Re: [dpdk-dev] [PATCH] eal: make rte_lcore_cpuset and rte_lcore_to_cpu_id stable
On 8/31/21 11:08 PM, Stephen Hemminger wrote:
> These were converted from inline to functions in 19.11
> and should be marked as stable now.
>
> Signed-off-by: Stephen Hemminger
> ---
> lib/eal/include/rte_lcore.h | 8
> lib/eal/version.map | 4 ++--
> 2 files changed, 2 insertions(+), 10 deletions(-)
>
> diff --git a/lib/eal/include/rte_lcore.h b/lib/eal/include/rte_lcore.h
> index 1550b75da0a5..cf6203a9af79 100644
> --- a/lib/eal/include/rte_lcore.h
> +++ b/lib/eal/include/rte_lcore.h
> @@ -172,9 +172,6 @@ unsigned int
> rte_lcore_to_socket_id(unsigned int lcore_id);
>
> /**
> - * @warning
> - * @b EXPERIMENTAL: this API may change without prior notice.
> - *
> * Return the id of the lcore on a socket starting from zero.
> *
> * @param lcore_id
> @@ -182,23 +179,18 @@ rte_lcore_to_socket_id(unsigned int lcore_id);
> * @return
> * The relative index, or -1 if not enabled.
> */
> -__rte_experimental
> int
> rte_lcore_to_cpu_id(int lcore_id);
>
> #ifdef RTE_HAS_CPUSET
>
> /**
> - * @warning
> - * @b EXPERIMENTAL: this API may change without prior notice.
> - *
> * Return the cpuset for a given lcore.
> * @param lcore_id
> * the targeted lcore, which MUST be between 0 and RTE_MAX_LCORE-1.
> * @return
> * The cpuset of that lcore
> */
> -__rte_experimental
> rte_cpuset_t
> rte_lcore_cpuset(unsigned int lcore_id);
I'm wondering why negative lcore_id is supported above
with special meaning, but not supported here.
>
> diff --git a/lib/eal/version.map b/lib/eal/version.map
> index beeb986adcaf..14565aa10df4 100644
> --- a/lib/eal/version.map
> +++ b/lib/eal/version.map
> @@ -98,9 +98,11 @@ DPDK_22 {
> rte_keepalive_register_core; # WINDOWS_NO_EXPORT
> rte_keepalive_register_relay_callback; # WINDOWS_NO_EXPORT
> rte_lcore_count;
> + rte_lcore_cpuset;
> rte_lcore_has_role;
> rte_lcore_index;
> rte_lcore_is_enabled;
> + rte_lcore_to_cpu_id;
> rte_lcore_to_socket_id;
> rte_log;
> rte_log_cur_msg_loglevel;
> @@ -322,8 +324,6 @@ EXPERIMENTAL {
>
> # added in 19.08
> rte_intr_ack;
> - rte_lcore_cpuset;
> - rte_lcore_to_cpu_id;
> rte_mcfg_timer_lock;
> rte_mcfg_timer_unlock;
> rte_rand_max; # WINDOWS_NO_EXPORT
>
Re: [dpdk-dev] [PATCH] ethdev: remove experimental flag from getting intr fd API
On 9/1/21 4:50 AM, Xiaoyun Li wrote: > Remove the experimental tag for rte_eth_dev_rx_intr_ctl_q_get_fd API > that was introduced in 18.11 and have been around for 11 releases. > > Signed-off-by: Xiaoyun Li Acked-by: Andrew Rybchenko
Re: [dpdk-dev] [PATCH] net: make rte_ether_unformat_addr stable
On 8/31/21 11:07 PM, Stephen Hemminger wrote: > This function has been in since 19.11. > > Signed-off-by: Stephen Hemminger Acked-by: Andrew Rybchenko
[dpdk-dev] [PATCH v3] net/ixgbe: fix mac resourece leak
In the eth_ixgbevf_dev_init and eth_ixgbe_dev_init functions, memory is
allocated for the MAC address, and the address is stored in the
eth_dev->data->mac_addrs member variable. If the subsequent function is
abnormal, you need to use the rte_free function to release the MAC
address memory.
Fixes: af75078fece3 ("first public release")
Cc: sta...@dpdk.org
Signed-off-by: Qiming Chen
---
v2:
Clear coding style warning.
v3:
Modify fixes commit.
---
drivers/net/ixgbe/ixgbe_ethdev.c | 4
1 file changed, 4 insertions(+)
diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/drivers/net/ixgbe/ixgbe_ethdev.c
index 7d3a821300..6a91f104e1 100644
--- a/drivers/net/ixgbe/ixgbe_ethdev.c
+++ b/drivers/net/ixgbe/ixgbe_ethdev.c
@@ -1218,6 +1218,8 @@ eth_ixgbe_dev_init(struct rte_eth_dev *eth_dev, void
*init_params __rte_unused)
PMD_INIT_LOG(ERR,
"Failed to allocate %d bytes needed to store MAC
addresses",
RTE_ETHER_ADDR_LEN * IXGBE_VMDQ_NUM_UC_MAC);
+ rte_free(eth_dev->data->mac_addrs);
+ eth_dev->data->mac_addrs = NULL;
return -ENOMEM;
}
@@ -1672,6 +1674,8 @@ eth_ixgbevf_dev_init(struct rte_eth_dev *eth_dev)
default:
PMD_INIT_LOG(ERR, "VF Initialization Failure: %d", diag);
+ rte_free(eth_dev->data->mac_addrs);
+ eth_dev->data->mac_addrs = NULL;
return -EIO;
}
--
2.30.1.windows.1
[dpdk-dev] [PATCH v3] net/ixgbe: fix mbuf leak
A local test found that repeated port start and stop operations during
the continuous SSE vector bufflist receiving process will cause the mbuf
resource to run out. The final positioning is when the port is stopped,
the mbuf of the pkt_first_seg pointer is not released. Resources leak.
The patch scheme is to judge whether the pointer is empty when the port
is stopped, and release the corresponding mbuf if it is not empty.
Fixes: af75078fece3 ("first public release")
Cc: sta...@dpdk.org
Signed-off-by: Qiming Chen
---
v2:
Sync to sta...@dpdk.org
v3:
Modify fixes commit
---
drivers/net/ixgbe/ixgbe_rxtx.c | 4
1 file changed, 4 insertions(+)
diff --git a/drivers/net/ixgbe/ixgbe_rxtx.c b/drivers/net/ixgbe/ixgbe_rxtx.c
index c814a28cb4..bfdfd5e755 100644
--- a/drivers/net/ixgbe/ixgbe_rxtx.c
+++ b/drivers/net/ixgbe/ixgbe_rxtx.c
@@ -2981,6 +2981,10 @@ ixgbe_reset_rx_queue(struct ixgbe_adapter *adapter,
struct ixgbe_rx_queue *rxq)
rxq->rx_free_trigger = (uint16_t)(rxq->rx_free_thresh - 1);
rxq->rx_tail = 0;
rxq->nb_rx_hold = 0;
+
+ if (rxq->pkt_first_seg != NULL)
+ rte_pktmbuf_free(rxq->pkt_first_seg);
+
rxq->pkt_first_seg = NULL;
rxq->pkt_last_seg = NULL;
--
2.30.1.windows.1
Re: [dpdk-dev] [PATCH v2] net/ixgbe: fix mac resourece leak
Ask a question: how to submit an issue while keeping the exchange log.
| |
Qiming Chen
|
|
chenqiming_hua...@163.com
|
签名由网易邮箱大师定制
On 9/1/2021 14:25，Wang, Haiyue wrote：
-Original Message-
From: Qiming Chen
Sent: Tuesday, August 31, 2021 22:22
To: dev@dpdk.org
Cc: Wang, Haiyue ; Qiming Chen
; sta...@dpdk.org
Subject: [PATCH v2] net/ixgbe: fix mac resourece leak
In the eth_ixgbevf_dev_init and eth_ixgbe_dev_init functions, memory is
allocated for the MAC address, and the address is stored in the
eth_dev->data->mac_addrs member variable. If the subsequent function is
abnormal, you need to use the rte_free function to release the MAC
address memory.
Fixes: abf7275bbaa2 ("ixgbe: move to drivers/net/")
Use this tag:
Fixes: af75078fece3 ("first public release")
Also, your ixgbe series fixes can be into patch set, no need to
send them one by one.
Thanks for your effort. ;-)
Cc: sta...@dpdk.org
Signed-off-by: Qiming Chen
---
v2:
Clear coding style warning.
---
drivers/net/ixgbe/ixgbe_ethdev.c | 4
1 file changed, 4 insertions(+)
diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/drivers/net/ixgbe/ixgbe_ethdev.c
index 7d3a821300..6a91f104e1 100644
--- a/drivers/net/ixgbe/ixgbe_ethdev.c
+++ b/drivers/net/ixgbe/ixgbe_ethdev.c
@@ -1218,6 +1218,8 @@ eth_ixgbe_dev_init(struct rte_eth_dev *eth_dev, void
*init_params __rte_unused)
PMD_INIT_LOG(ERR,
"Failed to allocate %d bytes needed to store MAC addresses",
RTE_ETHER_ADDR_LEN * IXGBE_VMDQ_NUM_UC_MAC);
+rte_free(eth_dev->data->mac_addrs);
+eth_dev->data->mac_addrs = NULL;
return -ENOMEM;
}
@@ -1672,6 +1674,8 @@ eth_ixgbevf_dev_init(struct rte_eth_dev *eth_dev)
default:
PMD_INIT_LOG(ERR, "VF Initialization Failure: %d", diag);
+rte_free(eth_dev->data->mac_addrs);
+eth_dev->data->mac_addrs = NULL;
return -EIO;
}
--
2.30.1.windows.1
[dpdk-dev] [PATCH] net/bnxt: fix double allocation of ring groups
From: Kalesh AP
After commit "d68249f88266", driver allocates ring groups in
bnxt_alloc_hwrm_rx_ring(). But during port start, driver invokes
bnxt_alloc_hwrm_rx_ring() followed by bnxt_alloc_all_hwrm_ring_grps().
This will cause the FW command failure in bnxt_alloc_all_hwrm_ring_grps().
To fix this, just don't create the ring group if it is already created.
Fixes: d68249f88266 ("net/bnxt: update ring group after ring stop start")
Cc: sta...@dpdk.org
Signed-off-by: Kalesh AP
Reviewed-by: Ajit Khaparde
Reviewed-by: Lance Richardson
---
drivers/net/bnxt/bnxt_hwrm.c | 4
1 file changed, 4 insertions(+)
diff --git a/drivers/net/bnxt/bnxt_hwrm.c b/drivers/net/bnxt/bnxt_hwrm.c
index f29d574..f65dc72 100644
--- a/drivers/net/bnxt/bnxt_hwrm.c
+++ b/drivers/net/bnxt/bnxt_hwrm.c
@@ -1864,6 +1864,10 @@ int bnxt_hwrm_ring_grp_alloc(struct bnxt *bp, unsigned
int idx)
struct hwrm_ring_grp_alloc_input req = {.req_type = 0 };
struct hwrm_ring_grp_alloc_output *resp = bp->hwrm_cmd_resp_addr;
+ /* Don't attempt to re-create the ring group if it is already created */
+ if (bp->grp_info[idx].fw_grp_id != INVALID_HW_RING_ID)
+ return 0;
+
HWRM_PREP(&req, HWRM_RING_GRP_ALLOC, BNXT_USE_CHIMP_MB);
req.cr = rte_cpu_to_le_16(bp->grp_info[idx].cp_fw_ring_id);
--
2.10.1
Re: [dpdk-dev] [PATCH] eal: promote random function with upper bound to stable
On Wed, Sep 1, 2021 at 9:35 AM Mattias Rönnblom
wrote:
>
> On 2021-09-01 09:29, Mattias Rönnblom wrote:
> > Remove experimental tag from rte_rand_max().
> >
> > Signed-off-by: Mattias Rönnblom
> > ---
> > lib/eal/include/rte_random.h | 1 -
> > lib/eal/version.map | 2 +-
> > 2 files changed, 1 insertion(+), 2 deletions(-)
> >
> > diff --git a/lib/eal/include/rte_random.h b/lib/eal/include/rte_random.h
> > index 2b30ec85c1..29f5f1325a 100644
> > --- a/lib/eal/include/rte_random.h
> > +++ b/lib/eal/include/rte_random.h
> > @@ -62,7 +62,6 @@ rte_rand(void);
> >* @return
> >* A pseudo-random value between 0 and (upper_bound-1).
> >*/
> > -__rte_experimental
> > uint64_t
> > rte_rand_max(uint64_t upper_bound);
> >
> > diff --git a/lib/eal/version.map b/lib/eal/version.map
> > index beeb986adc..92caaa2bd5 100644
> > --- a/lib/eal/version.map
> > +++ b/lib/eal/version.map
> > @@ -148,6 +148,7 @@ DPDK_22 {
> > rte_memzone_walk;
> > rte_openlog_stream;
> > rte_rand;
> > + rte_rand_max;
> > rte_realloc;
> > rte_reciprocal_value;
> > rte_reciprocal_value_u64;
> > @@ -326,7 +327,6 @@ EXPERIMENTAL {
> > rte_lcore_to_cpu_id;
> > rte_mcfg_timer_lock;
> > rte_mcfg_timer_unlock;
> > - rte_rand_max; # WINDOWS_NO_EXPORT
>
>
> Should I have kept the WINDOWS_NO_EXPORT tag?
Strictly speaking, you should keep it: this patch is about API promotion.
The windows exports are fixed in another patch in any case:
https://patchwork.dpdk.org/project/dpdk/patch/20210829021602.290466-3-dmitry.kozl...@gmail.com/
--
David Marchand
Re: [dpdk-dev] [RFC] Control packet event adapter and FIFO library
On Wed, Sep 1, 2021 at 11:55 AM Jerin Jacob wrote: > > On Wed, Sep 1, 2021 at 11:12 AM Kaladi, Ashok K > wrote: > > > > Dear dpdk-dev team, > > > > We would like to propose the following RFC for your review. > > > > A user space application may need access to the packets handled by eventdev > > based DPDK application. This application doesn't use mbuf or eventdev based > > DPDK APIs. Presently this is not possible without passing packets through > > DPDK KNI. > > > I think it is an innovative idea it is useful for multiple use cases > not just for eventdev. > > Some feedback on thoughts > > 1) The FIFO library should be generic it should not be specific to eventdev > 2) I think, This FIFO library should be generic and KNI also be a > consumer of this library > 3) I think, FIFO should not be a device instead it should be an > abstact object like rte_mempool * > 4) We need to consider User space app can be another DPDK primary > process or some non DPDK app > 4) I think, we can remove the Linux shared memory dependency instead > of introduce some > scheme of "exporting" memzone from DPDK application to another user > space app or another DPDK primary process. > I see the following reasons: > - It is backed by hugepage so better performance > - Is kernel do any memcpy when using Linux shm calls in kernel space? > > Thoughts? > > May be you can share set of API prototypes without any implementation > for the next level discussion if > others are OK this kind of library. Some thoughts for generic fifo library. // Create SPSC fifo of size struct rte_ipc_queue* rte_ipc_queue_create(const char *name, size_t size, int socket_id, unsigned int flags); void rte_ipc_queue_destroy(struct rte_ipc_queue* iq); // userspace app or another DPDK primary process to get the ipc_queue handle. struct rte_ipc_queue* rte_ipc_queue_lookup(const char *proc_name, const char *ipc_queue_name); // Get the memory of size to fill by producer size_t rte_ipc_queue_enq_peek(struct rte_ipc_queue* iq, void **obj, size_t size); // Commit the head pointer int rte_ipc_queue_enq_commit(struct rte_ipc_queue* iq, void *obj, size_t size); // Get the memory of size to to consumer by consumer size_t rte_ipc_queue_deq_peek(struct rte_ipc_queue* iq, void **obj, size_t size); // Commit the tail pointer int rte_ipc_queue_deq_commit(struct rte_ipc_queue* iq, void *obj, size_t size); int rte_ipc_queue_in_use(struct rte_ipc_queue* iq) int rte_ipc_queue_avilable(struct rte_ipc_queue* iq) except rte_ipc_queue_create() and rte_ipc_queue_destroy() all the remaning rte_ipc_* APIs can be used the other userspace appliction. Another alternative is to have ZeroMQ kind of messaging library for various transports. It may not as fast this, Needs a prototype to verify. > > > > > This RFC introduces control packet event adapter (CP adapter) and FIFO > > library > > which can be used by user space applications to access the packets > > handled by eventdev based DPDK application. Linux shared memory is used to > > keep mempool (mbufs), packets and FIFOs. This allows copy-free access of > > packets to both the applications. The Rx and Tx FIFOs are used to exchange > > packets between the applications. > > > > '--' > > | mempool | > > '-''---'|..|'--' > > '---' > > | || || <--- Tx FIFOs || | > > | | > > | User space || || <--- Alloc FIFO || Control | > > | DPDK| > > |App <> FIFO Lib <> ---> Rx FIFOs <> Packet > > <> Eventdev | > > | || || ---> Free FIFO || Adapter | > > | App | > > | || || || | > > | | > > '-''---'| |'--' > > '---' > > '--' > > > > CP adapter does the mbuf allocation and free for user space application. > > Mbufs > > allocated for user space application are put into alloc queue by CP adapter. > > The used mbufs are put into free queue by user space application. > > The FIFO lib translates packets between rte_mbuf format and a simplified > > format which will be used by the user application. FIFO lib doesn't use any > > other > > DPDK features than rte_mbuf structure. The simplified packet format > > contains only > > essential parameters like address of the payload, packet length, etc. > > > > - user space app send path: When user space application wants to send > > a packet, it takes an mbufs from Alloc FIFO, writes packet to it and puts > > it > > into Rx queue. CP adapter which polls this queue gets the packet and > > enqueues > > to eventdev. Then the mbuf is freed by CP adapter. > > > > - use space
Re: [dpdk-dev] [EXT] Re: [PATCH 5/8] crypto/cnxk: add cn9k crypto adapter fast path ops
> > > On 30/08/2021 12:09, Shijith Thotton wrote: > > Added crypto adapter enqueue and dequeue operations for CN9K. > > > > Signed-off-by: Shijith Thotton > > --- > > drivers/crypto/cnxk/cn9k_cryptodev_ops.c | 235 - > -- > > drivers/crypto/cnxk/cn9k_cryptodev_ops.h | 6 + > > drivers/crypto/cnxk/cnxk_cryptodev_ops.h | 28 +++ > > drivers/crypto/cnxk/meson.build | 2 +- > > drivers/crypto/cnxk/version.map | 5 + > > 5 files changed, 205 insertions(+), 71 deletions(-) > > > [SNIP] > > Acked-by: Ray Kinsella Acked-by: Anoob Joseph
Re: [dpdk-dev] [EXT] Re: [PATCH 7/8] crypto/cnxk: add cn10k crypto adapter fast path ops
> > > On 30/08/2021 12:09, Shijith Thotton wrote: > > Added crypto adapter enqueue and dequeue operations for CN10K. > > > > Signed-off-by: Shijith Thotton > > --- > > drivers/crypto/cnxk/cn10k_cryptodev_ops.c | 95 > > +++ > drivers/crypto/cnxk/cn10k_cryptodev_ops.h | 6 ++ > > drivers/crypto/cnxk/version.map | 2 + > > 3 files changed, 103 insertions(+) > > > [SNIP] > > Acked-by: Ray Kinsella Acked-by: Anoob Joseph
Re: [dpdk-dev] [PATCH] net: make rte_ether_unformat_addr stable
On 9/1/2021 8:11 AM, Andrew Rybchenko wrote: > On 8/31/21 11:07 PM, Stephen Hemminger wrote: >> This function has been in since 19.11. >> >> Signed-off-by: Stephen Hemminger > > Acked-by: Andrew Rybchenko > Acked-by: Ferruh Yigit
Re: [dpdk-dev] [PATCH] ethdev: remove experimental flag from getting intr fd API
On 9/1/2021 8:08 AM, Andrew Rybchenko wrote: > On 9/1/21 4:50 AM, Xiaoyun Li wrote: >> Remove the experimental tag for rte_eth_dev_rx_intr_ctl_q_get_fd API >> that was introduced in 18.11 and have been around for 11 releases. >> >> Signed-off-by: Xiaoyun Li > > Acked-by: Andrew Rybchenko > Acked-by: Ferruh Yigit
Re: [dpdk-dev] [PATCH v1 1/3] net/ixgbe: promote some API to stable
On 9/1/2021 6:07 AM, Haiyue Wang wrote: > The DPDK Symbol Bot reports: > Please note the symbols listed below have expired. In line with the > DPDK ABI policy, they should be scheduled for removal, in the next > DPDK release. > > Symbol > rte_pmd_ixgbe_mdio_lock > rte_pmd_ixgbe_mdio_unlock > rte_pmd_ixgbe_mdio_unlocked_read > rte_pmd_ixgbe_mdio_unlocked_write > rte_pmd_ixgbe_upd_fctrl_sbp I wonder if we should keep PMD specific APIs as experimental (Not talking about mbuf 'dynfield' / 'dynflag' APIs, we can promote them). If an application is using PMD specific API, not sure if it will concern about PMD specific APIs. And keeping PMD specific APIs lets us remove them as soon as we can, also adds additional discourage for users to use them. > > Signed-off-by: Haiyue Wang <...>
Re: [dpdk-dev] [PATCH v1 3/3] ethdev: promote burst mode API to stable
On 9/1/2021 6:07 AM, Haiyue Wang wrote: > The DPDK Symbol Bot reports: > Please note the symbols listed below have expired. In line with the > DPDK ABI policy, they should be scheduled for removal, in the next > DPDK release. > > Symbol > rte_eth_rx_burst_mode_get > rte_eth_tx_burst_mode_get > > Signed-off-by: Haiyue Wang Acked-by: Ferruh Yigit
Re: [dpdk-dev] [PATCH] eal: add log to print dpdk version at start
On Wed, Sep 01, 2021 at 11:48:21AM +0530, Aman Singh wrote: > From the logs it is difficult to get the DPDK version > that was used. So added a debug log to print the same. > The log has been added in eal_init so it gets printed > at startup for any application. > > Signed-off-by: Aman Singh > --- > lib/eal/freebsd/eal.c | 2 +- > lib/eal/linux/eal.c | 2 +- > lib/eal/windows/eal.c | 1 + > 3 files changed, 3 insertions(+), 2 deletions(-) > Note: EAL already has the "-v" flag which prints the version number at startup.
[dpdk-dev] 19.11.10 patches review and test
Hi all, Here is a list of patches targeted for stable release 19.11.10. The planned date for the final release is 6th of September. Please help with testing and validation of your use cases and report any issues/results with reply-all to this mail. For the final release the fixes and reported validations will be added to the release notes. You might think that time is a bit short compared to usually, but v19.11.10-rc2 has only two changes over -rc1 and both are not too complex. Therefore I'd be hoping that those that reported/tested these issues that are now: - https://bugs.dpdk.org/show_bug.cgi?id=731 - issues with vhost: fix missing guest pages table NUMA realloc could please re-test that particullar subset of cases. Everone else is welcome to fully test it as well or to do so in whatever subset of tests you consider appropriate. Therefore the time isn't as short as it might appear at first. And generally - if you are testing but need a bit more time, speak up and we usually can wait without much of a problem. In addition that also gives the two Teams that I'm still missing the -RC1 results a few more days to complete those. A release candidate tarball can be found at: https://dpdk.org/browse/dpdk-stable/tag/?id=v19.11.10-rc2 These patches are located at branch 19.11 of dpdk-stable repo: https://dpdk.org/browse/dpdk-stable/ Thanks. Christian Ehrhardt --- Abhinandan Gujjar (1): test/crypto: fix mempool size for session-less Ajit Khaparde (2): doc: fix default burst size in testpmd net/bnxt: fix Rx interrupt setting Akhil Goyal (1): crypto/octeontx: fix freeing after device release Alvin Zhang (1): telemetry: fix race in telemetry control thread creation Anatoly Burakov (1): power: fix namespace for internal struct Chenbo Xia (1): vhost: utilize dynamic memory allocator Chengchang Tang (1): net/hns3: fix VLAN strip log Chengwen Feng (1): net/hns3: fix filter parsing comment Christian Ehrhardt (1): version: 19.11.10-rc1 Ciara Power (1): cryptodev: fix freeing after device release Dana Vardi (5): crypto/mvsam: fix AES-GCM session parameters crypto/mvsam: fix session data reset crypto/mvsam: fix options parsing net/mvpp2: fix port speed overflow net/mvpp2: fix configured state dependency Dapeng Yu (5): net/ixgbe: fix flow entry access after freeing net/softnic: fix connection memory leak net/softnic: fix memory leak as profile is freed net/i40e: fix multi-process shared data net/ice: fix default RSS key generation David Hunt (1): distributor: fix 128-bit write alignment David Marchand (4): drivers/net: fix memzone allocations for DMA memory ipc: stop mp control thread on cleanup net/ice: fix memzone leak when firmware is missing net/octeontx/base: fix debug build with clang Dmitry Kozlyuk (3): doc: add limitation for ConnectX-4 with L2 in mlx5 guide net/mlx5: fix Rx/Tx queue checks bus: clarify log for non-NUMA-aware devices Feifei Wang (2): crypto/qat: fix Arm build with special memcpy net/mlx5: fix incorrect r/w lock usage in DMA unmap Ferruh Yigit (3): kni: fix crash on userspace VA for segmented packets app/testpmd: fix help string for port reset doc: announce common prefix for ethdev Ghalem Boudour (1): net/ena: enable multi-segment in Tx offload flags Gregory Etelson (3): app/testpmd: fix Tx checksum calculation for tunnel app/testpmd: fix IPv4 checksum net/mlx5: fix representor interrupt handler Guoyang Zhou (2): net/hinic: increase protection of the VLAN net/hinic/base: fix LRO Haiyue Wang (1): net/iavf: fix RSS key access out of bound Hemant Agrawal (1): bus/dpaa: fix freeing in FMAN interface destructor Henry Nadeau (1): doc: fix spelling Hongbo Zheng (1): net/hns3: increase VF reset retry maximum Huisong Li (3): net/hns3: fix delay for waiting to stop Rx/Tx app/testpmd: change port link speed without stopping all net/hns3: fix Tx prepare after stop Ivan Ilchenko (7): bitmap: fix buffer overrun in bitmap init net/sfc: fix MAC stats lock in xstats query by ID net/sfc: fix reading adapter state without locking net/sfc: fix xstats query by ID according to ethdev net/sfc: fix xstats query by unsorted list of IDs net/sfc: fix MAC stats update for stopped device net/virtio: report maximum MTU in device info Jan Viktorin (1): ethdev: fix doc of flow action Jiawei Wang (1): net/mlx5: fix RSS flow item expansion for GRE key Joyce Kong (1): net/i40e: fix descriptor scan on Arm Juraj Linkeš (1): net/virtio: fix aarch32 build Kai Ji (1): app/crypto-perf: fix out-of-place mempool allocation Kalesh AP (14): net/bnxt: fix error messages in VNIC prepare net/bnxt: set flow error when free filter not avail
[dpdk-dev] [PATCH v2 4/8] crypto/cnxk: use rlen from CPT result with lookaside
Use rlen from CPT result with lookaside operations
Signed-off-by: Tejasree Kondoj
---
drivers/crypto/cnxk/cn10k_cryptodev_ops.c | 40 ++-
drivers/crypto/cnxk/cn10k_ipsec.c | 4 +--
drivers/crypto/cnxk/cn10k_ipsec.h | 4 +--
drivers/crypto/cnxk/cn10k_ipsec_la_ops.h | 28 ++--
4 files changed, 15 insertions(+), 61 deletions(-)
diff --git a/drivers/crypto/cnxk/cn10k_cryptodev_ops.c
b/drivers/crypto/cnxk/cn10k_cryptodev_ops.c
index 15f66c2515..780a321cf7 100644
--- a/drivers/crypto/cnxk/cn10k_cryptodev_ops.c
+++ b/drivers/crypto/cnxk/cn10k_cryptodev_ops.c
@@ -48,7 +48,7 @@ cn10k_cpt_sym_temp_sess_create(struct cnxk_cpt_qp *qp, struct
rte_crypto_op *op)
static __rte_always_inline int __rte_hot
cpt_sec_inst_fill(struct rte_crypto_op *op, struct cn10k_sec_session *sess,
- struct cpt_inflight_req *infl_req, struct cpt_inst_s *inst)
+ struct cpt_inst_s *inst)
{
struct rte_crypto_sym_op *sym_op = op->sym;
union roc_ot_ipsec_sa_word2 *w2;
@@ -70,10 +70,8 @@ cpt_sec_inst_fill(struct rte_crypto_op *op, struct
cn10k_sec_session *sess,
if (w2->s.dir == ROC_IE_OT_SA_DIR_OUTBOUND)
ret = process_outb_sa(op, sa, inst);
- else {
- infl_req->op_flags |= CPT_OP_FLAGS_IPSEC_DIR_INBOUND;
+ else
ret = process_inb_sa(op, sa, inst);
- }
return ret;
}
@@ -122,8 +120,7 @@ cn10k_cpt_fill_inst(struct cnxk_cpt_qp *qp, struct
rte_crypto_op *ops[],
if (op->sess_type == RTE_CRYPTO_OP_SECURITY_SESSION) {
sec_sess = get_sec_session_private_data(
sym_op->sec_session);
- ret = cpt_sec_inst_fill(op, sec_sess, infl_req,
- &inst[0]);
+ ret = cpt_sec_inst_fill(op, sec_sess, &inst[0]);
if (unlikely(ret))
return 0;
w7 = sec_sess->sa.inst.w7;
@@ -334,30 +331,13 @@ cn10k_cpt_crypto_adapter_enqueue(uintptr_t tag_op, struct
rte_crypto_op *op)
static inline void
cn10k_cpt_sec_post_process(struct rte_crypto_op *cop,
- struct cpt_inflight_req *infl_req)
+ struct cpt_cn10k_res_s *res)
{
- struct rte_crypto_sym_op *sym_op = cop->sym;
- struct rte_mbuf *m = sym_op->m_src;
- struct rte_ipv6_hdr *ip6;
- struct rte_ipv4_hdr *ip;
- uint16_t m_len;
-
- if (infl_req->op_flags & CPT_OP_FLAGS_IPSEC_DIR_INBOUND) {
- ip = (struct rte_ipv4_hdr *)rte_pktmbuf_mtod(m, char *);
-
- if (((ip->version_ihl & 0xf0) >> RTE_IPV4_IHL_MULTIPLIER) ==
- IPVERSION) {
- m_len = rte_be_to_cpu_16(ip->total_length);
- } else {
- PLT_ASSERT(((ip->version_ihl & 0xf0) >>
- RTE_IPV4_IHL_MULTIPLIER) == 6);
- ip6 = (struct rte_ipv6_hdr *)ip;
- m_len = rte_be_to_cpu_16(ip6->payload_len) +
- sizeof(struct rte_ipv6_hdr);
- }
- m->data_len = m_len;
- m->pkt_len = m_len;
- }
+ struct rte_mbuf *m = cop->sym->m_src;
+ const uint16_t m_len = res->rlen;
+
+ m->data_len = m_len;
+ m->pkt_len = m_len;
}
static inline void
@@ -385,7 +365,7 @@ cn10k_cpt_dequeue_post_process(struct cnxk_cpt_qp *qp,
cop->status = RTE_CRYPTO_OP_STATUS_SUCCESS;
if (cop->type == RTE_CRYPTO_OP_TYPE_SYMMETRIC) {
if (cop->sess_type == RTE_CRYPTO_OP_SECURITY_SESSION) {
- cn10k_cpt_sec_post_process(cop, infl_req);
+ cn10k_cpt_sec_post_process(cop, res);
return;
}
diff --git a/drivers/crypto/cnxk/cn10k_ipsec.c
b/drivers/crypto/cnxk/cn10k_ipsec.c
index 944e0a7e3b..98110872a3 100644
--- a/drivers/crypto/cnxk/cn10k_ipsec.c
+++ b/drivers/crypto/cnxk/cn10k_ipsec.c
@@ -176,9 +176,7 @@ cn10k_ipsec_outb_sa_create(struct roc_cpt *roc_cpt,
if (ret)
return ret;
- sa->partial_len = rlens.partial_len;
- sa->roundup_byte = rlens.roundup_byte;
- sa->roundup_len = rlens.roundup_len;
+ sa->max_extended_len = rlens.max_extended_len;
/* pre-populate CPT INST word 4 */
inst_w4.u64 = 0;
diff --git a/drivers/crypto/cnxk/cn10k_ipsec.h
b/drivers/crypto/cnxk/cn10k_ipsec.h
index c30492e149..bc52c60179 100644
--- a/drivers/crypto/cnxk/cn10k_ipsec.h
+++ b/drivers/crypto/cnxk/cn10k_ipsec.h
@@ -20,9 +20,7 @@ struct cn10k_ipsec_sa {
};
/** Pre-populated CPT inst words */
struct cnxk_cpt_inst_tmpl inst;
- uint8_t partial_len;
- uint8_t roundup_len;
- uint8_t roundup_byte;
+ uin
[dpdk-dev] [PATCH v2 0/8] add lookaside IPsec additional features
This series adds cn10k lookaside IPsec AES-CBC-HMAC-SHA1, UDP encapsulation and transport mode support along with common cnxk pmd changes. The functionality has been tested with ipsec-secgw application running in lookaside protocol offload mode. v2: * Fixed release notes Archana Muniganti (2): crypto/cnxk: make IPsec verify functions common common/cnxk: make IPsec defines common Tejasree Kondoj (6): common/cnxk: add hash generation APIs crypto/cnxk: add lookaside IPsec AES-CBC-HMAC-SHA1 support crypto/cnxk: remove redundant code crypto/cnxk: use rlen from CPT result with lookaside crypto/cnxk: support cn10k transport mode crypto/cnxk: support UDP encap with lookaside IPsec doc/guides/cryptodevs/cnxk.rst| 3 + doc/guides/rel_notes/release_21_11.rst| 6 + drivers/common/cnxk/cnxk_security.c | 92 +- drivers/common/cnxk/meson.build | 1 + drivers/common/cnxk/roc_api.h | 4 + drivers/common/cnxk/roc_hash.c| 275 ++ drivers/common/cnxk/roc_hash.h| 16 + drivers/common/cnxk/roc_ie.h | 33 ++- drivers/common/cnxk/roc_ie_on.h | 26 -- drivers/common/cnxk/roc_ie_ot.h | 26 -- drivers/common/cnxk/version.map | 3 + drivers/crypto/cnxk/cn10k_cryptodev_ops.c | 44 +-- drivers/crypto/cnxk/cn10k_ipsec.c | 64 +--- drivers/crypto/cnxk/cn10k_ipsec.h | 6 +- drivers/crypto/cnxk/cn10k_ipsec_la_ops.h | 28 +- drivers/crypto/cnxk/cnxk_cryptodev.h | 2 +- .../crypto/cnxk/cnxk_cryptodev_capabilities.c | 79 - drivers/crypto/cnxk/cnxk_ipsec.h | 114 18 files changed, 620 insertions(+), 202 deletions(-) create mode 100644 drivers/common/cnxk/roc_hash.c create mode 100644 drivers/common/cnxk/roc_hash.h -- 2.27.0
[dpdk-dev] [PATCH v2 5/8] crypto/cnxk: make IPsec verify functions common
From: Archana Muniganti
IPsec verify functions can be made common
Signed-off-by: Archana Muniganti
---
drivers/crypto/cnxk/cn10k_ipsec.c | 116 +-
drivers/crypto/cnxk/cnxk_ipsec.h | 113 +
2 files changed, 114 insertions(+), 115 deletions(-)
diff --git a/drivers/crypto/cnxk/cn10k_ipsec.c
b/drivers/crypto/cnxk/cn10k_ipsec.c
index 98110872a3..5c57cf2818 100644
--- a/drivers/crypto/cnxk/cn10k_ipsec.c
+++ b/drivers/crypto/cnxk/cn10k_ipsec.c
@@ -17,120 +17,6 @@
#include "roc_api.h"
-static int
-ipsec_xform_cipher_verify(struct rte_crypto_sym_xform *xform)
-{
- if (xform->cipher.algo == RTE_CRYPTO_CIPHER_AES_CBC) {
- switch (xform->cipher.key.length) {
- case 16:
- case 24:
- case 32:
- break;
- default:
- return -ENOTSUP;
- }
- return 0;
- }
-
- return -ENOTSUP;
-}
-
-static int
-ipsec_xform_auth_verify(struct rte_crypto_sym_xform *xform)
-{
- uint16_t keylen = xform->auth.key.length;
-
- if (xform->auth.algo == RTE_CRYPTO_AUTH_SHA1_HMAC) {
- if (keylen >= 20 && keylen <= 64)
- return 0;
- }
-
- return -ENOTSUP;
-}
-
-static int
-ipsec_xform_aead_verify(struct rte_security_ipsec_xform *ipsec_xfrm,
- struct rte_crypto_sym_xform *crypto_xfrm)
-{
- if (ipsec_xfrm->direction == RTE_SECURITY_IPSEC_SA_DIR_EGRESS &&
- crypto_xfrm->aead.op != RTE_CRYPTO_AEAD_OP_ENCRYPT)
- return -EINVAL;
-
- if (ipsec_xfrm->direction == RTE_SECURITY_IPSEC_SA_DIR_INGRESS &&
- crypto_xfrm->aead.op != RTE_CRYPTO_AEAD_OP_DECRYPT)
- return -EINVAL;
-
- if (crypto_xfrm->aead.algo == RTE_CRYPTO_AEAD_AES_GCM) {
- switch (crypto_xfrm->aead.key.length) {
- case ROC_CPT_AES128_KEY_LEN:
- case ROC_CPT_AES192_KEY_LEN:
- case ROC_CPT_AES256_KEY_LEN:
- break;
- default:
- return -EINVAL;
- }
- return 0;
- }
-
- return -ENOTSUP;
-}
-
-static int
-cn10k_ipsec_xform_verify(struct rte_security_ipsec_xform *ipsec_xfrm,
-struct rte_crypto_sym_xform *crypto_xfrm)
-{
- struct rte_crypto_sym_xform *auth_xform, *cipher_xform;
- int ret;
-
- if ((ipsec_xfrm->direction != RTE_SECURITY_IPSEC_SA_DIR_INGRESS) &&
- (ipsec_xfrm->direction != RTE_SECURITY_IPSEC_SA_DIR_EGRESS))
- return -EINVAL;
-
- if ((ipsec_xfrm->proto != RTE_SECURITY_IPSEC_SA_PROTO_ESP) &&
- (ipsec_xfrm->proto != RTE_SECURITY_IPSEC_SA_PROTO_AH))
- return -EINVAL;
-
- if ((ipsec_xfrm->mode != RTE_SECURITY_IPSEC_SA_MODE_TRANSPORT) &&
- (ipsec_xfrm->mode != RTE_SECURITY_IPSEC_SA_MODE_TUNNEL))
- return -EINVAL;
-
- if ((ipsec_xfrm->tunnel.type != RTE_SECURITY_IPSEC_TUNNEL_IPV4) &&
- (ipsec_xfrm->tunnel.type != RTE_SECURITY_IPSEC_TUNNEL_IPV6))
- return -EINVAL;
-
- if (crypto_xfrm->type == RTE_CRYPTO_SYM_XFORM_AEAD)
- return ipsec_xform_aead_verify(ipsec_xfrm, crypto_xfrm);
-
- if (crypto_xfrm->next == NULL)
- return -EINVAL;
-
- if (ipsec_xfrm->direction == RTE_SECURITY_IPSEC_SA_DIR_INGRESS) {
- /* Ingress */
- if (crypto_xfrm->type != RTE_CRYPTO_SYM_XFORM_AUTH ||
- crypto_xfrm->next->type != RTE_CRYPTO_SYM_XFORM_CIPHER)
- return -EINVAL;
- auth_xform = crypto_xfrm;
- cipher_xform = crypto_xfrm->next;
- } else {
- /* Egress */
- if (crypto_xfrm->type != RTE_CRYPTO_SYM_XFORM_CIPHER ||
- crypto_xfrm->next->type != RTE_CRYPTO_SYM_XFORM_AUTH)
- return -EINVAL;
- cipher_xform = crypto_xfrm;
- auth_xform = crypto_xfrm->next;
- }
-
- ret = ipsec_xform_cipher_verify(cipher_xform);
- if (ret)
- return ret;
-
- ret = ipsec_xform_auth_verify(auth_xform);
- if (ret)
- return ret;
-
- return 0;
-}
-
static uint64_t
ipsec_cpt_inst_w7_get(struct roc_cpt *roc_cpt, void *sa)
{
@@ -245,7 +131,7 @@ cn10k_ipsec_session_create(void *dev,
return -EPERM;
}
- ret = cn10k_ipsec_xform_verify(ipsec_xfrm, crypto_xfrm);
+ ret = cnxk_ipsec_xform_verify(ipsec_xfrm, crypto_xfrm);
if (ret)
return ret;
diff --git a/drivers/crypto/cnxk/cnxk_ipsec.h b/drivers/crypto/cnxk/cnxk_ipsec.h
index f6897a0e14..d1eb74ebbe 100644
--- a/drivers/crypto/cnxk/cnxk_ipsec.h
+++ b/drivers/crypto/cnxk/cnxk_ipsec.h
@@ -17,4 +17,117 @@ struct cnxk_cpt_inst_tmpl {
uint64_t w7;
};
+static inline int
+ipsec_x
[dpdk-dev] [PATCH v2 1/8] common/cnxk: add hash generation APIs
Adding functions for hash generation that can be used
in hmac opad/ipad calculation.
