Re: [dpdk-dev] [PATCH v5 07/11] net/virtio: implement transmit path for packed queues
> -Original Message-
> From: dev On Behalf Of Jens Freimann
> Sent: Friday, September 7, 2018 2:20 AM
> To: dev@dpdk.org
> Cc: tiwei@intel.com; maxime.coque...@redhat.com
> Subject: [dpdk-dev] [PATCH v5 07/11] net/virtio: implement transmit path
> for packed queues
>
> This implements the transmit path for devices with support for packed
> virtqueues.
>
> Add the feature bit and enable code to
> add buffers to vring and mark descriptors as available.
>
> Signed-off-by: Jens Freiman
> ---
> drivers/net/virtio/virtio_ethdev.c | 8 +-
> drivers/net/virtio/virtio_ethdev.h | 2 +
> drivers/net/virtio/virtio_rxtx.c | 113 -
> 3 files changed, 121 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/net/virtio/virtio_ethdev.c
> b/drivers/net/virtio/virtio_ethdev.c
> index ad91f7f82..d2c5755bb 100644
> --- a/drivers/net/virtio/virtio_ethdev.c
> +++ b/drivers/net/virtio/virtio_ethdev.c
> @@ -384,6 +384,8 @@ virtio_init_queue(struct rte_eth_dev *dev, uint16_t
> vtpci_queue_idx)
> vq->hw = hw;
> vq->vq_queue_index = vtpci_queue_idx;
> vq->vq_nentries = vq_size;
> +if (vtpci_packed_queue(hw))
> +vq->vq_ring.avail_wrap_counter = 1;
>
> /*
> * Reserve a memzone for vring elements @@ -1338,7 +1340,11 @@
> set_rxtx_funcs(struct rte_eth_dev *eth_dev)
> eth_dev->rx_pkt_burst = &virtio_recv_pkts;
> }
>
> -if (hw->use_inorder_tx) {
> +if (vtpci_packed_queue(hw)) {
> +PMD_INIT_LOG(INFO, "virtio: using virtio 1.1 Tx path on
> port %u",
> +eth_dev->data->port_id);
> +eth_dev->tx_pkt_burst = virtio_xmit_pkts_packed;
> +} else if (hw->use_inorder_tx) {
> PMD_INIT_LOG(INFO, "virtio: using inorder Tx path on
> port %u",
> eth_dev->data->port_id);
> eth_dev->tx_pkt_burst = virtio_xmit_pkts_inorder; diff --git
> a/drivers/net/virtio/virtio_ethdev.h b/drivers/net/virtio/virtio_ethdev.h
> index b726ad108..04161b461 100644
> --- a/drivers/net/virtio/virtio_ethdev.h
> +++ b/drivers/net/virtio/virtio_ethdev.h
> @@ -79,6 +79,8 @@ uint16_t virtio_recv_mergeable_pkts_inorder(void
> *rx_queue,
>
> uint16_t virtio_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts,
> uint16_t nb_pkts);
> +uint16_t virtio_xmit_pkts_packed(void *tx_queue, struct rte_mbuf
> **tx_pkts,
> +uint16_t nb_pkts);
>
> uint16_t virtio_xmit_pkts_inorder(void *tx_queue, struct rte_mbuf
> **tx_pkts,
> uint16_t nb_pkts);
> diff --git a/drivers/net/virtio/virtio_rxtx.c
> b/drivers/net/virtio/virtio_rxtx.c
> index eb891433e..12787070e 100644
> --- a/drivers/net/virtio/virtio_rxtx.c
> +++ b/drivers/net/virtio/virtio_rxtx.c
> @@ -38,6 +38,112 @@
> #define VIRTIO_DUMP_PACKET(m, len) do { } while (0) #endif
>
> +
> +/* Cleanup from completed transmits. */ static void
> +virtio_xmit_cleanup_packed(struct virtqueue *vq) {
> +uint16_t idx;
> +uint16_t size = vq->vq_nentries;
> +struct vring_desc_packed *desc = vq->vq_ring.desc_packed;
> +struct vq_desc_extra *dxp;
> +
> +idx = vq->vq_used_cons_idx;
> +while (desc_is_used(&desc[idx], &vq->vq_ring) &&
> + vq->vq_free_cnt < size) {
> +dxp = &vq->vq_descx[idx];
> +vq->vq_free_cnt += dxp->ndescs;
> +idx = dxp->ndescs;
> +idx = idx >= size ? idx - size : idx;
> +}
> +}
> +
> +uint16_t
> +virtio_xmit_pkts_packed(void *tx_queue, struct rte_mbuf **tx_pkts,
> + uint16_t nb_pkts)
> +{
> +struct virtnet_tx *txvq = tx_queue;
> +struct virtqueue *vq = txvq->vq;
> +uint16_t i;
> +struct vring_desc_packed *desc = vq->vq_ring.desc_packed;
> +uint16_t idx, prev;
> +struct vq_desc_extra *dxp;
> +
> +if (unlikely(nb_pkts < 1))
> +return nb_pkts;
> +
> +PMD_TX_LOG(DEBUG, "%d packets to xmit", nb_pkts);
> +
> +if (likely(vq->vq_free_cnt < vq->vq_free_thresh))
> +virtio_xmit_cleanup_packed(vq);
> +
> +for (i = 0; i < nb_pkts; i++) {
> +struct rte_mbuf *txm = tx_pkts[i];
> +struct virtio_tx_region *txr = txvq->virtio_net_hdr_mz->addr;
> +uint16_t head_idx;
> +int wrap_counter;
> +int descs_used;
> +
> +if (unlikely(txm->nb_segs + 1 > vq->vq_free_cnt)) {
> +virtio_xmit_cleanup_packed(vq);
> +
> +if (unlikely(txm->nb_segs + 1 > vq->vq_free_cnt)) {
> +PMD_TX_LOG(ERR,
> + "No free tx descriptors to transmit");
> +break;
> +}
> +}
> +
> +txvq->stats.bytes += txm->pkt_len;
> +
> +vq->vq_free_cnt -= txm->nb_segs + 1;
> +
> +wrap_counter = vq->vq_ring.avail_wrap_counter;
> +idx = vq->vq_avail_idx;
> +head_idx = idx;
> +
> +dxp = &vq->vq_descx[idx];
> +if (dxp->cookie != NULL)
> +rte_pktmbuf_free(dxp->cookie);
> +dxp->cookie = txm;
> +
> +desc[idx].addr = txvq->virtio_net_hdr_mem +
> + RTE_PTR_DIFF(&txr[idx].tx_hdr, txr);
> +desc[idx].len = vq->hw->vtnet_hdr_size;
> +desc[idx].flags = VRING_DESC_F_NEXT |
> +VRING_DESC_F_AVAIL(vq-
> >vq_ring.avail_wrap_counter) |
> +VRING_DESC_F_USED(!vq-
> >vq_ring.avail_wrap_counter);
> +descs_used = 1;
> +
> +do {
> +idx = update_pq_avail_index(vq);
> +desc[idx].addr =
> VIRTIO_MBUF_DATA_DMA_ADDR(txm, vq);
> +desc[idx].len = txm->data_len;
> +desc[idx].flags = VRING_DESC_F_NEXT |
> +VRING_DESC_F_AVAIL(vq-
> >vq_ring.avail_wrap_counter) |
> +
[dpdk-dev] [PATCH v1] app/test-crypto-perf: fix double allocation of memory
The field, 'cipher_iv.data' is allocated twice when cipher is not null.
Ideally the allocation should depend only on the field
'cperf_options.cipher_iv_sz'. This will make sure this code path gets
valid for ciphers which doesn't require IV.
Fixes: 0fbd75a99fc9 ("cryptodev: move IV parameters to session")
Signed-off-by: Akash Saxena
Signed-off-by: Anoob Joseph
---
v1:
* Fixed typo
app/test-crypto-perf/cperf_test_vectors.c | 22 +-
1 file changed, 9 insertions(+), 13 deletions(-)
diff --git a/app/test-crypto-perf/cperf_test_vectors.c
b/app/test-crypto-perf/cperf_test_vectors.c
index 907a995..1af9524 100644
--- a/app/test-crypto-perf/cperf_test_vectors.c
+++ b/app/test-crypto-perf/cperf_test_vectors.c
@@ -419,13 +419,19 @@ cperf_test_vector_get_dummy(struct cperf_options *options)
t_vec->cipher_key.length = 0;
t_vec->ciphertext.data = plaintext;
t_vec->cipher_key.data = NULL;
- t_vec->cipher_iv.data = NULL;
} else {
t_vec->cipher_key.length = options->cipher_key_sz;
t_vec->ciphertext.data = ciphertext;
t_vec->cipher_key.data = cipher_key;
- t_vec->cipher_iv.data = rte_malloc(NULL,
options->cipher_iv_sz,
- 16);
+ }
+
+ /* Init IV data ptr */
+ t_vec->cipher_iv.data = NULL;
+
+ if (options->cipher_iv_sz != 0) {
+ /* Set IV parameters */
+ t_vec->cipher_iv.data = rte_malloc(NULL,
+ options->cipher_iv_sz, 16);
if (t_vec->cipher_iv.data == NULL) {
rte_free(t_vec);
return NULL;
@@ -433,17 +439,7 @@ cperf_test_vector_get_dummy(struct cperf_options *options)
memcpy(t_vec->cipher_iv.data, iv,
options->cipher_iv_sz);
}
t_vec->ciphertext.length = options->max_buffer_size;
-
- /* Set IV parameters */
- t_vec->cipher_iv.data = rte_malloc(NULL, options->cipher_iv_sz,
- 16);
- if (options->cipher_iv_sz && t_vec->cipher_iv.data == NULL) {
- rte_free(t_vec);
- return NULL;
- }
- memcpy(t_vec->cipher_iv.data, iv, options->cipher_iv_sz);
t_vec->cipher_iv.length = options->cipher_iv_sz;
-
t_vec->data.cipher_offset = 0;
t_vec->data.cipher_length = options->max_buffer_size;
--
2.7.4
Re: [dpdk-dev] [PATCH] vhost: fix vhost interrupt support
Hi Tiwei,
On 09/05/2018 01:55 AM, Tiwei Bie wrote:
When VIRTIO_RING_F_EVENT_IDX is negotiated, we need to
update the avail event to enable the notification.
Fixes: 3f8ff12821e4 ("vhost: support interrupt mode")
Cc: sta...@dpdk.org
Signed-off-by: Tiwei Bie
---
lib/librte_vhost/Makefile | 1 +
lib/librte_vhost/vhost.c | 18 --
lib/librte_vhost/vhost.h | 2 ++
3 files changed, 15 insertions(+), 6 deletions(-)
diff --git a/lib/librte_vhost/Makefile b/lib/librte_vhost/Makefile
index de431fbb7..531cf4832 100644
--- a/lib/librte_vhost/Makefile
+++ b/lib/librte_vhost/Makefile
@@ -13,6 +13,7 @@ LIBABIVER := 4
CFLAGS += -DALLOW_EXPERIMENTAL_API
CFLAGS += $(WERROR_FLAGS) -I$(SRCDIR) -O3 -D_FILE_OFFSET_BITS=64
CFLAGS += -I vhost_user
+CFLAGS += -fno-strict-aliasing
I'm not clear why this is needed looking at the code below,
could you please explain?
LDLIBS += -lpthread
ifeq ($(CONFIG_RTE_LIBRTE_VHOST_NUMA),y)
diff --git a/lib/librte_vhost/vhost.c b/lib/librte_vhost/vhost.c
index 3c9be10a0..88b1781d5 100644
--- a/lib/librte_vhost/vhost.c
+++ b/lib/librte_vhost/vhost.c
@@ -646,12 +646,18 @@ rte_vhost_avail_entries(int vid, uint16_t queue_id)
}
static inline void
-vhost_enable_notify_split(struct vhost_virtqueue *vq, int enable)
+vhost_enable_notify_split(struct virtio_net *dev,
+ struct vhost_virtqueue *vq, int enable)
{
- if (enable)
- vq->used->flags &= ~VRING_USED_F_NO_NOTIFY;
- else
- vq->used->flags |= VRING_USED_F_NO_NOTIFY;
+ if (!(dev->features & (1ULL << VIRTIO_RING_F_EVENT_IDX))) {
+ if (enable)
+ vq->used->flags &= ~VRING_USED_F_NO_NOTIFY;
+ else
+ vq->used->flags |= VRING_USED_F_NO_NOTIFY;
+ } else {
+ if (enable)
+ vhost_avail_event(vq) = vq->last_avail_idx;
+ }
}
static inline void
@@ -689,7 +695,7 @@ rte_vhost_enable_guest_notification(int vid, uint16_t
queue_id, int enable)
if (vq_is_packed(dev))
vhost_enable_notify_packed(dev, vq, enable);
else
- vhost_enable_notify_split(vq, enable);
+ vhost_enable_notify_split(dev, vq, enable);
return 0;
}
diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
index 760a09c0d..25ffd7614 100644
--- a/lib/librte_vhost/vhost.h
+++ b/lib/librte_vhost/vhost.h
@@ -648,6 +648,8 @@ vhost_iova_to_vva(struct virtio_net *dev, struct
vhost_virtqueue *vq,
return __vhost_iova_to_vva(dev, vq, iova, len, perm);
}
+#define vhost_avail_event(vr) \
+ (*(volatile uint16_t*)&(vr)->used->ring[(vr)->size])
#define vhost_used_event(vr) \
(*(volatile uint16_t*)&(vr)->avail->ring[(vr)->size])
Re: [dpdk-dev] [PATCH] build: add PPC64 Meson build
Checked on ppc64, the build process is successful.
> -Original Message-
> From: Luca Boccassi [mailto:bl...@debian.org]
> Sent: 2018年9月8日 2:35
> To: dev@dpdk.org
> Cc: chao...@linux.vnet.ibm.com; christian.ehrha...@canonical.com;
> bruce.richard...@intel.com; tho...@monjalon.net
> Subject: [PATCH] build: add PPC64 Meson build
>
> This has been only build-tested for now, on a native ppc64el POWER8E
machine
> running Debian sid.
>
> Signed-off-by: Luca Boccassi
> ---
> The build box cannot be used to run DPDK as it doesn't have supported NICs
> and root access. Would be great if someone could run-test it, but at this
point I
> think build support is enough to get started.
>
> config/meson.build | 8
> config/ppc_64/meson.build| 15
> +++
> lib/librte_eal/common/arch/ppc_64/meson.build| 5 +
> .../common/include/arch/ppc_64/meson.build | 16
>
> 4 files changed, 44 insertions(+)
> create mode 100644 config/ppc_64/meson.build create mode 100644
> lib/librte_eal/common/arch/ppc_64/meson.build
> create mode 100644 lib/librte_eal/common/include/arch/ppc_64/meson.build
>
> diff --git a/config/meson.build b/config/meson.build index
> 4d755323f4..8e87b344c2 100644
> --- a/config/meson.build
> +++ b/config/meson.build
> @@ -9,7 +9,13 @@ else
> endif
> dpdk_conf.set('RTE_MACHINE', machine)
> machine_args = []
> +# ppc64 does not support -march=native
> +if host_machine.cpu_family().startswith('ppc') and machine == 'native'
> +machine_args += '-mcpu=' + machine
> +machine_args += '-mtune=' + machine
> +else
> machine_args += '-march=' + machine
> +endif
>
> toolchain = cc.get_id()
> dpdk_conf.set_quoted('RTE_TOOLCHAIN', toolchain) @@ -84,6 +90,8 @@ if
> host_machine.cpu_family().startswith('x86')
> arch_subdir = 'x86'
> elif host_machine.cpu_family().startswith('arm') or
> host_machine.cpu_family().startswith('aarch')
> arch_subdir = 'arm'
> +elif host_machine.cpu_family().startswith('ppc')
> + arch_subdir = 'ppc_64'
> endif
> subdir(arch_subdir)
> dpdk_conf.set('RTE_COMPILE_TIME_CPUFLAGS',
> ','.join(compile_time_cpuflags)) diff --git a/config/ppc_64/meson.build
> b/config/ppc_64/meson.build new file mode 100644 index
> 00..d6faa7d64f
> --- /dev/null
> +++ b/config/ppc_64/meson.build
> @@ -0,0 +1,15 @@
> +# SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2018 Luca
> +Boccassi
> +
> +# for checking defines we need to use the correct compiler flags
> +march_opt = '-march=@0@'.format(machine)
> +
> +dpdk_conf.set('RTE_ARCH', 'ppc_64')
> +dpdk_conf.set('RTE_ARCH_PPC_64', 1)
> +dpdk_conf.set('RTE_ARCH_64', 1)
> +
> +# overrides specific to ppc64
> +dpdk_conf.set('RTE_MAX_LCORE', 256)
> +dpdk_conf.set('RTE_MAX_NUMA_NODES', 32)
> +dpdk_conf.set('RTE_CACHE_LINE_SIZE', 128)
> +dpdk_conf.set('RTE_MAX_LCORE', 256)
> diff --git a/lib/librte_eal/common/arch/ppc_64/meson.build
> b/lib/librte_eal/common/arch/ppc_64/meson.build
> new file mode 100644
> index 00..40b3dc533a
> --- /dev/null
> +++ b/lib/librte_eal/common/arch/ppc_64/meson.build
> @@ -0,0 +1,5 @@
> +# SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2018 Luca
> +Boccassi
> +
> +eal_common_arch_sources = files('rte_cpuflags.c',
> + 'rte_cycles.c', 'rte_hypervisor.c')
> diff --git a/lib/librte_eal/common/include/arch/ppc_64/meson.build
> b/lib/librte_eal/common/include/arch/ppc_64/meson.build
> new file mode 100644
> index 00..00f9611768
> --- /dev/null
> +++ b/lib/librte_eal/common/include/arch/ppc_64/meson.build
> @@ -0,0 +1,16 @@
> +# SPDX-License-Identifier: BSD-3-Clause # Copyright(c) 2018 Luca
> +Boccassi
> +
> +install_headers(
> + 'rte_atomic.h',
> + 'rte_byteorder.h',
> + 'rte_cpuflags.h',
> + 'rte_cycles.h',
> + 'rte_io.h',
> + 'rte_memcpy.h',
> + 'rte_pause.h',
> + 'rte_prefetch.h',
> + 'rte_rwlock.h',
> + 'rte_spinlock.h',
> + 'rte_vect.h',
> + subdir: get_option('include_subdir_arch'))
> --
> 2.18.0
Acked-by: Chao Zhu
Re: [dpdk-dev] [PATCH] vhost: fix vhost interrupt support
Hi Maxime,
On Mon, Sep 10, 2018 at 09:22:00AM +0200, Maxime Coquelin wrote:
> Hi Tiwei,
>
> On 09/05/2018 01:55 AM, Tiwei Bie wrote:
> > When VIRTIO_RING_F_EVENT_IDX is negotiated, we need to
> > update the avail event to enable the notification.
> >
> > Fixes: 3f8ff12821e4 ("vhost: support interrupt mode")
> > Cc: sta...@dpdk.org
> >
> > Signed-off-by: Tiwei Bie
> > ---
> > lib/librte_vhost/Makefile | 1 +
> > lib/librte_vhost/vhost.c | 18 --
> > lib/librte_vhost/vhost.h | 2 ++
> > 3 files changed, 15 insertions(+), 6 deletions(-)
> >
> > diff --git a/lib/librte_vhost/Makefile b/lib/librte_vhost/Makefile
> > index de431fbb7..531cf4832 100644
> > --- a/lib/librte_vhost/Makefile
> > +++ b/lib/librte_vhost/Makefile
> > @@ -13,6 +13,7 @@ LIBABIVER := 4
> > CFLAGS += -DALLOW_EXPERIMENTAL_API
> > CFLAGS += $(WERROR_FLAGS) -I$(SRCDIR) -O3 -D_FILE_OFFSET_BITS=64
> > CFLAGS += -I vhost_user
> > +CFLAGS += -fno-strict-aliasing
>
> I'm not clear why this is needed looking at the code below,
> could you please explain?
Without this, we will get below build error:
lib/librte_vhost/vhost.c: In function ‘vhost_enable_notify_split’:
lib/librte_vhost/vhost.h:656:4: error: dereferencing type-punned pointer will
break strict-aliasing rules [-Werror=strict-aliasing]
(*(volatile uint16_t*)&(vr)->used->ring[(vr)->size])
^
lib/librte_vhost/vhost.c:659:4: note: in expansion of macro ‘vhost_avail_event’
vhost_avail_event(vq) = vq->last_avail_idx;
^
cc1: all warnings being treated as errors
>
> > LDLIBS += -lpthread
> > ifeq ($(CONFIG_RTE_LIBRTE_VHOST_NUMA),y)
> > diff --git a/lib/librte_vhost/vhost.c b/lib/librte_vhost/vhost.c
> > index 3c9be10a0..88b1781d5 100644
> > --- a/lib/librte_vhost/vhost.c
> > +++ b/lib/librte_vhost/vhost.c
> > @@ -646,12 +646,18 @@ rte_vhost_avail_entries(int vid, uint16_t queue_id)
> > }
> > static inline void
> > -vhost_enable_notify_split(struct vhost_virtqueue *vq, int enable)
> > +vhost_enable_notify_split(struct virtio_net *dev,
> > + struct vhost_virtqueue *vq, int enable)
> > {
> > - if (enable)
> > - vq->used->flags &= ~VRING_USED_F_NO_NOTIFY;
> > - else
> > - vq->used->flags |= VRING_USED_F_NO_NOTIFY;
> > + if (!(dev->features & (1ULL << VIRTIO_RING_F_EVENT_IDX))) {
> > + if (enable)
> > + vq->used->flags &= ~VRING_USED_F_NO_NOTIFY;
> > + else
> > + vq->used->flags |= VRING_USED_F_NO_NOTIFY;
> > + } else {
> > + if (enable)
> > + vhost_avail_event(vq) = vq->last_avail_idx;
> > + }
> > }
> > static inline void
> > @@ -689,7 +695,7 @@ rte_vhost_enable_guest_notification(int vid, uint16_t
> > queue_id, int enable)
> > if (vq_is_packed(dev))
> > vhost_enable_notify_packed(dev, vq, enable);
> > else
> > - vhost_enable_notify_split(vq, enable);
> > + vhost_enable_notify_split(dev, vq, enable);
> > return 0;
> > }
> > diff --git a/lib/librte_vhost/vhost.h b/lib/librte_vhost/vhost.h
> > index 760a09c0d..25ffd7614 100644
> > --- a/lib/librte_vhost/vhost.h
> > +++ b/lib/librte_vhost/vhost.h
> > @@ -648,6 +648,8 @@ vhost_iova_to_vva(struct virtio_net *dev, struct
> > vhost_virtqueue *vq,
> > return __vhost_iova_to_vva(dev, vq, iova, len, perm);
> > }
> > +#define vhost_avail_event(vr) \
> > + (*(volatile uint16_t*)&(vr)->used->ring[(vr)->size])
> > #define vhost_used_event(vr) \
> > (*(volatile uint16_t*)&(vr)->avail->ring[(vr)->size])
> >
Re: [dpdk-dev] [PATCH] vhost: fix vhost interrupt support
On 09/10/2018 09:36 AM, Tiwei Bie wrote:
Hi Maxime,
On Mon, Sep 10, 2018 at 09:22:00AM +0200, Maxime Coquelin wrote:
Hi Tiwei,
On 09/05/2018 01:55 AM, Tiwei Bie wrote:
When VIRTIO_RING_F_EVENT_IDX is negotiated, we need to
update the avail event to enable the notification.
Fixes: 3f8ff12821e4 ("vhost: support interrupt mode")
Cc: sta...@dpdk.org
Signed-off-by: Tiwei Bie
---
lib/librte_vhost/Makefile | 1 +
lib/librte_vhost/vhost.c | 18 --
lib/librte_vhost/vhost.h | 2 ++
3 files changed, 15 insertions(+), 6 deletions(-)
diff --git a/lib/librte_vhost/Makefile b/lib/librte_vhost/Makefile
index de431fbb7..531cf4832 100644
--- a/lib/librte_vhost/Makefile
+++ b/lib/librte_vhost/Makefile
@@ -13,6 +13,7 @@ LIBABIVER := 4
CFLAGS += -DALLOW_EXPERIMENTAL_API
CFLAGS += $(WERROR_FLAGS) -I$(SRCDIR) -O3 -D_FILE_OFFSET_BITS=64
CFLAGS += -I vhost_user
+CFLAGS += -fno-strict-aliasing
I'm not clear why this is needed looking at the code below,
could you please explain?
Without this, we will get below build error:
lib/librte_vhost/vhost.c: In function ‘vhost_enable_notify_split’:
lib/librte_vhost/vhost.h:656:4: error: dereferencing type-punned pointer will
break strict-aliasing rules [-Werror=strict-aliasing]
(*(volatile uint16_t*)&(vr)->used->ring[(vr)->size])
^
lib/librte_vhost/vhost.c:659:4: note: in expansion of macro ‘vhost_avail_event’
vhost_avail_event(vq) = vq->last_avail_idx;
^
cc1: all warnings being treated as errors
OK, thanks for the info.
Reviewed-by: Maxime Coquelin
Maxime
Re: [dpdk-dev] [PATCH v2] ethdev: make default behavior CRC strip on Rx
On 09/04/2018 12:12 PM, Ferruh Yigit wrote:
diff --git a/drivers/net/vhost/rte_eth_vhost.c
b/drivers/net/vhost/rte_eth_vhost.c
index e58f32211..aa6052221 100644
--- a/drivers/net/vhost/rte_eth_vhost.c
+++ b/drivers/net/vhost/rte_eth_vhost.c
@@ -1070,8 +1070,7 @@ eth_dev_info(struct rte_eth_dev *dev,
dev_info->tx_offload_capa = DEV_TX_OFFLOAD_MULTI_SEGS |
DEV_TX_OFFLOAD_VLAN_INSERT;
- dev_info->rx_offload_capa = DEV_RX_OFFLOAD_VLAN_STRIP |
- DEV_RX_OFFLOAD_CRC_STRIP;
+ dev_info->rx_offload_capa = DEV_RX_OFFLOAD_VLAN_STRIP;
}
static int
diff --git a/drivers/net/virtio/virtio_ethdev.c
b/drivers/net/virtio/virtio_ethdev.c
index 614357da7..b81df0a99 100644
--- a/drivers/net/virtio/virtio_ethdev.c
+++ b/drivers/net/virtio/virtio_ethdev.c
@@ -2166,8 +2166,7 @@ virtio_dev_info_get(struct rte_eth_dev *dev, struct
rte_eth_dev_info *dev_info)
dev_info->max_mac_addrs = VIRTIO_MAX_MAC_ADDRS;
host_features = VTPCI_OPS(hw)->get_features(hw);
- dev_info->rx_offload_capa = DEV_RX_OFFLOAD_VLAN_STRIP |
- DEV_RX_OFFLOAD_CRC_STRIP;
+ dev_info->rx_offload_capa = DEV_RX_OFFLOAD_VLAN_STRIP;
if (host_features & (1ULL << VIRTIO_NET_F_GUEST_CSUM)) {
dev_info->rx_offload_capa |=
DEV_RX_OFFLOAD_TCP_CKSUM |
For Vhost & Virtio PMDs:
Reviewed-by: Maxime Coquelin
Thanks,
Maxime
Re: [dpdk-dev] [RFC] ethdev: complete closing to free all resources
On 09/08/2018 02:39 AM, Thomas Monjalon wrote:
After closing a port, it cannot be restarted.
So there is no reason to not free all associated resources.
The last step was done with rte_eth_dev_detach() which is deprecated.
Instead of removing the associated rte_device, the driver should check
if no more port (ethdev, cryptodev, etc) is still open for the device.
Then the device resources can be freed by the driver inside the
dev_close() driver callback operation.
The last ethdev freeing (dev_private and final release), which were done
by rte_eth_dev_detach(), are now done at the end of rte_eth_dev_close().
For me, it sounds more logical to kill dev_close and keep detach.
IMHO, dev_close is artificial and hardly useful. detach is a local pair
to attach.
Anyway it requires update of all drivers as I understand.
Signed-off-by: Thomas Monjalon
---
This patch contains only the change in the close function as RFC.
This idea was presented at Dublin during the "hotplug talk".
---
lib/librte_ethdev/rte_ethdev.c | 5 +
lib/librte_ethdev/rte_ethdev.h | 5 +++--
2 files changed, 8 insertions(+), 2 deletions(-)
diff --git a/lib/librte_ethdev/rte_ethdev.c b/lib/librte_ethdev/rte_ethdev.c
index 4c3202505..071fcbd23 100644
--- a/lib/librte_ethdev/rte_ethdev.c
+++ b/lib/librte_ethdev/rte_ethdev.c
@@ -1358,6 +1358,7 @@ void
rte_eth_dev_close(uint16_t port_id)
{
struct rte_eth_dev *dev;
+ struct rte_bus *bus;
RTE_ETH_VALID_PORTID_OR_RET(port_id);
dev = &rte_eth_devices[port_id];
@@ -1372,6 +1373,10 @@ rte_eth_dev_close(uint16_t port_id)
dev->data->nb_tx_queues = 0;
rte_free(dev->data->tx_queues);
dev->data->tx_queues = NULL;
+
+ rte_free(dev->data->dev_private);
+
+ rte_eth_dev_release_port(dev);
}
int
diff --git a/lib/librte_ethdev/rte_ethdev.h b/lib/librte_ethdev/rte_ethdev.h
index 7070e9ab4..37a757a7a 100644
--- a/lib/librte_ethdev/rte_ethdev.h
+++ b/lib/librte_ethdev/rte_ethdev.h
@@ -1797,8 +1797,9 @@ int rte_eth_dev_set_link_down(uint16_t port_id);
/**
* Close a stopped Ethernet device. The device cannot be restarted!
- * The function frees all resources except for needed by the
- * closed state. To free these resources, call rte_eth_dev_detach().
+ * The function frees all port resources.
+ * If there is no more port associated with the underlying device,
+ * the driver should free the device resources.
*
* @param port_id
* The port identifier of the Ethernet device.
Re: [dpdk-dev] [PATCH] mbuf: remove deprecated segment free functions
On 09/10/2018 08:18 AM, David Marchand wrote: __rte_mbuf_raw_free and __rte_pktmbuf_prefree_seg have been deprecated for a long time now (early 17.05), are not part of the abi and are easily replaced with existing api. Signed-off-by: David Marchand Acked-by: Andrew Rybchenko
Re: [dpdk-dev] [PATCH 2/3] mbuf: add a non fatal sanity check helper
On 09/10/2018 08:45 AM, David Marchand wrote: Let's add a little helper that does the same as rte_mbuf_sanity_check but without the panic. Signed-off-by: David Marchand --- <...> diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h index a50b05c64..e12a4c765 100644 --- a/lib/librte_mbuf/rte_mbuf.h +++ b/lib/librte_mbuf/rte_mbuf.h @@ -977,6 +977,29 @@ rte_mbuf_ext_refcnt_update(struct rte_mbuf_ext_shared_info *shinfo, void rte_mbuf_sanity_check(const struct rte_mbuf *m, int is_header); +/** + * Sanity checks on a mbuf. + * + * Almost like rte_mbuf_sanity_check(), but this function gives the reason + * if corruption is detected rather than panic. + * + * @param m + * The mbuf to be checked. + * @param is_header + * True if the mbuf is a packet header, false if it is a sub-segment + * of a packet (in this case, some fields like nb_segs are not checked) + * @param reason + * A reference to a string pointer where to store the reason why a mbuf is + * considered invalid. + * @return + * - 0 if no issue has been found, reason is left untouched. + * - -1 if a problem is detected, reason then points to a string describing + * the reason why the mbuf is deemed invalid. + */ +__rte_experimental +int rte_mbuf_check(const struct rte_mbuf *m, int is_header, + const char **reason); + May be it would be better to return reason as return value and if it is NULL everything is OK? <...>
Re: [dpdk-dev] [PATCH 2/3] mbuf: add a non fatal sanity check helper
On Mon, Sep 10, 2018 at 10:12 AM, Andrew Rybchenko wrote: > +/** > + * Sanity checks on a mbuf. > + * > + * Almost like rte_mbuf_sanity_check(), but this function gives the reason > + * if corruption is detected rather than panic. > + * > + * @param m > + * The mbuf to be checked. > + * @param is_header > + * True if the mbuf is a packet header, false if it is a sub-segment > + * of a packet (in this case, some fields like nb_segs are not checked) > + * @param reason > + * A reference to a string pointer where to store the reason why a mbuf > is > + * considered invalid. > + * @return > + * - 0 if no issue has been found, reason is left untouched. > + * - -1 if a problem is detected, reason then points to a string > describing > + * the reason why the mbuf is deemed invalid. > + */ > +__rte_experimental > +int rte_mbuf_check(const struct rte_mbuf *m, int is_header, > + const char **reason); > + > > > May be it would be better to return reason as return value and if it is NULL > everything is OK? This was what I had done with my first attempt. But given the name "rte_mbuf_check", having a int (used as a bool) returned makes sense to me. So no strong opinion here. -- David Marchand
Re: [dpdk-dev] [PATCH 2/3] mbuf: add a non fatal sanity check helper
On 09/10/2018 11:24 AM, David Marchand wrote: On Mon, Sep 10, 2018 at 10:12 AM, Andrew Rybchenko wrote: +/** + * Sanity checks on a mbuf. + * + * Almost like rte_mbuf_sanity_check(), but this function gives the reason + * if corruption is detected rather than panic. + * + * @param m + * The mbuf to be checked. + * @param is_header + * True if the mbuf is a packet header, false if it is a sub-segment + * of a packet (in this case, some fields like nb_segs are not checked) + * @param reason + * A reference to a string pointer where to store the reason why a mbuf is + * considered invalid. + * @return + * - 0 if no issue has been found, reason is left untouched. + * - -1 if a problem is detected, reason then points to a string describing + * the reason why the mbuf is deemed invalid. + */ +__rte_experimental +int rte_mbuf_check(const struct rte_mbuf *m, int is_header, + const char **reason); + May be it would be better to return reason as return value and if it is NULL everything is OK? This was what I had done with my first attempt. But given the name "rte_mbuf_check", having a int (used as a bool) returned makes sense to me. Yes, good point. So no strong opinion here. Me too.
Re: [dpdk-dev] [RFC] ethdev: complete closing to free all resources
10/09/2018 10:03, Andrew Rybchenko: > On 09/08/2018 02:39 AM, Thomas Monjalon wrote: > > After closing a port, it cannot be restarted. > > So there is no reason to not free all associated resources. > > > > The last step was done with rte_eth_dev_detach() which is deprecated. > > Instead of removing the associated rte_device, the driver should check > > if no more port (ethdev, cryptodev, etc) is still open for the device. > > Then the device resources can be freed by the driver inside the > > dev_close() driver callback operation. > > > > The last ethdev freeing (dev_private and final release), which were done > > by rte_eth_dev_detach(), are now done at the end of rte_eth_dev_close(). > > For me, it sounds more logical to kill dev_close and keep detach. > IMHO, dev_close is artificial and hardly useful. detach is a local pair > to attach. I don't get your point. In order to free a port, we need close + detach. We can keep only one. I choose close because: 1) attach/detach are deprecated 2) probe/close is a more obvious pair 3) we need the driver to free the lower level resources > Anyway it requires update of all drivers as I understand. Yes if we want to free all resources. But close operation in drivers can be completed in later steps.
Re: [dpdk-dev] [RFC] ethdev: complete closing to free all resources
On 09/10/2018 11:42 AM, Thomas Monjalon wrote: 10/09/2018 10:03, Andrew Rybchenko: On 09/08/2018 02:39 AM, Thomas Monjalon wrote: After closing a port, it cannot be restarted. So there is no reason to not free all associated resources. The last step was done with rte_eth_dev_detach() which is deprecated. Instead of removing the associated rte_device, the driver should check if no more port (ethdev, cryptodev, etc) is still open for the device. Then the device resources can be freed by the driver inside the dev_close() driver callback operation. The last ethdev freeing (dev_private and final release), which were done by rte_eth_dev_detach(), are now done at the end of rte_eth_dev_close(). For me, it sounds more logical to kill dev_close and keep detach. IMHO, dev_close is artificial and hardly useful. detach is a local pair to attach. I don't get your point. In order to free a port, we need close + detach. We can keep only one. I choose close because: 1) attach/detach are deprecated 2) probe/close is a more obvious pair 3) we need the driver to free the lower level resources Yes, I'm sorry I used bad terminology. We have probe/remove pair for both PCI and vdev drivers and I mean that remove is a better candidate to be kept (as a pair for probe which allocates all resources).
Re: [dpdk-dev] [PATCH] ethdev: fix missing names in Tx offload name array
On 09/06/2018 03:01 PM, Dekel Peled wrote:
Patch 5355f443 added two definitions of DEV_TX_OFFLOAD_xxx.
If new Tx offload capabilities are defined, they also must be mentioned
in rte_tx_offload_names in rte_ethdev.c file.
This patch adds the required lines in aray rte_tx_offload_names.
Fixes: 5355f4439e2e ("ethdev: introduce generic IP/UDP tunnel checksum and TSO")
Cc: xuemi...@mellanox.com
Signed-off-by: Dekel Peled
---
lib/librte_ethdev/rte_ethdev.c | 2 ++
lib/librte_ethdev/rte_ethdev.h | 4 ++--
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/lib/librte_ethdev/rte_ethdev.c b/lib/librte_ethdev/rte_ethdev.c
index 3f8de93..5004b9f 100644
--- a/lib/librte_ethdev/rte_ethdev.c
+++ b/lib/librte_ethdev/rte_ethdev.c
@@ -156,6 +156,8 @@ struct rte_eth_xstats_name_off {
RTE_TX_OFFLOAD_BIT2STR(MULTI_SEGS),
RTE_TX_OFFLOAD_BIT2STR(MBUF_FAST_FREE),
RTE_TX_OFFLOAD_BIT2STR(SECURITY),
+ RTE_TX_OFFLOAD_BIT2STR(UDP_TNL_TSO),
+ RTE_TX_OFFLOAD_BIT2STR(IP_TNL_TSO),
};
#undef RTE_TX_OFFLOAD_BIT2STR
diff --git a/lib/librte_ethdev/rte_ethdev.h b/lib/librte_ethdev/rte_ethdev.h
index fa2812b..5456ce2 100644
--- a/lib/librte_ethdev/rte_ethdev.h
+++ b/lib/librte_ethdev/rte_ethdev.h
@@ -941,18 +941,18 @@ struct rte_eth_conf {
* the same mempool and has refcnt = 1.
*/
#define DEV_TX_OFFLOAD_SECURITY 0x0002
+#define DEV_TX_OFFLOAD_UDP_TNL_TSO 0x0004
/**
* Device supports generic UDP tunneled packet TSO.
* Application must set PKT_TX_TUNNEL_UDP and other mbuf fields required
* for tunnel TSO.
*/
-#define DEV_TX_OFFLOAD_UDP_TNL_TSO 0x0004
+#define DEV_TX_OFFLOAD_IP_TNL_TSO 0x0008
/**
* Device supports generic IP tunneled packet TSO.
* Application must set PKT_TX_TUNNEL_IP and other mbuf fields required
* for tunnel TSO.
*/
-#define DEV_TX_OFFLOAD_IP_TNL_TSO 0x0008
I don't understand why it is changed. Comments should be before define.
#define RTE_ETH_DEV_CAPA_RUNTIME_RX_QUEUE_SETUP 0x0001
/**< Device supports Rx queue setup after device started*/
[dpdk-dev] [PATCH 04/37] net/sfc/base: highlight that image layout header generated
From: Andrew Jackson EF10 signed image layout header is generated from firmware sources. Signed-off-by: Andrew Jackson Signed-off-by: Andrew Rybchenko --- drivers/net/sfc/base/ef10_signed_image_layout.h | 8 1 file changed, 8 insertions(+) diff --git a/drivers/net/sfc/base/ef10_signed_image_layout.h b/drivers/net/sfc/base/ef10_signed_image_layout.h index a35d16012..c25ffe2f7 100644 --- a/drivers/net/sfc/base/ef10_signed_image_layout.h +++ b/drivers/net/sfc/base/ef10_signed_image_layout.h @@ -4,6 +4,14 @@ * All rights reserved. */ +/* + * This is NOT the original source file. Do NOT edit it. + * To update the image layout headers, please edit the copy in + * the sfregistry repo and then, in that repo, + * "make layout_headers" or "make export" to + * regenerate and export all types of headers. + */ + /* These structures define the layouts for the signed firmware image binary * saved in NVRAM. The original image is in the Cryptographic message * syntax (CMS) format which contains the bootable firmware binary plus the -- 2.17.1
[dpdk-dev] [PATCH 03/37] net/sfc/base: fix output buffer SAL annotation
From: Martin Harvey
Found by PreFAST warnings.
