* [PATCH v2 net-next 00/12] qed: Add iWARP support for unaligned MPA packets
From: Michal Kalderon @ 2017-10-03 8:54 UTC (permalink / raw)
To: davem-fT/PcQaiUtIeIZ0/mPfg9Q
Cc: netdev-u79uwXL29TY76Z2rM5mHXA, linux-rdma-u79uwXL29TY76Z2rM5mHXA,
dledford-H+wXaHxf7aLQT0dZR+AlfA, Michal Kalderon, Ariel Elior
This patch series adds support for handling unaligned MPA packets.
(FPDUs split over more than one tcp packet).
When FW detects a packet is unaligned it fowards the packet to
the driver via a light l2 dedicated connection. The driver then
stores this packet until the remainder of the packet is received.
Once the driver reconstructs the full FPDU, it sends it down
to fw via the ll2 connection. Driver also breaks down any packed
PDUs into separate packets for FW.
Patches 1-6 are all slight modifications to ll2 to support additional
requirements for the unaligned MPA ll2 client.
Patch 7 opens the additional ll2 connection for iWARP.
Patches 8-12 contain the algorithm for aligning packets.
Signed-off-by: Michal Kalderon <Michal.Kalderon-YGCgFSpz5w/QT0dZR+AlfA@public.gmane.org>
Signed-off-by: Ariel Elior <Ariel.Elior-YGCgFSpz5w/QT0dZR+AlfA@public.gmane.org>
---
Changes from v1:
patch #1: tx descq_array changed to void pointer and name
changed accordingly to descq_mem.
patch #2: modify boolean variables to be of type bool.
Michal Kalderon (12):
qed: Add ll2 option to limit the number of bds per packet
qed: Add ll2 ability of opening a secondary queue
qed: Add ll2 option for dropping a tx packet
qed: Fix initialization of ll2 offload feature
qed: Add the source of a packet sent on an iWARP ll2 connection
qed: Add LL2 slowpath handling
qed: Add ll2 connection for processing unaligned MPA packets
qed: Add mpa buffer descriptors for storing and processing mpa fpdus
qed: Add unaligned and packed packet processing
qed: Add support for freeing two ll2 buffers for corner cases
qed: Add support for MPA header being split over two tcp packets
qed: Add iWARP support for fpdu spanned over more than two tcp packets
drivers/net/ethernet/qlogic/qed/qed_iwarp.c | 690 ++++++++++++++++++++++++++++
drivers/net/ethernet/qlogic/qed/qed_iwarp.h | 33 ++
drivers/net/ethernet/qlogic/qed/qed_ll2.c | 102 +++-
drivers/net/ethernet/qlogic/qed/qed_ll2.h | 10 +-
include/linux/qed/qed_ll2_if.h | 7 +
5 files changed, 822 insertions(+), 20 deletions(-)
--
1.8.3.1
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply
* RE: [PATCH] fsl/fman: remove of_node
From: Madalin-cristian Bucur @ 2017-10-03 8:49 UTC (permalink / raw)
To: David Miller
Cc: netdev@vger.kernel.org, andrew@lunn.ch, f.fainelli@gmail.com,
linux-kernel@vger.kernel.org
In-Reply-To: <20171002.160445.737065995396076298.davem@davemloft.net>
> -----Original Message-----
> From: David Miller [mailto:davem@davemloft.net]
> Sent: Tuesday, October 03, 2017 2:05 AM
> To: Madalin-cristian Bucur <madalin.bucur@nxp.com>
> Subject: Re: [PATCH] fsl/fman: remove of_node
>
> From: Madalin Bucur <madalin.bucur@nxp.com>
> Date: Mon, 2 Oct 2017 13:31:37 +0300
>
> > The FMan MAC driver allocates a platform device for the Ethernet
> > driver to probe on. Setting pdev->dev.of_node with the MAC node
> > triggers the MAC driver probing of the new platform device. While
> > this fails quickly and does not affect the functionality of the
> > drivers, it is incorrect and must be removed. This was added to
> > address a report that DSA code using of_find_net_device_by_node()
> > is unable to use the DPAA interfaces. Error message seen before
> > this fix:
> >
> > fsl_mac dpaa-ethernet.0: __devm_request_mem_region(mac) failed
> > fsl_mac: probe of dpaa-ethernet.0 failed with error -16
> >
> > Signed-off-by: Madalin Bucur <madalin.bucur@nxp.com>
>
> Is the DSA issue no longer something we need to be concerned
> about? If not, why? You have to explain this.
My patch removes the of_node that was set to a device that was not an
of_device, preventing duplicated probing of both the real of_device
and the "fake" one created through this assignment.
I understand that the DSA issue that triggered the initial change
was related to DSA finding the network devices using
of_find_net_device_by_node(), something that will not work for the
DPAA case where the netdevice does not have an of_node. I do not know
enough about DSA to come up with a solution for this problem now.
Andrew, Florian, can you please comment on this?
Thanks,
Madalin
^ permalink raw reply
* (unknown),
From: morice.diane @ 2017-10-03 8:16 UTC (permalink / raw)
To: netdev
[-- Attachment #1: 747452.zip --]
[-- Type: application/zip, Size: 7300 bytes --]
^ permalink raw reply
* Re: [v2] cdc-ether: divorce initialisation with a filter reset and a generic method
From: Vignesh R @ 2017-10-03 8:09 UTC (permalink / raw)
To: davem; +Cc: Oliver Neukum, netdev, bjorn
In-Reply-To: <20170522125030.21266-1-oneukum@suse.com>
Hi Dave,
On Monday 22 May 2017 06:20 PM, Oliver Neukum wrote:
> Some devices need their multicast filter reset but others are crashed by that.
> So the methods need to be separated.
>
> Signed-off-by: Oliver Neukum <oneukum@suse.com>
> Reported-by: "Ridgway, Keith" <kridgway@harris.com>
> ---
I see this patch was merged and said to be queued for -stable, but I
don't see this fix in any of the stable kernels. Would you please ask
for this patch to be backported to -stable kernels?
Regards
Vignesh
> drivers/net/usb/cdc_ether.c | 31 ++++++++++++++++++++++++-------
> include/linux/usb/usbnet.h | 1 +
> 2 files changed, 25 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/net/usb/cdc_ether.c b/drivers/net/usb/cdc_ether.c
> index f3ae88fdf332..8ab281b478f2 100644
> --- a/drivers/net/usb/cdc_ether.c
> +++ b/drivers/net/usb/cdc_ether.c
> @@ -310,6 +310,26 @@ int usbnet_generic_cdc_bind(struct usbnet *dev, struct usb_interface *intf)
> return -ENODEV;
> }
>
> + return 0;
> +
> +bad_desc:
> + dev_info(&dev->udev->dev, "bad CDC descriptors\n");
> + return -ENODEV;
> +}
> +EXPORT_SYMBOL_GPL(usbnet_generic_cdc_bind);
> +
> +
> +/* like usbnet_generic_cdc_bind() but handles filter initialization
> + * correctly
> + */
> +int usbnet_ether_cdc_bind(struct usbnet *dev, struct usb_interface *intf)
> +{
> + int rv;
> +
> + rv = usbnet_generic_cdc_bind(dev, intf);
> + if (rv < 0)
> + goto bail_out;
> +
> /* Some devices don't initialise properly. In particular
> * the packet filter is not reset. There are devices that
> * don't do reset all the way. So the packet filter should
> @@ -317,13 +337,10 @@ int usbnet_generic_cdc_bind(struct usbnet *dev, struct usb_interface *intf)
> */
> usbnet_cdc_update_filter(dev);
>
> - return 0;
> -
> -bad_desc:
> - dev_info(&dev->udev->dev, "bad CDC descriptors\n");
> - return -ENODEV;
> +bail_out:
> + return rv;
> }
> -EXPORT_SYMBOL_GPL(usbnet_generic_cdc_bind);
> +EXPORT_SYMBOL_GPL(usbnet_ether_cdc_bind);
>
> void usbnet_cdc_unbind(struct usbnet *dev, struct usb_interface *intf)
> {
> @@ -417,7 +434,7 @@ int usbnet_cdc_bind(struct usbnet *dev, struct usb_interface *intf)
> BUILD_BUG_ON((sizeof(((struct usbnet *)0)->data)
> < sizeof(struct cdc_state)));
>
> - status = usbnet_generic_cdc_bind(dev, intf);
> + status = usbnet_ether_cdc_bind(dev, intf);
> if (status < 0)
> return status;
>
> diff --git a/include/linux/usb/usbnet.h b/include/linux/usb/usbnet.h
> index 7dffa5624ea6..97116379db5f 100644
> --- a/include/linux/usb/usbnet.h
> +++ b/include/linux/usb/usbnet.h
> @@ -206,6 +206,7 @@ struct cdc_state {
> };
>
> extern int usbnet_generic_cdc_bind(struct usbnet *, struct usb_interface *);
> +extern int usbnet_ether_cdc_bind(struct usbnet *dev, struct usb_interface *intf);
> extern int usbnet_cdc_bind(struct usbnet *, struct usb_interface *);
> extern void usbnet_cdc_unbind(struct usbnet *, struct usb_interface *);
> extern void usbnet_cdc_status(struct usbnet *, struct urb *);
>
--
Regards
Vignesh
^ permalink raw reply
* Re: [PATCH 05/18] net: use ARRAY_SIZE
From: Andy Shevchenko @ 2017-10-03 8:09 UTC (permalink / raw)
To: Jérémy Lefaure
Cc: Sathya Perla, Ajit Khaparde, Sriharsha Basavapatna, Somnath Kotur,
Jeff Kirsher, Arend van Spriel, Franky Lin, Hante Meuleman,
Chi-Hsien Lin, Wright Feng, Kalle Valo, Larry Finger, Chaoming Li,
David S. Miller, Alexey Kuznetsov, Hideaki YOSHIFUJI, netdev,
"linux-kernel@vger.kernel.org" <
In-Reply-To: <20171002212214.30d7013a@blatinox-laptop.localdomain>
On Tue, Oct 3, 2017 at 4:22 AM, Jérémy Lefaure
<jeremy.lefaure@lse.epita.fr> wrote:
> On Mon, 2 Oct 2017 16:07:36 +0300
> Andy Shevchenko <andy.shevchenko@gmail.com> wrote:
>
>> > + {&gainctrl_lut_core0_rev0, ARRAY_SIZE(gainctrl_lut_core0_rev0), 26, 192,
>> > + 32},
>>
>> For all such cases I would rather put on one line disregard checkpatch
>> warning for better readability.
> I agree that it would be better. I didn't know that it was possible to
> not follow this rule for anything else than a string.
IMO, it increases readability quite enough to overrule checkpatch recomendation.
--
With Best Regards,
Andy Shevchenko
^ permalink raw reply
* [patch net-next v2 7/7] mlxsw: spectrum: mr: Support trap-and-forward routes
From: Jiri Pirko @ 2017-10-03 7:58 UTC (permalink / raw)
To: netdev
Cc: davem, yotamg, idosch, mlxsw, nikolay, andrew, dsa, edumazet,
willemb, johannes.berg, dcaratti, pabeni, daniel, f.fainelli, fw,
gfree.wind
In-Reply-To: <20171003075812.1540-1-jiri@resnulli.us>
From: Yotam Gigi <yotamg@mellanox.com>
Add the support of trap-and-forward route action in the multicast routing
offloading logic. A route will be set to trap-and-forward action if one (or
more) of its output interfaces is not offload-able, i.e. does not have a
valid Spectrum RIF.
This way, a route with mixed output VIFs list, which contains both
offload-able and un-offload-able devices can go through partial offloading
in hardware, and the rest will be done in the kernel ipmr module.
Signed-off-by: Yotam Gigi <yotamg@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
drivers/net/ethernet/mellanox/mlxsw/spectrum_mr.c | 17 ++++++++---------
1 file changed, 8 insertions(+), 9 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_mr.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_mr.c
index 4aaf6ca..1f84bb8 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_mr.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_mr.c
@@ -114,9 +114,9 @@ static bool mlxsw_sp_mr_vif_valid(const struct mlxsw_sp_mr_vif *vif)
return mlxsw_sp_mr_vif_regular(vif) && vif->dev && vif->rif;
}
-static bool mlxsw_sp_mr_vif_rif_invalid(const struct mlxsw_sp_mr_vif *vif)
+static bool mlxsw_sp_mr_vif_exists(const struct mlxsw_sp_mr_vif *vif)
{
- return mlxsw_sp_mr_vif_regular(vif) && vif->dev && !vif->rif;
+ return vif->dev;
}
static bool
@@ -182,14 +182,13 @@ mlxsw_sp_mr_route_action(const struct mlxsw_sp_mr_route *mr_route)
if (!mlxsw_sp_mr_route_valid_evifs_num(mr_route))
return MLXSW_SP_MR_ROUTE_ACTION_TRAP;
- /* If either one of the eVIFs is not regular (VIF of type pimreg or
- * tunnel) or one of the VIFs has no matching RIF, trap the packet.
+ /* If one of the eVIFs has no RIF, trap-and-forward the route as there
+ * is some more routing to do in software too.
*/
- list_for_each_entry(rve, &mr_route->evif_list, route_node) {
- if (!mlxsw_sp_mr_vif_regular(rve->mr_vif) ||
- mlxsw_sp_mr_vif_rif_invalid(rve->mr_vif))
- return MLXSW_SP_MR_ROUTE_ACTION_TRAP;
- }
+ list_for_each_entry(rve, &mr_route->evif_list, route_node)
+ if (mlxsw_sp_mr_vif_exists(rve->mr_vif) && !rve->mr_vif->rif)
+ return MLXSW_SP_MR_ROUTE_ACTION_TRAP_AND_FORWARD;
+
return MLXSW_SP_MR_ROUTE_ACTION_FORWARD;
}
--
2.9.5
^ permalink raw reply related
* [patch net-next v2 5/7] mlxsw: spectrum: Add trap for multicast trap-and-forward routes
From: Jiri Pirko @ 2017-10-03 7:58 UTC (permalink / raw)
To: netdev
Cc: davem, yotamg, idosch, mlxsw, nikolay, andrew, dsa, edumazet,
willemb, johannes.berg, dcaratti, pabeni, daniel, f.fainelli, fw,
gfree.wind
In-Reply-To: <20171003075812.1540-1-jiri@resnulli.us>
From: Yotam Gigi <yotamg@mellanox.com>
When a multicast route is configured with trap-and-forward action, the
packets should be marked with skb->offload_mr_fwd_mark, in order to prevent
the packets from being forwarded again by the kernel ipmr module.
Due to this, it is not possible to use the already existing multicast trap
(MLXSW_TRAP_ID_ACL1) as the packet should be marked differently. Add the
MLXSW_TRAP_ID_ACL2 which is for trap-and-forward multicast routes, and set
the offload_mr_fwd_mark skb field in its handler.
Signed-off-by: Yotam Gigi <yotamg@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 13 +++++++++++++
drivers/net/ethernet/mellanox/mlxsw/trap.h | 2 ++
2 files changed, 15 insertions(+)
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
index e9b9443..3adf237 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.c
@@ -3312,6 +3312,14 @@ static void mlxsw_sp_rx_listener_mark_func(struct sk_buff *skb, u8 local_port,
return mlxsw_sp_rx_listener_no_mark_func(skb, local_port, priv);
}
+static void mlxsw_sp_rx_listener_mr_mark_func(struct sk_buff *skb,
+ u8 local_port, void *priv)
+{
+ skb->offload_mr_fwd_mark = 1;
+ skb->offload_fwd_mark = 1;
+ return mlxsw_sp_rx_listener_no_mark_func(skb, local_port, priv);
+}
+
static void mlxsw_sp_rx_listener_sample_func(struct sk_buff *skb, u8 local_port,
void *priv)
{
@@ -3355,6 +3363,10 @@ static void mlxsw_sp_rx_listener_sample_func(struct sk_buff *skb, u8 local_port,
MLXSW_RXL(mlxsw_sp_rx_listener_mark_func, _trap_id, _action, \
_is_ctrl, SP_##_trap_group, DISCARD)
+#define MLXSW_SP_RXL_MR_MARK(_trap_id, _action, _trap_group, _is_ctrl) \
+ MLXSW_RXL(mlxsw_sp_rx_listener_mr_mark_func, _trap_id, _action, \
+ _is_ctrl, SP_##_trap_group, DISCARD)
+
#define MLXSW_SP_EVENTL(_func, _trap_id) \
MLXSW_EVENTL(_func, _trap_id, SP_EVENT)
@@ -3425,6 +3437,7 @@ static const struct mlxsw_listener mlxsw_sp_listener[] = {
MLXSW_SP_RXL_MARK(IPV4_PIM, TRAP_TO_CPU, PIM, false),
MLXSW_SP_RXL_MARK(RPF, TRAP_TO_CPU, RPF, false),
MLXSW_SP_RXL_MARK(ACL1, TRAP_TO_CPU, MULTICAST, false),
+ MLXSW_SP_RXL_MR_MARK(ACL2, TRAP_TO_CPU, MULTICAST, false),
};
static int mlxsw_sp_cpu_policers_set(struct mlxsw_core *mlxsw_core)
diff --git a/drivers/net/ethernet/mellanox/mlxsw/trap.h b/drivers/net/ethernet/mellanox/mlxsw/trap.h
index a981035..ec6cef8 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/trap.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/trap.h
@@ -93,6 +93,8 @@ enum {
MLXSW_TRAP_ID_ACL0 = 0x1C0,
/* Multicast trap used for routes with trap action */
MLXSW_TRAP_ID_ACL1 = 0x1C1,
+ /* Multicast trap used for routes with trap-and-forward action */
+ MLXSW_TRAP_ID_ACL2 = 0x1C2,
MLXSW_TRAP_ID_MAX = 0x1FF
};
--
2.9.5
^ permalink raw reply related
* [patch net-next v2 6/7] mlxsw: spectrum: mr_tcam: Add trap-and-forward multicast route
From: Jiri Pirko @ 2017-10-03 7:58 UTC (permalink / raw)
To: netdev
Cc: davem, yotamg, idosch, mlxsw, nikolay, andrew, dsa, edumazet,
willemb, johannes.berg, dcaratti, pabeni, daniel, f.fainelli, fw,
gfree.wind
In-Reply-To: <20171003075812.1540-1-jiri@resnulli.us>
From: Yotam Gigi <yotamg@mellanox.com>
In addition to the current multicast route actions, which include trap
route action and a forward route action, add the trap-and-forward multicast
route action, and implement it in the multicast routing hardware logic.
To implement that, add a trap-and-forward ACL action as the last action in
the route flexible action set. The used trap is the ACL2 trap, which marks
the packets with offload_mr_forward_mark, to prevent the packet from being
forwarded again by the kernel.
Note: At that stage the offloading logic does not support trap-and-forward
multicast routes. This patch adds the support only in the hardware logic.
Signed-off-by: Yotam Gigi <yotamg@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
drivers/net/ethernet/mellanox/mlxsw/spectrum_mr.h | 1 +
drivers/net/ethernet/mellanox/mlxsw/spectrum_mr_tcam.c | 8 ++++++++
2 files changed, 9 insertions(+)
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_mr.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum_mr.h
index c851b23..5d26a12 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_mr.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_mr.h
@@ -42,6 +42,7 @@
enum mlxsw_sp_mr_route_action {
MLXSW_SP_MR_ROUTE_ACTION_FORWARD,
MLXSW_SP_MR_ROUTE_ACTION_TRAP,
+ MLXSW_SP_MR_ROUTE_ACTION_TRAP_AND_FORWARD,
};
enum mlxsw_sp_mr_route_prio {
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_mr_tcam.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_mr_tcam.c
index cda9e9a..3ffb28d 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_mr_tcam.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_mr_tcam.c
@@ -253,6 +253,7 @@ mlxsw_sp_mr_tcam_afa_block_create(struct mlxsw_sp *mlxsw_sp,
if (err)
goto err;
break;
+ case MLXSW_SP_MR_ROUTE_ACTION_TRAP_AND_FORWARD:
case MLXSW_SP_MR_ROUTE_ACTION_FORWARD:
/* If we are about to append a multicast router action, commit
* the erif_list.
@@ -266,6 +267,13 @@ mlxsw_sp_mr_tcam_afa_block_create(struct mlxsw_sp *mlxsw_sp,
erif_list->kvdl_index);
if (err)
goto err;
+
+ if (route_action == MLXSW_SP_MR_ROUTE_ACTION_TRAP_AND_FORWARD) {
+ err = mlxsw_afa_block_append_trap_and_forward(afa_block,
+ MLXSW_TRAP_ID_ACL2);
+ if (err)
+ goto err;
+ }
break;
default:
err = -EINVAL;
--
2.9.5
^ permalink raw reply related
* [patch net-next v2 4/7] mlxsw: acl: Introduce ACL trap and forward action
From: Jiri Pirko @ 2017-10-03 7:58 UTC (permalink / raw)
To: netdev
Cc: davem, yotamg, idosch, mlxsw, nikolay, andrew, dsa, edumazet,
willemb, johannes.berg, dcaratti, pabeni, daniel, f.fainelli, fw,
gfree.wind
In-Reply-To: <20171003075812.1540-1-jiri@resnulli.us>
From: Yotam Gigi <yotamg@mellanox.com>
Use trap/discard flex action to implement trap and forward. The action will
later be used for multicast routing, as the multicast routing mechanism is
done using ACL flexible actions in Spectrum hardware. Using that action, it
will be possible to implement a trap-and-forward route.