Signed-off-by: Tejasree Kondoj
---
drivers/common/cnxk/meson.build | 1 +
drivers/common/cnxk/roc_api.h | 3 +
drivers/common/cnxk/roc_hash.c | 275
drivers/common/cnxk/roc_hash.h | 16 ++
drivers/common/cnxk/version.map | 3 +
5 files changed, 298 insertions(+)
create mode 100644 drivers/common/cnxk/roc_hash.c
create mode 100644 drivers/common/cnxk/roc_hash.h
diff --git a/drivers/common/cnxk/meson.build b/drivers/common/cnxk/meson.build
index 6a7849f31c..8a551d15d6 100644
--- a/drivers/common/cnxk/meson.build
+++ b/drivers/common/cnxk/meson.build
@@ -19,6 +19,7 @@ sources = files(
'roc_cpt.c',
'roc_cpt_debug.c',
'roc_dev.c',
+'roc_hash.c',
'roc_idev.c',
'roc_irq.c',
'roc_mbox.c',
diff --git a/drivers/common/cnxk/roc_api.h b/drivers/common/cnxk/roc_api.h
index 52cb2f2d79..9c06cfee9a 100644
--- a/drivers/common/cnxk/roc_api.h
+++ b/drivers/common/cnxk/roc_api.h
@@ -125,4 +125,7 @@
#include "roc_ie_ot.h"
#include "roc_se.h"
+/* HASH computation */
+#include "roc_hash.h"
+
#endif /* _ROC_API_H_ */
diff --git a/drivers/common/cnxk/roc_hash.c b/drivers/common/cnxk/roc_hash.c
new file mode 100644
index 00..092286e41e
--- /dev/null
+++ b/drivers/common/cnxk/roc_hash.c
@@ -0,0 +1,275 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright (c) 2021 Marvell.
+ */
+
+#include "roc_api.h"
+
+#define lrot32(bits, word) (((word) << (bits)) | ((word) >> (32 - (bits
+#define rrot32(bits, word) lrot32(32 - (bits), word)
+#define lrot64(bits, word) (((word) << (bits)) | ((word) >> (64 - (bits
+#define rrot64(bits, word) lrot64(64 - (bits), word)
+
+/*
+ * Compute a partial hash with the assumption that msg is the first block.
+ * Based on implementation from RFC 3174
+ */
+void
+roc_hash_sha1_gen(uint8_t *msg, uint32_t *hash)
+{
+ const uint32_t _K[] = {/* Round Constants defined in SHA-1 */
+ 0x5A827999, 0x6ED9EBA1, 0x8F1BBCDC, 0xCA62C1D6};
+
+ const uint32_t _H[] = {/* Initial Hash constants defined in SHA-1 */
+ 0x67452301, 0xEFCDAB89, 0x98BADCFE, 0x10325476,
+ 0xC3D2E1F0};
+ int i;
+ uint32_t temp; /* Temporary word value */
+ uint32_t W[80]; /* Word sequence */
+ uint32_t A, B, C, D, E; /* Word buffers */
+
+ /* Initialize the first 16 words in the array W */
+ memcpy(&W[0], msg, 16 * sizeof(W[0]));
+
+ for (i = 0; i < 16; i++)
+ W[i] = htobe32(W[i]);
+
+ for (i = 16; i < 80; i++)
+ W[i] = lrot32(1, W[i - 3] ^ W[i - 8] ^ W[i - 14] ^ W[i - 16]);
+
+ A = _H[0];
+ B = _H[1];
+ C = _H[2];
+ D = _H[3];
+ E = _H[4];
+
+ for (i = 0; i < 80; i++) {
+ if (i >= 0 && i <= 19)
+ temp = ((B & C) | ((~B) & D)) + _K[0];
+ else if (i >= 20 && i <= 39)
+ temp = (B ^ C ^ D) + _K[1];
+ else if (i >= 40 && i <= 59)
+ temp = ((B & C) | (B & D) | (C & D)) + _K[2];
+ else if (i >= 60 && i <= 79)
+ temp = (B ^ C ^ D) + _K[3];
+
+ temp = lrot32(5, A) + temp + E + W[i];
+ E = D;
+ D = C;
+ C = lrot32(30, B);
+ B = A;
+ A = temp;
+ }
+
+ A += _H[0];
+ B += _H[1];
+ C += _H[2];
+ D += _H[3];
+ E += _H[4];
+ hash[0] = htobe32(A);
+ hash[1] = htobe32(B);
+ hash[2] = htobe32(C);
+ hash[3] = htobe32(D);
+ hash[4] = htobe32(E);
+}
+
+/*
+ * Compute a partial hash with the assumption that msg is the first block.
+ * Based on implementation from RFC 3174
+ */
+void
+roc_hash_sha256_gen(uint8_t *msg, uint32_t *hash)
+{
+ const uint32_t _K[] = {
+ /* Round Constants defined in SHA-256 */
+ 0x428a2f98, 0x71374491, 0xb5c0fbcf, 0xe9b5dba5, 0x3956c25b,
+ 0x59f111f1, 0x923f82a4, 0xab1c5ed5, 0xd807aa98, 0x12835b01,
+ 0x243185be, 0x550c7dc3, 0x72be5d74, 0x80deb1fe, 0x9bdc06a7,
+ 0xc19bf174, 0xe49b69c1, 0xefbe4786, 0x0fc19dc6, 0x240ca1cc,
+ 0x2de92c6f, 0x4a7484aa, 0x5cb0a9dc, 0x76f988da, 0x983e5152,
+ 0xa831c66d, 0xb00327c8, 0xbf597fc7, 0xc6e00bf3, 0xd5a79147,
+ 0x06ca6351, 0x14292967, 0x27b70a85, 0x2e1b2138, 0x4d2c6dfc,
+ 0x53380d13, 0x650a7354, 0x766a0abb, 0x81c2c92e, 0x92722c85,
+ 0xa2bfe8a1, 0xa81a664b, 0xc24b8b70, 0xc76c51a3, 0xd192e819,
+ 0xd6990624, 0xf40e3585, 0x106aa070, 0x19a4c116, 0x1e376c08,
+ 0x2748774c, 0x34b0bcb5, 0x391c0cb3, 0x4ed8aa4a, 0x5b9cca4f,
+ 0x682e6ff3, 0x748f82ee, 0x78a5636f, 0x84c87814, 0x8cc70208,
+ 0x90b
[dpdk-dev] [PATCH v2 2/8] crypto/cnxk: add lookaside IPsec AES-CBC-HMAC-SHA1 support
Adding lookaside IPsec AES-CBC-HMAC-SHA1 support to cnxk driver.
Signed-off-by: Tejasree Kondoj
---
doc/guides/cryptodevs/cnxk.rst| 1 +
doc/guides/rel_notes/release_21_11.rst| 4 ++
drivers/common/cnxk/cnxk_security.c | 68 ++-
drivers/crypto/cnxk/cn10k_ipsec.c | 63 -
.../crypto/cnxk/cnxk_cryptodev_capabilities.c | 44
5 files changed, 176 insertions(+), 4 deletions(-)
diff --git a/doc/guides/cryptodevs/cnxk.rst b/doc/guides/cryptodevs/cnxk.rst
index 98c7118d68..a40295c087 100644
--- a/doc/guides/cryptodevs/cnxk.rst
+++ b/doc/guides/cryptodevs/cnxk.rst
@@ -231,6 +231,7 @@ Features supported
* ESP
* Tunnel mode
* AES-128/192/256-GCM
+* AES-128/192/256-CBC-SHA1-HMAC
Limitations
---
diff --git a/doc/guides/rel_notes/release_21_11.rst
b/doc/guides/rel_notes/release_21_11.rst
index d707a554ef..130d676a11 100644
--- a/doc/guides/rel_notes/release_21_11.rst
+++ b/doc/guides/rel_notes/release_21_11.rst
@@ -55,6 +55,10 @@ New Features
Also, make sure to start the actual text at the margin.
===
+* **Updated Marvell cn10k_crypto PMD.**
+
+ * Added aes-cbc sha1-hmac in lookaside protocol (IPsec).
+
Removed Items
-
diff --git a/drivers/common/cnxk/cnxk_security.c
b/drivers/common/cnxk/cnxk_security.c
index 6c6728f570..fe64e70c81 100644
--- a/drivers/common/cnxk/cnxk_security.c
+++ b/drivers/common/cnxk/cnxk_security.c
@@ -6,12 +6,43 @@
#include "cnxk_security.h"
+static void
+ipsec_hmac_opad_ipad_gen(struct rte_crypto_sym_xform *auth_xform,
+uint8_t *hmac_opad_ipad)
+{
+ const uint8_t *key = auth_xform->auth.key.data;
+ uint32_t length = auth_xform->auth.key.length;
+ uint8_t opad[128] = {[0 ... 127] = 0x5c};
+ uint8_t ipad[128] = {[0 ... 127] = 0x36};
+ uint32_t i;
+
+ /* HMAC OPAD and IPAD */
+ for (i = 0; i < 127 && i < length; i++) {
+ opad[i] = opad[i] ^ key[i];
+ ipad[i] = ipad[i] ^ key[i];
+ }
+
+ /* Precompute hash of HMAC OPAD and IPAD to avoid
+* per packet computation
+*/
+ switch (auth_xform->auth.algo) {
+ case RTE_CRYPTO_AUTH_SHA1_HMAC:
+ roc_hash_sha1_gen(opad, (uint32_t *)&hmac_opad_ipad[0]);
+ roc_hash_sha1_gen(ipad, (uint32_t *)&hmac_opad_ipad[24]);
+ break;
+ default:
+ break;
+ }
+}
+
static int
ot_ipsec_sa_common_param_fill(union roc_ot_ipsec_sa_word2 *w2,
uint8_t *cipher_key, uint8_t *salt_key,
+ uint8_t *hmac_opad_ipad,
struct rte_security_ipsec_xform *ipsec_xfrm,
struct rte_crypto_sym_xform *crypto_xfrm)
{
+ struct rte_crypto_sym_xform *auth_xfrm, *cipher_xfrm;
const uint8_t *key;
uint32_t *tmp_salt;
uint64_t *tmp_key;
@@ -21,9 +52,13 @@ ot_ipsec_sa_common_param_fill(union roc_ot_ipsec_sa_word2
*w2,
switch (ipsec_xfrm->direction) {
case RTE_SECURITY_IPSEC_SA_DIR_INGRESS:
w2->s.dir = ROC_IE_OT_SA_DIR_INBOUND;
+ auth_xfrm = crypto_xfrm;
+ cipher_xfrm = crypto_xfrm->next;
break;
case RTE_SECURITY_IPSEC_SA_DIR_EGRESS:
w2->s.dir = ROC_IE_OT_SA_DIR_OUTBOUND;
+ cipher_xfrm = crypto_xfrm;
+ auth_xfrm = crypto_xfrm->next;
break;
default:
return -EINVAL;
@@ -70,7 +105,32 @@ ot_ipsec_sa_common_param_fill(union roc_ot_ipsec_sa_word2
*w2,
return -ENOTSUP;
}
} else {
- return -ENOTSUP;
+ switch (cipher_xfrm->cipher.algo) {
+ case RTE_CRYPTO_CIPHER_AES_CBC:
+ w2->s.enc_type = ROC_IE_OT_SA_ENC_AES_CBC;
+ break;
+ default:
+ return -ENOTSUP;
+ }
+
+ switch (auth_xfrm->auth.algo) {
+ case RTE_CRYPTO_AUTH_SHA1_HMAC:
+ w2->s.auth_type = ROC_IE_OT_SA_AUTH_SHA1;
+ break;
+ default:
+ return -ENOTSUP;
+ }
+
+ key = cipher_xfrm->cipher.key.data;
+ length = cipher_xfrm->cipher.key.length;
+
+ ipsec_hmac_opad_ipad_gen(auth_xfrm, hmac_opad_ipad);
+
+ tmp_key = (uint64_t *)hmac_opad_ipad;
+ for (i = 0;
+i < (int)(ROC_CTX_MAX_OPAD_IPAD_LEN / sizeof(uint64_t));
+i++)
+ tmp_key[i] = rte_be_to_cpu_64(tmp_key[i]);
}
/* Set encapsulation type */
@@ -129,7 +189,8 @@ cnxk_ot_ipsec_inb_sa_fill(struct roc_ot_ipsec_inb_sa *sa,
w2.u64 = 0;
rc = ot_ipsec_sa_
[dpdk-dev] [PATCH v2 3/8] crypto/cnxk: remove redundant code
Removing redundant code in cn10k lookaside IPsec.
Signed-off-by: Tejasree Kondoj
---
drivers/crypto/cnxk/cn10k_ipsec.c | 5 -
drivers/crypto/cnxk/cn10k_ipsec.h | 2 --
drivers/crypto/cnxk/cnxk_cryptodev_capabilities.c | 2 --
3 files changed, 9 deletions(-)
diff --git a/drivers/crypto/cnxk/cn10k_ipsec.c
b/drivers/crypto/cnxk/cn10k_ipsec.c
index 408a682b21..944e0a7e3b 100644
--- a/drivers/crypto/cnxk/cn10k_ipsec.c
+++ b/drivers/crypto/cnxk/cn10k_ipsec.c
@@ -270,9 +270,6 @@ cn10k_sec_session_create(void *device, struct
rte_security_session_conf *conf,
if (conf->action_type != RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL)
return -EINVAL;
- if (rte_security_dynfield_register() < 0)
- return -ENOTSUP;
-
if (rte_mempool_get(mempool, (void **)&priv)) {
plt_err("Could not allocate security session private data");
return -ENOMEM;
@@ -280,8 +277,6 @@ cn10k_sec_session_create(void *device, struct
rte_security_session_conf *conf,
set_sec_session_private_data(sess, priv);
- priv->userdata = conf->userdata;
-
if (conf->protocol != RTE_SECURITY_PROTOCOL_IPSEC) {
ret = -ENOTSUP;
goto mempool_put;
diff --git a/drivers/crypto/cnxk/cn10k_ipsec.h
b/drivers/crypto/cnxk/cn10k_ipsec.h
index 668282f7aa..c30492e149 100644
--- a/drivers/crypto/cnxk/cn10k_ipsec.h
+++ b/drivers/crypto/cnxk/cn10k_ipsec.h
@@ -27,8 +27,6 @@ struct cn10k_ipsec_sa {
struct cn10k_sec_session {
struct cn10k_ipsec_sa sa;
- void *userdata;
- /**< Userdata registered by the application */
} __rte_cache_aligned;
void cn10k_sec_ops_override(void);
diff --git a/drivers/crypto/cnxk/cnxk_cryptodev_capabilities.c
b/drivers/crypto/cnxk/cnxk_cryptodev_capabilities.c
index 47274b2c24..9430ca5d00 100644
--- a/drivers/crypto/cnxk/cnxk_cryptodev_capabilities.c
+++ b/drivers/crypto/cnxk/cnxk_cryptodev_capabilities.c
@@ -810,7 +810,6 @@ static const struct rte_security_capability
sec_caps_templ[] = {
.options = { 0 }
},
.crypto_capabilities = NULL,
- .ol_flags = RTE_SECURITY_TX_OLOAD_NEED_MDATA
},
{ /* IPsec Lookaside Protocol ESP Tunnel Egress */
.action = RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL,
@@ -822,7 +821,6 @@ static const struct rte_security_capability
sec_caps_templ[] = {
.options = { 0 }
},
.crypto_capabilities = NULL,
- .ol_flags = RTE_SECURITY_TX_OLOAD_NEED_MDATA
},
{
.action = RTE_SECURITY_ACTION_TYPE_NONE
--
2.27.0
[dpdk-dev] [PATCH v2 6/8] crypto/cnxk: support cn10k transport mode
Adding support for cn10k lookaside IPsec transport mode.
Signed-off-by: Tejasree Kondoj
---
doc/guides/cryptodevs/cnxk.rst| 1 +
doc/guides/rel_notes/release_21_11.rst| 1 +
drivers/crypto/cnxk/cnxk_cryptodev.h | 2 +-
.../crypto/cnxk/cnxk_cryptodev_capabilities.c | 22 +++
drivers/crypto/cnxk/cnxk_ipsec.h | 3 ++-
5 files changed, 27 insertions(+), 2 deletions(-)
diff --git a/doc/guides/cryptodevs/cnxk.rst b/doc/guides/cryptodevs/cnxk.rst
index a40295c087..0dd71135da 100644
--- a/doc/guides/cryptodevs/cnxk.rst
+++ b/doc/guides/cryptodevs/cnxk.rst
@@ -230,6 +230,7 @@ Features supported
* IPv4
* ESP
* Tunnel mode
+* Transport mode
* AES-128/192/256-GCM
* AES-128/192/256-CBC-SHA1-HMAC
diff --git a/doc/guides/rel_notes/release_21_11.rst
b/doc/guides/rel_notes/release_21_11.rst
index 130d676a11..0a024dda8f 100644
--- a/doc/guides/rel_notes/release_21_11.rst
+++ b/doc/guides/rel_notes/release_21_11.rst
@@ -58,6 +58,7 @@ New Features
* **Updated Marvell cn10k_crypto PMD.**
* Added aes-cbc sha1-hmac in lookaside protocol (IPsec).
+ * Added transport mode in lookaside protocol (IPsec).
Removed Items
diff --git a/drivers/crypto/cnxk/cnxk_cryptodev.h
b/drivers/crypto/cnxk/cnxk_cryptodev.h
index b3856f7eaa..8e051fa0fa 100644
--- a/drivers/crypto/cnxk/cnxk_cryptodev.h
+++ b/drivers/crypto/cnxk/cnxk_cryptodev.h
@@ -12,7 +12,7 @@
#define CNXK_CPT_MAX_CAPS 34
#define CNXK_SEC_CRYPTO_MAX_CAPS 4
-#define CNXK_SEC_MAX_CAPS 3
+#define CNXK_SEC_MAX_CAPS 5
#define CNXK_AE_EC_ID_MAX 8
/**
* Device private data
diff --git a/drivers/crypto/cnxk/cnxk_cryptodev_capabilities.c
b/drivers/crypto/cnxk/cnxk_cryptodev_capabilities.c
index 9430ca5d00..05bffa9759 100644
--- a/drivers/crypto/cnxk/cnxk_cryptodev_capabilities.c
+++ b/drivers/crypto/cnxk/cnxk_cryptodev_capabilities.c
@@ -822,6 +822,28 @@ static const struct rte_security_capability
sec_caps_templ[] = {
},
.crypto_capabilities = NULL,
},
+ { /* IPsec Lookaside Protocol ESP Transport Ingress */
+ .action = RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL,
+ .protocol = RTE_SECURITY_PROTOCOL_IPSEC,
+ .ipsec = {
+ .proto = RTE_SECURITY_IPSEC_SA_PROTO_ESP,
+ .mode = RTE_SECURITY_IPSEC_SA_MODE_TRANSPORT,
+ .direction = RTE_SECURITY_IPSEC_SA_DIR_INGRESS,
+ .options = { 0 },
+ },
+ .crypto_capabilities = NULL,
+ },
+ { /* IPsec Lookaside Protocol ESP Transport Egress */
+ .action = RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL,
+ .protocol = RTE_SECURITY_PROTOCOL_IPSEC,
+ .ipsec = {
+ .proto = RTE_SECURITY_IPSEC_SA_PROTO_ESP,
+ .mode = RTE_SECURITY_IPSEC_SA_MODE_TRANSPORT,
+ .direction = RTE_SECURITY_IPSEC_SA_DIR_EGRESS,
+ .options = { 0 },
+ },
+ .crypto_capabilities = NULL,
+ },
{
.action = RTE_SECURITY_ACTION_TYPE_NONE
}
diff --git a/drivers/crypto/cnxk/cnxk_ipsec.h b/drivers/crypto/cnxk/cnxk_ipsec.h
index d1eb74ebbe..ff396179ca 100644
--- a/drivers/crypto/cnxk/cnxk_ipsec.h
+++ b/drivers/crypto/cnxk/cnxk_ipsec.h
@@ -98,7 +98,8 @@ cnxk_ipsec_xform_verify(struct rte_security_ipsec_xform
*ipsec_xform,
(ipsec_xform->mode != RTE_SECURITY_IPSEC_SA_MODE_TUNNEL))
return -EINVAL;
- if ((ipsec_xform->tunnel.type != RTE_SECURITY_IPSEC_TUNNEL_IPV4) &&
+ if ((ipsec_xform->mode == RTE_SECURITY_IPSEC_SA_MODE_TUNNEL) &&
+ (ipsec_xform->tunnel.type != RTE_SECURITY_IPSEC_TUNNEL_IPV4) &&
(ipsec_xform->tunnel.type != RTE_SECURITY_IPSEC_TUNNEL_IPV6))
return -EINVAL;
--
2.27.0
[dpdk-dev] [PATCH v2 7/8] crypto/cnxk: support UDP encap with lookaside IPsec
Adding support for UDP encapsulation in lookaside IPsec.
Signed-off-by: Tejasree Kondoj
---
doc/guides/cryptodevs/cnxk.rst| 1 +
doc/guides/rel_notes/release_21_11.rst| 1 +
drivers/crypto/cnxk/cnxk_cryptodev_capabilities.c | 11 ++-
3 files changed, 12 insertions(+), 1 deletion(-)
diff --git a/doc/guides/cryptodevs/cnxk.rst b/doc/guides/cryptodevs/cnxk.rst
index 0dd71135da..1eb72282db 100644
--- a/doc/guides/cryptodevs/cnxk.rst
+++ b/doc/guides/cryptodevs/cnxk.rst
@@ -231,6 +231,7 @@ Features supported
* ESP
* Tunnel mode
* Transport mode
+* UDP Encapsulation
* AES-128/192/256-GCM
* AES-128/192/256-CBC-SHA1-HMAC
diff --git a/doc/guides/rel_notes/release_21_11.rst
b/doc/guides/rel_notes/release_21_11.rst
index 0a024dda8f..70dd1c52f7 100644
--- a/doc/guides/rel_notes/release_21_11.rst
+++ b/doc/guides/rel_notes/release_21_11.rst
@@ -59,6 +59,7 @@ New Features
* Added aes-cbc sha1-hmac in lookaside protocol (IPsec).
* Added transport mode in lookaside protocol (IPsec).
+ * Added UDP encapsulation in lookaside protocol (IPsec).
Removed Items
diff --git a/drivers/crypto/cnxk/cnxk_cryptodev_capabilities.c
b/drivers/crypto/cnxk/cnxk_cryptodev_capabilities.c
index 05bffa9759..c4f7824332 100644
--- a/drivers/crypto/cnxk/cnxk_cryptodev_capabilities.c
+++ b/drivers/crypto/cnxk/cnxk_cryptodev_capabilities.c
@@ -907,6 +907,12 @@ sec_crypto_caps_populate(struct rte_cryptodev_capabilities
cnxk_caps[],
sec_caps_add(cnxk_caps, &cur_pos, caps_end, RTE_DIM(caps_end));
}
+static void
+cnxk_sec_caps_update(struct rte_security_capability *sec_cap)
+{
+ sec_cap->ipsec.options.udp_encap = 1;
+}
+
void
cnxk_cpt_caps_populate(struct cnxk_cpt_vf *vf)
{
@@ -918,8 +924,11 @@ cnxk_cpt_caps_populate(struct cnxk_cpt_vf *vf)
PLT_STATIC_ASSERT(RTE_DIM(sec_caps_templ) <= RTE_DIM(vf->sec_caps));
memcpy(vf->sec_caps, sec_caps_templ, sizeof(sec_caps_templ));
- for (i = 0; i < RTE_DIM(sec_caps_templ) - 1; i++)
+ for (i = 0; i < RTE_DIM(sec_caps_templ) - 1; i++) {
vf->sec_caps[i].crypto_capabilities = vf->sec_crypto_caps;
+
+ cnxk_sec_caps_update(&vf->sec_caps[i]);
+ }
}
const struct rte_security_capability *
--
2.27.0
[dpdk-dev] [PATCH v2 8/8] common/cnxk: make IPsec defines common
From: Archana Muniganti
Make IPsec defines common and remove redundant macros.
Signed-off-by: Archana Muniganti
---
drivers/common/cnxk/cnxk_security.c | 24 +
drivers/common/cnxk/roc_api.h | 1 +
drivers/common/cnxk/roc_ie.h | 33 ---
drivers/common/cnxk/roc_ie_on.h | 26 --
drivers/common/cnxk/roc_ie_ot.h | 26 --
drivers/crypto/cnxk/cn10k_cryptodev_ops.c | 4 ++-
6 files changed, 41 insertions(+), 73 deletions(-)
diff --git a/drivers/common/cnxk/cnxk_security.c
b/drivers/common/cnxk/cnxk_security.c
index fe64e70c81..4f7fd1b3a0 100644
--- a/drivers/common/cnxk/cnxk_security.c
+++ b/drivers/common/cnxk/cnxk_security.c
@@ -6,6 +6,8 @@
#include "cnxk_security.h"
+#include "roc_api.h"
+
static void
ipsec_hmac_opad_ipad_gen(struct rte_crypto_sym_xform *auth_xform,
uint8_t *hmac_opad_ipad)
@@ -51,12 +53,12 @@ ot_ipsec_sa_common_param_fill(union roc_ot_ipsec_sa_word2
*w2,
/* Set direction */
switch (ipsec_xfrm->direction) {
case RTE_SECURITY_IPSEC_SA_DIR_INGRESS:
- w2->s.dir = ROC_IE_OT_SA_DIR_INBOUND;
+ w2->s.dir = ROC_IE_SA_DIR_INBOUND;
auth_xfrm = crypto_xfrm;
cipher_xfrm = crypto_xfrm->next;
break;
case RTE_SECURITY_IPSEC_SA_DIR_EGRESS:
- w2->s.dir = ROC_IE_OT_SA_DIR_OUTBOUND;
+ w2->s.dir = ROC_IE_SA_DIR_OUTBOUND;
cipher_xfrm = crypto_xfrm;
auth_xfrm = crypto_xfrm->next;
break;
@@ -67,10 +69,10 @@ ot_ipsec_sa_common_param_fill(union roc_ot_ipsec_sa_word2
*w2,
/* Set protocol - ESP vs AH */
switch (ipsec_xfrm->proto) {
case RTE_SECURITY_IPSEC_SA_PROTO_ESP:
- w2->s.protocol = ROC_IE_OT_SA_PROTOCOL_ESP;
+ w2->s.protocol = ROC_IE_SA_PROTOCOL_ESP;
break;
case RTE_SECURITY_IPSEC_SA_PROTO_AH:
- w2->s.protocol = ROC_IE_OT_SA_PROTOCOL_AH;
+ w2->s.protocol = ROC_IE_SA_PROTOCOL_AH;
break;
default:
return -EINVAL;
@@ -79,10 +81,10 @@ ot_ipsec_sa_common_param_fill(union roc_ot_ipsec_sa_word2
*w2,
/* Set mode - transport vs tunnel */
switch (ipsec_xfrm->mode) {
case RTE_SECURITY_IPSEC_SA_MODE_TRANSPORT:
- w2->s.mode = ROC_IE_OT_SA_MODE_TRANSPORT;
+ w2->s.mode = ROC_IE_SA_MODE_TRANSPORT;
break;
case RTE_SECURITY_IPSEC_SA_MODE_TUNNEL:
- w2->s.mode = ROC_IE_OT_SA_MODE_TUNNEL;
+ w2->s.mode = ROC_IE_SA_MODE_TUNNEL;
break;
default:
return -EINVAL;
@@ -147,13 +149,13 @@ ot_ipsec_sa_common_param_fill(union roc_ot_ipsec_sa_word2
*w2,
switch (length) {
case ROC_CPT_AES128_KEY_LEN:
- w2->s.aes_key_len = ROC_IE_OT_SA_AES_KEY_LEN_128;
+ w2->s.aes_key_len = ROC_IE_SA_AES_KEY_LEN_128;
break;
case ROC_CPT_AES192_KEY_LEN:
- w2->s.aes_key_len = ROC_IE_OT_SA_AES_KEY_LEN_192;
+ w2->s.aes_key_len = ROC_IE_SA_AES_KEY_LEN_192;
break;
case ROC_CPT_AES256_KEY_LEN:
- w2->s.aes_key_len = ROC_IE_OT_SA_AES_KEY_LEN_256;
+ w2->s.aes_key_len = ROC_IE_SA_AES_KEY_LEN_256;
break;
default:
return -EINVAL;
@@ -271,7 +273,7 @@ cnxk_ot_ipsec_outb_sa_fill(struct roc_ot_ipsec_outb_sa *sa,
/* Tunnel header info */
switch (tunnel->type) {
case RTE_SECURITY_IPSEC_TUNNEL_IPV4:
- sa->w2.s.outer_ip_ver = ROC_IE_OT_SA_IP_VERSION_4;
+ sa->w2.s.outer_ip_ver = ROC_IE_SA_IP_VERSION_4;
memcpy(&sa->outer_hdr.ipv4.src_addr, &tunnel->ipv4.src_ip,
sizeof(struct in_addr));
memcpy(&sa->outer_hdr.ipv4.dst_addr, &tunnel->ipv4.dst_ip,
@@ -302,7 +304,7 @@ cnxk_ot_ipsec_outb_sa_fill(struct roc_ot_ipsec_outb_sa *sa,
}
break;
case RTE_SECURITY_IPSEC_TUNNEL_IPV6:
- sa->w2.s.outer_ip_ver = ROC_IE_OT_SA_IP_VERSION_6;
+ sa->w2.s.outer_ip_ver = ROC_IE_SA_IP_VERSION_6;
memcpy(&sa->outer_hdr.ipv6.src_addr, &tunnel->ipv6.src_addr,
sizeof(struct in6_addr));
memcpy(&sa->outer_hdr.ipv6.dst_addr, &tunnel->ipv6.dst_addr,
diff --git a/drivers/common/cnxk/roc_api.h b/drivers/common/cnxk/roc_api.h
index 9c06cfee9a..7dec8453b4 100644
--- a/drivers/common/cnxk/roc_api.h
+++ b/drivers/common/cnxk/roc_api.h
@@ -121,6 +121,7 @@
/* CPT microcode */
#include "roc_ae.h"
#include "roc_ae_fpm_tables.h"
+#include "roc_ie.h"
#include "roc_ie_on.h"
#include "roc_ie_ot.h"
#include "roc_se.h"
diff --git a/drivers/common/cnxk/roc_ie.h b/drivers/common/cnxk/roc_ie.h
index a330
[dpdk-dev] [PATCH] eal: fix memory leak when saving arguments
This patch fixes a memleak which was reported in Bugzilla within the
eal_save_args function. This was caused by the function mistakenly
adding -- to the eal args instead of breaking beforehand.
