>>  ret, sz_hdr);
>> -goto fail;
>> +return -1;
>>  }
>>
>>  /* validate msg flags */
>>  if (msg->flags != (valid_flags)) {
>>  PMD_DRV_LOG(ERR, "Failed to recv msg: flags %x instead of %x.",
>>  msg->flags, valid_flags);
>> -goto fail;
>> +return -1;
> 
> Since you are here, also add '0x' before '%x' here?

Done.

> Thanks,
> Chenbo

Thanks,
Maxime

[dpdk-dev] [PATCH v2] ci: catch coredumps

[David Marchand]

Parts of the unit tests code rely on forked/secondary processes
(expectedly) failing.
A crash in those situations could be missed so add a check on coredumps
presence after unit tests have run.
When unit tests fail, it can also help checking for coredumps as it
could give more insights on what happened.

In some situations (like explicit call to rte_panic), coredump generation
must be disabled to avoid false positives.

Signed-off-by: David Marchand 
---
Changelog since RFC v1
- removed RFC,
- pushed coredumps into gdb for in-situ analysis,
- gdb presence is used to enable the check. Travis config is left
  untouched for now,

---
 .ci/linux-build.sh  | 37 ++---
 .github/workflows/build.yml |  4 
 app/test/test_debug.c   | 11 +--
 app/test/test_mbuf.c|  9 -
 4 files changed, 55 insertions(+), 6 deletions(-)

diff --git a/.ci/linux-build.sh b/.ci/linux-build.sh
index d2c821adf3..26c30a2301 100755
--- a/.ci/linux-build.sh
+++ b/.ci/linux-build.sh
@@ -4,7 +4,10 @@ on_error() {
 if [ $? = 0 ]; then
 exit
 fi
-FILES_TO_PRINT="build/meson-logs/testlog.txt build/.ninja_log 
build/meson-logs/meson-log.txt"
+FILES_TO_PRINT="build/meson-logs/testlog.txt"
+FILES_TO_PRINT="$FILES_TO_PRINT build/.ninja_log"
+FILES_TO_PRINT="$FILES_TO_PRINT build/meson-logs/meson-log.txt"
+FILES_TO_PRINT="$FILES_TO_PRINT build/gdb.log"
 
 for pr_file in $FILES_TO_PRINT; do
 if [ -e "$pr_file" ]; then
@@ -30,6 +33,26 @@ install_libabigail() {
 rm ${version}.tar.gz
 }
 
+configure_coredump() {
+# No point in configuring coredump without gdb
+which gdb >/dev/null || return 0
+ulimit -c unlimited
+sudo sysctl -w kernel.core_pattern=/tmp/dpdk-core.%e.%p
+}
+
+catch_coredump() {
+ls /tmp/dpdk-core.*.* 2>/dev/null || return 0
+for core in /tmp/dpdk-core.*.*; do
+binary=$(sudo readelf -n $core |grep $(pwd)/build/ 2>/dev/null |head 
-n1)
+[ -x $binary ] || binary=
+sudo gdb $binary -c $core \
+-ex 'info threads' \
+-ex 'thread apply all bt full' \
+-ex 'quit'
+done |tee -a build/gdb.log
+return 1
+}
+
 if [ "$AARCH64" = "true" ]; then
 # convert the arch specifier
 OPTS="$OPTS --cross-file config/arm/arm64_armv8_linux_gcc"
@@ -57,7 +80,11 @@ meson build --werror $OPTS
 ninja -C build
 
 if [ "$AARCH64" != "true" ]; then
-devtools/test-null.sh
+failed=
+configure_coredump
+devtools/test-null.sh || failed="true"
+catch_coredump
+[ "$failed" != "true" ]
 fi
 
 if [ "$ABI_CHECKS" = "true" ]; then
@@ -102,5 +129,9 @@ if [ "$ABI_CHECKS" = "true" ]; then
 fi
 
 if [ "$RUN_TESTS" = "true" ]; then
-sudo meson test -C build --suite fast-tests -t 3
+failed=
+configure_coredump
+sudo meson test -C build --suite fast-tests -t 3 || failed="true"
+catch_coredump
+[ "$failed" != "true" ]
 fi
diff --git a/.github/workflows/build.yml b/.github/workflows/build.yml
index a5b579adda..786525e5a3 100644
--- a/.github/workflows/build.yml
+++ b/.github/workflows/build.yml
@@ -104,6 +104,9 @@ jobs:
   if: env.AARCH64 == 'true'
   run: sudo apt install -y gcc-aarch64-linux-gnu libc6-dev-arm64-cross
 pkg-config-aarch64-linux-gnu
+- name: Install test tools packages
+  if: env.AARCH64 != 'true' || env.RUN_TESTS == 'true'
+  run: sudo apt install -y gdb
 - name: Install doc generation packages
   if: env.BUILD_DOCS == 'true'
   run: sudo apt install -y doxygen graphviz python3-sphinx
@@ -124,3 +127,4 @@ jobs:
   build/meson-logs/testlog.txt
   build/.ninja_log
   build/meson-logs/meson-log.txt
+  build/gdb.log
diff --git a/app/test/test_debug.c b/app/test/test_debug.c
index 834a7386f5..23b24db177 100644
--- a/app/test/test_debug.c
+++ b/app/test/test_debug.c
@@ -4,6 +4,8 @@
 
 #include 
 #include 
+#include 
+#include 
 #include 
 #include 
 
@@ -28,9 +30,14 @@ test_panic(void)
 
pid = fork();
 
-   if (pid == 0)
+   if (pid == 0) {
+   struct rlimit rl;
+
+   /* No need to generate a coredump when panicking. */
+   rl.rlim_cur = rl.rlim_max = 0;
+   setrlimit(RLIMIT_CORE, &rl);
rte_panic("Test Debug\n");
-   else if (pid < 0){
+   } else if (pid < 0) {
printf("Fork Failed\n");
return -1;
}
diff --git a/app/test/test_mbuf.c b/app/test/test_mbuf.c
index a40f7d4883..47a7b197d7 100644
--- a/app/test/test_mbuf.c
+++ b/app/test/test_mbuf.c
@@ -1174,6 +1174,8 @@ test_refcnt_mbuf(void)
 }
 
 #include 
+#include 
+#include 
 #include 
 
 /* use fork() to test mbuf errors panic */
@@ -1186,9 +1188,14 @@ verify_mbuf_check_panics(struct rte_mbuf *buf)
pid = fork();
 
if (pid == 0) {
+   struct rlimit rl;
+
+   /* No need to generate a coredump when panicking. */
+   rl.rl

[dpdk-dev] [PATCH] net/nfp: read chip model from PluDevice register

[Heinrich Kuhn]

For newer smartNIC NVRAM versions the chip model should be read from the
PluDevice register as it provides the authoritative chip model/revision.
This method of reading the chip model is backwards compatible with
legacy NVRAM versions too.

Since the model number is purely used for reporting purposes, follow the
hardware team convention of subtracting 0x10 from the PluDevice register
to obtain the chip model/revision number.

Fixes: c7e9729da ("net/nfp: support CPP")
Cc: sta...@dpdk.org

Signed-off-by: Heinrich Kuhn 
Reviewed-by: Louis Peens 
Signed-off-by: Simon Horman 
---
 drivers/net/nfp/nfpcore/nfp_cpp.h |  2 +-
 drivers/net/nfp/nfpcore/nfp_cppcore.c | 49 ---
 2 files changed, 23 insertions(+), 28 deletions(-)

diff --git a/drivers/net/nfp/nfpcore/nfp_cpp.h 
b/drivers/net/nfp/nfpcore/nfp_cpp.h
index 1427954c1..08d656da1 100644
--- a/drivers/net/nfp/nfpcore/nfp_cpp.h
+++ b/drivers/net/nfp/nfpcore/nfp_cpp.h
@@ -170,7 +170,7 @@ void *nfp_cpp_priv(struct nfp_cpp *cpp);
  */
 void *nfp_cpp_area_priv(struct nfp_cpp_area *cpp_area);
 
-uint32_t __nfp_cpp_model_autodetect(struct nfp_cpp *cpp);
+uint32_t __nfp_cpp_model_autodetect(struct nfp_cpp *cpp, uint32_t *model);
 
 /*
  * NFP CPP core interface for CPP clients.
diff --git a/drivers/net/nfp/nfpcore/nfp_cppcore.c 
b/drivers/net/nfp/nfpcore/nfp_cppcore.c
index dec4a8b6d..6d629430d 100644
--- a/drivers/net/nfp/nfpcore/nfp_cppcore.c
+++ b/drivers/net/nfp/nfpcore/nfp_cppcore.c
@@ -22,8 +22,9 @@
 
 #define NFP_PL_DEVICE_ID0x0004
 #define NFP_PL_DEVICE_ID_MASK   0xff
-
-#define NFP6000_ARM_GCSR_SOFTMODEL0 0x00400144
+#define NFP_PL_DEVICE_PART_MASK 0x
+#define NFP_PL_DEVICE_MODEL_MASK   (NFP_PL_DEVICE_PART_MASK | \
+   NFP_PL_DEVICE_ID_MASK)
 
 void
 nfp_cpp_priv_set(struct nfp_cpp *cpp, void *priv)
@@ -46,13 +47,18 @@ nfp_cpp_model_set(struct nfp_cpp *cpp, uint32_t model)
 uint32_t
 nfp_cpp_model(struct nfp_cpp *cpp)
 {
+   int err;
+   uint32_t model;
+
if (!cpp)
return NFP_CPP_MODEL_INVALID;
 
-   if (cpp->model == 0)
-   cpp->model = __nfp_cpp_model_autodetect(cpp);
+   err = __nfp_cpp_model_autodetect(cpp, &model);
 
-   return cpp->model;
+   if (err < 0)
+   return err;
+
+   return model;
 }
 
 void
@@ -389,9 +395,6 @@ nfp_xpb_to_cpp(struct nfp_cpp *cpp, uint32_t *xpb_addr)
uint32_t xpb;
int island;
 
-   if (!NFP_CPP_MODEL_IS_6000(cpp->model))
-   return 0;
-
xpb = NFP_CPP_ID(14, NFP_CPP_ACTION_RW, 0);
 
/*
@@ -796,29 +799,21 @@ nfp_cpp_area_fill(struct nfp_cpp_area *area, unsigned 
long offset,
  * as those are model-specific
  */
 uint32_t
-__nfp_cpp_model_autodetect(struct nfp_cpp *cpp)
+__nfp_cpp_model_autodetect(struct nfp_cpp *cpp, uint32_t *model)
 {
-   uint32_t arm_id = NFP_CPP_ID(NFP_CPP_TARGET_ARM, 0, 0);
-   uint32_t model = 0;
-
-   if (nfp_cpp_readl(cpp, arm_id, NFP6000_ARM_GCSR_SOFTMODEL0, &model))
-   return 0;
-
-   if (NFP_CPP_MODEL_IS_6000(model)) {
-   uint32_t tmp;
-
-   nfp_cpp_model_set(cpp, model);
+   uint32_t reg;
+   int err;
 
-   /* The PL's PluDeviceID revision code is authoratative */
-   model &= ~0xff;
-   if (nfp_xpb_readl(cpp, NFP_XPB_DEVICE(1, 1, 16) +
-  NFP_PL_DEVICE_ID, &tmp))
-   return 0;
+   err = nfp_xpb_readl(cpp, NFP_XPB_DEVICE(1, 1, 16) + NFP_PL_DEVICE_ID,
+   ®);
+   if (err < 0)
+   return err;
 
-   model |= (NFP_PL_DEVICE_ID_MASK & tmp) - 0x10;
-   }
+   *model = reg & NFP_PL_DEVICE_MODEL_MASK;
+   if (*model & NFP_PL_DEVICE_ID_MASK)
+   *model -= 0x10;
 
-   return model;
+   return 0;
 }
 
 /*
-- 
2.30.0

Re: [dpdk-dev] [PATCH v15 09/12] build: disable drivers in Arm builds

[Thomas Monjalon]

25/01/2021 15:58, Honnappa Nagarahalli:
> > 22/01/2021 10:07, Jerin Jacob:
> > > On Fri, Jan 22, 2021 at 2:28 PM Thomas Monjalon 
> > wrote:
> > > > 22/01/2021 09:39, Juraj Linkeš:
> > > > > > > > > disabled drivers, similarly how the command line option
> > > > > > > > > works and remove unneeded driver options ported from the
> > > > > > > > > old makefile system, since they don't work in the current 
> > > > > > > > > Meson
> > build system.
> > > > > > > > > Add support for removing drivers for cross builds so that
> > > > > > > > > we can disable them in cross files.
> > > > > > > >
> > > > > > > > Why disabling them?
> > > > > > > > If a driver is not supported it should disable itseld in its 
> > > > > > > > meson file.
> > > > > > > >
> > > > > > >
> > > > > > > This is helpful when building for an SoC where we don't want
> > > > > > > to build to build a driver, but the build machine actually 
> > > > > > > supports
> > the driver.
> > > > > > > I believe in this case the meson build system would find the
> > > > > > > dependencies and designate the driver to be build, but we
> > > > > > > don't want to build
> > > > > > the driver for that SoC.
> > > > > > >
> > > > > > > There may be other reasons as well - Honnappa or others from
> > > > > > > the Arm community may shed more light on this.
> > > > > > IMO, the assumption should be everything compiles on all the
> > > > > > platforms. Hence, the disables should be applied to the
> > > > > > platforms where the drivers do not compile.
> > > >
> > > > If a driver does not compile, it can disable itself.
> > > > No need for a configuration.
> > > >
> > > > > Would it be okay to leave the disabled as they're in this commit and
> > leave the updates to the plaform owners? Thomas, what do you think?
> > > >
> > > > I think this patch should not disable drivers but just add the infra to 
> > > > do it.
> > >
> > > IMO, If the SOC has "fixed" set of dpdk devices, probably better to
> > > have positive logic to enable only those in config file.
> > > I think, that will be portable and useful.
> > > IMO, We can have infrastructure code enable only specific drivers and
> > > config owners can later enable the required set.
> > 
> > Yes you're right, enabling makes more sense than disabling for SoCs.
> Every SoC also supports PCIe interfaces. This means, one could use them with 
> a PCIe based NIC (we do use these interfaces internally at Arm, I am not sure 
> from a deployment perspective).
> 
> If we use the enable logic, the list will be huge?

That's also right.
The last to talk will be right :)

Re: [dpdk-dev] [PATCH v3 0/4] add checking of header includes

[David Marchand]

On Mon, Jan 25, 2021 at 3:11 PM Bruce Richardson
 wrote:
>
> As a general principle, each header file should include any other
> headers it needs to provide data type definitions or macros. For
> example, any header using the uintX_t types in structures or function
> prototypes should include "stdint.h" to provide those type definitions.
>
> In practice, while many, but not all, headers in DPDK did include all
> necessary headers, it was never actually checked that each header could
> be included in a C file and compiled without having any compiler errors
> about missing definitions.  The script "check-includes.sh" could be used
> for this job, but it was not called out in the documentation, so many
> contributors may not have been aware of it's existance. It also was
> difficult to run from a source-code directory, as the script did not
> automatically allow finding of headers from one DPDK library directory
> to another [this was probably based on running it on a build created by
> the "make" build system, where all headers were in a single directory].
> To attempt to have a build-system integrated replacement, this patchset
> adds a "chkincs" app in the buildtools directory to verify this on an
> ongoing basis.
>
> This chkincs app does nothing when run, and is not installed as part of
> a DPDK "ninja install", it's for build-time checking only. Its source
> code consists of one C file per public DPDK header, where that C file
> contains nothing except an include for that header.  Therefore, if any
> header is added to the lib folder which fails to compile when included
> alone, the build of chkincs will fail with a suitable error message.
> Since this compile checking is not needed on most builds of DPDK, the
> building of chkincs is disabled by default, but can be enabled by the
> "test_includes" meson option. To catch errors with patch submissions,
> the final patch of this series enables it for a single build in
> test-meson-builds script.
>
> Future work could involve doing similar checks on headers for C++
> compatibility, which was something done by the check-includes.sh script
> but which is missing here..
>
> V3:
> * Shrunk patchset as most header fixes already applied
> * Moved chkincs from "apps" to the "buildtools" directory, which is a
>   better location for something not for installation for end-user use.
> * Added patch to drop check-includes script.
>
> V2:
> * Add maintainers file entry for new app
> * Drop patch for c11 ring header
> * Use build variable "headers_no_chkincs" for tracking exceptions
>
> Bruce Richardson (4):
>   eal: add missing include to mcslock
>   build: separate out headers for include checking
>   buildtools/chkincs: add app to verify header includes
>   devtools: remove check-includes script
>
>  MAINTAINERS  |   5 +-
>  buildtools/chkincs/gen_c_file_for_header.py  |  12 +
>  buildtools/chkincs/main.c|   4 +
>  buildtools/chkincs/meson.build   |  40 +++
>  devtools/check-includes.sh   | 259 ---
>  devtools/test-meson-builds.sh|   2 +-
>  doc/guides/contributing/coding_style.rst |  12 +
>  lib/librte_eal/include/generic/rte_mcslock.h |   1 +
>  lib/librte_eal/include/meson.build   |   2 +-
>  lib/librte_eal/x86/include/meson.build   |  14 +-
>  lib/librte_ethdev/meson.build|   4 +-
>  lib/librte_hash/meson.build  |   4 +-
>  lib/librte_ipsec/meson.build |   3 +-
>  lib/librte_lpm/meson.build   |   2 +-
>  lib/librte_regexdev/meson.build  |   2 +-
>  lib/librte_ring/meson.build  |   4 +-
>  lib/librte_stack/meson.build |   4 +-
>  lib/librte_table/meson.build |   7 +-
>  lib/meson.build  |   3 +
>  meson.build  |   6 +
>  meson_options.txt|   2 +
>  21 files changed, 112 insertions(+), 280 deletions(-)
>  create mode 100755 buildtools/chkincs/gen_c_file_for_header.py
>  create mode 100644 buildtools/chkincs/main.c
>  create mode 100644 buildtools/chkincs/meson.build
>  delete mode 100755 devtools/check-includes.sh

- clang is not happy when enabling the check:
$ meson configure $HOME/builds/build-clang-static -Dcheck_includes=true
$ devtools/test-meson-builds.sh
...
[362/464] Compiling C object
buildtools/chkincs/chkincs.p/meson-generated_rte_ethdev_vdev.c.o
FAILED: buildtools/chkincs/chkincs.p/meson-generated_rte_ethdev_vdev.c.o
clang -Ibuildtools/chkincs/chkincs.p -Ibuildtools/chkincs
-I../../dpdk/buildtools/chkincs -Idrivers/bus/pci
-I../../dpdk/drivers/bus/pci -Idrivers/bus/vdev
-I../../dpdk/drivers/bus/vdev -I. -I../../dpdk -Iconfig
-I../../dpdk/config -Ilib/librte_eal/include
-I../../dpdk/lib/librte_eal/include -Ilib/librte_eal/linux/include
-I../../dpdk/lib/librte_eal/linux/include -Ilib/librte_eal/x86/include
-I../../dpdk/

Re: [dpdk-dev] [PATCH v2 3/4] buildtools: support object file extraction for Windows

[Thomas Monjalon]

08/01/2021 03:47, Dmitry Kozlyuk:
> --- /dev/null
> +++ b/buildtools/gen-pmdinfo-cfile.py
[...]
> +with tempfile.TemporaryDirectory() as temp:
> +proc = subprocess.run(
> +# Don't use "ar p", because its output is corrupted on Windows.
> +[ar, "xv", os.path.abspath(archive)], capture_output=True, 
> check=True, cwd=temp
> +)

Not sure where is this temp file.
Please can you make sure it created inside the build directory?
It can be a follow-up patch, thanks.

Re: [dpdk-dev] [PATCH v2 0/4] pmdinfogen: support Windows

[Thomas Monjalon]

08/01/2021 03:47, Dmitry Kozlyuk:
> Dmitry Kozlyuk (4):
>   pmdinfogen: support COFF
>   pmdinfogen: allow multiple input files
>   buildtools: support object file extraction for Windows
>   build: enable pmdinfogen for Windows

Applied, thanks.

Please follow-up on temporary file location.

Re: [dpdk-dev] [PATCH v10 0/3] pmdinfogen: rewrite in Python

[Brandon Lo]

Hi Thomas,

We have updated all UNH-IOL environments to include pyelftools.

Thanks,
Brandon

On Mon, Jan 25, 2021 at 8:13 AM Thomas Monjalon  wrote:
>
> > Dmitry Kozlyuk (3):
> >   pmdinfogen: add Python implementation
> >   build: use Python pmdinfogen
> >   pmdinfogen: remove C implementation
>
> Applied, thanks.
>
> Reminder: it is adding a NEW DEPENDENCY on pyelftools,
> as highlighted already in CI and maintainers meetings:
> http://mails.dpdk.org/archives/ci/2021-January/000939.html
> http://mails.dpdk.org/archives/dev/2021-January/197814.html
>
> If not already done, please UPDATE your environment! Thanks
>
>


-- 

Brandon Lo

UNH InterOperability Laboratory

21 Madbury Rd, Suite 100, Durham, NH 03824

b...@iol.unh.edu

www.iol.unh.edu

Re: [dpdk-dev] [PATCH v15 09/12] build: disable drivers in Arm builds

[Jerin Jacob]

On Mon, Jan 25, 2021 at 8:28 PM Honnappa Nagarahalli
 wrote:
>
> 
>
> >
> > 22/01/2021 10:07, Jerin Jacob:
> > > On Fri, Jan 22, 2021 at 2:28 PM Thomas Monjalon 
> > wrote:
> > > > 22/01/2021 09:39, Juraj Linkeš:
> > > > > > > > > disabled drivers, similarly how the command line option
> > > > > > > > > works and remove unneeded driver options ported from the
> > > > > > > > > old makefile system, since they don't work in the current 
> > > > > > > > > Meson
> > build system.
> > > > > > > > > Add support for removing drivers for cross builds so that
> > > > > > > > > we can disable them in cross files.
> > > > > > > >
> > > > > > > > Why disabling them?
> > > > > > > > If a driver is not supported it should disable itseld in its 
> > > > > > > > meson file.
> > > > > > > >
> > > > > > >
> > > > > > > This is helpful when building for an SoC where we don't want
> > > > > > > to build to build a driver, but the build machine actually 
> > > > > > > supports
> > the driver.
> > > > > > > I believe in this case the meson build system would find the
> > > > > > > dependencies and designate the driver to be build, but we
> > > > > > > don't want to build
> > > > > > the driver for that SoC.
> > > > > > >
> > > > > > > There may be other reasons as well - Honnappa or others from
> > > > > > > the Arm community may shed more light on this.
> > > > > > IMO, the assumption should be everything compiles on all the
> > > > > > platforms. Hence, the disables should be applied to the
> > > > > > platforms where the drivers do not compile.
> > > >
> > > > If a driver does not compile, it can disable itself.
> > > > No need for a configuration.
> > > >
> > > > > Would it be okay to leave the disabled as they're in this commit and
> > leave the updates to the plaform owners? Thomas, what do you think?
> > > >
> > > > I think this patch should not disable drivers but just add the infra to 
> > > > do it.
> > >
> > > IMO, If the SOC has "fixed" set of dpdk devices, probably better to
> > > have positive logic to enable only those in config file.
> > > I think, that will be portable and useful.
> > > IMO, We can have infrastructure code enable only specific drivers and
> > > config owners can later enable the required set.
> >
> > Yes you're right, enabling makes more sense than disabling for SoCs.
> Every SoC also supports PCIe interfaces. This means, one could use them with 
> a PCIe based NIC (we do use these interfaces internally at Arm, I am not sure 
> from a deployment perspective).