Signed-off-by: Yotam Gigi <yotamg@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
.../net/ethernet/mellanox/mlxsw/core_acl_flex_actions.c | 17 +++++++++++++++++
.../net/ethernet/mellanox/mlxsw/core_acl_flex_actions.h | 2 ++
2 files changed, 19 insertions(+)
diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_actions.c b/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_actions.c
index bc55d0e..6a979a0 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_actions.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_actions.c
@@ -676,6 +676,7 @@ enum mlxsw_afa_trapdisc_trap_action {
MLXSW_ITEM32(afa, trapdisc, trap_action, 0x00, 24, 4);
enum mlxsw_afa_trapdisc_forward_action {
+ MLXSW_AFA_TRAPDISC_FORWARD_ACTION_FORWARD = 1,
MLXSW_AFA_TRAPDISC_FORWARD_ACTION_DISCARD = 3,
};
@@ -729,6 +730,22 @@ int mlxsw_afa_block_append_trap(struct mlxsw_afa_block *block, u16 trap_id)
}
EXPORT_SYMBOL(mlxsw_afa_block_append_trap);
+int mlxsw_afa_block_append_trap_and_forward(struct mlxsw_afa_block *block,
+ u16 trap_id)
+{
+ char *act = mlxsw_afa_block_append_action(block,
+ MLXSW_AFA_TRAPDISC_CODE,
+ MLXSW_AFA_TRAPDISC_SIZE);
+
+ if (!act)
+ return -ENOBUFS;
+ mlxsw_afa_trapdisc_pack(act, MLXSW_AFA_TRAPDISC_TRAP_ACTION_TRAP,
+ MLXSW_AFA_TRAPDISC_FORWARD_ACTION_FORWARD,
+ trap_id);
+ return 0;
+}
+EXPORT_SYMBOL(mlxsw_afa_block_append_trap_and_forward);
+
/* Forwarding Action
* -----------------
* Forwarding Action can be used to implement Policy Based Switching (PBS)
diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_actions.h b/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_actions.h
index 06b0be4..a8d3314 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_actions.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_acl_flex_actions.h
@@ -61,6 +61,8 @@ int mlxsw_afa_block_continue(struct mlxsw_afa_block *block);
int mlxsw_afa_block_jump(struct mlxsw_afa_block *block, u16 group_id);
int mlxsw_afa_block_append_drop(struct mlxsw_afa_block *block);
int mlxsw_afa_block_append_trap(struct mlxsw_afa_block *block, u16 trap_id);
+int mlxsw_afa_block_append_trap_and_forward(struct mlxsw_afa_block *block,
+ u16 trap_id);
int mlxsw_afa_block_append_fwd(struct mlxsw_afa_block *block,
u8 local_port, bool in_port);
int mlxsw_afa_block_append_vlan_modify(struct mlxsw_afa_block *block,
--
2.9.5
^ permalink raw reply related
* [patch net-next v2 3/7] ipv4: ipmr: Don't forward packets already forwarded by hardware
From: Jiri Pirko @ 2017-10-03 7:58 UTC (permalink / raw)
To: netdev
Cc: davem, yotamg, idosch, mlxsw, nikolay, andrew, dsa, edumazet,
willemb, johannes.berg, dcaratti, pabeni, daniel, f.fainelli, fw,
gfree.wind
In-Reply-To: <20171003075812.1540-1-jiri@resnulli.us>
From: Yotam Gigi <yotamg@mellanox.com>
Change the ipmr module to not forward packets if:
- The packet is marked with the offload_mr_fwd_mark, and
- Both input interface and output interface share the same parent ID.
This way, a packet can go through partial multicast forwarding in the
hardware, where it will be forwarded only to the devices that share the
same parent ID (AKA, reside inside the same hardware). The kernel will
forward the packet to all other interfaces.
To do this, add the ipmr_offload_forward helper, which per skb, ingress VIF
and egress VIF, returns whether the forwarding was offloaded to hardware.
The ipmr_queue_xmit frees the skb and does not forward it if the result is
a true value.
All the forwarding path code compiles out when the CONFIG_NET_SWITCHDEV is
not set.
Signed-off-by: Yotam Gigi <yotamg@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
v1->v2:
- Use dev_parent_id.len field instead of the dev_parent_id_valid field
---
net/ipv4/ipmr.c | 37 ++++++++++++++++++++++++++++++++-----
1 file changed, 32 insertions(+), 5 deletions(-)
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index 1b161ad..b3ee01b 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -1859,10 +1859,33 @@ static inline int ipmr_forward_finish(struct net *net, struct sock *sk,
return dst_output(net, sk, skb);
}
+#ifdef CONFIG_NET_SWITCHDEV
+static bool ipmr_forward_offloaded(struct sk_buff *skb, struct mr_table *mrt,
+ int in_vifi, int out_vifi)
+{
+ struct vif_device *out_vif = &mrt->vif_table[out_vifi];
+ struct vif_device *in_vif = &mrt->vif_table[in_vifi];
+
+ if (!skb->offload_mr_fwd_mark)
+ return false;
+ if (!out_vif->dev_parent_id.id_len || !in_vif->dev_parent_id.id_len)
+ return false;
+ return netdev_phys_item_id_same(&out_vif->dev_parent_id,
+ &in_vif->dev_parent_id);
+}
+#else
+static bool ipmr_forward_offloaded(struct sk_buff *skb, struct mr_table *mrt,
+ int in_vifi, int out_vifi)
+{
+ return false;
+}
+#endif
+
/* Processing handlers for ipmr_forward */
static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt,
- struct sk_buff *skb, struct mfc_cache *c, int vifi)
+ int in_vifi, struct sk_buff *skb,
+ struct mfc_cache *c, int vifi)
{
const struct iphdr *iph = ip_hdr(skb);
struct vif_device *vif = &mrt->vif_table[vifi];
@@ -1883,6 +1906,9 @@ static void ipmr_queue_xmit(struct net *net, struct mr_table *mrt,
goto out_free;
}
+ if (ipmr_forward_offloaded(skb, mrt, in_vifi, vifi))
+ goto out_free;
+
if (vif->flags & VIFF_TUNNEL) {
rt = ip_route_output_ports(net, &fl4, NULL,
vif->remote, vif->local,
@@ -2060,8 +2086,8 @@ static void ip_mr_forward(struct net *net, struct mr_table *mrt,
struct sk_buff *skb2 = skb_clone(skb, GFP_ATOMIC);
if (skb2)
- ipmr_queue_xmit(net, mrt, skb2, cache,
- psend);
+ ipmr_queue_xmit(net, mrt, true_vifi,
+ skb2, cache, psend);
}
psend = ct;
}
@@ -2072,9 +2098,10 @@ static void ip_mr_forward(struct net *net, struct mr_table *mrt,
struct sk_buff *skb2 = skb_clone(skb, GFP_ATOMIC);
if (skb2)
- ipmr_queue_xmit(net, mrt, skb2, cache, psend);
+ ipmr_queue_xmit(net, mrt, true_vifi, skb2,
+ cache, psend);
} else {
- ipmr_queue_xmit(net, mrt, skb, cache, psend);
+ ipmr_queue_xmit(net, mrt, true_vifi, skb, cache, psend);
return;
}
}
--
2.9.5
^ permalink raw reply related
* [patch net-next v2 2/7] ipv4: ipmr: Add the parent ID field to VIF struct
From: Jiri Pirko @ 2017-10-03 7:58 UTC (permalink / raw)
To: netdev
Cc: davem, yotamg, idosch, mlxsw, nikolay, andrew, dsa, edumazet,
willemb, johannes.berg, dcaratti, pabeni, daniel, f.fainelli, fw,
gfree.wind
In-Reply-To: <20171003075812.1540-1-jiri@resnulli.us>
From: Yotam Gigi <yotamg@mellanox.com>
In order to allow the ipmr module to do partial multicast forwarding
according to the device parent ID, add the device parent ID field to the
VIF struct. This way, the forwarding path can use the parent ID field
without invoking switchdev calls, which requires the RTNL lock.
When a new VIF is added, set the device parent ID field in it by invoking
the switchdev_port_attr_get call.
Signed-off-by: Yotam Gigi <yotamg@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
v1->v2:
- Set the vif->dev_parent_id.len field
- Remove the vif->dev_parent_id_valid field and use len instead
- Set the vif->dev_parent_id to invalid if the device does not support the
switchdev PARENT_ID attribute
---
include/linux/mroute.h | 1 +
net/ipv4/ipmr.c | 11 +++++++++++
2 files changed, 12 insertions(+)
diff --git a/include/linux/mroute.h b/include/linux/mroute.h
index b072a84..8242d05 100644
--- a/include/linux/mroute.h
+++ b/include/linux/mroute.h
@@ -57,6 +57,7 @@ static inline bool ipmr_rule_default(const struct fib_rule *rule)
struct vif_device {
struct net_device *dev; /* Device we are using */
+ struct netdev_phys_item_id dev_parent_id; /* Device parent ID */
unsigned long bytes_in,bytes_out;
unsigned long pkt_in,pkt_out; /* Statistics */
unsigned long rate_limit; /* Traffic shaping (NI) */
diff --git a/net/ipv4/ipmr.c b/net/ipv4/ipmr.c
index a844738..1b161ad 100644
--- a/net/ipv4/ipmr.c
+++ b/net/ipv4/ipmr.c
@@ -67,6 +67,7 @@
#include <net/fib_rules.h>
#include <linux/netconf.h>
#include <net/nexthop.h>
+#include <net/switchdev.h>
struct ipmr_rule {
struct fib_rule common;
@@ -868,6 +869,9 @@ static int vif_add(struct net *net, struct mr_table *mrt,
struct vifctl *vifc, int mrtsock)
{
int vifi = vifc->vifc_vifi;
+ struct switchdev_attr attr = {
+ .id = SWITCHDEV_ATTR_ID_PORT_PARENT_ID,
+ };
struct vif_device *v = &mrt->vif_table[vifi];
struct net_device *dev;
struct in_device *in_dev;
@@ -942,6 +946,13 @@ static int vif_add(struct net *net, struct mr_table *mrt,
/* Fill in the VIF structures */
+ attr.orig_dev = dev;
+ if (!switchdev_port_attr_get(dev, &attr)) {
+ memcpy(v->dev_parent_id.id, attr.u.ppid.id, attr.u.ppid.id_len);
+ v->dev_parent_id.id_len = attr.u.ppid.id_len;
+ } else {
+ v->dev_parent_id.id_len = 0;
+ }
v->rate_limit = vifc->vifc_rate_limit;
v->local = vifc->vifc_lcl_addr.s_addr;
v->remote = vifc->vifc_rmt_addr.s_addr;
--
2.9.5
^ permalink raw reply related
* [patch net-next v2 1/7] skbuff: Add the offload_mr_fwd_mark field
From: Jiri Pirko @ 2017-10-03 7:58 UTC (permalink / raw)
To: netdev
Cc: davem, yotamg, idosch, mlxsw, nikolay, andrew, dsa, edumazet,
willemb, johannes.berg, dcaratti, pabeni, daniel, f.fainelli, fw,
gfree.wind
In-Reply-To: <20171003075812.1540-1-jiri@resnulli.us>
From: Yotam Gigi <yotamg@mellanox.com>
Similarly to the offload_fwd_mark field, the offload_mr_fwd_mark field is
used to allow partial offloading of MFC multicast routes.
Switchdev drivers can offload MFC multicast routes to the hardware by
registering to the FIB notification chain. When one of the route output
interfaces is not offload-able, i.e. has different parent ID, the route
cannot be fully offloaded by the hardware. Examples to non-offload-able
devices are a management NIC, dummy device, pimreg device, etc.
Similar problem exists in the bridge module, as one bridge can hold
interfaces with different parent IDs. At the bridge, the problem is solved
by the offload_fwd_mark skb field.
Currently, when a route cannot go through full offload, the only solution
for a switchdev driver is not to offload it at all and let the packet go
through slow path.
Using the offload_mr_fwd_mark field, a driver can indicate that a packet
was already forwarded by hardware to all the devices with the same parent
ID as the input device. Further patches in this patch-set are going to
enhance ipmr to skip multicast forwarding to devices with the same parent
ID if a packets is marked with that field.
The reason why the already existing "offload_fwd_mark" bit cannot be used
is that a switchdev driver would want to make the distinction between a
packet that has already gone through L2 forwarding but did not go through
multicast forwarding, and a packet that has already gone through both L2
and multicast forwarding.
For example: when a packet is ingressing from a switchport enslaved to a
bridge, which is configured with multicast forwarding, the following
scenarios are possible:
- The packet can be trapped to the CPU due to exception while multicast
forwarding (for example, MTU error). In that case, it had already gone
through L2 forwarding in the hardware, thus A switchdev driver would
want to set the skb->offload_fwd_mark and not the
skb->offload_mr_fwd_mark.
- The packet can also be trapped due to a pimreg/dummy device used as one
of the output interfaces. In that case, it can go through both L2 and
(partial) multicast forwarding inside the hardware, thus a switchdev
driver would want to set both the skb->offload_fwd_mark and
skb->offload_mr_fwd_mark.
Signed-off-by: Yotam Gigi <yotamg@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellaox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
include/linux/skbuff.h | 1 +
1 file changed, 1 insertion(+)
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 19e64bf..ada8214 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -772,6 +772,7 @@ struct sk_buff {
__u8 remcsum_offload:1;
#ifdef CONFIG_NET_SWITCHDEV
__u8 offload_fwd_mark:1;
+ __u8 offload_mr_fwd_mark:1;
#endif
#ifdef CONFIG_NET_CLS_ACT
__u8 tc_skip_classify:1;
--
2.9.5
^ permalink raw reply related
* [patch net-next v2 0/7] mlxsw: Add support for partial multicast route offload
From: Jiri Pirko @ 2017-10-03 7:58 UTC (permalink / raw)
To: netdev
Cc: davem, yotamg, idosch, mlxsw, nikolay, andrew, dsa, edumazet,
willemb, johannes.berg, dcaratti, pabeni, daniel, f.fainelli, fw,
gfree.wind
From: Jiri Pirko <jiri@mellanox.com>
Yotam says:
Previous patchset introduced support for offloading multicast MFC routes to
the Spectrum hardware. As described in that patchset, no partial offloading
is supported, i.e if a route has one output interface which is not a valid
offloadable device (e.g. pimreg device, dummy device, management NIC), the
route is trapped to the CPU and the forwarding is done in slow-path.
Add support for partial offloading of multicast routes, by letting the
hardware to forward the packet to all the in-hardware devices, while the
kernel ipmr module will continue forwarding to all other interfaces.
Similarly to the bridge, the kernel ipmr module will forward a marked
packet to an interface only if the interface has a different parent ID than
the packet's ingress interfaces.
The first patch introduces the offload_mr_fwd_mark skb field, which can be
used by offloading drivers to indicate that a packet had already gone
through multicast forwarding in hardware, similarly to the offload_fwd_mark
field that indicates that a packet had already gone through L2 forwarding
in hardware.
Patches 2 and 3 change the ipmr module to not forward packets that had
already been forwarded by the hardware, i.e. packets that are marked with
offload_mr_fwd_mark and the ingress VIF shares the same parent ID with the
egress VIF.
Patches 4, 5, 6 and 7 add the support in the mlxsw Spectrum driver for trap
and forward routes, while marking the trapped packets with the
offload_mr_fwd_mark.
Yotam Gigi (7):
skbuff: Add the offload_mr_fwd_mark field
ipv4: ipmr: Add the parent ID field to VIF struct
ipv4: ipmr: Don't forward packets already forwarded by hardware
mlxsw: acl: Introduce ACL trap and forward action
mlxsw: spectrum: Add trap for multicast trap-and-forward routes
mlxsw: spectrum: mr_tcam: Add trap-and-forward multicast route
mlxsw: spectrum: mr: Support trap-and-forward routes
.../mellanox/mlxsw/core_acl_flex_actions.c | 17 ++++++++
.../mellanox/mlxsw/core_acl_flex_actions.h | 2 +
drivers/net/ethernet/mellanox/mlxsw/spectrum.c | 13 ++++++
drivers/net/ethernet/mellanox/mlxsw/spectrum_mr.c | 17 ++++----
drivers/net/ethernet/mellanox/mlxsw/spectrum_mr.h | 1 +
.../net/ethernet/mellanox/mlxsw/spectrum_mr_tcam.c | 8 ++++
drivers/net/ethernet/mellanox/mlxsw/trap.h | 2 +
include/linux/mroute.h | 1 +
include/linux/skbuff.h | 1 +
net/ipv4/ipmr.c | 48 +++++++++++++++++++---
10 files changed, 96 insertions(+), 14 deletions(-)
--
2.9.5
^ permalink raw reply
* Re: [PATCH 1/2] net: phonet: mark header_ops as const
From: kbuild test robot @ 2017-10-03 7:46 UTC (permalink / raw)
To: Lin Zhang; +Cc: kbuild-all, davem, netdev, courmisch, Lin Zhang
In-Reply-To: <1506762519-4709-1-git-send-email-xiaolou4617@gmail.com>
[-- Attachment #1: Type: text/plain, Size: 1396 bytes --]
Hi Lin,
[auto build test ERROR on net-next/master]
[also build test ERROR on v4.14-rc3 next-20170929]
[if your patch is applied to the wrong git tree, please drop us a note to help improve the system]
url: https://github.com/0day-ci/linux/commits/Lin-Zhang/net-phonet-mark-header_ops-as-const/20171003-145726
config: x86_64-randconfig-x000-201740 (attached as .config)
compiler: gcc-6 (Debian 6.2.0-3) 6.2.0 20160901
reproduce:
# save the attached .config to linux build tree
make ARCH=x86_64
All errors (new ones prefixed by >>):
>> net//phonet/af_phonet.c:152:25: error: conflicting type qualifiers for 'phonet_header_ops'
const struct header_ops phonet_header_ops = {
^~~~~~~~~~~~~~~~~
In file included from net//phonet/af_phonet.c:32:0:
include/linux/if_phonet.h:13:26: note: previous declaration of 'phonet_header_ops' was here
extern struct header_ops phonet_header_ops;
^~~~~~~~~~~~~~~~~
vim +/phonet_header_ops +152 net//phonet/af_phonet.c
151
> 152 const struct header_ops phonet_header_ops = {
153 .create = pn_header_create,
154 .parse = pn_header_parse,
155 };
156 EXPORT_SYMBOL(phonet_header_ops);
157
---
0-DAY kernel test infrastructure Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all Intel Corporation
[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 27699 bytes --]
^ permalink raw reply
* Re: [PATCH v4 net-next 0/8] flow_dissector: Protocol specific flow dissector offload
From: Jiri Pirko @ 2017-10-03 7:46 UTC (permalink / raw)
To: Tom Herbert
Cc: David Miller, Hannes Frederic Sowa, Tom Herbert,
Linux Kernel Network Developers, Rohit Seth
In-Reply-To: <CALx6S37DhT095fY92HWZAKQvMkqjxqq08nxJqL=wTP8JorHHDg@mail.gmail.com>
Fri, Sep 29, 2017 at 07:59:35PM CEST, tom@herbertland.com wrote:
>On Fri, Sep 29, 2017 at 10:42 AM, David Miller <davem@davemloft.net> wrote:
>> From: Tom Herbert <tom@herbertland.com>
>> Date: Fri, 29 Sep 2017 08:48:55 -0700
>>
>>> The flow_dissector interface is not a uAPI.
>>
>> That's not true, insofar as cls_flower.c uses the flow_dissector
>> therefore if you change the flow_dissector in certain ways then
>> cls_flower.c might have it's behavior changed and that is in fact UAPI
>> facing.
>
>Then I would suggest adding another flag like FLOW_DISSECTOR_F_FLOWER
>and when anyone puts new code into flow_dissector they can wrap it
>with "if !(flags & FLOW_DISSECTOR_F_FLOWER)". If the flower uAPI is
>subsequently update then the conditional can be removed. This way
>flower can support maintain its APIs, but we can still still extend
>and improve flow_dissector for othersuse cases.
This is not flower-specific problem. Flow_dissector is a servant of many.
As such, it is instructed what should it do. If you want to
change the way inner headers are parsed, you should either:
1) change the callers so they are behaving the same as before
2) make the flow_dissection change optional so the caller can say if he
wants original or new behaviour.
