* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
@ 2019-04-22 6:25 Akhil Goyal
2019-04-22 6:25 ` Akhil Goyal
0 siblings, 1 reply; 25+ messages in thread
From: Akhil Goyal @ 2019-04-22 6:25 UTC (permalink / raw)
To: Iremonger, Bernard, dev, Ananyev, Konstantin; +Cc: stable
Hi Bernard,
>
> Hi Akhil,
>
> <snip>
>
> > Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped
> > for inline crypto
>
> <snip>
> > > + if (sa->type ==
> > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > + ctx = (struct rte_security_ctx *)
> > > + rte_eth_dev_get_sec_ctx(sa->portid);
> >
> > This is breaking the lookaside mode. Ctx was retrieved using the ipsec_ctx-
> > >tbl struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > rte_cryptodev_get_sec_ctx(
> > ipsec_ctx->tbl[cdev_id_qp].id);
> >
> > I am looking into it, but I don't have time left to get it integrated in RC2. So
> > this has to be pushed to RC3
>
> <snip>
>
> Unfortunately we do not have the HW to test this feature.
> What HW are you using to test this?
>
> Having looked at the code previously
> ipsec_ctx->tbl[cdev_id_qp].id turned out to be the port_id.
>
> So we had expected it to work.
>
> We will need your help with this.
I am looking into this. Will let you know when I get the fix.
>
> Regards,
>
> Bernard.
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-04-22 6:25 [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto Akhil Goyal
@ 2019-04-22 6:25 ` Akhil Goyal
0 siblings, 0 replies; 25+ messages in thread
From: Akhil Goyal @ 2019-04-22 6:25 UTC (permalink / raw)
To: Iremonger, Bernard, dev, Ananyev, Konstantin; +Cc: stable
Hi Bernard,
>
> Hi Akhil,
>
> <snip>
>
> > Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped
> > for inline crypto
>
> <snip>
> > > + if (sa->type ==
> > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > + ctx = (struct rte_security_ctx *)
> > > + rte_eth_dev_get_sec_ctx(sa->portid);
> >
> > This is breaking the lookaside mode. Ctx was retrieved using the ipsec_ctx-
> > >tbl struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > rte_cryptodev_get_sec_ctx(
> > ipsec_ctx->tbl[cdev_id_qp].id);
> >
> > I am looking into it, but I don't have time left to get it integrated in RC2. So
> > this has to be pushed to RC3
>
> <snip>
>
> Unfortunately we do not have the HW to test this feature.
> What HW are you using to test this?
>
> Having looked at the code previously
> ipsec_ctx->tbl[cdev_id_qp].id turned out to be the port_id.
>
> So we had expected it to work.
>
> We will need your help with this.
I am looking into this. Will let you know when I get the fix.
>
> Regards,
>
> Bernard.
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-05-13 14:29 ` Ananyev, Konstantin
2019-05-13 14:29 ` Ananyev, Konstantin
@ 2019-05-27 8:58 ` Iremonger, Bernard
1 sibling, 0 replies; 25+ messages in thread
From: Iremonger, Bernard @ 2019-05-27 8:58 UTC (permalink / raw)
To: Ananyev, Konstantin, Akhil Goyal, dev; +Cc: stable
Hi Akhil,
<snip>
> Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for
> inline crypto
>
>
>
> > > > > > > > Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st
> > > > > > > > packet
> > > > dropped
> > > > > > for
> > > > > > > > inline crypto
> > > > > > > >
> > > > > > > > Hi Bernard,
> > > > > > > >
> > > > > > > > > - RTE_LOG_DP(DEBUG, IPSEC, "Create session for SA spi %u on
> > > > > > cryptodev "
> > > > > > > > > - "%u qp %u\n", sa->spi,
> > > > > > > > > - ipsec_ctx->tbl[cdev_id_qp].id,
> > > > > > > > > - ipsec_ctx->tbl[cdev_id_qp].qp);
> > > > > > > > > + if ((sa == NULL) || (pool == NULL))
> > > > > > > > > + return -EINVAL;
> > > > > > > > >
> > > > > > > > > - if (sa->type != RTE_SECURITY_ACTION_TYPE_NONE) {
> > > > > > > > > - struct rte_security_session_conf sess_conf = {
> > > > > > > > > + struct rte_security_session_conf sess_conf = {
> > > > > > > > > .action_type = sa->type,
> > > > > > > > > .protocol = RTE_SECURITY_PROTOCOL_IPSEC,
> > > > > > > > > {.ipsec = { @@ -90,247 +65,340
> > > > > > > > > @@ create_session(struct ipsec_ctx *ipsec_ctx,
> > > > > > struct
> > > > > > > > > ipsec_sa *sa)
> > > > > > > > > } },
> > > > > > > > > .crypto_xform = sa->xforms,
> > > > > > > > > .userdata = NULL,
> > > > > > > > > -
> > > > > > > > > };
> > > > > > > > >
> > > > > > > > > - if (sa->type ==
> > > > > > > > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL)
> > > > > > > > > {
> > > > > > > > > - struct rte_security_ctx *ctx = (struct rte_security_ctx
> *)
> > > > > > > > > - rte_cryptodev_get_sec_ctx(
> > > > > > > > > - ipsec_ctx->tbl[cdev_id_qp].id);
> > > > > > > > > -
> > > > > > > > > - /* Set IPsec parameters in conf */
> > > > > > > > > - set_ipsec_conf(sa, &(sess_conf.ipsec));
> > > > > > > > > -
> > > > > > > > > - sa->sec_session = rte_security_session_create(ctx,
> > > > > > > > > - &sess_conf, ipsec_ctx->session_pool);
> > > > > > > > > - if (sa->sec_session == NULL) {
> > > > > > > > > - RTE_LOG(ERR, IPSEC,
> > > > > > > > > - "SEC Session init failed: err: %d\n", ret);
> > > > > > > > > - return -1;
> > > > > > > > > - }
> > > > > > > > > - } else if (sa->type ==
> > > > > > RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO)
> > > > > > > > {
> > > > > > > > > - struct rte_flow_error err;
> > > > > > > > > - struct rte_security_ctx *ctx = (struct rte_security_ctx
> *)
> > > > > > > > > - rte_eth_dev_get_sec_ctx(
> > > > > > > > > - sa->portid);
> > > > > > > > > - const struct rte_security_capability *sec_cap;
> > > > > > > > > - int ret = 0;
> > > > > > > > > -
> > > > > > > > > - sa->sec_session = rte_security_session_create(ctx,
> > > > > > > > > - &sess_conf, ipsec_ctx->session_pool);
> > > > > > > > > - if (sa->sec_session == NULL) {
> > > > > > > > > - RTE_LOG(ERR, IPSEC,
> > > > > > > > > - "SEC Session init failed: err: %d\n", ret);
> > > > > > > > > - return -1;
> > > > > > > > > - }
> > > > > > > > > + if (sa->type ==
> > > > > > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > > > > > > > + ctx = (struct rte_security_ctx *)
> > > > > > > > > +
> > > > > > > > > + rte_eth_dev_get_sec_ctx(sa->portid);
> > > > > > > >
> > > > > > > > This is breaking the lookaside mode. Ctx was retrieved
> > > > > > > > using the
> > > > ipsec_ctx-
> > > > > > >tbl
> > > > > > > > struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > > > > rte_cryptodev_get_sec_ctx(
> > > > > > > > ipsec_ctx->tbl[cdev_id_qp].id);
> > > > > > > >
> > > > > > > > I am looking into it, but I don't have time left to get it integrated in
> RC2.
> > > > So
> > > > > > this
> > > > > > > > has to be pushed to RC3
> > > > > > >
> > > > > > > It looks like there are multiple issues in this patch wrt
> > > > > > > lookaside and none
> > > > cases.
> > > > > > Only the inline cases seem to be working.
> > > > > > >
> > > > > > > 1. the patch removes the cdev_mapping concept completely.
> > > > > > > Cdev_id_qp is
> > > > > > not getting used.
> > > > > >
> > > > > > Not exactly.
> > > > > > cdev_id_qp is still setup, and is still used to decide to
> > > > > > which crypto-dev to enqueuer the crypto-op:
> > > > > > ipsec_enqueue(...)
> > > > > > {
> > > > > > ...
> > > > > > enqueue_cop(&ipsec_ctx->tbl[sa->cdev_id_qp], &priv->cop);
> > > > >
> > > > > I don't see anybody filling "sa->cdev_id_qp". Please let me know
> > > > > if I have
> > > > missed it somewhere.
> > > > > It is memset to 0 I guess.
> > > >
> > > >
> > > > Yep, true we lost it somewhere during the rework.
> > > >
> > > > >
> > > > > >
> > > > > >
> > > > > > Same in ipsec_process().
> > > > > >
> > > > > > For initialization, yes cdev_id_qp is not used anymore.
> > > > > > As discussed here:
> > > > > >
> > > > https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2F
> > > > mail
> > > > s.dp
> > > > > > dk.org%2Farchives%2Fdev%2F2019-
> > > > > >
> > > >
> > >
> March%2F127725.html&data=02%7C01%7Cakhil.goyal%40nxp.com%7C04
> > > > > >
> > > >
> > >
> 194193cfc04c0b629008d6c7eea247%7C686ea1d3bc2b4c6fa92cd99c5c301635%
> > > > > >
> > > >
> > >
> 7C0%7C0%7C636916225072561313&sdata=ga9IiqhYRWOz9QkRDIXNiigInk
> > > > > > soVGgu1E5EetqvE%2FA%3D&reserved=0
> > > > > >
> > > > > > I think the problem you are hitting with lookaside-proto is
> > > > > > that for it we use 2 different values here:
> > > > > > a) In create_sec_session we use portid (it also should be
> > > > > > rte_cryptodev_get_sec_ctx() here)
> > > > > > if (sa->type ==
> RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > > > > ctx = (struct rte_security_ctx *)
> > > > > >
> > > > > > rte_eth_dev_get_sec_ctx(sa->portid);
> > > > > It should be rte_cryptodev_get_sec_ctx in the first place. And
> > > > > it needs a
> > > > cdev_id as input based on the cdev mapping done while initializing
> > > > > the cryptodev and neither the portid and nor cdev_id_qp.
> > > > > > b) in enqueue() we use cdev_id_qp
> > > > > >
> > > > > > Right now these values could be different.
> > > > > > As I understand we need to make sure that fro lookaside-proto
> > > > > > cdev_id_qp
> > > > ==
> > > > > > portid provided by user, correct?
> > > > > No it is not the case. Right now for lookaside there is no use
> > > > > of portid in case
> > > > of lookaside case.
> > > > > As the cdev/qp/core mappings are managed internally and the user
> > > > > cannot
> > > > tweak it from cfg file.
> > > > >
> > > >
> > > >
> > > > Hmm, then at least that line is wrong here:
> > > >
> https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdoc.
> > > > dpdk
> > > >
> > >
> .org%2Fguides%2Fsample_app_ug%2Fipsec_secgw.html&data=02%7C01%
> > > >
> > >
> 7Cakhil.goyal%40nxp.com%7Cb1e931a9967943887a0c08d6c7f49d28%7C686ea
> > > >
> > >
> 1d3bc2b4c6fa92cd99c5c301635%7C0%7C0%7C636916250754452306&sda
> > > >
> > >
> ta=tNjHCO0O2rfh8shhQXu93qrM0Hr0OZVXUVhMcsg53dw%3D&reserved=
> > > > 0
> > > >
> > > > sa out 5 cipher_algo aes-128-cbc cipher_key
> > > > 0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0 \ auth_algo sha1-hmac auth_key
> > > > 0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0 \ mode ipv4-tunnel src
> > > > 172.16.1.5 dst 172.16.2.5 \ type lookaside-protocol-offload
> > > > port_id 4
> > > >
> > > > And probably:
> > > > "Port/device ID of the ethernet/crypto accelerator for which the
> > > > SA is configured."
> > > > Need to be rephrased to remove crypto accelerator notice.
> > > Yes.
> > >
> > > >
> > > > Another question - why you guys don't consider using portid for
> > > > lookaside-
> > > proto?
> > > > As I can see add_mapping(function that fills cdev_id_qp) doesn't
> > > > bother to check which rte_security protocols are supported (only
> > > > crypto capabilities are checked).
> > > > So not sure does current code will work ok with a mix of
> > > > lookaside- none/lookaside-proto devices.
> > > > Forcing user to specify crypto-dev id for lookaside-proto (via
> > > > portid or so) will simplify things significantly.
> > > I will think about it. Actually initially when portid was
> > > introduced, the intent was same but it did not work well.
> > > Because there may be a case of a cryptodev with multiple queues, so
> > > there will be a mapping done internally in the application and
> > > matching it with the user provided portid will be difficult.
> > >
> > > I believe that this could be done but would need some rework.
> > >
> > > >
> > > > Konstantin
> >
> > Could we consider going back to the V2 version of this patch set which
> > fixed the issue with the inline code and left the lookaside code unchanged?
>
> Actually I have the same thoughts here:
> considering the problems with lookaside-proto, it seems more pragmatic to fix it
> as was suggested in V2:
> split create_session() into 2 functions, and invoke create_session_inline() at
> startup, while keeping create_session_lookaside() invocation at runtime.
> At least it would fix the issue in a clean way.
> If later ipsec-secgw will be reworked to allow lookaside session creation at init
> time, we can merge them back.
> Konstantin
>
> >
> > We are not in a position test any changes to the lookaside code.
> >
> > Regards,
> >
> > Bernard.
> >
Both Konstantin and I have proposed using the V2 patch set as a starting point to fix this issue instead of the V4 patch set which has caused issues with the lookaside code.
Have you an opinion on this proposal?
Regards,
Bernard.
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-04-24 10:40 ` Iremonger, Bernard
2019-04-24 10:40 ` Iremonger, Bernard
@ 2019-05-13 14:29 ` Ananyev, Konstantin
2019-05-13 14:29 ` Ananyev, Konstantin
2019-05-27 8:58 ` Iremonger, Bernard
1 sibling, 2 replies; 25+ messages in thread
From: Ananyev, Konstantin @ 2019-05-13 14:29 UTC (permalink / raw)
To: Iremonger, Bernard, Akhil Goyal, dev; +Cc: stable
> > > > > > > Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st
> > > > > > > packet
> > > dropped
> > > > > for
> > > > > > > inline crypto
> > > > > > >
> > > > > > > Hi Bernard,
> > > > > > >
> > > > > > > > - RTE_LOG_DP(DEBUG, IPSEC, "Create session for SA spi %u on
> > > > > cryptodev "
> > > > > > > > - "%u qp %u\n", sa->spi,
> > > > > > > > - ipsec_ctx->tbl[cdev_id_qp].id,
> > > > > > > > - ipsec_ctx->tbl[cdev_id_qp].qp);
> > > > > > > > + if ((sa == NULL) || (pool == NULL))
> > > > > > > > + return -EINVAL;
> > > > > > > >
> > > > > > > > - if (sa->type != RTE_SECURITY_ACTION_TYPE_NONE) {
> > > > > > > > - struct rte_security_session_conf sess_conf = {
> > > > > > > > + struct rte_security_session_conf sess_conf = {
> > > > > > > > .action_type = sa->type,
> > > > > > > > .protocol = RTE_SECURITY_PROTOCOL_IPSEC,
> > > > > > > > {.ipsec = { @@ -90,247 +65,340 @@
> > > > > > > > create_session(struct ipsec_ctx *ipsec_ctx,
> > > > > struct
> > > > > > > > ipsec_sa *sa)
> > > > > > > > } },
> > > > > > > > .crypto_xform = sa->xforms,
> > > > > > > > .userdata = NULL,
> > > > > > > > -
> > > > > > > > };
> > > > > > > >
> > > > > > > > - if (sa->type ==
> > > > > > > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL)
> > > > > > > > {
> > > > > > > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > > > > - rte_cryptodev_get_sec_ctx(
> > > > > > > > - ipsec_ctx->tbl[cdev_id_qp].id);
> > > > > > > > -
> > > > > > > > - /* Set IPsec parameters in conf */
> > > > > > > > - set_ipsec_conf(sa, &(sess_conf.ipsec));
> > > > > > > > -
> > > > > > > > - sa->sec_session = rte_security_session_create(ctx,
> > > > > > > > - &sess_conf, ipsec_ctx->session_pool);
> > > > > > > > - if (sa->sec_session == NULL) {
> > > > > > > > - RTE_LOG(ERR, IPSEC,
> > > > > > > > - "SEC Session init failed: err: %d\n", ret);
> > > > > > > > - return -1;
> > > > > > > > - }
> > > > > > > > - } else if (sa->type ==
> > > > > RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO)
> > > > > > > {
> > > > > > > > - struct rte_flow_error err;
> > > > > > > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > > > > - rte_eth_dev_get_sec_ctx(
> > > > > > > > - sa->portid);
> > > > > > > > - const struct rte_security_capability *sec_cap;
> > > > > > > > - int ret = 0;
> > > > > > > > -
> > > > > > > > - sa->sec_session = rte_security_session_create(ctx,
> > > > > > > > - &sess_conf, ipsec_ctx->session_pool);
> > > > > > > > - if (sa->sec_session == NULL) {
> > > > > > > > - RTE_LOG(ERR, IPSEC,
> > > > > > > > - "SEC Session init failed: err: %d\n", ret);
> > > > > > > > - return -1;
> > > > > > > > - }
> > > > > > > > + if (sa->type ==
> > > > > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > > > > > > + ctx = (struct rte_security_ctx *)
> > > > > > > > +
> > > > > > > > + rte_eth_dev_get_sec_ctx(sa->portid);
> > > > > > >
> > > > > > > This is breaking the lookaside mode. Ctx was retrieved using
> > > > > > > the
> > > ipsec_ctx-
> > > > > >tbl
> > > > > > > struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > > > rte_cryptodev_get_sec_ctx(
> > > > > > > ipsec_ctx->tbl[cdev_id_qp].id);
> > > > > > >
> > > > > > > I am looking into it, but I don't have time left to get it integrated in RC2.
> > > So
> > > > > this
> > > > > > > has to be pushed to RC3
> > > > > >
> > > > > > It looks like there are multiple issues in this patch wrt
> > > > > > lookaside and none
> > > cases.
> > > > > Only the inline cases seem to be working.
> > > > > >
> > > > > > 1. the patch removes the cdev_mapping concept completely.
> > > > > > Cdev_id_qp is
> > > > > not getting used.
> > > > >
> > > > > Not exactly.
> > > > > cdev_id_qp is still setup, and is still used to decide to which
> > > > > crypto-dev to enqueuer the crypto-op:
> > > > > ipsec_enqueue(...)
> > > > > {
> > > > > ...
> > > > > enqueue_cop(&ipsec_ctx->tbl[sa->cdev_id_qp], &priv->cop);
> > > >
> > > > I don't see anybody filling "sa->cdev_id_qp". Please let me know if
> > > > I have
> > > missed it somewhere.
> > > > It is memset to 0 I guess.
> > >
> > >
> > > Yep, true we lost it somewhere during the rework.
> > >
> > > >
> > > > >
> > > > >
> > > > > Same in ipsec_process().
> > > > >
> > > > > For initialization, yes cdev_id_qp is not used anymore.
> > > > > As discussed here:
> > > > >
> > > https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmail
> > > s.dp
> > > > > dk.org%2Farchives%2Fdev%2F2019-
> > > > >
> > >
> > March%2F127725.html&data=02%7C01%7Cakhil.goyal%40nxp.com%7C04
> > > > >
> > >
> > 194193cfc04c0b629008d6c7eea247%7C686ea1d3bc2b4c6fa92cd99c5c301635%
> > > > >
> > >
> > 7C0%7C0%7C636916225072561313&sdata=ga9IiqhYRWOz9QkRDIXNiigInk
> > > > > soVGgu1E5EetqvE%2FA%3D&reserved=0
> > > > >
> > > > > I think the problem you are hitting with lookaside-proto is that
> > > > > for it we use 2 different values here:
> > > > > a) In create_sec_session we use portid (it also should be
> > > > > rte_cryptodev_get_sec_ctx() here)
> > > > > if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > > > ctx = (struct rte_security_ctx *)
> > > > >
> > > > > rte_eth_dev_get_sec_ctx(sa->portid);
> > > > It should be rte_cryptodev_get_sec_ctx in the first place. And it
> > > > needs a
> > > cdev_id as input based on the cdev mapping done while initializing
> > > > the cryptodev and neither the portid and nor cdev_id_qp.
> > > > > b) in enqueue() we use cdev_id_qp
> > > > >
> > > > > Right now these values could be different.
> > > > > As I understand we need to make sure that fro lookaside-proto
> > > > > cdev_id_qp
> > > ==
> > > > > portid provided by user, correct?
> > > > No it is not the case. Right now for lookaside there is no use of
> > > > portid in case
> > > of lookaside case.
> > > > As the cdev/qp/core mappings are managed internally and the user
> > > > cannot
> > > tweak it from cfg file.
> > > >
> > >
> > >
> > > Hmm, then at least that line is wrong here:
> > > https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdoc.
> > > dpdk
> > >
> > .org%2Fguides%2Fsample_app_ug%2Fipsec_secgw.html&data=02%7C01%
> > >
> > 7Cakhil.goyal%40nxp.com%7Cb1e931a9967943887a0c08d6c7f49d28%7C686ea
> > >
> > 1d3bc2b4c6fa92cd99c5c301635%7C0%7C0%7C636916250754452306&sda
> > >
> > ta=tNjHCO0O2rfh8shhQXu93qrM0Hr0OZVXUVhMcsg53dw%3D&reserved=
> > > 0
> > >
> > > sa out 5 cipher_algo aes-128-cbc cipher_key
> > > 0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0 \ auth_algo sha1-hmac auth_key
> > > 0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0 \ mode ipv4-tunnel src
> > > 172.16.1.5 dst 172.16.2.5 \ type lookaside-protocol-offload port_id 4
> > >
> > > And probably:
> > > "Port/device ID of the ethernet/crypto accelerator for which the SA is
> > > configured."
> > > Need to be rephrased to remove crypto accelerator notice.
> > Yes.
> >
> > >
> > > Another question - why you guys don't consider using portid for lookaside-
> > proto?
> > > As I can see add_mapping(function that fills cdev_id_qp) doesn't
> > > bother to check which rte_security protocols are supported (only
> > > crypto capabilities are checked).
> > > So not sure does current code will work ok with a mix of lookaside-
> > > none/lookaside-proto devices.
> > > Forcing user to specify crypto-dev id for lookaside-proto (via portid
> > > or so) will simplify things significantly.
> > I will think about it. Actually initially when portid was introduced, the intent was
> > same but it did not work well.
> > Because there may be a case of a cryptodev with multiple queues, so there will
> > be a mapping done internally in the application and matching it with the user
> > provided portid will be difficult.
> >
> > I believe that this could be done but would need some rework.
> >
> > >
> > > Konstantin
>
> Could we consider going back to the V2 version of this patch set which fixed the issue with the inline code and left the lookaside code
> unchanged?
Actually I have the same thoughts here:
considering the problems with lookaside-proto, it seems more pragmatic
to fix it as was suggested in V2:
split create_session() into 2 functions, and invoke create_session_inline() at startup,
while keeping create_session_lookaside() invocation at runtime.
At least it would fix the issue in a clean way.
If later ipsec-secgw will be reworked to allow lookaside session creation at init time,
we can merge them back.
Konstantin
>
> We are not in a position test any changes to the lookaside code.
>
> Regards,
>
> Bernard.
>
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-05-13 14:29 ` Ananyev, Konstantin
@ 2019-05-13 14:29 ` Ananyev, Konstantin
2019-05-27 8:58 ` Iremonger, Bernard
1 sibling, 0 replies; 25+ messages in thread
From: Ananyev, Konstantin @ 2019-05-13 14:29 UTC (permalink / raw)
To: Iremonger, Bernard, Akhil Goyal, dev; +Cc: stable
> > > > > > > Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st
> > > > > > > packet
> > > dropped
> > > > > for
> > > > > > > inline crypto
> > > > > > >
> > > > > > > Hi Bernard,
> > > > > > >
> > > > > > > > - RTE_LOG_DP(DEBUG, IPSEC, "Create session for SA spi %u on
> > > > > cryptodev "
> > > > > > > > - "%u qp %u\n", sa->spi,
> > > > > > > > - ipsec_ctx->tbl[cdev_id_qp].id,
> > > > > > > > - ipsec_ctx->tbl[cdev_id_qp].qp);
> > > > > > > > + if ((sa == NULL) || (pool == NULL))
> > > > > > > > + return -EINVAL;
> > > > > > > >
> > > > > > > > - if (sa->type != RTE_SECURITY_ACTION_TYPE_NONE) {
> > > > > > > > - struct rte_security_session_conf sess_conf = {
> > > > > > > > + struct rte_security_session_conf sess_conf = {
> > > > > > > > .action_type = sa->type,
> > > > > > > > .protocol = RTE_SECURITY_PROTOCOL_IPSEC,
> > > > > > > > {.ipsec = { @@ -90,247 +65,340 @@
> > > > > > > > create_session(struct ipsec_ctx *ipsec_ctx,
> > > > > struct
> > > > > > > > ipsec_sa *sa)
> > > > > > > > } },
> > > > > > > > .crypto_xform = sa->xforms,
> > > > > > > > .userdata = NULL,
> > > > > > > > -
> > > > > > > > };
> > > > > > > >
> > > > > > > > - if (sa->type ==
> > > > > > > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL)
> > > > > > > > {
> > > > > > > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > > > > - rte_cryptodev_get_sec_ctx(
> > > > > > > > - ipsec_ctx->tbl[cdev_id_qp].id);
> > > > > > > > -
> > > > > > > > - /* Set IPsec parameters in conf */
> > > > > > > > - set_ipsec_conf(sa, &(sess_conf.ipsec));
> > > > > > > > -
> > > > > > > > - sa->sec_session = rte_security_session_create(ctx,
> > > > > > > > - &sess_conf, ipsec_ctx->session_pool);
> > > > > > > > - if (sa->sec_session == NULL) {
> > > > > > > > - RTE_LOG(ERR, IPSEC,
> > > > > > > > - "SEC Session init failed: err: %d\n", ret);
> > > > > > > > - return -1;
> > > > > > > > - }
> > > > > > > > - } else if (sa->type ==
> > > > > RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO)
> > > > > > > {
> > > > > > > > - struct rte_flow_error err;
> > > > > > > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > > > > - rte_eth_dev_get_sec_ctx(
> > > > > > > > - sa->portid);
> > > > > > > > - const struct rte_security_capability *sec_cap;
> > > > > > > > - int ret = 0;
> > > > > > > > -
> > > > > > > > - sa->sec_session = rte_security_session_create(ctx,
> > > > > > > > - &sess_conf, ipsec_ctx->session_pool);
> > > > > > > > - if (sa->sec_session == NULL) {
> > > > > > > > - RTE_LOG(ERR, IPSEC,
> > > > > > > > - "SEC Session init failed: err: %d\n", ret);
> > > > > > > > - return -1;
> > > > > > > > - }
> > > > > > > > + if (sa->type ==
> > > > > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > > > > > > + ctx = (struct rte_security_ctx *)
> > > > > > > > +
> > > > > > > > + rte_eth_dev_get_sec_ctx(sa->portid);
> > > > > > >
> > > > > > > This is breaking the lookaside mode. Ctx was retrieved using
> > > > > > > the
> > > ipsec_ctx-
> > > > > >tbl
> > > > > > > struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > > > rte_cryptodev_get_sec_ctx(
> > > > > > > ipsec_ctx->tbl[cdev_id_qp].id);
> > > > > > >
> > > > > > > I am looking into it, but I don't have time left to get it integrated in RC2.
> > > So
> > > > > this
> > > > > > > has to be pushed to RC3
> > > > > >
> > > > > > It looks like there are multiple issues in this patch wrt
> > > > > > lookaside and none
> > > cases.
> > > > > Only the inline cases seem to be working.
> > > > > >
> > > > > > 1. the patch removes the cdev_mapping concept completely.
> > > > > > Cdev_id_qp is
> > > > > not getting used.
> > > > >
> > > > > Not exactly.
> > > > > cdev_id_qp is still setup, and is still used to decide to which
> > > > > crypto-dev to enqueuer the crypto-op:
> > > > > ipsec_enqueue(...)
> > > > > {
> > > > > ...
> > > > > enqueue_cop(&ipsec_ctx->tbl[sa->cdev_id_qp], &priv->cop);
> > > >
> > > > I don't see anybody filling "sa->cdev_id_qp". Please let me know if
> > > > I have
> > > missed it somewhere.
> > > > It is memset to 0 I guess.
> > >
> > >
> > > Yep, true we lost it somewhere during the rework.
> > >
> > > >
> > > > >
> > > > >
> > > > > Same in ipsec_process().
> > > > >
> > > > > For initialization, yes cdev_id_qp is not used anymore.
> > > > > As discussed here:
> > > > >
> > > https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmail
> > > s.dp
> > > > > dk.org%2Farchives%2Fdev%2F2019-
> > > > >
> > >
> > March%2F127725.html&data=02%7C01%7Cakhil.goyal%40nxp.com%7C04
> > > > >
> > >
> > 194193cfc04c0b629008d6c7eea247%7C686ea1d3bc2b4c6fa92cd99c5c301635%
> > > > >
> > >
> > 7C0%7C0%7C636916225072561313&sdata=ga9IiqhYRWOz9QkRDIXNiigInk
> > > > > soVGgu1E5EetqvE%2FA%3D&reserved=0
> > > > >
> > > > > I think the problem you are hitting with lookaside-proto is that
> > > > > for it we use 2 different values here:
> > > > > a) In create_sec_session we use portid (it also should be
> > > > > rte_cryptodev_get_sec_ctx() here)
> > > > > if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > > > ctx = (struct rte_security_ctx *)
> > > > >
> > > > > rte_eth_dev_get_sec_ctx(sa->portid);
> > > > It should be rte_cryptodev_get_sec_ctx in the first place. And it
> > > > needs a
> > > cdev_id as input based on the cdev mapping done while initializing
> > > > the cryptodev and neither the portid and nor cdev_id_qp.