Not every SoC.

>
> If we use the enable logic, the list will be huge?

Yes. I think both support will be useful. IMO, Selecting a small
subset can be achieved via, have a flag in config file to disable all
the drivers and add the selective one as needed.

>
> >
>

Re: [dpdk-dev] [PATCH v2 44/44] net/virtio: handle Virtio-user setup failure properly

[Maxime Coquelin]

On 1/22/21 10:24 AM, Xia, Chenbo wrote:
> Hi Maxime,
> 
>> -Original Message-
>> From: Maxime Coquelin 
>> Sent: Wednesday, January 20, 2021 5:25 AM
>> To: dev@dpdk.org; Xia, Chenbo ; olivier.m...@6wind.com;
>> amore...@redhat.com; david.march...@redhat.com
>> Cc: Maxime Coquelin 
>> Subject: [PATCH v2 44/44] net/virtio: handle Virtio-user setup failure
>> properly
>>
>> This patch fixes virtio_user_dev_setup() error path,
>> by cleaning all resources it allocates.
>>
>> Suggested-by: Adrian Moreno 
>> Signed-off-by: Maxime Coquelin 
>> ---
>>  drivers/net/virtio/virtio_user/virtio_user_dev.c | 11 +--
>>  1 file changed, 9 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/net/virtio/virtio_user/virtio_user_dev.c
>> b/drivers/net/virtio/virtio_user/virtio_user_dev.c
>> index a1e32158bb..2f7dc574b6 100644
>> --- a/drivers/net/virtio/virtio_user/virtio_user_dev.c
>> +++ b/drivers/net/virtio/virtio_user/virtio_user_dev.c
>> @@ -427,15 +427,22 @@ virtio_user_dev_setup(struct virtio_user_dev *dev)
>>
>>  if (virtio_user_dev_init_notify(dev) < 0) {
>>  PMD_INIT_LOG(ERR, "(%s) Failed to init notifiers\n", dev->path);
>> -return -1;
>> +goto destroy;
>>  }
>>
>>  if (virtio_user_fill_intr_handle(dev) < 0) {
>>  PMD_INIT_LOG(ERR, "(%s) Failed to init interrupt handler\n", 
>> dev-
>>> path);
>> -return -1;
>> +goto uninit;
>>  }
>>
>>  return 0;
>> +
>> +uninit:
>> +virtio_user_dev_uninit(dev);
> 
> IMHO, it may not be the best choice to call virtio_user_dev_uninit here..
> 
> Logically when virtio_user_fill_intr_handle fails, we want to release the 
> resources
> which is allocated in virtio_user_dev_init_notify (i.e., fds), but 
> virtio_user_dev_uninit
> does more. For example, unregister the event callback that have not been 
> registered yet and
> it also leads to destroy callback called twice. Although things done but not 
> needed will
> not lead to error, but it looks not in the best way. What do you think?

I agree, I'm reworking it for v3.

Kick and call FDs will be initialized to -1 at virtio_user_dev_init()
time. I introduce a virtio_user_dev_uninit_notify that will close all
kick and call FDs is opened and reset their value to -1.

Thenb I call this function in the error path of virtio_user_dev_setup()
and also in virtio_user_dev_uninit().

Doing that, I can also simplifies virtio_user_dev_init_notify() and only
loop for max queue pairs instead of VIRTIO_MAX_VIRTQUEUES and so avoid
setting FDs to -1 a second time.

Thanks,
Maxime

> Thanks,
> Chenbo
> 
>> +destroy:
>> +dev->ops->destroy(dev);
>> +
>> +return -1;
>>  }
>>
>>  /* Use below macro to filter features from vhost backend */
>> --
>> 2.29.2
> 

Re: [dpdk-dev] [PATCH v2 1/3] cryptodev: add feature flag for SSL/TLS order

[Akhil Goyal]

Hi Tejasree,

> Subject: [PATCH v2 1/3] cryptodev: add feature flag for SSL/TLS order
> 
> Added device feature flag for SSL/TLS way of cipher-auth
> operations support i.e. auth generation followed by encryption
> and decryption followed by auth verify so that SSL/TLS tests
> are skipped if the device doesn't support the feature.
> 
> Signed-off-by: Tejasree Kondoj 
> ---

Is it something different than 
#define RTE_CRYPTODEV_FF_DIGEST_ENCRYPTED   (1ULL << 19)
/**< Support encrypted-digest operations where digest is appended to data */

@Fan: could you review this patchset.

>  doc/guides/cryptodevs/features/default.ini | 1 +
>  doc/guides/rel_notes/release_21_02.rst | 7 +++
>  lib/librte_cryptodev/rte_cryptodev.h   | 2 ++
>  3 files changed, 10 insertions(+)
> 
> diff --git a/doc/guides/cryptodevs/features/default.ini
> b/doc/guides/cryptodevs/features/default.ini
> index 17b177fc45..cd5f75d8f6 100644
> --- a/doc/guides/cryptodevs/features/default.ini
> +++ b/doc/guides/cryptodevs/features/default.ini
> @@ -31,6 +31,7 @@ CPU crypto =
>  Symmetric sessionless  =
>  Non-Byte aligned data  =
>  Sym raw data path API  =
> +SSL way of chaining=
> 
>  ;
>  ; Supported crypto algorithms of a default crypto driver.
> diff --git a/doc/guides/rel_notes/release_21_02.rst
> b/doc/guides/rel_notes/release_21_02.rst
> index e72a582b1b..45854d5d33 100644
> --- a/doc/guides/rel_notes/release_21_02.rst
> +++ b/doc/guides/rel_notes/release_21_02.rst
> @@ -82,6 +82,13 @@ New Features
>enable applications to add/remove user callbacks which gets called
>for every enqueue/dequeue operation.
> 
> +* **Added feature flag in cryptodev library.**
> +
> +  Added device feature flag for SSL/TLS way of cipher-auth operations support
> +  i.e. auth generation followed by encryption and decryption followed by
> +  auth verify in cryptodev library so that SSL/TLS tests are skipped if the
> +  device doesn't support the feature.
> +
>  * **Updated the OCTEON TX2 crypto PMD.**
> 
>* Updated the OCTEON TX2 crypto PMD lookaside protocol offload for IPsec
> with
> diff --git a/lib/librte_cryptodev/rte_cryptodev.h
> b/lib/librte_cryptodev/rte_cryptodev.h
> index ae34f33f69..6ee3f33c06 100644
> --- a/lib/librte_cryptodev/rte_cryptodev.h
> +++ b/lib/librte_cryptodev/rte_cryptodev.h
> @@ -461,6 +461,8 @@ rte_cryptodev_asym_get_xform_enum(enum
> rte_crypto_asym_xform_type *xform_enum,
>  /**< Support operations on data which is not byte aligned */
>  #define RTE_CRYPTODEV_FF_SYM_RAW_DP  (1ULL << 24)
>  /**< Support accelerator specific symmetric raw data-path APIs */
> +#define RTE_CRYPTODEV_FF_SSL_ORDER   (1ULL << 25)
> +/**< Support SSL order of cipher-auth xforms chaining */
> 
>  /**
>   * Get the name of a crypto device feature flag
> --
> 2.27.0

Re: [dpdk-dev] [PATCH] build: force pkg-config for dependency detection

[Martin Špinler]

libnfb and libsze2 works OK.
The libnfb is provided by netcope-common package, so it is possible to
leave it like this.

Tested-by: Martin Spinler 

Re: [dpdk-dev] [PATCH 0/2] support SSL/TLS way of cipher-auth operations

[Akhil Goyal]

Hi Tejasree,

Sorry I missed this mail earlier. I believe it is same as Digest encrypted case.

> -Original Message-
> From: Tejasree Kondoj 
> Sent: Thursday, January 21, 2021 12:57 PM
> To: Tejasree Kondoj ; Akhil Goyal
> ; Radu Nicolau 
> Cc: Anoob Joseph ; Ankur Dwivedi
> ; dev@dpdk.org; Zhang, Roy Fan
> ; Trahe, Fiona 
> Subject: RE: [PATCH 0/2] support SSL/TLS way of cipher-auth operations
> 
> Hi Akhil,
> 
> I'm planning to add device feature flag(RTE_CRYPTODEV_FF_SSL_ORDER)
> along with SSL test cases movement to chain array in v2. Would that be fine?
> Feature flag is needed to skip SSL test cases if device doesn't support them.
> 
> Thanks
> Tejasree
> 
> > -Original Message-
> > From: Tejasree Kondoj 
> > Sent: Friday, December 18, 2020 7:40 PM
> > To: Akhil Goyal ; Radu Nicolau
> > 
> > Cc: Tejasree Kondoj ; Anoob Joseph
> > ; Ankur Dwivedi ;
> > dev@dpdk.org
> > Subject: [PATCH 0/2] support SSL/TLS way of cipher-auth operations
> >
> > This series adds SSL/TLS way of cipher-auth operations support i.e. auth
> > generation followed by encryption and decryption followed by auth verify in
> > OCTEON TX, OCTEON TX2 PMDs and sample unit test application.
> >
> > Tejasree Kondoj (2):
> >   common/cpt: support SSL/TLS way of cipher-auth operations
> >   test/crypto: support SSL/TLS way of cipher-auth operations
> >
> >  app/test/test_cryptodev_aes_test_vectors.h| 576 ++
> >  app/test/test_cryptodev_blockcipher.c |  98 ++-
> >  app/test/test_cryptodev_blockcipher.h |   9 +
> >  doc/guides/rel_notes/release_21_02.rst|  12 +
> >  drivers/common/cpt/cpt_mcode_defines.h|   7 +-
> >  drivers/common/cpt/cpt_ucode.h|  42 +-
> >  drivers/crypto/octeontx/otx_cryptodev_ops.c   |   8 +-
> >  drivers/crypto/octeontx2/otx2_cryptodev_ops.c |   8 +-
> >  8 files changed, 736 insertions(+), 24 deletions(-)
> >
> > --
> > 2.27.0

Re: [dpdk-dev] [dpdklab] RE: [dpdk-stable] [PATCH v4] mbuf: fix reset on mbuf free

[Brandon Lo]

Hi,

This seems like a good task for us to do. I will see what it would
take in order to convert the difference into a decimal-formatted
percentage.
I have put this into bugzilla to keep track of this issue:
https://bugs.dpdk.org/show_bug.cgi?id=626

Thanks,
Brandon

On Sat, Jan 23, 2021 at 3:57 AM Morten Brørup  
wrote:
>
> > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Lincoln Lavoie
> > Sent: Thursday, January 21, 2021 5:35 PM
> > To: Morten Brørup
> >
> > Hi All,
> >
> > Trying to follow the specific conversation.  It is correct, the lab
> > does
> > not list the specific throughput values achieved by the hardware, as
> > that
> > data can be sensitive to the hardware vendors, etc. The purpose of the
> > lab
> > is to check for degradations caused by patches, so the difference is
> > really
> > the important factor.  The comparison is against a prior run on the
> > same
> > hardware, via the DPDK main branch, so any delta should be caused by
> > the
> > specific patch changes (excluding statistical "wiggle").
> >
>
> Thank you for clarifying, Lincoln.
>
> This sounds like a perfect solution to the meet the purpose.
>
> I request that you change the output to show the relative difference (i.e. 
> percentage above/below baseline), instead of the absolute difference (i.e. 
> number of packets per second).
>
> By showing a percentage, anyone reading the test report can understand if the 
> difference is insignificant, or big enough to require further discussion 
> before accepting the patch. Showing the difference in packets per second 
> requires the reader of the test report to have prior knowledge about the 
> expected performance.
>
> > If the group would prefer, we could calculate additional references if
> > desired (i.e. difference from the last official release, or a monthly
> > run
> > of the current, etc.).  We just need the community to define their
> > needs,
> > and we can add this to the development queue.
> >
>
> I retract my suggestion about adding additional references. You clearly 
> explained how your baselining works, and I think it fully serves the needs of 
> a regression test.
>
> So please just put this small change in your development queue: Output the 
> difference in percent with a few decimals after the comma, instead of the 
> difference in packets per second.
>
> For readability, performance drops should be shown as negative values, and 
> increases as positive, e.g.:
>
> Difference = (NewPPS - BaselinePPS) / BaselinePPS * 100.00 %.
>
>
> If we want to compare performance against official releases, current or 
> older, it should go elsewhere, not in the continuous testing environment. 
> E.g. the release notes could include a report showing differences from the 
> last few official releases. But that is a task for another day.
>
>
> > Cheers,
> > Lincoln
> >
> >
> > On Thu, Jan 21, 2021 at 4:29 AM Morten Brørup
> > 
> > wrote:
> >
> > > > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ferruh Yigit
> > > > Sent: Thursday, January 21, 2021 10:19 AM
> > > >
> > > > On 1/15/2021 6:39 PM, Ali Alnubani wrote:
> > > > > Hi,
> > > > > Adding Ferruh and Zhaoyan,
> > > > >
> > > > >> Ali,
> > > > >>
> > > > >> You reported some performance regression, did you confirm it?
> > > > >> If I get no reply by monday, I'll proceed with this patch.
> > > > >
> > > > > Sure I'll confirm by Monday.
> > > > >
> > > > > Doesn't the regression also reproduce on the Lab's Intel servers?
> > > > > Even though the check iol-intel-Performance isn't failing, I can
> > see
> > > > that the throughput differences from expected for this patch are
> > less
> > > > than those of another patch that was tested only 20 minutes
> > earlier.
> > > > Both patches were applied to the same tree:
> > > > >
> > > > > https://mails.dpdk.org/archives/test-report/2021-
> > January/173927.html
> > > > >> | 64 | 512 | 1.571   |
> > > > >
> > > > > https://mails.dpdk.org/archives/test-report/2021-
> > January/173919.html
> > > > >> | 64 | 512 | 2.698   |
> > > > >
> > > > > Assuming that pw86457 doesn't have an effect on this test, it
> > looks
> > > > to me that this patch caused a regression in Intel hardware as
> > well.
> > > > >
> > > > > Can someone update the baseline's expected values for the Intel
> > NICs
> > > > and rerun the test on this patch?
> > > > >
> > > >
> > > > Zhaoyan said that the baseline is calculated dynamically,
> > > > what I understand is baseline set based on previous days
> > performance
> > > > result, so
> > > > it shouldn't require updating.
> > >
> > > That sounds smart!
> > >
> > > Perhaps another reference baseline could be added, for informational
> > > purposes only:
> > > Deviation from the performance of the last official release.
> > >
> > > >
> > > > But cc'ed the lab for more details.
> > >
> > >
> >
> > --
> > *Lincoln Lavoie*
> > Senior Engineer, Broadband Technologies
> > 21 Madbury Rd.,

[dpdk-dev] [PATCH] common/mlx5: add GTP TEID modification field ID

[Alexander Kozyrev]

Define hardware ID for GTP TEID modification. This value
can be used later in the RTE modify field API.

Signed-off-by: Alexander Kozyrev 
---
 drivers/common/mlx5/mlx5_prm.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/common/mlx5/mlx5_prm.h b/drivers/common/mlx5/mlx5_prm.h
index fe9173bc5a..410c08999c 100644
--- a/drivers/common/mlx5/mlx5_prm.h
+++ b/drivers/common/mlx5/mlx5_prm.h
@@ -549,6 +549,7 @@ enum mlx5_modification_field {
MLX5_MODI_IN_TCP_SEQ_NUM,
MLX5_MODI_OUT_TCP_ACK_NUM,
MLX5_MODI_IN_TCP_ACK_NUM = 0x5C,
+   MLX5_MODI_GTP_TEID = 0x6E,
 };
 
 /* Total number of metadata reg_c's. */
-- 
2.24.1

[dpdk-dev] [PATCH] bus/pci: nvme on Windows requires class id and bus

[Nick Connolly]

Attaching to an NVMe disk on Windows using SPDK requires the
PCI class ID and device.bus fields. Decode the class ID from the PCI
device info strings if it is present and set device.bus.

Signed-off-by: Nick Connolly 
---
 drivers/bus/pci/windows/pci.c | 10 +-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/bus/pci/windows/pci.c b/drivers/bus/pci/windows/pci.c
index f66258452..3d11444c2 100644
--- a/drivers/bus/pci/windows/pci.c
+++ b/drivers/bus/pci/windows/pci.c
@@ -280,17 +280,24 @@ parse_pci_hardware_id(const char *buf, struct rte_pci_id 
*pci_id)
 {
int ids = 0;
uint16_t vendor_id, device_id;
-   uint32_t subvendor_id = 0;
+   uint32_t subvendor_id = 0, class_id = 0;
+   const char *cp;
 
ids = sscanf_s(buf, "PCI\\VEN_%" PRIx16 "&DEV_%" PRIx16 "&SUBSYS_%"
PRIx32, &vendor_id, &device_id, &subvendor_id);
if (ids != 3)
return -1;
 
+   /* Try and find PCI class ID */
+   for (cp = buf; !(cp[0] == 0 && cp[1] == 0); cp++)
+   if (*cp == '&' && sscanf_s(cp, "&CC_%" PRIx32, &class_id) == 1)
+   break;
+
pci_id->vendor_id = vendor_id;
pci_id->device_id = device_id;
pci_id->subsystem_device_id = subvendor_id >> 16;
pci_id->subsystem_vendor_id = subvendor_id & 0x;
+   pci_id->class_id = class_id;
return 0;
 }
 
@@ -339,6 +346,7 @@ pci_scan_one(HDEVINFO dev_info, PSP_DEVINFO_DATA 
device_info_data)
if (ret != 0)
goto end;
 
+   dev->device.bus = &rte_pci_bus.bus;
dev->addr = addr;
dev->id = pci_id;
dev->max_vfs = 0; /* TODO: get max_vfs */
-- 
2.25.1

[dpdk-dev] [PATCH v3 00/44] net/virtio: Virtio PMD rework

[Maxime Coquelin]

Thanks Chenbo, for reviewing the v2. All the comments you
made should be taken into account.

This series significantly rework Virtio PMD to improve
the Virtio-user PMD and its backends integration.

First part of the series removes the dependency of
Virtio-user ethdev on Virtio PCI, by creating generic
files, adding per-bus meta data, ...

Main (if not single) functionnal change of this first
part is to remove the hack for Virtio-user to work in
IOVA as PA mode, this hack being very fragile.

Second part of the series reworks Virtio-user internal,
by reworking the requests handling so that vDPA and Kernel
backends no more hack into being Vhost-user backend. It
implies implementing new ops for all the request types.
Also, all the backend specific actions are moved from the
virtio_user_dev.c and virtio_user_ethdev.c to their
backend files.

Only functionnal change in this second part is making the
Vhost-user server mode blocking at init time, as long as
a client is not connected. The goal of this change is to
make the Vhost-user support much more robust, as without
blocking, the driver has to assume features that are going
to be supported by the client, which is very fragile and
error prone. As a side-effect, it also simplifies the
logic nin several place of the virtio-user PMD.