^ permalink raw reply
* BUG in free_netdev() on ppp link deletion
From: Beniamino Galvani @ 2017-10-03 7:44 UTC (permalink / raw)
To: linux-ppp, netdev, Paul Mackerras, Guillaume Nault, David Ahern,
Gao Feng
Hi,
I see the following BUG on 4.14-rc2 and previous versions (reproduced
with 4.11 as well):
------------[ cut here ]------------
kernel BUG at net/core/dev.c:8141!
invalid opcode: 0000 [#1] SMP
Modules linked in: pppoe pppox ppp_generic slhc cfg80211 rfkill joydev uinput tun sunrpc snd_hda_codec_generic snd_hda_intel snd_hda_codec snd_hda_core snd_hwdep snd_seq snd_seq_device crct10dif_pclmul crc32_pclmul snd_pcm ghash_clmulni_intel snd_timer ppdev snd parport_pc pcspkr soundcore parport virtio_balloon i2c_piix4 8139too virtio_console qxl drm_kms_helper ttm drm crc32c_intel serio_raw e1000 virtio_pci 8139cp floppy virtio_ring qemu_fw_cfg mii virtio ata_generic pata_acpi
CPU: 1 PID: 1557 Comm: pppd Not tainted 4.14.0-rc2+ #4
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.10.2-1.fc26 04/01/2014
task: ffff9cd4ba5e4c80 task.stack: ffffc28a40570000
RIP: 0010:free_netdev+0x107/0x110
RSP: 0018:ffffc28a40573d88 EFLAGS: 00010297
RAX: 0000000000000002 RBX: ffff9cd4bb0178c0 RCX: 000000000001e94c
RDX: 0000000000000001 RSI: 0000000000000286 RDI: 0000000000000000
RBP: ffffc28a40573da0 R08: 0000000000000001 R09: 000000000000019b
R10: 000000000000059b R11: 0000000000000001 R12: ffff9cd4bb017000
R13: ffff9cd4bb017060 R14: ffff9cd4bb01793c R15: 0000000000000000
FS: 00007fa720aa2840(0000) GS:ffff9cd4bfc80000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007fffeefa3000 CR3: 0000000072b34002 CR4: 00000000001606e0
Call Trace:
ppp_destroy_interface+0xd8/0xe0 [ppp_generic]
ppp_disconnect_channel+0xda/0x110 [ppp_generic]
ppp_unregister_channel+0x5e/0x110 [ppp_generic]
pppox_unbind_sock+0x23/0x30 [pppox]
pppoe_connect+0x130/0x440 [pppoe]
SYSC_connect+0x98/0x110
? do_fcntl+0x2c0/0x5d0
SyS_connect+0xe/0x10
entry_SYSCALL_64_fastpath+0x1a/0xa5
RIP: 0033:0x7fa71f4af840
RSP: 002b:00007ffe4ea40bf8 EFLAGS: 00000246 ORIG_RAX: 000000000000002a
RAX: ffffffffffffffda RBX: 0000556d37ae0538 RCX: 00007fa71f4af840
RDX: 000000000000001e RSI: 00007ffe4ea40c00 RDI: 0000000000000008
RBP: 0000556d37b2a1b0 R08: 0000556d396e95b0 R09: 0000000000000008
R10: 00000000aaaaaaab R11: 0000000000000246 R12: 0000556d37adc008
R13: 0000556d37adc004 R14: 0000556d37b2a1a4 R15: 0000000000000000
Code: 04 00 00 04 e8 cb 52 e3 ff 5b 41 5c 41 5d 5d c3 41 0f b7 84 24 32 02 00 00 4c 89 e7 48 29 c7 e8 80 8b aa ff 5b 41 5c 41 5d 5d c3 <0f> 0b 0f 1f 80 00 00 00 00 0f 1f 44 00 00 55 48 89 e5 41 57 41
RIP: free_netdev+0x107/0x110 RSP: ffffc28a40573d88
---[ end trace ed294ff0cc40eeff ]---
To reproduce this, establish a PPP connection through pppd, then bring
down and delete the ppp interface:
# pppd nodetach lock user client plugin rp-pppoe.so ens11 noauth nodeflate password password &
Plugin rp-pppoe.so loaded.
RP-PPPoE plugin version 3.8p compiled against pppd 2.4.7
PPP session is 16
Connected to fe:54:00:5f:04:13 via interface ens11
Using interface ppp0
Connect: ppp0 <--> ens11
CHAP authentication succeeded: Access granted
CHAP authentication succeeded
peer from calling number FE:54:00:5F:04:13 authorized
local IP address 3.1.1.10
remote IP address 3.1.1.1
# ip l set ppp0 down
# ip l del ppp0
It does not happen every time but only when ppp_destroy_interface() is
called with dev->reg_state = UNREGISTERING, set by the concurrent
rtnl_delete_link().
Beniamino
^ permalink raw reply
* [PATCH 1/1] xdp: Sample xdp program implementing ip forward
From: cjacob @ 2017-10-03 7:37 UTC (permalink / raw)
To: netdev; +Cc: Christina.Jacob, linux-kernel, linux-arm-kernel
In-Reply-To: <1507016225-319-1-git-send-email-Christina.Jacob@cavium.com>
Implements port to port forwarding with route table and arp table
lookup for ipv4 packets using bpf_redirect helper function and
lpm_trie map.
Signed-off-by: cjacob <Christina.Jacob@cavium.com>
---
samples/bpf/Makefile | 4 +
samples/bpf/xdp3_kern.c | 204 +++++++++++++++
samples/bpf/xdp3_user.c | 649 +++++++++++++++++++++++++++++++++++++++++++++++
3 files changed, 857 insertions(+), 0 deletions(-)
diff --git a/samples/bpf/Makefile b/samples/bpf/Makefile
index cf17c79..cc9cc0b 100644
--- a/samples/bpf/Makefile
+++ b/samples/bpf/Makefile
@@ -28,6 +28,7 @@ hostprogs-y += test_cgrp2_sock
hostprogs-y += test_cgrp2_sock2
hostprogs-y += xdp1
hostprogs-y += xdp2
+hostprogs-y += xdp3
hostprogs-y += test_current_task_under_cgroup
hostprogs-y += trace_event
hostprogs-y += sampleip
@@ -73,6 +74,7 @@ test_cgrp2_sock2-objs := bpf_load.o $(LIBBPF) test_cgrp2_sock2.o
xdp1-objs := bpf_load.o $(LIBBPF) xdp1_user.o
# reuse xdp1 source intentionally
xdp2-objs := bpf_load.o $(LIBBPF) xdp1_user.o
+xdp3-objs := bpf_load.o $(LIBBPF) xdp3_user.o
test_current_task_under_cgroup-objs := bpf_load.o $(LIBBPF) cgroup_helpers.o \
test_current_task_under_cgroup_user.o
trace_event-objs := bpf_load.o $(LIBBPF) trace_event_user.o
@@ -114,6 +116,7 @@ always += parse_varlen.o parse_simple.o parse_ldabs.o
always += test_cgrp2_tc_kern.o
always += xdp1_kern.o
always += xdp2_kern.o
+always += xdp3_kern.o
always += test_current_task_under_cgroup_kern.o
always += trace_event_kern.o
always += sampleip_kern.o
@@ -160,6 +163,7 @@ HOSTLOADLIBES_map_perf_test += -lelf -lrt
HOSTLOADLIBES_test_overhead += -lelf -lrt
HOSTLOADLIBES_xdp1 += -lelf
HOSTLOADLIBES_xdp2 += -lelf
+HOSTLOADLIBES_xdp3 += -lelf
HOSTLOADLIBES_test_current_task_under_cgroup += -lelf
HOSTLOADLIBES_trace_event += -lelf
HOSTLOADLIBES_sampleip += -lelf
diff --git a/samples/bpf/xdp3_kern.c b/samples/bpf/xdp3_kern.c
new file mode 100644
index 0000000..62d905d
--- /dev/null
+++ b/samples/bpf/xdp3_kern.c
@@ -0,0 +1,204 @@
+/* Copyright (c) 2016 PLUMgrid
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ */
+#define KBUILD_MODNAME "foo"
+#include <uapi/linux/bpf.h>
+#include <linux/in.h>
+#include <linux/if_ether.h>
+#include <linux/if_packet.h>
+#include <linux/if_vlan.h>
+#include <linux/ip.h>
+#include <linux/ipv6.h>
+#include "bpf_helpers.h"
+#include <linux/slab.h>
+#include <net/ip_fib.h>
+
+struct trie_value {
+ __u8 prefix[4];
+ long value;
+ int gw;
+ int ifindex;
+ int metric;
+};
+
+union key_4 {
+ u32 b32[2];
+ u8 b8[8];
+};
+
+struct arp_entry {
+ int dst;
+ long mac;
+};
+
+struct direct_map {
+ long mac;
+ int ifindex;
+ struct arp_entry arp;
+};
+
+/* Map for trie implementation*/
+struct bpf_map_def SEC("maps") lpm_map = {
+ .type = BPF_MAP_TYPE_LPM_TRIE,
+ .key_size = 8,
+ .value_size =
+ sizeof(struct trie_value),
+ .max_entries = 50,
+ .map_flags = BPF_F_NO_PREALLOC,
+};
+
+/* Map for counter*/
+struct bpf_map_def SEC("maps") rxcnt = {
+ .type = BPF_MAP_TYPE_PERCPU_ARRAY,
+ .key_size = sizeof(u32),
+ .value_size = sizeof(long),
+ .max_entries = 256,
+};
+
+/* Map for ARP table*/
+struct bpf_map_def SEC("maps") arp_table = {
+ .type = BPF_MAP_TYPE_HASH,
+ .key_size = sizeof(int),
+ .value_size = sizeof(long),
+ .max_entries = 50,
+};
+
+/* Map to keep the exact match entries in the route table*/
+struct bpf_map_def SEC("maps") exact_match = {
+ .type = BPF_MAP_TYPE_HASH,
+ .key_size = sizeof(int),
+ .value_size = sizeof(struct direct_map),
+ .max_entries = 50,
+};
+
+/**
+ * Function to set source and destination mac of the packet
+ */
+static inline void set_src_dst_mac(void *data, void *src, void *dst)
+{
+ unsigned short *p = data;
+ unsigned short *dest = dst;
+ unsigned short *source = src;
+
+ p[3] = source[0];
+ p[4] = source[1];
+ p[5] = source[2];
+ p[0] = dest[0];
+ p[1] = dest[1];
+ p[2] = dest[2];
+}
+
+/**
+ * Parse IPV4 packet to get SRC, DST IP and protocol
+ */
+static inline int parse_ipv4(void *data, u64 nh_off, void *data_end,
+ unsigned int *src, unsigned int *dest)
+{
+ struct iphdr *iph = data + nh_off;
+
+ if (iph + 1 > data_end)
+ return 0;
+ *src = (unsigned int)iph->saddr;
+ *dest = (unsigned int)iph->daddr;
+ return iph->protocol;
+}
+
+SEC("xdp3")
+int xdp_prog3(struct xdp_md *ctx)
+{
+ void *data_end = (void *)(long)ctx->data_end;
+ void *data = (void *)(long)ctx->data;
+ struct ethhdr *eth = data;
+ int rc = XDP_DROP, forward_to;
+ long *value;
+ struct trie_value *prefix_value;
+ long *dest_mac = NULL, *src_mac = NULL;
+ u16 h_proto;
+ u64 nh_off;
+ u32 ipproto;
+ union key_4 key4;
+
+ nh_off = sizeof(*eth);
+ if (data + nh_off > data_end)
+ return rc;
+
+ h_proto = eth->h_proto;
+
+ if (h_proto == htons(ETH_P_8021Q) || h_proto == htons(ETH_P_8021AD)) {
+ struct vlan_hdr *vhdr;
+
+ vhdr = data + nh_off;
+ nh_off += sizeof(struct vlan_hdr);
+ if (data + nh_off > data_end)
+ return rc;
+ h_proto = vhdr->h_vlan_encapsulated_proto;
+ }
+ if (h_proto == htons(ETH_P_ARP)) {
+ return XDP_PASS;
+ } else if (h_proto == htons(ETH_P_IP)) {
+ int src_ip = 0, dest_ip = 0;
+ struct direct_map *direct_entry;
+
+ ipproto = parse_ipv4(data, nh_off, data_end, &src_ip, &dest_ip);
+ direct_entry = (struct direct_map *)bpf_map_lookup_elem
+ (&exact_match, &dest_ip);
+ /*check for exact match, this would give a faster lookup*/
+ if (direct_entry && direct_entry->mac &&
+ direct_entry->arp.mac) {
+ src_mac = &direct_entry->mac;
+ dest_mac = &direct_entry->arp.mac;
+ forward_to = direct_entry->ifindex;
+ } else {
+ /*Look up in the trie for lpm*/
+ // Key for trie
+ key4.b32[0] = 32;
+ key4.b8[4] = dest_ip % 0x100;
+ key4.b8[5] = (dest_ip >> 8) % 0x100;
+ key4.b8[6] = (dest_ip >> 16) % 0x100;
+ key4.b8[7] = (dest_ip >> 24) % 0x100;
+ prefix_value =
+ ((struct trie_value *)bpf_map_lookup_elem
+ (&lpm_map, &key4));
+ if (!prefix_value) {
+ return XDP_DROP;
+ } else {
+ src_mac = &prefix_value->value;
+ if (src_mac) {
+ dest_mac = (long *)bpf_map_lookup_elem
+ (&arp_table, &dest_ip);
+ if (!dest_mac) {
+ if (prefix_value->gw) {
+ dest_ip = *(unsigned int *)(&(prefix_value->gw));
+ dest_mac = (long *)bpf_map_lookup_elem
+ (&arp_table, &dest_ip);
+ } else {
+ return XDP_DROP;
+ }
+ }
+ forward_to = prefix_value->ifindex;
+ } else {
+ return XDP_DROP;
+ }
+ }
+ }
+ } else {
+ ipproto = 0;
+ }
+ if (src_mac && dest_mac) {
+ set_src_dst_mac(data, src_mac,
+ dest_mac);
+ value = bpf_map_lookup_elem
+ (&rxcnt, &ipproto);
+ if (value)
+ *value += 1;
+ return bpf_redirect(
+ forward_to,
+ 0);
+ }
+ return rc;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/samples/bpf/xdp3_user.c b/samples/bpf/xdp3_user.c
new file mode 100644
index 0000000..451b522
--- /dev/null
+++ b/samples/bpf/xdp3_user.c
@@ -0,0 +1,649 @@
+/* Copyright (c) 2016 PLUMgrid
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ */
+#include <linux/bpf.h>
+#include <linux/netlink.h>
+#include <linux/rtnetlink.h>
+#include <assert.h>
+#include <errno.h>
+#include <signal.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/socket.h>
+#include <unistd.h>
+#include "bpf_load.h"
+#include "libbpf.h"
+#include <arpa/inet.h>
+#include <fcntl.h>
+#include <poll.h>
+#include <net/if.h>
+#include <netdb.h>
+#include <sys/ioctl.h>
+#include "bpf_util.h"
+#include <sys/syscall.h>
+
+int sock, sock_arp, flags = 0;
+char buf[8192];
+static int total_ifindex;
+char **index_list;
+
+static int get_route_table(int rtm_family);
+static void int_exit(int sig)
+{
+ int i = 0, index;
+
+ for (i = 0; i < total_ifindex; i++) {
+ index = strtoul(index_list[i], NULL, 0);
+ set_link_xdp_fd(index, -1, flags);
+ }
+ exit(0);
+}
+
+static void close_and_exit(int sig)
+{
+ int i = 0, index;
+
+ close(sock);
+ close(sock_arp);
+
+ for (i = 0; i < total_ifindex; i++) {
+ index = strtoul(index_list[i], NULL, 0);
+ set_link_xdp_fd(index, -1, flags);
+ }
+ exit(0);
+}
+
+/* Get the mac address of the interface given interface name */
+static long *getmac(char *iface)
+{
+ int fd;
+ struct ifreq ifr;
+ long *mac = NULL;
+
+ fd = socket(AF_INET, SOCK_DGRAM, 0);
+ ifr.ifr_addr.sa_family = AF_INET;
+ strncpy(ifr.ifr_name, iface, IFNAMSIZ - 1);
+ ioctl(fd, SIOCGIFHWADDR, &ifr);
+ mac = (long *)ifr.ifr_hwaddr.sa_data;
+ close(fd);
+ return mac;
+}
+
+static int recv_msg(struct sockaddr_nl sock_addr, int sock)
+{
+ char *buf_ptr;
+ struct nlmsghdr *nh;
+ int len, nll = 0;
+
+ buf_ptr = buf;
+ while (1) {
+ len = recv(sock, buf_ptr, sizeof(buf) - nll, 0);
+ if (len < 0)
+ return len;
+
+ nh = (struct nlmsghdr *)buf_ptr;
+
+ if (nh->nlmsg_type == NLMSG_DONE)
+ break;
+ buf_ptr += len;
+ nll += len;
+ if ((sock_addr.nl_groups & RTMGRP_NEIGH) == RTMGRP_NEIGH)
+ break;
+
+ if ((sock_addr.nl_groups & RTMGRP_IPV4_ROUTE) ==
+ RTMGRP_IPV4_ROUTE)
+ break;
+ }
+ return nll;
+}
+
+/* Function to parse the route entry returned by netlink
+ * Updates the route entry related map entries
+ */
+static void read_route(struct nlmsghdr *nh, int nll)
+{
+ struct route_table {
+ int dst, gw, dst_len, iface, metric;
+ long *mac;
+ char iface_name[IFNAMSIZ];
+ } route;
+ struct arp_table {
+ int dst;
+ long mac;
+ };
+
+ struct direct_map {
+ long mac;
+ int ifindex;
+ struct arp_table arp;
+ } direct_entry;
+ int i;
+ int rtm_family;
+ struct bpf_lpm_trie_key *prefix_key;
+ char dsts[24], gws[24], ifs[16], dsts_len[24], metrics[24];
+ struct rtmsg *rt_msg;
+ int rtl;
+ struct rtattr *rt_attr;
+
+ if (nh->nlmsg_type == RTM_DELROUTE)
+ printf("DELETING Route entry\n");
+ else if (nh->nlmsg_type == RTM_GETROUTE)
+ printf("READING Route entry\n");
+ else if (nh->nlmsg_type == RTM_NEWROUTE)
+ printf("NEW Route entry\n");
+ else
+ printf("%d\n", nh->nlmsg_type);
+
+ bzero(&route, sizeof(route));
+ printf("Destination\tGateway\t\tGenmask\tMetric\tIface\n");
+ for (; NLMSG_OK(nh, nll); nh = NLMSG_NEXT(nh, nll)) {
+ rt_msg = (struct rtmsg *)NLMSG_DATA(nh);
+ rtm_family = rt_msg->rtm_family;
+ if (rtm_family == AF_INET)
+ if (rt_msg->rtm_table != RT_TABLE_MAIN)
+ continue;
+ rt_attr = (struct rtattr *)RTM_RTA(rt_msg);
+ rtl = RTM_PAYLOAD(nh);
+
+ for (; RTA_OK(rt_attr, rtl); rt_attr = RTA_NEXT(rt_attr, rtl)) {
+ switch (rt_attr->rta_type) {
+ case NDA_DST:
+ sprintf(dsts, "%d",
+ *((int *)RTA_DATA(rt_attr)));
+ break;
+ case RTA_GATEWAY:
+ sprintf(gws, "%d", *((int *)RTA_DATA(rt_attr)));
+ break;
+ case RTA_OIF:
+ sprintf(ifs, "%d", *((int *)RTA_DATA(rt_attr)));
+ break;
+ case RTA_METRICS:
+ sprintf(metrics, "%d",
+ *((int *)RTA_DATA(rt_attr)));
+ default:
+ break;
+ }
+ }
+ sprintf(dsts_len, "%d", rt_msg->rtm_dst_len);
+
+ route.dst = atoi(dsts);
+ route.dst_len = atoi(dsts_len);
+ route.gw = atoi(gws);
+ route.iface = atoi(ifs);
+ route.metric = atoi(metrics);
+ if_indextoname(route.iface, route.iface_name);
+ route.mac = getmac(route.iface_name);
+ printf("%x\t\t%x\t\t%d\t%d\t%d\n", route.dst, route.gw,
+ route.dst_len, route.metric, route.iface);
+ if (rtm_family == AF_INET) {
+ struct trie_value {
+ __u8 prefix[4];
+ long value;
+ int gw;
+ int ifindex;
+ int metric;
+ } *prefix_value;
+
+ prefix_key = alloca(sizeof(*prefix_key) + 3);
+ prefix_value = alloca(sizeof(*prefix_value));
+
+ prefix_key->prefixlen = 32;
+ prefix_key->prefixlen = route.dst_len;
+ direct_entry.mac = *route.mac & 0xffffffffffff;
+ direct_entry.ifindex = route.iface;
+ direct_entry.arp.mac = 0;
+ direct_entry.arp.dst = 0;
+ if (route.dst_len == 32) {
+ if (nh->nlmsg_type == RTM_DELROUTE) {
+ assert(bpf_map_delete_elem(
+ map_fd[3],
+ &route.dst
+ ) == 0);
+ } else {
+ if (bpf_map_lookup_elem(map_fd[2],
+ &route.dst,
+ &direct_entry.arp.mac
+ ) == 0)
+ direct_entry.arp.dst = route.dst;
+
+ assert(bpf_map_update_elem(map_fd[3],
+ &route.dst,
+ &direct_entry,
+ 0) == 0);
+ }
+ }
+ for (i = 0; i < 4; i++)
+ prefix_key->data[i] =
+ (route.dst >> i * 8) % 0x100;
+ if (bpf_map_lookup_elem(map_fd[0], prefix_key,
+ prefix_value) < 0) {
+ for (i = 0; i < 4; i++)
+ prefix_value->prefix[i] =
+ prefix_key->data[i];
+ prefix_value->value =
+ *route.mac & 0xffffffffffff;
+ prefix_value->ifindex = route.iface;
+ prefix_value->gw = route.gw;
+ prefix_value->metric = route.metric;
+
+ assert(bpf_map_update_elem(map_fd[0],
+ prefix_key,
+ prefix_value, 0
+ ) == 0);
+ } else {
+ if (nh->nlmsg_type == RTM_DELROUTE) {
+ printf("deleting entry\n");
+ printf("prefix key=%d.%d.%d.%d/%d",
+ prefix_key->data[0],
+ prefix_key->data[1],
+ prefix_key->data[2],
+ prefix_key->data[3],
+ prefix_key->prefixlen);
+ assert(bpf_map_delete_elem(map_fd[0],
+ prefix_key
+ ) == 0);
+ /* Rereading the route table to check if
+ * there is an entry with the same
+ * prefix but a different metric as the
+ * deleted enty.
