Re: [dpdk-dev] [PATCH 1/2] vhost: reference count fix for nb_started_ports
On Wed, Dec 28, 2016 at 04:10:51PM -0500, Charles (Chas) Williams wrote:
> From: Wen Chiu
>
> Only increment and decrement nb_started_ports on the first and last
> device start and stop. Otherwise, nb_started_ports can become negative
> if a device is stopped multiple times.
How could you be able to stop dev (precisely, invoke eth_dev_stop)
multiple times, judging that eth_dev_stop() will be invoked once
only?
void
rte_eth_dev_stop(uint8_t port_id)
{
struct rte_eth_dev *dev;
RTE_ETH_VALID_PORTID_OR_RET(port_id);
dev = &rte_eth_devices[port_id];
RTE_FUNC_PTR_OR_RET(*dev->dev_ops->dev_stop);
==> if (dev->data->dev_started == 0) {
RTE_PMD_DEBUG_TRACE("Device with port_id=%" PRIu8
" already stopped\n",
port_id);
return;
}
==> dev->data->dev_started = 0;
(*dev->dev_ops->dev_stop)(dev);
}
Multiple threads?
--yliu
Re: [dpdk-dev] [PATCH 2/2] vhost: start vhost servers once
On Wed, Dec 28, 2016 at 04:10:52PM -0500, Charles (Chas) Williams wrote: > Start a vhost server once during devinit instead of during device start > and stop. Some vhost clients, QEMU, don't re-attaching to sockets when > the vhost server is stopped and later started. Preserve existing behavior > for vhost clients. I didn't quite get the idea what you are going to fix. --yliu
Re: [dpdk-dev] [PATCH v2 3/7] net/virtio_user: move vhost user specific code
Hi Yuanhan, On 12/26/2016 3:57 PM, Yuanhan Liu wrote: On Mon, Dec 26, 2016 at 06:58:58AM +, Tan, Jianfeng wrote: -Original Message- From: Yuanhan Liu [mailto:yuanhan@linux.intel.com] Sent: Monday, December 26, 2016 2:28 PM To: Tan, Jianfeng Cc: dev@dpdk.org; Yigit, Ferruh; Liang, Cunming Subject: Re: [PATCH v2 3/7] net/virtio_user: move vhost user specific code On Fri, Dec 23, 2016 at 07:14:22AM +, Jianfeng Tan wrote: To support vhost kernel as the backend of net_virtio_user in coming patches, we move vhost_user specific structs and macros into vhost_user.c, and only keep common definitions in vhost.h. Good. Besides, remove VHOST_USER_MQ feature check, it will be added back in following multiqueue patch. Why then? Only vhost user recognizes this feature bit, and vhost kernel does not. My intension is to put those vhost user specific code inside vhost user. That's okay. But why you have to remove it and then add it back? After second thought , I agree that it's necessary to add it back. Currently, our vhost does not implement the semantics of VHOST_USER_F_PROTOCOL_FEATURES. Besides, all other comments from you on this patch series will be addressed on next version. Thank you very much for those suggestions! Thanks, Jianfeng --yliu And in fact, I forget to add it back. I should add it back in this patch. Thanks, Jianfeng --yliu
Re: [dpdk-dev] [PATCH v2 1/3] crypto/aesni_gcm: fix J0 padding bytes for GCM
Hi Arek,
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Arek Kusztal
> Sent: Friday, December 23, 2016 9:25 AM
> To: dev@dpdk.org
> Cc: Trahe, Fiona ; De Lara Guarch, Pablo
> ; Griffin, John ;
> Jain, Deepak K ; Doherty, Declan
> ; Kusztal, ArkadiuszX
>
> Subject: [dpdk-dev] [PATCH v2 1/3] crypto/aesni_gcm: fix J0 padding bytes
> for GCM
>
> This commit fixes pre-counter block (J0) padding by clearing four most
> significant bytes before setting initial counter value.
>
> Fixes: b2bb3597470c ("crypto/aesni_gcm: move pre-counter block to driver")
>
> Signed-off-by: Arek Kusztal
> ---
> drivers/crypto/aesni_gcm/aesni_gcm_pmd.c | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/crypto/aesni_gcm/aesni_gcm_pmd.c
> b/drivers/crypto/aesni_gcm/aesni_gcm_pmd.c
> index dba5e15..af3d60f 100644
> --- a/drivers/crypto/aesni_gcm/aesni_gcm_pmd.c
> +++ b/drivers/crypto/aesni_gcm/aesni_gcm_pmd.c
> @@ -40,6 +40,7 @@
> #include
> #include
> #include
> +#include
>
> #include "aesni_gcm_pmd_private.h"
>
> @@ -241,7 +242,8 @@ process_gcm_crypto_op(struct aesni_gcm_qp *qp,
> struct rte_crypto_sym_op *op,
>* to set BE LSB to 1, driver expects that 16B is allocated
I think that 16B expected by driver while only 12B IV is supported is not clear
from user perspective.
I think that we should expect 12B only and allocate 16B locally.
>*/
> if (op->cipher.iv.length == 12) {
> - op->cipher.iv.data[15] = 1;
> + uint32_t *iv_padd = (uint32_t *)&op->cipher.iv.data[12];
> + *iv_padd = rte_bswap32(1);
Should not be that the last byte (number 15) always be set to 1?
> }
>
> if (op->auth.aad.length != 12 && op->auth.aad.length != 8 &&
> --
> 2.1.0
Re: [dpdk-dev] [PATCH v2 2/3] crypto/aesni_gcm: fix iv size in PMD capabilities
> Subject: [dpdk-dev] [PATCH v2 2/3] crypto/aesni_gcm: fix iv size in PMD
> capabilities
>
> This patch sets iv size in aesni gcm PMD to 12 bytes to be conformant with
> nist SP800-38D.
>
> Fixes: eec136f3c54f ("aesni_gcm: add driver for AES-GCM crypto
> operations")
>
> Signed-off-by: Arek Kusztal
> ---
> drivers/crypto/aesni_gcm/aesni_gcm_pmd_ops.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/crypto/aesni_gcm/aesni_gcm_pmd_ops.c
> b/drivers/crypto/aesni_gcm/aesni_gcm_pmd_ops.c
> index e824d4b..c51f82a 100644
> --- a/drivers/crypto/aesni_gcm/aesni_gcm_pmd_ops.c
> +++ b/drivers/crypto/aesni_gcm/aesni_gcm_pmd_ops.c
> @@ -77,8 +77,8 @@ static const struct rte_cryptodev_capabilities
> aesni_gcm_pmd_capabilities[] = {
> .increment = 0
> },
> .iv_size = {
> - .min = 16,
> - .max = 16,
> + .min = 12,
> + .max = 12,
> .increment = 0
> }
> }, }
I think that we should also remove 16 na 0 bytes allowed in
process_gcm_crypto_op() function:
if (op->cipher.iv.length != 16 && op->cipher.iv.length != 12 &&
op->cipher.iv.length != 0) {
GCM_LOG_ERR("iv");
return -1;
}
Regards,
Piotr
[dpdk-dev] [PATCH v5 0/6] net/mlx5: support flow API
Changes in v5: - Fix masking when only spec is present in item structure. - Fix first element of flow items array. Changes in v4: - Simplify flow parsing by using a graph. - Add VXLAN flow item. - Add mark flow action. - Extend IPv4 filter item (Type of service, Next Protocol ID). Changes in v3: - Fix Ethernet ether type issue. Changes in v2: - Fix several issues. - Support VLAN filtering. Nelio Laranjeiro (6): net/mlx5: add preliminary flow API support net/mlx5: support basic flow items and actions net/mlx5: support VLAN flow item net/mlx5: support VXLAN flow item net/mlx5: support mark flow action net/mlx5: extend IPv4 flow item drivers/net/mlx5/Makefile |1 + drivers/net/mlx5/mlx5.h | 19 + drivers/net/mlx5/mlx5_fdir.c| 15 + drivers/net/mlx5/mlx5_flow.c| 1248 +++ drivers/net/mlx5/mlx5_prm.h | 70 ++- drivers/net/mlx5/mlx5_rxtx.c| 12 +- drivers/net/mlx5/mlx5_rxtx.h|3 +- drivers/net/mlx5/mlx5_trigger.c |2 + 8 files changed, 1367 insertions(+), 3 deletions(-) create mode 100644 drivers/net/mlx5/mlx5_flow.c -- 2.1.4
[dpdk-dev] [PATCH v5 1/6] net/mlx5: add preliminary flow API support
Signed-off-by: Nelio Laranjeiro
Acked-by: Adrien Mazarguil
---
drivers/net/mlx5/Makefile| 1 +
drivers/net/mlx5/mlx5.h | 16 ++
drivers/net/mlx5/mlx5_fdir.c | 15 ++
drivers/net/mlx5/mlx5_flow.c | 124 +++
4 files changed, 156 insertions(+)
create mode 100644 drivers/net/mlx5/mlx5_flow.c
diff --git a/drivers/net/mlx5/Makefile b/drivers/net/mlx5/Makefile
index cf87f0b..6d1338a 100644
--- a/drivers/net/mlx5/Makefile
+++ b/drivers/net/mlx5/Makefile
@@ -48,6 +48,7 @@ SRCS-$(CONFIG_RTE_LIBRTE_MLX5_PMD) += mlx5_stats.c
SRCS-$(CONFIG_RTE_LIBRTE_MLX5_PMD) += mlx5_rss.c
SRCS-$(CONFIG_RTE_LIBRTE_MLX5_PMD) += mlx5_fdir.c
SRCS-$(CONFIG_RTE_LIBRTE_MLX5_PMD) += mlx5_mr.c
+SRCS-$(CONFIG_RTE_LIBRTE_MLX5_PMD) += mlx5_flow.c
# Dependencies.
DEPDIRS-$(CONFIG_RTE_LIBRTE_MLX5_PMD) += lib/librte_ether
diff --git a/drivers/net/mlx5/mlx5.h b/drivers/net/mlx5/mlx5.h
index 79b7a60..04f4eaa 100644
--- a/drivers/net/mlx5/mlx5.h
+++ b/drivers/net/mlx5/mlx5.h
@@ -59,6 +59,7 @@
#include
#include
#include
+#include
#ifdef PEDANTIC
#pragma GCC diagnostic error "-Wpedantic"
#endif
@@ -268,4 +269,19 @@ void priv_fdir_enable(struct priv *);
int mlx5_dev_filter_ctrl(struct rte_eth_dev *, enum rte_filter_type,
enum rte_filter_op, void *);
+/* mlx5_flow.c */
+
+int mlx5_flow_validate(struct rte_eth_dev *, const struct rte_flow_attr *,
+ const struct rte_flow_item [],
+ const struct rte_flow_action [],
+ struct rte_flow_error *);
+struct rte_flow *mlx5_flow_create(struct rte_eth_dev *,
+ const struct rte_flow_attr *,
+ const struct rte_flow_item [],
+ const struct rte_flow_action [],
+ struct rte_flow_error *);
+int mlx5_flow_destroy(struct rte_eth_dev *, struct rte_flow *,
+ struct rte_flow_error *);
+int mlx5_flow_flush(struct rte_eth_dev *, struct rte_flow_error *);
+
#endif /* RTE_PMD_MLX5_H_ */
diff --git a/drivers/net/mlx5/mlx5_fdir.c b/drivers/net/mlx5/mlx5_fdir.c
index 1acf682..f80c58b 100644
--- a/drivers/net/mlx5/mlx5_fdir.c
+++ b/drivers/net/mlx5/mlx5_fdir.c
@@ -55,6 +55,8 @@
#include
#include
#include
+#include
+#include
#ifdef PEDANTIC
#pragma GCC diagnostic error "-Wpedantic"
#endif
@@ -1042,6 +1044,14 @@ priv_fdir_ctrl_func(struct priv *priv, enum
rte_filter_op filter_op, void *arg)
return ret;
}
+static const struct rte_flow_ops mlx5_flow_ops = {
+ .validate = mlx5_flow_validate,
+ .create = mlx5_flow_create,
+ .destroy = mlx5_flow_destroy,
+ .flush = mlx5_flow_flush,
+ .query = NULL,
+};
+
/**
* Manage filter operations.
*
@@ -1067,6 +1077,11 @@ mlx5_dev_filter_ctrl(struct rte_eth_dev *dev,
struct priv *priv = dev->data->dev_private;
switch (filter_type) {
+ case RTE_ETH_FILTER_GENERIC:
+ if (filter_op != RTE_ETH_FILTER_GET)
+ return -EINVAL;
+ *(const void **)arg = &mlx5_flow_ops;
+ return 0;
case RTE_ETH_FILTER_FDIR:
priv_lock(priv);
ret = priv_fdir_ctrl_func(priv, filter_op, arg);
diff --git a/drivers/net/mlx5/mlx5_flow.c b/drivers/net/mlx5/mlx5_flow.c
new file mode 100644
index 000..4fdefa0
--- /dev/null
+++ b/drivers/net/mlx5/mlx5_flow.c
@@ -0,0 +1,124 @@
+/*-
+ * BSD LICENSE
+ *
+ * Copyright 2016 6WIND S.A.
+ * Copyright 2016 Mellanox.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * * Neither the name of 6WIND S.A. nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN C
[dpdk-dev] [PATCH v5 2/6] net/mlx5: support basic flow items and actions
Introduce initial software for rte_flow rules.
VLAN, VXLAN are still not supported.
Signed-off-by: Nelio Laranjeiro
Acked-by: Adrien Mazarguil
---
drivers/net/mlx5/mlx5.h | 3 +
drivers/net/mlx5/mlx5_flow.c| 954 ++--
drivers/net/mlx5/mlx5_trigger.c | 2 +
3 files changed, 930 insertions(+), 29 deletions(-)
diff --git a/drivers/net/mlx5/mlx5.h b/drivers/net/mlx5/mlx5.h
index 04f4eaa..c415ce3 100644
--- a/drivers/net/mlx5/mlx5.h
+++ b/drivers/net/mlx5/mlx5.h
@@ -136,6 +136,7 @@ struct priv {
unsigned int reta_idx_n; /* RETA index size. */
struct fdir_filter_list *fdir_filter_list; /* Flow director rules. */
struct fdir_queue *fdir_drop_queue; /* Flow director drop queue. */
+ LIST_HEAD(mlx5_flows, rte_flow) flows; /* RTE Flow rules. */
uint32_t link_speed_capa; /* Link speed capabilities. */
rte_spinlock_t lock; /* Lock for control functions. */
};
@@ -283,5 +284,7 @@ struct rte_flow *mlx5_flow_create(struct rte_eth_dev *,
int mlx5_flow_destroy(struct rte_eth_dev *, struct rte_flow *,
struct rte_flow_error *);
int mlx5_flow_flush(struct rte_eth_dev *, struct rte_flow_error *);
+int priv_flow_start(struct priv *);
+void priv_flow_stop(struct priv *);
#endif /* RTE_PMD_MLX5_H_ */
diff --git a/drivers/net/mlx5/mlx5_flow.c b/drivers/net/mlx5/mlx5_flow.c
index 4fdefa0..4f6696e 100644
--- a/drivers/net/mlx5/mlx5_flow.c
+++ b/drivers/net/mlx5/mlx5_flow.c
@@ -31,12 +31,387 @@
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
+#include
+#include
+
+/* Verbs header. */
+/* ISO C doesn't support unnamed structs/unions, disabling -pedantic. */
+#ifdef PEDANTIC
+#pragma GCC diagnostic ignored "-Wpedantic"
+#endif
+#include
+#ifdef PEDANTIC
+#pragma GCC diagnostic error "-Wpedantic"
+#endif
+
#include
#include
#include
+#include
#include "mlx5.h"
+static int
+mlx5_flow_create_eth(const struct rte_flow_item *item,
+const void *default_mask,
+void *data);
+
+static int
+mlx5_flow_create_ipv4(const struct rte_flow_item *item,
+ const void *default_mask,
+ void *data);
+
+static int
+mlx5_flow_create_ipv6(const struct rte_flow_item *item,
+ const void *default_mask,
+ void *data);
+
+static int
+mlx5_flow_create_udp(const struct rte_flow_item *item,
+const void *default_mask,
+void *data);
+
+static int
+mlx5_flow_create_tcp(const struct rte_flow_item *item,
+const void *default_mask,
+void *data);
+
+struct rte_flow {
+ LIST_ENTRY(rte_flow) next; /**< Pointer to the next flow structure. */
+ struct ibv_exp_flow_attr *ibv_attr; /**< Pointer to Verbs attributes. */
+ struct ibv_exp_rwq_ind_table *ind_table; /**< Indirection table. */
+ struct ibv_qp *qp; /**< Verbs queue pair. */
+ struct ibv_exp_flow *ibv_flow; /**< Verbs flow. */
+ struct ibv_exp_wq *wq; /**< Verbs work queue. */
+ struct ibv_cq *cq; /**< Verbs completion queue. */
+ struct rxq *rxq; /**< Pointer to the queue, NULL if drop queue. */
+};
+
+/** Static initializer for items. */
+#define ITEMS(...) \
+ (const enum rte_flow_item_type []){ \
+ __VA_ARGS__, RTE_FLOW_ITEM_TYPE_END, \
+ }
+
+/** Structure to generate a simple graph of layers supported by the NIC. */
+struct mlx5_flow_items {
+ /** List of possible actions for these items. */
+ const enum rte_flow_action_type *const actions;
+ /** Bit-masks corresponding to the possibilities for the item. */
+ const void *mask;
+ /** Bit-masks size in bytes. */
+ const unsigned int mask_sz;
+ /**
+* Conversion function from rte_flow to NIC specific flow.
+*
+* @param item
+* rte_flow item to convert.
+* @param default_mask
+* Default bit-masks to use when item->mask is not provided.
+* @param data
+* Internal structure to store the conversion.
+*
+* @return
+* 0 on success, negative value otherwise.
+*/
+ int (*convert)(const struct rte_flow_item *item,
+ const void *default_mask,
+ void *data);
+ /** Size in bytes of the destination structure. */
+ const unsigned int dst_sz;
+ /** List of possible following items. */
+ const enum rte_flow_item_type *const items;
+};
+
+/** Valid action for this PMD. */
+static const enum rte_flow_action_type valid_actions[] = {
+ RTE_FLOW_ACTION_TYPE_DROP,
+ RTE_FLOW_ACTION_TYPE_QUEUE,
+ RTE_FLOW_ACTION_TYPE_END,
+};
+
+/** Graph of supported items and associated actions. */
+static const struct mlx5_flow_items mlx5_flow_items[] = {
+ [RTE_FLOW_ITEM_TYPE_END] = {
+ .items = ITEMS(RTE_F
[dpdk-dev] [PATCH v5 3/6] net/mlx5: support VLAN flow item
Signed-off-by: Nelio Laranjeiro
Acked-by: Adrien Mazarguil
---
drivers/net/mlx5/mlx5_flow.c | 62 +++-
1 file changed, 61 insertions(+), 1 deletion(-)
diff --git a/drivers/net/mlx5/mlx5_flow.c b/drivers/net/mlx5/mlx5_flow.c
index 4f6696e..8f2f4d5 100644
--- a/drivers/net/mlx5/mlx5_flow.c
+++ b/drivers/net/mlx5/mlx5_flow.c
@@ -57,6 +57,11 @@ mlx5_flow_create_eth(const struct rte_flow_item *item,
void *data);
static int
+mlx5_flow_create_vlan(const struct rte_flow_item *item,
+ const void *default_mask,
+ void *data);
+
+static int
mlx5_flow_create_ipv4(const struct rte_flow_item *item,
const void *default_mask,
void *data);
@@ -136,7 +141,8 @@ static const struct mlx5_flow_items mlx5_flow_items[] = {
.items = ITEMS(RTE_FLOW_ITEM_TYPE_ETH),
},
[RTE_FLOW_ITEM_TYPE_ETH] = {
- .items = ITEMS(RTE_FLOW_ITEM_TYPE_IPV4,
+ .items = ITEMS(RTE_FLOW_ITEM_TYPE_VLAN,
+ RTE_FLOW_ITEM_TYPE_IPV4,
RTE_FLOW_ITEM_TYPE_IPV6),
.actions = valid_actions,
.mask = &(const struct rte_flow_item_eth){
@@ -147,6 +153,17 @@ static const struct mlx5_flow_items mlx5_flow_items[] = {
.convert = mlx5_flow_create_eth,
.dst_sz = sizeof(struct ibv_exp_flow_spec_eth),
},
+ [RTE_FLOW_ITEM_TYPE_VLAN] = {
+ .items = ITEMS(RTE_FLOW_ITEM_TYPE_IPV4,
+ RTE_FLOW_ITEM_TYPE_IPV6),
+ .actions = valid_actions,
+ .mask = &(const struct rte_flow_item_vlan){
+ .tci = -1,
+ },
+ .mask_sz = sizeof(struct rte_flow_item_vlan),
+ .convert = mlx5_flow_create_vlan,
+ .dst_sz = 0,
+ },
[RTE_FLOW_ITEM_TYPE_IPV4] = {
.items = ITEMS(RTE_FLOW_ITEM_TYPE_UDP,
RTE_FLOW_ITEM_TYPE_TCP),
@@ -355,6 +372,17 @@ priv_flow_validate(struct priv *priv,
if (items->type == RTE_FLOW_ITEM_TYPE_VOID)
continue;
+ /* Handle special situation for VLAN. */
+ if (items->type == RTE_FLOW_ITEM_TYPE_VLAN) {
+ if (((const struct rte_flow_item_vlan *)items)->tci >
+ ETHER_MAX_VLAN_ID) {
+ rte_flow_error_set(error, ENOTSUP,
+ RTE_FLOW_ERROR_TYPE_ITEM,
+ items,
+ "wrong VLAN id value");
+ return -rte_errno;
+ }
+ }
for (i = 0;
cur_item->items &&
cur_item->items[i] != RTE_FLOW_ITEM_TYPE_END;
@@ -481,6 +509,38 @@ mlx5_flow_create_eth(const struct rte_flow_item *item,
}
/**
+ * Convert VLAN item to Verbs specification.
+ *
+ * @param item[in]
+ * Item specification.
+ * @param default_mask[in]
+ * Default bit-masks to use when item->mask is not provided.
+ * @param data[in, out]
+ * User structure.
+ */
+static int
+mlx5_flow_create_vlan(const struct rte_flow_item *item,
+ const void *default_mask,
+ void *data)
+{
+ const struct rte_flow_item_vlan *spec = item->spec;
+ const struct rte_flow_item_vlan *mask = item->mask;
+ struct mlx5_flow *flow = (struct mlx5_flow *)data;
+ struct ibv_exp_flow_spec_eth *eth;
+ const unsigned int eth_size = sizeof(struct ibv_exp_flow_spec_eth);
+
+ eth = (void *)((uintptr_t)flow->ibv_attr + flow->offset - eth_size);
+ if (!spec)
+ return 0;
+ if (!mask)
+ mask = default_mask;
+ eth->val.vlan_tag = spec->tci;
+ eth->mask.vlan_tag = mask->tci;
+ eth->val.vlan_tag &= eth->mask.vlan_tag;
+ return 0;
+}
+
+/**
* Convert IPv4 item to Verbs specification.
*
* @param item[in]
--
2.1.4
[dpdk-dev] [PATCH v5 5/6] net/mlx5: support mark flow action
Signed-off-by: Nelio Laranjeiro
Acked-by: Adrien Mazarguil
---
drivers/net/mlx5/mlx5_flow.c | 78
drivers/net/mlx5/mlx5_prm.h | 70 ++-
drivers/net/mlx5/mlx5_rxtx.c | 12 ++-
drivers/net/mlx5/mlx5_rxtx.h | 3 +-
4 files changed, 160 insertions(+), 3 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_flow.c b/drivers/net/mlx5/mlx5_flow.c
index 093c140..0e7ea99 100644
--- a/drivers/net/mlx5/mlx5_flow.c
+++ b/drivers/net/mlx5/mlx5_flow.c
@@ -50,6 +50,7 @@
#include
#include "mlx5.h"
+#include "mlx5_prm.h"
static int
mlx5_flow_create_eth(const struct rte_flow_item *item,
@@ -95,6 +96,7 @@ struct rte_flow {
struct ibv_exp_wq *wq; /**< Verbs work queue. */
struct ibv_cq *cq; /**< Verbs completion queue. */
struct rxq *rxq; /**< Pointer to the queue, NULL if drop queue. */
+ uint32_t mark:1; /**< Set if the flow is marked. */
};
/** Static initializer for items. */
@@ -137,6 +139,7 @@ struct mlx5_flow_items {
static const enum rte_flow_action_type valid_actions[] = {
RTE_FLOW_ACTION_TYPE_DROP,
RTE_FLOW_ACTION_TYPE_QUEUE,
+ RTE_FLOW_ACTION_TYPE_MARK,
RTE_FLOW_ACTION_TYPE_END,
};
@@ -255,7 +258,9 @@ struct mlx5_flow {
struct mlx5_flow_action {
uint32_t queue:1; /**< Target is a receive queue. */
uint32_t drop:1; /**< Target is a drop queue. */
+ uint32_t mark:1; /**< Mark is present in the flow. */
uint32_t queue_id; /**< Identifier of the queue. */
+ uint32_t mark_id; /**< Mark identifier. */
};
/**
@@ -352,6 +357,7 @@ priv_flow_validate(struct priv *priv,
struct mlx5_flow_action action = {
.queue = 0,
.drop = 0,
+ .mark = 0,
};
(void)priv;
@@ -438,10 +444,26 @@ priv_flow_validate(struct priv *priv,
if (!queue || (queue->index > (priv->rxqs_n - 1)))
goto exit_action_not_supported;
action.queue = 1;
+ } else if (actions->type == RTE_FLOW_ACTION_TYPE_MARK) {
+ const struct rte_flow_action_mark *mark =
+ (const struct rte_flow_action_mark *)
+ actions->conf;
+
+ if (mark && (mark->id >= MLX5_FLOW_MARK_MAX)) {
+ rte_flow_error_set(error, ENOTSUP,
+ RTE_FLOW_ERROR_TYPE_ACTION,
+ actions,
+ "mark must be between 0"
+ " and 16777199");
+ return -rte_errno;
+ }
+ action.mark = 1;
} else {
goto exit_action_not_supported;
}
}
+ if (action.mark && !flow->ibv_attr)
+ flow->offset += sizeof(struct ibv_exp_flow_spec_action_tag);
if (!action.queue && !action.drop) {
rte_flow_error_set(error, ENOTSUP, RTE_FLOW_ERROR_TYPE_HANDLE,
NULL, "no valid action");
@@ -785,6 +807,30 @@ mlx5_flow_create_vxlan(const struct rte_flow_item *item,
}
/**
+ * Convert mark/flag action to Verbs specification.
+ *
+ * @param flow
+ * Pointer to MLX5 flow structure.
+ * @param mark_id
+ * Mark identifier.
+ */
+static int
+mlx5_flow_create_flag_mark(struct mlx5_flow *flow, uint32_t mark_id)
+{
+ struct ibv_exp_flow_spec_action_tag *tag;
+ unsigned int size = sizeof(struct ibv_exp_flow_spec_action_tag);
+
+ tag = (void *)((uintptr_t)flow->ibv_attr + flow->offset);
+ *tag = (struct ibv_exp_flow_spec_action_tag){
+ .type = IBV_EXP_FLOW_SPEC_ACTION_TAG,
+ .size = size,
+ .tag_id = mlx5_flow_mark_set(mark_id),
+ };
+ ++flow->ibv_attr->num_of_specs;
+ return 0;
+}
+
+/**
* Complete flow rule creation.
*
* @param priv
@@ -840,8 +886,10 @@ priv_flow_create_action_queue(struct priv *priv,
rxq = container_of((*priv->rxqs)[action->queue_id],
struct rxq_ctrl, rxq);
rte_flow->rxq = &rxq->rxq;
+ rxq->rxq.mark |= action->mark;
rte_flow->wq = rxq->wq;
}
+ rte_flow->mark = action->mark;
rte_flow->ibv_attr = ibv_attr;
rte_flow->ind_table = ibv_exp_create_rwq_ind_table(
priv->ctx,
@@ -957,6 +1005,8 @@ priv_flow_create(struct priv *priv,
action = (struct mlx5_flow_action){
.queue = 0,
.drop = 0,
+ .mark = 0,
+ .mark_id = MLX5_FLOW_MARK_DEFAULT,
};
for (; actions->type != RTE_FLOW_ACTION_TYPE_END; ++actions) {
if (actions->type == RTE_FLO
[dpdk-dev] [PATCH v5 6/6] net/mlx5: extend IPv4 flow item
This commits adds:
- Type of service
- Next protocol ID
Signed-off-by: Nelio Laranjeiro
Acked-by: Adrien Mazarguil
---
drivers/net/mlx5/mlx5_flow.c | 40
1 file changed, 32 insertions(+), 8 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_flow.c b/drivers/net/mlx5/mlx5_flow.c
index 0e7ea99..d7ed686 100644
--- a/drivers/net/mlx5/mlx5_flow.c
+++ b/drivers/net/mlx5/mlx5_flow.c
@@ -111,6 +111,12 @@ struct mlx5_flow_items {
const enum rte_flow_action_type *const actions;
/** Bit-masks corresponding to the possibilities for the item. */
const void *mask;
+ /**
+* Default bit-masks to use when item->mask is not provided. When
+* \default_mask is also NULL, the full supported bit-mask (\mask) is
+* used instead.
+*/
+ const void *default_mask;
/** Bit-masks size in bytes. */
const unsigned int mask_sz;
/**
@@ -181,11 +187,19 @@ static const struct mlx5_flow_items mlx5_flow_items[] = {
.hdr = {
.src_addr = -1,
.dst_addr = -1,
+ .type_of_service = -1,
+ .next_proto_id = -1,
+ },
+ },
+ .default_mask = &(const struct rte_flow_item_ipv4){
+ .hdr = {
+ .src_addr = -1,
+ .dst_addr = -1,
},
},
.mask_sz = sizeof(struct rte_flow_item_ipv4),
.convert = mlx5_flow_create_ipv4,
- .dst_sz = sizeof(struct ibv_exp_flow_spec_ipv4),
+ .dst_sz = sizeof(struct ibv_exp_flow_spec_ipv4_ext),
},
[RTE_FLOW_ITEM_TYPE_IPV6] = {
.items = ITEMS(RTE_FLOW_ITEM_TYPE_UDP,
@@ -425,7 +439,11 @@ priv_flow_validate(struct priv *priv,
if (err)
goto exit_item_not_supported;
if (flow->ibv_attr && cur_item->convert) {
- err = cur_item->convert(items, cur_item->mask, flow);
+ err = cur_item->convert(items,
+ (cur_item->default_mask ?