Main changes in v3:
- Rename .intr_event to .intr_detect
- Rework last patch, properly clean allocated resources
  on failure.
- Rebase on top of latest net-next/main
- Minor typo fixes in comments and log improvements

Main changes in v2:
===
- Introduce vdev driver flag for drivers to require IOVA VA mode
- Rebase on top of -rc1 changes
- Fix regressions introduced in V1 (vhost-kernel broken, vhost-user 
reconnect...)
- Various minor issues & typos fixed
- Fix status feature issue introduced in v20.11, only reproduceable now that 
server
  mode is made blocking
- Improve failure handling in Virtio-user
- Improve logging

Testing coverage (All passed)
=
- Virtio-pci PMD
 * Virtio PMD in guest with Vhost-user backend in host
 * Virtio PMD in guest with Vhost-kernel backend in host
- Virtio-user PMD with Vhost-user backend
 * Vhost-user PMD server <-> Virtio-user client PMD IO loopback
 * Vhost-user PMD client <-> Virtio-user server PMD IO loopback
 * Vhost-user PMD client <-> Virtio-user server PMD reconnect
- Virtio-user PMD with Vhost-kernel backend
 * iperf test case
 * Txonly testpmd
- Virtio-user PMD with Vhost-vDPA backend
 * vdpa-sim (IO loopback)
 * CX-6 DX Kernel vDPA (Tx only)

Maxime Coquelin (44):
  bus/vdev: add helper to get vdev from ethdev
  bus/vdev: add driver IOVA VA mode requirement
  net/virtio: fix getting old status on reconnect
  net/virtio: introduce Virtio bus type
  net/virtio: refactor virtio-user device
  net/virtio: introduce PCI device metadata
  net/virtio: move PCI device init in dedicated file
  net/virtio: move PCI specific dev init to PCI ethdev init
  net/virtio: move MSIX detection to PCI ethdev
  net/virtio: force IOVA as VA mode for Virtio-user
  net/virtio: store PCI type in Virtio device metadata
  net/virtio: add callback for device closing
  net/virtio: validate features at bus level
  net/virtio: remove bus type enum
  net/virtio: move PCI-specific fields to PCI device
  net/virtio: pack virtio HW struct
  net/virtio: move legacy IO to Virtio PCI
  net/virtio: introduce generic virtio header
  net/virtio: move features definition to generic header
  net/virtio: move virtqueue defines in generic header
  net/virtio: move config definitions to generic header
  net/virtio: make interrupt handling more generic
  net/virtio: move vring alignment to generic header
  net/virtio: remove last PCI refs in non-PCI code
  net/virtio: make Vhost-user request sender consistent
  net/virtio: add Virtio-user ops to set owner
  net/virtio: add Virtio-user features ops
  net/virtio: add Virtio-user protocol features ops
  net/virtio: add Virtio-user memory tables ops
  net/virtio: add Virtio-user vring setting ops
  net/virtio: add Virtio-user vring file ops
  net/virtio: add Virtio-user vring address ops
  net/virtio: add Virtio-user status ops
  net/virtio: remove useless request ops
  net/virtio: improve Virtio-user errors handling
  net/virtio: move Vhost-user requests to Vhost-user backend
  net/virtio: make server mode blocking
  net/virtio: move protocol features to Vhost-user
  net/virtio: introduce backend data
  net/virtio: move Vhost-user specifics to its backend
  net/virtio: move Vhost-kernel data to its backend
  net/virtio: move Vhost-vDPA data to its backend
  net/virtio: improve Vhost-user error logging
  net/virtio: handle Virtio-user setup failure properly

 drivers/bus/vdev/rte_bus_vdev.h   |   6 +
 drivers/bus/vdev/vdev.c   |  29 +
 drivers/net/virtio/meson.build|   6 +-
 drivers/net/virtio/virtio.c   |  71 ++
 drivers/net/virtio/virtio.h   | 246 +
 drivers/n

[dpdk-dev] [PATCH v3 01/44] bus/vdev: add helper to get vdev from ethdev

[Maxime Coquelin]

This patch adds an helper macro to get the rte_vdev_device
pointer from a rte_eth_dev pointer.

This is similar to RTE_ETH_DEV_TO_PCI().

Signed-off-by: Maxime Coquelin 
Reviewed-by: Chenbo Xia 
Reviewed-by: David Marchand 
---
 drivers/bus/vdev/rte_bus_vdev.h | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/bus/vdev/rte_bus_vdev.h b/drivers/bus/vdev/rte_bus_vdev.h
index d14eeb41b0..f99a41f825 100644
--- a/drivers/bus/vdev/rte_bus_vdev.h
+++ b/drivers/bus/vdev/rte_bus_vdev.h
@@ -34,6 +34,8 @@ struct rte_vdev_device {
 #define RTE_DEV_TO_VDEV_CONST(ptr) \
container_of(ptr, const struct rte_vdev_device, device)
 
+#define RTE_ETH_DEV_TO_VDEV(eth_dev)   RTE_DEV_TO_VDEV((eth_dev)->device)
+
 static inline const char *
 rte_vdev_device_name(const struct rte_vdev_device *dev)
 {
-- 
2.29.2

[dpdk-dev] [PATCH v3 02/44] bus/vdev: add driver IOVA VA mode requirement

[Maxime Coquelin]

This patch adds driver flag in vdev bus driver so that
vdev drivers can require VA IOVA mode to be used, which
for example the case of Virtio-user PMD.

The patch implements the .get_iommu_class() callback, that
is called before devices probing to determine the IOVA mode
to be used.

It also adds a check right before the device is probed to
ensure compatible IOVa mode has been selected.

Signed-off-by: Maxime Coquelin 
---
 drivers/bus/vdev/rte_bus_vdev.h |  4 
 drivers/bus/vdev/vdev.c | 29 +
 2 files changed, 33 insertions(+)

diff --git a/drivers/bus/vdev/rte_bus_vdev.h b/drivers/bus/vdev/rte_bus_vdev.h
index f99a41f825..c8b41e649c 100644
--- a/drivers/bus/vdev/rte_bus_vdev.h
+++ b/drivers/bus/vdev/rte_bus_vdev.h
@@ -113,8 +113,12 @@ struct rte_vdev_driver {
rte_vdev_remove_t *remove;   /**< Virtual device remove function. */
rte_vdev_dma_map_t *dma_map; /**< Virtual device DMA map function. 
*/
rte_vdev_dma_unmap_t *dma_unmap; /**< Virtual device DMA unmap 
function. */
+   uint32_t drv_flags;/**< Flags RTE_VDEV_DRV_*. */
 };
 
+/** Device driver needs IOVA as VA and cannot work with IOVA as PA */
+#define RTE_VDEV_DRV_NEED_IOVA_AS_VA 0x0001
+
 /**
  * Register a virtual device driver.
  *
diff --git a/drivers/bus/vdev/vdev.c b/drivers/bus/vdev/vdev.c
index acfd78828f..9a673347ae 100644
--- a/drivers/bus/vdev/vdev.c
+++ b/drivers/bus/vdev/vdev.c
@@ -189,6 +189,7 @@ vdev_probe_all_drivers(struct rte_vdev_device *dev)
 {
const char *name;
struct rte_vdev_driver *driver;
+   enum rte_iova_mode iova_mode;
int ret;
 
if (rte_dev_is_probed(&dev->device))
@@ -199,6 +200,14 @@ vdev_probe_all_drivers(struct rte_vdev_device *dev)
 
if (vdev_parse(name, &driver))
return -1;
+
+   iova_mode = rte_eal_iova_mode();
+   if ((driver->drv_flags & RTE_VDEV_DRV_NEED_IOVA_AS_VA) && (iova_mode == 
RTE_IOVA_PA)) {
+   VDEV_LOG(ERR, "%s requires VA IOVA mode but current mode is PA, 
not initializing",
+   name);
+   return -1;
+   }
+
ret = driver->probe(dev);
if (ret == 0)
dev->device.driver = &driver->driver;
@@ -594,6 +603,25 @@ vdev_unplug(struct rte_device *dev)
return rte_vdev_uninit(dev->name);
 }
 
+static enum rte_iova_mode
+vdev_get_iommu_class(void)
+{
+   const char *name;
+   struct rte_vdev_device *dev;
+   struct rte_vdev_driver *driver;
+
+   TAILQ_FOREACH(dev, &vdev_device_list, next) {
+   name = rte_vdev_device_name(dev);
+   if (vdev_parse(name, &driver))
+   continue;
+
+   if (driver->drv_flags & RTE_VDEV_DRV_NEED_IOVA_AS_VA)
+   return RTE_IOVA_VA;
+   }
+
+   return RTE_IOVA_DC;
+}
+
 static struct rte_bus rte_vdev_bus = {
.scan = vdev_scan,
.probe = vdev_probe,
@@ -603,6 +631,7 @@ static struct rte_bus rte_vdev_bus = {
.parse = vdev_parse,
.dma_map = vdev_dma_map,
.dma_unmap = vdev_dma_unmap,
+   .get_iommu_class = vdev_get_iommu_class,
.dev_iterate = rte_vdev_dev_iterate,
 };
 
-- 
2.29.2

[dpdk-dev] [PATCH v3 03/44] net/virtio: fix getting old status on reconnect

[Maxime Coquelin]

This patch fixes getting reset status from the restarted
vhost-user backend in case of reconnection, instead of the
status at the time of the disconnection.

This issue was not spotted earlier because Vhost-user
protocol status feature was disabled in server mode.

Fixes: 47235f16505f ("net/virtio-user: set status on socket reconnect")
Cc: sta...@dpdk.org

Signed-off-by: Maxime Coquelin 
Reviewed-by: Chenbo Xia 
---
 drivers/net/virtio/virtio_user_ethdev.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/virtio/virtio_user_ethdev.c 
b/drivers/net/virtio/virtio_user_ethdev.c
index 241808cd8f..7abba1f1ee 100644
--- a/drivers/net/virtio/virtio_user_ethdev.c
+++ b/drivers/net/virtio/virtio_user_ethdev.c
@@ -77,7 +77,7 @@ virtio_user_server_reconnect(struct virtio_user_dev *dev)
return -1;
 
dev->vhostfd = connectfd;
-   old_status = vtpci_get_status(hw);
+   old_status = dev->status;
 
vtpci_reset(hw);
 
-- 
2.29.2

[dpdk-dev] [PATCH v3 04/44] net/virtio: introduce Virtio bus type

[Maxime Coquelin]

This patch is preliminary work for introducing a bus layer
in Virtio PMD, in order to improve Virtio-user integration.

A new bus type is added to provide a unified way to distinguish
which bus type is used (PCI modern, PCI legacy or Virtio-user).

Signed-off-by: Maxime Coquelin 
Reviewed-by: Chenbo Xia 
---
 drivers/net/virtio/virtio_ethdev.c  | 40 +
 drivers/net/virtio/virtio_pci.c |  4 +--
 drivers/net/virtio/virtio_pci.h |  9 +-
 drivers/net/virtio/virtio_user_ethdev.c |  2 +-
 4 files changed, 32 insertions(+), 23 deletions(-)

diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index 92a3d4efa5..0ba10c274a 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -592,9 +592,9 @@ virtio_init_queue(struct rte_eth_dev *dev, uint16_t 
vtpci_queue_idx)
 * we use virtual address. And we need properly set _offset_, please see
 * VIRTIO_MBUF_DATA_DMA_ADDR in virtqueue.h for more information.
 */
-   if (!hw->virtio_user_dev)
+   if (hw->bus_type == VIRTIO_BUS_PCI_LEGACY || hw->bus_type == 
VIRTIO_BUS_PCI_MODERN) {
vq->offset = offsetof(struct rte_mbuf, buf_iova);
-   else {
+   } else if (hw->bus_type == VIRTIO_BUS_USER) {
vq->vq_ring_mem = (uintptr_t)mz->addr;
vq->offset = offsetof(struct rte_mbuf, buf_addr);
if (queue_type == VTNET_TQ)
@@ -746,13 +746,13 @@ virtio_dev_close(struct rte_eth_dev *dev)
virtio_free_queues(hw);
 
 #ifdef RTE_VIRTIO_USER
-   if (hw->virtio_user_dev)
+   if (hw->bus_type == VIRTIO_BUS_USER)
virtio_user_dev_uninit(hw->virtio_user_dev);
else
 #endif
if (dev->device) {
rte_pci_unmap_device(RTE_ETH_DEV_TO_PCI(dev));
-   if (!hw->modern)
+   if (hw->bus_type == VIRTIO_BUS_PCI_LEGACY)
rte_pci_ioport_unmap(VTPCI_IO(hw));
}
 
@@ -1299,7 +1299,7 @@ virtio_intr_unmask(struct rte_eth_dev *dev)
if (rte_intr_ack(dev->intr_handle) < 0)
return -1;
 
-   if (!hw->virtio_user_dev)
+   if (hw->bus_type == VIRTIO_BUS_PCI_LEGACY || hw->bus_type == 
VIRTIO_BUS_PCI_MODERN)
hw->use_msix = vtpci_msix_detect(RTE_ETH_DEV_TO_PCI(dev));
 
return 0;
@@ -1313,7 +1313,7 @@ virtio_intr_enable(struct rte_eth_dev *dev)
if (rte_intr_enable(dev->intr_handle) < 0)
return -1;
 
-   if (!hw->virtio_user_dev)
+   if (hw->bus_type == VIRTIO_BUS_PCI_LEGACY || hw->bus_type == 
VIRTIO_BUS_PCI_MODERN)
hw->use_msix = vtpci_msix_detect(RTE_ETH_DEV_TO_PCI(dev));
 
return 0;
@@ -1327,7 +1327,7 @@ virtio_intr_disable(struct rte_eth_dev *dev)
if (rte_intr_disable(dev->intr_handle) < 0)
return -1;
 
-   if (!hw->virtio_user_dev)
+   if (hw->bus_type == VIRTIO_BUS_PCI_LEGACY || hw->bus_type == 
VIRTIO_BUS_PCI_MODERN)
hw->use_msix = vtpci_msix_detect(RTE_ETH_DEV_TO_PCI(dev));
 
return 0;
@@ -1368,13 +1368,13 @@ virtio_negotiate_features(struct virtio_hw *hw, 
uint64_t req_features)
PMD_INIT_LOG(DEBUG, "features after negotiate = %" PRIx64,
hw->guest_features);
 
-   if (hw->modern && !vtpci_with_feature(hw, VIRTIO_F_VERSION_1)) {
+   if (hw->bus_type == VIRTIO_BUS_PCI_MODERN && !vtpci_with_feature(hw, 
VIRTIO_F_VERSION_1)) {
PMD_INIT_LOG(ERR,
"VIRTIO_F_VERSION_1 features is not enabled.");
return -1;
}
 
-   if (hw->modern || hw->virtio_user_dev) {
+   if (hw->bus_type == VIRTIO_BUS_PCI_MODERN || hw->bus_type == 
VIRTIO_BUS_USER) {
vtpci_set_status(hw, VIRTIO_CONFIG_STATUS_FEATURES_OK);
if (!(vtpci_get_status(hw) & VIRTIO_CONFIG_STATUS_FEATURES_OK)) 
{
PMD_INIT_LOG(ERR,
@@ -1709,7 +1709,7 @@ virtio_init_device(struct rte_eth_dev *eth_dev, uint64_t 
req_features)
 
hw->weak_barriers = !vtpci_with_feature(hw, VIRTIO_F_ORDER_PLATFORM);
 
-   if (!hw->virtio_user_dev)
+   if (hw->bus_type == VIRTIO_BUS_PCI_LEGACY || hw->bus_type == 
VIRTIO_BUS_PCI_MODERN)
pci_dev = RTE_ETH_DEV_TO_PCI(eth_dev);
 
/* If host does not support both status and MSI-X then disable LSC */
@@ -1856,7 +1856,7 @@ virtio_init_device(struct rte_eth_dev *eth_dev, uint64_t 
req_features)
 static int
 virtio_remap_pci(struct rte_pci_device *pci_dev, struct virtio_hw *hw)
 {
-   if (hw->modern) {
+   if (hw->bus_type == VIRTIO_BUS_PCI_MODERN) {
/*
 * We don't have to re-parse the PCI config space, since
 * rte_pci_map_device() makes sure the mapped address
@@ -1872,7 +1872,7 @@ virtio_remap_pci(struct rte_pci_device *pci_dev, struct 
virtio_hw *hw)
PMD_INIT_LOG(DEBUG, "failed to map pci device!");

[dpdk-dev] [PATCH v3 05/44] net/virtio: refactor virtio-user device

[Maxime Coquelin]

This patch moves the virtio_hw structure into the virtio_user_dev
structure, with the goal of making virtio_hw bus-agnostic.

Signed-off-by: Maxime Coquelin 
Reviewed-by: Chenbo Xia 
---
 drivers/net/virtio/virtio_ethdev.c|  2 +-
 drivers/net/virtio/virtio_pci.h   |  1 -
 .../net/virtio/virtio_user/virtio_user_dev.h  |  1 +
 drivers/net/virtio/virtio_user_ethdev.c   | 64 ---
 4 files changed, 29 insertions(+), 39 deletions(-)

diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index 0ba10c274a..c8a01a46df 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -747,7 +747,7 @@ virtio_dev_close(struct rte_eth_dev *dev)
 
 #ifdef RTE_VIRTIO_USER
if (hw->bus_type == VIRTIO_BUS_USER)
-   virtio_user_dev_uninit(hw->virtio_user_dev);
+   virtio_user_dev_uninit(dev->data->dev_private);
else
 #endif
if (dev->device) {
diff --git a/drivers/net/virtio/virtio_pci.h b/drivers/net/virtio/virtio_pci.h
index 6388f0a74d..b35a596169 100644
--- a/drivers/net/virtio/virtio_pci.h
+++ b/drivers/net/virtio/virtio_pci.h
@@ -277,7 +277,6 @@ struct virtio_hw {
uint16_t*notify_base;
struct virtio_pci_common_cfg *common_cfg;
struct virtio_net_config *dev_cfg;
-   void*virtio_user_dev;
/*
 * App management thread and virtio interrupt handler thread
 * both can change device state, this lock is meant to avoid
diff --git a/drivers/net/virtio/virtio_user/virtio_user_dev.h 
b/drivers/net/virtio/virtio_user/virtio_user_dev.h
index 3b5b6bc3ae..b3776cc7a0 100644
--- a/drivers/net/virtio/virtio_user/virtio_user_dev.h
+++ b/drivers/net/virtio/virtio_user/virtio_user_dev.h
@@ -24,6 +24,7 @@ struct virtio_user_queue {
 };
 
 struct virtio_user_dev {
+   struct virtio_hw hw;
enum virtio_user_backend_type backend_type;
/* for vhost_user backend */
int vhostfd;
diff --git a/drivers/net/virtio/virtio_user_ethdev.c 
b/drivers/net/virtio/virtio_user_ethdev.c
index 441721b6ca..620ecb126c 100644
--- a/drivers/net/virtio/virtio_user_ethdev.c
+++ b/drivers/net/virtio/virtio_user_ethdev.c
@@ -26,13 +26,13 @@
 #include "virtio_user/virtio_user_dev.h"
 #include "virtio_user/vhost.h"
 
-#define virtio_user_get_dev(hw) \
-   ((struct virtio_user_dev *)(hw)->virtio_user_dev)
+#define virtio_user_get_dev(hwp) container_of(hwp, struct virtio_user_dev, hw)
 
 static void
-virtio_user_reset_queues_packed(struct rte_eth_dev *dev)
+virtio_user_reset_queues_packed(struct rte_eth_dev *eth_dev)
 {
-   struct virtio_hw *hw = dev->data->dev_private;
+   struct virtio_user_dev *dev = eth_dev->data->dev_private;
+   struct virtio_hw *hw = &dev->hw;
struct virtnet_rx *rxvq;
struct virtnet_tx *txvq;
uint16_t i;
@@ -48,14 +48,14 @@ virtio_user_reset_queues_packed(struct rte_eth_dev *dev)
rte_delay_ms(1);
 
/* Vring reset for each Tx queue and Rx queue. */
-   for (i = 0; i < dev->data->nb_rx_queues; i++) {
-   rxvq = dev->data->rx_queues[i];
+   for (i = 0; i < eth_dev->data->nb_rx_queues; i++) {
+   rxvq = eth_dev->data->rx_queues[i];
virtqueue_rxvq_reset_packed(rxvq->vq);
-   virtio_dev_rx_queue_setup_finish(dev, i);
+   virtio_dev_rx_queue_setup_finish(eth_dev, i);
}
 
-   for (i = 0; i < dev->data->nb_tx_queues; i++) {
-   txvq = dev->data->tx_queues[i];
+   for (i = 0; i < eth_dev->data->nb_tx_queues; i++) {
+   txvq = eth_dev->data->tx_queues[i];
virtqueue_txvq_reset_packed(txvq->vq);
}
 
@@ -69,7 +69,7 @@ virtio_user_server_reconnect(struct virtio_user_dev *dev)
 {
int ret, connectfd, old_status;
struct rte_eth_dev *eth_dev = &rte_eth_devices[dev->port_id];
-   struct virtio_hw *hw = eth_dev->data->dev_private;
+   struct virtio_hw *hw = &dev->hw;
uint64_t protocol_features;
 
connectfd = accept(dev->listenfd, NULL, NULL);
@@ -605,21 +605,15 @@ virtio_user_eth_dev_alloc(struct rte_vdev_device *vdev)
struct virtio_hw *hw;
struct virtio_user_dev *dev;
 
-   eth_dev = rte_eth_vdev_allocate(vdev, sizeof(*hw));
+   eth_dev = rte_eth_vdev_allocate(vdev, sizeof(*dev));
if (!eth_dev) {
PMD_INIT_LOG(ERR, "cannot alloc rte_eth_dev");
return NULL;
}
 
data = eth_dev->data;
-   hw = eth_dev->data->dev_private;
-
-   dev = rte_zmalloc(NULL, sizeof(*dev), 0);
-   if (!dev) {
-   PMD_INIT_LOG(ERR, "malloc virtio_user_dev failed");
-   rte_eth_dev_release_port(eth_dev);
-   return NULL;
-   }
+   dev = eth_dev->data->dev_private;
+   hw = &dev->hw;
 
hw->port_id = data->port_id;
dev->port_id = data->port_id;
@@ -634,17 +628,13 @@ virtio_user_

[dpdk-dev] [PATCH v3 06/44] net/virtio: introduce PCI device metadata

[Maxime Coquelin]

This patch initiate refactoring of Virtio PCI, by introducing
a new device structure for PCI-specific metadata.