> > > > > b) in enqueue() we use cdev_id_qp
> > > > >
> > > > > Right now these values could be different.
> > > > > As I understand we need to make sure that fro lookaside-proto
> > > > > cdev_id_qp
> > > ==
> > > > > portid provided by user, correct?
> > > > No it is not the case. Right now for lookaside there is no use of
> > > > portid in case
> > > of lookaside case.
> > > > As the cdev/qp/core mappings are managed internally and the user
> > > > cannot
> > > tweak it from cfg file.
> > > >
> > >
> > >
> > > Hmm, then at least that line is wrong here:
> > > https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdoc.
> > > dpdk
> > >
> > .org%2Fguides%2Fsample_app_ug%2Fipsec_secgw.html&data=02%7C01%
> > >
> > 7Cakhil.goyal%40nxp.com%7Cb1e931a9967943887a0c08d6c7f49d28%7C686ea
> > >
> > 1d3bc2b4c6fa92cd99c5c301635%7C0%7C0%7C636916250754452306&sda
> > >
> > ta=tNjHCO0O2rfh8shhQXu93qrM0Hr0OZVXUVhMcsg53dw%3D&reserved=
> > > 0
> > >
> > > sa out 5 cipher_algo aes-128-cbc cipher_key
> > > 0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0 \ auth_algo sha1-hmac auth_key
> > > 0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0 \ mode ipv4-tunnel src
> > > 172.16.1.5 dst 172.16.2.5 \ type lookaside-protocol-offload port_id 4
> > >
> > > And probably:
> > > "Port/device ID of the ethernet/crypto accelerator for which the SA is
> > > configured."
> > > Need to be rephrased to remove crypto accelerator notice.
> > Yes.
> >
> > >
> > > Another question - why you guys don't consider using portid for lookaside-
> > proto?
> > > As I can see add_mapping(function that fills cdev_id_qp) doesn't
> > > bother to check which rte_security protocols are supported (only
> > > crypto capabilities are checked).
> > > So not sure does current code will work ok with a mix of lookaside-
> > > none/lookaside-proto devices.
> > > Forcing user to specify crypto-dev id for lookaside-proto (via portid
> > > or so) will simplify things significantly.
> > I will think about it. Actually initially when portid was introduced, the intent was
> > same but it did not work well.
> > Because there may be a case of a cryptodev with multiple queues, so there will
> > be a mapping done internally in the application and matching it with the user
> > provided portid will be difficult.
> >
> > I believe that this could be done but would need some rework.
> >
> > >
> > > Konstantin
>
> Could we consider going back to the V2 version of this patch set which fixed the issue with the inline code and left the lookaside code
> unchanged?
Actually I have the same thoughts here:
considering the problems with lookaside-proto, it seems more pragmatic
to fix it as was suggested in V2:
split create_session() into 2 functions, and invoke create_session_inline() at startup,
while keeping create_session_lookaside() invocation at runtime.
At least it would fix the issue in a clean way.
If later ipsec-secgw will be reworked to allow lookaside session creation at init time,
we can merge them back.
Konstantin
>
> We are not in a position test any changes to the lookaside code.
>
> Regards,
>
> Bernard.
>
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-04-24 6:34 ` Akhil Goyal
2019-04-24 6:34 ` Akhil Goyal
@ 2019-04-24 10:40 ` Iremonger, Bernard
2019-04-24 10:40 ` Iremonger, Bernard
2019-05-13 14:29 ` Ananyev, Konstantin
1 sibling, 2 replies; 25+ messages in thread
From: Iremonger, Bernard @ 2019-04-24 10:40 UTC (permalink / raw)
To: Akhil Goyal, Ananyev, Konstantin, dev; +Cc: stable
Hi Akhil
<snip>
> > > > > > Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st
> > > > > > packet
> > dropped
> > > > for
> > > > > > inline crypto
> > > > > >
> > > > > > Hi Bernard,
> > > > > >
> > > > > > > - RTE_LOG_DP(DEBUG, IPSEC, "Create session for SA spi %u on
> > > > cryptodev "
> > > > > > > - "%u qp %u\n", sa->spi,
> > > > > > > - ipsec_ctx->tbl[cdev_id_qp].id,
> > > > > > > - ipsec_ctx->tbl[cdev_id_qp].qp);
> > > > > > > + if ((sa == NULL) || (pool == NULL))
> > > > > > > + return -EINVAL;
> > > > > > >
> > > > > > > - if (sa->type != RTE_SECURITY_ACTION_TYPE_NONE) {
> > > > > > > - struct rte_security_session_conf sess_conf = {
> > > > > > > + struct rte_security_session_conf sess_conf = {
> > > > > > > .action_type = sa->type,
> > > > > > > .protocol = RTE_SECURITY_PROTOCOL_IPSEC,
> > > > > > > {.ipsec = { @@ -90,247 +65,340 @@
> > > > > > > create_session(struct ipsec_ctx *ipsec_ctx,
> > > > struct
> > > > > > > ipsec_sa *sa)
> > > > > > > } },
> > > > > > > .crypto_xform = sa->xforms,
> > > > > > > .userdata = NULL,
> > > > > > > -
> > > > > > > };
> > > > > > >
> > > > > > > - if (sa->type ==
> > > > > > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL)
> > > > > > > {
> > > > > > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > > > - rte_cryptodev_get_sec_ctx(
> > > > > > > - ipsec_ctx->tbl[cdev_id_qp].id);
> > > > > > > -
> > > > > > > - /* Set IPsec parameters in conf */
> > > > > > > - set_ipsec_conf(sa, &(sess_conf.ipsec));
> > > > > > > -
> > > > > > > - sa->sec_session = rte_security_session_create(ctx,
> > > > > > > - &sess_conf, ipsec_ctx->session_pool);
> > > > > > > - if (sa->sec_session == NULL) {
> > > > > > > - RTE_LOG(ERR, IPSEC,
> > > > > > > - "SEC Session init failed: err: %d\n", ret);
> > > > > > > - return -1;
> > > > > > > - }
> > > > > > > - } else if (sa->type ==
> > > > RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO)
> > > > > > {
> > > > > > > - struct rte_flow_error err;
> > > > > > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > > > - rte_eth_dev_get_sec_ctx(
> > > > > > > - sa->portid);
> > > > > > > - const struct rte_security_capability *sec_cap;
> > > > > > > - int ret = 0;
> > > > > > > -
> > > > > > > - sa->sec_session = rte_security_session_create(ctx,
> > > > > > > - &sess_conf, ipsec_ctx->session_pool);
> > > > > > > - if (sa->sec_session == NULL) {
> > > > > > > - RTE_LOG(ERR, IPSEC,
> > > > > > > - "SEC Session init failed: err: %d\n", ret);
> > > > > > > - return -1;
> > > > > > > - }
> > > > > > > + if (sa->type ==
> > > > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > > > > > + ctx = (struct rte_security_ctx *)
> > > > > > > +
> > > > > > > + rte_eth_dev_get_sec_ctx(sa->portid);
> > > > > >
> > > > > > This is breaking the lookaside mode. Ctx was retrieved using
> > > > > > the
> > ipsec_ctx-
> > > > >tbl
> > > > > > struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > > rte_cryptodev_get_sec_ctx(
> > > > > > ipsec_ctx->tbl[cdev_id_qp].id);
> > > > > >
> > > > > > I am looking into it, but I don't have time left to get it integrated in RC2.
> > So
> > > > this
> > > > > > has to be pushed to RC3
> > > > >
> > > > > It looks like there are multiple issues in this patch wrt
> > > > > lookaside and none
> > cases.
> > > > Only the inline cases seem to be working.
> > > > >
> > > > > 1. the patch removes the cdev_mapping concept completely.
> > > > > Cdev_id_qp is
> > > > not getting used.
> > > >
> > > > Not exactly.
> > > > cdev_id_qp is still setup, and is still used to decide to which
> > > > crypto-dev to enqueuer the crypto-op:
> > > > ipsec_enqueue(...)
> > > > {
> > > > ...
> > > > enqueue_cop(&ipsec_ctx->tbl[sa->cdev_id_qp], &priv->cop);
> > >
> > > I don't see anybody filling "sa->cdev_id_qp". Please let me know if
> > > I have
> > missed it somewhere.
> > > It is memset to 0 I guess.
> >
> >
> > Yep, true we lost it somewhere during the rework.
> >
> > >
> > > >
> > > >
> > > > Same in ipsec_process().
> > > >
> > > > For initialization, yes cdev_id_qp is not used anymore.
> > > > As discussed here:
> > > >
> > https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmail
> > s.dp
> > > > dk.org%2Farchives%2Fdev%2F2019-
> > > >
> >
> March%2F127725.html&data=02%7C01%7Cakhil.goyal%40nxp.com%7C04
> > > >
> >
> 194193cfc04c0b629008d6c7eea247%7C686ea1d3bc2b4c6fa92cd99c5c301635%
> > > >
> >
> 7C0%7C0%7C636916225072561313&sdata=ga9IiqhYRWOz9QkRDIXNiigInk
> > > > soVGgu1E5EetqvE%2FA%3D&reserved=0
> > > >
> > > > I think the problem you are hitting with lookaside-proto is that
> > > > for it we use 2 different values here:
> > > > a) In create_sec_session we use portid (it also should be
> > > > rte_cryptodev_get_sec_ctx() here)
> > > > if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > > ctx = (struct rte_security_ctx *)
> > > >
> > > > rte_eth_dev_get_sec_ctx(sa->portid);
> > > It should be rte_cryptodev_get_sec_ctx in the first place. And it
> > > needs a
> > cdev_id as input based on the cdev mapping done while initializing
> > > the cryptodev and neither the portid and nor cdev_id_qp.
> > > > b) in enqueue() we use cdev_id_qp
> > > >
> > > > Right now these values could be different.
> > > > As I understand we need to make sure that fro lookaside-proto
> > > > cdev_id_qp
> > ==
> > > > portid provided by user, correct?
> > > No it is not the case. Right now for lookaside there is no use of
> > > portid in case
> > of lookaside case.
> > > As the cdev/qp/core mappings are managed internally and the user
> > > cannot
> > tweak it from cfg file.
> > >
> >
> >
> > Hmm, then at least that line is wrong here:
> > https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdoc.
> > dpdk
> >
> .org%2Fguides%2Fsample_app_ug%2Fipsec_secgw.html&data=02%7C01%
> >
> 7Cakhil.goyal%40nxp.com%7Cb1e931a9967943887a0c08d6c7f49d28%7C686ea
> >
> 1d3bc2b4c6fa92cd99c5c301635%7C0%7C0%7C636916250754452306&sda
> >
> ta=tNjHCO0O2rfh8shhQXu93qrM0Hr0OZVXUVhMcsg53dw%3D&reserved=
> > 0
> >
> > sa out 5 cipher_algo aes-128-cbc cipher_key
> > 0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0 \ auth_algo sha1-hmac auth_key
> > 0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0 \ mode ipv4-tunnel src
> > 172.16.1.5 dst 172.16.2.5 \ type lookaside-protocol-offload port_id 4
> >
> > And probably:
> > "Port/device ID of the ethernet/crypto accelerator for which the SA is
> > configured."
> > Need to be rephrased to remove crypto accelerator notice.
> Yes.
>
> >
> > Another question - why you guys don't consider using portid for lookaside-
> proto?
> > As I can see add_mapping(function that fills cdev_id_qp) doesn't
> > bother to check which rte_security protocols are supported (only
> > crypto capabilities are checked).
> > So not sure does current code will work ok with a mix of lookaside-
> > none/lookaside-proto devices.
> > Forcing user to specify crypto-dev id for lookaside-proto (via portid
> > or so) will simplify things significantly.
> I will think about it. Actually initially when portid was introduced, the intent was
> same but it did not work well.
> Because there may be a case of a cryptodev with multiple queues, so there will
> be a mapping done internally in the application and matching it with the user
> provided portid will be difficult.
>
> I believe that this could be done but would need some rework.
>
> >
> > Konstantin
Could we consider going back to the V2 version of this patch set which fixed the issue with the inline code and left the lookaside code unchanged?
We are not in a position test any changes to the lookaside code.
Regards,
Bernard.
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-04-24 10:40 ` Iremonger, Bernard
@ 2019-04-24 10:40 ` Iremonger, Bernard
2019-05-13 14:29 ` Ananyev, Konstantin
1 sibling, 0 replies; 25+ messages in thread
From: Iremonger, Bernard @ 2019-04-24 10:40 UTC (permalink / raw)
To: Akhil Goyal, Ananyev, Konstantin, dev; +Cc: stable
Hi Akhil
<snip>
> > > > > > Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st
> > > > > > packet
> > dropped
> > > > for
> > > > > > inline crypto
> > > > > >
> > > > > > Hi Bernard,
> > > > > >
> > > > > > > - RTE_LOG_DP(DEBUG, IPSEC, "Create session for SA spi %u on
> > > > cryptodev "
> > > > > > > - "%u qp %u\n", sa->spi,
> > > > > > > - ipsec_ctx->tbl[cdev_id_qp].id,
> > > > > > > - ipsec_ctx->tbl[cdev_id_qp].qp);
> > > > > > > + if ((sa == NULL) || (pool == NULL))
> > > > > > > + return -EINVAL;
> > > > > > >
> > > > > > > - if (sa->type != RTE_SECURITY_ACTION_TYPE_NONE) {
> > > > > > > - struct rte_security_session_conf sess_conf = {
> > > > > > > + struct rte_security_session_conf sess_conf = {
> > > > > > > .action_type = sa->type,
> > > > > > > .protocol = RTE_SECURITY_PROTOCOL_IPSEC,
> > > > > > > {.ipsec = { @@ -90,247 +65,340 @@
> > > > > > > create_session(struct ipsec_ctx *ipsec_ctx,
> > > > struct
> > > > > > > ipsec_sa *sa)
> > > > > > > } },
> > > > > > > .crypto_xform = sa->xforms,
> > > > > > > .userdata = NULL,
> > > > > > > -
> > > > > > > };
> > > > > > >
> > > > > > > - if (sa->type ==
> > > > > > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL)
> > > > > > > {
> > > > > > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > > > - rte_cryptodev_get_sec_ctx(
> > > > > > > - ipsec_ctx->tbl[cdev_id_qp].id);
> > > > > > > -
> > > > > > > - /* Set IPsec parameters in conf */
> > > > > > > - set_ipsec_conf(sa, &(sess_conf.ipsec));
> > > > > > > -
> > > > > > > - sa->sec_session = rte_security_session_create(ctx,
> > > > > > > - &sess_conf, ipsec_ctx->session_pool);
> > > > > > > - if (sa->sec_session == NULL) {
> > > > > > > - RTE_LOG(ERR, IPSEC,
> > > > > > > - "SEC Session init failed: err: %d\n", ret);
> > > > > > > - return -1;
> > > > > > > - }
> > > > > > > - } else if (sa->type ==
> > > > RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO)
> > > > > > {
> > > > > > > - struct rte_flow_error err;
> > > > > > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > > > - rte_eth_dev_get_sec_ctx(
> > > > > > > - sa->portid);
> > > > > > > - const struct rte_security_capability *sec_cap;
> > > > > > > - int ret = 0;
> > > > > > > -
> > > > > > > - sa->sec_session = rte_security_session_create(ctx,
> > > > > > > - &sess_conf, ipsec_ctx->session_pool);
> > > > > > > - if (sa->sec_session == NULL) {
> > > > > > > - RTE_LOG(ERR, IPSEC,
> > > > > > > - "SEC Session init failed: err: %d\n", ret);
> > > > > > > - return -1;
> > > > > > > - }
> > > > > > > + if (sa->type ==
> > > > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > > > > > + ctx = (struct rte_security_ctx *)
> > > > > > > +
> > > > > > > + rte_eth_dev_get_sec_ctx(sa->portid);
> > > > > >
> > > > > > This is breaking the lookaside mode. Ctx was retrieved using
> > > > > > the
> > ipsec_ctx-
> > > > >tbl
> > > > > > struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > > rte_cryptodev_get_sec_ctx(
> > > > > > ipsec_ctx->tbl[cdev_id_qp].id);
> > > > > >
> > > > > > I am looking into it, but I don't have time left to get it integrated in RC2.
> > So
> > > > this
> > > > > > has to be pushed to RC3
> > > > >
> > > > > It looks like there are multiple issues in this patch wrt
> > > > > lookaside and none
> > cases.
> > > > Only the inline cases seem to be working.
> > > > >
> > > > > 1. the patch removes the cdev_mapping concept completely.
> > > > > Cdev_id_qp is
> > > > not getting used.
> > > >
> > > > Not exactly.
> > > > cdev_id_qp is still setup, and is still used to decide to which
> > > > crypto-dev to enqueuer the crypto-op:
> > > > ipsec_enqueue(...)
> > > > {
> > > > ...
> > > > enqueue_cop(&ipsec_ctx->tbl[sa->cdev_id_qp], &priv->cop);
> > >
> > > I don't see anybody filling "sa->cdev_id_qp". Please let me know if
> > > I have
> > missed it somewhere.
> > > It is memset to 0 I guess.
> >
> >
> > Yep, true we lost it somewhere during the rework.
> >
> > >
> > > >
> > > >
> > > > Same in ipsec_process().
> > > >
> > > > For initialization, yes cdev_id_qp is not used anymore.
> > > > As discussed here:
> > > >
> > https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmail
> > s.dp
> > > > dk.org%2Farchives%2Fdev%2F2019-
> > > >
> >
> March%2F127725.html&data=02%7C01%7Cakhil.goyal%40nxp.com%7C04
> > > >
> >
> 194193cfc04c0b629008d6c7eea247%7C686ea1d3bc2b4c6fa92cd99c5c301635%
> > > >
> >
> 7C0%7C0%7C636916225072561313&sdata=ga9IiqhYRWOz9QkRDIXNiigInk
> > > > soVGgu1E5EetqvE%2FA%3D&reserved=0
> > > >
> > > > I think the problem you are hitting with lookaside-proto is that
> > > > for it we use 2 different values here:
> > > > a) In create_sec_session we use portid (it also should be
> > > > rte_cryptodev_get_sec_ctx() here)
> > > > if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > > ctx = (struct rte_security_ctx *)
> > > >
> > > > rte_eth_dev_get_sec_ctx(sa->portid);
> > > It should be rte_cryptodev_get_sec_ctx in the first place. And it
> > > needs a
> > cdev_id as input based on the cdev mapping done while initializing
> > > the cryptodev and neither the portid and nor cdev_id_qp.
> > > > b) in enqueue() we use cdev_id_qp
> > > >
> > > > Right now these values could be different.
> > > > As I understand we need to make sure that fro lookaside-proto
> > > > cdev_id_qp
> > ==
> > > > portid provided by user, correct?
> > > No it is not the case. Right now for lookaside there is no use of
> > > portid in case
> > of lookaside case.
> > > As the cdev/qp/core mappings are managed internally and the user
> > > cannot
> > tweak it from cfg file.
> > >
> >
> >
> > Hmm, then at least that line is wrong here:
> > https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdoc.
> > dpdk
> >
> .org%2Fguides%2Fsample_app_ug%2Fipsec_secgw.html&data=02%7C01%
> >
> 7Cakhil.goyal%40nxp.com%7Cb1e931a9967943887a0c08d6c7f49d28%7C686ea
> >
> 1d3bc2b4c6fa92cd99c5c301635%7C0%7C0%7C636916250754452306&sda
> >
> ta=tNjHCO0O2rfh8shhQXu93qrM0Hr0OZVXUVhMcsg53dw%3D&reserved=
> > 0
> >
> > sa out 5 cipher_algo aes-128-cbc cipher_key
> > 0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0 \ auth_algo sha1-hmac auth_key
> > 0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0 \ mode ipv4-tunnel src
> > 172.16.1.5 dst 172.16.2.5 \ type lookaside-protocol-offload port_id 4
> >
> > And probably:
> > "Port/device ID of the ethernet/crypto accelerator for which the SA is
> > configured."
> > Need to be rephrased to remove crypto accelerator notice.
> Yes.
>
> >
> > Another question - why you guys don't consider using portid for lookaside-
> proto?
> > As I can see add_mapping(function that fills cdev_id_qp) doesn't
> > bother to check which rte_security protocols are supported (only
> > crypto capabilities are checked).
> > So not sure does current code will work ok with a mix of lookaside-
> > none/lookaside-proto devices.
> > Forcing user to specify crypto-dev id for lookaside-proto (via portid
> > or so) will simplify things significantly.
> I will think about it. Actually initially when portid was introduced, the intent was
> same but it did not work well.
> Because there may be a case of a cryptodev with multiple queues, so there will
> be a mapping done internally in the application and matching it with the user
> provided portid will be difficult.
>
> I believe that this could be done but would need some rework.
>
> >
> > Konstantin
Could we consider going back to the V2 version of this patch set which fixed the issue with the inline code and left the lookaside code unchanged?
We are not in a position test any changes to the lookaside code.
Regards,
Bernard.
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-04-23 14:04 ` Ananyev, Konstantin
2019-04-23 14:04 ` Ananyev, Konstantin
@ 2019-04-24 6:34 ` Akhil Goyal
2019-04-24 6:34 ` Akhil Goyal
2019-04-24 10:40 ` Iremonger, Bernard
1 sibling, 2 replies; 25+ messages in thread
From: Akhil Goyal @ 2019-04-24 6:34 UTC (permalink / raw)
To: Ananyev, Konstantin, Iremonger, Bernard, dev; +Cc: stable
> -----Original Message-----
> From: Ananyev, Konstantin <konstantin.ananyev@intel.com>
> Sent: Tuesday, April 23, 2019 7:35 PM
> To: Akhil Goyal <akhil.goyal@nxp.com>; Iremonger, Bernard
> <bernard.iremonger@intel.com>; dev@dpdk.org
> Cc: stable@dpdk.org
> Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for
> inline crypto
>
>
>
> > >
> > > >
> > > > > -----Original Message-----
> > > > > From: Akhil Goyal
> > > > > Sent: Thursday, April 18, 2019 7:21 PM
> > > > > To: Bernard Iremonger <bernard.iremonger@intel.com>; dev@dpdk.org;
> > > > > konstantin.ananyev@intel.com
> > > > > Cc: stable@dpdk.org
> > > > > Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet
> dropped
> > > for
> > > > > inline crypto
> > > > >
> > > > > Hi Bernard,
> > > > >
> > > > > > - RTE_LOG_DP(DEBUG, IPSEC, "Create session for SA spi %u on
> > > cryptodev "
> > > > > > - "%u qp %u\n", sa->spi,
> > > > > > - ipsec_ctx->tbl[cdev_id_qp].id,
> > > > > > - ipsec_ctx->tbl[cdev_id_qp].qp);
> > > > > > + if ((sa == NULL) || (pool == NULL))
> > > > > > + return -EINVAL;
> > > > > >
> > > > > > - if (sa->type != RTE_SECURITY_ACTION_TYPE_NONE) {
> > > > > > - struct rte_security_session_conf sess_conf = {
> > > > > > + struct rte_security_session_conf sess_conf = {
> > > > > > .action_type = sa->type,
> > > > > > .protocol = RTE_SECURITY_PROTOCOL_IPSEC,
> > > > > > {.ipsec = {
> > > > > > @@ -90,247 +65,340 @@ create_session(struct ipsec_ctx *ipsec_ctx,
> > > struct
> > > > > > ipsec_sa *sa)
> > > > > > } },
> > > > > > .crypto_xform = sa->xforms,
> > > > > > .userdata = NULL,
> > > > > > -
> > > > > > };
> > > > > >
> > > > > > - if (sa->type ==
> > > > > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL)
> > > > > > {
> > > > > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > > - rte_cryptodev_get_sec_ctx(
> > > > > > - ipsec_ctx->tbl[cdev_id_qp].id);
> > > > > > -
> > > > > > - /* Set IPsec parameters in conf */
> > > > > > - set_ipsec_conf(sa, &(sess_conf.ipsec));
> > > > > > -
> > > > > > - sa->sec_session = rte_security_session_create(ctx,
> > > > > > - &sess_conf, ipsec_ctx->session_pool);
> > > > > > - if (sa->sec_session == NULL) {
> > > > > > - RTE_LOG(ERR, IPSEC,
> > > > > > - "SEC Session init failed: err: %d\n", ret);
> > > > > > - return -1;
> > > > > > - }
> > > > > > - } else if (sa->type ==
> > > RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO)
> > > > > {
> > > > > > - struct rte_flow_error err;
> > > > > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > > - rte_eth_dev_get_sec_ctx(
> > > > > > - sa->portid);
> > > > > > - const struct rte_security_capability *sec_cap;
> > > > > > - int ret = 0;
> > > > > > -
> > > > > > - sa->sec_session = rte_security_session_create(ctx,
> > > > > > - &sess_conf, ipsec_ctx->session_pool);
> > > > > > - if (sa->sec_session == NULL) {
> > > > > > - RTE_LOG(ERR, IPSEC,
> > > > > > - "SEC Session init failed: err: %d\n", ret);
> > > > > > - return -1;
> > > > > > - }
> > > > > > + if (sa->type ==
> > > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > > > > + ctx = (struct rte_security_ctx *)
> > > > > > + rte_eth_dev_get_sec_ctx(sa->portid);
> > > > >
> > > > > This is breaking the lookaside mode. Ctx was retrieved using the
> ipsec_ctx-
> > > >tbl
> > > > > struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > rte_cryptodev_get_sec_ctx(
> > > > > ipsec_ctx->tbl[cdev_id_qp].id);
> > > > >
> > > > > I am looking into it, but I don't have time left to get it integrated in RC2.
> So
> > > this
> > > > > has to be pushed to RC3
> > > >
> > > > It looks like there are multiple issues in this patch wrt lookaside and none
> cases.
> > > Only the inline cases seem to be working.
> > > >
> > > > 1. the patch removes the cdev_mapping concept completely. Cdev_id_qp is
> > > not getting used.
> > >
> > > Not exactly.
> > > cdev_id_qp is still setup, and is still used to decide to which crypto-dev to
> > > enqueuer the crypto-op:
> > > ipsec_enqueue(...)
> > > {
> > > ...
> > > enqueue_cop(&ipsec_ctx->tbl[sa->cdev_id_qp], &priv->cop);
> >
> > I don't see anybody filling "sa->cdev_id_qp". Please let me know if I have
> missed it somewhere.
> > It is memset to 0 I guess.
>
>
> Yep, true we lost it somewhere during the rework.
>
> >
> > >
> > >
> > > Same in ipsec_process().
> > >
> > > For initialization, yes cdev_id_qp is not used anymore.
> > > As discussed here:
> > >
> https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmails.dp
> > > dk.org%2Farchives%2Fdev%2F2019-
> > >
> March%2F127725.html&data=02%7C01%7Cakhil.goyal%40nxp.com%7C04
> > >
> 194193cfc04c0b629008d6c7eea247%7C686ea1d3bc2b4c6fa92cd99c5c301635%
> > >
> 7C0%7C0%7C636916225072561313&sdata=ga9IiqhYRWOz9QkRDIXNiigInk
> > > soVGgu1E5EetqvE%2FA%3D&reserved=0
> > >
> > > I think the problem you are hitting with lookaside-proto is that for it
> > > we use 2 different values here:
> > > a) In create_sec_session we use portid (it also should be
> > > rte_cryptodev_get_sec_ctx() here)
> > > if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > ctx = (struct rte_security_ctx *)
> > > rte_eth_dev_get_sec_ctx(sa->portid);
> > It should be rte_cryptodev_get_sec_ctx in the first place. And it needs a
> cdev_id as input based on the cdev mapping done while initializing
> > the cryptodev and neither the portid and nor cdev_id_qp.
> > > b) in enqueue() we use cdev_id_qp
> > >
> > > Right now these values could be different.
> > > As I understand we need to make sure that fro lookaside-proto cdev_id_qp
> ==
> > > portid provided by user, correct?
> > No it is not the case. Right now for lookaside there is no use of portid in case
> of lookaside case.
> > As the cdev/qp/core mappings are managed internally and the user cannot
> tweak it from cfg file.
> >
>
>
> Hmm, then at least that line is wrong here:
> https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdoc.dpdk
> .org%2Fguides%2Fsample_app_ug%2Fipsec_secgw.html&data=02%7C01%
> 7Cakhil.goyal%40nxp.com%7Cb1e931a9967943887a0c08d6c7f49d28%7C686ea
> 1d3bc2b4c6fa92cd99c5c301635%7C0%7C0%7C636916250754452306&sda
> ta=tNjHCO0O2rfh8shhQXu93qrM0Hr0OZVXUVhMcsg53dw%3D&reserved=
> 0
>
> sa out 5 cipher_algo aes-128-cbc cipher_key 0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0 \
> auth_algo sha1-hmac auth_key 0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0 \
> mode ipv4-tunnel src 172.16.1.5 dst 172.16.2.5 \
> type lookaside-protocol-offload port_id 4
>
> And probably:
> "Port/device ID of the ethernet/crypto accelerator for which the SA is
> configured."
> Need to be rephrased to remove crypto accelerator notice.
Yes.
>
> Another question - why you guys don't consider using portid for lookaside-proto?
> As I can see add_mapping(function that fills cdev_id_qp) doesn't bother to
> check
> which rte_security protocols are supported (only crypto capabilities are checked).
> So not sure does current code will work ok with a mix of lookaside-
> none/lookaside-proto devices.