+cur_item->default_mask :
+cur_item->mask),
+ flow);
if (err)
goto exit_item_not_supported;
}
@@ -598,31 +616,37 @@ mlx5_flow_create_ipv4(const struct rte_flow_item *item,
const struct rte_flow_item_ipv4 *spec = item->spec;
const struct rte_flow_item_ipv4 *mask = item->mask;
struct mlx5_flow *flow = (struct mlx5_flow *)data;
- struct ibv_exp_flow_spec_ipv4 *ipv4;
- unsigned int ipv4_size = sizeof(struct ibv_exp_flow_spec_ipv4);
+ struct ibv_exp_flow_spec_ipv4_ext *ipv4;
+ unsigned int ipv4_size = sizeof(struct ibv_exp_flow_spec_ipv4_ext);
++flow->ibv_attr->num_of_specs;
flow->ibv_attr->priority = 1;
ipv4 = (void *)((uintptr_t)flow->ibv_attr + flow->offset);
- *ipv4 = (struct ibv_exp_flow_spec_ipv4) {
- .type = flow->inner | IBV_EXP_FLOW_SPEC_IPV4,
+ *ipv4 = (struct ibv_exp_flow_spec_ipv4_ext) {
+ .type = flow->inner | IBV_EXP_FLOW_SPEC_IPV4_EXT,
.size = ipv4_size,
};
if (!spec)
return 0;
if (!mask)
mask = default_mask;
- ipv4->val = (struct ibv_exp_flow_ipv4_filter){
+ ipv4->val = (struct ibv_exp_flow_ipv4_ext_filter){
.src_ip = spec->hdr.src_addr,
.dst_ip = spec->hdr.dst_addr,
+ .proto = spec->hdr.next_proto_id,
+ .tos = spec->hdr.type_of_service,
};
- ipv4->mask = (struct ibv_exp_flow_ipv4_filter){
+ ipv4->mask = (struct ibv_exp_flow_ipv4_ext_filter){
.src_ip = mask->hdr.src_addr,
.dst_ip = mask->hdr.dst_addr,
+ .proto = mask->hdr.next_proto_id,
+ .tos = mask->hdr.type_of_service,
};
/* Remove unwanted bits from values. */
ipv4->val.src_ip &= ipv4->mask.src_ip;
ipv4->val.dst_ip &= ipv4->mask.dst_ip;
+ ipv4->val.proto &= ipv4->mask.proto;
+ ipv4->val.tos &= ipv4->mask.tos;
return 0;
}
--
2.1.4
[dpdk-dev] [PATCH v5 4/6] net/mlx5: support VXLAN flow item
Signed-off-by: Nelio Laranjeiro
Acked-by: Adrien Mazarguil
---
drivers/net/mlx5/mlx5_flow.c | 78
1 file changed, 72 insertions(+), 6 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_flow.c b/drivers/net/mlx5/mlx5_flow.c
index 8f2f4d5..093c140 100644
--- a/drivers/net/mlx5/mlx5_flow.c
+++ b/drivers/net/mlx5/mlx5_flow.c
@@ -81,6 +81,11 @@ mlx5_flow_create_tcp(const struct rte_flow_item *item,
const void *default_mask,
void *data);
+static int
+mlx5_flow_create_vxlan(const struct rte_flow_item *item,
+ const void *default_mask,
+ void *data);
+
struct rte_flow {
LIST_ENTRY(rte_flow) next; /**< Pointer to the next flow structure. */
struct ibv_exp_flow_attr *ibv_attr; /**< Pointer to Verbs attributes. */
@@ -138,7 +143,8 @@ static const enum rte_flow_action_type valid_actions[] = {
/** Graph of supported items and associated actions. */
static const struct mlx5_flow_items mlx5_flow_items[] = {
[RTE_FLOW_ITEM_TYPE_END] = {
- .items = ITEMS(RTE_FLOW_ITEM_TYPE_ETH),
+ .items = ITEMS(RTE_FLOW_ITEM_TYPE_ETH,
+ RTE_FLOW_ITEM_TYPE_VXLAN),
},
[RTE_FLOW_ITEM_TYPE_ETH] = {
.items = ITEMS(RTE_FLOW_ITEM_TYPE_VLAN,
@@ -203,6 +209,7 @@ static const struct mlx5_flow_items mlx5_flow_items[] = {
.dst_sz = sizeof(struct ibv_exp_flow_spec_ipv6),
},
[RTE_FLOW_ITEM_TYPE_UDP] = {
+ .items = ITEMS(RTE_FLOW_ITEM_TYPE_VXLAN),
.actions = valid_actions,
.mask = &(const struct rte_flow_item_udp){
.hdr = {
@@ -226,12 +233,23 @@ static const struct mlx5_flow_items mlx5_flow_items[] = {
.convert = mlx5_flow_create_tcp,
.dst_sz = sizeof(struct ibv_exp_flow_spec_tcp_udp),
},
+ [RTE_FLOW_ITEM_TYPE_VXLAN] = {
+ .items = ITEMS(RTE_FLOW_ITEM_TYPE_ETH),
+ .actions = valid_actions,
+ .mask = &(const struct rte_flow_item_vxlan){
+ .vni = "\xff\xff\xff",
+ },
+ .mask_sz = sizeof(struct rte_flow_item_vxlan),
+ .convert = mlx5_flow_create_vxlan,
+ .dst_sz = sizeof(struct ibv_exp_flow_spec_tunnel),
+ },
};
/** Structure to pass to the conversion function. */
struct mlx5_flow {
struct ibv_exp_flow_attr *ibv_attr; /**< Verbs attribute. */
unsigned int offset; /**< Offset in bytes in the ibv_attr buffer. */
+ uint32_t inner; /**< Set once VXLAN is encountered. */
};
struct mlx5_flow_action {
@@ -489,7 +507,7 @@ mlx5_flow_create_eth(const struct rte_flow_item *item,
flow->ibv_attr->priority = 2;
eth = (void *)((uintptr_t)flow->ibv_attr + flow->offset);
*eth = (struct ibv_exp_flow_spec_eth) {
- .type = IBV_EXP_FLOW_SPEC_ETH,
+ .type = flow->inner | IBV_EXP_FLOW_SPEC_ETH,
.size = eth_size,
};
if (!spec)
@@ -565,7 +583,7 @@ mlx5_flow_create_ipv4(const struct rte_flow_item *item,
flow->ibv_attr->priority = 1;
ipv4 = (void *)((uintptr_t)flow->ibv_attr + flow->offset);
*ipv4 = (struct ibv_exp_flow_spec_ipv4) {
- .type = IBV_EXP_FLOW_SPEC_IPV4,
+ .type = flow->inner | IBV_EXP_FLOW_SPEC_IPV4,
.size = ipv4_size,
};
if (!spec)
@@ -612,7 +630,7 @@ mlx5_flow_create_ipv6(const struct rte_flow_item *item,
flow->ibv_attr->priority = 1;
ipv6 = (void *)((uintptr_t)flow->ibv_attr + flow->offset);
*ipv6 = (struct ibv_exp_flow_spec_ipv6) {
- .type = IBV_EXP_FLOW_SPEC_IPV6,
+ .type = flow->inner | IBV_EXP_FLOW_SPEC_IPV6,
.size = ipv6_size,
};
if (!spec)
@@ -660,7 +678,7 @@ mlx5_flow_create_udp(const struct rte_flow_item *item,
flow->ibv_attr->priority = 0;
udp = (void *)((uintptr_t)flow->ibv_attr + flow->offset);
*udp = (struct ibv_exp_flow_spec_tcp_udp) {
- .type = IBV_EXP_FLOW_SPEC_UDP,
+ .type = flow->inner | IBV_EXP_FLOW_SPEC_UDP,
.size = udp_size,
};
if (!spec)
@@ -702,7 +720,7 @@ mlx5_flow_create_tcp(const struct rte_flow_item *item,
flow->ibv_attr->priority = 0;
tcp = (void *)((uintptr_t)flow->ibv_attr + flow->offset);
*tcp = (struct ibv_exp_flow_spec_tcp_udp) {
- .type = IBV_EXP_FLOW_SPEC_TCP,
+ .type = flow->inner | IBV_EXP_FLOW_SPEC_TCP,
.size = tcp_size,
};
if (!spec)
@@ -720,6 +738,53 @@ mlx5_flow_create_tcp(const struct rte_flow_item *item,
}
/**
+ * Convert VXLAN item to Verbs specification.
+ *
+ * @param item[in]
+ * Item specification.
+ * @param default_mask[in]
+ * Default bit-masks to use when
Re: [dpdk-dev] [PATCH 1/2] vhost: reference count fix for nb_started_ports
On 12/29/2016 03:51 AM, Yuanhan Liu wrote:
On Wed, Dec 28, 2016 at 04:10:51PM -0500, Charles (Chas) Williams wrote:
From: Wen Chiu
Only increment and decrement nb_started_ports on the first and last
device start and stop. Otherwise, nb_started_ports can become negative
if a device is stopped multiple times.
How could you be able to stop dev (precisely, invoke eth_dev_stop)
multiple times, judging that eth_dev_stop() will be invoked once
only?
void
rte_eth_dev_stop(uint8_t port_id)
{
struct rte_eth_dev *dev;
RTE_ETH_VALID_PORTID_OR_RET(port_id);
dev = &rte_eth_devices[port_id];
RTE_FUNC_PTR_OR_RET(*dev->dev_ops->dev_stop);
==> if (dev->data->dev_started == 0) {
RTE_PMD_DEBUG_TRACE("Device with port_id=%" PRIu8
" already stopped\n",
port_id);
return;
}
==> dev->data->dev_started = 0;
(*dev->dev_ops->dev_stop)(dev);
}
Multiple threads?
No, we aren't using multiple threads for control. But eth_dev_stop()
is called in rte_pmd_vhost_remove():
static int
rte_pmd_vhost_remove(const char *name)
{
...
pthread_mutex_lock(&internal_list_lock);
TAILQ_REMOVE(&internal_list, list, next);
pthread_mutex_unlock(&internal_list_lock);
rte_free(list);
eth_dev_stop(eth_dev);
...
So, if we .dev_stop() and deatch the virtual device, eth_dev_stop()
gets called twice. Calling .dev_stop() when you are about to detach
the device seems completely reasonable. It also seems reasonable to
call eth_dev_stop() inside rte_pmd_vhost_remove() in case the end
user didn't do a .dev_stop().
Re: [dpdk-dev] [PATCH 1/4] rte_mbuf: add rte_pktmbuf_coalesce
Hi Olivier,
> -Original Message-
> From: Olivier Matz [mailto:olivier.m...@6wind.com]
> Sent: Friday, December 16, 2016 11:06
> To: Kulasek, TomaszX
> Cc: dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 1/4] rte_mbuf: add rte_pktmbuf_coalesce
>
> Hi Tomasz,
>
> On Fri, 2 Dec 2016 18:07:43 +0100, Tomasz Kulasek
> wrote:
> > This patch adds function rte_pktmbuf_coalesce to let crypto PMD
> > coalesce chained mbuf before crypto operation and extend their
> > capabilities to support segmented mbufs when device cannot handle
> > them natively.
> >
> >
> > Signed-off-by: Tomasz Kulasek
> > ---
> > lib/librte_mbuf/rte_mbuf.h | 34 ++
> > 1 file changed, 34 insertions(+)
> >
> > diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h
> > index ead7c6e..f048681 100644
> > --- a/lib/librte_mbuf/rte_mbuf.h
> > +++ b/lib/librte_mbuf/rte_mbuf.h
> > @@ -1647,6 +1647,40 @@ static inline int rte_pktmbuf_chain(struct
> > rte_mbuf *head, struct rte_mbuf *tail }
> >
> > /**
> > + * Coalesce data from mbuf to the continuous buffer.
> > + *
> > + * @param mbuf_dst
> > + * Contiguous destination mbuf
> > + * @param mbuf_src
> > + * Uncontiguous source mbuf
> > + *
> > + * @return
> > + * - 0, on success
> > + * - -EINVAL, on error
> > + */
>
> I think the API should be clarified. In your case, it is expected that the
> destination mbuf is already filled with uninitialized data (i.e. that
> rte_pktmbuf_append() has been called).
>
> We could wonder if a better API wouldn't be to allocate the dst mbuf in
> the function, call append()/prepend(), and do the copy.
>
> Even better, we could have:
>
> int rte_pktmbuf_linearize(struct rte_mbuf *m)
>
> It will reuse the same mbuf (maybe moving the data).
>
>
> > +
> > +#include
>
> This should be removed.
>
> > +
> > +static inline int
> > +rte_pktmbuf_coalesce(struct rte_mbuf *mbuf_dst, struct rte_mbuf
> *mbuf_src) {
>
> Source mbuf should be const.
>
> > + char *dst;
> > +
> > + if (!rte_pktmbuf_is_contiguous(mbuf_dst) ||
> > + rte_pktmbuf_data_len(mbuf_dst) >=
> > + rte_pktmbuf_pkt_len(mbuf_src))
> > + return -EINVAL;
>
> Why >= ?
>
> > +
> > + dst = rte_pktmbuf_mtod(mbuf_dst, char *);
> > +
> > + if (!__rte_pktmbuf_read(mbuf_src, 0, rte_pktmbuf_pkt_len(mbuf_src),
> > + dst))
>
> When a function returns a pointer, I think it is clearer to do:
> if (func() == NULL)
> than:
> if (!func())
>
>
> > + return -EINVAL;
> > +
> > + return 0;
> > +}
> > +
> > +/**
> > * Dump an mbuf structure to a file.
> > *
> > * Dump all fields for the given packet mbuf and all its associated
>
>
> One more question, I don't see where this function is used in your
> patchset. What is your use-case?
>
> Regards,
> Olivier
This function is needed for crypto-perf application:
http://dpdk.org/dev/patchwork/patch/17492/ to compare performance of crypto
operations on segmented mbufs, when scatter gather is or is not supported by
crypto PMD. It will be introduced with v2.
When device doesn't support scatter-gather, we want to know an overhead of
manual coalescing mbuf.
struct rte_cryptodev_info dev_info;
int linearize = 0;
/* Check if source mbufs require coalescing */
if (ctx->options->segments_nb > 1) {
rte_cryptodev_info_get(ctx->dev_id, &dev_info);
if ((dev_info.feature_flags &
RTE_CRYPTODEV_FF_MBUF_SCATTER_GATHER) == 0)
linearize = 1;
}
// ...
if (linearize) {
/* PMD doesn't support scatter-gather and source buffer
* is segmented.
* We need to linearize it before enqueuing.
*/
for (i = 0; i < burst_size; i++)
rte_pktmbuf_linearize(ops[i]->sym->m_src);
}
/* Enqueue burst of ops on crypto device */
ops_enqd = rte_cryptodev_enqueue_burst(ctx->dev_id, ctx->qp_id,
ops, burst_size);
I have checked this use case (in crypto-perf application), and you're right,
using rte_pktmbuf_linearize() function here is better and more straightforward.
I will change it in v2.
Thanks for suggestions
Tomasz.
Re: [dpdk-dev] [PATCH 2/2] vhost: start vhost servers once
On 12/29/2016 03:52 AM, Yuanhan Liu wrote: On Wed, Dec 28, 2016 at 04:10:52PM -0500, Charles (Chas) Williams wrote: Start a vhost server once during devinit instead of during device start and stop. Some vhost clients, QEMU, don't re-attaching to sockets when the vhost server is stopped and later started. Preserve existing behavior for vhost clients. I didn't quite get the idea what you are going to fix. The issue I am trying to fix is QEMU interaction when DPDK's vhost is acting as a server to QEMU vhost clients. If you create a vhost server device, it doesn't create the actual datagram socket until you call .dev_start(). If you call .dev_stop() is also deletes those sockets. For QEMU, this is a problem since QEMU doesn't know how to re-attach to datagram sockets that have gone away. .dev_start()/.dev_stop() seems to roughly means link up and link down so I understand why you might want to add/remove the datagram sockets. However, in practice, this doesn't seem to make much sense for a DPDK vhost server. This doesn't seem like the right way to indicate link status to vhost clients. It seems like it would just be easier to do this for both clients and servers, but I don't know why it was done this way originally so I choose to keep the client behavior.
[dpdk-dev] [PATCH v3 00/17] net/i40e: consistent filter API
All pathes depend on Adrien's Generic flow API(rte_flow). The patches mainly finish following functions: 1) Store and restore all kinds of filters. 2) Parse all kinds of filters. 3) Add flow validate function. 4) Add flow create function. 5) Add flow destroy function. 6) Add flow flush function. v3 changes: Set the related cause pointer to a non-NULL value when error happens. Change return value when error happens. Modify filter_del parameter with key. Malloc filter after checking when delete a filter. Delete meaningless initialization. Add return value when there's error. Change global variable definition. Modify some function declaration. v2 changes: Add i40e_flow.c, all flow ops are implemented in the file. Change the whole implementation of all parse flow functions. Update error info for all flow ops. Add flow_list to store flows created. Beilei Xing (17): net/i40e: store ethertype filter net/i40e: store tunnel filter net/i40e: store flow director filter net/i40e: restore ethertype filter net/i40e: restore tunnel filter net/i40e: restore flow director filter net/i40e: add flow validate function net/i40e: parse flow director filter net/i40e: parse tunnel filter net/i40e: add flow create function net/i40e: add flow destroy function net/i40e: destroy ethertype filter net/i40e: destroy tunnel filter net/i40e: destroy flow directory filter net/i40e: add flow flush function net/i40e: flush ethertype filters net/i40e: flush tunnel filters drivers/net/i40e/Makefile |2 + drivers/net/i40e/i40e_ethdev.c | 519 ++-- drivers/net/i40e/i40e_ethdev.h | 173 drivers/net/i40e/i40e_fdir.c | 137 +++- drivers/net/i40e/i40e_flow.c | 1772 5 files changed, 2537 insertions(+), 66 deletions(-) create mode 100644 drivers/net/i40e/i40e_flow.c -- 2.5.5
[dpdk-dev] [PATCH v3 03/17] net/i40e: store flow director filter
Currently there's no flow director filter stored in SW. This
patch stores flow director filters in SW with cuckoo hash,
also adds protection if a flow director filter has been added.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_ethdev.c | 48 +++
drivers/net/i40e/i40e_ethdev.h | 14 ++
drivers/net/i40e/i40e_fdir.c | 105 +
3 files changed, 167 insertions(+)
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 2bdb4d6..fb7d794 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -953,6 +953,7 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
uint8_t aq_fail = 0;
struct i40e_ethertype_rule *ethertype_rule = &pf->ethertype;
struct i40e_tunnel_rule *tunnel_rule = &pf->tunnel;
+ struct i40e_fdir_info *fdir_info = &pf->fdir;
PMD_INIT_FUNC_TRACE();
@@ -972,6 +973,14 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
.hash_func = rte_hash_crc,
};
+ char fdir_hash_name[RTE_HASH_NAMESIZE];
+ struct rte_hash_parameters fdir_hash_params = {
+ .name = fdir_hash_name,
+ .entries = I40E_MAX_FDIR_FILTER_NUM,
+ .key_len = sizeof(struct rte_eth_fdir_input),
+ .hash_func = rte_hash_crc,
+ };
+
dev->dev_ops = &i40e_eth_dev_ops;
dev->rx_pkt_burst = i40e_recv_pkts;
dev->tx_pkt_burst = i40e_xmit_pkts;
@@ -1253,8 +1262,33 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
goto err_tunnel_hash_map_alloc;
}
+ /* Initialize flow director filter rule list and hash */
+ TAILQ_INIT(&fdir_info->fdir_list);
+ snprintf(fdir_hash_name, RTE_HASH_NAMESIZE,
+"fdir_%s", dev->data->name);
+ fdir_info->hash_table = rte_hash_create(&fdir_hash_params);
+ if (!fdir_info->hash_table) {
+ PMD_INIT_LOG(ERR, "Failed to create fdir hash table!");
+ ret = -EINVAL;
+ goto err_fdir_hash_table_create;
+ }
+ fdir_info->hash_map = rte_zmalloc("i40e_fdir_hash_map",
+ sizeof(struct i40e_fdir_filter *) *
+ I40E_MAX_FDIR_FILTER_NUM,
+ 0);
+ if (!fdir_info->hash_map) {
+ PMD_INIT_LOG(ERR,
+"Failed to allocate memory for fdir hash map!");
+ ret = -ENOMEM;
+ goto err_fdir_hash_map_alloc;
+ }
+
return 0;
+err_fdir_hash_map_alloc:
+ rte_hash_free(fdir_info->hash_table);
+err_fdir_hash_table_create:
+ rte_free(tunnel_rule->hash_map);
err_tunnel_hash_map_alloc:
rte_hash_free(tunnel_rule->hash_table);
err_tunnel_hash_table_create:
@@ -1291,10 +1325,12 @@ eth_i40e_dev_uninit(struct rte_eth_dev *dev)
struct i40e_filter_control_settings settings;
struct i40e_ethertype_filter *p_ethertype;
struct i40e_tunnel_filter *p_tunnel;
+ struct i40e_fdir_filter *p_fdir;
int ret;
uint8_t aq_fail = 0;
struct i40e_ethertype_rule *ethertype_rule;
struct i40e_tunnel_rule *tunnel_rule;
+ struct i40e_fdir_info *fdir_info;
PMD_INIT_FUNC_TRACE();
@@ -1306,6 +1342,7 @@ eth_i40e_dev_uninit(struct rte_eth_dev *dev)
pci_dev = dev->pci_dev;
ethertype_rule = &pf->ethertype;
tunnel_rule = &pf->tunnel;
+ fdir_info = &pf->fdir;
if (hw->adapter_stopped == 0)
i40e_dev_close(dev);
@@ -1333,6 +1370,17 @@ eth_i40e_dev_uninit(struct rte_eth_dev *dev)
rte_free(p_tunnel);
}
+ /* Remove all flow director rules and hash */
+ if (fdir_info->hash_map)
+ rte_free(fdir_info->hash_map);
+ if (fdir_info->hash_table)
+ rte_hash_free(fdir_info->hash_table);
+
+ while ((p_fdir = TAILQ_FIRST(&fdir_info->fdir_list))) {
+ TAILQ_REMOVE(&fdir_info->fdir_list, p_fdir, rules);
+ rte_free(p_fdir);
+ }
+
dev->dev_ops = NULL;
dev->rx_pkt_burst = NULL;
dev->tx_pkt_burst = NULL;
diff --git a/drivers/net/i40e/i40e_ethdev.h b/drivers/net/i40e/i40e_ethdev.h
index 83f3594..b79fbd6 100644
--- a/drivers/net/i40e/i40e_ethdev.h
+++ b/drivers/net/i40e/i40e_ethdev.h
@@ -377,6 +377,14 @@ struct i40e_fdir_flex_mask {
};
#define I40E_FILTER_PCTYPE_MAX 64
+#define I40E_MAX_FDIR_FILTER_NUM (1024 * 8)
+
+struct i40e_fdir_filter {
+ TAILQ_ENTRY(i40e_fdir_filter) rules;
+ struct rte_eth_fdir_filter fdir;
+};
+
+TAILQ_HEAD(i40e_fdir_filter_list, i40e_fdir_filter);
/*
* A structure used to define fields of a FDIR related info.
*/
@@ -395,6 +403,10 @@ struct i40e_fdir_info {
*/
struct i40e_fdir_flex_pit flex_set[I40E_MAX_FLXPLD_LAYER *
I40E_MAX_FLXPLD_FIED];
struct i40e_fdir_flex_mask flex_mask[I40E_FIL
[dpdk-dev] [PATCH v3 01/17] net/i40e: store ethertype filter
Currently there's no ethertype filter stored in SW.
This patch stores ethertype filter with cuckoo hash
in SW, also adds protection if an ethertype filter
has been added.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/Makefile | 1 +
drivers/net/i40e/i40e_ethdev.c | 166 -
drivers/net/i40e/i40e_ethdev.h | 31
3 files changed, 197 insertions(+), 1 deletion(-)
diff --git a/drivers/net/i40e/Makefile b/drivers/net/i40e/Makefile
index 66997b6..11175c4 100644
--- a/drivers/net/i40e/Makefile
+++ b/drivers/net/i40e/Makefile
@@ -117,5 +117,6 @@ DEPDIRS-$(CONFIG_RTE_LIBRTE_I40E_PMD) += lib/librte_eal
lib/librte_ether
DEPDIRS-$(CONFIG_RTE_LIBRTE_I40E_PMD) += lib/librte_mempool lib/librte_mbuf
DEPDIRS-$(CONFIG_RTE_LIBRTE_I40E_PMD) += lib/librte_net
DEPDIRS-$(CONFIG_RTE_LIBRTE_I40E_PMD) += lib/librte_kvargs
+DEPDIRS-$(CONFIG_RTE_LIBRTE_I40E_PMD) += lib/librte_hash
include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 8033c35..e43b4d9 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -51,6 +51,7 @@
#include
#include
#include
+#include
#include "i40e_logs.h"
#include "base/i40e_prototype.h"
@@ -461,6 +462,12 @@ static void i40e_set_default_mac_addr(struct rte_eth_dev
*dev,
static int i40e_dev_mtu_set(struct rte_eth_dev *dev, uint16_t mtu);
+static int i40e_ethertype_filter_convert(
+ const struct rte_eth_ethertype_filter *input,
+ struct i40e_ethertype_filter *filter);
+static int i40e_sw_ethertype_filter_insert(struct i40e_pf *pf,
+ struct i40e_ethertype_filter *filter);
+
static const struct rte_pci_id pci_id_i40e_map[] = {
{ RTE_PCI_DEVICE(I40E_INTEL_VENDOR_ID, I40E_DEV_ID_SFP_XL710) },
{ RTE_PCI_DEVICE(I40E_INTEL_VENDOR_ID, I40E_DEV_ID_QEMU) },
@@ -938,9 +945,18 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
int ret;
uint32_t len;
uint8_t aq_fail = 0;
+ struct i40e_ethertype_rule *ethertype_rule = &pf->ethertype;
PMD_INIT_FUNC_TRACE();
+ char ethertype_hash_name[RTE_HASH_NAMESIZE];
+ struct rte_hash_parameters ethertype_hash_params = {
+ .name = ethertype_hash_name,
+ .entries = I40E_MAX_ETHERTYPE_FILTER_NUM,
+ .key_len = sizeof(struct i40e_ethertype_filter_input),
+ .hash_func = rte_hash_crc,
+ };
+
dev->dev_ops = &i40e_eth_dev_ops;
dev->rx_pkt_burst = i40e_recv_pkts;
dev->tx_pkt_burst = i40e_xmit_pkts;
@@ -1180,8 +1196,33 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
pf->flags &= ~I40E_FLAG_DCB;
}
+ /* Initialize ethertype filter rule list and hash */
+ TAILQ_INIT(ðertype_rule->ethertype_list);
+ snprintf(ethertype_hash_name, RTE_HASH_NAMESIZE,
+"ethertype_%s", dev->data->name);
+ ethertype_rule->hash_table = rte_hash_create(ðertype_hash_params);
+ if (!ethertype_rule->hash_table) {
+ PMD_INIT_LOG(ERR, "Failed to create ethertype hash table!");
+ ret = -EINVAL;
+ goto err_ethertype_hash_table_create;
+ }
+ ethertype_rule->hash_map = rte_zmalloc("i40e_ethertype_hash_map",
+ sizeof(struct i40e_ethertype_filter *) *
+ I40E_MAX_ETHERTYPE_FILTER_NUM,
+ 0);
+ if (!ethertype_rule->hash_map) {
+ PMD_INIT_LOG(ERR,
+"Failed to allocate memory for ethertype hash map!");
+ ret = -ENOMEM;
+ goto err_ethertype_hash_map_alloc;
+ }
+
return 0;
+err_ethertype_hash_map_alloc:
+ rte_hash_free(ethertype_rule->hash_table);
+err_ethertype_hash_table_create:
+ rte_free(dev->data->mac_addrs);
err_mac_alloc:
i40e_vsi_release(pf->main_vsi);
err_setup_pf_switch:
@@ -1204,23 +1245,40 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
static int
eth_i40e_dev_uninit(struct rte_eth_dev *dev)
{
+ struct i40e_pf *pf;
struct rte_pci_device *pci_dev;
struct i40e_hw *hw;
struct i40e_filter_control_settings settings;
+ struct i40e_ethertype_filter *p_ethertype;
int ret;
uint8_t aq_fail = 0;
+ struct i40e_ethertype_rule *ethertype_rule;
PMD_INIT_FUNC_TRACE();
if (rte_eal_process_type() != RTE_PROC_PRIMARY)
return 0;
+ pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private);
hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private);
pci_dev = dev->pci_dev;
+ ethertype_rule = &pf->ethertype;
if (hw->adapter_stopped == 0)
i40e_dev_close(dev);
+ /* Remove all ethertype director rules and hash */
+ if (ethertype_rule->hash_map)
+ rte_free(ethertype_rule->hash_map);
+ if (ether
[dpdk-dev] [PATCH v3 02/17] net/i40e: store tunnel filter
Currently there's no tunnel filter stored in SW.
This patch stores tunnel filter in SW with cuckoo
hash, also adds protection if a tunnel filter has
been added.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_ethdev.c | 169 -
drivers/net/i40e/i40e_ethdev.h | 32
2 files changed, 198 insertions(+), 3 deletions(-)
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index e43b4d9..2bdb4d6 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -468,6 +468,12 @@ static int i40e_ethertype_filter_convert(
static int i40e_sw_ethertype_filter_insert(struct i40e_pf *pf,
struct i40e_ethertype_filter *filter);
+static int i40e_tunnel_filter_convert(
+ struct i40e_aqc_add_remove_cloud_filters_element_data *cld_filter,
+ struct i40e_tunnel_filter *tunnel_filter);
+static int i40e_sw_tunnel_filter_insert(struct i40e_pf *pf,
+ struct i40e_tunnel_filter *tunnel_filter);
+
static const struct rte_pci_id pci_id_i40e_map[] = {
{ RTE_PCI_DEVICE(I40E_INTEL_VENDOR_ID, I40E_DEV_ID_SFP_XL710) },
{ RTE_PCI_DEVICE(I40E_INTEL_VENDOR_ID, I40E_DEV_ID_QEMU) },
@@ -946,6 +952,7 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
uint32_t len;
uint8_t aq_fail = 0;
struct i40e_ethertype_rule *ethertype_rule = &pf->ethertype;
+ struct i40e_tunnel_rule *tunnel_rule = &pf->tunnel;
PMD_INIT_FUNC_TRACE();
@@ -957,6 +964,14 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
.hash_func = rte_hash_crc,
};
+ char tunnel_hash_name[RTE_HASH_NAMESIZE];
+ struct rte_hash_parameters tunnel_hash_params = {
+ .name = tunnel_hash_name,
+ .entries = I40E_MAX_TUNNEL_FILTER_NUM,
+ .key_len = sizeof(struct i40e_tunnel_filter_input),
+ .hash_func = rte_hash_crc,
+ };
+
dev->dev_ops = &i40e_eth_dev_ops;
dev->rx_pkt_burst = i40e_recv_pkts;
dev->tx_pkt_burst = i40e_xmit_pkts;
@@ -1217,8 +1232,33 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
goto err_ethertype_hash_map_alloc;
}
+ /* Initialize tunnel filter rule list and hash */
+ TAILQ_INIT(&tunnel_rule->tunnel_list);
+ snprintf(tunnel_hash_name, RTE_HASH_NAMESIZE,
+"tunnel_%s", dev->data->name);
+ tunnel_rule->hash_table = rte_hash_create(&tunnel_hash_params);
+ if (!tunnel_rule->hash_table) {
+ PMD_INIT_LOG(ERR, "Failed to create tunnel hash table!");
+ ret = -EINVAL;
+ goto err_tunnel_hash_table_create;
+ }
+ tunnel_rule->hash_map = rte_zmalloc("i40e_tunnel_hash_map",
+ sizeof(struct i40e_tunnel_filter *) *
+ I40E_MAX_TUNNEL_FILTER_NUM,
+ 0);
+ if (!tunnel_rule->hash_map) {
+ PMD_INIT_LOG(ERR,
+"Failed to allocate memory for tunnel hash map!");
+ ret = -ENOMEM;
+ goto err_tunnel_hash_map_alloc;
+ }
+
return 0;
+err_tunnel_hash_map_alloc:
+ rte_hash_free(tunnel_rule->hash_table);
+err_tunnel_hash_table_create:
+ rte_free(ethertype_rule->hash_map);
err_ethertype_hash_map_alloc:
rte_hash_free(ethertype_rule->hash_table);
err_ethertype_hash_table_create:
@@ -1250,9 +1290,11 @@ eth_i40e_dev_uninit(struct rte_eth_dev *dev)
struct i40e_hw *hw;
struct i40e_filter_control_settings settings;
struct i40e_ethertype_filter *p_ethertype;
+ struct i40e_tunnel_filter *p_tunnel;
int ret;
uint8_t aq_fail = 0;
struct i40e_ethertype_rule *ethertype_rule;
+ struct i40e_tunnel_rule *tunnel_rule;
PMD_INIT_FUNC_TRACE();
@@ -1263,6 +1305,7 @@ eth_i40e_dev_uninit(struct rte_eth_dev *dev)
hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private);
pci_dev = dev->pci_dev;
ethertype_rule = &pf->ethertype;
+ tunnel_rule = &pf->tunnel;
if (hw->adapter_stopped == 0)
i40e_dev_close(dev);
@@ -1279,6 +1322,17 @@ eth_i40e_dev_uninit(struct rte_eth_dev *dev)
rte_free(p_ethertype);
}
+ /* Remove all tunnel director rules and hash */
+ if (tunnel_rule->hash_map)
+ rte_free(tunnel_rule->hash_map);
+ if (tunnel_rule->hash_table)
+ rte_hash_free(tunnel_rule->hash_table);
+
+ while ((p_tunnel = TAILQ_FIRST(&tunnel_rule->tunnel_list))) {
+ TAILQ_REMOVE(&tunnel_rule->tunnel_list, p_tunnel, rules);
+ rte_free(p_tunnel);
+ }
+
dev->dev_ops = NULL;
dev->rx_pkt_burst = NULL;
dev->tx_pkt_burst = NULL;
@@ -6478,6 +6532,85 @@ i40e_dev_get_filter_type(uint16_t filter_type, uint16_t
*flag)
return 0;
}
+/*
[dpdk-dev] [PATCH v3 04/17] net/i40e: restore ethertype filter
Add support of restoring ethertype filter in case filter
dropped accidentally, as all filters need to be added and
removed by user obviously for generic filter API.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_ethdev.c | 44 ++
1 file changed, 44 insertions(+)
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index fb7d794..6cd8c06 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -474,6 +474,9 @@ static int i40e_tunnel_filter_convert(
static int i40e_sw_tunnel_filter_insert(struct i40e_pf *pf,
struct i40e_tunnel_filter *tunnel_filter);
+static void i40e_ethertype_filter_restore(struct i40e_pf *pf);
+static void i40e_filter_restore(struct i40e_pf *pf);
+
static const struct rte_pci_id pci_id_i40e_map[] = {
{ RTE_PCI_DEVICE(I40E_INTEL_VENDOR_ID, I40E_DEV_ID_SFP_XL710) },
{ RTE_PCI_DEVICE(I40E_INTEL_VENDOR_ID, I40E_DEV_ID_QEMU) },
@@ -1955,6 +1958,8 @@ i40e_dev_start(struct rte_eth_dev *dev)
/* enable uio intr after callback register */
rte_intr_enable(intr_handle);
+ i40e_filter_restore(pf);
+
return I40E_SUCCESS;
err_up:
@@ -10071,3 +10076,42 @@ i40e_dev_mtu_set(struct rte_eth_dev *dev, uint16_t mtu)
return ret;
}
+
+/* Restore ethertype filter */
+static void
+i40e_ethertype_filter_restore(struct i40e_pf *pf)
+{
+ struct i40e_hw *hw = I40E_PF_TO_HW(pf);
+ struct i40e_ethertype_filter_list
+ *ethertype_list = &pf->ethertype.ethertype_list;
+ struct i40e_ethertype_filter *f;
+ struct i40e_control_filter_stats stats;
+ uint16_t flags;
+
+ TAILQ_FOREACH(f, ethertype_list, rules) {
+ flags = 0;
+ if (!(f->flags & RTE_ETHTYPE_FLAGS_MAC))
+ flags |= I40E_AQC_ADD_CONTROL_PACKET_FLAGS_IGNORE_MAC;
+ if (f->flags & RTE_ETHTYPE_FLAGS_DROP)
+ flags |= I40E_AQC_ADD_CONTROL_PACKET_FLAGS_DROP;
+ flags |= I40E_AQC_ADD_CONTROL_PACKET_FLAGS_TO_QUEUE;
+
+ memset(&stats, 0, sizeof(stats));
+ i40e_aq_add_rem_control_packet_filter(hw,
+ f->input.mac_addr.addr_bytes,
+ f->input.ether_type,
+ flags, pf->main_vsi->seid,
+ f->queue, 1, &stats, NULL);
+ }
+ PMD_DRV_LOG(INFO, "add/rem control packet filter, return %d,"
+ " mac_etype_used = %u, etype_used = %u,"
+ " mac_etype_free = %u, etype_free = %u\n",
+ ret, stats.mac_etype_used, stats.etype_used,
+ stats.mac_etype_free, stats.etype_free);
+}
+
+static void
+i40e_filter_restore(struct i40e_pf *pf)
+{
+ i40e_ethertype_filter_restore(pf);
+}
--
2.5.5
[dpdk-dev] [PATCH v3 05/17] net/i40e: restore tunnel filter
Add support of restoring tunnel filter.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_ethdev.c | 21 +
1 file changed, 21 insertions(+)
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 6cd8c06..0d53c4e 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -475,6 +475,7 @@ static int i40e_sw_tunnel_filter_insert(struct i40e_pf *pf,
struct i40e_tunnel_filter *tunnel_filter);
static void i40e_ethertype_filter_restore(struct i40e_pf *pf);
+static void i40e_tunnel_filter_restore(struct i40e_pf *pf);
static void i40e_filter_restore(struct i40e_pf *pf);
static const struct rte_pci_id pci_id_i40e_map[] = {
@@ -10110,8 +10111,28 @@ i40e_ethertype_filter_restore(struct i40e_pf *pf)
stats.mac_etype_free, stats.etype_free);
}
+/* Restore tunnel filter */
+static void
+i40e_tunnel_filter_restore(struct i40e_pf *pf)
+{
+ struct i40e_hw *hw = I40E_PF_TO_HW(pf);
+ struct i40e_vsi *vsi = pf->main_vsi;
+ struct i40e_tunnel_filter_list
+ *tunnel_list = &pf->tunnel.tunnel_list;
+ struct i40e_tunnel_filter *f;
+ struct i40e_aqc_add_remove_cloud_filters_element_data cld_filter;
+
+ TAILQ_FOREACH(f, tunnel_list, rules) {
+ memset(&cld_filter, 0, sizeof(cld_filter));
+ rte_memcpy(&cld_filter, &f->input, sizeof(f->input));
+ cld_filter.queue_number = f->queue;
+ i40e_aq_add_cloud_filters(hw, vsi->seid, &cld_filter, 1);
+ }
+}
+
static void
i40e_filter_restore(struct i40e_pf *pf)
{
i40e_ethertype_filter_restore(pf);
+ i40e_tunnel_filter_restore(pf);
}
--
2.5.5
[dpdk-dev] [PATCH v3 06/17] net/i40e: restore flow director filter
Add support of restoring flow director filter.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_ethdev.c | 1 +
drivers/net/i40e/i40e_ethdev.h | 1 +
drivers/net/i40e/i40e_fdir.c | 28
3 files changed, 30 insertions(+)
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 0d53c4e..5bfa2e4 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -10135,4 +10135,5 @@ i40e_filter_restore(struct i40e_pf *pf)
{
i40e_ethertype_filter_restore(pf);
i40e_tunnel_filter_restore(pf);
+ i40e_fdir_filter_restore(pf);
}
diff --git a/drivers/net/i40e/i40e_ethdev.h b/drivers/net/i40e/i40e_ethdev.h
index b79fbd6..92f6f55 100644
--- a/drivers/net/i40e/i40e_ethdev.h
+++ b/drivers/net/i40e/i40e_ethdev.h
@@ -670,6 +670,7 @@ int i40e_fdir_ctrl_func(struct rte_eth_dev *dev,
int i40e_select_filter_input_set(struct i40e_hw *hw,
struct rte_eth_input_set_conf *conf,
enum rte_filter_type filter);
+void i40e_fdir_filter_restore(struct i40e_pf *pf);
int i40e_hash_filter_inset_select(struct i40e_hw *hw,
struct rte_eth_input_set_conf *conf);
int i40e_fdir_filter_inset_select(struct i40e_pf *pf,
diff --git a/drivers/net/i40e/i40e_fdir.c b/drivers/net/i40e/i40e_fdir.c
index 4a29b37..78ce92c 100644
--- a/drivers/net/i40e/i40e_fdir.c
+++ b/drivers/net/i40e/i40e_fdir.c
@@ -1586,3 +1586,31 @@ i40e_fdir_ctrl_func(struct rte_eth_dev *dev,
}
return ret;
}
+
+/* Restore flow director filter */
+void
+i40e_fdir_filter_restore(struct i40e_pf *pf)
+{
+ struct i40e_hw *hw = I40E_PF_TO_HW(pf);
+ struct rte_eth_dev *dev = I40E_VSI_TO_ETH_DEV(pf->main_vsi);
+ struct i40e_fdir_filter_list *fdir_list = &pf->fdir.fdir_list;
+ struct i40e_fdir_filter *f;
+ uint32_t fdstat;
+ uint32_t guarant_cnt; /**< Number of filters in guaranteed spaces. */
+ uint32_t best_cnt; /**< Number of filters in best effort spaces. */
+
+ TAILQ_FOREACH(f, fdir_list, rules)
+ i40e_add_del_fdir_filter(dev, &f->fdir, TRUE);
+
+ fdstat = I40E_READ_REG(hw, I40E_PFQF_FDSTAT);
+ guarant_cnt =
+ (uint32_t)((fdstat & I40E_PFQF_FDSTAT_GUARANT_CNT_MASK) >>
+ I40E_PFQF_FDSTAT_GUARANT_CNT_SHIFT);
+ best_cnt =
+ (uint32_t)((fdstat & I40E_PFQF_FDSTAT_BEST_CNT_MASK) >>
+ I40E_PFQF_FDSTAT_BEST_CNT_SHIFT);
+
+ printf("FDIR restore:"
+ "Guarant count: %d, Best count: %d\n",
+ guarant_cnt, best_cnt);
+}
--
2.5.5
[dpdk-dev] [PATCH v3 07/17] net/i40e: add flow validate function
This patch adds i40e_flow_validation function to check if
a flow is valid according to the flow pattern.