Signed-off-by: Maxime Coquelin 
Reviewed-by: Chenbo Xia 
---
 drivers/net/virtio/virtio_ethdev.c | 2 +-
 drivers/net/virtio/virtio_pci.c| 2 ++
 drivers/net/virtio/virtio_pci.h| 5 +
 3 files changed, 8 insertions(+), 1 deletion(-)

diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index c8a01a46df..aafba15ac2 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -2154,7 +2154,7 @@ static int eth_virtio_pci_probe(struct rte_pci_driver 
*pci_drv __rte_unused,
if (vdpa == 1)
return 1;
 
-   return rte_eth_dev_pci_generic_probe(pci_dev, sizeof(struct virtio_hw),
+   return rte_eth_dev_pci_generic_probe(pci_dev, sizeof(struct 
virtio_pci_dev),
eth_virtio_dev_init);
 }
 
diff --git a/drivers/net/virtio/virtio_pci.c b/drivers/net/virtio/virtio_pci.c
index 1692268f30..345d73f868 100644
--- a/drivers/net/virtio/virtio_pci.c
+++ b/drivers/net/virtio/virtio_pci.c
@@ -689,6 +689,8 @@ virtio_read_caps(struct rte_pci_device *dev, struct 
virtio_hw *hw)
 int
 vtpci_init(struct rte_pci_device *dev, struct virtio_hw *hw)
 {
+   RTE_BUILD_BUG_ON(offsetof(struct virtio_pci_dev, hw) != 0);
+
/*
 * Try if we can succeed reading virtio pci caps, which exists
 * only on modern pci device. If failed, we fallback to legacy
diff --git a/drivers/net/virtio/virtio_pci.h b/drivers/net/virtio/virtio_pci.h
index b35a596169..59f6688218 100644
--- a/drivers/net/virtio/virtio_pci.h
+++ b/drivers/net/virtio/virtio_pci.h
@@ -289,6 +289,11 @@ struct virtio_hw {
struct virtqueue **vqs;
 };
 
+struct virtio_pci_dev {
+   struct virtio_hw hw;
+};
+
+#define virtio_pci_get_dev(hwp) container_of(hwp, struct virtio_pci_dev, hw)
 
 /*
  * While virtio_hw is stored in shared memory, this structure stores
-- 
2.29.2

[dpdk-dev] [PATCH v3 07/44] net/virtio: move PCI device init in dedicated file

[Maxime Coquelin]

This patch moves the PCI Ethernet device registration bits
in a dedicated patch. In following patches, more code will
be moved there, with the goal of making virtio_ethdev.c
truly bus-agnostic.

Signed-off-by: Maxime Coquelin 
Reviewed-by: Chenbo Xia 
---
 drivers/net/virtio/meson.build |   1 +
 drivers/net/virtio/virtio_ethdev.c | 117 +--
 drivers/net/virtio/virtio_ethdev.h |   2 +
 drivers/net/virtio/virtio_pci_ethdev.c | 149 +
 4 files changed, 157 insertions(+), 112 deletions(-)
 create mode 100644 drivers/net/virtio/virtio_pci_ethdev.c

diff --git a/drivers/net/virtio/meson.build b/drivers/net/virtio/meson.build
index 0e78d1243b..07e085376b 100644
--- a/drivers/net/virtio/meson.build
+++ b/drivers/net/virtio/meson.build
@@ -8,6 +8,7 @@ if is_windows
 endif
 
 sources += files('virtio_ethdev.c',
+   'virtio_pci_ethdev.c',
'virtio_pci.c',
'virtio_rxtx.c',
'virtio_rxtx_simple.c',
diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index aafba15ac2..ad7e9efd5f 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -38,17 +38,14 @@
 #include "virtio_rxtx.h"
 #include "virtio_user/virtio_user_dev.h"
 
-static int eth_virtio_dev_uninit(struct rte_eth_dev *eth_dev);
 static int  virtio_dev_configure(struct rte_eth_dev *dev);
 static int  virtio_dev_start(struct rte_eth_dev *dev);
-static int  virtio_dev_stop(struct rte_eth_dev *dev);
 static int virtio_dev_promiscuous_enable(struct rte_eth_dev *dev);
 static int virtio_dev_promiscuous_disable(struct rte_eth_dev *dev);
 static int virtio_dev_allmulticast_enable(struct rte_eth_dev *dev);
 static int virtio_dev_allmulticast_disable(struct rte_eth_dev *dev);
 static uint32_t virtio_dev_speed_capa_get(uint32_t speed);
 static int virtio_dev_devargs_parse(struct rte_devargs *devargs,
-   int *vdpa,
uint32_t *speed,
int *vectorized);
 static int virtio_dev_info_get(struct rte_eth_dev *dev,
@@ -89,15 +86,6 @@ static int virtio_dev_queue_stats_mapping_set(
 static void virtio_notify_peers(struct rte_eth_dev *dev);
 static void virtio_ack_link_announce(struct rte_eth_dev *dev);
 
-/*
- * The set of PCI devices this driver supports
- */
-static const struct rte_pci_id pci_id_virtio_map[] = {
-   { RTE_PCI_DEVICE(VIRTIO_PCI_VENDORID, VIRTIO_PCI_LEGACY_DEVICEID_NET) },
-   { RTE_PCI_DEVICE(VIRTIO_PCI_VENDORID, VIRTIO_PCI_MODERN_DEVICEID_NET) },
-   { .vendor_id = 0, /* sentinel */ },
-};
-
 struct rte_virtio_xstats_name_off {
char name[RTE_ETH_XSTATS_NAME_SIZE];
unsigned offset;
@@ -714,7 +702,7 @@ virtio_alloc_queues(struct rte_eth_dev *dev)
 
 static void virtio_queues_unbind_intr(struct rte_eth_dev *dev);
 
-static int
+int
 virtio_dev_close(struct rte_eth_dev *dev)
 {
struct virtio_hw *hw = dev->data->dev_private;
@@ -1932,8 +1920,7 @@ eth_virtio_dev_init(struct rte_eth_dev *eth_dev)
 
return 0;
}
-   ret = virtio_dev_devargs_parse(eth_dev->device->devargs,
-NULL, &speed, &vectorized);
+   ret = virtio_dev_devargs_parse(eth_dev->device->devargs, &speed, 
&vectorized);
if (ret < 0)
return ret;
hw->speed = speed;
@@ -1995,36 +1982,6 @@ eth_virtio_dev_init(struct rte_eth_dev *eth_dev)
return ret;
 }
 
-static int
-eth_virtio_dev_uninit(struct rte_eth_dev *eth_dev)
-{
-   int ret;
-   PMD_INIT_FUNC_TRACE();
-
-   if (rte_eal_process_type() == RTE_PROC_SECONDARY)
-   return 0;
-
-   ret = virtio_dev_stop(eth_dev);
-   virtio_dev_close(eth_dev);
-
-   PMD_INIT_LOG(DEBUG, "dev_uninit completed");
-
-   return ret;
-}
-
-
-static int vdpa_check_handler(__rte_unused const char *key,
-   const char *value, void *ret_val)
-{
-   if (strcmp(value, "1") == 0)
-   *(int *)ret_val = 1;
-   else
-   *(int *)ret_val = 0;
-
-   return 0;
-}
-
-
 static uint32_t
 virtio_dev_speed_capa_get(uint32_t speed)
 {
@@ -2062,10 +2019,8 @@ static int vectorized_check_handler(__rte_unused const 
char *key,
 }
 
 #define VIRTIO_ARG_SPEED  "speed"
-#define VIRTIO_ARG_VDPA   "vdpa"
 #define VIRTIO_ARG_VECTORIZED "vectorized"
 
-
 static int
 link_speed_handler(const char *key __rte_unused,
const char *value, void *ret_val)
@@ -2084,8 +2039,7 @@ link_speed_handler(const char *key __rte_unused,
 
 
 static int
-virtio_dev_devargs_parse(struct rte_devargs *devargs, int *vdpa,
-   uint32_t *speed, int *vectorized)
+virtio_dev_devargs_parse(struct rte_devargs *devargs, uint32_t *speed, int 
*vectorized)
 {
struct rte_kvargs *kvlist;
int ret = 0;
@@ -2098,18 +2052,7 @@ virtio_dev_devargs_parse(struct rte_devargs *devargs, 
int *vdpa,
PMD_INIT_LOG(ERR, "error when parsing param");
return 0;
}
-   if (vdpa && rte_kvargs_count(kvlist, VIRTIO_A

[dpdk-dev] [PATCH v3 09/44] net/virtio: move MSIX detection to PCI ethdev

[Maxime Coquelin]

This patch introduces a new callback to notify the bus
driver some interrupt related operation was done.

This is used by Virtio PCI driver to check msix status.

Signed-off-by: Maxime Coquelin 
---
 drivers/net/virtio/virtio_ethdev.c |  12 +--
 drivers/net/virtio/virtio_pci.c| 120 ++---
 drivers/net/virtio/virtio_pci.h|   6 +-
 drivers/net/virtio/virtio_pci_ethdev.c |   2 +
 4 files changed, 82 insertions(+), 58 deletions(-)

diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index a3e81f336d..52eb878c42 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -1287,8 +1287,8 @@ virtio_intr_unmask(struct rte_eth_dev *dev)
if (rte_intr_ack(dev->intr_handle) < 0)
return -1;
 
-   if (hw->bus_type == VIRTIO_BUS_PCI_LEGACY || hw->bus_type == 
VIRTIO_BUS_PCI_MODERN)
-   hw->use_msix = vtpci_msix_detect(RTE_ETH_DEV_TO_PCI(dev));
+   if (VTPCI_OPS(hw)->intr_detect)
+   VTPCI_OPS(hw)->intr_detect(hw);
 
return 0;
 }
@@ -1301,8 +1301,8 @@ virtio_intr_enable(struct rte_eth_dev *dev)
if (rte_intr_enable(dev->intr_handle) < 0)
return -1;
 
-   if (hw->bus_type == VIRTIO_BUS_PCI_LEGACY || hw->bus_type == 
VIRTIO_BUS_PCI_MODERN)
-   hw->use_msix = vtpci_msix_detect(RTE_ETH_DEV_TO_PCI(dev));
+   if (VTPCI_OPS(hw)->intr_detect)
+   VTPCI_OPS(hw)->intr_detect(hw);
 
return 0;
 }
@@ -1315,8 +1315,8 @@ virtio_intr_disable(struct rte_eth_dev *dev)
if (rte_intr_disable(dev->intr_handle) < 0)
return -1;
 
-   if (hw->bus_type == VIRTIO_BUS_PCI_LEGACY || hw->bus_type == 
VIRTIO_BUS_PCI_MODERN)
-   hw->use_msix = vtpci_msix_detect(RTE_ETH_DEV_TO_PCI(dev));
+   if (VTPCI_OPS(hw)->intr_detect)
+   VTPCI_OPS(hw)->intr_detect(hw);
 
return 0;
 }
diff --git a/drivers/net/virtio/virtio_pci.c b/drivers/net/virtio/virtio_pci.c
index 345d73f868..556be1e3da 100644
--- a/drivers/net/virtio/virtio_pci.c
+++ b/drivers/net/virtio/virtio_pci.c
@@ -47,6 +47,56 @@ check_vq_phys_addr_ok(struct virtqueue *vq)
return 1;
 }
 
+#define PCI_MSIX_ENABLE 0x8000
+
+static enum virtio_msix_status
+vtpci_msix_detect(struct rte_pci_device *dev)
+{
+   uint8_t pos;
+   int ret;
+
+   ret = rte_pci_read_config(dev, &pos, 1, PCI_CAPABILITY_LIST);
+   if (ret != 1) {
+   PMD_INIT_LOG(DEBUG,
+"failed to read pci capability list, ret %d", ret);
+   return VIRTIO_MSIX_NONE;
+   }
+
+   while (pos) {
+   uint8_t cap[2];
+
+   ret = rte_pci_read_config(dev, cap, sizeof(cap), pos);
+   if (ret != sizeof(cap)) {
+   PMD_INIT_LOG(DEBUG,
+"failed to read pci cap at pos: %x ret %d",
+pos, ret);
+   break;
+   }
+
+   if (cap[0] == PCI_CAP_ID_MSIX) {
+   uint16_t flags;
+
+   ret = rte_pci_read_config(dev, &flags, sizeof(flags),
+   pos + sizeof(cap));
+   if (ret != sizeof(flags)) {
+   PMD_INIT_LOG(DEBUG,
+"failed to read pci cap at pos:"
+" %x ret %d", pos + 2, ret);
+   break;
+   }
+
+   if (flags & PCI_MSIX_ENABLE)
+   return VIRTIO_MSIX_ENABLED;
+   else
+   return VIRTIO_MSIX_DISABLED;
+   }
+
+   pos = cap[1];
+   }
+
+   return VIRTIO_MSIX_NONE;
+}
+
 /*
  * Since we are in legacy mode:
  * http://ozlabs.org/~rusty/virtio-spec/virtio-0.9.5.pdf
@@ -241,6 +291,12 @@ legacy_notify_queue(struct virtio_hw *hw, struct virtqueue 
*vq)
VIRTIO_PCI_QUEUE_NOTIFY);
 }
 
+static void
+legacy_intr_detect(struct virtio_hw *hw)
+{
+   hw->use_msix = vtpci_msix_detect(VTPCI_DEV(hw));
+}
+
 const struct virtio_pci_ops legacy_ops = {
.read_dev_cfg   = legacy_read_dev_config,
.write_dev_cfg  = legacy_write_dev_config,
@@ -255,6 +311,7 @@ const struct virtio_pci_ops legacy_ops = {
.setup_queue= legacy_setup_queue,
.del_queue  = legacy_del_queue,
.notify_queue   = legacy_notify_queue,
+   .intr_detect= legacy_intr_detect,
 };
 
 static inline void
@@ -446,6 +503,14 @@ modern_notify_queue(struct virtio_hw *hw, struct virtqueue 
*vq)
rte_write32(notify_data, vq->notify_addr);
 }
 
+
+
+static void
+modern_intr_detect(struct virtio_hw *hw)
+{
+   hw->use_msix = vtpci_msix_detect(VTPCI_DEV(hw));
+}
+
 const struct virtio_pci_ops modern_ops = {
.read_dev_cfg   = mode

[dpdk-dev] [PATCH v3 08/44] net/virtio: move PCI specific dev init to PCI ethdev init

[Maxime Coquelin]

This patch moves the PCI specific initialization from
eth_virtio_dev_init() to eth_virtio_pci_init().

Signed-off-by: Maxime Coquelin 
Reviewed-by: Chenbo Xia 
Reviewed-by: David Marchand 
---
 drivers/net/virtio/virtio_ethdev.c | 63 +--
 drivers/net/virtio/virtio_pci_ethdev.c | 71 +-
 2 files changed, 71 insertions(+), 63 deletions(-)

diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index ad7e9efd5f..a3e81f336d 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -1676,7 +1676,6 @@ virtio_init_device(struct rte_eth_dev *eth_dev, uint64_t 
req_features)
struct virtio_hw *hw = eth_dev->data->dev_private;
struct virtio_net_config *config;
struct virtio_net_config local_config;
-   struct rte_pci_device *pci_dev = NULL;
int ret;
 
/* Reset the device although not necessary at startup */
@@ -1697,9 +1696,6 @@ virtio_init_device(struct rte_eth_dev *eth_dev, uint64_t 
req_features)
 
hw->weak_barriers = !vtpci_with_feature(hw, VIRTIO_F_ORDER_PLATFORM);
 
-   if (hw->bus_type == VIRTIO_BUS_PCI_LEGACY || hw->bus_type == 
VIRTIO_BUS_PCI_MODERN)
-   pci_dev = RTE_ETH_DEV_TO_PCI(eth_dev);
-
/* If host does not support both status and MSI-X then disable LSC */
if (vtpci_with_feature(hw, VIRTIO_NET_F_STATUS) &&
hw->use_msix != VIRTIO_MSIX_NONE)
@@ -1828,45 +1824,9 @@ virtio_init_device(struct rte_eth_dev *eth_dev, uint64_t 
req_features)
 
vtpci_reinit_complete(hw);
 
-   if (pci_dev)
-   PMD_INIT_LOG(DEBUG, "port %d vendorID=0x%x deviceID=0x%x",
-   eth_dev->data->port_id, pci_dev->id.vendor_id,
-   pci_dev->id.device_id);
-
return 0;
 }
 
-/*
- * Remap the PCI device again (IO port map for legacy device and
- * memory map for modern device), so that the secondary process
- * could have the PCI initiated correctly.
- */
-static int
-virtio_remap_pci(struct rte_pci_device *pci_dev, struct virtio_hw *hw)
-{
-   if (hw->bus_type == VIRTIO_BUS_PCI_MODERN) {
-   /*
-* We don't have to re-parse the PCI config space, since
-* rte_pci_map_device() makes sure the mapped address
-* in secondary process would equal to the one mapped in
-* the primary process: error will be returned if that
-* requirement is not met.
-*
-* That said, we could simply reuse all cap pointers
-* (such as dev_cfg, common_cfg, etc.) parsed from the
-* primary process, which is stored in shared memory.
-*/
-   if (rte_pci_map_device(pci_dev)) {
-   PMD_INIT_LOG(DEBUG, "failed to map pci device!");
-   return -1;
-   }
-   } else if (hw->bus_type == VIRTIO_BUS_PCI_LEGACY) {
-   if (rte_pci_ioport_map(pci_dev, 0, VTPCI_IO(hw)) < 0)
-   return -1;
-   }
-
-   return 0;
-}
 
 static void
 virtio_set_vtpci_ops(struct virtio_hw *hw)
@@ -1909,17 +1869,11 @@ eth_virtio_dev_init(struct rte_eth_dev *eth_dev)
eth_dev->rx_descriptor_done = virtio_dev_rx_queue_done;
 
if (rte_eal_process_type() == RTE_PROC_SECONDARY) {
-   if (hw->bus_type != VIRTIO_BUS_USER) {
-   ret = virtio_remap_pci(RTE_ETH_DEV_TO_PCI(eth_dev), hw);
-   if (ret)
-   return ret;
-   }
-
virtio_set_vtpci_ops(hw);
set_rxtx_funcs(eth_dev);
-
return 0;
}
+
ret = virtio_dev_devargs_parse(eth_dev->device->devargs, &speed, 
&vectorized);
if (ret < 0)
return ret;
@@ -1936,15 +1890,6 @@ eth_virtio_dev_init(struct rte_eth_dev *eth_dev)
}
 
hw->port_id = eth_dev->data->port_id;
-   /* For virtio_user case the hw->virtio_user_dev is populated by
-* virtio_user_eth_dev_alloc() before eth_virtio_dev_init() is called.
-*/
-   if (hw->bus_type != VIRTIO_BUS_USER) {
-   ret = vtpci_init(RTE_ETH_DEV_TO_PCI(eth_dev), hw);
-   if (ret)
-   goto err_vtpci_init;
-   }
-
rte_spinlock_init(&hw->state_lock);
 
/* reset device and negotiate default features */
@@ -1971,12 +1916,6 @@ eth_virtio_dev_init(struct rte_eth_dev *eth_dev)
return 0;
 
 err_virtio_init:
-   if (hw->bus_type == VIRTIO_BUS_PCI_MODERN || hw->bus_type == 
VIRTIO_BUS_PCI_LEGACY) {
-   rte_pci_unmap_device(RTE_ETH_DEV_TO_PCI(eth_dev));
-   if (hw->bus_type == VIRTIO_BUS_PCI_LEGACY)
-   rte_pci_ioport_unmap(VTPCI_IO(hw));
-   }
-err_vtpci_init:
rte_free(eth_dev->data->mac_addrs);
eth_dev->data->mac_addrs = NULL;
   

[dpdk-dev] [PATCH v3 10/44] net/virtio: force IOVA as VA mode for Virtio-user

[Maxime Coquelin]

At least Vhost-user backend of Virtio-user PMD requires
IOVA as VA mode. Until now, it was implemented as a hack
by forcing to use mbuf's buf_addr field instead of buf_iova.

This patch removes all this logic and just fails probing
if IOVA as VA mode is not selected. It simplifies the
code overall, and removes some bus-specific logic from
generic virtio_ethdev.c.

Signed-off-by: Maxime Coquelin 
Reviewed-by: Chenbo Xia 
---
 drivers/net/virtio/virtio_ethdev.c   | 15 -
 drivers/net/virtio/virtio_rxtx.c | 34 
 drivers/net/virtio/virtio_rxtx_packed.h  |  2 +-
 drivers/net/virtio/virtio_rxtx_packed_avx.h  |  8 ++---
 drivers/net/virtio/virtio_rxtx_packed_neon.h |  8 ++---
 drivers/net/virtio/virtio_rxtx_simple.h  |  3 +-
 drivers/net/virtio/virtio_user_ethdev.c  |  1 +
 drivers/net/virtio/virtqueue.h   | 25 +-
 8 files changed, 26 insertions(+), 70 deletions(-)

diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index 52eb878c42..fb789460e8 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -576,21 +576,6 @@ virtio_init_queue(struct rte_eth_dev *dev, uint16_t 
vtpci_queue_idx)
hw->cvq = cvq;
}
 
-   /* For virtio_user case (that is when hw->virtio_user_dev is not NULL),
-* we use virtual address. And we need properly set _offset_, please see
-* VIRTIO_MBUF_DATA_DMA_ADDR in virtqueue.h for more information.
-*/
-   if (hw->bus_type == VIRTIO_BUS_PCI_LEGACY || hw->bus_type == 
VIRTIO_BUS_PCI_MODERN) {
-   vq->offset = offsetof(struct rte_mbuf, buf_iova);
-   } else if (hw->bus_type == VIRTIO_BUS_USER) {
-   vq->vq_ring_mem = (uintptr_t)mz->addr;
-   vq->offset = offsetof(struct rte_mbuf, buf_addr);
-   if (queue_type == VTNET_TQ)
-   txvq->virtio_net_hdr_mem = (uintptr_t)hdr_mz->addr;
-   else if (queue_type == VTNET_CQ)
-   cvq->virtio_net_hdr_mem = (uintptr_t)hdr_mz->addr;
-   }
-
if (queue_type == VTNET_TQ) {
struct virtio_tx_region *txr;
unsigned int i;
diff --git a/drivers/net/virtio/virtio_rxtx.c b/drivers/net/virtio/virtio_rxtx.c
index 622d4bf201..6875c8fbee 100644
--- a/drivers/net/virtio/virtio_rxtx.c
+++ b/drivers/net/virtio/virtio_rxtx.c
@@ -271,13 +271,10 @@ virtqueue_enqueue_refill_inorder(struct virtqueue *vq,
dxp->cookie = (void *)cookies[i];
dxp->ndescs = 1;
 
-   start_dp[idx].addr =
-   VIRTIO_MBUF_ADDR(cookies[i], vq) +
-   RTE_PKTMBUF_HEADROOM - hw->vtnet_hdr_size;
-   start_dp[idx].len =
-   cookies[i]->buf_len -
-   RTE_PKTMBUF_HEADROOM +
-   hw->vtnet_hdr_size;
+   start_dp[idx].addr = cookies[i]->buf_iova +
+   RTE_PKTMBUF_HEADROOM - hw->vtnet_hdr_size;
+   start_dp[idx].len = cookies[i]->buf_len -
+   RTE_PKTMBUF_HEADROOM + hw->vtnet_hdr_size;
start_dp[idx].flags =  VRING_DESC_F_WRITE;
 
vq_update_avail_ring(vq, idx);
@@ -313,12 +310,10 @@ virtqueue_enqueue_recv_refill(struct virtqueue *vq, 
struct rte_mbuf **cookie,
dxp->cookie = (void *)cookie[i];
dxp->ndescs = 1;
 
-   start_dp[idx].addr =
-   VIRTIO_MBUF_ADDR(cookie[i], vq) +
+   start_dp[idx].addr = cookie[i]->buf_iova +
RTE_PKTMBUF_HEADROOM - hw->vtnet_hdr_size;
-   start_dp[idx].len =
-   cookie[i]->buf_len - RTE_PKTMBUF_HEADROOM +
-   hw->vtnet_hdr_size;
+   start_dp[idx].len = cookie[i]->buf_len -
+   RTE_PKTMBUF_HEADROOM + hw->vtnet_hdr_size;
start_dp[idx].flags = VRING_DESC_F_WRITE;
vq->vq_desc_head_idx = start_dp[idx].next;
vq_update_avail_ring(vq, idx);
@@ -355,10 +350,10 @@ virtqueue_enqueue_recv_refill_packed(struct virtqueue *vq,
dxp->cookie = (void *)cookie[i];
dxp->ndescs = 1;
 
-   start_dp[idx].addr = VIRTIO_MBUF_ADDR(cookie[i], vq) +
-   RTE_PKTMBUF_HEADROOM - hw->vtnet_hdr_size;
-   start_dp[idx].len = cookie[i]->buf_len - RTE_PKTMBUF_HEADROOM
-   + hw->vtnet_hdr_size;
+   start_dp[idx].addr = cookie[i]->buf_iova +
+   RTE_PKTMBUF_HEADROOM - hw->vtnet_hdr_size;
+   start_dp[idx].len = cookie[i]->buf_len -
+   RTE_PKTMBUF_HEADROOM + hw->vtnet_hdr_size;
 
vq->vq_desc_head_idx = dxp->next;
if (vq->vq_desc_head_idx == VQ_RING_DESC_CHAIN_

[dpdk-dev] [PATCH v3 11/44] net/virtio: store PCI type in Virtio device metadata

[Maxime Coquelin]

Going further in making the Virtio ethdev layer bus agnostic,
this patch adds a boolean in the Virtio PCI device metadata.