Fixes: 3f2f0189dd44 ("net/sfc/base: add signed image layout support")
Cc: sta...@dpdk.org
Signed-off-by: Martin Harvey
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/ef10_image.c | 3 ++-
drivers/net/sfc/base/efx.h| 3 ++-
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/drivers/net/sfc/base/ef10_image.c
b/drivers/net/sfc/base/ef10_image.c
index 0d8898762..c035e0df6 100644
--- a/drivers/net/sfc/base/ef10_image.c
+++ b/drivers/net/sfc/base/ef10_image.c
@@ -577,7 +577,8 @@ efx_check_reflash_image(
__checkReturn efx_rc_t
efx_build_signed_image_write_buffer(
- __out uint8_t *bufferp,
+ __out_bcount(buffer_size)
+ uint8_t *bufferp,
__inuint32_tbuffer_size,
__inefx_image_info_t*infop,
__out efx_image_header_t **headerpp)
diff --git a/drivers/net/sfc/base/efx.h b/drivers/net/sfc/base/efx.h
index 5108b9b1f..53cbc9858 100644
--- a/drivers/net/sfc/base/efx.h
+++ b/drivers/net/sfc/base/efx.h
@@ -1689,7 +1689,8 @@ efx_check_reflash_image(
extern __checkReturn efx_rc_t
efx_build_signed_image_write_buffer(
- __out uint8_t *bufferp,
+ __out_bcount(buffer_size)
+ uint8_t *bufferp,
__inuint32_tbuffer_size,
__inefx_image_info_t*infop,
__out efx_image_header_t **headerpp);
--
2.17.1
[dpdk-dev] [PATCH 02/37] net/sfc/base: fix invalid order of memset arguments
From: Martin Harvey
Found by PreFAST.
Fixes: 3f2f0189dd44 ("net/sfc/base: add signed image layout support")
Cc: sta...@dpdk.org
Signed-off-by: Martin Harvey
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/ef10_image.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/sfc/base/ef10_image.c
b/drivers/net/sfc/base/ef10_image.c
index 6fb7e4764..0d8898762 100644
--- a/drivers/net/sfc/base/ef10_image.c
+++ b/drivers/net/sfc/base/ef10_image.c
@@ -704,7 +704,7 @@ efx_build_signed_image_write_buffer(
* results in the layout used for the data chunks and chunk headers.
*/
/* END CSTYLED */
- memset(bufferp, buffer_size, 0xFF);
+ memset(bufferp, 0xFF, buffer_size);
EFX_STATIC_ASSERT(sizeof (chunk_hdr) == SIGNED_IMAGE_CHUNK_HDR_LEN);
memset(&chunk_hdr, 0, SIGNED_IMAGE_CHUNK_HDR_LEN);
--
2.17.1
[dpdk-dev] [PATCH 00/37] net/sfc: update base driver
Update net/sfc base driver (aka libefx). There are a number of checkpatches.sh warnings/errors because of coding style difference. Andrew Jackson (1): net/sfc/base: highlight that image layout header generated Andrew Rybchenko (1): net/sfc/base: fix build failure because of no declaration Andy Moreton (6): net/sfc/base: properly align on line continuation net/sfc/base: add space after sizeof net/sfc/base: add routine to check for hardware presence net/sfc/base: add API to inform libefx of hardware removal net/sfc/base: fix ID retrival in v3 licensing net/sfc/base: fix MAC Tx stats for less or equal to 64 bytes Gautam Dawar (1): net/sfc/base: fix out of bounds read when dereferencing sdup Ivan Malov (6): net/sfc/base: fix name of the argument to store RSS flags net/sfc/base: fix a typo in unicast filter insertion comment net/sfc/base: use simpler code to check hash algorithm type net/sfc/base: check buffer size for hash flags net/sfc/base: simplify the code to parse RSS hash type net/sfc/base: improve handling of legacy RSS hash flags Mark Spender (3): net/sfc/base: remove probes when a Tx queue is too full net/sfc/base: add information if TSO workaround is required net/sfc/base: prevent access to the NIC config before probe Martin Harvey (10): net/sfc/base: fix PreFAST warnings because of unused return net/sfc/base: fix invalid order of memset arguments net/sfc/base: fix output buffer SAL annotation net/sfc/base: fix erroneous SAL annotation for input buffers net/sfc/base: move empty efsys definitions to EFX headers net/sfc/base: refactor monitors support net/sfc/base: add generated description of sensors net/sfc/base: check size of memory to read sensors data to net/sfc/base: add API to retrieve sensor limits net/sfc/base: avoid usage of too big arrays on stack Paul Fox (1): net/sfc/base: add more definitions of partitions Richard Houldsworth (4): net/sfc/base: add buffer editing functions to boot config net/sfc/base: add accessor for default port mode net/sfc/base: generalise EF10 NVRAM buffer interface net/sfc/base: modify phy caps to indicate FEC request Vijay Srivastava (4): net/sfc/base: fix outer IPID field in TSO option descriptors net/sfc/base: add check for TUNNEL module in NIC reset API net/sfc/base: add support to get active FEC type net/sfc/base: add helper API to make Geneve filter spec drivers/net/sfc/base/ef10_ev.c| 38 +- drivers/net/sfc/base/ef10_filter.c| 64 +- drivers/net/sfc/base/ef10_image.c | 5 +- drivers/net/sfc/base/ef10_impl.h | 62 +- drivers/net/sfc/base/ef10_intr.c | 5 +- drivers/net/sfc/base/ef10_mac.c | 24 +- drivers/net/sfc/base/ef10_nic.c | 155 ++-- drivers/net/sfc/base/ef10_nvram.c | 161 +++- drivers/net/sfc/base/ef10_phy.c | 93 ++- drivers/net/sfc/base/ef10_rx.c| 89 +-- .../net/sfc/base/ef10_signed_image_layout.h | 8 + drivers/net/sfc/base/ef10_tx.c| 38 +- drivers/net/sfc/base/efx.h| 305 +-- drivers/net/sfc/base/efx_annote.h | 103 +++ drivers/net/sfc/base/efx_bootcfg.c| 641 +-- drivers/net/sfc/base/efx_filter.c | 90 ++- drivers/net/sfc/base/efx_impl.h | 10 +- drivers/net/sfc/base/efx_lic.c| 91 +-- drivers/net/sfc/base/efx_mcdi.c | 88 +-- drivers/net/sfc/base/efx_mcdi.h | 15 + drivers/net/sfc/base/efx_mon.c| 745 -- drivers/net/sfc/base/efx_nic.c| 54 +- drivers/net/sfc/base/efx_nvram.c | 71 +- drivers/net/sfc/base/efx_phy.c| 31 + drivers/net/sfc/base/efx_port.c | 2 +- drivers/net/sfc/base/efx_rx.c | 225 +++--- drivers/net/sfc/base/efx_tunnel.c | 6 +- drivers/net/sfc/base/efx_tx.c | 19 +- drivers/net/sfc/base/hunt_nic.c | 6 +- drivers/net/sfc/base/mc_driver_pcol_strs.h| 102 +++ drivers/net/sfc/base/mcdi_mon.c | 381 + drivers/net/sfc/base/mcdi_mon.h | 5 + drivers/net/sfc/base/medford2_nic.c | 5 +- drivers/net/sfc/base/medford_nic.c| 5 +- drivers/net/sfc/base/siena_mac.c | 9 +- drivers/net/sfc/base/siena_nic.c | 5 +- drivers/net/sfc/base/siena_nvram.c| 5 +- drivers/net/sfc/base/siena_phy.c | 28 +- drivers/net/sfc/efsys.h | 38 - drivers/net/sfc/sfc_rx.c | 2 +- 40 files changed, 2886 insertions(+), 943 deletions(-) create mode 100644 drivers/net/sfc/base/efx_annote.h create mode 100644 drivers/net/sfc/base/mc_driver_pcol_strs.h -- 2.17.1
[dpdk-dev] [PATCH 06/37] net/sfc/base: properly align on line continuation
From: Andy Moreton
Fixes: 19b64c6ac35f ("net/sfc/base: import libefx base")
Cc: sta...@dpdk.org
Signed-off-by: Andy Moreton
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/efx_nic.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/sfc/base/efx_nic.c b/drivers/net/sfc/base/efx_nic.c
index 6c162e035..6314ae2ff 100644
--- a/drivers/net/sfc/base/efx_nic.c
+++ b/drivers/net/sfc/base/efx_nic.c
@@ -559,7 +559,7 @@ efx_nic_reset(
*/
mod_flags = enp->en_mod_flags;
mod_flags &= ~(EFX_MOD_MCDI | EFX_MOD_PROBE | EFX_MOD_NVRAM |
- EFX_MOD_VPD | EFX_MOD_MON);
+ EFX_MOD_VPD | EFX_MOD_MON);
EFSYS_ASSERT3U(mod_flags, ==, 0);
if (mod_flags != 0) {
rc = EINVAL;
--
2.17.1
[dpdk-dev] [PATCH 05/37] net/sfc/base: fix erroneous SAL annotation for input buffers
From: Martin Harvey
Fixes: d96a34d165b1 ("net/sfc/base: import NVRAM support")
Cc: sta...@dpdk.org
Signed-off-by: Martin Harvey
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/ef10_impl.h | 2 +-
drivers/net/sfc/base/ef10_nvram.c | 2 +-
drivers/net/sfc/base/efx_impl.h | 2 +-
drivers/net/sfc/base/efx_nvram.c | 2 +-
4 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/net/sfc/base/ef10_impl.h b/drivers/net/sfc/base/ef10_impl.h
index 4751faf16..fb0d98875 100644
--- a/drivers/net/sfc/base/ef10_impl.h
+++ b/drivers/net/sfc/base/ef10_impl.h
@@ -453,7 +453,7 @@ ef10_nvram_partn_write(
__inefx_nic_t *enp,
__inuint32_t partn,
__inunsigned int offset,
- __out_bcount(size) caddr_t data,
+ __in_bcount(size) caddr_t data,
__insize_t size);
extern __checkReturn efx_rc_t
diff --git a/drivers/net/sfc/base/ef10_nvram.c
b/drivers/net/sfc/base/ef10_nvram.c
index 0d885ccdf..46838dd75 100644
--- a/drivers/net/sfc/base/ef10_nvram.c
+++ b/drivers/net/sfc/base/ef10_nvram.c
@@ -2000,7 +2000,7 @@ ef10_nvram_partn_write(
__inefx_nic_t *enp,
__inuint32_t partn,
__inunsigned int offset,
- __out_bcount(size) caddr_t data,
+ __in_bcount(size) caddr_t data,
__insize_t size)
{
size_t chunk;
diff --git a/drivers/net/sfc/base/efx_impl.h b/drivers/net/sfc/base/efx_impl.h
index 548834f90..637e31e0c 100644
--- a/drivers/net/sfc/base/efx_impl.h
+++ b/drivers/net/sfc/base/efx_impl.h
@@ -583,7 +583,7 @@ efx_mcdi_nvram_write(
__inefx_nic_t *enp,
__inuint32_t partn,
__inuint32_t offset,
- __out_bcount(size) caddr_t data,
+ __in_bcount(size) caddr_t data,
__insize_t size);
__checkReturn efx_rc_t
diff --git a/drivers/net/sfc/base/efx_nvram.c b/drivers/net/sfc/base/efx_nvram.c
index be409c3af..f3107bbb5 100644
--- a/drivers/net/sfc/base/efx_nvram.c
+++ b/drivers/net/sfc/base/efx_nvram.c
@@ -865,7 +865,7 @@ efx_mcdi_nvram_write(
__inefx_nic_t *enp,
__inuint32_t partn,
__inuint32_t offset,
- __out_bcount(size) caddr_t data,
+ __in_bcount(size) caddr_t data,
__insize_t size)
{
efx_mcdi_req_t req;
--
2.17.1
[dpdk-dev] [PATCH 01/37] net/sfc/base: fix PreFAST warnings because of unused return
From: Martin Harvey
Fixes: 19b64c6ac35f ("net/sfc/base: import libefx base")
Fixes: d96a34d165b1 ("net/sfc/base: import NVRAM support")
Fixes: e7cd430c864f ("net/sfc/base: import SFN7xxx family support")
Cc: sta...@dpdk.org
Signed-off-by: Martin Harvey
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/ef10_filter.c | 47 --
drivers/net/sfc/base/ef10_mac.c| 2 +-
drivers/net/sfc/base/ef10_nic.c| 4 +--
drivers/net/sfc/base/ef10_nvram.c | 4 +--
drivers/net/sfc/base/ef10_tx.c | 8 +++--
drivers/net/sfc/base/efx_port.c| 2 +-
6 files changed, 49 insertions(+), 18 deletions(-)
diff --git a/drivers/net/sfc/base/ef10_filter.c
b/drivers/net/sfc/base/ef10_filter.c
index ae872853d..a4d97f99c 100644
--- a/drivers/net/sfc/base/ef10_filter.c
+++ b/drivers/net/sfc/base/ef10_filter.c
@@ -1144,12 +1144,15 @@ ef10_filter_insert_unicast(
efx_filter_spec_init_rx(&spec, EFX_FILTER_PRI_AUTO,
filter_flags,
eftp->eft_default_rxq);
- efx_filter_spec_set_eth_local(&spec, EFX_FILTER_SPEC_VID_UNSPEC, addr);
+ rc = efx_filter_spec_set_eth_local(&spec, EFX_FILTER_SPEC_VID_UNSPEC,
+ addr);
+ if (rc != 0)
+ goto fail1;
rc = ef10_filter_add_internal(enp, &spec, B_TRUE,
&eftp->eft_unicst_filter_indexes[eftp->eft_unicst_filter_count]);
if (rc != 0)
- goto fail1;
+ goto fail2;
eftp->eft_unicst_filter_count++;
EFSYS_ASSERT(eftp->eft_unicst_filter_count <=
@@ -1157,6 +1160,8 @@ ef10_filter_insert_unicast(
return (0);
+fail2:
+ EFSYS_PROBE(fail2);
fail1:
EFSYS_PROBE1(fail1, efx_rc_t, rc);
return (rc);
@@ -1175,11 +1180,13 @@ ef10_filter_insert_all_unicast(
efx_filter_spec_init_rx(&spec, EFX_FILTER_PRI_AUTO,
filter_flags,
eftp->eft_default_rxq);
- efx_filter_spec_set_uc_def(&spec);
+ rc = efx_filter_spec_set_uc_def(&spec);
+ if (rc != 0)
+ goto fail1;
rc = ef10_filter_add_internal(enp, &spec, B_TRUE,
&eftp->eft_unicst_filter_indexes[eftp->eft_unicst_filter_count]);
if (rc != 0)
- goto fail1;
+ goto fail2;
eftp->eft_unicst_filter_count++;
EFSYS_ASSERT(eftp->eft_unicst_filter_count <=
@@ -1187,6 +1194,8 @@ ef10_filter_insert_all_unicast(
return (0);
+fail2:
+ EFSYS_PROBE(fail2);
fail1:
EFSYS_PROBE1(fail1, efx_rc_t, rc);
return (rc);
@@ -1228,9 +1237,21 @@ ef10_filter_insert_multicast_list(
filter_flags,
eftp->eft_default_rxq);
- efx_filter_spec_set_eth_local(&spec,
+ rc = efx_filter_spec_set_eth_local(&spec,
EFX_FILTER_SPEC_VID_UNSPEC,
&addrs[i * EFX_MAC_ADDR_LEN]);
+ if (rc != 0) {
+ if (rollback == B_TRUE) {
+ /* Only stop upon failure if told to rollback */
+ goto rollback;
+ } else {
+ /*
+* Don't try to add a filter with a corrupt
+* specification.
+*/
+ continue;
+ }
+ }
rc = ef10_filter_add_internal(enp, &spec, B_TRUE,
&filter_index);
@@ -1253,8 +1274,12 @@ ef10_filter_insert_multicast_list(
eftp->eft_default_rxq);
EFX_MAC_BROADCAST_ADDR_SET(addr);
- efx_filter_spec_set_eth_local(&spec, EFX_FILTER_SPEC_VID_UNSPEC,
- addr);
+ rc = efx_filter_spec_set_eth_local(&spec,
+ EFX_FILTER_SPEC_VID_UNSPEC, addr);
+ if ((rc != 0) && (rollback == B_TRUE)) {
+ /* Only stop upon failure if told to rollback */
+ goto rollback;
+ }
rc = ef10_filter_add_internal(enp, &spec, B_TRUE,
&filter_index);
@@ -1302,12 +1327,14 @@ ef10_filter_insert_all_multicast(
efx_filter_spec_init_rx(&spec, EFX_FILTER_PRI_AUTO,
filter_flags,
eftp->eft_default_rxq);
- efx_filter_spec_set_mc_def(&spec);
+ rc = efx_filter_spec_set_mc_def(&spec);
+ if (rc != 0)
+ goto fail1;
rc = ef10_filter_add_internal(enp, &spec, B_TRUE,
&eftp->eft_mulcst_filter_indexes[0]);
if (rc != 0)
- goto fail1;
+ goto fail2;
eftp->eft_mulcst_filter_count = 1;
eftp->eft_using_all_mulcst = B_TRUE;
@@ -1318,6 +1345,8 @@ ef10_filter_insert_all_multicast(
return (0);
+fail2:
+ EFSYS_PROBE(fail2);
fail1:
EFSYS_PROBE1(fail1, efx_r
[dpdk-dev] [PATCH 11/37] net/sfc/base: move empty efsys definitions to EFX headers
From: Martin Harvey
Move empty definitions for platform-specific annotations from efsys.h
to EFX headers.
Signed-off-by: Martin Harvey
Signed-off-by: Andrew Lee
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/efx.h| 1 +
drivers/net/sfc/base/efx_annote.h | 101 ++
drivers/net/sfc/efsys.h | 38 ---
3 files changed, 102 insertions(+), 38 deletions(-)
create mode 100644 drivers/net/sfc/base/efx_annote.h
diff --git a/drivers/net/sfc/base/efx.h b/drivers/net/sfc/base/efx.h
index baeffd934..4c8983387 100644
--- a/drivers/net/sfc/base/efx.h
+++ b/drivers/net/sfc/base/efx.h
@@ -7,6 +7,7 @@
#ifndef_SYS_EFX_H
#define_SYS_EFX_H
+#include "efx_annote.h"
#include "efsys.h"
#include "efx_check.h"
#include "efx_phy_ids.h"
diff --git a/drivers/net/sfc/base/efx_annote.h
b/drivers/net/sfc/base/efx_annote.h
new file mode 100644
index 0..603260e22
--- /dev/null
+++ b/drivers/net/sfc/base/efx_annote.h
@@ -0,0 +1,101 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ *
+ * Copyright (c) 2018 Solarflare Communications Inc.
+ * All rights reserved.
+ */
+
+#ifndef_SYS_EFX_ANNOTE_H
+#define_SYS_EFX_ANNOTE_H
+
+#if defined(_WIN32) || defined(_WIN64)
+#defineEFX_HAVE_WINDOWS_ANNOTATIONS 1
+#else
+#defineEFX_HAVE_WINDOWS_ANNOTATIONS 0
+#endif /* defined(_WIN32) || defined(_WIN64) */
+
+#if defined(__sun)
+#defineEFX_HAVE_SOLARIS_ANNOTATIONS 1
+#else
+#defineEFX_HAVE_SOLARIS_ANNOTATIONS 0
+#endif /* defined(__sun) */
+
+#if !EFX_HAVE_WINDOWS_ANNOTATIONS
+
+/* Ignore Windows SAL annotations on other platforms */
+#define__in
+#define__in_opt
+#define__in_ecount(_n)
+#define__in_ecount_opt(_n)
+#define__in_bcount(_n)
+#define__in_bcount_opt(_n)
+
+#define__out
+#define__out_opt
+#define__out_ecount(_n)
+#define__out_ecount_opt(_n)
+#define__out_bcount(_n)
+#define__out_bcount_opt(_n)
+#define__out_bcount_part(_n, _l)
+#define__out_bcount_part_opt(_n, _l)
+
+#define__deref_out
+
+#define__inout
+#define__inout_opt
+#define__inout_ecount(_n)
+#define__inout_ecount_opt(_n)
+#define__inout_bcount(_n)
+#define__inout_bcount_opt(_n)
+#define__inout_bcount_full_opt(_n)
+
+#define__deref_out_bcount_opt(n)
+
+#define__checkReturn
+#define__success(_x)
+
+#define__drv_when(_p, _c)
+
+#endif /* !EFX_HAVE_WINDOWS_ANNOTATIONS */
+
+#if !EFX_HAVE_SOLARIS_ANNOTATIONS
+
+#if EFX_HAVE_WINDOWS_ANNOTATIONS
+
+/*
+ * Support some SunOS/Solaris style _NOTE() annotations
+ *
+ * At present with the facilities provided in the WDL and the SAL we can only
+ * easily act upon _NOTE(ARGUNUSED(arglist)) annotations.
+ *
+ * Intermediate macros to expand individual _NOTE annotation types into
+ * something the WDK or SAL can understand. They shouldn't be used directly,
+ * for example EFX_NOTE_ARGUNUSED() is only used as an intermediate step on the
+ * transformation of _NOTE(ARGUNSED(arg1, arg2)) into
+ * UNREFERENCED_PARAMETER((arg1, arg2));
+ */
+#defineEFX_NOTE_ALIGNMENT(_fname, _n)
+#defineEFX_NOTE_ARGUNUSED(...)
UNREFERENCED_PARAMETER((__VA_ARGS__));
+#defineEFX_NOTE_CONSTANTCONDITION
+#defineEFX_NOTE_CONSTCOND
+#defineEFX_NOTE_EMPTY
+#defineEFX_NOTE_FALLTHROUGH
+#defineEFX_NOTE_FALLTHRU
+#defineEFX_NOTE_LINTED(_msg)
+#defineEFX_NOTE_NOTREACHED
+#defineEFX_NOTE_PRINTFLIKE(_n)
+#defineEFX_NOTE_SCANFLIKE(_n)
+#defineEFX_NOTE_VARARGS(_n)
+
+#define_NOTE(_annotation) EFX_NOTE_ ## _annotation
+
+#else
+
+/* Ignore Solaris annotations on other platforms */
+
+#define_NOTE(_annotation)
+
+#endif /* EFX_HAVE_WINDOWS_ANNOTATIONS */
+
+#endif /* !EFX_HAVE_SOLARIS_ANNOTATIONS */
+
+#endif /* _SYS_EFX_ANNOTE_H */
diff --git a/drivers/net/sfc/efsys.h b/drivers/net/sfc/efsys.h
index b9d2df581..0b4795da1 100644
--- a/drivers/net/sfc/efsys.h
+++ b/drivers/net/sfc/efsys.h
@@ -48,10 +48,6 @@ extern "C" {
#include "efx_types.h"
-#ifndef _NOTE
-#define _NOTE(s)
-#endif
-
typedef bool boolean_t;
#ifndef B_FALSE
@@ -106,40 +102,6 @@ prefetch_read_once(const volatile void *addr)
rte_prefetch_non_temporal(addr);
}
-/* Modifiers used for Windows builds */
-#define __in
-#define __in_opt
-#define __in_ecount(_n)
-#define __in_ecount_opt(_n)
-#define __in_bcount(_n)
-#define __in_bcount_opt(_n)
-
-#define __out
-#define __out_opt
-#define __out_ecount(_n)
-#define __out_ecount_opt(_n)
-#define __out_bcount(_n)
-#define __out_bcount_opt(_n)
-#define __out_bcount_part(_n, _l)
-#define __out_bcount_part_opt(_n, _l)
-
-#define __deref_out
-
-#define __inout
-#define __inout_opt
-#define __inout_ecount(_n)
-#define __inout_ecount_opt(_n)
-#define __inout_bcount(_n)
-#define __inout_bcou
[dpdk-dev] [PATCH 08/37] net/sfc/base: fix build failure because of no declaration
Functions declared in mcdi_mon.h are implemented in mcdi_mon.c.
The build fails if compiler options require declaration before definition.
Fixes: dfb3b1ce15f6 ("net/sfc/base: import monitors access via MCDI")
Cc: sta...@dpdk.org
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/mcdi_mon.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/sfc/base/mcdi_mon.c b/drivers/net/sfc/base/mcdi_mon.c
index 940bd0265..8c0b6f0d9 100644
--- a/drivers/net/sfc/base/mcdi_mon.c
+++ b/drivers/net/sfc/base/mcdi_mon.c
@@ -6,6 +6,7 @@
#include "efx.h"
#include "efx_impl.h"
+#include "mcdi_mon.h"
#if EFSYS_OPT_MON_MCDI
--
2.17.1
[dpdk-dev] [PATCH 10/37] net/sfc/base: fix outer IPID field in TSO option descriptors
From: Vijay Srivastava
Fixes: 912e603706c5 ("net/sfc/base: add outer IP ID parameter to TSOv2
descriptor")
Cc: sta...@dpdk.org
Signed-off-by: Vijay Srivastava
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/ef10_tx.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/net/sfc/base/ef10_tx.c b/drivers/net/sfc/base/ef10_tx.c
index b92cadcbf..e74d39540 100644
--- a/drivers/net/sfc/base/ef10_tx.c
+++ b/drivers/net/sfc/base/ef10_tx.c
@@ -639,22 +639,22 @@ ef10_tx_qdesc_tso2_create(
EFSYS_ASSERT(count >= EFX_TX_FATSOV2_OPT_NDESCS);
- EFX_POPULATE_QWORD_6(edp[0].ed_eq,
+ EFX_POPULATE_QWORD_5(edp[0].ed_eq,
ESF_DZ_TX_DESC_IS_OPT, 1,
ESF_DZ_TX_OPTION_TYPE,
ESE_DZ_TX_OPTION_DESC_TSO,
ESF_DZ_TX_TSO_OPTION_TYPE,
ESE_DZ_TX_TSO_OPTION_DESC_FATSO2A,
ESF_DZ_TX_TSO_IP_ID, ipv4_id,
- ESF_DZ_TX_TSO_OUTER_IPID, outer_ipv4_id,
ESF_DZ_TX_TSO_TCP_SEQNO, tcp_seq);
- EFX_POPULATE_QWORD_4(edp[1].ed_eq,
+ EFX_POPULATE_QWORD_5(edp[1].ed_eq,
ESF_DZ_TX_DESC_IS_OPT, 1,
ESF_DZ_TX_OPTION_TYPE,
ESE_DZ_TX_OPTION_DESC_TSO,
ESF_DZ_TX_TSO_OPTION_TYPE,
ESE_DZ_TX_TSO_OPTION_DESC_FATSO2B,
- ESF_DZ_TX_TSO_TCP_MSS, tcp_mss);
+ ESF_DZ_TX_TSO_TCP_MSS, tcp_mss,
+ ESF_DZ_TX_TSO_OUTER_IPID, outer_ipv4_id);
}
void
--
2.17.1
[dpdk-dev] [PATCH 07/37] net/sfc/base: add space after sizeof
From: Andy Moreton
Required by GLD cstyle.
Fixes: d4f4b8f9d260 ("net/sfc/base: make RxQ type data an union")
Cc: sta...@dpdk.org
Signed-off-by: Andy Moreton
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/efx_rx.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/net/sfc/base/efx_rx.c b/drivers/net/sfc/base/efx_rx.c
index 4fd73bab3..09933b41a 100644
--- a/drivers/net/sfc/base/efx_rx.c
+++ b/drivers/net/sfc/base/efx_rx.c
@@ -831,7 +831,7 @@ efx_rx_qcreate_packed_stream(
{
efx_rxq_type_data_t type_data;
- memset(&type_data, 0, sizeof(type_data));
+ memset(&type_data, 0, sizeof (type_data));
type_data.ertd_packed_stream.eps_buf_size = ps_buf_size;
@@ -867,7 +867,7 @@ efx_rx_qcreate_es_super_buffer(
goto fail1;
}
- memset(&type_data, 0, sizeof(type_data));
+ memset(&type_data, 0, sizeof (type_data));
type_data.ertd_es_super_buffer.eessb_bufs_per_desc = n_bufs_per_desc;
type_data.ertd_es_super_buffer.eessb_max_dma_len = max_dma_len;
--
2.17.1
[dpdk-dev] [PATCH 12/37] net/sfc/base: add check for TUNNEL module in NIC reset API
From: Vijay Srivastava
Fixes: 17551f6dffcc ("net/sfc/base: add API to control UDP tunnel ports")
Cc: sta...@dpdk.org
Signed-off-by: Vijay Srivastava
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/efx_nic.c | 5 -
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/drivers/net/sfc/base/efx_nic.c b/drivers/net/sfc/base/efx_nic.c
index 6314ae2ff..0e8ed9c2a 100644
--- a/drivers/net/sfc/base/efx_nic.c
+++ b/drivers/net/sfc/base/efx_nic.c
@@ -549,7 +549,7 @@ efx_nic_reset(
EFSYS_ASSERT3U(enp->en_magic, ==, EFX_NIC_MAGIC);
EFSYS_ASSERT(enp->en_mod_flags & EFX_MOD_PROBE);
/*
-* All modules except the MCDI, PROBE, NVRAM, VPD, MON
+* All modules except the MCDI, PROBE, NVRAM, VPD, MON, TUNNEL
* (which we do not reset here) must have been shut down or never
* initialized.
*
@@ -560,6 +560,9 @@ efx_nic_reset(
mod_flags = enp->en_mod_flags;
mod_flags &= ~(EFX_MOD_MCDI | EFX_MOD_PROBE | EFX_MOD_NVRAM |
EFX_MOD_VPD | EFX_MOD_MON);
+#if EFSYS_OPT_TUNNEL
+ mod_flags &= ~EFX_MOD_TUNNEL;
+#endif /* EFSYS_OPT_TUNNEL */
EFSYS_ASSERT3U(mod_flags, ==, 0);
if (mod_flags != 0) {
rc = EINVAL;
--
2.17.1
[dpdk-dev] [PATCH 14/37] net/sfc/base: remove probes when a Tx queue is too full
From: Mark Spender
No need for probe messages when a TxQ is too full for a post to be done.
Existing drivers check if there is room in the queue before posting
descriptors, even though efx_tx_qdesc_post() does the check itself.
The new SFN Windows driver doesn't perform the check before calling
efx_tx_qdesc_post(), but that means these probes can get frequently
printed out. It's normal driver behaviour so there's no need to print
an error.
Signed-off-by: Mark Spender
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/ef10_tx.c | 12 ++--
drivers/net/sfc/base/efx_tx.c | 19 ++-
2 files changed, 4 insertions(+), 27 deletions(-)
diff --git a/drivers/net/sfc/base/ef10_tx.c b/drivers/net/sfc/base/ef10_tx.c
index e74d39540..4d77d76d7 100644
--- a/drivers/net/sfc/base/ef10_tx.c
+++ b/drivers/net/sfc/base/ef10_tx.c
@@ -541,12 +541,9 @@ ef10_tx_qdesc_post(
{
unsigned int added = *addedp;
unsigned int i;
- efx_rc_t rc;
- if (added - completed + ndescs > EFX_TXQ_LIMIT(etp->et_mask + 1)) {
- rc = ENOSPC;
- goto fail1;
- }
+ if (added - completed + ndescs > EFX_TXQ_LIMIT(etp->et_mask + 1))
+ return (ENOSPC);
for (i = 0; i < ndescs; i++) {
efx_desc_t *edp = &ed[i];
@@ -566,11 +563,6 @@ ef10_tx_qdesc_post(
*addedp = added;
return (0);
-
-fail1:
- EFSYS_PROBE1(fail1, efx_rc_t, rc);
-
- return (rc);
}
void
diff --git a/drivers/net/sfc/base/efx_tx.c b/drivers/net/sfc/base/efx_tx.c
index da37580a6..bf1180a1e 100644
--- a/drivers/net/sfc/base/efx_tx.c
+++ b/drivers/net/sfc/base/efx_tx.c
@@ -572,19 +572,10 @@ efx_tx_qdesc_post(
{
efx_nic_t *enp = etp->et_enp;
const efx_tx_ops_t *etxop = enp->en_etxop;
- efx_rc_t rc;
EFSYS_ASSERT3U(etp->et_magic, ==, EFX_TXQ_MAGIC);
- if ((rc = etxop->etxo_qdesc_post(etp, ed,
- ndescs, completed, addedp)) != 0)
- goto fail1;
-
- return (0);
-
-fail1:
- EFSYS_PROBE1(fail1, efx_rc_t, rc);
- return (rc);
+ return (etxop->etxo_qdesc_post(etp, ed, ndescs, completed, addedp));
}
void
@@ -763,10 +754,9 @@ siena_tx_qpost(
{
unsigned int added = *addedp;
unsigned int i;
- int rc = ENOSPC;
if (added - completed + ndescs > EFX_TXQ_LIMIT(etp->et_mask + 1))
- goto fail1;
+ return (ENOSPC);
for (i = 0; i < ndescs; i++) {
efx_buffer_t *ebp = &eb[i];
@@ -788,11 +778,6 @@ siena_tx_qpost(
*addedp = added;
return (0);
-
-fail1:
- EFSYS_PROBE1(fail1, efx_rc_t, rc);
-
- return (rc);
}
static void
--
2.17.1
[dpdk-dev] [PATCH 09/37] net/sfc/base: add more definitions of partitions
From: Paul Fox
Add definitions of dynamic config and expansion ROM backup
partitions.
Signed-off-by: Paul Fox
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/ef10_nvram.c | 2 ++
drivers/net/sfc/base/efx.h| 2 ++
2 files changed, 4 insertions(+)
diff --git a/drivers/net/sfc/base/ef10_nvram.c
b/drivers/net/sfc/base/ef10_nvram.c
index 46838dd75..96ea5624a 100644
--- a/drivers/net/sfc/base/ef10_nvram.c
+++ b/drivers/net/sfc/base/ef10_nvram.c
@@ -2168,6 +2168,8 @@ static ef10_parttbl_entry_t medford2_parttbl[] = {
PARTN_MAP_ENTRY(LICENSE,ALL,LICENSE),
PARTN_MAP_ENTRY(EXPANSION_UEFI, ALL,UEFIROM),
PARTN_MAP_ENTRY(MUM_FIRMWARE, ALL,MUM_FIRMWARE),
+ PARTN_MAP_ENTRY(DYNCONFIG_DEFAULTS, ALL,DYNCONFIG_DEFAULTS),
+ PARTN_MAP_ENTRY(ROMCONFIG_DEFAULTS, ALL,ROMCONFIG_DEFAULTS),
};
static __checkReturn efx_rc_t
diff --git a/drivers/net/sfc/base/efx.h b/drivers/net/sfc/base/efx.h
index 53cbc9858..baeffd934 100644
--- a/drivers/net/sfc/base/efx.h
+++ b/drivers/net/sfc/base/efx.h
@@ -1483,6 +1483,8 @@ typedef enum efx_nvram_type_e {
EFX_NVRAM_LICENSE,
EFX_NVRAM_UEFIROM,
EFX_NVRAM_MUM_FIRMWARE,
+ EFX_NVRAM_DYNCONFIG_DEFAULTS,
+ EFX_NVRAM_ROMCONFIG_DEFAULTS,
EFX_NVRAM_NTYPES,
} efx_nvram_type_t;
--
2.17.1
[dpdk-dev] [PATCH 13/37] net/sfc/base: refactor monitors support
From: Martin Harvey
Remove obsolete monitor types since Falcon SFN4000 series adapters
no longer supported by libefx.
Rename MCDI monitors to be consistent with YML.
The code may be simplified and generalized since only MCDI monitors
remain.
Signed-off-by: Martin Harvey
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/efx.h | 158
drivers/net/sfc/base/efx_mon.c | 623
drivers/net/sfc/base/mcdi_mon.c | 187 ++
3 files changed, 695 insertions(+), 273 deletions(-)
diff --git a/drivers/net/sfc/base/efx.h b/drivers/net/sfc/base/efx.h
index 4c8983387..ffb6aad94 100644
--- a/drivers/net/sfc/base/efx.h
+++ b/drivers/net/sfc/base/efx.h
@@ -662,77 +662,74 @@ efx_mon_init(
#defineEFX_MON_STATS_PAGE_SIZE 0x100
#defineEFX_MON_MASK_ELEMENT_SIZE 32
-/* START MKCONFIG GENERATED MonitorHeaderStatsBlock 400fdb0517af1fca */
+/* START MKCONFIG GENERATED MonitorHeaderStatsBlock 78b65c8d5af9747b */
typedef enum efx_mon_stat_e {
- EFX_MON_STAT_2_5V,
- EFX_MON_STAT_VCCP1,
- EFX_MON_STAT_VCC,
- EFX_MON_STAT_5V,
- EFX_MON_STAT_12V,
- EFX_MON_STAT_VCCP2,
- EFX_MON_STAT_EXT_TEMP,
- EFX_MON_STAT_INT_TEMP,
- EFX_MON_STAT_AIN1,
- EFX_MON_STAT_AIN2,
- EFX_MON_STAT_INT_COOLING,
- EFX_MON_STAT_EXT_COOLING,
- EFX_MON_STAT_1V,
- EFX_MON_STAT_1_2V,
- EFX_MON_STAT_1_8V,
- EFX_MON_STAT_3_3V,
- EFX_MON_STAT_1_2VA,
- EFX_MON_STAT_VREF,
- EFX_MON_STAT_VAOE,
+ EFX_MON_STAT_CONTROLLER_TEMP,
+ EFX_MON_STAT_PHY_COMMON_TEMP,
+ EFX_MON_STAT_CONTROLLER_COOLING,
+ EFX_MON_STAT_PHY0_TEMP,
+ EFX_MON_STAT_PHY0_COOLING,
+ EFX_MON_STAT_PHY1_TEMP,
+ EFX_MON_STAT_PHY1_COOLING,
+ EFX_MON_STAT_IN_1V0,
+ EFX_MON_STAT_IN_1V2,
+ EFX_MON_STAT_IN_1V8,
+ EFX_MON_STAT_IN_2V5,
+ EFX_MON_STAT_IN_3V3,
+ EFX_MON_STAT_IN_12V0,
+ EFX_MON_STAT_IN_1V2A,
+ EFX_MON_STAT_IN_VREF,
+ EFX_MON_STAT_OUT_VAOE,
EFX_MON_STAT_AOE_TEMP,
EFX_MON_STAT_PSU_AOE_TEMP,
EFX_MON_STAT_PSU_TEMP,
- EFX_MON_STAT_FAN0,
- EFX_MON_STAT_FAN1,
- EFX_MON_STAT_FAN2,
- EFX_MON_STAT_FAN3,
- EFX_MON_STAT_FAN4,
- EFX_MON_STAT_VAOE_IN,
- EFX_MON_STAT_IAOE,
- EFX_MON_STAT_IAOE_IN,
+ EFX_MON_STAT_FAN_0,
+ EFX_MON_STAT_FAN_1,
+ EFX_MON_STAT_FAN_2,
+ EFX_MON_STAT_FAN_3,
+ EFX_MON_STAT_FAN_4,
+ EFX_MON_STAT_IN_VAOE,
+ EFX_MON_STAT_OUT_IAOE,
+ EFX_MON_STAT_IN_IAOE,
EFX_MON_STAT_NIC_POWER,
- EFX_MON_STAT_0_9V,
- EFX_MON_STAT_I0_9V,
- EFX_MON_STAT_I1_2V,
- EFX_MON_STAT_0_9V_ADC,
- EFX_MON_STAT_INT_TEMP2,
- EFX_MON_STAT_VREG_TEMP,
- EFX_MON_STAT_VREG_0_9V_TEMP,
- EFX_MON_STAT_VREG_1_2V_TEMP,
- EFX_MON_STAT_INT_VPTAT,
- EFX_MON_STAT_INT_ADC_TEMP,
- EFX_MON_STAT_EXT_VPTAT,
- EFX_MON_STAT_EXT_ADC_TEMP,
+ EFX_MON_STAT_IN_0V9,
+ EFX_MON_STAT_IN_I0V9,
+ EFX_MON_STAT_IN_I1V2,
+ EFX_MON_STAT_IN_0V9_ADC,
+ EFX_MON_STAT_CONTROLLER_2_TEMP,
+ EFX_MON_STAT_VREG_INTERNAL_TEMP,
+ EFX_MON_STAT_VREG_0V9_TEMP,
+ EFX_MON_STAT_VREG_1V2_TEMP,
+ EFX_MON_STAT_CONTROLLER_VPTAT,
+ EFX_MON_STAT_CONTROLLER_INTERNAL_TEMP,
+ EFX_MON_STAT_CONTROLLER_VPTAT_EXTADC,
+ EFX_MON_STAT_CONTROLLER_INTERNAL_TEMP_EXTADC,
EFX_MON_STAT_AMBIENT_TEMP,
EFX_MON_STAT_AIRFLOW,
EFX_MON_STAT_VDD08D_VSS08D_CSR,
EFX_MON_STAT_VDD08D_VSS08D_CSR_EXTADC,
EFX_MON_STAT_HOTPOINT_TEMP,
- EFX_MON_STAT_PHY_POWER_SWITCH_PORT0,
- EFX_MON_STAT_PHY_POWER_SWITCH_PORT1,
+ EFX_MON_STAT_PHY_POWER_PORT0,
+ EFX_MON_STAT_PHY_POWER_PORT1,
EFX_MON_STAT_MUM_VCC,
- EFX_MON_STAT_0V9_A,
- EFX_MON_STAT_I0V9_A,
- EFX_MON_STAT_0V9_A_TEMP,
- EFX_MON_STAT_0V9_B,
- EFX_MON_STAT_I0V9_B,
- EFX_MON_STAT_0V9_B_TEMP,
+ EFX_MON_STAT_IN_0V9_A,
+ EFX_MON_STAT_IN_I0V9_A,
+ EFX_MON_STAT_VREG_0V9_A_TEMP,
+ EFX_MON_STAT_IN_0V9_B,
+ EFX_MON_STAT_IN_I0V9_B,
+ EFX_MON_STAT_VREG_0V9_B_TEMP,
EFX_MON_STAT_CCOM_AVREG_1V2_SUPPLY,
- EFX_MON_STAT_CCOM_AVREG_1V2_SUPPLY_EXT_ADC,
+ EFX_MON_STAT_CCOM_AVREG_1V2_SUPPLY_EXTADC,
EFX_MON_STAT_CCOM_AVREG_1V8_SUPPLY,
- EFX_MON_STAT_CCOM_AVREG_1V8_SUPPLY_EXT_ADC,
+ EFX_MON_STAT_CCOM_AVREG_1V8_SUPPLY_EXTADC,
EFX_MON_STAT_CONTROLLER_MASTER_VPTAT,
EFX_MON_STAT_CONTROLLER_MASTER_INTERNAL_TEMP,
- EFX_MON_STAT_CONTROLLER_MASTER_VPTAT_EXT_ADC,
- EFX_MON_STAT_CONTROLLER_MASTER_INTERNAL_TEMP_EXT_ADC,
+ EFX_MON_STAT_CONTROLLER_MASTER_VPTAT_EXTADC,
+ EFX_MON_STAT_CONTROLLER_MASTER_INTERNAL_TEMP_EXTADC,
EFX_MON_STAT_CONTROLLER_SLAVE_VPTAT,
EFX_MON_STAT_CONTROLLER_SLAVE_INTERNAL_TEMP,
- EFX_MON_STAT_
[dpdk-dev] [PATCH 16/37] net/sfc/base: check size of memory to read sensors data to
From: Martin Harvey
Size of provided memory should be consistent with specified size.