> Forcing user to specify crypto-dev id for lookaside-proto (via portid or so)
> will simplify things significantly.
I will think about it. Actually initially when portid was introduced, the intent was same but it did not work well.
Because there may be a case of a cryptodev with multiple queues, so there will be a mapping done internally in the application and matching it with the user provided portid will be difficult.
I believe that this could be done but would need some rework.
>
> Konstantin
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-04-24 6:34 ` Akhil Goyal
@ 2019-04-24 6:34 ` Akhil Goyal
2019-04-24 10:40 ` Iremonger, Bernard
1 sibling, 0 replies; 25+ messages in thread
From: Akhil Goyal @ 2019-04-24 6:34 UTC (permalink / raw)
To: Ananyev, Konstantin, Iremonger, Bernard, dev; +Cc: stable
> -----Original Message-----
> From: Ananyev, Konstantin <konstantin.ananyev@intel.com>
> Sent: Tuesday, April 23, 2019 7:35 PM
> To: Akhil Goyal <akhil.goyal@nxp.com>; Iremonger, Bernard
> <bernard.iremonger@intel.com>; dev@dpdk.org
> Cc: stable@dpdk.org
> Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for
> inline crypto
>
>
>
> > >
> > > >
> > > > > -----Original Message-----
> > > > > From: Akhil Goyal
> > > > > Sent: Thursday, April 18, 2019 7:21 PM
> > > > > To: Bernard Iremonger <bernard.iremonger@intel.com>; dev@dpdk.org;
> > > > > konstantin.ananyev@intel.com
> > > > > Cc: stable@dpdk.org
> > > > > Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet
> dropped
> > > for
> > > > > inline crypto
> > > > >
> > > > > Hi Bernard,
> > > > >
> > > > > > - RTE_LOG_DP(DEBUG, IPSEC, "Create session for SA spi %u on
> > > cryptodev "
> > > > > > - "%u qp %u\n", sa->spi,
> > > > > > - ipsec_ctx->tbl[cdev_id_qp].id,
> > > > > > - ipsec_ctx->tbl[cdev_id_qp].qp);
> > > > > > + if ((sa == NULL) || (pool == NULL))
> > > > > > + return -EINVAL;
> > > > > >
> > > > > > - if (sa->type != RTE_SECURITY_ACTION_TYPE_NONE) {
> > > > > > - struct rte_security_session_conf sess_conf = {
> > > > > > + struct rte_security_session_conf sess_conf = {
> > > > > > .action_type = sa->type,
> > > > > > .protocol = RTE_SECURITY_PROTOCOL_IPSEC,
> > > > > > {.ipsec = {
> > > > > > @@ -90,247 +65,340 @@ create_session(struct ipsec_ctx *ipsec_ctx,
> > > struct
> > > > > > ipsec_sa *sa)
> > > > > > } },
> > > > > > .crypto_xform = sa->xforms,
> > > > > > .userdata = NULL,
> > > > > > -
> > > > > > };
> > > > > >
> > > > > > - if (sa->type ==
> > > > > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL)
> > > > > > {
> > > > > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > > - rte_cryptodev_get_sec_ctx(
> > > > > > - ipsec_ctx->tbl[cdev_id_qp].id);
> > > > > > -
> > > > > > - /* Set IPsec parameters in conf */
> > > > > > - set_ipsec_conf(sa, &(sess_conf.ipsec));
> > > > > > -
> > > > > > - sa->sec_session = rte_security_session_create(ctx,
> > > > > > - &sess_conf, ipsec_ctx->session_pool);
> > > > > > - if (sa->sec_session == NULL) {
> > > > > > - RTE_LOG(ERR, IPSEC,
> > > > > > - "SEC Session init failed: err: %d\n", ret);
> > > > > > - return -1;
> > > > > > - }
> > > > > > - } else if (sa->type ==
> > > RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO)
> > > > > {
> > > > > > - struct rte_flow_error err;
> > > > > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > > - rte_eth_dev_get_sec_ctx(
> > > > > > - sa->portid);
> > > > > > - const struct rte_security_capability *sec_cap;
> > > > > > - int ret = 0;
> > > > > > -
> > > > > > - sa->sec_session = rte_security_session_create(ctx,
> > > > > > - &sess_conf, ipsec_ctx->session_pool);
> > > > > > - if (sa->sec_session == NULL) {
> > > > > > - RTE_LOG(ERR, IPSEC,
> > > > > > - "SEC Session init failed: err: %d\n", ret);
> > > > > > - return -1;
> > > > > > - }
> > > > > > + if (sa->type ==
> > > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > > > > + ctx = (struct rte_security_ctx *)
> > > > > > + rte_eth_dev_get_sec_ctx(sa->portid);
> > > > >
> > > > > This is breaking the lookaside mode. Ctx was retrieved using the
> ipsec_ctx-
> > > >tbl
> > > > > struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > rte_cryptodev_get_sec_ctx(
> > > > > ipsec_ctx->tbl[cdev_id_qp].id);
> > > > >
> > > > > I am looking into it, but I don't have time left to get it integrated in RC2.
> So
> > > this
> > > > > has to be pushed to RC3
> > > >
> > > > It looks like there are multiple issues in this patch wrt lookaside and none
> cases.
> > > Only the inline cases seem to be working.
> > > >
> > > > 1. the patch removes the cdev_mapping concept completely. Cdev_id_qp is
> > > not getting used.
> > >
> > > Not exactly.
> > > cdev_id_qp is still setup, and is still used to decide to which crypto-dev to
> > > enqueuer the crypto-op:
> > > ipsec_enqueue(...)
> > > {
> > > ...
> > > enqueue_cop(&ipsec_ctx->tbl[sa->cdev_id_qp], &priv->cop);
> >
> > I don't see anybody filling "sa->cdev_id_qp". Please let me know if I have
> missed it somewhere.
> > It is memset to 0 I guess.
>
>
> Yep, true we lost it somewhere during the rework.
>
> >
> > >
> > >
> > > Same in ipsec_process().
> > >
> > > For initialization, yes cdev_id_qp is not used anymore.
> > > As discussed here:
> > >
> https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmails.dp
> > > dk.org%2Farchives%2Fdev%2F2019-
> > >
> March%2F127725.html&data=02%7C01%7Cakhil.goyal%40nxp.com%7C04
> > >
> 194193cfc04c0b629008d6c7eea247%7C686ea1d3bc2b4c6fa92cd99c5c301635%
> > >
> 7C0%7C0%7C636916225072561313&sdata=ga9IiqhYRWOz9QkRDIXNiigInk
> > > soVGgu1E5EetqvE%2FA%3D&reserved=0
> > >
> > > I think the problem you are hitting with lookaside-proto is that for it
> > > we use 2 different values here:
> > > a) In create_sec_session we use portid (it also should be
> > > rte_cryptodev_get_sec_ctx() here)
> > > if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > ctx = (struct rte_security_ctx *)
> > > rte_eth_dev_get_sec_ctx(sa->portid);
> > It should be rte_cryptodev_get_sec_ctx in the first place. And it needs a
> cdev_id as input based on the cdev mapping done while initializing
> > the cryptodev and neither the portid and nor cdev_id_qp.
> > > b) in enqueue() we use cdev_id_qp
> > >
> > > Right now these values could be different.
> > > As I understand we need to make sure that fro lookaside-proto cdev_id_qp
> ==
> > > portid provided by user, correct?
> > No it is not the case. Right now for lookaside there is no use of portid in case
> of lookaside case.
> > As the cdev/qp/core mappings are managed internally and the user cannot
> tweak it from cfg file.
> >
>
>
> Hmm, then at least that line is wrong here:
> https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdoc.dpdk
> .org%2Fguides%2Fsample_app_ug%2Fipsec_secgw.html&data=02%7C01%
> 7Cakhil.goyal%40nxp.com%7Cb1e931a9967943887a0c08d6c7f49d28%7C686ea
> 1d3bc2b4c6fa92cd99c5c301635%7C0%7C0%7C636916250754452306&sda
> ta=tNjHCO0O2rfh8shhQXu93qrM0Hr0OZVXUVhMcsg53dw%3D&reserved=
> 0
>
> sa out 5 cipher_algo aes-128-cbc cipher_key 0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0 \
> auth_algo sha1-hmac auth_key 0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0 \
> mode ipv4-tunnel src 172.16.1.5 dst 172.16.2.5 \
> type lookaside-protocol-offload port_id 4
>
> And probably:
> "Port/device ID of the ethernet/crypto accelerator for which the SA is
> configured."
> Need to be rephrased to remove crypto accelerator notice.
Yes.
>
> Another question - why you guys don't consider using portid for lookaside-proto?
> As I can see add_mapping(function that fills cdev_id_qp) doesn't bother to
> check
> which rte_security protocols are supported (only crypto capabilities are checked).
> So not sure does current code will work ok with a mix of lookaside-
> none/lookaside-proto devices.
> Forcing user to specify crypto-dev id for lookaside-proto (via portid or so)
> will simplify things significantly.
I will think about it. Actually initially when portid was introduced, the intent was same but it did not work well.
Because there may be a case of a cryptodev with multiple queues, so there will be a mapping done internally in the application and matching it with the user provided portid will be difficult.
I believe that this could be done but would need some rework.
>
> Konstantin
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-04-23 13:32 ` Akhil Goyal
2019-04-23 13:32 ` Akhil Goyal
@ 2019-04-23 14:04 ` Ananyev, Konstantin
2019-04-23 14:04 ` Ananyev, Konstantin
2019-04-24 6:34 ` Akhil Goyal
1 sibling, 2 replies; 25+ messages in thread
From: Ananyev, Konstantin @ 2019-04-23 14:04 UTC (permalink / raw)
To: Akhil Goyal, Iremonger, Bernard, dev; +Cc: stable
> >
> > >
> > > > -----Original Message-----
> > > > From: Akhil Goyal
> > > > Sent: Thursday, April 18, 2019 7:21 PM
> > > > To: Bernard Iremonger <bernard.iremonger@intel.com>; dev@dpdk.org;
> > > > konstantin.ananyev@intel.com
> > > > Cc: stable@dpdk.org
> > > > Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped
> > for
> > > > inline crypto
> > > >
> > > > Hi Bernard,
> > > >
> > > > > - RTE_LOG_DP(DEBUG, IPSEC, "Create session for SA spi %u on
> > cryptodev "
> > > > > - "%u qp %u\n", sa->spi,
> > > > > - ipsec_ctx->tbl[cdev_id_qp].id,
> > > > > - ipsec_ctx->tbl[cdev_id_qp].qp);
> > > > > + if ((sa == NULL) || (pool == NULL))
> > > > > + return -EINVAL;
> > > > >
> > > > > - if (sa->type != RTE_SECURITY_ACTION_TYPE_NONE) {
> > > > > - struct rte_security_session_conf sess_conf = {
> > > > > + struct rte_security_session_conf sess_conf = {
> > > > > .action_type = sa->type,
> > > > > .protocol = RTE_SECURITY_PROTOCOL_IPSEC,
> > > > > {.ipsec = {
> > > > > @@ -90,247 +65,340 @@ create_session(struct ipsec_ctx *ipsec_ctx,
> > struct
> > > > > ipsec_sa *sa)
> > > > > } },
> > > > > .crypto_xform = sa->xforms,
> > > > > .userdata = NULL,
> > > > > -
> > > > > };
> > > > >
> > > > > - if (sa->type ==
> > > > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL)
> > > > > {
> > > > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > - rte_cryptodev_get_sec_ctx(
> > > > > - ipsec_ctx->tbl[cdev_id_qp].id);
> > > > > -
> > > > > - /* Set IPsec parameters in conf */
> > > > > - set_ipsec_conf(sa, &(sess_conf.ipsec));
> > > > > -
> > > > > - sa->sec_session = rte_security_session_create(ctx,
> > > > > - &sess_conf, ipsec_ctx->session_pool);
> > > > > - if (sa->sec_session == NULL) {
> > > > > - RTE_LOG(ERR, IPSEC,
> > > > > - "SEC Session init failed: err: %d\n", ret);
> > > > > - return -1;
> > > > > - }
> > > > > - } else if (sa->type ==
> > RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO)
> > > > {
> > > > > - struct rte_flow_error err;
> > > > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > - rte_eth_dev_get_sec_ctx(
> > > > > - sa->portid);
> > > > > - const struct rte_security_capability *sec_cap;
> > > > > - int ret = 0;
> > > > > -
> > > > > - sa->sec_session = rte_security_session_create(ctx,
> > > > > - &sess_conf, ipsec_ctx->session_pool);
> > > > > - if (sa->sec_session == NULL) {
> > > > > - RTE_LOG(ERR, IPSEC,
> > > > > - "SEC Session init failed: err: %d\n", ret);
> > > > > - return -1;
> > > > > - }
> > > > > + if (sa->type ==
> > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > > > + ctx = (struct rte_security_ctx *)
> > > > > + rte_eth_dev_get_sec_ctx(sa->portid);
> > > >
> > > > This is breaking the lookaside mode. Ctx was retrieved using the ipsec_ctx-
> > >tbl
> > > > struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > rte_cryptodev_get_sec_ctx(
> > > > ipsec_ctx->tbl[cdev_id_qp].id);
> > > >
> > > > I am looking into it, but I don't have time left to get it integrated in RC2. So
> > this
> > > > has to be pushed to RC3
> > >
> > > It looks like there are multiple issues in this patch wrt lookaside and none cases.
> > Only the inline cases seem to be working.
> > >
> > > 1. the patch removes the cdev_mapping concept completely. Cdev_id_qp is
> > not getting used.
> >
> > Not exactly.
> > cdev_id_qp is still setup, and is still used to decide to which crypto-dev to
> > enqueuer the crypto-op:
> > ipsec_enqueue(...)
> > {
> > ...
> > enqueue_cop(&ipsec_ctx->tbl[sa->cdev_id_qp], &priv->cop);
>
> I don't see anybody filling "sa->cdev_id_qp". Please let me know if I have missed it somewhere.
> It is memset to 0 I guess.
Yep, true we lost it somewhere during the rework.
>
> >
> >
> > Same in ipsec_process().
> >
> > For initialization, yes cdev_id_qp is not used anymore.
> > As discussed here:
> > https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmails.dp
> > dk.org%2Farchives%2Fdev%2F2019-
> > March%2F127725.html&data=02%7C01%7Cakhil.goyal%40nxp.com%7C04
> > 194193cfc04c0b629008d6c7eea247%7C686ea1d3bc2b4c6fa92cd99c5c301635%
> > 7C0%7C0%7C636916225072561313&sdata=ga9IiqhYRWOz9QkRDIXNiigInk
> > soVGgu1E5EetqvE%2FA%3D&reserved=0
> >
> > I think the problem you are hitting with lookaside-proto is that for it
> > we use 2 different values here:
> > a) In create_sec_session we use portid (it also should be
> > rte_cryptodev_get_sec_ctx() here)
> > if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > ctx = (struct rte_security_ctx *)
> > rte_eth_dev_get_sec_ctx(sa->portid);
> It should be rte_cryptodev_get_sec_ctx in the first place. And it needs a cdev_id as input based on the cdev mapping done while initializing
> the cryptodev and neither the portid and nor cdev_id_qp.
> > b) in enqueue() we use cdev_id_qp
> >
> > Right now these values could be different.
> > As I understand we need to make sure that fro lookaside-proto cdev_id_qp ==
> > portid provided by user, correct?
> No it is not the case. Right now for lookaside there is no use of portid in case of lookaside case.
> As the cdev/qp/core mappings are managed internally and the user cannot tweak it from cfg file.
>
Hmm, then at least that line is wrong here:
https://doc.dpdk.org/guides/sample_app_ug/ipsec_secgw.html
sa out 5 cipher_algo aes-128-cbc cipher_key 0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0 \
auth_algo sha1-hmac auth_key 0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0 \
mode ipv4-tunnel src 172.16.1.5 dst 172.16.2.5 \
type lookaside-protocol-offload port_id 4
And probably:
"Port/device ID of the ethernet/crypto accelerator for which the SA is configured."
Need to be rephrased to remove crypto accelerator notice.
Another question - why you guys don't consider using portid for lookaside-proto?
As I can see add_mapping(function that fills cdev_id_qp) doesn't bother to check
which rte_security protocols are supported (only crypto capabilities are checked).
So not sure does current code will work ok with a mix of lookaside-none/lookaside-proto devices.
Forcing user to specify crypto-dev id for lookaside-proto (via portid or so)
will simplify things significantly.
Konstantin
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-04-23 14:04 ` Ananyev, Konstantin
@ 2019-04-23 14:04 ` Ananyev, Konstantin
2019-04-24 6:34 ` Akhil Goyal
1 sibling, 0 replies; 25+ messages in thread
From: Ananyev, Konstantin @ 2019-04-23 14:04 UTC (permalink / raw)
To: Akhil Goyal, Iremonger, Bernard, dev; +Cc: stable
> >
> > >
> > > > -----Original Message-----
> > > > From: Akhil Goyal
> > > > Sent: Thursday, April 18, 2019 7:21 PM
> > > > To: Bernard Iremonger <bernard.iremonger@intel.com>; dev@dpdk.org;
> > > > konstantin.ananyev@intel.com
> > > > Cc: stable@dpdk.org
> > > > Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped
> > for
> > > > inline crypto
> > > >
> > > > Hi Bernard,
> > > >
> > > > > - RTE_LOG_DP(DEBUG, IPSEC, "Create session for SA spi %u on
> > cryptodev "
> > > > > - "%u qp %u\n", sa->spi,
> > > > > - ipsec_ctx->tbl[cdev_id_qp].id,
> > > > > - ipsec_ctx->tbl[cdev_id_qp].qp);
> > > > > + if ((sa == NULL) || (pool == NULL))
> > > > > + return -EINVAL;
> > > > >
> > > > > - if (sa->type != RTE_SECURITY_ACTION_TYPE_NONE) {
> > > > > - struct rte_security_session_conf sess_conf = {
> > > > > + struct rte_security_session_conf sess_conf = {
> > > > > .action_type = sa->type,
> > > > > .protocol = RTE_SECURITY_PROTOCOL_IPSEC,
> > > > > {.ipsec = {
> > > > > @@ -90,247 +65,340 @@ create_session(struct ipsec_ctx *ipsec_ctx,
> > struct
> > > > > ipsec_sa *sa)
> > > > > } },
> > > > > .crypto_xform = sa->xforms,
> > > > > .userdata = NULL,
> > > > > -
> > > > > };
> > > > >
> > > > > - if (sa->type ==
> > > > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL)
> > > > > {
> > > > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > - rte_cryptodev_get_sec_ctx(
> > > > > - ipsec_ctx->tbl[cdev_id_qp].id);
> > > > > -
> > > > > - /* Set IPsec parameters in conf */
> > > > > - set_ipsec_conf(sa, &(sess_conf.ipsec));
> > > > > -
> > > > > - sa->sec_session = rte_security_session_create(ctx,
> > > > > - &sess_conf, ipsec_ctx->session_pool);
> > > > > - if (sa->sec_session == NULL) {
> > > > > - RTE_LOG(ERR, IPSEC,
> > > > > - "SEC Session init failed: err: %d\n", ret);
> > > > > - return -1;
> > > > > - }
> > > > > - } else if (sa->type ==
> > RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO)
> > > > {
> > > > > - struct rte_flow_error err;
> > > > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > > - rte_eth_dev_get_sec_ctx(
> > > > > - sa->portid);
> > > > > - const struct rte_security_capability *sec_cap;
> > > > > - int ret = 0;
> > > > > -
> > > > > - sa->sec_session = rte_security_session_create(ctx,
> > > > > - &sess_conf, ipsec_ctx->session_pool);
> > > > > - if (sa->sec_session == NULL) {
> > > > > - RTE_LOG(ERR, IPSEC,
> > > > > - "SEC Session init failed: err: %d\n", ret);
> > > > > - return -1;
> > > > > - }
> > > > > + if (sa->type ==
> > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > > > + ctx = (struct rte_security_ctx *)
> > > > > + rte_eth_dev_get_sec_ctx(sa->portid);
> > > >
> > > > This is breaking the lookaside mode. Ctx was retrieved using the ipsec_ctx-
> > >tbl
> > > > struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > rte_cryptodev_get_sec_ctx(
> > > > ipsec_ctx->tbl[cdev_id_qp].id);
> > > >
> > > > I am looking into it, but I don't have time left to get it integrated in RC2. So
> > this
> > > > has to be pushed to RC3
> > >
> > > It looks like there are multiple issues in this patch wrt lookaside and none cases.
> > Only the inline cases seem to be working.
> > >
> > > 1. the patch removes the cdev_mapping concept completely. Cdev_id_qp is
> > not getting used.
> >
> > Not exactly.
> > cdev_id_qp is still setup, and is still used to decide to which crypto-dev to
> > enqueuer the crypto-op:
> > ipsec_enqueue(...)
> > {
> > ...
> > enqueue_cop(&ipsec_ctx->tbl[sa->cdev_id_qp], &priv->cop);
>
> I don't see anybody filling "sa->cdev_id_qp". Please let me know if I have missed it somewhere.
> It is memset to 0 I guess.
Yep, true we lost it somewhere during the rework.
>
> >
> >
> > Same in ipsec_process().
> >
> > For initialization, yes cdev_id_qp is not used anymore.
> > As discussed here:
> > https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmails.dp
> > dk.org%2Farchives%2Fdev%2F2019-
> > March%2F127725.html&data=02%7C01%7Cakhil.goyal%40nxp.com%7C04
> > 194193cfc04c0b629008d6c7eea247%7C686ea1d3bc2b4c6fa92cd99c5c301635%
> > 7C0%7C0%7C636916225072561313&sdata=ga9IiqhYRWOz9QkRDIXNiigInk
> > soVGgu1E5EetqvE%2FA%3D&reserved=0
> >
> > I think the problem you are hitting with lookaside-proto is that for it
> > we use 2 different values here:
> > a) In create_sec_session we use portid (it also should be
> > rte_cryptodev_get_sec_ctx() here)
> > if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > ctx = (struct rte_security_ctx *)
> > rte_eth_dev_get_sec_ctx(sa->portid);
> It should be rte_cryptodev_get_sec_ctx in the first place. And it needs a cdev_id as input based on the cdev mapping done while initializing
> the cryptodev and neither the portid and nor cdev_id_qp.
> > b) in enqueue() we use cdev_id_qp
> >
> > Right now these values could be different.
> > As I understand we need to make sure that fro lookaside-proto cdev_id_qp ==
> > portid provided by user, correct?
> No it is not the case. Right now for lookaside there is no use of portid in case of lookaside case.
> As the cdev/qp/core mappings are managed internally and the user cannot tweak it from cfg file.
>
Hmm, then at least that line is wrong here:
https://doc.dpdk.org/guides/sample_app_ug/ipsec_secgw.html
sa out 5 cipher_algo aes-128-cbc cipher_key 0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0 \
auth_algo sha1-hmac auth_key 0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0:0 \
mode ipv4-tunnel src 172.16.1.5 dst 172.16.2.5 \
type lookaside-protocol-offload port_id 4
And probably:
"Port/device ID of the ethernet/crypto accelerator for which the SA is configured."
Need to be rephrased to remove crypto accelerator notice.
Another question - why you guys don't consider using portid for lookaside-proto?
As I can see add_mapping(function that fills cdev_id_qp) doesn't bother to check
which rte_security protocols are supported (only crypto capabilities are checked).
So not sure does current code will work ok with a mix of lookaside-none/lookaside-proto devices.
Forcing user to specify crypto-dev id for lookaside-proto (via portid or so)
will simplify things significantly.
Konstantin
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-04-23 13:21 ` Ananyev, Konstantin
2019-04-23 13:21 ` Ananyev, Konstantin
@ 2019-04-23 13:32 ` Akhil Goyal
2019-04-23 13:32 ` Akhil Goyal
2019-04-23 14:04 ` Ananyev, Konstantin
1 sibling, 2 replies; 25+ messages in thread
From: Akhil Goyal @ 2019-04-23 13:32 UTC (permalink / raw)
To: Ananyev, Konstantin, Iremonger, Bernard, dev; +Cc: stable
Hi Konstantin,
> Hi Akhil,
>
> >
> > > -----Original Message-----
> > > From: Akhil Goyal
> > > Sent: Thursday, April 18, 2019 7:21 PM
> > > To: Bernard Iremonger <bernard.iremonger@intel.com>; dev@dpdk.org;
> > > konstantin.ananyev@intel.com
> > > Cc: stable@dpdk.org
> > > Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped
> for
> > > inline crypto
> > >
> > > Hi Bernard,
> > >
> > > > - RTE_LOG_DP(DEBUG, IPSEC, "Create session for SA spi %u on
> cryptodev "
> > > > - "%u qp %u\n", sa->spi,
> > > > - ipsec_ctx->tbl[cdev_id_qp].id,
> > > > - ipsec_ctx->tbl[cdev_id_qp].qp);
> > > > + if ((sa == NULL) || (pool == NULL))
> > > > + return -EINVAL;
> > > >
> > > > - if (sa->type != RTE_SECURITY_ACTION_TYPE_NONE) {
> > > > - struct rte_security_session_conf sess_conf = {
> > > > + struct rte_security_session_conf sess_conf = {
> > > > .action_type = sa->type,
> > > > .protocol = RTE_SECURITY_PROTOCOL_IPSEC,
> > > > {.ipsec = {
> > > > @@ -90,247 +65,340 @@ create_session(struct ipsec_ctx *ipsec_ctx,
> struct
> > > > ipsec_sa *sa)
> > > > } },
> > > > .crypto_xform = sa->xforms,
> > > > .userdata = NULL,
> > > > -
> > > > };
> > > >
> > > > - if (sa->type ==
> > > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL)
> > > > {
> > > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > - rte_cryptodev_get_sec_ctx(
> > > > - ipsec_ctx->tbl[cdev_id_qp].id);
> > > > -
> > > > - /* Set IPsec parameters in conf */
> > > > - set_ipsec_conf(sa, &(sess_conf.ipsec));
> > > > -
> > > > - sa->sec_session = rte_security_session_create(ctx,
> > > > - &sess_conf, ipsec_ctx->session_pool);
> > > > - if (sa->sec_session == NULL) {
> > > > - RTE_LOG(ERR, IPSEC,
> > > > - "SEC Session init failed: err: %d\n", ret);
> > > > - return -1;
> > > > - }
> > > > - } else if (sa->type ==
> RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO)
> > > {
> > > > - struct rte_flow_error err;
> > > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > - rte_eth_dev_get_sec_ctx(
> > > > - sa->portid);
> > > > - const struct rte_security_capability *sec_cap;
> > > > - int ret = 0;
> > > > -
> > > > - sa->sec_session = rte_security_session_create(ctx,
> > > > - &sess_conf, ipsec_ctx->session_pool);
> > > > - if (sa->sec_session == NULL) {
> > > > - RTE_LOG(ERR, IPSEC,
> > > > - "SEC Session init failed: err: %d\n", ret);
> > > > - return -1;
> > > > - }
> > > > + if (sa->type ==
> RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > > + ctx = (struct rte_security_ctx *)
> > > > + rte_eth_dev_get_sec_ctx(sa->portid);
> > >
> > > This is breaking the lookaside mode. Ctx was retrieved using the ipsec_ctx-
> >tbl
> > > struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > rte_cryptodev_get_sec_ctx(
> > > ipsec_ctx->tbl[cdev_id_qp].id);
> > >
> > > I am looking into it, but I don't have time left to get it integrated in RC2. So
> this
> > > has to be pushed to RC3
> >
> > It looks like there are multiple issues in this patch wrt lookaside and none cases.
> Only the inline cases seem to be working.
> >
> > 1. the patch removes the cdev_mapping concept completely. Cdev_id_qp is
> not getting used.
>
> Not exactly.
> cdev_id_qp is still setup, and is still used to decide to which crypto-dev to
> enqueuer the crypto-op:
> ipsec_enqueue(...)
> {
> ...
> enqueue_cop(&ipsec_ctx->tbl[sa->cdev_id_qp], &priv->cop);
I don't see anybody filling "sa->cdev_id_qp". Please let me know if I have missed it somewhere.
It is memset to 0 I guess.
>
>
> Same in ipsec_process().
>
> For initialization, yes cdev_id_qp is not used anymore.
> As discussed here:
> https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmails.dp
> dk.org%2Farchives%2Fdev%2F2019-
> March%2F127725.html&data=02%7C01%7Cakhil.goyal%40nxp.com%7C04
> 194193cfc04c0b629008d6c7eea247%7C686ea1d3bc2b4c6fa92cd99c5c301635%
> 7C0%7C0%7C636916225072561313&sdata=ga9IiqhYRWOz9QkRDIXNiigInk
> soVGgu1E5EetqvE%2FA%3D&reserved=0
>
> I think the problem you are hitting with lookaside-proto is that for it
> we use 2 different values here:
> a) In create_sec_session we use portid (it also should be
> rte_cryptodev_get_sec_ctx() here)
> if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> ctx = (struct rte_security_ctx *)
> rte_eth_dev_get_sec_ctx(sa->portid);
It should be rte_cryptodev_get_sec_ctx in the first place. And it needs a cdev_id as input based on the cdev mapping done while initializing the cryptodev and neither the portid and nor cdev_id_qp.
> b) in enqueue() we use cdev_id_qp
>
> Right now these values could be different.
> As I understand we need to make sure that fro lookaside-proto cdev_id_qp ==
> portid provided by user, correct?
No it is not the case. Right now for lookaside there is no use of portid in case of lookaside case.
As the cdev/qp/core mappings are managed internally and the user cannot tweak it from cfg file.
>
>
> > The port_id cannot be used in case of crypto, the mapping of cdev/qp/core
> is done differently for inbound and outbound ports which is
> > missed in this patch.