Signed-off-by: Maxime Coquelin 
Reviewed-by: Chenbo Xia 
Reviewed-by: David Marchand 
---
 drivers/net/virtio/virtio_pci.c| 18 +++---
 drivers/net/virtio/virtio_pci.h|  3 ++-
 drivers/net/virtio/virtio_pci_ethdev.c | 12 +++-
 3 files changed, 20 insertions(+), 13 deletions(-)

diff --git a/drivers/net/virtio/virtio_pci.c b/drivers/net/virtio/virtio_pci.c
index 556be1e3da..6d9c712fd5 100644
--- a/drivers/net/virtio/virtio_pci.c
+++ b/drivers/net/virtio/virtio_pci.c
@@ -751,8 +751,10 @@ virtio_read_caps(struct rte_pci_device *dev, struct 
virtio_hw *hw)
  * Return 0 on success.
  */
 int
-vtpci_init(struct rte_pci_device *dev, struct virtio_hw *hw)
+vtpci_init(struct rte_pci_device *pci_dev, struct virtio_pci_dev *dev)
 {
+   struct virtio_hw *hw = &dev->hw;
+
RTE_BUILD_BUG_ON(offsetof(struct virtio_pci_dev, hw) != 0);
 
/*
@@ -760,19 +762,20 @@ vtpci_init(struct rte_pci_device *dev, struct virtio_hw 
*hw)
 * only on modern pci device. If failed, we fallback to legacy
 * virtio handling.
 */
-   if (virtio_read_caps(dev, hw) == 0) {
+   if (virtio_read_caps(pci_dev, hw) == 0) {
PMD_INIT_LOG(INFO, "modern virtio pci detected.");
virtio_hw_internal[hw->port_id].vtpci_ops = &modern_ops;
hw->bus_type = VIRTIO_BUS_PCI_MODERN;
+   dev->modern = true;
goto msix_detect;
}
 
PMD_INIT_LOG(INFO, "trying with legacy virtio pci.");
-   if (rte_pci_ioport_map(dev, 0, VTPCI_IO(hw)) < 0) {
-   rte_pci_unmap_device(dev);
-   if (dev->kdrv == RTE_PCI_KDRV_UNKNOWN &&
-   (!dev->device.devargs ||
-dev->device.devargs->bus !=
+   if (rte_pci_ioport_map(pci_dev, 0, VTPCI_IO(hw)) < 0) {
+   rte_pci_unmap_device(pci_dev);
+   if (pci_dev->kdrv == RTE_PCI_KDRV_UNKNOWN &&
+   (!pci_dev->device.devargs ||
+pci_dev->device.devargs->bus !=
 rte_bus_find_by_name("pci"))) {
PMD_INIT_LOG(INFO,
"skip kernel managed virtio device.");
@@ -783,6 +786,7 @@ vtpci_init(struct rte_pci_device *dev, struct virtio_hw *hw)
 
virtio_hw_internal[hw->port_id].vtpci_ops = &legacy_ops;
hw->bus_type = VIRTIO_BUS_PCI_LEGACY;
+   dev->modern = false;
 
 msix_detect:
VTPCI_OPS(hw)->intr_detect(hw);
diff --git a/drivers/net/virtio/virtio_pci.h b/drivers/net/virtio/virtio_pci.h
index b29bbb8074..4c22692414 100644
--- a/drivers/net/virtio/virtio_pci.h
+++ b/drivers/net/virtio/virtio_pci.h
@@ -292,6 +292,7 @@ struct virtio_hw {
 
 struct virtio_pci_dev {
struct virtio_hw hw;
+   bool modern;
 };
 
 #define virtio_pci_get_dev(hwp) container_of(hwp, struct virtio_pci_dev, hw)
@@ -371,7 +372,7 @@ vtpci_packed_queue(struct virtio_hw *hw)
 /*
  * Function declaration from virtio_pci.c
  */
-int vtpci_init(struct rte_pci_device *dev, struct virtio_hw *hw);
+int vtpci_init(struct rte_pci_device *pci_dev, struct virtio_pci_dev *dev);
 void vtpci_reset(struct virtio_hw *);
 
 void vtpci_reinit_complete(struct virtio_hw *);
diff --git a/drivers/net/virtio/virtio_pci_ethdev.c 
b/drivers/net/virtio/virtio_pci_ethdev.c
index 045b134ef2..076a5dbced 100644
--- a/drivers/net/virtio/virtio_pci_ethdev.c
+++ b/drivers/net/virtio/virtio_pci_ethdev.c
@@ -39,9 +39,11 @@ static const struct rte_pci_id pci_id_virtio_map[] = {
  * could have the PCI initiated correctly.
  */
 static int
-virtio_remap_pci(struct rte_pci_device *pci_dev, struct virtio_hw *hw)
+virtio_remap_pci(struct rte_pci_device *pci_dev, struct virtio_pci_dev *dev)
 {
-   if (hw->bus_type == VIRTIO_BUS_PCI_MODERN) {
+   struct virtio_hw *hw = &dev->hw;
+
+   if (dev->modern) {
/*
 * We don't have to re-parse the PCI config space, since
 * rte_pci_map_device() makes sure the mapped address
@@ -57,7 +59,7 @@ virtio_remap_pci(struct rte_pci_device *pci_dev, struct 
virtio_hw *hw)
PMD_INIT_LOG(DEBUG, "failed to map pci device!");
return -1;
}
-   } else if (hw->bus_type == VIRTIO_BUS_PCI_LEGACY) {
+   } else {
if (rte_pci_ioport_map(pci_dev, 0, VTPCI_IO(hw)) < 0)
return -1;
}
@@ -76,13 +78,13 @@ eth_virtio_pci_init(struct rte_eth_dev *eth_dev)
VTPCI_DEV(hw) = pci_dev;
 
if (rte_eal_process_type() == RTE_PROC_PRIMARY) {
-   ret = vtpci_init(RTE_ETH_DEV_TO_PCI(eth_dev), hw);
+   ret = vtpci_init(RTE_ETH_DEV_TO_PCI(eth_dev), dev);
if (ret) {
PMD_INIT_LOG(ERR, "Failed to init PCI device\n");
return -1;
 

[dpdk-dev] [PATCH v3 12/44] net/virtio: add callback for device closing

[Maxime Coquelin]

This patch introduces a new callback for device closing,
making virtio_dev_close() bus-agnostic.

Signed-off-by: Maxime Coquelin 
Reviewed-by: Chenbo Xia 
Reviewed-by: David Marchand 
---
 drivers/net/virtio/meson.build  |  2 --
 drivers/net/virtio/virtio_ethdev.c  | 13 +
 drivers/net/virtio/virtio_pci.c | 25 +
 drivers/net/virtio/virtio_pci.h |  2 ++
 drivers/net/virtio/virtio_user_ethdev.c | 11 +++
 5 files changed, 39 insertions(+), 14 deletions(-)

diff --git a/drivers/net/virtio/meson.build b/drivers/net/virtio/meson.build
index 07e085376b..f2873d6180 100644
--- a/drivers/net/virtio/meson.build
+++ b/drivers/net/virtio/meson.build
@@ -44,8 +44,6 @@ elif arch_subdir == 'arm' and 
host_machine.cpu_family().startswith('aarch64')
 endif
 
 if is_linux
-   dpdk_conf.set('RTE_VIRTIO_USER', 1)
-
sources += files('virtio_user_ethdev.c',
'virtio_user/vhost_kernel.c',
'virtio_user/vhost_kernel_tap.c',
diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index fb789460e8..84edcd4724 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -718,18 +718,7 @@ virtio_dev_close(struct rte_eth_dev *dev)
virtio_dev_free_mbufs(dev);
virtio_free_queues(hw);
 
-#ifdef RTE_VIRTIO_USER
-   if (hw->bus_type == VIRTIO_BUS_USER)
-   virtio_user_dev_uninit(dev->data->dev_private);
-   else
-#endif
-   if (dev->device) {
-   rte_pci_unmap_device(RTE_ETH_DEV_TO_PCI(dev));
-   if (hw->bus_type == VIRTIO_BUS_PCI_LEGACY)
-   rte_pci_ioport_unmap(VTPCI_IO(hw));
-   }
-
-   return 0;
+   return VTPCI_OPS(hw)->dev_close(hw);
 }
 
 static int
diff --git a/drivers/net/virtio/virtio_pci.c b/drivers/net/virtio/virtio_pci.c
index 6d9c712fd5..ea4ab381a6 100644
--- a/drivers/net/virtio/virtio_pci.c
+++ b/drivers/net/virtio/virtio_pci.c
@@ -297,6 +297,17 @@ legacy_intr_detect(struct virtio_hw *hw)
hw->use_msix = vtpci_msix_detect(VTPCI_DEV(hw));
 }
 
+static int
+legacy_dev_close(struct virtio_hw *hw)
+{
+   struct virtio_pci_dev *dev = virtio_pci_get_dev(hw);
+
+   rte_pci_unmap_device(dev->pci_dev);
+   rte_pci_ioport_unmap(VTPCI_IO(hw));
+
+   return 0;
+}
+
 const struct virtio_pci_ops legacy_ops = {
.read_dev_cfg   = legacy_read_dev_config,
.write_dev_cfg  = legacy_write_dev_config,
@@ -312,6 +323,7 @@ const struct virtio_pci_ops legacy_ops = {
.del_queue  = legacy_del_queue,
.notify_queue   = legacy_notify_queue,
.intr_detect= legacy_intr_detect,
+   .dev_close  = legacy_dev_close,
 };
 
 static inline void
@@ -511,6 +523,16 @@ modern_intr_detect(struct virtio_hw *hw)
hw->use_msix = vtpci_msix_detect(VTPCI_DEV(hw));
 }
 
+static int
+modern_dev_close(struct virtio_hw *hw)
+{
+   struct virtio_pci_dev *dev = virtio_pci_get_dev(hw);
+
+   rte_pci_unmap_device(dev->pci_dev);
+
+   return 0;
+}
+
 const struct virtio_pci_ops modern_ops = {
.read_dev_cfg   = modern_read_dev_config,
.write_dev_cfg  = modern_write_dev_config,
@@ -526,6 +548,7 @@ const struct virtio_pci_ops modern_ops = {
.del_queue  = modern_del_queue,
.notify_queue   = modern_notify_queue,
.intr_detect= modern_intr_detect,
+   .dev_close  = modern_dev_close,
 };
 
 
@@ -757,6 +780,8 @@ vtpci_init(struct rte_pci_device *pci_dev, struct 
virtio_pci_dev *dev)
 
RTE_BUILD_BUG_ON(offsetof(struct virtio_pci_dev, hw) != 0);
 
+   dev->pci_dev = pci_dev;
+
/*
 * Try if we can succeed reading virtio pci caps, which exists
 * only on modern pci device. If failed, we fallback to legacy
diff --git a/drivers/net/virtio/virtio_pci.h b/drivers/net/virtio/virtio_pci.h
index 4c22692414..0515bbb247 100644
--- a/drivers/net/virtio/virtio_pci.h
+++ b/drivers/net/virtio/virtio_pci.h
@@ -240,6 +240,7 @@ struct virtio_pci_ops {
void (*del_queue)(struct virtio_hw *hw, struct virtqueue *vq);
void (*notify_queue)(struct virtio_hw *hw, struct virtqueue *vq);
void (*intr_detect)(struct virtio_hw *hw);
+   int (*dev_close)(struct virtio_hw *hw);
 };
 
 struct virtio_net_config;
@@ -292,6 +293,7 @@ struct virtio_hw {
 
 struct virtio_pci_dev {
struct virtio_hw hw;
+   struct rte_pci_device *pci_dev;
bool modern;
 };
 
diff --git a/drivers/net/virtio/virtio_user_ethdev.c 
b/drivers/net/virtio/virtio_user_ethdev.c
index 241fe373b9..3cbf310c03 100644
--- a/drivers/net/virtio/virtio_user_ethdev.c
+++ b/drivers/net/virtio/virtio_user_ethdev.c
@@ -462,6 +462,16 @@ virtio_user_notify_queue(struct virtio_hw *hw, struct 
virtqueue *vq)
strerror(errno));
 }
 
+static int
+virtio_user_dev_close(struct virtio_hw *hw)
+{
+   struct virtio_user_dev *dev = virtio_user_get_dev(hw)

[dpdk-dev] [PATCH v3 13/44] net/virtio: validate features at bus level

[Maxime Coquelin]

This patch provides a new callback for the bus type
to validate negotiated features are compatible with it.

Only user for now is PCI modern bus type, which implies
that the device supports Virtio 1.0+.

Signed-off-by: Maxime Coquelin 
Reviewed-by: Chenbo Xia 
---
 drivers/net/virtio/virtio_ethdev.c  | 11 ---
 drivers/net/virtio/virtio_pci.c | 19 +++
 drivers/net/virtio/virtio_pci.h |  1 +
 drivers/net/virtio/virtio_user_ethdev.c |  7 +++
 4 files changed, 31 insertions(+), 7 deletions(-)

diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index 84edcd4724..72f527144e 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -1330,17 +1330,14 @@ virtio_negotiate_features(struct virtio_hw *hw, 
uint64_t req_features)
PMD_INIT_LOG(DEBUG, "features after negotiate = %" PRIx64,
hw->guest_features);
 
-   if (hw->bus_type == VIRTIO_BUS_PCI_MODERN && !vtpci_with_feature(hw, 
VIRTIO_F_VERSION_1)) {
-   PMD_INIT_LOG(ERR,
-   "VIRTIO_F_VERSION_1 features is not enabled.");
+   if (VTPCI_OPS(hw)->features_ok(hw) < 0)
return -1;
-   }
 
-   if (hw->bus_type == VIRTIO_BUS_PCI_MODERN || hw->bus_type == 
VIRTIO_BUS_USER) {
+   if (vtpci_with_feature(hw, VIRTIO_F_VERSION_1)) {
vtpci_set_status(hw, VIRTIO_CONFIG_STATUS_FEATURES_OK);
+
if (!(vtpci_get_status(hw) & VIRTIO_CONFIG_STATUS_FEATURES_OK)) 
{
-   PMD_INIT_LOG(ERR,
-   "failed to set FEATURES_OK status!");
+   PMD_INIT_LOG(ERR, "Failed to set FEATURES_OK status!");
return -1;
}
}
diff --git a/drivers/net/virtio/virtio_pci.c b/drivers/net/virtio/virtio_pci.c
index ea4ab381a6..1c8d4b7ddd 100644
--- a/drivers/net/virtio/virtio_pci.c
+++ b/drivers/net/virtio/virtio_pci.c
@@ -201,6 +201,12 @@ legacy_set_features(struct virtio_hw *hw, uint64_t 
features)
VIRTIO_PCI_GUEST_FEATURES);
 }
 
+static int
+legacy_features_ok(struct virtio_hw *hw __rte_unused)
+{
+   return 0;
+}
+
 static uint8_t
 legacy_get_status(struct virtio_hw *hw)
 {
@@ -315,6 +321,7 @@ const struct virtio_pci_ops legacy_ops = {
.set_status = legacy_set_status,
.get_features   = legacy_get_features,
.set_features   = legacy_set_features,
+   .features_ok= legacy_features_ok,
.get_isr= legacy_get_isr,
.set_config_irq = legacy_set_config_irq,
.set_queue_irq  = legacy_set_queue_irq,
@@ -389,6 +396,17 @@ modern_set_features(struct virtio_hw *hw, uint64_t 
features)
&hw->common_cfg->guest_feature);
 }
 
+static int
+modern_features_ok(struct virtio_hw *hw)
+{
+   if (!vtpci_with_feature(hw, VIRTIO_F_VERSION_1)) {
+   PMD_INIT_LOG(ERR, "Version 1+ required with modern devices\n");
+   return -1;
+   }
+
+   return 0;
+}
+
 static uint8_t
 modern_get_status(struct virtio_hw *hw)
 {
@@ -540,6 +558,7 @@ const struct virtio_pci_ops modern_ops = {
.set_status = modern_set_status,
.get_features   = modern_get_features,
.set_features   = modern_set_features,
+   .features_ok= modern_features_ok,
.get_isr= modern_get_isr,
.set_config_irq = modern_set_config_irq,
.set_queue_irq  = modern_set_queue_irq,
diff --git a/drivers/net/virtio/virtio_pci.h b/drivers/net/virtio/virtio_pci.h
index 0515bbb247..6aacc942fc 100644
--- a/drivers/net/virtio/virtio_pci.h
+++ b/drivers/net/virtio/virtio_pci.h
@@ -227,6 +227,7 @@ struct virtio_pci_ops {
 
uint64_t (*get_features)(struct virtio_hw *hw);
void (*set_features)(struct virtio_hw *hw, uint64_t features);
+   int  (*features_ok)(struct virtio_hw *hw);
 
uint8_t (*get_isr)(struct virtio_hw *hw);
 
diff --git a/drivers/net/virtio/virtio_user_ethdev.c 
b/drivers/net/virtio/virtio_user_ethdev.c
index 3cbf310c03..bf958de571 100644
--- a/drivers/net/virtio/virtio_user_ethdev.c
+++ b/drivers/net/virtio/virtio_user_ethdev.c
@@ -327,6 +327,12 @@ virtio_user_set_features(struct virtio_hw *hw, uint64_t 
features)
dev->features = features & dev->device_features;
 }
 
+static int
+virtio_user_features_ok(struct virtio_hw *hw __rte_unused)
+{
+   return 0;
+}
+
 static uint8_t
 virtio_user_get_isr(struct virtio_hw *hw __rte_unused)
 {
@@ -479,6 +485,7 @@ const struct virtio_pci_ops virtio_user_ops = {
.set_status = virtio_user_set_status,
.get_features   = virtio_user_get_features,
.set_features   = virtio_user_set_features,
+   .features_ok= virtio_user_features_ok,
.get_isr= virtio_user_get_isr,
.set_config_irq = virtio_user_set_config_irq,
.set_queue_irq  = virtio_user_set_queue_irq,
-- 
2.29.2

[dpdk-dev] [PATCH v3 14/44] net/virtio: remove bus type enum

[Maxime Coquelin]

Bus type awareness at the generic ethdev level is no
more needed as previous patches have made it bus-agnostic.

This patch removes it from struct virtio_hw.