i40e_parse_ethertype_filter is added first, it also gets
the ethertype info.
i40e_flow.c is added to handle all generic filter events.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/Makefile | 1 +
drivers/net/i40e/i40e_ethdev.c | 7 +
drivers/net/i40e/i40e_ethdev.h | 18 ++
drivers/net/i40e/i40e_flow.c | 447 +
4 files changed, 473 insertions(+)
create mode 100644 drivers/net/i40e/i40e_flow.c
diff --git a/drivers/net/i40e/Makefile b/drivers/net/i40e/Makefile
index 11175c4..89bd85a 100644
--- a/drivers/net/i40e/Makefile
+++ b/drivers/net/i40e/Makefile
@@ -105,6 +105,7 @@ endif
SRCS-$(CONFIG_RTE_LIBRTE_I40E_PMD) += i40e_ethdev_vf.c
SRCS-$(CONFIG_RTE_LIBRTE_I40E_PMD) += i40e_pf.c
SRCS-$(CONFIG_RTE_LIBRTE_I40E_PMD) += i40e_fdir.c
+SRCS-$(CONFIG_RTE_LIBRTE_I40E_PMD) += i40e_flow.c
# vector PMD driver needs SSE4.1 support
ifeq ($(findstring RTE_MACHINE_CPUFLAG_SSE4_1,$(CFLAGS)),)
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 5bfa2e4..710631f 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -8426,6 +8426,8 @@ i40e_ethertype_filter_handle(struct rte_eth_dev *dev,
return ret;
}
+const struct rte_flow_ops i40e_flow_ops;
+
static int
i40e_dev_filter_ctrl(struct rte_eth_dev *dev,
enum rte_filter_type filter_type,
@@ -8457,6 +8459,11 @@ i40e_dev_filter_ctrl(struct rte_eth_dev *dev,
case RTE_ETH_FILTER_FDIR:
ret = i40e_fdir_ctrl_func(dev, filter_op, arg);
break;
+ case RTE_ETH_FILTER_GENERIC:
+ if (filter_op != RTE_ETH_FILTER_GET)
+ return -EINVAL;
+ *(const void **)arg = &i40e_flow_ops;
+ break;
default:
PMD_DRV_LOG(WARNING, "Filter type (%d) not supported",
filter_type);
diff --git a/drivers/net/i40e/i40e_ethdev.h b/drivers/net/i40e/i40e_ethdev.h
index 92f6f55..23f360b 100644
--- a/drivers/net/i40e/i40e_ethdev.h
+++ b/drivers/net/i40e/i40e_ethdev.h
@@ -38,6 +38,7 @@
#include
#include
#include
+#include
#define I40E_VLAN_TAG_SIZE4
@@ -629,6 +630,23 @@ struct i40e_adapter {
struct rte_timecounter tx_tstamp_tc;
};
+union i40e_filter_t {
+ struct rte_eth_ethertype_filter ethertype_filter;
+ struct rte_eth_fdir_filter fdir_filter;
+ struct rte_eth_tunnel_filter_conf tunnel_filter;
+};
+
+typedef int (*parse_filter_t)(struct rte_eth_dev *dev,
+ const struct rte_flow_attr *attr,
+ const struct rte_flow_item pattern[],
+ const struct rte_flow_action actions[],
+ struct rte_flow_error *error,
+ union i40e_filter_t *filter);
+struct i40e_valid_pattern {
+ enum rte_flow_item_type *items;
+ parse_filter_t parse_filter;
+};
+
int i40e_dev_switch_queues(struct i40e_pf *pf, bool on);
int i40e_vsi_release(struct i40e_vsi *vsi);
struct i40e_vsi *i40e_vsi_setup(struct i40e_pf *pf,
diff --git a/drivers/net/i40e/i40e_flow.c b/drivers/net/i40e/i40e_flow.c
new file mode 100644
index 000..a9ff73f
--- /dev/null
+++ b/drivers/net/i40e/i40e_flow.c
@@ -0,0 +1,447 @@
+/*-
+ * BSD LICENSE
+ *
+ * Copyright (c) 2016 Intel Corporation. All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * * Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (I
[dpdk-dev] [PATCH v3 10/17] net/i40e: add flow create function
This patch adds i40e_flow_create function to create a
rule. It will check if a flow matches ethertype filter
or flow director filter or tunnel filter, if the flow
matches some kind of filter, then set the filter to HW.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_ethdev.c | 9 +++--
drivers/net/i40e/i40e_ethdev.h | 21
drivers/net/i40e/i40e_fdir.c | 2 +-
drivers/net/i40e/i40e_flow.c | 77 ++
4 files changed, 103 insertions(+), 6 deletions(-)
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 4a057b4..fa59ee4 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -353,9 +353,6 @@ static int i40e_dev_udp_tunnel_port_add(struct rte_eth_dev
*dev,
static int i40e_dev_udp_tunnel_port_del(struct rte_eth_dev *dev,
struct rte_eth_udp_tunnel *udp_tunnel);
static void i40e_filter_input_set_init(struct i40e_pf *pf);
-static int i40e_ethertype_filter_set(struct i40e_pf *pf,
- struct rte_eth_ethertype_filter *filter,
- bool add);
static int i40e_ethertype_filter_handle(struct rte_eth_dev *dev,
enum rte_filter_op filter_op,
void *arg);
@@ -1233,6 +1230,8 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
goto err_fdir_hash_map_alloc;
}
+ TAILQ_INIT(&pf->flow_list);
+
return 0;
err_fdir_hash_map_alloc:
@@ -6611,7 +6610,7 @@ i40e_sw_tunnel_filter_del(struct i40e_pf *pf,
return 0;
}
-static int
+int
i40e_dev_tunnel_filter_set(struct i40e_pf *pf,
struct rte_eth_tunnel_filter_conf *tunnel_filter,
uint8_t add)
@@ -8256,7 +8255,7 @@ i40e_sw_ethertype_filter_del(struct i40e_pf *pf,
* Configure ethertype filter, which can director packet by filtering
* with mac address and ether_type or only ether_type
*/
-static int
+int
i40e_ethertype_filter_set(struct i40e_pf *pf,
struct rte_eth_ethertype_filter *filter,
bool add)
diff --git a/drivers/net/i40e/i40e_ethdev.h b/drivers/net/i40e/i40e_ethdev.h
index 9e3a48d..b33910d 100644
--- a/drivers/net/i40e/i40e_ethdev.h
+++ b/drivers/net/i40e/i40e_ethdev.h
@@ -536,6 +536,17 @@ struct i40e_mirror_rule {
TAILQ_HEAD(i40e_mirror_rule_list, i40e_mirror_rule);
/*
+ * Struct to store flow created.
+ */
+struct i40e_flow {
+ TAILQ_ENTRY(i40e_flow) node;
+ enum rte_filter_type filter_type;
+ void *rule;
+};
+
+TAILQ_HEAD(i40e_flow_list, i40e_flow);
+
+/*
* Structure to store private data specific for PF instance.
*/
struct i40e_pf {
@@ -592,6 +603,7 @@ struct i40e_pf {
bool floating_veb; /* The flag to use the floating VEB */
/* The floating enable flag for the specific VF */
bool floating_veb_list[I40E_MAX_VF];
+ struct i40e_flow_list flow_list;
};
enum pending_msg {
@@ -767,6 +779,15 @@ i40e_sw_tunnel_filter_lookup(struct i40e_tunnel_rule
*tunnel_rule,
int i40e_sw_tunnel_filter_del(struct i40e_pf *pf,
struct i40e_tunnel_filter_input *input);
uint64_t i40e_get_default_input_set(uint16_t pctype);
+int i40e_ethertype_filter_set(struct i40e_pf *pf,
+ struct rte_eth_ethertype_filter *filter,
+ bool add);
+int i40e_add_del_fdir_filter(struct rte_eth_dev *dev,
+const struct rte_eth_fdir_filter *filter,
+bool add);
+int i40e_dev_tunnel_filter_set(struct i40e_pf *pf,
+ struct rte_eth_tunnel_filter_conf *tunnel_filter,
+ uint8_t add);
/* I40E_DEV_PRIVATE_TO */
#define I40E_DEV_PRIVATE_TO_PF(adapter) \
diff --git a/drivers/net/i40e/i40e_fdir.c b/drivers/net/i40e/i40e_fdir.c
index 78ce92c..e1f97bb 100644
--- a/drivers/net/i40e/i40e_fdir.c
+++ b/drivers/net/i40e/i40e_fdir.c
@@ -1099,7 +1099,7 @@ i40e_sw_fdir_filter_del(struct i40e_pf *pf, struct
rte_eth_fdir_input *input)
* @filter: fdir filter entry
* @add: 0 - delete, 1 - add
*/
-static int
+int
i40e_add_del_fdir_filter(struct rte_eth_dev *dev,
const struct rte_eth_fdir_filter *filter,
bool add)
diff --git a/drivers/net/i40e/i40e_flow.c b/drivers/net/i40e/i40e_flow.c
index 53bfb2b..8db8e0f 100644
--- a/drivers/net/i40e/i40e_flow.c
+++ b/drivers/net/i40e/i40e_flow.c
@@ -62,6 +62,11 @@ static int i40e_flow_validate(struct rte_eth_dev *dev,
const struct rte_flow_item pattern[],
const struct rte_flow_action actions[],
struct rte_flow_error *error);
+static struct rte_flow *i40e_flow_create(struct rte_eth_dev *dev,
+const struct rte_flow_attr *attr,
+
[dpdk-dev] [PATCH v3 09/17] net/i40e: parse tunnel filter
This patch adds i40e_parse_tunnel_filter to check if
a rule is a tunnel rule according to items of the flow
pattern, and the function also gets the tunnel info.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_flow.c | 394 +++
1 file changed, 394 insertions(+)
diff --git a/drivers/net/i40e/i40e_flow.c b/drivers/net/i40e/i40e_flow.c
index 7b872c9..53bfb2b 100644
--- a/drivers/net/i40e/i40e_flow.c
+++ b/drivers/net/i40e/i40e_flow.c
@@ -54,6 +54,8 @@
#define I40E_IPV4_TC_SHIFT 4
#define I40E_IPV6_TC_MASK (0x00FF << I40E_IPV4_TC_SHIFT)
#define I40E_IPV6_FRAG_HEADER 44
+#define I40E_TENANT_ARRAY_NUM 3
+#define I40E_TCI_MASK 0x0FFF
static int i40e_flow_validate(struct rte_eth_dev *dev,
const struct rte_flow_attr *attr,
@@ -76,6 +78,14 @@ static int i40e_parse_fdir_act(struct rte_eth_dev *dev,
const struct rte_flow_action *actions,
struct rte_flow_error *error,
struct rte_eth_fdir_filter *filter);
+static int i40e_parse_tunnel_pattern(__rte_unused struct rte_eth_dev *dev,
+const struct rte_flow_item *pattern,
+struct rte_flow_error *error,
+struct rte_eth_tunnel_filter_conf *filter);
+static int i40e_parse_tunnel_act(struct rte_eth_dev *dev,
+const struct rte_flow_action *actions,
+struct rte_flow_error *error,
+struct rte_eth_tunnel_filter_conf *filter);
static int i40e_parse_attr(const struct rte_flow_attr *attr,
struct rte_flow_error *error);
@@ -192,6 +202,45 @@ static enum rte_flow_item_type
pattern_fdir_ipv6_sctp_ext[] = {
RTE_FLOW_ITEM_TYPE_END,
};
+/* Pattern matched tunnel filter */
+static enum rte_flow_item_type pattern_vxlan_1[] = {
+ RTE_FLOW_ITEM_TYPE_ETH,
+ RTE_FLOW_ITEM_TYPE_IPV4,
+ RTE_FLOW_ITEM_TYPE_UDP,
+ RTE_FLOW_ITEM_TYPE_VXLAN,
+ RTE_FLOW_ITEM_TYPE_ETH,
+ RTE_FLOW_ITEM_TYPE_END,
+};
+
+static enum rte_flow_item_type pattern_vxlan_2[] = {
+ RTE_FLOW_ITEM_TYPE_ETH,
+ RTE_FLOW_ITEM_TYPE_IPV6,
+ RTE_FLOW_ITEM_TYPE_UDP,
+ RTE_FLOW_ITEM_TYPE_VXLAN,
+ RTE_FLOW_ITEM_TYPE_ETH,
+ RTE_FLOW_ITEM_TYPE_END,
+};
+
+static enum rte_flow_item_type pattern_vxlan_3[] = {
+ RTE_FLOW_ITEM_TYPE_ETH,
+ RTE_FLOW_ITEM_TYPE_IPV4,
+ RTE_FLOW_ITEM_TYPE_UDP,
+ RTE_FLOW_ITEM_TYPE_VXLAN,
+ RTE_FLOW_ITEM_TYPE_ETH,
+ RTE_FLOW_ITEM_TYPE_VLAN,
+ RTE_FLOW_ITEM_TYPE_END,
+};
+
+static enum rte_flow_item_type pattern_vxlan_4[] = {
+ RTE_FLOW_ITEM_TYPE_ETH,
+ RTE_FLOW_ITEM_TYPE_IPV6,
+ RTE_FLOW_ITEM_TYPE_UDP,
+ RTE_FLOW_ITEM_TYPE_VXLAN,
+ RTE_FLOW_ITEM_TYPE_ETH,
+ RTE_FLOW_ITEM_TYPE_VLAN,
+ RTE_FLOW_ITEM_TYPE_END,
+};
+
static int
i40e_parse_ethertype_filter(struct rte_eth_dev *dev,
const struct rte_flow_attr *attr,
@@ -257,6 +306,33 @@ i40e_parse_fdir_filter(struct rte_eth_dev *dev,
return 0;
}
+static int
+i40e_parse_tunnel_filter(struct rte_eth_dev *dev,
+const struct rte_flow_attr *attr,
+const struct rte_flow_item pattern[],
+const struct rte_flow_action actions[],
+struct rte_flow_error *error,
+union i40e_filter_t *filter)
+{
+ struct rte_eth_tunnel_filter_conf *tunnel_filter =
+ &filter->tunnel_filter;
+ int ret;
+
+ ret = i40e_parse_tunnel_pattern(dev, pattern, error, tunnel_filter);
+ if (ret)
+ return ret;
+
+ ret = i40e_parse_tunnel_act(dev, actions, error, tunnel_filter);
+ if (ret)
+ return ret;
+
+ ret = i40e_parse_attr(attr, error);
+ if (ret)
+ return ret;
+
+ return ret;
+}
+
static struct i40e_valid_pattern i40e_supported_patterns[] = {
/* Ethertype */
{ pattern_ethertype, i40e_parse_ethertype_filter },
@@ -277,6 +353,11 @@ static struct i40e_valid_pattern i40e_supported_patterns[]
= {
{ pattern_fdir_ipv6_tcp_ext, i40e_parse_fdir_filter },
{ pattern_fdir_ipv6_sctp, i40e_parse_fdir_filter },
{ pattern_fdir_ipv6_sctp_ext, i40e_parse_fdir_filter },
+ /* tunnel */
+ { pattern_vxlan_1, i40e_parse_tunnel_filter },
+ { pattern_vxlan_2, i40e_parse_tunnel_filter },
+ { pattern_vxlan_3, i40e_parse_tunnel_filter },
+ { pattern_vxlan_4, i40e_parse_tunnel_filter },
};
#define NEXT_ITEM_OF_ACTION(act, actions, index)\
@@ -991,6 +1072,319 @@ i40e_parse_fdir_act(struct rte_eth_dev *dev,
return 0;
}
+/* Parse to get the action info of a tunnle filter */
+static int
[dpdk-dev] [PATCH v3 13/17] net/i40e: destroy tunnel filter
This patch adds i40e_dev_destroy_tunnel_filter function
to destroy a tunnel filter for users.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_flow.c | 41 +
1 file changed, 41 insertions(+)
diff --git a/drivers/net/i40e/i40e_flow.c b/drivers/net/i40e/i40e_flow.c
index a41d4c2..e442c5c 100644
--- a/drivers/net/i40e/i40e_flow.c
+++ b/drivers/net/i40e/i40e_flow.c
@@ -99,6 +99,8 @@ static int i40e_parse_attr(const struct rte_flow_attr *attr,
struct rte_flow_error *error);
static int i40e_dev_destroy_ethertype_filter(struct i40e_pf *pf,
struct i40e_ethertype_filter *filter);
+static int i40e_dev_destroy_tunnel_filter(struct i40e_pf *pf,
+ struct i40e_tunnel_filter *filter);
const struct rte_flow_ops i40e_flow_ops = {
.validate = i40e_flow_validate,
@@ -1546,6 +1548,10 @@ i40e_flow_destroy(struct rte_eth_dev *dev,
ret = i40e_dev_destroy_ethertype_filter(pf,
(struct i40e_ethertype_filter *)pmd_flow->rule);
break;
+ case RTE_ETH_FILTER_TUNNEL:
+ ret = i40e_dev_destroy_tunnel_filter(pf,
+(struct i40e_tunnel_filter *)pmd_flow->rule);
+ break;
default:
PMD_DRV_LOG(WARNING, "Filter type (%d) not supported",
filter_type);
@@ -1598,3 +1604,38 @@ i40e_dev_destroy_ethertype_filter(struct i40e_pf *pf,
return ret;
}
+
+static int
+i40e_dev_destroy_tunnel_filter(struct i40e_pf *pf,
+ struct i40e_tunnel_filter *filter)
+{
+ struct i40e_hw *hw = I40E_PF_TO_HW(pf);
+ struct i40e_vsi *vsi = pf->main_vsi;
+ struct i40e_aqc_add_remove_cloud_filters_element_data cld_filter;
+ struct i40e_tunnel_rule *tunnel_rule = &pf->tunnel;
+ struct i40e_tunnel_filter *node;
+ int ret = 0;
+
+ memset(&cld_filter, 0, sizeof(cld_filter));
+ ether_addr_copy((struct ether_addr *)&filter->input.outer_mac,
+ (struct ether_addr *)&cld_filter.outer_mac);
+ ether_addr_copy((struct ether_addr *)&filter->input.inner_mac,
+ (struct ether_addr *)&cld_filter.inner_mac);
+ cld_filter.inner_vlan = filter->input.inner_vlan;
+ cld_filter.flags = filter->input.flags;
+ cld_filter.tenant_id = filter->input.tenant_id;
+ cld_filter.queue_number = filter->queue;
+
+ ret = i40e_aq_remove_cloud_filters(hw, vsi->seid,
+ &cld_filter, 1);
+ if (ret < 0)
+ return ret;
+
+ node = i40e_sw_tunnel_filter_lookup(tunnel_rule, &filter->input);
+ if (!node)
+ return -EINVAL;
+
+ ret = i40e_sw_tunnel_filter_del(pf, &node->input);
+
+ return ret;
+}
--
2.5.5
[dpdk-dev] [PATCH v3 08/17] net/i40e: parse flow director filter
This patch adds i40e_parse_fdir_filter to check if a rule
is a flow director rule according to the flow pattern,
and the function also gets the flow director info.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_ethdev.c | 56 +---
drivers/net/i40e/i40e_ethdev.h | 55
drivers/net/i40e/i40e_flow.c | 607 +
3 files changed, 663 insertions(+), 55 deletions(-)
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 710631f..4a057b4 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -139,60 +139,6 @@
#define I40E_DEFAULT_DCB_APP_NUM1
#define I40E_DEFAULT_DCB_APP_PRIO 3
-#define I40E_INSET_NONE0x0ULL
-
-/* bit0 ~ bit 7 */
-#define I40E_INSET_DMAC0x0001ULL
-#define I40E_INSET_SMAC0x0002ULL
-#define I40E_INSET_VLAN_OUTER 0x0004ULL
-#define I40E_INSET_VLAN_INNER 0x0008ULL
-#define I40E_INSET_VLAN_TUNNEL 0x0010ULL
-
-/* bit 8 ~ bit 15 */
-#define I40E_INSET_IPV4_SRC0x0100ULL
-#define I40E_INSET_IPV4_DST0x0200ULL
-#define I40E_INSET_IPV6_SRC0x0400ULL
-#define I40E_INSET_IPV6_DST0x0800ULL
-#define I40E_INSET_SRC_PORT0x1000ULL
-#define I40E_INSET_DST_PORT0x2000ULL
-#define I40E_INSET_SCTP_VT 0x4000ULL
-
-/* bit 16 ~ bit 31 */
-#define I40E_INSET_IPV4_TOS0x0001ULL
-#define I40E_INSET_IPV4_PROTO 0x0002ULL
-#define I40E_INSET_IPV4_TTL0x0004ULL
-#define I40E_INSET_IPV6_TC 0x0008ULL
-#define I40E_INSET_IPV6_FLOW 0x0010ULL
-#define I40E_INSET_IPV6_NEXT_HDR 0x0020ULL
-#define I40E_INSET_IPV6_HOP_LIMIT 0x0040ULL
-#define I40E_INSET_TCP_FLAGS 0x0080ULL
-
-/* bit 32 ~ bit 47, tunnel fields */
-#define I40E_INSET_TUNNEL_IPV4_DST 0x0001ULL
-#define I40E_INSET_TUNNEL_IPV6_DST 0x0002ULL
-#define I40E_INSET_TUNNEL_DMAC 0x0004ULL
-#define I40E_INSET_TUNNEL_SRC_PORT 0x0008ULL
-#define I40E_INSET_TUNNEL_DST_PORT 0x0010ULL
-#define I40E_INSET_TUNNEL_ID 0x0020ULL
-
-/* bit 48 ~ bit 55 */
-#define I40E_INSET_LAST_ETHER_TYPE 0x0001ULL
-
-/* bit 56 ~ bit 63, Flex Payload */
-#define I40E_INSET_FLEX_PAYLOAD_W1 0x0100ULL
-#define I40E_INSET_FLEX_PAYLOAD_W2 0x0200ULL
-#define I40E_INSET_FLEX_PAYLOAD_W3 0x0400ULL
-#define I40E_INSET_FLEX_PAYLOAD_W4 0x0800ULL
-#define I40E_INSET_FLEX_PAYLOAD_W5 0x1000ULL
-#define I40E_INSET_FLEX_PAYLOAD_W6 0x2000ULL
-#define I40E_INSET_FLEX_PAYLOAD_W7 0x4000ULL
-#define I40E_INSET_FLEX_PAYLOAD_W8 0x8000ULL
-#define I40E_INSET_FLEX_PAYLOAD \
- (I40E_INSET_FLEX_PAYLOAD_W1 | I40E_INSET_FLEX_PAYLOAD_W2 | \
- I40E_INSET_FLEX_PAYLOAD_W3 | I40E_INSET_FLEX_PAYLOAD_W4 | \
- I40E_INSET_FLEX_PAYLOAD_W5 | I40E_INSET_FLEX_PAYLOAD_W6 | \
- I40E_INSET_FLEX_PAYLOAD_W7 | I40E_INSET_FLEX_PAYLOAD_W8)
-
/**
* Below are values for writing un-exposed registers suggested
* by silicon experts
@@ -7617,7 +7563,7 @@ i40e_validate_input_set(enum i40e_filter_pctype pctype,
}
/* default input set fields combination per pctype */
-static uint64_t
+uint64_t
i40e_get_default_input_set(uint16_t pctype)
{
static const uint64_t default_inset_table[] = {
diff --git a/drivers/net/i40e/i40e_ethdev.h b/drivers/net/i40e/i40e_ethdev.h
index 23f360b..9e3a48d 100644
--- a/drivers/net/i40e/i40e_ethdev.h
+++ b/drivers/net/i40e/i40e_ethdev.h
@@ -190,6 +190,60 @@ enum i40e_flxpld_layer_idx {
#define FLOATING_VEB_SUPPORTED_FW_MAJ 5
#define FLOATING_VEB_SUPPORTED_FW_MIN 0
+#define I40E_INSET_NONE0x0ULL
+
+/* bit0 ~ bit 7 */
+#define I40E_INSET_DMAC0x0001ULL
+#define I40E_INSET_SMAC0x0002ULL
+#define I40E_INSET_VLAN_OUTER 0x0004ULL
+#define I40E_INSET_VLAN_INNER 0x0008ULL
+#define I40E_INSET_VLAN_TUNNEL 0x0010ULL
+
+/* bit 8 ~ bit 15 */
+#define I40E_INSET_IPV4_SRC0x0100ULL
+#define I40E_INSET_IPV4_DST0x0200ULL
+#define I40E_INSET_IPV6_SRC0x0400ULL
+#define I40E_INSET_IPV6_DST0x0800ULL
+#define I40E_INSET_SRC_PORT0x1000ULL
+#define I40E_INSET_DST_PORT0x2000ULL
+#define I40E_INSET_SCTP_VT 0x4000ULL
+
+/* bit 16 ~ bit 31 */
+#define I40E_INSET_IPV4_TOS0x0001ULL
+#define I40E_INSET_IPV4_PROTO 0x0002ULL
+#define I40E_INSET_IPV4_TTL0x0004ULL
+#define I40E_INS
[dpdk-dev] [PATCH v3 14/17] net/i40e: destroy flow directory filter
This patch supports destroying a flow directory filter for users. Signed-off-by: Beilei Xing --- drivers/net/i40e/i40e_flow.c | 4 1 file changed, 4 insertions(+) diff --git a/drivers/net/i40e/i40e_flow.c b/drivers/net/i40e/i40e_flow.c index e442c5c..605ee70 100644 --- a/drivers/net/i40e/i40e_flow.c +++ b/drivers/net/i40e/i40e_flow.c @@ -1552,6 +1552,10 @@ i40e_flow_destroy(struct rte_eth_dev *dev, ret = i40e_dev_destroy_tunnel_filter(pf, (struct i40e_tunnel_filter *)pmd_flow->rule); break; + case RTE_ETH_FILTER_FDIR: + ret = i40e_add_del_fdir_filter(dev, + &((struct i40e_fdir_filter *)pmd_flow->rule)->fdir, 0); + break; default: PMD_DRV_LOG(WARNING, "Filter type (%d) not supported", filter_type); -- 2.5.5
[dpdk-dev] [PATCH v3 12/17] net/i40e: destroy ethertype filter
This patch adds i40e_dev_destroy_ethertype_filter function
to destroy a ethertype filter for users.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_flow.c | 42 ++
1 file changed, 42 insertions(+)
diff --git a/drivers/net/i40e/i40e_flow.c b/drivers/net/i40e/i40e_flow.c
index a64ef0e..a41d4c2 100644
--- a/drivers/net/i40e/i40e_flow.c
+++ b/drivers/net/i40e/i40e_flow.c
@@ -49,6 +49,7 @@
#include "i40e_logs.h"
#include "base/i40e_type.h"
+#include "base/i40e_prototype.h"
#include "i40e_ethdev.h"
#define I40E_IPV4_TC_SHIFT 4
@@ -96,6 +97,8 @@ static int i40e_parse_tunnel_act(struct rte_eth_dev *dev,
struct rte_eth_tunnel_filter_conf *filter);
static int i40e_parse_attr(const struct rte_flow_attr *attr,
struct rte_flow_error *error);
+static int i40e_dev_destroy_ethertype_filter(struct i40e_pf *pf,
+struct i40e_ethertype_filter *filter);
const struct rte_flow_ops i40e_flow_ops = {
.validate = i40e_flow_validate,
@@ -1539,6 +1542,10 @@ i40e_flow_destroy(struct rte_eth_dev *dev,
int ret = 0;
switch (filter_type) {
+ case RTE_ETH_FILTER_ETHERTYPE:
+ ret = i40e_dev_destroy_ethertype_filter(pf,
+ (struct i40e_ethertype_filter *)pmd_flow->rule);
+ break;
default:
PMD_DRV_LOG(WARNING, "Filter type (%d) not supported",
filter_type);
@@ -1556,3 +1563,38 @@ i40e_flow_destroy(struct rte_eth_dev *dev,
return ret;
}
+
+static int
+i40e_dev_destroy_ethertype_filter(struct i40e_pf *pf,
+ struct i40e_ethertype_filter *filter)
+{
+ struct i40e_hw *hw = I40E_PF_TO_HW(pf);
+ struct i40e_ethertype_rule *ethertype_rule = &pf->ethertype;
+ struct i40e_ethertype_filter *node;
+ struct i40e_control_filter_stats stats;
+ uint16_t flags = 0;
+ int ret = 0;
+
+ if (!(filter->flags & RTE_ETHTYPE_FLAGS_MAC))
+ flags |= I40E_AQC_ADD_CONTROL_PACKET_FLAGS_IGNORE_MAC;
+ if (filter->flags & RTE_ETHTYPE_FLAGS_DROP)
+ flags |= I40E_AQC_ADD_CONTROL_PACKET_FLAGS_DROP;
+ flags |= I40E_AQC_ADD_CONTROL_PACKET_FLAGS_TO_QUEUE;
+
+ memset(&stats, 0, sizeof(stats));
+ ret = i40e_aq_add_rem_control_packet_filter(hw,
+ filter->input.mac_addr.addr_bytes,
+ filter->input.ether_type,
+ flags, pf->main_vsi->seid,
+ filter->queue, 0, &stats, NULL);
+ if (ret < 0)
+ return ret;
+
+ node = i40e_sw_ethertype_filter_lookup(ethertype_rule, &filter->input);
+ if (!node)
+ return -EINVAL;
+
+ ret = i40e_sw_ethertype_filter_del(pf, &node->input);
+
+ return ret;
+}
--
2.5.5
[dpdk-dev] [PATCH v3 16/17] net/i40e: flush ethertype filters
This patch adds i40e_ethertype_filter_flush function
to flush all ethertype filters, including filters in
SW and HW.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_flow.c | 41 -
1 file changed, 40 insertions(+), 1 deletion(-)
diff --git a/drivers/net/i40e/i40e_flow.c b/drivers/net/i40e/i40e_flow.c
index ee463c9..947199d 100644
--- a/drivers/net/i40e/i40e_flow.c
+++ b/drivers/net/i40e/i40e_flow.c
@@ -104,6 +104,7 @@ static int i40e_dev_destroy_ethertype_filter(struct i40e_pf
*pf,
static int i40e_dev_destroy_tunnel_filter(struct i40e_pf *pf,