Bugzilla ID: 722
Fixes: 293c53d8b23c ("eal: add telemetry callbacks")
Reported-by: Zhihong Peng
Signed-off-by: Conor Walsh
Signed-off-by: Conor Fogarty
---
lib/eal/common/eal_common_options.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/lib/eal/common/eal_common_options.c
b/lib/eal/common/eal_common_options.c
index ff5861b5f3..bee716a714 100644
--- a/lib/eal/common/eal_common_options.c
+++ b/lib/eal/common/eal_common_options.c
@@ -229,9 +229,9 @@ eal_save_args(int argc, char **argv)
return -1;
for (i = 0; i < argc; i++) {
- eal_args[i] = strdup(argv[i]);
if (strcmp(argv[i], "--") == 0)
break;
+ eal_args[i] = strdup(argv[i]);
}
eal_args[i++] = NULL; /* always finish with NULL */
--
2.25.1
Re: [dpdk-dev] [PATCH] doc: announce change in vfio dma mapping
On 9/1/2021 2:41 AM, Ding, Xuan wrote: > Hi Ferruh, > >> -Original Message- >> From: Yigit, Ferruh >> Sent: Wednesday, September 1, 2021 12:01 AM >> To: Ding, Xuan ; dev@dpdk.org; Burakov, Anatoly >> >> Cc: maxime.coque...@redhat.com; Xia, Chenbo ; Hu, >> Jiayu ; Richardson, Bruce >> Subject: Re: [PATCH] doc: announce change in vfio dma mapping >> >> On 8/31/2021 2:10 PM, Xuan Ding wrote: >>> Currently, the VFIO subsystem will compact adjacent DMA regions for the >>> purposes of saving space in the internal list of mappings. This has a >>> side effect of compacting two separate mappings that just happen to be >>> adjacent in memory. Since VFIO implementation on IA platforms also does >>> not allow partial unmapping of memory mapped for DMA, the current >> DPDK >>> VFIO implementation will prevent unmapping of accidentally adjacent >>> maps even though it could have been unmapped [1]. >>> >>> The proper fix for this issue is to change the VFIO DMA mapping API to >>> also include page size, and always map memory page-by-page. >>> >>> [1] https://mails.dpdk.org/archives/dev/2021-July/213493.html >>> >>> Signed-off-by: Xuan Ding >>> --- >>> doc/guides/rel_notes/deprecation.rst | 3 +++ >>> 1 file changed, 3 insertions(+) >>> >>> diff --git a/doc/guides/rel_notes/deprecation.rst >> b/doc/guides/rel_notes/deprecation.rst >>> index 76a4abfd6b..1234420caf 100644 >>> --- a/doc/guides/rel_notes/deprecation.rst >>> +++ b/doc/guides/rel_notes/deprecation.rst >>> @@ -287,3 +287,6 @@ Deprecation Notices >>>reserved bytes to 2 (from 3), and use 1 byte to indicate warnings and >> other >>>information from the crypto/security operation. This field will be used >>> to >>>communicate events such as soft expiry with IPsec in lookaside mode. >>> + >>> +* vfio: the functions `rte_vfio_container_dma_map` will be amended to >>> + include page size. This change is targeted for DPDK 22.02. >>> >> >> Is this means adding a new parameter to API? >> If so this is an ABI/API break and we can't do this change in the 22.02. > > Our original plan is add a new parameter in order not to use a new function > name, so you mean, any changes to the API can only be done in the LTS version? > If so, we can only add a new API and retire the old one in 22.11. > We can add a new API anytime. Adding new parameter to an existing API can be done on the ABI break release. You can add the new API in this release, and start using it. And mark the old API as deprecated in this release. This lets existing binaries to keep using it, but app needs to switch to new API for compilation. Old API can be removed on 22.11, and you will need a deprecation notice before 22.11 for it. Is above plan works for you?
Re: [dpdk-dev] [PATCH] eal: fix memory leak when saving arguments
On Wed, Sep 01, 2021 at 09:38:47AM +, Conor Walsh wrote:
> This patch fixes a memleak which was reported in Bugzilla within the
> eal_save_args function. This was caused by the function mistakenly
> adding -- to the eal args instead of breaking beforehand.
>
> Bugzilla ID: 722
> Fixes: 293c53d8b23c ("eal: add telemetry callbacks")
>
> Reported-by: Zhihong Peng
> Signed-off-by: Conor Walsh
> Signed-off-by: Conor Fogarty
> ---
Acked-by: Bruce Richardson
[dpdk-dev] [PATCH v1] net/iavf: fix QFI fields of GTPU UL and DL for FDIR
Fix QFI fields matching of GTPU UL/DL for FDIR.
Fixes: 78e8a87f6324 ("net/iavf: fix GTPU UL and DL support for flow director")
Cc: sta...@dpdk.org
Signed-off-by: Junfeng Guo
Signed-off-by: Lingyu Liu
---
drivers/net/iavf/iavf_fdir.c | 7 ++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/drivers/net/iavf/iavf_fdir.c b/drivers/net/iavf/iavf_fdir.c
index 32b06044f2..134f887965 100644
--- a/drivers/net/iavf/iavf_fdir.c
+++ b/drivers/net/iavf/iavf_fdir.c
@@ -1171,7 +1171,12 @@ iavf_fdir_parse_pattern(__rte_unused struct iavf_adapter
*ad,
if (gtp_psc_spec && gtp_psc_mask) {
if (gtp_psc_mask->qfi == UINT8_MAX) {
input_set |= IAVF_INSET_GTPU_QFI;
- VIRTCHNL_ADD_PROTO_HDR_FIELD_BIT(hdr,
GTPU_EH, QFI);
+ if (gtp_psc_spec->pdu_type ==
IAVF_GTPU_EH_UPLINK)
+
VIRTCHNL_ADD_PROTO_HDR_FIELD_BIT(hdr, GTPU_UP, QFI);
+ else if (gtp_psc_spec->pdu_type ==
IAVF_GTPU_EH_DWLINK)
+
VIRTCHNL_ADD_PROTO_HDR_FIELD_BIT(hdr, GTPU_DWN, QFI);
+ else
+
VIRTCHNL_ADD_PROTO_HDR_FIELD_BIT(hdr, GTPU_EH, QFI);
}
rte_memcpy(hdr->buffer, gtp_psc_spec,
--
2.25.1
Re: [dpdk-dev] [PATCH] doc: announce change in vfio dma mapping
On 01-Sep-21 10:56 AM, Ferruh Yigit wrote: On 9/1/2021 2:41 AM, Ding, Xuan wrote: Hi Ferruh, -Original Message- From: Yigit, Ferruh Sent: Wednesday, September 1, 2021 12:01 AM To: Ding, Xuan ; dev@dpdk.org; Burakov, Anatoly Cc: maxime.coque...@redhat.com; Xia, Chenbo ; Hu, Jiayu ; Richardson, Bruce Subject: Re: [PATCH] doc: announce change in vfio dma mapping On 8/31/2021 2:10 PM, Xuan Ding wrote: Currently, the VFIO subsystem will compact adjacent DMA regions for the purposes of saving space in the internal list of mappings. This has a side effect of compacting two separate mappings that just happen to be adjacent in memory. Since VFIO implementation on IA platforms also does not allow partial unmapping of memory mapped for DMA, the current DPDK VFIO implementation will prevent unmapping of accidentally adjacent maps even though it could have been unmapped [1]. The proper fix for this issue is to change the VFIO DMA mapping API to also include page size, and always map memory page-by-page. [1] https://mails.dpdk.org/archives/dev/2021-July/213493.html Signed-off-by: Xuan Ding --- doc/guides/rel_notes/deprecation.rst | 3 +++ 1 file changed, 3 insertions(+) diff --git a/doc/guides/rel_notes/deprecation.rst b/doc/guides/rel_notes/deprecation.rst index 76a4abfd6b..1234420caf 100644 --- a/doc/guides/rel_notes/deprecation.rst +++ b/doc/guides/rel_notes/deprecation.rst @@ -287,3 +287,6 @@ Deprecation Notices reserved bytes to 2 (from 3), and use 1 byte to indicate warnings and other information from the crypto/security operation. This field will be used to communicate events such as soft expiry with IPsec in lookaside mode. + +* vfio: the functions `rte_vfio_container_dma_map` will be amended to + include page size. This change is targeted for DPDK 22.02. Is this means adding a new parameter to API? If so this is an ABI/API break and we can't do this change in the 22.02. Our original plan is add a new parameter in order not to use a new function name, so you mean, any changes to the API can only be done in the LTS version? If so, we can only add a new API and retire the old one in 22.11. We can add a new API anytime. Adding new parameter to an existing API can be done on the ABI break release. So, 22.11 then? You can add the new API in this release, and start using it. And mark the old API as deprecated in this release. This lets existing binaries to keep using it, but app needs to switch to new API for compilation. Old API can be removed on 22.11, and you will need a deprecation notice before 22.11 for it. Is above plan works for you? We have slightly rethought our approach, and the functionality that Xuan requires does not rely on this API. They can, for all intents and purposes, be considered unrelated issues. I still think it's a good idea to update the API that way, so I would like to do that, and if we have to wait until 22.11 to fix it, I'm OK with that. Since there no longer is any urgency here, it's acceptable to wait for the next LTS to break it. -- Thanks, Anatoly
Re: [dpdk-dev] [PATCH v2 3/3] doc: remove i40evf related documentation
On 01/09/2021 04:30, Robin Zhang wrote: > As announced in the deprecation note, i40evf has been removed. So removing > deprecation notice and other i40evf related documentation. > > Signed-off-by: Robin Zhang > --- > doc/guides/howto/lm_bond_virtio_sriov.rst | 4 ++-- > doc/guides/nics/intel_vf.rst | 8 +--- > doc/guides/rel_notes/deprecation.rst | 8 > doc/guides/rel_notes/release_21_11.rst| 2 ++ > 4 files changed, 5 insertions(+), 17 deletions(-) > Acked-by: Ray Kinsella
Re: [dpdk-dev] [PATCH v1 1/3] net/ixgbe: promote some API to stable
> -Original Message- > From: Yigit, Ferruh > Sent: Wednesday, September 1, 2021 17:02 > To: Wang, Haiyue ; dev@dpdk.org > Cc: m...@ashroe.eu; tho...@monjalon.net > Subject: Re: [dpdk-dev] [PATCH v1 1/3] net/ixgbe: promote some API to stable > > On 9/1/2021 6:07 AM, Haiyue Wang wrote: > > The DPDK Symbol Bot reports: > > Please note the symbols listed below have expired. In line with the > > DPDK ABI policy, they should be scheduled for removal, in the next > > DPDK release. > > > > Symbol > > rte_pmd_ixgbe_mdio_lock > > rte_pmd_ixgbe_mdio_unlock > > rte_pmd_ixgbe_mdio_unlocked_read > > rte_pmd_ixgbe_mdio_unlocked_write > > rte_pmd_ixgbe_upd_fctrl_sbp > > I wonder if we should keep PMD specific APIs as experimental (Not talking > about > mbuf 'dynfield' / 'dynflag' APIs, we can promote them). Yes, makes sense. > > If an application is using PMD specific API, not sure if it will concern about > PMD specific APIs. > And keeping PMD specific APIs lets us remove them as soon as we can, also adds > additional discourage for users to use them. Can update this to DPDK ABI Policy, section 3.5.3. https://doc.dpdk.org/guides/contributing/abi_policy.html > > > > > Signed-off-by: Haiyue Wang > > > <...> >
Re: [dpdk-dev] [PATCH v2 2/3] net/iavf: remove i40evf devargs option
On 01/09/2021 04:30, Robin Zhang wrote:
> Due to i40evf will be removed, so there's no need to keep the devargs
> option "driver=i40evf" in iavf.
>
> Signed-off-by: Robin Zhang
> ---
[SNIP]
> @@ -2556,7 +2508,7 @@ static struct rte_pci_driver rte_iavf_pmd = {
> RTE_PMD_REGISTER_PCI(net_iavf, rte_iavf_pmd);
> RTE_PMD_REGISTER_PCI_TABLE(net_iavf, pci_id_iavf_map);
> RTE_PMD_REGISTER_KMOD_DEP(net_iavf, "* igb_uio | vfio-pci");
> -RTE_PMD_REGISTER_PARAM_STRING(net_iavf, "cap=dcf driver=i40evf");
> +RTE_PMD_REGISTER_PARAM_STRING(net_iavf, "cap=dcf");
Is "cap=dcf driver=iavf" required?
> RTE_LOG_REGISTER_SUFFIX(iavf_logtype_init, init, NOTICE);
> RTE_LOG_REGISTER_SUFFIX(iavf_logtype_driver, driver, NOTICE);
> #ifdef RTE_ETHDEV_DEBUG_RX
>
Re: [dpdk-dev] [PATCH v2 1/3] net/i40e: remove i40evf
On 01/09/2021 04:30, Robin Zhang wrote: > The default VF driver for Intel 700 Series Ethernet Controller already > switch to iavf in DPDK 21.05. And i40evf is no need to maintain now, > so remove i40evf related code. > > Signed-off-by: Robin Zhang > --- > drivers/net/i40e/i40e_ethdev.h|1 - > drivers/net/i40e/i40e_ethdev_vf.c | 3015 - > drivers/net/i40e/meson.build |1 - > drivers/net/i40e/rte_pmd_i40e.c |9 +- > 4 files changed, 3 insertions(+), 3023 deletions(-) > delete mode 100644 drivers/net/i40e/i40e_ethdev_vf.c > Acked-by: Ray Kinsella
Re: [dpdk-dev] [PATCH v1] bbdev: remove experimental tag from API
On 01/09/2021 04:52, Hemant Agrawal wrote: >> >> On Tue, Aug 31, 2021 at 6:27 PM Nicolas Chautru >> wrote: >>> >>> This was previously suggested last year >>> >> https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpatc >>> hes.dpdk.org%2Fproject%2Fdpdk%2Fpatch%2F1593213242-157394-2-git- >> send-e >>> mail- >> nicolas.chautru%40intel.com%2F&data=04%7C01%7Chemant.agrawal% >>> >> 40nxp.com%7Cfcdfb339ec284057db1508d96c9dd048%7C686ea1d3bc2b4c6f >> a92cd99 >>> >> c5c301635%7C0%7C0%7C637660247324212288%7CUnknown%7CTWFpbGZ >> sb3d8eyJWIjo >>> >> iMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1 >> 000& >>> ;sdata=%2FYwaT29I4XlUxedcdCeyvM5z1R5QeNGTLLPY3XQBKjU%3D& >> ;reserved=0 >>> but there was request from community to wait another year to confirm >> formally this api is mature. >> >> The request was to wait for a new vendor (i.e. non Intel) to implement a >> bbdev driver for their hw. >> >> NXP proposed a driver for this class: >> https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpatch >> work.dpdk.org%2Fproject%2Fdpdk%2Flist%2F%3Fseries%3D16642%26state >> %3D%252A%26archive%3Dboth&data=04%7C01%7Chemant.agrawal% >> 40nxp.com%7Cfcdfb339ec284057db1508d96c9dd048%7C686ea1d3bc2b4c6f >> a92cd99c5c301635%7C0%7C0%7C637660247324212288%7CUnknown%7CT >> WFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLC >> JXVCI6Mn0%3D%7C1000&sdata=iorsQ7wAXyU2%2FaqeV5CrDp9asl8ztZ >> nSrbDuJpCv9Gk%3D&reserved=0 >> >> What is the status of this driver? > > [Hemant] We are in process of resending the patches in next few days. Ok - just so we are clear - the net-net of that, is that the bbdev APIs will be 'experimental' for another year, right? > >> I see no update on the bbdev API, which seems good, but a confirmation is >> needed. >> >> >> -- >> David Marchand >
Re: [dpdk-dev] [PATCH] common/cnxk: add ROC API to merge base steering rule
On 31/08/2021 05:16, psathe...@marvell.com wrote: > From: Satheesh Paul > > This patch adds an ROC API to merge base steering rule with rules > added by VF. > > Signed-off-by: Satheesh Paul > Reviewed-by: Kiran Kumar Kokkilagadda > --- > drivers/common/cnxk/roc_npc.c | 27 +++ > drivers/common/cnxk/roc_npc.h | 5 ++--- > drivers/common/cnxk/version.map | 1 + > 3 files changed, 30 insertions(+), 3 deletions(-) > Acked-by: Ray Kinsella
Re: [dpdk-dev] [PATCH] eal: promote random function with upper bound to stable
On 01/09/2021 08:29, Mattias Rönnblom wrote: > Remove experimental tag from rte_rand_max(). > > Signed-off-by: Mattias Rönnblom Acked-by: Ray Kinsella
Re: [dpdk-dev] [PATCH] ethdev: remove experimental flag from getting intr fd API
On 01/09/2021 09:53, Ferruh Yigit wrote: > On 9/1/2021 8:08 AM, Andrew Rybchenko wrote: >> On 9/1/21 4:50 AM, Xiaoyun Li wrote: >>> Remove the experimental tag for rte_eth_dev_rx_intr_ctl_q_get_fd API >>> that was introduced in 18.11 and have been around for 11 releases. >>> >>> Signed-off-by: Xiaoyun Li >> >> Acked-by: Andrew Rybchenko >> > > Acked-by: Ferruh Yigit > Acked-by: Ray Kinsella
[dpdk-dev] [PATCH] eal: promote random function with upper bound to stable
Remove experimental tag from rte_rand_max().
Signed-off-by: Mattias Rönnblom
---
lib/eal/include/rte_random.h | 1 -
lib/eal/version.map | 2 +-
2 files changed, 1 insertion(+), 2 deletions(-)
diff --git a/lib/eal/include/rte_random.h b/lib/eal/include/rte_random.h
index 2b30ec85c1..29f5f1325a 100644
--- a/lib/eal/include/rte_random.h
+++ b/lib/eal/include/rte_random.h
@@ -62,7 +62,6 @@ rte_rand(void);
* @return
* A pseudo-random value between 0 and (upper_bound-1).
*/
-__rte_experimental
uint64_t
rte_rand_max(uint64_t upper_bound);
diff --git a/lib/eal/version.map b/lib/eal/version.map
index beeb986adc..92caaa2bd5 100644
--- a/lib/eal/version.map
+++ b/lib/eal/version.map
@@ -148,6 +148,7 @@ DPDK_22 {
rte_memzone_walk;
rte_openlog_stream;
rte_rand;
+ rte_rand_max;
rte_realloc;
rte_reciprocal_value;
rte_reciprocal_value_u64;
@@ -326,7 +327,6 @@ EXPERIMENTAL {
rte_lcore_to_cpu_id;
rte_mcfg_timer_lock;
rte_mcfg_timer_unlock;
- rte_rand_max; # WINDOWS_NO_EXPORT
# added in 19.11
rte_mcfg_get_single_file_segments;
--
2.25.1
Re: [dpdk-dev] [PATCH] eal: promote random function with upper bound to stable
On 2021-09-01 09:29, Mattias Rönnblom wrote:
> Remove experimental tag from rte_rand_max().
>
> Signed-off-by: Mattias Rönnblom
> ---
> lib/eal/include/rte_random.h | 1 -
> lib/eal/version.map | 2 +-
> 2 files changed, 1 insertion(+), 2 deletions(-)
>
> diff --git a/lib/eal/include/rte_random.h b/lib/eal/include/rte_random.h
> index 2b30ec85c1..29f5f1325a 100644
> --- a/lib/eal/include/rte_random.h
> +++ b/lib/eal/include/rte_random.h
> @@ -62,7 +62,6 @@ rte_rand(void);
>* @return
>* A pseudo-random value between 0 and (upper_bound-1).
>*/
> -__rte_experimental
> uint64_t
> rte_rand_max(uint64_t upper_bound);
>
> diff --git a/lib/eal/version.map b/lib/eal/version.map
> index beeb986adc..92caaa2bd5 100644
> --- a/lib/eal/version.map
> +++ b/lib/eal/version.map
> @@ -148,6 +148,7 @@ DPDK_22 {
> rte_memzone_walk;
> rte_openlog_stream;
> rte_rand;
> + rte_rand_max;
> rte_realloc;
> rte_reciprocal_value;
> rte_reciprocal_value_u64;
> @@ -326,7 +327,6 @@ EXPERIMENTAL {
> rte_lcore_to_cpu_id;
> rte_mcfg_timer_lock;
> rte_mcfg_timer_unlock;
> - rte_rand_max; # WINDOWS_NO_EXPORT
Should I have kept the WINDOWS_NO_EXPORT tag?
>
> # added in 19.11
> rte_mcfg_get_single_file_segments;
Re: [dpdk-dev] [PATCH] doc: announce change in vfio dma mapping
On 9/1/2021 12:01 PM, Burakov, Anatoly wrote: > On 01-Sep-21 10:56 AM, Ferruh Yigit wrote: >> On 9/1/2021 2:41 AM, Ding, Xuan wrote: >>> Hi Ferruh, >>> -Original Message- From: Yigit, Ferruh Sent: Wednesday, September 1, 2021 12:01 AM To: Ding, Xuan ; dev@dpdk.org; Burakov, Anatoly Cc: maxime.coque...@redhat.com; Xia, Chenbo ; Hu, Jiayu ; Richardson, Bruce Subject: Re: [PATCH] doc: announce change in vfio dma mapping On 8/31/2021 2:10 PM, Xuan Ding wrote: > Currently, the VFIO subsystem will compact adjacent DMA regions for the > purposes of saving space in the internal list of mappings. This has a > side effect of compacting two separate mappings that just happen to be > adjacent in memory. Since VFIO implementation on IA platforms also does > not allow partial unmapping of memory mapped for DMA, the current DPDK > VFIO implementation will prevent unmapping of accidentally adjacent > maps even though it could have been unmapped [1]. > > The proper fix for this issue is to change the VFIO DMA mapping API to > also include page size, and always map memory page-by-page. > > [1] https://mails.dpdk.org/archives/dev/2021-July/213493.html > > Signed-off-by: Xuan Ding > --- > doc/guides/rel_notes/deprecation.rst | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/doc/guides/rel_notes/deprecation.rst b/doc/guides/rel_notes/deprecation.rst > index 76a4abfd6b..1234420caf 100644 > --- a/doc/guides/rel_notes/deprecation.rst > +++ b/doc/guides/rel_notes/deprecation.rst > @@ -287,3 +287,6 @@ Deprecation Notices > reserved bytes to 2 (from 3), and use 1 byte to indicate warnings and other > information from the crypto/security operation. This field will be > used to > communicate events such as soft expiry with IPsec in lookaside mode. > + > +* vfio: the functions `rte_vfio_container_dma_map` will be amended to > + include page size. This change is targeted for DPDK 22.02. > Is this means adding a new parameter to API? If so this is an ABI/API break and we can't do this change in the 22.02. >>> >>> Our original plan is add a new parameter in order not to use a new function >>> name, so you mean, any changes to the API can only be done in the LTS >>> version? >>> If so, we can only add a new API and retire the old one in 22.11. >>> >> >> We can add a new API anytime. Adding new parameter to an existing API can be >> done on the ABI break release. > > So, 22.11 then? > Yes. >> >> You can add the new API in this release, and start using it. >> And mark the old API as deprecated in this release. This lets existing >> binaries >> to keep using it, but app needs to switch to new API for compilation. >> Old API can be removed on 22.11, and you will need a deprecation notice >> before >> 22.11 for it. >> >> Is above plan works for you? >> > > We have slightly rethought our approach, and the functionality that Xuan > requires does not rely on this API. They can, for all intents and purposes, be > considered unrelated issues. > > I still think it's a good idea to update the API that way, so I would like to > do > that, and if we have to wait until 22.11 to fix it, I'm OK with that. Since > there no longer is any urgency here, it's acceptable to wait for the next LTS > to > break it. > Got it. As far as I understand, main motivation in techboard decision was to prevent the ABI break as much as possible (main reason of decision wasn't deprecation notice being late). But if the correct thing to do is to rename the API (and break the ABI), I don't see the benefit to wait one more year, it is just delaying the impact and adding overhead to us. I am for being pragmatic and doing the change in this release if API rename is better option, perhaps we can visit the issue again in techboard. Can you please describe why renaming API is better option, comparing to adding new API with new parameter? Thanks, ferruh
Re: [dpdk-dev] [PATCH 0/6] mlx5: some independent fixes
From: Michael Baum > Some independent fixes in mlx5 net and common driver. > > Michael Baum (6): > net/mlx5: fix shared device context creation error flow > net/mlx5: fix PCI probing error flow > net/mlx5: fix allow duplicate pattern devarg default > common/mlx5: fix class combination validation > common/mlx5: fix device list operation concurrency > common/mlx5: fix resource cleanliness in a device remove > > drivers/common/mlx5/mlx5_common.c | 26 +- > drivers/net/mlx5/linux/mlx5_os.c | 15 +-- > drivers/net/mlx5/mlx5.c | 4 +++- > drivers/net/mlx5/mlx5.h | 1 + > 4 files changed, 34 insertions(+), 12 deletions(-) Series-acked-by: Matan Azrad > -- > 2.25.1
Re: [dpdk-dev] [PATCH] regex/mlx5: fix RegEx probing error flow
From: Michael Baum
> In RegEx device probing, there is register read trying after context device
> creation.
>
> When the reading fails, the context device was not freed what caused a
> memory leak.
>
> Free it.
>
> Fixes: f324162e8e77 ("regex/mlx5: support combined rule file")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Michael Baum
Acked-by: Matan Azrad
Re: [dpdk-dev] [PATCH] crypto/mlx5: support timestamp format
From: Michael Baum
> This patch adds support for the timestamp format settings for the receive
> and send queues. If the firmware version x.30.1000 or above is installed and
> the NIC timestamps are configured with the real-time format, the default
> zero values for newly added fields cause the queue creation to fail.
>
> The patch queries the timestamp formats supported by the hardware and
> sets the configuration values in queue context accordingly.
>
> Fixes: 6152534e211e ("crypto/mlx5: support queue pairs operations")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Michael Baum
Acked-by: Matan Azrad
Re: [dpdk-dev] [PATCH] compress/mlx5: fix QP setup error flow
From: Michael Baum
> Sent: Tuesday, August 31, 2021 11:40 PM
> To: dev@dpdk.org
> Cc: Matan Azrad ; sta...@dpdk.org
> Subject: [PATCH] compress/mlx5: fix QP setup error flow
>
> The QP setup function allocates buffer for its opaque MR and register it into
> MR structure.
>
> After buffer alloction and before MR registration, it tries allocate MR Btree.
> When the MR Btree allocation fails, the buffer was not freed what caused a
> memory leak.
>
> Allocate the MR Btree before buffer alloction.
>
> Fixes: 0165bccdb45f ("compress/mlx5: add memory region management")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Michael Baum
Acked-by: Matan Azrad
[dpdk-dev] [PATCH] net/mlx5: fix shared RSS destruction
Shared RSS resources were released before checking that the shared RSS
has no more references. If it had, the destruction was aborted, leaving
the shared RSS in an invalid state where it could no longer be used.
Move reference counter check before resource release.
Fixes: d2046c09aa64 ("net/mlx5: support shared action for RSS")
Cc: andr...@nvidia.com
Cc: sta...@dpdk.org
Signed-off-by: Dmitry Kozlyuk
Acked-by: Viacheslav Ovsiienko
---
drivers/net/mlx5/mlx5_flow_dv.c | 12 ++--
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_flow_dv.c b/drivers/net/mlx5/mlx5_flow_dv.c
index 3f6f5dcfba..d59fa1c5f8 100644
--- a/drivers/net/mlx5/mlx5_flow_dv.c
+++ b/drivers/net/mlx5/mlx5_flow_dv.c
@@ -14742,12 +14742,6 @@ __flow_dv_action_rss_release(struct rte_eth_dev *dev,
uint32_t idx,
return rte_flow_error_set(error, EINVAL,
RTE_FLOW_ERROR_TYPE_ACTION, NULL,
"invalid shared action");
- remaining = __flow_dv_action_rss_hrxqs_release(dev, shared_rss);
- if (remaining)
- return rte_flow_error_set(error, EBUSY,
- RTE_FLOW_ERROR_TYPE_ACTION,
- NULL,
- "shared rss hrxq has references");
if (!__atomic_compare_exchange_n(&shared_rss->refcnt, &old_refcnt,
0, 0, __ATOMIC_ACQUIRE,
__ATOMIC_RELAXED))
@@ -14755,6 +14749,12 @@ __flow_dv_action_rss_release(struct rte_eth_dev *dev,
uint32_t idx,
RTE_FLOW_ERROR_TYPE_ACTION,
NULL,
"shared rss has references");
+ remaining = __flow_dv_action_rss_hrxqs_release(dev, shared_rss);
+ if (remaining)
+ return rte_flow_error_set(error, EBUSY,
+ RTE_FLOW_ERROR_TYPE_ACTION,
+ NULL,
+ "shared rss hrxq has references");
queue = shared_rss->ind_tbl->queues;
remaining = mlx5_ind_table_obj_release(dev, shared_rss->ind_tbl, true);
if (remaining)
--
2.25.1
Re: [dpdk-dev] [PATCH] regex/mlx5: fix RegEx probing error flow
> -Original Message-
> From: Michael Baum
>
> In RegEx device probing, there is register read trying after context device
> creation.
>
> When the reading fails, the context device was not freed what caused a
> memory leak.
>
> Free it.
>
> Fixes: f324162e8e77 ("regex/mlx5: support combined rule file")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Michael Baum
> ---
> drivers/regex/mlx5/mlx5_regex.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/regex/mlx5/mlx5_regex.c
> b/drivers/regex/mlx5/mlx5_regex.c index f17b6df47f..8866a4d0c6 100644
> --- a/drivers/regex/mlx5/mlx5_regex.c
> +++ b/drivers/regex/mlx5/mlx5_regex.c
> @@ -171,7 +171,7 @@ mlx5_regex_dev_probe(struct rte_device *rte_dev)
> MLX5_RXP_CSR_IDENTIFIER, &val);
> if (ret) {
> DRV_LOG(ERR, "CSR read failed!");
> - return -1;
> + goto dev_error;
> }
> if (val == MLX5_RXP_BF2_IDENTIFIER)
> priv->is_bf2 = 1;
> --
> 2.25.1
Thanks,
Acked-by: Ori Kam
[dpdk-dev] [PATCH] sched: remove experimental tag from the API
This API was introduced in 18.05, therefore removing
experimental tag to promote it to stable state
Signed-off-by: Jasvinder Singh
---
lib/sched/rte_sched.h | 4
lib/sched/version.map | 2 +-
2 files changed, 1 insertion(+), 5 deletions(-)
diff --git a/lib/sched/rte_sched.h b/lib/sched/rte_sched.h
index c1a772b70c..cb851301e9 100644
--- a/lib/sched/rte_sched.h
+++ b/lib/sched/rte_sched.h
@@ -301,9 +301,6 @@ void
rte_sched_port_free(struct rte_sched_port *port);
/**
- * @warning
- * @b EXPERIMENTAL: this API may change without prior notice.
- *
* Hierarchical scheduler pipe profile add
*
* @param port
@@ -317,7 +314,6 @@ rte_sched_port_free(struct rte_sched_port *port);
* @return
* 0 upon success, error code otherwise
*/
-__rte_experimental
int
rte_sched_subport_pipe_profile_add(struct rte_sched_port *port,
uint32_t subport_id,
diff --git a/lib/sched/version.map b/lib/sched/version.map
index 53c337b143..a6e505c8ac 100644
--- a/lib/sched/version.map
+++ b/lib/sched/version.map
@@ -19,6 +19,7 @@ DPDK_22 {
rte_sched_port_pkt_write;
rte_sched_queue_read_stats;
rte_sched_subport_config;
+ rte_sched_subport_pipe_profile_add;
rte_sched_subport_read_stats;
local: *;
@@ -27,7 +28,6 @@ DPDK_22 {
EXPERIMENTAL {
global:
- rte_sched_subport_pipe_profile_add;
# added in 20.11
rte_sched_port_subport_profile_add;
};
--
2.26.3
[dpdk-dev] [PATCH] pipeline: remove experimental tag from API
These APIs were introduced in 18.05, therefore removing experimental tag to promote them to stable state. Signed-off-by: Jasvinder Singh --- lib/pipeline/rte_port_in_action.h | 10 -- lib/pipeline/rte_table_action.h | 18 -- lib/pipeline/version.map | 16 ++-- 3 files changed, 6 insertions(+), 38 deletions(-) diff --git a/lib/pipeline/rte_port_in_action.h b/lib/pipeline/rte_port_in_action.h index d6b063cf58..de611cbfab 100644 --- a/lib/pipeline/rte_port_in_action.h +++ b/lib/pipeline/rte_port_in_action.h @@ -42,8 +42,6 @@ *used by the input port action handler based on given action parameters. *API functions: rte_port_in_action_apply(). * - * @warning - * @b EXPERIMENTAL: this API may change without prior notice */ #ifdef __cplusplus @@ -172,7 +170,6 @@ struct rte_port_in_action_profile; * @return * Input port action profile handle on success, NULL otherwise. */ -__rte_experimental struct rte_port_in_action_profile * rte_port_in_action_profile_create(uint32_t socket_id); @@ -184,7 +181,6 @@ rte_port_in_action_profile_create(uint32_t socket_id); * @return * Zero on success, non-zero error code otherwise. */ -__rte_experimental int rte_port_in_action_profile_free(struct rte_port_in_action_profile *profile); @@ -203,7 +199,6 @@ rte_port_in_action_profile_free(struct rte_port_in_action_profile *profile); * @return * Zero on success, non-zero error code otherwise. */ -__rte_experimental int rte_port_in_action_profile_action_register( struct rte_port_in_action_profile *profile, @@ -226,7 +221,6 @@ rte_port_in_action_profile_action_register( * * @see rte_port_in_action_create() */ -__rte_experimental int rte_port_in_action_profile_freeze(struct rte_port_in_action_profile *profile); @@ -249,7 +243,6 @@ struct rte_port_in_action; * @return * Handle to input port action object on success, NULL on error. */ -__rte_experimental struct rte_port_in_action * rte_port_in_action_create(struct rte_port_in_action_profile *profile, uint32_t socket_id); @@ -262,7 +255,6 @@ rte_port_in_action_create(struct rte_port_in_action_profile *profile, * @return * Zero on success, non-zero error code otherwise. */ -__rte_experimental int rte_port_in_action_free(struct rte_port_in_action *action); @@ -276,7 +268,6 @@ rte_port_in_action_free(struct rte_port_in_action *action); * @return * Zero on success, non-zero error code otherwise. */ -__rte_experimental int rte_port_in_action_params_get(struct rte_port_in_action *action, struct rte_pipeline_port_in_params *params); @@ -296,7 +287,6 @@ rte_port_in_action_params_get(struct rte_port_in_action *action, * @return * Zero on success, non-zero error code otherwise. */ -__rte_experimental int rte_port_in_action_apply(struct rte_port_in_action *action, enum rte_port_in_action_type type, diff --git a/lib/pipeline/rte_table_action.h b/lib/pipeline/rte_table_action.h index 82bc9d9ac9..aac6b8ce8d 100644 --- a/lib/pipeline/rte_table_action.h +++ b/lib/pipeline/rte_table_action.h @@ -48,8 +48,6 @@ * - Use the table action object to read action data (e.g. stats counters) for *any given rule. API functions: rte_table_action_XYZ_read(). * - * @warning - * @b EXPERIMENTAL: this API may change without prior notice */ #ifdef __cplusplus @@ -823,7 +821,6 @@ struct rte_table_action_profile; * @return * Table action profile handle on success, NULL otherwise. */ -__rte_experimental struct rte_table_action_profile * rte_table_action_profile_create(struct rte_table_action_common_config *common); @@ -835,7 +832,6 @@ rte_table_action_profile_create(struct rte_table_action_common_config *common); * @return * Zero on success, non-zero error code otherwise. */ -__rte_experimental int rte_table_action_profile_free(struct rte_table_action_profile *profile); @@ -854,7 +850,6 @@ rte_table_action_profile_free(struct rte_table_action_profile *profile); * @return * Zero on success, non-zero error code otherwise. */ -__rte_experimental int rte_table_action_profile_action_register(struct rte_table_action_profile *profile, enum rte_table_action_type type, @@ -875,7 +870,6 @@ rte_table_action_profile_action_register(struct rte_table_action_profile *profil * * @see rte_table_action_create() */ -__rte_experimental int rte_table_action_profile_freeze(struct rte_table_action_profile *profile); @@ -899,7 +893,6 @@ struct rte_table_action; * * @see rte_table_action_create() */ -__rte_experimental struct rte_table_action * rte_table_action_create(struct rte_table_action_profile *profile, uint32_t socket_id); @@ -912,7 +905,6 @@ rte_table_action_create(struct rte_table_action_profile *profile, * @return * Zero on success, non-zero error code otherwise. */ -__rte_experimental int rte_table_action_free(struct rte_table_action *action); @@ -926,7 +918
[dpdk-dev] [PATCH 0/2] net/mlx5: fix flow indirect action reference counting
Dmitry Kozlyuk (2): net/mlx5: report error on indirect CT action destroy net/mlx5: fix flow indirect action reference counting drivers/net/mlx5/mlx5_flow_dv.c | 59 ++--- 1 file changed, 39 insertions(+), 20 deletions(-) -- 2.25.1
[dpdk-dev] [PATCH 1/2] net/mlx5: report error on indirect CT action destroy
When an indirect CT action of mlx5 PMD could not be destroyed,
rte_action_handle_destroy() was returning (-1), but the error
structure was not filled. This lead to a segfault in testpmd
on an attempt to print it. Fill the details for each possible
cause of this error.