> >
> > 2. crypto sessions are created using the session mempool and the private data
> is allocated using the session priv_mempool which is
> > removed in this patch. This will break cases where the priv data is more than
> the size of sess_mp element size.
> > Also the security sessions need to be allocated using the session_priv_mp
> instead of the session_mp.
> > Please check this one.
> >
> https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fpatches.d
> pdk.org%2Fpatch%2F52981%2F&data=02%7C01%7Cakhil.goyal%40nxp.co
> m%7C04194193cfc04c0b629008d6c7eea247%7C686ea1d3bc2b4c6fa92cd99c5c
> 301635%7C0%7C0%7C636916225072561313&sdata=libTy%2F%2Bj23gGru
> vxhdlVUGIOeVq%2BlM2PIF1ZsgN%2FaSY%3D&reserved=0
>
> Yes, I think you right, we need to use sess_private_pool here.
>
> >
> > Ideally this issue should be resolved by adding another parameter in
> rte_security_session_create which can take another mempool pointer
> > for private data allocation. But this cannot be done in this release as it would
> need a deprecation notice.
> >
> > With the above issues I don't see your patch going in 19.05 release cycle.
> >
> > Regards,
> > Akhil
> >
> > >
> > >
> > >
> > > >
> > > > - sec_cap = rte_security_capabilities_get(ctx);
> > > > + /* Set IPsec parameters in conf */
> > > > + set_ipsec_conf(sa, &(sess_conf.ipsec));
> > > >
> > > > - /* iterate until ESP tunnel*/
> > > > - while (sec_cap->action !=
> > > > - RTE_SECURITY_ACTION_TYPE_NONE) {
> > > > + sa->sec_session = rte_security_session_create(ctx,
> > > > + &sess_conf, pool);
> > > > + if (sa->sec_session == NULL) {
> > > > + RTE_LOG(ERR, IPSEC,
> > > > + "SEC Session init failed: err: %d\n",
> > > > + ret);
> > > > + return -1;
> > > > + }
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-04-23 13:32 ` Akhil Goyal
@ 2019-04-23 13:32 ` Akhil Goyal
2019-04-23 14:04 ` Ananyev, Konstantin
1 sibling, 0 replies; 25+ messages in thread
From: Akhil Goyal @ 2019-04-23 13:32 UTC (permalink / raw)
To: Ananyev, Konstantin, Iremonger, Bernard, dev; +Cc: stable
Hi Konstantin,
> Hi Akhil,
>
> >
> > > -----Original Message-----
> > > From: Akhil Goyal
> > > Sent: Thursday, April 18, 2019 7:21 PM
> > > To: Bernard Iremonger <bernard.iremonger@intel.com>; dev@dpdk.org;
> > > konstantin.ananyev@intel.com
> > > Cc: stable@dpdk.org
> > > Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped
> for
> > > inline crypto
> > >
> > > Hi Bernard,
> > >
> > > > - RTE_LOG_DP(DEBUG, IPSEC, "Create session for SA spi %u on
> cryptodev "
> > > > - "%u qp %u\n", sa->spi,
> > > > - ipsec_ctx->tbl[cdev_id_qp].id,
> > > > - ipsec_ctx->tbl[cdev_id_qp].qp);
> > > > + if ((sa == NULL) || (pool == NULL))
> > > > + return -EINVAL;
> > > >
> > > > - if (sa->type != RTE_SECURITY_ACTION_TYPE_NONE) {
> > > > - struct rte_security_session_conf sess_conf = {
> > > > + struct rte_security_session_conf sess_conf = {
> > > > .action_type = sa->type,
> > > > .protocol = RTE_SECURITY_PROTOCOL_IPSEC,
> > > > {.ipsec = {
> > > > @@ -90,247 +65,340 @@ create_session(struct ipsec_ctx *ipsec_ctx,
> struct
> > > > ipsec_sa *sa)
> > > > } },
> > > > .crypto_xform = sa->xforms,
> > > > .userdata = NULL,
> > > > -
> > > > };
> > > >
> > > > - if (sa->type ==
> > > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL)
> > > > {
> > > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > - rte_cryptodev_get_sec_ctx(
> > > > - ipsec_ctx->tbl[cdev_id_qp].id);
> > > > -
> > > > - /* Set IPsec parameters in conf */
> > > > - set_ipsec_conf(sa, &(sess_conf.ipsec));
> > > > -
> > > > - sa->sec_session = rte_security_session_create(ctx,
> > > > - &sess_conf, ipsec_ctx->session_pool);
> > > > - if (sa->sec_session == NULL) {
> > > > - RTE_LOG(ERR, IPSEC,
> > > > - "SEC Session init failed: err: %d\n", ret);
> > > > - return -1;
> > > > - }
> > > > - } else if (sa->type ==
> RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO)
> > > {
> > > > - struct rte_flow_error err;
> > > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > > - rte_eth_dev_get_sec_ctx(
> > > > - sa->portid);
> > > > - const struct rte_security_capability *sec_cap;
> > > > - int ret = 0;
> > > > -
> > > > - sa->sec_session = rte_security_session_create(ctx,
> > > > - &sess_conf, ipsec_ctx->session_pool);
> > > > - if (sa->sec_session == NULL) {
> > > > - RTE_LOG(ERR, IPSEC,
> > > > - "SEC Session init failed: err: %d\n", ret);
> > > > - return -1;
> > > > - }
> > > > + if (sa->type ==
> RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > > + ctx = (struct rte_security_ctx *)
> > > > + rte_eth_dev_get_sec_ctx(sa->portid);
> > >
> > > This is breaking the lookaside mode. Ctx was retrieved using the ipsec_ctx-
> >tbl
> > > struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > rte_cryptodev_get_sec_ctx(
> > > ipsec_ctx->tbl[cdev_id_qp].id);
> > >
> > > I am looking into it, but I don't have time left to get it integrated in RC2. So
> this
> > > has to be pushed to RC3
> >
> > It looks like there are multiple issues in this patch wrt lookaside and none cases.
> Only the inline cases seem to be working.
> >
> > 1. the patch removes the cdev_mapping concept completely. Cdev_id_qp is
> not getting used.
>
> Not exactly.
> cdev_id_qp is still setup, and is still used to decide to which crypto-dev to
> enqueuer the crypto-op:
> ipsec_enqueue(...)
> {
> ...
> enqueue_cop(&ipsec_ctx->tbl[sa->cdev_id_qp], &priv->cop);
I don't see anybody filling "sa->cdev_id_qp". Please let me know if I have missed it somewhere.
It is memset to 0 I guess.
>
>
> Same in ipsec_process().
>
> For initialization, yes cdev_id_qp is not used anymore.
> As discussed here:
> https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmails.dp
> dk.org%2Farchives%2Fdev%2F2019-
> March%2F127725.html&data=02%7C01%7Cakhil.goyal%40nxp.com%7C04
> 194193cfc04c0b629008d6c7eea247%7C686ea1d3bc2b4c6fa92cd99c5c301635%
> 7C0%7C0%7C636916225072561313&sdata=ga9IiqhYRWOz9QkRDIXNiigInk
> soVGgu1E5EetqvE%2FA%3D&reserved=0
>
> I think the problem you are hitting with lookaside-proto is that for it
> we use 2 different values here:
> a) In create_sec_session we use portid (it also should be
> rte_cryptodev_get_sec_ctx() here)
> if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> ctx = (struct rte_security_ctx *)
> rte_eth_dev_get_sec_ctx(sa->portid);
It should be rte_cryptodev_get_sec_ctx in the first place. And it needs a cdev_id as input based on the cdev mapping done while initializing the cryptodev and neither the portid and nor cdev_id_qp.
> b) in enqueue() we use cdev_id_qp
>
> Right now these values could be different.
> As I understand we need to make sure that fro lookaside-proto cdev_id_qp ==
> portid provided by user, correct?
No it is not the case. Right now for lookaside there is no use of portid in case of lookaside case.
As the cdev/qp/core mappings are managed internally and the user cannot tweak it from cfg file.
>
>
> > The port_id cannot be used in case of crypto, the mapping of cdev/qp/core
> is done differently for inbound and outbound ports which is
> > missed in this patch.
> >
> > 2. crypto sessions are created using the session mempool and the private data
> is allocated using the session priv_mempool which is
> > removed in this patch. This will break cases where the priv data is more than
> the size of sess_mp element size.
> > Also the security sessions need to be allocated using the session_priv_mp
> instead of the session_mp.
> > Please check this one.
> >
> https://eur01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fpatches.d
> pdk.org%2Fpatch%2F52981%2F&data=02%7C01%7Cakhil.goyal%40nxp.co
> m%7C04194193cfc04c0b629008d6c7eea247%7C686ea1d3bc2b4c6fa92cd99c5c
> 301635%7C0%7C0%7C636916225072561313&sdata=libTy%2F%2Bj23gGru
> vxhdlVUGIOeVq%2BlM2PIF1ZsgN%2FaSY%3D&reserved=0
>
> Yes, I think you right, we need to use sess_private_pool here.
>
> >
> > Ideally this issue should be resolved by adding another parameter in
> rte_security_session_create which can take another mempool pointer
> > for private data allocation. But this cannot be done in this release as it would
> need a deprecation notice.
> >
> > With the above issues I don't see your patch going in 19.05 release cycle.
> >
> > Regards,
> > Akhil
> >
> > >
> > >
> > >
> > > >
> > > > - sec_cap = rte_security_capabilities_get(ctx);
> > > > + /* Set IPsec parameters in conf */
> > > > + set_ipsec_conf(sa, &(sess_conf.ipsec));
> > > >
> > > > - /* iterate until ESP tunnel*/
> > > > - while (sec_cap->action !=
> > > > - RTE_SECURITY_ACTION_TYPE_NONE) {
> > > > + sa->sec_session = rte_security_session_create(ctx,
> > > > + &sess_conf, pool);
> > > > + if (sa->sec_session == NULL) {
> > > > + RTE_LOG(ERR, IPSEC,
> > > > + "SEC Session init failed: err: %d\n",
> > > > + ret);
> > > > + return -1;
> > > > + }
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-04-23 11:14 ` Akhil Goyal
2019-04-23 11:14 ` Akhil Goyal
@ 2019-04-23 13:21 ` Ananyev, Konstantin
2019-04-23 13:21 ` Ananyev, Konstantin
2019-04-23 13:32 ` Akhil Goyal
1 sibling, 2 replies; 25+ messages in thread
From: Ananyev, Konstantin @ 2019-04-23 13:21 UTC (permalink / raw)
To: Akhil Goyal, Iremonger, Bernard, dev; +Cc: stable
Hi Akhil,
>
> > -----Original Message-----
> > From: Akhil Goyal
> > Sent: Thursday, April 18, 2019 7:21 PM
> > To: Bernard Iremonger <bernard.iremonger@intel.com>; dev@dpdk.org;
> > konstantin.ananyev@intel.com
> > Cc: stable@dpdk.org
> > Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for
> > inline crypto
> >
> > Hi Bernard,
> >
> > > - RTE_LOG_DP(DEBUG, IPSEC, "Create session for SA spi %u on cryptodev "
> > > - "%u qp %u\n", sa->spi,
> > > - ipsec_ctx->tbl[cdev_id_qp].id,
> > > - ipsec_ctx->tbl[cdev_id_qp].qp);
> > > + if ((sa == NULL) || (pool == NULL))
> > > + return -EINVAL;
> > >
> > > - if (sa->type != RTE_SECURITY_ACTION_TYPE_NONE) {
> > > - struct rte_security_session_conf sess_conf = {
> > > + struct rte_security_session_conf sess_conf = {
> > > .action_type = sa->type,
> > > .protocol = RTE_SECURITY_PROTOCOL_IPSEC,
> > > {.ipsec = {
> > > @@ -90,247 +65,340 @@ create_session(struct ipsec_ctx *ipsec_ctx, struct
> > > ipsec_sa *sa)
> > > } },
> > > .crypto_xform = sa->xforms,
> > > .userdata = NULL,
> > > -
> > > };
> > >
> > > - if (sa->type ==
> > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL)
> > > {
> > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > - rte_cryptodev_get_sec_ctx(
> > > - ipsec_ctx->tbl[cdev_id_qp].id);
> > > -
> > > - /* Set IPsec parameters in conf */
> > > - set_ipsec_conf(sa, &(sess_conf.ipsec));
> > > -
> > > - sa->sec_session = rte_security_session_create(ctx,
> > > - &sess_conf, ipsec_ctx->session_pool);
> > > - if (sa->sec_session == NULL) {
> > > - RTE_LOG(ERR, IPSEC,
> > > - "SEC Session init failed: err: %d\n", ret);
> > > - return -1;
> > > - }
> > > - } else if (sa->type == RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO)
> > {
> > > - struct rte_flow_error err;
> > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > - rte_eth_dev_get_sec_ctx(
> > > - sa->portid);
> > > - const struct rte_security_capability *sec_cap;
> > > - int ret = 0;
> > > -
> > > - sa->sec_session = rte_security_session_create(ctx,
> > > - &sess_conf, ipsec_ctx->session_pool);
> > > - if (sa->sec_session == NULL) {
> > > - RTE_LOG(ERR, IPSEC,
> > > - "SEC Session init failed: err: %d\n", ret);
> > > - return -1;
> > > - }
> > > + if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > + ctx = (struct rte_security_ctx *)
> > > + rte_eth_dev_get_sec_ctx(sa->portid);
> >
> > This is breaking the lookaside mode. Ctx was retrieved using the ipsec_ctx->tbl
> > struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > rte_cryptodev_get_sec_ctx(
> > ipsec_ctx->tbl[cdev_id_qp].id);
> >
> > I am looking into it, but I don't have time left to get it integrated in RC2. So this
> > has to be pushed to RC3
>
> It looks like there are multiple issues in this patch wrt lookaside and none cases. Only the inline cases seem to be working.
>
> 1. the patch removes the cdev_mapping concept completely. Cdev_id_qp is not getting used.
Not exactly.
cdev_id_qp is still setup, and is still used to decide to which crypto-dev to enqueuer the crypto-op:
ipsec_enqueue(...)
{
...
enqueue_cop(&ipsec_ctx->tbl[sa->cdev_id_qp], &priv->cop);
Same in ipsec_process().
For initialization, yes cdev_id_qp is not used anymore.
As discussed here:
https://mails.dpdk.org/archives/dev/2019-March/127725.html
I think the problem you are hitting with lookaside-proto is that for it
we use 2 different values here:
a) In create_sec_session we use portid (it also should be rte_cryptodev_get_sec_ctx() here)
if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
ctx = (struct rte_security_ctx *)
rte_eth_dev_get_sec_ctx(sa->portid);
b) in enqueue() we use cdev_id_qp
Right now these values could be different.
As I understand we need to make sure that fro lookaside-proto cdev_id_qp == portid provided by user, correct?
> The port_id cannot be used in case of crypto, the mapping of cdev/qp/core is done differently for inbound and outbound ports which is
> missed in this patch.
>
> 2. crypto sessions are created using the session mempool and the private data is allocated using the session priv_mempool which is
> removed in this patch. This will break cases where the priv data is more than the size of sess_mp element size.
> Also the security sessions need to be allocated using the session_priv_mp instead of the session_mp.
> Please check this one.
> http://patches.dpdk.org/patch/52981/
Yes, I think you right, we need to use sess_private_pool here.
>
> Ideally this issue should be resolved by adding another parameter in rte_security_session_create which can take another mempool pointer
> for private data allocation. But this cannot be done in this release as it would need a deprecation notice.
>
> With the above issues I don't see your patch going in 19.05 release cycle.
>
> Regards,
> Akhil
>
> >
> >
> >
> > >
> > > - sec_cap = rte_security_capabilities_get(ctx);
> > > + /* Set IPsec parameters in conf */
> > > + set_ipsec_conf(sa, &(sess_conf.ipsec));
> > >
> > > - /* iterate until ESP tunnel*/
> > > - while (sec_cap->action !=
> > > - RTE_SECURITY_ACTION_TYPE_NONE) {
> > > + sa->sec_session = rte_security_session_create(ctx,
> > > + &sess_conf, pool);
> > > + if (sa->sec_session == NULL) {
> > > + RTE_LOG(ERR, IPSEC,
> > > + "SEC Session init failed: err: %d\n",
> > > + ret);
> > > + return -1;
> > > + }
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-04-23 13:21 ` Ananyev, Konstantin
@ 2019-04-23 13:21 ` Ananyev, Konstantin
2019-04-23 13:32 ` Akhil Goyal
1 sibling, 0 replies; 25+ messages in thread
From: Ananyev, Konstantin @ 2019-04-23 13:21 UTC (permalink / raw)
To: Akhil Goyal, Iremonger, Bernard, dev; +Cc: stable
Hi Akhil,
>
> > -----Original Message-----
> > From: Akhil Goyal
> > Sent: Thursday, April 18, 2019 7:21 PM
> > To: Bernard Iremonger <bernard.iremonger@intel.com>; dev@dpdk.org;
> > konstantin.ananyev@intel.com
> > Cc: stable@dpdk.org
> > Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for
> > inline crypto
> >
> > Hi Bernard,
> >
> > > - RTE_LOG_DP(DEBUG, IPSEC, "Create session for SA spi %u on cryptodev "
> > > - "%u qp %u\n", sa->spi,
> > > - ipsec_ctx->tbl[cdev_id_qp].id,
> > > - ipsec_ctx->tbl[cdev_id_qp].qp);
> > > + if ((sa == NULL) || (pool == NULL))
> > > + return -EINVAL;
> > >
> > > - if (sa->type != RTE_SECURITY_ACTION_TYPE_NONE) {
> > > - struct rte_security_session_conf sess_conf = {
> > > + struct rte_security_session_conf sess_conf = {
> > > .action_type = sa->type,
> > > .protocol = RTE_SECURITY_PROTOCOL_IPSEC,
> > > {.ipsec = {
> > > @@ -90,247 +65,340 @@ create_session(struct ipsec_ctx *ipsec_ctx, struct
> > > ipsec_sa *sa)
> > > } },
> > > .crypto_xform = sa->xforms,
> > > .userdata = NULL,
> > > -
> > > };
> > >
> > > - if (sa->type ==
> > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL)
> > > {
> > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > - rte_cryptodev_get_sec_ctx(
> > > - ipsec_ctx->tbl[cdev_id_qp].id);
> > > -
> > > - /* Set IPsec parameters in conf */
> > > - set_ipsec_conf(sa, &(sess_conf.ipsec));
> > > -
> > > - sa->sec_session = rte_security_session_create(ctx,
> > > - &sess_conf, ipsec_ctx->session_pool);
> > > - if (sa->sec_session == NULL) {
> > > - RTE_LOG(ERR, IPSEC,
> > > - "SEC Session init failed: err: %d\n", ret);
> > > - return -1;
> > > - }
> > > - } else if (sa->type == RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO)
> > {
> > > - struct rte_flow_error err;
> > > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > > - rte_eth_dev_get_sec_ctx(
> > > - sa->portid);
> > > - const struct rte_security_capability *sec_cap;
> > > - int ret = 0;
> > > -
> > > - sa->sec_session = rte_security_session_create(ctx,
> > > - &sess_conf, ipsec_ctx->session_pool);
> > > - if (sa->sec_session == NULL) {
> > > - RTE_LOG(ERR, IPSEC,
> > > - "SEC Session init failed: err: %d\n", ret);
> > > - return -1;
> > > - }
> > > + if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > + ctx = (struct rte_security_ctx *)
> > > + rte_eth_dev_get_sec_ctx(sa->portid);
> >
> > This is breaking the lookaside mode. Ctx was retrieved using the ipsec_ctx->tbl
> > struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > rte_cryptodev_get_sec_ctx(
> > ipsec_ctx->tbl[cdev_id_qp].id);
> >
> > I am looking into it, but I don't have time left to get it integrated in RC2. So this
> > has to be pushed to RC3
>
> It looks like there are multiple issues in this patch wrt lookaside and none cases. Only the inline cases seem to be working.
>
> 1. the patch removes the cdev_mapping concept completely. Cdev_id_qp is not getting used.
Not exactly.
cdev_id_qp is still setup, and is still used to decide to which crypto-dev to enqueuer the crypto-op:
ipsec_enqueue(...)
{
...
enqueue_cop(&ipsec_ctx->tbl[sa->cdev_id_qp], &priv->cop);
Same in ipsec_process().
For initialization, yes cdev_id_qp is not used anymore.
As discussed here:
https://mails.dpdk.org/archives/dev/2019-March/127725.html
I think the problem you are hitting with lookaside-proto is that for it
we use 2 different values here:
a) In create_sec_session we use portid (it also should be rte_cryptodev_get_sec_ctx() here)
if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
ctx = (struct rte_security_ctx *)
rte_eth_dev_get_sec_ctx(sa->portid);
b) in enqueue() we use cdev_id_qp
Right now these values could be different.
As I understand we need to make sure that fro lookaside-proto cdev_id_qp == portid provided by user, correct?
> The port_id cannot be used in case of crypto, the mapping of cdev/qp/core is done differently for inbound and outbound ports which is
> missed in this patch.
>
> 2. crypto sessions are created using the session mempool and the private data is allocated using the session priv_mempool which is
> removed in this patch. This will break cases where the priv data is more than the size of sess_mp element size.
> Also the security sessions need to be allocated using the session_priv_mp instead of the session_mp.
> Please check this one.
> http://patches.dpdk.org/patch/52981/
Yes, I think you right, we need to use sess_private_pool here.
>
> Ideally this issue should be resolved by adding another parameter in rte_security_session_create which can take another mempool pointer
> for private data allocation. But this cannot be done in this release as it would need a deprecation notice.
>
> With the above issues I don't see your patch going in 19.05 release cycle.
>
> Regards,
> Akhil
>
> >
> >
> >
> > >
> > > - sec_cap = rte_security_capabilities_get(ctx);
> > > + /* Set IPsec parameters in conf */
> > > + set_ipsec_conf(sa, &(sess_conf.ipsec));
> > >
> > > - /* iterate until ESP tunnel*/
> > > - while (sec_cap->action !=
> > > - RTE_SECURITY_ACTION_TYPE_NONE) {
> > > + sa->sec_session = rte_security_session_create(ctx,
> > > + &sess_conf, pool);
> > > + if (sa->sec_session == NULL) {
> > > + RTE_LOG(ERR, IPSEC,
> > > + "SEC Session init failed: err: %d\n",
> > > + ret);
> > > + return -1;
> > > + }
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-04-18 13:51 Akhil Goyal
2019-04-18 13:51 ` Akhil Goyal
2019-04-18 14:58 ` Iremonger, Bernard
@ 2019-04-23 11:14 ` Akhil Goyal
2019-04-23 11:14 ` Akhil Goyal
2019-04-23 13:21 ` Ananyev, Konstantin
2 siblings, 2 replies; 25+ messages in thread
From: Akhil Goyal @ 2019-04-23 11:14 UTC (permalink / raw)
To: Bernard Iremonger, dev, konstantin.ananyev; +Cc: stable
Hi Bernard,
> -----Original Message-----
> From: Akhil Goyal
> Sent: Thursday, April 18, 2019 7:21 PM
> To: Bernard Iremonger <bernard.iremonger@intel.com>; dev@dpdk.org;
> konstantin.ananyev@intel.com
> Cc: stable@dpdk.org
> Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for
> inline crypto
>
> Hi Bernard,
>
> > - RTE_LOG_DP(DEBUG, IPSEC, "Create session for SA spi %u on cryptodev "
> > - "%u qp %u\n", sa->spi,
> > - ipsec_ctx->tbl[cdev_id_qp].id,
> > - ipsec_ctx->tbl[cdev_id_qp].qp);
> > + if ((sa == NULL) || (pool == NULL))
> > + return -EINVAL;
> >
> > - if (sa->type != RTE_SECURITY_ACTION_TYPE_NONE) {
> > - struct rte_security_session_conf sess_conf = {
> > + struct rte_security_session_conf sess_conf = {
> > .action_type = sa->type,
> > .protocol = RTE_SECURITY_PROTOCOL_IPSEC,
> > {.ipsec = {
> > @@ -90,247 +65,340 @@ create_session(struct ipsec_ctx *ipsec_ctx, struct
> > ipsec_sa *sa)
> > } },
> > .crypto_xform = sa->xforms,
> > .userdata = NULL,
> > -
> > };
> >
> > - if (sa->type ==
> RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL)
> > {
> > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > - rte_cryptodev_get_sec_ctx(
> > - ipsec_ctx->tbl[cdev_id_qp].id);
> > -
> > - /* Set IPsec parameters in conf */
> > - set_ipsec_conf(sa, &(sess_conf.ipsec));
> > -
> > - sa->sec_session = rte_security_session_create(ctx,
> > - &sess_conf, ipsec_ctx->session_pool);
> > - if (sa->sec_session == NULL) {
> > - RTE_LOG(ERR, IPSEC,
> > - "SEC Session init failed: err: %d\n", ret);
> > - return -1;
> > - }
> > - } else if (sa->type == RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO)
> {
> > - struct rte_flow_error err;
> > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > - rte_eth_dev_get_sec_ctx(
> > - sa->portid);
> > - const struct rte_security_capability *sec_cap;
> > - int ret = 0;
> > -
> > - sa->sec_session = rte_security_session_create(ctx,
> > - &sess_conf, ipsec_ctx->session_pool);
> > - if (sa->sec_session == NULL) {
> > - RTE_LOG(ERR, IPSEC,
> > - "SEC Session init failed: err: %d\n", ret);
> > - return -1;
> > - }
> > + if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > + ctx = (struct rte_security_ctx *)
> > + rte_eth_dev_get_sec_ctx(sa->portid);
>
> This is breaking the lookaside mode. Ctx was retrieved using the ipsec_ctx->tbl
> struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> rte_cryptodev_get_sec_ctx(
> ipsec_ctx->tbl[cdev_id_qp].id);
>
> I am looking into it, but I don't have time left to get it integrated in RC2. So this
> has to be pushed to RC3
It looks like there are multiple issues in this patch wrt lookaside and none cases. Only the inline cases seem to be working.
1. the patch removes the cdev_mapping concept completely. Cdev_id_qp is not getting used.
The port_id cannot be used in case of crypto, the mapping of cdev/qp/core is done differently for inbound and outbound ports which is missed in this patch.
2. crypto sessions are created using the session mempool and the private data is allocated using the session priv_mempool which is removed in this patch. This will break cases where the priv data is more than the size of sess_mp element size.
Also the security sessions need to be allocated using the session_priv_mp instead of the session_mp.
Please check this one.
http://patches.dpdk.org/patch/52981/
Ideally this issue should be resolved by adding another parameter in rte_security_session_create which can take another mempool pointer for private data allocation. But this cannot be done in this release as it would need a deprecation notice.
With the above issues I don't see your patch going in 19.05 release cycle.
Regards,
Akhil
>
>
>
> >
> > - sec_cap = rte_security_capabilities_get(ctx);
> > + /* Set IPsec parameters in conf */
> > + set_ipsec_conf(sa, &(sess_conf.ipsec));
> >
> > - /* iterate until ESP tunnel*/
> > - while (sec_cap->action !=
> > - RTE_SECURITY_ACTION_TYPE_NONE) {
> > + sa->sec_session = rte_security_session_create(ctx,
> > + &sess_conf, pool);
> > + if (sa->sec_session == NULL) {
> > + RTE_LOG(ERR, IPSEC,
> > + "SEC Session init failed: err: %d\n",
> > + ret);
> > + return -1;
> > + }
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-04-23 11:14 ` Akhil Goyal
@ 2019-04-23 11:14 ` Akhil Goyal
2019-04-23 13:21 ` Ananyev, Konstantin
1 sibling, 0 replies; 25+ messages in thread
From: Akhil Goyal @ 2019-04-23 11:14 UTC (permalink / raw)
To: Bernard Iremonger, dev, konstantin.ananyev; +Cc: stable
Hi Bernard,
> -----Original Message-----
> From: Akhil Goyal
> Sent: Thursday, April 18, 2019 7:21 PM
> To: Bernard Iremonger <bernard.iremonger@intel.com>; dev@dpdk.org;
> konstantin.ananyev@intel.com
> Cc: stable@dpdk.org
> Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for
> inline crypto
>
> Hi Bernard,
>
> > - RTE_LOG_DP(DEBUG, IPSEC, "Create session for SA spi %u on cryptodev "
> > - "%u qp %u\n", sa->spi,
> > - ipsec_ctx->tbl[cdev_id_qp].id,
> > - ipsec_ctx->tbl[cdev_id_qp].qp);
> > + if ((sa == NULL) || (pool == NULL))
> > + return -EINVAL;
> >
> > - if (sa->type != RTE_SECURITY_ACTION_TYPE_NONE) {
> > - struct rte_security_session_conf sess_conf = {
> > + struct rte_security_session_conf sess_conf = {
> > .action_type = sa->type,
> > .protocol = RTE_SECURITY_PROTOCOL_IPSEC,
> > {.ipsec = {
> > @@ -90,247 +65,340 @@ create_session(struct ipsec_ctx *ipsec_ctx, struct
> > ipsec_sa *sa)
> > } },
> > .crypto_xform = sa->xforms,
> > .userdata = NULL,
> > -
> > };
> >
> > - if (sa->type ==
> RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL)
> > {
> > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > - rte_cryptodev_get_sec_ctx(
> > - ipsec_ctx->tbl[cdev_id_qp].id);
> > -
> > - /* Set IPsec parameters in conf */
> > - set_ipsec_conf(sa, &(sess_conf.ipsec));
> > -
> > - sa->sec_session = rte_security_session_create(ctx,
> > - &sess_conf, ipsec_ctx->session_pool);
> > - if (sa->sec_session == NULL) {
> > - RTE_LOG(ERR, IPSEC,
> > - "SEC Session init failed: err: %d\n", ret);
> > - return -1;
> > - }
> > - } else if (sa->type == RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO)
> {
> > - struct rte_flow_error err;
> > - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > - rte_eth_dev_get_sec_ctx(
> > - sa->portid);
> > - const struct rte_security_capability *sec_cap;
> > - int ret = 0;
> > -
> > - sa->sec_session = rte_security_session_create(ctx,
> > - &sess_conf, ipsec_ctx->session_pool);
> > - if (sa->sec_session == NULL) {
> > - RTE_LOG(ERR, IPSEC,
> > - "SEC Session init failed: err: %d\n", ret);
> > - return -1;
> > - }
> > + if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > + ctx = (struct rte_security_ctx *)
> > + rte_eth_dev_get_sec_ctx(sa->portid);
>
> This is breaking the lookaside mode. Ctx was retrieved using the ipsec_ctx->tbl
> struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> rte_cryptodev_get_sec_ctx(
> ipsec_ctx->tbl[cdev_id_qp].id);
>
> I am looking into it, but I don't have time left to get it integrated in RC2. So this
> has to be pushed to RC3
It looks like there are multiple issues in this patch wrt lookaside and none cases. Only the inline cases seem to be working.