Fixes: dfb3b1ce15f6 ("net/sfc/base: import monitors access via MCDI")
Cc: sta...@dpdk.org
Signed-off-by: Martin Harvey
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/mcdi_mon.c | 11 +++
1 file changed, 11 insertions(+)
diff --git a/drivers/net/sfc/base/mcdi_mon.c b/drivers/net/sfc/base/mcdi_mon.c
index 93e6b1e35..68bbc575d 100644
--- a/drivers/net/sfc/base/mcdi_mon.c
+++ b/drivers/net/sfc/base/mcdi_mon.c
@@ -194,6 +194,12 @@ efx_mcdi_read_sensors(
uint8_t payload[MAX(MC_CMD_READ_SENSORS_EXT_IN_LEN,
MC_CMD_READ_SENSORS_EXT_OUT_LEN)];
uint32_t addr_lo, addr_hi;
+ efx_rc_t rc;
+
+ if (EFSYS_MEM_SIZE(esmp) < size) {
+ rc = EINVAL;
+ goto fail1;
+ }
req.emr_cmd = MC_CMD_READ_SENSORS;
req.emr_in_buf = payload;
@@ -211,6 +217,11 @@ efx_mcdi_read_sensors(
efx_mcdi_execute(enp, &req);
return (req.emr_rc);
+
+fail1:
+ EFSYS_PROBE1(fail1, efx_rc_t, rc);
+
+ return (rc);
}
static __checkReturn efx_rc_t
--
2.17.1
[dpdk-dev] [PATCH 21/37] net/sfc/base: avoid usage of too big arrays on stack
From: Martin Harvey
Found by PreFAST static analysis.
Fixes: 1dae25112a54 ("net/sfc/base: import built-in selftest")
Fixes: d96a34d165b1 ("net/sfc/base: import NVRAM support")
Cc: sta...@dpdk.org
Signed-off-by: Martin Harvey
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/ef10_phy.c | 18 +++---
drivers/net/sfc/base/efx_nvram.c | 27 ++-
2 files changed, 33 insertions(+), 12 deletions(-)
diff --git a/drivers/net/sfc/base/ef10_phy.c b/drivers/net/sfc/base/ef10_phy.c
index 84acb70a1..e9c7b40e4 100644
--- a/drivers/net/sfc/base/ef10_phy.c
+++ b/drivers/net/sfc/base/ef10_phy.c
@@ -583,14 +583,26 @@ ef10_bist_poll(
unsigned long *valuesp,
__insize_t count)
{
+ /*
+* MCDI_CTL_SDU_LEN_MAX_V1 is large enough cover all BIST results,
+* whilst not wasting stack.
+*/
+ uint8_t payload[MAX(MC_CMD_POLL_BIST_IN_LEN, MCDI_CTL_SDU_LEN_MAX_V1)];
efx_nic_cfg_t *encp = &(enp->en_nic_cfg);
efx_mcdi_req_t req;
- uint8_t payload[MAX(MC_CMD_POLL_BIST_IN_LEN,
- MCDI_CTL_SDU_LEN_MAX)];
uint32_t value_mask = 0;
uint32_t result;
efx_rc_t rc;
+ EFX_STATIC_ASSERT(MC_CMD_POLL_BIST_OUT_LEN <=
+ MCDI_CTL_SDU_LEN_MAX_V1);
+ EFX_STATIC_ASSERT(MC_CMD_POLL_BIST_OUT_SFT9001_LEN <=
+ MCDI_CTL_SDU_LEN_MAX_V1);
+ EFX_STATIC_ASSERT(MC_CMD_POLL_BIST_OUT_MRSFP_LEN <=
+ MCDI_CTL_SDU_LEN_MAX_V1);
+ EFX_STATIC_ASSERT(MC_CMD_POLL_BIST_OUT_MEM_LEN <=
+ MCDI_CTL_SDU_LEN_MAX_V1);
+
_NOTE(ARGUNUSED(type))
(void) memset(payload, 0, sizeof (payload));
@@ -598,7 +610,7 @@ ef10_bist_poll(
req.emr_in_buf = payload;
req.emr_in_length = MC_CMD_POLL_BIST_IN_LEN;
req.emr_out_buf = payload;
- req.emr_out_length = MCDI_CTL_SDU_LEN_MAX;
+ req.emr_out_length = MCDI_CTL_SDU_LEN_MAX_V1;
efx_mcdi_execute(enp, &req);
diff --git a/drivers/net/sfc/base/efx_nvram.c b/drivers/net/sfc/base/efx_nvram.c
index 9000fe886..d7b1a6778 100644
--- a/drivers/net/sfc/base/efx_nvram.c
+++ b/drivers/net/sfc/base/efx_nvram.c
@@ -869,23 +869,27 @@ efx_mcdi_nvram_write(
__insize_t size)
{
efx_mcdi_req_t req;
- uint8_t payload[MAX(MCDI_CTL_SDU_LEN_MAX_V1,
- MCDI_CTL_SDU_LEN_MAX_V2)];
+ uint8_t *payload;
efx_rc_t rc;
size_t max_data_size;
+ size_t payload_len = enp->en_nic_cfg.enc_mcdi_max_payload_length;
- max_data_size = enp->en_nic_cfg.enc_mcdi_max_payload_length
- - MC_CMD_NVRAM_WRITE_IN_LEN(0);
- EFSYS_ASSERT3U(enp->en_nic_cfg.enc_mcdi_max_payload_length, >, 0);
- EFSYS_ASSERT3U(max_data_size, <,
- enp->en_nic_cfg.enc_mcdi_max_payload_length);
+ max_data_size = payload_len - MC_CMD_NVRAM_WRITE_IN_LEN(0);
+ EFSYS_ASSERT3U(payload_len, >, 0);
+ EFSYS_ASSERT3U(max_data_size, <, payload_len);
if (size > max_data_size) {
rc = EINVAL;
goto fail1;
}
- (void) memset(payload, 0, sizeof (payload));
+ EFSYS_KMEM_ALLOC(enp->en_esip, payload_len, payload);
+ if (payload == NULL) {
+ rc = ENOMEM;
+ goto fail2;
+ }
+
+ (void) memset(payload, 0, payload_len);
req.emr_cmd = MC_CMD_NVRAM_WRITE;
req.emr_in_buf = payload;
req.emr_in_length = MC_CMD_NVRAM_WRITE_IN_LEN(size);
@@ -903,11 +907,16 @@ efx_mcdi_nvram_write(
if (req.emr_rc != 0) {
rc = req.emr_rc;
- goto fail2;
+ goto fail3;
}
+ EFSYS_KMEM_FREE(enp->en_esip, payload_len, payload);
+
return (0);
+fail3:
+ EFSYS_PROBE(fail3);
+ EFSYS_KMEM_FREE(enp->en_esip, payload_len, payload);
fail2:
EFSYS_PROBE(fail2);
fail1:
--
2.17.1
[dpdk-dev] [PATCH 15/37] net/sfc/base: add generated description of sensors
From: Martin Harvey
Description of sensors is generated from firmware sources.
Signed-off-by: Martin Harvey
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/efx.h | 5 +
drivers/net/sfc/base/efx_mcdi.h| 4 +
drivers/net/sfc/base/efx_mon.c | 133 ++---
drivers/net/sfc/base/mc_driver_pcol_strs.h | 102
4 files changed, 229 insertions(+), 15 deletions(-)
create mode 100644 drivers/net/sfc/base/mc_driver_pcol_strs.h
diff --git a/drivers/net/sfc/base/efx.h b/drivers/net/sfc/base/efx.h
index ffb6aad94..dae90bd6e 100644
--- a/drivers/net/sfc/base/efx.h
+++ b/drivers/net/sfc/base/efx.h
@@ -791,6 +791,11 @@ efx_mon_stat_name(
__inefx_nic_t *enp,
__inefx_mon_stat_t id);
+extern const char *
+efx_mon_stat_description(
+ __inefx_nic_t *enp,
+ __inefx_mon_stat_t id);
+
#endif /* EFSYS_OPT_NAMES */
extern __checkReturn boolean_t
diff --git a/drivers/net/sfc/base/efx_mcdi.h b/drivers/net/sfc/base/efx_mcdi.h
index 253a9e60b..d2cd1e9e1 100644
--- a/drivers/net/sfc/base/efx_mcdi.h
+++ b/drivers/net/sfc/base/efx_mcdi.h
@@ -10,6 +10,10 @@
#include "efx.h"
#include "efx_regs_mcdi.h"
+#if EFSYS_OPT_NAMES
+#include "mc_driver_pcol_strs.h"
+#endif /* EFSYS_OPT_NAMES */
+
#ifdef __cplusplus
extern "C" {
#endif
diff --git a/drivers/net/sfc/base/efx_mon.c b/drivers/net/sfc/base/efx_mon.c
index 34689921d..91fa16ca0 100644
--- a/drivers/net/sfc/base/efx_mon.c
+++ b/drivers/net/sfc/base/efx_mon.c
@@ -185,7 +185,124 @@ static const char * const __mon_stat_name[] = {
/* END MKCONFIG GENERATED MonitorStatNamesBlock */
-/* START MKCONFIG GENERATED MonitorMcdiMappingBlock 362875db87a4e7da */
+ const char *
+efx_mon_stat_name(
+ __inefx_nic_t *enp,
+ __inefx_mon_stat_t id)
+{
+ _NOTE(ARGUNUSED(enp))
+ EFSYS_ASSERT3U(enp->en_magic, ==, EFX_NIC_MAGIC);
+
+ EFSYS_ASSERT3U(id, <, EFX_MON_NSTATS);
+ return (__mon_stat_name[id]);
+}
+
+typedef struct _stat_description_t {
+ efx_mon_stat_t stat;
+ const char *desc;
+} stat_description_t;
+
+/* START MKCONFIG GENERATED MonitorStatDescriptionsBlock f072138f16d2e1f8 */
+static const char *__mon_stat_description[] = {
+ MC_CMD_SENSOR_CONTROLLER_TEMP_ENUM_STR,
+ MC_CMD_SENSOR_PHY_COMMON_TEMP_ENUM_STR,
+ MC_CMD_SENSOR_CONTROLLER_COOLING_ENUM_STR,
+ MC_CMD_SENSOR_PHY0_TEMP_ENUM_STR,
+ MC_CMD_SENSOR_PHY0_COOLING_ENUM_STR,
+ MC_CMD_SENSOR_PHY1_TEMP_ENUM_STR,
+ MC_CMD_SENSOR_PHY1_COOLING_ENUM_STR,
+ MC_CMD_SENSOR_IN_1V0_ENUM_STR,
+ MC_CMD_SENSOR_IN_1V2_ENUM_STR,
+ MC_CMD_SENSOR_IN_1V8_ENUM_STR,
+ MC_CMD_SENSOR_IN_2V5_ENUM_STR,
+ MC_CMD_SENSOR_IN_3V3_ENUM_STR,
+ MC_CMD_SENSOR_IN_12V0_ENUM_STR,
+ MC_CMD_SENSOR_IN_1V2A_ENUM_STR,
+ MC_CMD_SENSOR_IN_VREF_ENUM_STR,
+ MC_CMD_SENSOR_OUT_VAOE_ENUM_STR,
+ MC_CMD_SENSOR_AOE_TEMP_ENUM_STR,
+ MC_CMD_SENSOR_PSU_AOE_TEMP_ENUM_STR,
+ MC_CMD_SENSOR_PSU_TEMP_ENUM_STR,
+ MC_CMD_SENSOR_FAN_0_ENUM_STR,
+ MC_CMD_SENSOR_FAN_1_ENUM_STR,
+ MC_CMD_SENSOR_FAN_2_ENUM_STR,
+ MC_CMD_SENSOR_FAN_3_ENUM_STR,
+ MC_CMD_SENSOR_FAN_4_ENUM_STR,
+ MC_CMD_SENSOR_IN_VAOE_ENUM_STR,
+ MC_CMD_SENSOR_OUT_IAOE_ENUM_STR,
+ MC_CMD_SENSOR_IN_IAOE_ENUM_STR,
+ MC_CMD_SENSOR_NIC_POWER_ENUM_STR,
+ MC_CMD_SENSOR_IN_0V9_ENUM_STR,
+ MC_CMD_SENSOR_IN_I0V9_ENUM_STR,
+ MC_CMD_SENSOR_IN_I1V2_ENUM_STR,
+ MC_CMD_SENSOR_IN_0V9_ADC_ENUM_STR,
+ MC_CMD_SENSOR_CONTROLLER_2_TEMP_ENUM_STR,
+ MC_CMD_SENSOR_VREG_INTERNAL_TEMP_ENUM_STR,
+ MC_CMD_SENSOR_VREG_0V9_TEMP_ENUM_STR,
+ MC_CMD_SENSOR_VREG_1V2_TEMP_ENUM_STR,
+ MC_CMD_SENSOR_CONTROLLER_VPTAT_ENUM_STR,
+ MC_CMD_SENSOR_CONTROLLER_INTERNAL_TEMP_ENUM_STR,
+ MC_CMD_SENSOR_CONTROLLER_VPTAT_EXTADC_ENUM_STR,
+ MC_CMD_SENSOR_CONTROLLER_INTERNAL_TEMP_EXTADC_ENUM_STR,
+ MC_CMD_SENSOR_AMBIENT_TEMP_ENUM_STR,
+ MC_CMD_SENSOR_AIRFLOW_ENUM_STR,
+ MC_CMD_SENSOR_VDD08D_VSS08D_CSR_ENUM_STR,
+ MC_CMD_SENSOR_VDD08D_VSS08D_CSR_EXTADC_ENUM_STR,
+ MC_CMD_SENSOR_HOTPOINT_TEMP_ENUM_STR,
+ MC_CMD_SENSOR_PHY_POWER_PORT0_ENUM_STR,
+ MC_CMD_SENSOR_PHY_POWER_PORT1_ENUM_STR,
+ MC_CMD_SENSOR_MUM_VCC_ENUM_STR,
+ MC_CMD_SENSOR_IN_0V9_A_ENUM_STR,
+ MC_CMD_SENSOR_IN_I0V9_A_ENUM_STR,
+ MC_CMD_SENSOR_VREG_0V9_A_TEMP_ENUM_STR,
+ MC_CMD_SENSOR_IN_0V9_B_ENUM_STR,
+ MC_CMD_SENSOR_IN_I0V9_B_ENUM_STR,
+ MC_CMD_SENSOR_VREG_0V9_B_TEMP_ENUM_STR,
+ MC_CMD_SENSOR_CCOM_AVREG_1V2_SUPPLY_ENUM_STR,
+ MC_CMD_SENSOR_CCOM_AVREG_1V2_SUPPLY_EXTADC_ENUM_STR,
+
[dpdk-dev] [PATCH 19/37] net/sfc/base: add accessor for default port mode
From: Richard Houldsworth
Extend efx_mcdi_get_port_modes() to optionally pass on the default
port mode field. This provides a more direct way of handling the case
where the dynamic config does not specify the port mode than the
alternative of a lookup table indexed by MCFW subtype.
Signed-off-by: Richard Houldsworth
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/ef10_impl.h| 3 ++-
drivers/net/sfc/base/ef10_nic.c | 13 ++---
drivers/net/sfc/base/hunt_nic.c | 3 ++-
drivers/net/sfc/base/medford2_nic.c | 2 +-
drivers/net/sfc/base/medford_nic.c | 2 +-
5 files changed, 16 insertions(+), 7 deletions(-)
diff --git a/drivers/net/sfc/base/ef10_impl.h b/drivers/net/sfc/base/ef10_impl.h
index fb0d98875..2abd699a0 100644
--- a/drivers/net/sfc/base/ef10_impl.h
+++ b/drivers/net/sfc/base/ef10_impl.h
@@ -1128,7 +1128,8 @@ extern__checkReturn efx_rc_t
efx_mcdi_get_port_modes(
__inefx_nic_t *enp,
__out uint32_t *modesp,
- __out_opt uint32_t *current_modep);
+ __out_opt uint32_t *current_modep,
+ __out_opt uint32_t *default_modep);
extern __checkReturn efx_rc_t
ef10_nic_get_port_mode_bandwidth(
diff --git a/drivers/net/sfc/base/ef10_nic.c b/drivers/net/sfc/base/ef10_nic.c
index d1985b3c5..9145c389c 100644
--- a/drivers/net/sfc/base/ef10_nic.c
+++ b/drivers/net/sfc/base/ef10_nic.c
@@ -63,7 +63,8 @@ efx_mcdi_get_port_assignment(
efx_mcdi_get_port_modes(
__inefx_nic_t *enp,
__out uint32_t *modesp,
- __out_opt uint32_t *current_modep)
+ __out_opt uint32_t *current_modep,
+ __out_opt uint32_t *default_modep)
{
efx_mcdi_req_t req;
uint8_t payload[MAX(MC_CMD_GET_PORT_MODES_IN_LEN,
@@ -110,6 +111,11 @@ efx_mcdi_get_port_modes(
GET_PORT_MODES_OUT_CURRENT_MODE);
}
+ if (default_modep != NULL) {
+ *default_modep = MCDI_OUT_DWORD(req,
+ GET_PORT_MODES_OUT_DEFAULT_MODE);
+ }
+
return (0);
fail3:
@@ -1635,13 +1641,14 @@ ef10_external_port_mapping(
int32_t count = 1; /* Default 1-1 mapping */
int32_t offset = 1; /* Default starting external port number */
- if ((rc = efx_mcdi_get_port_modes(enp, &port_modes, ¤t)) != 0) {
+ if ((rc = efx_mcdi_get_port_modes(enp, &port_modes, ¤t,
+ NULL)) != 0) {
/*
* No current port mode information (i.e. Huntington)
* - infer mapping from available modes
*/
if ((rc = efx_mcdi_get_port_modes(enp,
- &port_modes, NULL)) != 0) {
+ &port_modes, NULL, NULL)) != 0) {
/*
* No port mode information available
* - use default mapping
diff --git a/drivers/net/sfc/base/hunt_nic.c b/drivers/net/sfc/base/hunt_nic.c
index 16ea81d23..1bec3c485 100644
--- a/drivers/net/sfc/base/hunt_nic.c
+++ b/drivers/net/sfc/base/hunt_nic.c
@@ -30,7 +30,8 @@ hunt_nic_get_required_pcie_bandwidth(
* capable mode is in use.
*/
- if ((rc = efx_mcdi_get_port_modes(enp, &port_modes, NULL)) != 0) {
+ if ((rc = efx_mcdi_get_port_modes(enp, &port_modes,
+ NULL, NULL)) != 0) {
/* No port mode info available */
bandwidth = 0;
goto out;
diff --git a/drivers/net/sfc/base/medford2_nic.c
b/drivers/net/sfc/base/medford2_nic.c
index 7f5ad1757..b36e54bab 100644
--- a/drivers/net/sfc/base/medford2_nic.c
+++ b/drivers/net/sfc/base/medford2_nic.c
@@ -23,7 +23,7 @@ medford2_nic_get_required_pcie_bandwidth(
/* FIXME: support new Medford2 dynamic port modes */
if ((rc = efx_mcdi_get_port_modes(enp, &port_modes,
- ¤t_mode)) != 0) {
+ ¤t_mode, NULL)) != 0) {
/* No port mode info available. */
bandwidth = 0;
goto out;
diff --git a/drivers/net/sfc/base/medford_nic.c
b/drivers/net/sfc/base/medford_nic.c
index 6dc895f5b..96f3a1204 100644
--- a/drivers/net/sfc/base/medford_nic.c
+++ b/drivers/net/sfc/base/medford_nic.c
@@ -21,7 +21,7 @@ medford_nic_get_required_pcie_bandwidth(
efx_rc_t rc;
if ((rc = efx_mcdi_get_port_modes(enp, &port_modes,
- ¤t_mode)) != 0) {
+ ¤t_mode, NULL)) != 0) {
/* No port mode info available. */
bandwidth = 0;
goto out;
--
2.17.1
[dpdk-dev] [PATCH 25/37] net/sfc/base: add API to inform libefx of hardware removal
From: Andy Moreton
The efx_nic_hw_unavailable() checks ensure that if the NIC hardware
has failed or has been physically removed then libefx will stop
further attempts to access the hardware.
Add an interface for libefx clients to force unavailability, so the
hardware is treated as dead or removed even if still physically present.
Signed-off-by: Andy Moreton
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/ef10_impl.h | 4
drivers/net/sfc/base/ef10_nic.c | 12 ++--
drivers/net/sfc/base/efx.h | 4
drivers/net/sfc/base/efx_impl.h | 1 +
drivers/net/sfc/base/efx_nic.c | 18 --
5 files changed, 35 insertions(+), 4 deletions(-)
diff --git a/drivers/net/sfc/base/ef10_impl.h b/drivers/net/sfc/base/ef10_impl.h
index 0214a75ef..2819ae6ed 100644
--- a/drivers/net/sfc/base/ef10_impl.h
+++ b/drivers/net/sfc/base/ef10_impl.h
@@ -194,6 +194,10 @@ extern __checkReturn boolean_t
ef10_nic_hw_unavailable(
__inefx_nic_t *enp);
+extern void
+ef10_nic_set_hw_unavailable(
+ __inefx_nic_t *enp);
+
#if EFSYS_OPT_DIAG
extern __checkReturn efx_rc_t
diff --git a/drivers/net/sfc/base/ef10_nic.c b/drivers/net/sfc/base/ef10_nic.c
index ff96a7ff2..0a2474f3e 100644
--- a/drivers/net/sfc/base/ef10_nic.c
+++ b/drivers/net/sfc/base/ef10_nic.c
@@ -2320,12 +2320,20 @@ ef10_nic_hw_unavailable(
return (B_FALSE);
unavail:
- EFSYS_PROBE(hw_unavail);
- enp->en_reset_flags |= EFX_RESET_HW_UNAVAIL;
+ ef10_nic_set_hw_unavailable(enp);
return (B_TRUE);
}
+ void
+ef10_nic_set_hw_unavailable(
+ __inefx_nic_t *enp)
+{
+ EFSYS_PROBE(hw_unavail);
+ enp->en_reset_flags |= EFX_RESET_HW_UNAVAIL;
+}
+
+
void
ef10_nic_fini(
__inefx_nic_t *enp)
diff --git a/drivers/net/sfc/base/efx.h b/drivers/net/sfc/base/efx.h
index fce519037..0982a34d6 100644
--- a/drivers/net/sfc/base/efx.h
+++ b/drivers/net/sfc/base/efx.h
@@ -159,6 +159,10 @@ extern __checkReturn boolean_t
efx_nic_hw_unavailable(
__inefx_nic_t *enp);
+extern void
+efx_nic_set_hw_unavailable(
+ __inefx_nic_t *enp);
+
#if EFSYS_OPT_DIAG
extern __checkReturn efx_rc_t
diff --git a/drivers/net/sfc/base/efx_impl.h b/drivers/net/sfc/base/efx_impl.h
index 8a7dc8cf6..2c95571b1 100644
--- a/drivers/net/sfc/base/efx_impl.h
+++ b/drivers/net/sfc/base/efx_impl.h
@@ -358,6 +358,7 @@ typedef struct efx_nic_ops_s {
efx_rc_t(*eno_get_bar_region)(efx_nic_t *, efx_nic_region_t,
uint32_t *, size_t *);
boolean_t (*eno_hw_unavailable)(efx_nic_t *);
+ void(*eno_set_hw_unavailable)(efx_nic_t *);
#if EFSYS_OPT_DIAG
efx_rc_t(*eno_register_test)(efx_nic_t *);
#endif /* EFSYS_OPT_DIAG */
diff --git a/drivers/net/sfc/base/efx_nic.c b/drivers/net/sfc/base/efx_nic.c
index 22e464a4e..e5cb0105f 100644
--- a/drivers/net/sfc/base/efx_nic.c
+++ b/drivers/net/sfc/base/efx_nic.c
@@ -101,6 +101,7 @@ static const efx_nic_ops_t __efx_nic_siena_ops = {
NULL, /* eno_get_vi_pool */
NULL, /* eno_get_bar_region */
NULL, /* eno_hw_unavailable */
+ NULL, /* eno_set_hw_unavailable */
#if EFSYS_OPT_DIAG
siena_nic_register_test,/* eno_register_test */
#endif /* EFSYS_OPT_DIAG */
@@ -121,6 +122,7 @@ static const efx_nic_ops_t __efx_nic_hunt_ops = {
ef10_nic_get_vi_pool, /* eno_get_vi_pool */
ef10_nic_get_bar_region,/* eno_get_bar_region */
ef10_nic_hw_unavailable,/* eno_hw_unavailable */
+ ef10_nic_set_hw_unavailable,/* eno_set_hw_unavailable */
#if EFSYS_OPT_DIAG
ef10_nic_register_test, /* eno_register_test */
#endif /* EFSYS_OPT_DIAG */
@@ -141,6 +143,7 @@ static const efx_nic_ops_t __efx_nic_medford_ops = {
ef10_nic_get_vi_pool, /* eno_get_vi_pool */
ef10_nic_get_bar_region,/* eno_get_bar_region */
ef10_nic_hw_unavailable,/* eno_hw_unavailable */
+ ef10_nic_set_hw_unavailable,/* eno_set_hw_unavailable */
#if EFSYS_OPT_DIAG
ef10_nic_register_test, /* eno_register_test */
#endif /* EFSYS_OPT_DIAG */
@@ -161,6 +164,7 @@ static const efx_nic_ops_t __efx_nic_medford2_ops = {
ef10_nic_get_vi_pool, /* eno_get_vi_pool */
ef10_nic_get_bar_region,/* eno_get_bar_region */
ef10_nic_hw_unavailable,/* eno_hw_unavailable */
+ ef10_nic_set_hw_unavailable,/* eno_set_hw_unavailable */
#if EFSYS_OPT_DIAG
ef10_nic_register_test, /* eno_register_test */
#endif /* EFSYS_OPT_DIAG */
@@ -673,11 +677,21 @@ efx_nic_hw_unavailable(
return (B_FALSE);
u
[dpdk-dev] [PATCH 17/37] net/sfc/base: add API to retrieve sensor limits
From: Martin Harvey
Signed-off-by: Martin Harvey
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/efx.h | 12 +++
drivers/net/sfc/base/efx_impl.h | 2 +
drivers/net/sfc/base/efx_mon.c | 17 +++-
drivers/net/sfc/base/mcdi_mon.c | 171
drivers/net/sfc/base/mcdi_mon.h | 5 +
5 files changed, 206 insertions(+), 1 deletion(-)
diff --git a/drivers/net/sfc/base/efx.h b/drivers/net/sfc/base/efx.h
index dae90bd6e..099f9df67 100644
--- a/drivers/net/sfc/base/efx.h
+++ b/drivers/net/sfc/base/efx.h
@@ -774,6 +774,13 @@ typedef struct efx_mon_stat_value_s {
efx_mon_stat_unit_t emsv_unit;
} efx_mon_stat_value_t;
+typedef struct efx_mon_limit_value_s {
+ uint16_temlv_warning_min;
+ uint16_temlv_warning_max;
+ uint16_temlv_fatal_min;
+ uint16_temlv_fatal_max;
+} efx_mon_stat_limits_t;
+
typedef enum efx_mon_stat_portmask_e {
EFX_MON_STAT_PORTMAP_NONE = 0,
EFX_MON_STAT_PORTMAP_PORT0 = 1,
@@ -819,6 +826,11 @@ efx_mon_stats_update(
__inefsys_mem_t *esmp,
__inout_ecount(EFX_MON_NSTATS) efx_mon_stat_value_t *values);
+extern __checkReturn efx_rc_t
+efx_mon_limits_update(
+ __inefx_nic_t *enp,
+ __inout_ecount(EFX_MON_NSTATS) efx_mon_stat_limits_t *values);
+
#endif /* EFSYS_OPT_MON_STATS */
extern void
diff --git a/drivers/net/sfc/base/efx_impl.h b/drivers/net/sfc/base/efx_impl.h
index 637e31e0c..0764df5de 100644
--- a/drivers/net/sfc/base/efx_impl.h
+++ b/drivers/net/sfc/base/efx_impl.h
@@ -317,6 +317,8 @@ typedef struct efx_mon_ops_s {
#if EFSYS_OPT_MON_STATS
efx_rc_t(*emo_stats_update)(efx_nic_t *, efsys_mem_t *,
efx_mon_stat_value_t *);
+ efx_rc_t(*emo_limits_update)(efx_nic_t *,
+efx_mon_stat_limits_t *);
#endif /* EFSYS_OPT_MON_STATS */
} efx_mon_ops_t;
diff --git a/drivers/net/sfc/base/efx_mon.c b/drivers/net/sfc/base/efx_mon.c
index 91fa16ca0..f28775d04 100644
--- a/drivers/net/sfc/base/efx_mon.c
+++ b/drivers/net/sfc/base/efx_mon.c
@@ -38,7 +38,8 @@ efx_mon_name(
#if EFSYS_OPT_MON_MCDI
static const efx_mon_ops_t __efx_mon_mcdi_ops = {
#if EFSYS_OPT_MON_STATS
- mcdi_mon_stats_update /* emo_stats_update */
+ mcdi_mon_stats_update, /* emo_stats_update */
+ mcdi_mon_limits_update, /* emo_limits_update */
#endif /* EFSYS_OPT_MON_STATS */
};
#endif
@@ -815,6 +816,20 @@ efx_mon_stats_update(
return (emop->emo_stats_update(enp, esmp, values));
}
+ __checkReturn efx_rc_t
+efx_mon_limits_update(
+ __inefx_nic_t *enp,
+ __inout_ecount(EFX_MON_NSTATS) efx_mon_stat_limits_t *values)
+{
+ efx_mon_t *emp = &(enp->en_mon);
+ const efx_mon_ops_t *emop = emp->em_emop;
+
+ EFSYS_ASSERT3U(enp->en_magic, ==, EFX_NIC_MAGIC);
+ EFSYS_ASSERT3U(enp->en_mod_flags, &, EFX_MOD_MON);
+
+ return (emop->emo_limits_update(enp, values));
+}
+
#endif /* EFSYS_OPT_MON_STATS */
void
diff --git a/drivers/net/sfc/base/mcdi_mon.c b/drivers/net/sfc/base/mcdi_mon.c
index 68bbc575d..0e860168a 100644
--- a/drivers/net/sfc/base/mcdi_mon.c
+++ b/drivers/net/sfc/base/mcdi_mon.c
@@ -334,6 +334,87 @@ efx_mcdi_sensor_info(
return (rc);
}
+static __checkReturn efx_rc_t
+efx_mcdi_sensor_info_page(
+ __inefx_nic_t *enp,
+ __inuint32_t page,
+ __out uint32_t *mask_part,
+ __out_ecount((sizeof (*mask_part) * 8) - 1)
+ efx_mon_stat_limits_t *limits)
+{
+ efx_mcdi_req_t req;
+ uint8_t payload[MAX(MC_CMD_SENSOR_INFO_EXT_IN_LEN,
+ MC_CMD_SENSOR_INFO_OUT_LENMAX)];
+ efx_rc_t rc;
+ uint32_t mask_copy;
+ efx_dword_t *maskp;
+ efx_qword_t *limit_info;
+
+ EFSYS_ASSERT(mask_part != NULL);
+ EFSYS_ASSERT(limits != NULL);
+
+ memset(limits, 0,
+ ((sizeof (*mask_part) * 8) - 1) * sizeof (efx_mon_stat_limits_t));
+
+ (void) memset(payload, 0, sizeof (payload));
+ req.emr_cmd = MC_CMD_SENSOR_INFO;
+ req.emr_in_buf = payload;
+ req.emr_in_length = MC_CMD_SENSOR_INFO_EXT_IN_LEN;
+ req.emr_out_buf = payload;
+ req.emr_out_length = MC_CMD_SENSOR_INFO_OUT_LENMAX;
+
+ MCDI_IN_SET_DWORD(req, SENSOR_INFO_EXT_IN_PAGE, page);
+
+ efx_mcdi_execute(enp, &req);
+
+ rc = req.emr_rc;
+
+ if (rc != 0)
+ goto fail1;
+
+ EFSYS_ASSERT(sizeof (*limit_info) ==
+ MC_CMD_SENSOR_INFO_ENTRY_TYPEDEF_LEN);
+ maskp = MCDI_OUT2(req, efx_dword_t, SENSOR_INFO_OUT_MASK);
+ limit_info = (ef
[dpdk-dev] [PATCH 20/37] net/sfc/base: generalise EF10 NVRAM buffer interface
From: Richard Houldsworth
The SFN driver's PartitionControl WMI object requires an API to parse
and filter partition data in TLV format, particularly for the Dynamic
Config partition. The ef10_nvram_buffer functions provide this
functionality but are tied to use with license partition only.
Modify functions so they are applicable to all TLV partitions and add
functions to support in-place tag modification.
Signed-off-by: Richard Houldsworth
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/ef10_impl.h | 43 +++--
drivers/net/sfc/base/ef10_nvram.c | 153 --
drivers/net/sfc/base/efx_impl.h | 2 +-
drivers/net/sfc/base/efx_lic.c| 15 ++-
drivers/net/sfc/base/efx_nvram.c | 2 +-
5 files changed, 174 insertions(+), 41 deletions(-)
diff --git a/drivers/net/sfc/base/ef10_impl.h b/drivers/net/sfc/base/ef10_impl.h
index 2abd699a0..503f02362 100644
--- a/drivers/net/sfc/base/ef10_impl.h
+++ b/drivers/net/sfc/base/ef10_impl.h
@@ -477,17 +477,21 @@ ef10_nvram_partn_set_version(
extern __checkReturn efx_rc_t
ef10_nvram_buffer_validate(
- __inefx_nic_t *enp,
__inuint32_t partn,
__in_bcount(buffer_size)
caddr_t bufferp,
__insize_t buffer_size);
+extern void
+ef10_nvram_buffer_init(
+ __out_bcount(buffer_size)
+ caddr_t bufferp,
+ __insize_t buffer_size);
+
extern __checkReturn efx_rc_t
ef10_nvram_buffer_create(
- __inefx_nic_t *enp,
- __inuint16_t partn_type,
- __in_bcount(buffer_size)
+ __inuint32_t partn_type,
+ __out_bcount(buffer_size)
caddr_t bufferp,
__insize_t buffer_size);
@@ -515,6 +519,16 @@ ef10_nvram_buffer_find_item(
__out uint32_t *startp,
__out uint32_t *lengthp);
+extern __checkReturn efx_rc_t
+ef10_nvram_buffer_peek_item(
+ __in_bcount(buffer_size)
+ caddr_t bufferp,
+ __insize_t buffer_size,
+ __inuint32_t offset,
+ __out uint32_t *tagp,
+ __out uint32_t *lengthp,
+ __out uint32_t *value_offsetp);
+
extern __checkReturn efx_rc_t
ef10_nvram_buffer_get_item(
__in_bcount(buffer_size)
@@ -522,9 +536,10 @@ ef10_nvram_buffer_get_item(
__insize_t buffer_size,
__inuint32_t offset,
__inuint32_t length,
- __out_bcount_part(item_max_size, *lengthp)
- caddr_t itemp,
- __insize_t item_max_size,
+ __out uint32_t *tagp,
+ __out_bcount_part(value_max_size, *lengthp)
+ caddr_t valuep,
+ __insize_t value_max_size,
__out uint32_t *lengthp);
extern __checkReturn efx_rc_t
@@ -533,7 +548,19 @@ ef10_nvram_buffer_insert_item(
caddr_t bufferp,
__insize_t buffer_size,
__inuint32_t offset,
- __in_bcount(length) caddr_t keyp,
+ __inuint32_t tag,
+ __in_bcount(length) caddr_t valuep,
+ __inuint32_t length,
+ __out uint32_t *lengthp);
+
+extern __checkReturn efx_rc_t
+ef10_nvram_buffer_modify_item(
+ __in_bcount(buffer_size)
+ caddr_t bufferp,
+ __insize_t buffer_size,
+ __inuint32_t offset,
+ __inuint32_t tag,
+ __in_bcount(length) caddr_t valuep,
__inuint32_t length,
__out uint32_t *lengthp);
diff --git a/drivers/net/sfc/base/ef10_nvram.c
b/drivers/net/sfc/base/ef10_nvram.c
index 96ea5624a..8d1b64f25 100644
--- a/drivers/net/sfc/base/ef10_nvram.c
+++ b/drivers/net/sfc/base/ef10_nvram.c
@@ -203,14 +203,14 @@ tlv_validate_state(
if (tlv_tag(cursor) != TLV_TAG_END) {
/* Check current item has space for tag and length */
- if (cursor->current > (cursor->limit - 2)) {
+ if (cursor->current > (cursor->limit - 1)) {
cursor->current = NULL;
rc = EFAULT;
goto fail3;
}
- /* Check we have value data for current item and another tag */
- if (tlv_next_item_ptr(cursor) > (cursor->limit - 1)) {
+ /* Check we have value data for current item and an END tag */
[dpdk-dev] [PATCH 26/37] net/sfc/base: fix ID retrival in v3 licensing
From: Andy Moreton
Fixes: 05fce2ce8451 ("net/sfc/base: import libefx licensing")
Fixes: f67e4719147d ("net/sfc/base: fix coding style")
Cc: sta...@dpdk.org
Signed-off-by: Andy Moreton
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/efx_lic.c | 39 --
1 file changed, 9 insertions(+), 30 deletions(-)
diff --git a/drivers/net/sfc/base/efx_lic.c b/drivers/net/sfc/base/efx_lic.c
index 2a6da2647..4081aef1b 100644
--- a/drivers/net/sfc/base/efx_lic.c
+++ b/drivers/net/sfc/base/efx_lic.c
@@ -983,26 +983,14 @@ efx_mcdi_licensing_v3_get_id(
{
efx_mcdi_req_t req;
EFX_MCDI_DECLARE_BUF(payload, MC_CMD_LICENSING_GET_ID_V3_IN_LEN,
- MC_CMD_LICENSING_GET_ID_V3_OUT_LENMIN);
+ MC_CMD_LICENSING_GET_ID_V3_OUT_LENMAX);
efx_rc_t rc;
req.emr_cmd = MC_CMD_LICENSING_GET_ID_V3;
-
- if (bufferp == NULL) {
- /* Request id type and length only */
- req.emr_in_buf = bufferp;
- req.emr_in_length = MC_CMD_LICENSING_GET_ID_V3_IN_LEN;
- req.emr_out_buf = bufferp;
- req.emr_out_length = MC_CMD_LICENSING_GET_ID_V3_OUT_LENMIN;
- } else {
- /* Request full buffer */
- req.emr_in_buf = bufferp;
- req.emr_in_length = MC_CMD_LICENSING_GET_ID_V3_IN_LEN;
- req.emr_out_buf = bufferp;
- req.emr_out_length =
- MIN(buffer_size, MC_CMD_LICENSING_GET_ID_V3_OUT_LENMAX);
- (void) memset(bufferp, 0, req.emr_out_length);
- }
+ req.emr_in_buf = payload;
+ req.emr_in_length = MC_CMD_LICENSING_GET_ID_V3_IN_LEN;
+ req.emr_out_buf = payload;
+ req.emr_out_length = MC_CMD_LICENSING_GET_ID_V3_OUT_LENMAX;
efx_mcdi_execute_quiet(enp, &req);
@@ -1020,19 +1008,10 @@ efx_mcdi_licensing_v3_get_id(
*lengthp =
MCDI_OUT_DWORD(req, LICENSING_GET_ID_V3_OUT_LICENSE_ID_LENGTH);
- if (bufferp == NULL) {
- /*
-* Modify length requirements to indicate to caller the extra
-* buffering needed to read the complete output.