Fixes: c5a49265fc23 ("net/mlx5: add ASO connection tracking destroy")
Cc: bi...@nvidia.com
Signed-off-by: Dmitry Kozlyuk
Acked-by: Viacheslav Ovsiienko
---
drivers/net/mlx5/mlx5_flow_dv.c | 22 --
1 file changed, 16 insertions(+), 6 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_flow_dv.c b/drivers/net/mlx5/mlx5_flow_dv.c
index 3f6f5dcfba..b769ae4007 100644
--- a/drivers/net/mlx5/mlx5_flow_dv.c
+++ b/drivers/net/mlx5/mlx5_flow_dv.c
@@ -12304,17 +12304,27 @@ flow_dv_aso_ct_dev_release(struct rte_eth_dev *dev,
uint32_t idx)
}
static inline int
-flow_dv_aso_ct_release(struct rte_eth_dev *dev, uint32_t own_idx)
+flow_dv_aso_ct_release(struct rte_eth_dev *dev, uint32_t own_idx,
+ struct rte_flow_error *error)
{
uint16_t owner = (uint16_t)MLX5_INDIRECT_ACT_CT_GET_OWNER(own_idx);
uint32_t idx = MLX5_INDIRECT_ACT_CT_GET_IDX(own_idx);
struct rte_eth_dev *owndev = &rte_eth_devices[owner];
- RTE_SET_USED(dev);
+ int ret;
MLX5_ASSERT(owner < RTE_MAX_ETHPORTS);
if (dev->data->dev_started != 1)
- return -1;
- return flow_dv_aso_ct_dev_release(owndev, idx);
+ return rte_flow_error_set(error, EAGAIN,
+ RTE_FLOW_ERROR_TYPE_ACTION,
+ NULL,
+ "Indirect CT action cannot be
destroyed when the port is stopped");
+ ret = flow_dv_aso_ct_dev_release(owndev, idx);
+ if (ret < 0)
+ return rte_flow_error_set(error, EAGAIN,
+ RTE_FLOW_ERROR_TYPE_ACTION,
+ NULL,
+ "Current state prevents indirect CT
action from being destroyed");
+ return ret;
}
/*
@@ -14365,7 +14375,7 @@ flow_dv_destroy(struct rte_eth_dev *dev, struct
rte_flow *flow)
}
/* Keep the current age handling by default. */
if (flow->indirect_type == MLX5_INDIRECT_ACTION_TYPE_CT && flow->ct)
- flow_dv_aso_ct_release(dev, flow->ct);
+ flow_dv_aso_ct_release(dev, flow->ct, NULL);
else if (flow->age)
flow_dv_aso_age_release(dev, flow->age);
if (flow->geneve_tlv_option) {
@@ -14900,7 +14910,7 @@ flow_dv_action_destroy(struct rte_eth_dev *dev,
" released with references %d.", idx, ret);
return 0;
case MLX5_INDIRECT_ACTION_TYPE_CT:
- ret = flow_dv_aso_ct_release(dev, idx);
+ ret = flow_dv_aso_ct_release(dev, idx, error);
if (ret < 0)
return ret;
if (ret > 0)
--
2.25.1
[dpdk-dev] [PATCH] net/softnic: remove experimental table from API
This API was introduced in 18.08, therefore removing
experimental tag to promote it to stable state.
Signed-off-by: Jasvinder Singh
---
drivers/net/softnic/rte_eth_softnic.h | 1 -
drivers/net/softnic/version.map | 7 +--
2 files changed, 1 insertion(+), 7 deletions(-)
diff --git a/drivers/net/softnic/rte_eth_softnic.h
b/drivers/net/softnic/rte_eth_softnic.h
index 3f01161777..6c11d89ab6 100644
--- a/drivers/net/softnic/rte_eth_softnic.h
+++ b/drivers/net/softnic/rte_eth_softnic.h
@@ -75,7 +75,6 @@ rte_pmd_softnic_run(uint16_t port_id);
* @return
*Zero on success, error code otherwise.
*/
-__rte_experimental
int
rte_pmd_softnic_manage(uint16_t port_id);
diff --git a/drivers/net/softnic/version.map b/drivers/net/softnic/version.map
index 6784318f77..cd5afcf155 100644
--- a/drivers/net/softnic/version.map
+++ b/drivers/net/softnic/version.map
@@ -2,12 +2,7 @@ DPDK_22 {
global:
rte_pmd_softnic_run;
+ rte_pmd_softnic_manage;
local: *;
};
-
-EXPERIMENTAL {
- global:
-
- rte_pmd_softnic_manage;
-};
--
2.26.3
[dpdk-dev] [PATCH 2/2] net/mlx5: fix flow indirect action reference counting
When an indirect action is used in a flow rule with a pattern that
causes RSS expansion, each device flow generated by the expansion
incremented the reference counter of the action. When such a flow was
destroyed, its action reference counter had been decremented only once.
The action remained marked as being used and could not be destroyed.
COUNT, AGE, and CONNTRACK indirect actions have been affected
(for AGE the error was not immediately observable).
Increment action counter only once for the original flow rule.
Fixes: 81073e1f8ce1 ("net/mlx5: support shared age action")
Fixes: 2d084f69aa26 ("net/mlx5: add translation of connection tracking action")
Fixes: f3191849f2c2 ("net/mlx5: support flow count action handle")
Cc: ma...@nvidia.com
Cc: bi...@nvidia.com
Cc: michae...@nvidia.com
Cc: sta...@dpdk.org
Signed-off-by: Dmitry Kozlyuk
Acked-by: Viacheslav Ovsiienko
---
drivers/net/mlx5/mlx5_flow_dv.c | 37 -
1 file changed, 23 insertions(+), 14 deletions(-)
diff --git a/drivers/net/mlx5/mlx5_flow_dv.c b/drivers/net/mlx5/mlx5_flow_dv.c
index b769ae4007..626513ce56 100644
--- a/drivers/net/mlx5/mlx5_flow_dv.c
+++ b/drivers/net/mlx5/mlx5_flow_dv.c
@@ -12840,10 +12840,13 @@ flow_dv_translate(struct rte_eth_dev *dev,
MLX5_FLOW_FATE_QUEUE;
break;
case MLX5_RTE_FLOW_ACTION_TYPE_AGE:
- flow->age = (uint32_t)(uintptr_t)(action->conf);
- age_act = flow_aso_age_get_by_idx(dev, flow->age);
- __atomic_fetch_add(&age_act->refcnt, 1,
- __ATOMIC_RELAXED);
+ owner_idx = (uint32_t)(uintptr_t)action->conf;
+ age_act = flow_aso_age_get_by_idx(dev, owner_idx);
+ if (flow->age == 0) {
+ flow->age = owner_idx;
+ __atomic_fetch_add(&age_act->refcnt, 1,
+ __ATOMIC_RELAXED);
+ }
age_act_pos = actions_n++;
action_flags |= MLX5_FLOW_ACTION_AGE;
break;
@@ -12853,9 +12856,9 @@ flow_dv_translate(struct rte_eth_dev *dev,
action_flags |= MLX5_FLOW_ACTION_AGE;
break;
case MLX5_RTE_FLOW_ACTION_TYPE_COUNT:
- cnt_act = flow_dv_counter_get_by_idx(dev,
- (uint32_t)(uintptr_t)action->conf,
- NULL);
+ owner_idx = (uint32_t)(uintptr_t)action->conf;
+ cnt_act = flow_dv_counter_get_by_idx(dev, owner_idx,
+NULL);
MLX5_ASSERT(cnt_act != NULL);
/**
* When creating meter drop flow in drop table, the
@@ -12866,10 +12869,12 @@ flow_dv_translate(struct rte_eth_dev *dev,
dev_flow->dv.actions[actions_n++] =
cnt_act->action;
} else {
- flow->counter =
- (uint32_t)(uintptr_t)(action->conf);
- __atomic_fetch_add(&cnt_act->shared_info.refcnt,
- 1, __ATOMIC_RELAXED);
+ if (flow->counter == 0) {
+ flow->counter = owner_idx;
+ __atomic_fetch_add
+ (&cnt_act->shared_info.refcnt,
+1, __ATOMIC_RELAXED);
+ }
/* Save information first, will apply later. */
action_flags |= MLX5_FLOW_ACTION_COUNT;
}
@@ -13192,9 +13197,13 @@ flow_dv_translate(struct rte_eth_dev *dev,
else
dev_flow->dv.actions[actions_n] =
ct->dr_action_rply;
- flow->indirect_type = MLX5_INDIRECT_ACTION_TYPE_CT;
- flow->ct = owner_idx;
- __atomic_fetch_add(&ct->refcnt, 1, __ATOMIC_RELAXED);
+ if (flow->ct == 0) {
+ flow->indirect_type =
+ MLX5_INDIRECT_ACTION_TYPE_CT;
+ flow->ct = owner_idx;
+ __atomic_fetch_add(&ct->refcnt, 1,
+ __ATOMIC_RELAXED);
+ }
actions_n++;
action_flags |= MLX5_FLOW_
Re: [dpdk-dev] [PATCH v10 0/3] devtools: scripts to count and track symbols
Ray Kinsella writes: > Scripts to count and track the lifecycle of DPDK symbols. > > The symbol-tool script reports on the growth of symbols over releases > and list expired symbols. The notify-symbol-maintainers script > consumes the input from symbol-tool and generates email notifications > of expired symbols. > > v2: reworked to fix pylint errors > v3: sent with the correct in-reply-to > v4: fix typos picked up by the CI > v5: fix terminal_size & directory args > v6: added list-expired, to list expired experimental symbols > v7: fix typo in comments > v8: added tool to notify maintainers of expired symbols > v9: removed hardcoded emails addressed and script names > v10: added ability to identify and notify the original contributors > > Ray Kinsella (3): > devtools: script to track symbols over releases > devtools: script to send notifications of expired symbols > maintainers: add new abi scripts > > MAINTAINERS | 2 + > devtools/notify-symbol-maintainers.py | 256 ++ > devtools/symbol-tool.py | 482 ++ > 3 files changed, 740 insertions(+) > create mode 100755 devtools/notify-symbol-maintainers.py > create mode 100755 devtools/symbol-tool.py I get a whole mess of flake8 issues from this series (mostly 'backslash is redundant' and whitespace issues). I'm using flake8 because it pretty well enforces PEP8 style guide. I would like to see it addressed, but also I see that many of the python files in the DPDK tree don't actually pass. Example:: $ flake8 ./usertools/dpdk-devbind.py | wc -l 34 $ flake8 ./usertools/dpdk-devbind.py | sed 's@./usertools/dpdk-devbind.py[:0-9]* @@' | sort -u E128 continuation line under-indented for visual indent E302 expected 2 blank lines, found 1 E305 expected 2 blank lines after class or function definition, found 1 E501 line too long (105 > 79 characters) E501 line too long (80 > 79 characters) E501 line too long (82 > 79 characters) E501 line too long (83 > 79 characters) E501 line too long (84 > 79 characters) E501 line too long (85 > 79 characters) E501 line too long (86 > 79 characters) E501 line too long (91 > 79 characters) E502 the backslash is redundant between brackets E722 do not use bare 'except' Looks like we repeat the same kinds of errors everywhere (this is on multiple tools). Some of our in-tree python is better than others (like app/test/autotest.py which only has 1 flake). Maybe we can address this. Other comments inline on the patches.
Re: [dpdk-dev] [PATCH v10 2/3] devtools: script to send notifications of expired symbols
Ray Kinsella writes:
> Use this script with the output of the DPDK symbol tool, to notify
> maintainers of expired symbols by email. You need to define the environment
> variable DPDK_GETMAINTAINER_PATH for this tool to work.
>
> Use terminal output to review the emails before sending.
> e.g.
> $ devtools/symbol-tool.py list-expired --format-output csv \
> | DPDK_GETMAINTAINER_PATH=/get_maintainer.pl \
> devtools/notify_expired_symbols.py --format-output terminal
>
> Then use email output to send the emails to the maintainers.
> e.g.
> $ devtools/symbol-tool.py list-expired --format-output csv \
> | DPDK_GETMAINTAINER_PATH=/get_maintainer.pl \
> devtools/notify_expired_symbols.py --format-output email \
> --smtp-server --sender \
> --password --cc
>
> Signed-off-by: Ray Kinsella
> ---
> devtools/notify-symbol-maintainers.py | 256 ++
> 1 file changed, 256 insertions(+)
> create mode 100755 devtools/notify-symbol-maintainers.py
>
> diff --git a/devtools/notify-symbol-maintainers.py
> b/devtools/notify-symbol-maintainers.py
> new file mode 100755
> index 00..ee554687ff
> --- /dev/null
> +++ b/devtools/notify-symbol-maintainers.py
> @@ -0,0 +1,256 @@
> +#!/usr/bin/env python3
> +# SPDX-License-Identifier: BSD-3-Clause
> +# Copyright(c) 2021 Intel Corporation
> +# pylint: disable=invalid-name
> +'''Tool to notify maintainers of expired symbols'''
> +import smtplib
> +import ssl
> +import sys
> +import subprocess
> +import argparse
> +from argparse import RawTextHelpFormatter
> +import time
> +from email.message import EmailMessage
> +
> +DESCRIPTION = '''
> +Use this script with the output of the DPDK symbol tool, to notify
> maintainers
> +and contributors of expired symbols by email. You need to define the
> environment
> +variable DPDK_GETMAINTAINER_PATH for this tool to work.
> +
> +Use terminal output to review the emails before sending.
> +e.g.
> +$ devtools/symbol-tool.py list-expired --format-output csv \\
> +| DPDK_GETMAINTAINER_PATH=/get_maintainer.pl \\
> +{s} --format-output terminal
> +
> +Then use email output to send the emails to the maintainers.
> +e.g.
> +$ devtools/symbol-tool.py list-expired --format-output csv \\
> +| DPDK_GETMAINTAINER_PATH=/get_maintainer.pl \\
> +{s} --format-output email \\
> +--smtp-server --sender --password
> \\
> +--cc
> +'''
> +
> +EMAIL_TEMPLATE = '''Hi there,
> +
> +Please note the symbols listed below have expired. In line with the DPDK ABI
> +policy, they should be scheduled for removal, in the next DPDK release.
> +
> +For more information, please see the DPDK ABI Policy, section 3.5.3.
> +https://doc.dpdk.org/guides/contributing/abi_policy.html
> +
> +Thanks,
> +
> +The DPDK Symbol Bot
> +
> +'''
> +
> +ABI_POLICY = 'doc/guides/contributing/abi_policy.rst'
> +MAINTAINERS = 'MAINTAINERS'
> +get_maintainer = ['devtools/get-maintainer.sh', \
> + '--email', '-f']
Maybe it's best to make this something that can be overridden. There's
a series to change the .sh files to .py files. Perhaps an environment
variable or argument?
> +def _get_maintainers(libpath):
> +'''Get the maintainers for given library'''
> +try:
> +cmd = get_maintainer + [libpath]
> +result = subprocess.run(cmd, \
> +stdout=subprocess.PIPE, \
> +stderr=subprocess.PIPE,
> +check=True)
> +except subprocess.CalledProcessError:
> +return None
You might consider handling
except FileNotFoundError:
With a graceful exit and error message. In case the get_maintainers
path changes.
> +if result is None:
> +return None
> +
> +email = result.stdout.decode('utf-8')
> +if email == '':
> +return None
> +
> +email = list(filter(None,email.split('\n')))
> +return email
> +
> +default_maintainers = _get_maintainers(ABI_POLICY) + \
> +_get_maintainers(MAINTAINERS)
> +
> +def get_maintainers(libpath):
> +'''Get the maintainers for given library'''
> +maintainers=_get_maintainers(libpath)
> +
> +if maintainers is None:
> +maintainers = default_maintainers
> +
> +return maintainers
> +
> +def get_message(library, symbols, config):
> +'''Build email message from symbols, config and maintainers'''
> +contributors = {}
> +message = {}
> +maintainers = get_maintainers(library)
> +
> +if maintainers != default_maintainers:
> +message['CC'] = default_maintainers.copy()
> +
> +if 'CC' in config:
> +message.setdefault('CC',[]).append(config['CC'])
> +
> +message['Subject'] = 'Expired symbols in {}\n'.format(library)
> +
> +body = EMAIL_TEMPLATE
> +body += '{:<50}{:<25}{:<25}\n'.format('Symbol','Contributor','Email')
> +for sym in symbols:
> +body += ('{:<50}{:<25}{:<25}\n'.format(sym,\
> + symbols[sym]['name'],
> +
Re: [dpdk-dev] [PATCH v5] build: optional NUMA and cpu counts detection
On 21-08-31 07:54:16, Juraj Linkeš wrote:
>
>
> > -Original Message-
> > From: Piotr Kubaj
> > Sent: Tuesday, August 31, 2021 2:55 AM
> > To: Juraj Linkeš
> > Cc: David Christensen ; tho...@monjalon.net;
> > david.march...@redhat.com; bruce.richard...@intel.com;
> > honnappa.nagaraha...@arm.com; ruifeng.w...@arm.com;
> > ferruh.yi...@intel.com; jerinjac...@gmail.com; jer...@marvell.com;
> > step...@networkplumber.org; dev@dpdk.org
> > Subject: Re: [PATCH v5] build: optional NUMA and cpu counts detection
> >
> > Hi,
> >
> > sorry for the late answer.
>
> Thanks for the answer anyway, better late than never.
>
> >
> > I suppose you mean sysctl command, not systemctl.
> >
>
> That's right. What does lscpu say? Are the NUMA nodes non-contiguous like
> this?:
> NUMA node0 CPU(s): 0-63
> NUMA node8 CPU(s): 64-127
> NUMA node252 CPU(s):
> NUMA node253 CPU(s):
> NUMA node254 CPU(s):
> NUMA node255 CPU(s):
>
> > On dual CPU systems, it returns 2. On single CPU ones, 1.
>
> I asked the previous question so that we know the actual numa node number of
> the second CPU. If it's 8, then sysctl does some renumeration and we can't
> use it.
>
> Bruce, maybe we should just parse lscpu output? That introduces a dependency,
> but that may not be such a big deal as lscpu is pretty common.
There's no lscpu on FreeBSD in the base system. There is one in packages, but
it's written only for amd64 / i386 since it uses CPUID.
https://github.com/NanXiao/lscpu
>
> >
> > On 21-08-03 10:21:50, Juraj Linkeš wrote:
> > >
> > >
> > > > -Original Message-
> > > > From: David Christensen
> > > > Sent: Tuesday, August 3, 2021 1:29 AM
> > > > To: Juraj Linkeš ; tho...@monjalon.net;
> > > > david.march...@redhat.com; bruce.richard...@intel.com;
> > > > honnappa.nagaraha...@arm.com; ruifeng.w...@arm.com;
> > > > ferruh.yi...@intel.com; jerinjac...@gmail.com; jer...@marvell.com;
> > > > step...@networkplumber.org; Piotr Kubaj
> > > > Cc: dev@dpdk.org
> > > > Subject: Re: [PATCH v5] build: optional NUMA and cpu counts
> > > > detection
> > > >
> > > >
> > > >
> > > > On 8/2/21 5:44 AM, Juraj Linkeš wrote:
> > > > >> +if os.name == 'posix':
> > > > >> +if os.path.isdir('/sys/devices/system/node'):
> > > > >> +numa_nodes = glob.glob('/sys/devices/system/node/node*')
> > > > >> +numa_nodes.sort()
> > > > >> +print(int(os.path.basename(numa_nodes[-1])[4:]) + 1)
> > > > >> +else:
> > > > >> +subprocess.run(['sysctl', '-n', 'vm.ndomains'],
> > > > >> +check=False)
> > > > >> +
> > > > >
> > > > > Bruce, David, Thomas,
> > > > >
> > > > > Is DPDK actually supported on Power9 FreeBSD? Is anyone using this
> > > > combination? How can we address the open question of what exactly
> > > > does sysctl -n vm.ndomains return on a Power9 FreeBSD system? Or
> > > > should we just leave it as is? Or maybe add 1 to the output (as we do
> > > > in other
> > cases)?
> > > >
> > > > Not supported within IBM, but you can buy OpenPOWER boxes from 3rd
> > > > parties such as Raptor Computing Systems so there may be customers
> > > > using DPDK on POWER with FreeBSD that I don't track. Adding Piotr
> > > > Kubaj who has commented on POWER/FreeBSD issues in this past.
> > > >
> > > > Dave
> > >
> > > Thanks, David.
> > >
> > > Piotr, to provide more context, we're trying to figure out what the
> > > highest
> > NUMA node on a system is.
> > > On P9 systems, here's how NUMA nodes look like in Linux:
> > > NUMA node0 CPU(s): 0-63
> > > NUMA node8 CPU(s): 64-127
> > > NUMA node252 CPU(s):
> > > NUMA node253 CPU(s):
> > > NUMA node254 CPU(s):
> > > NUMA node255 CPU(s):
> > > The highest NUMA with CPUs is node8.
> > >
> > > We're trying to get the highest NUMA with CPUs on P9 FreeBSD systems, but
> > we don't know whether FreeBSD NUMA layout looks the same (does FreeBSD
> > report non-contiguous NUMA nodes as Linxu above, or does it renumerate) a
> > what does "systemctl -n vm.ndomains" return. Could you check these for us?
--
[dpdk-dev] [RFC PATCH 0/2] Flow entities behavior across port restart
It is currently unspecified if flow rules and indirect actions persist across port restart and reconfiguration. Actual behavior differs between PMDs, for example, i40e and mlx5. There are patches that try to fix particular behavior [1] or at least document it for specific PMDs [2]. We propose to codify the least demanding behavior, i.e. no persistence, and to add device capabilities for persistence of rules and actions. This continues the discussion thread [3]. [1]: http://patchwork.dpdk.org/project/dpdk/list/?series=18065 [2]: http://patchwork.dpdk.org/project/dpdk/list/?series=17939 [3]: http://inbox.dpdk.org/dev/20210727073121.895620-5-dkozl...@nvidia.com/ Dmitry Kozlyuk (2): ethdev: add capability to keep flow rules on restart ethdev: add capability to keep indirect actions on restart doc/guides/prog_guide/rte_flow.rst | 21 + lib/ethdev/rte_ethdev.h| 7 +++ 2 files changed, 28 insertions(+) -- 2.25.1
[dpdk-dev] [RFC PATCH 1/2] ethdev: add capability to keep flow rules on restart
Currently, it is not specified what happens to the flow rules when
the device is stopped, possibly reconfigured, then started.
If flow rules were kept, it can be convenient for application
developers, because they wouldn't need to save and restore them.
However, due to the number of flows and possible creation rate it is
impractical to save all flow rules in DPDK layer. This means that flow
rules persistence really depends on whether PMD and HW can implement it
efficiently. It is proposed for PMDs to advertise this capability
if supported using a new flag.
If the device is being reconfigured in a way that is incompatible with
existing flow rules, PMD is required to report an error.
This is mandatory, because flow API does not supply users with
capabilities, so this is the only way for a user to learn that
configuration is invalid. For example, if queue count changes and the
action of a flow rule specifies queues that are going away, the user
must update or remove the flow rule before removing the queues.
Signed-off-by: Dmitry Kozlyuk
Acked-by: Matan Azrad
Acked-by: Ori Kam
---
doc/guides/prog_guide/rte_flow.rst | 9 +
lib/ethdev/rte_ethdev.h| 2 ++
2 files changed, 11 insertions(+)
diff --git a/doc/guides/prog_guide/rte_flow.rst
b/doc/guides/prog_guide/rte_flow.rst
index 2b42d5ec8c..0a03097a7c 100644
--- a/doc/guides/prog_guide/rte_flow.rst
+++ b/doc/guides/prog_guide/rte_flow.rst
@@ -87,6 +87,15 @@ To avoid resource leaks on the PMD side, handles must be
explicitly
destroyed by the application before releasing associated resources such as
queues and ports.
+By default flow rules are implicitly destroyed when the device is stopped.
+If the device advertises ``RTE_DEV_CAPA_FLOW_RULE_KEEP``, flow rules persist
+across device stop and start with possible reconfiguration in between.
+Some configuration changes may be incompatible with existing flow rules,
+in this case ``rte_eth_dev_configure()`` or ``rte_eth_rx/tx_queue_setup()``
+will fail. At this point PMD developers are encouraged to log errors identical
+to the ones that would be emitted by ``rte_flow_create()`` if the new
+configuration was active.
+
The following sections cover:
- **Attributes** (represented by ``struct rte_flow_attr``): properties of a
diff --git a/lib/ethdev/rte_ethdev.h b/lib/ethdev/rte_ethdev.h
index d2b27c351f..1616bdf2dd 100644
--- a/lib/ethdev/rte_ethdev.h
+++ b/lib/ethdev/rte_ethdev.h
@@ -1448,6 +1448,8 @@ struct rte_eth_conf {
#define RTE_ETH_DEV_CAPA_RUNTIME_RX_QUEUE_SETUP 0x0001
/** Device supports Tx queue setup after device started. */
#define RTE_ETH_DEV_CAPA_RUNTIME_TX_QUEUE_SETUP 0x0002
+/** Device keeps flow rules across restart and reconfiguration. */
+#define RTE_ETH_DEV_CAPA_FLOW_RULE_KEEP 0x0004
/**@}*/
/*
--
2.25.1
[dpdk-dev] [RFC PATCH 2/2] ethdev: add capability to keep indirect actions on restart
rte_flow_action_handle_create() did not mention what happens
with an indirect action when a device is stopped, possibly reconfigured,
and started again. It is natural for some indirect actions to be
persistent, like counters and meters; keeping others just saves
application time and complexity. However, not all PMDs can support it.
It is proposed to add a device capability to indicate if indirect actions
are kept across the above sequence or implicitly destroyed.
It may happen that in the future a PMD acquires support for a type of
indirect actions that it cannot keep across a restart. It is undesirable
to stop advertising the capability so that applications that don't use
actions of the problematic type can still take advantage of it.
This is why PMDs are allowed to keep only a subset of indirect actions
provided that the vendor mandatorily documents it.
If the device is being reconfigured in a way that is incompatible with
an existing indirect action, PMD is required to report an error.
This is mandatory, because flow API does not supply users with
capabilities, so this is the only way for a user to learn that
configuration is invalid. For example, if queue count changes and RSS
indirect action specifies queues that are going away, the user must
update the action before removing the queues or remove the action and
all flow rules that were using it.
Signed-off-by: Dmitry Kozlyuk
Acked-by: Matan Azrad
Acked-by: Ori Kam
---
doc/guides/prog_guide/rte_flow.rst | 12
lib/ethdev/rte_ethdev.h| 5 +
2 files changed, 17 insertions(+)
diff --git a/doc/guides/prog_guide/rte_flow.rst
b/doc/guides/prog_guide/rte_flow.rst
index 0a03097a7c..da90b52f48 100644
--- a/doc/guides/prog_guide/rte_flow.rst
+++ b/doc/guides/prog_guide/rte_flow.rst
@@ -2794,6 +2794,18 @@ updated depend on the type of the ``action`` and
different for every type.
The indirect action specified data (e.g. counter) can be queried by
``rte_flow_action_handle_query()``.
+By default indirect actions are destroyed when the device is stopped.
+If the device advertises ``RTE_ETH_DEV_CAPA_FLOW_INDIRECT_ACTION_KEEP``,
+indirect actions persist across the device stop and start with possible
+reconfiguration in between. Some configuration changes may be incompatible
+with existing indirect actions, in this case ``rte_eth_dev_configure()`` and/or
+``rte_eth_rx/tx_queue_setup()`` will fail. At this point PMD developers
+are encouraged to log errors identical to the ones that would be emitted by
+``rte_flow_action_handle_create()`` if the new configuration was active.
+Even if this capability is advertised, there may be kinds of indirect actions
+that the device cannot keep. They are implicitly destroyed at device stop.
+PMD developers must document such kinds of actions if applicable.
+
.. _table_rte_flow_action_handle:
.. table:: INDIRECT
diff --git a/lib/ethdev/rte_ethdev.h b/lib/ethdev/rte_ethdev.h
index 1616bdf2dd..c3be5afcb2 100644
--- a/lib/ethdev/rte_ethdev.h
+++ b/lib/ethdev/rte_ethdev.h
@@ -1450,6 +1450,11 @@ struct rte_eth_conf {
#define RTE_ETH_DEV_CAPA_RUNTIME_TX_QUEUE_SETUP 0x0002
/** Device keeps flow rules across restart and reconfiguration. */
#define RTE_ETH_DEV_CAPA_FLOW_RULE_KEEP 0x0004
+/**
+ * Device keeps indirect actions across restart and reconfiguration.
+ * For a specific PMD this may not be applicable to certain action types.
+ */
+#define RTE_ETH_DEV_CAPA_FLOW_INDIRECT_ACTION_KEEP 0x0008
/**@}*/
/*
--
2.25.1
Re: [dpdk-dev] [PATCH v10 2/3] devtools: script to send notifications of expired symbols
Hello Ray, On Tue, Aug 31, 2021 at 4:51 PM Ray Kinsella wrote: > > Use this script with the output of the DPDK symbol tool, to notify > maintainers of expired symbols by email. You need to define the environment > variable DPDK_GETMAINTAINER_PATH for this tool to work. > > Use terminal output to review the emails before sending. Two comments: - there are references of a previous name for the script, %s/notify_expired_symbols.py/notify-symbol-maintainers.py/g - and a reminder for the empty report that we received yesterday. I think this can be reproduced with: $ DPDK_GETMAINTAINER_PATH=devtools/get_maintainer.pl devtools/notify-symbol-maintainers.py --format-output terminal < mapfile,expired (v21.08,v19.11),contributor name,contributor email > lib/rib,rte_rib6_get_ip,Stephen Hemminger,step...@networkplumber.org > EOF To:Ray Kinsella , Thomas Monjalon Reply-To:no-re...@dpdk.org Subject:Expired symbols in Body:Hi there, Please note the symbols listed below have expired. In line with the DPDK ABI policy, they should be scheduled for removal, in the next DPDK release. For more information, please see the DPDK ABI Policy, section 3.5.3. https://doc.dpdk.org/guides/contributing/abi_policy.html Thanks, The DPDK Symbol Bot SymbolContributor Email Here, empty report. To:Vladimir Medvedkin , step...@networkplumber.org Reply-To:no-re...@dpdk.org CC:Ray Kinsella , Thomas Monjalon Subject:Expired symbols in lib/rib Body:Hi there, Please note the symbols listed below have expired. In line with the DPDK ABI policy, they should be scheduled for removal, in the next DPDK release. For more information, please see the DPDK ABI Policy, section 3.5.3. https://doc.dpdk.org/guides/contributing/abi_policy.html Thanks, The DPDK Symbol Bot SymbolContributor Email rte_rib6_get_ip Stephen Hemminger step...@networkplumber.org -- David Marchand
Re: [dpdk-dev] [PATCH] doc: announce change in vfio dma mapping
On 01-Sep-21 12:42 PM, Ferruh Yigit wrote: On 9/1/2021 12:01 PM, Burakov, Anatoly wrote: On 01-Sep-21 10:56 AM, Ferruh Yigit wrote: On 9/1/2021 2:41 AM, Ding, Xuan wrote: Hi Ferruh, -Original Message- From: Yigit, Ferruh Sent: Wednesday, September 1, 2021 12:01 AM To: Ding, Xuan ; dev@dpdk.org; Burakov, Anatoly Cc: maxime.coque...@redhat.com; Xia, Chenbo ; Hu, Jiayu ; Richardson, Bruce Subject: Re: [PATCH] doc: announce change in vfio dma mapping On 8/31/2021 2:10 PM, Xuan Ding wrote: Currently, the VFIO subsystem will compact adjacent DMA regions for the purposes of saving space in the internal list of mappings. This has a side effect of compacting two separate mappings that just happen to be adjacent in memory. Since VFIO implementation on IA platforms also does not allow partial unmapping of memory mapped for DMA, the current DPDK VFIO implementation will prevent unmapping of accidentally adjacent maps even though it could have been unmapped [1]. The proper fix for this issue is to change the VFIO DMA mapping API to also include page size, and always map memory page-by-page. [1] https://mails.dpdk.org/archives/dev/2021-July/213493.html Signed-off-by: Xuan Ding --- doc/guides/rel_notes/deprecation.rst | 3 +++ 1 file changed, 3 insertions(+) diff --git a/doc/guides/rel_notes/deprecation.rst b/doc/guides/rel_notes/deprecation.rst index 76a4abfd6b..1234420caf 100644 --- a/doc/guides/rel_notes/deprecation.rst +++ b/doc/guides/rel_notes/deprecation.rst @@ -287,3 +287,6 @@ Deprecation Notices reserved bytes to 2 (from 3), and use 1 byte to indicate warnings and other information from the crypto/security operation. This field will be used to communicate events such as soft expiry with IPsec in lookaside mode. + +* vfio: the functions `rte_vfio_container_dma_map` will be amended to + include page size. This change is targeted for DPDK 22.02. Is this means adding a new parameter to API? If so this is an ABI/API break and we can't do this change in the 22.02. Our original plan is add a new parameter in order not to use a new function name, so you mean, any changes to the API can only be done in the LTS version? If so, we can only add a new API and retire the old one in 22.11. We can add a new API anytime. Adding new parameter to an existing API can be done on the ABI break release. So, 22.11 then? Yes. You can add the new API in this release, and start using it. And mark the old API as deprecated in this release. This lets existing binaries to keep using it, but app needs to switch to new API for compilation. Old API can be removed on 22.11, and you will need a deprecation notice before 22.11 for it. Is above plan works for you? We have slightly rethought our approach, and the functionality that Xuan requires does not rely on this API. They can, for all intents and purposes, be considered unrelated issues. I still think it's a good idea to update the API that way, so I would like to do that, and if we have to wait until 22.11 to fix it, I'm OK with that. Since there no longer is any urgency here, it's acceptable to wait for the next LTS to break it. Got it. As far as I understand, main motivation in techboard decision was to prevent the ABI break as much as possible (main reason of decision wasn't deprecation notice being late). But if the correct thing to do is to rename the API (and break the ABI), I don't see the benefit to wait one more year, it is just delaying the impact and adding overhead to us. I am for being pragmatic and doing the change in this release if API rename is better option, perhaps we can visit the issue again in techboard. Can you please describe why renaming API is better option, comparing to adding new API with new parameter? I take it you meant "why renaming API *isn't* a better option". The problem we're solving is that the API in question does not know about page sizes and thus can't map segments page-by-page. I mean I /guess/ we could have two API's (one paged, one not paged), but then we get into all kinds of hairy things about the API leaking the details of underlying platform. Bottom line: i like current API function name. It's concise, it's descriptive. It's only missing a parameter, which i would like to add. A rename has been suggested (deprecate old API, add new API with a different name, and with added parameter), but honestly, I don't see why we have to do that because this is predicated upon the assumption that we *can't* break ABI at all, under any circumstances. Can you please explain to me what is wrong with keeping a versioned symbol? Like, keep the old function around to keep ABI compatibility, but break the API compatibility for those who target 22.02 or later? That's what symbol versioning is *for*, is it not? -- Thanks, Anatoly
Re: [dpdk-dev] [PATCH v2 1/6] bbdev: add capability for CRC16 check
On 8/19/21 2:10 PM, Nicolas Chautru wrote:
Adding a missing operation when CRC16
is being used for TB CRC check.
Signed-off-by: Nicolas Chautru
---
app/test-bbdev/test_bbdev_vector.c | 2 ++
doc/guides/prog_guide/bbdev.rst| 3 +++
doc/guides/rel_notes/release_21_11.rst | 1 +
lib/bbdev/rte_bbdev_op.h | 34 ++
4 files changed, 24 insertions(+), 16 deletions(-)
diff --git a/app/test-bbdev/test_bbdev_vector.c
b/app/test-bbdev/test_bbdev_vector.c
index 614dbd1..8d796b1 100644
--- a/app/test-bbdev/test_bbdev_vector.c
+++ b/app/test-bbdev/test_bbdev_vector.c
@@ -167,6 +167,8 @@
*op_flag_value = RTE_BBDEV_LDPC_CRC_TYPE_24B_CHECK;
else if (!strcmp(token, "RTE_BBDEV_LDPC_CRC_TYPE_24B_DROP"))
*op_flag_value = RTE_BBDEV_LDPC_CRC_TYPE_24B_DROP;
+ else if (!strcmp(token, "RTE_BBDEV_LDPC_CRC_TYPE_16_CHECK"))
+ *op_flag_value = RTE_BBDEV_LDPC_CRC_TYPE_16_CHECK;
else if (!strcmp(token, "RTE_BBDEV_LDPC_DEINTERLEAVER_BYPASS"))
*op_flag_value = RTE_BBDEV_LDPC_DEINTERLEAVER_BYPASS;
else if (!strcmp(token, "RTE_BBDEV_LDPC_HQ_COMBINE_IN_ENABLE"))
diff --git a/doc/guides/prog_guide/bbdev.rst b/doc/guides/prog_guide/bbdev.rst
index 9619280..8bd7cba 100644
--- a/doc/guides/prog_guide/bbdev.rst
+++ b/doc/guides/prog_guide/bbdev.rst
@@ -891,6 +891,9 @@ given below.
|RTE_BBDEV_LDPC_CRC_TYPE_24B_DROP|
| Set to drop the last CRC bits decoding output |
++
+|RTE_BBDEV_LDPC_CRC_TYPE_16_CHECK|
+| Set for code block CRC-16 checking |
+++
|RTE_BBDEV_LDPC_DEINTERLEAVER_BYPASS |
| Set for bit-level de-interleaver bypass on input stream|
++
diff --git a/doc/guides/rel_notes/release_21_11.rst
b/doc/guides/rel_notes/release_21_11.rst
index d707a55..69dd518 100644
--- a/doc/guides/rel_notes/release_21_11.rst
+++ b/doc/guides/rel_notes/release_21_11.rst
@@ -84,6 +84,7 @@ API Changes
Also, make sure to start the actual text at the margin.