1. the patch removes the cdev_mapping concept completely. Cdev_id_qp is not getting used.
The port_id cannot be used in case of crypto, the mapping of cdev/qp/core is done differently for inbound and outbound ports which is missed in this patch.
2. crypto sessions are created using the session mempool and the private data is allocated using the session priv_mempool which is removed in this patch. This will break cases where the priv data is more than the size of sess_mp element size.
Also the security sessions need to be allocated using the session_priv_mp instead of the session_mp.
Please check this one.
http://patches.dpdk.org/patch/52981/
Ideally this issue should be resolved by adding another parameter in rte_security_session_create which can take another mempool pointer for private data allocation. But this cannot be done in this release as it would need a deprecation notice.
With the above issues I don't see your patch going in 19.05 release cycle.
Regards,
Akhil
>
>
>
> >
> > - sec_cap = rte_security_capabilities_get(ctx);
> > + /* Set IPsec parameters in conf */
> > + set_ipsec_conf(sa, &(sess_conf.ipsec));
> >
> > - /* iterate until ESP tunnel*/
> > - while (sec_cap->action !=
> > - RTE_SECURITY_ACTION_TYPE_NONE) {
> > + sa->sec_session = rte_security_session_create(ctx,
> > + &sess_conf, pool);
> > + if (sa->sec_session == NULL) {
> > + RTE_LOG(ERR, IPSEC,
> > + "SEC Session init failed: err: %d\n",
> > + ret);
> > + return -1;
> > + }
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-04-18 14:58 ` Iremonger, Bernard
2019-04-18 14:58 ` Iremonger, Bernard
@ 2019-04-18 15:23 ` Iremonger, Bernard
2019-04-18 15:23 ` Iremonger, Bernard
1 sibling, 1 reply; 25+ messages in thread
From: Iremonger, Bernard @ 2019-04-18 15:23 UTC (permalink / raw)
To: Iremonger, Bernard, Akhil Goyal, dev, Ananyev, Konstantin; +Cc: stable
Hi Akhil,
> -----Original Message-----
> From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Iremonger, Bernard
> Sent: Thursday, April 18, 2019 3:59 PM
> To: Akhil Goyal <akhil.goyal@nxp.com>; dev@dpdk.org; Ananyev,
> Konstantin <konstantin.ananyev@intel.com>
> Cc: stable@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet
> dropped for inline crypto
>
> Hi Akhil,
>
> <snip>
>
> > Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet
> > dropped for inline crypto
>
> <snip>
> > > + if (sa->type ==
> > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > + ctx = (struct rte_security_ctx *)
> > > + rte_eth_dev_get_sec_ctx(sa->portid);
> >
> > This is breaking the lookaside mode. Ctx was retrieved using the
> > ipsec_ctx-
> > >tbl struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > rte_cryptodev_get_sec_ctx(
> > ipsec_ctx->tbl[cdev_id_qp].id);
> >
> > I am looking into it, but I don't have time left to get it integrated
> > in RC2. So this has to be pushed to RC3
>
> <snip>
>
> Unfortunately we do not have the HW to test this feature.
> What HW are you using to test this?
>
> Having looked at the code previously
> ipsec_ctx->tbl[cdev_id_qp].id turned out to be the port_id.
>
> So we had expected it to work.
>
> We will need your help with this.
>
> Regards,
>
> Bernard.
Just had another look at the 19.05.rc1 code
Line 1546 in ipsec-secgw.c:
ipsec_ctx->tbl[i].id = cdev_id;
The id is the cryptodev id, not the port_id
Regards,
Bernard.
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-04-18 15:23 ` Iremonger, Bernard
@ 2019-04-18 15:23 ` Iremonger, Bernard
0 siblings, 0 replies; 25+ messages in thread
From: Iremonger, Bernard @ 2019-04-18 15:23 UTC (permalink / raw)
To: Iremonger, Bernard, Akhil Goyal, dev, Ananyev, Konstantin; +Cc: stable
Hi Akhil,
> -----Original Message-----
> From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Iremonger, Bernard
> Sent: Thursday, April 18, 2019 3:59 PM
> To: Akhil Goyal <akhil.goyal@nxp.com>; dev@dpdk.org; Ananyev,
> Konstantin <konstantin.ananyev@intel.com>
> Cc: stable@dpdk.org
> Subject: Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet
> dropped for inline crypto
>
> Hi Akhil,
>
> <snip>
>
> > Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet
> > dropped for inline crypto
>
> <snip>
> > > + if (sa->type ==
> > RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > > + ctx = (struct rte_security_ctx *)
> > > + rte_eth_dev_get_sec_ctx(sa->portid);
> >
> > This is breaking the lookaside mode. Ctx was retrieved using the
> > ipsec_ctx-
> > >tbl struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> > rte_cryptodev_get_sec_ctx(
> > ipsec_ctx->tbl[cdev_id_qp].id);
> >
> > I am looking into it, but I don't have time left to get it integrated
> > in RC2. So this has to be pushed to RC3
>
> <snip>
>
> Unfortunately we do not have the HW to test this feature.
> What HW are you using to test this?
>
> Having looked at the code previously
> ipsec_ctx->tbl[cdev_id_qp].id turned out to be the port_id.
>
> So we had expected it to work.
>
> We will need your help with this.
>
> Regards,
>
> Bernard.
Just had another look at the 19.05.rc1 code
Line 1546 in ipsec-secgw.c:
ipsec_ctx->tbl[i].id = cdev_id;
The id is the cryptodev id, not the port_id
Regards,
Bernard.
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-04-18 13:51 Akhil Goyal
2019-04-18 13:51 ` Akhil Goyal
@ 2019-04-18 14:58 ` Iremonger, Bernard
2019-04-18 14:58 ` Iremonger, Bernard
2019-04-18 15:23 ` Iremonger, Bernard
2019-04-23 11:14 ` Akhil Goyal
2 siblings, 2 replies; 25+ messages in thread
From: Iremonger, Bernard @ 2019-04-18 14:58 UTC (permalink / raw)
To: Akhil Goyal, dev, Ananyev, Konstantin; +Cc: stable
Hi Akhil,
<snip>
> Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped
> for inline crypto
<snip>
> > + if (sa->type ==
> RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > + ctx = (struct rte_security_ctx *)
> > + rte_eth_dev_get_sec_ctx(sa->portid);
>
> This is breaking the lookaside mode. Ctx was retrieved using the ipsec_ctx-
> >tbl struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> rte_cryptodev_get_sec_ctx(
> ipsec_ctx->tbl[cdev_id_qp].id);
>
> I am looking into it, but I don't have time left to get it integrated in RC2. So
> this has to be pushed to RC3
<snip>
Unfortunately we do not have the HW to test this feature.
What HW are you using to test this?
Having looked at the code previously
ipsec_ctx->tbl[cdev_id_qp].id turned out to be the port_id.
So we had expected it to work.
We will need your help with this.
Regards,
Bernard.
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-04-18 14:58 ` Iremonger, Bernard
@ 2019-04-18 14:58 ` Iremonger, Bernard
2019-04-18 15:23 ` Iremonger, Bernard
1 sibling, 0 replies; 25+ messages in thread
From: Iremonger, Bernard @ 2019-04-18 14:58 UTC (permalink / raw)
To: Akhil Goyal, dev, Ananyev, Konstantin; +Cc: stable
Hi Akhil,
<snip>
> Subject: RE: [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped
> for inline crypto
<snip>
> > + if (sa->type ==
> RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> > + ctx = (struct rte_security_ctx *)
> > + rte_eth_dev_get_sec_ctx(sa->portid);
>
> This is breaking the lookaside mode. Ctx was retrieved using the ipsec_ctx-
> >tbl struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> rte_cryptodev_get_sec_ctx(
> ipsec_ctx->tbl[cdev_id_qp].id);
>
> I am looking into it, but I don't have time left to get it integrated in RC2. So
> this has to be pushed to RC3
<snip>
Unfortunately we do not have the HW to test this feature.
What HW are you using to test this?
Having looked at the code previously
ipsec_ctx->tbl[cdev_id_qp].id turned out to be the port_id.
So we had expected it to work.
We will need your help with this.
Regards,
Bernard.
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
@ 2019-04-18 13:51 Akhil Goyal
2019-04-18 13:51 ` Akhil Goyal
` (2 more replies)
0 siblings, 3 replies; 25+ messages in thread
From: Akhil Goyal @ 2019-04-18 13:51 UTC (permalink / raw)
To: Bernard Iremonger, dev, konstantin.ananyev; +Cc: stable
Hi Bernard,
> - RTE_LOG_DP(DEBUG, IPSEC, "Create session for SA spi %u on cryptodev "
> - "%u qp %u\n", sa->spi,
> - ipsec_ctx->tbl[cdev_id_qp].id,
> - ipsec_ctx->tbl[cdev_id_qp].qp);
> + if ((sa == NULL) || (pool == NULL))
> + return -EINVAL;
>
> - if (sa->type != RTE_SECURITY_ACTION_TYPE_NONE) {
> - struct rte_security_session_conf sess_conf = {
> + struct rte_security_session_conf sess_conf = {
> .action_type = sa->type,
> .protocol = RTE_SECURITY_PROTOCOL_IPSEC,
> {.ipsec = {
> @@ -90,247 +65,340 @@ create_session(struct ipsec_ctx *ipsec_ctx, struct
> ipsec_sa *sa)
> } },
> .crypto_xform = sa->xforms,
> .userdata = NULL,
> -
> };
>
> - if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL)
> {
> - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> - rte_cryptodev_get_sec_ctx(
> - ipsec_ctx->tbl[cdev_id_qp].id);
> -
> - /* Set IPsec parameters in conf */
> - set_ipsec_conf(sa, &(sess_conf.ipsec));
> -
> - sa->sec_session = rte_security_session_create(ctx,
> - &sess_conf, ipsec_ctx->session_pool);
> - if (sa->sec_session == NULL) {
> - RTE_LOG(ERR, IPSEC,
> - "SEC Session init failed: err: %d\n", ret);
> - return -1;
> - }
> - } else if (sa->type == RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO) {
> - struct rte_flow_error err;
> - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> - rte_eth_dev_get_sec_ctx(
> - sa->portid);
> - const struct rte_security_capability *sec_cap;
> - int ret = 0;
> -
> - sa->sec_session = rte_security_session_create(ctx,
> - &sess_conf, ipsec_ctx->session_pool);
> - if (sa->sec_session == NULL) {
> - RTE_LOG(ERR, IPSEC,
> - "SEC Session init failed: err: %d\n", ret);
> - return -1;
> - }
> + if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> + ctx = (struct rte_security_ctx *)
> + rte_eth_dev_get_sec_ctx(sa->portid);
This is breaking the lookaside mode. Ctx was retrieved using the ipsec_ctx->tbl
struct rte_security_ctx *ctx = (struct rte_security_ctx *)
rte_cryptodev_get_sec_ctx(
ipsec_ctx->tbl[cdev_id_qp].id);
I am looking into it, but I don't have time left to get it integrated in RC2. So this has to be pushed to RC3
>
> - sec_cap = rte_security_capabilities_get(ctx);
> + /* Set IPsec parameters in conf */
> + set_ipsec_conf(sa, &(sess_conf.ipsec));
>
> - /* iterate until ESP tunnel*/
> - while (sec_cap->action !=
> - RTE_SECURITY_ACTION_TYPE_NONE) {
> + sa->sec_session = rte_security_session_create(ctx,
> + &sess_conf, pool);
> + if (sa->sec_session == NULL) {
> + RTE_LOG(ERR, IPSEC,
> + "SEC Session init failed: err: %d\n",
> + ret);
> + return -1;
> + }
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-04-18 13:51 Akhil Goyal
@ 2019-04-18 13:51 ` Akhil Goyal
2019-04-18 14:58 ` Iremonger, Bernard
2019-04-23 11:14 ` Akhil Goyal
2 siblings, 0 replies; 25+ messages in thread
From: Akhil Goyal @ 2019-04-18 13:51 UTC (permalink / raw)
To: Bernard Iremonger, dev, konstantin.ananyev; +Cc: stable
Hi Bernard,
> - RTE_LOG_DP(DEBUG, IPSEC, "Create session for SA spi %u on cryptodev "
> - "%u qp %u\n", sa->spi,
> - ipsec_ctx->tbl[cdev_id_qp].id,
> - ipsec_ctx->tbl[cdev_id_qp].qp);
> + if ((sa == NULL) || (pool == NULL))
> + return -EINVAL;
>
> - if (sa->type != RTE_SECURITY_ACTION_TYPE_NONE) {
> - struct rte_security_session_conf sess_conf = {
> + struct rte_security_session_conf sess_conf = {
> .action_type = sa->type,
> .protocol = RTE_SECURITY_PROTOCOL_IPSEC,
> {.ipsec = {
> @@ -90,247 +65,340 @@ create_session(struct ipsec_ctx *ipsec_ctx, struct
> ipsec_sa *sa)
> } },
> .crypto_xform = sa->xforms,
> .userdata = NULL,
> -
> };
>
> - if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL)
> {
> - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> - rte_cryptodev_get_sec_ctx(
> - ipsec_ctx->tbl[cdev_id_qp].id);
> -
> - /* Set IPsec parameters in conf */
> - set_ipsec_conf(sa, &(sess_conf.ipsec));
> -
> - sa->sec_session = rte_security_session_create(ctx,
> - &sess_conf, ipsec_ctx->session_pool);
> - if (sa->sec_session == NULL) {
> - RTE_LOG(ERR, IPSEC,
> - "SEC Session init failed: err: %d\n", ret);
> - return -1;
> - }
> - } else if (sa->type == RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO) {
> - struct rte_flow_error err;
> - struct rte_security_ctx *ctx = (struct rte_security_ctx *)
> - rte_eth_dev_get_sec_ctx(
> - sa->portid);
> - const struct rte_security_capability *sec_cap;
> - int ret = 0;
> -
> - sa->sec_session = rte_security_session_create(ctx,
> - &sess_conf, ipsec_ctx->session_pool);
> - if (sa->sec_session == NULL) {
> - RTE_LOG(ERR, IPSEC,
> - "SEC Session init failed: err: %d\n", ret);
> - return -1;
> - }
> + if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
> + ctx = (struct rte_security_ctx *)
> + rte_eth_dev_get_sec_ctx(sa->portid);
This is breaking the lookaside mode. Ctx was retrieved using the ipsec_ctx->tbl
struct rte_security_ctx *ctx = (struct rte_security_ctx *)
rte_cryptodev_get_sec_ctx(
ipsec_ctx->tbl[cdev_id_qp].id);
I am looking into it, but I don't have time left to get it integrated in RC2. So this has to be pushed to RC3
>
> - sec_cap = rte_security_capabilities_get(ctx);
> + /* Set IPsec parameters in conf */
> + set_ipsec_conf(sa, &(sess_conf.ipsec));
>
> - /* iterate until ESP tunnel*/
> - while (sec_cap->action !=
> - RTE_SECURITY_ACTION_TYPE_NONE) {
> + sa->sec_session = rte_security_session_create(ctx,
> + &sess_conf, pool);
> + if (sa->sec_session == NULL) {
> + RTE_LOG(ERR, IPSEC,
> + "SEC Session init failed: err: %d\n",
> + ret);
> + return -1;
> + }
^ permalink raw reply [flat|nested] 25+ messages in thread
* [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-04-04 13:28 [dpdk-dev] [PATCH v3 0/2] examples/ipsec-secgw: fix 1st pkt dropped Bernard Iremonger
@ 2019-04-17 13:42 ` Bernard Iremonger
2019-04-17 13:42 ` Bernard Iremonger
0 siblings, 1 reply; 25+ messages in thread
From: Bernard Iremonger @ 2019-04-17 13:42 UTC (permalink / raw)
To: dev, konstantin.ananyev, akhil.goyal; +Cc: Bernard Iremonger, stable
Inline crypto installs a flow rule in the NIC. This flow
rule must be installed before the first inbound packet is
received.
The create_session() function installs the flow rule.
Refactor ipsec-secgw.c, sa.c, ipsec.h and ipsec.c to create
sessions at startup to fix the issue of the first packet
being dropped for inline crypto.
The create_session() function is now called at initialisation in
sa_add_rules() which is called from sa_init().
The return code for add_rules is checked.
Calls to create_session() in other functions are dropped.
Add crypto_devid_fill() in ipsec-secgw.c
Add max_session_size() in ipsec-secgw.c
Add check_cryptodev_capability() in ipsec.c
Add check_cryptodev_aead_capability() in ipsec.c
Add create_sec_session() and create_crypto_session() in ipsec.c
The crypto_dev_fill() function has been added to find the
enabled crypto devices.
The max_session_size() function has been added to calculate memory
requirements.
The check_cryptodev_capability() and check_cryptodev_aead_capability()
functions have been added to check that the SA is supported by the
crypto device.
The create_session() function is refactored to use the
create_sec_session() and create_crypto_session() functions.
The cryprodev_init() function has been refactored to drop calls to
rte_mempool_create() and to drop calculation of memory requirements.
The main() function has been refactored to call crypto_devid_fill()
and max_session_size() and to call session_pool_init() and
session_priv_pool_init().
The ports are started now before adding a flow rule in main().
The sa_init(), sp4_init(), sp6_init() and rt_init() functions are
now called after the ports have been started.
Fixes: ec17993a145a ("examples/ipsec-secgw: support security offload")
Fixes: d299106e8e31 ("examples/ipsec-secgw: add IPsec sample application")
Cc: stable@dpdk.org
Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
examples/ipsec-secgw/ipsec-secgw.c | 271 ++++++++--------
examples/ipsec-secgw/ipsec.c | 581 +++++++++++++++++++----------------
examples/ipsec-secgw/ipsec.h | 10 +-
examples/ipsec-secgw/ipsec_process.c | 38 +--
examples/ipsec-secgw/sa.c | 42 ++-
5 files changed, 501 insertions(+), 441 deletions(-)
diff --git a/examples/ipsec-secgw/ipsec-secgw.c b/examples/ipsec-secgw/ipsec-secgw.c
index ffbd00b..cc8bb57 100644
--- a/examples/ipsec-secgw/ipsec-secgw.c
+++ b/examples/ipsec-secgw/ipsec-secgw.c
@@ -182,6 +182,14 @@ struct lcore_params {
uint8_t lcore_id;
} __rte_cache_aligned;
+/*
+ * Number of enabled crypto devices
+ * This number is needed when checking crypto device capabilities
+ */
+uint8_t crypto_dev_num;
+/* array of crypto device ID's */
+uint8_t crypto_devid[RTE_CRYPTO_MAX_DEVS];
+
static struct lcore_params lcore_params_array[MAX_LCORE_PARAMS];
static struct lcore_params *lcore_params;
@@ -1623,13 +1631,27 @@ check_cryptodev_mask(uint8_t cdev_id)
return -1;
}
+static void
+crypto_devid_fill(void)
+{
+ uint32_t i, n;
+
+ n = rte_cryptodev_count();
+
+ for (i = 0; i != n; i++) {
+ if (check_cryptodev_mask(i) == 0)
+ crypto_devid[crypto_dev_num++] = i;
+ }
+}
+
static int32_t
cryptodevs_init(void)
{
struct rte_cryptodev_config dev_conf;
struct rte_cryptodev_qp_conf qp_conf;
uint16_t idx, max_nb_qps, qp, i;
- int16_t cdev_id, port_id;
+ int16_t cdev_id;
+ uint32_t dev_max_sess;
struct rte_hash_parameters params = { 0 };
params.entries = CDEV_MAP_ENTRIES;
@@ -1652,45 +1674,6 @@ cryptodevs_init(void)
printf("lcore/cryptodev/qp mappings:\n");
- uint32_t max_sess_sz = 0, sess_sz;
- for (cdev_id = 0; cdev_id < rte_cryptodev_count(); cdev_id++) {
- void *sec_ctx;
-
- /* Get crypto priv session size */
- sess_sz = rte_cryptodev_sym_get_private_session_size(cdev_id);
- if (sess_sz > max_sess_sz)
- max_sess_sz = sess_sz;
-
- /*
- * If crypto device is security capable, need to check the
- * size of security session as well.
- */
-
- /* Get security context of the crypto device */
- sec_ctx = rte_cryptodev_get_sec_ctx(cdev_id);
- if (sec_ctx == NULL)
- continue;
-
- /* Get size of security session */
- sess_sz = rte_security_session_get_size(sec_ctx);
- if (sess_sz > max_sess_sz)
- max_sess_sz = sess_sz;
- }
- RTE_ETH_FOREACH_DEV(port_id) {
- void *sec_ctx;
-
- if ((enabled_port_mask & (1 << port_id)) == 0)
- continue;
-
- sec_ctx = rte_eth_dev_get_sec_ctx(port_id);
- if (sec_ctx == NULL)
- continue;
-
- sess_sz = rte_security_session_get_size(sec_ctx);
- if (sess_sz > max_sess_sz)
- max_sess_sz = sess_sz;
- }
-
idx = 0;
for (cdev_id = 0; cdev_id < rte_cryptodev_count(); cdev_id++) {
struct rte_cryptodev_info cdev_info;
@@ -1722,51 +1705,12 @@ cryptodevs_init(void)
dev_conf.socket_id = rte_cryptodev_socket_id(cdev_id);
dev_conf.nb_queue_pairs = qp;
- uint32_t dev_max_sess = cdev_info.sym.max_nb_sessions;
+ dev_max_sess = cdev_info.sym.max_nb_sessions;
if (dev_max_sess != 0 && dev_max_sess < CDEV_MP_NB_OBJS)
rte_exit(EXIT_FAILURE,
"Device does not support at least %u "
"sessions", CDEV_MP_NB_OBJS);
- if (!socket_ctx[dev_conf.socket_id].session_pool) {
- char mp_name[RTE_MEMPOOL_NAMESIZE];
- struct rte_mempool *sess_mp;
-
- snprintf(mp_name, RTE_MEMPOOL_NAMESIZE,
- "sess_mp_%u", dev_conf.socket_id);
- sess_mp = rte_cryptodev_sym_session_pool_create(
- mp_name, CDEV_MP_NB_OBJS,
- 0, CDEV_MP_CACHE_SZ, 0,
- dev_conf.socket_id);
- socket_ctx[dev_conf.socket_id].session_pool = sess_mp;
- }
-
- if (!socket_ctx[dev_conf.socket_id].session_priv_pool) {
- char mp_name[RTE_MEMPOOL_NAMESIZE];
- struct rte_mempool *sess_mp;
-
- snprintf(mp_name, RTE_MEMPOOL_NAMESIZE,
- "sess_mp_priv_%u", dev_conf.socket_id);
- sess_mp = rte_mempool_create(mp_name,
- CDEV_MP_NB_OBJS,
- max_sess_sz,
- CDEV_MP_CACHE_SZ,
- 0, NULL, NULL, NULL,
- NULL, dev_conf.socket_id,
- 0);
- socket_ctx[dev_conf.socket_id].session_priv_pool =
- sess_mp;
- }
-
- if (!socket_ctx[dev_conf.socket_id].session_priv_pool ||
- !socket_ctx[dev_conf.socket_id].session_pool)
- rte_exit(EXIT_FAILURE,
- "Cannot create session pool on socket %d\n",
- dev_conf.socket_id);
- else
- printf("Allocated session pool on socket %d\n",
- dev_conf.socket_id);
-
if (rte_cryptodev_configure(cdev_id, &dev_conf))
rte_panic("Failed to initialize cryptodev %u\n",
cdev_id);
@@ -1787,38 +1731,6 @@ cryptodevs_init(void)
cdev_id);
}
- /* create session pools for eth devices that implement security */
- RTE_ETH_FOREACH_DEV(port_id) {
- if ((enabled_port_mask & (1 << port_id)) &&
- rte_eth_dev_get_sec_ctx(port_id)) {
- int socket_id = rte_eth_dev_socket_id(port_id);
-
- if (!socket_ctx[socket_id].session_pool) {
- char mp_name[RTE_MEMPOOL_NAMESIZE];
- struct rte_mempool *sess_mp;
-
- snprintf(mp_name, RTE_MEMPOOL_NAMESIZE,
- "sess_mp_%u", socket_id);
- sess_mp = rte_mempool_create(mp_name,
- (CDEV_MP_NB_OBJS * 2),
- max_sess_sz,
- CDEV_MP_CACHE_SZ,
- 0, NULL, NULL, NULL,
- NULL, socket_id,
- 0);
- if (sess_mp == NULL)
- rte_exit(EXIT_FAILURE,
- "Cannot create session pool "
- "on socket %d\n", socket_id);
- else
- printf("Allocated session pool "
- "on socket %d\n", socket_id);
- socket_ctx[socket_id].session_pool = sess_mp;
- }
- }
- }
-
-
printf("\n");
return 0;
@@ -1984,6 +1896,98 @@ port_init(uint16_t portid, uint64_t req_rx_offloads, uint64_t req_tx_offloads)
printf("\n");
}
+static size_t
+max_session_size(void)
+{
+ size_t max_sz, sz;
+ void *sec_ctx;
+ int16_t cdev_id, port_id, n;
+
+ max_sz = 0;
+ n = rte_cryptodev_count();
+ for (cdev_id = 0; cdev_id != n; cdev_id++) {
+ sz = rte_cryptodev_sym_get_private_session_size(cdev_id);
+ if (sz > max_sz)
+ max_sz = sz;
+ /*
+ * If crypto device is security capable, need to check the
+ * size of security session as well.
+ */
+
+ /* Get security context of the crypto device */
+ sec_ctx = rte_cryptodev_get_sec_ctx(cdev_id);
+ if (sec_ctx == NULL)
+ continue;
+
+ /* Get size of security session */
+ sz = rte_security_session_get_size(sec_ctx);
+ if (sz > max_sz)
+ max_sz = sz;
+ }
+
+ RTE_ETH_FOREACH_DEV(port_id) {
+ if ((enabled_port_mask & (1 << port_id)) == 0)
+ continue;
+
+ sec_ctx = rte_eth_dev_get_sec_ctx(port_id);
+ if (sec_ctx == NULL)
+ continue;
+
+ sz = rte_security_session_get_size(sec_ctx);
+ if (sz > max_sz)
+ max_sz = sz;
+ }
+
+ return max_sz;
+}
+
+static void
+session_pool_init(struct socket_ctx *ctx, int32_t socket_id, size_t sess_sz)
+{
+ char mp_name[RTE_MEMPOOL_NAMESIZE];
+ struct rte_mempool *sess_mp;
+
+ snprintf(mp_name, RTE_MEMPOOL_NAMESIZE,
+ "sess_mp_%u", socket_id);
+ sess_mp = rte_cryptodev_sym_session_pool_create(
+ mp_name, CDEV_MP_NB_OBJS,
+ sess_sz, CDEV_MP_CACHE_SZ, 0,
+ socket_id);
+ ctx->session_pool = sess_mp;
+
+ if (ctx->session_pool == NULL)
+ rte_exit(EXIT_FAILURE,
+ "Cannot init session pool on socket %d\n", socket_id);
+ else
+ printf("Allocated session pool on socket %d\n", socket_id);
+}
+
+static void
+session_priv_pool_init(struct socket_ctx *ctx, int32_t socket_id,
+ size_t sess_sz)
+{
+ char mp_name[RTE_MEMPOOL_NAMESIZE];
+ struct rte_mempool *sess_mp;
+
+ snprintf(mp_name, RTE_MEMPOOL_NAMESIZE,
+ "sess_mp_priv_%u", socket_id);
+ sess_mp = rte_mempool_create(mp_name,
+ CDEV_MP_NB_OBJS,
+ sess_sz,
+ CDEV_MP_CACHE_SZ,
+ 0, NULL, NULL, NULL,
+ NULL, socket_id,
+ 0);
+ ctx->session_priv_pool = sess_mp;
+ if (ctx->session_priv_pool == NULL)
+ rte_exit(EXIT_FAILURE,
+ "Cannot init session priv pool on socket %d\n",
+ socket_id);
+ else
+ printf("Allocated session priv pool on socket %d\n",
+ socket_id);
+}
+
static void
pool_init(struct socket_ctx *ctx, int32_t socket_id, uint32_t nb_mbuf)
{
@@ -2064,9 +2068,11 @@ main(int32_t argc, char **argv)
{
int32_t ret;
uint32_t lcore_id;
+ uint32_t i;
uint8_t socket_id;
uint16_t portid;
uint64_t req_rx_offloads, req_tx_offloads;
+ size_t sess_sz;
/* init EAL */
ret = rte_eal_init(argc, argv);
@@ -2094,7 +2100,10 @@ main(int32_t argc, char **argv)
nb_lcores = rte_lcore_count();
- /* Replicate each context per socket */
+ crypto_devid_fill();
+
+ sess_sz = max_session_size();
+
for (lcore_id = 0; lcore_id < RTE_MAX_LCORE; lcore_id++) {
if (rte_lcore_is_enabled(lcore_id) == 0)
continue;
@@ -2104,20 +2113,17 @@ main(int32_t argc, char **argv)
else
socket_id = 0;
+ /* mbuf_pool is initialised by the pool_init() function*/
if (socket_ctx[socket_id].mbuf_pool)
continue;
- /* initilaze SPD */
- sp4_init(&socket_ctx[socket_id], socket_id);
-
- sp6_init(&socket_ctx[socket_id], socket_id);
-
- /* initilaze SAD */
- sa_init(&socket_ctx[socket_id], socket_id);
-
- rt_init(&socket_ctx[socket_id], socket_id);
-
pool_init(&socket_ctx[socket_id], socket_id, NB_MBUF);
+ session_pool_init(&socket_ctx[socket_id], socket_id, sess_sz);
+ session_priv_pool_init(&socket_ctx[socket_id], socket_id,
+ sess_sz);
+
+ if (!numa_on)
+ break;
}
RTE_ETH_FOREACH_DEV(portid) {
@@ -2135,7 +2141,11 @@ main(int32_t argc, char **argv)
if ((enabled_port_mask & (1 << portid)) == 0)
continue;
- /* Start device */
+ /*
+ * Start device
+ * note: device must be started before a flow rule
+ * can be installed.