-*/
- *lengthp += MC_CMD_LICENSING_GET_ID_V3_OUT_LENMIN;
- } else {
- /* Shift ID down to start of buffer */
- memmove(bufferp,
- bufferp + MC_CMD_LICENSING_GET_ID_V3_OUT_LICENSE_ID_OFST,
- *lengthp);
- memset(bufferp + (*lengthp), 0,
- MC_CMD_LICENSING_GET_ID_V3_OUT_LICENSE_ID_OFST);
+ if (bufferp != NULL) {
+ memcpy(bufferp,
+ payload + MC_CMD_LICENSING_GET_ID_V3_OUT_LICENSE_ID_OFST,
+ MIN(buffer_size, *lengthp));
}
return (0);
--
2.17.1
[dpdk-dev] [PATCH 24/37] net/sfc/base: add routine to check for hardware presence
From: Andy Moreton
Add efx_nic_hw_unavailable() routine to check for hardware presence
before continuing with NIC operations.
Signed-off-by: Andy Moreton
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/ef10_ev.c | 10 ++
drivers/net/sfc/base/ef10_impl.h | 4
drivers/net/sfc/base/ef10_nic.c | 22 ++
drivers/net/sfc/base/efx.h | 4
drivers/net/sfc/base/efx_impl.h | 2 ++
drivers/net/sfc/base/efx_mcdi.c | 6 ++
drivers/net/sfc/base/efx_nic.c | 27 +++
7 files changed, 71 insertions(+), 4 deletions(-)
diff --git a/drivers/net/sfc/base/ef10_ev.c b/drivers/net/sfc/base/ef10_ev.c
index 287550605..cdf835f03 100644
--- a/drivers/net/sfc/base/ef10_ev.c
+++ b/drivers/net/sfc/base/ef10_ev.c
@@ -863,8 +863,9 @@ ef10_ev_rx(
EFX_EV_QSTAT_INCR(eep, EV_RX);
- /* Discard events after RXQ/TXQ errors */
- if (enp->en_reset_flags & (EFX_RESET_RXQ_ERR | EFX_RESET_TXQ_ERR))
+ /* Discard events after RXQ/TXQ errors, or hardware not available */
+ if (enp->en_reset_flags &
+ (EFX_RESET_RXQ_ERR | EFX_RESET_TXQ_ERR | EFX_RESET_HW_UNAVAIL))
return (B_FALSE);
/* Basic packet information */
@@ -1064,8 +1065,9 @@ ef10_ev_tx(
EFX_EV_QSTAT_INCR(eep, EV_TX);
- /* Discard events after RXQ/TXQ errors */
- if (enp->en_reset_flags & (EFX_RESET_RXQ_ERR | EFX_RESET_TXQ_ERR))
+ /* Discard events after RXQ/TXQ errors, or hardware not available */
+ if (enp->en_reset_flags &
+ (EFX_RESET_RXQ_ERR | EFX_RESET_TXQ_ERR | EFX_RESET_HW_UNAVAIL))
return (B_FALSE);
if (EFX_QWORD_FIELD(*eqp, ESF_DZ_TX_DROP_EVENT) != 0) {
diff --git a/drivers/net/sfc/base/ef10_impl.h b/drivers/net/sfc/base/ef10_impl.h
index 503f02362..0214a75ef 100644
--- a/drivers/net/sfc/base/ef10_impl.h
+++ b/drivers/net/sfc/base/ef10_impl.h
@@ -190,6 +190,10 @@ extern __checkReturn efx_rc_t
ef10_nic_init(
__inefx_nic_t *enp);
+extern __checkReturn boolean_t
+ef10_nic_hw_unavailable(
+ __inefx_nic_t *enp);
+
#if EFSYS_OPT_DIAG
extern __checkReturn efx_rc_t
diff --git a/drivers/net/sfc/base/ef10_nic.c b/drivers/net/sfc/base/ef10_nic.c
index 332f6ef81..ff96a7ff2 100644
--- a/drivers/net/sfc/base/ef10_nic.c
+++ b/drivers/net/sfc/base/ef10_nic.c
@@ -2304,6 +2304,28 @@ ef10_nic_get_bar_region(
return (rc);
}
+ __checkReturn boolean_t
+ef10_nic_hw_unavailable(
+ __inefx_nic_t *enp)
+{
+ efx_dword_t dword;
+
+ if (enp->en_reset_flags & EFX_RESET_HW_UNAVAIL)
+ return (B_TRUE);
+
+ EFX_BAR_READD(enp, ER_DZ_BIU_MC_SFT_STATUS_REG, &dword, B_FALSE);
+ if (EFX_DWORD_FIELD(dword, EFX_DWORD_0) == 0x)
+ goto unavail;
+
+ return (B_FALSE);
+
+unavail:
+ EFSYS_PROBE(hw_unavail);
+ enp->en_reset_flags |= EFX_RESET_HW_UNAVAIL;
+
+ return (B_TRUE);
+}
+
void
ef10_nic_fini(
__inefx_nic_t *enp)
diff --git a/drivers/net/sfc/base/efx.h b/drivers/net/sfc/base/efx.h
index 2356a9294..fce519037 100644
--- a/drivers/net/sfc/base/efx.h
+++ b/drivers/net/sfc/base/efx.h
@@ -155,6 +155,10 @@ extern __checkReturn efx_rc_t
efx_nic_reset(
__inefx_nic_t *enp);
+extern __checkReturn boolean_t
+efx_nic_hw_unavailable(
+ __inefx_nic_t *enp);
+
#if EFSYS_OPT_DIAG
extern __checkReturn efx_rc_t
diff --git a/drivers/net/sfc/base/efx_impl.h b/drivers/net/sfc/base/efx_impl.h
index 78df43018..8a7dc8cf6 100644
--- a/drivers/net/sfc/base/efx_impl.h
+++ b/drivers/net/sfc/base/efx_impl.h
@@ -59,6 +59,7 @@ extern "C" {
#defineEFX_RESET_PHY 0x0001
#defineEFX_RESET_RXQ_ERR 0x0002
#defineEFX_RESET_TXQ_ERR 0x0004
+#defineEFX_RESET_HW_UNAVAIL0x0008
typedef enum efx_mac_type_e {
EFX_MAC_INVALID = 0,
@@ -356,6 +357,7 @@ typedef struct efx_nic_ops_s {
efx_rc_t(*eno_get_vi_pool)(efx_nic_t *, uint32_t *);
efx_rc_t(*eno_get_bar_region)(efx_nic_t *, efx_nic_region_t,
uint32_t *, size_t *);
+ boolean_t (*eno_hw_unavailable)(efx_nic_t *);
#if EFSYS_OPT_DIAG
efx_rc_t(*eno_register_test)(efx_nic_t *);
#endif /* EFSYS_OPT_DIAG */
diff --git a/drivers/net/sfc/base/efx_mcdi.c b/drivers/net/sfc/base/efx_mcdi.c
index 84d8452e4..c8d670c23 100644
--- a/drivers/net/sfc/base/efx_mcdi.c
+++ b/drivers/net/sfc/base/efx_mcdi.c
@@ -496,6 +496,12 @@ efx_mcdi_request_poll(
EFSYS_ASSERT(!emip->emi_ev_cpl);
emrp = emip->emi_pending_req;
+ /* Check if hardware is unavailable */
+ if (efx_nic_hw_unavailable(enp)) {
+ EFSYS_UNLOCK(enp->en_eslp, state);
+ return (B_FALSE);
+ }
+
/* Check for reboot atomically w.
[dpdk-dev] [PATCH 22/37] net/sfc/base: add information if TSO workaround is required
From: Mark Spender
In SF bug 61297 it's been confirmed that the hardware does not always
calculate the TCP checksum correctly with TSO sends.
The value of the Total Length field (IPv4) or Payload Length field
(IPv6) is the critical factor. We're sufficiently confident that if
these fields are zero then the checksum will be calculated correctly.
The information may be used by the drivers to check if the workaround is
required when FATSOv2 is implemented.
Signed-off-by: Mark Spender
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/efx.h | 1 +
drivers/net/sfc/base/hunt_nic.c | 3 +++
drivers/net/sfc/base/medford2_nic.c | 3 +++
drivers/net/sfc/base/medford_nic.c | 3 +++
4 files changed, 10 insertions(+)
diff --git a/drivers/net/sfc/base/efx.h b/drivers/net/sfc/base/efx.h
index 3c3227739..2356a9294 100644
--- a/drivers/net/sfc/base/efx.h
+++ b/drivers/net/sfc/base/efx.h
@@ -1292,6 +1292,7 @@ typedef struct efx_nic_cfg_s {
boolean_t enc_bug35388_workaround;
boolean_t enc_bug41750_workaround;
boolean_t enc_bug61265_workaround;
+ boolean_t enc_bug61297_workaround;
boolean_t enc_rx_batching_enabled;
/* Maximum number of descriptors completed in an rx event. */
uint32_tenc_rx_batch_max;
diff --git a/drivers/net/sfc/base/hunt_nic.c b/drivers/net/sfc/base/hunt_nic.c
index 1bec3c485..70c042f3f 100644
--- a/drivers/net/sfc/base/hunt_nic.c
+++ b/drivers/net/sfc/base/hunt_nic.c
@@ -190,6 +190,9 @@ hunt_board_cfg(
encp->enc_bug61265_workaround = B_FALSE; /* Medford only */
+ /* Checksums for TSO sends can be incorrect on Huntington. */
+ encp->enc_bug61297_workaround = B_TRUE;
+
/* Alignment for receive packet DMA buffers */
encp->enc_rx_buf_align_start = 1;
encp->enc_rx_buf_align_end = 64; /* RX DMA end padding */
diff --git a/drivers/net/sfc/base/medford2_nic.c
b/drivers/net/sfc/base/medford2_nic.c
index b36e54bab..3efc35886 100644
--- a/drivers/net/sfc/base/medford2_nic.c
+++ b/drivers/net/sfc/base/medford2_nic.c
@@ -96,6 +96,9 @@ medford2_board_cfg(
else
goto fail1;
+ /* Checksums for TSO sends should always be correct on Medford2. */
+ encp->enc_bug61297_workaround = B_FALSE;
+
/* Get clock frequencies (in MHz). */
if ((rc = efx_mcdi_get_clock(enp, &sysclk, &dpcpu_clk)) != 0)
goto fail2;
diff --git a/drivers/net/sfc/base/medford_nic.c
b/drivers/net/sfc/base/medford_nic.c
index 96f3a1204..4f1896343 100644
--- a/drivers/net/sfc/base/medford_nic.c
+++ b/drivers/net/sfc/base/medford_nic.c
@@ -94,6 +94,9 @@ medford_board_cfg(
else
goto fail1;
+ /* Checksums for TSO sends can be incorrect on Medford. */
+ encp->enc_bug61297_workaround = B_TRUE;
+
/* Get clock frequencies (in MHz). */
if ((rc = efx_mcdi_get_clock(enp, &sysclk, &dpcpu_clk)) != 0)
goto fail2;
--
2.17.1
[dpdk-dev] [PATCH 18/37] net/sfc/base: add buffer editing functions to boot config
From: Richard Houldsworth Functions to process the DHCP option list format used by the expansion ROM config buffers, to support extracting and updating of individual options. The initial use case is the driver presenting the global and per-PF options as separate items, with the driver implementing the synchronization of global options across the configuration buffers for all PFs. Signed-off-by: Richard Houldsworth Signed-off-by: Andrew Rybchenko --- drivers/net/sfc/base/efx.h | 81 drivers/net/sfc/base/efx_annote.h | 1 + drivers/net/sfc/base/efx_bootcfg.c | 641 ++--- 3 files changed, 672 insertions(+), 51 deletions(-) diff --git a/drivers/net/sfc/base/efx.h b/drivers/net/sfc/base/efx.h index 099f9df67..3c3227739 100644 --- a/drivers/net/sfc/base/efx.h +++ b/drivers/net/sfc/base/efx.h @@ -1662,6 +1662,87 @@ efx_bootcfg_write( __in_bcount(size) uint8_t *data, __insize_t size); + +/* + * Processing routines for buffers arranged in the DHCP/BOOTP option format + * (see https://tools.ietf.org/html/rfc1533) + * + * Summarising the format: the buffer is a sequence of options. All options + * begin with a tag octet, which uniquely identifies the option. Fixed- + * length options without data consist of only a tag octet. Only options PAD + * (0) and END (255) are fixed length. All other options are variable-length + * with a length octet following the tag octet. The value of the length + * octet does not include the two octets specifying the tag and length. The + * length octet is followed by "length" octets of data. + * + * Option data may be a sequence of sub-options in the same format. The data + * content of the encapsulating option is one or more encapsulated sub-options, + * with no terminating END tag is required. + * + * To be valid, the top-level sequence of options should be terminated by an + * END tag. The buffer should be padded with the PAD byte. + * + * When stored to NVRAM, the DHCP option format buffer is preceded by a + * checksum octet. The full buffer (including after the END tag) contributes + * to the checksum, hence the need to fill the buffer to the end with PAD. + */ + +#defineEFX_DHCP_END ((uint8_t)0xff) +#defineEFX_DHCP_PAD ((uint8_t)0) + +#defineEFX_DHCP_ENCAP_OPT(encapsulator, encapsulated) \ + (uint16_t)(((encapsulator) << 8) | (encapsulated)) + +extern __checkReturn uint8_t +efx_dhcp_csum( + __in_bcount(size) uint8_t const *data, + __insize_t size); + +extern __checkReturn efx_rc_t +efx_dhcp_verify( + __in_bcount(size) uint8_t const *data, + __insize_t size, + __out_opt size_t *usedp); + +extern __checkReturn efx_rc_t +efx_dhcp_find_tag( + __in_bcount(buffer_length) uint8_t *bufferp, + __insize_t buffer_length, + __inuint16_t opt, + __deref_out uint8_t **valuepp, + __out size_t *value_lengthp); + +extern __checkReturn efx_rc_t +efx_dhcp_find_end( + __in_bcount(buffer_length) uint8_t *bufferp, + __insize_t buffer_length, + __deref_out uint8_t **endpp); + + +extern __checkReturn efx_rc_t +efx_dhcp_delete_tag( + __inout_bcount(buffer_length) uint8_t *bufferp, + __insize_t buffer_length, + __inuint16_t opt); + +extern __checkReturn efx_rc_t +efx_dhcp_add_tag( + __inout_bcount(buffer_length) uint8_t *bufferp, + __insize_t buffer_length, + __inuint16_t opt, + __in_bcount_opt(value_length) uint8_t *valuep, + __insize_t value_length); + +extern __checkReturn efx_rc_t +efx_dhcp_update_tag( + __inout_bcount(buffer_length) uint8_t *bufferp, + __insize_t buffer_length, + __inuint16_t opt, + __inuint8_t *value_locationp, + __in_bcount_opt(value_length) uint8_t *valuep, + __insize_t value_length); + + #endif /* EFSYS_OPT_BOOTCFG */ #if EFSYS_OPT_IMAGE_LAYOUT diff --git a/drivers/net/sfc/base/efx_annote.h b/drivers/net/sfc/base/efx_annote.h index 603260e22..671aaed3b 100644 --- a/drivers/net/sfc/base/efx_annote.h +++ b/drivers/net/sfc/base/efx_annote.h @@ -39,6 +39,7 @@ #define__out_bcount_part_opt(_n, _l) #define__deref_out +#define__deref_inout #define__inout #define__inout_opt diff --git a/drivers/net/sfc/base/efx_bootcfg.c b/drivers/net/sfc/base/efx_bootcfg.c index 715e18e8c..3b0401e89 100644 --- a/drivers/net/sfc/base/efx_bootcfg.c +++ b/drivers/net/sfc
[dpdk-dev] [PATCH 28/37] net/sfc/base: fix name of the argument to store RSS flags
From: Ivan Malov
The function used to retrieve supported RSS flags has an
argument which should be named properly to indicate
that it's a pointer.
Fixes: 613cbe75ae99 ("net/sfc/base: add a new means to control RSS hash")
Cc: sta...@dpdk.org
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/efx.h| 2 +-
drivers/net/sfc/base/efx_rx.c | 8
2 files changed, 5 insertions(+), 5 deletions(-)
diff --git a/drivers/net/sfc/base/efx.h b/drivers/net/sfc/base/efx.h
index 0982a34d6..15b3882dd 100644
--- a/drivers/net/sfc/base/efx.h
+++ b/drivers/net/sfc/base/efx.h
@@ -2368,7 +2368,7 @@ extern__checkReturn efx_rc_t
efx_rx_scale_hash_flags_get(
__inefx_nic_t *enp,
__inefx_rx_hash_alg_t hash_alg,
- __inout_ecount(EFX_RX_HASH_NFLAGS) unsigned int *flags,
+ __inout_ecount(EFX_RX_HASH_NFLAGS) unsigned int *flagsp,
__out unsigned int *nflagsp);
extern __checkReturn efx_rc_t
diff --git a/drivers/net/sfc/base/efx_rx.c b/drivers/net/sfc/base/efx_rx.c
index 09933b41a..7b9f28664 100644
--- a/drivers/net/sfc/base/efx_rx.c
+++ b/drivers/net/sfc/base/efx_rx.c
@@ -298,16 +298,16 @@ efx_rx_scatter_enable(
efx_rx_scale_hash_flags_get(
__inefx_nic_t *enp,
__inefx_rx_hash_alg_t hash_alg,
- __inout_ecount(EFX_RX_HASH_NFLAGS) unsigned int *flags,
+ __inout_ecount(EFX_RX_HASH_NFLAGS) unsigned int *flagsp,
__out unsigned int *nflagsp)
{
efx_nic_cfg_t *encp = &enp->en_nic_cfg;
boolean_t l4;
boolean_t additional_modes;
- unsigned int *entryp = flags;
+ unsigned int *entryp = flagsp;
efx_rc_t rc;
- if (flags == NULL || nflagsp == NULL) {
+ if (flagsp == NULL || nflagsp == NULL) {
rc = EINVAL;
goto fail1;
}
@@ -368,7 +368,7 @@ efx_rx_scale_hash_flags_get(
#undef LIST_FLAGS
- *nflagsp = (unsigned int)(entryp - flags);
+ *nflagsp = (unsigned int)(entryp - flagsp);
EFSYS_ASSERT3U(*nflagsp, <=, EFX_RX_HASH_NFLAGS);
return (0);
--
2.17.1
[dpdk-dev] [PATCH 29/37] net/sfc/base: fix a typo in unicast filter insertion comment
From: Ivan Malov
Fixes: e7cd430c864f ("net/sfc/base: import SFN7xxx family support")
Cc: sta...@dpdk.org
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/ef10_filter.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/sfc/base/ef10_filter.c
b/drivers/net/sfc/base/ef10_filter.c
index 30a4892df..afe4064d9 100644
--- a/drivers/net/sfc/base/ef10_filter.c
+++ b/drivers/net/sfc/base/ef10_filter.c
@@ -1578,7 +1578,7 @@ ef10_filter_reconfigure(
/*
* Insert or renew unicast filters.
*
-* Frimware does not perform chaining on unicast filters. As traffic is
+* Firmware does not perform chaining on unicast filters. As traffic is
* therefore only delivered to the first matching filter, we should
* always insert the specific filter for our MAC address, to try and
* ensure we get that traffic.
--
2.17.1
[dpdk-dev] [PATCH 27/37] net/sfc/base: prevent access to the NIC config before probe
From: Mark Spender
NIC config is initialized during NIC probe.
Fixes: 19b64c6ac35f ("net/sfc/base: import libefx base")
Cc: sta...@dpdk.org
Signed-off-by: Mark Spender
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/efx_nic.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/sfc/base/efx_nic.c b/drivers/net/sfc/base/efx_nic.c
index e5cb0105f..cea32b792 100644
--- a/drivers/net/sfc/base/efx_nic.c
+++ b/drivers/net/sfc/base/efx_nic.c
@@ -595,6 +595,7 @@ efx_nic_cfg_get(
__inefx_nic_t *enp)
{
EFSYS_ASSERT3U(enp->en_magic, ==, EFX_NIC_MAGIC);
+ EFSYS_ASSERT3U(enp->en_mod_flags, &, EFX_MOD_PROBE);
return (&(enp->en_nic_cfg));
}
--
2.17.1
[dpdk-dev] [PATCH 31/37] net/sfc/base: use simpler code to check hash algorithm type
From: Ivan Malov
The API which is used to list supported hash flags verifies
hash algorithm choice before writing the output. This check
is based on a switch() statement which has only two options
and no distinctive actions to be conducted for each of them.
Use simpler code instead of switch() to improve readability.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/efx_rx.c | 39 ---
1 file changed, 13 insertions(+), 26 deletions(-)
diff --git a/drivers/net/sfc/base/efx_rx.c b/drivers/net/sfc/base/efx_rx.c
index 7b9f28664..dfd3974da 100644
--- a/drivers/net/sfc/base/efx_rx.c
+++ b/drivers/net/sfc/base/efx_rx.c
@@ -312,6 +312,11 @@ efx_rx_scale_hash_flags_get(
goto fail1;
}
+ if ((encp->enc_rx_scale_hash_alg_mask & (1U << hash_alg)) == 0) {
+ *nflagsp = 0;
+ return 0;
+ }
+
l4 = encp->enc_rx_scale_l4_hash_supported;
additional_modes = encp->enc_rx_scale_additional_modes_supported;
@@ -340,32 +345,17 @@ efx_rx_scale_hash_flags_get(
_NOTE(CONSTANTCONDITION)\
} while (B_FALSE)
- switch (hash_alg) {
- case EFX_RX_HASHALG_PACKED_STREAM:
- if ((encp->enc_rx_scale_hash_alg_mask & (1U << hash_alg)) == 0)
- break;
- /* FALLTHRU */
- case EFX_RX_HASHALG_TOEPLITZ:
- if ((encp->enc_rx_scale_hash_alg_mask & (1U << hash_alg)) == 0)
- break;
+ LIST_FLAGS(entryp, IPV4_TCP, l4, additional_modes);
+ LIST_FLAGS(entryp, IPV6_TCP, l4, additional_modes);
- LIST_FLAGS(entryp, IPV4_TCP, l4, additional_modes);
- LIST_FLAGS(entryp, IPV6_TCP, l4, additional_modes);
-
- if (additional_modes) {
- LIST_FLAGS(entryp, IPV4_UDP, l4, additional_modes);
- LIST_FLAGS(entryp, IPV6_UDP, l4, additional_modes);
- }
-
- LIST_FLAGS(entryp, IPV4, B_FALSE, additional_modes);
- LIST_FLAGS(entryp, IPV6, B_FALSE, additional_modes);
- break;
-
- default:
- rc = EINVAL;
- goto fail2;
+ if (additional_modes) {
+ LIST_FLAGS(entryp, IPV4_UDP, l4, additional_modes);
+ LIST_FLAGS(entryp, IPV6_UDP, l4, additional_modes);
}
+ LIST_FLAGS(entryp, IPV4, B_FALSE, additional_modes);
+ LIST_FLAGS(entryp, IPV6, B_FALSE, additional_modes);
+
#undef LIST_FLAGS
*nflagsp = (unsigned int)(entryp - flagsp);
@@ -373,9 +363,6 @@ efx_rx_scale_hash_flags_get(
return (0);
-fail2:
- EFSYS_PROBE(fail2);
-
fail1:
EFSYS_PROBE1(fail1, efx_rc_t, rc);
--
2.17.1
[dpdk-dev] [PATCH 23/37] net/sfc/base: fix out of bounds read when dereferencing sdup
From: Gautam Dawar
Introduce and use macro to make sure that MCDI buffers allocated
on stack are rounded up properly.
Fixes: 6f619653b9b1 ("net/sfc/base: import MCDI implementation")
Fixes: f7dc06bf35f2 ("net/sfc/base: import 5xxx/6xxx family support")
Fixes: e7cd430c864f ("net/sfc/base: import SFN7xxx family support")
Fixes: 1dae25112a54 ("net/sfc/base: import built-in selftest")
Fixes: 0a7864349106 ("net/sfc/base: import PHY statistics")
Fixes: 8c7c723dfe7c ("net/sfc/base: import MAC statistics")
Fixes: 5935cd8c47d3 ("net/sfc/base: import RSS support")
Fixes: 9ee64bd404fc ("net/sfc/base: import loopback control")
Fixes: dfb3b1ce15f6 ("net/sfc/base: import monitors access via MCDI")
Fixes: d96a34d165b1 ("net/sfc/base: import NVRAM support")
Fixes: 05fce2ce8451 ("net/sfc/base: import libefx licensing")
Fixes: ba6afee9a81e ("net/sfc/base: add advanced function to extract FW
version")
Fixes: c7815c1d1f20 ("net/sfc/base: use proper MCDI command for encap filters")
Fixes: 17551f6dffcc ("net/sfc/base: add API to control UDP tunnel ports")
Fixes: eff9b666eae5 ("net/sfc/base: move RxDP config get to EF10 NIC code")
Fixes: 4aab7f07a645 ("net/sfc/base: refactor EF10 get datapath capabilities")
Fixes: 480a13044b8b ("net/sfc/base: support FW subvariant choice")
Fixes: 6f60cc4a78b6 ("net/sfc/base: support equal stride super-buffer Rx mode")
Fixes: 9a733758c046 ("net/sfc/base: support MARK and FLAG actions in filters")
Cc: sta...@dpdk.org
Signed-off-by: Gautam Dawar
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/ef10_ev.c | 28 -
drivers/net/sfc/base/ef10_filter.c | 15 ++---
drivers/net/sfc/base/ef10_intr.c | 5 +-
drivers/net/sfc/base/ef10_mac.c| 20 +++
drivers/net/sfc/base/ef10_nic.c| 93 --
drivers/net/sfc/base/ef10_phy.c| 19 +++---
drivers/net/sfc/base/ef10_rx.c | 35 +--
drivers/net/sfc/base/ef10_tx.c | 10 ++--
drivers/net/sfc/base/efx_lic.c | 39 +
drivers/net/sfc/base/efx_mcdi.c| 82 +++---
drivers/net/sfc/base/efx_mcdi.h| 11
drivers/net/sfc/base/efx_nic.c | 5 +-
drivers/net/sfc/base/efx_nvram.c | 40 +
drivers/net/sfc/base/efx_tunnel.c | 6 +-
drivers/net/sfc/base/mcdi_mon.c| 17 +++---
drivers/net/sfc/base/siena_mac.c | 9 ++-
drivers/net/sfc/base/siena_nic.c | 5 +-
drivers/net/sfc/base/siena_nvram.c | 5 +-
drivers/net/sfc/base/siena_phy.c | 28 -
19 files changed, 199 insertions(+), 273 deletions(-)
diff --git a/drivers/net/sfc/base/ef10_ev.c b/drivers/net/sfc/base/ef10_ev.c
index 7f89a7bf0..287550605 100644
--- a/drivers/net/sfc/base/ef10_ev.c
+++ b/drivers/net/sfc/base/ef10_ev.c
@@ -73,11 +73,10 @@ efx_mcdi_set_evq_tmr(
__inuint32_t timer_ns)
{
efx_mcdi_req_t req;
- uint8_t payload[MAX(MC_CMD_SET_EVQ_TMR_IN_LEN,
- MC_CMD_SET_EVQ_TMR_OUT_LEN)];
+ EFX_MCDI_DECLARE_BUF(payload, MC_CMD_SET_EVQ_TMR_IN_LEN,
+ MC_CMD_SET_EVQ_TMR_OUT_LEN);
efx_rc_t rc;
- (void) memset(payload, 0, sizeof (payload));
req.emr_cmd = MC_CMD_SET_EVQ_TMR;
req.emr_in_buf = payload;
req.emr_in_length = MC_CMD_SET_EVQ_TMR_IN_LEN;
@@ -123,9 +122,9 @@ efx_mcdi_init_evq(
__inboolean_t low_latency)
{
efx_mcdi_req_t req;
- uint8_t payload[
- MAX(MC_CMD_INIT_EVQ_IN_LEN(EFX_EVQ_NBUFS(EFX_EVQ_MAXNEVS)),
- MC_CMD_INIT_EVQ_OUT_LEN)];
+ EFX_MCDI_DECLARE_BUF(payload,
+ MC_CMD_INIT_EVQ_IN_LEN(EFX_EVQ_NBUFS(EFX_EVQ_MAXNEVS)),
+ MC_CMD_INIT_EVQ_OUT_LEN);
efx_qword_t *dma_addr;
uint64_t addr;
int npages;
@@ -140,7 +139,6 @@ efx_mcdi_init_evq(
goto fail1;
}
- (void) memset(payload, 0, sizeof (payload));
req.emr_cmd = MC_CMD_INIT_EVQ;
req.emr_in_buf = payload;
req.emr_in_length = MC_CMD_INIT_EVQ_IN_LEN(npages);
@@ -260,9 +258,9 @@ efx_mcdi_init_evq_v2(
__inuint32_t flags)
{
efx_mcdi_req_t req;
- uint8_t payload[
- MAX(MC_CMD_INIT_EVQ_V2_IN_LEN(EFX_EVQ_NBUFS(EFX_EVQ_MAXNEVS)),
- MC_CMD_INIT_EVQ_V2_OUT_LEN)];
+ EFX_MCDI_DECLARE_BUF(payload,
+ MC_CMD_INIT_EVQ_V2_IN_LEN(EFX_EVQ_NBUFS(EFX_EVQ_MAXNEVS)),
+ MC_CMD_INIT_EVQ_V2_OUT_LEN);
boolean_t interrupting;
unsigned int evq_type;
efx_qword_t *dma_addr;
@@ -277,7 +275,6 @@ efx_mcdi_init_evq_v2(
goto fail1;
}
- (void) memset(payload, 0, sizeof (payload));
req.emr_cmd = MC_CMD_INIT_EVQ;
req.emr_in_buf = payload;
req.emr_in_length = MC_CMD_INIT_EVQ_V2_IN_LEN(npages);
@@ -384,11 +381,10 @@ efx_mcdi_fini_evq(
__inuint32_t instance)
{
efx_mcdi_req_t req;
- uint8_t payload[MAX(MC_CMD_FINI_EVQ_IN_LEN,
- MC_C
[dpdk-dev] [PATCH 33/37] net/sfc/base: simplify the code to parse RSS hash type
From: Ivan Malov
RSS mode bits can be accessed a lot easier in the hash
type value provided that the variable type is uint32_t.
The macro helper can be removed to enhance readability.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/ef10_rx.c | 24 ++--
drivers/net/sfc/base/efx.h | 2 +-
2 files changed, 11 insertions(+), 15 deletions(-)
diff --git a/drivers/net/sfc/base/ef10_rx.c b/drivers/net/sfc/base/ef10_rx.c
index 17678b517..1444eca81 100644
--- a/drivers/net/sfc/base/ef10_rx.c
+++ b/drivers/net/sfc/base/ef10_rx.c
@@ -377,12 +377,6 @@ efx_mcdi_rss_context_set_flags(
if (encp->enc_rx_scale_additional_modes_supported == B_FALSE)
modes = 0;
-#defineEXTRACT_RSS_MODE(_type, _class) \
- (EFX_EXTRACT_NATIVE(_type, 0, 31, \
- EFX_LOW_BIT(EFX_RX_CLASS_##_class), \
- EFX_HIGH_BIT(EFX_RX_CLASS_##_class)) & \
- EFX_MASK32(EFX_RX_CLASS_##_class))
-
MCDI_IN_POPULATE_DWORD_10(req, RSS_CONTEXT_SET_FLAGS_IN_FLAGS,
RSS_CONTEXT_SET_FLAGS_IN_TOEPLITZ_IPV4_EN,
((type & type_ipv4) == type_ipv4) ? 1 : 0,
@@ -393,19 +387,21 @@ efx_mcdi_rss_context_set_flags(
RSS_CONTEXT_SET_FLAGS_IN_TOEPLITZ_TCPV6_EN,
((type & type_ipv6_tcp) == type_ipv6_tcp) ? 1 : 0,
RSS_CONTEXT_SET_FLAGS_IN_TCP_IPV4_RSS_MODE,
- EXTRACT_RSS_MODE(modes, IPV4_TCP),
+ (modes >> EFX_RX_CLASS_IPV4_TCP_LBN) &
+ EFX_MASK32(EFX_RX_CLASS_IPV4_TCP),
RSS_CONTEXT_SET_FLAGS_IN_UDP_IPV4_RSS_MODE,
- EXTRACT_RSS_MODE(modes, IPV4_UDP),
+ (modes >> EFX_RX_CLASS_IPV4_UDP_LBN) &
+ EFX_MASK32(EFX_RX_CLASS_IPV4_UDP),
RSS_CONTEXT_SET_FLAGS_IN_OTHER_IPV4_RSS_MODE,
- EXTRACT_RSS_MODE(modes, IPV4),
+ (modes >> EFX_RX_CLASS_IPV4_LBN) & EFX_MASK32(EFX_RX_CLASS_IPV4),
RSS_CONTEXT_SET_FLAGS_IN_TCP_IPV6_RSS_MODE,
- EXTRACT_RSS_MODE(modes, IPV6_TCP),
+ (modes >> EFX_RX_CLASS_IPV6_TCP_LBN) &
+ EFX_MASK32(EFX_RX_CLASS_IPV6_TCP),
RSS_CONTEXT_SET_FLAGS_IN_UDP_IPV6_RSS_MODE,
- EXTRACT_RSS_MODE(modes, IPV6_UDP),
+ (modes >> EFX_RX_CLASS_IPV6_UDP_LBN) &
+ EFX_MASK32(EFX_RX_CLASS_IPV6_UDP),
RSS_CONTEXT_SET_FLAGS_IN_OTHER_IPV6_RSS_MODE,
- EXTRACT_RSS_MODE(modes, IPV6));
-
-#undef EXTRACT_RSS_MODE
+ (modes >> EFX_RX_CLASS_IPV6_LBN) & EFX_MASK32(EFX_RX_CLASS_IPV6));
efx_mcdi_execute(enp, &req);
diff --git a/drivers/net/sfc/base/efx.h b/drivers/net/sfc/base/efx.h
index de62b7d50..62c56e6e1 100644
--- a/drivers/net/sfc/base/efx.h
+++ b/drivers/net/sfc/base/efx.h
@@ -2267,7 +2267,7 @@ typedef enum efx_rx_hash_alg_e {
* - a combination of legacy flags
* - a combination of EFX_RX_HASH() flags
*/
-typedef unsigned int efx_rx_hash_type_t;
+typedef uint32_t efx_rx_hash_type_t;
typedef enum efx_rx_hash_support_e {
EFX_RX_HASH_UNAVAILABLE = 0,/* Hardware hash not inserted */
--
2.17.1
[dpdk-dev] [PATCH 35/37] net/sfc/base: modify phy caps to indicate FEC request
From: Richard Houldsworth
The capability bits to request FEC modes are implicitly valid
when the corresponding FEC mode is a supported capability.
Drivers expect that it is only valid to advertise those
capabilities explicitly marked as supported. The capabilities
reported by firmware is modified with the implicit capabilities
to present the explicit model to drivers.
Signed-off-by: Richard Houldsworth
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/ef10_nic.c | 15 +++
drivers/net/sfc/base/efx_phy.c | 8 +---
2 files changed, 16 insertions(+), 7 deletions(-)
diff --git a/drivers/net/sfc/base/ef10_nic.c b/drivers/net/sfc/base/ef10_nic.c
index 0a2474f3e..b54cd3940 100644
--- a/drivers/net/sfc/base/ef10_nic.c
+++ b/drivers/net/sfc/base/ef10_nic.c
@@ -1772,6 +1772,21 @@ ef10_nic_board_cfg(
if ((rc = efx_mcdi_get_phy_cfg(enp)) != 0)
goto fail6;
+ /*
+* Firmware with support for *_FEC capability bits does not
+* report that the corresponding *_FEC_REQUESTED bits are supported.
+* Add them here so that drivers understand that they are supported.