===
+* bbdev: Added capability related to more comprehensive CRC options.
ABI Changes
---
diff --git a/lib/bbdev/rte_bbdev_op.h b/lib/bbdev/rte_bbdev_op.h
index f946842..7c44ddd 100644
--- a/lib/bbdev/rte_bbdev_op.h
+++ b/lib/bbdev/rte_bbdev_op.h
@@ -142,51 +142,53 @@ enum rte_bbdev_op_ldpcdec_flag_bitmasks {
RTE_BBDEV_LDPC_CRC_TYPE_24B_CHECK = (1ULL << 1),
/** Set to drop the last CRC bits decoding output */
RTE_BBDEV_LDPC_CRC_TYPE_24B_DROP = (1ULL << 2),
+ /** Set for transport block CRC-16 checking */
+ RTE_BBDEV_LDPC_CRC_TYPE_16_CHECK = (1ULL << 3),
Changing these enums will break the abi backwards.
Why not add the new one at the end ?
Tom
/** Set for bit-level de-interleaver bypass on Rx stream. */
- RTE_BBDEV_LDPC_DEINTERLEAVER_BYPASS = (1ULL << 3),
+ RTE_BBDEV_LDPC_DEINTERLEAVER_BYPASS = (1ULL << 4),
/** Set for HARQ combined input stream enable. */
- RTE_BBDEV_LDPC_HQ_COMBINE_IN_ENABLE = (1ULL << 4),
+ RTE_BBDEV_LDPC_HQ_COMBINE_IN_ENABLE = (1ULL << 5),
/** Set for HARQ combined output stream enable. */
- RTE_BBDEV_LDPC_HQ_COMBINE_OUT_ENABLE = (1ULL << 5),
+ RTE_BBDEV_LDPC_HQ_COMBINE_OUT_ENABLE = (1ULL << 6),
/** Set for LDPC decoder bypass.
* RTE_BBDEV_LDPC_HQ_COMBINE_OUT_ENABLE must be set.
*/
- RTE_BBDEV_LDPC_DECODE_BYPASS = (1ULL << 6),
+ RTE_BBDEV_LDPC_DECODE_BYPASS = (1ULL << 7),
/** Set for soft-output stream enable */
- RTE_BBDEV_LDPC_SOFT_OUT_ENABLE = (1ULL << 7),
+ RTE_BBDEV_LDPC_SOFT_OUT_ENABLE = (1ULL << 8),
/** Set for Rate-Matching bypass on soft-out stream. */
- RTE_BBDEV_LDPC_SOFT_OUT_RM_BYPASS = (1ULL << 8),
+ RTE_BBDEV_LDPC_SOFT_OUT_RM_BYPASS = (1ULL << 9),
/** Set for bit-level de-interleaver bypass on soft-output stream. */
- RTE_BBDEV_LDPC_SOFT_OUT_DEINTERLEAVER_BYPASS = (1ULL << 9),
+ RTE_BBDEV_LDPC_SOFT_OUT_DEINTERLEAVER_BYPASS = (1ULL << 10),
/** Set for iteration stopping on successful decode condition
* i.e. a successful syndrome check.
*/
- RTE_BBDEV_LDPC_ITERATION_STOP_ENABLE = (1ULL << 10),
+ RTE_BBDEV_LDPC_ITERATION_STOP_ENABLE = (1ULL << 11),
/** Set if a device supports decoder dequeue interrupts. */
- RTE_BBDEV_LDPC_DEC_INTERRUPTS = (1ULL << 11),
+ RTE_BBDEV_LDPC_DEC_INTERRUPTS = (1ULL << 12),
/** Set if a device supports scatter-gather functionality. */
- RTE_BBDEV_LDPC_DEC_SCATTER_GA
Re: [dpdk-dev] [PATCH] sched: remove experimental tag from the API
On 01/09/2021 13:19, Jasvinder Singh wrote: > This API was introduced in 18.05, therefore removing > experimental tag to promote it to stable state > > Signed-off-by: Jasvinder Singh > --- > lib/sched/rte_sched.h | 4 > lib/sched/version.map | 2 +- > 2 files changed, 1 insertion(+), 5 deletions(-) > Acked-by: Ray Kinsella
Re: [dpdk-dev] [PATCH] pipeline: remove experimental tag from API
On 01/09/2021 13:20, Jasvinder Singh wrote: > These APIs were introduced in 18.05, therefore removing > experimental tag to promote them to stable state. > > Signed-off-by: Jasvinder Singh > --- > lib/pipeline/rte_port_in_action.h | 10 -- > lib/pipeline/rte_table_action.h | 18 -- > lib/pipeline/version.map | 16 ++-- > 3 files changed, 6 insertions(+), 38 deletions(-) Acked-by: Ray Kinsella
Re: [dpdk-dev] [PATCH] net/softnic: remove experimental table from API
On 01/09/2021 13:20, Jasvinder Singh wrote: > This API was introduced in 18.08, therefore removing > experimental tag to promote it to stable state. > > Signed-off-by: Jasvinder Singh > --- > drivers/net/softnic/rte_eth_softnic.h | 1 - > drivers/net/softnic/version.map | 7 +-- > 2 files changed, 1 insertion(+), 7 deletions(-) > Acked-by: Ray Kinsella
Re: [dpdk-dev] [PATCH v1 3/3] ethdev: promote burst mode API to stable
On 01/09/2021 10:07, Ferruh Yigit wrote: > On 9/1/2021 6:07 AM, Haiyue Wang wrote: >> The DPDK Symbol Bot reports: >> Please note the symbols listed below have expired. In line with the >> DPDK ABI policy, they should be scheduled for removal, in the next >> DPDK release. >> >> Symbol >> rte_eth_rx_burst_mode_get >> rte_eth_tx_burst_mode_get >> >> Signed-off-by: Haiyue Wang > > Acked-by: Ferruh Yigit > Acked-by: Ray Kinsella
Re: [dpdk-dev] [PATCH v1 1/3] net/ixgbe: promote some API to stable
On 01/09/2021 12:13, Wang, Haiyue wrote: >> -Original Message- >> From: Yigit, Ferruh >> Sent: Wednesday, September 1, 2021 17:02 >> To: Wang, Haiyue ; dev@dpdk.org >> Cc: m...@ashroe.eu; tho...@monjalon.net >> Subject: Re: [dpdk-dev] [PATCH v1 1/3] net/ixgbe: promote some API to stable >> >> On 9/1/2021 6:07 AM, Haiyue Wang wrote: >>> The DPDK Symbol Bot reports: >>> Please note the symbols listed below have expired. In line with the >>> DPDK ABI policy, they should be scheduled for removal, in the next >>> DPDK release. >>> >>> Symbol >>> rte_pmd_ixgbe_mdio_lock >>> rte_pmd_ixgbe_mdio_unlock >>> rte_pmd_ixgbe_mdio_unlocked_read >>> rte_pmd_ixgbe_mdio_unlocked_write >>> rte_pmd_ixgbe_upd_fctrl_sbp >> >> I wonder if we should keep PMD specific APIs as experimental (Not talking >> about >> mbuf 'dynfield' / 'dynflag' APIs, we can promote them). > > Yes, makes sense. > >> >> If an application is using PMD specific API, not sure if it will concern >> about >> PMD specific APIs. >> And keeping PMD specific APIs lets us remove them as soon as we can, also >> adds >> additional discourage for users to use them. > > Can update this to DPDK ABI Policy, section 3.5.3. > https://doc.dpdk.org/guides/contributing/abi_policy.html I understand and agree. However we never made any exceptions for PMD specific APIs in the policy. Leave them as experimental for the moment. I will add a clause to the policy Thomas / David - any opinion? > >> >>> >>> Signed-off-by: Haiyue Wang >> >> >> <...> >> >
Re: [dpdk-dev] [PATCH v2 2/6] baseband/turbo_sw: add support for CRC16
On 8/19/21 2:10 PM, Nicolas Chautru wrote:
This is to support the case for operation
where CRC16 is to be appended or checked.
Signed-off-by: Nicolas Chautru
---
doc/guides/rel_notes/release_21_11.rst | 3 +++
drivers/baseband/turbo_sw/bbdev_turbo_software.c | 17 +
2 files changed, 20 insertions(+)
diff --git a/doc/guides/rel_notes/release_21_11.rst
b/doc/guides/rel_notes/release_21_11.rst
index 69dd518..8ca59b7 100644
--- a/doc/guides/rel_notes/release_21_11.rst
+++ b/doc/guides/rel_notes/release_21_11.rst
@@ -55,6 +55,9 @@ New Features
Also, make sure to start the actual text at the margin.
===
+* **Updated the turbo_sw bbdev PMD.**
+
+ Added support for more comprehensive CRC options.
Removed Items
-
diff --git a/drivers/baseband/turbo_sw/bbdev_turbo_software.c
b/drivers/baseband/turbo_sw/bbdev_turbo_software.c
index 77e9a2e..e570044 100644
--- a/drivers/baseband/turbo_sw/bbdev_turbo_software.c
+++ b/drivers/baseband/turbo_sw/bbdev_turbo_software.c
@@ -199,6 +199,7 @@ struct turbo_sw_queue {
.cap.ldpc_enc = {
.capability_flags =
RTE_BBDEV_LDPC_RATE_MATCH |
+ RTE_BBDEV_LDPC_CRC_16_ATTACH |
RTE_BBDEV_LDPC_CRC_24A_ATTACH |
RTE_BBDEV_LDPC_CRC_24B_ATTACH,
.num_buffers_src =
@@ -211,6 +212,7 @@ struct turbo_sw_queue {
.type = RTE_BBDEV_OP_LDPC_DEC,
.cap.ldpc_dec = {
.capability_flags =
+ RTE_BBDEV_LDPC_CRC_TYPE_16_CHECK |
RTE_BBDEV_LDPC_CRC_TYPE_24B_CHECK |
RTE_BBDEV_LDPC_CRC_TYPE_24A_CHECK |
RTE_BBDEV_LDPC_CRC_TYPE_24B_DROP |
@@ -880,6 +882,12 @@ struct turbo_sw_queue {
crc_req.len = in_length_in_bits - 24;
crc_resp.data = q->enc_in;
bblib_lte_crc24b_gen(&crc_req, &crc_resp);
+ } else if (enc->op_flags & RTE_BBDEV_LDPC_CRC_16_ATTACH) {
The 'else if' assumes the new flag is mutually exclusive wrt the other
crc flags.
At least a debug check should be added to verify.
+ rte_memcpy(q->enc_in, in, in_length_in_bytes - 2);
+ crc_req.data = in;
+ crc_req.len = in_length_in_bits - 16;
+ crc_resp.data = q->enc_in;
+ bblib_lte_crc16_gen(&crc_req, &crc_resp);
} else
rte_memcpy(q->enc_in, in, in_length_in_bytes);
@@ -1492,6 +1500,15 @@ struct turbo_sw_queue {
if (!crc_resp.check_passed)
op->status |= 1 << RTE_BBDEV_CRC_ERROR;
}
+ if (check_bit(dec->op_flags, RTE_BBDEV_LDPC_CRC_TYPE_16_CHECK)) {
The series of 'if-statements' means the new flag is not mutually
exclusive wrt the other crc flags.
doing both 24a and 16 would create a mess.
this should likely change to an else-if-statement similar to above.
Tom
+ crc_req.data = adapter_input;
+ crc_req.len = K - dec->n_filler - 16;
+ crc_resp.check_passed = false;
+ crc_resp.data = adapter_input;
+ bblib_lte_crc16_check(&crc_req, &crc_resp);
+ if (!crc_resp.check_passed)
+ op->status |= 1 << RTE_BBDEV_CRC_ERROR;
+ }
#ifdef RTE_BBDEV_OFFLOAD_COST
q_stats->acc_offload_cycles += rte_rdtsc_precise() - start_time;
Re: [dpdk-dev] [PATCH v2 4/6] baseband/acc100: add support for 4G CRC drop
On 8/19/21 2:10 PM, Nicolas Chautru wrote:
This implements in PMD the option to drop the CB CRC
after 4G decoding to help transport block concatenation.
Signed-off-by: Nicolas Chautru
---
doc/guides/bbdevs/acc100.rst | 1 +
doc/guides/rel_notes/release_21_11.rst | 4
drivers/baseband/acc100/rte_acc100_pmd.c | 12 +---
3 files changed, 14 insertions(+), 3 deletions(-)
diff --git a/doc/guides/bbdevs/acc100.rst b/doc/guides/bbdevs/acc100.rst
index ff0fa4b..9fff6ab 100644
--- a/doc/guides/bbdevs/acc100.rst
+++ b/doc/guides/bbdevs/acc100.rst
@@ -58,6 +58,7 @@ ACC100 5G/4G FEC PMD supports the following BBDEV
capabilities:
- ``RTE_BBDEV_TURBO_NEG_LLR_1_BIT_IN`` : set if negative LLR encoder i/p
is supported
- ``RTE_BBDEV_TURBO_POS_LLR_1_BIT_IN`` : set if positive LLR encoder i/p
is supported
- ``RTE_BBDEV_TURBO_DEC_TB_CRC_24B_KEEP`` : keep CRC24B bits appended
while decoding
+ - ``RTE_BBDEV_TURBO_DEC_CRC_24B_DROP`` : option to drop the code block CRC
after decoding
- ``RTE_BBDEV_TURBO_EARLY_TERMINATION`` : set early termination feature
- ``RTE_BBDEV_TURBO_DEC_SCATTER_GATHER`` : supports scatter-gather for
input/output data
- ``RTE_BBDEV_TURBO_HALF_ITERATION_EVEN`` : set half iteration granularity
diff --git a/doc/guides/rel_notes/release_21_11.rst
b/doc/guides/rel_notes/release_21_11.rst
index 8ca59b7..f7843bc 100644
--- a/doc/guides/rel_notes/release_21_11.rst
+++ b/doc/guides/rel_notes/release_21_11.rst
@@ -59,6 +59,10 @@ New Features
Added support for more comprehensive CRC options.
+* **Updated the ACC100 bbdev PMD.**
+
+ Added support for more comprehensive CRC options.
+
Removed Items
-
diff --git a/drivers/baseband/acc100/rte_acc100_pmd.c b/drivers/baseband/acc100/rte_acc100_pmd.c
index 68ba523..891be81 100644
--- a/drivers/baseband/acc100/rte_acc100_pmd.c
+++ b/drivers/baseband/acc100/rte_acc100_pmd.c
@@ -980,6 +980,7 @@
RTE_BBDEV_TURBO_NEG_LLR_1_BIT_IN |
RTE_BBDEV_TURBO_MAP_DEC |
RTE_BBDEV_TURBO_DEC_TB_CRC_24B_KEEP |
+ RTE_BBDEV_TURBO_DEC_CRC_24B_DROP |
RTE_BBDEV_TURBO_DEC_SCATTER_GATHER,
.max_llr_modulus = INT8_MAX,
.num_buffers_src =
@@ -1708,8 +1709,12 @@
}
if ((op->turbo_dec.code_block_mode == RTE_BBDEV_TRANSPORT_BLOCK)
- && !check_bit(op->turbo_dec.op_flags,
- RTE_BBDEV_TURBO_DEC_TB_CRC_24B_KEEP))
+ && !check_bit(op->turbo_dec.op_flags,
+ RTE_BBDEV_TURBO_DEC_TB_CRC_24B_KEEP))
+ crc24_overlap = 24;
+ if ((op->turbo_dec.code_block_mode == RTE_BBDEV_CODE_BLOCK)
These two if-statements can be combined.
+ && check_bit(op->turbo_dec.op_flags,
+ RTE_BBDEV_TURBO_DEC_CRC_24B_DROP))
crc24_overlap = 24;
/* Calculates circular buffer size.
@@ -1744,7 +1749,8 @@
next_triplet = acc100_dma_fill_blk_type_out(
desc, h_output, *h_out_offset,
- k >> 3, next_triplet, ACC100_DMA_BLKID_OUT_HARD);
+ (k - crc24_overlap) >> 3, next_triplet,
crc24_overlap had been set before this patch in the above if-statement
for crc_24b_keep.
so this looks like a bug.
If it is a bug, it should be separated out as its own patch.
Tom
+ ACC100_DMA_BLKID_OUT_HARD);
if (unlikely(next_triplet < 0)) {
rte_bbdev_log(ERR,
"Mismatch between data to process and mbuf data
length in bbdev_op: %p",
Re: [dpdk-dev] [PATCH v2 6/6] bbdev: reduce warning level for one scenario
On 8/19/21 2:10 PM, Nicolas Chautru wrote:
Queue setup may genuinely fail when adding incremental queues
for a given priority level. In that case application would
attempt to configure a queue at a different priority level.
Not an actual error.
Signed-off-by: Nicolas Chautru
---
lib/bbdev/rte_bbdev.c | 7 ---
1 file changed, 4 insertions(+), 3 deletions(-)
diff --git a/lib/bbdev/rte_bbdev.c b/lib/bbdev/rte_bbdev.c
index fc37236..defddcf 100644
--- a/lib/bbdev/rte_bbdev.c
+++ b/lib/bbdev/rte_bbdev.c
@@ -528,9 +528,10 @@ struct rte_bbdev *
ret = dev->dev_ops->queue_setup(dev, queue_id, (conf != NULL) ?
conf : &dev_info.default_queue_conf);
if (ret < 0) {
- rte_bbdev_log(ERR,
- "Device %u queue %u setup failed", dev_id,
- queue_id);
+ /* This may happen when trying different priority levels */
+ rte_bbdev_log(INFO,
+ "Device %u queue %u setup failed",
+ dev_id, queue_id);
Earlier code tears down the exiting plumbing.
If it is ok to fail, should this block move earlier before the teardown ?
Tom
return ret;
}
[dpdk-dev] [PATCH 0/4] drivers/qat: isolate implementations of qat generations
This patchset introduces new qat driver structure and updates existing symmetric crypto qat PMD. The purpose of the change is to isolate QAT generation specific implementations from one to another. It is expected the changes to the specific generation driver code does minimum impact to other generations' implementations. Also adding the support to new features or new qat generation hardware will have zero impact to existing functionalities. Arek Kusztal (4): common/qat: isolate implementations of qat generations crypto/qat: isolate implementations of symmetric operations crypto/qat: move capabilities initialization to spec files common/qat: add extra data to qat pci dev drivers/common/qat/dev/qat_dev_gen1.c | 252 + drivers/common/qat/dev/qat_dev_gen1.h | 55 ++ drivers/common/qat/dev/qat_dev_gen2.c | 39 ++ drivers/common/qat/dev/qat_dev_gen3.c | 77 +++ drivers/common/qat/dev/qat_dev_gen4.c | 285 ++ drivers/common/qat/dev/qat_dev_gen4.h | 18 + drivers/common/qat/meson.build| 12 +- drivers/common/qat/qat_common.h | 2 + drivers/common/qat/qat_device.c | 183 +++--- drivers/common/qat/qat_device.h | 28 +- drivers/common/qat/qat_qp.c | 641 -- drivers/common/qat/qat_qp.h | 54 +- drivers/crypto/qat/dev/qat_sym_pmd_gen1.c | 78 +++ drivers/crypto/qat/dev/qat_sym_pmd_gen1.h | 15 + drivers/crypto/qat/dev/qat_sym_pmd_gen2.c | 103 drivers/crypto/qat/dev/qat_sym_pmd_gen3.c | 63 +++ drivers/crypto/qat/dev/qat_sym_pmd_gen4.c | 107 drivers/crypto/qat/qat_sym_pmd.c | 188 ++- drivers/crypto/qat/qat_sym_pmd.h | 40 ++ 19 files changed, 1540 insertions(+), 700 deletions(-) create mode 100644 drivers/common/qat/dev/qat_dev_gen1.c create mode 100644 drivers/common/qat/dev/qat_dev_gen1.h create mode 100644 drivers/common/qat/dev/qat_dev_gen2.c create mode 100644 drivers/common/qat/dev/qat_dev_gen3.c create mode 100644 drivers/common/qat/dev/qat_dev_gen4.c create mode 100644 drivers/common/qat/dev/qat_dev_gen4.h create mode 100644 drivers/crypto/qat/dev/qat_sym_pmd_gen1.c create mode 100644 drivers/crypto/qat/dev/qat_sym_pmd_gen1.h create mode 100644 drivers/crypto/qat/dev/qat_sym_pmd_gen2.c create mode 100644 drivers/crypto/qat/dev/qat_sym_pmd_gen3.c create mode 100644 drivers/crypto/qat/dev/qat_sym_pmd_gen4.c -- 2.30.2
[dpdk-dev] [PATCH 1/4] common/qat: isolate implementations of qat generations
This commit isolates implementations of common part in QAT PMD.
When changing/expanding particular generation code, other generations
code should left intact. Generation code in drivers common is
invisible to other generations code.
Signed-off-by: Arek Kusztal
---
drivers/common/qat/dev/qat_dev_gen1.c | 245 ++
drivers/common/qat/dev/qat_dev_gen1.h | 52 +++
drivers/common/qat/dev/qat_dev_gen2.c | 38 ++
drivers/common/qat/dev/qat_dev_gen3.c | 76 +++
drivers/common/qat/dev/qat_dev_gen4.c | 258 +++
drivers/common/qat/meson.build| 4 +
drivers/common/qat/qat_common.h | 2 +
drivers/common/qat/qat_device.c | 117 ++---
drivers/common/qat/qat_device.h | 24 +-
drivers/common/qat/qat_qp.c | 641 +-
drivers/common/qat/qat_qp.h | 45 +-
11 files changed, 983 insertions(+), 519 deletions(-)
create mode 100644 drivers/common/qat/dev/qat_dev_gen1.c
create mode 100644 drivers/common/qat/dev/qat_dev_gen1.h
create mode 100644 drivers/common/qat/dev/qat_dev_gen2.c
create mode 100644 drivers/common/qat/dev/qat_dev_gen3.c
create mode 100644 drivers/common/qat/dev/qat_dev_gen4.c
diff --git a/drivers/common/qat/dev/qat_dev_gen1.c
b/drivers/common/qat/dev/qat_dev_gen1.c
new file mode 100644
index 00..4d60c2a051
--- /dev/null
+++ b/drivers/common/qat/dev/qat_dev_gen1.c
@@ -0,0 +1,245 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2021 Intel Corporation
+ */
+
+#include "qat_device.h"
+#include "qat_qp.h"
+#include "adf_transport_access_macros.h"
+#include "qat_dev_gen1.h"
+
+#include
+
+#define ADF_ARB_REG_SLOT 0x1000
+
+#define WRITE_CSR_ARB_RINGSRVARBEN(csr_addr, index, value) \
+ ADF_CSR_WR(csr_addr, ADF_ARB_RINGSRVARBEN_OFFSET + \
+ (ADF_ARB_REG_SLOT * index), value)
+
+__extension__
+const struct qat_qp_hw_data qat_gen1_qps[QAT_MAX_SERVICES]
+[ADF_MAX_QPS_ON_ANY_SERVICE] = {
+ /* queue pairs which provide an asymmetric crypto service */
+ [QAT_SERVICE_ASYMMETRIC] = {
+ {
+ .service_type = QAT_SERVICE_ASYMMETRIC,
+ .hw_bundle_num = 0,
+ .tx_ring_num = 0,
+ .rx_ring_num = 8,
+ .tx_msg_size = 64,
+ .rx_msg_size = 32,
+
+ }, {
+ .service_type = QAT_SERVICE_ASYMMETRIC,
+ .hw_bundle_num = 0,
+ .tx_ring_num = 1,
+ .rx_ring_num = 9,
+ .tx_msg_size = 64,
+ .rx_msg_size = 32,
+ }
+ },
+ /* queue pairs which provide a symmetric crypto service */
+ [QAT_SERVICE_SYMMETRIC] = {
+ {
+ .service_type = QAT_SERVICE_SYMMETRIC,
+ .hw_bundle_num = 0,
+ .tx_ring_num = 2,
+ .rx_ring_num = 10,
+ .tx_msg_size = 128,
+ .rx_msg_size = 32,
+ },
+ {
+ .service_type = QAT_SERVICE_SYMMETRIC,
+ .hw_bundle_num = 0,
+ .tx_ring_num = 3,
+ .rx_ring_num = 11,
+ .tx_msg_size = 128,
+ .rx_msg_size = 32,
+ }
+ },
+ /* queue pairs which provide a compression service */
+ [QAT_SERVICE_COMPRESSION] = {
+ {
+ .service_type = QAT_SERVICE_COMPRESSION,
+ .hw_bundle_num = 0,
+ .tx_ring_num = 6,
+ .rx_ring_num = 14,
+ .tx_msg_size = 128,
+ .rx_msg_size = 32,
+ }, {
+ .service_type = QAT_SERVICE_COMPRESSION,
+ .hw_bundle_num = 0,
+ .tx_ring_num = 7,
+ .rx_ring_num = 15,
+ .tx_msg_size = 128,
+ .rx_msg_size = 32,
+ }
+ }
+};
+
+int
+qat_qp_rings_per_service_gen1(struct qat_pci_device *qat_dev,
+ enum qat_service_type service)
+{
+ int i = 0, count = 0, max_ops_per_srv = 0;
+ const struct qat_qp_hw_data *sym_hw_qps =
+ qat_gen_config[qat_dev->qat_dev_gen]
+ .qp_hw_data[service];
+
+ max_ops_per_srv = ADF_MAX_QPS_ON_ANY_SERVICE;
+ for (i = 0, count = 0; i < max_ops_per_srv; i++)
+ if (sym_hw_qps[i].service_type == service)
+ count++;
+ return count;
+}
+
+void
+qat_qp_csr_build_ring_base_gen1(void *io_addr,
+ struct qat_queue *queue)
+{
+ uint64_t queue_base;
+
+ queue_base = BUILD_RING_BASE_ADDR(queue->base_phys_addr,
+ queue->
[dpdk-dev] [PATCH 2/4] crypto/qat: isolate implementations of symmetric operations
This commit isolates implementations of symmetric part in QAT PMD.
When changing/expanding particular generation code, other generations
code should left intact. Generation code in drivers crypto is
invisible to other generations code.
Signed-off-by: Arek Kusztal
---
drivers/common/qat/meson.build| 6 +-
drivers/crypto/qat/dev/qat_sym_pmd_gen1.c | 55 ++
drivers/crypto/qat/dev/qat_sym_pmd_gen1.h | 15 +++
drivers/crypto/qat/dev/qat_sym_pmd_gen2.c | 80 +++
drivers/crypto/qat/dev/qat_sym_pmd_gen3.c | 39 +++
drivers/crypto/qat/dev/qat_sym_pmd_gen4.c | 82 +++
drivers/crypto/qat/qat_sym_pmd.c | 120 +-
drivers/crypto/qat/qat_sym_pmd.h | 23 +
8 files changed, 322 insertions(+), 98 deletions(-)
create mode 100644 drivers/crypto/qat/dev/qat_sym_pmd_gen1.c
create mode 100644 drivers/crypto/qat/dev/qat_sym_pmd_gen1.h
create mode 100644 drivers/crypto/qat/dev/qat_sym_pmd_gen2.c
create mode 100644 drivers/crypto/qat/dev/qat_sym_pmd_gen3.c
create mode 100644 drivers/crypto/qat/dev/qat_sym_pmd_gen4.c
diff --git a/drivers/common/qat/meson.build b/drivers/common/qat/meson.build
index 532e0fabb3..de54004b4c 100644
--- a/drivers/common/qat/meson.build
+++ b/drivers/common/qat/meson.build
@@ -69,7 +69,11 @@ endif
if qat_crypto
foreach f: ['qat_sym_pmd.c', 'qat_sym.c', 'qat_sym_session.c',
-'qat_sym_hw_dp.c', 'qat_asym_pmd.c', 'qat_asym.c']
+'qat_sym_hw_dp.c', 'qat_asym_pmd.c', 'qat_asym.c',
+'dev/qat_sym_pmd_gen1.c',
+'dev/qat_sym_pmd_gen2.c',
+'dev/qat_sym_pmd_gen3.c',
+'dev/qat_sym_pmd_gen4.c']
sources += files(join_paths(qat_crypto_relpath, f))
endforeach
deps += ['security']
diff --git a/drivers/crypto/qat/dev/qat_sym_pmd_gen1.c
b/drivers/crypto/qat/dev/qat_sym_pmd_gen1.c
new file mode 100644
index 00..4a4dc9ab55
--- /dev/null
+++ b/drivers/crypto/qat/dev/qat_sym_pmd_gen1.c
@@ -0,0 +1,55 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2021 Intel Corporation
+ */
+
+#include
+#include
+#include "qat_sym_pmd.h"
+#include "qat_sym_session.h"
+#include "qat_sym.h"
+#include "qat_sym_pmd_gen1.h"
+
+int qat_sym_qp_setup_gen1(struct rte_cryptodev *dev, uint16_t qp_id,
+ const struct rte_cryptodev_qp_conf *qp_conf,
+ int socket_id)
+{
+ struct qat_qp_config qat_qp_conf = { };
+ const struct qat_qp_hw_data *sym_hw_qps = NULL;
+ struct qat_sym_dev_private *qat_sym_private = dev->data->dev_private;
+ struct qat_pci_device *qat_dev = qat_sym_private->qat_dev;
+
+ sym_hw_qps = qat_gen_config[qat_dev->qat_dev_gen]
+ .qp_hw_data[QAT_SERVICE_SYMMETRIC];
+ qat_qp_conf.hw = sym_hw_qps + qp_id;
+
+ return qat_sym_qp_setup(dev, qp_id, qp_conf, qat_qp_conf, socket_id);
+}
+
+struct rte_cryptodev_ops crypto_qat_gen1_ops = {
+
+ /* Device related operations */
+ .dev_configure = qat_sym_dev_config,
+ .dev_start = qat_sym_dev_start,
+ .dev_stop = qat_sym_dev_stop,
+ .dev_close = qat_sym_dev_close,
+ .dev_infos_get = qat_sym_dev_info_get,
+
+ .stats_get = qat_sym_stats_get,
+ .stats_reset= qat_sym_stats_reset,
+ .queue_pair_setup = qat_sym_qp_setup_gen1,
+ .queue_pair_release = qat_sym_qp_release,
+
+ /* Crypto related operations */
+ .sym_session_get_size = qat_sym_session_get_private_size,
+ .sym_session_configure = qat_sym_session_configure,
+ .sym_session_clear = qat_sym_session_clear,
+
+ /* Raw data-path API related operations */
+ .sym_get_raw_dp_ctx_size = qat_sym_get_dp_ctx_size,
+ .sym_configure_raw_dp_ctx = qat_sym_configure_dp_ctx,
+};
+
+RTE_INIT(qat_sym_pmd_gen1_init)
+{
+ QAT_CRYPTODEV_OPS[QAT_GEN1] = &crypto_qat_gen1_ops;
+}
diff --git a/drivers/crypto/qat/dev/qat_sym_pmd_gen1.h
b/drivers/crypto/qat/dev/qat_sym_pmd_gen1.h
new file mode 100644
index 00..397faab0b0
--- /dev/null
+++ b/drivers/crypto/qat/dev/qat_sym_pmd_gen1.h
@@ -0,0 +1,15 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2021 Intel Corporation
+ */
+
+#include
+#include
+
+#ifndef _QAT_DEV_GEN_H_
+#define _QAT_DEV_GEN_H_
+
+int qat_sym_qp_setup_gen1(struct rte_cryptodev *dev, uint16_t qp_id,
+ const struct rte_cryptodev_qp_conf *qp_conf,
+ int socket_id);
+
+#endif
diff --git a/drivers/crypto/qat/dev/qat_sym_pmd_gen2.c
b/drivers/crypto/qat/dev/qat_sym_pmd_gen2.c
new file mode 100644
index 00..6344d7de13
--- /dev/null
+++ b/drivers/crypto/qat/dev/qat_sym_pmd_gen2.c
@@ -0,0 +1,80 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2021 Intel Corporation
+ */
+
+#include
+#incl
[dpdk-dev] [PATCH 3/4] crypto/qat: move capabilities initialization to spec files
Move capabilites static struct to particular generations into
separate translation units that it can be isolated from each other.
Signed-off-by: Arek Kusztal
---
drivers/crypto/qat/dev/qat_sym_pmd_gen1.c | 27 -
drivers/crypto/qat/dev/qat_sym_pmd_gen2.c | 25 -
drivers/crypto/qat/dev/qat_sym_pmd_gen3.c | 26 -
drivers/crypto/qat/dev/qat_sym_pmd_gen4.c | 24 +++-
drivers/crypto/qat/qat_sym_pmd.c | 68 +++
drivers/crypto/qat/qat_sym_pmd.h | 19 ++-
6 files changed, 135 insertions(+), 54 deletions(-)
diff --git a/drivers/crypto/qat/dev/qat_sym_pmd_gen1.c
b/drivers/crypto/qat/dev/qat_sym_pmd_gen1.c
index 4a4dc9ab55..40ec77f846 100644
--- a/drivers/crypto/qat/dev/qat_sym_pmd_gen1.c
+++ b/drivers/crypto/qat/dev/qat_sym_pmd_gen1.c
@@ -8,6 +8,12 @@
#include "qat_sym_session.h"
#include "qat_sym.h"
#include "qat_sym_pmd_gen1.h"
+#include "qat_sym_capabilities.h"
+
+static struct rte_cryptodev_capabilities qat_gen1_sym_capabilities[] = {
+ QAT_BASE_GEN1_SYM_CAPABILITIES,
+ RTE_CRYPTODEV_END_OF_CAPABILITIES_LIST()
+};
int qat_sym_qp_setup_gen1(struct rte_cryptodev *dev, uint16_t qp_id,
const struct rte_cryptodev_qp_conf *qp_conf,
@@ -49,7 +55,24 @@ struct rte_cryptodev_ops crypto_qat_gen1_ops = {
.sym_configure_raw_dp_ctx = qat_sym_configure_dp_ctx,
};
-RTE_INIT(qat_sym_pmd_gen1_init)
+static struct
+qat_capabilities_info get_capabilties_gen1(
+ struct qat_pci_device *qat_dev __rte_unused)
{
- QAT_CRYPTODEV_OPS[QAT_GEN1] = &crypto_qat_gen1_ops;
+ struct qat_capabilities_info capa_info;
+
+ capa_info.data = qat_gen1_sym_capabilities;
+ capa_info.size = sizeof(qat_gen1_sym_capabilities);
+ return capa_info;
}
+
+static struct
+qat_sym_pmd_dev_ops qat_sym_pmd_ops_gen1 = {
+ .qat_sym_get_capabilities = get_capabilties_gen1,
+};
+
+RTE_INIT(qat_sym_pmd_gen1_init)
+{
+ QAT_CRYPTODEV_OPS[QAT_GEN1] = &crypto_qat_gen1_ops;
+ qat_sym_pmd_ops[QAT_GEN1] = &qat_sym_pmd_ops_gen1;
+}
\ No newline at end of file
diff --git a/drivers/crypto/qat/dev/qat_sym_pmd_gen2.c
b/drivers/crypto/qat/dev/qat_sym_pmd_gen2.c
index 6344d7de13..18dfca3a84 100644
--- a/drivers/crypto/qat/dev/qat_sym_pmd_gen2.c
+++ b/drivers/crypto/qat/dev/qat_sym_pmd_gen2.c
@@ -7,9 +7,16 @@
#include "qat_sym_pmd.h"
#include "qat_sym_session.h"
#include "qat_sym.h"
+#include "qat_sym_capabilities.h"
#define MIXED_CRYPTO_MIN_FW_VER 0x0409
+static struct rte_cryptodev_capabilities qat_gen2_sym_capabilities[] = {
+ QAT_BASE_GEN1_SYM_CAPABILITIES,
+ QAT_EXTRA_GEN2_SYM_CAPABILITIES,
+ RTE_CRYPTODEV_END_OF_CAPABILITIES_LIST()
+};
+
static int qat_sym_qp_setup_gen2(struct rte_cryptodev *dev, uint16_t qp_id,
const struct rte_cryptodev_qp_conf *qp_conf,
int socket_id)
@@ -74,7 +81,23 @@ struct rte_cryptodev_ops crypto_qat_gen2_ops = {
.sym_configure_raw_dp_ctx = qat_sym_configure_dp_ctx,
};
+static struct
+qat_capabilities_info get_capabilties_gen2(
+ struct qat_pci_device *qat_dev __rte_unused)
+{
+ struct qat_capabilities_info capa_info;
+ capa_info.data = qat_gen2_sym_capabilities;
+ capa_info.size = sizeof(qat_gen2_sym_capabilities);
+ return capa_info;
+}
+
+static struct
+qat_sym_pmd_dev_ops qat_sym_pmd_ops_gen2 = {
+ .qat_sym_get_capabilities = get_capabilties_gen2,
+};
+
RTE_INIT(qat_sym_pmd_gen2)
{
- QAT_CRYPTODEV_OPS[QAT_GEN2] = &crypto_qat_gen2_ops;
+ QAT_CRYPTODEV_OPS[QAT_GEN2] = &crypto_qat_gen2_ops;
+ qat_sym_pmd_ops[QAT_GEN2] = &qat_sym_pmd_ops_gen2;
}
diff --git a/drivers/crypto/qat/dev/qat_sym_pmd_gen3.c
b/drivers/crypto/qat/dev/qat_sym_pmd_gen3.c
index f8488cd122..e914a09362 100644
--- a/drivers/crypto/qat/dev/qat_sym_pmd_gen3.c
+++ b/drivers/crypto/qat/dev/qat_sym_pmd_gen3.c
@@ -9,6 +9,13 @@
#include "qat_sym.h"
#include "qat_sym_pmd_gen1.h"
+static struct rte_cryptodev_capabilities qat_gen3_sym_capabilities[] = {
+ QAT_BASE_GEN1_SYM_CAPABILITIES,
+ QAT_EXTRA_GEN2_SYM_CAPABILITIES,
+ QAT_EXTRA_GEN3_SYM_CAPABILITIES,
+ RTE_CRYPTODEV_END_OF_CAPABILITIES_LIST()
+};
+
struct rte_cryptodev_ops crypto_qat_gen3_ops = {
/* Device related operations */
@@ -33,7 +40,24 @@ struct rte_cryptodev_ops crypto_qat_gen3_ops = {
.sym_configure_raw_dp_ctx = qat_sym_configure_dp_ctx,
};
+static struct
+qat_capabilities_info get_capabilties_gen3(
+ struct qat_pci_device *qat_dev __rte_unused)
+{
+ struct qat_capabilities_info capa_info;
+ capa_info.data = qat_gen3_sym_capabilities;
+ capa_info.size = sizeof(qat_gen3_sym_capabilities);
+ return capa_info;
+}
+
+static struct
+qat_sym_pmd_dev_ops qat_sym_pmd_ops_gen3 = {
+ .qat_sym_get_capabilities = get_capabilties_gen3,
+};
+
+
RTE_INIT(qat_sym_pmd_g
[dpdk-dev] [PATCH 4/4] common/qat: add extra data to qat pci dev
Add private data to qat_pci_device struct that will be
visible only by specific generation it belongs to.