+ */
ret = rte_eth_dev_start(portid);
if (ret < 0)
rte_exit(EXIT_FAILURE, "rte_eth_dev_start: "
@@ -2153,6 +2163,19 @@ main(int32_t argc, char **argv)
RTE_ETH_EVENT_IPSEC, inline_ipsec_event_callback, NULL);
}
+ /* Replicate each context per socket */
+ for (i = 0; i < NB_SOCKETS && i < rte_socket_count(); i++) {
+ socket_id = rte_socket_id_by_idx(i);
+ if ((socket_ctx[socket_id].mbuf_pool != NULL) &&
+ (socket_ctx[socket_id].sa_in == NULL) &&
+ (socket_ctx[socket_id].sa_out == NULL)) {
+ sa_init(&socket_ctx[socket_id], socket_id);
+ sp4_init(&socket_ctx[socket_id], socket_id);
+ sp6_init(&socket_ctx[socket_id], socket_id);
+ rt_init(&socket_ctx[socket_id], socket_id);
+ }
+ }
+
check_all_ports_link_status(enabled_port_mask);
/* launch per-lcore init on every lcore */
diff --git a/examples/ipsec-secgw/ipsec.c b/examples/ipsec-secgw/ipsec.c
index 4352cb8..83080f7 100644
--- a/examples/ipsec-secgw/ipsec.c
+++ b/examples/ipsec-secgw/ipsec.c
@@ -39,42 +39,17 @@ set_ipsec_conf(struct ipsec_sa *sa, struct rte_security_ipsec_xform *ipsec)
ipsec->esn_soft_limit = IPSEC_OFFLOAD_ESN_SOFTLIMIT;
}
-int
-create_session(struct ipsec_ctx *ipsec_ctx, struct ipsec_sa *sa)
+static int
+create_sec_session(struct ipsec_sa *sa, struct rte_mempool *pool)
{
- struct rte_cryptodev_info cdev_info;
- unsigned long cdev_id_qp = 0;
+ const struct rte_security_capability *sec_cap;
+ struct rte_security_ctx *ctx;
int32_t ret = 0;
- struct cdev_key key = { 0 };
-
- key.lcore_id = (uint8_t)rte_lcore_id();
-
- key.cipher_algo = (uint8_t)sa->cipher_algo;
- key.auth_algo = (uint8_t)sa->auth_algo;
- key.aead_algo = (uint8_t)sa->aead_algo;
-
- if (sa->type == RTE_SECURITY_ACTION_TYPE_NONE) {
- ret = rte_hash_lookup_data(ipsec_ctx->cdev_map, &key,
- (void **)&cdev_id_qp);
- if (ret < 0) {
- RTE_LOG(ERR, IPSEC,
- "No cryptodev: core %u, cipher_algo %u, "
- "auth_algo %u, aead_algo %u\n",
- key.lcore_id,
- key.cipher_algo,
- key.auth_algo,
- key.aead_algo);
- return -1;
- }
- }
- RTE_LOG_DP(DEBUG, IPSEC, "Create session for SA spi %u on cryptodev "
- "%u qp %u\n", sa->spi,
- ipsec_ctx->tbl[cdev_id_qp].id,
- ipsec_ctx->tbl[cdev_id_qp].qp);
+ if ((sa == NULL) || (pool == NULL))
+ return -EINVAL;
- if (sa->type != RTE_SECURITY_ACTION_TYPE_NONE) {
- struct rte_security_session_conf sess_conf = {
+ struct rte_security_session_conf sess_conf = {
.action_type = sa->type,
.protocol = RTE_SECURITY_PROTOCOL_IPSEC,
{.ipsec = {
@@ -90,247 +65,340 @@ create_session(struct ipsec_ctx *ipsec_ctx, struct ipsec_sa *sa)
} },
.crypto_xform = sa->xforms,
.userdata = NULL,
-
};
- if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
- struct rte_security_ctx *ctx = (struct rte_security_ctx *)
- rte_cryptodev_get_sec_ctx(
- ipsec_ctx->tbl[cdev_id_qp].id);
-
- /* Set IPsec parameters in conf */
- set_ipsec_conf(sa, &(sess_conf.ipsec));
-
- sa->sec_session = rte_security_session_create(ctx,
- &sess_conf, ipsec_ctx->session_pool);
- if (sa->sec_session == NULL) {
- RTE_LOG(ERR, IPSEC,
- "SEC Session init failed: err: %d\n", ret);
- return -1;
- }
- } else if (sa->type == RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO) {
- struct rte_flow_error err;
- struct rte_security_ctx *ctx = (struct rte_security_ctx *)
- rte_eth_dev_get_sec_ctx(
- sa->portid);
- const struct rte_security_capability *sec_cap;
- int ret = 0;
-
- sa->sec_session = rte_security_session_create(ctx,
- &sess_conf, ipsec_ctx->session_pool);
- if (sa->sec_session == NULL) {
- RTE_LOG(ERR, IPSEC,
- "SEC Session init failed: err: %d\n", ret);
- return -1;
- }
+ if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
+ ctx = (struct rte_security_ctx *)
+ rte_eth_dev_get_sec_ctx(sa->portid);
- sec_cap = rte_security_capabilities_get(ctx);
+ /* Set IPsec parameters in conf */
+ set_ipsec_conf(sa, &(sess_conf.ipsec));
- /* iterate until ESP tunnel*/
- while (sec_cap->action !=
- RTE_SECURITY_ACTION_TYPE_NONE) {
+ sa->sec_session = rte_security_session_create(ctx,
+ &sess_conf, pool);
+ if (sa->sec_session == NULL) {
+ RTE_LOG(ERR, IPSEC,
+ "SEC Session init failed: err: %d\n",
+ ret);
+ return -1;
+ }
+ } else if (sa->type == RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO) {
+ struct rte_flow_error err;
+ ctx = (struct rte_security_ctx *)
+ rte_eth_dev_get_sec_ctx(sa->portid);
+ sa->sec_session = rte_security_session_create(ctx,
+ &sess_conf, pool);
+ if (sa->sec_session == NULL) {
+ RTE_LOG(ERR, IPSEC, "SEC Session init failed\n");
+ return -1;
+ }
- if (sec_cap->action == sa->type &&
- sec_cap->protocol ==
- RTE_SECURITY_PROTOCOL_IPSEC &&
- sec_cap->ipsec.mode ==
- RTE_SECURITY_IPSEC_SA_MODE_TUNNEL &&
- sec_cap->ipsec.direction == sa->direction)
- break;
- sec_cap++;
- }
+ sec_cap = rte_security_capabilities_get(ctx);
+
+ /* iterate until ESP tunnel*/
+ while (sec_cap->action != RTE_SECURITY_ACTION_TYPE_NONE) {
+ if (sec_cap->action == sa->type &&
+ sec_cap->protocol ==
+ RTE_SECURITY_PROTOCOL_IPSEC &&
+ sec_cap->ipsec.mode ==
+ RTE_SECURITY_IPSEC_SA_MODE_TUNNEL &&
+ sec_cap->ipsec.direction == sa->direction)
+ break;
+ sec_cap++;
+ }
- if (sec_cap->action == RTE_SECURITY_ACTION_TYPE_NONE) {
- RTE_LOG(ERR, IPSEC,
+ if (sec_cap->action == RTE_SECURITY_ACTION_TYPE_NONE) {
+ RTE_LOG(ERR, IPSEC,
"No suitable security capability found\n");
return -1;
- }
+ }
- sa->ol_flags = sec_cap->ol_flags;
- sa->security_ctx = ctx;
- sa->pattern[0].type = RTE_FLOW_ITEM_TYPE_ETH;
-
- sa->pattern[1].type = RTE_FLOW_ITEM_TYPE_IPV4;
- sa->pattern[1].mask = &rte_flow_item_ipv4_mask;
- if (sa->flags & IP6_TUNNEL) {
- sa->pattern[1].spec = &sa->ipv6_spec;
- memcpy(sa->ipv6_spec.hdr.dst_addr,
- sa->dst.ip.ip6.ip6_b, 16);
- memcpy(sa->ipv6_spec.hdr.src_addr,
- sa->src.ip.ip6.ip6_b, 16);
- } else {
- sa->pattern[1].spec = &sa->ipv4_spec;
- sa->ipv4_spec.hdr.dst_addr = sa->dst.ip.ip4;
- sa->ipv4_spec.hdr.src_addr = sa->src.ip.ip4;
- }
+ sa->ol_flags = sec_cap->ol_flags;
+ sa->security_ctx = ctx;
+ sa->pattern[0].type = RTE_FLOW_ITEM_TYPE_ETH;
+
+ sa->pattern[1].type = RTE_FLOW_ITEM_TYPE_IPV4;
+ sa->pattern[1].mask = &rte_flow_item_ipv4_mask;
+ if (sa->flags & IP6_TUNNEL) {
+ sa->pattern[1].spec = &sa->ipv6_spec;
+ memcpy(sa->ipv6_spec.hdr.dst_addr,
+ sa->dst.ip.ip6.ip6_b, 16);
+ memcpy(sa->ipv6_spec.hdr.src_addr,
+ sa->src.ip.ip6.ip6_b, 16);
+ } else {
+ sa->pattern[1].spec = &sa->ipv4_spec;
+ sa->ipv4_spec.hdr.dst_addr = sa->dst.ip.ip4;
+ sa->ipv4_spec.hdr.src_addr = sa->src.ip.ip4;
+ }
- sa->pattern[2].type = RTE_FLOW_ITEM_TYPE_ESP;
- sa->pattern[2].spec = &sa->esp_spec;
- sa->pattern[2].mask = &rte_flow_item_esp_mask;
- sa->esp_spec.hdr.spi = rte_cpu_to_be_32(sa->spi);
+ sa->pattern[2].type = RTE_FLOW_ITEM_TYPE_ESP;
+ sa->pattern[2].spec = &sa->esp_spec;
+ sa->pattern[2].mask = &rte_flow_item_esp_mask;
+ sa->esp_spec.hdr.spi = rte_cpu_to_be_32(sa->spi);
- sa->pattern[3].type = RTE_FLOW_ITEM_TYPE_END;
+ sa->pattern[3].type = RTE_FLOW_ITEM_TYPE_END;
- sa->action[0].type = RTE_FLOW_ACTION_TYPE_SECURITY;
- sa->action[0].conf = sa->sec_session;
+ sa->action[0].type = RTE_FLOW_ACTION_TYPE_SECURITY;
+ sa->action[0].conf = sa->sec_session;
- sa->action[1].type = RTE_FLOW_ACTION_TYPE_END;
+ sa->action[1].type = RTE_FLOW_ACTION_TYPE_END;
- sa->attr.egress = (sa->direction ==
+ sa->attr.egress = (sa->direction ==
RTE_SECURITY_IPSEC_SA_DIR_EGRESS);
- sa->attr.ingress = (sa->direction ==
+ sa->attr.ingress = (sa->direction ==
RTE_SECURITY_IPSEC_SA_DIR_INGRESS);
- if (sa->attr.ingress) {
- uint8_t rss_key[40];
- struct rte_eth_rss_conf rss_conf = {
- .rss_key = rss_key,
- .rss_key_len = 40,
- };
- struct rte_eth_dev *eth_dev;
- uint16_t queue[RTE_MAX_QUEUES_PER_PORT];
- struct rte_flow_action_rss action_rss;
- unsigned int i;
- unsigned int j;
-
- sa->action[2].type = RTE_FLOW_ACTION_TYPE_END;
- /* Try RSS. */
- sa->action[1].type = RTE_FLOW_ACTION_TYPE_RSS;
- sa->action[1].conf = &action_rss;
- eth_dev = ctx->device;
- rte_eth_dev_rss_hash_conf_get(sa->portid,
- &rss_conf);
- for (i = 0, j = 0;
- i < eth_dev->data->nb_rx_queues; ++i)
- if (eth_dev->data->rx_queues[i])
- queue[j++] = i;
- action_rss = (struct rte_flow_action_rss){
- .types = rss_conf.rss_hf,
- .key_len = rss_conf.rss_key_len,
- .queue_num = j,
- .key = rss_key,
- .queue = queue,
- };
- ret = rte_flow_validate(sa->portid, &sa->attr,
- sa->pattern, sa->action,
- &err);
- if (!ret)
- goto flow_create;
- /* Try Queue. */
- sa->action[1].type = RTE_FLOW_ACTION_TYPE_QUEUE;
- sa->action[1].conf =
- &(struct rte_flow_action_queue){
- .index = 0,
- };
- ret = rte_flow_validate(sa->portid, &sa->attr,
- sa->pattern, sa->action,
- &err);
- /* Try End. */
- sa->action[1].type = RTE_FLOW_ACTION_TYPE_END;
- sa->action[1].conf = NULL;
- ret = rte_flow_validate(sa->portid, &sa->attr,
+ if (sa->attr.ingress) {
+ uint8_t rss_key[40];
+ struct rte_eth_rss_conf rss_conf = {
+ .rss_key = rss_key,
+ .rss_key_len = 40,
+ };
+ struct rte_eth_dev *eth_dev;
+ uint16_t queue[RTE_MAX_QUEUES_PER_PORT];
+ struct rte_flow_action_rss action_rss;
+ unsigned int i;
+ unsigned int j;
+
+ sa->action[2].type = RTE_FLOW_ACTION_TYPE_END;
+ /* Try RSS. */
+ sa->action[1].type = RTE_FLOW_ACTION_TYPE_RSS;
+ sa->action[1].conf = &action_rss;
+ eth_dev = ctx->device;
+ rte_eth_dev_rss_hash_conf_get(sa->portid,
+ &rss_conf);
+ for (i = 0, j = 0;
+ i < eth_dev->data->nb_rx_queues; ++i)
+ if (eth_dev->data->rx_queues[i])
+ queue[j++] = i;
+ action_rss = (struct rte_flow_action_rss){
+ .types = rss_conf.rss_hf,
+ .key_len = rss_conf.rss_key_len,
+ .queue_num = j,
+ .key = rss_key,
+ .queue = queue,
+ };
+ ret = rte_flow_validate(sa->portid, &sa->attr,
+ sa->pattern, sa->action,
+ &err);
+ if (!ret)
+ goto flow_create;
+ /* Try Queue. */
+ sa->action[1].type = RTE_FLOW_ACTION_TYPE_QUEUE;
+ sa->action[1].conf =
+ &(struct rte_flow_action_queue){
+ .index = 0,
+ };
+ ret = rte_flow_validate(sa->portid, &sa->attr,
+ sa->pattern, sa->action,
+ &err);
+ /* Try End. */
+ sa->action[1].type = RTE_FLOW_ACTION_TYPE_END;
+ sa->action[1].conf = NULL;
+ ret = rte_flow_validate(sa->portid, &sa->attr,
sa->pattern, sa->action,
&err);
- if (ret)
- goto flow_create_failure;
- } else if (sa->attr.egress &&
- (sa->ol_flags &
- RTE_SECURITY_TX_HW_TRAILER_OFFLOAD)) {
- sa->action[1].type =
- RTE_FLOW_ACTION_TYPE_PASSTHRU;
- sa->action[2].type =
- RTE_FLOW_ACTION_TYPE_END;
- }
+ if (ret)
+ goto flow_create_failure;
+ } else if (sa->attr.egress &&
+ (sa->ol_flags &
+ RTE_SECURITY_TX_HW_TRAILER_OFFLOAD)) {
+ sa->action[1].type =
+ RTE_FLOW_ACTION_TYPE_PASSTHRU;
+ sa->action[2].type =
+ RTE_FLOW_ACTION_TYPE_END;
+ }
flow_create:
- sa->flow = rte_flow_create(sa->portid,
- &sa->attr, sa->pattern, sa->action, &err);
- if (sa->flow == NULL) {
+ sa->flow = rte_flow_create(sa->portid,
+ &sa->attr, sa->pattern, sa->action, &err);
+ if (sa->flow == NULL) {
flow_create_failure:
- RTE_LOG(ERR, IPSEC,
- "Failed to create ipsec flow msg: %s\n",
- err.message);
- return -1;
- }
- } else if (sa->type ==
- RTE_SECURITY_ACTION_TYPE_INLINE_PROTOCOL) {
- struct rte_security_ctx *ctx =
- (struct rte_security_ctx *)
- rte_eth_dev_get_sec_ctx(sa->portid);
- const struct rte_security_capability *sec_cap;
-
- if (ctx == NULL) {
- RTE_LOG(ERR, IPSEC,
- "Ethernet device doesn't have security features registered\n");
- return -1;
- }
+ RTE_LOG(ERR, IPSEC,
+ "Failed to create ipsec flow msg: %s\n",
+ err.message);
+ return -1;
+ }
+ } else if (sa->type == RTE_SECURITY_ACTION_TYPE_INLINE_PROTOCOL) {
+ struct rte_security_ctx *ctx =
+ (struct rte_security_ctx *)
+ rte_eth_dev_get_sec_ctx(sa->portid);
+ const struct rte_security_capability *sec_cap;
- /* Set IPsec parameters in conf */
- set_ipsec_conf(sa, &(sess_conf.ipsec));
-
- /* Save SA as userdata for the security session. When
- * the packet is received, this userdata will be
- * retrieved using the metadata from the packet.
- *
- * The PMD is expected to set similar metadata for other
- * operations, like rte_eth_event, which are tied to
- * security session. In such cases, the userdata could
- * be obtained to uniquely identify the security
- * parameters denoted.
- */
-
- sess_conf.userdata = (void *) sa;
-
- sa->sec_session = rte_security_session_create(ctx,
- &sess_conf, ipsec_ctx->session_pool);
- if (sa->sec_session == NULL) {
- RTE_LOG(ERR, IPSEC,
- "SEC Session init failed: err: %d\n", ret);
- return -1;
- }
+ if (ctx == NULL) {
+ RTE_LOG(ERR, IPSEC,
+ "Ethernet device doesn't have security features registered\n");
+ return -1;
+ }
+
+ /* Set IPsec parameters in conf */
+ set_ipsec_conf(sa, &(sess_conf.ipsec));
+
+ /* Save SA as userdata for the security session. When
+ * the packet is received, this userdata will be
+ * retrieved using the metadata from the packet.
+ *
+ * The PMD is expected to set similar metadata for other
+ * operations, like rte_eth_event, which are tied to
+ * security session. In such cases, the userdata could
+ * be obtained to uniquely identify the security
+ * parameters denoted.
+ */
+
+ sess_conf.userdata = (void *) sa;
+
+ sa->sec_session = rte_security_session_create(ctx,
+ &sess_conf, pool);
+ if (sa->sec_session == NULL) {
+ RTE_LOG(ERR, IPSEC,
+ "SEC Session init failed: err: %d\n", ret);
+ return -1;
+ }
- sec_cap = rte_security_capabilities_get(ctx);
+ sec_cap = rte_security_capabilities_get(ctx);
- if (sec_cap == NULL) {
- RTE_LOG(ERR, IPSEC,
- "No capabilities registered\n");
- return -1;
- }
+ if (sec_cap == NULL) {
+ RTE_LOG(ERR, IPSEC,
+ "No capabilities registered\n");
+ return -1;
+ }
- /* iterate until ESP tunnel*/
- while (sec_cap->action !=
+ /* iterate until ESP tunnel*/
+ while (sec_cap->action !=
RTE_SECURITY_ACTION_TYPE_NONE) {
- if (sec_cap->action == sa->type &&
- sec_cap->protocol ==
- RTE_SECURITY_PROTOCOL_IPSEC &&
- sec_cap->ipsec.mode ==
- RTE_SECURITY_IPSEC_SA_MODE_TUNNEL &&
- sec_cap->ipsec.direction == sa->direction)
- break;
- sec_cap++;
- }
+ if (sec_cap->action == sa->type &&
+ sec_cap->protocol ==
+ RTE_SECURITY_PROTOCOL_IPSEC &&
+ sec_cap->ipsec.mode ==
+ RTE_SECURITY_IPSEC_SA_MODE_TUNNEL &&
+ sec_cap->ipsec.direction == sa->direction)
+ break;
+ sec_cap++;
+ }
- if (sec_cap->action == RTE_SECURITY_ACTION_TYPE_NONE) {
- RTE_LOG(ERR, IPSEC,
- "No suitable security capability found\n");
- return -1;
- }
+ if (sec_cap->action == RTE_SECURITY_ACTION_TYPE_NONE) {
+ RTE_LOG(ERR, IPSEC,
+ "No suitable security capability found\n");
+ return -1;
+ }
+
+ sa->ol_flags = sec_cap->ol_flags;
+ sa->security_ctx = ctx;
+ } else
+ return -EINVAL;
+ return 0;
+}
+
+#define CDEV_IV_SIZE 12
+
+static int
+check_cryptodev_aead_capablity(const struct ipsec_sa *ss, uint8_t dev_id)
+{
+ struct rte_cryptodev_sym_capability_idx cap_idx;
+ const struct rte_cryptodev_symmetric_capability *cap;
- sa->ol_flags = sec_cap->ol_flags;
- sa->security_ctx = ctx;
+ cap_idx.type = RTE_CRYPTO_SYM_XFORM_AEAD;
+ cap_idx.algo.aead = ss->aead_algo;
+
+ cap = rte_cryptodev_sym_capability_get(dev_id, &cap_idx);
+ if (cap == NULL)
+ return -ENOENT;
+
+ return rte_cryptodev_sym_capability_check_aead(cap,
+ ss->cipher_key_len,
+ ss->digest_len,
+ ss->aad_len,
+ CDEV_IV_SIZE);
+}
+
+static int
+check_cryptodev_capablity(const struct ipsec_sa *ss, uint8_t dev_id)
+{
+ struct rte_cryptodev_sym_capability_idx cap_idx;
+ const struct rte_cryptodev_symmetric_capability *cap;
+ uint16_t auth_iv_len;
+ int rc = -1;
+
+ if (ss == NULL)
+ return rc;
+
+ if (ss->aead_algo == RTE_CRYPTO_AEAD_AES_GCM)
+ return check_cryptodev_aead_capablity(ss, dev_id);
+
+ auth_iv_len = 0;
+
+ cap_idx.type = RTE_CRYPTO_SYM_XFORM_AUTH;
+ cap_idx.algo.auth = ss->auth_algo;
+ cap = rte_cryptodev_sym_capability_get(dev_id, &cap_idx);
+ if (cap != NULL) {
+ rc = rte_cryptodev_sym_capability_check_auth(
+ cap, ss->auth_key_len, ss->digest_len,
+ auth_iv_len);
+ if (rc == 0) {
+ cap_idx.type = RTE_CRYPTO_SYM_XFORM_CIPHER;
+ cap_idx.algo.cipher = ss->cipher_algo;
+ cap = rte_cryptodev_sym_capability_get(dev_id,
+ &cap_idx);
+ if (cap != NULL)
+ rc = rte_cryptodev_sym_capability_check_cipher(
+ cap, ss->cipher_key_len,
+ ss->iv_len);
}
- } else {
- sa->crypto_session = rte_cryptodev_sym_session_create(
- ipsec_ctx->session_pool);
- rte_cryptodev_sym_session_init(ipsec_ctx->tbl[cdev_id_qp].id,
- sa->crypto_session, sa->xforms,
- ipsec_ctx->session_priv_pool);
-
- rte_cryptodev_info_get(ipsec_ctx->tbl[cdev_id_qp].id,
- &cdev_info);
}
- sa->cdev_id_qp = cdev_id_qp;
- return 0;
+ return rc;
+}
+
+static int
+create_crypto_session(struct ipsec_sa *sa, struct rte_mempool *pool)
+{
+ int32_t rc;
+ uint32_t devnum, i;
+ struct rte_cryptodev_sym_session *s;
+ uint8_t devid[RTE_CRYPTO_MAX_DEVS];
+
+ /* check which cryptodevs support SA */
+ devnum = 0;
+ for (i = 0; i < crypto_dev_num; i++) {
+ rc = check_cryptodev_capablity(sa, crypto_devid[i]);
+ if (rc == 0)
+ devid[devnum++] = crypto_devid[i];
+ }
+
+ if (devnum == 0)
+ return -ENODEV;
+
+ s = rte_cryptodev_sym_session_create(pool);
+ if (s == NULL)
+ return -ENOMEM;
+
+ /* initialize SA crypto session for all supported devices */
+ for (i = 0; i != devnum; i++) {
+ rc = rte_cryptodev_sym_session_init(devid[i], s, sa->xforms,
+ pool);
+ if (rc != 0)
+ break;
+ }
+
+ if (i == devnum) {
+ sa->crypto_session = s;
+ return 0;
+ }
+
+ /* failure, do cleanup */
+ while (i-- != 0)
+ rte_cryptodev_sym_session_clear(devid[i], s);
+
+ rte_cryptodev_sym_session_free(s);
+ return rc;
+}
+
+int
+create_session(struct ipsec_sa *sa, struct rte_mempool *pool)
+{
+ if (sa->type != RTE_SECURITY_ACTION_TYPE_NONE)
+ return create_sec_session(sa, pool);
+ else
+ return create_crypto_session(sa, pool);
}
/*
@@ -393,13 +461,6 @@ ipsec_enqueue(ipsec_xform_fn xform_func, struct ipsec_ctx *ipsec_ctx,
priv->cop.status = RTE_CRYPTO_OP_STATUS_NOT_PROCESSED;
rte_prefetch0(&priv->sym_cop);
-
- if ((unlikely(sa->sec_session == NULL)) &&
- create_session(ipsec_ctx, sa)) {
- rte_pktmbuf_free(pkts[i]);
- continue;
- }
-
sym_cop = get_sym_cop(&priv->cop);
sym_cop->m_src = pkts[i];
@@ -412,13 +473,6 @@ ipsec_enqueue(ipsec_xform_fn xform_func, struct ipsec_ctx *ipsec_ctx,
priv->cop.status = RTE_CRYPTO_OP_STATUS_NOT_PROCESSED;
rte_prefetch0(&priv->sym_cop);
-
- if ((unlikely(sa->crypto_session == NULL)) &&
- create_session(ipsec_ctx, sa)) {
- rte_pktmbuf_free(pkts[i]);
- continue;
- }
-
rte_crypto_op_attach_sym_session(&priv->cop,
sa->crypto_session);
@@ -429,12 +483,7 @@ ipsec_enqueue(ipsec_xform_fn xform_func, struct ipsec_ctx *ipsec_ctx,
}
break;
case RTE_SECURITY_ACTION_TYPE_INLINE_PROTOCOL:
- if ((unlikely(sa->sec_session == NULL)) &&
- create_session(ipsec_ctx, sa)) {
- rte_pktmbuf_free(pkts[i]);
- continue;
- }
-
+ RTE_ASSERT(sa->sec_session != NULL);
ipsec_ctx->ol_pkts[ipsec_ctx->ol_pkts_cnt++] = pkts[i];
if (sa->ol_flags & RTE_SECURITY_TX_OLOAD_NEED_MDATA)
rte_security_set_pkt_metadata(
@@ -442,17 +491,11 @@ ipsec_enqueue(ipsec_xform_fn xform_func, struct ipsec_ctx *ipsec_ctx,
sa->sec_session, pkts[i], NULL);
continue;
case RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO:
+ RTE_ASSERT(sa->sec_session != NULL);
priv->cop.type = RTE_CRYPTO_OP_TYPE_SYMMETRIC;
priv->cop.status = RTE_CRYPTO_OP_STATUS_NOT_PROCESSED;
rte_prefetch0(&priv->sym_cop);
-
- if ((unlikely(sa->sec_session == NULL)) &&
- create_session(ipsec_ctx, sa)) {
- rte_pktmbuf_free(pkts[i]);
- continue;
- }
-
rte_security_attach_session(&priv->cop,
sa->sec_session);
diff --git a/examples/ipsec-secgw/ipsec.h b/examples/ipsec-secgw/ipsec.h
index 99f49d6..804330c 100644
--- a/examples/ipsec-secgw/ipsec.h
+++ b/examples/ipsec-secgw/ipsec.h
@@ -83,6 +83,14 @@ struct app_sa_prm {
extern struct app_sa_prm app_sa_prm;
+/*
+ * Number of enabled crypto devices
+ * This number is needed when checking crypto device capabilities
+ */
+extern uint8_t crypto_dev_num;
+/* array of crypto device ID's */
+extern uint8_t crypto_devid[RTE_CRYPTO_MAX_DEVS];
+
struct ipsec_sa {
struct rte_ipsec_session ips; /* one session per sa for now */
uint32_t spi;
@@ -306,6 +314,6 @@ void
enqueue_cop_burst(struct cdev_qp *cqp);
int
-create_session(struct ipsec_ctx *ipsec_ctx, struct ipsec_sa *sa);
+create_session(struct ipsec_sa *sa, struct rte_mempool *pool);
#endif /* __IPSEC_H__ */
diff --git a/examples/ipsec-secgw/ipsec_process.c b/examples/ipsec-secgw/ipsec_process.c
index 3f9cacb..0df6969 100644
--- a/examples/ipsec-secgw/ipsec_process.c
+++ b/examples/ipsec-secgw/ipsec_process.c
@@ -86,39 +86,6 @@ enqueue_cop_bulk(struct cdev_qp *cqp, struct rte_crypto_op *cop[], uint32_t num)
cqp->len = len;
}
-static inline int
-fill_ipsec_session(struct rte_ipsec_session *ss, struct ipsec_ctx *ctx,
- struct ipsec_sa *sa)
-{
- int32_t rc;
-
- /* setup crypto section */
- if (ss->type == RTE_SECURITY_ACTION_TYPE_NONE) {
- if (sa->crypto_session == NULL) {
- rc = create_session(ctx, sa);
- if (rc != 0)
- return rc;
- }
- ss->crypto.ses = sa->crypto_session;
- /* setup session action type */
- } else {
- if (sa->sec_session == NULL) {
- rc = create_session(ctx, sa);
- if (rc != 0)
- return rc;
- }
- ss->security.ses = sa->sec_session;
- ss->security.ctx = sa->security_ctx;
- ss->security.ol_flags = sa->ol_flags;
- }
-
- rc = rte_ipsec_session_prepare(ss);
- if (rc != 0)
- memset(ss, 0, sizeof(*ss));
-
- return rc;
-}
-
/*
* group input packets byt the SA they belong to.