+ */
+ get_route_table(AF_INET);
+ } else if (prefix_key->data[0] ==
+ prefix_value->prefix[0] &&
+ prefix_key->data[1] ==
+ prefix_value->prefix[1] &&
+ prefix_key->data[2] ==
+ prefix_value->prefix[2] &&
+ prefix_key->data[3] ==
+ prefix_value->prefix[3] &&
+ route.metric >= prefix_value->metric) {
+ continue;
+ } else {
+ for (i = 0; i < 4; i++)
+ prefix_value->prefix[i] =
+ prefix_key->data[i];
+ prefix_value->value =
+ *route.mac & 0xffffffffffff;
+ prefix_value->ifindex = route.iface;
+ prefix_value->gw = route.gw;
+ prefix_value->metric = route.metric;
+ assert(bpf_map_update_elem(
+ map_fd[0],
+ prefix_key,
+ prefix_value,
+ 0) == 0);
+ }
+ }
+ }
+ bzero(&route, sizeof(route));
+ bzero(dsts, sizeof(dsts));
+ bzero(dsts_len, sizeof(dsts_len));
+ bzero(gws, sizeof(gws));
+ bzero(ifs, sizeof(ifs));
+ bzero(&route, sizeof(route));
+ }
+}
+
+/* Function to read the existing route table when the process is launched*/
+static int get_route_table(int rtm_family)
+{
+ struct {
+ struct nlmsghdr nl;
+ struct rtmsg rt;
+ char buf[8192];
+ } req;
+
+ int sock, seq = 0;
+ struct sockaddr_nl sa;
+ struct msghdr msg;
+ struct iovec iov;
+ int ret = 0;
+ struct nlmsghdr *nh;
+ int nll;
+
+ sock = socket(AF_NETLINK, SOCK_RAW, NETLINK_ROUTE);
+ if (sock < 0) {
+ printf("open netlink socket: %s\n", strerror(errno));
+ return -1;
+ }
+ bzero(&sa, sizeof(sa));
+ sa.nl_family = AF_NETLINK;
+ if (bind(sock, (struct sockaddr *)&sa, sizeof(sa)) < 0) {
+ printf("bind to netlink: %s\n", strerror(errno));
+ ret = -1;
+ goto cleanup;
+ }
+ bzero(&req, sizeof(req));
+ req.nl.nlmsg_len = NLMSG_LENGTH(sizeof(struct rtmsg));
+ req.nl.nlmsg_flags = NLM_F_REQUEST | NLM_F_DUMP;
+ req.nl.nlmsg_type = RTM_GETROUTE;
+
+ req.rt.rtm_family = rtm_family;
+ req.rt.rtm_table = RT_TABLE_MAIN;
+ req.nl.nlmsg_pid = 0;
+ req.nl.nlmsg_seq = ++seq;
+ bzero(&msg, sizeof(msg));
+ iov.iov_base = (void *)&req.nl;
+ iov.iov_len = req.nl.nlmsg_len;
+ msg.msg_iov = &iov;
+ msg.msg_iovlen = 1;
+ ret = sendmsg(sock, &msg, 0);
+ if (ret < 0) {
+ printf("send to netlink: %s\n", strerror(errno));
+ ret = -1;
+ goto cleanup;
+ }
+ bzero(buf, sizeof(buf));
+ nll = recv_msg(sa, sock);
+ if (nll < 0) {
+ printf("recv from netlink: %s\n", strerror(nll));
+ ret = -1;
+ goto cleanup;
+ }
+ nh = (struct nlmsghdr *)buf;
+ read_route(nh, nll);
+cleanup:
+ close(sock);
+ return ret;
+}
+
+/* Function to parse the arp entry returned by netlink
+ * Updates the arp entry related map entries
+ */
+static void read_arp(struct nlmsghdr *nh, int nll)
+{
+ struct arp_table {
+ int dst;
+ long mac;
+ } arp_entry;
+ struct direct_map {
+ long mac;
+ int ifindex;
+ struct arp_table arp;
+ } direct_entry;
+
+ char dsts[24], mac[24];
+ struct ndmsg *rt_msg;
+ int rtl, i = 0, ndm_family;
+ struct rtattr *rt_attr;
+
+ if (nh->nlmsg_type == RTM_GETNEIGH)
+ printf("READING arp entry\n");
+ printf("Address\tHwAddress\n");
+ for (; NLMSG_OK(nh, nll); nh = NLMSG_NEXT(nh, nll)) {
+ i++;
+ rt_msg = (struct ndmsg *)NLMSG_DATA(nh);
+ rt_attr = (struct rtattr *)RTM_RTA(rt_msg);
+ ndm_family = rt_msg->ndm_family;
+ rtl = RTM_PAYLOAD(nh);
+ for (; RTA_OK(rt_attr, rtl); rt_attr = RTA_NEXT(rt_attr, rtl)) {
+ switch (rt_attr->rta_type) {
+ case NDA_DST:
+ sprintf(dsts, "%d",
+ *((int *)RTA_DATA(rt_attr)));
+ break;
+ case NDA_LLADDR:
+ sprintf(mac, "%ld",
+ *((long *)RTA_DATA(rt_attr)));
+ break;
+ default:
+ break;
+ }
+ }
+ arp_entry.dst = atoi(dsts);
+ arp_entry.mac = atol(mac);
+ printf("%x\t\t%lx\n", arp_entry.dst, arp_entry.mac);
+ if (ndm_family == AF_INET) {
+ if (bpf_map_lookup_elem(map_fd[3], &arp_entry.dst,
+ &direct_entry) == 0) {
+ if (nh->nlmsg_type == RTM_DELNEIGH) {
+ direct_entry.arp.dst = 0;
+ direct_entry.arp.mac = 0;
+ } else if (nh->nlmsg_type == RTM_NEWNEIGH) {
+ direct_entry.arp.dst = arp_entry.dst;
+ direct_entry.arp.mac = arp_entry.mac;
+ }
+ assert(bpf_map_update_elem(map_fd[3],
+ &arp_entry.dst,
+ &direct_entry, 0
+ ) == 0);
+ bzero(&direct_entry, sizeof(direct_entry));
+ }
+ if (nh->nlmsg_type == RTM_DELNEIGH) {
+ assert(bpf_map_delete_elem(map_fd[2],
+ &arp_entry.dst) == 0);
+ } else if (nh->nlmsg_type == RTM_NEWNEIGH) {
+ assert(bpf_map_update_elem(map_fd[2],
+ &arp_entry.dst,
+ &arp_entry.mac, 0
+ ) == 0);
+ }
+ }
+ bzero(&arp_entry, sizeof(arp_entry));
+ bzero(dsts, sizeof(dsts));
+ }
+}
+
+/* Function to read the existing arp table when the process is launched*/
+static int get_arp_table(int rtm_family)
+{
+ struct {
+ struct nlmsghdr nl;
+ struct ndmsg rt;
+ char buf[8192];
+ } req;
+
+ int sock, seq = 0;
+ struct sockaddr_nl sa;
+ struct msghdr msg;
+ struct iovec iov;
+ int ret = 0;
+ struct nlmsghdr *nh;
+ int nll;
+
+ sock = socket(AF_NETLINK, SOCK_RAW, NETLINK_ROUTE);
+ if (sock < 0) {
+ printf("open netlink socket: %s\n", strerror(errno));
+ return -1;
+ }
+ bzero(&sa, sizeof(sa));
+ sa.nl_family = AF_NETLINK;
+ if (bind(sock, (struct sockaddr *)&sa, sizeof(sa)) < 0) {
+ printf("bind to netlink: %s\n", strerror(errno));
+ ret = -1;
+ goto cleanup;
+ }
+ bzero(&req, sizeof(req));
+ req.nl.nlmsg_len = NLMSG_LENGTH(sizeof(struct rtmsg));
+ req.nl.nlmsg_flags = NLM_F_REQUEST | NLM_F_DUMP;
+ req.nl.nlmsg_type = RTM_GETNEIGH;
+ req.rt.ndm_state = NUD_REACHABLE;
+ req.rt.ndm_family = rtm_family;
+ req.nl.nlmsg_pid = 0;
+ req.nl.nlmsg_seq = ++seq;
+ bzero(&msg, sizeof(msg));
+ iov.iov_base = (void *)&req.nl;
+ iov.iov_len = req.nl.nlmsg_len;
+ msg.msg_iov = &iov;
+ msg.msg_iovlen = 1;
+ ret = sendmsg(sock, &msg, 0);
+ if (ret < 0) {
+ printf("send to netlink: %s\n", strerror(errno));
+ ret = -1;
+ goto cleanup;
+ }
+ bzero(buf, sizeof(buf));
+ nll = recv_msg(sa, sock);
+ if (nll < 0) {
+ printf("recv from netlink: %s\n", strerror(nll));
+ ret = -1;
+ goto cleanup;
+ }
+ nh = (struct nlmsghdr *)buf;
+ read_arp(nh, nll);
+cleanup:
+ close(sock);
+ return ret;
+}
+
+/* Function to keep track and update changes in route and arp table
+ * Give regular statistics of packets forwarded
+ */
+static int monitor_route(void)
+{
+ struct sockaddr_nl la, lr;
+ struct nlmsghdr *nh;
+ int nll, ret = 0;
+ const unsigned int nr_keys = 256;
+ int interval = 5;
+ unsigned int nr_cpus = bpf_num_possible_cpus();
+ __u64 values[nr_cpus], prev[nr_keys][nr_cpus];
+ __u32 key;
+ int i;
+ struct pollfd fds_route, fds_arp;
+
+ sock = socket(AF_NETLINK, SOCK_RAW, NETLINK_ROUTE);
+ if (sock < 0) {
+ printf("open netlink socket: %s\n", strerror(errno));
+ return -1;
+ }
+
+ fcntl(sock, F_SETFL, O_NONBLOCK);
+ bzero(&lr, sizeof(lr));
+ lr.nl_family = AF_NETLINK;
+ lr.nl_groups = RTMGRP_IPV6_ROUTE | RTMGRP_IPV4_ROUTE | RTMGRP_NOTIFY;
+ if (bind(sock, (struct sockaddr *)&lr, sizeof(lr)) < 0) {
+ printf("bind to netlink: %s\n", strerror(errno));
+ ret = -1;
+ goto cleanup;
+ }
+ fds_route.fd = sock;
+ fds_route.events = POLL_IN;
+
+ sock_arp = socket(AF_NETLINK, SOCK_RAW, NETLINK_ROUTE);
+ if (sock_arp < 0) {
+ printf("open netlink socket: %s\n", strerror(errno));
+ return -1;
+ }
+
+ fcntl(sock_arp, F_SETFL, O_NONBLOCK);
+ bzero(&la, sizeof(la));
+ la.nl_family = AF_NETLINK;
+ la.nl_groups = RTMGRP_NEIGH | RTMGRP_NOTIFY;
+ if (bind(sock_arp, (struct sockaddr *)&la, sizeof(la)) < 0) {
+ printf("bind to netlink: %s\n", strerror(errno));
+ ret = -1;
+ goto cleanup;
+ }
+ fds_arp.fd = sock_arp;
+ fds_arp.events = POLL_IN;
+
+ memset(prev, 0, sizeof(prev));
+ do {
+ signal(SIGINT, close_and_exit);
+ signal(SIGTERM, close_and_exit);
+
+ sleep(interval);
+ for (key = 0; key < nr_keys; key++) {
+ __u64 sum = 0;
+
+ assert(bpf_map_lookup_elem(map_fd[1], &key, values) == 0);
+ for (i = 0; i < nr_cpus; i++)
+ sum += (values[i] - prev[key][i]);
+ if (sum)
+ printf("proto %u: %10llu pkt/s\n",
+ key, sum / interval);
+ memcpy(prev[key], values, sizeof(values));
+ }
+
+ bzero(buf, sizeof(buf));
+ if (poll(&fds_route, 1, 3) == POLL_IN) {
+ nll = recv_msg(lr, sock);
+ if (nll < 0) {
+ printf("recv from netlink: %s\n",
+ strerror(nll));
+ ret = -1;
+ goto cleanup;
+ }
+
+ nh = (struct nlmsghdr *)buf;
+ printf("Routing table updated.\n");
+ read_route(nh, nll);
+ }
+ bzero(buf, sizeof(buf));
+ if (poll(&fds_arp, 1, 3) == POLL_IN) {
+ nll = recv_msg(la, sock_arp);
+ if (nll < 0) {
+ printf("recv from netlink: %s\n",
+ strerror(nll));
+ ret = -1;
+ goto cleanup;
+ }
+
+ nh = (struct nlmsghdr *)buf;
+ read_arp(nh, nll);
+ }
+
+ } while (1);
+cleanup:
+ close(sock);
+ return ret;
+}
+
+int main(int ac, char **argv)
+{
+ char filename[256];
+ int i = 1, index;
+
+ snprintf(filename, sizeof(filename), "%s_kern.o", argv[0]);
+ printf("Entering user program\n");
+ if (ac < 2) {
+ printf("usage: %s [-S] IFINDEX\n", argv[0]);
+ return 1;
+ }
+ if (!strcmp(argv[1], "-S")) {
+ flags = XDP_FLAGS_SKB_MODE;
+ total_ifindex = ac - 2;
+ index_list = (argv + 2);
+ } else {
+ flags = 0;
+ total_ifindex = ac - 1;
+ index_list = (argv + 1);
+ }
+printf("Loading bpf program\n");
+ if (load_bpf_file(filename)) {
+ printf("%s", bpf_log_buf);
+ return 1;
+ }
+ printf("\n**************loading bpf file*********************\n\n\n");
+ if (!prog_fd[0]) {
+ printf("load_bpf_file: %s\n", strerror(errno));
+ return 1;
+ }
+
+ for (i = 0; i < total_ifindex; i++) {
+ index = strtoul(index_list[i], NULL, 0);
+ if (set_link_xdp_fd(index, prog_fd[0], flags) < 0) {
+ printf("link set xdp fd failed\n");
+ return 1;
+ }
+ printf("Attached to %d\n", index);
+ }
+ signal(SIGINT, int_exit);
+ signal(SIGTERM, int_exit);
+
+ printf("*******************ROUTE TABLE*************************\n\n\n");
+ get_route_table(AF_INET);
+ printf("*******************ARP TABLE***************************\n\n\n");
+ get_arp_table(AF_INET);
+ if (monitor_route() < 0) {
+ printf("Error in receiving route update");
+ return 1;
+ }
+
+ return 0;
+}
--
1.7.1
^ permalink raw reply related
* [PATCH 0/1] XDP Program for Ip forward
From: cjacob @ 2017-10-03 7:37 UTC (permalink / raw)
To: netdev; +Cc: Christina.Jacob, linux-kernel, linux-arm-kernel
The patch below implements port to port forwarding through route table and arp
table lookup for ipv4 packets using bpf_redirect helper function and lpm_trie
map. This has an improved performance over the normal kernel stack ip forward.
Implementation details.
-----------------------
The program uses one map each for arp table, route table and packet count.
The number of entries the program can process is limited by the size of the
map used.
In the xdp3_user.c,
initially, the routing table is read and stored in an lpm trie map.
The arp table is read and stored in an array map. There are two netlink sockets
that listen to any change in the route table and arp table.
There are two types of changes to the route table.
1.New
The new entries are added to the lpm trie with proper key and prefix
length. If there is a another entry in the route table with a different
metric(only metric is considered), then the values are compared and the
one with lowest metric is added to the trie node.
2.Deletion
On deletion from the route table, the particular node is removed and the
entire route table is read again to check if there is another entry with
the same key and prefix length but a different metric. If it exists it
is added to the lpm trie.
This implementation depends on Patch bpf: Implement map_delete_elem for
BPF_MAP_TYPE_LPM_TRIE which is not yet upstreamed.
There are two types of changes to the route table
1.New
The new arp entries are added in the in the array map directly with the
ip address as the key and the destination mac address as the value.
2.Delete
The entry corresponding to the particular ip is deleted from the
arp table map.
Another map is maintained for entries in the route table having 32 bit mask.
Such entries can have a corresponding arp entry which if stored together with
the route entry in an array map and can be accessed in O(1) time, thereby
eliminating the trie lookup and arp lookup.
In the xdp3_kern.c,
The array map for the 32 bit mask entries is checked to see if there is a key
that exactly matches the destination ip. If it has a non zero destination mac
entry then the xdp data is updated accordingly. Otherwise a proper route and
arp table lookup is done using the lpm_trie and the arp table array map.
Usage: ./xdp3 [-S] <ifindex1...ifindexn>
-S to choose generic xdp implementation
[Default is driver xdp implementation]
ifindex - the index of the interface to which
the xdp program has to be attached.
in 4.14-rc3 kernel.