struct i40e_tunnel_filter *filter);
static int i40e_fdir_filter_flush(struct i40e_pf *pf);
+static int i40e_ethertype_filter_flush(struct i40e_pf *pf);
const struct rte_flow_ops i40e_flow_ops = {
.validate = i40e_flow_validate,
@@ -1655,10 +1656,20 @@ i40e_flow_flush(struct rte_eth_dev *dev, struct
rte_flow_error *error)
int ret;
ret = i40e_fdir_filter_flush(pf);
- if (ret)
+ if (ret) {
rte_flow_error_set(error, -ret,
RTE_FLOW_ERROR_TYPE_HANDLE, NULL,
"Failed to flush FDIR flows.");
+ return -rte_errno;
+ }
+
+ ret = i40e_ethertype_filter_flush(pf);
+ if (ret) {
+ rte_flow_error_set(error, -ret,
+ RTE_FLOW_ERROR_TYPE_HANDLE, NULL,
+ "Failed to ethertype flush flows.");
+ return -rte_errno;
+ }
return ret;
}
@@ -1694,3 +1705,31 @@ i40e_fdir_filter_flush(struct i40e_pf *pf)
return ret;
}
+
+/* Flush all ethertype filters */
+static int
+i40e_ethertype_filter_flush(struct i40e_pf *pf)
+{
+ struct i40e_ethertype_filter_list
+ *ethertype_list = &pf->ethertype.ethertype_list;
+ struct i40e_ethertype_filter *f;
+ struct i40e_flow *flow;
+ void *temp;
+ int ret = 0;
+
+ while ((f = TAILQ_FIRST(ethertype_list))) {
+ ret = i40e_dev_destroy_ethertype_filter(pf, f);
+ if (ret)
+ return ret;
+ }
+
+ /* Delete ethertype flows in flow list. */
+ TAILQ_FOREACH_SAFE(flow, &pf->flow_list, node, temp) {
+ if (flow->filter_type == RTE_ETH_FILTER_ETHERTYPE) {
+ TAILQ_REMOVE(&pf->flow_list, flow, node);
+ rte_free(flow);
+ }
+ }
+
+ return ret;
+}
--
2.5.5
[dpdk-dev] [PATCH v3 15/17] net/i40e: add flow flush function
This patch adds i40e_flow_flush function to flush all
filters for users. And flow director flush function
is involved first.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_ethdev.h | 1 +
drivers/net/i40e/i40e_fdir.c | 4 +---
drivers/net/i40e/i40e_flow.c | 51 ++
3 files changed, 53 insertions(+), 3 deletions(-)
diff --git a/drivers/net/i40e/i40e_ethdev.h b/drivers/net/i40e/i40e_ethdev.h
index b33910d..57fd796 100644
--- a/drivers/net/i40e/i40e_ethdev.h
+++ b/drivers/net/i40e/i40e_ethdev.h
@@ -788,6 +788,7 @@ int i40e_add_del_fdir_filter(struct rte_eth_dev *dev,
int i40e_dev_tunnel_filter_set(struct i40e_pf *pf,
struct rte_eth_tunnel_filter_conf *tunnel_filter,
uint8_t add);
+int i40e_fdir_flush(struct rte_eth_dev *dev);
/* I40E_DEV_PRIVATE_TO */
#define I40E_DEV_PRIVATE_TO_PF(adapter) \
diff --git a/drivers/net/i40e/i40e_fdir.c b/drivers/net/i40e/i40e_fdir.c
index e1f97bb..965e3b2 100644
--- a/drivers/net/i40e/i40e_fdir.c
+++ b/drivers/net/i40e/i40e_fdir.c
@@ -119,8 +119,6 @@ static int i40e_fdir_filter_programming(struct i40e_pf *pf,
enum i40e_filter_pctype pctype,
const struct rte_eth_fdir_filter *filter,
bool add);
-static int i40e_fdir_flush(struct rte_eth_dev *dev);
-
static int i40e_fdir_filter_convert(const struct rte_eth_fdir_filter *input,
struct i40e_fdir_filter *filter);
static struct i40e_fdir_filter *
@@ -1325,7 +1323,7 @@ i40e_fdir_filter_programming(struct i40e_pf *pf,
* i40e_fdir_flush - clear all filters of Flow Director table
* @pf: board private structure
*/
-static int
+int
i40e_fdir_flush(struct rte_eth_dev *dev)
{
struct i40e_pf *pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private);
diff --git a/drivers/net/i40e/i40e_flow.c b/drivers/net/i40e/i40e_flow.c
index 605ee70..ee463c9 100644
--- a/drivers/net/i40e/i40e_flow.c
+++ b/drivers/net/i40e/i40e_flow.c
@@ -68,6 +68,8 @@ static struct rte_flow *i40e_flow_create(struct rte_eth_dev
*dev,
const struct rte_flow_item pattern[],
const struct rte_flow_action actions[],
struct rte_flow_error *error);
+static int i40e_flow_flush(struct rte_eth_dev *dev,
+ struct rte_flow_error *error);
static int i40e_flow_destroy(struct rte_eth_dev *dev,
struct rte_flow *flow,
struct rte_flow_error *error);
@@ -101,11 +103,13 @@ static int i40e_dev_destroy_ethertype_filter(struct
i40e_pf *pf,
struct i40e_ethertype_filter *filter);
static int i40e_dev_destroy_tunnel_filter(struct i40e_pf *pf,
struct i40e_tunnel_filter *filter);
+static int i40e_fdir_filter_flush(struct i40e_pf *pf);
const struct rte_flow_ops i40e_flow_ops = {
.validate = i40e_flow_validate,
.create = i40e_flow_create,
.destroy = i40e_flow_destroy,
+ .flush = i40e_flow_flush,
};
union i40e_filter_t cons_filter;
@@ -1643,3 +1647,50 @@ i40e_dev_destroy_tunnel_filter(struct i40e_pf *pf,
return ret;
}
+
+static int
+i40e_flow_flush(struct rte_eth_dev *dev, struct rte_flow_error *error)
+{
+ struct i40e_pf *pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private);
+ int ret;
+
+ ret = i40e_fdir_filter_flush(pf);
+ if (ret)
+ rte_flow_error_set(error, -ret,
+ RTE_FLOW_ERROR_TYPE_HANDLE, NULL,
+ "Failed to flush FDIR flows.");
+
+ return ret;
+}
+
+static int
+i40e_fdir_filter_flush(struct i40e_pf *pf)
+{
+ struct rte_eth_dev *dev = pf->adapter->eth_dev;
+ struct i40e_fdir_info *fdir_info = &pf->fdir;
+ struct i40e_fdir_filter *fdir_filter;
+ struct i40e_flow *flow;
+ void *temp;
+ int ret;
+
+ ret = i40e_fdir_flush(dev);
+ if (!ret) {
+ /* Delete FDIR filters in FDIR list. */
+ while ((fdir_filter = TAILQ_FIRST(&fdir_info->fdir_list))) {
+ ret = i40e_sw_fdir_filter_del(pf,
+ &fdir_filter->fdir.input);
+ if (ret < 0)
+ return ret;
+ }
+
+ /* Delete FDIR flows in flow list. */
+ TAILQ_FOREACH_SAFE(flow, &pf->flow_list, node, temp) {
+ if (flow->filter_type == RTE_ETH_FILTER_FDIR) {
+ TAILQ_REMOVE(&pf->flow_list, flow, node);
+ rte_free(flow);
+ }
+ }
+ }
+
+ return ret;
+}
--
2.5.5
[dpdk-dev] [PATCH v3 11/17] net/i40e: add flow destroy function
This patch adds i40e_flow_destroy function to destroy
a flow for users.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_flow.c | 33 +
1 file changed, 33 insertions(+)
diff --git a/drivers/net/i40e/i40e_flow.c b/drivers/net/i40e/i40e_flow.c
index 8db8e0f..a64ef0e 100644
--- a/drivers/net/i40e/i40e_flow.c
+++ b/drivers/net/i40e/i40e_flow.c
@@ -67,6 +67,9 @@ static struct rte_flow *i40e_flow_create(struct rte_eth_dev
*dev,
const struct rte_flow_item pattern[],
const struct rte_flow_action actions[],
struct rte_flow_error *error);
+static int i40e_flow_destroy(struct rte_eth_dev *dev,
+struct rte_flow *flow,
+struct rte_flow_error *error);
static int i40e_parse_ethertype_pattern(__rte_unused struct rte_eth_dev *dev,
const struct rte_flow_item *pattern,
struct rte_flow_error *error,
@@ -97,6 +100,7 @@ static int i40e_parse_attr(const struct rte_flow_attr *attr,
const struct rte_flow_ops i40e_flow_ops = {
.validate = i40e_flow_validate,
.create = i40e_flow_create,
+ .destroy = i40e_flow_destroy,
};
union i40e_filter_t cons_filter;
@@ -1523,3 +1527,32 @@ i40e_flow_create(struct rte_eth_dev *dev,
rte_free(flow);
return NULL;
}
+
+static int
+i40e_flow_destroy(struct rte_eth_dev *dev,
+ struct rte_flow *flow,
+ struct rte_flow_error *error)
+{
+ struct i40e_pf *pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private);
+ struct i40e_flow *pmd_flow = (struct i40e_flow *)flow;
+ enum rte_filter_type filter_type = pmd_flow->filter_type;
+ int ret = 0;
+
+ switch (filter_type) {
+ default:
+ PMD_DRV_LOG(WARNING, "Filter type (%d) not supported",
+ filter_type);
+ ret = -EINVAL;
+ break;
+ }
+
+ if (!ret) {
+ TAILQ_REMOVE(&pf->flow_list, pmd_flow, node);
+ rte_free(pmd_flow);
+ } else
+ rte_flow_error_set(error, -ret,
+ RTE_FLOW_ERROR_TYPE_HANDLE, NULL,
+ "Failed to destroy flow.");
+
+ return ret;
+}
--
2.5.5
[dpdk-dev] [PATCH v3 17/17] net/i40e: flush tunnel filters
This patch adds i40e_tunnel_filter_flush function
to flush all tunnel filters, including filters in
SW and HW.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_flow.c | 37 +
1 file changed, 37 insertions(+)
diff --git a/drivers/net/i40e/i40e_flow.c b/drivers/net/i40e/i40e_flow.c
index 947199d..d19d952 100644
--- a/drivers/net/i40e/i40e_flow.c
+++ b/drivers/net/i40e/i40e_flow.c
@@ -105,6 +105,7 @@ static int i40e_dev_destroy_tunnel_filter(struct i40e_pf
*pf,
struct i40e_tunnel_filter *filter);
static int i40e_fdir_filter_flush(struct i40e_pf *pf);
static int i40e_ethertype_filter_flush(struct i40e_pf *pf);
+static int i40e_tunnel_filter_flush(struct i40e_pf *pf);
const struct rte_flow_ops i40e_flow_ops = {
.validate = i40e_flow_validate,
@@ -1671,6 +1672,14 @@ i40e_flow_flush(struct rte_eth_dev *dev, struct
rte_flow_error *error)
return -rte_errno;
}
+ ret = i40e_tunnel_filter_flush(pf);
+ if (ret) {
+ rte_flow_error_set(error, -ret,
+ RTE_FLOW_ERROR_TYPE_HANDLE, NULL,
+ "Failed to flush tunnel flows.");
+ return -rte_errno;
+ }
+
return ret;
}
@@ -1733,3 +1742,31 @@ i40e_ethertype_filter_flush(struct i40e_pf *pf)
return ret;
}
+
+/* Flush all tunnel filters */
+static int
+i40e_tunnel_filter_flush(struct i40e_pf *pf)
+{
+ struct i40e_tunnel_filter_list
+ *tunnel_list = &pf->tunnel.tunnel_list;
+ struct i40e_tunnel_filter *f;
+ struct i40e_flow *flow;
+ void *temp;
+ int ret = 0;
+
+ while ((f = TAILQ_FIRST(tunnel_list))) {
+ ret = i40e_dev_destroy_tunnel_filter(pf, f);
+ if (ret)
+ return ret;
+ }
+
+ /* Delete tunnel flows in flow list. */
+ TAILQ_FOREACH_SAFE(flow, &pf->flow_list, node, temp) {
+ if (flow->filter_type == RTE_ETH_FILTER_TUNNEL) {
+ TAILQ_REMOVE(&pf->flow_list, flow, node);
+ rte_free(flow);
+ }
+ }
+
+ return ret;
+}
--
2.5.5
[dpdk-dev] [PATCH v2 0/5] Chained Mbufs support in SW PMDs
This patch set adds support of scattered-gather list for SW PMDs. As of now, application needs to reserve continuous block of memory for mbufs which is not always the case. Hence needed to support chaining of mbufs which are smaller in size but can be used if chained. Above work involves: a) Create mbuf functions to coalesce mbuf chains into a single mbuf. b) For each software poll mode driver code to detect chained mbufs support and coalesce these before preforming crypto. c) Add relevant unit tests to test the functionality. Known limitations for openssl PMD: While libcrypto library expects continuous destination buffer for output of cipher operations, implementation of openssl PMD is limited the same way, and requires contigous destination mbuf. changes in v2: - add support for sgl in openssl PMD - rte_pktmbuf_coalesce replaced with rte_pktmbuf_linearize - extended test vector data for aes gcm from 60 to 2048 bytes Tomasz Kulasek (5): rte_mbuf: add rte_pktmbuf_linearize test: add rte_pktmbuf_linearize unit tests crypto: add sgl support in sw PMDs crypto: add sgl support in openssl PMD test: add sgl unit tests for crypto devices app/test/test_cryptodev.c | 589 +++- app/test/test_cryptodev.h | 139 +++ app/test/test_cryptodev_aes_test_vectors.h | 54 ++- app/test/test_cryptodev_blockcipher.c | 181 ++--- app/test/test_cryptodev_blockcipher.h |3 +- app/test/test_cryptodev_des_test_vectors.h | 29 ++ app/test/test_cryptodev_gcm_test_vectors.h | 557 +- app/test/test_mbuf.c | 123 ++ doc/guides/cryptodevs/openssl.rst |2 +- drivers/crypto/aesni_gcm/aesni_gcm_pmd.c | 14 + drivers/crypto/aesni_mb/rte_aesni_mb_pmd.c | 19 +- drivers/crypto/kasumi/rte_kasumi_pmd.c | 13 + drivers/crypto/null/null_crypto_pmd.c |3 +- drivers/crypto/openssl/rte_openssl_pmd.c | 329 drivers/crypto/snow3g/rte_snow3g_pmd.c | 15 + drivers/crypto/zuc/rte_zuc_pmd.c | 13 + lib/librte_cryptodev/rte_cryptodev.c |4 +- lib/librte_cryptodev/rte_cryptodev.h |3 +- lib/librte_mbuf/rte_mbuf.h | 56 +++ 19 files changed, 1991 insertions(+), 155 deletions(-) -- 1.7.9.5
[dpdk-dev] [PATCH v2 1/5] rte_mbuf: add rte_pktmbuf_linearize
This patch adds function rte_pktmbuf_linearize to let crypto PMD coalesce
chained mbuf before crypto operation and extend their capabilities to
support segmented mbufs when device cannot handle them natively.
Signed-off-by: Tomasz Kulasek
---
lib/librte_mbuf/rte_mbuf.h | 56
1 file changed, 56 insertions(+)
diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h
index ead7c6e..b11a31d 100644
--- a/lib/librte_mbuf/rte_mbuf.h
+++ b/lib/librte_mbuf/rte_mbuf.h
@@ -1647,6 +1647,62 @@ static inline int rte_pktmbuf_chain(struct rte_mbuf
*head, struct rte_mbuf *tail
}
/**
+ * Linearize data in mbuf.
+ *
+ * This function coalesce mbuf merging data in the first segment, unchaining
+ * rest, and then frees them.
+ *
+ * All operations are done in-place, so the structure of incoming mbuf
+ * is changed.
+ *
+ * @param mbuf
+ * mbuf to linearize
+ * @return
+ * - 0, on success
+ * - -1, on error
+ */
+static inline int
+rte_pktmbuf_linearize(struct rte_mbuf *mbuf)
+{
+ int l, n;
+ struct rte_mbuf *m;
+ struct rte_mbuf *m_next;
+ char *buffer;
+
+ if (rte_pktmbuf_is_contiguous(mbuf))
+ return 0;
+
+ /* Extend first segment to the total packet length
+*/
+ n = rte_pktmbuf_pkt_len(mbuf) - rte_pktmbuf_data_len(mbuf);
+
+ if (unlikely(n > rte_pktmbuf_tailroom(mbuf)))
+ return -1;
+
+ buffer = rte_pktmbuf_mtod_offset(mbuf, char *, mbuf->data_len);
+ mbuf->data_len = (uint16_t)(mbuf->pkt_len);
+
+ /* Append data from next segments to the first one
+*/
+ m = mbuf->next;
+ while (m != NULL) {
+ m_next = m->next;
+
+ l = rte_pktmbuf_data_len(m);
+ rte_memcpy(buffer, rte_pktmbuf_mtod(m, char *), l);
+ buffer += l;
+
+ rte_pktmbuf_free_seg(m);
+ m = m_next;
+ }
+
+ mbuf->next = NULL;
+ mbuf->nb_segs = 1;
+
+ return 0;
+}
+
+/**
* Dump an mbuf structure to a file.
*
* Dump all fields for the given packet mbuf and all its associated
--
1.7.9.5
[dpdk-dev] [PATCH v2 2/5] test: add rte_pktmbuf_linearize unit tests
This patch tests rte_pktmbuf_coalesce functionality:
1) Creates banch of segmented mbufs with different size and number of
segments.
2) Generates pkt_len bytes of random data.
3) Fills noncontigouos mbuf with randomly generated data.
4) Uses rte_pktmbuf_linearize to coalesce segmented buffer into one
contiguous.
5) Verifies data in linearized buffer.
Signed-off-by: Tomasz Kulasek
---
app/test/test_mbuf.c | 123 ++
1 file changed, 123 insertions(+)
diff --git a/app/test/test_mbuf.c b/app/test/test_mbuf.c
index c0823ea..39577e7 100644
--- a/app/test/test_mbuf.c
+++ b/app/test/test_mbuf.c
@@ -930,6 +930,124 @@
return 0;
}
+static int
+test_mbuf_linearize(int pkt_len, int nb_segs) {
+
+ struct rte_mbuf *m = NULL, *mbuf_src = NULL;
+ uint8_t data[pkt_len], *src, *dst;
+ int data_len = 0;
+ int i, size;
+ int t_len;
+
+ if (pkt_len < 1) {
+ printf("Packet size must be 1 or more (is %d)\n", pkt_len);
+ return -1;
+ }
+
+ if (nb_segs < 1) {
+ printf("Number of segments must be 1 or more (is %d)\n",
+ nb_segs);
+ return -1;
+ }
+
+ /* Setup buffer */
+ for (i = 0; i < pkt_len; i++)
+ data[i] = (uint8_t) rte_rand();
+
+ t_len = pkt_len >= nb_segs ? pkt_len / nb_segs : 1;
+ src = data;
+ size = pkt_len;
+
+ /* Create chained mbuf_src and fill it generated data */
+ for (i = 0; size > 0; i++) {
+
+ m = rte_pktmbuf_alloc(pktmbuf_pool);
+ if (i == 0)
+ mbuf_src = m;
+
+ if (!m) {
+ printf("Cannot create segment for source mbuf");
+ goto fail;
+ }
+
+ /* Make sure if tailroom is zeroed */
+ memset(rte_pktmbuf_mtod(m, uint8_t *), 0,
+ rte_pktmbuf_tailroom(m));
+
+ data_len = size > t_len ? t_len : size;
+ dst = (uint8_t *)rte_pktmbuf_append(m, data_len);
+ if (!dst) {
+ printf("Cannot append %d bytes to the mbuf\n",
+ data_len);
+ goto fail;
+ }
+
+ rte_memcpy(dst, src, data_len);
+ src += data_len;
+
+ if (mbuf_src != m)
+ rte_pktmbuf_chain(mbuf_src, m);
+
+ size -= data_len;
+
+ }
+
+ /* Create destination buffer to store coalesced data */
+ if (rte_pktmbuf_linearize(mbuf_src)) {
+ printf("Mbuf linearization failed\n");
+ goto fail;
+ }
+
+ if (!rte_pktmbuf_is_contiguous(mbuf_src)) {
+ printf("Source buffer should be contiguous after "
+ "linearization\n");
+ goto fail;
+ }
+
+ src = rte_pktmbuf_mtod(mbuf_src, uint8_t *);
+
+ if (memcmp(src, data, rte_pktmbuf_pkt_len(mbuf_src))) {
+ printf("Incorrect data in coalesced mbuf\n");
+ goto fail;
+ }
+
+ if (mbuf_src)
+ rte_pktmbuf_free(mbuf_src);
+ return 0;
+
+fail:
+ if (mbuf_src)
+ rte_pktmbuf_free(mbuf_src);
+ return -1;
+}
+
+static int
+test_mbuf_linearize_check(void)
+{
+ struct test_mbuf_array {
+ int size;
+ int nb_segs;
+ } mbuf_array[5] = {
+ { 128, 1 },
+ { 64, 64 },
+ { 512, 10 },
+ { 250, 11 },
+ { 123, 8 },
+ };
+ unsigned int i;
+
+ printf("Test mbuf linearize API\n");
+
+ for (i = 0; i < RTE_DIM(mbuf_array); i++)
+ if (test_mbuf_linearize(mbuf_array[i].size,
+ mbuf_array[i].nb_segs)) {
+ printf("Test failed for %d, %d\n", mbuf_array[i].size,
+ mbuf_array[i].nb_segs);
+ return -1;
+ }
+
+ return 0;
+}
static int
test_mbuf(void)
@@ -1023,6 +1141,11 @@
printf("test_failing_mbuf_sanity_check() failed\n");
return -1;
}
+
+ if (test_mbuf_linearize_check() < 0) {
+ printf("test_mbuf_linearize_check() failed\n");
+ return -1;
+ }
return 0;
}
--
1.7.9.5
[dpdk-dev] [PATCH v2 3/5] crypto: add sgl support in sw PMDs
This patch introduces RTE_CRYPTODEV_FF_MBUF_SCATTER_GATHER feature flag
informing that selected crypto device supports segmented mbufs natively
and doesn't need to be coalesced before crypto operation.
While using segmented buffers in crypto devices may have unpredictable
results, for PMDs which doesn't support it natively, additional check is
made for debug compilation.
Signed-off-by: Tomasz Kulasek
---
drivers/crypto/aesni_gcm/aesni_gcm_pmd.c | 14 ++
drivers/crypto/aesni_mb/rte_aesni_mb_pmd.c | 19 ---
drivers/crypto/kasumi/rte_kasumi_pmd.c | 13 +
drivers/crypto/null/null_crypto_pmd.c |3 ++-
drivers/crypto/snow3g/rte_snow3g_pmd.c | 15 +++
drivers/crypto/zuc/rte_zuc_pmd.c | 13 +
lib/librte_cryptodev/rte_cryptodev.c |4 ++--
lib/librte_cryptodev/rte_cryptodev.h |3 ++-
8 files changed, 77 insertions(+), 7 deletions(-)
diff --git a/drivers/crypto/aesni_gcm/aesni_gcm_pmd.c
b/drivers/crypto/aesni_gcm/aesni_gcm_pmd.c
index dba5e15..1a6120c 100644
--- a/drivers/crypto/aesni_gcm/aesni_gcm_pmd.c
+++ b/drivers/crypto/aesni_gcm/aesni_gcm_pmd.c
@@ -375,6 +375,20 @@
break;
}
+#ifdef RTE_LIBRTE_PMD_AESNI_GCM_DEBUG
+ if (!rte_pktmbuf_is_contiguous(ops[i]->sym->m_src) ||
+ (ops[i]->sym->m_dst != NULL &&
+ !rte_pktmbuf_is_contiguous(
+ ops[i]->sym->m_dst))) {
+ ops[i]->status = RTE_CRYPTO_OP_STATUS_INVALID_ARGS;
+ GCM_LOG_ERR("PMD supports only contiguous mbufs, "
+ "op (%p) provides noncontiguous mbuf as "
+ "source/destination buffer.\n", ops[i]);
+ qp->qp_stats.enqueue_err_count++;
+ break;
+ }
+#endif
+
retval = process_gcm_crypto_op(qp, ops[i]->sym, sess);
if (retval < 0) {
ops[i]->status = RTE_CRYPTO_OP_STATUS_INVALID_ARGS;
diff --git a/drivers/crypto/aesni_mb/rte_aesni_mb_pmd.c
b/drivers/crypto/aesni_mb/rte_aesni_mb_pmd.c
index f07cd07..b5e115e 100644
--- a/drivers/crypto/aesni_mb/rte_aesni_mb_pmd.c
+++ b/drivers/crypto/aesni_mb/rte_aesni_mb_pmd.c
@@ -529,15 +529,28 @@
int i, processed_jobs = 0;
for (i = 0; i < nb_ops; i++) {
-#ifdef RTE_LIBRTE_AESNI_MB_DEBUG
- if (unlikely(op->type != RTE_CRYPTO_OP_TYPE_SYMMETRIC)) {
+#ifdef RTE_LIBRTE_PMD_AESNI_MB_DEBUG
+ if (unlikely(ops[i]->type != RTE_CRYPTO_OP_TYPE_SYMMETRIC)) {
MB_LOG_ERR("PMD only supports symmetric crypto "
"operation requests, op (%p) is not a "
- "symmetric operation.", op);
+ "symmetric operation.", ops[i]);
+ qp->stats.enqueue_err_count++;
+ goto flush_jobs;
+ }
+
+ if (!rte_pktmbuf_is_contiguous(ops[i]->sym->m_src) ||
+ (ops[i]->sym->m_dst != NULL &&
+ !rte_pktmbuf_is_contiguous(
+ ops[i]->sym->m_dst))) {
+ MB_LOG_ERR("PMD supports only contiguous mbufs, "
+ "op (%p) provides noncontiguous mbuf as "
+ "source/destination buffer.\n", ops[i]);
+ ops[i]->status = RTE_CRYPTO_OP_STATUS_INVALID_ARGS;
qp->stats.enqueue_err_count++;
goto flush_jobs;
}
#endif
+
sess = get_session(qp, ops[i]);
if (unlikely(sess == NULL)) {
qp->stats.enqueue_err_count++;
diff --git a/drivers/crypto/kasumi/rte_kasumi_pmd.c
b/drivers/crypto/kasumi/rte_kasumi_pmd.c
index b119da2..4bdd7bb 100644
--- a/drivers/crypto/kasumi/rte_kasumi_pmd.c
+++ b/drivers/crypto/kasumi/rte_kasumi_pmd.c
@@ -455,6 +455,19 @@
for (i = 0; i < nb_ops; i++) {
curr_c_op = ops[i];
+#ifdef RTE_LIBRTE_PMD_KASUMI_DEBUG
+ if (!rte_pktmbuf_is_contiguous(curr_c_op->sym->m_src) ||
+ (curr_c_op->sym->m_dst != NULL &&
+ !rte_pktmbuf_is_contiguous(
+ curr_c_op->sym->m_dst))) {
+ KASUMI_LOG_ERR("PMD supports only contiguous mbufs, "
+ "op (%p) provides noncontiguous mbuf as "
+ "source/destination buffer.\n", curr_c_op);
+ curr_c_op->status = RTE_CRYPTO_OP_STATUS_INVALID_ARGS;
+ break;
+ }
+#endif
+
/* Set status as enqueued (not processed yet) by default. */
[dpdk-dev] [PATCH v2 4/5] crypto: add sgl support in openssl PMD
Previous implementation uses EVP_EncryptUpdate() on whole source buffer
limiting its usage to the contiguous buffers.
This implementation calls EVP_EncryptUpdate() on each segment for cipher
operations in continous mode, before finalization allowing to provide
chained mbuf as a source.
However, libcrypto library expects continuous destination buffer for
output of cipher operations and implementation of openssl PMD is limited
the same way, requiring contiguous destination mbuf.
Signed-off-by: Tomasz Kulasek
---
doc/guides/cryptodevs/openssl.rst|2 +-
drivers/crypto/openssl/rte_openssl_pmd.c | 329 +++---
2 files changed, 258 insertions(+), 73 deletions(-)
diff --git a/doc/guides/cryptodevs/openssl.rst
b/doc/guides/cryptodevs/openssl.rst
index d2b5906..05e6d67 100644
--- a/doc/guides/cryptodevs/openssl.rst
+++ b/doc/guides/cryptodevs/openssl.rst
@@ -112,6 +112,6 @@ Limitations
---
* Maximum number of sessions is 2048.
-* Chained mbufs are not supported.
+* Chained mbufs are supported only for source mbuf (destination must be
contiguous).
* Hash only is not supported for GCM and GMAC.