*/
@@ -219,9 +186,8 @@ ipsec_process(struct ipsec_ctx *ctx, struct ipsec_traffic *trf)
ips = &sa->ips;
- /* no valid HW session for that SA, try to create one */
- if (sa == NULL || (ips->crypto.ses == NULL &&
- fill_ipsec_session(ips, ctx, sa) != 0))
+ /* no valid HW session for that SA */
+ if (sa == NULL || ips->crypto.ses == NULL)
k = 0;
/* process packets inline */
diff --git a/examples/ipsec-secgw/sa.c b/examples/ipsec-secgw/sa.c
index a7298a3..0f36f5b 100644
--- a/examples/ipsec-secgw/sa.c
+++ b/examples/ipsec-secgw/sa.c
@@ -774,14 +774,14 @@ check_eth_dev_caps(uint16_t portid, uint32_t inbound)
return 0;
}
-
static int
sa_add_rules(struct sa_ctx *sa_ctx, const struct ipsec_sa entries[],
- uint32_t nb_entries, uint32_t inbound)
+ uint32_t nb_entries, uint32_t inbound, struct socket_ctx *skt_ctx)
{
struct ipsec_sa *sa;
uint32_t i, idx;
uint16_t iv_length, aad_length;
+ int32_t rc;
/* for ESN upper 32 bits of SQN also need to be part of AAD */
aad_length = (app_sa_prm.enable_esn != 0) ? sizeof(uint32_t) : 0;
@@ -902,6 +902,12 @@ sa_add_rules(struct sa_ctx *sa_ctx, const struct ipsec_sa entries[],
print_one_sa_rule(sa, inbound);
}
+ rc = create_session(sa, skt_ctx->session_pool);
+ if (rc != 0) {
+ RTE_LOG(ERR, IPSEC_ESP,
+ "create_session() failed\n");
+ return -EINVAL;
+ }
}
return 0;
@@ -909,16 +915,16 @@ sa_add_rules(struct sa_ctx *sa_ctx, const struct ipsec_sa entries[],
static inline int
sa_out_add_rules(struct sa_ctx *sa_ctx, const struct ipsec_sa entries[],
- uint32_t nb_entries)
+ uint32_t nb_entries, struct socket_ctx *skt_ctx)
{
- return sa_add_rules(sa_ctx, entries, nb_entries, 0);
+ return sa_add_rules(sa_ctx, entries, nb_entries, 0, skt_ctx);
}
static inline int
sa_in_add_rules(struct sa_ctx *sa_ctx, const struct ipsec_sa entries[],
- uint32_t nb_entries)
+ uint32_t nb_entries, struct socket_ctx *skt_ctx)
{
- return sa_add_rules(sa_ctx, entries, nb_entries, 1);
+ return sa_add_rules(sa_ctx, entries, nb_entries, 1, skt_ctx);
}
/*
@@ -1012,10 +1018,12 @@ fill_ipsec_sa_prm(struct rte_ipsec_sa_prm *prm, const struct ipsec_sa *ss,
return 0;
}
-static void
+static int
fill_ipsec_session(struct rte_ipsec_session *ss, struct rte_ipsec_sa *sa,
const struct ipsec_sa *lsa)
{
+ int32_t rc = 0;
+
ss->sa = sa;
ss->type = lsa->type;
@@ -1028,6 +1036,12 @@ fill_ipsec_session(struct rte_ipsec_session *ss, struct rte_ipsec_sa *sa,
ss->security.ctx = lsa->security_ctx;
ss->security.ol_flags = lsa->ol_flags;
}
+
+ rc = rte_ipsec_session_prepare(ss);
+ if (rc != 0)
+ memset(ss, 0, sizeof(*ss));
+
+ return rc;
}
/*
@@ -1062,8 +1076,8 @@ ipsec_sa_init(struct ipsec_sa *lsa, struct rte_ipsec_sa *sa, uint32_t sa_size)
if (rc < 0)
return rc;
- fill_ipsec_session(&lsa->ips, sa, lsa);
- return 0;
+ rc = fill_ipsec_session(&lsa->ips, sa, lsa);
+ return rc;
}
/*
@@ -1141,7 +1155,10 @@ sa_init(struct socket_ctx *ctx, int32_t socket_id)
"context %s in socket %d\n", rte_errno,
name, socket_id);
- sa_in_add_rules(ctx->sa_in, sa_in, nb_sa_in);
+ rc = sa_in_add_rules(ctx->sa_in, sa_in, nb_sa_in, ctx);
+ if (rc != 0)
+ rte_exit(EXIT_FAILURE,
+ "failed to add inbound rules\n");
if (app_sa_prm.enable != 0) {
rc = ipsec_satbl_init(ctx->sa_in, sa_in, nb_sa_in,
@@ -1161,7 +1178,10 @@ sa_init(struct socket_ctx *ctx, int32_t socket_id)
"context %s in socket %d\n", rte_errno,
name, socket_id);
- sa_out_add_rules(ctx->sa_out, sa_out, nb_sa_out);
+ rc = sa_out_add_rules(ctx->sa_out, sa_out, nb_sa_out, ctx);
+ if (rc != 0)
+ rte_exit(EXIT_FAILURE,
+ "failed to add outbound rules\n");
if (app_sa_prm.enable != 0) {
rc = ipsec_satbl_init(ctx->sa_out, sa_out, nb_sa_out,
--
2.7.4
^ permalink raw reply [flat|nested] 25+ messages in thread
* [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto
2019-04-17 13:42 ` [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto Bernard Iremonger
@ 2019-04-17 13:42 ` Bernard Iremonger
0 siblings, 0 replies; 25+ messages in thread
From: Bernard Iremonger @ 2019-04-17 13:42 UTC (permalink / raw)
To: dev, konstantin.ananyev, akhil.goyal; +Cc: Bernard Iremonger, stable
Inline crypto installs a flow rule in the NIC. This flow
rule must be installed before the first inbound packet is
received.
The create_session() function installs the flow rule.
Refactor ipsec-secgw.c, sa.c, ipsec.h and ipsec.c to create
sessions at startup to fix the issue of the first packet
being dropped for inline crypto.
The create_session() function is now called at initialisation in
sa_add_rules() which is called from sa_init().
The return code for add_rules is checked.
Calls to create_session() in other functions are dropped.
Add crypto_devid_fill() in ipsec-secgw.c
Add max_session_size() in ipsec-secgw.c
Add check_cryptodev_capability() in ipsec.c
Add check_cryptodev_aead_capability() in ipsec.c
Add create_sec_session() and create_crypto_session() in ipsec.c
The crypto_dev_fill() function has been added to find the
enabled crypto devices.
The max_session_size() function has been added to calculate memory
requirements.
The check_cryptodev_capability() and check_cryptodev_aead_capability()
functions have been added to check that the SA is supported by the
crypto device.
The create_session() function is refactored to use the
create_sec_session() and create_crypto_session() functions.
The cryprodev_init() function has been refactored to drop calls to
rte_mempool_create() and to drop calculation of memory requirements.
The main() function has been refactored to call crypto_devid_fill()
and max_session_size() and to call session_pool_init() and
session_priv_pool_init().
The ports are started now before adding a flow rule in main().
The sa_init(), sp4_init(), sp6_init() and rt_init() functions are
now called after the ports have been started.
Fixes: ec17993a145a ("examples/ipsec-secgw: support security offload")
Fixes: d299106e8e31 ("examples/ipsec-secgw: add IPsec sample application")
Cc: stable@dpdk.org
Signed-off-by: Bernard Iremonger <bernard.iremonger@intel.com>
Acked-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
---
examples/ipsec-secgw/ipsec-secgw.c | 271 ++++++++--------
examples/ipsec-secgw/ipsec.c | 581 +++++++++++++++++++----------------
examples/ipsec-secgw/ipsec.h | 10 +-
examples/ipsec-secgw/ipsec_process.c | 38 +--
examples/ipsec-secgw/sa.c | 42 ++-
5 files changed, 501 insertions(+), 441 deletions(-)
diff --git a/examples/ipsec-secgw/ipsec-secgw.c b/examples/ipsec-secgw/ipsec-secgw.c
index ffbd00b..cc8bb57 100644
--- a/examples/ipsec-secgw/ipsec-secgw.c
+++ b/examples/ipsec-secgw/ipsec-secgw.c
@@ -182,6 +182,14 @@ struct lcore_params {
uint8_t lcore_id;
} __rte_cache_aligned;
+/*
+ * Number of enabled crypto devices
+ * This number is needed when checking crypto device capabilities
+ */
+uint8_t crypto_dev_num;
+/* array of crypto device ID's */
+uint8_t crypto_devid[RTE_CRYPTO_MAX_DEVS];
+
static struct lcore_params lcore_params_array[MAX_LCORE_PARAMS];
static struct lcore_params *lcore_params;
@@ -1623,13 +1631,27 @@ check_cryptodev_mask(uint8_t cdev_id)
return -1;
}
+static void
+crypto_devid_fill(void)
+{
+ uint32_t i, n;
+
+ n = rte_cryptodev_count();
+
+ for (i = 0; i != n; i++) {
+ if (check_cryptodev_mask(i) == 0)
+ crypto_devid[crypto_dev_num++] = i;
+ }
+}
+
static int32_t
cryptodevs_init(void)
{
struct rte_cryptodev_config dev_conf;
struct rte_cryptodev_qp_conf qp_conf;
uint16_t idx, max_nb_qps, qp, i;
- int16_t cdev_id, port_id;
+ int16_t cdev_id;
+ uint32_t dev_max_sess;
struct rte_hash_parameters params = { 0 };
params.entries = CDEV_MAP_ENTRIES;
@@ -1652,45 +1674,6 @@ cryptodevs_init(void)
printf("lcore/cryptodev/qp mappings:\n");
- uint32_t max_sess_sz = 0, sess_sz;
- for (cdev_id = 0; cdev_id < rte_cryptodev_count(); cdev_id++) {
- void *sec_ctx;
-
- /* Get crypto priv session size */
- sess_sz = rte_cryptodev_sym_get_private_session_size(cdev_id);
- if (sess_sz > max_sess_sz)
- max_sess_sz = sess_sz;
-
- /*
- * If crypto device is security capable, need to check the
- * size of security session as well.
- */
-
- /* Get security context of the crypto device */
- sec_ctx = rte_cryptodev_get_sec_ctx(cdev_id);
- if (sec_ctx == NULL)
- continue;
-
- /* Get size of security session */
- sess_sz = rte_security_session_get_size(sec_ctx);
- if (sess_sz > max_sess_sz)
- max_sess_sz = sess_sz;
- }
- RTE_ETH_FOREACH_DEV(port_id) {
- void *sec_ctx;
-
- if ((enabled_port_mask & (1 << port_id)) == 0)
- continue;
-
- sec_ctx = rte_eth_dev_get_sec_ctx(port_id);
- if (sec_ctx == NULL)
- continue;
-
- sess_sz = rte_security_session_get_size(sec_ctx);
- if (sess_sz > max_sess_sz)
- max_sess_sz = sess_sz;
- }
-
idx = 0;
for (cdev_id = 0; cdev_id < rte_cryptodev_count(); cdev_id++) {
struct rte_cryptodev_info cdev_info;
@@ -1722,51 +1705,12 @@ cryptodevs_init(void)
dev_conf.socket_id = rte_cryptodev_socket_id(cdev_id);
dev_conf.nb_queue_pairs = qp;
- uint32_t dev_max_sess = cdev_info.sym.max_nb_sessions;
+ dev_max_sess = cdev_info.sym.max_nb_sessions;
if (dev_max_sess != 0 && dev_max_sess < CDEV_MP_NB_OBJS)
rte_exit(EXIT_FAILURE,
"Device does not support at least %u "
"sessions", CDEV_MP_NB_OBJS);
- if (!socket_ctx[dev_conf.socket_id].session_pool) {
- char mp_name[RTE_MEMPOOL_NAMESIZE];
- struct rte_mempool *sess_mp;
-
- snprintf(mp_name, RTE_MEMPOOL_NAMESIZE,
- "sess_mp_%u", dev_conf.socket_id);
- sess_mp = rte_cryptodev_sym_session_pool_create(
- mp_name, CDEV_MP_NB_OBJS,
- 0, CDEV_MP_CACHE_SZ, 0,
- dev_conf.socket_id);
- socket_ctx[dev_conf.socket_id].session_pool = sess_mp;
- }
-
- if (!socket_ctx[dev_conf.socket_id].session_priv_pool) {
- char mp_name[RTE_MEMPOOL_NAMESIZE];
- struct rte_mempool *sess_mp;
-
- snprintf(mp_name, RTE_MEMPOOL_NAMESIZE,
- "sess_mp_priv_%u", dev_conf.socket_id);
- sess_mp = rte_mempool_create(mp_name,
- CDEV_MP_NB_OBJS,
- max_sess_sz,
- CDEV_MP_CACHE_SZ,
- 0, NULL, NULL, NULL,
- NULL, dev_conf.socket_id,
- 0);
- socket_ctx[dev_conf.socket_id].session_priv_pool =
- sess_mp;
- }
-
- if (!socket_ctx[dev_conf.socket_id].session_priv_pool ||
- !socket_ctx[dev_conf.socket_id].session_pool)
- rte_exit(EXIT_FAILURE,
- "Cannot create session pool on socket %d\n",
- dev_conf.socket_id);
- else
- printf("Allocated session pool on socket %d\n",
- dev_conf.socket_id);
-
if (rte_cryptodev_configure(cdev_id, &dev_conf))
rte_panic("Failed to initialize cryptodev %u\n",
cdev_id);
@@ -1787,38 +1731,6 @@ cryptodevs_init(void)
cdev_id);
}
- /* create session pools for eth devices that implement security */
- RTE_ETH_FOREACH_DEV(port_id) {
- if ((enabled_port_mask & (1 << port_id)) &&
- rte_eth_dev_get_sec_ctx(port_id)) {
- int socket_id = rte_eth_dev_socket_id(port_id);
-
- if (!socket_ctx[socket_id].session_pool) {
- char mp_name[RTE_MEMPOOL_NAMESIZE];
- struct rte_mempool *sess_mp;
-
- snprintf(mp_name, RTE_MEMPOOL_NAMESIZE,
- "sess_mp_%u", socket_id);
- sess_mp = rte_mempool_create(mp_name,
- (CDEV_MP_NB_OBJS * 2),
- max_sess_sz,
- CDEV_MP_CACHE_SZ,
- 0, NULL, NULL, NULL,
- NULL, socket_id,
- 0);
- if (sess_mp == NULL)
- rte_exit(EXIT_FAILURE,
- "Cannot create session pool "
- "on socket %d\n", socket_id);
- else
- printf("Allocated session pool "
- "on socket %d\n", socket_id);
- socket_ctx[socket_id].session_pool = sess_mp;
- }
- }
- }
-
-
printf("\n");
return 0;
@@ -1984,6 +1896,98 @@ port_init(uint16_t portid, uint64_t req_rx_offloads, uint64_t req_tx_offloads)
printf("\n");
}
+static size_t
+max_session_size(void)
+{
+ size_t max_sz, sz;
+ void *sec_ctx;
+ int16_t cdev_id, port_id, n;
+
+ max_sz = 0;
+ n = rte_cryptodev_count();
+ for (cdev_id = 0; cdev_id != n; cdev_id++) {
+ sz = rte_cryptodev_sym_get_private_session_size(cdev_id);
+ if (sz > max_sz)
+ max_sz = sz;
+ /*
+ * If crypto device is security capable, need to check the
+ * size of security session as well.
+ */
+
+ /* Get security context of the crypto device */
+ sec_ctx = rte_cryptodev_get_sec_ctx(cdev_id);
+ if (sec_ctx == NULL)
+ continue;
+
+ /* Get size of security session */
+ sz = rte_security_session_get_size(sec_ctx);
+ if (sz > max_sz)
+ max_sz = sz;
+ }
+
+ RTE_ETH_FOREACH_DEV(port_id) {
+ if ((enabled_port_mask & (1 << port_id)) == 0)
+ continue;
+
+ sec_ctx = rte_eth_dev_get_sec_ctx(port_id);
+ if (sec_ctx == NULL)
+ continue;
+
+ sz = rte_security_session_get_size(sec_ctx);
+ if (sz > max_sz)
+ max_sz = sz;
+ }
+
+ return max_sz;
+}
+
+static void
+session_pool_init(struct socket_ctx *ctx, int32_t socket_id, size_t sess_sz)
+{
+ char mp_name[RTE_MEMPOOL_NAMESIZE];
+ struct rte_mempool *sess_mp;
+
+ snprintf(mp_name, RTE_MEMPOOL_NAMESIZE,
+ "sess_mp_%u", socket_id);
+ sess_mp = rte_cryptodev_sym_session_pool_create(
+ mp_name, CDEV_MP_NB_OBJS,
+ sess_sz, CDEV_MP_CACHE_SZ, 0,
+ socket_id);
+ ctx->session_pool = sess_mp;
+
+ if (ctx->session_pool == NULL)
+ rte_exit(EXIT_FAILURE,
+ "Cannot init session pool on socket %d\n", socket_id);
+ else
+ printf("Allocated session pool on socket %d\n", socket_id);
+}
+
+static void
+session_priv_pool_init(struct socket_ctx *ctx, int32_t socket_id,
+ size_t sess_sz)
+{
+ char mp_name[RTE_MEMPOOL_NAMESIZE];
+ struct rte_mempool *sess_mp;
+
+ snprintf(mp_name, RTE_MEMPOOL_NAMESIZE,
+ "sess_mp_priv_%u", socket_id);
+ sess_mp = rte_mempool_create(mp_name,
+ CDEV_MP_NB_OBJS,
+ sess_sz,
+ CDEV_MP_CACHE_SZ,
+ 0, NULL, NULL, NULL,
+ NULL, socket_id,
+ 0);
+ ctx->session_priv_pool = sess_mp;
+ if (ctx->session_priv_pool == NULL)
+ rte_exit(EXIT_FAILURE,
+ "Cannot init session priv pool on socket %d\n",
+ socket_id);
+ else
+ printf("Allocated session priv pool on socket %d\n",
+ socket_id);
+}
+
static void
pool_init(struct socket_ctx *ctx, int32_t socket_id, uint32_t nb_mbuf)
{
@@ -2064,9 +2068,11 @@ main(int32_t argc, char **argv)
{
int32_t ret;
uint32_t lcore_id;
+ uint32_t i;
uint8_t socket_id;
uint16_t portid;
uint64_t req_rx_offloads, req_tx_offloads;
+ size_t sess_sz;
/* init EAL */
ret = rte_eal_init(argc, argv);
@@ -2094,7 +2100,10 @@ main(int32_t argc, char **argv)
nb_lcores = rte_lcore_count();
- /* Replicate each context per socket */
+ crypto_devid_fill();
+
+ sess_sz = max_session_size();
+
for (lcore_id = 0; lcore_id < RTE_MAX_LCORE; lcore_id++) {
if (rte_lcore_is_enabled(lcore_id) == 0)
continue;
@@ -2104,20 +2113,17 @@ main(int32_t argc, char **argv)
else
socket_id = 0;
+ /* mbuf_pool is initialised by the pool_init() function*/
if (socket_ctx[socket_id].mbuf_pool)
continue;
- /* initilaze SPD */
- sp4_init(&socket_ctx[socket_id], socket_id);
-
- sp6_init(&socket_ctx[socket_id], socket_id);
-
- /* initilaze SAD */
- sa_init(&socket_ctx[socket_id], socket_id);
-
- rt_init(&socket_ctx[socket_id], socket_id);
-
pool_init(&socket_ctx[socket_id], socket_id, NB_MBUF);
+ session_pool_init(&socket_ctx[socket_id], socket_id, sess_sz);
+ session_priv_pool_init(&socket_ctx[socket_id], socket_id,
+ sess_sz);
+
+ if (!numa_on)
+ break;
}
RTE_ETH_FOREACH_DEV(portid) {
@@ -2135,7 +2141,11 @@ main(int32_t argc, char **argv)
if ((enabled_port_mask & (1 << portid)) == 0)
continue;
- /* Start device */
+ /*
+ * Start device
+ * note: device must be started before a flow rule
+ * can be installed.