cjacob (1):
xdp: Sample xdp program implementing ip forward
samples/bpf/Makefile | 4 +
samples/bpf/xdp3_kern.c | 204 +++++++++++++++
samples/bpf/xdp3_user.c | 649 +++++++++++++++++++++++++++++++++++++++++++++++
3 files changed, 857 insertions(+), 0 deletions(-)
create mode 100644 samples/bpf/xdp3_kern.c
create mode 100644 samples/bpf/xdp3_user.c
^ permalink raw reply
* Re: [PATCH 1/2] net: phonet: mark header_ops as const
From: kbuild test robot @ 2017-10-03 7:36 UTC (permalink / raw)
To: Lin Zhang; +Cc: kbuild-all, davem, netdev, courmisch, Lin Zhang
In-Reply-To: <1506762519-4709-1-git-send-email-xiaolou4617@gmail.com>
[-- Attachment #1: Type: text/plain, Size: 11254 bytes --]
Hi Lin,
[auto build test WARNING on net-next/master]
[also build test WARNING on v4.14-rc3 next-20170929]
[if your patch is applied to the wrong git tree, please drop us a note to help improve the system]
url: https://github.com/0day-ci/linux/commits/Lin-Zhang/net-phonet-mark-header_ops-as-const/20171003-145726
config: i386-randconfig-x009-201740 (attached as .config)
compiler: gcc-6 (Debian 6.2.0-3) 6.2.0 20160901
reproduce:
# save the attached .config to linux build tree
make ARCH=i386
All warnings (new ones prefixed by >>):
net/phonet/af_phonet.c:152:25: error: conflicting type qualifiers for 'phonet_header_ops'
const struct header_ops phonet_header_ops = {
^~~~~~~~~~~~~~~~~
In file included from net/phonet/af_phonet.c:32:0:
include/linux/if_phonet.h:13:26: note: previous declaration of 'phonet_header_ops' was here
extern struct header_ops phonet_header_ops;
^~~~~~~~~~~~~~~~~
In file included from include/linux/linkage.h:6:0,
from include/linux/kernel.h:6,
from net/phonet/af_phonet.c:26:
net/phonet/af_phonet.c:156:15: error: conflicting type qualifiers for 'phonet_header_ops'
EXPORT_SYMBOL(phonet_header_ops);
^
include/linux/export.h:65:21: note: in definition of macro '___EXPORT_SYMBOL'
extern typeof(sym) sym; \
^~~
>> net/phonet/af_phonet.c:156:1: note: in expansion of macro 'EXPORT_SYMBOL'
EXPORT_SYMBOL(phonet_header_ops);
^~~~~~~~~~~~~
In file included from net/phonet/af_phonet.c:32:0:
include/linux/if_phonet.h:13:26: note: previous declaration of 'phonet_header_ops' was here
extern struct header_ops phonet_header_ops;
^~~~~~~~~~~~~~~~~
vim +/EXPORT_SYMBOL +156 net/phonet/af_phonet.c
4b07b3f69 Remi Denis-Courmont 2008-09-22 @26 #include <linux/kernel.h>
4b07b3f69 Remi Denis-Courmont 2008-09-22 27 #include <linux/module.h>
5a0e3ad6a Tejun Heo 2010-03-24 28 #include <linux/slab.h>
4b07b3f69 Remi Denis-Courmont 2008-09-22 29 #include <asm/unaligned.h>
4b07b3f69 Remi Denis-Courmont 2008-09-22 30 #include <net/sock.h>
4b07b3f69 Remi Denis-Courmont 2008-09-22 31
4b07b3f69 Remi Denis-Courmont 2008-09-22 32 #include <linux/if_phonet.h>
4b07b3f69 Remi Denis-Courmont 2008-09-22 33 #include <linux/phonet.h>
4b07b3f69 Remi Denis-Courmont 2008-09-22 34 #include <net/phonet/phonet.h>
f8ff60283 Remi Denis-Courmont 2008-09-22 35 #include <net/phonet/pn_dev.h>
4b07b3f69 Remi Denis-Courmont 2008-09-22 36
566521d63 Alexey Dobriyan 2008-11-19 37 /* Transport protocol registration */
566521d63 Alexey Dobriyan 2008-11-19 38 static struct phonet_protocol *proto_tab[PHONET_NPROTO] __read_mostly;
566521d63 Alexey Dobriyan 2008-11-19 39
facb4edc1 Dan Carpenter 2011-01-10 40 static struct phonet_protocol *phonet_proto_get(unsigned int protocol)
566521d63 Alexey Dobriyan 2008-11-19 41 {
566521d63 Alexey Dobriyan 2008-11-19 42 struct phonet_protocol *pp;
566521d63 Alexey Dobriyan 2008-11-19 43
566521d63 Alexey Dobriyan 2008-11-19 44 if (protocol >= PHONET_NPROTO)
566521d63 Alexey Dobriyan 2008-11-19 45 return NULL;
566521d63 Alexey Dobriyan 2008-11-19 46
7ed0132f2 Rémi Denis-Courmont 2009-11-13 47 rcu_read_lock();
b2a5decdd Rémi Denis-Courmont 2009-11-16 48 pp = rcu_dereference(proto_tab[protocol]);
566521d63 Alexey Dobriyan 2008-11-19 49 if (pp && !try_module_get(pp->prot->owner))
566521d63 Alexey Dobriyan 2008-11-19 50 pp = NULL;
7ed0132f2 Rémi Denis-Courmont 2009-11-13 51 rcu_read_unlock();
566521d63 Alexey Dobriyan 2008-11-19 52
566521d63 Alexey Dobriyan 2008-11-19 53 return pp;
566521d63 Alexey Dobriyan 2008-11-19 54 }
566521d63 Alexey Dobriyan 2008-11-19 55
566521d63 Alexey Dobriyan 2008-11-19 56 static inline void phonet_proto_put(struct phonet_protocol *pp)
566521d63 Alexey Dobriyan 2008-11-19 57 {
566521d63 Alexey Dobriyan 2008-11-19 58 module_put(pp->prot->owner);
566521d63 Alexey Dobriyan 2008-11-19 59 }
4b07b3f69 Remi Denis-Courmont 2008-09-22 60
4b07b3f69 Remi Denis-Courmont 2008-09-22 61 /* protocol family functions */
4b07b3f69 Remi Denis-Courmont 2008-09-22 62
3f378b684 Eric Paris 2009-11-05 63 static int pn_socket_create(struct net *net, struct socket *sock, int protocol,
3f378b684 Eric Paris 2009-11-05 64 int kern)
4b07b3f69 Remi Denis-Courmont 2008-09-22 65 {
ba113a94b Remi Denis-Courmont 2008-09-22 66 struct sock *sk;
ba113a94b Remi Denis-Courmont 2008-09-22 67 struct pn_sock *pn;
4b07b3f69 Remi Denis-Courmont 2008-09-22 68 struct phonet_protocol *pnp;
4b07b3f69 Remi Denis-Courmont 2008-09-22 69 int err;
4b07b3f69 Remi Denis-Courmont 2008-09-22 70
4b07b3f69 Remi Denis-Courmont 2008-09-22 71 if (!capable(CAP_SYS_ADMIN))
4b07b3f69 Remi Denis-Courmont 2008-09-22 72 return -EPERM;
4b07b3f69 Remi Denis-Courmont 2008-09-22 73
4b07b3f69 Remi Denis-Courmont 2008-09-22 74 if (protocol == 0) {
4b07b3f69 Remi Denis-Courmont 2008-09-22 75 /* Default protocol selection */
4b07b3f69 Remi Denis-Courmont 2008-09-22 76 switch (sock->type) {
4b07b3f69 Remi Denis-Courmont 2008-09-22 77 case SOCK_DGRAM:
4b07b3f69 Remi Denis-Courmont 2008-09-22 78 protocol = PN_PROTO_PHONET;
4b07b3f69 Remi Denis-Courmont 2008-09-22 79 break;
9641458d3 Rémi Denis-Courmont 2008-10-05 80 case SOCK_SEQPACKET:
9641458d3 Rémi Denis-Courmont 2008-10-05 81 protocol = PN_PROTO_PIPE;
9641458d3 Rémi Denis-Courmont 2008-10-05 82 break;
4b07b3f69 Remi Denis-Courmont 2008-09-22 83 default:
4b07b3f69 Remi Denis-Courmont 2008-09-22 84 return -EPROTONOSUPPORT;
4b07b3f69 Remi Denis-Courmont 2008-09-22 85 }
4b07b3f69 Remi Denis-Courmont 2008-09-22 86 }
4b07b3f69 Remi Denis-Courmont 2008-09-22 87
4b07b3f69 Remi Denis-Courmont 2008-09-22 88 pnp = phonet_proto_get(protocol);
25532824f Rémi Denis-Courmont 2008-10-05 89 if (pnp == NULL &&
25532824f Rémi Denis-Courmont 2008-10-05 90 request_module("net-pf-%d-proto-%d", PF_PHONET, protocol) == 0)
25532824f Rémi Denis-Courmont 2008-10-05 91 pnp = phonet_proto_get(protocol);
95a5afca4 Johannes Berg 2008-10-16 92
4b07b3f69 Remi Denis-Courmont 2008-09-22 93 if (pnp == NULL)
4b07b3f69 Remi Denis-Courmont 2008-09-22 94 return -EPROTONOSUPPORT;
4b07b3f69 Remi Denis-Courmont 2008-09-22 95 if (sock->type != pnp->sock_type) {
4b07b3f69 Remi Denis-Courmont 2008-09-22 96 err = -EPROTONOSUPPORT;
4b07b3f69 Remi Denis-Courmont 2008-09-22 97 goto out;
4b07b3f69 Remi Denis-Courmont 2008-09-22 98 }
4b07b3f69 Remi Denis-Courmont 2008-09-22 99
11aa9c28b Eric W. Biederman 2015-05-08 100 sk = sk_alloc(net, PF_PHONET, GFP_KERNEL, pnp->prot, kern);
ba113a94b Remi Denis-Courmont 2008-09-22 101 if (sk == NULL) {
ba113a94b Remi Denis-Courmont 2008-09-22 102 err = -ENOMEM;
ba113a94b Remi Denis-Courmont 2008-09-22 103 goto out;
ba113a94b Remi Denis-Courmont 2008-09-22 104 }
ba113a94b Remi Denis-Courmont 2008-09-22 105
ba113a94b Remi Denis-Courmont 2008-09-22 106 sock_init_data(sock, sk);
ba113a94b Remi Denis-Courmont 2008-09-22 107 sock->state = SS_UNCONNECTED;
ba113a94b Remi Denis-Courmont 2008-09-22 108 sock->ops = pnp->ops;
ba113a94b Remi Denis-Courmont 2008-09-22 109 sk->sk_backlog_rcv = sk->sk_prot->backlog_rcv;
ba113a94b Remi Denis-Courmont 2008-09-22 110 sk->sk_protocol = protocol;
ba113a94b Remi Denis-Courmont 2008-09-22 111 pn = pn_sk(sk);
ba113a94b Remi Denis-Courmont 2008-09-22 112 pn->sobject = 0;
a8059512b Rémi Denis-Courmont 2011-02-24 113 pn->dobject = 0;
ba113a94b Remi Denis-Courmont 2008-09-22 114 pn->resource = 0;
ba113a94b Remi Denis-Courmont 2008-09-22 115 sk->sk_prot->init(sk);
ba113a94b Remi Denis-Courmont 2008-09-22 116 err = 0;
4b07b3f69 Remi Denis-Courmont 2008-09-22 117
4b07b3f69 Remi Denis-Courmont 2008-09-22 118 out:
4b07b3f69 Remi Denis-Courmont 2008-09-22 119 phonet_proto_put(pnp);
4b07b3f69 Remi Denis-Courmont 2008-09-22 120 return err;
4b07b3f69 Remi Denis-Courmont 2008-09-22 121 }
4b07b3f69 Remi Denis-Courmont 2008-09-22 122
ec1b4cf74 Stephen Hemminger 2009-10-05 123 static const struct net_proto_family phonet_proto_family = {
25532824f Rémi Denis-Courmont 2008-10-05 124 .family = PF_PHONET,
4b07b3f69 Remi Denis-Courmont 2008-09-22 125 .create = pn_socket_create,
4b07b3f69 Remi Denis-Courmont 2008-09-22 126 .owner = THIS_MODULE,
4b07b3f69 Remi Denis-Courmont 2008-09-22 127 };
4b07b3f69 Remi Denis-Courmont 2008-09-22 128
5f77076d7 Remi Denis-Courmont 2008-09-22 129 /* Phonet device header operations */
5f77076d7 Remi Denis-Courmont 2008-09-22 130 static int pn_header_create(struct sk_buff *skb, struct net_device *dev,
5f77076d7 Remi Denis-Courmont 2008-09-22 131 unsigned short type, const void *daddr,
95c961747 Eric Dumazet 2012-04-15 132 const void *saddr, unsigned int len)
5f77076d7 Remi Denis-Courmont 2008-09-22 133 {
5f77076d7 Remi Denis-Courmont 2008-09-22 134 u8 *media = skb_push(skb, 1);
5f77076d7 Remi Denis-Courmont 2008-09-22 135
5f77076d7 Remi Denis-Courmont 2008-09-22 136 if (type != ETH_P_PHONET)
5f77076d7 Remi Denis-Courmont 2008-09-22 137 return -1;
5f77076d7 Remi Denis-Courmont 2008-09-22 138
5f77076d7 Remi Denis-Courmont 2008-09-22 139 if (!saddr)
5f77076d7 Remi Denis-Courmont 2008-09-22 140 saddr = dev->dev_addr;
5f77076d7 Remi Denis-Courmont 2008-09-22 141 *media = *(const u8 *)saddr;
5f77076d7 Remi Denis-Courmont 2008-09-22 142 return 1;
5f77076d7 Remi Denis-Courmont 2008-09-22 143 }
5f77076d7 Remi Denis-Courmont 2008-09-22 144
5f77076d7 Remi Denis-Courmont 2008-09-22 145 static int pn_header_parse(const struct sk_buff *skb, unsigned char *haddr)
5f77076d7 Remi Denis-Courmont 2008-09-22 146 {
5f77076d7 Remi Denis-Courmont 2008-09-22 147 const u8 *media = skb_mac_header(skb);
5f77076d7 Remi Denis-Courmont 2008-09-22 148 *haddr = *media;
5f77076d7 Remi Denis-Courmont 2008-09-22 149 return 1;
5f77076d7 Remi Denis-Courmont 2008-09-22 150 }
5f77076d7 Remi Denis-Courmont 2008-09-22 151
9ffaa93ed Lin Zhang 2017-09-30 152 const struct header_ops phonet_header_ops = {
5f77076d7 Remi Denis-Courmont 2008-09-22 153 .create = pn_header_create,
5f77076d7 Remi Denis-Courmont 2008-09-22 154 .parse = pn_header_parse,
5f77076d7 Remi Denis-Courmont 2008-09-22 155 };
5f77076d7 Remi Denis-Courmont 2008-09-22 @156 EXPORT_SYMBOL(phonet_header_ops);
5f77076d7 Remi Denis-Courmont 2008-09-22 157
:::::: The code at line 156 was first introduced by commit
:::::: 5f77076d75d35c9f5619e1f9d7e7428a627f65e6 Phonet: provide MAC header operations
:::::: TO: Remi Denis-Courmont <remi.denis-courmont@nokia.com>
:::::: CC: David S. Miller <davem@davemloft.net>
---
0-DAY kernel test infrastructure Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all Intel Corporation
[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 24579 bytes --]
^ permalink raw reply
* Re: Fw: [Bug 197099] New: Kernel panic in interrupt [l2tp_ppp]
From: James Chapman @ 2017-10-03 7:27 UTC (permalink / raw)
To: SviMik; +Cc: netdev
In-Reply-To: <CA++DawY6CPh5S8A7pPE+-nwwGxJ4=pKL_MkXT_Q5TZb8xFnCjA@mail.gmail.com>
On 2 October 2017 at 19:35, SviMik <svimik@gmail.com> wrote:
> Hi, James!
>
> No, I'm suffering from kernel panics since I started using 4.x
> kernels.
It's interesting that you are seeing l2tp issues since switching to
4.x kernels. Are you able to try earlier kernels to find the latest
version that works? I'm curious whether things broke at v3.15.
> See my current collection:
> http://svimik.com/hdmmsk1kp1.png
This one is another crash implicating l2tp socket shutdown, this time
when the session pppol2tp socket is closed. Unfortunately the
screenshot doesn't show the full oops text. I'll investigate this one
too if you can get a full oops capture.
> http://svimik.com/hdmmsk2kp2.png
> http://svimik.com/hdmmsk2kp3.png
These are both the same crash as the oops of this bug report.
> http://svimik.com/hdmmsk2kp4.png
> http://svimik.com/hdmmsk2kp5.png
These are truncated oops.
> http://svimik.com/hdmmsk7kp1.png
Same crash as hdmmsk1kp1.png
> Screenshots are from three different machines, kernels from 4.8.13 to 4.13.4.
For capturing complete oops messages, have you tried setting up
netconsole? You might also find the full text in the syslog on reboot.
> 2017-10-02 16:56 GMT+03:00 Eric Dumazet <eric.dumazet@gmail.com>:
>> CC svimik@gmail.com so that he is aware of this netdev thread.
>>
>> On Mon, 2017-10-02 at 14:32 +0100, James Chapman wrote:
>>> This seems to be a NULL pointer exception caused by tunnel->sock being
>>> NULL at the call to bh_lock_sock() in l2tp_xmit_skb() at
>>> l2tp_core.c:1135.
>>>
>>> tunnel->sock is set NULL in l2tp_core's tunnel socket destructor.
>>>
>>> At the moment, I don't understand how this happens because
>>> pppol2tp_xmit() does a sock_hold() on the tunnel socket before
>>> l2tp_xmit_skb() is called. I'm still looking at this.
>>>
>>> Has this problem only recently started happening?
>>>
>>>
>>>
>>>
>>>
>>> On 1 October 2017 at 18:21, Stephen Hemminger
>>> <stephen@networkplumber.org> wrote:
>>> >
>>> >
>>> > Begin forwarded message:
>>> >
>>> > Date: Sun, 01 Oct 2017 16:22:33 +0000
>>> > From: bugzilla-daemon@bugzilla.kernel.org
>>> > To: stephen@networkplumber.org
>>> > Subject: [Bug 197099] New: Kernel panic in interrupt [l2tp_ppp]
>>> >
>>> >
>>> > https://bugzilla.kernel.org/show_bug.cgi?id=197099
>>> >
>>> > Bug ID: 197099
>>> > Summary: Kernel panic in interrupt [l2tp_ppp]
>>> > Product: Networking
>>> > Version: 2.5
>>> > Kernel Version: 4.8.13-1.el6.elrepo.x86_64
>>> > Hardware: x86-64
>>> > OS: Linux
>>> > Tree: Mainline
>>> > Status: NEW
>>> > Severity: normal
>>> > Priority: P1
>>> > Component: Other
>>> > Assignee: stephen@networkplumber.org
>>> > Reporter: svimik@gmail.com
>>> > Regression: No
>>> >
>>> > Created attachment 258685
>>> > --> https://bugzilla.kernel.org/attachment.cgi?id=258685&action=edit
>>> > stacktrace screenshot
>>> >
>>> > Hello!
>>> >
>>> > Getting kernel panics on multiple servers. Since it mentions l2tp_core,
>>> > l2tp_ppp and ppp_generic, I decided to report it to Networking (correct me if
>>> > I'm wrong).
>>> >
>>> > Unfortunately I'm still struggling with making kdump work, so the trace
>>> > screenshot is all I have at this moment. The only hope is that this stacktrace
>>> > means something to the guys that wrote the code.
>>> >
>>> > --
>>> > You are receiving this mail because:
>>> > You are the assignee for the bug.
>>
>>
^ permalink raw reply
* Re: [net-next V3 PATCH 3/5] bpf: cpumap xdp_buff to skb conversion and allocation
From: Jesper Dangaard Brouer @ 2017-10-03 6:58 UTC (permalink / raw)
To: Alexei Starovoitov
Cc: netdev, jakub.kicinski, Michael S. Tsirkin, pavel.odintsov,
Jason Wang, mchan, John Fastabend, peter.waskiewicz.jr,
Daniel Borkmann, Andy Gospodarek, brouer
In-Reply-To: <20171003010245.f3op4t56crbjc4ke@ast-mbp>
On Mon, 2 Oct 2017 18:02:46 -0700
Alexei Starovoitov <alexei.starovoitov@gmail.com> wrote:
> On Mon, Oct 02, 2017 at 06:05:29PM +0200, Jesper Dangaard Brouer wrote:
> > + while ((xdp_pkt = __ptr_ring_consume(rcpu->queue))) {
> > + struct sk_buff *skb;
> > + int ret;
> > +
> > + /* Allow busy polling again */
> > + empty_cnt = 0;
> > +
> > + skb = cpu_map_build_skb(rcpu, xdp_pkt);
> > + if (!skb) {
> > + page_frag_free(xdp_pkt);
> > + continue;
> > + }
> > +
> > + /* Inject into network stack */
> > + ret = netif_receive_skb(skb);
> > + if (ret == NET_RX_DROP)
> > + drops++;
>
> I thought the whole thing is an alternative to RPS,
> but netif_receive_skb_internal() will call into RPS logic.
> So the user has to make sure it disabled or they will
> conflict in some weird way?
In this patchset, cpumap and RPS are independent, and there is nothing
wrong with running RPS after cpumap have placed the SKB on a CPU.
Combining the two does seem a little weird. Especially since cpumap
doesn't (yet) transfer the HW-rxhash, thus extra SW-rxhash work will be
done by RPS.
I like you ABI argument. While combining RPS+cpumap is technically
possible, there isn't a good use-case for this. Thus, we should not
open this possibility, as we would need to support this combination
forever.
> Or you're calling netif_receive_skb() to be able to call
> generic XDP on that cpu again ?
That should not (currently) be possible. AFAIK we (Daniel) choose to
not allow Native and Generic XDP to be loaded on the same net_device.
(With the same ABI argument as here)
> But that prog can do cpumap redirect again?
> sort-of recursive redirect? Is it really useful?
> May be call into __netif_receive_skb_core() directly?
> not sure.
I like the idea of calling __netif_receive_skb_core() directly. I'll
send a V4 (after running my different benchmarks).
> I'm asking all these questions to make sure we think through
> these implications before it becomes an abi.
I fully follow your ABI argument. Thank you for bringing this up!
Do notice, that I expect to change this code path (later), to support
GRO. But it would be beneficial to get the HW-rxhash working first, as
it will speedup the GRO "same_flow" check, and allow cpumap to
distribute packets better.
--
Best regards,
Jesper Dangaard Brouer
MSc.CS, Principal Kernel Engineer at Red Hat
LinkedIn: http://www.linkedin.com/in/brouer
^ permalink raw reply
* [PATCH 6/7] crypto:chelsio:Move DMA un/mapping to chcr from lld cxgb4 driver
From: Harsh Jain @ 2017-10-03 6:46 UTC (permalink / raw)
To: herbert, linux-crypto, netdev; +Cc: Harsh Jain, Ganesh Goudar
In-Reply-To: <cover.1507010612.git.harsh@chelsio.com>
Allow chcr to do DMA mapping/Unmapping instead of lld cxgb4.
It moves "Copy AAD to dst buffer" requirement from driver to
firmware.