* Cipher only is not supported for GCM and GMAC.
diff --git a/drivers/crypto/openssl/rte_openssl_pmd.c
b/drivers/crypto/openssl/rte_openssl_pmd.c
index 5f8fa33..22639b4 100644
--- a/drivers/crypto/openssl/rte_openssl_pmd.c
+++ b/drivers/crypto/openssl/rte_openssl_pmd.c
@@ -484,22 +484,110 @@
* Process Operations
*--
*/
+static inline int
+process_openssl_encryption_update(struct rte_mbuf *mbuf_src, int offset,
+ uint8_t **dst, int srclen, EVP_CIPHER_CTX *ctx)
+{
+ struct rte_mbuf *m;
+ int dstlen;
+ int l, n = srclen;
+ uint8_t *src;
+
+ for (m = mbuf_src; m != NULL && offset > rte_pktmbuf_data_len(m);
+ m = m->next)
+ offset -= rte_pktmbuf_data_len(m);
+
+ if (m == 0)
+ return -1;
+
+ src = rte_pktmbuf_mtod_offset(m, uint8_t *, offset);
+
+ l = rte_pktmbuf_data_len(m) - offset;
+ if (srclen <= l) {
+ if (EVP_EncryptUpdate(ctx, *dst, &dstlen, src, srclen) <= 0)
+ return -1;
+ *dst += l;
+ return 0;
+ }
+
+ if (EVP_EncryptUpdate(ctx, *dst, &dstlen, src, l) <= 0)
+ return -1;
+
+ *dst += dstlen;
+ n -= l;
+
+ for (m = m->next; (m != NULL) && (n > 0); m = m->next) {
+ src = rte_pktmbuf_mtod(m, uint8_t *);
+ l = rte_pktmbuf_data_len(m) < n ? rte_pktmbuf_data_len(m) : n;
+ if (EVP_EncryptUpdate(ctx, *dst, &dstlen, src, l) <= 0)
+ return -1;
+ *dst += dstlen;
+ n -= l;
+ }
+
+ return 0;
+}
+
+static inline int
+process_openssl_decryption_update(struct rte_mbuf *mbuf_src, int offset,
+ uint8_t **dst, int srclen, EVP_CIPHER_CTX *ctx)
+{
+ struct rte_mbuf *m;
+ int dstlen;
+ int l, n = srclen;
+ uint8_t *src;
+
+ for (m = mbuf_src; m != NULL && offset > rte_pktmbuf_data_len(m);
+ m = m->next)
+ offset -= rte_pktmbuf_data_len(m);
+
+ if (m == 0)
+ return -1;
+
+ src = rte_pktmbuf_mtod_offset(m, uint8_t *, offset);
+
+ l = rte_pktmbuf_data_len(m) - offset;
+ if (srclen <= l) {
+ if (EVP_DecryptUpdate(ctx, *dst, &dstlen, src, srclen) <= 0)
+ return -1;
+ *dst += l;
+ return 0;
+ }
+
+ if (EVP_DecryptUpdate(ctx, *dst, &dstlen, src, l) <= 0)
+ return -1;
+
+ *dst += dstlen;
+ n -= l;
+
+ for (m = m->next; (m != NULL) && (n > 0); m = m->next) {
+ src = rte_pktmbuf_mtod(m, uint8_t *);
+ l = rte_pktmbuf_data_len(m) < n ? rte_pktmbuf_data_len(m) : n;
+ if (EVP_DecryptUpdate(ctx, *dst, &dstlen, src, l) <= 0)
+ return -1;
+ *dst += dstlen;
+ n -= l;
+ }
+
+ return 0;
+}
/** Process standard openssl cipher encryption */
static int
-process_openssl_cipher_encrypt(uint8_t *src, uint8_t *dst,
- uint8_t *iv, uint8_t *key, int srclen,
+process_openssl_cipher_encrypt(struct rte_mbuf *mbuf_src, uint8_t *dst,
+ int offset, uint8_t *iv, uint8_t *key, int srclen,
EVP_CIPHER_CTX *ctx, const EVP_CIPHER *algo)
{
- int dstlen, totlen;
+ int totlen;
if (EVP_EncryptInit_ex(ctx, algo, NULL, key, iv) <= 0)
goto process_cipher_encrypt_err;
- if (EVP_EncryptUpdate(ctx, dst, &dstlen, src, srclen) <= 0)
+ if (process_openssl_encryption_update(mbuf_src, offset, &dst,
+ srclen, ctx))
goto process_cipher_encrypt_err;
- if (EVP_EncryptFinal_e
[dpdk-dev] [PATCH v2 5/5] test: add sgl unit tests for crypto devices
This patch provides unit tests for set of cipher/hash combinations covering
currently implemented crypto PMD's and allowing to verify scatter gather
support.
Signed-off-by: Tomasz Kulasek
---
app/test/test_cryptodev.c | 589 +++-
app/test/test_cryptodev.h | 139 +++
app/test/test_cryptodev_aes_test_vectors.h | 54 ++-
app/test/test_cryptodev_blockcipher.c | 181 ++---
app/test/test_cryptodev_blockcipher.h |3 +-
app/test/test_cryptodev_des_test_vectors.h | 29 ++
app/test/test_cryptodev_gcm_test_vectors.h | 557 +-
7 files changed, 1477 insertions(+), 75 deletions(-)
diff --git a/app/test/test_cryptodev.c b/app/test/test_cryptodev.c
index 872f8b4..e419f80 100644
--- a/app/test/test_cryptodev.c
+++ b/app/test/test_cryptodev.c
@@ -1688,6 +1688,10 @@ struct crypto_unittest_params {
TEST_ASSERT_NOT_NULL(sym_op->cipher.iv.data, "no room to prepend iv");
+ /* For OOP operation both buffers must have the same size */
+ if (ut_params->obuf)
+ rte_pktmbuf_prepend(ut_params->obuf, iv_pad_len);
+
memset(sym_op->cipher.iv.data, 0, iv_pad_len);
sym_op->cipher.iv.phys_addr = rte_pktmbuf_mtophys(ut_params->ibuf);
sym_op->cipher.iv.length = iv_pad_len;
@@ -2509,6 +2513,84 @@ struct crypto_unittest_params {
}
static int
+test_kasumi_encryption_sgl(const struct kasumi_test_data *tdata)
+{
+ struct crypto_testsuite_params *ts_params = &testsuite_params;
+ struct crypto_unittest_params *ut_params = &unittest_params;
+
+ int retval;
+
+ unsigned int plaintext_pad_len;
+ unsigned int plaintext_len;
+
+ uint8_t buffer[1];
+ const uint8_t *ciphertext;
+
+ struct rte_cryptodev_info dev_info;
+
+ rte_cryptodev_info_get(ts_params->valid_devs[0], &dev_info);
+ if (!(dev_info.feature_flags & RTE_CRYPTODEV_FF_MBUF_SCATTER_GATHER)) {
+ printf("Device doesn't support scatter-gather. "
+ "Test Skipped.\n");
+ return 0;
+ }
+
+ /* Create KASUMI session */
+ retval = create_wireless_algo_cipher_session(ts_params->valid_devs[0],
+ RTE_CRYPTO_CIPHER_OP_ENCRYPT,
+ RTE_CRYPTO_CIPHER_KASUMI_F8,
+ tdata->key.data, tdata->key.len);
+ if (retval < 0)
+ return retval;
+
+ plaintext_len = ceil_byte_length(tdata->plaintext.len);
+
+
+ /* Append data which is padded to a multiple */
+ /* of the algorithms block size */
+ plaintext_pad_len = RTE_ALIGN_CEIL(plaintext_len, 8);
+
+ ut_params->ibuf = create_segmented_mbuf(ts_params->mbuf_pool,
+ plaintext_pad_len, 10);
+
+ pktmbuf_write(ut_params->ibuf, 0, plaintext_len, tdata->plaintext.data);
+
+ /* Create KASUMI operation */
+ retval = create_wireless_algo_cipher_operation(tdata->iv.data,
+ tdata->iv.len,
+ tdata->plaintext.len,
+ tdata->validCipherOffsetLenInBits.len,
+ RTE_CRYPTO_CIPHER_KASUMI_F8);
+ if (retval < 0)
+ return retval;
+
+ ut_params->op = process_crypto_request(ts_params->valid_devs[0],
+ ut_params->op);
+ TEST_ASSERT_NOT_NULL(ut_params->op, "failed to retrieve obuf");
+
+ ut_params->obuf = ut_params->op->sym->m_dst;
+
+ if (ut_params->obuf)
+ ciphertext = rte_pktmbuf_read(ut_params->obuf, tdata->iv.len,
+ plaintext_len, buffer);
+ else
+ ciphertext = rte_pktmbuf_read(ut_params->ibuf, tdata->iv.len,
+ plaintext_len, buffer);
+
+ /* Validate obuf */
+ TEST_HEXDUMP(stdout, "ciphertext:", ciphertext, plaintext_len);
+
+ /* Validate obuf */
+ TEST_ASSERT_BUFFERS_ARE_EQUAL_BIT(
+ ciphertext,
+ tdata->ciphertext.data,
+ tdata->validCipherLenInBits.len,
+ "KASUMI Ciphertext data not as expected");
+ return 0;
+}
+
+
+static int
test_kasumi_encryption_oop(const struct kasumi_test_data *tdata)
{
struct crypto_testsuite_params *ts_params = &testsuite_params;
@@ -2577,6 +2659,81 @@ struct crypto_unittest_params {
}
static int
+test_kasumi_encryption_oop_sgl(const struct kasumi_test_data *tdata)
+{
+ struct crypto_testsuite_params *ts_params = &testsuite_params;
+ struct crypto_unittest_params *ut_params = &unittest_params;
+
+ int retval;
+ unsigned int plaintext_pad_len;
+ unsigned int plaintext_len;
+
+ const uint8_t *ciphertext;
+ uint8_t buffer[2048];
+
+
[dpdk-dev] Running DPDK as an unprivileged user
Hi all, I've been digging in to what it would take to run DPDK as an unprivileged user and I have some findings that I thought were worthy of discussion. The assumptions here are that I'm using a very recent Linux kernel (4.8.15 to be specific) and I'm using vfio with my IOMMU enabled. I'm only interested in making it possible to run as an unprivileged user in this type of environment. There are a few key things that DPDK needs to do in order to run as an unprivileged user: 1) Allocate hugepages 2) Map device resources 3) Map hugepage virtual addresses to DMA addresses. For #1 and #2, DPDK works just fine today. You simply chown the relevant resources in sysfs to the desired user and everything is happy. The problem is #3. This currently relies on looking up the mappings in /proc/self/pagemap, but the ability to get physical addresses in /proc/self/pagemap as an unprivileged user was removed from the kernel in the 4.x timeframe due to the Rowhammer vulnerability. At this time, it is not possible to run DPDK as an unprivileged user on a 4.x Linux kernel. There is a way to make this work though, which I'll outline now. Unfortunately, I think it is going to require some very significant changes to the initialization flow in the EAL. One bit of of background before I go into how to fix this - there are three types of memory addresses - virtual addresses, physical addresses, and DMA addresses. Sometimes DMA addresses are called bus addresses or I/O addresses, but I'll call them DMA addresses because I think that's the clearest name. In a system without an IOMMU, DMA addresses and physical addresses are equivalent, but in a system with an IOMMU any arbitrary DMA address can be chosen by the user to map to a given physical address. For security reasons (rowhammer), it is no longer considered safe to expose physical addresses to userspace, but it is perfectly fine to expose DMA addresses when an IOMMU is present. DPDK today begins by allocating all of the required hugepages, then finds all of the physical addresses for those hugepages using /proc/self/pagemap, sorts the hugepages by physical address, then remaps the pages to contiguous virtual addresses. Later on and if vfio is enabled, it asks vfio to pin the hugepages and to set their DMA addresses in the IOMMU to be the physical addresses discovered earlier. Of course, running as an unprivileged user means all of the physical addresses in /proc/self/pagemap are just 0, so this doesn't end up working. Further, there is no real reason to choose the physical address as the DMA address in the IOMMU - it would be better to just count up starting at 0. Also, because the pages are pinned after the virtual to physical mapping is looked up, there is a window where a page could be moved. Hugepage mappings can be moved on more recent kernels (at least 4.x), and the reliability of hugepages having static mappings decreases with every kernel release. Note that this probably means that using uio on recent kernels is subtly broken and cannot be supported going forward because there is no uio mechanism to pin the memory. The first open question I have is whether DPDK should allow uio at all on recent (4.x) kernels. My current understanding is that there is no way to pin memory and hugepages can now be moved around, so uio would be unsafe. What does the community think here? My second question is whether the user should be allowed to mix uio and vfio usage simultaneously. For vfio, the physical addresses are really DMA addresses and are best when arbitrarily chosen to appear sequential relative to their virtual addresses. For uio, they are physical addresses and are not chosen at all. It seems that these two things are in conflict and that it will be difficult, ugly, and maybe impossible to resolve the simultaneous use of both. Once we agree on the above two things, we can try to talk through some solutions in the code. Thanks, Ben
Re: [dpdk-dev] Running DPDK as an unprivileged user
On Thu, 29 Dec 2016 20:41:21 + "Walker, Benjamin" wrote: > The first open question I have is whether DPDK should allow > uio at all on recent (4.x) kernels. My current understanding > is that there is no way to pin memory and hugepages can now > be moved around, so uio would be unsafe. What does the > community think here? DMA access without IOMMU (ie UIO) is not safe from a security point of view. A malicious app could program device (like Ethernet NIC) to change its current privledge level in kernel memory. Therefore ignore UIO as an option if you want to allow unprivileged access. But there are many many systems without working IOMMU. Not just broken motherboards, but virtualization environments (Xen, Hyper-V, and KVM until very recently) where IOMMU is no going to work. In these environments, DPDK is still useful where the security risks are known. If kernel broke pinning of hugepages, then it is an upstream kernel bug. > > My second question is whether the user should be allowed to > mix uio and vfio usage simultaneously. For vfio, the > physical addresses are really DMA addresses and are best > when arbitrarily chosen to appear sequential relative to > their virtual addresses. For uio, they are physical > addresses and are not chosen at all. It seems that these two > things are in conflict and that it will be difficult, ugly, > and maybe impossible to resolve the simultaneous use of > both. Unless application is running as privileged user (ie root), UIO is not going to work. Therefore don't worry about mixed environment.
Re: [dpdk-dev] [PATCH 2/2] vhost: start vhost servers once
On Thu, Dec 29, 2016 at 10:58:11AM -0500, Charles (Chas) Williams wrote: > On 12/29/2016 03:52 AM, Yuanhan Liu wrote: > >On Wed, Dec 28, 2016 at 04:10:52PM -0500, Charles (Chas) Williams wrote: > >>Start a vhost server once during devinit instead of during device start > >>and stop. Some vhost clients, QEMU, don't re-attaching to sockets when > >>the vhost server is stopped and later started. Preserve existing behavior > >>for vhost clients. > > > >I didn't quite get the idea what you are going to fix. > > The issue I am trying to fix is QEMU interaction when DPDK's vhost is > acting as a server to QEMU vhost clients. If you create a vhost server > device, it doesn't create the actual datagram socket until you call > .dev_start(). If you call .dev_stop() is also deletes those sockets. > For QEMU, this is a problem since QEMU doesn't know how to re-attach to > datagram sockets that have gone away. Thanks! And I'd appreciate it if you could have written the commit log this way firstly. > .dev_start()/.dev_stop() seems to roughly means link up and link down > so I understand why you might want to add/remove the datagram sockets. > However, in practice, this doesn't seem to make much sense for a DPDK > vhost server. Agree. > This doesn't seem like the right way to indicate link > status to vhost clients. > > It seems like it would just be easier to do this for both clients and > servers, but I don't know why it was done this way originally so I > choose to keep the client behavior. I don't think there are any differences between DPDK acting as client or server. To me, the right logic seems to be (for both DPDK as server and client). For register, - register the vhost socket at probe stage (either at rte_pmd_vhost_probe or at eth_dev_vhost_create). - start the vhost session right after the register when we haven't started it before. For unregister, - invoke rte_vhost_driver_unregister() at rte_pmd_vhost_remove(). For dev_start/stop, - set allow_queuing to 1/0 for start/stop, respectively. --yliu
[dpdk-dev] [PATCH v4 00/17] net/i40e: consistent filter API
The patch set depends on Adrien's Generic flow API(rte_flow). The patches mainly finish following functions: 1) Store and restore all kinds of filters. 2) Parse all kinds of filters. 3) Add flow validate function. 4) Add flow create function. 5) Add flow destroy function. 6) Add flow flush function. v4 changes: Change I40E_TCI_MASK with 0x to align with testpmd. Modidy the stats show when restoring filters. v3 changes: Set the related cause pointer to a non-NULL value when error happens. Change return value when error happens. Modify filter_del parameter with key. Malloc filter after checking when delete a filter. Delete meaningless initialization. Add return value when there's error. Change global variable definition. Modify some function declaration. v2 changes: Add i40e_flow.c, all flow ops are implemented in the file. Change the whole implementation of all parse flow functions. Update error info for all flow ops. Add flow_list to store flows created. Beilei Xing (17): net/i40e: store ethertype filter net/i40e: store tunnel filter net/i40e: store flow director filter net/i40e: restore ethertype filter net/i40e: restore tunnel filter net/i40e: restore flow director filter net/i40e: add flow validate function net/i40e: parse flow director filter net/i40e: parse tunnel filter net/i40e: add flow create function net/i40e: add flow destroy function net/i40e: destroy ethertype filter net/i40e: destroy tunnel filter net/i40e: destroy flow directory filter net/i40e: add flow flush function net/i40e: flush ethertype filters net/i40e: flush tunnel filters drivers/net/i40e/Makefile |2 + drivers/net/i40e/i40e_ethdev.c | 519 ++-- drivers/net/i40e/i40e_ethdev.h | 173 drivers/net/i40e/i40e_fdir.c | 140 +++- drivers/net/i40e/i40e_flow.c | 1772 5 files changed, 2540 insertions(+), 66 deletions(-) create mode 100644 drivers/net/i40e/i40e_flow.c -- 2.5.5
[dpdk-dev] [PATCH v4 05/17] net/i40e: restore tunnel filter
Add support of restoring tunnel filter.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_ethdev.c | 21 +
1 file changed, 21 insertions(+)
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 189d110..67e1b37 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -475,6 +475,7 @@ static int i40e_sw_tunnel_filter_insert(struct i40e_pf *pf,
struct i40e_tunnel_filter *tunnel_filter);
static void i40e_ethertype_filter_restore(struct i40e_pf *pf);
+static void i40e_tunnel_filter_restore(struct i40e_pf *pf);
static void i40e_filter_restore(struct i40e_pf *pf);
static const struct rte_pci_id pci_id_i40e_map[] = {
@@ -10110,8 +10111,28 @@ i40e_ethertype_filter_restore(struct i40e_pf *pf)
stats.mac_etype_free, stats.etype_free);
}
+/* Restore tunnel filter */
+static void
+i40e_tunnel_filter_restore(struct i40e_pf *pf)
+{
+ struct i40e_hw *hw = I40E_PF_TO_HW(pf);
+ struct i40e_vsi *vsi = pf->main_vsi;
+ struct i40e_tunnel_filter_list
+ *tunnel_list = &pf->tunnel.tunnel_list;
+ struct i40e_tunnel_filter *f;
+ struct i40e_aqc_add_remove_cloud_filters_element_data cld_filter;
+
+ TAILQ_FOREACH(f, tunnel_list, rules) {
+ memset(&cld_filter, 0, sizeof(cld_filter));
+ rte_memcpy(&cld_filter, &f->input, sizeof(f->input));
+ cld_filter.queue_number = f->queue;
+ i40e_aq_add_cloud_filters(hw, vsi->seid, &cld_filter, 1);
+ }
+}
+
static void
i40e_filter_restore(struct i40e_pf *pf)
{
i40e_ethertype_filter_restore(pf);
+ i40e_tunnel_filter_restore(pf);
}
--
2.5.5
[dpdk-dev] [PATCH v4 01/17] net/i40e: store ethertype filter
Currently there's no ethertype filter stored in SW.
This patch stores ethertype filter with cuckoo hash
in SW, also adds protection if an ethertype filter
has been added.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/Makefile | 1 +
drivers/net/i40e/i40e_ethdev.c | 166 -
drivers/net/i40e/i40e_ethdev.h | 31
3 files changed, 197 insertions(+), 1 deletion(-)
diff --git a/drivers/net/i40e/Makefile b/drivers/net/i40e/Makefile
index 66997b6..11175c4 100644
--- a/drivers/net/i40e/Makefile
+++ b/drivers/net/i40e/Makefile
@@ -117,5 +117,6 @@ DEPDIRS-$(CONFIG_RTE_LIBRTE_I40E_PMD) += lib/librte_eal
lib/librte_ether
DEPDIRS-$(CONFIG_RTE_LIBRTE_I40E_PMD) += lib/librte_mempool lib/librte_mbuf
DEPDIRS-$(CONFIG_RTE_LIBRTE_I40E_PMD) += lib/librte_net
DEPDIRS-$(CONFIG_RTE_LIBRTE_I40E_PMD) += lib/librte_kvargs
+DEPDIRS-$(CONFIG_RTE_LIBRTE_I40E_PMD) += lib/librte_hash
include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 8033c35..e43b4d9 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -51,6 +51,7 @@
#include
#include
#include
+#include
#include "i40e_logs.h"
#include "base/i40e_prototype.h"
@@ -461,6 +462,12 @@ static void i40e_set_default_mac_addr(struct rte_eth_dev
*dev,
static int i40e_dev_mtu_set(struct rte_eth_dev *dev, uint16_t mtu);
+static int i40e_ethertype_filter_convert(
+ const struct rte_eth_ethertype_filter *input,
+ struct i40e_ethertype_filter *filter);
+static int i40e_sw_ethertype_filter_insert(struct i40e_pf *pf,
+ struct i40e_ethertype_filter *filter);
+
static const struct rte_pci_id pci_id_i40e_map[] = {
{ RTE_PCI_DEVICE(I40E_INTEL_VENDOR_ID, I40E_DEV_ID_SFP_XL710) },
{ RTE_PCI_DEVICE(I40E_INTEL_VENDOR_ID, I40E_DEV_ID_QEMU) },
@@ -938,9 +945,18 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
int ret;
uint32_t len;
uint8_t aq_fail = 0;
+ struct i40e_ethertype_rule *ethertype_rule = &pf->ethertype;
PMD_INIT_FUNC_TRACE();
+ char ethertype_hash_name[RTE_HASH_NAMESIZE];
+ struct rte_hash_parameters ethertype_hash_params = {
+ .name = ethertype_hash_name,
+ .entries = I40E_MAX_ETHERTYPE_FILTER_NUM,
+ .key_len = sizeof(struct i40e_ethertype_filter_input),
+ .hash_func = rte_hash_crc,
+ };
+
dev->dev_ops = &i40e_eth_dev_ops;
dev->rx_pkt_burst = i40e_recv_pkts;
dev->tx_pkt_burst = i40e_xmit_pkts;
@@ -1180,8 +1196,33 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
pf->flags &= ~I40E_FLAG_DCB;
}
+ /* Initialize ethertype filter rule list and hash */
+ TAILQ_INIT(ðertype_rule->ethertype_list);
+ snprintf(ethertype_hash_name, RTE_HASH_NAMESIZE,
+"ethertype_%s", dev->data->name);
+ ethertype_rule->hash_table = rte_hash_create(ðertype_hash_params);
+ if (!ethertype_rule->hash_table) {
+ PMD_INIT_LOG(ERR, "Failed to create ethertype hash table!");
+ ret = -EINVAL;
+ goto err_ethertype_hash_table_create;
+ }
+ ethertype_rule->hash_map = rte_zmalloc("i40e_ethertype_hash_map",
+ sizeof(struct i40e_ethertype_filter *) *
+ I40E_MAX_ETHERTYPE_FILTER_NUM,
+ 0);
+ if (!ethertype_rule->hash_map) {
+ PMD_INIT_LOG(ERR,
+"Failed to allocate memory for ethertype hash map!");
+ ret = -ENOMEM;
+ goto err_ethertype_hash_map_alloc;
+ }
+
return 0;
+err_ethertype_hash_map_alloc:
+ rte_hash_free(ethertype_rule->hash_table);
+err_ethertype_hash_table_create:
+ rte_free(dev->data->mac_addrs);
err_mac_alloc:
i40e_vsi_release(pf->main_vsi);
err_setup_pf_switch:
@@ -1204,23 +1245,40 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
static int
eth_i40e_dev_uninit(struct rte_eth_dev *dev)
{
+ struct i40e_pf *pf;
struct rte_pci_device *pci_dev;
struct i40e_hw *hw;
struct i40e_filter_control_settings settings;
+ struct i40e_ethertype_filter *p_ethertype;
int ret;
uint8_t aq_fail = 0;
+ struct i40e_ethertype_rule *ethertype_rule;
PMD_INIT_FUNC_TRACE();
if (rte_eal_process_type() != RTE_PROC_PRIMARY)
return 0;
+ pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private);
hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private);
pci_dev = dev->pci_dev;
+ ethertype_rule = &pf->ethertype;
if (hw->adapter_stopped == 0)
i40e_dev_close(dev);
+ /* Remove all ethertype director rules and hash */
+ if (ethertype_rule->hash_map)
+ rte_free(ethertype_rule->hash_map);
+ if (ether
[dpdk-dev] [PATCH v4 03/17] net/i40e: store flow director filter
Currently there's no flow director filter stored in SW. This
patch stores flow director filters in SW with cuckoo hash,
also adds protection if a flow director filter has been added.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_ethdev.c | 48 +++
drivers/net/i40e/i40e_ethdev.h | 14 ++
drivers/net/i40e/i40e_fdir.c | 105 +
3 files changed, 167 insertions(+)
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 2bdb4d6..fb7d794 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -953,6 +953,7 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
uint8_t aq_fail = 0;
struct i40e_ethertype_rule *ethertype_rule = &pf->ethertype;
struct i40e_tunnel_rule *tunnel_rule = &pf->tunnel;
+ struct i40e_fdir_info *fdir_info = &pf->fdir;
PMD_INIT_FUNC_TRACE();
@@ -972,6 +973,14 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
.hash_func = rte_hash_crc,
};
+ char fdir_hash_name[RTE_HASH_NAMESIZE];
+ struct rte_hash_parameters fdir_hash_params = {
+ .name = fdir_hash_name,
+ .entries = I40E_MAX_FDIR_FILTER_NUM,
+ .key_len = sizeof(struct rte_eth_fdir_input),
+ .hash_func = rte_hash_crc,
+ };
+
dev->dev_ops = &i40e_eth_dev_ops;
dev->rx_pkt_burst = i40e_recv_pkts;
dev->tx_pkt_burst = i40e_xmit_pkts;
@@ -1253,8 +1262,33 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
goto err_tunnel_hash_map_alloc;
}
+ /* Initialize flow director filter rule list and hash */
+ TAILQ_INIT(&fdir_info->fdir_list);
+ snprintf(fdir_hash_name, RTE_HASH_NAMESIZE,
+"fdir_%s", dev->data->name);
+ fdir_info->hash_table = rte_hash_create(&fdir_hash_params);
+ if (!fdir_info->hash_table) {
+ PMD_INIT_LOG(ERR, "Failed to create fdir hash table!");
+ ret = -EINVAL;
+ goto err_fdir_hash_table_create;
+ }
+ fdir_info->hash_map = rte_zmalloc("i40e_fdir_hash_map",
+ sizeof(struct i40e_fdir_filter *) *
+ I40E_MAX_FDIR_FILTER_NUM,
+ 0);
+ if (!fdir_info->hash_map) {
+ PMD_INIT_LOG(ERR,
+"Failed to allocate memory for fdir hash map!");
+ ret = -ENOMEM;
+ goto err_fdir_hash_map_alloc;
+ }
+
return 0;
+err_fdir_hash_map_alloc:
+ rte_hash_free(fdir_info->hash_table);
+err_fdir_hash_table_create:
+ rte_free(tunnel_rule->hash_map);
err_tunnel_hash_map_alloc:
rte_hash_free(tunnel_rule->hash_table);
err_tunnel_hash_table_create:
@@ -1291,10 +1325,12 @@ eth_i40e_dev_uninit(struct rte_eth_dev *dev)
struct i40e_filter_control_settings settings;
struct i40e_ethertype_filter *p_ethertype;
struct i40e_tunnel_filter *p_tunnel;
+ struct i40e_fdir_filter *p_fdir;
int ret;
uint8_t aq_fail = 0;
struct i40e_ethertype_rule *ethertype_rule;
struct i40e_tunnel_rule *tunnel_rule;
+ struct i40e_fdir_info *fdir_info;
PMD_INIT_FUNC_TRACE();
@@ -1306,6 +1342,7 @@ eth_i40e_dev_uninit(struct rte_eth_dev *dev)
pci_dev = dev->pci_dev;
ethertype_rule = &pf->ethertype;
tunnel_rule = &pf->tunnel;
+ fdir_info = &pf->fdir;
if (hw->adapter_stopped == 0)
i40e_dev_close(dev);
@@ -1333,6 +1370,17 @@ eth_i40e_dev_uninit(struct rte_eth_dev *dev)
rte_free(p_tunnel);
}
+ /* Remove all flow director rules and hash */
+ if (fdir_info->hash_map)
+ rte_free(fdir_info->hash_map);
+ if (fdir_info->hash_table)
+ rte_hash_free(fdir_info->hash_table);
+
+ while ((p_fdir = TAILQ_FIRST(&fdir_info->fdir_list))) {
+ TAILQ_REMOVE(&fdir_info->fdir_list, p_fdir, rules);
+ rte_free(p_fdir);
+ }
+
dev->dev_ops = NULL;
dev->rx_pkt_burst = NULL;
dev->tx_pkt_burst = NULL;
diff --git a/drivers/net/i40e/i40e_ethdev.h b/drivers/net/i40e/i40e_ethdev.h
index 83f3594..b79fbd6 100644
--- a/drivers/net/i40e/i40e_ethdev.h
+++ b/drivers/net/i40e/i40e_ethdev.h
@@ -377,6 +377,14 @@ struct i40e_fdir_flex_mask {
};
#define I40E_FILTER_PCTYPE_MAX 64
+#define I40E_MAX_FDIR_FILTER_NUM (1024 * 8)
+
+struct i40e_fdir_filter {
+ TAILQ_ENTRY(i40e_fdir_filter) rules;
+ struct rte_eth_fdir_filter fdir;
+};
+
+TAILQ_HEAD(i40e_fdir_filter_list, i40e_fdir_filter);
/*
* A structure used to define fields of a FDIR related info.
*/
@@ -395,6 +403,10 @@ struct i40e_fdir_info {
*/
struct i40e_fdir_flex_pit flex_set[I40E_MAX_FLXPLD_LAYER *
I40E_MAX_FLXPLD_FIED];
struct i40e_fdir_flex_mask flex_mask[I40E_FIL
[dpdk-dev] [PATCH v4 02/17] net/i40e: store tunnel filter
Currently there's no tunnel filter stored in SW.