+ */
ret = rte_eth_dev_start(portid);
if (ret < 0)
rte_exit(EXIT_FAILURE, "rte_eth_dev_start: "
@@ -2153,6 +2163,19 @@ main(int32_t argc, char **argv)
RTE_ETH_EVENT_IPSEC, inline_ipsec_event_callback, NULL);
}
+ /* Replicate each context per socket */
+ for (i = 0; i < NB_SOCKETS && i < rte_socket_count(); i++) {
+ socket_id = rte_socket_id_by_idx(i);
+ if ((socket_ctx[socket_id].mbuf_pool != NULL) &&
+ (socket_ctx[socket_id].sa_in == NULL) &&
+ (socket_ctx[socket_id].sa_out == NULL)) {
+ sa_init(&socket_ctx[socket_id], socket_id);
+ sp4_init(&socket_ctx[socket_id], socket_id);
+ sp6_init(&socket_ctx[socket_id], socket_id);
+ rt_init(&socket_ctx[socket_id], socket_id);
+ }
+ }
+
check_all_ports_link_status(enabled_port_mask);
/* launch per-lcore init on every lcore */
diff --git a/examples/ipsec-secgw/ipsec.c b/examples/ipsec-secgw/ipsec.c
index 4352cb8..83080f7 100644
--- a/examples/ipsec-secgw/ipsec.c
+++ b/examples/ipsec-secgw/ipsec.c
@@ -39,42 +39,17 @@ set_ipsec_conf(struct ipsec_sa *sa, struct rte_security_ipsec_xform *ipsec)
ipsec->esn_soft_limit = IPSEC_OFFLOAD_ESN_SOFTLIMIT;
}
-int
-create_session(struct ipsec_ctx *ipsec_ctx, struct ipsec_sa *sa)
+static int
+create_sec_session(struct ipsec_sa *sa, struct rte_mempool *pool)
{
- struct rte_cryptodev_info cdev_info;
- unsigned long cdev_id_qp = 0;
+ const struct rte_security_capability *sec_cap;
+ struct rte_security_ctx *ctx;
int32_t ret = 0;
- struct cdev_key key = { 0 };
-
- key.lcore_id = (uint8_t)rte_lcore_id();
-
- key.cipher_algo = (uint8_t)sa->cipher_algo;
- key.auth_algo = (uint8_t)sa->auth_algo;
- key.aead_algo = (uint8_t)sa->aead_algo;
-
- if (sa->type == RTE_SECURITY_ACTION_TYPE_NONE) {
- ret = rte_hash_lookup_data(ipsec_ctx->cdev_map, &key,
- (void **)&cdev_id_qp);
- if (ret < 0) {
- RTE_LOG(ERR, IPSEC,
- "No cryptodev: core %u, cipher_algo %u, "
- "auth_algo %u, aead_algo %u\n",
- key.lcore_id,
- key.cipher_algo,
- key.auth_algo,
- key.aead_algo);
- return -1;
- }
- }
- RTE_LOG_DP(DEBUG, IPSEC, "Create session for SA spi %u on cryptodev "
- "%u qp %u\n", sa->spi,
- ipsec_ctx->tbl[cdev_id_qp].id,
- ipsec_ctx->tbl[cdev_id_qp].qp);
+ if ((sa == NULL) || (pool == NULL))
+ return -EINVAL;
- if (sa->type != RTE_SECURITY_ACTION_TYPE_NONE) {
- struct rte_security_session_conf sess_conf = {
+ struct rte_security_session_conf sess_conf = {
.action_type = sa->type,
.protocol = RTE_SECURITY_PROTOCOL_IPSEC,
{.ipsec = {
@@ -90,247 +65,340 @@ create_session(struct ipsec_ctx *ipsec_ctx, struct ipsec_sa *sa)
} },
.crypto_xform = sa->xforms,
.userdata = NULL,
-
};
- if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
- struct rte_security_ctx *ctx = (struct rte_security_ctx *)
- rte_cryptodev_get_sec_ctx(
- ipsec_ctx->tbl[cdev_id_qp].id);
-
- /* Set IPsec parameters in conf */
- set_ipsec_conf(sa, &(sess_conf.ipsec));
-
- sa->sec_session = rte_security_session_create(ctx,
- &sess_conf, ipsec_ctx->session_pool);
- if (sa->sec_session == NULL) {
- RTE_LOG(ERR, IPSEC,
- "SEC Session init failed: err: %d\n", ret);
- return -1;
- }
- } else if (sa->type == RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO) {
- struct rte_flow_error err;
- struct rte_security_ctx *ctx = (struct rte_security_ctx *)
- rte_eth_dev_get_sec_ctx(
- sa->portid);
- const struct rte_security_capability *sec_cap;
- int ret = 0;
-
- sa->sec_session = rte_security_session_create(ctx,
- &sess_conf, ipsec_ctx->session_pool);
- if (sa->sec_session == NULL) {
- RTE_LOG(ERR, IPSEC,
- "SEC Session init failed: err: %d\n", ret);
- return -1;
- }
+ if (sa->type == RTE_SECURITY_ACTION_TYPE_LOOKASIDE_PROTOCOL) {
+ ctx = (struct rte_security_ctx *)
+ rte_eth_dev_get_sec_ctx(sa->portid);
- sec_cap = rte_security_capabilities_get(ctx);
+ /* Set IPsec parameters in conf */
+ set_ipsec_conf(sa, &(sess_conf.ipsec));
- /* iterate until ESP tunnel*/
- while (sec_cap->action !=
- RTE_SECURITY_ACTION_TYPE_NONE) {
+ sa->sec_session = rte_security_session_create(ctx,
+ &sess_conf, pool);
+ if (sa->sec_session == NULL) {
+ RTE_LOG(ERR, IPSEC,
+ "SEC Session init failed: err: %d\n",
+ ret);
+ return -1;
+ }
+ } else if (sa->type == RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO) {
+ struct rte_flow_error err;
+ ctx = (struct rte_security_ctx *)
+ rte_eth_dev_get_sec_ctx(sa->portid);
+ sa->sec_session = rte_security_session_create(ctx,
+ &sess_conf, pool);
+ if (sa->sec_session == NULL) {
+ RTE_LOG(ERR, IPSEC, "SEC Session init failed\n");
+ return -1;
+ }
- if (sec_cap->action == sa->type &&
- sec_cap->protocol ==
- RTE_SECURITY_PROTOCOL_IPSEC &&
- sec_cap->ipsec.mode ==
- RTE_SECURITY_IPSEC_SA_MODE_TUNNEL &&
- sec_cap->ipsec.direction == sa->direction)
- break;
- sec_cap++;
- }
+ sec_cap = rte_security_capabilities_get(ctx);
+
+ /* iterate until ESP tunnel*/
+ while (sec_cap->action != RTE_SECURITY_ACTION_TYPE_NONE) {
+ if (sec_cap->action == sa->type &&
+ sec_cap->protocol ==
+ RTE_SECURITY_PROTOCOL_IPSEC &&
+ sec_cap->ipsec.mode ==
+ RTE_SECURITY_IPSEC_SA_MODE_TUNNEL &&
+ sec_cap->ipsec.direction == sa->direction)
+ break;
+ sec_cap++;
+ }
- if (sec_cap->action == RTE_SECURITY_ACTION_TYPE_NONE) {
- RTE_LOG(ERR, IPSEC,
+ if (sec_cap->action == RTE_SECURITY_ACTION_TYPE_NONE) {
+ RTE_LOG(ERR, IPSEC,
"No suitable security capability found\n");
return -1;
- }
+ }
- sa->ol_flags = sec_cap->ol_flags;
- sa->security_ctx = ctx;
- sa->pattern[0].type = RTE_FLOW_ITEM_TYPE_ETH;
-
- sa->pattern[1].type = RTE_FLOW_ITEM_TYPE_IPV4;
- sa->pattern[1].mask = &rte_flow_item_ipv4_mask;
- if (sa->flags & IP6_TUNNEL) {
- sa->pattern[1].spec = &sa->ipv6_spec;
- memcpy(sa->ipv6_spec.hdr.dst_addr,
- sa->dst.ip.ip6.ip6_b, 16);
- memcpy(sa->ipv6_spec.hdr.src_addr,
- sa->src.ip.ip6.ip6_b, 16);
- } else {
- sa->pattern[1].spec = &sa->ipv4_spec;
- sa->ipv4_spec.hdr.dst_addr = sa->dst.ip.ip4;
- sa->ipv4_spec.hdr.src_addr = sa->src.ip.ip4;
- }
+ sa->ol_flags = sec_cap->ol_flags;
+ sa->security_ctx = ctx;
+ sa->pattern[0].type = RTE_FLOW_ITEM_TYPE_ETH;
+
+ sa->pattern[1].type = RTE_FLOW_ITEM_TYPE_IPV4;
+ sa->pattern[1].mask = &rte_flow_item_ipv4_mask;
+ if (sa->flags & IP6_TUNNEL) {
+ sa->pattern[1].spec = &sa->ipv6_spec;
+ memcpy(sa->ipv6_spec.hdr.dst_addr,
+ sa->dst.ip.ip6.ip6_b, 16);
+ memcpy(sa->ipv6_spec.hdr.src_addr,
+ sa->src.ip.ip6.ip6_b, 16);
+ } else {
+ sa->pattern[1].spec = &sa->ipv4_spec;
+ sa->ipv4_spec.hdr.dst_addr = sa->dst.ip.ip4;
+ sa->ipv4_spec.hdr.src_addr = sa->src.ip.ip4;
+ }
- sa->pattern[2].type = RTE_FLOW_ITEM_TYPE_ESP;
- sa->pattern[2].spec = &sa->esp_spec;
- sa->pattern[2].mask = &rte_flow_item_esp_mask;
- sa->esp_spec.hdr.spi = rte_cpu_to_be_32(sa->spi);
+ sa->pattern[2].type = RTE_FLOW_ITEM_TYPE_ESP;
+ sa->pattern[2].spec = &sa->esp_spec;
+ sa->pattern[2].mask = &rte_flow_item_esp_mask;
+ sa->esp_spec.hdr.spi = rte_cpu_to_be_32(sa->spi);
- sa->pattern[3].type = RTE_FLOW_ITEM_TYPE_END;
+ sa->pattern[3].type = RTE_FLOW_ITEM_TYPE_END;
- sa->action[0].type = RTE_FLOW_ACTION_TYPE_SECURITY;
- sa->action[0].conf = sa->sec_session;
+ sa->action[0].type = RTE_FLOW_ACTION_TYPE_SECURITY;
+ sa->action[0].conf = sa->sec_session;
- sa->action[1].type = RTE_FLOW_ACTION_TYPE_END;
+ sa->action[1].type = RTE_FLOW_ACTION_TYPE_END;
- sa->attr.egress = (sa->direction ==
+ sa->attr.egress = (sa->direction ==
RTE_SECURITY_IPSEC_SA_DIR_EGRESS);
- sa->attr.ingress = (sa->direction ==
+ sa->attr.ingress = (sa->direction ==
RTE_SECURITY_IPSEC_SA_DIR_INGRESS);
- if (sa->attr.ingress) {
- uint8_t rss_key[40];
- struct rte_eth_rss_conf rss_conf = {
- .rss_key = rss_key,
- .rss_key_len = 40,
- };
- struct rte_eth_dev *eth_dev;
- uint16_t queue[RTE_MAX_QUEUES_PER_PORT];
- struct rte_flow_action_rss action_rss;
- unsigned int i;
- unsigned int j;
-
- sa->action[2].type = RTE_FLOW_ACTION_TYPE_END;
- /* Try RSS. */
- sa->action[1].type = RTE_FLOW_ACTION_TYPE_RSS;
- sa->action[1].conf = &action_rss;
- eth_dev = ctx->device;
- rte_eth_dev_rss_hash_conf_get(sa->portid,
- &rss_conf);
- for (i = 0, j = 0;
- i < eth_dev->data->nb_rx_queues; ++i)
- if (eth_dev->data->rx_queues[i])
- queue[j++] = i;
- action_rss = (struct rte_flow_action_rss){
- .types = rss_conf.rss_hf,
- .key_len = rss_conf.rss_key_len,
- .queue_num = j,
- .key = rss_key,
- .queue = queue,
- };
- ret = rte_flow_validate(sa->portid, &sa->attr,
- sa->pattern, sa->action,
- &err);
- if (!ret)
- goto flow_create;
- /* Try Queue. */
- sa->action[1].type = RTE_FLOW_ACTION_TYPE_QUEUE;
- sa->action[1].conf =
- &(struct rte_flow_action_queue){
- .index = 0,
- };
- ret = rte_flow_validate(sa->portid, &sa->attr,
- sa->pattern, sa->action,
- &err);
- /* Try End. */
- sa->action[1].type = RTE_FLOW_ACTION_TYPE_END;
- sa->action[1].conf = NULL;
- ret = rte_flow_validate(sa->portid, &sa->attr,
+ if (sa->attr.ingress) {
+ uint8_t rss_key[40];
+ struct rte_eth_rss_conf rss_conf = {
+ .rss_key = rss_key,
+ .rss_key_len = 40,
+ };
+ struct rte_eth_dev *eth_dev;
+ uint16_t queue[RTE_MAX_QUEUES_PER_PORT];
+ struct rte_flow_action_rss action_rss;
+ unsigned int i;
+ unsigned int j;
+
+ sa->action[2].type = RTE_FLOW_ACTION_TYPE_END;
+ /* Try RSS. */
+ sa->action[1].type = RTE_FLOW_ACTION_TYPE_RSS;
+ sa->action[1].conf = &action_rss;
+ eth_dev = ctx->device;
+ rte_eth_dev_rss_hash_conf_get(sa->portid,
+ &rss_conf);
+ for (i = 0, j = 0;
+ i < eth_dev->data->nb_rx_queues; ++i)
+ if (eth_dev->data->rx_queues[i])
+ queue[j++] = i;
+ action_rss = (struct rte_flow_action_rss){
+ .types = rss_conf.rss_hf,
+ .key_len = rss_conf.rss_key_len,
+ .queue_num = j,
+ .key = rss_key,
+ .queue = queue,
+ };
+ ret = rte_flow_validate(sa->portid, &sa->attr,
+ sa->pattern, sa->action,
+ &err);
+ if (!ret)
+ goto flow_create;
+ /* Try Queue. */
+ sa->action[1].type = RTE_FLOW_ACTION_TYPE_QUEUE;
+ sa->action[1].conf =
+ &(struct rte_flow_action_queue){
+ .index = 0,
+ };
+ ret = rte_flow_validate(sa->portid, &sa->attr,
+ sa->pattern, sa->action,
+ &err);
+ /* Try End. */
+ sa->action[1].type = RTE_FLOW_ACTION_TYPE_END;
+ sa->action[1].conf = NULL;
+ ret = rte_flow_validate(sa->portid, &sa->attr,
sa->pattern, sa->action,
&err);
- if (ret)
- goto flow_create_failure;
- } else if (sa->attr.egress &&
- (sa->ol_flags &
- RTE_SECURITY_TX_HW_TRAILER_OFFLOAD)) {
- sa->action[1].type =
- RTE_FLOW_ACTION_TYPE_PASSTHRU;
- sa->action[2].type =
- RTE_FLOW_ACTION_TYPE_END;
- }
+ if (ret)
+ goto flow_create_failure;
+ } else if (sa->attr.egress &&
+ (sa->ol_flags &
+ RTE_SECURITY_TX_HW_TRAILER_OFFLOAD)) {
+ sa->action[1].type =
+ RTE_FLOW_ACTION_TYPE_PASSTHRU;
+ sa->action[2].type =
+ RTE_FLOW_ACTION_TYPE_END;
+ }
flow_create:
- sa->flow = rte_flow_create(sa->portid,
- &sa->attr, sa->pattern, sa->action, &err);
- if (sa->flow == NULL) {
+ sa->flow = rte_flow_create(sa->portid,
+ &sa->attr, sa->pattern, sa->action, &err);
+ if (sa->flow == NULL) {
flow_create_failure:
- RTE_LOG(ERR, IPSEC,
- "Failed to create ipsec flow msg: %s\n",
- err.message);
- return -1;
- }
- } else if (sa->type ==
- RTE_SECURITY_ACTION_TYPE_INLINE_PROTOCOL) {
- struct rte_security_ctx *ctx =
- (struct rte_security_ctx *)
- rte_eth_dev_get_sec_ctx(sa->portid);
- const struct rte_security_capability *sec_cap;
-
- if (ctx == NULL) {
- RTE_LOG(ERR, IPSEC,
- "Ethernet device doesn't have security features registered\n");
- return -1;
- }
+ RTE_LOG(ERR, IPSEC,
+ "Failed to create ipsec flow msg: %s\n",
+ err.message);
+ return -1;
+ }
+ } else if (sa->type == RTE_SECURITY_ACTION_TYPE_INLINE_PROTOCOL) {
+ struct rte_security_ctx *ctx =
+ (struct rte_security_ctx *)
+ rte_eth_dev_get_sec_ctx(sa->portid);
+ const struct rte_security_capability *sec_cap;
- /* Set IPsec parameters in conf */
- set_ipsec_conf(sa, &(sess_conf.ipsec));
-
- /* Save SA as userdata for the security session. When
- * the packet is received, this userdata will be
- * retrieved using the metadata from the packet.
- *
- * The PMD is expected to set similar metadata for other
- * operations, like rte_eth_event, which are tied to
- * security session. In such cases, the userdata could
- * be obtained to uniquely identify the security
- * parameters denoted.
- */
-
- sess_conf.userdata = (void *) sa;
-
- sa->sec_session = rte_security_session_create(ctx,
- &sess_conf, ipsec_ctx->session_pool);
- if (sa->sec_session == NULL) {
- RTE_LOG(ERR, IPSEC,
- "SEC Session init failed: err: %d\n", ret);
- return -1;
- }
+ if (ctx == NULL) {
+ RTE_LOG(ERR, IPSEC,
+ "Ethernet device doesn't have security features registered\n");
+ return -1;
+ }
+
+ /* Set IPsec parameters in conf */
+ set_ipsec_conf(sa, &(sess_conf.ipsec));
+
+ /* Save SA as userdata for the security session. When
+ * the packet is received, this userdata will be
+ * retrieved using the metadata from the packet.
+ *
+ * The PMD is expected to set similar metadata for other
+ * operations, like rte_eth_event, which are tied to
+ * security session. In such cases, the userdata could
+ * be obtained to uniquely identify the security
+ * parameters denoted.
+ */
+
+ sess_conf.userdata = (void *) sa;
+
+ sa->sec_session = rte_security_session_create(ctx,
+ &sess_conf, pool);
+ if (sa->sec_session == NULL) {
+ RTE_LOG(ERR, IPSEC,
+ "SEC Session init failed: err: %d\n", ret);
+ return -1;
+ }
- sec_cap = rte_security_capabilities_get(ctx);
+ sec_cap = rte_security_capabilities_get(ctx);
- if (sec_cap == NULL) {
- RTE_LOG(ERR, IPSEC,
- "No capabilities registered\n");
- return -1;
- }
+ if (sec_cap == NULL) {
+ RTE_LOG(ERR, IPSEC,
+ "No capabilities registered\n");
+ return -1;
+ }
- /* iterate until ESP tunnel*/
- while (sec_cap->action !=
+ /* iterate until ESP tunnel*/
+ while (sec_cap->action !=
RTE_SECURITY_ACTION_TYPE_NONE) {
- if (sec_cap->action == sa->type &&
- sec_cap->protocol ==
- RTE_SECURITY_PROTOCOL_IPSEC &&
- sec_cap->ipsec.mode ==
- RTE_SECURITY_IPSEC_SA_MODE_TUNNEL &&
- sec_cap->ipsec.direction == sa->direction)
- break;
- sec_cap++;
- }
+ if (sec_cap->action == sa->type &&
+ sec_cap->protocol ==
+ RTE_SECURITY_PROTOCOL_IPSEC &&
+ sec_cap->ipsec.mode ==
+ RTE_SECURITY_IPSEC_SA_MODE_TUNNEL &&
+ sec_cap->ipsec.direction == sa->direction)
+ break;
+ sec_cap++;
+ }
- if (sec_cap->action == RTE_SECURITY_ACTION_TYPE_NONE) {
- RTE_LOG(ERR, IPSEC,
- "No suitable security capability found\n");
- return -1;
- }
+ if (sec_cap->action == RTE_SECURITY_ACTION_TYPE_NONE) {
+ RTE_LOG(ERR, IPSEC,
+ "No suitable security capability found\n");
+ return -1;
+ }
+
+ sa->ol_flags = sec_cap->ol_flags;
+ sa->security_ctx = ctx;
+ } else
+ return -EINVAL;
+ return 0;
+}
+
+#define CDEV_IV_SIZE 12
+
+static int
+check_cryptodev_aead_capablity(const struct ipsec_sa *ss, uint8_t dev_id)
+{
+ struct rte_cryptodev_sym_capability_idx cap_idx;
+ const struct rte_cryptodev_symmetric_capability *cap;
- sa->ol_flags = sec_cap->ol_flags;
- sa->security_ctx = ctx;
+ cap_idx.type = RTE_CRYPTO_SYM_XFORM_AEAD;
+ cap_idx.algo.aead = ss->aead_algo;
+
+ cap = rte_cryptodev_sym_capability_get(dev_id, &cap_idx);
+ if (cap == NULL)
+ return -ENOENT;
+
+ return rte_cryptodev_sym_capability_check_aead(cap,
+ ss->cipher_key_len,
+ ss->digest_len,
+ ss->aad_len,
+ CDEV_IV_SIZE);
+}
+
+static int
+check_cryptodev_capablity(const struct ipsec_sa *ss, uint8_t dev_id)
+{
+ struct rte_cryptodev_sym_capability_idx cap_idx;
+ const struct rte_cryptodev_symmetric_capability *cap;
+ uint16_t auth_iv_len;
+ int rc = -1;
+
+ if (ss == NULL)
+ return rc;
+
+ if (ss->aead_algo == RTE_CRYPTO_AEAD_AES_GCM)
+ return check_cryptodev_aead_capablity(ss, dev_id);
+
+ auth_iv_len = 0;
+
+ cap_idx.type = RTE_CRYPTO_SYM_XFORM_AUTH;
+ cap_idx.algo.auth = ss->auth_algo;
+ cap = rte_cryptodev_sym_capability_get(dev_id, &cap_idx);
+ if (cap != NULL) {
+ rc = rte_cryptodev_sym_capability_check_auth(
+ cap, ss->auth_key_len, ss->digest_len,
+ auth_iv_len);
+ if (rc == 0) {
+ cap_idx.type = RTE_CRYPTO_SYM_XFORM_CIPHER;
+ cap_idx.algo.cipher = ss->cipher_algo;
+ cap = rte_cryptodev_sym_capability_get(dev_id,
+ &cap_idx);
+ if (cap != NULL)
+ rc = rte_cryptodev_sym_capability_check_cipher(
+ cap, ss->cipher_key_len,
+ ss->iv_len);
}
- } else {
- sa->crypto_session = rte_cryptodev_sym_session_create(
- ipsec_ctx->session_pool);
- rte_cryptodev_sym_session_init(ipsec_ctx->tbl[cdev_id_qp].id,
- sa->crypto_session, sa->xforms,
- ipsec_ctx->session_priv_pool);
-
- rte_cryptodev_info_get(ipsec_ctx->tbl[cdev_id_qp].id,
- &cdev_info);
}
- sa->cdev_id_qp = cdev_id_qp;
- return 0;
+ return rc;
+}
+
+static int
+create_crypto_session(struct ipsec_sa *sa, struct rte_mempool *pool)
+{
+ int32_t rc;
+ uint32_t devnum, i;
+ struct rte_cryptodev_sym_session *s;
+ uint8_t devid[RTE_CRYPTO_MAX_DEVS];
+
+ /* check which cryptodevs support SA */
+ devnum = 0;
+ for (i = 0; i < crypto_dev_num; i++) {
+ rc = check_cryptodev_capablity(sa, crypto_devid[i]);
+ if (rc == 0)
+ devid[devnum++] = crypto_devid[i];
+ }
+
+ if (devnum == 0)
+ return -ENODEV;
+
+ s = rte_cryptodev_sym_session_create(pool);
+ if (s == NULL)
+ return -ENOMEM;
+
+ /* initialize SA crypto session for all supported devices */
+ for (i = 0; i != devnum; i++) {
+ rc = rte_cryptodev_sym_session_init(devid[i], s, sa->xforms,
+ pool);
+ if (rc != 0)
+ break;
+ }
+
+ if (i == devnum) {
+ sa->crypto_session = s;
+ return 0;
+ }
+
+ /* failure, do cleanup */
+ while (i-- != 0)
+ rte_cryptodev_sym_session_clear(devid[i], s);
+
+ rte_cryptodev_sym_session_free(s);
+ return rc;
+}
+
+int
+create_session(struct ipsec_sa *sa, struct rte_mempool *pool)
+{
+ if (sa->type != RTE_SECURITY_ACTION_TYPE_NONE)
+ return create_sec_session(sa, pool);
+ else
+ return create_crypto_session(sa, pool);
}
/*
@@ -393,13 +461,6 @@ ipsec_enqueue(ipsec_xform_fn xform_func, struct ipsec_ctx *ipsec_ctx,
priv->cop.status = RTE_CRYPTO_OP_STATUS_NOT_PROCESSED;
rte_prefetch0(&priv->sym_cop);
-
- if ((unlikely(sa->sec_session == NULL)) &&
- create_session(ipsec_ctx, sa)) {
- rte_pktmbuf_free(pkts[i]);
- continue;
- }
-
sym_cop = get_sym_cop(&priv->cop);
sym_cop->m_src = pkts[i];
@@ -412,13 +473,6 @@ ipsec_enqueue(ipsec_xform_fn xform_func, struct ipsec_ctx *ipsec_ctx,
priv->cop.status = RTE_CRYPTO_OP_STATUS_NOT_PROCESSED;
rte_prefetch0(&priv->sym_cop);
-
- if ((unlikely(sa->crypto_session == NULL)) &&
- create_session(ipsec_ctx, sa)) {
- rte_pktmbuf_free(pkts[i]);
- continue;
- }
-
rte_crypto_op_attach_sym_session(&priv->cop,
sa->crypto_session);
@@ -429,12 +483,7 @@ ipsec_enqueue(ipsec_xform_fn xform_func, struct ipsec_ctx *ipsec_ctx,
}
break;
case RTE_SECURITY_ACTION_TYPE_INLINE_PROTOCOL:
- if ((unlikely(sa->sec_session == NULL)) &&
- create_session(ipsec_ctx, sa)) {
- rte_pktmbuf_free(pkts[i]);
- continue;
- }
-
+ RTE_ASSERT(sa->sec_session != NULL);
ipsec_ctx->ol_pkts[ipsec_ctx->ol_pkts_cnt++] = pkts[i];
if (sa->ol_flags & RTE_SECURITY_TX_OLOAD_NEED_MDATA)
rte_security_set_pkt_metadata(
@@ -442,17 +491,11 @@ ipsec_enqueue(ipsec_xform_fn xform_func, struct ipsec_ctx *ipsec_ctx,
sa->sec_session, pkts[i], NULL);
continue;
case RTE_SECURITY_ACTION_TYPE_INLINE_CRYPTO:
+ RTE_ASSERT(sa->sec_session != NULL);
priv->cop.type = RTE_CRYPTO_OP_TYPE_SYMMETRIC;
priv->cop.status = RTE_CRYPTO_OP_STATUS_NOT_PROCESSED;
rte_prefetch0(&priv->sym_cop);
-
- if ((unlikely(sa->sec_session == NULL)) &&
- create_session(ipsec_ctx, sa)) {
- rte_pktmbuf_free(pkts[i]);
- continue;
- }
-
rte_security_attach_session(&priv->cop,
sa->sec_session);
diff --git a/examples/ipsec-secgw/ipsec.h b/examples/ipsec-secgw/ipsec.h
index 99f49d6..804330c 100644
--- a/examples/ipsec-secgw/ipsec.h
+++ b/examples/ipsec-secgw/ipsec.h
@@ -83,6 +83,14 @@ struct app_sa_prm {
extern struct app_sa_prm app_sa_prm;
+/*
+ * Number of enabled crypto devices
+ * This number is needed when checking crypto device capabilities
+ */
+extern uint8_t crypto_dev_num;
+/* array of crypto device ID's */
+extern uint8_t crypto_devid[RTE_CRYPTO_MAX_DEVS];
+
struct ipsec_sa {
struct rte_ipsec_session ips; /* one session per sa for now */
uint32_t spi;
@@ -306,6 +314,6 @@ void
enqueue_cop_burst(struct cdev_qp *cqp);
int
-create_session(struct ipsec_ctx *ipsec_ctx, struct ipsec_sa *sa);
+create_session(struct ipsec_sa *sa, struct rte_mempool *pool);
#endif /* __IPSEC_H__ */
diff --git a/examples/ipsec-secgw/ipsec_process.c b/examples/ipsec-secgw/ipsec_process.c
index 3f9cacb..0df6969 100644
--- a/examples/ipsec-secgw/ipsec_process.c
+++ b/examples/ipsec-secgw/ipsec_process.c
@@ -86,39 +86,6 @@ enqueue_cop_bulk(struct cdev_qp *cqp, struct rte_crypto_op *cop[], uint32_t num)
cqp->len = len;
}
-static inline int
-fill_ipsec_session(struct rte_ipsec_session *ss, struct ipsec_ctx *ctx,
- struct ipsec_sa *sa)
-{
- int32_t rc;
-
- /* setup crypto section */
- if (ss->type == RTE_SECURITY_ACTION_TYPE_NONE) {
- if (sa->crypto_session == NULL) {
- rc = create_session(ctx, sa);
- if (rc != 0)
- return rc;
- }
- ss->crypto.ses = sa->crypto_session;
- /* setup session action type */
- } else {
- if (sa->sec_session == NULL) {
- rc = create_session(ctx, sa);
- if (rc != 0)
- return rc;
- }
- ss->security.ses = sa->sec_session;
- ss->security.ctx = sa->security_ctx;
- ss->security.ol_flags = sa->ol_flags;
- }
-
- rc = rte_ipsec_session_prepare(ss);
- if (rc != 0)
- memset(ss, 0, sizeof(*ss));
-
- return rc;
-}
-
/*
* group input packets byt the SA they belong to.
*/
@@ -219,9 +186,8 @@ ipsec_process(struct ipsec_ctx *ctx, struct ipsec_traffic *trf)
ips = &sa->ips;
- /* no valid HW session for that SA, try to create one */
- if (sa == NULL || (ips->crypto.ses == NULL &&
- fill_ipsec_session(ips, ctx, sa) != 0))
+ /* no valid HW session for that SA */
+ if (sa == NULL || ips->crypto.ses == NULL)
k = 0;
/* process packets inline */
diff --git a/examples/ipsec-secgw/sa.c b/examples/ipsec-secgw/sa.c
index a7298a3..0f36f5b 100644
--- a/examples/ipsec-secgw/sa.c
+++ b/examples/ipsec-secgw/sa.c
@@ -774,14 +774,14 @@ check_eth_dev_caps(uint16_t portid, uint32_t inbound)
return 0;
}
-
static int
sa_add_rules(struct sa_ctx *sa_ctx, const struct ipsec_sa entries[],
- uint32_t nb_entries, uint32_t inbound)
+ uint32_t nb_entries, uint32_t inbound, struct socket_ctx *skt_ctx)
{
struct ipsec_sa *sa;
uint32_t i, idx;
uint16_t iv_length, aad_length;
+ int32_t rc;
/* for ESN upper 32 bits of SQN also need to be part of AAD */
aad_length = (app_sa_prm.enable_esn != 0) ? sizeof(uint32_t) : 0;
@@ -902,6 +902,12 @@ sa_add_rules(struct sa_ctx *sa_ctx, const struct ipsec_sa entries[],
print_one_sa_rule(sa, inbound);
}
+ rc = create_session(sa, skt_ctx->session_pool);
+ if (rc != 0) {
+ RTE_LOG(ERR, IPSEC_ESP,
+ "create_session() failed\n");
+ return -EINVAL;
+ }
}
return 0;
@@ -909,16 +915,16 @@ sa_add_rules(struct sa_ctx *sa_ctx, const struct ipsec_sa entries[],
static inline int
sa_out_add_rules(struct sa_ctx *sa_ctx, const struct ipsec_sa entries[],
- uint32_t nb_entries)
+ uint32_t nb_entries, struct socket_ctx *skt_ctx)
{
- return sa_add_rules(sa_ctx, entries, nb_entries, 0);
+ return sa_add_rules(sa_ctx, entries, nb_entries, 0, skt_ctx);
}
static inline int
sa_in_add_rules(struct sa_ctx *sa_ctx, const struct ipsec_sa entries[],
- uint32_t nb_entries)
+ uint32_t nb_entries, struct socket_ctx *skt_ctx)
{
- return sa_add_rules(sa_ctx, entries, nb_entries, 1);
+ return sa_add_rules(sa_ctx, entries, nb_entries, 1, skt_ctx);
}
/*
@@ -1012,10 +1018,12 @@ fill_ipsec_sa_prm(struct rte_ipsec_sa_prm *prm, const struct ipsec_sa *ss,
return 0;
}
-static void
+static int
fill_ipsec_session(struct rte_ipsec_session *ss, struct rte_ipsec_sa *sa,
const struct ipsec_sa *lsa)
{
+ int32_t rc = 0;
+
ss->sa = sa;
ss->type = lsa->type;
@@ -1028,6 +1036,12 @@ fill_ipsec_session(struct rte_ipsec_session *ss, struct rte_ipsec_sa *sa,
ss->security.ctx = lsa->security_ctx;
ss->security.ol_flags = lsa->ol_flags;
}
+
+ rc = rte_ipsec_session_prepare(ss);
+ if (rc != 0)
+ memset(ss, 0, sizeof(*ss));
+
+ return rc;
}
/*
@@ -1062,8 +1076,8 @@ ipsec_sa_init(struct ipsec_sa *lsa, struct rte_ipsec_sa *sa, uint32_t sa_size)
if (rc < 0)
return rc;
- fill_ipsec_session(&lsa->ips, sa, lsa);
- return 0;
+ rc = fill_ipsec_session(&lsa->ips, sa, lsa);
+ return rc;
}
/*
@@ -1141,7 +1155,10 @@ sa_init(struct socket_ctx *ctx, int32_t socket_id)
"context %s in socket %d\n", rte_errno,
name, socket_id);
- sa_in_add_rules(ctx->sa_in, sa_in, nb_sa_in);
+ rc = sa_in_add_rules(ctx->sa_in, sa_in, nb_sa_in, ctx);
+ if (rc != 0)
+ rte_exit(EXIT_FAILURE,
+ "failed to add inbound rules\n");
if (app_sa_prm.enable != 0) {
rc = ipsec_satbl_init(ctx->sa_in, sa_in, nb_sa_in,
@@ -1161,7 +1178,10 @@ sa_init(struct socket_ctx *ctx, int32_t socket_id)
"context %s in socket %d\n", rte_errno,
name, socket_id);
- sa_out_add_rules(ctx->sa_out, sa_out, nb_sa_out);
+ rc = sa_out_add_rules(ctx->sa_out, sa_out, nb_sa_out, ctx);
+ if (rc != 0)
+ rte_exit(EXIT_FAILURE,
+ "failed to add outbound rules\n");
if (app_sa_prm.enable != 0) {
rc = ipsec_satbl_init(ctx->sa_out, sa_out, nb_sa_out,
--
2.7.4
^ permalink raw reply [flat|nested] 25+ messages in thread
end of thread, other threads:[~2019-05-27 8:58 UTC | newest]
Thread overview: 25+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-04-22 6:25 [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto Akhil Goyal
2019-04-22 6:25 ` Akhil Goyal
-- strict thread matches above, loose matches on Subject: below --
2019-04-18 13:51 Akhil Goyal
2019-04-18 13:51 ` Akhil Goyal
2019-04-18 14:58 ` Iremonger, Bernard
2019-04-18 14:58 ` Iremonger, Bernard
2019-04-18 15:23 ` Iremonger, Bernard
2019-04-18 15:23 ` Iremonger, Bernard
2019-04-23 11:14 ` Akhil Goyal
2019-04-23 11:14 ` Akhil Goyal
2019-04-23 13:21 ` Ananyev, Konstantin
2019-04-23 13:21 ` Ananyev, Konstantin
2019-04-23 13:32 ` Akhil Goyal
2019-04-23 13:32 ` Akhil Goyal
2019-04-23 14:04 ` Ananyev, Konstantin
2019-04-23 14:04 ` Ananyev, Konstantin
2019-04-24 6:34 ` Akhil Goyal
2019-04-24 6:34 ` Akhil Goyal
2019-04-24 10:40 ` Iremonger, Bernard
2019-04-24 10:40 ` Iremonger, Bernard
2019-05-13 14:29 ` Ananyev, Konstantin
2019-05-13 14:29 ` Ananyev, Konstantin
2019-05-27 8:58 ` Iremonger, Bernard
2019-04-04 13:28 [dpdk-dev] [PATCH v3 0/2] examples/ipsec-secgw: fix 1st pkt dropped Bernard Iremonger
2019-04-17 13:42 ` [dpdk-dev] [PATCH v4 1/2] examples/ipsec-secgw: fix 1st packet dropped for inline crypto Bernard Iremonger
2019-04-17 13:42 ` Bernard Iremonger
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).