+*/
+ if (epp->ep_phy_cap_mask & (1u << EFX_PHY_CAP_BASER_FEC))
+ epp->ep_phy_cap_mask |=
+ (1u << EFX_PHY_CAP_BASER_FEC_REQUESTED);
+ if (epp->ep_phy_cap_mask & (1u << EFX_PHY_CAP_RS_FEC))
+ epp->ep_phy_cap_mask |=
+ (1u << EFX_PHY_CAP_RS_FEC_REQUESTED);
+ if (epp->ep_phy_cap_mask & (1u << EFX_PHY_CAP_25G_BASER_FEC))
+ epp->ep_phy_cap_mask |=
+ (1u << EFX_PHY_CAP_25G_BASER_FEC_REQUESTED);
+
/* Obtain the default PHY advertised capabilities */
if ((rc = ef10_phy_get_link(enp, &els)) != 0)
goto fail7;
diff --git a/drivers/net/sfc/base/efx_phy.c b/drivers/net/sfc/base/efx_phy.c
index 7c341e429..25059dfe1 100644
--- a/drivers/net/sfc/base/efx_phy.c
+++ b/drivers/net/sfc/base/efx_phy.c
@@ -192,11 +192,6 @@ efx_phy_adv_cap_get(
}
}
-#defineEFX_PHY_CAP_FEC_REQ_MASK\
- (1U << EFX_PHY_CAP_BASER_FEC_REQUESTED) | \
- (1U << EFX_PHY_CAP_RS_FEC_REQUESTED)| \
- (1U << EFX_PHY_CAP_25G_BASER_FEC_REQUESTED)
-
__checkReturn efx_rc_t
efx_phy_adv_cap_set(
__inefx_nic_t *enp,
@@ -210,8 +205,7 @@ efx_phy_adv_cap_set(
EFSYS_ASSERT3U(enp->en_magic, ==, EFX_NIC_MAGIC);
EFSYS_ASSERT3U(enp->en_mod_flags, &, EFX_MOD_PORT);
- /* Ignore don't care bits of FEC (FEC EFX_PHY_CAP_*_REQUESTED) */
- if ((mask & ~(epp->ep_phy_cap_mask | EFX_PHY_CAP_FEC_REQ_MASK)) != 0) {
+ if ((mask & ~epp->ep_phy_cap_mask) != 0) {
rc = ENOTSUP;
goto fail1;
}
--
2.17.1
[dpdk-dev] [PATCH 30/37] net/sfc/base: add support to get active FEC type
From: Vijay Srivastava
Signed-off-by: Vijay Srivastava
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/ef10_impl.h | 6
drivers/net/sfc/base/ef10_phy.c | 60 +---
drivers/net/sfc/base/efx.h | 11 ++
drivers/net/sfc/base/efx_impl.h | 1 +
drivers/net/sfc/base/efx_phy.c | 39 -
5 files changed, 111 insertions(+), 6 deletions(-)
diff --git a/drivers/net/sfc/base/ef10_impl.h b/drivers/net/sfc/base/ef10_impl.h
index 2819ae6ed..b72e7d256 100644
--- a/drivers/net/sfc/base/ef10_impl.h
+++ b/drivers/net/sfc/base/ef10_impl.h
@@ -596,6 +596,7 @@ typedef struct ef10_link_state_s {
uint32_tels_adv_cap_mask;
uint32_tels_lp_cap_mask;
unsigned intels_fcntl;
+ efx_phy_fec_type_t els_fec;
efx_link_mode_t els_link_mode;
#if EFSYS_OPT_LOOPBACK
efx_loopback_type_t els_loopback;
@@ -632,6 +633,11 @@ ef10_phy_oui_get(
__inefx_nic_t *enp,
__out uint32_t *ouip);
+extern __checkReturn efx_rc_t
+ef10_phy_fec_type_get(
+ __inefx_nic_t *enp,
+ __out efx_phy_fec_type_t *fecp);
+
#if EFSYS_OPT_PHY_STATS
extern __checkReturn efx_rc_t
diff --git a/drivers/net/sfc/base/ef10_phy.c b/drivers/net/sfc/base/ef10_phy.c
index a1f59ff1c..ec3600e96 100644
--- a/drivers/net/sfc/base/ef10_phy.c
+++ b/drivers/net/sfc/base/ef10_phy.c
@@ -98,8 +98,10 @@ mcdi_phy_decode_link_mode(
__inuint32_t link_flags,
__inunsigned int speed,
__inunsigned int fcntl,
+ __inuint32_t fec,
__out efx_link_mode_t *link_modep,
- __out unsigned int *fcntlp)
+ __out unsigned int *fcntlp,
+ __out efx_phy_fec_type_t *fecp)
{
boolean_t fd = !!(link_flags &
(1 << MC_CMD_GET_LINK_OUT_FULL_DUPLEX_LBN));
@@ -141,6 +143,22 @@ mcdi_phy_decode_link_mode(
EFSYS_PROBE1(mc_pcol_error, int, fcntl);
*fcntlp = 0;
}
+
+ switch (fec) {
+ case MC_CMD_FEC_NONE:
+ *fecp = EFX_PHY_FEC_NONE;
+ break;
+ case MC_CMD_FEC_BASER:
+ *fecp = EFX_PHY_FEC_BASER;
+ break;
+ case MC_CMD_FEC_RS:
+ *fecp = EFX_PHY_FEC_RS;
+ break;
+ default:
+ EFSYS_PROBE1(mc_pcol_error, int, fec);
+ *fecp = EFX_PHY_FEC_NONE;
+ break;
+ }
}
@@ -154,6 +172,7 @@ ef10_phy_link_ev(
unsigned int link_flags;
unsigned int speed;
unsigned int fcntl;
+ efx_phy_fec_type_t fec = MC_CMD_FEC_NONE;
efx_link_mode_t link_mode;
uint32_t lp_cap_mask;
@@ -191,7 +210,8 @@ ef10_phy_link_ev(
link_flags = MCDI_EV_FIELD(eqp, LINKCHANGE_LINK_FLAGS);
mcdi_phy_decode_link_mode(enp, link_flags, speed,
MCDI_EV_FIELD(eqp, LINKCHANGE_FCNTL),
- &link_mode, &fcntl);
+ MC_CMD_FEC_NONE, &link_mode,
+ &fcntl, &fec);
mcdi_phy_decode_cap(MCDI_EV_FIELD(eqp, LINKCHANGE_LP_CAP),
&lp_cap_mask);
@@ -242,15 +262,16 @@ ef10_phy_get_link(
__out ef10_link_state_t *elsp)
{
efx_mcdi_req_t req;
+ uint32_t fec;
EFX_MCDI_DECLARE_BUF(payload, MC_CMD_GET_LINK_IN_LEN,
- MC_CMD_GET_LINK_OUT_LEN);
+ MC_CMD_GET_LINK_OUT_V2_LEN);
efx_rc_t rc;
req.emr_cmd = MC_CMD_GET_LINK;
req.emr_in_buf = payload;
req.emr_in_length = MC_CMD_GET_LINK_IN_LEN;
req.emr_out_buf = payload;
- req.emr_out_length = MC_CMD_GET_LINK_OUT_LEN;
+ req.emr_out_length = MC_CMD_GET_LINK_OUT_V2_LEN;
efx_mcdi_execute(enp, &req);
@@ -269,10 +290,16 @@ ef10_phy_get_link(
mcdi_phy_decode_cap(MCDI_OUT_DWORD(req, GET_LINK_OUT_LP_CAP),
&elsp->els_lp_cap_mask);
+ if (req.emr_out_length_used < MC_CMD_GET_LINK_OUT_V2_LEN)
+ fec = MC_CMD_FEC_NONE;
+ else
+ fec = MCDI_OUT_DWORD(req, GET_LINK_OUT_V2_FEC_TYPE);
+
mcdi_phy_decode_link_mode(enp, MCDI_OUT_DWORD(req, GET_LINK_OUT_FLAGS),
MCDI_OUT_DWORD(req, GET_LINK_OUT_LINK_SPEED),
MCDI_OUT_DWORD(req, GET_LINK_OUT_FCNTL),
- &elsp->els_link_mode, &elsp->els_fcntl);
+ fec, &elsp->els_link_mode,
+ &elsp->els_fcntl, &elsp->els_fec);
#if EFSYS_OPT_LOOPBACK
/*
@@ -515,6 +542,29 @@ ef10_phy_oui_get(
return (ENOTSUP);
}
+ __checkReturn efx_rc_t
+ef10_phy_fec_type_get(
+ __inefx_nic_t *enp,
+ __out
[dpdk-dev] [PATCH 32/37] net/sfc/base: check buffer size for hash flags
From: Ivan Malov
The efx_rx_scale_hash_flags_get interface is unsafe, as it does not
have an argument for the size of the output buffer used to return
the flags. While the only caller currently supplies a sufficiently
large buffer, this should be checked at runtime to avoid writing
past the end of the buffer.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/efx.h| 3 +-
drivers/net/sfc/base/efx_annote.h | 1 +
drivers/net/sfc/base/efx_rx.c | 120 +++---
drivers/net/sfc/sfc_rx.c | 2 +-
4 files changed, 81 insertions(+), 45 deletions(-)
diff --git a/drivers/net/sfc/base/efx.h b/drivers/net/sfc/base/efx.h
index cc68f744e..de62b7d50 100644
--- a/drivers/net/sfc/base/efx.h
+++ b/drivers/net/sfc/base/efx.h
@@ -2368,7 +2368,8 @@ extern__checkReturn efx_rc_t
efx_rx_scale_hash_flags_get(
__inefx_nic_t *enp,
__inefx_rx_hash_alg_t hash_alg,
- __inout_ecount(EFX_RX_HASH_NFLAGS) unsigned int *flagsp,
+ __out_ecount_part(max_nflags, *nflagsp) unsigned int *flagsp,
+ __inunsigned int max_nflags,
__out unsigned int *nflagsp);
extern __checkReturn efx_rc_t
diff --git a/drivers/net/sfc/base/efx_annote.h
b/drivers/net/sfc/base/efx_annote.h
index 671aaed3b..607b43c77 100644
--- a/drivers/net/sfc/base/efx_annote.h
+++ b/drivers/net/sfc/base/efx_annote.h
@@ -33,6 +33,7 @@
#define__out_opt
#define__out_ecount(_n)
#define__out_ecount_opt(_n)
+#define__out_ecount_part(_n, _l)
#define__out_bcount(_n)
#define__out_bcount_opt(_n)
#define__out_bcount_part(_n, _l)
diff --git a/drivers/net/sfc/base/efx_rx.c b/drivers/net/sfc/base/efx_rx.c
index dfd3974da..bb0c144d7 100644
--- a/drivers/net/sfc/base/efx_rx.c
+++ b/drivers/net/sfc/base/efx_rx.c
@@ -298,13 +298,12 @@ efx_rx_scatter_enable(
efx_rx_scale_hash_flags_get(
__inefx_nic_t *enp,
__inefx_rx_hash_alg_t hash_alg,
- __inout_ecount(EFX_RX_HASH_NFLAGS) unsigned int *flagsp,
+ __out_ecount_part(max_nflags, *nflagsp) unsigned int *flagsp,
+ __inunsigned int max_nflags,
__out unsigned int *nflagsp)
{
efx_nic_cfg_t *encp = &enp->en_nic_cfg;
- boolean_t l4;
- boolean_t additional_modes;
- unsigned int *entryp = flagsp;
+ unsigned int nflags = 0;
efx_rc_t rc;
if (flagsp == NULL || nflagsp == NULL) {
@@ -313,56 +312,90 @@ efx_rx_scale_hash_flags_get(
}
if ((encp->enc_rx_scale_hash_alg_mask & (1U << hash_alg)) == 0) {
- *nflagsp = 0;
- return 0;
+ nflags = 0;
+ goto done;
}
- l4 = encp->enc_rx_scale_l4_hash_supported;
- additional_modes = encp->enc_rx_scale_additional_modes_supported;
-
-#defineLIST_FLAGS(_entryp, _class, _l4_hashing, _additional_modes)
\
- do {\
- if (_l4_hashing) { \
- *(_entryp++) = EFX_RX_HASH(_class, 4TUPLE); \
- \
- if (_additional_modes) {\
- *(_entryp++) = \
- EFX_RX_HASH(_class, 2TUPLE_DST);\
- *(_entryp++) = \
- EFX_RX_HASH(_class, 2TUPLE_SRC);\
- } \
- } \
- \
- *(_entryp++) = EFX_RX_HASH(_class, 2TUPLE); \
- \
- if (_additional_modes) {\
- *(_entryp++) = EFX_RX_HASH(_class, 1TUPLE_DST); \
- *(_entryp++) = EFX_RX_HASH(_class, 1TUPLE_SRC); \
- } \
- \
- *(_entryp++) = EFX_RX_HASH(_class, DISABLE);\
- \
- _NOTE(CONSTANTCONDITION)\
+ /* Helper to add flags word to flags array without buffer overflow */
+#defineINSERT_FLAGS(_flags)
[dpdk-dev] [PATCH 34/37] net/sfc/base: improve handling of legacy RSS hash flags
From: Ivan Malov
Client drivers may use either legacy flags, for example,
EFX_RX_HASH_TCPIPV4, or generalised flags, for example,
EFX_RX_HASH(IPV4_TCP, 4TUPLE), to configure RSS hash.
The libefx is able to recognise what scheme is used.
Legacy flags may be consumed directly by a chip-specific handler to
configure the NIC, that is, on EF10, these flags can be used to fill
in legacy RSS mode field in MCDI request. Generalised flags can also
be directly used in EF10-specific handler as they are fully compatible
with additional fields of the same MCDI request.
Legacy flags undergo conversion to generalised flags before they
are consumed by a chip-specific handler. This conversion is used to
make sure that chip-specific handlers expect only generalised flags
in the input for the sake of clarity of the code.
Depending on firmware capabilities, a chip-specififc handler either
supplies the input to the NIC directly, for example,
EFX_RX_HASH(IPV4_TCP, 4TUPLE) flag will enable 4 bits in
RSS_CONTEXT_SET_FLAGS_IN_TCP_IPV4_RSS_MODE field on EF10, or takes
the opportunity to translate the input to enable bits which don't map
to the generic flag, like setting
RSS_CONTEXT_SET_FLAGS_IN_TOEPLITZ_TCPV4_EN on EF10 when the firmware
claims no support for additional modes.
However, this approach has introduced a severe problem which can be
reproduced with ultra-low-latency firmware variant. In order to enable
IP hash, EF10-specific handler requires the user to request 2-tuple
hash for IP-other, TCP and UDP traffic classes, unconditionally.
In example, IPv4 hash can be enabled using the following input:
EFX_RX_HASH(IPV4_TCP, 2TUPLE) | EFX_RX_HASH(IPV4_UDP, 2TUPLE) |
EFX_RX_HASH(IPV4, 2TUPLE).
At the same time, on ultra-low-latency firmware, the common code will
never report support for any UDP tuple to the client driver. That is,
in the same example, the driver will use EFX_RX_HASH(IPV4_TCP, 2TUPLE) |
EFX_RX_HASH(IPV4, 2TUPLE). This input will not be recognised by
EF10-specific handler, and RSS_CONTEXT_SET_FLAGS_IN_TOEPLITZ_IPV4_EN
bit will not be set in the MCDI request.
In order to solve the problem, the patch removes conversion code
from chip-specific handlers and adds appropriate code to convert
EFX_RX_HASH() flags to their legacy counterparts to the common scale
mode set function. If the firmware does not support additional modes,
the function will convert generalised flags to legacy flags correctly
without any demand for UDP flags and pass the result to a chip-specific
handler.
Signed-off-by: Ivan Malov
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/ef10_rx.c | 42 ---
drivers/net/sfc/base/efx_rx.c | 94 ++
2 files changed, 62 insertions(+), 74 deletions(-)
diff --git a/drivers/net/sfc/base/ef10_rx.c b/drivers/net/sfc/base/ef10_rx.c
index 1444eca81..3c8f4f3b9 100644
--- a/drivers/net/sfc/base/ef10_rx.c
+++ b/drivers/net/sfc/base/ef10_rx.c
@@ -314,11 +314,6 @@ efx_mcdi_rss_context_set_flags(
__inefx_rx_hash_type_t type)
{
efx_nic_cfg_t *encp = &enp->en_nic_cfg;
- efx_rx_hash_type_t type_ipv4;
- efx_rx_hash_type_t type_ipv4_tcp;
- efx_rx_hash_type_t type_ipv6;
- efx_rx_hash_type_t type_ipv6_tcp;
- efx_rx_hash_type_t modes;
efx_mcdi_req_t req;
EFX_MCDI_DECLARE_BUF(payload, MC_CMD_RSS_CONTEXT_SET_FLAGS_IN_LEN,
MC_CMD_RSS_CONTEXT_SET_FLAGS_OUT_LEN);
@@ -355,53 +350,38 @@ efx_mcdi_rss_context_set_flags(
MCDI_IN_SET_DWORD(req, RSS_CONTEXT_SET_FLAGS_IN_RSS_CONTEXT_ID,
rss_context);
- type_ipv4 = EFX_RX_HASH(IPV4, 2TUPLE) | EFX_RX_HASH(IPV4_TCP, 2TUPLE) |
- EFX_RX_HASH(IPV4_UDP, 2TUPLE);
- type_ipv4_tcp = EFX_RX_HASH(IPV4_TCP, 4TUPLE);
- type_ipv6 = EFX_RX_HASH(IPV6, 2TUPLE) | EFX_RX_HASH(IPV6_TCP, 2TUPLE) |
- EFX_RX_HASH(IPV6_UDP, 2TUPLE);
- type_ipv6_tcp = EFX_RX_HASH(IPV6_TCP, 4TUPLE);
-
- /*
-* Create a copy of the original hash type.
-* The copy will be used to fill in RSS_MODE bits and
-* may be cleared beforehand. The original variable
-* and, thus, EN bits will remain unaffected.
-*/
- modes = type;
-
/*
* If the firmware lacks support for additional modes, RSS_MODE
* fields must contain zeros, otherwise the operation will fail.
*/
if (encp->enc_rx_scale_additional_modes_supported == B_FALSE)
- modes = 0;
+ type &= EFX_RX_HASH_LEGACY_MASK;
MCDI_IN_POPULATE_DWORD_10(req, RSS_CONTEXT_SET_FLAGS_IN_FLAGS,
RSS_CONTEXT_SET_FLAGS_IN_TOEPLITZ_IPV4_EN,
- ((type & type_ipv4) == type_ipv4) ? 1 : 0,
+ (type & EFX_RX_HASH_IPV4) ? 1 : 0,
RSS_CONTEXT_SET_FLAGS_IN_TOEPLITZ_TCPV4_EN,
- ((type & type_ipv4_tcp) == type_ipv4_tcp) ? 1 : 0,
+ (type & EFX_RX_HASH_TCPIPV4) ? 1 : 0,
RSS_CONTEXT_SET_FLAGS_IN_
[dpdk-dev] [PATCH 36/37] net/sfc/base: fix MAC Tx stats for less or equal to 64 bytes
From: Andy Moreton
This statistic should include 64byte and smaller frames.
Fix EF10 calculation to match Siena code.
Fixes: 8c7c723dfe7c ("net/sfc/base: import MAC statistics")
Cc: sta...@dpdk.org
Signed-off-by: Andy Moreton
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/ef10_mac.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/sfc/base/ef10_mac.c b/drivers/net/sfc/base/ef10_mac.c
index a4a6d9ec8..ab73828f1 100644
--- a/drivers/net/sfc/base/ef10_mac.c
+++ b/drivers/net/sfc/base/ef10_mac.c
@@ -650,7 +650,7 @@ ef10_mac_stats_update(
EF10_MAC_STAT_READ(esmp, MC_CMD_MAC_TX_LT64_PKTS, &value);
EFSYS_STAT_SET_QWORD(&(stat[EFX_MAC_TX_LE_64_PKTS]), &value);
EF10_MAC_STAT_READ(esmp, MC_CMD_MAC_TX_64_PKTS, &value);
- EFSYS_STAT_SET_QWORD(&(stat[EFX_MAC_TX_LE_64_PKTS]), &value);
+ EFSYS_STAT_INCR_QWORD(&(stat[EFX_MAC_TX_LE_64_PKTS]), &value);
EF10_MAC_STAT_READ(esmp, MC_CMD_MAC_TX_65_TO_127_PKTS, &value);
EFSYS_STAT_SET_QWORD(&(stat[EFX_MAC_TX_65_TO_127_PKTS]), &value);
--
2.17.1
[dpdk-dev] [PATCH 37/37] net/sfc/base: add helper API to make Geneve filter spec
From: Vijay Srivastava
Signed-off-by: Vijay Srivastava
Signed-off-by: Andrew Rybchenko
---
drivers/net/sfc/base/efx.h| 18 ++-
drivers/net/sfc/base/efx_filter.c | 90 +++
2 files changed, 95 insertions(+), 13 deletions(-)
diff --git a/drivers/net/sfc/base/efx.h b/drivers/net/sfc/base/efx.h
index 62c56e6e1..fd68d69c7 100644
--- a/drivers/net/sfc/base/efx.h
+++ b/drivers/net/sfc/base/efx.h
@@ -2959,9 +2959,23 @@ efx_filter_spec_set_encap_type(
__inefx_filter_inner_frame_match_t inner_frame_match);
extern __checkReturn efx_rc_t
-efx_filter_spec_set_vxlan_full(
+efx_filter_spec_set_vxlan(
__inout efx_filter_spec_t *spec,
- __inconst uint8_t *vxlan_id,
+ __inconst uint8_t *vni,
+ __inconst uint8_t *inner_addr,
+ __inconst uint8_t *outer_addr);
+
+extern __checkReturn efx_rc_t
+efx_filter_spec_set_geneve(
+ __inout efx_filter_spec_t *spec,
+ __inconst uint8_t *vni,
+ __inconst uint8_t *inner_addr,
+ __inconst uint8_t *outer_addr);
+
+extern __checkReturn efx_rc_t
+efx_filter_spec_set_nvgre(
+ __inout efx_filter_spec_t *spec,
+ __inconst uint8_t *vsid,
__inconst uint8_t *inner_addr,
__inconst uint8_t *outer_addr);
diff --git a/drivers/net/sfc/base/efx_filter.c
b/drivers/net/sfc/base/efx_filter.c
index 412298acf..a7523b38b 100644
--- a/drivers/net/sfc/base/efx_filter.c
+++ b/drivers/net/sfc/base/efx_filter.c
@@ -490,27 +490,42 @@ efx_filter_spec_set_encap_type(
}
/*
- * Specify inner and outer Ethernet address and VXLAN ID in filter
+ * Specify inner and outer Ethernet address and VNI or VSID in tunnel filter
* specification.
*/
- __checkReturn efx_rc_t
-efx_filter_spec_set_vxlan_full(
- __inout efx_filter_spec_t *spec,
- __inconst uint8_t *vxlan_id,
+static __checkReturn efx_rc_t
+efx_filter_spec_set_tunnel(
+ __inout efx_filter_spec_t *spec,
+ __inefx_tunnel_protocol_t encap_type,
+ __inconst uint8_t *vni_or_vsid,
__inconst uint8_t *inner_addr,
__inconst uint8_t *outer_addr)
{
+ efx_rc_t rc;
+
EFSYS_ASSERT3P(spec, !=, NULL);
- EFSYS_ASSERT3P(vxlan_id, !=, NULL);
+ EFSYS_ASSERT3P(vni_or_vsid, !=, NULL);
EFSYS_ASSERT3P(inner_addr, !=, NULL);
EFSYS_ASSERT3P(outer_addr, !=, NULL);
- if ((inner_addr == NULL) && (outer_addr == NULL))
- return (EINVAL);
+ switch (encap_type) {
+ case EFX_TUNNEL_PROTOCOL_VXLAN:
+ case EFX_TUNNEL_PROTOCOL_GENEVE:
+ case EFX_TUNNEL_PROTOCOL_NVGRE:
+ break;
+ default:
+ rc = EINVAL;
+ goto fail1;
+ }
+
+ if ((inner_addr == NULL) && (outer_addr == NULL)) {
+ rc = EINVAL;
+ goto fail2;
+ }
- if (vxlan_id != NULL) {
+ if (vni_or_vsid != NULL) {
spec->efs_match_flags |= EFX_FILTER_MATCH_VNI_OR_VSID;
- memcpy(spec->efs_vni_or_vsid, vxlan_id, EFX_VNI_OR_VSID_LEN);
+ memcpy(spec->efs_vni_or_vsid, vni_or_vsid, EFX_VNI_OR_VSID_LEN);
}
if (outer_addr != NULL) {
spec->efs_match_flags |= EFX_FILTER_MATCH_LOC_MAC;
@@ -520,10 +535,63 @@ efx_filter_spec_set_vxlan_full(
spec->efs_match_flags |= EFX_FILTER_MATCH_IFRM_LOC_MAC;
memcpy(spec->efs_ifrm_loc_mac, inner_addr, EFX_MAC_ADDR_LEN);
}
+
spec->efs_match_flags |= EFX_FILTER_MATCH_ENCAP_TYPE;
- spec->efs_encap_type = EFX_TUNNEL_PROTOCOL_VXLAN;
+ spec->efs_encap_type = encap_type;
return (0);
+
+fail2:
+ EFSYS_PROBE(fail2);
+fail1:
+ EFSYS_PROBE1(fail1, efx_rc_t, rc);
+
+ return (rc);
+}
+
+/*
+ * Specify inner and outer Ethernet address and VNI in VXLAN filter
+ * specification.
+ */
+__checkReturn efx_rc_t
+efx_filter_spec_set_vxlan(
+ __inout efx_filter_spec_t *spec,
+ __inconst uint8_t *vni,
+ __inconst uint8_t *inner_addr,
+ __inconst uint8_t *outer_addr)
+{
+ return efx_filter_spec_set_tunnel(spec, EFX_TUNNEL_PROTOCOL_VXLAN,
+ vni, inner_addr, outer_addr);
+}
+
+/*
+ * Specify inner and outer Ethernet address and VNI in Geneve filter
+ * specification.
+ */
+__checkReturn efx_rc_t
+efx_filter_spec_set_geneve(
+ __inout efx_filter_spec_t *spec,
+ __inconst uint8_t *vni,
+ __inconst uint8_t *inner_addr,
+ __inconst uint8_t *outer_addr)
+{
+ return efx_filter_spec_set_tunnel(spec, EFX_TUNNEL_PROTOCOL_GENEVE,
+ vni, inner_addr, outer_addr);
+}
+
+/*
+ * Specify inner and outer Ethernet ad
Re: [dpdk-dev] [PATCH v5 07/11] net/virtio: implement transmit path for packed queues
One more comment:
> -Original Message-
> From: dev On Behalf Of Jens Freimann
> Sent: Friday, September 7, 2018 2:20 AM
> To: dev@dpdk.org
> Cc: tiwei@intel.com; maxime.coque...@redhat.com
> Subject: [dpdk-dev] [PATCH v5 07/11] net/virtio: implement transmit path
> for packed queues
>
> This implements the transmit path for devices with support for packed
> virtqueues.
>
> Add the feature bit and enable code to
> add buffers to vring and mark descriptors as available.
>
> Signed-off-by: Jens Freiman
> ---
> drivers/net/virtio/virtio_ethdev.c | 8 +-
> drivers/net/virtio/virtio_ethdev.h | 2 +
> drivers/net/virtio/virtio_rxtx.c | 113 -
> 3 files changed, 121 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/net/virtio/virtio_ethdev.c
> b/drivers/net/virtio/virtio_ethdev.c
> index ad91f7f82..d2c5755bb 100644
> --- a/drivers/net/virtio/virtio_ethdev.c
> +++ b/drivers/net/virtio/virtio_ethdev.c
> @@ -384,6 +384,8 @@ virtio_init_queue(struct rte_eth_dev *dev, uint16_t
> vtpci_queue_idx)
> vq->hw = hw;
> vq->vq_queue_index = vtpci_queue_idx;
> vq->vq_nentries = vq_size;
> +if (vtpci_packed_queue(hw))
> +vq->vq_ring.avail_wrap_counter = 1;
>
> /*
> * Reserve a memzone for vring elements @@ -1338,7 +1340,11 @@
> set_rxtx_funcs(struct rte_eth_dev *eth_dev)
> eth_dev->rx_pkt_burst = &virtio_recv_pkts;
> }
>
> -if (hw->use_inorder_tx) {
> +if (vtpci_packed_queue(hw)) {
> +PMD_INIT_LOG(INFO, "virtio: using virtio 1.1 Tx path on
> port %u",
> +eth_dev->data->port_id);
> +eth_dev->tx_pkt_burst = virtio_xmit_pkts_packed;
> +} else if (hw->use_inorder_tx) {
> PMD_INIT_LOG(INFO, "virtio: using inorder Tx path on
> port %u",
> eth_dev->data->port_id);
> eth_dev->tx_pkt_burst = virtio_xmit_pkts_inorder; diff --git
> a/drivers/net/virtio/virtio_ethdev.h b/drivers/net/virtio/virtio_ethdev.h
> index b726ad108..04161b461 100644
> --- a/drivers/net/virtio/virtio_ethdev.h
> +++ b/drivers/net/virtio/virtio_ethdev.h
> @@ -79,6 +79,8 @@ uint16_t virtio_recv_mergeable_pkts_inorder(void
> *rx_queue,
>
> uint16_t virtio_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts,
> uint16_t nb_pkts);
> +uint16_t virtio_xmit_pkts_packed(void *tx_queue, struct rte_mbuf
> **tx_pkts,
> +uint16_t nb_pkts);
>
> uint16_t virtio_xmit_pkts_inorder(void *tx_queue, struct rte_mbuf
> **tx_pkts,
> uint16_t nb_pkts);
> diff --git a/drivers/net/virtio/virtio_rxtx.c
> b/drivers/net/virtio/virtio_rxtx.c
> index eb891433e..12787070e 100644
> --- a/drivers/net/virtio/virtio_rxtx.c
> +++ b/drivers/net/virtio/virtio_rxtx.c
> @@ -38,6 +38,112 @@
> #define VIRTIO_DUMP_PACKET(m, len) do { } while (0) #endif
>
> +
> +/* Cleanup from completed transmits. */ static void
> +virtio_xmit_cleanup_packed(struct virtqueue *vq) {
> +uint16_t idx;
> +uint16_t size = vq->vq_nentries;
> +struct vring_desc_packed *desc = vq->vq_ring.desc_packed;
> +struct vq_desc_extra *dxp;
> +
> +idx = vq->vq_used_cons_idx;
> +while (desc_is_used(&desc[idx], &vq->vq_ring) &&
> + vq->vq_free_cnt < size) {
> +dxp = &vq->vq_descx[idx];
> +vq->vq_free_cnt += dxp->ndescs;
> +idx = dxp->ndescs;
Should be "+=" here?
> +idx = idx >= size ? idx - size : idx;
> +}
> +}
> +
> +uint16_t
> +virtio_xmit_pkts_packed(void *tx_queue, struct rte_mbuf **tx_pkts,
> + uint16_t nb_pkts)
> +{
> +struct virtnet_tx *txvq = tx_queue;
> +struct virtqueue *vq = txvq->vq;
> +uint16_t i;
> +struct vring_desc_packed *desc = vq->vq_ring.desc_packed;
> +uint16_t idx, prev;
> +struct vq_desc_extra *dxp;
> +
> +if (unlikely(nb_pkts < 1))
> +return nb_pkts;
> +
> +PMD_TX_LOG(DEBUG, "%d packets to xmit", nb_pkts);
> +
> +if (likely(vq->vq_free_cnt < vq->vq_free_thresh))
> +virtio_xmit_cleanup_packed(vq);
> +
> +for (i = 0; i < nb_pkts; i++) {
> +struct rte_mbuf *txm = tx_pkts[i];
> +struct virtio_tx_region *txr = txvq->virtio_net_hdr_mz->addr;
> +uint16_t head_idx;
> +int wrap_counter;
> +int descs_used;
> +
> +if (unlikely(txm->nb_segs + 1 > vq->vq_free_cnt)) {
> +virtio_xmit_cleanup_packed(vq);
> +
> +if (unlikely(txm->nb_segs + 1 > vq->vq_free_cnt)) {
> +PMD_TX_LOG(ERR,
> + "No free tx descriptors to transmit");
> +break;
> +}
> +}
> +
> +txvq->stats.bytes += txm->pkt_len;
> +
> +vq->vq_free_cnt -= txm->nb_segs + 1;
> +
> +wrap_counter = vq->vq_ring.avail_wrap_counter;
> +idx = vq->vq_avail_idx;
> +head_idx = idx;
> +
> +dxp = &vq->vq_descx[idx];
> +if (dxp->cookie != NULL)
> +rte_pktmbuf_free(dxp->cookie);
> +dxp->cookie = txm;
> +
> +desc[idx].addr = txvq->virtio_net_hdr_mem +
> + RTE_PTR_DIFF(&txr[idx].tx_hdr, txr);
> +desc[idx].len = vq->hw->vtnet_hdr_size;
> +desc[idx].flags = VRING_DESC_F_NEXT |
> +VRING_DESC_F_AVAIL(vq-
> >vq_ring.avail_wrap_counter) |
> +VRING_DESC_F_USED(!vq-
> >vq_ring.avail_wrap_counter);
> +descs_used = 1;
> +
> +do {
> +idx = update_pq_avail_index(vq);
> +desc[idx].addr =
> VIRTIO_MBUF_DATA_DMA_ADDR(txm, vq);
> +desc[idx].len = txm->data_len;
> +desc[idx].flags = VRING_DESC_F_NEXT |
> +VRING_DESC_F_AVAIL(v
[dpdk-dev] [PATCH v4] net/i40e: add interface to choose latest vector path
Right now, vector path is limited to only use on later platform.
This patch adds a devarg use-latest-vec to allow the users to
use the latest vector path that the platform supported. Namely,
using AVX2 vector path on broadwell is possible.
Signed-off-by: Xiaoyun Li
---
v4:
* Polish the codes.
v3:
* Polish the doc and commit log.
v2:
* Correct the calling of the wrong function last time.
* Fix seg fault bug.
---
doc/guides/nics/i40e.rst | 8 ++
doc/guides/rel_notes/release_18_11.rst | 4 +
drivers/net/i40e/i40e_ethdev.c | 46 ++-
drivers/net/i40e/i40e_ethdev.h | 3 +
drivers/net/i40e/i40e_rxtx.c | 103 -
5 files changed, 128 insertions(+), 36 deletions(-)
diff --git a/doc/guides/nics/i40e.rst b/doc/guides/nics/i40e.rst
index 65d87f869..643e6a062 100644
--- a/doc/guides/nics/i40e.rst
+++ b/doc/guides/nics/i40e.rst
@@ -163,6 +163,14 @@ Runtime Config Options
Currently hot-plugging of representor ports is not supported so all required
representors must be specified on the creation of the PF.
+- ``Use latest vector`` (default ``disable``)
+
+ Vector path was limited to use only on later platform. But users may want the
+ latest vector path. For example, VPP users may want to use AVX2 vector path
on HSW/BDW
+ because it can get better perf. So ``devargs`` parameter ``use-latest-vec``
is
+ introduced, for example::
+-w 84:00.0,use-latest-vec=1
+
Driver compilation and testing
--
diff --git a/doc/guides/rel_notes/release_18_11.rst
b/doc/guides/rel_notes/release_18_11.rst
index 3ae6b3f58..34af591a2 100644
--- a/doc/guides/rel_notes/release_18_11.rst
+++ b/doc/guides/rel_notes/release_18_11.rst
@@ -54,6 +54,10 @@ New Features
Also, make sure to start the actual text at the margin.
=
+* **Added a devarg to use the latest vector path.**
+ A new devarg ``use-latest-vec`` was introduced to allow users to choose
+ the latest vector path that the platform supported. For example, VPP users
+ can use AVX2 vector path on BDW/HSW to get better performance.
API Changes
---
diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c
index 85a6a867f..72377d0b6 100644
--- a/drivers/net/i40e/i40e_ethdev.c
+++ b/drivers/net/i40e/i40e_ethdev.c
@@ -44,6 +44,7 @@
#define ETH_I40E_FLOATING_VEB_LIST_ARG "floating_veb_list"
#define ETH_I40E_SUPPORT_MULTI_DRIVER "support-multi-driver"
#define ETH_I40E_QUEUE_NUM_PER_VF_ARG "queue-num-per-vf"
+#define ETH_I40E_USE_LATEST_VEC"use-latest-vec"
#define I40E_CLEAR_PXE_WAIT_MS 200
@@ -408,6 +409,7 @@ static const char *const valid_keys[] = {
ETH_I40E_FLOATING_VEB_LIST_ARG,
ETH_I40E_SUPPORT_MULTI_DRIVER,
ETH_I40E_QUEUE_NUM_PER_VF_ARG,
+ ETH_I40E_USE_LATEST_VEC,
NULL};
static const struct rte_pci_id pci_id_i40e_map[] = {
@@ -1201,6 +1203,46 @@ i40e_aq_debug_write_global_register(struct i40e_hw *hw,
return i40e_aq_debug_write_register(hw, reg_addr, reg_val, cmd_details);
}
+static int
+i40e_parse_latest_vec(struct rte_eth_dev *dev)
+{
+ struct i40e_adapter *ad =
+ I40E_DEV_PRIVATE_TO_ADAPTER(dev->data->dev_private);
+ int kvargs_count, use_latest_vec;
+ struct rte_kvargs *kvlist;
+
+ ad->use_latest_vec = false;
+
+ if (!dev->device->devargs)
+ return 0;
+
+ kvlist = rte_kvargs_parse(dev->device->devargs->args, valid_keys);
+ if (!kvlist)
+ return -EINVAL;
+
+ kvargs_count = rte_kvargs_count(kvlist, ETH_I40E_USE_LATEST_VEC);
+ if (!kvargs_count) {
+ rte_kvargs_free(kvlist);
+ return 0;
+ }
+
+ if (kvargs_count > 1)
+ PMD_DRV_LOG(WARNING, "More than one argument \"%s\" and only "
+ "the first one is used !",
+ ETH_I40E_USE_LATEST_VEC);
+
+ use_latest_vec = atoi((&kvlist->pairs[0])->value);
+
+ rte_kvargs_free(kvlist);
+
+ if (use_latest_vec != 0 && use_latest_vec != 1)
+ PMD_DRV_LOG(WARNING, "Value should be 0 or 1, set it as 1!");
+
+ ad->use_latest_vec = (bool)use_latest_vec;
+
+ return 0;
+}
+
static int
eth_i40e_dev_init(struct rte_eth_dev *dev, void *init_params __rte_unused)
{
@@ -1263,6 +1305,7 @@ eth_i40e_dev_init(struct rte_eth_dev *dev, void
*init_params __rte_unused)
/* Check if need to support multi-driver */
i40e_support_multi_driver(dev);
+ i40e_parse_latest_vec(dev);
/* Make sure all is clean before doing PF reset */
i40e_clear_hw(hw);
@@ -12527,4 +12570,5 @@ RTE_PMD_REGISTER_PARAM_STRING(net_i40e,
ETH_I40E_FLOATING_VEB_ARG "=1"
ETH_I40E_FLOATING_VEB_LIST_ARG "="
ETH_I40E_QUEUE_NUM_PER_VF_ARG "=1|2|4|8|1
Re: [dpdk-dev] DPDK Shared lib: No Ethernet ports
On Fri, Sep 07, 2018 at 05:11:21PM +, Shubhachint, Chaitanya wrote: > Thank you for your reply Bruce. > Still having issues with shared-lib DPDK. I tried both options and get two > distinct issues. > I tried running the l2fwd example with e1000 driver and get little farther. > This time it does find the devices but there is an error with MBUF > > root:build$./l2fwd -d > /home/sp2/dpdk-18.08/x86_64-native-linuxapp-gcc/lib/librte_pmd_e1000.so > EAL: Detected 6 lcore(s) > EAL: Detected 1 NUMA nodes > EAL: Multi-process socket /var/run/dpdk/rte/mp_socket > EAL: No free hugepages reported in hugepages-1048576kB > EAL: Probing VFIO support... > EAL: PCI device :00:19.0 on NUMA socket -1 > EAL: Invalid NUMA socket, default to 0 > EAL: probe driver: 8086:1502 net_e1000_em > EAL: PCI device :06:00.0 on NUMA socket -1 > EAL: Invalid NUMA socket, default to 0 > EAL: probe driver: 8086:1521 net_e1000_igb > EAL: PCI device :06:00.1 on NUMA socket -1 > EAL: Invalid NUMA socket, default to 0 > EAL: probe driver: 8086:1521 net_e1000_igb > EAL: PCI device :06:00.2 on NUMA socket -1 > EAL: Invalid NUMA socket, default to 0 > EAL: probe driver: 8086:1521 net_e1000_igb > EAL: PCI device :06:00.3 on NUMA socket -1 > EAL: Invalid NUMA socket, default to 0 > EAL: probe driver: 8086:1521 net_e1000_igb > EAL: PCI device :08:00.0 on NUMA socket -1 > EAL: Invalid NUMA socket, default to 0 > EAL: probe driver: 8086:10d3 net_e1000_em > MAC updating enabled > MBUF: error setting mempool handler > EAL: Error - exiting with code: 1 > Cause: Cannot init mbuf pool > > 2nd Option of setting CONFIG_RTE_EAL_PMD_PATH option is ideal, but I run into > issue with as well. I am configuring and installing the DPDK with > CONFIG_RTE_EAL_PMD_PATH set as > CONFIG_RTE_EAL_PMD_PATH=$(HOME)/dpdk-18.08/x86_64-native-linuxapp-gcc/lib > > I this case I get following error when I run l2fwd > > root:build$./l2fwd > EAL: Detected 6 lcore(s) > EAL: Detected 1 NUMA nodes > EAL: /home/sp2/dpdk-18.08/x86_64-native-linuxapp-gcc/lib/libdpdk.so: invalid > ELF header > EAL: FATAL: Cannot init plugins > > EAL: Cannot init plugins > > EAL: Error - exiting with code: 1 > Cause: Invalid EAL arguments > To use CONFIG_RTE_EAL_PMD_PATH, you really need to have the drivers all in a separate directory, without any additional libs present. Other non-PMD libraries will cause the error above when they are loaded. /Bruce
Re: [dpdk-dev] [PATCH] build: add PPC64 Meson build
On Fri, Sep 07, 2018 at 07:35:02PM +0100, Luca Boccassi wrote:
> This has been only build-tested for now, on a native ppc64el POWER8E
> machine running Debian sid.
>
> Signed-off-by: Luca Boccassi
> ---
> The build box cannot be used to run DPDK as it doesn't have supported
> NICs and root access. Would be great if someone could run-test it, but
> at this point I think build support is enough to get started.
>
> config/meson.build | 8
> config/ppc_64/meson.build| 15 +++
> lib/librte_eal/common/arch/ppc_64/meson.build| 5 +
> .../common/include/arch/ppc_64/meson.build | 16
> 4 files changed, 44 insertions(+)
> create mode 100644 config/ppc_64/meson.build
> create mode 100644 lib/librte_eal/common/arch/ppc_64/meson.build
> create mode 100644 lib/librte_eal/common/include/arch/ppc_64/meson.build
>
> diff --git a/config/meson.build b/config/meson.build
> index 4d755323f4..8e87b344c2 100644
> --- a/config/meson.build
> +++ b/config/meson.build
> @@ -9,7 +9,13 @@ else
> endif
> dpdk_conf.set('RTE_MACHINE', machine)
> machine_args = []
> +# ppc64 does not support -march=native
> +if host_machine.cpu_family().startswith('ppc') and machine == 'native'
> +machine_args += '-mcpu=' + machine
> +machine_args += '-mtune=' + machine
> +else
> machine_args += '-march=' + machine
> +endif
Indentation?
>
> toolchain = cc.get_id()
> dpdk_conf.set_quoted('RTE_TOOLCHAIN', toolchain)
> @@ -84,6 +90,8 @@ if host_machine.cpu_family().startswith('x86')
> arch_subdir = 'x86'
> elif host_machine.cpu_family().startswith('arm') or
> host_machine.cpu_family().startswith('aarch')
> arch_subdir = 'arm'
> +elif host_machine.cpu_family().startswith('ppc')
> + arch_subdir = 'ppc_64'
> endif
> subdir(arch_subdir)
> dpdk_conf.set('RTE_COMPILE_TIME_CPUFLAGS', ','.join(compile_time_cpuflags))
> diff --git a/config/ppc_64/meson.build b/config/ppc_64/meson.build
> new file mode 100644
> index 00..d6faa7d64f
> --- /dev/null
> +++ b/config/ppc_64/meson.build
> @@ -0,0 +1,15 @@
> +# SPDX-License-Identifier: BSD-3-Clause
> +# Copyright(c) 2018 Luca Boccassi
> +
> +# for checking defines we need to use the correct compiler flags
> +march_opt = '-march=@0@'.format(machine)
This contradicts the statement above in config/meson.build where you state
that ppc64 doesn't support "-march=native"?
> +
> +dpdk_conf.set('RTE_ARCH', 'ppc_64')
> +dpdk_conf.set('RTE_ARCH_PPC_64', 1)
> +dpdk_conf.set('RTE_ARCH_64', 1)
> +
> +# overrides specific to ppc64
> +dpdk_conf.set('RTE_MAX_LCORE', 256)
> +dpdk_conf.set('RTE_MAX_NUMA_NODES', 32)
> +dpdk_conf.set('RTE_CACHE_LINE_SIZE', 128)
> +dpdk_conf.set('RTE_MAX_LCORE', 256)
/Bruce
Re: [dpdk-dev] [PATCH v5 08/11] net/virtio: implement receive path for packed queues
> -Original Message-
> From: dev On Behalf Of Jens Freimann
> Sent: Friday, September 7, 2018 2:20 AM
> To: dev@dpdk.org
> Cc: tiwei@intel.com; maxime.coque...@redhat.com
> Subject: [dpdk-dev] [PATCH v5 08/11] net/virtio: implement receive path for
> packed queues
>
> Implement the receive part.