This patch stores tunnel filter in SW with cuckoo
hash, also adds protection if a tunnel filter has
been added.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_ethdev.c | 169 -
drivers/net/i40e/i40e_ethdev.h | 32
2 files changed, 198 insertions(+), 3 deletions(-)
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index e43b4d9..2bdb4d6 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -468,6 +468,12 @@ static int i40e_ethertype_filter_convert(
static int i40e_sw_ethertype_filter_insert(struct i40e_pf *pf,
struct i40e_ethertype_filter *filter);
+static int i40e_tunnel_filter_convert(
+ struct i40e_aqc_add_remove_cloud_filters_element_data *cld_filter,
+ struct i40e_tunnel_filter *tunnel_filter);
+static int i40e_sw_tunnel_filter_insert(struct i40e_pf *pf,
+ struct i40e_tunnel_filter *tunnel_filter);
+
static const struct rte_pci_id pci_id_i40e_map[] = {
{ RTE_PCI_DEVICE(I40E_INTEL_VENDOR_ID, I40E_DEV_ID_SFP_XL710) },
{ RTE_PCI_DEVICE(I40E_INTEL_VENDOR_ID, I40E_DEV_ID_QEMU) },
@@ -946,6 +952,7 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
uint32_t len;
uint8_t aq_fail = 0;
struct i40e_ethertype_rule *ethertype_rule = &pf->ethertype;
+ struct i40e_tunnel_rule *tunnel_rule = &pf->tunnel;
PMD_INIT_FUNC_TRACE();
@@ -957,6 +964,14 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
.hash_func = rte_hash_crc,
};
+ char tunnel_hash_name[RTE_HASH_NAMESIZE];
+ struct rte_hash_parameters tunnel_hash_params = {
+ .name = tunnel_hash_name,
+ .entries = I40E_MAX_TUNNEL_FILTER_NUM,
+ .key_len = sizeof(struct i40e_tunnel_filter_input),
+ .hash_func = rte_hash_crc,
+ };
+
dev->dev_ops = &i40e_eth_dev_ops;
dev->rx_pkt_burst = i40e_recv_pkts;
dev->tx_pkt_burst = i40e_xmit_pkts;
@@ -1217,8 +1232,33 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
goto err_ethertype_hash_map_alloc;
}
+ /* Initialize tunnel filter rule list and hash */
+ TAILQ_INIT(&tunnel_rule->tunnel_list);
+ snprintf(tunnel_hash_name, RTE_HASH_NAMESIZE,
+"tunnel_%s", dev->data->name);
+ tunnel_rule->hash_table = rte_hash_create(&tunnel_hash_params);
+ if (!tunnel_rule->hash_table) {
+ PMD_INIT_LOG(ERR, "Failed to create tunnel hash table!");
+ ret = -EINVAL;
+ goto err_tunnel_hash_table_create;
+ }
+ tunnel_rule->hash_map = rte_zmalloc("i40e_tunnel_hash_map",
+ sizeof(struct i40e_tunnel_filter *) *
+ I40E_MAX_TUNNEL_FILTER_NUM,
+ 0);
+ if (!tunnel_rule->hash_map) {
+ PMD_INIT_LOG(ERR,
+"Failed to allocate memory for tunnel hash map!");
+ ret = -ENOMEM;
+ goto err_tunnel_hash_map_alloc;
+ }
+
return 0;
+err_tunnel_hash_map_alloc:
+ rte_hash_free(tunnel_rule->hash_table);
+err_tunnel_hash_table_create:
+ rte_free(ethertype_rule->hash_map);
err_ethertype_hash_map_alloc:
rte_hash_free(ethertype_rule->hash_table);
err_ethertype_hash_table_create:
@@ -1250,9 +1290,11 @@ eth_i40e_dev_uninit(struct rte_eth_dev *dev)
struct i40e_hw *hw;
struct i40e_filter_control_settings settings;
struct i40e_ethertype_filter *p_ethertype;
+ struct i40e_tunnel_filter *p_tunnel;
int ret;
uint8_t aq_fail = 0;
struct i40e_ethertype_rule *ethertype_rule;
+ struct i40e_tunnel_rule *tunnel_rule;
PMD_INIT_FUNC_TRACE();
@@ -1263,6 +1305,7 @@ eth_i40e_dev_uninit(struct rte_eth_dev *dev)
hw = I40E_DEV_PRIVATE_TO_HW(dev->data->dev_private);
pci_dev = dev->pci_dev;
ethertype_rule = &pf->ethertype;
+ tunnel_rule = &pf->tunnel;
if (hw->adapter_stopped == 0)
i40e_dev_close(dev);
@@ -1279,6 +1322,17 @@ eth_i40e_dev_uninit(struct rte_eth_dev *dev)
rte_free(p_ethertype);
}
+ /* Remove all tunnel director rules and hash */
+ if (tunnel_rule->hash_map)
+ rte_free(tunnel_rule->hash_map);
+ if (tunnel_rule->hash_table)
+ rte_hash_free(tunnel_rule->hash_table);
+
+ while ((p_tunnel = TAILQ_FIRST(&tunnel_rule->tunnel_list))) {
+ TAILQ_REMOVE(&tunnel_rule->tunnel_list, p_tunnel, rules);
+ rte_free(p_tunnel);
+ }
+
dev->dev_ops = NULL;
dev->rx_pkt_burst = NULL;
dev->tx_pkt_burst = NULL;
@@ -6478,6 +6532,85 @@ i40e_dev_get_filter_type(uint16_t filter_type, uint16_t
*flag)
return 0;
}
+/*
[dpdk-dev] [PATCH v4 06/17] net/i40e: restore flow director filter
Add support of restoring flow director filter.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_ethdev.c | 1 +
drivers/net/i40e/i40e_ethdev.h | 1 +
drivers/net/i40e/i40e_fdir.c | 31 +++
3 files changed, 33 insertions(+)
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 67e1b37..153322a 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -10135,4 +10135,5 @@ i40e_filter_restore(struct i40e_pf *pf)
{
i40e_ethertype_filter_restore(pf);
i40e_tunnel_filter_restore(pf);
+ i40e_fdir_filter_restore(pf);
}
diff --git a/drivers/net/i40e/i40e_ethdev.h b/drivers/net/i40e/i40e_ethdev.h
index b79fbd6..92f6f55 100644
--- a/drivers/net/i40e/i40e_ethdev.h
+++ b/drivers/net/i40e/i40e_ethdev.h
@@ -670,6 +670,7 @@ int i40e_fdir_ctrl_func(struct rte_eth_dev *dev,
int i40e_select_filter_input_set(struct i40e_hw *hw,
struct rte_eth_input_set_conf *conf,
enum rte_filter_type filter);
+void i40e_fdir_filter_restore(struct i40e_pf *pf);
int i40e_hash_filter_inset_select(struct i40e_hw *hw,
struct rte_eth_input_set_conf *conf);
int i40e_fdir_filter_inset_select(struct i40e_pf *pf,
diff --git a/drivers/net/i40e/i40e_fdir.c b/drivers/net/i40e/i40e_fdir.c
index 4a29b37..f89dbc9 100644
--- a/drivers/net/i40e/i40e_fdir.c
+++ b/drivers/net/i40e/i40e_fdir.c
@@ -1586,3 +1586,34 @@ i40e_fdir_ctrl_func(struct rte_eth_dev *dev,
}
return ret;
}
+
+/* Restore flow director filter */
+void
+i40e_fdir_filter_restore(struct i40e_pf *pf)
+{
+ struct rte_eth_dev *dev = I40E_VSI_TO_ETH_DEV(pf->main_vsi);
+ struct i40e_fdir_filter_list *fdir_list = &pf->fdir.fdir_list;
+ struct i40e_fdir_filter *f;
+#ifdef RTE_LIBRTE_I40E_DEBUG_DRIVER
+ struct i40e_hw *hw = I40E_PF_TO_HW(pf);
+ uint32_t fdstat;
+ uint32_t guarant_cnt; /**< Number of filters in guaranteed spaces. */
+ uint32_t best_cnt; /**< Number of filters in best effort spaces. */
+#endif /* RTE_LIBRTE_I40E_DEBUG_DRIVER */
+
+ TAILQ_FOREACH(f, fdir_list, rules)
+ i40e_add_del_fdir_filter(dev, &f->fdir, TRUE);
+
+#ifdef RTE_LIBRTE_I40E_DEBUG_DRIVER
+ fdstat = I40E_READ_REG(hw, I40E_PFQF_FDSTAT);
+ guarant_cnt =
+ (uint32_t)((fdstat & I40E_PFQF_FDSTAT_GUARANT_CNT_MASK) >>
+ I40E_PFQF_FDSTAT_GUARANT_CNT_SHIFT);
+ best_cnt =
+ (uint32_t)((fdstat & I40E_PFQF_FDSTAT_BEST_CNT_MASK) >>
+ I40E_PFQF_FDSTAT_BEST_CNT_SHIFT);
+#endif /* RTE_LIBRTE_I40E_DEBUG_DRIVER */
+
+ PMD_DRV_LOG(INFO, "FDIR: Guarant count: %d, Best count: %d\n",
+ guarant_cnt, best_cnt);
+}
--
2.5.5
[dpdk-dev] [PATCH v4 04/17] net/i40e: restore ethertype filter
Add support of restoring ethertype filter in case filter
dropped accidentally, as all filters need to be added and
removed by user obviously for generic filter API.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_ethdev.c | 44 ++
1 file changed, 44 insertions(+)
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index fb7d794..189d110 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -474,6 +474,9 @@ static int i40e_tunnel_filter_convert(
static int i40e_sw_tunnel_filter_insert(struct i40e_pf *pf,
struct i40e_tunnel_filter *tunnel_filter);
+static void i40e_ethertype_filter_restore(struct i40e_pf *pf);
+static void i40e_filter_restore(struct i40e_pf *pf);
+
static const struct rte_pci_id pci_id_i40e_map[] = {
{ RTE_PCI_DEVICE(I40E_INTEL_VENDOR_ID, I40E_DEV_ID_SFP_XL710) },
{ RTE_PCI_DEVICE(I40E_INTEL_VENDOR_ID, I40E_DEV_ID_QEMU) },
@@ -1955,6 +1958,8 @@ i40e_dev_start(struct rte_eth_dev *dev)
/* enable uio intr after callback register */
rte_intr_enable(intr_handle);
+ i40e_filter_restore(pf);
+
return I40E_SUCCESS;
err_up:
@@ -10071,3 +10076,42 @@ i40e_dev_mtu_set(struct rte_eth_dev *dev, uint16_t mtu)
return ret;
}
+
+/* Restore ethertype filter */
+static void
+i40e_ethertype_filter_restore(struct i40e_pf *pf)
+{
+ struct i40e_hw *hw = I40E_PF_TO_HW(pf);
+ struct i40e_ethertype_filter_list
+ *ethertype_list = &pf->ethertype.ethertype_list;
+ struct i40e_ethertype_filter *f;
+ struct i40e_control_filter_stats stats;
+ uint16_t flags;
+
+ TAILQ_FOREACH(f, ethertype_list, rules) {
+ flags = 0;
+ if (!(f->flags & RTE_ETHTYPE_FLAGS_MAC))
+ flags |= I40E_AQC_ADD_CONTROL_PACKET_FLAGS_IGNORE_MAC;
+ if (f->flags & RTE_ETHTYPE_FLAGS_DROP)
+ flags |= I40E_AQC_ADD_CONTROL_PACKET_FLAGS_DROP;
+ flags |= I40E_AQC_ADD_CONTROL_PACKET_FLAGS_TO_QUEUE;
+
+ memset(&stats, 0, sizeof(stats));
+ i40e_aq_add_rem_control_packet_filter(hw,
+ f->input.mac_addr.addr_bytes,
+ f->input.ether_type,
+ flags, pf->main_vsi->seid,
+ f->queue, 1, &stats, NULL);
+ }
+ PMD_DRV_LOG(INFO, "Ethertype filter:"
+ " mac_etype_used = %u, etype_used = %u,"
+ " mac_etype_free = %u, etype_free = %u\n",
+ stats.mac_etype_used, stats.etype_used,
+ stats.mac_etype_free, stats.etype_free);
+}
+
+static void
+i40e_filter_restore(struct i40e_pf *pf)
+{
+ i40e_ethertype_filter_restore(pf);
+}
--
2.5.5
[dpdk-dev] [PATCH v4 07/17] net/i40e: add flow validate function
This patch adds i40e_flow_validation function to check if
a flow is valid according to the flow pattern.
i40e_parse_ethertype_filter is added first, it also gets
the ethertype info.
i40e_flow.c is added to handle all generic filter events.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/Makefile | 1 +
drivers/net/i40e/i40e_ethdev.c | 7 +
drivers/net/i40e/i40e_ethdev.h | 18 ++
drivers/net/i40e/i40e_flow.c | 447 +
4 files changed, 473 insertions(+)
create mode 100644 drivers/net/i40e/i40e_flow.c
diff --git a/drivers/net/i40e/Makefile b/drivers/net/i40e/Makefile
index 11175c4..89bd85a 100644
--- a/drivers/net/i40e/Makefile
+++ b/drivers/net/i40e/Makefile
@@ -105,6 +105,7 @@ endif
SRCS-$(CONFIG_RTE_LIBRTE_I40E_PMD) += i40e_ethdev_vf.c
SRCS-$(CONFIG_RTE_LIBRTE_I40E_PMD) += i40e_pf.c
SRCS-$(CONFIG_RTE_LIBRTE_I40E_PMD) += i40e_fdir.c
+SRCS-$(CONFIG_RTE_LIBRTE_I40E_PMD) += i40e_flow.c
# vector PMD driver needs SSE4.1 support
ifeq ($(findstring RTE_MACHINE_CPUFLAG_SSE4_1,$(CFLAGS)),)
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 153322a..edfd52b 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -8426,6 +8426,8 @@ i40e_ethertype_filter_handle(struct rte_eth_dev *dev,
return ret;
}
+const struct rte_flow_ops i40e_flow_ops;
+
static int
i40e_dev_filter_ctrl(struct rte_eth_dev *dev,
enum rte_filter_type filter_type,
@@ -8457,6 +8459,11 @@ i40e_dev_filter_ctrl(struct rte_eth_dev *dev,
case RTE_ETH_FILTER_FDIR:
ret = i40e_fdir_ctrl_func(dev, filter_op, arg);
break;
+ case RTE_ETH_FILTER_GENERIC:
+ if (filter_op != RTE_ETH_FILTER_GET)
+ return -EINVAL;
+ *(const void **)arg = &i40e_flow_ops;
+ break;
default:
PMD_DRV_LOG(WARNING, "Filter type (%d) not supported",
filter_type);
diff --git a/drivers/net/i40e/i40e_ethdev.h b/drivers/net/i40e/i40e_ethdev.h
index 92f6f55..23f360b 100644
--- a/drivers/net/i40e/i40e_ethdev.h
+++ b/drivers/net/i40e/i40e_ethdev.h
@@ -38,6 +38,7 @@
#include
#include
#include
+#include
#define I40E_VLAN_TAG_SIZE4
@@ -629,6 +630,23 @@ struct i40e_adapter {
struct rte_timecounter tx_tstamp_tc;
};
+union i40e_filter_t {
+ struct rte_eth_ethertype_filter ethertype_filter;
+ struct rte_eth_fdir_filter fdir_filter;
+ struct rte_eth_tunnel_filter_conf tunnel_filter;
+};
+
+typedef int (*parse_filter_t)(struct rte_eth_dev *dev,
+ const struct rte_flow_attr *attr,
+ const struct rte_flow_item pattern[],
+ const struct rte_flow_action actions[],
+ struct rte_flow_error *error,
+ union i40e_filter_t *filter);
+struct i40e_valid_pattern {
+ enum rte_flow_item_type *items;
+ parse_filter_t parse_filter;
+};
+
int i40e_dev_switch_queues(struct i40e_pf *pf, bool on);
int i40e_vsi_release(struct i40e_vsi *vsi);
struct i40e_vsi *i40e_vsi_setup(struct i40e_pf *pf,
diff --git a/drivers/net/i40e/i40e_flow.c b/drivers/net/i40e/i40e_flow.c
new file mode 100644
index 000..a9ff73f
--- /dev/null
+++ b/drivers/net/i40e/i40e_flow.c
@@ -0,0 +1,447 @@
+/*-
+ * BSD LICENSE
+ *
+ * Copyright (c) 2016 Intel Corporation. All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * are met:
+ *
+ * * Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in
+ * the documentation and/or other materials provided with the
+ * distribution.
+ * * Neither the name of Intel Corporation nor the names of its
+ * contributors may be used to endorse or promote products derived
+ * from this software without specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ * A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ * OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ * (I
[dpdk-dev] [PATCH v4 08/17] net/i40e: parse flow director filter
This patch adds i40e_parse_fdir_filter to check if a rule
is a flow director rule according to the flow pattern,
and the function also gets the flow director info.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_ethdev.c | 56 +---
drivers/net/i40e/i40e_ethdev.h | 55
drivers/net/i40e/i40e_flow.c | 607 +
3 files changed, 663 insertions(+), 55 deletions(-)
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index edfd52b..bcf28cf 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -139,60 +139,6 @@
#define I40E_DEFAULT_DCB_APP_NUM1
#define I40E_DEFAULT_DCB_APP_PRIO 3
-#define I40E_INSET_NONE0x0ULL
-
-/* bit0 ~ bit 7 */
-#define I40E_INSET_DMAC0x0001ULL
-#define I40E_INSET_SMAC0x0002ULL
-#define I40E_INSET_VLAN_OUTER 0x0004ULL
-#define I40E_INSET_VLAN_INNER 0x0008ULL
-#define I40E_INSET_VLAN_TUNNEL 0x0010ULL
-
-/* bit 8 ~ bit 15 */
-#define I40E_INSET_IPV4_SRC0x0100ULL
-#define I40E_INSET_IPV4_DST0x0200ULL
-#define I40E_INSET_IPV6_SRC0x0400ULL
-#define I40E_INSET_IPV6_DST0x0800ULL
-#define I40E_INSET_SRC_PORT0x1000ULL
-#define I40E_INSET_DST_PORT0x2000ULL
-#define I40E_INSET_SCTP_VT 0x4000ULL
-
-/* bit 16 ~ bit 31 */
-#define I40E_INSET_IPV4_TOS0x0001ULL
-#define I40E_INSET_IPV4_PROTO 0x0002ULL
-#define I40E_INSET_IPV4_TTL0x0004ULL
-#define I40E_INSET_IPV6_TC 0x0008ULL
-#define I40E_INSET_IPV6_FLOW 0x0010ULL
-#define I40E_INSET_IPV6_NEXT_HDR 0x0020ULL
-#define I40E_INSET_IPV6_HOP_LIMIT 0x0040ULL
-#define I40E_INSET_TCP_FLAGS 0x0080ULL
-
-/* bit 32 ~ bit 47, tunnel fields */
-#define I40E_INSET_TUNNEL_IPV4_DST 0x0001ULL
-#define I40E_INSET_TUNNEL_IPV6_DST 0x0002ULL
-#define I40E_INSET_TUNNEL_DMAC 0x0004ULL
-#define I40E_INSET_TUNNEL_SRC_PORT 0x0008ULL
-#define I40E_INSET_TUNNEL_DST_PORT 0x0010ULL
-#define I40E_INSET_TUNNEL_ID 0x0020ULL
-
-/* bit 48 ~ bit 55 */
-#define I40E_INSET_LAST_ETHER_TYPE 0x0001ULL
-
-/* bit 56 ~ bit 63, Flex Payload */
-#define I40E_INSET_FLEX_PAYLOAD_W1 0x0100ULL
-#define I40E_INSET_FLEX_PAYLOAD_W2 0x0200ULL
-#define I40E_INSET_FLEX_PAYLOAD_W3 0x0400ULL
-#define I40E_INSET_FLEX_PAYLOAD_W4 0x0800ULL
-#define I40E_INSET_FLEX_PAYLOAD_W5 0x1000ULL
-#define I40E_INSET_FLEX_PAYLOAD_W6 0x2000ULL
-#define I40E_INSET_FLEX_PAYLOAD_W7 0x4000ULL
-#define I40E_INSET_FLEX_PAYLOAD_W8 0x8000ULL
-#define I40E_INSET_FLEX_PAYLOAD \
- (I40E_INSET_FLEX_PAYLOAD_W1 | I40E_INSET_FLEX_PAYLOAD_W2 | \
- I40E_INSET_FLEX_PAYLOAD_W3 | I40E_INSET_FLEX_PAYLOAD_W4 | \
- I40E_INSET_FLEX_PAYLOAD_W5 | I40E_INSET_FLEX_PAYLOAD_W6 | \
- I40E_INSET_FLEX_PAYLOAD_W7 | I40E_INSET_FLEX_PAYLOAD_W8)
-
/**
* Below are values for writing un-exposed registers suggested
* by silicon experts
@@ -7617,7 +7563,7 @@ i40e_validate_input_set(enum i40e_filter_pctype pctype,
}
/* default input set fields combination per pctype */
-static uint64_t
+uint64_t
i40e_get_default_input_set(uint16_t pctype)
{
static const uint64_t default_inset_table[] = {
diff --git a/drivers/net/i40e/i40e_ethdev.h b/drivers/net/i40e/i40e_ethdev.h
index 23f360b..9e3a48d 100644
--- a/drivers/net/i40e/i40e_ethdev.h
+++ b/drivers/net/i40e/i40e_ethdev.h
@@ -190,6 +190,60 @@ enum i40e_flxpld_layer_idx {
#define FLOATING_VEB_SUPPORTED_FW_MAJ 5
#define FLOATING_VEB_SUPPORTED_FW_MIN 0
+#define I40E_INSET_NONE0x0ULL
+
+/* bit0 ~ bit 7 */
+#define I40E_INSET_DMAC0x0001ULL
+#define I40E_INSET_SMAC0x0002ULL
+#define I40E_INSET_VLAN_OUTER 0x0004ULL
+#define I40E_INSET_VLAN_INNER 0x0008ULL
+#define I40E_INSET_VLAN_TUNNEL 0x0010ULL
+
+/* bit 8 ~ bit 15 */
+#define I40E_INSET_IPV4_SRC0x0100ULL
+#define I40E_INSET_IPV4_DST0x0200ULL
+#define I40E_INSET_IPV6_SRC0x0400ULL
+#define I40E_INSET_IPV6_DST0x0800ULL
+#define I40E_INSET_SRC_PORT0x1000ULL
+#define I40E_INSET_DST_PORT0x2000ULL
+#define I40E_INSET_SCTP_VT 0x4000ULL
+
+/* bit 16 ~ bit 31 */
+#define I40E_INSET_IPV4_TOS0x0001ULL
+#define I40E_INSET_IPV4_PROTO 0x0002ULL
+#define I40E_INSET_IPV4_TTL0x0004ULL
+#define I40E_INS
[dpdk-dev] [PATCH v4 09/17] net/i40e: parse tunnel filter
This patch adds i40e_parse_tunnel_filter to check if
a rule is a tunnel rule according to items of the flow
pattern, and the function also gets the tunnel info.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_flow.c | 394 +++
1 file changed, 394 insertions(+)
diff --git a/drivers/net/i40e/i40e_flow.c b/drivers/net/i40e/i40e_flow.c
index 7b872c9..063f8e2 100644
--- a/drivers/net/i40e/i40e_flow.c
+++ b/drivers/net/i40e/i40e_flow.c
@@ -54,6 +54,8 @@
#define I40E_IPV4_TC_SHIFT 4
#define I40E_IPV6_TC_MASK (0x00FF << I40E_IPV4_TC_SHIFT)
#define I40E_IPV6_FRAG_HEADER 44
+#define I40E_TENANT_ARRAY_NUM 3
+#define I40E_TCI_MASK 0x
static int i40e_flow_validate(struct rte_eth_dev *dev,
const struct rte_flow_attr *attr,
@@ -76,6 +78,14 @@ static int i40e_parse_fdir_act(struct rte_eth_dev *dev,
const struct rte_flow_action *actions,
struct rte_flow_error *error,
struct rte_eth_fdir_filter *filter);
+static int i40e_parse_tunnel_pattern(__rte_unused struct rte_eth_dev *dev,
+const struct rte_flow_item *pattern,
+struct rte_flow_error *error,
+struct rte_eth_tunnel_filter_conf *filter);
+static int i40e_parse_tunnel_act(struct rte_eth_dev *dev,
+const struct rte_flow_action *actions,
+struct rte_flow_error *error,
+struct rte_eth_tunnel_filter_conf *filter);
static int i40e_parse_attr(const struct rte_flow_attr *attr,
struct rte_flow_error *error);
@@ -192,6 +202,45 @@ static enum rte_flow_item_type
pattern_fdir_ipv6_sctp_ext[] = {
RTE_FLOW_ITEM_TYPE_END,
};
+/* Pattern matched tunnel filter */
+static enum rte_flow_item_type pattern_vxlan_1[] = {
+ RTE_FLOW_ITEM_TYPE_ETH,
+ RTE_FLOW_ITEM_TYPE_IPV4,
+ RTE_FLOW_ITEM_TYPE_UDP,
+ RTE_FLOW_ITEM_TYPE_VXLAN,
+ RTE_FLOW_ITEM_TYPE_ETH,
+ RTE_FLOW_ITEM_TYPE_END,
+};
+
+static enum rte_flow_item_type pattern_vxlan_2[] = {
+ RTE_FLOW_ITEM_TYPE_ETH,
+ RTE_FLOW_ITEM_TYPE_IPV6,
+ RTE_FLOW_ITEM_TYPE_UDP,
+ RTE_FLOW_ITEM_TYPE_VXLAN,
+ RTE_FLOW_ITEM_TYPE_ETH,
+ RTE_FLOW_ITEM_TYPE_END,
+};
+
+static enum rte_flow_item_type pattern_vxlan_3[] = {
+ RTE_FLOW_ITEM_TYPE_ETH,
+ RTE_FLOW_ITEM_TYPE_IPV4,
+ RTE_FLOW_ITEM_TYPE_UDP,
+ RTE_FLOW_ITEM_TYPE_VXLAN,
+ RTE_FLOW_ITEM_TYPE_ETH,
+ RTE_FLOW_ITEM_TYPE_VLAN,
+ RTE_FLOW_ITEM_TYPE_END,
+};
+
+static enum rte_flow_item_type pattern_vxlan_4[] = {
+ RTE_FLOW_ITEM_TYPE_ETH,
+ RTE_FLOW_ITEM_TYPE_IPV6,
+ RTE_FLOW_ITEM_TYPE_UDP,
+ RTE_FLOW_ITEM_TYPE_VXLAN,
+ RTE_FLOW_ITEM_TYPE_ETH,
+ RTE_FLOW_ITEM_TYPE_VLAN,
+ RTE_FLOW_ITEM_TYPE_END,
+};
+
static int
i40e_parse_ethertype_filter(struct rte_eth_dev *dev,
const struct rte_flow_attr *attr,
@@ -257,6 +306,33 @@ i40e_parse_fdir_filter(struct rte_eth_dev *dev,
return 0;
}
+static int
+i40e_parse_tunnel_filter(struct rte_eth_dev *dev,
+const struct rte_flow_attr *attr,
+const struct rte_flow_item pattern[],
+const struct rte_flow_action actions[],
+struct rte_flow_error *error,
+union i40e_filter_t *filter)
+{
+ struct rte_eth_tunnel_filter_conf *tunnel_filter =
+ &filter->tunnel_filter;
+ int ret;
+
+ ret = i40e_parse_tunnel_pattern(dev, pattern, error, tunnel_filter);
+ if (ret)
+ return ret;
+
+ ret = i40e_parse_tunnel_act(dev, actions, error, tunnel_filter);
+ if (ret)
+ return ret;
+
+ ret = i40e_parse_attr(attr, error);
+ if (ret)
+ return ret;
+
+ return ret;
+}
+
static struct i40e_valid_pattern i40e_supported_patterns[] = {
/* Ethertype */
{ pattern_ethertype, i40e_parse_ethertype_filter },
@@ -277,6 +353,11 @@ static struct i40e_valid_pattern i40e_supported_patterns[]
= {
{ pattern_fdir_ipv6_tcp_ext, i40e_parse_fdir_filter },
{ pattern_fdir_ipv6_sctp, i40e_parse_fdir_filter },
{ pattern_fdir_ipv6_sctp_ext, i40e_parse_fdir_filter },
+ /* tunnel */
+ { pattern_vxlan_1, i40e_parse_tunnel_filter },
+ { pattern_vxlan_2, i40e_parse_tunnel_filter },
+ { pattern_vxlan_3, i40e_parse_tunnel_filter },
+ { pattern_vxlan_4, i40e_parse_tunnel_filter },
};
#define NEXT_ITEM_OF_ACTION(act, actions, index)\
@@ -991,6 +1072,319 @@ i40e_parse_fdir_act(struct rte_eth_dev *dev,
return 0;
}
+/* Parse to get the action info of a tunnle filter */
+static int
[dpdk-dev] [PATCH v4 13/17] net/i40e: destroy tunnel filter
This patch adds i40e_dev_destroy_tunnel_filter function
to destroy a tunnel filter for users.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_flow.c | 41 +
1 file changed, 41 insertions(+)
diff --git a/drivers/net/i40e/i40e_flow.c b/drivers/net/i40e/i40e_flow.c
index 8d1e2ff..9fc311d 100644
--- a/drivers/net/i40e/i40e_flow.c
+++ b/drivers/net/i40e/i40e_flow.c
@@ -99,6 +99,8 @@ static int i40e_parse_attr(const struct rte_flow_attr *attr,
struct rte_flow_error *error);
static int i40e_dev_destroy_ethertype_filter(struct i40e_pf *pf,
struct i40e_ethertype_filter *filter);
+static int i40e_dev_destroy_tunnel_filter(struct i40e_pf *pf,
+ struct i40e_tunnel_filter *filter);
const struct rte_flow_ops i40e_flow_ops = {
.validate = i40e_flow_validate,
@@ -1546,6 +1548,10 @@ i40e_flow_destroy(struct rte_eth_dev *dev,
ret = i40e_dev_destroy_ethertype_filter(pf,
(struct i40e_ethertype_filter *)pmd_flow->rule);
break;
+ case RTE_ETH_FILTER_TUNNEL:
+ ret = i40e_dev_destroy_tunnel_filter(pf,
+(struct i40e_tunnel_filter *)pmd_flow->rule);
+ break;
default:
PMD_DRV_LOG(WARNING, "Filter type (%d) not supported",
filter_type);
@@ -1598,3 +1604,38 @@ i40e_dev_destroy_ethertype_filter(struct i40e_pf *pf,
return ret;
}
+
+static int
+i40e_dev_destroy_tunnel_filter(struct i40e_pf *pf,
+ struct i40e_tunnel_filter *filter)
+{
+ struct i40e_hw *hw = I40E_PF_TO_HW(pf);
+ struct i40e_vsi *vsi = pf->main_vsi;
+ struct i40e_aqc_add_remove_cloud_filters_element_data cld_filter;
+ struct i40e_tunnel_rule *tunnel_rule = &pf->tunnel;
+ struct i40e_tunnel_filter *node;
+ int ret = 0;
+
+ memset(&cld_filter, 0, sizeof(cld_filter));
+ ether_addr_copy((struct ether_addr *)&filter->input.outer_mac,
+ (struct ether_addr *)&cld_filter.outer_mac);
+ ether_addr_copy((struct ether_addr *)&filter->input.inner_mac,
+ (struct ether_addr *)&cld_filter.inner_mac);
+ cld_filter.inner_vlan = filter->input.inner_vlan;
+ cld_filter.flags = filter->input.flags;
+ cld_filter.tenant_id = filter->input.tenant_id;
+ cld_filter.queue_number = filter->queue;
+
+ ret = i40e_aq_remove_cloud_filters(hw, vsi->seid,
+ &cld_filter, 1);
+ if (ret < 0)
+ return ret;
+
+ node = i40e_sw_tunnel_filter_lookup(tunnel_rule, &filter->input);
+ if (!node)
+ return -EINVAL;
+
+ ret = i40e_sw_tunnel_filter_del(pf, &node->input);
+
+ return ret;
+}
--
2.5.5
[dpdk-dev] [PATCH v4 10/17] net/i40e: add flow create function
This patch adds i40e_flow_create function to create a
rule. It will check if a flow matches ethertype filter
or flow director filter or tunnel filter, if the flow
matches some kind of filter, then set the filter to HW.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_ethdev.c | 9 +++--
drivers/net/i40e/i40e_ethdev.h | 21
drivers/net/i40e/i40e_fdir.c | 2 +-
drivers/net/i40e/i40e_flow.c | 77 ++
4 files changed, 103 insertions(+), 6 deletions(-)
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index bcf28cf..e49045a 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -353,9 +353,6 @@ static int i40e_dev_udp_tunnel_port_add(struct rte_eth_dev
*dev,
static int i40e_dev_udp_tunnel_port_del(struct rte_eth_dev *dev,
struct rte_eth_udp_tunnel *udp_tunnel);
static void i40e_filter_input_set_init(struct i40e_pf *pf);
-static int i40e_ethertype_filter_set(struct i40e_pf *pf,
- struct rte_eth_ethertype_filter *filter,
- bool add);
static int i40e_ethertype_filter_handle(struct rte_eth_dev *dev,
enum rte_filter_op filter_op,
void *arg);
@@ -1233,6 +1230,8 @@ eth_i40e_dev_init(struct rte_eth_dev *dev)
goto err_fdir_hash_map_alloc;
}
+ TAILQ_INIT(&pf->flow_list);
+
return 0;
err_fdir_hash_map_alloc:
@@ -6611,7 +6610,7 @@ i40e_sw_tunnel_filter_del(struct i40e_pf *pf,
return 0;
}
-static int
+int
i40e_dev_tunnel_filter_set(struct i40e_pf *pf,
struct rte_eth_tunnel_filter_conf *tunnel_filter,
uint8_t add)
@@ -8256,7 +8255,7 @@ i40e_sw_ethertype_filter_del(struct i40e_pf *pf,
* Configure ethertype filter, which can director packet by filtering
* with mac address and ether_type or only ether_type
*/
-static int
+int
i40e_ethertype_filter_set(struct i40e_pf *pf,
struct rte_eth_ethertype_filter *filter,
bool add)
diff --git a/drivers/net/i40e/i40e_ethdev.h b/drivers/net/i40e/i40e_ethdev.h
index 9e3a48d..b33910d 100644
--- a/drivers/net/i40e/i40e_ethdev.h
+++ b/drivers/net/i40e/i40e_ethdev.h
@@ -536,6 +536,17 @@ struct i40e_mirror_rule {
TAILQ_HEAD(i40e_mirror_rule_list, i40e_mirror_rule);
/*
+ * Struct to store flow created.
+ */
+struct i40e_flow {
+ TAILQ_ENTRY(i40e_flow) node;
+ enum rte_filter_type filter_type;
+ void *rule;
+};
+
+TAILQ_HEAD(i40e_flow_list, i40e_flow);
+
+/*
* Structure to store private data specific for PF instance.