Signed-off-by: Ganesh Goudar <ganeshgr@chelsio.com>
Signed-off-by: Harsh Jain <harsh@chelsio.com>
---
drivers/crypto/chelsio/chcr_algo.c | 1645 ++++++++++++++++++------------
drivers/crypto/chelsio/chcr_algo.h | 44 +-
drivers/crypto/chelsio/chcr_crypto.h | 114 ++-
drivers/net/ethernet/chelsio/cxgb4/sge.c | 8 +-
4 files changed, 1116 insertions(+), 695 deletions(-)
diff --git a/drivers/crypto/chelsio/chcr_algo.c b/drivers/crypto/chelsio/chcr_algo.c
index b13991d..646dfff 100644
--- a/drivers/crypto/chelsio/chcr_algo.c
+++ b/drivers/crypto/chelsio/chcr_algo.c
@@ -70,6 +70,8 @@
#include "chcr_algo.h"
#include "chcr_crypto.h"
+#define IV AES_BLOCK_SIZE
+
static inline struct chcr_aead_ctx *AEAD_CTX(struct chcr_context *ctx)
{
return ctx->crypto_ctx->aeadctx;
@@ -102,7 +104,7 @@ static inline struct uld_ctx *ULD_CTX(struct chcr_context *ctx)
static inline int is_ofld_imm(const struct sk_buff *skb)
{
- return (skb->len <= CRYPTO_MAX_IMM_TX_PKT_LEN);
+ return (skb->len <= SGE_MAX_WR_LEN);
}
/*
@@ -117,21 +119,92 @@ static inline unsigned int sgl_len(unsigned int n)
return (3 * n) / 2 + (n & 1) + 2;
}
-static int dstsg_2k(struct scatterlist *sgl, unsigned int reqlen)
+static int sg_nents_xlen(struct scatterlist *sg, unsigned int reqlen,
+ unsigned int entlen,
+ unsigned int skip)
{
int nents = 0;
unsigned int less;
+ unsigned int skip_len = 0;
- while (sgl && reqlen) {
- less = min(reqlen, sgl->length);
- nents += DIV_ROUND_UP(less, CHCR_SG_SIZE);
- reqlen -= less;
- sgl = sg_next(sgl);
+ while (sg && skip) {
+ if (sg_dma_len(sg) <= skip) {
+ skip -= sg_dma_len(sg);
+ skip_len = 0;
+ sg = sg_next(sg);
+ } else {
+ skip_len = skip;
+ skip = 0;
+ }
}
+ while (sg && reqlen) {
+ less = min(reqlen, sg_dma_len(sg) - skip_len);
+ nents += DIV_ROUND_UP(less, entlen);
+ reqlen -= less;
+ skip_len = 0;
+ sg = sg_next(sg);
+ }
return nents;
}
+static inline void chcr_handle_ahash_resp(struct ahash_request *req,
+ unsigned char *input,
+ int err)
+{
+ struct chcr_ahash_req_ctx *reqctx = ahash_request_ctx(req);
+ int digestsize, updated_digestsize;
+ struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
+ struct uld_ctx *u_ctx = ULD_CTX(h_ctx(tfm));
+
+ if (input == NULL)
+ goto out;
+ reqctx = ahash_request_ctx(req);
+ digestsize = crypto_ahash_digestsize(crypto_ahash_reqtfm(req));
+ if (reqctx->is_sg_map)
+ chcr_hash_dma_unmap(&u_ctx->lldi.pdev->dev, req);
+ if (reqctx->dma_addr)
+ dma_unmap_single(&u_ctx->lldi.pdev->dev, reqctx->dma_addr,
+ reqctx->dma_len, DMA_TO_DEVICE);
+ reqctx->dma_addr = 0;
+ updated_digestsize = digestsize;
+ if (digestsize == SHA224_DIGEST_SIZE)
+ updated_digestsize = SHA256_DIGEST_SIZE;
+ else if (digestsize == SHA384_DIGEST_SIZE)
+ updated_digestsize = SHA512_DIGEST_SIZE;
+ if (reqctx->result == 1) {
+ reqctx->result = 0;
+ memcpy(req->result, input + sizeof(struct cpl_fw6_pld),
+ digestsize);
+ } else {
+ memcpy(reqctx->partial_hash, input + sizeof(struct cpl_fw6_pld),
+ updated_digestsize);
+ }
+out:
+ req->base.complete(&req->base, err);
+
+ }
+
+static inline void chcr_handle_aead_resp(struct aead_request *req,
+ unsigned char *input,
+ int err)
+{
+ struct chcr_aead_reqctx *reqctx = aead_request_ctx(req);
+ struct crypto_aead *tfm = crypto_aead_reqtfm(req);
+ struct uld_ctx *u_ctx = ULD_CTX(a_ctx(tfm));
+
+
+ chcr_aead_dma_unmap(&u_ctx->lldi.pdev->dev, req, reqctx->op);
+ if (reqctx->b0_dma)
+ dma_unmap_single(&u_ctx->lldi.pdev->dev, reqctx->b0_dma,
+ reqctx->b0_len, DMA_BIDIRECTIONAL);
+ if (reqctx->verify == VERIFY_SW) {
+ chcr_verify_tag(req, input, &err);
+ reqctx->verify = VERIFY_HW;
+}
+ req->base.complete(&req->base, err);
+
+}
static void chcr_verify_tag(struct aead_request *req, u8 *input, int *err)
{
u8 temp[SHA512_DIGEST_SIZE];
@@ -166,27 +239,11 @@ int chcr_handle_resp(struct crypto_async_request *req, unsigned char *input,
{
struct crypto_tfm *tfm = req->tfm;
struct chcr_context *ctx = crypto_tfm_ctx(tfm);
- struct uld_ctx *u_ctx = ULD_CTX(ctx);
- struct chcr_req_ctx ctx_req;
- unsigned int digestsize, updated_digestsize;
struct adapter *adap = padap(ctx->dev);
switch (tfm->__crt_alg->cra_flags & CRYPTO_ALG_TYPE_MASK) {
case CRYPTO_ALG_TYPE_AEAD:
- ctx_req.req.aead_req = aead_request_cast(req);
- ctx_req.ctx.reqctx = aead_request_ctx(ctx_req.req.aead_req);
- dma_unmap_sg(&u_ctx->lldi.pdev->dev, ctx_req.ctx.reqctx->dst,
- ctx_req.ctx.reqctx->dst_nents, DMA_FROM_DEVICE);
- if (ctx_req.ctx.reqctx->skb) {
- kfree_skb(ctx_req.ctx.reqctx->skb);
- ctx_req.ctx.reqctx->skb = NULL;
- }
- if (ctx_req.ctx.reqctx->verify == VERIFY_SW) {
- chcr_verify_tag(ctx_req.req.aead_req, input,
- &err);
- ctx_req.ctx.reqctx->verify = VERIFY_HW;
- }
- ctx_req.req.aead_req->base.complete(req, err);
+ chcr_handle_aead_resp(aead_request_cast(req), input, err);
break;
case CRYPTO_ALG_TYPE_ABLKCIPHER:
@@ -195,60 +252,13 @@ int chcr_handle_resp(struct crypto_async_request *req, unsigned char *input,
break;
case CRYPTO_ALG_TYPE_AHASH:
- ctx_req.req.ahash_req = ahash_request_cast(req);
- ctx_req.ctx.ahash_ctx =
- ahash_request_ctx(ctx_req.req.ahash_req);
- digestsize =
- crypto_ahash_digestsize(crypto_ahash_reqtfm(
- ctx_req.req.ahash_req));
- updated_digestsize = digestsize;
- if (digestsize == SHA224_DIGEST_SIZE)
- updated_digestsize = SHA256_DIGEST_SIZE;
- else if (digestsize == SHA384_DIGEST_SIZE)
- updated_digestsize = SHA512_DIGEST_SIZE;
- if (ctx_req.ctx.ahash_ctx->skb) {
- kfree_skb(ctx_req.ctx.ahash_ctx->skb);
- ctx_req.ctx.ahash_ctx->skb = NULL;
- }
- if (ctx_req.ctx.ahash_ctx->result == 1) {
- ctx_req.ctx.ahash_ctx->result = 0;
- memcpy(ctx_req.req.ahash_req->result, input +
- sizeof(struct cpl_fw6_pld),
- digestsize);
- } else {
- memcpy(ctx_req.ctx.ahash_ctx->partial_hash, input +
- sizeof(struct cpl_fw6_pld),
- updated_digestsize);
+ chcr_handle_ahash_resp(ahash_request_cast(req), input, err);
}
- ctx_req.req.ahash_req->base.complete(req, err);
- break;
- }
atomic_inc(&adap->chcr_stats.complete);
return err;
}
-/*
- * calc_tx_flits_ofld - calculate # of flits for an offload packet
- * @skb: the packet
- * Returns the number of flits needed for the given offload packet.
- * These packets are already fully constructed and no additional headers
- * will be added.
- */
-static inline unsigned int calc_tx_flits_ofld(const struct sk_buff *skb)
-{
- unsigned int flits, cnt;
-
- if (is_ofld_imm(skb))
- return DIV_ROUND_UP(skb->len, 8);
-
- flits = skb_transport_offset(skb) / 8; /* headers */
- cnt = skb_shinfo(skb)->nr_frags;
- if (skb_tail_pointer(skb) != skb_transport_header(skb))
- cnt++;
- return flits + sgl_len(cnt);
-}
-
-static inline void get_aes_decrypt_key(unsigned char *dec_key,
+static void get_aes_decrypt_key(unsigned char *dec_key,
const unsigned char *key,
unsigned int keylength)
{
@@ -395,64 +405,193 @@ static inline int is_hmac(struct crypto_tfm *tfm)
return 0;
}
-static void write_phys_cpl(struct cpl_rx_phys_dsgl *phys_cpl,
- struct scatterlist *sg,
- struct phys_sge_parm *sg_param)
+static inline void dsgl_walk_init(struct dsgl_walk *walk,
+ struct cpl_rx_phys_dsgl *dsgl)
{
- struct phys_sge_pairs *to;
- unsigned int len = 0, left_size = sg_param->obsize;
- unsigned int j = 0;
- int offset, ent_len;
+ walk->dsgl = dsgl;
+ walk->nents = 0;
+ walk->to = (struct phys_sge_pairs *)(dsgl + 1);
+}
+
+static inline void dsgl_walk_end(struct dsgl_walk *walk, unsigned short qid)
+{
+ struct cpl_rx_phys_dsgl *phys_cpl;
+
+ phys_cpl = walk->dsgl;
phys_cpl->op_to_tid = htonl(CPL_RX_PHYS_DSGL_OPCODE_V(CPL_RX_PHYS_DSGL)
| CPL_RX_PHYS_DSGL_ISRDMA_V(0));
- to = (struct phys_sge_pairs *)((unsigned char *)phys_cpl +
- sizeof(struct cpl_rx_phys_dsgl));
+ phys_cpl->pcirlxorder_to_noofsgentr =
+ htonl(CPL_RX_PHYS_DSGL_PCIRLXORDER_V(0) |
+ CPL_RX_PHYS_DSGL_PCINOSNOOP_V(0) |
+ CPL_RX_PHYS_DSGL_PCITPHNTENB_V(0) |
+ CPL_RX_PHYS_DSGL_PCITPHNT_V(0) |
+ CPL_RX_PHYS_DSGL_DCAID_V(0) |
+ CPL_RX_PHYS_DSGL_NOOFSGENTR_V(walk->nents));
+ phys_cpl->rss_hdr_int.opcode = CPL_RX_PHYS_ADDR;
+ phys_cpl->rss_hdr_int.qid = htons(qid);
+ phys_cpl->rss_hdr_int.hash_val = 0;
+}
+
+static inline void dsgl_walk_add_page(struct dsgl_walk *walk,
+ size_t size,
+ dma_addr_t *addr)
+{
+ int j;
+
+ if (!size)
+ return;
+ j = walk->nents;
+ walk->to->len[j % 8] = htons(size);
+ walk->to->addr[j % 8] = cpu_to_be64(*addr);
+ j++;
+ if ((j % 8) == 0)
+ walk->to++;
+ walk->nents = j;
+}
+
+static void dsgl_walk_add_sg(struct dsgl_walk *walk,
+ struct scatterlist *sg,
+ unsigned int slen,
+ unsigned int skip)
+{
+ int skip_len = 0;
+ unsigned int left_size = slen, len = 0;
+ unsigned int j = walk->nents;
+ int offset, ent_len;
+
+ if (!slen)
+ return;
+ while (sg && skip) {
+ if (sg_dma_len(sg) <= skip) {
+ skip -= sg_dma_len(sg);
+ skip_len = 0;
+ sg = sg_next(sg);
+ } else {
+ skip_len = skip;
+ skip = 0;
+ }
+ }
+
while (left_size && sg) {
- len = min_t(u32, left_size, sg_dma_len(sg));
+ len = min_t(u32, left_size, sg_dma_len(sg) - skip_len);
offset = 0;
while (len) {
- ent_len = min_t(u32, len, CHCR_SG_SIZE);
- to->len[j % 8] = htons(ent_len);
- to->addr[j % 8] = cpu_to_be64(sg_dma_address(sg) +
- offset);
+ ent_len = min_t(u32, len, CHCR_DST_SG_SIZE);
+ walk->to->len[j % 8] = htons(ent_len);
+ walk->to->addr[j % 8] = cpu_to_be64(sg_dma_address(sg) +
+ offset + skip_len);
offset += ent_len;
len -= ent_len;
j++;
if ((j % 8) == 0)
- to++;
+ walk->to++;
}
- left_size -= min(left_size, sg_dma_len(sg));
+ walk->last_sg = sg;
+ walk->last_sg_len = min_t(u32, left_size, sg_dma_len(sg) -
+ skip_len) + skip_len;
+ left_size -= min_t(u32, left_size, sg_dma_len(sg) - skip_len);
+ skip_len = 0;
sg = sg_next(sg);
}
- phys_cpl->pcirlxorder_to_noofsgentr =
- htonl(CPL_RX_PHYS_DSGL_PCIRLXORDER_V(0) |
- CPL_RX_PHYS_DSGL_PCINOSNOOP_V(0) |
- CPL_RX_PHYS_DSGL_PCITPHNTENB_V(0) |
- CPL_RX_PHYS_DSGL_PCITPHNT_V(0) |
- CPL_RX_PHYS_DSGL_DCAID_V(0) |
- CPL_RX_PHYS_DSGL_NOOFSGENTR_V(j));
- phys_cpl->rss_hdr_int.opcode = CPL_RX_PHYS_ADDR;
- phys_cpl->rss_hdr_int.qid = htons(sg_param->qid);
- phys_cpl->rss_hdr_int.hash_val = 0;
+ walk->nents = j;
+}
+
+static inline void ulptx_walk_init(struct ulptx_walk *walk,
+ struct ulptx_sgl *ulp)
+{
+ walk->sgl = ulp;
+ walk->nents = 0;
+ walk->pair_idx = 0;
+ walk->pair = ulp->sge;
+ walk->last_sg = NULL;
+ walk->last_sg_len = 0;
+}
+
+static inline void ulptx_walk_end(struct ulptx_walk *walk)
+{
+ walk->sgl->cmd_nsge = htonl(ULPTX_CMD_V(ULP_TX_SC_DSGL) |
+ ULPTX_NSGE_V(walk->nents));
+}
+
+static inline void ulptx_walk_add_page(struct ulptx_walk *walk,
+ size_t size,
+ dma_addr_t *addr)
+{
+ if (!size)
+ return;
+
+ if (walk->nents == 0) {
+ walk->sgl->len0 = cpu_to_be32(size);
+ walk->sgl->addr0 = cpu_to_be64(*addr);
+ } else {
+ walk->pair->addr[walk->pair_idx] = cpu_to_be64(*addr);
+ walk->pair->len[walk->pair_idx] = cpu_to_be32(size);
+ walk->pair_idx = !walk->pair_idx;
+ if (!walk->pair_idx)
+ walk->pair++;
+ }
+ walk->nents++;
}
-static inline int map_writesg_phys_cpl(struct device *dev,
- struct cpl_rx_phys_dsgl *phys_cpl,
+static void ulptx_walk_add_sg(struct ulptx_walk *walk,
struct scatterlist *sg,
- struct phys_sge_parm *sg_param)
+ unsigned int len,
+ unsigned int skip)
{
- if (!sg || !sg_param->nents)
- return -EINVAL;
+ int small;
+ int skip_len = 0;
+ unsigned int sgmin;
- sg_param->nents = dma_map_sg(dev, sg, sg_param->nents, DMA_FROM_DEVICE);
- if (sg_param->nents == 0) {
- pr_err("CHCR : DMA mapping failed\n");
- return -EINVAL;
+ if (!len)
+ return;
+
+ while (sg && skip) {
+ if (sg_dma_len(sg) <= skip) {
+ skip -= sg_dma_len(sg);
+ skip_len = 0;
+ sg = sg_next(sg);
+ } else {
+ skip_len = skip;
+ skip = 0;
+ }
+ }
+ if (walk->nents == 0) {
+ small = min_t(unsigned int, sg_dma_len(sg) - skip_len, len);
+ sgmin = min_t(unsigned int, small, CHCR_SRC_SG_SIZE);
+ walk->sgl->len0 = cpu_to_be32(sgmin);
+ walk->sgl->addr0 = cpu_to_be64(sg_dma_address(sg) + skip_len);
+ walk->nents++;
+ len -= sgmin;
+ walk->last_sg = sg;
+ walk->last_sg_len = sgmin + skip_len;
+ skip_len += sgmin;
+ if (sg_dma_len(sg) == skip_len) {
+ sg = sg_next(sg);
+ skip_len = 0;
+ }
+ }
+
+ while (sg && len) {
+ small = min(sg_dma_len(sg) - skip_len, len);
+ sgmin = min_t(unsigned int, small, CHCR_SRC_SG_SIZE);
+ walk->pair->len[walk->pair_idx] = cpu_to_be32(sgmin);
+ walk->pair->addr[walk->pair_idx] =
+ cpu_to_be64(sg_dma_address(sg) + skip_len);
+ walk->pair_idx = !walk->pair_idx;
+ walk->nents++;
+ if (!walk->pair_idx)
+ walk->pair++;
+ len -= sgmin;
+ skip_len += sgmin;
+ walk->last_sg = sg;
+ walk->last_sg_len = skip_len;
+ if (sg_dma_len(sg) == skip_len) {
+ sg = sg_next(sg);
+ skip_len = 0;
+ }
}
- write_phys_cpl(phys_cpl, sg, sg_param);
- return 0;
}
static inline int get_aead_subtype(struct crypto_aead *aead)
@@ -472,45 +611,6 @@ static inline int get_cryptoalg_subtype(struct crypto_tfm *tfm)
return chcr_crypto_alg->type & CRYPTO_ALG_SUB_TYPE_MASK;
}
-static inline void write_buffer_to_skb(struct sk_buff *skb,
- unsigned int *frags,
- char *bfr,
- u8 bfr_len)
-{
- skb->len += bfr_len;
- skb->data_len += bfr_len;
- skb->truesize += bfr_len;
- get_page(virt_to_page(bfr));
- skb_fill_page_desc(skb, *frags, virt_to_page(bfr),
- offset_in_page(bfr), bfr_len);
- (*frags)++;
-}
-
-
-static inline void
-write_sg_to_skb(struct sk_buff *skb, unsigned int *frags,
- struct scatterlist *sg, unsigned int count)
-{
- struct page *spage;
- unsigned int page_len;
-
- skb->len += count;
- skb->data_len += count;
- skb->truesize += count;
-
- while (count > 0) {
- if (!sg || (!(sg->length)))
- break;
- spage = sg_page(sg);
- get_page(spage);
- page_len = min(sg->length, count);
- skb_fill_page_desc(skb, *frags, spage, sg->offset, page_len);
- (*frags)++;
- count -= page_len;
- sg = sg_next(sg);
- }
-}
-
static int cxgb4_is_crypto_q_full(struct net_device *dev, unsigned int idx)
{
struct adapter *adap = netdev2adap(dev);
@@ -546,32 +646,46 @@ static int generate_copy_rrkey(struct ablk_ctx *ablkctx,
static int chcr_sg_ent_in_wr(struct scatterlist *src,
struct scatterlist *dst,
unsigned int minsg,
- unsigned int space)
+ unsigned int space,
+ unsigned int srcskip,
+ unsigned int dstskip)
{
int srclen = 0, dstlen = 0;
- int srcsg = minsg, dstsg = 0;
+ int srcsg = minsg, dstsg = minsg;
int offset = 0, less;
- while (src && dst && ((srcsg + 1) <= MAX_SKB_FRAGS) &&
+ if (sg_dma_len(src) == srcskip) {
+ src = sg_next(src);
+ srcskip = 0;
+ }
+
+ if (sg_dma_len(dst) == dstskip) {
+ dst = sg_next(dst);
+ dstskip = 0;
+ }
+
+ while (src && dst &&
space > (sgl_ent_len[srcsg + 1] + dsgl_ent_len[dstsg])) {
- srclen += src->length;
+ srclen += (sg_dma_len(src) - srcskip);
srcsg++;
offset = 0;
while (dst && ((dstsg + 1) <= MAX_DSGL_ENT) &&
space > (sgl_ent_len[srcsg] + dsgl_ent_len[dstsg + 1])) {
if (srclen <= dstlen)
break;
- less = min_t(unsigned int, dst->length - offset,
- CHCR_SG_SIZE);
+ less = min_t(unsigned int, sg_dma_len(dst) - offset -
+ dstskip, CHCR_DST_SG_SIZE);
dstlen += less;
offset += less;
- if (offset == dst->length) {
+ if (offset == sg_dma_len(dst)) {
dst = sg_next(dst);
offset = 0;
}
dstsg++;
+ dstskip = 0;
}
src = sg_next(src);
+ srcskip = 0;
}
return min(srclen, dstlen);
}
@@ -601,24 +715,22 @@ static int chcr_cipher_fallback(struct crypto_skcipher *cipher,
}
static inline void create_wreq(struct chcr_context *ctx,
struct chcr_wr *chcr_req,
- void *req, struct sk_buff *skb,
+ struct crypto_async_request *req,
+ unsigned int imm,
int hash_sz,
+ unsigned int len16,
unsigned int sc_len,
unsigned int lcb)
{
struct uld_ctx *u_ctx = ULD_CTX(ctx);
int qid = u_ctx->lldi.rxq_ids[ctx->rx_qidx];
- unsigned int immdatalen = 0;
- if (is_ofld_imm(skb))
- immdatalen = skb->data_len;
chcr_req->wreq.op_to_cctx_size = FILL_WR_OP_CCTX_SIZE;
chcr_req->wreq.pld_size_hash_size =
htonl(FW_CRYPTO_LOOKASIDE_WR_HASH_SIZE_V(hash_sz));
chcr_req->wreq.len16_pkd =
- htonl(FW_CRYPTO_LOOKASIDE_WR_LEN16_V(DIV_ROUND_UP(
- (calc_tx_flits_ofld(skb) * 8), 16)));
+ htonl(FW_CRYPTO_LOOKASIDE_WR_LEN16_V(DIV_ROUND_UP(len16, 16)));
chcr_req->wreq.cookie = cpu_to_be64((uintptr_t)req);
chcr_req->wreq.rx_chid_to_rx_q_id =
FILL_WR_RX_Q_ID(ctx->dev->rx_channel_id, qid,
@@ -626,13 +738,12 @@ static inline void create_wreq(struct chcr_context *ctx,
chcr_req->ulptx.cmd_dest = FILL_ULPTX_CMD_DEST(ctx->dev->tx_channel_id,
qid);
- chcr_req->ulptx.len = htonl((DIV_ROUND_UP((calc_tx_flits_ofld(skb) * 8),
- 16) - ((sizeof(chcr_req->wreq)) >> 4)));
+ chcr_req->ulptx.len = htonl((DIV_ROUND_UP(len16, 16) -
+ ((sizeof(chcr_req->wreq)) >> 4)));
- chcr_req->sc_imm.cmd_more = FILL_CMD_MORE(immdatalen);
+ chcr_req->sc_imm.cmd_more = FILL_CMD_MORE(!imm);
chcr_req->sc_imm.len = cpu_to_be32(sizeof(struct cpl_tx_sec_pdu) +
- sizeof(chcr_req->key_ctx) +
- sc_len + immdatalen);
+ sizeof(chcr_req->key_ctx) + sc_len);
}
/**
@@ -645,49 +756,52 @@ static inline void create_wreq(struct chcr_context *ctx,
static struct sk_buff *create_cipher_wr(struct cipher_wr_param *wrparam)
{
struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(wrparam->req);
- struct chcr_context *ctx = crypto_ablkcipher_ctx(tfm);
- struct uld_ctx *u_ctx = ULD_CTX(ctx);
- struct ablk_ctx *ablkctx = ABLK_CTX(ctx);
+ struct ablk_ctx *ablkctx = ABLK_CTX(c_ctx(tfm));
struct sk_buff *skb = NULL;
struct chcr_wr *chcr_req;
struct cpl_rx_phys_dsgl *phys_cpl;
+ struct ulptx_sgl *ulptx;
struct chcr_blkcipher_req_ctx *reqctx =
ablkcipher_request_ctx(wrparam->req);
- struct phys_sge_parm sg_param;
- unsigned int frags = 0, transhdr_len, phys_dsgl;
+ unsigned int temp = 0, transhdr_len, dst_size;
int error;
int nents;
- unsigned int ivsize = AES_BLOCK_SIZE, kctx_len;
+ unsigned int kctx_len;
gfp_t flags = wrparam->req->base.flags & CRYPTO_TFM_REQ_MAY_SLEEP ?