Signed-off-by: Arek Kusztal
---
drivers/common/qat/dev/qat_dev_gen1.c | 7 +++
drivers/common/qat/dev/qat_dev_gen1.h | 3 ++
drivers/common/qat/dev/qat_dev_gen2.c | 1 +
drivers/common/qat/dev/qat_dev_gen3.c | 1 +
drivers/common/qat/dev/qat_dev_gen4.c | 31 ++-
drivers/common/qat/dev/qat_dev_gen4.h | 18 +++
drivers/common/qat/meson.build| 2 +
drivers/common/qat/qat_device.c | 66 +++
drivers/common/qat/qat_device.h | 10 ++--
drivers/common/qat/qat_qp.h | 9
drivers/crypto/qat/dev/qat_sym_pmd_gen4.c | 7 ++-
11 files changed, 113 insertions(+), 42 deletions(-)
create mode 100644 drivers/common/qat/dev/qat_dev_gen4.h
diff --git a/drivers/common/qat/dev/qat_dev_gen1.c
b/drivers/common/qat/dev/qat_dev_gen1.c
index 4d60c2a051..3c7a558959 100644
--- a/drivers/common/qat/dev/qat_dev_gen1.c
+++ b/drivers/common/qat/dev/qat_dev_gen1.c
@@ -227,11 +227,18 @@ qat_dev_read_config_gen1(struct qat_pci_device *qat_dev
__rte_unused)
return 0;
}
+int
+qat_dev_get_extra_size_gen1(void)
+{
+ return 0;
+}
+
static struct qat_dev_hw_spec_funcs qat_dev_hw_spec_gen1 = {
.qat_dev_reset_ring_pairs = qat_reset_ring_pairs_gen1,
.qat_dev_get_transport_bar = qat_dev_get_transport_bar_gen1,
.qat_dev_get_misc_bar = qat_dev_get_misc_bar_gen1,
.qat_dev_read_config= qat_dev_read_config_gen1,
+ .qat_dev_get_extra_size = qat_dev_get_extra_size_gen1,
};
RTE_INIT(qat_dev_gen_gen1_init)
diff --git a/drivers/common/qat/dev/qat_dev_gen1.h
b/drivers/common/qat/dev/qat_dev_gen1.h
index 9bf4fcf01b..ec0af94655 100644
--- a/drivers/common/qat/dev/qat_dev_gen1.h
+++ b/drivers/common/qat/dev/qat_dev_gen1.h
@@ -13,6 +13,9 @@
extern const struct qat_qp_hw_data qat_gen1_qps[QAT_MAX_SERVICES]
[ADF_MAX_QPS_ON_ANY_SERVICE];
+int
+qat_dev_get_extra_size_gen1(void);
+
int
qat_qp_rings_per_service_gen1(struct qat_pci_device *qat_dev,
enum qat_service_type service);
diff --git a/drivers/common/qat/dev/qat_dev_gen2.c
b/drivers/common/qat/dev/qat_dev_gen2.c
index ad1b643e00..856463c06f 100644
--- a/drivers/common/qat/dev/qat_dev_gen2.c
+++ b/drivers/common/qat/dev/qat_dev_gen2.c
@@ -25,6 +25,7 @@ static struct qat_dev_hw_spec_funcs qat_dev_hw_spec_gen2 = {
.qat_dev_get_transport_bar = qat_dev_get_transport_bar_gen1,
.qat_dev_get_misc_bar = qat_dev_get_misc_bar_gen1,
.qat_dev_read_config= qat_dev_read_config_gen1,
+ .qat_dev_get_extra_size = qat_dev_get_extra_size_gen1,
};
RTE_INIT(qat_dev_gen_gen2_init)
diff --git a/drivers/common/qat/dev/qat_dev_gen3.c
b/drivers/common/qat/dev/qat_dev_gen3.c
index 407d21576b..237712f1ef 100644
--- a/drivers/common/qat/dev/qat_dev_gen3.c
+++ b/drivers/common/qat/dev/qat_dev_gen3.c
@@ -63,6 +63,7 @@ static struct qat_dev_hw_spec_funcs qat_dev_hw_spec_gen3 = {
.qat_dev_get_transport_bar = qat_dev_get_transport_bar_gen1,
.qat_dev_get_misc_bar = qat_dev_get_misc_bar_gen1,
.qat_dev_read_config= qat_dev_read_config_gen1,
+ .qat_dev_get_extra_size = qat_dev_get_extra_size_gen1,
};
RTE_INIT(qat_dev_gen_gen3_init)
diff --git a/drivers/common/qat/dev/qat_dev_gen4.c
b/drivers/common/qat/dev/qat_dev_gen4.c
index 6394e17dde..aecdedf375 100644
--- a/drivers/common/qat/dev/qat_dev_gen4.c
+++ b/drivers/common/qat/dev/qat_dev_gen4.c
@@ -10,9 +10,27 @@
#include "adf_transport_access_macros_gen4vf.h"
#include "adf_pf2vf_msg.h"
#include "qat_pf2vf.h"
+#include "qat_dev_gen4.h"
#include
+struct qat_dev_gen4_extra {
+ struct qat_qp_hw_data qp_gen4_data[QAT_GEN4_BUNDLE_NUM]
+ [QAT_GEN4_QPS_PER_BUNDLE_NUM];
+};
+
+enum qat_service_type qat_dev4_get_qp_serv(
+ struct qat_dev_gen4_extra *dev_extra, int ring_pair)
+{
+ return dev_extra->qp_gen4_data[ring_pair][0].service_type;
+}
+
+const struct qat_qp_hw_data *qat_dev4_get_hw(
+ struct qat_dev_gen4_extra *dev_extra, int ring_pair)
+{
+ return &dev_extra->qp_gen4_data[ring_pair][0];
+}
+
static struct qat_pf2vf_dev qat_pf2vf_gen4 = {
.pf2vf_offset = ADF_4XXXIOV_PF2VM_OFFSET,
.vf2pf_offset = ADF_4XXXIOV_VM2PF_OFFSET,
@@ -38,10 +56,11 @@ qat_qp_rings_per_service_gen4(struct qat_pci_device
*qat_dev,
enum qat_service_type service)
{
int i = 0, count = 0, max_ops_per_srv = 0;
+ struct qat_dev_gen4_extra *dev_extra = qat_dev->dev_private;
max_ops_per_srv = QAT_GEN4_BUNDLE_NUM;
for (i = 0, count = 0; i < max_ops_per_srv; i++)
- if (qat_dev->qp_gen4_data[i][0].service_type == service)
+ if (dev_extra->qp_gen4_data[i][0].service_ty
Re: [dpdk-dev] [PATCH v1] bbdev: remove experimental tag from API
> -Original Message- > From: Hemant Agrawal > Sent: Tuesday, August 31, 2021 8:53 PM > To: David Marchand ; Chautru, Nicolas > > Cc: dev ; Akhil Goyal ; Thomas > Monjalon ; Tom Rix ; Zhang, > Mingshan ; Joshi, Arun > ; Ray Kinsella ; Maxime Coquelin > > Subject: RE: [PATCH v1] bbdev: remove experimental tag from API > > > > > On Tue, Aug 31, 2021 at 6:27 PM Nicolas Chautru > > wrote: > > > > > > This was previously suggested last year > > > > > > https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpatc > > > hes.dpdk.org%2Fproject%2Fdpdk%2Fpatch%2F1593213242-157394-2-git- > > send-e > > > mail- > > > nicolas.chautru%40intel.com%2F&data=04%7C01%7Chemant.agrawal% > > > > > 40nxp.com%7Cfcdfb339ec284057db1508d96c9dd048%7C686ea1d3bc2b4c6f > > a92cd99 > > > > > c5c301635%7C0%7C0%7C637660247324212288%7CUnknown%7CTWFpbGZ > > sb3d8eyJWIjo > > > > > iMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1 > > 000& > > > > ;sdata=%2FYwaT29I4XlUxedcdCeyvM5z1R5QeNGTLLPY3XQBKjU%3D& > > ;reserved=0 > > > but there was request from community to wait another year to confirm > > formally this api is mature. > > > > The request was to wait for a new vendor (i.e. non Intel) to implement > > a bbdev driver for their hw. > > > > NXP proposed a driver for this class: > > > https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpatc > > h > work.dpdk.org%2Fproject%2Fdpdk%2Flist%2F%3Fseries%3D16642%26state > > %3D%252A%26archive%3Dboth&data=04%7C01%7Chemant.agrawal% > > 40nxp.com%7Cfcdfb339ec284057db1508d96c9dd048%7C686ea1d3bc2b4c6f > > a92cd99c5c301635%7C0%7C0%7C637660247324212288%7CUnknown%7CT > > > WFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLC > > > JXVCI6Mn0%3D%7C1000&sdata=iorsQ7wAXyU2%2FaqeV5CrDp9asl8ztZ > > nSrbDuJpCv9Gk%3D&reserved=0 > > > > What is the status of this driver? > > [Hemant] We are in process of resending the patches in next few days. Hi Hemant, You are not suggesting any change to the actual API and exposed bbdev symbols right? Based on this please ack the patch to remove the experimental patch. Thanks Nic > > > I see no update on the bbdev API, which seems good, but a confirmation > > is needed. > > > > > > -- > > David Marchand
Re: [dpdk-dev] [PATCH v1] bbdev: remove experimental tag from API
> -Original Message- > From: Kinsella, Ray > Sent: Wednesday, September 1, 2021 4:15 AM > To: Hemant Agrawal ; David Marchand > ; Chautru, Nicolas > > Cc: dev ; Akhil Goyal ; Thomas > Monjalon ; Tom Rix ; Zhang, > Mingshan ; Joshi, Arun > ; Maxime Coquelin > Subject: Re: [PATCH v1] bbdev: remove experimental tag from API > > > > On 01/09/2021 04:52, Hemant Agrawal wrote: > >> > >> On Tue, Aug 31, 2021 at 6:27 PM Nicolas Chautru > >> wrote: > >>> > >>> This was previously suggested last year > >>> > >> > https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpat > >> c > >>> hes.dpdk.org%2Fproject%2Fdpdk%2Fpatch%2F1593213242-157394-2- > git- > >> send-e > >>> mail- > >> > nicolas.chautru%40intel.com%2F&data=04%7C01%7Chemant.agrawal% > >>> > >> > 40nxp.com%7Cfcdfb339ec284057db1508d96c9dd048%7C686ea1d3bc2b4c6f > >> a92cd99 > >>> > >> c5c301635%7C0%7C0%7C637660247324212288%7CUnknown%7CTWFpbGZ > >> sb3d8eyJWIjo > >>> > >> > iMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1 > >> 000& > >>> > ;sdata=%2FYwaT29I4XlUxedcdCeyvM5z1R5QeNGTLLPY3XQBKjU%3D& > >> ;reserved=0 > >>> but there was request from community to wait another year to confirm > >> formally this api is mature. > >> > >> The request was to wait for a new vendor (i.e. non Intel) to > >> implement a bbdev driver for their hw. > >> > >> NXP proposed a driver for this class: > >> > https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fpat > >> ch > work.dpdk.org%2Fproject%2Fdpdk%2Flist%2F%3Fseries%3D16642%26state > >> > %3D%252A%26archive%3Dboth&data=04%7C01%7Chemant.agrawal% > >> > 40nxp.com%7Cfcdfb339ec284057db1508d96c9dd048%7C686ea1d3bc2b4c6f > >> a92cd99c5c301635%7C0%7C0%7C637660247324212288%7CUnknown%7CT > >> > WFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLC > >> > JXVCI6Mn0%3D%7C1000&sdata=iorsQ7wAXyU2%2FaqeV5CrDp9asl8ztZ > >> nSrbDuJpCv9Gk%3D&reserved=0 > >> > >> What is the status of this driver? > > > > [Hemant] We are in process of resending the patches in next few days. > > Ok - just so we are clear - the net-net of that, is that the bbdev APIs will > be > 'experimental' > for another year, right? There is little reason to defer further I believe. Henant please confirm you are not looking at any ABI change based on discussion for the last few months. > > > > >> I see no update on the bbdev API, which seems good, but a > >> confirmation is needed. Indeed that's all we need. Confirmation no change is require for new PMD. > >> > >> > >> -- > >> David Marchand > >
Re: [dpdk-dev] [PATCH v4] ethdev: fix representor port ID search by name
On 8/31/2021 5:06 PM, Andrew Rybchenko wrote:
> From: Viacheslav Galaktionov
>
> Getting a list of representors from a representor does not make sense.
> Instead, a parent device should be used.
>
Which code is getting list of the representors?
As far as I can see impacted APIs are:
'rte_eth_representor_id_get()'
'rte_eth_representor_info_get()'
Which are now getting 'parent_port_id' as argument, instead of representro port
id.
'rte_eth_representor_info_get()' is using 'representor_info_get()' dev_ops,
which is only implemented by 'mlx5', so is this problem only valid for 'mlx5'
and can it be solved within PMD implementation?
> To this end, extend the rte_eth_dev_data structure to include the port ID
> of the backing device for representors.
>
> Signed-off-by: Viacheslav Galaktionov
> Signed-off-by: Andrew Rybchenko
> ---
> The new field is added into the hole in rte_eth_dev_data structure.
> The patch does not change ABI, but extra care is required since ABI
> check is disabled for the structure because of the libabigail bug [1].
>
> Potentially it is bad for out-of-tree drivers which implement
> representors but do not fill in a new parert_port_id field in
> rte_eth_dev_data structure. Do we care?
>
> mlx5 changes should be reviwed by maintainers very carefully, since
> we are not sure if we patch it correctly.
>
> [1] https://sourceware.org/bugzilla/show_bug.cgi?id=28060
>
> v4:
> - apply mlx5 review notes: remove fallback from generic ethdev
> code and add fallback to mlx5 code to handle legacy usecase
>
> v3:
> - fix mlx5 build breakage
>
> v2:
> - fix mlx5 review notes
> - try device port ID first before parent in order to address
> backward compatibility issue
>
<...>
> index edf96de2dc..72fefa59c2 100644
> --- a/lib/ethdev/rte_ethdev_core.h
> +++ b/lib/ethdev/rte_ethdev_core.h
> @@ -185,6 +185,12 @@ struct rte_eth_dev_data {
> /**< Switch-specific identifier.
>* Valid if RTE_ETH_DEV_REPRESENTOR in dev_flags.
>*/
> + uint16_t parent_port_id;
Why 'parent'? Isn't this for the port that port representator represents, does
it called 'parent'? Above that device mentioned as 'backing device' a few times,
so would something like 'peer_port_id' better?
> + /**< Port ID of the backing device.
> + * This device will be used to query representor
> + * info and calculate representor IDs.
> + * Valid if RTE_ETH_DEV_REPRESENTOR in dev_flags.
> + */
>
Overall I am not feeling good about adding representor port related information
withing the device data struct.
I wonder if this information can be kept in the device private data.
Or, it is hard to explain but can we use something like inheritance, a
representor specific dev_data derived from original dev_data. We can store
dev_data pointers in 'struct rte_eth_dev' but can cast it to representor
specific dev_data when type is representor.
struct rte_eth_dev_data_rep
struct rte_eth_dev_data
This brings lots of complexity though, specially in allocating/freeing this
struct, not sure if it worth to the effort.
Re: [dpdk-dev] [PATCH v2 1/6] bbdev: add capability for CRC16 check
> -Original Message-
> From: Tom Rix
> Sent: Wednesday, September 1, 2021 6:37 AM
> To: Chautru, Nicolas ; dev@dpdk.org;
> gak...@marvell.com
> Cc: tho...@monjalon.net; hemant.agra...@nxp.com; Zhang, Mingshan
> ; Joshi, Arun
> Subject: Re: [PATCH v2 1/6] bbdev: add capability for CRC16 check
>
>
> On 8/19/21 2:10 PM, Nicolas Chautru wrote:
> > Adding a missing operation when CRC16
> > is being used for TB CRC check.
> >
> > Signed-off-by: Nicolas Chautru
> > ---
> > app/test-bbdev/test_bbdev_vector.c | 2 ++
> > doc/guides/prog_guide/bbdev.rst| 3 +++
> > doc/guides/rel_notes/release_21_11.rst | 1 +
> > lib/bbdev/rte_bbdev_op.h | 34
> > ++--
> --
> > 4 files changed, 24 insertions(+), 16 deletions(-)
> >
> > diff --git a/app/test-bbdev/test_bbdev_vector.c
> > b/app/test-bbdev/test_bbdev_vector.c
> > index 614dbd1..8d796b1 100644
> > --- a/app/test-bbdev/test_bbdev_vector.c
> > +++ b/app/test-bbdev/test_bbdev_vector.c
> > @@ -167,6 +167,8 @@
> > *op_flag_value = RTE_BBDEV_LDPC_CRC_TYPE_24B_CHECK;
> > else if (!strcmp(token, "RTE_BBDEV_LDPC_CRC_TYPE_24B_DROP"))
> > *op_flag_value = RTE_BBDEV_LDPC_CRC_TYPE_24B_DROP;
> > + else if (!strcmp(token, "RTE_BBDEV_LDPC_CRC_TYPE_16_CHECK"))
> > + *op_flag_value = RTE_BBDEV_LDPC_CRC_TYPE_16_CHECK;
> > else if (!strcmp(token,
> "RTE_BBDEV_LDPC_DEINTERLEAVER_BYPASS"))
> > *op_flag_value =
> RTE_BBDEV_LDPC_DEINTERLEAVER_BYPASS;
> > else if (!strcmp(token,
> "RTE_BBDEV_LDPC_HQ_COMBINE_IN_ENABLE"))
> > diff --git a/doc/guides/prog_guide/bbdev.rst
> > b/doc/guides/prog_guide/bbdev.rst index 9619280..8bd7cba 100644
> > --- a/doc/guides/prog_guide/bbdev.rst
> > +++ b/doc/guides/prog_guide/bbdev.rst
> > @@ -891,6 +891,9 @@ given below.
> > |RTE_BBDEV_LDPC_CRC_TYPE_24B_DROP|
> > | Set to drop the last CRC bits decoding output |
> >
> > ++
> > +|RTE_BBDEV_LDPC_CRC_TYPE_16_CHECK|
> > +| Set for code block CRC-16 checking |
> > +++
> > |RTE_BBDEV_LDPC_DEINTERLEAVER_BYPASS |
> > | Set for bit-level de-interleaver bypass on input stream|
> >
> > ++
> > diff --git a/doc/guides/rel_notes/release_21_11.rst
> > b/doc/guides/rel_notes/release_21_11.rst
> > index d707a55..69dd518 100644
> > --- a/doc/guides/rel_notes/release_21_11.rst
> > +++ b/doc/guides/rel_notes/release_21_11.rst
> > @@ -84,6 +84,7 @@ API Changes
> > Also, make sure to start the actual text at the margin.
> > ===
> >
> > +* bbdev: Added capability related to more comprehensive CRC options.
> >
> > ABI Changes
> > ---
> > diff --git a/lib/bbdev/rte_bbdev_op.h b/lib/bbdev/rte_bbdev_op.h index
> > f946842..7c44ddd 100644
> > --- a/lib/bbdev/rte_bbdev_op.h
> > +++ b/lib/bbdev/rte_bbdev_op.h
> > @@ -142,51 +142,53 @@ enum rte_bbdev_op_ldpcdec_flag_bitmasks {
> > RTE_BBDEV_LDPC_CRC_TYPE_24B_CHECK = (1ULL << 1),
> > /** Set to drop the last CRC bits decoding output */
> > RTE_BBDEV_LDPC_CRC_TYPE_24B_DROP = (1ULL << 2),
> > + /** Set for transport block CRC-16 checking */
> > + RTE_BBDEV_LDPC_CRC_TYPE_16_CHECK = (1ULL << 3),
>
> Changing these enums will break the abi backwards.
>
> Why not add the new one at the end ?
To keep all the CRC related flags next to each other for better readability and
logical clarity. The ABI is still marked as experimental.
>
> Tom
>
> > /** Set for bit-level de-interleaver bypass on Rx stream. */
> > - RTE_BBDEV_LDPC_DEINTERLEAVER_BYPASS = (1ULL << 3),
> > + RTE_BBDEV_LDPC_DEINTERLEAVER_BYPASS = (1ULL << 4),
> > /** Set for HARQ combined input stream enable. */
> > - RTE_BBDEV_LDPC_HQ_COMBINE_IN_ENABLE = (1ULL << 4),
> > + RTE_BBDEV_LDPC_HQ_COMBINE_IN_ENABLE = (1ULL << 5),
> > /** Set for HARQ combined output stream enable. */
> > - RTE_BBDEV_LDPC_HQ_COMBINE_OUT_ENABLE = (1ULL << 5),
> > + RTE_BBDEV_LDPC_HQ_COMBINE_OUT_ENABLE = (1ULL << 6),
> > /** Set for LDPC decoder bypass.
> > * RTE_BBDEV_LDPC_HQ_COMBINE_OUT_ENABLE must be set.
> > */
> > - RTE_BBDEV_LDPC_DECODE_BYPASS = (1ULL << 6),
> > + RTE_BBDEV_LDPC_DECODE_BYPASS = (1ULL << 7),
> > /** Set for soft-output stream enable */
> > - RTE_BBDEV_LDPC_SOFT_OUT_ENABLE = (1ULL << 7),
> > + RTE_BBDEV_LDPC_SOFT_OUT_ENABLE = (1ULL << 8),
> > /** Set for Rate-Matching bypass on soft-out stream. */
> > - RTE_BBDEV_LDPC_SOFT_OUT_RM_BYPASS = (1ULL << 8),
> > + RTE_BBDEV_LDPC_SOFT_OUT_RM_BYPASS = (1ULL << 9),
> > /** Set for bit-level de-interleaver bypass on soft-output stream. */
> >
[dpdk-dev] [PATCH] doc: clarify implicit filtering in transfer rules
As per existing documentation, attribute "transfer", quote, "complements the behavior of some pattern items such as RTE_FLOW_ITEM_TYPE_PHY_PORT and is meaningless without them". That effectively confronts the idea of implicit filtering imposed by port_id argument passed by flow create API. This bit of documentation is vague, and having no implicit filtering is unfriendly to applications which insert flow rules on specific ports based on the source port IDs of the (not offloaded) incoming packets. In order to address the problem, document the existence of the implicit filtering. Use the term "weak" for this filtering as it implies the possibility to override it by including explicit traffic source criteria in the flow pattern (PORT_ID, PHY_PORT and the likes). Signed-off-by: Andrew Rybchenko --- The topic was briefly discussed in mail thread [1]. I'm not sure if the patch should have "Fixes:" tag. If it is really behaviour intended from the very beginning, it should be backported and corresponding fixes in drivers should be backported as well. [1] https://patches.dpdk.org/project/dpdk/patch/2021060420.5549-1-ivan.ma...@oktetlabs.ru/ doc/guides/prog_guide/rte_flow.rst | 17 ++--- doc/guides/rel_notes/deprecation.rst | 5 - 2 files changed, 14 insertions(+), 8 deletions(-) diff --git a/doc/guides/prog_guide/rte_flow.rst b/doc/guides/prog_guide/rte_flow.rst index 2b42d5ec8c..af54939418 100644 --- a/doc/guides/prog_guide/rte_flow.rst +++ b/doc/guides/prog_guide/rte_flow.rst @@ -171,13 +171,24 @@ When supported, this effectively enables an application to reroute traffic not necessarily intended for it (e.g. coming from or addressed to different physical ports, VFs or applications) at the device level. -It complements the behavior of some pattern items such as `Item: PHY_PORT`_ -and is meaningless without them. - When transferring flow rules, **ingress** and **egress** attributes (`Attribute: Traffic direction`_) keep their original meaning, as if processing traffic emitted or received by the application. +DPDK port used to create transfer rule is important since it implicitly adds +filtering by it (similar to `Item: PORT_ID` with ``spec.id`` equal to +the port ID and exact match mask) if no other items which specify source +are present in the rule pattern (e.g. `Item: PHY_PORT`, `Item: VF` or +explicit `Item: PORT_ID`). It means that by default ingress rules apply to +traffic which comes from associated upstream switch port, i.e. physical +network port for PF DPDK port, VF for VF representor. Egress rules +transfer traffic transmitted via corresponding DPDK port, i.e. PF DPDK +port or VF representor itself. + +It is still possible to apply transfer rule on a traffic originating from +any switch port using wildcard mask in corresponding pattern item if +underlying PMD supports it. + Pattern item diff --git a/doc/guides/rel_notes/deprecation.rst b/doc/guides/rel_notes/deprecation.rst index 76a4abfd6b..f1d290a911 100644 --- a/doc/guides/rel_notes/deprecation.rst +++ b/doc/guides/rel_notes/deprecation.rst @@ -134,11 +134,6 @@ Deprecation Notices traffic direction to the represented entity or ethdev port itself in DPDK 21.11. -* ethdev: Flow API documentation is unclear if ethdev port used to create - a flow rule adds any implicit match criteria in the case of transfer rules. - The semantics will be clarified in DPDK 21.11 and it will require fixes in - drivers and applications which interpret it in a different way. - * ethdev: The flow API matching pattern structures, ``struct rte_flow_item_*``, should start with relevant protocol header. Some matching pattern structures implements this by duplicating protocol header -- 2.30.2
Re: [dpdk-dev] [PATCH] doc: announce changes in EFD function
> -Original Message- > From: dev On Behalf Of Pablo de Lara > Sent: Tuesday, August 31, 2021 3:31 PM > To: Marohn, Byron ; Wang, Yipeng1 > > Cc: dev@dpdk.org; De Lara Guarch, Pablo > Subject: [dpdk-dev] [PATCH] doc: announce changes in EFD function This change is required to fix Coverity defect 366390. https://scan4.coverity.com/reports.htm#v34998/p10075/fileInstanceId=107231002&defectInstanceId=14260754&mergedDefectId=366390 Acked-by: John McNamara
Re: [dpdk-dev] Running DPDK application with non-previlege mode
Thanks Bruce. BTW we are trying to run our application as docker container. We may need to pass the parameters(system capability) along with docker command line to mount the hugepages ? Like this : https://github.com/usnistgov/ndn-dpdk/blob/main/docs/Docker.md Please share your thoughts. Thanks, Kamaraj On Wed, Sep 1, 2021 at 1:14 AM Bruce Richardson wrote: > On Tue, Aug 31, 2021 at 09:50:43PM +0530, Kamaraj P wrote: > > Thanks David. Will check it out. > > > > BTW are there any steps where DPDK testpmd can run as a non privileged > user > > mode ( with any driver )? > > Please share if you have any pointers. > > > > Thanks, > > Kamaraj > > > Not sure if this is generally applicable but I ran a test using testpmd as > a non-root user on my Ubuntu 21.04 system. > > I bound a couple of ports using the i40e driver to DPDK using vfio-pci > kernel driver, and then did a chmod on the /dev/vfio/ nodes for them to > give the user access. I also gave the user rw access to the hugepage mount > point I had created. Thereafter testpmd ran fine as a non-root user and I > was able to put traffic through it. > > Regards, > /Bruce >
[dpdk-dev] [PATCH v2 0/6] add test suite for DMA drivers
This patchset adds a fairly comprehensive set of tests for basic dmadev functionality. Tests are added to verify basic copy operation in each device, using both submit function and submit flag, and verifying completion gathering using both "completed()" and "completed_status()" functions. Beyond that, tests are then added for the error reporting and handling, as is a suite of tests for the fill() operation for devices that support those. NOTE: This patchset depends on series v17 of the dmadev set [1]. [1] http://patches.dpdk.org/project/dpdk/list/?series=18502 V2: * added into dmadev a API to check for a device being idle * removed the hard-coded timeout delays before checking completions, and instead wait for device to be idle * added in checks for statistics updates as part of some tests * fixed issue identified by internal coverity scan * other minor miscellaneous changes and fixes. Bruce Richardson (5): dmadev: add device idle check for testing use app/test: add basic dmadev instance tests app/test: add basic dmadev copy tests app/test: add more comprehensive dmadev copy tests app/test: test dmadev instance failure handling Kevin Laatz (1): app/test: add dmadev fill tests app/test/test_dmadev.c | 893 +++ lib/dmadev/rte_dmadev.c | 16 + lib/dmadev/rte_dmadev.h | 21 + lib/dmadev/rte_dmadev_core.h | 4 + lib/dmadev/version.map | 1 + 5 files changed, 935 insertions(+) -- 2.30.2
[dpdk-dev] [PATCH v2 1/6] dmadev: add device idle check for testing use
Add in a function to check if a device or vchan has completed all jobs
assigned to it, without gathering in the results. This is primarily for
use in testing, to allow the hardware to be in a known-state prior to
gathering completions.
Signed-off-by: Bruce Richardson
---
lib/dmadev/rte_dmadev.c | 16
lib/dmadev/rte_dmadev.h | 21 +
lib/dmadev/rte_dmadev_core.h | 4
lib/dmadev/version.map | 1 +
4 files changed, 42 insertions(+)
diff --git a/lib/dmadev/rte_dmadev.c b/lib/dmadev/rte_dmadev.c
index 1c946402db..e249411631 100644
--- a/lib/dmadev/rte_dmadev.c
+++ b/lib/dmadev/rte_dmadev.c
@@ -555,3 +555,19 @@ rte_dmadev_dump(uint16_t dev_id, FILE *f)
return 0;
}
+
+int
+rte_dmadev_vchan_idle(uint16_t dev_id, uint16_t vchan)
+{
+ struct rte_dmadev *dev = &rte_dmadevices[dev_id];
+
+ RTE_DMADEV_VALID_DEV_ID_OR_ERR_RET(dev_id, -EINVAL);
+ if (vchan >= dev->data->dev_conf.nb_vchans) {
+ RTE_DMADEV_LOG(ERR,
+ "Device %u vchan %u out of range\n", dev_id, vchan);
+ return -EINVAL;
+ }
+
+ RTE_FUNC_PTR_OR_ERR_RET(*dev->dev_ops->vchan_idle, -ENOTSUP);
+ return (*dev->dev_ops->vchan_idle)(dev, vchan);
+}
diff --git a/lib/dmadev/rte_dmadev.h b/lib/dmadev/rte_dmadev.h
index e8f58e9213..350e7defc8 100644
--- a/lib/dmadev/rte_dmadev.h
+++ b/lib/dmadev/rte_dmadev.h
@@ -1028,6 +1028,27 @@ rte_dmadev_completed_status(uint16_t dev_id, uint16_t
vchan,
return (*dev->completed_status)(dev, vchan, nb_cpls, last_idx, status);
}
+/**
+ * @warning
+ * @b EXPERIMENTAL: this API may change without prior notice.
+ *
+ * Determine if all jobs have completed on a device channel.
+ * This function is primarily designed for testing use, as it allows a process
to check if
+ * all jobs are completed, without actually gathering completions from those
jobs.
+ *
+ * @param dev_id
+ * The identifier of the device.
+ * @param vchan
+ * The identifier of virtual DMA channel.
+ * @return
+ * 1 - if all jobs have completed and the device vchan is idle
+ * 0 - if there are still outstanding jobs yet to complete
+ * < 0 - error code indicating there was a problem calling the API
+ */
+__rte_experimental
+int
+rte_dmadev_vchan_idle(uint16_t dev_id, uint16_t vchan);
+
#ifdef __cplusplus
}
#endif
diff --git a/lib/dmadev/rte_dmadev_core.h b/lib/dmadev/rte_dmadev_core.h
index e94aa1c457..7ec5a5b572 100644
--- a/lib/dmadev/rte_dmadev_core.h
+++ b/lib/dmadev/rte_dmadev_core.h
@@ -53,6 +53,9 @@ typedef int (*rte_dmadev_stats_reset_t)(struct rte_dmadev
*dev, uint16_t vchan);
typedef int (*rte_dmadev_dump_t)(const struct rte_dmadev *dev, FILE *f);
/**< @internal Used to dump internal information. */
+typedef int (*rte_dmadev_vchan_idle_t)(const struct rte_dmadev *dev, uint16_t
vchan);
+/**< @internal Used to check if a virtual channel has finished all jobs. */
+
typedef int (*rte_dmadev_copy_t)(struct rte_dmadev *dev, uint16_t vchan,
rte_iova_t src, rte_iova_t dst,
uint32_t length, uint64_t flags);
@@ -106,6 +109,7 @@ struct rte_dmadev_ops {
rte_dmadev_stats_get_t stats_get;
rte_dmadev_stats_reset_t stats_reset;
rte_dmadev_dump_t dev_dump;
+ rte_dmadev_vchan_idle_t vchan_idle;
};
/**
diff --git a/lib/dmadev/version.map b/lib/dmadev/version.map
index 80be592713..b7e52fda3d 100644
--- a/lib/dmadev/version.map
+++ b/lib/dmadev/version.map
@@ -18,6 +18,7 @@ EXPERIMENTAL {
rte_dmadev_stats_reset;
rte_dmadev_stop;
rte_dmadev_submit;
+ rte_dmadev_vchan_idle;
rte_dmadev_vchan_setup;
local: *;
--
2.30.2
[dpdk-dev] [PATCH v2 2/6] app/test: add basic dmadev instance tests
Run basic sanity tests for configuring, starting and stopping a dmadev
instance to help validate drivers. This also provides the framework for
future tests for data-path operation.
Signed-off-by: Bruce Richardson
---
app/test/test_dmadev.c | 81 ++
1 file changed, 81 insertions(+)
diff --git a/app/test/test_dmadev.c b/app/test/test_dmadev.c
index bb01e86483..12f7c69629 100644
--- a/app/test/test_dmadev.c
+++ b/app/test/test_dmadev.c
@@ -2,6 +2,7 @@
* Copyright(c) 2021 HiSilicon Limited.
* Copyright(c) 2021 Intel Corporation.
*/
+#include
#include
#include
@@ -13,6 +14,77 @@
/* from test_dmadev_api.c */
extern int test_dmadev_api(uint16_t dev_id);
+#define PRINT_ERR(...) print_err(__func__, __LINE__, __VA_ARGS__)
+
+static inline int
+__rte_format_printf(3, 4)
+print_err(const char *func, int lineno, const char *format, ...)