>
> Signed-off-by: Jens Freimann
> ---
> drivers/net/virtio/virtio_ethdev.c | 15 +++-
> drivers/net/virtio/virtio_ethdev.h | 2 +
> drivers/net/virtio/virtio_rxtx.c | 131 +
> 3 files changed, 145 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/net/virtio/virtio_ethdev.c
> b/drivers/net/virtio/virtio_ethdev.c
> index d2c5755bb..a2bb726ba 100644
> --- a/drivers/net/virtio/virtio_ethdev.c
> +++ b/drivers/net/virtio/virtio_ethdev.c
> @@ -384,8 +384,10 @@ virtio_init_queue(struct rte_eth_dev *dev, uint16_t
> vtpci_queue_idx)
> vq->hw = hw;
> vq->vq_queue_index = vtpci_queue_idx;
> vq->vq_nentries = vq_size;
> -if (vtpci_packed_queue(hw))
> +if (vtpci_packed_queue(hw)) {
> vq->vq_ring.avail_wrap_counter = 1;
> +vq->vq_ring.used_wrap_counter = 1;
> +}
>
> /*
> * Reserve a memzone for vring elements @@ -1320,7 +1322,13 @@
> set_rxtx_funcs(struct rte_eth_dev *eth_dev) {
> struct virtio_hw *hw = eth_dev->data->dev_private;
>
> -if (hw->use_simple_rx) {
> +/*
> + * workarount for packed vqs which don't support
> + * mrg_rxbuf at this point
> + */
> +if (vtpci_packed_queue(hw)) {
> +eth_dev->rx_pkt_burst = &virtio_recv_pkts_packed;
> +} else if (hw->use_simple_rx) {
> PMD_INIT_LOG(INFO, "virtio: using simple Rx path on
> port %u",
> eth_dev->data->port_id);
> eth_dev->rx_pkt_burst = virtio_recv_pkts_vec; @@ -1484,7
> +1492,8 @@ virtio_init_device(struct rte_eth_dev *eth_dev, uint64_t
> req_features)
>
> /* Setting up rx_header size for the device */
> if (vtpci_with_feature(hw, VIRTIO_NET_F_MRG_RXBUF) ||
> -vtpci_with_feature(hw, VIRTIO_F_VERSION_1))
> +vtpci_with_feature(hw, VIRTIO_F_VERSION_1) ||
> +vtpci_with_feature(hw, VIRTIO_F_RING_PACKED))
> hw->vtnet_hdr_size = sizeof(struct
> virtio_net_hdr_mrg_rxbuf);
> else
> hw->vtnet_hdr_size = sizeof(struct virtio_net_hdr); diff --git
> a/drivers/net/virtio/virtio_ethdev.h b/drivers/net/virtio/virtio_ethdev.h
> index 04161b461..25eaff224 100644
> --- a/drivers/net/virtio/virtio_ethdev.h
> +++ b/drivers/net/virtio/virtio_ethdev.h
> @@ -70,6 +70,8 @@ int virtio_dev_tx_queue_setup_finish(struct
> rte_eth_dev *dev,
>
> uint16_t virtio_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
> uint16_t nb_pkts);
> +uint16_t virtio_recv_pkts_packed(void *rx_queue, struct rte_mbuf
> **rx_pkts,
> +uint16_t nb_pkts);
>
> uint16_t virtio_recv_mergeable_pkts(void *rx_queue, struct rte_mbuf
> **rx_pkts,
> uint16_t nb_pkts);
> diff --git a/drivers/net/virtio/virtio_rxtx.c
> b/drivers/net/virtio/virtio_rxtx.c
> index 12787070e..3f5fa7366 100644
> --- a/drivers/net/virtio/virtio_rxtx.c
> +++ b/drivers/net/virtio/virtio_rxtx.c
> @@ -31,6 +31,7 @@
> #include "virtqueue.h"
> #include "virtio_rxtx.h"
> #include "virtio_rxtx_simple.h"
> +#include "virtio_ring.h"
>
> #ifdef RTE_LIBRTE_VIRTIO_DEBUG_DUMP
> #define VIRTIO_DUMP_PACKET(m, len) rte_pktmbuf_dump(stdout, m, len)
> @@ -710,6 +711,34 @@ virtio_dev_rx_queue_setup_finish(struct
> rte_eth_dev *dev, uint16_t queue_idx)
>
> PMD_INIT_FUNC_TRACE();
>
> +if (vtpci_packed_queue(hw)) {
> +struct vring_desc_packed *desc;
> +struct vq_desc_extra *dxp;
> +
> +for (desc_idx = 0; desc_idx < vq->vq_nentries;
> +desc_idx++) {
> +m = rte_mbuf_raw_alloc(rxvq->mpool);
> +if (unlikely(m == NULL))
> +return -ENOMEM;
> +
> +dxp = &vq->vq_descx[desc_idx];
> +dxp->cookie = m;
> +dxp->ndescs = 1;
> +
> +desc = &vq->vq_ring.desc_packed[desc_idx];
> +desc->addr = VIRTIO_MBUF_ADDR(m, vq) +
> +RTE_PKTMBUF_HEADROOM - hw-
> >vtnet_hdr_size;
> +desc->len = m->buf_len -
> RTE_PKTMBUF_HEADROOM +
> +hw->vtnet_hdr_size;
> +desc->flags |= VRING_DESC_F_WRITE;
> +rte_smp_wmb();
> +set_desc_avail(&vq->vq_ring, desc);
> +}
> +vq->vq_ring.avail_wrap_counter ^= 1;
> +nbufs = desc_idx;
> +goto out;
> +}
> +
> /* Allocate blank mbufs for the each rx descriptor */
> nbufs = 0;
>
> @@ -773,6 +802,7 @@ virtio_dev_rx_queue_setup_finish(struct
> rte_eth_dev *dev, uint16_t queue_idx)
> vq_update_avail_idx(vq);
> }
>
> +out:
> PMD_INIT_LOG(DEBUG, "Allocated %d bufs", nbufs);
>
> VIRTQUEUE_DUMP(vq);
> @@ -993,6 +1023,107 @@ virtio_rx_offload(struct rte_mbuf *m, struct
> virtio_net_hdr *hdr)
> return 0;
> }
>
> +uint16_t
> +virtio_recv_pkts_packed(void *rx_queue, struct rte_mbuf **rx_pkts,
> + uint16_t nb_pkts)
> +{
> +struct virtnet_rx *rxvq = rx_queue;
> +struct virtqueue *vq = rxvq->vq;
> +struct virtio_hw *hw = vq->hw;
> +struct rte_mbuf *rxm, *nmb;
> +uint16_t nb_rx;
> +uint32_t len;
> +uint32_t i;
> +uint32_t hdr_size;
> +struct virtio_net_hdr *hdr;
> +struct vring_desc_packed *descs = vq->vq_ring.desc_packed;
> +struct vring_desc_packed *desc;
> +uint16_t used_idx, id;
> +struct vq_de
Re: [dpdk-dev] [dpdk-stable] [PATCH] net/mlx5: fix interrupt completion queue index wrapping
On 9/4/2018 7:21 AM, Shahaf Shuler wrote:
> Tuesday, August 28, 2018 9:59 PM, Yongseok Koh:
>> wrapping
>>
>>> On Aug 23, 2018, at 4:10 PM, Xueming Li wrote:
>>>
>>> Rxq cq_ci was 16 bits while hardware is expecting to wrap around 24
>>> bits, this caused interrupt failure after burst of packets.
>>>
>>> Fixes: 43e9d9794cde ("net/mlx5: support upstream rdma-core")
>>> Cc: Shahaf Shuler
I assume you would want to backport the fix to stable releases, so adding:
Cc: sta...@dpdk.org
Please shout if that is not the intention.
>>>
>>> Signed-off-by: Xueming Li
>>> ---
>> Acked-by: Yongseok Koh
>
> Applied to next-net-mlx, thanks.
>
>>
>> Thanks
[dpdk-dev] [PATCH] net/ifc: add live migration support
IFCVF can help to log dirty page in live migration stage,
each queue's index can be read and configured to support
VHOST_USER_GET_VRING_BASE and VHOST_USER_SET_VRING_BASE.
Signed-off-by: Xiao Wang
---
drivers/net/ifc/base/ifcvf.c | 33 +++-
drivers/net/ifc/base/ifcvf.h | 7 +
drivers/net/ifc/ifcvf_vdpa.c | 71 ++--
3 files changed, 108 insertions(+), 3 deletions(-)
diff --git a/drivers/net/ifc/base/ifcvf.c b/drivers/net/ifc/base/ifcvf.c
index 4b22d9ed1..3c0b2dff6 100644
--- a/drivers/net/ifc/base/ifcvf.c
+++ b/drivers/net/ifc/base/ifcvf.c
@@ -249,7 +249,7 @@ ifcvf_hw_disable(struct ifcvf_hw *hw)
IFCVF_WRITE_REG16(IFCVF_MSI_NO_VECTOR, &cfg->queue_msix_vector);
ring_state = *(u32 *)(hw->lm_cfg + IFCVF_LM_RING_STATE_OFFSET +
(i / 2) * IFCVF_LM_CFG_SIZE + (i % 2) * 4);
- hw->vring[i].last_avail_idx = (u16)ring_state;
+ hw->vring[i].last_avail_idx = (u16)(ring_state >> 16);
hw->vring[i].last_used_idx = (u16)(ring_state >> 16);
}
}
@@ -278,6 +278,37 @@ ifcvf_stop_hw(struct ifcvf_hw *hw)
ifcvf_reset(hw);
}
+void
+ifcvf_enable_logging(struct ifcvf_hw *hw, u64 log_base, u64 log_size)
+{
+ u8 *lm_cfg;
+
+ lm_cfg = hw->lm_cfg;
+
+ *(u32 *)(lm_cfg + IFCVF_LM_BASE_ADDR_LOW) =
+ log_base & IFCVF_32_BIT_MASK;
+
+ *(u32 *)(lm_cfg + IFCVF_LM_BASE_ADDR_HIGH) =
+ (log_base >> 32) & IFCVF_32_BIT_MASK;
+
+ *(u32 *)(lm_cfg + IFCVF_LM_END_ADDR_LOW) =
+ (log_base + log_size) & IFCVF_32_BIT_MASK;
+
+ *(u32 *)(lm_cfg + IFCVF_LM_END_ADDR_HIGH) =
+ ((log_base + log_size) >> 32) & IFCVF_32_BIT_MASK;
+
+ *(u32 *)(lm_cfg + IFCVF_LM_LOGGING_CTRL) = IFCVF_LM_ENABLE_VF;
+}
+
+void
+ifcvf_disable_logging(struct ifcvf_hw *hw)
+{
+ u8 *lm_cfg;
+
+ lm_cfg = hw->lm_cfg;
+ *(u32 *)(lm_cfg + IFCVF_LM_LOGGING_CTRL) = IFCVF_LM_DISABLE;
+}
+
void
ifcvf_notify_queue(struct ifcvf_hw *hw, u16 qid)
{
diff --git a/drivers/net/ifc/base/ifcvf.h b/drivers/net/ifc/base/ifcvf.h
index badacb615..f026c70ab 100644
--- a/drivers/net/ifc/base/ifcvf.h
+++ b/drivers/net/ifc/base/ifcvf.h
@@ -49,6 +49,7 @@
#define IFCVF_LM_DISABLE 0x0
#define IFCVF_LM_ENABLE_VF 0x1
#define IFCVF_LM_ENABLE_PF 0x3
+#define IFCVF_LOG_BASE 0x1000
#define IFCVF_32_BIT_MASK 0x
@@ -142,6 +143,12 @@ ifcvf_start_hw(struct ifcvf_hw *hw);
void
ifcvf_stop_hw(struct ifcvf_hw *hw);
+void
+ifcvf_enable_logging(struct ifcvf_hw *hw, u64 log_base, u64 log_size);
+
+void
+ifcvf_disable_logging(struct ifcvf_hw *hw);
+
void
ifcvf_notify_queue(struct ifcvf_hw *hw, u16 qid);
diff --git a/drivers/net/ifc/ifcvf_vdpa.c b/drivers/net/ifc/ifcvf_vdpa.c
index 88d814037..3c5430dc0 100644
--- a/drivers/net/ifc/ifcvf_vdpa.c
+++ b/drivers/net/ifc/ifcvf_vdpa.c
@@ -7,6 +7,7 @@
#include
#include
#include
+#include
#include
#include
@@ -276,12 +277,30 @@ vdpa_ifcvf_start(struct ifcvf_internal *internal)
return ifcvf_start_hw(&internal->hw);
}
+static void
+ifcvf_used_ring_log(struct ifcvf_hw *hw, uint32_t queue, uint8_t *log_buf)
+{
+ uint32_t i, size;
+ uint64_t pfn;
+
+ pfn = hw->vring[queue].used / PAGE_SIZE;
+ size = hw->vring[queue].size * sizeof(struct vring_used_elem) +
+ sizeof(__virtio16) * 3;
+
+ for (i = 0; i <= size / PAGE_SIZE; i++)
+ __sync_fetch_and_or_8(&log_buf[(pfn + i) / 8],
+ 1 << ((pfn + i) % 8));
+}
+
static void
vdpa_ifcvf_stop(struct ifcvf_internal *internal)
{
struct ifcvf_hw *hw = &internal->hw;
uint32_t i;
int vid;
+ uint64_t features;
+ uint64_t log_base, log_size;
+ uint8_t *log_buf;
vid = internal->vid;
ifcvf_stop_hw(hw);
@@ -289,6 +308,21 @@ vdpa_ifcvf_stop(struct ifcvf_internal *internal)
for (i = 0; i < hw->nr_vring; i++)
rte_vhost_set_vring_base(vid, i, hw->vring[i].last_avail_idx,
hw->vring[i].last_used_idx);
+
+ rte_vhost_get_negotiated_features(vid, &features);
+ if (RTE_VHOST_NEED_LOG(features)) {
+ ifcvf_disable_logging(hw);
+ rte_vhost_get_log_base(internal->vid, &log_base, &log_size);
+ rte_vfio_container_dma_unmap(internal->vfio_container_fd,
+ log_base, IFCVF_LOG_BASE, log_size);
+ /*
+* IFCVF marks dirty memory pages for only packet buffer,
+* SW helps to mark the used ring as dirty after device stops.
+*/
+ log_buf = (uint8_t *)(uintptr_t)log_base;
+ for (i = 0; i < hw->nr_vring; i++)
+ ifcvf_used_ring_log(hw, i, log_buf);
+ }
}
#define M
Re: [dpdk-dev] [PATCH v2] ethdev: make default behavior CRC strip on Rx
For szedata2: Acked-by: Jan Remes Thanks, Jan > --- a/drivers/net/szedata2/rte_eth_szedata2.c > +++ b/drivers/net/szedata2/rte_eth_szedata2.c > @@ -1056,8 +1056,7 @@ eth_dev_info(struct rte_eth_dev *dev, > dev_info->max_rx_queues = internals->max_rx_queues; > dev_info->max_tx_queues = internals->max_tx_queues; > dev_info->min_rx_bufsize = 0; > - dev_info->rx_offload_capa = DEV_RX_OFFLOAD_SCATTER | > - DEV_RX_OFFLOAD_CRC_STRIP; > + dev_info->rx_offload_capa = DEV_RX_OFFLOAD_SCATTER; > dev_info->tx_offload_capa = 0; > dev_info->rx_queue_offload_capa = 0; > dev_info->tx_queue_offload_capa = 0;
[dpdk-dev] [PATCH] net/ifc: do not notify before HW ready
Fixes: a3f8150eac6d ("net/ifcvf: add ifcvf vDPA driver")
Signed-off-by: Xiao Wang
---
drivers/net/ifc/ifcvf_vdpa.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ifc/ifcvf_vdpa.c b/drivers/net/ifc/ifcvf_vdpa.c
index 3c5430dc0..7d3085d8d 100644
--- a/drivers/net/ifc/ifcvf_vdpa.c
+++ b/drivers/net/ifc/ifcvf_vdpa.c
@@ -503,11 +503,11 @@ update_datapath(struct ifcvf_internal *internal)
if (ret)
goto err;
- ret = setup_notify_relay(internal);
+ ret = vdpa_ifcvf_start(internal);
if (ret)
goto err;
- ret = vdpa_ifcvf_start(internal);
+ ret = setup_notify_relay(internal);
if (ret)
goto err;
@@ -515,12 +515,12 @@ update_datapath(struct ifcvf_internal *internal)
} else if (rte_atomic32_read(&internal->running) &&
(!rte_atomic32_read(&internal->started) ||
!rte_atomic32_read(&internal->dev_attached))) {
- vdpa_ifcvf_stop(internal);
-
ret = unset_notify_relay(internal);
if (ret)
goto err;
+ vdpa_ifcvf_stop(internal);
+
ret = vdpa_disable_vfio_intr(internal);
if (ret)
goto err;
--
2.15.1
[dpdk-dev] [PATCH v2] build: add PPC64 Meson build
This has been only build-tested for now, on a native ppc64el POWER8E
machine running Debian sid.
Signed-off-by: Luca Boccassi
---
v2: fix indentation, drop march_opt in ppc build file
config/meson.build | 10 +-
config/ppc_64/meson.build| 11 +++
lib/librte_eal/common/arch/ppc_64/meson.build| 5 +
.../common/include/arch/ppc_64/meson.build | 16
4 files changed, 41 insertions(+), 1 deletion(-)
create mode 100644 config/ppc_64/meson.build
create mode 100644 lib/librte_eal/common/arch/ppc_64/meson.build
create mode 100644 lib/librte_eal/common/include/arch/ppc_64/meson.build
diff --git a/config/meson.build b/config/meson.build
index 4d755323f4..6f9228c874 100644
--- a/config/meson.build
+++ b/config/meson.build
@@ -9,7 +9,13 @@ else
endif
dpdk_conf.set('RTE_MACHINE', machine)
machine_args = []
-machine_args += '-march=' + machine
+# ppc64 does not support -march=native
+if host_machine.cpu_family().startswith('ppc') and machine == 'native'
+ machine_args += '-mcpu=' + machine
+ machine_args += '-mtune=' + machine
+else
+ machine_args += '-march=' + machine
+endif
toolchain = cc.get_id()
dpdk_conf.set_quoted('RTE_TOOLCHAIN', toolchain)
@@ -84,6 +90,8 @@ if host_machine.cpu_family().startswith('x86')
arch_subdir = 'x86'
elif host_machine.cpu_family().startswith('arm') or
host_machine.cpu_family().startswith('aarch')
arch_subdir = 'arm'
+elif host_machine.cpu_family().startswith('ppc')
+ arch_subdir = 'ppc_64'
endif
subdir(arch_subdir)
dpdk_conf.set('RTE_COMPILE_TIME_CPUFLAGS', ','.join(compile_time_cpuflags))
diff --git a/config/ppc_64/meson.build b/config/ppc_64/meson.build
new file mode 100644
index 00..e207c438bf
--- /dev/null
+++ b/config/ppc_64/meson.build
@@ -0,0 +1,11 @@
+# SPDX-License-Identifier: BSD-3-Clause
+# Copyright(c) 2018 Luca Boccassi
+
+dpdk_conf.set('RTE_ARCH', 'ppc_64')
+dpdk_conf.set('RTE_ARCH_PPC_64', 1)
+dpdk_conf.set('RTE_ARCH_64', 1)
+
+# overrides specific to ppc64
+dpdk_conf.set('RTE_MAX_LCORE', 256)
+dpdk_conf.set('RTE_MAX_NUMA_NODES', 32)
+dpdk_conf.set('RTE_CACHE_LINE_SIZE', 128)
diff --git a/lib/librte_eal/common/arch/ppc_64/meson.build
b/lib/librte_eal/common/arch/ppc_64/meson.build
new file mode 100644
index 00..40b3dc533a
--- /dev/null
+++ b/lib/librte_eal/common/arch/ppc_64/meson.build
@@ -0,0 +1,5 @@
+# SPDX-License-Identifier: BSD-3-Clause
+# Copyright(c) 2018 Luca Boccassi
+
+eal_common_arch_sources = files('rte_cpuflags.c',
+ 'rte_cycles.c', 'rte_hypervisor.c')
diff --git a/lib/librte_eal/common/include/arch/ppc_64/meson.build
b/lib/librte_eal/common/include/arch/ppc_64/meson.build
new file mode 100644
index 00..00f9611768
--- /dev/null
+++ b/lib/librte_eal/common/include/arch/ppc_64/meson.build
@@ -0,0 +1,16 @@
+# SPDX-License-Identifier: BSD-3-Clause
+# Copyright(c) 2018 Luca Boccassi
+
+install_headers(
+ 'rte_atomic.h',
+ 'rte_byteorder.h',
+ 'rte_cpuflags.h',
+ 'rte_cycles.h',
+ 'rte_io.h',
+ 'rte_memcpy.h',
+ 'rte_pause.h',
+ 'rte_prefetch.h',
+ 'rte_rwlock.h',
+ 'rte_spinlock.h',
+ 'rte_vect.h',
+ subdir: get_option('include_subdir_arch'))
--
2.18.0
Re: [dpdk-dev] [PATCH] build: add PPC64 Meson build
On Mon, 2018-09-10 at 11:50 +0100, Bruce Richardson wrote:
> On Fri, Sep 07, 2018 at 07:35:02PM +0100, Luca Boccassi wrote:
> > This has been only build-tested for now, on a native ppc64el
> > POWER8E
> > machine running Debian sid.
> >
> > Signed-off-by: Luca Boccassi
> > ---
> > The build box cannot be used to run DPDK as it doesn't have
> > supported
> > NICs and root access. Would be great if someone could run-test it,
> > but
> > at this point I think build support is enough to get started.
> >
> > config/meson.build | 8
> > config/ppc_64/meson.build| 15
> > +++
> > lib/librte_eal/common/arch/ppc_64/meson.build| 5 +
> > .../common/include/arch/ppc_64/meson.build | 16
> >
> > 4 files changed, 44 insertions(+)
> > create mode 100644 config/ppc_64/meson.build
> > create mode 100644 lib/librte_eal/common/arch/ppc_64/meson.build
> > create mode 100644
> > lib/librte_eal/common/include/arch/ppc_64/meson.build
> >
> > diff --git a/config/meson.build b/config/meson.build
> > index 4d755323f4..8e87b344c2 100644
> > --- a/config/meson.build
> > +++ b/config/meson.build
> > @@ -9,7 +9,13 @@ else
> > endif
> > dpdk_conf.set('RTE_MACHINE', machine)
> > machine_args = []
> > +# ppc64 does not support -march=native
> > +if host_machine.cpu_family().startswith('ppc') and machine ==
> > 'native'
> > +machine_args += '-mcpu=' + machine
> > +machine_args += '-mtune=' + machine
> > +else
> > machine_args += '-march=' + machine
> > +endif
>
> Indentation?
Was left in the keyboard
> >
> > toolchain = cc.get_id()
> > dpdk_conf.set_quoted('RTE_TOOLCHAIN', toolchain)
> > @@ -84,6 +90,8 @@ if host_machine.cpu_family().startswith('x86')
> > arch_subdir = 'x86'
> > elif host_machine.cpu_family().startswith('arm') or
> > host_machine.cpu_family().startswith('aarch')
> > arch_subdir = 'arm'
> > +elif host_machine.cpu_family().startswith('ppc')
> > + arch_subdir = 'ppc_64'
> > endif
> > subdir(arch_subdir)
> > dpdk_conf.set('RTE_COMPILE_TIME_CPUFLAGS',
> > ','.join(compile_time_cpuflags))
> > diff --git a/config/ppc_64/meson.build b/config/ppc_64/meson.build
> > new file mode 100644
> > index 00..d6faa7d64f
> > --- /dev/null
> > +++ b/config/ppc_64/meson.build
> > @@ -0,0 +1,15 @@
> > +# SPDX-License-Identifier: BSD-3-Clause
> > +# Copyright(c) 2018 Luca Boccassi
> > +
> > +# for checking defines we need to use the correct compiler flags
> > +march_opt = '-march=@0@'.format(machine)
>
> This contradicts the statement above in config/meson.build where you
> state
> that ppc64 doesn't support "-march=native"?
Yeah I copy-pasted the arm file to get started, but it doesn't seem to
me march_opt is used anywhere else other than in the x86 specific file,
so I've dropped that line in v2
--
Kind regards,
Luca Boccassi
Re: [dpdk-dev] [PATCH v5] linuxapp, eal: Fix the memory leak issue of logid
Hi
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ziye Yang
> Sent: Wednesday, September 5, 2018 6:39 AM
> To: dev@dpdk.org
> Cc: Ziye Yang
> Subject: [dpdk-dev] [PATCH v5] linuxapp, eal: Fix the memory leak issue of
> logid
>
> From: Ziye Yang
>
> This patch is used to fix the memory leak issue of logid.
> We use the ASAN test in SPDK when intergrating DPDK and
> find this memory leak issue.
>
> Signed-off-by: Ziye Yang
> ---
> lib/librte_eal/linuxapp/eal/eal.c | 21 -
> lib/librte_eal/linuxapp/eal/eal_log.c | 11 ++-
> 2 files changed, 30 insertions(+), 2 deletions(-)
>
> diff --git a/lib/librte_eal/linuxapp/eal/eal.c
> b/lib/librte_eal/linuxapp/eal/eal.c
> index e59ac65..e150200 100644
> --- a/lib/librte_eal/linuxapp/eal/eal.c
> +++ b/lib/librte_eal/linuxapp/eal/eal.c
> @@ -793,7 +793,7 @@ static void rte_eal_init_alert(const char *msg)
> int i, fctret, ret;
> pthread_t thread_id;
> static rte_atomic32_t run_once = RTE_ATOMIC32_INIT(0);
> - const char *logid;
> + char *logid;
> char cpuset[RTE_CPU_AFFINITY_STR_LEN];
> char thread_name[RTE_MAX_THREAD_NAME_LEN];
>
> @@ -812,6 +812,12 @@ static void rte_eal_init_alert(const char *msg)
>
> logid = strrchr(argv[0], '/');
> logid = strdup(logid ? logid + 1: argv[0]);
> + if (!logid) {
> + rte_eal_init_alert("Cannot allocate memory for logid\n");
> + rte_errno = ENOMEM;
> + rte_atomic32_clear(&run_once);
> + return -1;
> + }
>
> thread_id = pthread_self();
>
> @@ -823,6 +829,8 @@ static void rte_eal_init_alert(const char *msg)
> if (rte_eal_cpu_init() < 0) {
> rte_eal_init_alert("Cannot detect lcores.");
> rte_errno = ENOTSUP;
> + rte_atomic32_clear(&run_once);
> + free(logid);
> return -1;
> }
>
> @@ -831,6 +839,7 @@ static void rte_eal_init_alert(const char *msg)
> rte_eal_init_alert("Invalid 'command line' arguments.");
> rte_errno = EINVAL;
> rte_atomic32_clear(&run_once);
> + free(logid);
> return -1;
> }
>
> @@ -838,12 +847,14 @@ static void rte_eal_init_alert(const char *msg)
> rte_eal_init_alert("Cannot init plugins\n");
> rte_errno = EINVAL;
> rte_atomic32_clear(&run_once);
> + free(logid);
> return -1;
> }
>
> if (eal_option_device_parse()) {
> rte_errno = ENODEV;
> rte_atomic32_clear(&run_once);
> + free(logid);
> return -1;
> }
>
> @@ -851,6 +862,8 @@ static void rte_eal_init_alert(const char *msg)
>
> if (rte_eal_intr_init() < 0) {
> rte_eal_init_alert("Cannot init interrupt-handling thread\n");
> + rte_atomic32_clear(&run_once);
> + free(logid);
> return -1;
> }
>
> @@ -861,6 +874,8 @@ static void rte_eal_init_alert(const char *msg)
> rte_eal_init_alert("failed to init mp channel\n");
> if (rte_eal_process_type() == RTE_PROC_PRIMARY) {
> rte_errno = EFAULT;
> + rte_atomic32_clear(&run_once);
> + free(logid);
> return -1;
> }
> }
> @@ -869,6 +884,7 @@ static void rte_eal_init_alert(const char *msg)
> rte_eal_init_alert("Cannot scan the buses for devices\n");
> rte_errno = ENODEV;
> rte_atomic32_clear(&run_once);
> + free(logid);
> return -1;
> }
>
> @@ -893,6 +909,7 @@ static void rte_eal_init_alert(const char *msg)
> rte_eal_init_alert("Cannot get hugepage information.");
> rte_errno = EACCES;
> rte_atomic32_clear(&run_once);
> + free(logid);
> return -1;
> }
> }
> @@ -919,8 +936,10 @@ static void rte_eal_init_alert(const char *msg)
> rte_eal_init_alert("Cannot init logging.");
> rte_errno = ENOMEM;
> rte_atomic32_clear(&run_once);
> + free(logid);
> return -1;
> }
> + free(logid);
>
> #ifdef VFIO_PRESENT
> if (rte_eal_vfio_setup() < 0) {
> diff --git a/lib/librte_eal/linuxapp/eal/eal_log.c
> b/lib/librte_eal/linuxapp/eal/eal_log.c
> index 9d02ddd..076b1b9 100644
> --- a/lib/librte_eal/linuxapp/eal/eal_log.c
> +++ b/lib/librte_eal/linuxapp/eal/eal_log.c
> @@ -19,6 +19,9 @@
>
> #include "eal_private.h"
>
> +#define DEFAULT_LOG_ID_LEN 128
> +char g_logid[DEFAULT_LOG_ID_LEN];
> +
> /*
> * default log function
> */
> @@ -49,12 +52,18 @@
> rte_eal_log_init(const char *id, int facility)
> {
> FILE *log_stream;
> + int str_len;
>
> log_stream = fopencookie(NULL, "w
Re: [dpdk-dev] [PATCH v5] linuxapp, eal: Fix the memory leak issue of logid
Hi Konstantin,
The global logid is used to prevent the memory leak. If we do not do this, but
directly free(logid), this pointer will still be used by openlog, which means
that we cannot free the logid anymore. And by adding the global variable, it
will not only solve the memory leak but also makes the openlog function still
work.
Best Regards
Ziye Yang
-Original Message-
From: Ananyev, Konstantin
Sent: Monday, September 10, 2018 7:55 PM
To: Yang, Ziye ; dev@dpdk.org
Cc: Ziye Yang
Subject: RE: [dpdk-dev] [PATCH v5] linuxapp, eal: Fix the memory leak issue of
logid
Hi
> -Original Message-
> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ziye Yang
> Sent: Wednesday, September 5, 2018 6:39 AM
> To: dev@dpdk.org
> Cc: Ziye Yang
> Subject: [dpdk-dev] [PATCH v5] linuxapp, eal: Fix the memory leak
> issue of logid
>
> From: Ziye Yang
>
> This patch is used to fix the memory leak issue of logid.
> We use the ASAN test in SPDK when intergrating DPDK and find this
> memory leak issue.
>
> Signed-off-by: Ziye Yang
> ---
> lib/librte_eal/linuxapp/eal/eal.c | 21 -
> lib/librte_eal/linuxapp/eal/eal_log.c | 11 ++-
> 2 files changed, 30 insertions(+), 2 deletions(-)
>
> diff --git a/lib/librte_eal/linuxapp/eal/eal.c
> b/lib/librte_eal/linuxapp/eal/eal.c
> index e59ac65..e150200 100644
> --- a/lib/librte_eal/linuxapp/eal/eal.c
> +++ b/lib/librte_eal/linuxapp/eal/eal.c
> @@ -793,7 +793,7 @@ static void rte_eal_init_alert(const char *msg)
> int i, fctret, ret;
> pthread_t thread_id;
> static rte_atomic32_t run_once = RTE_ATOMIC32_INIT(0);
> - const char *logid;
> + char *logid;
> char cpuset[RTE_CPU_AFFINITY_STR_LEN];
> char thread_name[RTE_MAX_THREAD_NAME_LEN];
>
> @@ -812,6 +812,12 @@ static void rte_eal_init_alert(const char *msg)
>
> logid = strrchr(argv[0], '/');
> logid = strdup(logid ? logid + 1: argv[0]);
> + if (!logid) {
> + rte_eal_init_alert("Cannot allocate memory for logid\n");
> + rte_errno = ENOMEM;
> + rte_atomic32_clear(&run_once);
> + return -1;
> + }
>
> thread_id = pthread_self();
>
> @@ -823,6 +829,8 @@ static void rte_eal_init_alert(const char *msg)
> if (rte_eal_cpu_init() < 0) {
> rte_eal_init_alert("Cannot detect lcores.");
> rte_errno = ENOTSUP;
> + rte_atomic32_clear(&run_once);
> + free(logid);
> return -1;
> }
>
> @@ -831,6 +839,7 @@ static void rte_eal_init_alert(const char *msg)
> rte_eal_init_alert("Invalid 'command line' arguments.");
> rte_errno = EINVAL;
> rte_atomic32_clear(&run_once);
> + free(logid);
> return -1;
> }
>
> @@ -838,12 +847,14 @@ static void rte_eal_init_alert(const char *msg)
> rte_eal_init_alert("Cannot init plugins\n");
> rte_errno = EINVAL;
> rte_atomic32_clear(&run_once);
> + free(logid);
> return -1;
> }
>
> if (eal_option_device_parse()) {
> rte_errno = ENODEV;
> rte_atomic32_clear(&run_once);
> + free(logid);
> return -1;
> }
>
> @@ -851,6 +862,8 @@ static void rte_eal_init_alert(const char *msg)
>
> if (rte_eal_intr_init() < 0) {
> rte_eal_init_alert("Cannot init interrupt-handling thread\n");
> + rte_atomic32_clear(&run_once);
> + free(logid);
> return -1;
> }
>
> @@ -861,6 +874,8 @@ static void rte_eal_init_alert(const char *msg)
> rte_eal_init_alert("failed to init mp channel\n");
> if (rte_eal_process_type() == RTE_PROC_PRIMARY) {
> rte_errno = EFAULT;
> + rte_atomic32_clear(&run_once);
> + free(logid);
> return -1;
> }
> }
> @@ -869,6 +884,7 @@ static void rte_eal_init_alert(const char *msg)
> rte_eal_init_alert("Cannot scan the buses for devices\n");
> rte_errno = ENODEV;
> rte_atomic32_clear(&run_once);
> + free(logid);
> return -1;
> }
>
> @@ -893,6 +909,7 @@ static void rte_eal_init_alert(const char *msg)
> rte_eal_init_alert("Cannot get hugepage information.");
> rte_errno = EACCES;
> rte_atomic32_clear(&run_once);
> + free(logid);
> return -1;
> }
> }
> @@ -919,8 +936,10 @@ static void rte_eal_init_alert(const char *msg)
> rte_eal_init_alert("Cannot init logging.");
> rte_errno = ENOMEM;
> rte_atomic32_clear(&run_once);
> + free(logid);
> return -1;
> }
> + free(logid);
>
> #ifdef VFIO_PRESENT
>
Re: [dpdk-dev] [PATCH v5] net/pcap: physical interface MAC address support
On 9/6/2018 5:56 PM, Juhamatti Kuusisaari wrote:
> Support for PCAP physical interface MAC with phy_mac=1 devarg.
>
> Signed-off-by: Juhamatti Kuusisaari
Hi Juhamatti,
Thanks for the patch.
Can you please add a little more details into commit log that why this new
feature is enabled, what is the use case?
Also can you please send new version of the patches as a reply to previous one,
- via "--in-reply-to" git send-email argument. And a history after commit log
what changed in new version. These helps tracing the multiple versions.
> ---
> doc/guides/rel_notes/release_18_11.rst | 4 +
Please update doc/guides/nics/pcap_ring.rst to document new devargs.
> drivers/net/pcap/rte_eth_pcap.c| 119 +++--
> 2 files changed, 118 insertions(+), 5 deletions(-)
>
> diff --git a/doc/guides/rel_notes/release_18_11.rst
> b/doc/guides/rel_notes/release_18_11.rst
> index 3ae6b3f58..70966740a 100644
> --- a/doc/guides/rel_notes/release_18_11.rst
> +++ b/doc/guides/rel_notes/release_18_11.rst
> @@ -54,6 +54,10 @@ New Features
> Also, make sure to start the actual text at the margin.
> =
>
> +* **Added a devarg to use PCAP interface physical MAC address.**
> + A new devarg ``phy_mac`` was introduced to allow users to use physical
> + MAC address of the selected PCAP interface.
> +
>
> API Changes
> ---
> diff --git a/drivers/net/pcap/rte_eth_pcap.c b/drivers/net/pcap/rte_eth_pcap.c
> index e8810a171..8917c4c4d 100644
> --- a/drivers/net/pcap/rte_eth_pcap.c
> +++ b/drivers/net/pcap/rte_eth_pcap.c
> @@ -7,6 +7,14 @@
> #include
>
> #include
> +#include
> +#include
> +#include
> +
> +#ifdef __FreeBSD__
> +#include
> +#include
> +#endif
>
> #include
>
> @@ -17,6 +25,7 @@
> #include
> #include
> #include
> +#include
>
> #define RTE_ETH_PCAP_SNAPSHOT_LEN 65535
> #define RTE_ETH_PCAP_SNAPLEN ETHER_MAX_JUMBO_FRAME_LEN
> @@ -29,6 +38,7 @@
> #define ETH_PCAP_RX_IFACE_IN_ARG "rx_iface_in"
> #define ETH_PCAP_TX_IFACE_ARG "tx_iface"
> #define ETH_PCAP_IFACE_ARG"iface"
> +#define ETH_PCAP_PHY_MAC_ARG "phy_mac"
>
> #define ETH_PCAP_ARG_MAXLEN 64
>
> @@ -87,6 +97,7 @@ static const char *valid_arguments[] = {
> ETH_PCAP_RX_IFACE_IN_ARG,
> ETH_PCAP_TX_IFACE_ARG,
> ETH_PCAP_IFACE_ARG,
> + ETH_PCAP_PHY_MAC_ARG,
> NULL
> };
>
> @@ -904,12 +915,79 @@ pmd_init_internals(struct rte_vdev_device *vdev,
> return 0;
> }
>
> +static void eth_pcap_update_mac(const char *if_name, struct rte_eth_dev
> **eth_dev,
> + const unsigned int numa_node)
Please follow convention, to have return type in a separate line above.
> +{
> + void *mac_addrs;
> + PMD_LOG(INFO, "Setting phy MAC for %s\n",
> + if_name);
This can be single line. And please leave an empty line between variable
declaration and code.
> +#ifndef __FreeBSD__
> + int if_fd = socket(AF_INET, SOCK_DGRAM, 0);
> + if (if_fd != -1) {
To reduce the indentation of rest of the code, it helps to reverse the logic,
exit on failure and continue without indentation.
> + struct ifreq ifr;
> + strlcpy(ifr.ifr_name, if_name, sizeof(ifr.ifr_name));
> + if (!ioctl(if_fd, SIOCGIFHWADDR, &ifr)) {
> + mac_addrs = rte_zmalloc_socket(NULL, ETHER_ADDR_LEN,
> + 0, numa_node);
> + if (mac_addrs) {
> + (*eth_dev)->data->mac_addrs = mac_addrs;
> + rte_memcpy((*eth_dev)->data->mac_addrs,
> + ifr.ifr_addr.sa_data,
> + ETHER_ADDR_LEN);
> + }
> + }
> + close(if_fd);
> + }
> +#else
There is an assumption that it is either Linux or FreeBSD, that is true for now
but Windows is coming :) Better to be safe here and do an "else if" for FreeBSD.
> + int mib[6];
> + size_t len = 0;
> + char *buf = NULL;
> +
> + mib[0] = CTL_NET;
> + mib[1] = AF_ROUTE;
> + mib[2] = 0;
> + mib[3] = AF_LINK;
> + mib[4] = NET_RT_IFLIST;
> + mib[5] = if_nametoindex(if_name);
> +
> + if (sysctl(mib, 6, NULL, &len, NULL, 0) < 0)
> + return;
> +
> + if (len > 0) {
> + struct if_msghdr*ifm;
> + struct sockaddr_dl *sdl;
> +
> + buf = rte_zmalloc_socket(NULL, len,
> + 0, numa_node);
This is an interim buffer, no need to use rte_zmalloc_socket API, rte_malloc or
regular malloc will do the same.
Also no need to break the line.
> + if (buf) {
> + if (sysctl(mib, 6, buf, &len, NULL, 0) < 0) {
> + rte_free(buf);
> + return;
> + }
> +
> + ifm = (struct if_ms
Re: [dpdk-dev] [PATCH v2] ethdev: make default behavior CRC strip on Rx
-Original Message- > > Removed DEV_RX_OFFLOAD_CRC_STRIP offload flag. > Without any specific Rx offload flag, default behavior by PMDs is to > strip CRC. > > PMDs that support keeping CRC should advertise DEV_RX_OFFLOAD_KEEP_CRC > Rx offload capability. > > Applications that require keeping CRC should check PMD capability first > and if it is supported can enable this feature by setting > DEV_RX_OFFLOAD_KEEP_CRC in Rx offload flag in rte_eth_dev_configure() > > Signed-off-by: Ferruh Yigit > Acked-by: Tomasz Duszynski > --- > v2: > * fix flag check > * add KEEP_CRC flag into "show port cap #" > > Note "show port cap #" and > "show port # [r/t]x_offload capabilities/configuration" > does same thing, in long term I suggest removing "show port cap" one For thunderx and octeontx Acked-by: Jerin Jacob
Re: [dpdk-dev] [PATCH v5] linuxapp, eal: Fix the memory leak issue of logid
> -Original Message-
> From: Yang, Ziye
> Sent: Monday, September 10, 2018 12:58 PM
> To: Ananyev, Konstantin ; dev@dpdk.org
> Cc: Ziye Yang
> Subject: RE: [dpdk-dev] [PATCH v5] linuxapp, eal: Fix the memory leak issue
> of logid
>
> Hi Konstantin,
>
> The global logid is used to prevent the memory leak. If we do not do this,
> but directly free(logid), this pointer will still be used by
> openlog, which means that we cannot free the logid anymore. And by adding the
> global variable, it will not only solve the memory
> leak but also makes the openlog function still work.