GFP_KERNEL : GFP_ATOMIC;
- struct adapter *adap = padap(ctx->dev);
+ struct adapter *adap = padap(c_ctx(tfm)->dev);
- reqctx->dst_nents = sg_nents_for_len(reqctx->dst, wrparam->bytes);
- nents = dstsg_2k(reqctx->dst, wrparam->bytes);
- phys_dsgl = get_space_for_phys_dsgl(nents);
+ nents = sg_nents_xlen(reqctx->dstsg, wrparam->bytes, CHCR_DST_SG_SIZE,
+ reqctx->dst_ofst);
+ dst_size = get_space_for_phys_dsgl(nents + 1);
kctx_len = (DIV_ROUND_UP(ablkctx->enckey_len, 16) * 16);
- transhdr_len = CIPHER_TRANSHDR_SIZE(kctx_len, phys_dsgl);
- skb = alloc_skb((transhdr_len + sizeof(struct sge_opaque_hdr)), flags);
+ transhdr_len = CIPHER_TRANSHDR_SIZE(kctx_len, dst_size);
+ nents = sg_nents_xlen(reqctx->srcsg, wrparam->bytes,
+ CHCR_SRC_SG_SIZE, reqctx->src_ofst);
+ temp = reqctx->imm ? (DIV_ROUND_UP((IV + wrparam->req->nbytes), 16)
+ * 16) : (sgl_len(nents + MIN_CIPHER_SG) * 8);
+ transhdr_len += temp;
+ transhdr_len = DIV_ROUND_UP(transhdr_len, 16) * 16;
+ skb = alloc_skb(SGE_MAX_WR_LEN, flags);
if (!skb) {
error = -ENOMEM;
goto err;
}
- skb_reserve(skb, sizeof(struct sge_opaque_hdr));
chcr_req = __skb_put_zero(skb, transhdr_len);
chcr_req->sec_cpl.op_ivinsrtofst =
- FILL_SEC_CPL_OP_IVINSR(ctx->dev->rx_channel_id, 2, 1);
+ FILL_SEC_CPL_OP_IVINSR(c_ctx(tfm)->dev->rx_channel_id, 2, 1);
- chcr_req->sec_cpl.pldlen = htonl(ivsize + wrparam->bytes);
+ chcr_req->sec_cpl.pldlen = htonl(IV + wrparam->bytes);
chcr_req->sec_cpl.aadstart_cipherstop_hi =
- FILL_SEC_CPL_CIPHERSTOP_HI(0, 0, ivsize + 1, 0);
+ FILL_SEC_CPL_CIPHERSTOP_HI(0, 0, IV + 1, 0);
chcr_req->sec_cpl.cipherstop_lo_authinsert =
FILL_SEC_CPL_AUTHINSERT(0, 0, 0, 0);
chcr_req->sec_cpl.seqno_numivs = FILL_SEC_CPL_SCMD0_SEQNO(reqctx->op, 0,
ablkctx->ciph_mode,
- 0, 0, ivsize >> 1);
+ 0, 0, IV >> 1);
chcr_req->sec_cpl.ivgen_hdrlen = FILL_SEC_CPL_IVGEN_HDRLEN(0, 0, 0,
- 0, 1, phys_dsgl);
+ 0, 0, dst_size);
chcr_req->key_ctx.ctx_hdr = ablkctx->key_ctx_hdr;
if ((reqctx->op == CHCR_DECRYPT_OP) &&
@@ -712,26 +826,18 @@ static struct sk_buff *create_cipher_wr(struct cipher_wr_param *wrparam)
}
}
phys_cpl = (struct cpl_rx_phys_dsgl *)((u8 *)(chcr_req + 1) + kctx_len);
- sg_param.nents = reqctx->dst_nents;
- sg_param.obsize = wrparam->bytes;
- sg_param.qid = wrparam->qid;
- error = map_writesg_phys_cpl(&u_ctx->lldi.pdev->dev, phys_cpl,
- reqctx->dst, &sg_param);
- if (error)
- goto map_fail1;
+ ulptx = (struct ulptx_sgl *)((u8 *)(phys_cpl + 1) + dst_size);
+ chcr_add_cipher_src_ent(wrparam->req, ulptx, wrparam);
+ chcr_add_cipher_dst_ent(wrparam->req, phys_cpl, wrparam, wrparam->qid);
- skb_set_transport_header(skb, transhdr_len);
- write_buffer_to_skb(skb, &frags, reqctx->iv, ivsize);
- write_sg_to_skb(skb, &frags, wrparam->srcsg, wrparam->bytes);
atomic_inc(&adap->chcr_stats.cipher_rqst);
- create_wreq(ctx, chcr_req, &(wrparam->req->base), skb, 0,
- sizeof(struct cpl_rx_phys_dsgl) + phys_dsgl + kctx_len,
+ temp = sizeof(struct cpl_rx_phys_dsgl) + dst_size + kctx_len
+ +(reqctx->imm ? (IV + wrparam->bytes) : 0);
+ create_wreq(c_ctx(tfm), chcr_req, &(wrparam->req->base), reqctx->imm, 0,
+ transhdr_len, temp,
ablkctx->ciph_mode == CHCR_SCMD_CIPHER_MODE_AES_CBC);
reqctx->skb = skb;
- skb_get(skb);
return skb;
-map_fail1:
- kfree_skb(skb);
err:
return ERR_PTR(error);
}
@@ -756,8 +862,7 @@ static int chcr_cipher_fallback_setkey(struct crypto_ablkcipher *cipher,
unsigned int keylen)
{
struct crypto_tfm *tfm = crypto_ablkcipher_tfm(cipher);
- struct chcr_context *ctx = crypto_ablkcipher_ctx(cipher);
- struct ablk_ctx *ablkctx = ABLK_CTX(ctx);
+ struct ablk_ctx *ablkctx = ABLK_CTX(c_ctx(cipher));
int err = 0;
crypto_skcipher_clear_flags(ablkctx->sw_cipher, CRYPTO_TFM_REQ_MASK);
@@ -775,8 +880,7 @@ static int chcr_aes_cbc_setkey(struct crypto_ablkcipher *cipher,
const u8 *key,
unsigned int keylen)
{
- struct chcr_context *ctx = crypto_ablkcipher_ctx(cipher);
- struct ablk_ctx *ablkctx = ABLK_CTX(ctx);
+ struct ablk_ctx *ablkctx = ABLK_CTX(c_ctx(cipher));
unsigned int ck_size, context_size;
u16 alignment = 0;
int err;
@@ -808,8 +912,7 @@ static int chcr_aes_ctr_setkey(struct crypto_ablkcipher *cipher,
const u8 *key,
unsigned int keylen)
{
- struct chcr_context *ctx = crypto_ablkcipher_ctx(cipher);
- struct ablk_ctx *ablkctx = ABLK_CTX(ctx);
+ struct ablk_ctx *ablkctx = ABLK_CTX(c_ctx(cipher));
unsigned int ck_size, context_size;
u16 alignment = 0;
int err;
@@ -840,8 +943,7 @@ static int chcr_aes_rfc3686_setkey(struct crypto_ablkcipher *cipher,
const u8 *key,
unsigned int keylen)
{
- struct chcr_context *ctx = crypto_ablkcipher_ctx(cipher);
- struct ablk_ctx *ablkctx = ABLK_CTX(ctx);
+ struct ablk_ctx *ablkctx = ABLK_CTX(c_ctx(cipher));
unsigned int ck_size, context_size;
u16 alignment = 0;
int err;
@@ -908,8 +1010,7 @@ static unsigned int adjust_ctr_overflow(u8 *iv, u32 bytes)
static int chcr_update_tweak(struct ablkcipher_request *req, u8 *iv)
{
struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req);
- struct chcr_context *ctx = crypto_ablkcipher_ctx(tfm);
- struct ablk_ctx *ablkctx = ABLK_CTX(ctx);
+ struct ablk_ctx *ablkctx = ABLK_CTX(c_ctx(tfm));
struct chcr_blkcipher_req_ctx *reqctx = ablkcipher_request_ctx(req);
struct crypto_cipher *cipher;
int ret, i;
@@ -926,8 +1027,7 @@ static int chcr_update_tweak(struct ablkcipher_request *req, u8 *iv)
ret = crypto_cipher_setkey(cipher, key, keylen);
if (ret)
goto out;
-
- crypto_cipher_encrypt_one(cipher, iv, iv);
+ /*H/W sends the encrypted IV in dsgl when AADIVDROP bit is 0*/
for (i = 0; i < round8; i++)
gf128mul_x8_ble((le128 *)iv, (le128 *)iv);
@@ -1005,64 +1105,60 @@ static int chcr_handle_cipher_resp(struct ablkcipher_request *req,
unsigned char *input, int err)
{
struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req);
- struct chcr_context *ctx = crypto_ablkcipher_ctx(tfm);
- struct uld_ctx *u_ctx = ULD_CTX(ctx);
- struct ablk_ctx *ablkctx = ABLK_CTX(ctx);
+ struct uld_ctx *u_ctx = ULD_CTX(c_ctx(tfm));
+ struct ablk_ctx *ablkctx = ABLK_CTX(c_ctx(tfm));
struct sk_buff *skb;
struct cpl_fw6_pld *fw6_pld = (struct cpl_fw6_pld *)input;
struct chcr_blkcipher_req_ctx *reqctx = ablkcipher_request_ctx(req);
struct cipher_wr_param wrparam;
int bytes;
- dma_unmap_sg(&u_ctx->lldi.pdev->dev, reqctx->dst, reqctx->dst_nents,
- DMA_FROM_DEVICE);
-
- if (reqctx->skb) {
- kfree_skb(reqctx->skb);
- reqctx->skb = NULL;
- }
if (err)
- goto complete;
-
+ goto unmap;
if (req->nbytes == reqctx->processed) {
+ chcr_cipher_dma_unmap(&ULD_CTX(c_ctx(tfm))->lldi.pdev->dev,
+ req);
err = chcr_final_cipher_iv(req, fw6_pld, req->info);
goto complete;
}
if (unlikely(cxgb4_is_crypto_q_full(u_ctx->lldi.ports[0],
- ctx->tx_qidx))) {
+ c_ctx(tfm)->tx_qidx))) {
if (!(req->base.flags & CRYPTO_TFM_REQ_MAY_BACKLOG)) {
err = -EBUSY;
- goto complete;
+ goto unmap;
}
}
- wrparam.srcsg = scatterwalk_ffwd(reqctx->srcffwd, req->src,
- reqctx->processed);
- reqctx->dst = scatterwalk_ffwd(reqctx->dstffwd, reqctx->dstsg,
- reqctx->processed);
- if (!wrparam.srcsg || !reqctx->dst) {
- pr_err("Input sg list length less that nbytes\n");
- err = -EINVAL;
- goto complete;
- }
- bytes = chcr_sg_ent_in_wr(wrparam.srcsg, reqctx->dst, 1,
- SPACE_LEFT(ablkctx->enckey_len));
+ if (!reqctx->imm) {
+ bytes = chcr_sg_ent_in_wr(reqctx->srcsg, reqctx->dstsg, 1,
+ SPACE_LEFT(ablkctx->enckey_len),
+ reqctx->src_ofst, reqctx->dst_ofst);
if ((bytes + reqctx->processed) >= req->nbytes)
bytes = req->nbytes - reqctx->processed;
else
bytes = ROUND_16(bytes);
+ } else {
+ /*CTR mode counter overfloa*/
+ bytes = req->nbytes - reqctx->processed;
+ }
+ dma_sync_single_for_cpu(&ULD_CTX(c_ctx(tfm))->lldi.pdev->dev,
+ reqctx->iv_dma, IV, DMA_BIDIRECTIONAL);
err = chcr_update_cipher_iv(req, fw6_pld, reqctx->iv);
+ dma_sync_single_for_device(&ULD_CTX(c_ctx(tfm))->lldi.pdev->dev,
+ reqctx->iv_dma, IV, DMA_BIDIRECTIONAL);
if (err)
- goto complete;
+ goto unmap;
if (unlikely(bytes == 0)) {
+ chcr_cipher_dma_unmap(&ULD_CTX(c_ctx(tfm))->lldi.pdev->dev,
+ req);
err = chcr_cipher_fallback(ablkctx->sw_cipher,
req->base.flags,
- wrparam.srcsg,
- reqctx->dst,
- req->nbytes - reqctx->processed,
- reqctx->iv,
+ req->src,
+ req->dst,
+ req->nbytes,
+ req->info,
reqctx->op);
goto complete;
}
@@ -1070,21 +1166,23 @@ static int chcr_handle_cipher_resp(struct ablkcipher_request *req,
if (get_cryptoalg_subtype(crypto_ablkcipher_tfm(tfm)) ==
CRYPTO_ALG_SUB_TYPE_CTR)
bytes = adjust_ctr_overflow(reqctx->iv, bytes);
- reqctx->processed += bytes;
- reqctx->last_req_len = bytes;
- wrparam.qid = u_ctx->lldi.rxq_ids[ctx->rx_qidx];
+ wrparam.qid = u_ctx->lldi.rxq_ids[c_ctx(tfm)->rx_qidx];
wrparam.req = req;
wrparam.bytes = bytes;
skb = create_cipher_wr(&wrparam);
if (IS_ERR(skb)) {
pr_err("chcr : %s : Failed to form WR. No memory\n", __func__);
err = PTR_ERR(skb);
- goto complete;
+ goto unmap;
}
skb->dev = u_ctx->lldi.ports[0];
- set_wr_txq(skb, CPL_PRIORITY_DATA, ctx->tx_qidx);
+ set_wr_txq(skb, CPL_PRIORITY_DATA, c_ctx(tfm)->tx_qidx);
chcr_send_wr(skb);
+ reqctx->last_req_len = bytes;
+ reqctx->processed += bytes;
return 0;
+unmap:
+ chcr_cipher_dma_unmap(&ULD_CTX(c_ctx(tfm))->lldi.pdev->dev, req);
complete:
req->base.complete(&req->base, err);
return err;
@@ -1098,8 +1196,7 @@ static int process_cipher(struct ablkcipher_request *req,
struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req);
unsigned int ivsize = crypto_ablkcipher_ivsize(tfm);
struct chcr_blkcipher_req_ctx *reqctx = ablkcipher_request_ctx(req);
- struct chcr_context *ctx = crypto_ablkcipher_ctx(tfm);
- struct ablk_ctx *ablkctx = ABLK_CTX(ctx);
+ struct ablk_ctx *ablkctx = ABLK_CTX(c_ctx(tfm));
struct cipher_wr_param wrparam;
int bytes, err = -EINVAL;
@@ -1113,16 +1210,41 @@ static int process_cipher(struct ablkcipher_request *req,
ablkctx->enckey_len, req->nbytes, ivsize);
goto error;
}
- wrparam.srcsg = req->src;
- reqctx->dstsg = req->dst;
- bytes = chcr_sg_ent_in_wr(wrparam.srcsg, reqctx->dstsg, MIN_CIPHER_SG,
- SPACE_LEFT(ablkctx->enckey_len));
+ chcr_cipher_dma_map(&ULD_CTX(c_ctx(tfm))->lldi.pdev->dev, req);
+ if (req->nbytes < (SGE_MAX_WR_LEN - (sizeof(struct chcr_wr) +
+ AES_MIN_KEY_SIZE +
+ sizeof(struct cpl_rx_phys_dsgl) +
+ /*Min dsgl size*/
+ 32))) {
+ /* Can be sent as Imm*/
+ unsigned int dnents = 0, transhdr_len, phys_dsgl, kctx_len;
+
+ dnents = sg_nents_xlen(req->dst, req->nbytes,
+ CHCR_DST_SG_SIZE, 0);
+ dnents += 1; // IV
+ phys_dsgl = get_space_for_phys_dsgl(dnents);
+ kctx_len = (DIV_ROUND_UP(ablkctx->enckey_len, 16) * 16);
+ transhdr_len = CIPHER_TRANSHDR_SIZE(kctx_len, phys_dsgl);
+ reqctx->imm = (transhdr_len + IV + req->nbytes) <=
+ SGE_MAX_WR_LEN;
+ bytes = IV + req->nbytes;
+
+ } else {
+ reqctx->imm = 0;
+ }
+
+ if (!reqctx->imm) {
+ bytes = chcr_sg_ent_in_wr(req->src, req->dst,
+ MIN_CIPHER_SG,
+ SPACE_LEFT(ablkctx->enckey_len),
+ 0, 0);
if ((bytes + reqctx->processed) >= req->nbytes)
bytes = req->nbytes - reqctx->processed;
else
bytes = ROUND_16(bytes);
- if (unlikely(bytes > req->nbytes))
+ } else {
bytes = req->nbytes;
+ }
if (get_cryptoalg_subtype(crypto_ablkcipher_tfm(tfm)) ==
CRYPTO_ALG_SUB_TYPE_CTR) {
bytes = adjust_ctr_overflow(req->info, bytes);
@@ -1139,9 +1261,11 @@ static int process_cipher(struct ablkcipher_request *req,
} else {
- memcpy(reqctx->iv, req->info, ivsize);
+ memcpy(reqctx->iv, req->info, IV);
}
if (unlikely(bytes == 0)) {
+ chcr_cipher_dma_unmap(&ULD_CTX(c_ctx(tfm))->lldi.pdev->dev,
+ req);
err = chcr_cipher_fallback(ablkctx->sw_cipher,
req->base.flags,
req->src,
@@ -1151,20 +1275,25 @@ static int process_cipher(struct ablkcipher_request *req,
op_type);
goto error;
}
- reqctx->processed = bytes;
- reqctx->last_req_len = bytes;
- reqctx->dst = reqctx->dstsg;
reqctx->op = op_type;
+ reqctx->srcsg = req->src;
+ reqctx->dstsg = req->dst;
+ reqctx->src_ofst = 0;
+ reqctx->dst_ofst = 0;
wrparam.qid = qid;
wrparam.req = req;
wrparam.bytes = bytes;
*skb = create_cipher_wr(&wrparam);
if (IS_ERR(*skb)) {
err = PTR_ERR(*skb);
- goto error;
+ goto unmap;
}
+ reqctx->processed = bytes;
+ reqctx->last_req_len = bytes;
return 0;
+unmap:
+ chcr_cipher_dma_unmap(&ULD_CTX(c_ctx(tfm))->lldi.pdev->dev, req);
error:
return err;
}
@@ -1172,23 +1301,22 @@ static int process_cipher(struct ablkcipher_request *req,
static int chcr_aes_encrypt(struct ablkcipher_request *req)
{
struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req);
- struct chcr_context *ctx = crypto_ablkcipher_ctx(tfm);
struct sk_buff *skb = NULL;
int err;
- struct uld_ctx *u_ctx = ULD_CTX(ctx);
+ struct uld_ctx *u_ctx = ULD_CTX(c_ctx(tfm));
if (unlikely(cxgb4_is_crypto_q_full(u_ctx->lldi.ports[0],
- ctx->tx_qidx))) {
+ c_ctx(tfm)->tx_qidx))) {
if (!(req->base.flags & CRYPTO_TFM_REQ_MAY_BACKLOG))
return -EBUSY;
}
- err = process_cipher(req, u_ctx->lldi.rxq_ids[ctx->rx_qidx], &skb,
- CHCR_ENCRYPT_OP);
+ err = process_cipher(req, u_ctx->lldi.rxq_ids[c_ctx(tfm)->rx_qidx],
+ &skb, CHCR_ENCRYPT_OP);
if (err || !skb)
return err;
skb->dev = u_ctx->lldi.ports[0];
- set_wr_txq(skb, CPL_PRIORITY_DATA, ctx->tx_qidx);
+ set_wr_txq(skb, CPL_PRIORITY_DATA, c_ctx(tfm)->tx_qidx);
chcr_send_wr(skb);
return -EINPROGRESS;
}
@@ -1196,23 +1324,22 @@ static int chcr_aes_encrypt(struct ablkcipher_request *req)
static int chcr_aes_decrypt(struct ablkcipher_request *req)
{
struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req);
- struct chcr_context *ctx = crypto_ablkcipher_ctx(tfm);
- struct uld_ctx *u_ctx = ULD_CTX(ctx);
+ struct uld_ctx *u_ctx = ULD_CTX(c_ctx(tfm));
struct sk_buff *skb = NULL;
int err;
if (unlikely(cxgb4_is_crypto_q_full(u_ctx->lldi.ports[0],
- ctx->tx_qidx))) {
+ c_ctx(tfm)->tx_qidx))) {
if (!(req->base.flags & CRYPTO_TFM_REQ_MAY_BACKLOG))
return -EBUSY;
}
- err = process_cipher(req, u_ctx->lldi.rxq_ids[ctx->rx_qidx], &skb,
- CHCR_DECRYPT_OP);
+ err = process_cipher(req, u_ctx->lldi.rxq_ids[c_ctx(tfm)->rx_qidx],
+ &skb, CHCR_DECRYPT_OP);
if (err || !skb)
return err;
skb->dev = u_ctx->lldi.ports[0];
- set_wr_txq(skb, CPL_PRIORITY_DATA, ctx->tx_qidx);
+ set_wr_txq(skb, CPL_PRIORITY_DATA, c_ctx(tfm)->tx_qidx);
chcr_send_wr(skb);
return -EINPROGRESS;
}
@@ -1360,17 +1487,19 @@ static struct sk_buff *create_hash_wr(struct ahash_request *req,
{
struct chcr_ahash_req_ctx *req_ctx = ahash_request_ctx(req);