*/
struct i40e_pf {
@@ -592,6 +603,7 @@ struct i40e_pf {
bool floating_veb; /* The flag to use the floating VEB */
/* The floating enable flag for the specific VF */
bool floating_veb_list[I40E_MAX_VF];
+ struct i40e_flow_list flow_list;
};
enum pending_msg {
@@ -767,6 +779,15 @@ i40e_sw_tunnel_filter_lookup(struct i40e_tunnel_rule
*tunnel_rule,
int i40e_sw_tunnel_filter_del(struct i40e_pf *pf,
struct i40e_tunnel_filter_input *input);
uint64_t i40e_get_default_input_set(uint16_t pctype);
+int i40e_ethertype_filter_set(struct i40e_pf *pf,
+ struct rte_eth_ethertype_filter *filter,
+ bool add);
+int i40e_add_del_fdir_filter(struct rte_eth_dev *dev,
+const struct rte_eth_fdir_filter *filter,
+bool add);
+int i40e_dev_tunnel_filter_set(struct i40e_pf *pf,
+ struct rte_eth_tunnel_filter_conf *tunnel_filter,
+ uint8_t add);
/* I40E_DEV_PRIVATE_TO */
#define I40E_DEV_PRIVATE_TO_PF(adapter) \
diff --git a/drivers/net/i40e/i40e_fdir.c b/drivers/net/i40e/i40e_fdir.c
index f89dbc9..91d91aa 100644
--- a/drivers/net/i40e/i40e_fdir.c
+++ b/drivers/net/i40e/i40e_fdir.c
@@ -1099,7 +1099,7 @@ i40e_sw_fdir_filter_del(struct i40e_pf *pf, struct
rte_eth_fdir_input *input)
* @filter: fdir filter entry
* @add: 0 - delete, 1 - add
*/
-static int
+int
i40e_add_del_fdir_filter(struct rte_eth_dev *dev,
const struct rte_eth_fdir_filter *filter,
bool add)
diff --git a/drivers/net/i40e/i40e_flow.c b/drivers/net/i40e/i40e_flow.c
index 063f8e2..67ea83d 100644
--- a/drivers/net/i40e/i40e_flow.c
+++ b/drivers/net/i40e/i40e_flow.c
@@ -62,6 +62,11 @@ static int i40e_flow_validate(struct rte_eth_dev *dev,
const struct rte_flow_item pattern[],
const struct rte_flow_action actions[],
struct rte_flow_error *error);
+static struct rte_flow *i40e_flow_create(struct rte_eth_dev *dev,
+const struct rte_flow_attr *attr,
+
[dpdk-dev] [PATCH v4 12/17] net/i40e: destroy ethertype filter
This patch adds i40e_dev_destroy_ethertype_filter function
to destroy a ethertype filter for users.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_flow.c | 42 ++
1 file changed, 42 insertions(+)
diff --git a/drivers/net/i40e/i40e_flow.c b/drivers/net/i40e/i40e_flow.c
index cda308d..8d1e2ff 100644
--- a/drivers/net/i40e/i40e_flow.c
+++ b/drivers/net/i40e/i40e_flow.c
@@ -49,6 +49,7 @@
#include "i40e_logs.h"
#include "base/i40e_type.h"
+#include "base/i40e_prototype.h"
#include "i40e_ethdev.h"
#define I40E_IPV4_TC_SHIFT 4
@@ -96,6 +97,8 @@ static int i40e_parse_tunnel_act(struct rte_eth_dev *dev,
struct rte_eth_tunnel_filter_conf *filter);
static int i40e_parse_attr(const struct rte_flow_attr *attr,
struct rte_flow_error *error);
+static int i40e_dev_destroy_ethertype_filter(struct i40e_pf *pf,
+struct i40e_ethertype_filter *filter);
const struct rte_flow_ops i40e_flow_ops = {
.validate = i40e_flow_validate,
@@ -1539,6 +1542,10 @@ i40e_flow_destroy(struct rte_eth_dev *dev,
int ret = 0;
switch (filter_type) {
+ case RTE_ETH_FILTER_ETHERTYPE:
+ ret = i40e_dev_destroy_ethertype_filter(pf,
+ (struct i40e_ethertype_filter *)pmd_flow->rule);
+ break;
default:
PMD_DRV_LOG(WARNING, "Filter type (%d) not supported",
filter_type);
@@ -1556,3 +1563,38 @@ i40e_flow_destroy(struct rte_eth_dev *dev,
return ret;
}
+
+static int
+i40e_dev_destroy_ethertype_filter(struct i40e_pf *pf,
+ struct i40e_ethertype_filter *filter)
+{
+ struct i40e_hw *hw = I40E_PF_TO_HW(pf);
+ struct i40e_ethertype_rule *ethertype_rule = &pf->ethertype;
+ struct i40e_ethertype_filter *node;
+ struct i40e_control_filter_stats stats;
+ uint16_t flags = 0;
+ int ret = 0;
+
+ if (!(filter->flags & RTE_ETHTYPE_FLAGS_MAC))
+ flags |= I40E_AQC_ADD_CONTROL_PACKET_FLAGS_IGNORE_MAC;
+ if (filter->flags & RTE_ETHTYPE_FLAGS_DROP)
+ flags |= I40E_AQC_ADD_CONTROL_PACKET_FLAGS_DROP;
+ flags |= I40E_AQC_ADD_CONTROL_PACKET_FLAGS_TO_QUEUE;
+
+ memset(&stats, 0, sizeof(stats));
+ ret = i40e_aq_add_rem_control_packet_filter(hw,
+ filter->input.mac_addr.addr_bytes,
+ filter->input.ether_type,
+ flags, pf->main_vsi->seid,
+ filter->queue, 0, &stats, NULL);
+ if (ret < 0)
+ return ret;
+
+ node = i40e_sw_ethertype_filter_lookup(ethertype_rule, &filter->input);
+ if (!node)
+ return -EINVAL;
+
+ ret = i40e_sw_ethertype_filter_del(pf, &node->input);
+
+ return ret;
+}
--
2.5.5
[dpdk-dev] [PATCH v4 11/17] net/i40e: add flow destroy function
This patch adds i40e_flow_destroy function to destroy
a flow for users.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_flow.c | 33 +
1 file changed, 33 insertions(+)
diff --git a/drivers/net/i40e/i40e_flow.c b/drivers/net/i40e/i40e_flow.c
index 67ea83d..cda308d 100644
--- a/drivers/net/i40e/i40e_flow.c
+++ b/drivers/net/i40e/i40e_flow.c
@@ -67,6 +67,9 @@ static struct rte_flow *i40e_flow_create(struct rte_eth_dev
*dev,
const struct rte_flow_item pattern[],
const struct rte_flow_action actions[],
struct rte_flow_error *error);
+static int i40e_flow_destroy(struct rte_eth_dev *dev,
+struct rte_flow *flow,
+struct rte_flow_error *error);
static int i40e_parse_ethertype_pattern(__rte_unused struct rte_eth_dev *dev,
const struct rte_flow_item *pattern,
struct rte_flow_error *error,
@@ -97,6 +100,7 @@ static int i40e_parse_attr(const struct rte_flow_attr *attr,
const struct rte_flow_ops i40e_flow_ops = {
.validate = i40e_flow_validate,
.create = i40e_flow_create,
+ .destroy = i40e_flow_destroy,
};
union i40e_filter_t cons_filter;
@@ -1523,3 +1527,32 @@ i40e_flow_create(struct rte_eth_dev *dev,
rte_free(flow);
return NULL;
}
+
+static int
+i40e_flow_destroy(struct rte_eth_dev *dev,
+ struct rte_flow *flow,
+ struct rte_flow_error *error)
+{
+ struct i40e_pf *pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private);
+ struct i40e_flow *pmd_flow = (struct i40e_flow *)flow;
+ enum rte_filter_type filter_type = pmd_flow->filter_type;
+ int ret = 0;
+
+ switch (filter_type) {
+ default:
+ PMD_DRV_LOG(WARNING, "Filter type (%d) not supported",
+ filter_type);
+ ret = -EINVAL;
+ break;
+ }
+
+ if (!ret) {
+ TAILQ_REMOVE(&pf->flow_list, pmd_flow, node);
+ rte_free(pmd_flow);
+ } else
+ rte_flow_error_set(error, -ret,
+ RTE_FLOW_ERROR_TYPE_HANDLE, NULL,
+ "Failed to destroy flow.");
+
+ return ret;
+}
--
2.5.5
[dpdk-dev] [PATCH v4 14/17] net/i40e: destroy flow directory filter
This patch supports destroying a flow directory filter for users. Signed-off-by: Beilei Xing --- drivers/net/i40e/i40e_flow.c | 4 1 file changed, 4 insertions(+) diff --git a/drivers/net/i40e/i40e_flow.c b/drivers/net/i40e/i40e_flow.c index 9fc311d..e56e8b8 100644 --- a/drivers/net/i40e/i40e_flow.c +++ b/drivers/net/i40e/i40e_flow.c @@ -1552,6 +1552,10 @@ i40e_flow_destroy(struct rte_eth_dev *dev, ret = i40e_dev_destroy_tunnel_filter(pf, (struct i40e_tunnel_filter *)pmd_flow->rule); break; + case RTE_ETH_FILTER_FDIR: + ret = i40e_add_del_fdir_filter(dev, + &((struct i40e_fdir_filter *)pmd_flow->rule)->fdir, 0); + break; default: PMD_DRV_LOG(WARNING, "Filter type (%d) not supported", filter_type); -- 2.5.5
[dpdk-dev] [PATCH v4 15/17] net/i40e: add flow flush function
This patch adds i40e_flow_flush function to flush all
filters for users. And flow director flush function
is involved first.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_ethdev.h | 1 +
drivers/net/i40e/i40e_fdir.c | 4 +---
drivers/net/i40e/i40e_flow.c | 51 ++
3 files changed, 53 insertions(+), 3 deletions(-)
diff --git a/drivers/net/i40e/i40e_ethdev.h b/drivers/net/i40e/i40e_ethdev.h
index b33910d..57fd796 100644
--- a/drivers/net/i40e/i40e_ethdev.h
+++ b/drivers/net/i40e/i40e_ethdev.h
@@ -788,6 +788,7 @@ int i40e_add_del_fdir_filter(struct rte_eth_dev *dev,
int i40e_dev_tunnel_filter_set(struct i40e_pf *pf,
struct rte_eth_tunnel_filter_conf *tunnel_filter,
uint8_t add);
+int i40e_fdir_flush(struct rte_eth_dev *dev);
/* I40E_DEV_PRIVATE_TO */
#define I40E_DEV_PRIVATE_TO_PF(adapter) \
diff --git a/drivers/net/i40e/i40e_fdir.c b/drivers/net/i40e/i40e_fdir.c
index 91d91aa..67d63ff 100644
--- a/drivers/net/i40e/i40e_fdir.c
+++ b/drivers/net/i40e/i40e_fdir.c
@@ -119,8 +119,6 @@ static int i40e_fdir_filter_programming(struct i40e_pf *pf,
enum i40e_filter_pctype pctype,
const struct rte_eth_fdir_filter *filter,
bool add);
-static int i40e_fdir_flush(struct rte_eth_dev *dev);
-
static int i40e_fdir_filter_convert(const struct rte_eth_fdir_filter *input,
struct i40e_fdir_filter *filter);
static struct i40e_fdir_filter *
@@ -1325,7 +1323,7 @@ i40e_fdir_filter_programming(struct i40e_pf *pf,
* i40e_fdir_flush - clear all filters of Flow Director table
* @pf: board private structure
*/
-static int
+int
i40e_fdir_flush(struct rte_eth_dev *dev)
{
struct i40e_pf *pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private);
diff --git a/drivers/net/i40e/i40e_flow.c b/drivers/net/i40e/i40e_flow.c
index e56e8b8..eacea68 100644
--- a/drivers/net/i40e/i40e_flow.c
+++ b/drivers/net/i40e/i40e_flow.c
@@ -68,6 +68,8 @@ static struct rte_flow *i40e_flow_create(struct rte_eth_dev
*dev,
const struct rte_flow_item pattern[],
const struct rte_flow_action actions[],
struct rte_flow_error *error);
+static int i40e_flow_flush(struct rte_eth_dev *dev,
+ struct rte_flow_error *error);
static int i40e_flow_destroy(struct rte_eth_dev *dev,
struct rte_flow *flow,
struct rte_flow_error *error);
@@ -101,11 +103,13 @@ static int i40e_dev_destroy_ethertype_filter(struct
i40e_pf *pf,
struct i40e_ethertype_filter *filter);
static int i40e_dev_destroy_tunnel_filter(struct i40e_pf *pf,
struct i40e_tunnel_filter *filter);
+static int i40e_fdir_filter_flush(struct i40e_pf *pf);
const struct rte_flow_ops i40e_flow_ops = {
.validate = i40e_flow_validate,
.create = i40e_flow_create,
.destroy = i40e_flow_destroy,
+ .flush = i40e_flow_flush,
};
union i40e_filter_t cons_filter;
@@ -1643,3 +1647,50 @@ i40e_dev_destroy_tunnel_filter(struct i40e_pf *pf,
return ret;
}
+
+static int
+i40e_flow_flush(struct rte_eth_dev *dev, struct rte_flow_error *error)
+{
+ struct i40e_pf *pf = I40E_DEV_PRIVATE_TO_PF(dev->data->dev_private);
+ int ret;
+
+ ret = i40e_fdir_filter_flush(pf);
+ if (ret)
+ rte_flow_error_set(error, -ret,
+ RTE_FLOW_ERROR_TYPE_HANDLE, NULL,
+ "Failed to flush FDIR flows.");
+
+ return ret;
+}
+
+static int
+i40e_fdir_filter_flush(struct i40e_pf *pf)
+{
+ struct rte_eth_dev *dev = pf->adapter->eth_dev;
+ struct i40e_fdir_info *fdir_info = &pf->fdir;
+ struct i40e_fdir_filter *fdir_filter;
+ struct i40e_flow *flow;
+ void *temp;
+ int ret;
+
+ ret = i40e_fdir_flush(dev);
+ if (!ret) {
+ /* Delete FDIR filters in FDIR list. */
+ while ((fdir_filter = TAILQ_FIRST(&fdir_info->fdir_list))) {
+ ret = i40e_sw_fdir_filter_del(pf,
+ &fdir_filter->fdir.input);
+ if (ret < 0)
+ return ret;
+ }
+
+ /* Delete FDIR flows in flow list. */
+ TAILQ_FOREACH_SAFE(flow, &pf->flow_list, node, temp) {
+ if (flow->filter_type == RTE_ETH_FILTER_FDIR) {
+ TAILQ_REMOVE(&pf->flow_list, flow, node);
+ rte_free(flow);
+ }
+ }
+ }
+
+ return ret;
+}
--
2.5.5
[dpdk-dev] [PATCH v4 17/17] net/i40e: flush tunnel filters
This patch adds i40e_tunnel_filter_flush function
to flush all tunnel filters, including filters in
SW and HW.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_flow.c | 37 +
1 file changed, 37 insertions(+)
diff --git a/drivers/net/i40e/i40e_flow.c b/drivers/net/i40e/i40e_flow.c
index 5c16fb8..fd0f0ec 100644
--- a/drivers/net/i40e/i40e_flow.c
+++ b/drivers/net/i40e/i40e_flow.c
@@ -105,6 +105,7 @@ static int i40e_dev_destroy_tunnel_filter(struct i40e_pf
*pf,
struct i40e_tunnel_filter *filter);
static int i40e_fdir_filter_flush(struct i40e_pf *pf);
static int i40e_ethertype_filter_flush(struct i40e_pf *pf);
+static int i40e_tunnel_filter_flush(struct i40e_pf *pf);
const struct rte_flow_ops i40e_flow_ops = {
.validate = i40e_flow_validate,
@@ -1671,6 +1672,14 @@ i40e_flow_flush(struct rte_eth_dev *dev, struct
rte_flow_error *error)
return -rte_errno;
}
+ ret = i40e_tunnel_filter_flush(pf);
+ if (ret) {
+ rte_flow_error_set(error, -ret,
+ RTE_FLOW_ERROR_TYPE_HANDLE, NULL,
+ "Failed to flush tunnel flows.");
+ return -rte_errno;
+ }
+
return ret;
}
@@ -1733,3 +1742,31 @@ i40e_ethertype_filter_flush(struct i40e_pf *pf)
return ret;
}
+
+/* Flush all tunnel filters */
+static int
+i40e_tunnel_filter_flush(struct i40e_pf *pf)
+{
+ struct i40e_tunnel_filter_list
+ *tunnel_list = &pf->tunnel.tunnel_list;
+ struct i40e_tunnel_filter *f;
+ struct i40e_flow *flow;
+ void *temp;
+ int ret = 0;
+
+ while ((f = TAILQ_FIRST(tunnel_list))) {
+ ret = i40e_dev_destroy_tunnel_filter(pf, f);
+ if (ret)
+ return ret;
+ }
+
+ /* Delete tunnel flows in flow list. */
+ TAILQ_FOREACH_SAFE(flow, &pf->flow_list, node, temp) {
+ if (flow->filter_type == RTE_ETH_FILTER_TUNNEL) {
+ TAILQ_REMOVE(&pf->flow_list, flow, node);
+ rte_free(flow);
+ }
+ }
+
+ return ret;
+}
--
2.5.5
[dpdk-dev] [PATCH v4 16/17] net/i40e: flush ethertype filters
This patch adds i40e_ethertype_filter_flush function
to flush all ethertype filters, including filters in
SW and HW.
Signed-off-by: Beilei Xing
---
drivers/net/i40e/i40e_flow.c | 41 -
1 file changed, 40 insertions(+), 1 deletion(-)
diff --git a/drivers/net/i40e/i40e_flow.c b/drivers/net/i40e/i40e_flow.c
index eacea68..5c16fb8 100644
--- a/drivers/net/i40e/i40e_flow.c
+++ b/drivers/net/i40e/i40e_flow.c
@@ -104,6 +104,7 @@ static int i40e_dev_destroy_ethertype_filter(struct i40e_pf
*pf,
static int i40e_dev_destroy_tunnel_filter(struct i40e_pf *pf,
struct i40e_tunnel_filter *filter);
static int i40e_fdir_filter_flush(struct i40e_pf *pf);
+static int i40e_ethertype_filter_flush(struct i40e_pf *pf);
const struct rte_flow_ops i40e_flow_ops = {
.validate = i40e_flow_validate,
@@ -1655,10 +1656,20 @@ i40e_flow_flush(struct rte_eth_dev *dev, struct
rte_flow_error *error)
int ret;
ret = i40e_fdir_filter_flush(pf);
- if (ret)
+ if (ret) {
rte_flow_error_set(error, -ret,
RTE_FLOW_ERROR_TYPE_HANDLE, NULL,
"Failed to flush FDIR flows.");
+ return -rte_errno;
+ }
+
+ ret = i40e_ethertype_filter_flush(pf);
+ if (ret) {
+ rte_flow_error_set(error, -ret,
+ RTE_FLOW_ERROR_TYPE_HANDLE, NULL,
+ "Failed to ethertype flush flows.");
+ return -rte_errno;
+ }
return ret;
}
@@ -1694,3 +1705,31 @@ i40e_fdir_filter_flush(struct i40e_pf *pf)
return ret;
}
+
+/* Flush all ethertype filters */
+static int
+i40e_ethertype_filter_flush(struct i40e_pf *pf)
+{
+ struct i40e_ethertype_filter_list
+ *ethertype_list = &pf->ethertype.ethertype_list;
+ struct i40e_ethertype_filter *f;
+ struct i40e_flow *flow;
+ void *temp;
+ int ret = 0;
+
+ while ((f = TAILQ_FIRST(ethertype_list))) {
+ ret = i40e_dev_destroy_ethertype_filter(pf, f);
+ if (ret)
+ return ret;
+ }
+
+ /* Delete ethertype flows in flow list. */
+ TAILQ_FOREACH_SAFE(flow, &pf->flow_list, node, temp) {
+ if (flow->filter_type == RTE_ETH_FILTER_ETHERTYPE) {
+ TAILQ_REMOVE(&pf->flow_list, flow, node);
+ rte_free(flow);
+ }
+ }
+
+ return ret;
+}
--
2.5.5
Re: [dpdk-dev] [PATCH v2 0/9] rxq interrupt mode for virtio PMD
On Thu, Dec 29, 2016 at 07:42:51AM +, Tan, Jianfeng wrote: > Forget to mention a key part, when we bind virtio to vfio-pci driver in the > VM, we need to insert vfio kernel module like this: > > $ modprobe vfio enable_unsafe_noiommu_mode=1 > $ modprobe vfio-pci Such info should be documated somewhere, say maybe at doc/guides/nics/virtio.rst --yliu
Re: [dpdk-dev] [PATCH v2 3/9] net/virtio: add PCI ops for queue/irq binding
On Thu, Dec 29, 2016 at 07:30:37AM +, Jianfeng Tan wrote:
> Add handler in virtio_pci_ops to set queue/irq bind.
>
> Signed-off-by: Jianfeng Tan
> ---
> drivers/net/virtio/virtio_pci.c | 29 +
> drivers/net/virtio/virtio_pci.h | 5 +
> 2 files changed, 34 insertions(+)
>
> diff --git a/drivers/net/virtio/virtio_pci.c b/drivers/net/virtio/virtio_pci.c
> index 9b47165..4e706cf 100644
> --- a/drivers/net/virtio/virtio_pci.c
> +++ b/drivers/net/virtio/virtio_pci.c
> @@ -217,6 +217,18 @@ legacy_set_config_irq(struct virtio_hw *hw, uint16_t vec)
> }
>
> static uint16_t
> +legacy_set_queue_irq(struct virtio_hw *hw, struct virtqueue *vq, uint16_t
> vec)
> +{
> + uint16_t dst;
> +
> + rte_eal_pci_ioport_write(&hw->io, &vq->vq_queue_index, 2,
> + VIRTIO_PCI_QUEUE_SEL);
> + rte_eal_pci_ioport_write(&hw->io, &vec, 2, VIRTIO_MSI_QUEUE_VECTOR);
> + rte_eal_pci_ioport_read(&hw->io, &dst, 2, VIRTIO_MSI_QUEUE_VECTOR);
> + return dst;
> +}
> +
> +static uint16_t
> legacy_get_queue_num(struct virtio_hw *hw, uint16_t queue_id)
> {
> uint16_t dst;
> @@ -310,6 +322,7 @@ static const struct virtio_pci_ops legacy_ops = {
> .set_features = legacy_set_features,
> .get_isr= legacy_get_isr,
> .set_config_irq = legacy_set_config_irq,
> + .set_queue_irq = legacy_set_queue_irq,
Please keep the vertical alignment here.
> .get_queue_num = legacy_get_queue_num,
> .setup_queue= legacy_setup_queue,
> .del_queue = legacy_del_queue,
> @@ -449,6 +462,15 @@ modern_set_config_irq(struct virtio_hw *hw, uint16_t vec)
> }
>
> static uint16_t
> +modern_set_queue_irq(struct virtio_hw *hw, struct virtqueue *vq, uint16_t
> vec)
> +{
> + PMD_INIT_LOG(INFO, "vq: %u, vec: %u", vq->vq_queue_index, vec);
Such debug info should not exist only in one callback. It should be
in the caller of this callback, before invoke it.
> + io_write16(vq->vq_queue_index, &hw->common_cfg->queue_select);
> + io_write16(vec, &hw->common_cfg->queue_msix_vector);
> + return io_read16(&hw->common_cfg->queue_msix_vector);
> +}
> +
> +static uint16_t
> modern_get_queue_num(struct virtio_hw *hw, uint16_t queue_id)
> {
> io_write16(queue_id, &hw->common_cfg->queue_select);
> @@ -526,6 +548,7 @@ static const struct virtio_pci_ops modern_ops = {
> .set_features = modern_set_features,
> .get_isr= modern_get_isr,
> .set_config_irq = modern_set_config_irq,
> + .set_queue_irq = modern_set_queue_irq,
Ditto.
> .get_queue_num = modern_get_queue_num,
> .setup_queue= modern_setup_queue,
> .del_queue = modern_del_queue,
> @@ -605,6 +628,12 @@ vtpci_irq_config(struct virtio_hw *hw, uint16_t vec)
> return hw->vtpci_ops->set_config_irq(hw, vec);
> }
>
> +uint16_t
> +vtpci_irq_queue(struct virtqueue *vq, uint16_t vec)
> +{
> + return vq->hw->vtpci_ops->set_queue_irq(vq->hw, vq, vec);
> +}
No need to introduce a wrapper for a vtpci_ops, instead, just invoke
it directly.
--yliu
Re: [dpdk-dev] [PATCH v2 4/9] net/virtio: add Rx queue intr enable/disable functions
On Thu, Dec 29, 2016 at 07:30:38AM +, Jianfeng Tan wrote:
> /*
> * Two types of mbuf to be cleaned:
> * 1) mbuf that has been consumed by backend but not used by virtio.
> diff --git a/drivers/net/virtio/virtqueue.h b/drivers/net/virtio/virtqueue.h
> index f0bb089..62be136 100644
> --- a/drivers/net/virtio/virtqueue.h
> +++ b/drivers/net/virtio/virtqueue.h
> @@ -274,7 +274,29 @@ vring_desc_init(struct vring_desc *dp, uint16_t n)
> /**
> * Tell the backend not to interrupt us.
> */
> -void virtqueue_disable_intr(struct virtqueue *vq);
> +static inline void
> +virtqueue_disable_intr(struct virtqueue *vq)
> +{
> + /* Set VRING_AVAIL_F_NO_INTERRUPT to hint host
> + * not to interrupt when it consumes packets
> + * Note: this is only considered a hint to the host
> + */
> + vq->vq_ring.avail->flags |= VRING_AVAIL_F_NO_INTERRUPT;
> +}
> +
> +/**
> + * Tell the backend to interrupt us.
> + */
> +static inline void
> +virtqueue_enable_intr(struct virtqueue *vq)
> +{
> + /* Unset VRING_AVAIL_F_NO_INTERRUPT to hint host
> + * to interrupt when it consumes packets
> + * Note: this is only considered a hint to the host
> + */
This kind of comment is a bit weird to me; it becomes weird-er when it
appears twice. Maybe we could just drop it, as the code already expains
it well (disabling interrupt regarding to NO_INTERRUPT). Or, you could
combine the two and write one comment before the two functions.
--yliu
> + vq->vq_ring.avail->flags &= (~VRING_AVAIL_F_NO_INTERRUPT);
> +}
> +
> /**
> * Dump virtqueue internal structures, for debug purpose only.
> */
> --
> 2.7.4
Re: [dpdk-dev] [PATCH v2 5/9] net/virtio: setup rxq interrupts
On Thu, Dec 29, 2016 at 07:30:39AM +, Jianfeng Tan wrote:
> This patch mainly allocates structure to store queue/irq mapping,
> and configure queue/irq mapping down through PCI ops. It also creates
> eventfds for each Rx queue and tell the kernel about the eventfd/intr
> binding.
>
> Mostly importantly, different from previous NICs (usually implements
> these logic in dev_start()), virtio's interrupt settings should be
> configured down to QEMU before sending DRIVER_OK notification.
Isn't it obvious we have to have all driver stuff (including interrupt
settings) configured properly before setting DRIVER_OK? :) That said,
it's meanless to state the fact that virtio acts differently than other
nics here on dev_start/stop.
> Note: We only support 1:1 queue/irq mapping so far, which means, each
> rx queue has one exclusive interrupt (corresponding to irqfd in the
> qemu/kvm) to get notified when packets are available on that queue.
That means you have to setup the "vectors=N" option has to set correctly
in QEMU, otherwise it won't work? If so, you also have to doc it somewhere.
> Signed-off-by: Jianfeng Tan
> ---
> drivers/net/virtio/virtio_ethdev.c | 89
> ++
> 1 file changed, 89 insertions(+)
>
> diff --git a/drivers/net/virtio/virtio_ethdev.c
> b/drivers/net/virtio/virtio_ethdev.c
> index 3f8b90c..082346b 100644
> --- a/drivers/net/virtio/virtio_ethdev.c
> +++ b/drivers/net/virtio/virtio_ethdev.c
> @@ -1206,6 +1206,76 @@ rx_func_get(struct rte_eth_dev *eth_dev)
> eth_dev->rx_pkt_burst = &virtio_recv_pkts;
> }
>
> +/* Only support 1:1 queue/interrupt mapping so far.
> + * TODO: under below cases, lsc and rxq interrupt share one interrupt.
> + * a) binded to uio, igb_uio, vfio (type1);
> + * b) device only has one vec, see _vectors_ option in -device
> virtio-net-pci.
> + * TODO: support n:1 queue/interrupt mapping.
Both TODOs are actually the same: supporting n:1 mapping. That said, you
don't have to write 2 TODOs here. Please, don't be mean by adding some
whitespace lines.
> + */
> +static int
> +virtio_queues_bind_intr(struct rte_eth_dev *dev)
> +{
> + uint32_t i;
> + struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
> + struct virtio_hw *hw = dev->data->dev_private;
> +
> + PMD_INIT_LOG(INFO, "queue/interrupt binding\n");
> + for (i = 0; i < dev->data->nb_rx_queues; ++i) {
> + intr_handle->intr_vec[i] = i + 1;
> + if (vtpci_irq_queue(hw->vqs[i * VTNET_CQ], i + 1) ==
It's logically wrong to use VTNET_CQ to get the Rx queue index. You
could either use 2 simply, or define a macro for that.
> + VIRTIO_MSI_NO_VECTOR) {
> + PMD_DRV_LOG(ERR, "failed to set queue vector");
> + return -EBUSY;
> + }
> + }
> +
> + return 0;
> +}
> +
> +static int
> +virtio_configure_intr(struct rte_eth_dev *dev)
> +{
> + uint32_t intr_vector;
> + struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
> +
> + /* check if rxq interrupt is enabled */
Unnecessary comment, the function name and the error log explains it
well.
> + if (!rte_intr_cap_multiple(intr_handle)) {
> + PMD_INIT_LOG(ERR, "Multiple intr vector not supported");
> + return -ENOTSUP;
> + }
> +
> + intr_vector = dev->data->nb_rx_queues;
> + if (rte_intr_efd_enable(intr_handle, intr_vector)) {
> + PMD_INIT_LOG(ERR, "Fail to create eventfd");
> + return -1;
> + }
> +
> + if (!intr_handle->intr_vec) {
> + intr_handle->intr_vec =
> + rte_zmalloc("intr_vec", intr_vector * sizeof(int), 0);
> + if (!intr_handle->intr_vec) {
> + PMD_INIT_LOG(ERR, "Failed to allocate %d rxq vectors",
> + intr_vector);
> + return -ENOMEM;
> + }
> + }
> +
> + if (virtio_queues_bind_intr(dev) < 0) {
> + PMD_INIT_LOG(ERR, "Failed to bind queue/interrupt");
> + return -1;
You have to free intr_handle->intr_vec, otherwise, memory leak occurs.
> + }
> +
> + /* DO NOT try remove this! This function will enable msix, or QEMU
> + * will encounter SIGSEGV.
> + */
Looks like a QEMU bug to me. I mean, even though the driver is badly
configured, it should not crash QEMU.
> + if (rte_intr_enable(intr_handle) < 0) {
> + PMD_DRV_LOG(ERR, "interrupt enable failed");
> + return -1;
> + }
> +
> + return 0;
> +}
> +
> /* reset device and renegotiate features if needed */
> static int
> virtio_init_device(struct rte_eth_dev *eth_dev, uint64_t req_features)
> @@ -1299,6 +1369,17 @@ virtio_init_device(struct rte_eth_dev *eth_dev,
> uint64_t req_features)
> ret = virtio_alloc_queues(eth_dev);
> if (ret < 0)
> return ret;
> +
> + /* Make sure rxq interrupt is configure
Re: [dpdk-dev] [PATCH v2 7/9] net/virtio: unmapping queue/irq when close device
On Thu, Dec 29, 2016 at 07:30:41AM +, Jianfeng Tan wrote:
> When closing virtio devices, close eventfds, free the struct to
> store queue/irq mapping.
>
> Signed-off-by: Jianfeng Tan
> ---
> drivers/net/virtio/virtio_ethdev.c | 27 ++-
> 1 file changed, 26 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/net/virtio/virtio_ethdev.c
> b/drivers/net/virtio/virtio_ethdev.c
> index 0754ba0..d1a0afb 100644
> --- a/drivers/net/virtio/virtio_ethdev.c
> +++ b/drivers/net/virtio/virtio_ethdev.c
> @@ -591,16 +591,30 @@ virtio_alloc_queues(struct rte_eth_dev *dev)
> return 0;
> }
>
> +static void virtio_queues_unbind_intr(struct rte_eth_dev *dev);
> +
> static void
> virtio_dev_close(struct rte_eth_dev *dev)
> {
> struct virtio_hw *hw = dev->data->dev_private;
> + struct rte_intr_conf *intr_conf = &dev->data->dev_conf.intr_conf;
> + struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
I saw quite many occurrence of the two, maybe you could define few
macros for that, say VIRTIO_INTR_ENABLED, VIRTIO_RX_INTR_ENABLED,
VIRTIO_LSC_INTR_ENABLED?
--yliu
Re: [dpdk-dev] [PATCH v2 8/9] examples/l3fwd: add parse-ptype option
On Thu, Dec 29, 2016 at 07:30:42AM +, Jianfeng Tan wrote:
> To support those devices that do not provide packet type info when
> receiving packets, add a new option, --parse-ptype, to analyze
> packet type in the Rx callback.
I think this would be needed for all PMD drivers don't have the PTYPE
support. For these, --parse-ptype looks like a mandatory option in
the l3fwd example. I didn't find such option given in your test guide
though, which is weird. Mistake?
Besides that, is there a way to query whether a PMD supports PTYPE
or not? If not, we should add a software one by our own. This could
be done without introducing yet another option.
>
> Signed-off-by: Jianfeng Tan
> ---
> examples/l3fwd-power/main.c | 60
> -
> 1 file changed, 59 insertions(+), 1 deletion(-)
>
> diff --git a/examples/l3fwd-power/main.c b/examples/l3fwd-power/main.c
> index b65d683..44843ec 100644
> --- a/examples/l3fwd-power/main.c
> +++ b/examples/l3fwd-power/main.c
> @@ -164,6 +164,8 @@ static uint32_t enabled_port_mask = 0;
> static int promiscuous_on = 0;
> /* NUMA is enabled by default. */
> static int numa_on = 1;
> +static int parse_ptype; /**< Parse packet type using rx callback, and */
> + /**< disabled by default */
>
> enum freq_scale_hint_t
> {
> @@ -607,6 +609,48 @@ get_ipv4_dst_port(struct ipv4_hdr *ipv4_hdr, uint8_t
> portid,
> #endif
>
> static inline void
> +parse_ptype_one(struct rte_mbuf *m)
> +{
> + struct ether_hdr *eth_hdr;
> + uint32_t packet_type = RTE_PTYPE_UNKNOWN;
> + uint16_t ether_type;
> +
> + eth_hdr = rte_pktmbuf_mtod(m, struct ether_hdr *);
> + ether_type = eth_hdr->ether_type;
> + if (ether_type == rte_cpu_to_be_16(ETHER_TYPE_IPv4))
> + packet_type |= RTE_PTYPE_L3_IPV4_EXT_UNKNOWN;
> + else if (ether_type == rte_cpu_to_be_16(ETHER_TYPE_IPv6))
> + packet_type |= RTE_PTYPE_L3_IPV6_EXT_UNKNOWN;
BTW, we have rte_net_get_ptype(). Looks like a better option?