+{
+ va_list ap;
+ int ret;
+
+ ret = fprintf(stderr, "In %s:%d - ", func, lineno);
+ va_start(ap, format);
+ ret += vfprintf(stderr, format, ap);
+ va_end(ap);
+
+ return ret;
+}
+
+static int
+test_dmadev_instance(uint16_t dev_id)
+{
+#define TEST_RINGSIZE 512
+ struct rte_dmadev_stats stats;
+ struct rte_dmadev_info info;
+ const struct rte_dmadev_conf conf = { .nb_vchans = 1};
+ const struct rte_dmadev_vchan_conf qconf = {
+ .direction = RTE_DMA_DIR_MEM_TO_MEM,
+ .nb_desc = TEST_RINGSIZE,
+ };
+ const int vchan = 0;
+
+ printf("\n### Test dmadev instance %u\n", dev_id);
+
+ rte_dmadev_info_get(dev_id, &info);
+ if (info.max_vchans < 1) {
+ PRINT_ERR("Error, no channels available on device id %u\n",
dev_id);
+ return -1;
+ }
+ if (rte_dmadev_configure(dev_id, &conf) != 0) {
+ PRINT_ERR("Error with rte_dmadev_configure()\n");
+ return -1;
+ }
+ if (rte_dmadev_vchan_setup(dev_id, vchan, &qconf) < 0) {
+ PRINT_ERR("Error with queue configuration\n");
+ return -1;
+ }
+
+ rte_dmadev_info_get(dev_id, &info);
+ if (info.nb_vchans != 1) {
+ PRINT_ERR("Error, no configured queues reported on device id
%u\n", dev_id);
+ return -1;
+ }
+
+ if (rte_dmadev_start(dev_id) != 0) {
+ PRINT_ERR("Error with rte_dmadev_start()\n");
+ return -1;
+ }
+ if (rte_dmadev_stats_get(dev_id, vchan, &stats) != 0) {
+ PRINT_ERR("Error with rte_dmadev_stats_get()\n");
+ return -1;
+ }
+ if (stats.completed != 0 || stats.submitted != 0 || stats.errors != 0) {
+ PRINT_ERR("Error device stats are not all zero: completed =
%"PRIu64", submitted = %"PRIu64", errors = %"PRIu64"\n",
+ stats.completed, stats.submitted, stats.errors);
+ return -1;
+ }
+
+ rte_dmadev_stop(dev_id);
+ rte_dmadev_stats_reset(dev_id, vchan);
+ return 0;
+}
+
static int
test_apis(void)
{
@@ -35,10 +107,19 @@ test_apis(void)
static int
test_dmadev(void)
{
+ int i;
+
/* basic sanity on dmadev infrastructure */
if (test_apis() < 0)
return -1;
+ if (rte_dmadev_count() == 0)
+ return TEST_SKIPPED;
+
+ for (i = 0; i < RTE_DMADEV_MAX_DEVS; i++)
+ if (rte_dmadevices[i].state == RTE_DMADEV_ATTACHED &&
test_dmadev_instance(i) < 0)
+ return -1;
+
return 0;
}
--
2.30.2
[dpdk-dev] [PATCH v2 3/6] app/test: add basic dmadev copy tests
For each dmadev instance, perform some basic copy tests to validate that
functionality.
Signed-off-by: Bruce Richardson
---
app/test/test_dmadev.c | 174 +
1 file changed, 174 insertions(+)
diff --git a/app/test/test_dmadev.c b/app/test/test_dmadev.c
index 12f7c69629..261f45db71 100644
--- a/app/test/test_dmadev.c
+++ b/app/test/test_dmadev.c
@@ -2,12 +2,15 @@
* Copyright(c) 2021 HiSilicon Limited.
* Copyright(c) 2021 Intel Corporation.
*/
+#include
#include
#include
#include
#include
#include
+#include
+#include
#include "test.h"
@@ -16,6 +19,11 @@ extern int test_dmadev_api(uint16_t dev_id);
#define PRINT_ERR(...) print_err(__func__, __LINE__, __VA_ARGS__)
+#define COPY_LEN 1024
+
+static struct rte_mempool *pool;
+static uint16_t id_count;
+
static inline int
__rte_format_printf(3, 4)
print_err(const char *func, int lineno, const char *format, ...)
@@ -31,6 +39,134 @@ print_err(const char *func, int lineno, const char *format,
...)
return ret;
}
+static inline void
+await_hw(int dev_id, uint16_t vchan)
+{
+ int idle = rte_dmadev_vchan_idle(dev_id, vchan);
+ if (idle < 0) {
+ /* for drivers that don't support this op, just sleep for 25
microseconds */
+ usleep(25);
+ return;
+ }
+
+ /* for those that do, *max* end time is one second from now, but all
should be faster */
+ const uint64_t end_cycles = rte_get_timer_cycles() + rte_get_timer_hz();
+ while (!idle && rte_get_timer_cycles() < end_cycles) {
+ rte_pause();
+ idle = rte_dmadev_vchan_idle(dev_id, vchan);
+ }
+}
+
+static int
+test_enqueue_copies(int dev_id, uint16_t vchan)
+{
+ unsigned int i;
+ uint16_t id;
+
+ /* test doing a single copy */
+ do {
+ struct rte_mbuf *src, *dst;
+ char *src_data, *dst_data;
+
+ src = rte_pktmbuf_alloc(pool);
+ dst = rte_pktmbuf_alloc(pool);
+ src_data = rte_pktmbuf_mtod(src, char *);
+ dst_data = rte_pktmbuf_mtod(dst, char *);
+
+ for (i = 0; i < COPY_LEN; i++)
+ src_data[i] = rte_rand() & 0xFF;
+
+ id = rte_dmadev_copy(dev_id, vchan, src->buf_iova +
src->data_off,
+ dst->buf_iova + dst->data_off, COPY_LEN,
RTE_DMA_OP_FLAG_SUBMIT);
+ if (id != id_count) {
+ PRINT_ERR("Error with rte_dmadev_copy, got %u, expected
%u\n",
+ id, id_count);
+ return -1;
+ }
+
+ /* give time for copy to finish, then check it was done */
+ await_hw(dev_id, vchan);
+
+ for (i = 0; i < COPY_LEN; i++) {
+ if (dst_data[i] != src_data[i]) {
+ PRINT_ERR("Data mismatch at char %u [Got %02x
not %02x]\n", i,
+ dst_data[i], src_data[i]);
+ rte_dmadev_dump(dev_id, stderr);
+ return -1;
+ }
+ }
+
+ /* now check completion works */
+ if (rte_dmadev_completed(dev_id, vchan, 1, &id, NULL) != 1) {
+ PRINT_ERR("Error with rte_dmadev_completed\n");
+ return -1;
+ }
+ if (id != id_count) {
+ PRINT_ERR("Error:incorrect job id received, %u
[expected %u]\n",
+ id, id_count);
+ return -1;
+ }
+
+ rte_pktmbuf_free(src);
+ rte_pktmbuf_free(dst);
+
+ /* now check completion works */
+ if (rte_dmadev_completed(dev_id, 0, 1, NULL, NULL) != 0) {
+ PRINT_ERR("Error with rte_dmadev_completed in empty
check\n");
+ return -1;
+ }
+ id_count++;
+
+ } while (0);
+
+ /* test doing a multiple single copies */
+ do {
+ const uint16_t max_ops = 4;
+ struct rte_mbuf *src, *dst;
+ char *src_data, *dst_data;
+
+ src = rte_pktmbuf_alloc(pool);
+ dst = rte_pktmbuf_alloc(pool);
+ src_data = rte_pktmbuf_mtod(src, char *);
+ dst_data = rte_pktmbuf_mtod(dst, char *);
+
+ for (i = 0; i < COPY_LEN; i++)
+ src_data[i] = rte_rand() & 0xFF;
+
+ /* perform the same copy times */
+ for (i = 0; i < max_ops; i++) {
+ if (rte_dmadev_copy(dev_id, vchan,
+ src->buf_iova + src->data_off,
+ dst->buf_iova + dst->data_off,
+ COPY_LEN, RTE_DMA_
[dpdk-dev] [PATCH v2 4/6] app/test: add more comprehensive dmadev copy tests
Add unit tests for various combinations of use for dmadev, copying
bursts of packets in various formats, e.g.
1. enqueuing two smaller bursts and completing them as one burst
2. enqueuing one burst and gathering completions in smaller bursts
3. using completed_status() function to gather completions rather than
just completed()
Signed-off-by: Bruce Richardson
---
app/test/test_dmadev.c | 142 -
1 file changed, 140 insertions(+), 2 deletions(-)
diff --git a/app/test/test_dmadev.c b/app/test/test_dmadev.c
index 261f45db71..7a808a9cba 100644
--- a/app/test/test_dmadev.c
+++ b/app/test/test_dmadev.c
@@ -39,6 +39,20 @@ print_err(const char *func, int lineno, const char *format,
...)
return ret;
}
+static inline int
+check_stats(struct rte_dmadev_stats *stats, bool check_errors)
+{
+ if (stats->completed != stats->submitted) {
+ PRINT_ERR("Error, not all submitted jobs are reported as
completed\n");
+ return -1;
+ }
+ if (check_errors && stats->errors != 0) {
+ PRINT_ERR("Errors reported during copy processing, aborting
tests\n");
+ return -1;
+ }
+ return 0;
+}
+
static inline void
await_hw(int dev_id, uint16_t vchan)
{
@@ -57,6 +71,120 @@ await_hw(int dev_id, uint16_t vchan)
}
}
+/* run a series of copy tests just using some different options for enqueues
and completions */
+static int
+do_multi_copies(int dev_id, uint16_t vchan,
+ int split_batches, /* submit 2 x 16 or 1 x 32 burst */
+ int split_completions, /* gather 2 x 16 or 1 x 32 completions */
+ int use_completed_status) /* use completed or completed_status
function */
+{
+ struct rte_mbuf *srcs[32], *dsts[32];
+ enum rte_dma_status_code sc[32];
+ unsigned int i, j;
+ bool dma_err = false;
+
+ /* Enqueue burst of copies and hit doorbell */
+ for (i = 0; i < RTE_DIM(srcs); i++) {
+ uint64_t *src_data;
+
+ if (split_batches && i == RTE_DIM(srcs) / 2)
+ rte_dmadev_submit(dev_id, vchan);
+
+ srcs[i] = rte_pktmbuf_alloc(pool);
+ dsts[i] = rte_pktmbuf_alloc(pool);
+ if (srcs[i] == NULL || dsts[i] == NULL) {
+ PRINT_ERR("Error allocating buffers\n");
+ return -1;
+ }
+ src_data = rte_pktmbuf_mtod(srcs[i], uint64_t *);
+
+ for (j = 0; j < COPY_LEN/sizeof(uint64_t); j++)
+ src_data[j] = rte_rand();
+
+ if (rte_dmadev_copy(dev_id, vchan, srcs[i]->buf_iova +
srcs[i]->data_off,
+ dsts[i]->buf_iova + dsts[i]->data_off,
COPY_LEN, 0) != id_count++) {
+ PRINT_ERR("Error with rte_dmadev_copy for buffer %u\n",
i);
+ return -1;
+ }
+ }
+ rte_dmadev_submit(dev_id, vchan);
+
+ await_hw(dev_id, vchan);
+
+ if (split_completions) {
+ /* gather completions in two halves */
+ uint16_t half_len = RTE_DIM(srcs) / 2;
+ int ret = rte_dmadev_completed(dev_id, vchan, half_len, NULL,
&dma_err);
+ if (ret != half_len || dma_err) {
+ PRINT_ERR("Error with rte_dmadev_completed - first
half. ret = %d, expected ret = %u, dma_err = %d\n",
+ ret, half_len, dma_err);
+ rte_dmadev_dump(dev_id, stdout);
+ return -1;
+ }
+ ret = rte_dmadev_completed(dev_id, vchan, half_len, NULL,
&dma_err);
+ if (ret != half_len || dma_err) {
+ PRINT_ERR("Error with rte_dmadev_completed - second
half. ret = %d, expected ret = %u, dma_err = %d\n",
+ ret, half_len, dma_err);
+ rte_dmadev_dump(dev_id, stdout);
+ return -1;
+ }
+ } else {
+ /* gather all completions in one go, using either
+* completed or completed_status fns
+*/
+ if (!use_completed_status) {
+ int n = rte_dmadev_completed(dev_id, vchan,
RTE_DIM(srcs), NULL, &dma_err);
+ if (n != RTE_DIM(srcs) || dma_err) {
+ PRINT_ERR("Error with rte_dmadev_completed, %u
[expected: %zu], dma_err = %d\n",
+ n, RTE_DIM(srcs), dma_err);
+ rte_dmadev_dump(dev_id, stdout);
+ return -1;
+ }
+ } else {
+ int n = rte_dmadev_completed_status(dev_id, vchan,
RTE_DIM(srcs), NULL, sc);
+ if (n != RTE_DIM(srcs)) {
+ PRINT_ERR("Error with
rte_
[dpdk-dev] [PATCH v2 5/6] app/test: test dmadev instance failure handling
Add a series of tests to inject bad copy operations into a dmadev to
test the error handling and reporting capabilities. Various combinations
of errors in various positions in a burst are tested, as are errors in
bursts with fence flag set, and multiple errors in a single burst.
Signed-off-by: Bruce Richardson
---
app/test/test_dmadev.c | 427 +
1 file changed, 427 insertions(+)
diff --git a/app/test/test_dmadev.c b/app/test/test_dmadev.c
index 7a808a9cba..5d7b6ddd87 100644
--- a/app/test/test_dmadev.c
+++ b/app/test/test_dmadev.c
@@ -302,6 +302,414 @@ test_enqueue_copies(int dev_id, uint16_t vchan)
|| do_multi_copies(dev_id, vchan, 0, 0, 1);
}
+/* Failure handling test cases - global macros and variables for those tests*/
+#define COMP_BURST_SZ 16
+#define OPT_FENCE(idx) ((fence && idx == 8) ? RTE_DMA_OP_FLAG_FENCE : 0)
+
+static int
+test_failure_in_full_burst(int dev_id, uint16_t vchan, bool fence,
+ struct rte_mbuf **srcs, struct rte_mbuf **dsts, unsigned int
fail_idx)
+{
+ /* Test single full batch statuses with failures */
+ enum rte_dma_status_code status[COMP_BURST_SZ];
+ struct rte_dmadev_stats baseline, stats;
+ uint16_t invalid_addr_id = 0;
+ uint16_t idx;
+ uint16_t count, status_count;
+ unsigned int i;
+ bool error = 0;
+ int err_count = 0;
+
+ rte_dmadev_stats_get(dev_id, vchan, &baseline); /* get a baseline set
of stats */
+ for (i = 0; i < COMP_BURST_SZ; i++) {
+ int id = rte_dmadev_copy(dev_id, vchan,
+ (i == fail_idx ? 0 : (srcs[i]->buf_iova +
srcs[i]->data_off)),
+ dsts[i]->buf_iova + dsts[i]->data_off,
+ COPY_LEN, OPT_FENCE(i));
+ if (id < 0) {
+ PRINT_ERR("Error with rte_dmadev_copy for buffer %u\n",
i);
+ return -1;
+ }
+ if (i == fail_idx)
+ invalid_addr_id = id;
+ }
+ rte_dmadev_submit(dev_id, vchan);
+ rte_dmadev_stats_get(dev_id, vchan, &stats);
+ if (stats.submitted != baseline.submitted + COMP_BURST_SZ) {
+ PRINT_ERR("Submitted stats value not as expected, %"PRIu64" not
%"PRIu64"\n",
+ stats.submitted, baseline.submitted +
COMP_BURST_SZ);
+ return -1;
+ }
+
+ await_hw(dev_id, vchan);
+
+ count = rte_dmadev_completed(dev_id, vchan, COMP_BURST_SZ, &idx,
&error);
+ if (count != fail_idx) {
+ PRINT_ERR("Error with rte_dmadev_completed for failure test.
Got returned %u not %u.\n",
+ count, fail_idx);
+ rte_dmadev_dump(dev_id, stdout);
+ return -1;
+ }
+ if (error == false) {
+ PRINT_ERR("Error, missing expected failed copy, %u. has_error
is not set\n",
+ fail_idx);
+ return -1;
+ }
+ if (idx != invalid_addr_id - 1) {
+ PRINT_ERR("Error, missing expected failed copy, %u. Got last
idx %u, not %u\n",
+ fail_idx, idx, invalid_addr_id - 1);
+ return -1;
+ }
+
+ /* all checks ok, now verify calling completed() again always returns 0
*/
+ for (i = 0; i < 10; i++) {
+ if (rte_dmadev_completed(dev_id, vchan, COMP_BURST_SZ, &idx,
&error) != 0
+ || error == false || idx != (invalid_addr_id -
1)) {
+ PRINT_ERR("Error with follow-up completed calls for
fail idx %u\n",
+ fail_idx);
+ return -1;
+ }
+ }
+
+ status_count = rte_dmadev_completed_status(dev_id, vchan, COMP_BURST_SZ,
+ &idx, status);
+ /* some HW may stop on error and be restarted after getting error
status for single value
+* To handle this case, if we get just one error back, wait for more
completions and get
+* status for rest of the burst
+*/
+ if (status_count == 1) {
+ await_hw(dev_id, vchan);
+ status_count += rte_dmadev_completed_status(dev_id, vchan,
COMP_BURST_SZ - 1,
+ &idx, &status[1]);
+ }
+ /* check that at this point we have all status values */
+ if (status_count != COMP_BURST_SZ - count) {
+ PRINT_ERR("Error with completed_status calls for fail idx %u.
Got %u not %u\n",
+ fail_idx, status_count, COMP_BURST_SZ - count);
+ return -1;
+ }
+ /* now verify just one failure followed by multiple successful or
skipped entries */
+ if (status[0] == RTE_DMA_STATUS_SUCCESSFUL) {
+ PRINT_ERR("Error with status returned for fail idx %u. First
status
[dpdk-dev] [PATCH v2 6/6] app/test: add dmadev fill tests
From: Kevin Laatz
For dma devices which support the fill operation, run unit tests to
verify fill behaviour is correct.
Signed-off-by: Kevin Laatz
Signed-off-by: Bruce Richardson
---
app/test/test_dmadev.c | 73 ++
1 file changed, 73 insertions(+)
diff --git a/app/test/test_dmadev.c b/app/test/test_dmadev.c
index 5d7b6ddd87..c44c3ad9db 100644
--- a/app/test/test_dmadev.c
+++ b/app/test/test_dmadev.c
@@ -710,6 +710,62 @@ test_completion_status(int dev_id, uint16_t vchan, bool
fence)
return 0;
}
+static int
+test_enqueue_fill(int dev_id, uint16_t vchan)
+{
+ const unsigned int lengths[] = {8, 64, 1024, 50, 100, 89};
+ struct rte_mbuf *dst;
+ char *dst_data;
+ uint64_t pattern = 0xfedcba9876543210;
+ unsigned int i, j;
+
+ dst = rte_pktmbuf_alloc(pool);
+ if (dst == NULL) {
+ PRINT_ERR("Failed to allocate mbuf\n");
+ return -1;
+ }
+ dst_data = rte_pktmbuf_mtod(dst, char *);
+
+ for (i = 0; i < RTE_DIM(lengths); i++) {
+ /* reset dst_data */
+ memset(dst_data, 0, rte_pktmbuf_data_len(dst));
+
+ /* perform the fill operation */
+ int id = rte_dmadev_fill(dev_id, vchan, pattern,
+ rte_pktmbuf_iova(dst), lengths[i],
RTE_DMA_OP_FLAG_SUBMIT);
+ if (id < 0) {
+ PRINT_ERR("Error with rte_ioat_enqueue_fill\n");
+ return -1;
+ }
+ await_hw(dev_id, vchan);
+
+ if (rte_dmadev_completed(dev_id, vchan, 1, NULL, NULL) != 1) {
+ PRINT_ERR("Error: fill operation failed (length:
%u)\n", lengths[i]);
+ return -1;
+ }
+ /* check the data from the fill operation is correct */
+ for (j = 0; j < lengths[i]; j++) {
+ char pat_byte = ((char *)&pattern)[j % 8];
+ if (dst_data[j] != pat_byte) {
+ PRINT_ERR("Error with fill operation (lengths =
%u): got (%x), not (%x)\n",
+ lengths[i], dst_data[j],
pat_byte);
+ return -1;
+ }
+ }
+ /* check that the data after the fill operation was not written
to */
+ for (; j < rte_pktmbuf_data_len(dst); j++) {
+ if (dst_data[j] != 0) {
+ PRINT_ERR("Error, fill operation wrote too far
(lengths = %u): got (%x), not (%x)\n",
+ lengths[i], dst_data[j], 0);
+ return -1;
+ }
+ }
+ }
+
+ rte_pktmbuf_free(dst);
+ return 0;
+}
+
static int
test_dmadev_instance(uint16_t dev_id)
{
@@ -813,6 +869,23 @@ test_dmadev_instance(uint16_t dev_id)
goto err;
}
+ if ((info.dev_capa & RTE_DMADEV_CAPA_OPS_FILL) == 0)
+ printf("DMA Dev: %u, No device fill support - skipping fill
tests\n", dev_id);
+ else {
+ rte_dmadev_stats_reset(dev_id, vchan);
+ printf("DMA Dev: %u, Running Fill Tests\n", dev_id);
+
+ if (test_enqueue_fill(dev_id, vchan) != 0)
+ goto err;
+
+ rte_dmadev_stats_get(dev_id, 0, &stats);
+ printf("Ops submitted: %"PRIu64"\t", stats.submitted);
+ printf("Ops completed: %"PRIu64"\t", stats.completed);
+ printf("Errors: %"PRIu64"\n", stats.errors);
+ if (check_stats(&stats, true) < 0)
+ goto err;
+ }
+
rte_mempool_free(pool);
rte_dmadev_stop(dev_id);
rte_dmadev_stats_reset(dev_id, vchan);
--
2.30.2
Re: [dpdk-dev] [PATCH] net/af_packet: fix ignoring full ring on tx
On 8/20/2021 2:39 PM, Tudor Cornea wrote:
> The poll call can return POLLERR which is ignored, or it can return
> POLLOUT, even if there are no free frames in the mmap-ed area.
>
> We can account for both of these cases by re-checking if the next
> frame is empty before writing into it.
>
> We also now eliminate the timestamp status from the frame status.
>
Hi Tudor,
Would you mind separate timestamp status fix to its own patch? I think better to
fix 'ignoring full Tx ring' first, to not make it dependent to timestamp patch.
> Signed-off-by: Mihai Pogonaru
> Signed-off-by: Tudor Cornea
> ---
> drivers/net/af_packet/rte_eth_af_packet.c | 47
> +--
> 1 file changed, 45 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/net/af_packet/rte_eth_af_packet.c
> b/drivers/net/af_packet/rte_eth_af_packet.c
> index b73b211..3845df5 100644
> --- a/drivers/net/af_packet/rte_eth_af_packet.c
> +++ b/drivers/net/af_packet/rte_eth_af_packet.c
> @@ -167,6 +167,12 @@ eth_af_packet_rx(void *queue, struct rte_mbuf **bufs,
> uint16_t nb_pkts)
> return num_rx;
> }
>
> +static inline __u32 tx_ring_status_remove_ts(volatile __u32 *tp_status)
> +{
> + return *tp_status &
> + ~(TP_STATUS_TS_SOFTWARE | TP_STATUS_TS_RAW_HARDWARE);
I can see 'TP_STATUS_TS_SYS_HARDWARE' is deprecated, and I assume in the kernel
versions the bug exists, this flag is not set, but can you please confirm?
> +}
> +
> /*
> * Callback to handle sending packets through a real NIC.
> */
> @@ -211,9 +217,41 @@ eth_af_packet_tx(void *queue, struct rte_mbuf **bufs,
> uint16_t nb_pkts)
> }
> }
>
> + /*
> + * We must eliminate the timestamp status from the packet
> + * status. This should only matter if timestamping is enabled
> + * on the socket, but there is a BUG in the kernel which is
> + * fixed in newer releases.
> +
> + * For interfaces of type 'veth', the sent skb is forwarded
> + * to the peer and back into the network stack which timestamps
> + * it on the RX path if timestamping is enabled globally
> + * (which happens if any socket enables timestamping).
> +
> + * When the skb is destructed, tpacket_destruct_skb() is called
> + * and it calls __packet_set_timestamp() which doesn't check
> + * the flags on the socket and returns the timestamp if it is
> + * set in the skb (and for veth it is, as mentioned above).
> + */
> +
Can you give some more details on this bug, any link etc..
And does it only seen with veth, if so I wonder if we can ignore it, not sure
how common to use af_packet PMD over veth interface, do you have this usecase?
And if only specific kernel versions impacted from this bug, what do you think
about adding kernel version check to reduce the scope of the fix in af_packet
PMD.
> /* point at the next incoming frame */
> - if ((ppd->tp_status != TP_STATUS_AVAILABLE) &&
> - (poll(&pfd, 1, -1) < 0))
> + if ((tx_ring_status_remove_ts(&ppd->tp_status)
> + != TP_STATUS_AVAILABLE) && (poll(&pfd, 1, -1) < 0))
> + break;
> +
> + /*
> + * Poll can return POLLERR if the interface is down or POLLOUT,
> + * even if there are no extra buffers available.
> + * This happens, because packet_poll() calls datagram_poll()
> + * which checks the space left in the socket buffer and in the
> + * case of packet_mmap the default socket buffer length
> + * doesn't match the requested size for the tx_ring so there
> + * is always space left in socket buffer, which doesn't seem
> + * to be correlated to the requested size for the tx_ring
> + * in packet_mmap.
> + */
> + if (tx_ring_status_remove_ts(&ppd->tp_status)
> + != TP_STATUS_AVAILABLE)
> break;
In this case should we break or poll again?
If 'POLLERR' is received when interface is down, it makes sense to break. Do you
know if is there any other case that 'POLLERR' is returned?
And for 'POLLOUT', when exactly this event sent? If 'POLLOUT' received, can we
poll again to wait Tx ring available to send more packets?
>
> /* copy the tx frame data */
> @@ -242,6 +280,11 @@ eth_af_packet_tx(void *queue, struct rte_mbuf **bufs,
> uint16_t nb_pkts)
> rte_pktmbuf_free(mbuf);
> }
>
> + /*
> + * We might have to ignore a few more errnos here since the packets
> + * remain in the mmap-ed queue and will be sent later, presumably.
> + */
> +
Can you please describe above comment more? What errors ignored?
And won't all packets sent will below sendto() call?
> /* kick-off transm
[dpdk-dev] [PATCH 1/8] common/cnxk: use different macros for sdp and lbk max frames
From: Satha Rao
For SDP interface all platforms supports up to 65535 frame size.
Updated api with new check for SDP interface.
Signed-off-by: Satha Rao
---
drivers/common/cnxk/hw/nix.h | 1 +
drivers/common/cnxk/roc_nix.c | 5 -
2 files changed, 5 insertions(+), 1 deletion(-)
diff --git a/drivers/common/cnxk/hw/nix.h b/drivers/common/cnxk/hw/nix.h
index 6b86002ead..a0ffd25660 100644
--- a/drivers/common/cnxk/hw/nix.h
+++ b/drivers/common/cnxk/hw/nix.h
@@ -2102,6 +2102,7 @@ struct nix_lso_format {
#define NIX_CN9K_MAX_HW_FRS 9212UL
#define NIX_LBK_MAX_HW_FRS 65535UL
+#define NIX_SDP_MAX_HW_FRS 65535UL
#define NIX_RPM_MAX_HW_FRS 16380UL
#define NIX_MIN_HW_FRS 60UL
diff --git a/drivers/common/cnxk/roc_nix.c b/drivers/common/cnxk/roc_nix.c
index 23d508b941..d1e8c2d4af 100644
--- a/drivers/common/cnxk/roc_nix.c
+++ b/drivers/common/cnxk/roc_nix.c
@@ -113,10 +113,13 @@ roc_nix_max_pkt_len(struct roc_nix *roc_nix)
{
struct nix *nix = roc_nix_to_nix_priv(roc_nix);
+ if (roc_nix_is_sdp(roc_nix))
+ return NIX_SDP_MAX_HW_FRS;
+
if (roc_model_is_cn9k())
return NIX_CN9K_MAX_HW_FRS;
- if (nix->lbk_link || roc_nix_is_sdp(roc_nix))
+ if (nix->lbk_link)
return NIX_LBK_MAX_HW_FRS;
return NIX_RPM_MAX_HW_FRS;
--
2.25.1
[dpdk-dev] [PATCH 2/8] common/cnxk: flush smq
From: Satha Rao
Added new API to flush all SMQs related nix interface
Signed-off-by: Satha Rao
---
drivers/common/cnxk/hw/nix.h | 6
drivers/common/cnxk/roc_nix.h| 1 +
drivers/common/cnxk/roc_nix_tm_ops.c | 50
drivers/common/cnxk/version.map | 1 +
4 files changed, 58 insertions(+)
diff --git a/drivers/common/cnxk/hw/nix.h b/drivers/common/cnxk/hw/nix.h
index a0ffd25660..bc908c25b1 100644
--- a/drivers/common/cnxk/hw/nix.h
+++ b/drivers/common/cnxk/hw/nix.h
@@ -2189,4 +2189,10 @@ struct nix_lso_format {
#define NIX_LSO_FORMAT_IDX_TSOV4 0
#define NIX_LSO_FORMAT_IDX_TSOV6 1
+/* [CN10K, .) */
+#define NIX_SENDSTATALG_MASK 0x7
+#define NIX_SENDSTATALG_SEL_MASK 0x8
+#define NIX_SENDSTAT_IOFFSET_MASK 0xFFF
+#define NIX_SENDSTAT_OOFFSET_MASK 0xFFF
+
#endif /* __NIX_HW_H__ */
diff --git a/drivers/common/cnxk/roc_nix.h b/drivers/common/cnxk/roc_nix.h
index 822c1900e2..a4c522ec3e 100644
--- a/drivers/common/cnxk/roc_nix.h
+++ b/drivers/common/cnxk/roc_nix.h
@@ -464,6 +464,7 @@ int __roc_api roc_nix_tm_rsrc_count(struct roc_nix *roc_nix,
int __roc_api roc_nix_tm_node_name_get(struct roc_nix *roc_nix,
uint32_t node_id, char *buf,
size_t buflen);
+int __roc_api roc_nix_smq_flush(struct roc_nix *roc_nix);
/* MAC */
int __roc_api roc_nix_mac_rxtx_start_stop(struct roc_nix *roc_nix, bool start);
diff --git a/drivers/common/cnxk/roc_nix_tm_ops.c
b/drivers/common/cnxk/roc_nix_tm_ops.c
index ed244d4214..d9741f542f 100644
--- a/drivers/common/cnxk/roc_nix_tm_ops.c
+++ b/drivers/common/cnxk/roc_nix_tm_ops.c
@@ -310,6 +310,56 @@ roc_nix_tm_node_delete(struct roc_nix *roc_nix, uint32_t
node_id, bool free)
return nix_tm_node_delete(roc_nix, node_id, ROC_NIX_TM_USER, free);
}
+int
+roc_nix_smq_flush(struct roc_nix *roc_nix)
+{
+ struct nix *nix = roc_nix_to_nix_priv(roc_nix);
+ struct nix_tm_node_list *list;
+ enum roc_nix_tm_tree tree;
+ struct nix_tm_node *node;
+ int rc = 0;
+
+ if (!(nix->tm_flags & NIX_TM_HIERARCHY_ENA))
+ return 0;
+
+ tree = nix->tm_tree;
+ list = nix_tm_node_list(nix, tree);
+
+ /* XOFF & Flush all SMQ's. HRM mandates
+* all SQ's empty before SMQ flush is issued.
+*/
+ TAILQ_FOREACH(node, list, node) {
+ if (node->hw_lvl != NIX_TXSCH_LVL_SMQ)
+ continue;
+ if (!(node->flags & NIX_TM_NODE_HWRES))
+ continue;
+
+ rc = nix_tm_smq_xoff(nix, node, true);
+ if (rc) {
+ plt_err("Failed to enable smq %u, rc=%d", node->hw_id,
+ rc);
+ goto exit;
+ }
+ }
+
+ /* XON all SMQ's */
+ TAILQ_FOREACH(node, list, node) {
+ if (node->hw_lvl != NIX_TXSCH_LVL_SMQ)
+ continue;
+ if (!(node->flags & NIX_TM_NODE_HWRES))
+ continue;
+
+ rc = nix_tm_smq_xoff(nix, node, false);
+ if (rc) {
+ plt_err("Failed to enable smq %u, rc=%d", node->hw_id,
+ rc);
+ goto exit;
+ }
+ }
+exit:
+ return rc;
+}
+
int
roc_nix_tm_hierarchy_disable(struct roc_nix *roc_nix)
{
diff --git a/drivers/common/cnxk/version.map b/drivers/common/cnxk/version.map
index 2cbcc4b93a..6705d13edf 100644
--- a/drivers/common/cnxk/version.map
+++ b/drivers/common/cnxk/version.map
@@ -165,6 +165,7 @@ INTERNAL {
roc_nix_xstats_names_get;
roc_nix_switch_hdr_set;
roc_nix_eeprom_info_get;
+ roc_nix_smq_flush;
roc_nix_tm_dump;
roc_nix_tm_fini;
roc_nix_tm_free_resources;
--
2.25.1
[dpdk-dev] [PATCH 3/8] common/cnxk: increase sched weight and shaper burst limit
From: Nithin Dabilpuram
Increase sched weight and shaper burst limit for cn10k.