Signed-off-by: Maxime Coquelin 
Reviewed-by: Chenbo Xia 
Reviewed-by: David Marchand 
---
 drivers/net/virtio/virtio_ethdev.c  | 18 --
 drivers/net/virtio/virtio_pci.c |  2 --
 drivers/net/virtio/virtio_pci.h |  8 
 drivers/net/virtio/virtio_pci_ethdev.c  |  7 ++-
 drivers/net/virtio/virtio_user_ethdev.c |  5 -
 5 files changed, 10 insertions(+), 30 deletions(-)

diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index 72f527144e..c46fe4adf6 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -1798,23 +1798,6 @@ virtio_init_device(struct rte_eth_dev *eth_dev, uint64_t 
req_features)
return 0;
 }
 
-
-static void
-virtio_set_vtpci_ops(struct virtio_hw *hw)
-{
-#ifdef RTE_VIRTIO_USER
-   if (hw->bus_type == VIRTIO_BUS_USER)
-   VTPCI_OPS(hw) = &virtio_user_ops;
-   else
-#endif
-   if (hw->bus_type == VIRTIO_BUS_PCI_MODERN)
-   VTPCI_OPS(hw) = &modern_ops;
-   else if (hw->bus_type == VIRTIO_BUS_PCI_LEGACY)
-   VTPCI_OPS(hw) = &legacy_ops;
-
-   return;
-}
-
 /*
  * This function is based on probe() function in virtio_pci.c
  * It returns 0 on success.
@@ -1840,7 +1823,6 @@ eth_virtio_dev_init(struct rte_eth_dev *eth_dev)
eth_dev->rx_descriptor_done = virtio_dev_rx_queue_done;
 
if (rte_eal_process_type() == RTE_PROC_SECONDARY) {
-   virtio_set_vtpci_ops(hw);
set_rxtx_funcs(eth_dev);
return 0;
}
diff --git a/drivers/net/virtio/virtio_pci.c b/drivers/net/virtio/virtio_pci.c
index 1c8d4b7ddd..20599774a7 100644
--- a/drivers/net/virtio/virtio_pci.c
+++ b/drivers/net/virtio/virtio_pci.c
@@ -809,7 +809,6 @@ vtpci_init(struct rte_pci_device *pci_dev, struct 
virtio_pci_dev *dev)
if (virtio_read_caps(pci_dev, hw) == 0) {
PMD_INIT_LOG(INFO, "modern virtio pci detected.");
virtio_hw_internal[hw->port_id].vtpci_ops = &modern_ops;
-   hw->bus_type = VIRTIO_BUS_PCI_MODERN;
dev->modern = true;
goto msix_detect;
}
@@ -829,7 +828,6 @@ vtpci_init(struct rte_pci_device *pci_dev, struct 
virtio_pci_dev *dev)
}
 
virtio_hw_internal[hw->port_id].vtpci_ops = &legacy_ops;
-   hw->bus_type = VIRTIO_BUS_PCI_LEGACY;
dev->modern = false;
 
 msix_detect:
diff --git a/drivers/net/virtio/virtio_pci.h b/drivers/net/virtio/virtio_pci.h
index 6aacc942fc..2cede4a100 100644
--- a/drivers/net/virtio/virtio_pci.h
+++ b/drivers/net/virtio/virtio_pci.h
@@ -246,15 +246,7 @@ struct virtio_pci_ops {
 
 struct virtio_net_config;
 
-enum virtio_bus_type {
-   VIRTIO_BUS_UNKNOWN,
-   VIRTIO_BUS_PCI_LEGACY,
-   VIRTIO_BUS_PCI_MODERN,
-   VIRTIO_BUS_USER,
-};
-
 struct virtio_hw {
-   enum virtio_bus_type bus_type;
struct virtnet_ctl *cvq;
uint64_treq_guest_features;
uint64_tguest_features;
diff --git a/drivers/net/virtio/virtio_pci_ethdev.c 
b/drivers/net/virtio/virtio_pci_ethdev.c
index 076a5dbced..8cc6561914 100644
--- a/drivers/net/virtio/virtio_pci_ethdev.c
+++ b/drivers/net/virtio/virtio_pci_ethdev.c
@@ -84,6 +84,11 @@ eth_virtio_pci_init(struct rte_eth_dev *eth_dev)
return -1;
}
} else {
+   if (dev->modern)
+   VTPCI_OPS(hw) = &modern_ops;
+   else
+   VTPCI_OPS(hw) = &legacy_ops;
+
ret = virtio_remap_pci(RTE_ETH_DEV_TO_PCI(eth_dev), dev);
if (ret < 0) {
PMD_INIT_LOG(ERR, "Failed to remap PCI device\n");
@@ -105,7 +110,7 @@ eth_virtio_pci_init(struct rte_eth_dev *eth_dev)
 
 err_unmap:
rte_pci_unmap_device(RTE_ETH_DEV_TO_PCI(eth_dev));
-   if (hw->bus_type == VIRTIO_BUS_PCI_LEGACY)
+   if (!dev->modern)
rte_pci_ioport_unmap(VTPCI_IO(hw));
 
return ret;
diff --git a/drivers/net/virtio/virtio_user_ethdev.c 
b/drivers/net/virtio/virtio_user_ethdev.c
index bf958de571..61880a8e02 100644
--- a/drivers/net/virtio/virtio_user_ethdev.c
+++ b/drivers/net/virtio/virtio_user_ethdev.c
@@ -641,7 +641,6 @@ virtio_user_eth_dev_alloc(struct rte_vdev_device *vdev)
 * Here just pretend that we support msix.
 */
hw->use_msix = 1;
-   hw->bus_type = VIRTIO_BUS_USER;
hw->use_vec_rx = 0;
hw->use_vec_tx = 0;
hw->use_inorder_rx = 0;
@@ -691,6 +690,10 @@ virtio_user_pmd_probe(struct rte_vdev_device *vdev)
return -1;
}
 
+   dev = eth_dev->data->dev_private;
+   hw = &dev->hw;
+   VTPCI_OPS(hw) = &virtio_user_ops;
+
if (eth_virtio_dev_init(eth_dev)

[dpdk-dev] [PATCH v3 15/44] net/virtio: move PCI-specific fields to PCI device

[Maxime Coquelin]

This patch moves the fields from virtio_hw structure that
are PCI-specific to virtio_pci_dev_struct.

Signed-off-by: Maxime Coquelin 
Reviewed-by: Chenbo Xia 
Reviewed-by: David Marchand 
---
 drivers/net/virtio/virtio_pci.c | 139 ++--
 drivers/net/virtio/virtio_pci.h |  10 +--
 2 files changed, 85 insertions(+), 64 deletions(-)

diff --git a/drivers/net/virtio/virtio_pci.c b/drivers/net/virtio/virtio_pci.c
index 20599774a7..b347e5fbc0 100644
--- a/drivers/net/virtio/virtio_pci.c
+++ b/drivers/net/virtio/virtio_pci.c
@@ -344,18 +344,19 @@ static void
 modern_read_dev_config(struct virtio_hw *hw, size_t offset,
   void *dst, int length)
 {
+   struct virtio_pci_dev *dev = virtio_pci_get_dev(hw);
int i;
uint8_t *p;
uint8_t old_gen, new_gen;
 
do {
-   old_gen = rte_read8(&hw->common_cfg->config_generation);
+   old_gen = rte_read8(&dev->common_cfg->config_generation);
 
p = dst;
for (i = 0;  i < length; i++)
-   *p++ = rte_read8((uint8_t *)hw->dev_cfg + offset + i);
+   *p++ = rte_read8((uint8_t *)dev->dev_cfg + offset + i);
 
-   new_gen = rte_read8(&hw->common_cfg->config_generation);
+   new_gen = rte_read8(&dev->common_cfg->config_generation);
} while (old_gen != new_gen);
 }
 
@@ -363,23 +364,25 @@ static void
 modern_write_dev_config(struct virtio_hw *hw, size_t offset,
const void *src, int length)
 {
+   struct virtio_pci_dev *dev = virtio_pci_get_dev(hw);
int i;
const uint8_t *p = src;
 
for (i = 0;  i < length; i++)
-   rte_write8((*p++), (((uint8_t *)hw->dev_cfg) + offset + i));
+   rte_write8((*p++), (((uint8_t *)dev->dev_cfg) + offset + i));
 }
 
 static uint64_t
 modern_get_features(struct virtio_hw *hw)
 {
+   struct virtio_pci_dev *dev = virtio_pci_get_dev(hw);
uint32_t features_lo, features_hi;
 
-   rte_write32(0, &hw->common_cfg->device_feature_select);
-   features_lo = rte_read32(&hw->common_cfg->device_feature);
+   rte_write32(0, &dev->common_cfg->device_feature_select);
+   features_lo = rte_read32(&dev->common_cfg->device_feature);
 
-   rte_write32(1, &hw->common_cfg->device_feature_select);
-   features_hi = rte_read32(&hw->common_cfg->device_feature);
+   rte_write32(1, &dev->common_cfg->device_feature_select);
+   features_hi = rte_read32(&dev->common_cfg->device_feature);
 
return ((uint64_t)features_hi << 32) | features_lo;
 }
@@ -387,13 +390,15 @@ modern_get_features(struct virtio_hw *hw)
 static void
 modern_set_features(struct virtio_hw *hw, uint64_t features)
 {
-   rte_write32(0, &hw->common_cfg->guest_feature_select);
+   struct virtio_pci_dev *dev = virtio_pci_get_dev(hw);
+
+   rte_write32(0, &dev->common_cfg->guest_feature_select);
rte_write32(features & ((1ULL << 32) - 1),
-   &hw->common_cfg->guest_feature);
+   &dev->common_cfg->guest_feature);
 
-   rte_write32(1, &hw->common_cfg->guest_feature_select);
+   rte_write32(1, &dev->common_cfg->guest_feature_select);
rte_write32(features >> 32,
-   &hw->common_cfg->guest_feature);
+   &dev->common_cfg->guest_feature);
 }
 
 static int
@@ -410,46 +415,59 @@ modern_features_ok(struct virtio_hw *hw)
 static uint8_t
 modern_get_status(struct virtio_hw *hw)
 {
-   return rte_read8(&hw->common_cfg->device_status);
+   struct virtio_pci_dev *dev = virtio_pci_get_dev(hw);
+
+   return rte_read8(&dev->common_cfg->device_status);
 }
 
 static void
 modern_set_status(struct virtio_hw *hw, uint8_t status)
 {
-   rte_write8(status, &hw->common_cfg->device_status);
+   struct virtio_pci_dev *dev = virtio_pci_get_dev(hw);
+
+   rte_write8(status, &dev->common_cfg->device_status);
 }
 
 static uint8_t
 modern_get_isr(struct virtio_hw *hw)
 {
-   return rte_read8(hw->isr);
+   struct virtio_pci_dev *dev = virtio_pci_get_dev(hw);
+
+   return rte_read8(dev->isr);
 }
 
 static uint16_t
 modern_set_config_irq(struct virtio_hw *hw, uint16_t vec)
 {
-   rte_write16(vec, &hw->common_cfg->msix_config);
-   return rte_read16(&hw->common_cfg->msix_config);
+   struct virtio_pci_dev *dev = virtio_pci_get_dev(hw);
+
+   rte_write16(vec, &dev->common_cfg->msix_config);
+   return rte_read16(&dev->common_cfg->msix_config);
 }
 
 static uint16_t
 modern_set_queue_irq(struct virtio_hw *hw, struct virtqueue *vq, uint16_t vec)
 {
-   rte_write16(vq->vq_queue_index, &hw->common_cfg->queue_select);
-   rte_write16(vec, &hw->common_cfg->queue_msix_vector);
-   return rte_read16(&hw->common_cfg->queue_msix_vector);
+   struct virtio_pci_dev *dev = virtio_pci_get_dev(hw);
+
+   rte_write16(vq->vq_queue_index, &dev->common_cfg->queue_select);
+   rte

[dpdk-dev] [PATCH v3 16/44] net/virtio: pack virtio HW struct

[Maxime Coquelin]

This patch improves the virtio_hw struct packing,
going from 88 down to 80 bytes with a 6 bytes hole in
the end of the first cacheline. Fields only used in the
slow path are placed in the end, so that hot path only
uses the first cacheline.

The patch also changes booleans fields to uint8_t type, and
fix inconsistencies in their assignments.

Signed-off-by: Maxime Coquelin 
Reviewed-by: Chenbo Xia 
---
 drivers/net/virtio/virtio_ethdev.c | 12 
 drivers/net/virtio/virtio_pci.h| 45 +++---
 drivers/net/virtio/virtqueue.h |  2 +-
 3 files changed, 29 insertions(+), 30 deletions(-)

diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index c46fe4adf6..c1d7b14dda 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -699,7 +699,7 @@ virtio_dev_close(struct rte_eth_dev *dev)
 
if (!hw->opened)
return 0;
-   hw->opened = false;
+   hw->opened = 0;
 
/* reset the NIC */
if (dev->data->dev_flags & RTE_ETH_DEV_INTR_LSC)
@@ -1864,7 +1864,7 @@ eth_virtio_dev_init(struct rte_eth_dev *eth_dev)
}
}
 
-   hw->opened = true;
+   hw->opened = 1;
 
return 0;
 
@@ -1973,7 +1973,7 @@ virtio_dev_devargs_parse(struct rte_devargs *devargs, 
uint32_t *speed, int *vect
return ret;
 }
 
-static bool
+static uint8_t
 rx_offload_enabled(struct virtio_hw *hw)
 {
return vtpci_with_feature(hw, VIRTIO_NET_F_GUEST_CSUM) ||
@@ -1981,7 +1981,7 @@ rx_offload_enabled(struct virtio_hw *hw)
vtpci_with_feature(hw, VIRTIO_NET_F_GUEST_TSO6);
 }
 
-static bool
+static uint8_t
 tx_offload_enabled(struct virtio_hw *hw)
 {
return vtpci_with_feature(hw, VIRTIO_NET_F_CSUM) ||
@@ -2267,7 +2267,7 @@ virtio_dev_start(struct rte_eth_dev *dev)
}
 
set_rxtx_funcs(dev);
-   hw->started = true;
+   hw->started = 1;
 
/* Initialize Link state */
virtio_dev_link_update(dev, 0);
@@ -2336,7 +2336,7 @@ virtio_dev_stop(struct rte_eth_dev *dev)
rte_spinlock_lock(&hw->state_lock);
if (!hw->started)
goto out_unlock;
-   hw->started = false;
+   hw->started = 0;
 
if (intr_conf->lsc || intr_conf->rxq) {
virtio_intr_disable(dev);
diff --git a/drivers/net/virtio/virtio_pci.h b/drivers/net/virtio/virtio_pci.h
index 4f3690032b..15f8144fc6 100644
--- a/drivers/net/virtio/virtio_pci.h
+++ b/drivers/net/virtio/virtio_pci.h
@@ -247,26 +247,25 @@ struct virtio_pci_ops {
 struct virtio_net_config;
 
 struct virtio_hw {
-   struct virtnet_ctl *cvq;
-   uint64_treq_guest_features;
-   uint64_tguest_features;
-   uint32_tmax_queue_pairs;
-   boolstarted;
-   uint16_tmax_mtu;
-   uint16_tvtnet_hdr_size;
-   uint8_t vlan_strip;
-   uint8_t use_msix;
-   uint8_t use_vec_rx;
-   uint8_t use_vec_tx;
-   uint8_t use_inorder_rx;
-   uint8_t use_inorder_tx;
-   uint8_t weak_barriers;
-   boolhas_tx_offload;
-   boolhas_rx_offload;
-   uint16_tport_id;
-   uint8_t mac_addr[RTE_ETHER_ADDR_LEN];
-   uint32_tspeed;  /* link speed in MB */
-   uint8_t duplex;
+   struct virtqueue **vqs;
+   uint64_t guest_features;
+   uint16_t vtnet_hdr_size;
+   uint8_t started;
+   uint8_t weak_barriers;
+   uint8_t vlan_strip;
+   uint8_t has_tx_offload;
+   uint8_t has_rx_offload;
+   uint8_t use_vec_rx;
+   uint8_t use_vec_tx;
+   uint8_t use_inorder_rx;
+   uint8_t use_inorder_tx;
+   uint8_t opened;
+   uint16_t port_id;
+   uint8_t mac_addr[RTE_ETHER_ADDR_LEN];
+   uint32_t speed;  /* link speed in MB */
+   uint8_t duplex;
+   uint8_t use_msix;
+   uint16_t max_mtu;
/*
 * App management thread and virtio interrupt handler thread
 * both can change device state, this lock is meant to avoid
@@ -274,9 +273,9 @@ struct virtio_hw {
 */
rte_spinlock_t state_lock;
struct rte_mbuf **inject_pkts;
-   boolopened;
-
-   struct virtqueue **vqs;
+   uint16_t max_queue_pairs;
+   uint64_t req_guest_features;
+   struct virtnet_ctl *cvq;
 };
 
 struct virtio_pci_dev {
diff --git a/drivers/net/virtio/virtqueue.h b/drivers/net/virtio/virtqueue.h
index 7611317581..3a9ce29069 100644
--- a/drivers/net/virtio/virtqueue.h
+++ b/drivers/net/virtio/virtqueue.h
@@ -615,7 +615,7 @@ virtqueue_notify(struct virtqueue *vq)
 static inline void
 virtqueue_xmit_offload(struct virtio_net_hdr *hdr,
struct rte_mbuf *cookie,
-   bool offload)
+   uint8_t offload)
 {
if (offload) {
if (cookie->ol_flags & PKT_TX_TCP_SEG)
-- 
2.29.2

[dpdk-dev] [PATCH v3 17/44] net/virtio: move legacy IO to Virtio PCI

[Maxime Coquelin]

This patch moves Virtio PCI legacy IO handling to
virtio_pci.c. Two functions are created so that
virtio_pci_ethdev does not have to care about it.

Signed-off-by: Maxime Coquelin 
Reviewed-by: Chenbo Xia 
Reviewed-by: David Marchand 
---
 drivers/net/virtio/virtio_pci.c| 28 --
 drivers/net/virtio/virtio_pci.h|  9 +++--
 drivers/net/virtio/virtio_pci_ethdev.c |  6 ++
 3 files changed, 31 insertions(+), 12 deletions(-)

diff --git a/drivers/net/virtio/virtio_pci.c b/drivers/net/virtio/virtio_pci.c
index b347e5fbc0..3fe0631a30 100644
--- a/drivers/net/virtio/virtio_pci.c
+++ b/drivers/net/virtio/virtio_pci.c
@@ -31,6 +31,15 @@
 #define VIRTIO_PCI_CONFIG(hw) \
(((hw)->use_msix == VIRTIO_MSIX_ENABLED) ? 24 : 20)
 
+
+struct virtio_pci_internal {
+   struct rte_pci_ioport io;
+};
+
+#define VTPCI_IO(hw) (&virtio_pci_internal[(hw)->port_id].io)
+
+struct virtio_pci_internal virtio_pci_internal[RTE_MAX_ETHPORTS];
+
 static inline int
 check_vq_phys_addr_ok(struct virtqueue *vq)
 {
@@ -300,7 +309,9 @@ legacy_notify_queue(struct virtio_hw *hw, struct virtqueue 
*vq)
 static void
 legacy_intr_detect(struct virtio_hw *hw)
 {
-   hw->use_msix = vtpci_msix_detect(VTPCI_DEV(hw));
+   struct virtio_pci_dev *dev = virtio_pci_get_dev(hw);
+
+   hw->use_msix = vtpci_msix_detect(dev->pci_dev);
 }
 
 static int
@@ -558,7 +569,9 @@ modern_notify_queue(struct virtio_hw *hw, struct virtqueue 
*vq)
 static void
 modern_intr_detect(struct virtio_hw *hw)
 {
-   hw->use_msix = vtpci_msix_detect(VTPCI_DEV(hw));
+   struct virtio_pci_dev *dev = virtio_pci_get_dev(hw);
+
+   hw->use_msix = vtpci_msix_detect(dev->pci_dev);
 }
 
 static int
@@ -857,3 +870,14 @@ vtpci_init(struct rte_pci_device *pci_dev, struct 
virtio_pci_dev *dev)
return 0;
 }
 
+void vtpci_legacy_ioport_unmap(struct virtio_hw *hw)
+{
+   rte_pci_ioport_unmap(VTPCI_IO(hw));
+}
+
+int vtpci_legacy_ioport_map(struct virtio_hw *hw)
+{
+   struct virtio_pci_dev *dev = virtio_pci_get_dev(hw);
+
+   return rte_pci_ioport_map(dev->pci_dev, 0, VTPCI_IO(hw));
+}
diff --git a/drivers/net/virtio/virtio_pci.h b/drivers/net/virtio/virtio_pci.h
index 15f8144fc6..def8faca72 100644
--- a/drivers/net/virtio/virtio_pci.h
+++ b/drivers/net/virtio/virtio_pci.h
@@ -298,18 +298,12 @@ struct virtio_pci_dev {
  */
 struct virtio_hw_internal {
const struct virtio_pci_ops *vtpci_ops;
-   struct rte_pci_ioport io;
-   struct rte_pci_device *dev;
 };
 
 #define VTPCI_OPS(hw)  (virtio_hw_internal[(hw)->port_id].vtpci_ops)
-#define VTPCI_IO(hw)   (&virtio_hw_internal[(hw)->port_id].io)
-#define VTPCI_DEV(hw)  (virtio_hw_internal[(hw)->port_id].dev)
-
 
 extern struct virtio_hw_internal virtio_hw_internal[RTE_MAX_ETHPORTS];
 
-
 /*
  * This structure is just a reference to read
  * net device specific config space; it just a chodu structure
@@ -382,6 +376,9 @@ void vtpci_read_dev_config(struct virtio_hw *, size_t, void 
*, int);
 
 uint8_t vtpci_isr(struct virtio_hw *);
 
+void vtpci_legacy_ioport_unmap(struct virtio_hw *hw);
+int vtpci_legacy_ioport_map(struct virtio_hw *hw);
+
 extern const struct virtio_pci_ops legacy_ops;
 extern const struct virtio_pci_ops modern_ops;
 extern const struct virtio_pci_ops virtio_user_ops;
diff --git a/drivers/net/virtio/virtio_pci_ethdev.c 
b/drivers/net/virtio/virtio_pci_ethdev.c
index 8cc6561914..9fe59feb51 100644
--- a/drivers/net/virtio/virtio_pci_ethdev.c
+++ b/drivers/net/virtio/virtio_pci_ethdev.c
@@ -60,7 +60,7 @@ virtio_remap_pci(struct rte_pci_device *pci_dev, struct 
virtio_pci_dev *dev)
return -1;
}
} else {
-   if (rte_pci_ioport_map(pci_dev, 0, VTPCI_IO(hw)) < 0)
+   if (vtpci_legacy_ioport_map(hw) < 0)
return -1;
}
 
@@ -75,8 +75,6 @@ eth_virtio_pci_init(struct rte_eth_dev *eth_dev)
struct rte_pci_device *pci_dev = RTE_ETH_DEV_TO_PCI(eth_dev);
int ret;
 
-   VTPCI_DEV(hw) = pci_dev;
-
if (rte_eal_process_type() == RTE_PROC_PRIMARY) {
ret = vtpci_init(RTE_ETH_DEV_TO_PCI(eth_dev), dev);
if (ret) {
@@ -111,7 +109,7 @@ eth_virtio_pci_init(struct rte_eth_dev *eth_dev)
 err_unmap:
rte_pci_unmap_device(RTE_ETH_DEV_TO_PCI(eth_dev));
if (!dev->modern)
-   rte_pci_ioport_unmap(VTPCI_IO(hw));
+   vtpci_legacy_ioport_unmap(hw);
 
return ret;
 }
-- 
2.29.2

[dpdk-dev] [PATCH v3 18/44] net/virtio: introduce generic virtio header

[Maxime Coquelin]

This patch moves virtio_hw and virtio callbacks into
a generic virtio header, now that they have been
curated from PCI references.