--yliu
Re: [dpdk-dev] [PATCH v2 9/9] examples/l3fwd-power: fix not stop and close device
On Thu, Dec 29, 2016 at 07:30:43AM +, Jianfeng Tan wrote: > As it gets killed, in SIGINT signal handler, device is not stopped > and closed. In virtio's case, vector assignment in the KVM is not > deassigned. What wrong could happen then? > This patch will invoke dev_stop() and dev_close() in signal handler. I will just say, it may workaround the bug you encountered, particulary, for this example only. If some people want to use the virtio interrupt somewhere at another app, he also has to do similar thing. Is there a more clean way to handle such case in the driver? --yliu
Re: [dpdk-dev] [PATCH v2 0/9] rxq interrupt mode for virtio PMD
On Thu, Dec 29, 2016 at 07:30:34AM +, Jianfeng Tan wrote: > v2: > - Add PCI queue/irq config ops. > - Move rxq interrupt settings before sending DRIVER OK. > > Historically, virtio PMD can only be binded to igb_uio or > uio_pci_generic, and not for vfio-pci. Besides, quote from > http://dpdk.org/doc/guides-16.11/rel_notes/release_2_1.html: > "Per queue RX interrupt events are only allowed in VFIO >which supports multiple MSI-X vectors." > > As Linux starts to support vfio noiommu mode since 4.8.0, it's > a good chance to enable rxq interrupt for virtio PMD. This cover letter (as well as all the commit logs) lacks an _overall_ introducation of how the Rx interrupt works in virtio, say how things are setup, etc. --yliu > > How to test: > > Step 1, prepare a VM image with kernel version >= 4.8.0. > > Step 2, on the host, start a testpmd with a vhost port: > $ testpmd -c 0x7 -m 1024 --vdev 'eth_vhost0,iface=/tmp/sock0,queues=2' \ > --no-pci -- -i --rxq=2 --txq=2 --nb-cores=2 > > Step 3, boot the VM: > $ qemu ... -chardev socket,id=chr1,path=/tmp/sock0 \ > -netdev vhost-user,id=net1,chardev=chr1,vhostforce,queues=2 \ > -device virtio-net-pci,netdev=net1,mq=on,vectors=5 ... > > Step 4, start l3fwd-power in VM: > $ l3fwd-power -c 0x3 -n 4 -- -p 1 -P --config="(0,0,1),(0,1,1)" --no-numa > > Step 5, send packets from testpmd on the host: > $ start tx_first > > Then l3fwd-power outputs: > L3FWD_POWER: lcore 1 is waked up from rx interrupt on port 0 queue 0 > L3FWD_POWER: lcore 1 is waked up from rx interrupt on port 0 queue 1 > > Signed-off-by: Jianfeng Tan > Jianfeng Tan (9): > net/virtio: fix rewriting LSC flag > net/virtio: add Rx descriptor check > net/virtio: add PCI ops for queue/irq binding > net/virtio: add Rx queue intr enable/disable functions > net/virtio: setup rxq interrupts > net/virtio: unbind intr/eventfd when stop device > net/virtio: unmapping queue/irq when close device > examples/l3fwd: add parse-ptype option > examples/l3fwd-power: fix not stop and close device > > drivers/net/virtio/virtio_ethdev.c | 149 > +++-- > drivers/net/virtio/virtio_ethdev.h | 3 + > drivers/net/virtio/virtio_pci.c| 29 > drivers/net/virtio/virtio_pci.h| 5 ++ > drivers/net/virtio/virtio_rxtx.c | 9 +++ > drivers/net/virtio/virtqueue.c | 11 --- > drivers/net/virtio/virtqueue.h | 24 +- > examples/l3fwd-power/main.c| 70 - > 8 files changed, 282 insertions(+), 18 deletions(-) > > -- > 2.7.4
Re: [dpdk-dev] [PATCH v2 9/9] examples/l3fwd-power: fix not stop and close device
> -Original Message- > From: Yuanhan Liu [mailto:yuanhan@linux.intel.com] > Sent: Friday, December 30, 2016 2:45 PM > To: Tan, Jianfeng > Cc: dev@dpdk.org; step...@networkplumber.org > Subject: Re: [PATCH v2 9/9] examples/l3fwd-power: fix not stop and close > device > > On Thu, Dec 29, 2016 at 07:30:43AM +, Jianfeng Tan wrote: > > As it gets killed, in SIGINT signal handler, device is not stopped > > and closed. In virtio's case, vector assignment in the KVM is not > > deassigned. > > What wrong could happen then? Actually, no strange things happen so far. My purpose for this patch is to verify that irqfd is deassigned. And as calling dev_stop() and dev_close() is a good practice, so I use the word "fix" here. > > > This patch will invoke dev_stop() and dev_close() in signal handler. > > I will just say, it may workaround the bug you encountered, particulary, > for this example only. There's no bug to work around. > If some people want to use the virtio interrupt > somewhere at another app, he also has to do similar thing. Vfio-pci is the best place to put device into original state, but is it really necessary to do that? If another program takes over that device, it will be re-initialized. > > Is there a more clean way to handle such case in the driver? Let's do with the necessity firstly. Thanks, Jianfeng > > --yliu
Re: [dpdk-dev] [PATCH v2 8/9] examples/l3fwd: add parse-ptype option
Hi,
> -Original Message-
> From: Yuanhan Liu [mailto:yuanhan@linux.intel.com]
> Sent: Friday, December 30, 2016 2:40 PM
> To: Tan, Jianfeng
> Cc: dev@dpdk.org; step...@networkplumber.org
> Subject: Re: [PATCH v2 8/9] examples/l3fwd: add parse-ptype option
>
> On Thu, Dec 29, 2016 at 07:30:42AM +, Jianfeng Tan wrote:
> > To support those devices that do not provide packet type info when
> > receiving packets, add a new option, --parse-ptype, to analyze
> > packet type in the Rx callback.
>
> I think this would be needed for all PMD drivers don't have the PTYPE
> support. For these, --parse-ptype looks like a mandatory option in
> the l3fwd example. I didn't find such option given in your test guide
> though, which is weird. Mistake?
Oops, my fault, it should be used with this option. As I just cared about if
packets are received, instead of what types of packets are received, so I
missed that. I'll fix it.
>
> Besides that, is there a way to query whether a PMD supports PTYPE
> or not?
Yes, we have API rte_eth_dev_get_supported_ptypes() to do that. This patch is
to emulate the commit 71a7e2424e07 ("examples/l3fwd: fix using packet type
blindly").
As we talked offline, I'll leverage this API to query if device support needed
ptypes, if not, register callback to analysis ptypes. This avoids to use
another option. But for record, this also leads to un-consistent behavior with
l3fwd.
Thanks,
Jianfeng
> If not, we should add a software one by our own. This could
> be done without introducing yet another option.
>
> >
> > Signed-off-by: Jianfeng Tan
> > ---
> > examples/l3fwd-power/main.c | 60
> -
> > 1 file changed, 59 insertions(+), 1 deletion(-)
> >
> > diff --git a/examples/l3fwd-power/main.c b/examples/l3fwd-
> power/main.c
> > index b65d683..44843ec 100644
> > --- a/examples/l3fwd-power/main.c
> > +++ b/examples/l3fwd-power/main.c
> > @@ -164,6 +164,8 @@ static uint32_t enabled_port_mask = 0;
> > static int promiscuous_on = 0;
> > /* NUMA is enabled by default. */
> > static int numa_on = 1;
> > +static int parse_ptype; /**< Parse packet type using rx callback, and */
> > + /**< disabled by default */
> >
> > enum freq_scale_hint_t
> > {
> > @@ -607,6 +609,48 @@ get_ipv4_dst_port(struct ipv4_hdr *ipv4_hdr,
> uint8_t portid,
> > #endif
> >
> > static inline void
> > +parse_ptype_one(struct rte_mbuf *m)
> > +{
> > + struct ether_hdr *eth_hdr;
> > + uint32_t packet_type = RTE_PTYPE_UNKNOWN;
> > + uint16_t ether_type;
> > +
> > + eth_hdr = rte_pktmbuf_mtod(m, struct ether_hdr *);
> > + ether_type = eth_hdr->ether_type;
> > + if (ether_type == rte_cpu_to_be_16(ETHER_TYPE_IPv4))
> > + packet_type |= RTE_PTYPE_L3_IPV4_EXT_UNKNOWN;
> > + else if (ether_type == rte_cpu_to_be_16(ETHER_TYPE_IPv6))
> > + packet_type |= RTE_PTYPE_L3_IPV6_EXT_UNKNOWN;
>
> BTW, we have rte_net_get_ptype(). Looks like a better option?
>
> --yliu
[dpdk-dev] [PATCH v2 02/18] net/ixgbe: store flow director filter
Add support for storing flow director filter in SW.
Signed-off-by: Wenzhuo Lu
Signed-off-by: Wei Zhao
---
v2:
--add a fdir initialization function in device start process
---
drivers/net/ixgbe/ixgbe_ethdev.c | 55
drivers/net/ixgbe/ixgbe_ethdev.h | 19 ++-
drivers/net/ixgbe/ixgbe_fdir.c | 105 ++-
3 files changed, 176 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/drivers/net/ixgbe/ixgbe_ethdev.c
index 316e560..de27a73 100644
--- a/drivers/net/ixgbe/ixgbe_ethdev.c
+++ b/drivers/net/ixgbe/ixgbe_ethdev.c
@@ -60,6 +60,7 @@
#include
#include
#include
+#include
#include "ixgbe_logs.h"
#include "base/ixgbe_api.h"
@@ -165,6 +166,7 @@ enum ixgbevf_xcast_modes {
static int eth_ixgbe_dev_init(struct rte_eth_dev *eth_dev);
static int eth_ixgbe_dev_uninit(struct rte_eth_dev *eth_dev);
+static int ixgbe_fdir_filter_init(struct rte_eth_dev *eth_dev);
static int ixgbe_dev_configure(struct rte_eth_dev *dev);
static int ixgbe_dev_start(struct rte_eth_dev *dev);
static void ixgbe_dev_stop(struct rte_eth_dev *dev);
@@ -1276,6 +1278,9 @@ eth_ixgbe_dev_init(struct rte_eth_dev *eth_dev)
/* initialize SYN filter */
filter_info->syn_info = 0;
+ /* initialize flow director filter list & hash */
+ ixgbe_fdir_filter_init(eth_dev);
+
return 0;
}
@@ -1284,6 +1289,9 @@ eth_ixgbe_dev_uninit(struct rte_eth_dev *eth_dev)
{
struct rte_pci_device *pci_dev;
struct ixgbe_hw *hw;
+ struct ixgbe_hw_fdir_info *fdir_info =
+ IXGBE_DEV_PRIVATE_TO_FDIR_INFO(eth_dev->data->dev_private);
+ struct ixgbe_fdir_filter *fdir_filter;
PMD_INIT_FUNC_TRACE();
@@ -1317,9 +1325,56 @@ eth_ixgbe_dev_uninit(struct rte_eth_dev *eth_dev)
rte_free(eth_dev->data->hash_mac_addrs);
eth_dev->data->hash_mac_addrs = NULL;
+ /* remove all the fdir filters & hash */
+ if (fdir_info->hash_map)
+ rte_free(fdir_info->hash_map);
+ if (fdir_info->hash_handle)
+ rte_hash_free(fdir_info->hash_handle);
+
+ while ((fdir_filter = TAILQ_FIRST(&fdir_info->fdir_list))) {
+ TAILQ_REMOVE(&fdir_info->fdir_list,
+fdir_filter,
+entries);
+ rte_free(fdir_filter);
+ }
+
return 0;
}
+static int ixgbe_fdir_filter_init(struct rte_eth_dev *eth_dev)
+{
+ struct ixgbe_hw_fdir_info *fdir_info =
+ IXGBE_DEV_PRIVATE_TO_FDIR_INFO(eth_dev->data->dev_private);
+ char fdir_hash_name[RTE_HASH_NAMESIZE];
+ struct rte_hash_parameters fdir_hash_params = {
+ .name = fdir_hash_name,
+ .entries = IXGBE_MAX_FDIR_FILTER_NUM,
+ .key_len = sizeof(union ixgbe_atr_input),
+ .hash_func = rte_hash_crc,
+ .hash_func_init_val = 0,
+ .socket_id = rte_socket_id(),
+ };
+
+ TAILQ_INIT(&fdir_info->fdir_list);
+ snprintf(fdir_hash_name, RTE_HASH_NAMESIZE,
+"fdir_%s", eth_dev->data->name);
+ fdir_info->hash_handle = rte_hash_create(&fdir_hash_params);
+ if (!fdir_info->hash_handle) {
+ PMD_INIT_LOG(ERR, "Failed to create fdir hash table!");
+ return -EINVAL;
+ }
+ fdir_info->hash_map = rte_zmalloc("ixgbe",
+ sizeof(struct ixgbe_fdir_filter *) *
+ IXGBE_MAX_FDIR_FILTER_NUM,
+ 0);
+ if (!fdir_info->hash_map) {
+ PMD_INIT_LOG(ERR,
+"Failed to allocate memory for fdir hash map!");
+ return -ENOMEM;
+ }
+
+ return 0;
+}
/*
* Negotiate mailbox API version with the PF.
* After reset API version is always set to the basic one (ixgbe_mbox_api_10).
diff --git a/drivers/net/ixgbe/ixgbe_ethdev.h b/drivers/net/ixgbe/ixgbe_ethdev.h
index 827026c..8310220 100644
--- a/drivers/net/ixgbe/ixgbe_ethdev.h
+++ b/drivers/net/ixgbe/ixgbe_ethdev.h
@@ -38,6 +38,7 @@
#include "base/ixgbe_dcb_82598.h"
#include "ixgbe_bypass.h"
#include
+#include
/* need update link, bit flag */
#define IXGBE_FLAG_NEED_LINK_UPDATE (uint32_t)(1 << 0)
@@ -130,10 +131,11 @@
#define IXGBE_MISC_VEC_ID RTE_INTR_VEC_ZERO_OFFSET
#define IXGBE_RX_VEC_START RTE_INTR_VEC_RXTX_OFFSET
+#define IXGBE_MAX_FDIR_FILTER_NUM (1024 * 32)
+
/*
* Information about the fdir mode.
*/
-
struct ixgbe_hw_fdir_mask {
uint16_t vlan_tci_mask;
uint32_t src_ipv4_mask;
@@ -148,6 +150,17 @@ struct ixgbe_hw_fdir_mask {
uint8_t tunnel_type_mask;
};
+struct ixgbe_fdir_filter {
+ TAILQ_ENTRY(ixgbe_fdir_filter) entries;
+ union ixgbe_atr_input ixgbe_fdir; /* key of fdir filter*/
+ uint32_t fdirflags; /* drop or forward */
+ uint32_t fd
[dpdk-dev] [PATCH v2 00/18] net/ixgbe: Consistent filter API
The patches mainly finish following functions: 1) Store and restore all kinds of filters. 2) Parse all kinds of filters. 3) Add flow validate function. 4) Add flow create function. 5) Add flow destroy function. 6) Add flow flush function. v2 changes: fix git log error. Modify some function call relationship. Change return value type of all parse flow functions. Update error info for all flow ops. Add ixgbe_filterlist_flush to flush flows and rules created. zhao wei (18): net/ixgbe: store SYN filter net/ixgbe: store flow director filter net/ixgbe: store L2 tunnel filter net/ixgbe: restore n-tuple filter net/ixgbe: restore ether type filter net/ixgbe: restore TCP SYN filter net/ixgbe: restore flow director filter net/ixgbe: restore L2 tunnel filter net/ixgbe: store and restore L2 tunnel configuration net/ixgbe: flush all the filters net/ixgbe: parse n-tuple filter net/ixgbe: parse ethertype filter net/ixgbe: parse TCP SYN filter net/ixgbe: parse L2 tunnel filter net/ixgbe: parse flow director filter net/ixgbe: create consistent filter net/ixgbe: destroy consistent filter net/ixgbe: flush all the filter list drivers/net/ixgbe/ixgbe_ethdev.c | 3285 -- drivers/net/ixgbe/ixgbe_ethdev.h | 136 +- drivers/net/ixgbe/ixgbe_fdir.c | 403 - drivers/net/ixgbe/ixgbe_pf.c | 26 +- lib/librte_ether/rte_flow.h | 55 + 5 files changed, 3713 insertions(+), 192 deletions(-) -- 2.5.5
[dpdk-dev] [PATCH v2 04/18] net/ixgbe: restore n-tuple filter
Add support for restoring n-tuple filter in SW.
Signed-off-by: Wenzhuo Lu
Signed-off-by: Wei Zhao
---
drivers/net/ixgbe/ixgbe_ethdev.c | 131 +--
1 file changed, 83 insertions(+), 48 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/drivers/net/ixgbe/ixgbe_ethdev.c
index 204e531..4dd8d52 100644
--- a/drivers/net/ixgbe/ixgbe_ethdev.c
+++ b/drivers/net/ixgbe/ixgbe_ethdev.c
@@ -384,6 +384,7 @@ static int ixgbe_dev_udp_tunnel_port_add(struct rte_eth_dev
*dev,
struct rte_eth_udp_tunnel *udp_tunnel);
static int ixgbe_dev_udp_tunnel_port_del(struct rte_eth_dev *dev,
struct rte_eth_udp_tunnel *udp_tunnel);
+static int ixgbe_filter_restore(struct rte_eth_dev *dev);
/*
* Define VF Stats MACRO for Non "cleared on read" register
@@ -1292,6 +1293,9 @@ eth_ixgbe_dev_uninit(struct rte_eth_dev *eth_dev)
{
struct rte_pci_device *pci_dev;
struct ixgbe_hw *hw;
+ struct ixgbe_filter_info *filter_info =
+ IXGBE_DEV_PRIVATE_TO_FILTER_INFO(eth_dev->data->dev_private);
+ struct ixgbe_5tuple_filter *p_5tuple;
struct ixgbe_hw_fdir_info *fdir_info =
IXGBE_DEV_PRIVATE_TO_FDIR_INFO(eth_dev->data->dev_private);
struct ixgbe_l2_tn_info *l2_tn_info =
@@ -1331,6 +1335,16 @@ eth_ixgbe_dev_uninit(struct rte_eth_dev *eth_dev)
rte_free(eth_dev->data->hash_mac_addrs);
eth_dev->data->hash_mac_addrs = NULL;
+ /* Remove all ntuple filters of the device */
+ while ((p_5tuple = TAILQ_FIRST(&filter_info->fivetuple_list))) {
+ TAILQ_REMOVE(&filter_info->fivetuple_list,
+p_5tuple,
+entries);
+ rte_free(p_5tuple);
+ }
+ memset(filter_info->fivetuple_mask, 0,
+ sizeof(uint32_t) * IXGBE_5TUPLE_ARRAY_SIZE);
+
/* remove all the fdir filters & hash */
if (fdir_info->hash_map)
rte_free(fdir_info->hash_map);
@@ -2483,6 +2497,7 @@ ixgbe_dev_start(struct rte_eth_dev *dev)
/* resume enabled intr since hw reset */
ixgbe_enable_intr(dev);
+ ixgbe_filter_restore(dev);
return 0;
@@ -2503,9 +2518,6 @@ ixgbe_dev_stop(struct rte_eth_dev *dev)
IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
struct ixgbe_vf_info *vfinfo =
*IXGBE_DEV_PRIVATE_TO_P_VFDATA(dev->data->dev_private);
- struct ixgbe_filter_info *filter_info =
- IXGBE_DEV_PRIVATE_TO_FILTER_INFO(dev->data->dev_private);
- struct ixgbe_5tuple_filter *p_5tuple, *p_5tuple_next;
struct rte_intr_handle *intr_handle = &dev->pci_dev->intr_handle;
int vf;
@@ -2543,17 +2555,6 @@ ixgbe_dev_stop(struct rte_eth_dev *dev)
memset(&link, 0, sizeof(link));
rte_ixgbe_dev_atomic_write_link_status(dev, &link);
- /* Remove all ntuple filters of the device */
- for (p_5tuple = TAILQ_FIRST(&filter_info->fivetuple_list);
-p_5tuple != NULL; p_5tuple = p_5tuple_next) {
- p_5tuple_next = TAILQ_NEXT(p_5tuple, entries);
- TAILQ_REMOVE(&filter_info->fivetuple_list,
-p_5tuple, entries);
- rte_free(p_5tuple);
- }
- memset(filter_info->fivetuple_mask, 0,
- sizeof(uint32_t) * IXGBE_5TUPLE_ARRAY_SIZE);
-
if (!rte_intr_allow_others(intr_handle))
/* resume to the default handler */
rte_intr_callback_register(intr_handle,
@@ -5795,6 +5796,52 @@ convert_protocol_type(uint8_t protocol_value)
return IXGBE_FILTER_PROTOCOL_NONE;
}
+/* inject a 5-tuple filter to HW */
+static inline void
+ixgbe_inject_5tuple_filter(struct rte_eth_dev *dev,
+ struct ixgbe_5tuple_filter *filter)
+{
+ struct ixgbe_hw *hw = IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+ int i;
+ uint32_t ftqf, sdpqf;
+ uint32_t l34timir = 0;
+ uint8_t mask = 0xff;
+
+ i = filter->index;
+
+ sdpqf = (uint32_t)(filter->filter_info.dst_port <<
+ IXGBE_SDPQF_DSTPORT_SHIFT);
+ sdpqf = sdpqf | (filter->filter_info.src_port & IXGBE_SDPQF_SRCPORT);
+
+ ftqf = (uint32_t)(filter->filter_info.proto &
+ IXGBE_FTQF_PROTOCOL_MASK);
+ ftqf |= (uint32_t)((filter->filter_info.priority &
+ IXGBE_FTQF_PRIORITY_MASK) << IXGBE_FTQF_PRIORITY_SHIFT);
+ if (filter->filter_info.src_ip_mask == 0) /* 0 means compare. */
+ mask &= IXGBE_FTQF_SOURCE_ADDR_MASK;
+ if (filter->filter_info.dst_ip_mask == 0)
+ mask &= IXGBE_FTQF_DEST_ADDR_MASK;
+ if (filter->filter_info.src_port_mask == 0)
+ mask &= IXGBE_FTQF_SOURCE_PORT_MASK;
+ if (filter->filter_info.dst_port_mask == 0)
+ mask &= IXGBE_FTQF_DEST_PORT_MASK;
+
[dpdk-dev] [PATCH v2 01/18] net/ixgbe: store SYN filter
Add support for storing SYN filter in SW.
Signed-off-by: Wenzhuo Lu
Signed-off-by: Wei Zhao
---
v2:
--synqf assignment location change
---
drivers/net/ixgbe/ixgbe_ethdev.c | 14 +++---
drivers/net/ixgbe/ixgbe_ethdev.h | 2 ++
2 files changed, 13 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/drivers/net/ixgbe/ixgbe_ethdev.c
index a25bac8..316e560 100644
--- a/drivers/net/ixgbe/ixgbe_ethdev.c
+++ b/drivers/net/ixgbe/ixgbe_ethdev.c
@@ -1274,6 +1274,8 @@ eth_ixgbe_dev_init(struct rte_eth_dev *eth_dev)
memset(filter_info->fivetuple_mask, 0,
sizeof(uint32_t) * IXGBE_5TUPLE_ARRAY_SIZE);
+ /* initialize SYN filter */
+ filter_info->syn_info = 0;
return 0;
}
@@ -5580,15 +5582,18 @@ ixgbe_syn_filter_set(struct rte_eth_dev *dev,
bool add)
{
struct ixgbe_hw *hw = IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+ struct ixgbe_filter_info *filter_info =
+ IXGBE_DEV_PRIVATE_TO_FILTER_INFO(dev->data->dev_private);
+ uint32_t syn_info;
uint32_t synqf;
if (filter->queue >= IXGBE_MAX_RX_QUEUE_NUM)
return -EINVAL;
- synqf = IXGBE_READ_REG(hw, IXGBE_SYNQF);
+ syn_info = filter_info->syn_info;
if (add) {
- if (synqf & IXGBE_SYN_FILTER_ENABLE)
+ if (syn_info & IXGBE_SYN_FILTER_ENABLE)
return -EINVAL;
synqf = (uint32_t)(((filter->queue <<
IXGBE_SYN_FILTER_QUEUE_SHIFT) &
IXGBE_SYN_FILTER_QUEUE) | IXGBE_SYN_FILTER_ENABLE);
@@ -5598,10 +5603,13 @@ ixgbe_syn_filter_set(struct rte_eth_dev *dev,
else
synqf &= ~IXGBE_SYN_FILTER_SYNQFP;
} else {
- if (!(synqf & IXGBE_SYN_FILTER_ENABLE))
+ synqf = IXGBE_READ_REG(hw, IXGBE_SYNQF);
+ if (!(syn_info & IXGBE_SYN_FILTER_ENABLE))
return -ENOENT;
synqf &= ~(IXGBE_SYN_FILTER_QUEUE | IXGBE_SYN_FILTER_ENABLE);
}
+
+ filter_info->syn_info = synqf;
IXGBE_WRITE_REG(hw, IXGBE_SYNQF, synqf);
IXGBE_WRITE_FLUSH(hw);
return 0;
diff --git a/drivers/net/ixgbe/ixgbe_ethdev.h b/drivers/net/ixgbe/ixgbe_ethdev.h
index 4ff6338..827026c 100644
--- a/drivers/net/ixgbe/ixgbe_ethdev.h
+++ b/drivers/net/ixgbe/ixgbe_ethdev.h
@@ -262,6 +262,8 @@ struct ixgbe_filter_info {
/* Bit mask for every used 5tuple filter */
uint32_t fivetuple_mask[IXGBE_5TUPLE_ARRAY_SIZE];
struct ixgbe_5tuple_filter_list fivetuple_list;
+ /* store the SYN filter info */
+ uint32_t syn_info;
};
/*
--
2.5.5
[dpdk-dev] [PATCH v2 03/18] net/ixgbe: store L2 tunnel filter
Add support for storing L2 tunnel filter in SW.
Signed-off-by: Wenzhuo Lu
Signed-off-by: Wei Zhao
---
v2:
--add a L2 tunnel initialization function in device start process
---
drivers/net/ixgbe/ixgbe_ethdev.c | 157 +++
drivers/net/ixgbe/ixgbe_ethdev.h | 24 ++
2 files changed, 181 insertions(+)
diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/drivers/net/ixgbe/ixgbe_ethdev.c
index de27a73..204e531 100644
--- a/drivers/net/ixgbe/ixgbe_ethdev.c
+++ b/drivers/net/ixgbe/ixgbe_ethdev.c
@@ -167,6 +167,7 @@ enum ixgbevf_xcast_modes {
static int eth_ixgbe_dev_init(struct rte_eth_dev *eth_dev);
static int eth_ixgbe_dev_uninit(struct rte_eth_dev *eth_dev);
static int ixgbe_fdir_filter_init(struct rte_eth_dev *eth_dev);
+static int ixgbe_l2_tn_filter_init(struct rte_eth_dev *eth_dev);
static int ixgbe_dev_configure(struct rte_eth_dev *dev);
static int ixgbe_dev_start(struct rte_eth_dev *dev);
static void ixgbe_dev_stop(struct rte_eth_dev *dev);
@@ -1281,6 +1282,8 @@ eth_ixgbe_dev_init(struct rte_eth_dev *eth_dev)
/* initialize flow director filter list & hash */
ixgbe_fdir_filter_init(eth_dev);
+ /* initialize l2 tunnel filter list & hash */
+ ixgbe_l2_tn_filter_init(eth_dev);
return 0;
}
@@ -1291,7 +1294,10 @@ eth_ixgbe_dev_uninit(struct rte_eth_dev *eth_dev)
struct ixgbe_hw *hw;
struct ixgbe_hw_fdir_info *fdir_info =
IXGBE_DEV_PRIVATE_TO_FDIR_INFO(eth_dev->data->dev_private);
+ struct ixgbe_l2_tn_info *l2_tn_info =
+ IXGBE_DEV_PRIVATE_TO_L2_TN_INFO(eth_dev->data->dev_private);
struct ixgbe_fdir_filter *fdir_filter;
+ struct ixgbe_l2_tn_filter *l2_tn_filter;
PMD_INIT_FUNC_TRACE();
@@ -1338,6 +1344,19 @@ eth_ixgbe_dev_uninit(struct rte_eth_dev *eth_dev)
rte_free(fdir_filter);
}
+ /* remove all the L2 tunnel filters & hash */
+ if (l2_tn_info->hash_map)
+ rte_free(l2_tn_info->hash_map);
+ if (l2_tn_info->hash_handle)
+ rte_hash_free(l2_tn_info->hash_handle);
+
+ while ((l2_tn_filter = TAILQ_FIRST(&l2_tn_info->l2_tn_list))) {
+ TAILQ_REMOVE(&l2_tn_info->l2_tn_list,
+l2_tn_filter,
+entries);
+ rte_free(l2_tn_filter);
+ }
+
return 0;
}
@@ -1372,6 +1391,40 @@ static int ixgbe_fdir_filter_init(struct rte_eth_dev
*eth_dev)
"Failed to allocate memory for fdir hash map!");
return -ENOMEM;
}
+ return 0;
+}
+
+static int ixgbe_l2_tn_filter_init(struct rte_eth_dev *eth_dev)
+{
+ struct ixgbe_l2_tn_info *l2_tn_info =
+ IXGBE_DEV_PRIVATE_TO_L2_TN_INFO(eth_dev->data->dev_private);
+ char l2_tn_hash_name[RTE_HASH_NAMESIZE];
+ struct rte_hash_parameters l2_tn_hash_params = {
+ .name = l2_tn_hash_name,
+ .entries = IXGBE_MAX_L2_TN_FILTER_NUM,
+ .key_len = sizeof(struct ixgbe_l2_tn_key),
+ .hash_func = rte_hash_crc,
+ .hash_func_init_val = 0,
+ .socket_id = rte_socket_id(),
+ };
+
+ TAILQ_INIT(&l2_tn_info->l2_tn_list);
+ snprintf(l2_tn_hash_name, RTE_HASH_NAMESIZE,
+"l2_tn_%s", eth_dev->data->name);
+ l2_tn_info->hash_handle = rte_hash_create(&l2_tn_hash_params);
+ if (!l2_tn_info->hash_handle) {
+ PMD_INIT_LOG(ERR, "Failed to create L2 TN hash table!");
+ return -EINVAL;
+ }
+ l2_tn_info->hash_map = rte_zmalloc("ixgbe",
+ sizeof(struct ixgbe_l2_tn_filter *) *
+ IXGBE_MAX_L2_TN_FILTER_NUM,
+ 0);
+ if (!l2_tn_info->hash_map) {
+ PMD_INIT_LOG(ERR,
+ "Failed to allocate memory for L2 TN hash map!");
+ return -ENOMEM;
+ }
return 0;
}
@@ -7178,12 +7231,104 @@ ixgbe_e_tag_filter_add(struct rte_eth_dev *dev,
return -EINVAL;
}
+static inline struct ixgbe_l2_tn_filter *
+ixgbe_l2_tn_filter_lookup(struct ixgbe_l2_tn_info *l2_tn_info,
+ struct ixgbe_l2_tn_key *key)
+{
+ int ret = 0;
+
+ ret = rte_hash_lookup(l2_tn_info->hash_handle, (const void *)key);
+ if (ret < 0)
+ return NULL;
+
+ return l2_tn_info->hash_map[ret];
+}
+
+static inline int
+ixgbe_insert_l2_tn_filter(struct ixgbe_l2_tn_info *l2_tn_info,
+ struct ixgbe_l2_tn_filter *l2_tn_filter)
+{
+ int ret = 0;
+
+ ret = rte_hash_add_key(l2_tn_info->hash_handle,
+ &l2_tn_filter->key);
+
+ if (ret < 0) {
+ PMD_DRV_LOG(ERR,
+ "Failed to insert L2 tunnel filter"
+ " to hash table %d!",
+ ret)
[dpdk-dev] [PATCH v2 06/18] net/ixgbe: restore TCP SYN filter
Add support for restoring TCP SYN filter in SW.
Signed-off-by: Wenzhuo Lu
Signed-off-by: Wei Zhao
---
v2:
--change git log expression
---
drivers/net/ixgbe/ixgbe_ethdev.c | 18 ++
1 file changed, 18 insertions(+)
diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/drivers/net/ixgbe/ixgbe_ethdev.c
index e6e5f72..7281ba5 100644
--- a/drivers/net/ixgbe/ixgbe_ethdev.c
+++ b/drivers/net/ixgbe/ixgbe_ethdev.c
@@ -7899,11 +7899,29 @@ ixgbe_ethertype_filter_restore(struct rte_eth_dev *dev)
}
}
+/* restore SYN filter */
+static inline void
+ixgbe_syn_filter_restore(struct rte_eth_dev *dev)
+{
+ struct ixgbe_hw *hw = IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
+ struct ixgbe_filter_info *filter_info =
+ IXGBE_DEV_PRIVATE_TO_FILTER_INFO(dev->data->dev_private);
+ uint32_t synqf;
+
+ synqf = filter_info->syn_info;
+
+ if (synqf & IXGBE_SYN_FILTER_ENABLE) {
+ IXGBE_WRITE_REG(hw, IXGBE_SYNQF, synqf);
+ IXGBE_WRITE_FLUSH(hw);
+ }
+}
+
static int
ixgbe_filter_restore(struct rte_eth_dev *dev)
{
ixgbe_ntuple_filter_restore(dev);
ixgbe_ethertype_filter_restore(dev);
+ ixgbe_syn_filter_restore(dev);
return 0;
}
--
2.5.5