Signed-off-by: Nithin Dabilpuram
---
drivers/common/cnxk/hw/nix.h | 13 ++---
drivers/common/cnxk/roc_nix.h | 23 ++-
drivers/common/cnxk/roc_nix_priv.h | 11 ---
drivers/common/cnxk/roc_nix_tm.c | 2 +-
drivers/common/cnxk/roc_nix_tm_ops.c | 10 ---
drivers/common/cnxk/roc_nix_tm_utils.c | 40 +++---
6 files changed, 75 insertions(+), 24 deletions(-)
diff --git a/drivers/common/cnxk/hw/nix.h b/drivers/common/cnxk/hw/nix.h
index bc908c25b1..d2054385c2 100644
--- a/drivers/common/cnxk/hw/nix.h
+++ b/drivers/common/cnxk/hw/nix.h
@@ -2134,8 +2134,9 @@ struct nix_lso_format {
0)
/* NIX burst limits */
-#define NIX_TM_MAX_BURST_EXPONENT 0xf
-#define NIX_TM_MAX_BURST_MANTISSA 0xff
+#define NIX_TM_MAX_BURST_EXPONENT 0xful
+#define NIX_TM_MAX_BURST_MANTISSA 0x7ffful
+#define NIX_CN9K_TM_MAX_BURST_MANTISSA 0xfful
/* NIX burst calculation
* PIR_BURST = ((256 + NIX_*_PIR[BURST_MANTISSA])
@@ -2147,7 +2148,7 @@ struct nix_lso_format {
* / 256
*/
#define NIX_TM_SHAPER_BURST(exponent, mantissa)
\
- (((256 + (mantissa)) << ((exponent) + 1)) / 256)
+ (((256ul + (mantissa)) << ((exponent) + 1)) / 256ul)
/* Burst limit in Bytes */
#define NIX_TM_MIN_SHAPER_BURST NIX_TM_SHAPER_BURST(0, 0)
@@ -2156,13 +2157,17 @@ struct nix_lso_format {
NIX_TM_SHAPER_BURST(NIX_TM_MAX_BURST_EXPONENT, \
NIX_TM_MAX_BURST_MANTISSA)
+#define NIX_CN9K_TM_MAX_SHAPER_BURST
\
+ NIX_TM_SHAPER_BURST(NIX_TM_MAX_BURST_EXPONENT, \
+ NIX_CN9K_TM_MAX_BURST_MANTISSA)
+
/* Min is limited so that NIX_AF_SMQX_CFG[MINLEN]+ADJUST is not -ve */
#define NIX_TM_LENGTH_ADJUST_MIN ((int)-NIX_MIN_HW_FRS + 1)
#define NIX_TM_LENGTH_ADJUST_MAX 255
#define NIX_TM_TLX_SP_PRIO_MAX10
#define NIX_CN9K_TM_RR_QUANTUM_MAX (BIT_ULL(24) - 1)
-#define NIX_TM_RR_QUANTUM_MAX (BIT_ULL(14) - 1)
+#define NIX_TM_RR_WEIGHT_MAX (BIT_ULL(14) - 1)
/* [CN9K, CN10K) */
#define NIX_CN9K_TXSCH_LVL_SMQ_MAX 512
diff --git a/drivers/common/cnxk/roc_nix.h b/drivers/common/cnxk/roc_nix.h
index a4c522ec3e..d947fe0900 100644
--- a/drivers/common/cnxk/roc_nix.h
+++ b/drivers/common/cnxk/roc_nix.h
@@ -273,6 +273,28 @@ enum roc_nix_lso_tun_type {
ROC_NIX_LSO_TUN_MAX,
};
+/* Restrict CN9K sched weight to have a minimum quantum */
+#define ROC_NIX_CN9K_TM_RR_WEIGHT_MAX 255u
+
+/* NIX TM Inlines */
+static inline uint64_t
+roc_nix_tm_max_sched_wt_get(void)
+{
+ if (roc_model_is_cn9k())
+ return ROC_NIX_CN9K_TM_RR_WEIGHT_MAX;
+ else
+ return NIX_TM_RR_WEIGHT_MAX;
+}
+
+static inline uint64_t
+roc_nix_tm_max_shaper_burst_get(void)
+{
+ if (roc_model_is_cn9k())
+ return NIX_CN9K_TM_MAX_SHAPER_BURST;
+ else
+ return NIX_TM_MAX_SHAPER_BURST;
+}
+
/* Dev */
int __roc_api roc_nix_dev_init(struct roc_nix *roc_nix);
int __roc_api roc_nix_dev_fini(struct roc_nix *roc_nix);
@@ -320,7 +342,6 @@ int __roc_api roc_nix_register_cq_irqs(struct roc_nix
*roc_nix);
void __roc_api roc_nix_unregister_cq_irqs(struct roc_nix *roc_nix);
/* Traffic Management */
-#define ROC_NIX_TM_MAX_SCHED_WT ((uint8_t)~0)
#define ROC_NIX_TM_SHAPER_PROFILE_NONE UINT32_MAX
#define ROC_NIX_TM_NODE_ID_INVALID UINT32_MAX
diff --git a/drivers/common/cnxk/roc_nix_priv.h
b/drivers/common/cnxk/roc_nix_priv.h
index 9dc0c88a6f..cc8e822427 100644
--- a/drivers/common/cnxk/roc_nix_priv.h
+++ b/drivers/common/cnxk/roc_nix_priv.h
@@ -256,11 +256,14 @@ struct nix_tm_shaper_data {
static inline uint64_t
nix_tm_weight_to_rr_quantum(uint64_t weight)
{
- uint64_t max = (roc_model_is_cn9k() ? NIX_CN9K_TM_RR_QUANTUM_MAX :
- NIX_TM_RR_QUANTUM_MAX);
+ uint64_t max = NIX_CN9K_TM_RR_QUANTUM_MAX;
- weight &= (uint64_t)ROC_NIX_TM_MAX_SCHED_WT;
- return (weight * max) / ROC_NIX_TM_MAX_SCHED_WT;
+ /* From CN10K onwards, we only configure RR weight */
+ if (!roc_model_is_cn9k())
+ return weight;
+
+ weight &= (uint64_t)max;
+ return (weight * max) / ROC_NIX_CN9K_TM_RR_WEIGHT_MAX;
}
static inline bool
diff --git a/drivers/common/cnxk/roc_nix_tm.c b/drivers/common/cnxk/roc_nix_tm.c
index ad54e17a28..947320ae63 100644
--- a/drivers/common/cnxk/roc_nix_tm.c
+++ b/drivers/common/cnxk/roc_nix_tm.c
@@ -223,7 +223,7 @@ nix_tm_node_add(struct roc_nix *roc_nix, struct nix_tm_node
*node)
if (rc)
return rc;
- if (node->weight > ROC_NIX_TM_MAX_SCHED_WT)
+ if (node->weight > roc_nix_tm_max_sched_wt_get())
return NIX_ERR_TM_WEIGHT_EXCEED;
[dpdk-dev] [PATCH 4/8] common/cnxk: handle packet mode shaper limits
From: Satha Rao
Add new macros to reflect HW shaper PPS limits. New API to validate
input rates for packet mode. Increase adjust value to support lesser
PPS (<61).
Signed-off-by: Satha Rao
---
drivers/common/cnxk/hw/nix.h | 3 +
drivers/common/cnxk/roc_nix_priv.h | 1 +
drivers/common/cnxk/roc_nix_tm_ops.c | 76 ++
drivers/common/cnxk/roc_nix_tm_utils.c | 4 +-
4 files changed, 60 insertions(+), 24 deletions(-)
diff --git a/drivers/common/cnxk/hw/nix.h b/drivers/common/cnxk/hw/nix.h
index d2054385c2..6a0eb019ac 100644
--- a/drivers/common/cnxk/hw/nix.h
+++ b/drivers/common/cnxk/hw/nix.h
@@ -2133,6 +2133,9 @@ struct nix_lso_format {
NIX_TM_SHAPER_RATE(NIX_TM_MAX_RATE_EXPONENT, NIX_TM_MAX_RATE_MANTISSA, \
0)
+#define NIX_TM_MIN_SHAPER_PPS_RATE 25
+#define NIX_TM_MAX_SHAPER_PPS_RATE (100ul << 20)
+
/* NIX burst limits */
#define NIX_TM_MAX_BURST_EXPONENT 0xful
#define NIX_TM_MAX_BURST_MANTISSA 0x7ffful
diff --git a/drivers/common/cnxk/roc_nix_priv.h
b/drivers/common/cnxk/roc_nix_priv.h
index cc8e822427..3412bf25e5 100644
--- a/drivers/common/cnxk/roc_nix_priv.h
+++ b/drivers/common/cnxk/roc_nix_priv.h
@@ -90,6 +90,7 @@ struct nix_tm_shaper_profile {
struct nix_tm_tb commit;
struct nix_tm_tb peak;
int32_t pkt_len_adj;
+ int32_t pkt_mode_adj;
bool pkt_mode;
uint32_t id;
void (*free_fn)(void *profile);
diff --git a/drivers/common/cnxk/roc_nix_tm_ops.c
b/drivers/common/cnxk/roc_nix_tm_ops.c
index a313023be2..69d58376ec 100644
--- a/drivers/common/cnxk/roc_nix_tm_ops.c
+++ b/drivers/common/cnxk/roc_nix_tm_ops.c
@@ -77,6 +77,51 @@ roc_nix_tm_free_resources(struct roc_nix *roc_nix, bool
hw_only)
return nix_tm_free_resources(roc_nix, BIT(ROC_NIX_TM_USER), hw_only);
}
+static int
+nix_tm_adjust_shaper_pps_rate(struct nix_tm_shaper_profile *profile)
+{
+ uint64_t min_rate = profile->commit.rate;
+
+ if (!profile->pkt_mode)
+ return 0;
+
+ profile->pkt_mode_adj = 1;
+
+ if (profile->commit.rate &&
+ (profile->commit.rate < NIX_TM_MIN_SHAPER_PPS_RATE ||
+profile->commit.rate > NIX_TM_MAX_SHAPER_PPS_RATE))
+ return NIX_ERR_TM_INVALID_COMMIT_RATE;
+
+ if (profile->peak.rate &&
+ (profile->peak.rate < NIX_TM_MIN_SHAPER_PPS_RATE ||
+profile->peak.rate > NIX_TM_MAX_SHAPER_PPS_RATE))
+ return NIX_ERR_TM_INVALID_PEAK_RATE;
+
+ if (profile->peak.rate && min_rate > profile->peak.rate)
+ min_rate = profile->peak.rate;
+
+ /* Each packet accomulate single count, whereas HW
+* considers each unit as Byte, so we need convert
+* user pps to bps
+*/
+ profile->commit.rate = profile->commit.rate * 8;
+ profile->peak.rate = profile->peak.rate * 8;
+ min_rate = min_rate * 8;
+
+ if (min_rate && (min_rate < NIX_TM_MIN_SHAPER_RATE)) {
+ int adjust = NIX_TM_MIN_SHAPER_RATE / min_rate;
+
+ if (adjust > NIX_TM_LENGTH_ADJUST_MAX)
+ return NIX_ERR_TM_SHAPER_PKT_LEN_ADJUST;
+
+ profile->pkt_mode_adj += adjust;
+ profile->commit.rate += (adjust * profile->commit.rate);
+ profile->peak.rate += (adjust * profile->peak.rate);
+ }
+
+ return 0;
+}
+
static int
nix_tm_shaper_profile_add(struct roc_nix *roc_nix,
struct nix_tm_shaper_profile *profile, int skip_ins)
@@ -86,8 +131,13 @@ nix_tm_shaper_profile_add(struct roc_nix *roc_nix,
uint64_t min_burst, max_burst;
uint64_t peak_rate, peak_sz;
uint32_t id;
+ int rc;
id = profile->id;
+ rc = nix_tm_adjust_shaper_pps_rate(profile);
+ if (rc)
+ return rc;
+
commit_rate = profile->commit.rate;
commit_sz = profile->commit.size;
peak_rate = profile->peak.rate;
@@ -157,17 +207,8 @@ roc_nix_tm_shaper_profile_add(struct roc_nix *roc_nix,
profile->ref_cnt = 0;
profile->id = roc_profile->id;
- if (roc_profile->pkt_mode) {
- /* Each packet accomulate single count, whereas HW
-* considers each unit as Byte, so we need convert
-* user pps to bps
-*/
- profile->commit.rate = roc_profile->commit_rate * 8;
- profile->peak.rate = roc_profile->peak_rate * 8;
- } else {
- profile->commit.rate = roc_profile->commit_rate;
- profile->peak.rate = roc_profile->peak_rate;
- }
+ profile->commit.rate = roc_profile->commit_rate;
+ profile->peak.rate = roc_profile->peak_rate;
profile->commit.size = roc_profile->commit_sz;
profile->peak.size = roc_profile->peak_sz;
profile->pkt_len_adj = roc_profile->pkt_len_adj;
@@ -185,17 +226,8 @@ roc_nix_tm_shaper_profile_update(struct roc_nix
[dpdk-dev] [PATCH 5/8] common/cnxk: handler to get rte tm error type
From: Satha Rao
Different TM handlers returns various platform specific errors,
this patch introduces new API to convert these internal error
types to RTE_TM* error types.
Also updated error message API with missed TM error types.
Signed-off-by: Satha Rao
---
drivers/common/cnxk/cnxk_utils.c | 68
drivers/common/cnxk/cnxk_utils.h | 11 ++
drivers/common/cnxk/meson.build | 5 +++
drivers/common/cnxk/roc_utils.c | 6 +++
drivers/common/cnxk/version.map | 1 +
5 files changed, 91 insertions(+)
create mode 100644 drivers/common/cnxk/cnxk_utils.c
create mode 100644 drivers/common/cnxk/cnxk_utils.h
diff --git a/drivers/common/cnxk/cnxk_utils.c b/drivers/common/cnxk/cnxk_utils.c
new file mode 100644
index 00..4e56adc659
--- /dev/null
+++ b/drivers/common/cnxk/cnxk_utils.c
@@ -0,0 +1,68 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(C) 2021 Marvell.
+ */
+#include
+#include
+
+#include "roc_api.h"
+#include "roc_priv.h"
+
+#include "cnxk_utils.h"
+
+int
+roc_nix_tm_err_to_rte_err(int errorcode)
+{
+ int err_type;
+
+ switch (errorcode) {
+ case NIX_ERR_TM_SHAPER_PKT_LEN_ADJUST:
+ err_type = RTE_TM_ERROR_TYPE_SHAPER_PROFILE_PKT_ADJUST_LEN;
+ break;
+ case NIX_ERR_TM_INVALID_COMMIT_SZ:
+ err_type = RTE_TM_ERROR_TYPE_SHAPER_PROFILE_COMMITTED_SIZE;
+ break;
+ case NIX_ERR_TM_INVALID_COMMIT_RATE:
+ err_type = RTE_TM_ERROR_TYPE_SHAPER_PROFILE_COMMITTED_RATE;
+ break;
+ case NIX_ERR_TM_INVALID_PEAK_SZ:
+ err_type = RTE_TM_ERROR_TYPE_SHAPER_PROFILE_PEAK_SIZE;
+ break;
+ case NIX_ERR_TM_INVALID_PEAK_RATE:
+ err_type = RTE_TM_ERROR_TYPE_SHAPER_PROFILE_PEAK_RATE;
+ break;
+ case NIX_ERR_TM_INVALID_SHAPER_PROFILE:
+ err_type = RTE_TM_ERROR_TYPE_SHAPER_PROFILE_ID;
+ break;
+ case NIX_ERR_TM_SHAPER_PROFILE_IN_USE:
+ err_type = RTE_TM_ERROR_TYPE_SHAPER_PROFILE;
+ break;
+ case NIX_ERR_TM_INVALID_NODE:
+ err_type = RTE_TM_ERROR_TYPE_NODE_ID;
+ break;
+ case NIX_ERR_TM_PKT_MODE_MISMATCH:
+ err_type = RTE_TM_ERROR_TYPE_SHAPER_PROFILE_ID;
+ break;
+ case NIX_ERR_TM_INVALID_PARENT:
+ case NIX_ERR_TM_PARENT_PRIO_UPDATE:
+ err_type = RTE_TM_ERROR_TYPE_NODE_PARENT_NODE_ID;
+ break;
+ case NIX_ERR_TM_PRIO_ORDER:
+ case NIX_ERR_TM_MULTIPLE_RR_GROUPS:
+ err_type = RTE_TM_ERROR_TYPE_NODE_PRIORITY;
+ break;
+ case NIX_ERR_TM_PRIO_EXCEEDED:
+ err_type = RTE_TM_ERROR_TYPE_CAPABILITIES;
+ break;
+ default:
+ /**
+* Handle general error (as defined in linux errno.h)
+*/
+ if (abs(errorcode) < 300)
+ err_type = errorcode;
+ else
+ err_type = RTE_TM_ERROR_TYPE_UNSPECIFIED;
+ break;
+ }
+
+ return err_type;
+}
diff --git a/drivers/common/cnxk/cnxk_utils.h b/drivers/common/cnxk/cnxk_utils.h
new file mode 100644
index 00..5463cd49c4
--- /dev/null
+++ b/drivers/common/cnxk/cnxk_utils.h
@@ -0,0 +1,11 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(C) 2021 Marvell.
+ */
+#ifndef _CNXK_UTILS_H_
+#define _CNXK_UTILS_H_
+
+#include "roc_platform.h"
+
+int __roc_api roc_nix_tm_err_to_rte_err(int errorcode);
+
+#endif /* _CNXK_UTILS_H_ */
diff --git a/drivers/common/cnxk/meson.build b/drivers/common/cnxk/meson.build
index 6a7849f31c..3a0399dffa 100644
--- a/drivers/common/cnxk/meson.build
+++ b/drivers/common/cnxk/meson.build
@@ -60,5 +60,10 @@ sources = files(
# Security common code
sources += files('cnxk_security.c')
+# common DPDK utilities code
+sources += files('cnxk_utils.c')
+
includes += include_directories('../../bus/pci')
includes += include_directories('../../../lib/net')
+includes += include_directories('../../../lib/ethdev')
+includes += include_directories('../../../lib/meter')
diff --git a/drivers/common/cnxk/roc_utils.c b/drivers/common/cnxk/roc_utils.c
index 9cb8708a74..751486f503 100644
--- a/drivers/common/cnxk/roc_utils.c
+++ b/drivers/common/cnxk/roc_utils.c
@@ -64,6 +64,9 @@ roc_error_msg_get(int errorcode)
case NIX_ERR_TM_INVALID_SHAPER_PROFILE:
err_msg = "TM shaper profile invalid";
break;
+ case NIX_ERR_TM_PKT_MODE_MISMATCH:
+ err_msg = "shaper profile pkt mode mismatch";
+ break;
case NIX_ERR_TM_WEIGHT_EXCEED:
err_msg = "TM DWRR weight exceeded";
break;
@@ -88,6 +91,9 @@ roc_error_msg_get(int errorcode)
case NIX_ERR_TM_SHAPER_PROFILE_EXISTS:
err_msg = "TM shaper profile exists";
break;
+ case NIX_ERR_TM_SHAPE
[dpdk-dev] [PATCH 6/8] common/cnxk: set of handlers to get tm hierarchy internals
From: Satha Rao
Platform specific TM tree hierarchy details are part of common cnxk
driver. This patch introduces missing HAL apis to return state of
TM hierarchy required to support ethdev TM operations inside cnxk PMD.
Signed-off-by: Satha Rao
---
drivers/common/cnxk/roc_model.h| 6 ++
drivers/common/cnxk/roc_nix.h | 10 +++
drivers/common/cnxk/roc_nix_priv.h | 1 -
drivers/common/cnxk/roc_nix_tm.c | 22 ++-
drivers/common/cnxk/roc_nix_tm_ops.c | 11 +---
drivers/common/cnxk/roc_nix_tm_utils.c | 86 --
drivers/common/cnxk/version.map| 8 +++
7 files changed, 127 insertions(+), 17 deletions(-)
diff --git a/drivers/common/cnxk/roc_model.h b/drivers/common/cnxk/roc_model.h
index c1d11b77c6..856a570ab9 100644
--- a/drivers/common/cnxk/roc_model.h
+++ b/drivers/common/cnxk/roc_model.h
@@ -105,6 +105,12 @@ roc_model_is_cn96_ax(void)
return (roc_model->flag & ROC_MODEL_CN96xx_Ax);
}
+static inline uint64_t
+roc_model_is_cn96_cx(void)
+{
+ return (roc_model->flag & ROC_MODEL_CN96xx_C0);
+}
+
static inline uint64_t
roc_model_is_cn95_a0(void)
{
diff --git a/drivers/common/cnxk/roc_nix.h b/drivers/common/cnxk/roc_nix.h
index d947fe0900..59410257ad 100644
--- a/drivers/common/cnxk/roc_nix.h
+++ b/drivers/common/cnxk/roc_nix.h
@@ -486,6 +486,16 @@ int __roc_api roc_nix_tm_node_name_get(struct roc_nix
*roc_nix,
uint32_t node_id, char *buf,
size_t buflen);
int __roc_api roc_nix_smq_flush(struct roc_nix *roc_nix);
+int __roc_api roc_nix_tm_max_prio(struct roc_nix *roc_nix, int lvl);
+int __roc_api roc_nix_tm_lvl_is_leaf(struct roc_nix *roc_nix, int lvl);
+void __roc_api
+roc_nix_tm_shaper_default_red_algo(struct roc_nix_tm_node *node,
+ struct roc_nix_tm_shaper_profile *profile);
+int __roc_api roc_nix_tm_lvl_cnt_get(struct roc_nix *roc_nix);
+int __roc_api roc_nix_tm_lvl_have_link_access(struct roc_nix *roc_nix, int
lvl);
+int __roc_api roc_nix_tm_prepare_rate_limited_tree(struct roc_nix *roc_nix);
+bool __roc_api roc_nix_tm_is_user_hierarchy_enabled(struct roc_nix *nix);
+int __roc_api roc_nix_tm_tree_type_get(struct roc_nix *nix);
/* MAC */
int __roc_api roc_nix_mac_rxtx_start_stop(struct roc_nix *roc_nix, bool start);
diff --git a/drivers/common/cnxk/roc_nix_priv.h
b/drivers/common/cnxk/roc_nix_priv.h
index 3412bf25e5..b67f648e5a 100644
--- a/drivers/common/cnxk/roc_nix_priv.h
+++ b/drivers/common/cnxk/roc_nix_priv.h
@@ -350,7 +350,6 @@ int nix_tm_txsch_reg_config(struct nix *nix, enum
roc_nix_tm_tree tree);
int nix_tm_update_parent_info(struct nix *nix, enum roc_nix_tm_tree tree);
int nix_tm_sq_sched_conf(struct nix *nix, struct nix_tm_node *node,
bool rr_quantum_only);
-int nix_tm_prepare_rate_limited_tree(struct roc_nix *roc_nix);
/*
* TM priv utils.
diff --git a/drivers/common/cnxk/roc_nix_tm.c b/drivers/common/cnxk/roc_nix_tm.c
index 947320ae63..08d6e866fe 100644
--- a/drivers/common/cnxk/roc_nix_tm.c
+++ b/drivers/common/cnxk/roc_nix_tm.c
@@ -155,6 +155,20 @@ nix_tm_update_parent_info(struct nix *nix, enum
roc_nix_tm_tree tree)
return 0;
}
+static int
+nix_tm_root_node_get(struct nix *nix, int tree)
+{
+ struct nix_tm_node_list *list = nix_tm_node_list(nix, tree);
+ struct nix_tm_node *tm_node;
+
+ TAILQ_FOREACH(tm_node, list, node) {
+ if (tm_node->hw_lvl == nix->tm_root_lvl)
+ return 1;
+ }
+
+ return 0;
+}
+
int
nix_tm_node_add(struct roc_nix *roc_nix, struct nix_tm_node *node)
{
@@ -207,6 +221,10 @@ nix_tm_node_add(struct roc_nix *roc_nix, struct
nix_tm_node *node)
if (nix_tm_node_search(nix, node_id, tree))
return NIX_ERR_TM_NODE_EXISTS;
+ /* Check if root node exists */
+ if (hw_lvl == nix->tm_root_lvl && nix_tm_root_node_get(nix, tree))
+ return NIX_ERR_TM_NODE_EXISTS;
+
profile = nix_tm_shaper_profile_search(nix, profile_id);
if (!nix_tm_is_leaf(nix, lvl)) {
/* Check if shaper profile exists for non leaf node */
@@ -1157,7 +1175,7 @@ nix_tm_prepare_default_tree(struct roc_nix *roc_nix)
}
int
-nix_tm_prepare_rate_limited_tree(struct roc_nix *roc_nix)
+roc_nix_tm_prepare_rate_limited_tree(struct roc_nix *roc_nix)
{
struct nix *nix = roc_nix_to_nix_priv(roc_nix);
uint32_t nonleaf_id = nix->nb_tx_queues;
@@ -1227,7 +1245,7 @@ nix_tm_prepare_rate_limited_tree(struct roc_nix *roc_nix)
goto error;
node->id = i;
- node->parent_id = parent;
+ node->parent_id = parent + i;
node->priority = 0;
node->weight = NIX_TM_DFLT_RR_WT;
node->shaper_profile_id = ROC_NIX_TM_SHAPER_PROFILE_NONE;
diff --git a/drivers/common/cnxk/roc_nix_tm_ops.c
b/drivers/common/cnxk/roc_nix_tm_ops.c
[dpdk-dev] [PATCH 7/8] net/cnxk: tm capabilities and queue rate limit handlers
From: Satha Rao
Initial version of TM implementation added basic infrastructure,
tm node_get, capabilities operations and rate limit queue operation.
Signed-off-by: Satha Rao
---
drivers/net/cnxk/cnxk_ethdev.c | 2 +
drivers/net/cnxk/cnxk_ethdev.h | 3 +
drivers/net/cnxk/cnxk_tm.c | 322 +
drivers/net/cnxk/cnxk_tm.h | 18 ++
drivers/net/cnxk/meson.build | 1 +
5 files changed, 346 insertions(+)
create mode 100644 drivers/net/cnxk/cnxk_tm.c
create mode 100644 drivers/net/cnxk/cnxk_tm.h
diff --git a/drivers/net/cnxk/cnxk_ethdev.c b/drivers/net/cnxk/cnxk_ethdev.c
index 0e3652ed51..fb4a4e8c97 100644
--- a/drivers/net/cnxk/cnxk_ethdev.c
+++ b/drivers/net/cnxk/cnxk_ethdev.c
@@ -1276,6 +1276,8 @@ struct eth_dev_ops cnxk_eth_dev_ops = {
.rss_hash_update = cnxk_nix_rss_hash_update,
.rss_hash_conf_get = cnxk_nix_rss_hash_conf_get,
.set_mc_addr_list = cnxk_nix_mc_addr_list_configure,
+ .set_queue_rate_limit = cnxk_nix_tm_set_queue_rate_limit,
+ .tm_ops_get = cnxk_nix_tm_ops_get,
};
static int
diff --git a/drivers/net/cnxk/cnxk_ethdev.h b/drivers/net/cnxk/cnxk_ethdev.h
index 2528b3cdaa..80e144d183 100644
--- a/drivers/net/cnxk/cnxk_ethdev.h
+++ b/drivers/net/cnxk/cnxk_ethdev.h
@@ -330,6 +330,9 @@ int cnxk_nix_tsc_convert(struct cnxk_eth_dev *dev);
int cnxk_nix_read_clock(struct rte_eth_dev *eth_dev, uint64_t *clock);
uint64_t cnxk_nix_rxq_mbuf_setup(struct cnxk_eth_dev *dev);
+int cnxk_nix_tm_ops_get(struct rte_eth_dev *eth_dev, void *ops);
+int cnxk_nix_tm_set_queue_rate_limit(struct rte_eth_dev *eth_dev,
+uint16_t queue_idx, uint16_t tx_rate);
/* RSS */
uint32_t cnxk_rss_ethdev_to_nix(struct cnxk_eth_dev *dev, uint64_t ethdev_rss,
diff --git a/drivers/net/cnxk/cnxk_tm.c b/drivers/net/cnxk/cnxk_tm.c
new file mode 100644
index 00..33dab15b55
--- /dev/null
+++ b/drivers/net/cnxk/cnxk_tm.c
@@ -0,0 +1,322 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(C) 2021 Marvell.
+ */
+#include
+#include
+#include
+
+static int
+cnxk_nix_tm_node_type_get(struct rte_eth_dev *eth_dev, uint32_t node_id,
+ int *is_leaf, struct rte_tm_error *error)
+{
+ struct cnxk_eth_dev *dev = cnxk_eth_pmd_priv(eth_dev);
+ struct roc_nix *nix = &dev->nix;
+ struct roc_nix_tm_node *node;
+
+ if (is_leaf == NULL) {
+ error->type = RTE_TM_ERROR_TYPE_UNSPECIFIED;
+ return -EINVAL;
+ }
+
+ node = roc_nix_tm_node_get(nix, node_id);
+ if (node_id == RTE_TM_NODE_ID_NULL || !node) {
+ error->type = RTE_TM_ERROR_TYPE_NODE_ID;
+ return -EINVAL;
+ }
+
+ if (roc_nix_tm_lvl_is_leaf(nix, node->lvl))
+ *is_leaf = true;
+ else
+ *is_leaf = false;
+
+ return 0;
+}
+
+static int
+cnxk_nix_tm_capa_get(struct rte_eth_dev *eth_dev,
+struct rte_tm_capabilities *cap,
+struct rte_tm_error *error)
+{
+ struct cnxk_eth_dev *dev = cnxk_eth_pmd_priv(eth_dev);
+ int rc, max_nr_nodes = 0, i, n_lvl;
+ struct roc_nix *nix = &dev->nix;
+ uint16_t schq[ROC_TM_LVL_MAX];
+
+ memset(cap, 0, sizeof(*cap));
+
+ rc = roc_nix_tm_rsrc_count(nix, schq);
+ if (rc) {
+ error->type = RTE_TM_ERROR_TYPE_UNSPECIFIED;
+ error->message = "unexpected fatal error";
+ return rc;
+ }
+
+ for (i = 0; i < NIX_TXSCH_LVL_TL1; i++)
+ max_nr_nodes += schq[i];
+
+ cap->n_nodes_max = max_nr_nodes + dev->nb_txq;
+
+ n_lvl = roc_nix_tm_lvl_cnt_get(nix);
+ /* Consider leaf level */
+ cap->n_levels_max = n_lvl + 1;
+ cap->non_leaf_nodes_identical = 1;
+ cap->leaf_nodes_identical = 1;
+
+ /* Shaper Capabilities */
+ cap->shaper_private_n_max = max_nr_nodes;
+ cap->shaper_n_max = max_nr_nodes;
+ cap->shaper_private_dual_rate_n_max = max_nr_nodes;
+ cap->shaper_private_rate_min = NIX_TM_MIN_SHAPER_RATE / 8;
+ cap->shaper_private_rate_max = NIX_TM_MAX_SHAPER_RATE / 8;
+ cap->shaper_private_packet_mode_supported = 1;
+ cap->shaper_private_byte_mode_supported = 1;
+ cap->shaper_pkt_length_adjust_min = NIX_TM_LENGTH_ADJUST_MIN;
+ cap->shaper_pkt_length_adjust_max = NIX_TM_LENGTH_ADJUST_MAX;
+
+ /* Schedule Capabilities */
+ cap->sched_n_children_max = schq[n_lvl - 1];
+ cap->sched_sp_n_priorities_max = NIX_TM_TLX_SP_PRIO_MAX;
+ cap->sched_wfq_n_children_per_group_max = cap->sched_n_children_max;
+ cap->sched_wfq_n_groups_max = 1;
+ cap->sched_wfq_weight_max = roc_nix_tm_max_sched_wt_get();
+ cap->sched_wfq_packet_mode_supported = 1;
+ cap->sched_wfq_byte_mode_supported = 1;
+
+ cap->dynamic_update_mask = RTE_TM_UPDATE_NODE_PARENT_KEEP_LEVEL |
+ RTE_TM_UPDATE_NODE_SUS
[dpdk-dev] [PATCH 8/8] net/cnxk: tm shaper and node operations
From: Satha Rao
Implemented TM node, shaper profile, hierarchy_commit and
statistic operations.
Signed-off-by: Satha Rao
---
drivers/net/cnxk/cnxk_tm.c | 353 +
drivers/net/cnxk/cnxk_tm.h | 5 +
2 files changed, 358 insertions(+)
diff --git a/drivers/net/cnxk/cnxk_tm.c b/drivers/net/cnxk/cnxk_tm.c
index 33dab15b55..82f0613a74 100644
--- a/drivers/net/cnxk/cnxk_tm.c
+++ b/drivers/net/cnxk/cnxk_tm.c
@@ -259,11 +259,364 @@ cnxk_nix_tm_node_capa_get(struct rte_eth_dev *eth_dev,
uint32_t node_id,
return 0;
}
+static int
+cnxk_nix_tm_shaper_profile_add(struct rte_eth_dev *eth_dev, uint32_t id,
+ struct rte_tm_shaper_params *params,
+ struct rte_tm_error *error)
+{
+ struct cnxk_eth_dev *dev = cnxk_eth_pmd_priv(eth_dev);
+ struct cnxk_nix_tm_shaper_profile *profile;
+ struct roc_nix *nix = &dev->nix;
+ int rc;
+
+ if (roc_nix_tm_shaper_profile_get(nix, id)) {
+ error->type = RTE_TM_ERROR_TYPE_SHAPER_PROFILE_ID;
+ error->message = "shaper profile ID exist";
+ return -EINVAL;
+ }
+
+ profile = rte_zmalloc("cnxk_nix_tm_shaper_profile",
+ sizeof(struct cnxk_nix_tm_shaper_profile), 0);
+ if (!profile)
+ return -ENOMEM;
+ profile->profile.id = id;
+ profile->profile.commit_rate = params->committed.rate;
+ profile->profile.peak_rate = params->peak.rate;
+ profile->profile.commit_sz = params->committed.size;
+ profile->profile.peak_sz = params->peak.size;
+ /* If Byte mode, then convert to bps */
+ if (!params->packet_mode) {
+ profile->profile.commit_rate *= 8;
+ profile->profile.peak_rate *= 8;
+ profile->profile.commit_sz *= 8;
+ profile->profile.peak_sz *= 8;
+ }
+ profile->profile.pkt_len_adj = params->pkt_length_adjust;
+ profile->profile.pkt_mode = params->packet_mode;
+ profile->profile.free_fn = rte_free;
+ rte_memcpy(&profile->params, params,
+ sizeof(struct rte_tm_shaper_params));
+
+ rc = roc_nix_tm_shaper_profile_add(nix, &profile->profile);
+
+ /* fill error information based on return value */
+ if (rc) {
+ error->type = roc_nix_tm_err_to_rte_err(rc);
+ error->message = roc_error_msg_get(rc);
+ }
+
+ return rc;
+}
+
+static int
+cnxk_nix_tm_shaper_profile_delete(struct rte_eth_dev *eth_dev,
+ uint32_t profile_id,
+ struct rte_tm_error *error)
+{
+ struct cnxk_eth_dev *dev = cnxk_eth_pmd_priv(eth_dev);
+ struct roc_nix *nix = &dev->nix;
+ int rc;
+
+ rc = roc_nix_tm_shaper_profile_delete(nix, profile_id);
+ if (rc) {
+ error->type = roc_nix_tm_err_to_rte_err(rc);
+ error->message = roc_error_msg_get(rc);
+ }
+
+ return rc;
+}
+
+static int
+cnxk_nix_tm_node_add(struct rte_eth_dev *eth_dev, uint32_t node_id,
+uint32_t parent_node_id, uint32_t priority,
+uint32_t weight, uint32_t lvl,
+struct rte_tm_node_params *params,
+struct rte_tm_error *error)
+{
+ struct cnxk_eth_dev *dev = cnxk_eth_pmd_priv(eth_dev);
+ struct roc_nix_tm_shaper_profile *profile;
+ struct roc_nix_tm_node *parent_node;
+ struct roc_nix *nix = &dev->nix;
+ struct cnxk_nix_tm_node *node;
+ int rc;
+
+ /* we don't support dynamic updates */
+ if (roc_nix_tm_is_user_hierarchy_enabled(nix)) {
+ error->type = RTE_TM_ERROR_TYPE_CAPABILITIES;
+ error->message = "dynamic update not supported";
+ return -EIO;
+ }
+
+ parent_node = roc_nix_tm_node_get(nix, parent_node_id);
+ /* find the right level */
+ if (lvl == RTE_TM_NODE_LEVEL_ID_ANY) {
+ if (parent_node_id == RTE_TM_NODE_ID_NULL) {
+ lvl = ROC_TM_LVL_ROOT;
+ } else if (parent_node) {
+ lvl = parent_node->lvl + 1;
+ } else {
+ /* Neither proper parent nor proper level id given */
+ error->type = RTE_TM_ERROR_TYPE_NODE_PARENT_NODE_ID;
+ error->message = "invalid parent node id";
+ return -ERANGE;
+ }
+ }
+
+ node = rte_zmalloc("cnxk_nix_tm_node", sizeof(struct cnxk_nix_tm_node),
+ 0);
+ if (!node)
+ return -ENOMEM;
+
+ rte_memcpy(&node->params, params, sizeof(struct rte_tm_node_params));
+
+ node->nix_node.id = node_id;
+ node->nix_node.parent_id = parent_node_id;
+ node->nix_node.priority = priority;
+ node->nix_node.weight = weight;
+ node->nix_node.lvl =
Re: [dpdk-dev] [PATCH] eal: add log to print dpdk version at start
On Wed, 1 Sep 2021 11:48:21 +0530 Aman Singh wrote: > From the logs it is difficult to get the DPDK version > that was used. So added a debug log to print the same. > The log has been added in eal_init so it gets printed > at startup for any application. > > Signed-off-by: Aman Singh > --- > lib/eal/freebsd/eal.c | 2 +- > lib/eal/linux/eal.c | 2 +- > lib/eal/windows/eal.c | 1 + > 3 files changed, 3 insertions(+), 2 deletions(-) > > diff --git a/lib/eal/freebsd/eal.c b/lib/eal/freebsd/eal.c > index 6cee5ae369..a14a205f4d 100644 > --- a/lib/eal/freebsd/eal.c > +++ b/lib/eal/freebsd/eal.c > @@ -958,7 +958,7 @@ rte_eal_init(int argc, char **argv) > } > > eal_mcfg_complete(); > - > + rte_log(RTE_LOG_DEBUG, RTE_LOGTYPE_EAL, "DPDK version: %s\n", > rte_version()); > return fctret; > } > > diff --git a/lib/eal/linux/eal.c b/lib/eal/linux/eal.c > index 3577eaeaa4..a50960cc78 100644 > --- a/lib/eal/linux/eal.c > +++ b/lib/eal/linux/eal.c > @@ -1332,7 +1332,7 @@ rte_eal_init(int argc, char **argv) > } > > eal_mcfg_complete(); > - > + rte_log(RTE_LOG_DEBUG, RTE_LOGTYPE_EAL, "DPDK version: %s\n", > rte_version()); > return fctret; > } > > diff --git a/lib/eal/windows/eal.c b/lib/eal/windows/eal.c > index 3d8c520412..5e6d5d8930 100644 > --- a/lib/eal/windows/eal.c > +++ b/lib/eal/windows/eal.c > @@ -419,6 +419,7 @@ rte_eal_init(int argc, char **argv) >*/ > rte_eal_mp_remote_launch(sync_func, NULL, SKIP_MAIN); > rte_eal_mp_wait_lcore(); > + rte_log(RTE_LOG_DEBUG, RTE_LOGTYPE_EAL, "DPDK version: %s\n", > rte_version()); > return fctret; > } > NAK DPDK is already too chatty at startup. If your application needs it then do-it-yourself.