Signed-off-by: Maxime Coquelin 
Reviewed-by: Chenbo Xia 
---
 drivers/net/virtio/virtio.h | 74 ++
 drivers/net/virtio/virtio_ethdev.c  | 32 +-
 drivers/net/virtio/virtio_pci.c | 28 -
 drivers/net/virtio/virtio_pci.h | 84 ++---
 drivers/net/virtio/virtio_pci_ethdev.c  |  5 +-
 drivers/net/virtio/virtio_user_ethdev.c |  8 +--
 drivers/net/virtio/virtqueue.h  |  2 +-
 7 files changed, 116 insertions(+), 117 deletions(-)
 create mode 100644 drivers/net/virtio/virtio.h

diff --git a/drivers/net/virtio/virtio.h b/drivers/net/virtio/virtio.h
new file mode 100644
index 00..e55e3329f9
--- /dev/null
+++ b/drivers/net/virtio/virtio.h
@@ -0,0 +1,74 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2010-2014 Intel Corporation
+ * Copyright(c) 2021 Red Hat, Inc.
+ */
+
+#ifndef _VIRTIO_H_
+#define _VIRTIO_H_
+
+#include 
+
+struct virtio_hw {
+   struct virtqueue **vqs;
+   uint64_t guest_features;
+   uint16_t vtnet_hdr_size;
+   uint8_t started;
+   uint8_t weak_barriers;
+   uint8_t vlan_strip;
+   uint8_t has_tx_offload;
+   uint8_t has_rx_offload;
+   uint8_t use_vec_rx;
+   uint8_t use_vec_tx;
+   uint8_t use_inorder_rx;
+   uint8_t use_inorder_tx;
+   uint8_t opened;
+   uint16_t port_id;
+   uint8_t mac_addr[RTE_ETHER_ADDR_LEN];
+   uint32_t speed;  /* link speed in MB */
+   uint8_t duplex;
+   uint8_t use_msix;
+   uint16_t max_mtu;
+   /*
+* App management thread and virtio interrupt handler thread
+* both can change device state, this lock is meant to avoid
+* such a contention.
+*/
+   rte_spinlock_t state_lock;
+   struct rte_mbuf **inject_pkts;
+   uint16_t max_queue_pairs;
+   uint64_t req_guest_features;
+   struct virtnet_ctl *cvq;
+};
+
+struct virtio_ops {
+   void (*read_dev_cfg)(struct virtio_hw *hw, size_t offset, void *dst, 
int len);
+   void (*write_dev_cfg)(struct virtio_hw *hw, size_t offset, const void 
*src, int len);
+   uint8_t (*get_status)(struct virtio_hw *hw);
+   void (*set_status)(struct virtio_hw *hw, uint8_t status);
+   uint64_t (*get_features)(struct virtio_hw *hw);
+   void (*set_features)(struct virtio_hw *hw, uint64_t features);
+   int (*features_ok)(struct virtio_hw *hw);
+   uint8_t (*get_isr)(struct virtio_hw *hw);
+   uint16_t (*set_config_irq)(struct virtio_hw *hw, uint16_t vec);
+   uint16_t (*set_queue_irq)(struct virtio_hw *hw, struct virtqueue *vq, 
uint16_t vec);
+   uint16_t (*get_queue_num)(struct virtio_hw *hw, uint16_t queue_id);
+   int (*setup_queue)(struct virtio_hw *hw, struct virtqueue *vq);
+   void (*del_queue)(struct virtio_hw *hw, struct virtqueue *vq);
+   void (*notify_queue)(struct virtio_hw *hw, struct virtqueue *vq);
+   void (*intr_detect)(struct virtio_hw *hw);
+   int (*dev_close)(struct virtio_hw *hw);
+};
+
+/*
+ * This structure stores per-process data. Only virtio_ops for now.
+ */
+struct virtio_hw_internal {
+   const struct virtio_ops *virtio_ops;
+};
+
+#define VIRTIO_OPS(hw) (virtio_hw_internal[(hw)->port_id].virtio_ops)
+
+extern struct virtio_hw_internal virtio_hw_internal[RTE_MAX_ETHPORTS];
+
+
+#endif /* _VIRTIO_H_ */
diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index c1d7b14dda..4b22622d2d 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -446,7 +446,7 @@ virtio_init_queue(struct rte_eth_dev *dev, uint16_t 
vtpci_queue_idx)
 * Read the virtqueue size from the Queue Size field
 * Always power of 2 and if 0 virtqueue does not exist
 */
-   vq_size = VTPCI_OPS(hw)->get_queue_num(hw, vtpci_queue_idx);
+   vq_size = VIRTIO_OPS(hw)->get_queue_num(hw, vtpci_queue_idx);
PMD_INIT_LOG(DEBUG, "vq_size: %u", vq_size);
if (vq_size == 0) {
PMD_INIT_LOG(ERR, "virtqueue does not exist");
@@ -608,7 +608,7 @@ virtio_init_queue(struct rte_eth_dev *dev, uint16_t 
vtpci_queue_idx)
}
}
 
-   if (VTPCI_OPS(hw)->setup_queue(hw, vq) < 0) {
+   if (VIRTIO_OPS(hw)->setup_queue(hw, vq) < 0) {
PMD_INIT_LOG(ERR, "setup_queue failed");
return -EINVAL;
}
@@ -703,7 +703,7 @@ virtio_dev_close(struct rte_eth_dev *dev)
 
/* reset the NIC */
if (dev->data->dev_flags & RTE_ETH_DEV_INTR_LSC)
-   VTPCI_OPS(hw)->set_config_irq(hw, VIRTIO_MSI_NO_VECTOR);
+   VIRTIO_OPS(hw)->set_config_irq(hw, VIRTIO_MSI_NO_VECTOR);
if (intr_conf->rxq)
virtio_queues_unbind_intr(dev);
 
@@ -718,7 +718,7 @@ virtio_dev_close(struct rte_eth_dev *dev)
virtio_dev_free_mb

[dpdk-dev] [PATCH v3 19/44] net/virtio: move features definition to generic header

[Maxime Coquelin]

This patch moves all the Virtio definition to the generic
header. It also renames some helpers to no more reference
PCI.

Signed-off-by: Maxime Coquelin 
---
 drivers/net/virtio/meson.build|   3 +-
 drivers/net/virtio/virtio.c   |  22 
 drivers/net/virtio/virtio.h   |  94 +++
 drivers/net/virtio/virtio_ethdev.c| 114 +-
 drivers/net/virtio/virtio_pci.c   |  21 +---
 drivers/net/virtio/virtio_pci.h   |  90 --
 drivers/net/virtio/virtio_ring.h  |   2 +-
 drivers/net/virtio/virtio_rxtx.c  |  38 +++---
 drivers/net/virtio/virtio_rxtx_packed.h   |   6 +-
 .../net/virtio/virtio_user/vhost_kernel_tap.c |   2 +-
 drivers/net/virtio/virtio_user_ethdev.c   |   6 +-
 drivers/net/virtio/virtqueue.c|   4 +-
 drivers/net/virtio/virtqueue.h|   8 +-
 13 files changed, 211 insertions(+), 199 deletions(-)
 create mode 100644 drivers/net/virtio/virtio.c

diff --git a/drivers/net/virtio/meson.build b/drivers/net/virtio/meson.build
index f2873d6180..d595cfdcab 100644
--- a/drivers/net/virtio/meson.build
+++ b/drivers/net/virtio/meson.build
@@ -7,7 +7,8 @@ if is_windows
subdir_done()
 endif
 
-sources += files('virtio_ethdev.c',
+sources += files('virtio.c',
+   'virtio_ethdev.c',
'virtio_pci_ethdev.c',
'virtio_pci.c',
'virtio_rxtx.c',
diff --git a/drivers/net/virtio/virtio.c b/drivers/net/virtio/virtio.c
new file mode 100644
index 00..d8d6bf7add
--- /dev/null
+++ b/drivers/net/virtio/virtio.c
@@ -0,0 +1,22 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2010-2014 Intel Corporation
+ * Copyright(c) 2020 Red Hat, Inc.
+ */
+
+#include "virtio.h"
+
+uint64_t
+virtio_negotiate_features(struct virtio_hw *hw, uint64_t host_features)
+{
+   uint64_t features;
+
+   /*
+* Limit negotiated features to what the driver, virtqueue, and
+* host all support.
+*/
+   features = host_features & hw->guest_features;
+   VIRTIO_OPS(hw)->set_features(hw, features);
+
+   return features;
+}
+
diff --git a/drivers/net/virtio/virtio.h b/drivers/net/virtio/virtio.h
index e55e3329f9..cdfee5d182 100644
--- a/drivers/net/virtio/virtio.h
+++ b/drivers/net/virtio/virtio.h
@@ -8,6 +8,86 @@
 
 #include 
 
+/* The feature bitmap for virtio net */
+#define VIRTIO_NET_F_CSUM  0   /* Host handles pkts w/ partial csum */
+#define VIRTIO_NET_F_GUEST_CSUM1   /* Guest handles pkts w/ 
partial csum */
+#define VIRTIO_NET_F_MTU   3   /* Initial MTU advice. */
+#define VIRTIO_NET_F_MAC   5   /* Host has given MAC address. */
+#define VIRTIO_NET_F_GUEST_TSO47   /* Guest can handle TSOv4 in. */
+#define VIRTIO_NET_F_GUEST_TSO68   /* Guest can handle TSOv6 in. */
+#define VIRTIO_NET_F_GUEST_ECN 9   /* Guest can handle TSO[6] w/ ECN in. */
+#define VIRTIO_NET_F_GUEST_UFO 10  /* Guest can handle UFO in. */
+#define VIRTIO_NET_F_HOST_TSO4 11  /* Host can handle TSOv4 in. */
+#define VIRTIO_NET_F_HOST_TSO6 12  /* Host can handle TSOv6 in. */
+#define VIRTIO_NET_F_HOST_ECN  13  /* Host can handle TSO[6] w/ ECN in. */
+#define VIRTIO_NET_F_HOST_UFO  14  /* Host can handle UFO in. */
+#define VIRTIO_NET_F_MRG_RXBUF 15  /* Host can merge receive buffers. */
+#define VIRTIO_NET_F_STATUS16  /* virtio_net_config.status available */
+#define VIRTIO_NET_F_CTRL_VQ   17  /* Control channel available */
+#define VIRTIO_NET_F_CTRL_RX   18  /* Control channel RX mode support */
+#define VIRTIO_NET_F_CTRL_VLAN 19  /* Control channel VLAN filtering */
+#define VIRTIO_NET_F_CTRL_RX_EXTRA 20  /* Extra RX mode control support */
+#define VIRTIO_NET_F_GUEST_ANNOUNCE 21 /* Guest can announce device on the 
network */
+#define VIRTIO_NET_F_MQ22  /* Device supports Receive Flow 
Steering */
+#define VIRTIO_NET_F_CTRL_MAC_ADDR 23  /* Set MAC address */
+
+/*
+ * Do we get callbacks when the ring is completely used,
+ * even if we've suppressed them?
+ */
+#define VIRTIO_F_NOTIFY_ON_EMPTY   24
+
+/* Can the device handle any descriptor layout? */
+#define VIRTIO_F_ANY_LAYOUT27
+
+/* We support indirect buffer descriptors */
+#define VIRTIO_RING_F_INDIRECT_DESC28
+
+#define VIRTIO_F_VERSION_1 32
+#define VIRTIO_F_IOMMU_PLATFORM33
+#define VIRTIO_F_RING_PACKED   34
+
+/*
+ * Some VirtIO feature bits (currently bits 28 through 31) are
+ * reserved for the transport being used (eg. virtio_ring), the
+ * rest are per-device feature bits.
+ */
+#define VIRTIO_TRANSPORT_F_START 28
+#define VIRTIO_TRANSPORT_F_END   34
+
+/*
+ * Inorder feature indicates that all buffers are used by the device
+ * in the same order in which they have been made available.
+ */
+#define VIRTIO_F_IN_ORDER 35
+
+/*
+ * This feature indicates that memory accesses by th

[dpdk-dev] [PATCH v3 20/44] net/virtio: move virtqueue defines in generic header

[Maxime Coquelin]

This patch moves the virtqueues defines from PCI header
to the generic one.

Signed-off-by: Maxime Coquelin 
Reviewed-by: Chenbo Xia 
---
 drivers/net/virtio/virtio.h| 18 ++
 drivers/net/virtio/virtio_ethdev.h |  4 +++-
 drivers/net/virtio/virtio_pci.h| 17 -
 .../net/virtio/virtio_user/virtio_user_dev.h   |  3 ++-
 4 files changed, 23 insertions(+), 19 deletions(-)

diff --git a/drivers/net/virtio/virtio.h b/drivers/net/virtio/virtio.h
index cdfee5d182..0c2b3525d5 100644
--- a/drivers/net/virtio/virtio.h
+++ b/drivers/net/virtio/virtio.h
@@ -88,6 +88,24 @@
 #define VIRTIO_NET_S_LINK_UP   1   /* Link is up */
 #define VIRTIO_NET_S_ANNOUNCE  2   /* Announcement is needed */
 
+/*
+ * Each virtqueue indirect descriptor list must be physically contiguous.
+ * To allow us to malloc(9) each list individually, limit the number
+ * supported to what will fit in one page. With 4KB pages, this is a limit
+ * of 256 descriptors. If there is ever a need for more, we can switch to
+ * contigmalloc(9) for the larger allocations, similar to what
+ * bus_dmamem_alloc(9) does.
+ *
+ * Note the sizeof(struct vring_desc) is 16 bytes.
+ */
+#define VIRTIO_MAX_INDIRECT ((int)(PAGE_SIZE / 16))
+
+/*
+ * Maximum number of virtqueues per device.
+ */
+#define VIRTIO_MAX_VIRTQUEUE_PAIRS 8
+#define VIRTIO_MAX_VIRTQUEUES (VIRTIO_MAX_VIRTQUEUE_PAIRS * 2 + 1)
+
 struct virtio_hw {
struct virtqueue **vqs;
uint64_t guest_features;
diff --git a/drivers/net/virtio/virtio_ethdev.h 
b/drivers/net/virtio/virtio_ethdev.h
index 13395937c8..d82856df9b 100644
--- a/drivers/net/virtio/virtio_ethdev.h
+++ b/drivers/net/virtio/virtio_ethdev.h
@@ -7,7 +7,9 @@
 
 #include 
 
-#include "virtio_pci.h"
+#include 
+
+#include "virtio.h"
 
 #ifndef PAGE_SIZE
 #define PAGE_SIZE 4096
diff --git a/drivers/net/virtio/virtio_pci.h b/drivers/net/virtio/virtio_pci.h
index 3d93e74f36..2c27c906fb 100644
--- a/drivers/net/virtio/virtio_pci.h
+++ b/drivers/net/virtio/virtio_pci.h
@@ -67,23 +67,6 @@ struct virtnet_ctl;
 #define VIRTIO_CONFIG_STATUS_DEV_NEED_RESET0x40
 #define VIRTIO_CONFIG_STATUS_FAILED0x80
 
-/*
- * Each virtqueue indirect descriptor list must be physically contiguous.
- * To allow us to malloc(9) each list individually, limit the number
- * supported to what will fit in one page. With 4KB pages, this is a limit
- * of 256 descriptors. If there is ever a need for more, we can switch to
- * contigmalloc(9) for the larger allocations, similar to what
- * bus_dmamem_alloc(9) does.
- *
- * Note the sizeof(struct vring_desc) is 16 bytes.
- */
-#define VIRTIO_MAX_INDIRECT ((int) (PAGE_SIZE / 16))
-
-/*
- * Maximum number of virtqueues per device.
- */
-#define VIRTIO_MAX_VIRTQUEUE_PAIRS 8
-#define VIRTIO_MAX_VIRTQUEUES (VIRTIO_MAX_VIRTQUEUE_PAIRS * 2 + 1)
 
 /* Common configuration */
 #define VIRTIO_PCI_CAP_COMMON_CFG  1
diff --git a/drivers/net/virtio/virtio_user/virtio_user_dev.h 
b/drivers/net/virtio/virtio_user/virtio_user_dev.h
index b3776cc7a0..ab62463a5b 100644
--- a/drivers/net/virtio/virtio_user/virtio_user_dev.h
+++ b/drivers/net/virtio/virtio_user/virtio_user_dev.h
@@ -7,7 +7,8 @@
 
 #include 
 #include 
-#include "../virtio_pci.h"
+
+#include "../virtio.h"
 #include "../virtio_ring.h"
 
 enum virtio_user_backend_type {
-- 
2.29.2

[dpdk-dev] [PATCH v3 22/44] net/virtio: make interrupt handling more generic

[Maxime Coquelin]

This patch aims at isolating MSIX notion into PCI
layer.

Signed-off-by: Maxime Coquelin 
Reviewed-by: David Marchand 
---
 drivers/net/virtio/virtio.c |  6 
 drivers/net/virtio/virtio.h | 11 +--
 drivers/net/virtio/virtio_ethdev.c  |  7 ++---
 drivers/net/virtio/virtio_pci.c | 38 -
 drivers/net/virtio/virtio_pci.h | 24 +---
 drivers/net/virtio/virtio_user_ethdev.c |  9 ++
 6 files changed, 46 insertions(+), 49 deletions(-)

diff --git a/drivers/net/virtio/virtio.c b/drivers/net/virtio/virtio.c
index ba3203e68b..7e1e77797f 100644
--- a/drivers/net/virtio/virtio.c
+++ b/drivers/net/virtio/virtio.c
@@ -63,3 +63,9 @@ virtio_get_status(struct virtio_hw *hw)
 {
return VIRTIO_OPS(hw)->get_status(hw);
 }
+
+uint8_t
+virtio_get_isr(struct virtio_hw *hw)
+{
+   return VIRTIO_OPS(hw)->get_isr(hw);
+}
diff --git a/drivers/net/virtio/virtio.h b/drivers/net/virtio/virtio.h
index 0ac5328c57..a7629ad16b 100644
--- a/drivers/net/virtio/virtio.h
+++ b/drivers/net/virtio/virtio.h
@@ -124,6 +124,13 @@
 #define VIRTIO_CONFIG_STATUS_DEV_NEED_RESET0x40
 #define VIRTIO_CONFIG_STATUS_FAILED0x80
 
+/* The bit of the ISR which indicates a device has an interrupt. */
+#define VIRTIO_ISR_INTR   0x1
+/* The bit of the ISR which indicates a device configuration change. */
+#define VIRTIO_ISR_CONFIG 0x2
+/* Vector value used to disable MSI for queue. */
+#define VIRTIO_MSI_NO_VECTOR 0x
+
 /*
  * This structure is just a reference to read net device specific
  * config space; it is just a shadow structure.
@@ -168,7 +175,7 @@ struct virtio_hw {
uint8_t mac_addr[RTE_ETHER_ADDR_LEN];
uint32_t speed;  /* link speed in MB */
uint8_t duplex;
-   uint8_t use_msix;
+   uint8_t intr_lsc;
uint16_t max_mtu;
/*
 * App management thread and virtio interrupt handler thread
@@ -232,5 +239,5 @@ void virtio_write_dev_config(struct virtio_hw *hw, size_t 
offset, const void *sr
 void virtio_read_dev_config(struct virtio_hw *hw, size_t offset, void *dst, 
int length);
 void virtio_reset(struct virtio_hw *hw);
 void virtio_reinit_complete(struct virtio_hw *hw);
-
+uint8_t virtio_get_isr(struct virtio_hw *hw);
 #endif /* _VIRTIO_H_ */
diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index dfd8d29f60..51eb5aebf3 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -1455,13 +1455,13 @@ virtio_interrupt_handler(void *param)
uint16_t status;
 
/* Read interrupt status which clears interrupt */
-   isr = vtpci_isr(hw);
+   isr = virtio_get_isr(hw);
PMD_DRV_LOG(INFO, "interrupt status = %#x", isr);
 
if (virtio_intr_unmask(dev) < 0)
PMD_DRV_LOG(ERR, "interrupt enable failed");
 
-   if (isr & VIRTIO_PCI_ISR_CONFIG) {
+   if (isr & VIRTIO_ISR_CONFIG) {
if (virtio_dev_link_update(dev, 0) == 0)
rte_eth_dev_callback_process(dev,
 RTE_ETH_EVENT_INTR_LSC,
@@ -1668,8 +1668,7 @@ virtio_init_device(struct rte_eth_dev *eth_dev, uint64_t 
req_features)
hw->weak_barriers = !virtio_with_feature(hw, VIRTIO_F_ORDER_PLATFORM);
 
/* If host does not support both status and MSI-X then disable LSC */
-   if (virtio_with_feature(hw, VIRTIO_NET_F_STATUS) &&
-   hw->use_msix != VIRTIO_MSIX_NONE)
+   if (virtio_with_feature(hw, VIRTIO_NET_F_STATUS) && hw->intr_lsc)
eth_dev->data->dev_flags |= RTE_ETH_DEV_INTR_LSC;
else
eth_dev->data->dev_flags &= ~RTE_ETH_DEV_INTR_LSC;
diff --git a/drivers/net/virtio/virtio_pci.c b/drivers/net/virtio/virtio_pci.c
index baadc79c43..905534ac72 100644
--- a/drivers/net/virtio/virtio_pci.c
+++ b/drivers/net/virtio/virtio_pci.c
@@ -28,8 +28,8 @@
  * The remaining space is defined by each driver as the per-driver
  * configuration space.
  */
-#define VIRTIO_PCI_CONFIG(hw) \
-   (((hw)->use_msix == VIRTIO_MSIX_ENABLED) ? 24 : 20)
+#define VIRTIO_PCI_CONFIG(dev) \
+   (((dev)->msix_status == VIRTIO_MSIX_ENABLED) ? 24 : 20)
 
 
 struct virtio_pci_internal {
@@ -121,6 +121,7 @@ static void
 legacy_read_dev_config(struct virtio_hw *hw, size_t offset,
   void *dst, int length)
 {
+   struct virtio_pci_dev *dev = virtio_pci_get_dev(hw);
 #ifdef RTE_ARCH_PPC_64
int size;
 
@@ -128,17 +129,17 @@ legacy_read_dev_config(struct virtio_hw *hw, size_t 
offset,
if (length >= 4) {
size = 4;
rte_pci_ioport_read(VTPCI_IO(hw), dst, size,
-   VIRTIO_PCI_CONFIG(hw) + offset);
+   VIRTIO_PCI_CONFIG(dev) + offset);
*(uint32_t *)dst = rte_be_to_cpu_32(*(uint32_t *)dst);
} else if (length >= 2) {

[dpdk-dev] [PATCH v3 21/44] net/virtio: move config definitions to generic header

[Maxime Coquelin]

This patch moves config and status definitions from the PCI
header to the generic one.