But then I suppose you can avoid any dynamic memory allocation at all with
something like that:
rte_eal_init(int argc, char **argv)
{
...
static char logid[PATH_MAX];
...
const char *p= strrchr(argv[0], '/');
snprintf(logid, sizeof(logid), "%s", p);
...
rte_eal_log_init(logid, ...);
?
Konstantin
>
>
> Best Regards
> Ziye Yang
>
> -Original Message-
> From: Ananyev, Konstantin
> Sent: Monday, September 10, 2018 7:55 PM
> To: Yang, Ziye ; dev@dpdk.org
> Cc: Ziye Yang
> Subject: RE: [dpdk-dev] [PATCH v5] linuxapp, eal: Fix the memory leak issue
> of logid
>
> Hi
>
> > -Original Message-
> > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ziye Yang
> > Sent: Wednesday, September 5, 2018 6:39 AM
> > To: dev@dpdk.org
> > Cc: Ziye Yang
> > Subject: [dpdk-dev] [PATCH v5] linuxapp, eal: Fix the memory leak
> > issue of logid
> >
> > From: Ziye Yang
> >
> > This patch is used to fix the memory leak issue of logid.
> > We use the ASAN test in SPDK when intergrating DPDK and find this
> > memory leak issue.
> >
> > Signed-off-by: Ziye Yang
> > ---
> > lib/librte_eal/linuxapp/eal/eal.c | 21 -
> > lib/librte_eal/linuxapp/eal/eal_log.c | 11 ++-
> > 2 files changed, 30 insertions(+), 2 deletions(-)
> >
> > diff --git a/lib/librte_eal/linuxapp/eal/eal.c
> > b/lib/librte_eal/linuxapp/eal/eal.c
> > index e59ac65..e150200 100644
> > --- a/lib/librte_eal/linuxapp/eal/eal.c
> > +++ b/lib/librte_eal/linuxapp/eal/eal.c
> > @@ -793,7 +793,7 @@ static void rte_eal_init_alert(const char *msg)
> > int i, fctret, ret;
> > pthread_t thread_id;
> > static rte_atomic32_t run_once = RTE_ATOMIC32_INIT(0);
> > - const char *logid;
> > + char *logid;
> > char cpuset[RTE_CPU_AFFINITY_STR_LEN];
> > char thread_name[RTE_MAX_THREAD_NAME_LEN];
> >
> > @@ -812,6 +812,12 @@ static void rte_eal_init_alert(const char *msg)
> >
> > logid = strrchr(argv[0], '/');
> > logid = strdup(logid ? logid + 1: argv[0]);
> > + if (!logid) {
> > + rte_eal_init_alert("Cannot allocate memory for logid\n");
> > + rte_errno = ENOMEM;
> > + rte_atomic32_clear(&run_once);
> > + return -1;
> > + }
> >
> > thread_id = pthread_self();
> >
> > @@ -823,6 +829,8 @@ static void rte_eal_init_alert(const char *msg)
> > if (rte_eal_cpu_init() < 0) {
> > rte_eal_init_alert("Cannot detect lcores.");
> > rte_errno = ENOTSUP;
> > + rte_atomic32_clear(&run_once);
> > + free(logid);
> > return -1;
> > }
> >
> > @@ -831,6 +839,7 @@ static void rte_eal_init_alert(const char *msg)
> > rte_eal_init_alert("Invalid 'command line' arguments.");
> > rte_errno = EINVAL;
> > rte_atomic32_clear(&run_once);
> > + free(logid);
> > return -1;
> > }
> >
> > @@ -838,12 +847,14 @@ static void rte_eal_init_alert(const char *msg)
> > rte_eal_init_alert("Cannot init plugins\n");
> > rte_errno = EINVAL;
> > rte_atomic32_clear(&run_once);
> > + free(logid);
> > return -1;
> > }
> >
> > if (eal_option_device_parse()) {
> > rte_errno = ENODEV;
> > rte_atomic32_clear(&run_once);
> > + free(logid);
> > return -1;
> > }
> >
> > @@ -851,6 +862,8 @@ static void rte_eal_init_alert(const char *msg)
> >
> > if (rte_eal_intr_init() < 0) {
> > rte_eal_init_alert("Cannot init interrupt-handling thread\n");
> > + rte_atomic32_clear(&run_once);
> > + free(logid);
> > return -1;
> > }
> >
> > @@ -861,6 +874,8 @@ static void rte_eal_init_alert(const char *msg)
> > rte_eal_init_alert("failed to init mp channel\n");
> > if (rte_eal_process_type() == RTE_PROC_PRIMARY) {
> > rte_errno = EFAULT;
> > + rte_atomic32_clear(&run_once);
> > + free(logid);
> > return -1;
> > }
> > }
> > @@ -869,6 +884,7 @@ static void rte_eal_init_alert(const char *msg)
> > rte_eal_init_alert("Cannot scan the buses for devices\n");
> > rte_errno = ENODEV;
> > rte_atomic32_clear(&run_once);
> > + free(logid);
> > return -1;
> > }
> >
> >
Re: [dpdk-dev] [PATCH v5] linuxapp, eal: Fix the memory leak issue of logid
Hi Konstantin,
Thanks for your comments. I will revise the patch and submit another patch
later.
Best Regards
Ziye Yang
-Original Message-
From: Ananyev, Konstantin
Sent: Monday, September 10, 2018 8:19 PM
To: Yang, Ziye ; dev@dpdk.org
Cc: Ziye Yang
Subject: RE: [dpdk-dev] [PATCH v5] linuxapp, eal: Fix the memory leak issue of
logid
> -Original Message-
> From: Yang, Ziye
> Sent: Monday, September 10, 2018 12:58 PM
> To: Ananyev, Konstantin ; dev@dpdk.org
> Cc: Ziye Yang
> Subject: RE: [dpdk-dev] [PATCH v5] linuxapp, eal: Fix the memory leak
> issue of logid
>
> Hi Konstantin,
>
> The global logid is used to prevent the memory leak. If we do not do
> this, but directly free(logid), this pointer will still be used by
> openlog, which means that we cannot free the logid anymore. And by adding the
> global variable, it will not only solve the memory leak but also makes the
> openlog function still work.
But then I suppose you can avoid any dynamic memory allocation at all with
something like that:
rte_eal_init(int argc, char **argv)
{
...
static char logid[PATH_MAX];
...
const char *p= strrchr(argv[0], '/');
snprintf(logid, sizeof(logid), "%s", p);
...
rte_eal_log_init(logid, ...);
?
Konstantin
>
>
> Best Regards
> Ziye Yang
>
> -Original Message-
> From: Ananyev, Konstantin
> Sent: Monday, September 10, 2018 7:55 PM
> To: Yang, Ziye ; dev@dpdk.org
> Cc: Ziye Yang
> Subject: RE: [dpdk-dev] [PATCH v5] linuxapp, eal: Fix the memory leak
> issue of logid
>
> Hi
>
> > -Original Message-
> > From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ziye Yang
> > Sent: Wednesday, September 5, 2018 6:39 AM
> > To: dev@dpdk.org
> > Cc: Ziye Yang
> > Subject: [dpdk-dev] [PATCH v5] linuxapp, eal: Fix the memory leak
> > issue of logid
> >
> > From: Ziye Yang
> >
> > This patch is used to fix the memory leak issue of logid.
> > We use the ASAN test in SPDK when intergrating DPDK and find this
> > memory leak issue.
> >
> > Signed-off-by: Ziye Yang
> > ---
> > lib/librte_eal/linuxapp/eal/eal.c | 21 -
> > lib/librte_eal/linuxapp/eal/eal_log.c | 11 ++-
> > 2 files changed, 30 insertions(+), 2 deletions(-)
> >
> > diff --git a/lib/librte_eal/linuxapp/eal/eal.c
> > b/lib/librte_eal/linuxapp/eal/eal.c
> > index e59ac65..e150200 100644
> > --- a/lib/librte_eal/linuxapp/eal/eal.c
> > +++ b/lib/librte_eal/linuxapp/eal/eal.c
> > @@ -793,7 +793,7 @@ static void rte_eal_init_alert(const char *msg)
> > int i, fctret, ret;
> > pthread_t thread_id;
> > static rte_atomic32_t run_once = RTE_ATOMIC32_INIT(0);
> > - const char *logid;
> > + char *logid;
> > char cpuset[RTE_CPU_AFFINITY_STR_LEN];
> > char thread_name[RTE_MAX_THREAD_NAME_LEN];
> >
> > @@ -812,6 +812,12 @@ static void rte_eal_init_alert(const char *msg)
> >
> > logid = strrchr(argv[0], '/');
> > logid = strdup(logid ? logid + 1: argv[0]);
> > + if (!logid) {
> > + rte_eal_init_alert("Cannot allocate memory for logid\n");
> > + rte_errno = ENOMEM;
> > + rte_atomic32_clear(&run_once);
> > + return -1;
> > + }
> >
> > thread_id = pthread_self();
> >
> > @@ -823,6 +829,8 @@ static void rte_eal_init_alert(const char *msg)
> > if (rte_eal_cpu_init() < 0) {
> > rte_eal_init_alert("Cannot detect lcores.");
> > rte_errno = ENOTSUP;
> > + rte_atomic32_clear(&run_once);
> > + free(logid);
> > return -1;
> > }
> >
> > @@ -831,6 +839,7 @@ static void rte_eal_init_alert(const char *msg)
> > rte_eal_init_alert("Invalid 'command line' arguments.");
> > rte_errno = EINVAL;
> > rte_atomic32_clear(&run_once);
> > + free(logid);
> > return -1;
> > }
> >
> > @@ -838,12 +847,14 @@ static void rte_eal_init_alert(const char *msg)
> > rte_eal_init_alert("Cannot init plugins\n");
> > rte_errno = EINVAL;
> > rte_atomic32_clear(&run_once);
> > + free(logid);
> > return -1;
> > }
> >
> > if (eal_option_device_parse()) {
> > rte_errno = ENODEV;
> > rte_atomic32_clear(&run_once);
> > + free(logid);
> > return -1;
> > }
> >
> > @@ -851,6 +862,8 @@ static void rte_eal_init_alert(const char *msg)
> >
> > if (rte_eal_intr_init() < 0) {
> > rte_eal_init_alert("Cannot init interrupt-handling thread\n");
> > + rte_atomic32_clear(&run_once);
> > + free(logid);
> > return -1;
> > }
> >
> > @@ -861,6 +874,8 @@ static void rte_eal_init_alert(const char *msg)
> > rte_eal_init_alert("failed to init mp channel\n");
> > if (rte_eal_process_type() == RTE_PROC_PRIMARY) {
> > rte_errno = EFAULT;
> > + rte_atomic32_clear(&run_once);
> > + free(logid);
> >
Re: [dpdk-dev] [PATCH 10/10] eventdev: include DSW event device documentation
Will DSW be included in 18.11?
/M
On 2018-08-30 16:27, Mattias Rönnblom wrote:
The DSW event device is documented in DPDK Programmer's Guide.
Signed-off-by: Mattias Rönnblom
---
doc/guides/eventdevs/dsw.rst | 97 ++
doc/guides/eventdevs/index.rst | 1 +
2 files changed, 98 insertions(+)
create mode 100644 doc/guides/eventdevs/dsw.rst
diff --git a/doc/guides/eventdevs/dsw.rst b/doc/guides/eventdevs/dsw.rst
new file mode 100644
index 0..de41ae9d3
--- /dev/null
+++ b/doc/guides/eventdevs/dsw.rst
@@ -0,0 +1,97 @@
+.. SPDX-License-Identifier: BSD-3-Clause
+Copyright(c) 2017 Intel Corporation.
+Copyright(c) 2018 Ericsson AB
+
+Distributed Software Eventdev Poll Mode Driver
+==
+
+The distributed software eventdev is a parallel implementation of the
+eventdev API, which distributes the task of scheduling events among
+all the eventdev ports and the lcore threads using them.
+
+Features
+
+
+Queues
+ * Atomic
+ * Parallel
+ * Single-Link
+
+Ports
+ * Load balanced (for Atomic, Ordered, Parallel queues)
+ * Single Link (for single-link queues)
+
+Configuration and Options
+-
+
+The distributed software eventdev is a vdev device, and as such can be
+created from the application code, or from the EAL command line:
+
+* Call ``rte_vdev_init("event_dsw0")`` from the application
+
+* Use ``--vdev="event_dsw0"`` in the EAL options, which will call
+ rte_vdev_init() internally
+
+Example:
+
+.. code-block:: console
+
+./your_eventdev_application --vdev="event_dsw0"
+
+Limitations
+---
+
+Unattended Ports
+
+
+The distributed software eventdev uses an internal signaling schema
+between the ports to achieve load balancing. In order for this to
+work, the application must perform enqueue and/or dequeue operations
+on all ports.
+
+Producer-only ports which currently have no events to enqueue should
+periodically call rte_event_enqueue_burst() with a zero-sized burst.
+
+Ports left unattended for longer periods of time will prevent load
+balancing, and also cause traffic interruptions on the flows which
+are in the process of being migrated.
+
+Output Buffering
+
+
+For efficiency reasons, the distributed software eventdev might not
+send enqueued events immediately to the destination port, but instead
+store them in an internal buffer in the source port.
+
+In case no more events are enqueued on a port with buffered events,
+these events will be sent after the application has performed a number
+of enqueue and/or dequeue operations.
+
+For explicit flushing, an application may call
+rte_event_enqueue_burst() with a zero-sized burst.
+
+
+Priorities
+~~
+
+The distributed software eventdev does not support event priorities.
+
+Ordered Queues
+~~
+
+The distributed software eventdev does not support the ordered queue type.
+
+
+"All Types" Queues
+~~
+
+The distributed software eventdev does not support queues of type
+RTE_EVENT_QUEUE_CFG_ALL_TYPES, which allow both atomic, ordered, and
+parallel events on the same queue.
+
+Dynamic Link/Unlink
+~~~
+
+The distributed software eventdev does not support calls to
+rte_event_port_link() or rte_event_port_unlink() after
+rte_event_dev_start() has been called.
diff --git a/doc/guides/eventdevs/index.rst b/doc/guides/eventdevs/index.rst
index 18ec8e462..984eea5f4 100644
--- a/doc/guides/eventdevs/index.rst
+++ b/doc/guides/eventdevs/index.rst
@@ -14,5 +14,6 @@ application trough the eventdev API.
dpaa
dpaa2
sw
+dsw
octeontx
opdl
Re: [dpdk-dev] [PATCH] doc: fix wrong usage of bind command
On 08/24/2018 09:43 AM, Rami Rosen wrote: This patch fixes wrong usage of bind command in vhost.rst. Using "dpdk-devbind.py -b=uio_pci_generic :00:04.0" gives an error of "unbind failed". It should be "-b uio_pci_generic" so it will work correctly. Signed-off-by: Rami Rosen --- doc/guides/sample_app_ug/vhost.rst | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/doc/guides/sample_app_ug/vhost.rst b/doc/guides/sample_app_ug/vhost.rst index fd42cb3f7..df4d6f9a0 100644 --- a/doc/guides/sample_app_ug/vhost.rst +++ b/doc/guides/sample_app_ug/vhost.rst @@ -78,7 +78,7 @@ could be done by: .. code-block:: console modprobe uio_pci_generic - $RTE_SDK/usertools/dpdk-devbind.py -b=uio_pci_generic :00:04.0 + $RTE_SDK/usertools/dpdk-devbind.py -b uio_pci_generic :00:04.0 Then start testpmd for packet forwarding testing. Reviewed-by: Maxime Coquelin And applied to dpdk-next-virtio/master Thanks, Maxime
Re: [dpdk-dev] [PATCH 10/10] eventdev: include DSW event device documentation
-Original Message-
> Date: Mon, 10 Sep 2018 14:59:51 +0200
> From: Mattias Rönnblom
> To: jerin.ja...@caviumnetworks.com
> CC: bruce.richard...@intel.com, dev@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH 10/10] eventdev: include DSW event device
> documentation
> User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101
> Thunderbird/52.9.1
>
>
> Will DSW be included in 18.11?
Yes.
I may have minor comments on individual patches. But overall the patches looks
good.
Please fix the following check-git-log.sh issues
[dpdk-next-eventdev] $ ./devtools/check-git-log.sh
Wrong headline prefix:
eventdev: add DSW device and queue configuration
eventdev: add DSW port configuration
eventdev: add support in DSW for linking/unlinking ports
eventdev: add DSW event scheduling and device start/stop
eventdev: add DSW port load measurements
eventdev: add load balancing to the DSW event device
eventdev: let DSW event device sort events on dequeue
eventdev: implement eventdev 'xstats' counters in DSW
>
> /M
>
> On 2018-08-30 16:27, Mattias Rönnblom wrote:
> > The DSW event device is documented in DPDK Programmer's Guide.
> >
> > Signed-off-by: Mattias Rönnblom
> > ---
> > doc/guides/eventdevs/dsw.rst | 97 ++
> > doc/guides/eventdevs/index.rst | 1 +
> > 2 files changed, 98 insertions(+)
> > create mode 100644 doc/guides/eventdevs/dsw.rst
> >
> > diff --git a/doc/guides/eventdevs/dsw.rst b/doc/guides/eventdevs/dsw.rst
> > new file mode 100644
> > index 0..de41ae9d3
> > --- /dev/null
> > +++ b/doc/guides/eventdevs/dsw.rst
> > @@ -0,0 +1,97 @@
> > +.. SPDX-License-Identifier: BSD-3-Clause
> > +Copyright(c) 2017 Intel Corporation.
> > +Copyright(c) 2018 Ericsson AB
> > +
> > +Distributed Software Eventdev Poll Mode Driver
> > +==
> > +
> > +The distributed software eventdev is a parallel implementation of the
> > +eventdev API, which distributes the task of scheduling events among
> > +all the eventdev ports and the lcore threads using them.
> > +
> > +Features
> > +
> > +
> > +Queues
> > + * Atomic
> > + * Parallel
> > + * Single-Link
> > +
> > +Ports
> > + * Load balanced (for Atomic, Ordered, Parallel queues)
> > + * Single Link (for single-link queues)
> > +
> > +Configuration and Options
> > +-
> > +
> > +The distributed software eventdev is a vdev device, and as such can be
> > +created from the application code, or from the EAL command line:
> > +
> > +* Call ``rte_vdev_init("event_dsw0")`` from the application
> > +
> > +* Use ``--vdev="event_dsw0"`` in the EAL options, which will call
> > + rte_vdev_init() internally
> > +
> > +Example:
> > +
> > +.. code-block:: console
> > +
> > +./your_eventdev_application --vdev="event_dsw0"
> > +
> > +Limitations
> > +---
> > +
> > +Unattended Ports
> > +
> > +
> > +The distributed software eventdev uses an internal signaling schema
> > +between the ports to achieve load balancing. In order for this to
> > +work, the application must perform enqueue and/or dequeue operations
> > +on all ports.
> > +
> > +Producer-only ports which currently have no events to enqueue should
> > +periodically call rte_event_enqueue_burst() with a zero-sized burst.
> > +
> > +Ports left unattended for longer periods of time will prevent load
> > +balancing, and also cause traffic interruptions on the flows which
> > +are in the process of being migrated.
> > +
> > +Output Buffering
> > +
> > +
> > +For efficiency reasons, the distributed software eventdev might not
> > +send enqueued events immediately to the destination port, but instead
> > +store them in an internal buffer in the source port.
> > +
> > +In case no more events are enqueued on a port with buffered events,
> > +these events will be sent after the application has performed a number
> > +of enqueue and/or dequeue operations.
> > +
> > +For explicit flushing, an application may call
> > +rte_event_enqueue_burst() with a zero-sized burst.
> > +
> > +
> > +Priorities
> > +~~
> > +
> > +The distributed software eventdev does not support event priorities.
> > +
> > +Ordered Queues
> > +~~
> > +
> > +The distributed software eventdev does not support the ordered queue type.
> > +
> > +
> > +"All Types" Queues
> > +~~
> > +
> > +The distributed software eventdev does not support queues of type
> > +RTE_EVENT_QUEUE_CFG_ALL_TYPES, which allow both atomic, ordered, and
> > +parallel events on the same queue.
> > +
> > +Dynamic Link/Unlink
> > +~~~
> > +
> > +The distributed software eventdev does not support calls to
> > +rte_event_port_link() or rte_event_port_unlink() after
> > +rte_event_dev_start() has been called.
> > diff --git a/doc/guides/eventdevs/index.rst b/doc/guides/eventdevs/index.rst
> > index 18ec8e462..984eea5f4
Re: [dpdk-dev] [PATCH v2] build: add PPC64 Meson build
On Mon, Sep 10, 2018 at 12:32:43PM +0100, Luca Boccassi wrote: > This has been only build-tested for now, on a native ppc64el POWER8E > machine running Debian sid. > > Signed-off-by: Luca Boccassi > --- > v2: fix indentation, drop march_opt in ppc build file > LGTM Acked-by: Bruce Richardson
Re: [dpdk-dev] [PATCH] examples/vhost: remove unnecessary MAX_PRINT_BUFF defintion
On 08/25/2018 09:22 PM, Rami Rosen wrote:
This patch removes an unnecessary definition of MAX_PRINT_BUFF
in examples/vhost/main.c, since it is no longer being used.
Fixes: 68363d85857d ("examples/vhost: remove the non-working zero copy code").
Signed-off-by: Rami Rosen
---
examples/vhost/main.c | 3 ---
1 file changed, 3 deletions(-)
diff --git a/examples/vhost/main.c b/examples/vhost/main.c
index 2175c1186..5d28d03f6 100644
--- a/examples/vhost/main.c
+++ b/examples/vhost/main.c
@@ -58,9 +58,6 @@
/* Max number of devices. Limited by vmdq. */
#define MAX_DEVICES 64
-/* Size of buffers used for snprintfs. */
-#define MAX_PRINT_BUFF 6072
-
/* Maximum long option length for option parsing. */
#define MAX_LONG_OPT_SZ 64
Applied to dpdk-next-virtio/master, with changing the title to
pass check-git-log.sh script.
Thanks,
Maxime
Re: [dpdk-dev] [PATCH 2/2] event/dpaa: add select based event support
-Original Message-
> Date: Thu, 30 Aug 2018 11:03:16 +0530
> From: Hemant Agrawal
> To: dev@dpdk.org
> CC: jerin.ja...@caviumnetworks.com, nipun.gu...@nxp.com
> Subject: [PATCH 2/2] event/dpaa: add select based event support
> X-Mailer: git-send-email 2.7.4
>
> External Email
>
> Signed-off-by: Hemant Agrawal
> ---
> config/common_base | 1 +
> config/defconfig_arm64-dpaa-linuxapp-gcc | 1 +
> drivers/event/dpaa/dpaa_eventdev.c | 148
> +++
> drivers/event/dpaa/dpaa_eventdev.h | 8 +-
> 4 files changed, 115 insertions(+), 43 deletions(-)
>
> diff --git a/config/common_base b/config/common_base
> index 4bcbaf9..01a6f17 100644
> --- a/config/common_base
> +++ b/config/common_base
> @@ -199,6 +199,7 @@ CONFIG_RTE_LIBRTE_DPAA_BUS=n
> CONFIG_RTE_LIBRTE_DPAA_MEMPOOL=n
> CONFIG_RTE_LIBRTE_DPAA_PMD=n
> CONFIG_RTE_LIBRTE_DPAA_HWDEBUG=n
> +CONFIG_RTE_LIBRTE_DPAA_EVENT_INTR_MODE=n
> +#ifdef RTE_LIBRTE_DPAA_EVENT_INTR_MODE
Please don't add new compile time options. You can use
devargs to select this mode and have different function
pointer to choose this mode at runtime.
> +static void drain_4_bytes(int fd, fd_set *fdset)
> +{
> + if (FD_ISSET(fd, fdset)) {
> + /* drain 4 bytes */
> + uint32_t junk;
> + ssize_t sjunk = read(qman_thread_fd(), &junk, sizeof(junk));
> + if (sjunk != sizeof(junk))
> + DPAA_EVENTDEV_ERR("UIO irq read error");
> + }
> +}
> +
> +static inline int
> +dpaa_event_dequeue_wait(uint64_t timeout_ticks)
> +{
> + int fd_qman, nfds;
> + int ret;
> + fd_set readset;
> +
> + /* Go into (and back out of) IRQ mode for each select,
> +* it simplifies exit-path considerations and other
> +* potential nastiness.
> +*/
> + struct timeval tv = {
> + .tv_sec = timeout_ticks / 100,
> + .tv_usec = timeout_ticks % 100
> + };
> +
> + fd_qman = qman_thread_fd();
> + nfds = fd_qman + 1;
> + FD_ZERO(&readset);
> + FD_SET(fd_qman, &readset);
> +
> + qman_irqsource_add(QM_PIRQ_DQRI);
> +
> + ret = select(nfds, &readset, NULL, NULL, &tv);
> + if (ret < 0)
> + return ret;
> + /* Calling irqsource_remove() prior to thread_irq()
> +* means thread_irq() will not process whatever caused
> +* the interrupts, however it does ensure that, once
> +* thread_irq() re-enables interrupts, they won't fire
> +* again immediately.
> +*/
> + qman_irqsource_remove(~0);
> + drain_4_bytes(fd_qman, &readset);
> + qman_thread_irq();
> +
> + return ret;
> +}
> +#endif
> +
> static uint16_t
> dpaa_event_dequeue_burst(void *port, struct rte_event ev[],
> uint16_t nb_events, uint64_t timeout_ticks)
> @@ -107,8 +163,8 @@ dpaa_event_dequeue_burst(void *port, struct rte_event
> ev[],
> int ret;
> u16 ch_id;
> void *buffers[8];
> - u32 num_frames, i;
> - uint64_t wait_time, cur_ticks, start_ticks;
> + u32 num_frames, i, irq = 0;
> + uint64_t cur_ticks = 0, wait_time_ticks = 0;
> struct dpaa_port *portal = (struct dpaa_port *)port;
> struct rte_mbuf *mbuf;
>
> @@ -147,20 +203,32 @@ dpaa_event_dequeue_burst(void *port, struct rte_event
> ev[],
> }
> DPAA_PER_LCORE_DQRR_HELD = 0;
>
> - if (portal->timeout == DPAA_EVENT_PORT_DEQUEUE_TIMEOUT_INVALID)
> - wait_time = timeout_ticks;
> + if (timeout_ticks)
> + wait_time_ticks = timeout_ticks;
> else
> - wait_time = portal->timeout;
> + wait_time_ticks = portal->timeout_us;
>
> - /* Lets dequeue the frames */
> - start_ticks = rte_get_timer_cycles();
> - wait_time += start_ticks;
> +#ifndef RTE_LIBRTE_DPAA_EVENT_INTR_MODE
> + wait_time_ticks += rte_get_timer_cycles();
> +#endif
> do {
> + /* Lets dequeue the frames */
> num_frames = qman_portal_dequeue(ev, nb_events, buffers);
> - if (num_frames != 0)
> + if (irq)
> + irq = 0;
> + if (num_frames)
> break;
> +#ifdef RTE_LIBRTE_DPAA_EVENT_INTR_MODE
> + if (wait_time_ticks) { /* wait for time */
> + if (dpaa_event_dequeue_wait(wait_time_ticks) > 0) {
> + irq = 1;
> + continue;
> + }
> + break; /* no event after waiting */
> + }
> +#endif
> cur_ticks = rte_get_timer_cycles();
> - } while (cur_ticks < wait_time);
> + } while (cur_ticks < wait_time_ticks);
>
> return num_frames;
> }
> @@ -184,7 +252,7 @@ dpaa_event_dev_info_get(struct rte_eventdev *d
Re: [dpdk-dev] [PATCH 3/5] event/dpaa2: enchance timeout handling
-Original Message- > Date: Thu, 30 Aug 2018 11:33:57 +0530 > From: Hemant Agrawal > To: dev@dpdk.org > CC: jerin.ja...@caviumnetworks.com, nipun.gu...@nxp.com > Subject: [PATCH 3/5] event/dpaa2: enchance timeout handling > X-Mailer: git-send-email 2.7.4 > Missing git commit log description about timeout handling enhancement.
Re: [dpdk-dev] [PATCH] vhost: fix zmbufs array leak after NUMA realloc
On 08/15/2018 04:54 PM, Ilya Maximets wrote:
'numa_realloc()' allocates 'zmbufs' even if zero copy mode
is not configured. This leads to memory leak, because array
is freed only for zero copy case.
Fixes: 2651726defb7 ("vhost: do deep copy while reallocating queue")
CC: sta...@dpdk.org
Signed-off-by: Ilya Maximets
---
lib/librte_vhost/vhost_user.c | 12 +++-
1 file changed, 7 insertions(+), 5 deletions(-)
diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c
index a2d4c9ffc..9aa1ce118 100644
--- a/lib/librte_vhost/vhost_user.c
+++ b/lib/librte_vhost/vhost_user.c
@@ -357,11 +357,13 @@ numa_realloc(struct virtio_net *dev, int index)
memcpy(vq, old_vq, sizeof(*vq));
TAILQ_INIT(&vq->zmbuf_list);
- new_zmbuf = rte_malloc_socket(NULL, vq->zmbuf_size *
- sizeof(struct zcopy_mbuf), 0, newnode);
- if (new_zmbuf) {
- rte_free(vq->zmbufs);
- vq->zmbufs = new_zmbuf;
+ if (dev->dequeue_zero_copy) {
+ new_zmbuf = rte_malloc_socket(NULL, vq->zmbuf_size *
+ sizeof(struct zcopy_mbuf), 0, newnode);
+ if (new_zmbuf) {
+ rte_free(vq->zmbufs);
+ vq->zmbufs = new_zmbuf;
+ }
}
if (vq_is_packed(dev)) {
Applied to dpdk-next-virtio/master.
Thanks!
Maxime
Re: [dpdk-dev] [PATCH v3] net/virtio-user: check negotiated features before set
On 08/29/2018 05:55 PM, eric zhang wrote: This patch checks negotiated features to see if necessary to offload before set the tap device offload capabilities. It also checks if kernel support the TUNSETOFFLOAD operation. Signed-off-by: eric zhang --- v3: * make other offloading features depend on CSUM * check IFF_VNET_HDR support when handling VHOST_GET_FEATURES --- v2: * don't return failure when failed to set offload to tap * check if offloads available when handling VHOST_GET_FEATURES --- drivers/net/virtio/virtio_user/vhost_kernel.c | 18 +--- drivers/net/virtio/virtio_user/vhost_kernel_tap.c | 56 +-- drivers/net/virtio/virtio_user/vhost_kernel_tap.h | 2 +- 3 files changed, 54 insertions(+), 22 deletions(-) Applied to dpdk-next-virtio/master, adding Fixes: line as suggested by Tiwei. Thanks, Maxime
Re: [dpdk-dev] [PATCH v2 1/2] test/event: fix eth Rx adapter autotest for skeleton PMD
-Original Message-
> Date: Wed, 5 Sep 2018 19:21:05 +0530
> From: Nikhil Rao
> To: jerin.ja...@caviumnetworks.com
> CC: dev@dpdk.org, Nikhil Rao ,
> vipin.vargh...@intel.com, sta...@dpdk.org
> Subject: [PATCH v2 1/2] test/event: fix eth Rx adapter autotest for
> skeleton PMD
> X-Mailer: git-send-email 1.8.3.1
>
> External Email
>
> skeleton PMD does not support
> RTE_EVENT_ETH_RX_ADAPTER_CAP_MULTI_EVENTQ and
> implicit_release_disable so make the Rx queue_id = -1 and
> initialize the event port configuration to zero.
>
> Fixes: ec36d881f56d ("eventdev: add implicit release disable capability")
> Fixes: 2a9c83ae3b2e ("test/eventdev: add multi-ports test")
> Cc: vipin.vargh...@intel.com
> CC: sta...@dpdk.org
> Signed-off-by: Nikhil Rao
> ---
>
> v2:
> * add include patch to fix Rx adapter intr autotest as it is
> dependent on the first patch.
>
> test/test/test_event_eth_rx_adapter.c | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/test/test/test_event_eth_rx_adapter.c
> b/test/test/test_event_eth_rx_adapter.c
> index 2337e54..29d0ff5 100644
> --- a/test/test/test_event_eth_rx_adapter.c
> +++ b/test/test/test_event_eth_rx_adapter.c
> @@ -317,7 +317,7 @@ struct event_eth_rx_adapter_test_params {
> {
> int err;
> struct rte_event_dev_info dev_info;
> - struct rte_event_port_conf rx_p_conf;
> + struct rte_event_port_conf rx_p_conf = {0};
Some old compiler does not support this scheme. Can you do memset to
zero to make more portable.
>
> err = rte_event_dev_info_get(TEST_DEV_ID, &dev_info);
> TEST_ASSERT(err == 0, "Expected 0 got %d", err);
> @@ -503,7 +503,7 @@ struct event_eth_rx_adapter_test_params {
> port_index = 0;
> for (; port_index < rte_eth_dev_count_total(); port_index += 1) {
> err = rte_event_eth_rx_adapter_queue_add(TEST_INST_ID,
> - port_index, 0,
> + port_index, -1,
> &queue_config);
> TEST_ASSERT(err == 0, "Expected 0 got %d", err);
> }
> @@ -512,7 +512,7 @@ struct event_eth_rx_adapter_test_params {
> port_index = 0;
> for (; port_index < rte_eth_dev_count_total(); port_index += 1) {
> err = rte_event_eth_rx_adapter_queue_del(TEST_INST_ID,
> - port_index, 0);
> + port_index, -1);
> TEST_ASSERT(err == 0, "Expected 0 got %d", err);
> }
>
> --
> 1.8.3.1
>
[dpdk-dev] [RFC v2 0/2] ethdev: claim device reset as async
Device reset may have the dependency, for example, a VF reset expects PF ready, or a NIC function as a part of a SOC need to wait for other parts of the system be ready, these are time-consuming tasks and will block current thread. So we claimed rte_eth_dev_reset as an async API, that makes things easy for an application that what to reset the device from the interrupt thread since typically a RTE_ETH_EVENT_INTR_RESET handler is invoked in interrupt thread. rte_eth_dev_reset will spawn a new thread to call ops->dev_reset, once it is finished, it will raise the RTE_ETH_EVENT_RESET_COMPLETE event to notify the application. Application should not assume device reset is finished after rte_eth_dev_reset return, it should always wait for a RTE_ETH_EVENT_RESET_COMPLETE event and check the reset result. v2: - rte_eth_dev_reset will spawn a thread. Qi Zhang (2): ethdev: claim device reset as async testpmd: enable async device reset app/test-pmd/testpmd.c | 50 +- lib/librte_ethdev/rte_ethdev.c | 48 ++-- lib/librte_ethdev/rte_ethdev.h | 48 3 files changed, 124 insertions(+), 22 deletions(-) -- 2.13.6
[dpdk-dev] [PATCH v6] linuxapp, eal: Fix the memory leak issue of logid
From: Ziye Yang
This patch is used to fix the memory leak issue of logid.
We use the ASAN test in SPDK when intergrating DPDK and
find this memory leak issue.
By the way, we also fix several missed function call of
rte_atomic32_clear.
Signed-off-by: Ziye Yang
---
lib/librte_eal/linuxapp/eal/eal.c | 11 +++
1 file changed, 7 insertions(+), 4 deletions(-)
diff --git a/lib/librte_eal/linuxapp/eal/eal.c
b/lib/librte_eal/linuxapp/eal/eal.c
index e59ac65..a1b218a 100644
--- a/lib/librte_eal/linuxapp/eal/eal.c
+++ b/lib/librte_eal/linuxapp/eal/eal.c
@@ -793,7 +793,8 @@ static void rte_eal_init_alert(const char *msg)
int i, fctret, ret;
pthread_t thread_id;
static rte_atomic32_t run_once = RTE_ATOMIC32_INIT(0);
- const char *logid;
+ const char *p;
+ static char logid[PATH_MAX];
char cpuset[RTE_CPU_AFFINITY_STR_LEN];
char thread_name[RTE_MAX_THREAD_NAME_LEN];
@@ -810,9 +811,8 @@ static void rte_eal_init_alert(const char *msg)
return -1;
}
- logid = strrchr(argv[0], '/');
- logid = strdup(logid ? logid + 1: argv[0]);
-
+ p = strrchr(argv[0], '/');
+ snprintf(logid, sizeof(logid), "%s", (p ? p + 1: argv[0]));
thread_id = pthread_self();
eal_reset_internal_config(&internal_config);
@@ -823,6 +823,7 @@ static void rte_eal_init_alert(const char *msg)
if (rte_eal_cpu_init() < 0) {
rte_eal_init_alert("Cannot detect lcores.");
rte_errno = ENOTSUP;
+ rte_atomic32_clear(&run_once);
return -1;
}
@@ -851,6 +852,7 @@ static void rte_eal_init_alert(const char *msg)
if (rte_eal_intr_init() < 0) {
rte_eal_init_alert("Cannot init interrupt-handling thread\n");
+ rte_atomic32_clear(&run_once);
return -1;
}
@@ -861,6 +863,7 @@ static void rte_eal_init_alert(const char *msg)
rte_eal_init_alert("failed to init mp channel\n");
if (rte_eal_process_type() == RTE_PROC_PRIMARY) {
rte_errno = EFAULT;
+ rte_atomic32_clear(&run_once);
return -1;
}
}
--
1.9.3
Re: [dpdk-dev] [PATCH] vhost: fix vhost interrupt support
On 09/05/2018 01:55 AM, Tiwei Bie wrote:
When VIRTIO_RING_F_EVENT_IDX is negotiated, we need to
update the avail event to enable the notification.
Fixes: 3f8ff12821e4 ("vhost: support interrupt mode")
Cc:sta...@dpdk.org
Signed-off-by: Tiwei Bie
---
lib/librte_vhost/Makefile | 1 +
lib/librte_vhost/vhost.c | 18 --
lib/librte_vhost/vhost.h | 2 ++
3 files changed, 15 insertions(+), 6 deletions(-)
Applied to dpdk-next-virtio/master.
Thanks!
Maxime
[dpdk-dev] [RFC v2 0/2] ethdev: claim device reset as async
Device reset may have the dependency, for example, a VF reset expects PF ready, or a NIC function as a part of a SOC need to wait for other parts of the system be ready, these are time-consuming tasks and will block current thread. So we claimed rte_eth_dev_reset as an async API, that makes things easy for an application that what to reset the device from the interrupt thread since typically a RTE_ETH_EVENT_INTR_RESET handler is invoked in interrupt thread. rte_eth_dev_reset will spawn a new thread to call ops->dev_reset, once it is finished, it will raise the RTE_ETH_EVENT_RESET_COMPLETE event to notify the application. Application should not assume device reset is finished after rte_eth_dev_reset return, it should always wait for a RTE_ETH_EVENT_RESET_COMPLETE event and check the reset result. v2: - rte_eth_dev_reset will spawn a thread. Qi Zhang (2): ethdev: claim device reset as async testpmd: enable async device reset app/test-pmd/testpmd.c | 50 +- lib/librte_ethdev/rte_ethdev.c | 48 ++-- lib/librte_ethdev/rte_ethdev.h | 48 3 files changed, 124 insertions(+), 22 deletions(-) -- 2.13.6
[dpdk-dev] [RFC v2 2/2] testpmd: enable async device reset
rte_eth_dev_reset is claimed as an async API, so re-work
on the device reset handling.
Signed-off-by: Qi Zhang
---
app/test-pmd/testpmd.c | 50 +-
1 file changed, 49 insertions(+), 1 deletion(-)
diff --git a/app/test-pmd/testpmd.c b/app/test-pmd/testpmd.c
index ee48db2a3..8d4d7b4f9 100644
--- a/app/test-pmd/testpmd.c
+++ b/app/test-pmd/testpmd.c
@@ -1918,6 +1918,54 @@ close_port(portid_t pid)
printf("Done\n");
}
+static int reset_ret;
+static int reset_done;
+
+static int
+on_reset_complete(__rte_unused uint16_t port_id,
+ __rte_unused enum rte_eth_event_type event,
+ __rte_unused void *cb_arg,
+ void *ret_param)
+{
+ RTE_ASSERT(event == RTE_ETH_EVENT_RESET_COMPLETE);
+
+ reset_ret = *(int *)ret_param;
+ reset_done = 1;
+ return 0;
+}
+
+static int
+do_dev_reset_sync(portid_t pid)
+{
+ int ret;
+
+ ret = rte_eth_dev_callback_register(pid,
+ RTE_ETH_EVENT_RESET_COMPLETE,
+ on_reset_complete, NULL);
+
+ if (ret) {
+ printf("Fail to reigster callback function\n");
+ return ret;
+ }
+
+ reset_done = 0;
+ ret = rte_eth_dev_reset(pid);
+ if (ret)
+ goto finish;
+
+ while (!reset_done)
+ rte_delay_ms(10);
+
+ ret = reset_ret;
+
+finish:
+ rte_eth_dev_callback_unregister(pid,
+ RTE_ETH_EVENT_RESET_COMPLETE,
+ on_reset_complete, NULL);
+
+ return ret;
+}
+
void
reset_port(portid_t pid)
{
@@ -1946,7 +1994,7 @@ reset_port(portid_t pid)
continue;
}
- diag = rte_eth_dev_reset(pi);
+ diag = do_dev_reset_sync(pi);
if (diag == 0) {
port = &ports[pi];
port->need_reconfig = 1;
--
2.13.6
[dpdk-dev] [RFC v2 1/2] ethdev: claim device reset as async
rte_eth_dev_reset should be implemented in an async way since it is
possible to be invoked in interrupt thread and sometimes to reset a
device need to wait for some dependency, for example, a VF expects for
PF ready or a NIC function as part of a SOC wait for the whole system
reset complete, and all these time-consuming tasks will block the
interrupt thread.
The patch claims rte_eth_dev_reset is an async function it will spawn
a new thread which will call ops->dev_reset, and when finished it will
raise the event RTE_ETH_EVENT_RESET_COMPLETE. The application should
always wait for this event before it continues to configure and restart
the device.
Signed-off-by: Qi Zhang
---
lib/librte_ethdev/rte_ethdev.c | 48 --
lib/librte_ethdev/rte_ethdev.h | 48 ++
2 files changed, 76 insertions(+), 20 deletions(-)
diff --git a/lib/librte_ethdev/rte_ethdev.c b/lib/librte_ethdev/rte_ethdev.c
index 4c3202505..4c07dabcf 100644
--- a/lib/librte_ethdev/rte_ethdev.c
+++ b/lib/librte_ethdev/rte_ethdev.c
@@ -1374,10 +1374,38 @@ rte_eth_dev_close(uint16_t port_id)
dev->data->tx_queues = NULL;
}
+struct dev_reset_args {
+ struct rte_eth_dev *dev;
+ enum rte_eth_dev_state pre_state;
+};
+
+static void *
+do_dev_reset(void *args)
+{
+ struct dev_reset_args *reset_args = args;
+ struct rte_eth_dev *dev = reset_args->dev;
+ int ret;
+
+ ret = dev->dev_ops->dev_reset(dev);
+
+ /*restore previous device state */
+ dev->state = reset_args->pre_state;
+
+ /* notify users */
+ _rte_eth_dev_callback_process(dev, RTE_ETH_EVENT_RESET_COMPLETE,
+ &ret);
+
+ free(args);
+ return NULL;
+}
+
int
rte_eth_dev_reset(uint16_t port_id)
{
struct rte_eth_dev *dev;
+ struct dev_reset_args *args;
+ char pthread_name[20];
+ pthread_t tid;
int ret;
RTE_ETH_VALID_PORTID_OR_ERR_RET(port_id, -EINVAL);
@@ -1385,10 +1413,26 @@ rte_eth_dev_reset(uint16_t port_id)
RTE_FUNC_PTR_OR_ERR_RET(*dev->dev_ops->dev_reset, -ENOTSUP);
+ /* already on resetting */
+ if (dev->state == RTE_ETH_DEV_RESETTING)
+ return 0;
+
+ args = calloc(1, sizeof(struct dev_reset_args));
+ if (!args)
+ return -ENOMEM;
+
rte_eth_dev_stop(port_id);
- ret = dev->dev_ops->dev_reset(dev);
- return eth_err(port_id, ret);
+ /* store previous device state temporary */
+ args->pre_state = dev->state;
+
+ dev->state = RTE_ETH_DEV_RESETTING;
+ snprintf(pthread_name, sizeof(pthread_name),
+ "do_dev_reset_%d", port_id);
+ ret = rte_ctrl_thread_create(&tid, pthread_name, NULL,
+ do_dev_reset, args);
+
+ return ret;
}
int __rte_experimental
diff --git a/lib/librte_ethdev/rte_ethdev.h b/lib/librte_ethdev/rte_ethdev.h
index 7070e9ab4..0739d9fe7 100644
--- a/lib/librte_ethdev/rte_ethdev.h
+++ b/lib/librte_ethdev/rte_ethdev.h
@@ -1239,6 +1239,8 @@ enum rte_eth_dev_state {
RTE_ETH_DEV_DEFERRED,
/** Device is in removed state when plug-out is detected. */
RTE_ETH_DEV_REMOVED,
+ /** Device is on resetting when rte_eth_dev_reset is called. */
+ RTE_ETH_DEV_RESETTING,
};
struct rte_eth_dev_sriov {
@@ -1814,21 +1816,31 @@ void rte_eth_dev_close(uint16_t port_id);
* RTE_ETH_EVENT_INTR_RESET event is detected, but can also use it to start
* a port reset in other circumstances.
*
- * When this function is called, it first stops the port and then calls the
- * PMD specific dev_uninit( ) and dev_init( ) to return the port to initial
- * state, in which no Tx and Rx queues are setup, as if the port has been
- * reset and not started. The port keeps the port id it had before the
- * function call.
- *
- * After calling rte_eth_dev_reset( ), the application should use
- * rte_eth_dev_configure( ), rte_eth_rx_queue_setup( ),
- * rte_eth_tx_queue_setup( ), and rte_eth_dev_start( )
- * to reconfigure the device as appropriate.
- *
- * Note: To avoid unexpected behavior, the application should stop calling
- * Tx and Rx functions before calling rte_eth_dev_reset( ). For thread
- * safety, all these controlling functions should be called from the same
- * thread.
+ * @note
+ * Device reset may have the dependency, for example, a VF reset expects
+ * PF ready, or a NIC function as a part of a SOC need to wait for other
+ * parts of the system be ready, these are time-consuming tasks and will
+ * block current thread.
+ *
+ * So we claimed rte_eth_dev_reset as an async API, that makes things easy
+ * for an application that what to reset the device from the interrupt
+ * thread since typically a RTE_ETH_EVENT_INTR_RESET handler is invoked in
+ * interrupt thread.
+ *
+ * rte_eth_dev_reset will spawn a new thread to call ops->dev_reset, once it
+ * is finis
Re: [dpdk-dev] [PATCH v2] vhost-user: drop connection on message handling failures
On 09/03/2018 12:12 PM, Ilya Maximets wrote:
There are a lot of cases where vhost-user massage handling
could fail and end up in a fully not recoverable state. For
example, allocation failures of shadow used ring and batched
copy array are not recoverable and leads to the segmentation
faults like this on the receiving/transmission path:
Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x7f913fecf0 (LWP 43625)]
in copy_desc_to_mbuf () at /lib/librte_vhost/virtio_net.c:760
760 batch_copy[vq->batch_copy_nb_elems].dst =
This could be easily reproduced in case of low memory or big
number of vhost-user ports.
Fix that by propagating error to the upper layer which will
end up with disconnection in case we can not report to
the message sender when the error happens.
Fixes: f689586bc060 ("vhost: shadow used ring update")
Cc:sta...@dpdk.org
Signed-off-by: Ilya Maximets
---
v2:
* Patch changed to cover most of possible failures at once. [Tiwei Bie]
lib/librte_vhost/vhost_user.c | 51 +--
1 file changed, 31 insertions(+), 20 deletions(-)
Reviewed-by: Maxime Coquelin
Thanks!
Maxime
[dpdk-dev] [PATCH] doc: update API deprecation for device reset
Device reset may have the dependency, for example, a VF reset expects PF ready, or a NIC function as a part of a SOC need to wait for other parts of the system be ready, these are time-consuming tasks and will block current thread. So we claimed rte_eth_dev_reset as an async API, that makes things easy for an application that what to reset the device from the interrupt thread since typically a RTE_ETH_EVENT_INTR_RESET handler is invoked in interrupt thread. RFC patch: http://patchwork.dpdk.org/patch/44513/ Signed-off-by: Qi Zhang --- doc/guides/rel_notes/deprecation.rst | 5 + 1 file changed, 5 insertions(+) diff --git a/doc/guides/rel_notes/deprecation.rst b/doc/guides/rel_notes/deprecation.rst index bade1e4c4..3490aac87 100644 --- a/doc/guides/rel_notes/deprecation.rst +++ b/doc/guides/rel_notes/deprecation.rst @@ -95,3 +95,8 @@ Deprecation Notices - ``rte_pdump_set_socket_dir`` will be removed; - The parameter, ``path``, of ``rte_pdump_init`` will be removed; - The enum ``rte_pdump_socktype`` will be removed. + +* ethdev: In v19.02 ``rte_eth_dev_reset`` is claimed as an async API. + Application should not assume device reset is finished after + ``rte_eth_dev_reset`` return, it should always wait for a + RTE_ETH_EVENT_RESET_COMPLETE event and check the reset result. -- 2.13.6
[dpdk-dev] [PATCH] crypto action
Change-Id: If1aecaac3335685f3bef79f47768ba8ae7765fbb
Signed-off-by: Zhang, Roy Fan
---
examples/ip_pipeline/Makefile | 1 +
examples/ip_pipeline/action.c | 11 +
examples/ip_pipeline/action.h | 1 +
examples/ip_pipeline/cli.c| 744 +-
examples/ip_pipeline/examples/flow_crypto.cli | 76 +++
examples/ip_pipeline/main.c | 17 +
examples/ip_pipeline/mempool.c| 2 +-
examples/ip_pipeline/mempool.h| 1 +
examples/ip_pipeline/pipeline.c | 61 +++
examples/ip_pipeline/pipeline.h | 13 +
examples/ip_pipeline/sym_crypto.c | 320 +++
examples/ip_pipeline/sym_crypto.h | 104
examples/ip_pipeline/thread.c | 10 +
lib/librte_pipeline/rte_table_action.c| 183 ---
lib/librte_pipeline/rte_table_action.h| 28 +-
15 files changed, 1472 insertions(+), 100 deletions(-)
create mode 100644 examples/ip_pipeline/examples/flow_crypto.cli
create mode 100644 examples/ip_pipeline/sym_crypto.c
create mode 100644 examples/ip_pipeline/sym_crypto.h
diff --git a/examples/ip_pipeline/Makefile b/examples/ip_pipeline/Makefile
index 3fb98ce3e..819625632 100644
--- a/examples/ip_pipeline/Makefile
+++ b/examples/ip_pipeline/Makefile
@@ -18,6 +18,7 @@ SRCS-y += swq.c
SRCS-y += tap.c
SRCS-y += thread.c
SRCS-y += tmgr.c
+SRCS-y += sym_crypto.c
# Build using pkg-config variables if possible
$(shell pkg-config --exists libdpdk)
diff --git a/examples/ip_pipeline/action.c b/examples/ip_pipeline/action.c
index a29c2b368..d97423568 100644
--- a/examples/ip_pipeline/action.c
+++ b/examples/ip_pipeline/action.c
@@ -333,6 +333,17 @@ table_action_profile_create(const char *name,
}
}
+ if (params->action_mask & (1LLU << RTE_TABLE_ACTION_SYM_CRYPTO)) {
+ status = rte_table_action_profile_action_register(ap,
+ RTE_TABLE_ACTION_SYM_CRYPTO,
+ ¶ms->sym_crypto);
+
+ if (status) {
+ rte_table_action_profile_free(ap);
+ return NULL;
+ }
+ }
+
status = rte_table_action_profile_freeze(ap);
if (status) {
rte_table_action_profile_free(ap);
diff --git a/examples/ip_pipeline/action.h b/examples/ip_pipeline/action.h
index 417200e86..cde17e69a 100644
--- a/examples/ip_pipeline/action.h
+++ b/examples/ip_pipeline/action.h
@@ -53,6 +53,7 @@ struct table_action_profile_params {
struct rte_table_action_nat_config nat;
struct rte_table_action_ttl_config ttl;
struct rte_table_action_stats_config stats;
+ struct rte_table_action_sym_crypto_config sym_crypto;
};
struct table_action_profile {
diff --git a/examples/ip_pipeline/cli.c b/examples/ip_pipeline/cli.c
index 102a1d6b7..c73403378 100644
--- a/examples/ip_pipeline/cli.c
+++ b/examples/ip_pipeline/cli.c
@@ -17,6 +17,7 @@
#include "mempool.h"
#include "parser.h"
#include "pipeline.h"
+#include "sym_crypto.h"
#include "swq.h"
#include "tap.h"
#include "thread.h"
@@ -66,7 +67,7 @@ cmd_mempool(char **tokens,
char *name;
struct mempool *mempool;
- if (n_tokens != 10) {
+ if (n_tokens != 10 && n_tokens != 12) {
snprintf(out, out_size, MSG_ARG_MISMATCH, tokens[0]);
return;
}
@@ -113,6 +114,18 @@ cmd_mempool(char **tokens,
return;
}
+ if (n_tokens == 12) {
+ if (strcmp(tokens[10], "priv") != 0) {
+ snprintf(out, out_size, MSG_ARG_INVALID, "priv");
+ return;
+ }
+
+ if (parser_read_uint32(&p.priv_size, tokens[11]) != 0) {
+ snprintf(out, out_size, MSG_ARG_INVALID, "priv");
+ return;
+ }
+ }
+
mempool = mempool_create(name, &p);
if (mempool == NULL) {
snprintf(out, out_size, MSG_CMD_FAIL, tokens[0]);
@@ -785,6 +798,569 @@ cmd_kni(char **tokens,
}
}
+static const char cmd_sym_crypto_help[] =
+"sym_crypto \n"
+" cryptodev | cryptodev_id \n"
+" q \n"
+" offset \n"
+" cpu \n";
+
+static void
+cmd_sym_crypto(char **tokens,
+ uint32_t n_tokens,
+ char *out,
+ size_t out_size)
+{
+ struct sym_crypto_params params;
+ char *name;
+
+ memset(¶ms, 0, sizeof(params));
+ if (n_tokens != 11) {
+ snprintf(out, out_size, MSG_ARG_MISMATCH, tokens[0]);
+ return;
+ }
+
+ name = tokens[1];
+
+ if (strcmp(tokens[2], "cryptodev") == 0)
+ params.dev_name = tokens[3];
+ else if (strcmp(tokens[2], "cryptodev_id") == 0) {
+ if (parser_read_uint32(¶ms.cryptodev_id, tokens[3]) < 0) {
+ snprintf(out, out_size, MSG_ARG_INVALID
Re: [dpdk-dev] [PATCH v2 1/5] vhost: unify VhostUserMsg usage
Hi Nikolay, On 07/19/2018 09:13 PM, Nikolay Nikolaev wrote: Use the typedef version of struct VhostUserMsg. Also unify the related parameter name. Signed-off-by: Nikolay Nikolaev --- lib/librte_vhost/vhost_user.c | 50 + 1 file changed, 25 insertions(+), 25 deletions(-) Reviewed-by: Maxime Coquelin Thanks, Maxime
Re: [dpdk-dev] [PATCH] net/ixgbe: fix busy polling while fiber link update
On 04.09.2018 09:08, Zhang, Qi Z wrote:
> Hi Ilya:
>
>> -Original Message-
>> From: dev [mailto:dev-boun...@dpdk.org] On Behalf Of Ilya Maximets
>> Sent: Friday, August 31, 2018 8:40 PM
>> To: dev@dpdk.org
>> Cc: Lu, Wenzhuo ; Ananyev, Konstantin
>> ; Laurent Hardy
>> ; Dai, Wei ; Ilya Maximets
>> ; sta...@dpdk.org
>> Subject: [dpdk-dev] [PATCH] net/ixgbe: fix busy polling while fiber link
>> update
>>
>> If the multispeed fiber link is in DOWN state, ixgbe_setup_link could take
>> around a second of busy polling. This is highly inconvenient for the case
>> where
>> single thread periodically checks the link statuses. For example, OVS main
>> thread periodically updates the link statuses and hangs for a really long
>> time
>> busy waiting on ixgbe_setup_link() for a DOWN fiber ports. For case with 3
>> down ports it hangs for a 3 seconds and unable to do anything including
>> packet processing.
>> Fix that by shifting that workaround to a separate thread by alarm handler
>> that
>> will try to set up link if it is DOWN.
>
> Does that mean we will block the interrupt thread for 3 seconds?
Three times for one second. Other work could be scheduled between.
IMHO, it's much better than blocking usual caller for 3 seconds.
> Also, can we guarantee there will not be any race condition if we call
> ixgbe_setup_link at another thread, the base code API is not assumed to be
> thread-safe as I know.
The only user of 'ixgbe_setup_link' is 'ixgbe_dev_start', but it could be
called only if device stopped. 'ixgbe_dev_stop' cancels the alarm.
Race with 'link_update' avoided by 'IXGBE_FLAG_NEED_LINK_CONFIG' flag.
>
> Regards
> Qi
>
>>
>> Fixes: c12d22f65b13 ("net/ixgbe: ensure link status is updated")
>> CC: sta...@dpdk.org
>>
>> Signed-off-by: Ilya Maximets
>> ---
>> drivers/net/ixgbe/ixgbe_ethdev.c | 43
>> 1 file changed, 32 insertions(+), 11 deletions(-)
>>
>> diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c
>> b/drivers/net/ixgbe/ixgbe_ethdev.c
>> index 26b192737..a33b9a6e8 100644
>> --- a/drivers/net/ixgbe/ixgbe_ethdev.c
>> +++ b/drivers/net/ixgbe/ixgbe_ethdev.c
>> @@ -221,6 +221,8 @@ static int ixgbe_dev_interrupt_action(struct
>> rte_eth_dev *dev,
>>struct rte_intr_handle *handle); static
>> void
>> ixgbe_dev_interrupt_handler(void *param); static void
>> ixgbe_dev_interrupt_delayed_handler(void *param);
>> +static void ixgbe_dev_setup_link_alarm_handler(void *param);
>> +
>> static int ixgbe_add_rar(struct rte_eth_dev *dev, struct ether_addr
>> *mac_addr,
>> uint32_t index, uint32_t pool);
>> static void ixgbe_remove_rar(struct rte_eth_dev *dev, uint32_t index); @@
>> -2791,6 +2793,8 @@ ixgbe_dev_stop(struct rte_eth_dev *dev)
>>
>> PMD_INIT_FUNC_TRACE();
>>
>> +rte_eal_alarm_cancel(ixgbe_dev_setup_link_alarm_handler, dev);
>> +
>> /* disable interrupts */
>> ixgbe_disable_intr(hw);
>>
>> @@ -3969,6 +3973,25 @@ ixgbevf_check_link(struct ixgbe_hw *hw,
>> ixgbe_link_speed *speed,
>> return ret_val;
>> }
>>
>> +static void
>> +ixgbe_dev_setup_link_alarm_handler(void *param) {
>> +struct rte_eth_dev *dev = (struct rte_eth_dev *)param;
>> +struct ixgbe_hw *hw =
>> IXGBE_DEV_PRIVATE_TO_HW(dev->data->dev_private);
>> +struct ixgbe_interrupt *intr =
>> +IXGBE_DEV_PRIVATE_TO_INTR(dev->data->dev_private);
>> +u32 speed;
>> +bool autoneg = false;
>> +
>> +speed = hw->phy.autoneg_advertised;
>> +if (!speed)
>> +ixgbe_get_link_capabilities(hw, &speed, &autoneg);
>> +
>> +ixgbe_setup_link(hw, speed, true);
>> +
>> +intr->flags &= ~IXGBE_FLAG_NEED_LINK_CONFIG; }
>> +
>> /* return 0 means link status changed, -1 means not changed */ int
>> ixgbe_dev_link_update_share(struct rte_eth_dev *dev, @@ -3981,9 +4004,7
>> @@ ixgbe_dev_link_update_share(struct rte_eth_dev *dev,
>> IXGBE_DEV_PRIVATE_TO_INTR(dev->data->dev_private);
>> int link_up;
>> int diag;
>> -u32 speed = 0;
>> int wait = 1;
>> -bool autoneg = false;
>>
>> memset(&link, 0, sizeof(link));
>> link.link_status = ETH_LINK_DOWN;
>> @@ -3993,13 +4014,8 @@ ixgbe_dev_link_update_share(struct rte_eth_dev
>> *dev,
>>
>> hw->mac.get_link_status = true;
>>
>> -if ((intr->flags & IXGBE_FLAG_NEED_LINK_CONFIG) &&
>> -ixgbe_get_media_type(hw) == ixgbe_media_type_fiber) {
>> -speed = hw->phy.autoneg_advertised;
>> -if (!speed)
>> -ixgbe_get_link_capabilities(hw, &speed, &autoneg);
>> -ixgbe_setup_link(hw, speed, true);
>> -}
>> +if (intr->flags & IXGBE_FLAG_NEED_LINK_CONFIG)
>> +return rte_eth_linkstatus_set(dev, &link);
>>
>> /* check if it needs to wait to complete, if lsc interrupt is enabled */
>> if (wait_to_complete == 0 || dev->data->dev_conf.intr_conf.lsc != 0) @@
>> -4017,11 +4033,14 @@ ixgbe_dev_link_update_share(struct rte_eth
Re: [dpdk-dev] [PATCH v2 2/5] vhost: make message handling functions prepare the reply
On 07/19/2018 09:13 PM, Nikolay Nikolaev wrote: As VhostUserMsg structure is resued to generate the reply, move the s/resued/reused/ relevant fields update into the respective message handling functions. Signed-off-by: Nikolay Nikolaev --- lib/librte_vhost/vhost_user.c | 26 -- 1 file changed, 16 insertions(+), 10 deletions(-) Agree this is cleaner: Reviewed-by: Maxime Coquelin For the typo, don't resend the series just for this if this is the only thing to fix. I suspect a rebase may be required anyway as it may conflict with a patch from Ilya. Thanks, Maxime
Re: [dpdk-dev] [PATCH v1 1/2] eal: add nanosleep based delay function
On 05.09.2018 16:10, Stephen Hemminger wrote:
> On Mon, 3 Sep 2018 17:47:42 +0300
> Ilya Maximets wrote:
>
>>
>> +void __rte_experimental
>> +rte_delay_us_sleep(unsigned int us)
>> +{
>> +struct timespec wait[2];
>> +int ind = 0;
>> +
>> +wait[0].tv_sec = 0;
>> +if (us >= US_PER_S) {
>> +wait[0].tv_sec = us / US_PER_S;
>> +us -= wait[0].tv_sec * US_PER_S;
>> +}
>> +wait[0].tv_nsec = 1000 * us;
>> +
>> +while (nanosleep(&wait[ind], &wait[1 - ind]) == EINTR)
>> +ind = 1 - ind;
>> +}
>> +
>
> This seems like a complex/tricky way to handle the case where
> nanosleep is interrupted. It needs a comment.
OK. I can add some comment here.
Best regards, Ilya Maximets.
Re: [dpdk-dev] [PATCH v2 3/5] vhost: handle unsupported message types in functions
On 07/19/2018 09:13 PM, Nikolay Nikolaev wrote: Add new functions to handle the unsupported vhost message types: - vhost_user_set_vring_err - vhost_user_set_log_fd Signed-off-by: Nikolay Nikolaev --- lib/librte_vhost/vhost_user.c | 22 +- 1 file changed, 17 insertions(+), 5 deletions(-) Reviewed-by: Maxime Coquelin Thanks, Maxime
Re: [dpdk-dev] [PATCH] doc: update API deprecation for device reset
On 09/10/2018 03:04 PM, Qi Zhang wrote: > Device reset may have the dependency, for example, a VF reset expects > PF ready, or a NIC function as a part of a SOC need to wait for other > parts of the system be ready, these are time-consuming tasks and will > block current thread. > > So we claimed rte_eth_dev_reset as an async API, that makes things > easy for an application that what to reset the device from the interrupt > thread since typically a RTE_ETH_EVENT_INTR_RESET handler is invoked > in interrupt thread. > I think it may be better to give the function with new async behaviour a new name. Otherwise, it could be confusing for users who are currently taking some action based on the return value from the current blocking version of the function. What do you think? > RFC patch: > http://patchwork.dpdk.org/patch/44513/ > > Signed-off-by: Qi Zhang > --- > doc/guides/rel_notes/deprecation.rst | 5 + > 1 file changed, 5 insertions(+) > > diff --git a/doc/guides/rel_notes/deprecation.rst > b/doc/guides/rel_notes/deprecation.rst > index bade1e4c4..3490aac87 100644 > --- a/doc/guides/rel_notes/deprecation.rst > +++ b/doc/guides/rel_notes/deprecation.rst > @@ -95,3 +95,8 @@ Deprecation Notices >- ``rte_pdump_set_socket_dir`` will be removed; >- The parameter, ``path``, of ``rte_pdump_init`` will be removed; >- The enum ``rte_pdump_socktype`` will be removed. > + > +* ethdev: In v19.02 ``rte_eth_dev_reset`` is claimed as an async API. > + Application should not assume device reset is finished after > + ``rte_eth_dev_reset`` return, it should always wait for a > + RTE_ETH_EVENT_RESET_COMPLETE event and check the reset result. >
Re: [dpdk-dev] [PATCH v2 4/5] vhost: unify message handling function signature
On 07/19/2018 09:13 PM, Nikolay Nikolaev wrote:
Each vhost-user message handling function will return an int result
which is described in the new enum vh_result: error, OK and reply.
All functions will now have two arguments, virtio_net double pointer
and VhostUserMsg pointer.
Signed-off-by: Nikolay Nikolaev
---
lib/librte_vhost/vhost_user.c | 219 -
1 file changed, 130 insertions(+), 89 deletions(-)
Adding Ilya in cc, who has submitted a patch that seems to be
complementary with yours: http://patches.dpdk.org/patch/44168/
Other than that, it looks good to me:
Reviewed-by: Maxime Coquelin
Thanks,
Maxime
diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c
index b408460f0..6b39d1e30 100644
--- a/lib/librte_vhost/vhost_user.c
+++ b/lib/librte_vhost/vhost_user.c
@@ -71,6 +71,16 @@ static const char *vhost_message_str[VHOST_USER_MAX] = {
[VHOST_USER_CRYPTO_CLOSE_SESS] = "VHOST_USER_CRYPTO_CLOSE_SESS",
};
+/* The possible results of a message handling function */
+enum vh_result {
+ /* Message handling failed */
+ VH_RESULT_ERR = -1,
+ /* Message handling successful */
+ VH_RESULT_OK= 0,
+ /* Message handling successful and reply prepared */
+ VH_RESULT_REPLY = 1,
+};
+
static uint64_t
get_blk_size(int fd)
{
@@ -127,27 +137,31 @@ vhost_backend_cleanup(struct virtio_net *dev)
* the device hasn't been initialised.
*/
static int
-vhost_user_set_owner(void)
+vhost_user_set_owner(struct virtio_net **pdev __rte_unused,
+ VhostUserMsg * msg __rte_unused)
{
- return 0;
+ return VH_RESULT_OK;
}
static int
-vhost_user_reset_owner(struct virtio_net *dev)
+vhost_user_reset_owner(struct virtio_net **pdev,
+ VhostUserMsg * msg __rte_unused)
{
+ struct virtio_net *dev = *pdev;
vhost_destroy_device_notify(dev);
cleanup_device(dev, 0);
reset_device(dev);
- return 0;
+ return VH_RESULT_OK;
}
/*
* The features that we support are requested.
*/
-static uint64_t
-vhost_user_get_features(struct virtio_net *dev, VhostUserMsg *msg)
+static int
+vhost_user_get_features(struct virtio_net **pdev, VhostUserMsg *msg)
{
+ struct virtio_net *dev = *pdev;
uint64_t features = 0;
rte_vhost_driver_get_features(dev->ifname, &features);
@@ -155,15 +169,16 @@ vhost_user_get_features(struct virtio_net *dev,
VhostUserMsg *msg)
msg->payload.u64 = features;
msg->size = sizeof(msg->payload.u64);
- return features;
+ return VH_RESULT_REPLY;
}
/*
* The queue number that we support are requested.
*/
-static uint32_t
-vhost_user_get_queue_num(struct virtio_net *dev, VhostUserMsg *msg)
+static int
+vhost_user_get_queue_num(struct virtio_net **pdev, VhostUserMsg *msg)
{
+ struct virtio_net *dev = *pdev;
uint32_t queue_num = 0;
rte_vhost_driver_get_queue_num(dev->ifname, &queue_num);
@@ -171,15 +186,17 @@ vhost_user_get_queue_num(struct virtio_net *dev,
VhostUserMsg *msg)
msg->payload.u64 = (uint64_t)queue_num;
msg->size = sizeof(msg->payload.u64);
- return queue_num;
+ return VH_RESULT_REPLY;
}
/*
* We receive the negotiated features supported by us and the virtio device.
*/
static int
-vhost_user_set_features(struct virtio_net *dev, uint64_t features)
+vhost_user_set_features(struct virtio_net **pdev, VhostUserMsg *msg)
{
+ struct virtio_net *dev = *pdev;
+ uint64_t features = msg->payload.u64;
uint64_t vhost_features = 0;
struct rte_vdpa_device *vdpa_dev;
int did = -1;
@@ -189,12 +206,12 @@ vhost_user_set_features(struct virtio_net *dev, uint64_t
features)
RTE_LOG(ERR, VHOST_CONFIG,
"(%d) received invalid negotiated features.\n",
dev->vid);
- return -1;
+ return VH_RESULT_ERR;
}
if (dev->flags & VIRTIO_DEV_RUNNING) {
if (dev->features == features)
- return 0;
+ return VH_RESULT_OK;
/*
* Error out if master tries to change features while device is
@@ -205,7 +222,7 @@ vhost_user_set_features(struct virtio_net *dev, uint64_t
features)
RTE_LOG(ERR, VHOST_CONFIG,
"(%d) features changed while device is
running.\n",
dev->vid);
- return -1;
+ return VH_RESULT_ERR;
}
if (dev->notify_ops->features_changed)
@@ -250,16 +267,17 @@ vhost_user_set_features(struct virtio_net *dev, uint64_t
features)
if (vdpa_dev && vdpa_dev->ops->set_features)
vdpa_dev->ops->set_features(dev->vid);
- return 0;
+ return VH_RESULT_OK;
}
/*
* The virtio device sends us the size of the
Re: [dpdk-dev] [PATCH v2 5/5] vhost: message handling implemented as a callback array
On 07/19/2018 09:13 PM, Nikolay Nikolaev wrote:
Introduce vhost_message_handlers, which maps the message request
type to the message handler. Then replace the switch construct
with a map and call.
Signed-off-by: Nikolay Nikolaev
---
lib/librte_vhost/vhost_user.c | 143 +++--
1 file changed, 54 insertions(+), 89 deletions(-)
Reviewed-by: Maxime Coquelin
diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c
index 6b39d1e30..1b164df27 100644
--- a/lib/librte_vhost/vhost_user.c
+++ b/lib/librte_vhost/vhost_user.c
@@ -1466,6 +1466,34 @@ vhost_user_iotlb_msg(struct virtio_net **pdev,
VhostUserMsg *msg)
return VH_RESULT_OK;
}
+typedef int (*vhost_message_handler_t)(struct virtio_net **pdev, VhostUserMsg * msg);
+static vhost_message_handler_t vhost_message_handlers[VHOST_USER_MAX] = {
+ [VHOST_USER_NONE] = NULL,
+ [VHOST_USER_GET_FEATURES] = vhost_user_get_features,
+ [VHOST_USER_SET_FEATURES] = vhost_user_set_features,
+ [VHOST_USER_SET_OWNER] = vhost_user_set_owner,
+ [VHOST_USER_RESET_OWNER] = vhost_user_reset_owner,
+ [VHOST_USER_SET_MEM_TABLE] = vhost_user_set_mem_table,
+ [VHOST_USER_SET_LOG_BASE] = vhost_user_set_log_base,
+ [VHOST_USER_SET_LOG_FD] = vhost_user_set_log_fd,
+ [VHOST_USER_SET_VRING_NUM] = vhost_user_set_vring_num,
+ [VHOST_USER_SET_VRING_ADDR] = vhost_user_set_vring_addr,
+ [VHOST_USER_SET_VRING_BASE] = vhost_user_set_vring_base,
+ [VHOST_USER_GET_VRING_BASE] = vhost_user_get_vring_base,
+ [VHOST_USER_SET_VRING_KICK] = vhost_user_set_vring_kick,
+ [VHOST_USER_SET_VRING_CALL] = vhost_user_set_vring_call,
+ [VHOST_USER_SET_VRING_ERR] = vhost_user_set_vring_err,
+ [VHOST_USER_GET_PROTOCOL_FEATURES] = vhost_user_get_protocol_features,
+ [VHOST_USER_SET_PROTOCOL_FEATURES] = vhost_user_set_protocol_features,
+ [VHOST_USER_GET_QUEUE_NUM] = vhost_user_get_queue_num,
+ [VHOST_USER_SET_VRING_ENABLE] = vhost_user_set_vring_enable,
+ [VHOST_USER_SEND_RARP] = vhost_user_send_rarp,
+ [VHOST_USER_NET_SET_MTU] = vhost_user_net_set_mtu,
+ [VHOST_USER_SET_SLAVE_REQ_FD] = vhost_user_set_req_fd,
+ [VHOST_USER_IOTLB_MSG] = vhost_user_iotlb_msg,
+};
+
+
/* return bytes# of read on success or negative val on failure. */
static int
read_vhost_message(int sockfd, VhostUserMsg *msg)
@@ -1618,6 +1646,7 @@ vhost_user_msg_handler(int vid, int fd)
int ret;
int unlock_required = 0;
uint32_t skip_master = 0;
+ int request;
dev = get_device(vid);
if (dev == NULL)
@@ -1710,97 +1739,33 @@ vhost_user_msg_handler(int vid, int fd)
goto skip_to_post_handle;
}
- switch (msg.request.master) {
- case VHOST_USER_GET_FEATURES:
- ret = vhost_user_get_features(&dev, &msg);
- send_vhost_reply(fd, &msg);
- break;
- case VHOST_USER_SET_FEATURES:
- ret = vhost_user_set_features(&dev, &msg);
- if (ret)
- return -1;
- break;
-
- case VHOST_USER_GET_PROTOCOL_FEATURES:
- ret = vhost_user_get_protocol_features(&dev, &msg);
- send_vhost_reply(fd, &msg);
- break;
- case VHOST_USER_SET_PROTOCOL_FEATURES:
- ret = vhost_user_set_protocol_features(&dev, &msg);
- break;
-
- case VHOST_USER_SET_OWNER:
- ret = vhost_user_set_owner(&dev, &msg);
- break;
- case VHOST_USER_RESET_OWNER:
- ret = vhost_user_reset_owner(&dev, &msg);
- break;
-
- case VHOST_USER_SET_MEM_TABLE:
- ret = vhost_user_set_mem_table(&dev, &msg);
- break;
-
- case VHOST_USER_SET_LOG_BASE:
- ret = vhost_user_set_log_base(&dev, &msg);
- send_vhost_reply(fd, &msg);
- break;
- case VHOST_USER_SET_LOG_FD:
- ret = vhost_user_set_log_fd(&dev, &msg);
- break;
-
- case VHOST_USER_SET_VRING_NUM:
- ret = vhost_user_set_vring_num(&dev, &msg);
- break;
- case VHOST_USER_SET_VRING_ADDR:
- ret = vhost_user_set_vring_addr(&dev, &msg);
- break;
- case VHOST_USER_SET_VRING_BASE:
- ret = vhost_user_set_vring_base(&dev, &msg);
- break;
-
- case VHOST_USER_GET_VRING_BASE:
- ret = vhost_user_get_vring_base(&dev, &msg);
- send_vhost_reply(fd, &msg);
- break;
-
- case VHOST_USER_SET_VRING_KICK:
- ret = vhost_user_set_vring_kick(&dev, &msg);
- break;
- case VHOST_USER_SET_VRING_CALL:
- ret = vhost_user_set_vring_call(&dev, &msg);
- break;
-
- case VHOST_USER_SET_VRING_ERR:
- ret = vhost_user_se
Re: [dpdk-dev] [PATCH] app/testpmd: add commands for TM to mark pkts
-Original Message- > Date: Fri, 17 Aug 2018 13:39:19 +0200 > From: Krzysztof Kanas > To: dev@dpdk.org > Cc: krzysztof.ka...@caviumnetworks.com, > jerin.jacobkollanukka...@cavium.com, nithin.dabilpu...@cavium.com > Subject: [dpdk-dev] [PATCH] app/testpmd: add commands for TM to mark pkts > X-Mailer: git-send-email 2.18.0 > > Add following testpmd run-time commands to support test of TM packet > marking: > > set port tm mark ip_ecn > set port tm mark ip_dscp > set port tm mark vlan_dei > > Signed-off-by: Krzysztof Kanas > --- > app/test-pmd/cmdline.c | 3 + > app/test-pmd/cmdline_tm.c | 260 > app/test-pmd/cmdline_tm.h | 3 + + Cristian, Wenzhuo, Jingjing, Bernard In addition to this slow path change, IMO, we need test engine similar to checksum(app/test-pmd/csumonly.c) to generate traffic(app/test-pmd/markonly.c) with struct rte_mbuf::tx_offload populated so that we can test the packet marking. Cristian, Any thoughts as a rte_tm maintainer?
Re: [dpdk-dev] [PATCH v2 4/4] build: generate API documentation with Meson
On Fri, Sep 07, 2018 at 05:55:24PM +0100, Luca Boccassi wrote:
> Signed-off-by: Luca Boccassi
> ---
> v2: made doxygen dependency optional, skip doxygen build when not found
>
> doc/api/generate_doxygen.sh | 10 +++
> doc/api/meson.build | 54 +
> doc/build-sdk-meson.txt | 2 ++
> doc/meson.build | 4 +++
> meson.build | 3 +++
> meson_options.txt | 2 ++
> 6 files changed, 75 insertions(+)
> create mode 100755 doc/api/generate_doxygen.sh
> create mode 100644 doc/api/meson.build
> create mode 100644 doc/meson.build
>
> diff --git a/doc/api/generate_doxygen.sh b/doc/api/generate_doxygen.sh
> new file mode 100755
> index 00..ab57660958
> --- /dev/null
> +++ b/doc/api/generate_doxygen.sh
> @@ -0,0 +1,10 @@
> +#! /bin/sh -e
> +# SPDX-License-Identifier: BSD-3-Clause
> +# Copyright 2018 Luca Boccassi
> +
> +DOXYCONF=$1
> +OUTDIR=$2
> +SCRIPTCSS=$3
> +
> +doxygen "${DOXYCONF}"
> +"${SCRIPTCSS}" "${OUTDIR}"/doxygen.css
> diff --git a/doc/api/meson.build b/doc/api/meson.build
> new file mode 100644
> index 00..5dfa0fe044
> --- /dev/null
> +++ b/doc/api/meson.build
> @@ -0,0 +1,54 @@
> +# SPDX-License-Identifier: BSD-3-Clause
> +# Copyright(c) 2018 Luca Boccassi
> +
> +doxygen = find_program('doxygen', required: false)
Thinking about it, I think that "required" should be set to
"get_option(enable_docs)", since if someone explicitly enables the docs
they should be built and cause error if they can't be.
> +
> +if doxygen.found()
> + # due to the CSS customisation script, which needs to run on a file that
> + # is in a subdirectory that is created at build time and thus it cannot
> + # be an individual custom_target, we need to wrap the doxygen call in a
> + # script to run the CSS modification afterwards
> + generate_doxygen = find_program('generate_doxygen.sh')
> + generate_examples = find_program('generate_examples.sh')
> + generate_css = find_program('doxy-html-custom.sh')
> +
> + inputdir = join_paths(meson.source_root(), 'examples')
> + htmldir = join_paths('doc', 'html')
> +
> + # due to the following bug:
> https://github.com/mesonbuild/meson/issues/4107
> + # if install is set to true it will override build_by_default and it
> will
> + # cause the target to always be built. If install were to be always set
> to
> + # false it would be impossible to install the docs.
> + # So use a configure option for now.
> + example = custom_target('examples.dox',
> + input: inputdir,
> + output: 'examples.dox',
> + command: [generate_examples, '@INPUT@', '@OUTPUT@'],
> + install: get_option('enable_docs'),
> + install_dir: htmldir,
> + build_by_default: false)
> +
> + cdata = configuration_data()
> + cdata.set('VERSION', meson.project_version())
> + cdata.set('API_EXAMPLES', join_paths(meson.build_root(), 'doc', 'api',
> 'examples.dox'))
> + cdata.set('OUTPUT', join_paths(meson.build_root(), 'doc', 'api'))
> + cdata.set('HTML_OUTPUT', 'api')
> + cdata.set('TOPDIR', meson.source_root())
> + cdata.set('STRIP_FROM_PATH', meson.source_root())
> +
> + doxy_conf = configure_file(input: 'doxy-api.conf.in',
> + output: 'doxy-api.conf',
> + configuration: cdata,
> + install: false)
> +
> + doxy_build = custom_target('doxygen',
> + depends: example,
> + input: doxy_conf,
> + output: 'api',
> + command: [generate_doxygen, '@INPUT@', '@OUTPUT@',
> generate_css],
> + install: get_option('enable_docs'),
> + install_dir: htmldir,
> + build_by_default: false)
> +
> + run_target('doc', command: 'true', depends: doxy_build)
> +endif
Suggestion: add a run_target in an else leg to just print out "no
doxygen found" or similar message when ninja doc is called.
Re: [dpdk-dev] [PATCH v2 1/4] mk: use script to generate examples.dox
On Fri, Sep 07, 2018 at 05:55:21PM +0100, Luca Boccassi wrote: > This will make it possible to generate the file in the same way from > Meson as well. > > Signed-off-by: Luca Boccassi > --- Couple of comments on patch 4, otherwise looks ok to me. Series-acked-by: Bruce Richardson