Signed-off-by: Maxime Coquelin 
Reviewed-by: Chenbo Xia 
---
 drivers/net/virtio/virtio.c | 43 +++
 drivers/net/virtio/virtio.h | 50 ++
 drivers/net/virtio/virtio_ethdev.c  | 36 
 drivers/net/virtio/virtio_pci.c | 44 
 drivers/net/virtio/virtio_pci.h | 55 -
 drivers/net/virtio/virtio_user_ethdev.c | 10 ++---
 6 files changed, 116 insertions(+), 122 deletions(-)

diff --git a/drivers/net/virtio/virtio.c b/drivers/net/virtio/virtio.c
index d8d6bf7add..ba3203e68b 100644
--- a/drivers/net/virtio/virtio.c
+++ b/drivers/net/virtio/virtio.c
@@ -20,3 +20,46 @@ virtio_negotiate_features(struct virtio_hw *hw, uint64_t 
host_features)
return features;
 }
 
+
+void
+virtio_read_dev_config(struct virtio_hw *hw, size_t offset,
+ void *dst, int length)
+{
+   VIRTIO_OPS(hw)->read_dev_cfg(hw, offset, dst, length);
+}
+
+void
+virtio_write_dev_config(struct virtio_hw *hw, size_t offset,
+  const void *src, int length)
+{
+   VIRTIO_OPS(hw)->write_dev_cfg(hw, offset, src, length);
+}
+
+void
+virtio_reset(struct virtio_hw *hw)
+{
+   VIRTIO_OPS(hw)->set_status(hw, VIRTIO_CONFIG_STATUS_RESET);
+   /* flush status write */
+   VIRTIO_OPS(hw)->get_status(hw);
+}
+
+void
+virtio_reinit_complete(struct virtio_hw *hw)
+{
+   virtio_set_status(hw, VIRTIO_CONFIG_STATUS_DRIVER_OK);
+}
+
+void
+virtio_set_status(struct virtio_hw *hw, uint8_t status)
+{
+   if (status != VIRTIO_CONFIG_STATUS_RESET)
+   status |= VIRTIO_OPS(hw)->get_status(hw);
+
+   VIRTIO_OPS(hw)->set_status(hw, status);
+}
+
+uint8_t
+virtio_get_status(struct virtio_hw *hw)
+{
+   return VIRTIO_OPS(hw)->get_status(hw);
+}
diff --git a/drivers/net/virtio/virtio.h b/drivers/net/virtio/virtio.h
index 0c2b3525d5..0ac5328c57 100644
--- a/drivers/net/virtio/virtio.h
+++ b/drivers/net/virtio/virtio.h
@@ -106,6 +106,50 @@
 #define VIRTIO_MAX_VIRTQUEUE_PAIRS 8
 #define VIRTIO_MAX_VIRTQUEUES (VIRTIO_MAX_VIRTQUEUE_PAIRS * 2 + 1)
 
+/* VirtIO device IDs. */
+#define VIRTIO_ID_NETWORK  0x01
+#define VIRTIO_ID_BLOCK0x02
+#define VIRTIO_ID_CONSOLE  0x03
+#define VIRTIO_ID_ENTROPY  0x04
+#define VIRTIO_ID_BALLOON  0x05
+#define VIRTIO_ID_IOMEMORY 0x06
+#define VIRTIO_ID_9P   0x09
+
+/* Status byte for guest to report progress. */
+#define VIRTIO_CONFIG_STATUS_RESET 0x00
+#define VIRTIO_CONFIG_STATUS_ACK   0x01
+#define VIRTIO_CONFIG_STATUS_DRIVER0x02
+#define VIRTIO_CONFIG_STATUS_DRIVER_OK 0x04
+#define VIRTIO_CONFIG_STATUS_FEATURES_OK   0x08
+#define VIRTIO_CONFIG_STATUS_DEV_NEED_RESET0x40
+#define VIRTIO_CONFIG_STATUS_FAILED0x80
+
+/*
+ * This structure is just a reference to read net device specific
+ * config space; it is just a shadow structure.
+ *
+ */
+struct virtio_net_config {
+   /* The config defining mac address (if VIRTIO_NET_F_MAC) */
+   uint8_tmac[RTE_ETHER_ADDR_LEN];
+   /* See VIRTIO_NET_F_STATUS and VIRTIO_NET_S_* above */
+   uint16_t   status;
+   uint16_t   max_virtqueue_pairs;
+   uint16_t   mtu;
+   /*
+* speed, in units of 1Mb. All values 0 to INT_MAX are legal.
+* Any other value stands for unknown.
+*/
+   uint32_t speed;
+   /*
+* 0x00 - half duplex
+* 0x01 - full duplex
+* Any other value stands for unknown.
+*/
+   uint8_t duplex;
+
+} __rte_packed;
+
 struct virtio_hw {
struct virtqueue **vqs;
uint64_t guest_features;
@@ -182,5 +226,11 @@ virtio_with_packed_queue(struct virtio_hw *hw)
 }
 
 uint64_t virtio_negotiate_features(struct virtio_hw *hw, uint64_t 
host_features);
+uint8_t virtio_get_status(struct virtio_hw *hw);
+void virtio_set_status(struct virtio_hw *hw, uint8_t status);
+void virtio_write_dev_config(struct virtio_hw *hw, size_t offset, const void 
*src, int length);
+void virtio_read_dev_config(struct virtio_hw *hw, size_t offset, void *dst, 
int length);
+void virtio_reset(struct virtio_hw *hw);
+void virtio_reinit_complete(struct virtio_hw *hw);
 
 #endif /* _VIRTIO_H_ */
diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index 1771844fe7..dfd8d29f60 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -714,7 +714,7 @@ virtio_dev_close(struct rte_eth_dev *dev)
dev->intr_handle->intr_vec = NULL;
}
 
-   vtpci_reset(hw);
+   virtio_reset(hw);
virtio_dev_free_mbufs(dev);
virtio_free_queues(hw);
 
@@ -1096,7 +1096,7 @@ virtio_dev_stats_reset(struct rte_eth_dev *dev)
 static void
 virtio_set_hwaddr(struct virtio_hw *hw)
 {
-   vtpci_write_dev_config(hw,
+   virtio_write_dev_config(hw,
of

[dpdk-dev] [PATCH v3 24/44] net/virtio: remove last PCI refs in non-PCI code

[Maxime Coquelin]

This patch finalizes the bus isolation part of this
refactoring.

Signed-off-by: Maxime Coquelin 
Reviewed-by: Chenbo Xia 
Reviewed-by: David Marchand 
---
 drivers/net/virtio/virtio_ethdev.c   | 21 +---
 drivers/net/virtio/virtio_rxtx.c | 18 -
 drivers/net/virtio/virtio_rxtx_packed.h  |  2 +-
 drivers/net/virtio/virtio_rxtx_packed_avx.h  |  2 +-
 drivers/net/virtio/virtio_rxtx_packed_neon.h |  2 +-
 drivers/net/virtio/virtio_user/vhost.h   |  4 +++-
 drivers/net/virtio/virtqueue.c   |  2 +-
 drivers/net/virtio/virtqueue.h   |  8 
 8 files changed, 29 insertions(+), 30 deletions(-)

diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index a2195bdda8..5f7ea390c8 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -9,14 +9,11 @@
 #include 
 
 #include 
-#include 
 #include 
 #include 
 #include 
 #include 
 #include 
-#include 
-#include 
 #include 
 #include 
 #include 
@@ -32,7 +29,7 @@
 #include 
 
 #include "virtio_ethdev.h"
-#include "virtio_pci.h"
+#include "virtio.h"
 #include "virtio_logs.h"
 #include "virtqueue.h"
 #include "virtio_rxtx.h"
@@ -422,7 +419,7 @@ virtio_init_vring(struct virtqueue *vq)
 }
 
 static int
-virtio_init_queue(struct rte_eth_dev *dev, uint16_t vtpci_queue_idx)
+virtio_init_queue(struct rte_eth_dev *dev, uint16_t queue_idx)
 {
char vq_name[VIRTQUEUE_MAX_NAME_SZ];
char vq_hdr_name[VIRTQUEUE_MAX_NAME_SZ];
@@ -435,18 +432,18 @@ virtio_init_queue(struct rte_eth_dev *dev, uint16_t 
vtpci_queue_idx)
struct virtqueue *vq;
size_t sz_hdr_mz = 0;
void *sw_ring = NULL;
-   int queue_type = virtio_get_queue_type(hw, vtpci_queue_idx);
+   int queue_type = virtio_get_queue_type(hw, queue_idx);
int ret;
int numa_node = dev->device->numa_node;
 
PMD_INIT_LOG(INFO, "setting up queue: %u on NUMA node %d",
-   vtpci_queue_idx, numa_node);
+   queue_idx, numa_node);
 
/*
 * Read the virtqueue size from the Queue Size field
 * Always power of 2 and if 0 virtqueue does not exist
 */
-   vq_size = VIRTIO_OPS(hw)->get_queue_num(hw, vtpci_queue_idx);
+   vq_size = VIRTIO_OPS(hw)->get_queue_num(hw, queue_idx);
PMD_INIT_LOG(DEBUG, "vq_size: %u", vq_size);
if (vq_size == 0) {
PMD_INIT_LOG(ERR, "virtqueue does not exist");
@@ -459,7 +456,7 @@ virtio_init_queue(struct rte_eth_dev *dev, uint16_t 
vtpci_queue_idx)
}
 
snprintf(vq_name, sizeof(vq_name), "port%d_vq%d",
-dev->data->port_id, vtpci_queue_idx);
+dev->data->port_id, queue_idx);
 
size = RTE_ALIGN_CEIL(sizeof(*vq) +
vq_size * sizeof(struct vq_desc_extra),
@@ -481,10 +478,10 @@ virtio_init_queue(struct rte_eth_dev *dev, uint16_t 
vtpci_queue_idx)
PMD_INIT_LOG(ERR, "can not allocate vq");
return -ENOMEM;
}
-   hw->vqs[vtpci_queue_idx] = vq;
+   hw->vqs[queue_idx] = vq;
 
vq->hw = hw;
-   vq->vq_queue_index = vtpci_queue_idx;
+   vq->vq_queue_index = queue_idx;
vq->vq_nentries = vq_size;
if (virtio_with_packed_queue(hw)) {
vq->vq_packed.used_wrap_counter = 1;
@@ -527,7 +524,7 @@ virtio_init_queue(struct rte_eth_dev *dev, uint16_t 
vtpci_queue_idx)
 
if (sz_hdr_mz) {
snprintf(vq_hdr_name, sizeof(vq_hdr_name), "port%d_vq%d_hdr",
-dev->data->port_id, vtpci_queue_idx);
+dev->data->port_id, queue_idx);
hdr_mz = rte_memzone_reserve_aligned(vq_hdr_name, sz_hdr_mz,
numa_node, RTE_MEMZONE_IOVA_CONTIG,
RTE_CACHE_LINE_SIZE);
diff --git a/drivers/net/virtio/virtio_rxtx.c b/drivers/net/virtio/virtio_rxtx.c
index 728085e99b..b5657d5c83 100644
--- a/drivers/net/virtio/virtio_rxtx.c
+++ b/drivers/net/virtio/virtio_rxtx.c
@@ -27,7 +27,7 @@
 
 #include "virtio_logs.h"
 #include "virtio_ethdev.h"
-#include "virtio_pci.h"
+#include "virtio.h"
 #include "virtqueue.h"
 #include "virtio_rxtx.h"
 #include "virtio_rxtx_simple.h"
@@ -628,9 +628,9 @@ virtio_dev_rx_queue_setup(struct rte_eth_dev *dev,
const struct rte_eth_rxconf *rx_conf,
struct rte_mempool *mp)
 {
-   uint16_t vtpci_queue_idx = 2 * queue_idx + VTNET_SQ_RQ_QUEUE_IDX;
+   uint16_t vq_idx = 2 * queue_idx + VTNET_SQ_RQ_QUEUE_IDX;
struct virtio_hw *hw = dev->data->dev_private;
-   struct virtqueue *vq = hw->vqs[vtpci_queue_idx];
+   struct virtqueue *vq = hw->vqs[vq_idx];
struct virtnet_rx *rxvq;
uint16_t rx_free_thresh;
 
@@ -678,9 +678,9 @@ virtio_dev_rx_queue_setup(struct rte_eth_dev *dev,
 int
 virtio_dev_rx_queue_setup_finish(struct rte_eth_dev *dev, u

[dpdk-dev] [PATCH v3 23/44] net/virtio: move vring alignment to generic header

[Maxime Coquelin]

This patch moves vring alignment define to the generic
Virtio header.

Signed-off-by: Maxime Coquelin 
Reviewed-by: Chenbo Xia 
Reviewed-by: David Marchand 
---
 drivers/net/virtio/virtio.h |  3 +++
 drivers/net/virtio/virtio_ethdev.c  | 10 +-
 drivers/net/virtio/virtio_pci.c |  2 +-
 drivers/net/virtio/virtio_pci.h |  3 ---
 drivers/net/virtio/virtio_user_ethdev.c |  4 ++--
 5 files changed, 11 insertions(+), 11 deletions(-)

diff --git a/drivers/net/virtio/virtio.h b/drivers/net/virtio/virtio.h
index a7629ad16b..21d54904e7 100644
--- a/drivers/net/virtio/virtio.h
+++ b/drivers/net/virtio/virtio.h
@@ -131,6 +131,9 @@
 /* Vector value used to disable MSI for queue. */
 #define VIRTIO_MSI_NO_VECTOR 0x
 
+/* The alignment to use between consumer and producer parts of vring. */
+#define VIRTIO_VRING_ALIGN 4096
+
 /*
  * This structure is just a reference to read net device specific
  * config space; it is just a shadow structure.
diff --git a/drivers/net/virtio/virtio_ethdev.c 
b/drivers/net/virtio/virtio_ethdev.c
index 51eb5aebf3..a2195bdda8 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -407,12 +407,12 @@ virtio_init_vring(struct virtqueue *vq)
memset(vq->vq_descx, 0, sizeof(struct vq_desc_extra) * vq->vq_nentries);
if (virtio_with_packed_queue(vq->hw)) {
vring_init_packed(&vq->vq_packed.ring, ring_mem,
- VIRTIO_PCI_VRING_ALIGN, size);
+ VIRTIO_VRING_ALIGN, size);
vring_desc_init_packed(vq, size);
} else {
struct vring *vr = &vq->vq_split.ring;
 
-   vring_init_split(vr, ring_mem, VIRTIO_PCI_VRING_ALIGN, size);
+   vring_init_split(vr, ring_mem, VIRTIO_VRING_ALIGN, size);
vring_desc_init_split(vr->desc, size);
}
/*
@@ -497,14 +497,14 @@ virtio_init_queue(struct rte_eth_dev *dev, uint16_t 
vtpci_queue_idx)
/*
 * Reserve a memzone for vring elements
 */
-   size = vring_size(hw, vq_size, VIRTIO_PCI_VRING_ALIGN);
-   vq->vq_ring_size = RTE_ALIGN_CEIL(size, VIRTIO_PCI_VRING_ALIGN);
+   size = vring_size(hw, vq_size, VIRTIO_VRING_ALIGN);
+   vq->vq_ring_size = RTE_ALIGN_CEIL(size, VIRTIO_VRING_ALIGN);
PMD_INIT_LOG(DEBUG, "vring_size: %d, rounded_vring_size: %d",
 size, vq->vq_ring_size);
 
mz = rte_memzone_reserve_aligned(vq_name, vq->vq_ring_size,
numa_node, RTE_MEMZONE_IOVA_CONTIG,
-   VIRTIO_PCI_VRING_ALIGN);
+   VIRTIO_VRING_ALIGN);
if (mz == NULL) {
if (rte_errno == EEXIST)
mz = rte_memzone_lookup(vq_name);
diff --git a/drivers/net/virtio/virtio_pci.c b/drivers/net/virtio/virtio_pci.c
index 905534ac72..d27e9eb515 100644
--- a/drivers/net/virtio/virtio_pci.c
+++ b/drivers/net/virtio/virtio_pci.c
@@ -492,7 +492,7 @@ modern_setup_queue(struct virtio_hw *hw, struct virtqueue 
*vq)
avail_addr = desc_addr + vq->vq_nentries * sizeof(struct vring_desc);
used_addr = RTE_ALIGN_CEIL(avail_addr + offsetof(struct vring_avail,
 ring[vq->vq_nentries]),
-  VIRTIO_PCI_VRING_ALIGN);
+  VIRTIO_VRING_ALIGN);
 
rte_write16(vq->vq_queue_index, &dev->common_cfg->queue_select);
 
diff --git a/drivers/net/virtio/virtio_pci.h b/drivers/net/virtio/virtio_pci.h
index 68f36dc0dd..6cdf2c55dd 100644
--- a/drivers/net/virtio/virtio_pci.h
+++ b/drivers/net/virtio/virtio_pci.h
@@ -122,9 +122,6 @@ struct virtio_pci_dev {
  */
 #define VIRTIO_PCI_QUEUE_ADDR_SHIFT 12
 
-/* The alignment to use between consumer and producer parts of vring. */
-#define VIRTIO_PCI_VRING_ALIGN 4096
-
 /*
  * Function declaration from virtio_pci.c
  */
diff --git a/drivers/net/virtio/virtio_user_ethdev.c 
b/drivers/net/virtio/virtio_user_ethdev.c
index c3dc3908d6..91794b8645 100644
--- a/drivers/net/virtio/virtio_user_ethdev.c
+++ b/drivers/net/virtio/virtio_user_ethdev.c
@@ -388,7 +388,7 @@ virtio_user_setup_queue_packed(struct virtqueue *vq,
sizeof(struct vring_packed_desc);
used_addr = RTE_ALIGN_CEIL(avail_addr +
   sizeof(struct vring_packed_desc_event),
-  VIRTIO_PCI_VRING_ALIGN);
+  VIRTIO_VRING_ALIGN);
vring->num = vq->vq_nentries;
vring->desc = (void *)(uintptr_t)desc_addr;
vring->driver = (void *)(uintptr_t)avail_addr;
@@ -410,7 +410,7 @@ virtio_user_setup_queue_split(struct virtqueue *vq, struct 
virtio_user_dev *dev)
avail_addr = desc_addr + vq->vq_nentries * sizeof(struct vring_desc);
used_addr = RTE_ALIGN_CEIL(avail_addr + offsetof(struct vring_avail,
 ring[vq->v

[dpdk-dev] [PATCH v3 25/44] net/virtio: make Vhost-user request sender consistent

[Maxime Coquelin]

This patch makes vhost_user_write() consistent with
vhost_user_read(), by passing a Vhost-user message
pointer instead of a buffer pointer and its length, which
is now calculated in the function.

Signed-off-by: Maxime Coquelin 
Reviewed-by: Chenbo Xia 
---
 drivers/net/virtio/virtio_user/vhost_user.c | 10 --
 1 file changed, 4 insertions(+), 6 deletions(-)

diff --git a/drivers/net/virtio/virtio_user/vhost_user.c 
b/drivers/net/virtio/virtio_user/vhost_user.c
index 350eed4182..55c81a 100644
--- a/drivers/net/virtio/virtio_user/vhost_user.c
+++ b/drivers/net/virtio/virtio_user/vhost_user.c
@@ -51,7 +51,7 @@ struct vhost_user_msg {
(sizeof(struct vhost_user_msg) - VHOST_USER_HDR_SIZE)
 
 static int
-vhost_user_write(int fd, void *buf, int len, int *fds, int fd_num)
+vhost_user_write(int fd, struct vhost_user_msg *msg, int *fds, int fd_num)
 {
int r;
struct msghdr msgh;
@@ -63,8 +63,8 @@ vhost_user_write(int fd, void *buf, int len, int *fds, int 
fd_num)
memset(&msgh, 0, sizeof(msgh));
memset(control, 0, sizeof(control));
 
-   iov.iov_base = (uint8_t *)buf;
-   iov.iov_len = len;
+   iov.iov_base = (uint8_t *)msg;
+   iov.iov_len = VHOST_USER_HDR_SIZE + msg->size;
 
msgh.msg_iov = &iov;
msgh.msg_iovlen = 1;
@@ -259,7 +259,6 @@ vhost_user_sock(struct virtio_user_dev *dev,
int has_reply_ack = 0;
int fds[VHOST_MEMORY_MAX_NREGIONS];
int fd_num = 0;
-   int len;
int vhostfd = dev->vhostfd;
 
RTE_SET_USED(m);
@@ -364,8 +363,7 @@ vhost_user_sock(struct virtio_user_dev *dev,
return -1;
}
 
-   len = VHOST_USER_HDR_SIZE + msg.size;
-   if (vhost_user_write(vhostfd, &msg, len, fds, fd_num) < 0) {
+   if (vhost_user_write(vhostfd, &msg, fds, fd_num) < 0) {
PMD_DRV_LOG(ERR, "%s failed: %s",
vhost_msg_strings[req], strerror(errno));
return -1;
-- 
2.29.2