struct crypto_ahash *tfm = crypto_ahash_reqtfm(req);
- struct chcr_context *ctx = crypto_tfm_ctx(crypto_ahash_tfm(tfm));
- struct hmac_ctx *hmacctx = HMAC_CTX(ctx);
+ struct hmac_ctx *hmacctx = HMAC_CTX(h_ctx(tfm));
struct sk_buff *skb = NULL;
+ struct uld_ctx *u_ctx = ULD_CTX(h_ctx(tfm));
struct chcr_wr *chcr_req;
- unsigned int frags = 0, transhdr_len, iopad_alignment = 0;
+ struct ulptx_sgl *ulptx;
+ unsigned int nents = 0, transhdr_len, iopad_alignment = 0;
unsigned int digestsize = crypto_ahash_digestsize(tfm);
- unsigned int kctx_len = 0;
+ unsigned int kctx_len = 0, temp = 0;
u8 hash_size_in_response = 0;
gfp_t flags = req->base.flags & CRYPTO_TFM_REQ_MAY_SLEEP ? GFP_KERNEL :
GFP_ATOMIC;
- struct adapter *adap = padap(ctx->dev);
+ struct adapter *adap = padap(h_ctx(tfm)->dev);
+ int error = 0;
iopad_alignment = KEYCTX_ALIGN_PAD(digestsize);
kctx_len = param->alg_prm.result_size + iopad_alignment;
@@ -1382,15 +1511,22 @@ static struct sk_buff *create_hash_wr(struct ahash_request *req,
else
hash_size_in_response = param->alg_prm.result_size;
transhdr_len = HASH_TRANSHDR_SIZE(kctx_len);
- skb = alloc_skb((transhdr_len + sizeof(struct sge_opaque_hdr)), flags);
+ req_ctx->imm = (transhdr_len + param->bfr_len + param->sg_len) <=
+ SGE_MAX_WR_LEN;
+ nents = sg_nents_xlen(req->src, param->sg_len, CHCR_SRC_SG_SIZE, 0);
+ nents += param->bfr_len ? 1 : 0;
+ transhdr_len += req_ctx->imm ? (DIV_ROUND_UP((param->bfr_len +
+ param->sg_len), 16) * 16) :
+ (sgl_len(nents) * 8);
+ transhdr_len = DIV_ROUND_UP(transhdr_len, 16) * 16;
+
+ skb = alloc_skb(SGE_MAX_WR_LEN, flags);
if (!skb)
- return skb;
-
- skb_reserve(skb, sizeof(struct sge_opaque_hdr));
+ return ERR_PTR(-ENOMEM);
chcr_req = __skb_put_zero(skb, transhdr_len);
chcr_req->sec_cpl.op_ivinsrtofst =
- FILL_SEC_CPL_OP_IVINSR(ctx->dev->rx_channel_id, 2, 0);
+ FILL_SEC_CPL_OP_IVINSR(h_ctx(tfm)->dev->rx_channel_id, 2, 0);
chcr_req->sec_cpl.pldlen = htonl(param->bfr_len + param->sg_len);
chcr_req->sec_cpl.aadstart_cipherstop_hi =
@@ -1419,37 +1555,52 @@ static struct sk_buff *create_hash_wr(struct ahash_request *req,
((kctx_len +
sizeof(chcr_req->key_ctx)) >> 4));
chcr_req->sec_cpl.scmd1 = cpu_to_be64((u64)param->scmd1);
-
- skb_set_transport_header(skb, transhdr_len);
- if (param->bfr_len != 0)
- write_buffer_to_skb(skb, &frags, req_ctx->reqbfr,
- param->bfr_len);
- if (param->sg_len != 0)
- write_sg_to_skb(skb, &frags, req->src, param->sg_len);
+ ulptx = (struct ulptx_sgl *)((u8 *)(chcr_req + 1) + kctx_len +
+ DUMMY_BYTES);
+ if (param->bfr_len != 0) {
+ req_ctx->dma_addr = dma_map_single(&u_ctx->lldi.pdev->dev,
+ req_ctx->reqbfr, param->bfr_len,
+ DMA_TO_DEVICE);
+ if (dma_mapping_error(&u_ctx->lldi.pdev->dev,
+ req_ctx->dma_addr)) {
+ error = -ENOMEM;
+ goto err;
+ }
+ req_ctx->dma_len = param->bfr_len;
+ } else {
+ req_ctx->dma_addr = 0;
+ }
+ chcr_add_hash_src_ent(req, ulptx, param);
+ /* Request upto max wr size */
+ temp = kctx_len + DUMMY_BYTES + (req_ctx->imm ? (param->sg_len
+ + param->bfr_len) : 0);
atomic_inc(&adap->chcr_stats.digest_rqst);
- create_wreq(ctx, chcr_req, &req->base, skb, hash_size_in_response,
- DUMMY_BYTES + kctx_len, 0);
+ create_wreq(h_ctx(tfm), chcr_req, &req->base, req_ctx->imm,
+ hash_size_in_response, transhdr_len,
+ temp, 0);
req_ctx->skb = skb;
- skb_get(skb);
return skb;
+err:
+ kfree_skb(skb);
+ return ERR_PTR(error);
}
static int chcr_ahash_update(struct ahash_request *req)
{
struct chcr_ahash_req_ctx *req_ctx = ahash_request_ctx(req);
struct crypto_ahash *rtfm = crypto_ahash_reqtfm(req);
- struct chcr_context *ctx = crypto_tfm_ctx(crypto_ahash_tfm(rtfm));
struct uld_ctx *u_ctx = NULL;
struct sk_buff *skb;
u8 remainder = 0, bs;
unsigned int nbytes = req->nbytes;
struct hash_wr_param params;
+ int error;
bs = crypto_tfm_alg_blocksize(crypto_ahash_tfm(rtfm));
- u_ctx = ULD_CTX(ctx);
+ u_ctx = ULD_CTX(h_ctx(rtfm));
if (unlikely(cxgb4_is_crypto_q_full(u_ctx->lldi.ports[0],
- ctx->tx_qidx))) {
+ h_ctx(rtfm)->tx_qidx))) {
if (!(req->base.flags & CRYPTO_TFM_REQ_MAY_BACKLOG))
return -EBUSY;
}
@@ -1463,7 +1614,9 @@ static int chcr_ahash_update(struct ahash_request *req)
req_ctx->reqlen += nbytes;
return 0;
}
-
+ error = chcr_hash_dma_map(&u_ctx->lldi.pdev->dev, req);
+ if (error)
+ return -ENOMEM;
params.opad_needed = 0;
params.more = 1;
params.last = 0;
@@ -1474,8 +1627,10 @@ static int chcr_ahash_update(struct ahash_request *req)
req_ctx->result = 0;
req_ctx->data_len += params.sg_len + params.bfr_len;
skb = create_hash_wr(req, ¶ms);
- if (IS_ERR(skb))
- return PTR_ERR(skb);
+ if (IS_ERR(skb)) {
+ error = PTR_ERR(skb);
+ goto unmap;
+ }
if (remainder) {
u8 *temp;
@@ -1489,10 +1644,13 @@ static int chcr_ahash_update(struct ahash_request *req)
}
req_ctx->reqlen = remainder;
skb->dev = u_ctx->lldi.ports[0];
- set_wr_txq(skb, CPL_PRIORITY_DATA, ctx->tx_qidx);
+ set_wr_txq(skb, CPL_PRIORITY_DATA, h_ctx(rtfm)->tx_qidx);
chcr_send_wr(skb);
return -EINPROGRESS;
+unmap:
+ chcr_hash_dma_unmap(&u_ctx->lldi.pdev->dev, req);
+ return error;
}
static void create_last_hash_block(char *bfr_ptr, unsigned int bs, u64 scmd1)
@@ -1509,13 +1667,12 @@ static int chcr_ahash_final(struct ahash_request *req)
{
struct chcr_ahash_req_ctx *req_ctx = ahash_request_ctx(req);
struct crypto_ahash *rtfm = crypto_ahash_reqtfm(req);
- struct chcr_context *ctx = crypto_tfm_ctx(crypto_ahash_tfm(rtfm));
struct hash_wr_param params;
struct sk_buff *skb;
struct uld_ctx *u_ctx = NULL;
u8 bs = crypto_tfm_alg_blocksize(crypto_ahash_tfm(rtfm));
- u_ctx = ULD_CTX(ctx);
+ u_ctx = ULD_CTX(h_ctx(rtfm));
if (is_hmac(crypto_ahash_tfm(rtfm)))
params.opad_needed = 1;
else
@@ -1542,7 +1699,7 @@ static int chcr_ahash_final(struct ahash_request *req)
return PTR_ERR(skb);
skb->dev = u_ctx->lldi.ports[0];
- set_wr_txq(skb, CPL_PRIORITY_DATA, ctx->tx_qidx);
+ set_wr_txq(skb, CPL_PRIORITY_DATA, h_ctx(rtfm)->tx_qidx);
chcr_send_wr(skb);
return -EINPROGRESS;
}
@@ -1551,17 +1708,17 @@ static int chcr_ahash_finup(struct ahash_request *req)
{
struct chcr_ahash_req_ctx *req_ctx = ahash_request_ctx(req);
struct crypto_ahash *rtfm = crypto_ahash_reqtfm(req);
- struct chcr_context *ctx = crypto_tfm_ctx(crypto_ahash_tfm(rtfm));
struct uld_ctx *u_ctx = NULL;
struct sk_buff *skb;
struct hash_wr_param params;
u8 bs;
+ int error;
bs = crypto_tfm_alg_blocksize(crypto_ahash_tfm(rtfm));
- u_ctx = ULD_CTX(ctx);
+ u_ctx = ULD_CTX(h_ctx(rtfm));
if (unlikely(cxgb4_is_crypto_q_full(u_ctx->lldi.ports[0],
- ctx->tx_qidx))) {
+ h_ctx(rtfm)->tx_qidx))) {
if (!(req->base.flags & CRYPTO_TFM_REQ_MAY_BACKLOG))
return -EBUSY;
}
@@ -1587,34 +1744,41 @@ static int chcr_ahash_finup(struct ahash_request *req)
params.last = 1;
params.more = 0;
}
+ error = chcr_hash_dma_map(&u_ctx->lldi.pdev->dev, req);
+ if (error)
+ return -ENOMEM;
skb = create_hash_wr(req, ¶ms);
- if (IS_ERR(skb))
- return PTR_ERR(skb);
-
+ if (IS_ERR(skb)) {
+ error = PTR_ERR(skb);
+ goto unmap;
+ }
skb->dev = u_ctx->lldi.ports[0];
- set_wr_txq(skb, CPL_PRIORITY_DATA, ctx->tx_qidx);
+ set_wr_txq(skb, CPL_PRIORITY_DATA, h_ctx(rtfm)->tx_qidx);
chcr_send_wr(skb);
return -EINPROGRESS;
+unmap:
+ chcr_hash_dma_unmap(&u_ctx->lldi.pdev->dev, req);
+ return error;
}
static int chcr_ahash_digest(struct ahash_request *req)
{
struct chcr_ahash_req_ctx *req_ctx = ahash_request_ctx(req);
struct crypto_ahash *rtfm = crypto_ahash_reqtfm(req);
- struct chcr_context *ctx = crypto_tfm_ctx(crypto_ahash_tfm(rtfm));
struct uld_ctx *u_ctx = NULL;
struct sk_buff *skb;
struct hash_wr_param params;
u8 bs;
+ int error;
rtfm->init(req);
bs = crypto_tfm_alg_blocksize(crypto_ahash_tfm(rtfm));
- u_ctx = ULD_CTX(ctx);
+ u_ctx = ULD_CTX(h_ctx(rtfm));
if (unlikely(cxgb4_is_crypto_q_full(u_ctx->lldi.ports[0],
- ctx->tx_qidx))) {
+ h_ctx(rtfm)->tx_qidx))) {
if (!(req->base.flags & CRYPTO_TFM_REQ_MAY_BACKLOG))
return -EBUSY;
}
@@ -1623,6 +1787,9 @@ static int chcr_ahash_digest(struct ahash_request *req)
params.opad_needed = 1;
else
params.opad_needed = 0;
+ error = chcr_hash_dma_map(&u_ctx->lldi.pdev->dev, req);
+ if (error)
+ return -ENOMEM;
params.last = 0;
params.more = 0;
@@ -1640,13 +1807,17 @@ static int chcr_ahash_digest(struct ahash_request *req)
}
skb = create_hash_wr(req, ¶ms);
- if (IS_ERR(skb))
- return PTR_ERR(skb);
-
+ if (IS_ERR(skb)) {
+ error = PTR_ERR(skb);
+ goto unmap;
+ }
skb->dev = u_ctx->lldi.ports[0];
- set_wr_txq(skb, CPL_PRIORITY_DATA, ctx->tx_qidx);
+ set_wr_txq(skb, CPL_PRIORITY_DATA, h_ctx(rtfm)->tx_qidx);
chcr_send_wr(skb);
return -EINPROGRESS;
+unmap:
+ chcr_hash_dma_unmap(&u_ctx->lldi.pdev->dev, req);
+ return error;
}
static int chcr_ahash_export(struct ahash_request *areq, void *out)
@@ -1656,6 +1827,8 @@ static int chcr_ahash_export(struct ahash_request *areq, void *out)
state->reqlen = req_ctx->reqlen;
state->data_len = req_ctx->data_len;
+ state->is_sg_map = 0;
+ state->result = 0;
memcpy(state->bfr1, req_ctx->reqbfr, req_ctx->reqlen);
memcpy(state->partial_hash, req_ctx->partial_hash,
CHCR_HASH_MAX_DIGEST_SIZE);
@@ -1671,6 +1844,8 @@ static int chcr_ahash_import(struct ahash_request *areq, const void *in)
req_ctx->data_len = state->data_len;
req_ctx->reqbfr = req_ctx->bfr1;
req_ctx->skbfr = req_ctx->bfr2;
+ req_ctx->is_sg_map = 0;
+ req_ctx->result = 0;
memcpy(req_ctx->bfr1, state->bfr1, CHCR_HASH_MAX_BLOCK_SIZE_128);
memcpy(req_ctx->partial_hash, state->partial_hash,
CHCR_HASH_MAX_DIGEST_SIZE);
@@ -1680,8 +1855,7 @@ static int chcr_ahash_import(struct ahash_request *areq, const void *in)
static int chcr_ahash_setkey(struct crypto_ahash *tfm, const u8 *key,
unsigned int keylen)
{
- struct chcr_context *ctx = crypto_tfm_ctx(crypto_ahash_tfm(tfm));
- struct hmac_ctx *hmacctx = HMAC_CTX(ctx);
+ struct hmac_ctx *hmacctx = HMAC_CTX(h_ctx(tfm));
unsigned int digestsize = crypto_ahash_digestsize(tfm);
unsigned int bs = crypto_tfm_alg_blocksize(crypto_ahash_tfm(tfm));
unsigned int i, err = 0, updated_digestsize;
@@ -1734,8 +1908,7 @@ static int chcr_ahash_setkey(struct crypto_ahash *tfm, const u8 *key,
static int chcr_aes_xts_setkey(struct crypto_ablkcipher *cipher, const u8 *key,
unsigned int key_len)
{
- struct chcr_context *ctx = crypto_ablkcipher_ctx(cipher);
- struct ablk_ctx *ablkctx = ABLK_CTX(ctx);
+ struct ablk_ctx *ablkctx = ABLK_CTX(c_ctx(cipher));
unsigned short context_size = 0;
int err;
@@ -1774,6 +1947,7 @@ static int chcr_sha_init(struct ahash_request *areq)
req_ctx->skbfr = req_ctx->bfr2;
req_ctx->skb = NULL;
req_ctx->result = 0;
+ req_ctx->is_sg_map = 0;
copy_hash_init_values(req_ctx->partial_hash, digestsize);
return 0;
}
@@ -1789,8 +1963,7 @@ static int chcr_hmac_init(struct ahash_request *areq)
{
struct chcr_ahash_req_ctx *req_ctx = ahash_request_ctx(areq);
struct crypto_ahash *rtfm = crypto_ahash_reqtfm(areq);
- struct chcr_context *ctx = crypto_tfm_ctx(crypto_ahash_tfm(rtfm));
- struct hmac_ctx *hmacctx = HMAC_CTX(ctx);
+ struct hmac_ctx *hmacctx = HMAC_CTX(h_ctx(rtfm));
unsigned int digestsize = crypto_ahash_digestsize(rtfm);
unsigned int bs = crypto_tfm_alg_blocksize(crypto_ahash_tfm(rtfm));
@@ -1836,29 +2009,48 @@ static void chcr_hmac_cra_exit(struct crypto_tfm *tfm)
}
}
-static int chcr_copy_assoc(struct aead_request *req,
- struct chcr_aead_ctx *ctx)
+static int chcr_aead_common_init(struct aead_request *req,
+ unsigned short op_type)
{
- SKCIPHER_REQUEST_ON_STACK(skreq, ctx->null);
-
- skcipher_request_set_tfm(skreq, ctx->null);
- skcipher_request_set_callback(skreq, aead_request_flags(req),
- NULL, NULL);
- skcipher_request_set_crypt(skreq, req->src, req->dst, req->assoclen,
- NULL);
+ struct crypto_aead *tfm = crypto_aead_reqtfm(req);
+ struct chcr_aead_ctx *aeadctx = AEAD_CTX(a_ctx(tfm));
+ struct chcr_aead_reqctx *reqctx = aead_request_ctx(req);
+ int error = -EINVAL;
+ unsigned int dst_size;
+ unsigned int authsize = crypto_aead_authsize(tfm);
- return crypto_skcipher_encrypt(skreq);
+ dst_size = req->assoclen + req->cryptlen + (op_type ?
+ -authsize : authsize);
+ /* validate key size */
+ if (aeadctx->enckey_len == 0)
+ goto err;
+ if (op_type && req->cryptlen < authsize)
+ goto err;
+ error = chcr_aead_dma_map(&ULD_CTX(a_ctx(tfm))->lldi.pdev->dev, req,
+ op_type);
+ if (error) {
+ error = -ENOMEM;
+ goto err;
+ }
+ reqctx->aad_nents = sg_nents_xlen(req->src, req->assoclen,
+ CHCR_SRC_SG_SIZE, 0);
+ reqctx->src_nents = sg_nents_xlen(req->src, req->cryptlen,
+ CHCR_SRC_SG_SIZE, req->assoclen);
+ return 0;
+err:
+ return error;
}
-static int chcr_aead_need_fallback(struct aead_request *req, int src_nent,
+
+static int chcr_aead_need_fallback(struct aead_request *req, int dst_nents,
int aadmax, int wrlen,
unsigned short op_type)
{
unsigned int authsize = crypto_aead_authsize(crypto_aead_reqtfm(req));
if (((req->cryptlen - (op_type ? authsize : 0)) == 0) ||
+ dst_nents > MAX_DSGL_ENT ||
(req->assoclen > aadmax) ||
- (src_nent > MAX_SKB_FRAGS) ||
- (wrlen > MAX_WR_SIZE))
+ (wrlen > SGE_MAX_WR_LEN))
return 1;
return 0;
}
@@ -1866,8 +2058,7 @@ static int chcr_aead_need_fallback(struct aead_request *req, int src_nent,
static int chcr_aead_fallback(struct aead_request *req, unsigned short op_type)
{
struct crypto_aead *tfm = crypto_aead_reqtfm(req);
- struct chcr_context *ctx = crypto_aead_ctx(tfm);
- struct chcr_aead_ctx *aeadctx = AEAD_CTX(ctx);
+ struct chcr_aead_ctx *aeadctx = AEAD_CTX(a_ctx(tfm));
struct aead_request *subreq = aead_request_ctx(req);
aead_request_set_tfm(subreq, aeadctx->sw_cipher);
@@ -1886,84 +2077,75 @@ static struct sk_buff *create_authenc_wr(struct aead_request *req,
unsigned short op_type)
{
struct crypto_aead *tfm = crypto_aead_reqtfm(req);
- struct chcr_context *ctx = crypto_aead_ctx(tfm);
- struct uld_ctx *u_ctx = ULD_CTX(ctx);
- struct chcr_aead_ctx *aeadctx = AEAD_CTX(ctx);
+ struct chcr_aead_ctx *aeadctx = AEAD_CTX(a_ctx(tfm));
struct chcr_authenc_ctx *actx = AUTHENC_CTX(aeadctx);
struct chcr_aead_reqctx *reqctx = aead_request_ctx(req);
struct sk_buff *skb = NULL;
struct chcr_wr *chcr_req;
struct cpl_rx_phys_dsgl *phys_cpl;
- struct phys_sge_parm sg_param;
- struct scatterlist *src;
- unsigned int frags = 0, transhdr_len;
- unsigned int ivsize = crypto_aead_ivsize(tfm), dst_size = 0;
- unsigned int kctx_len = 0, nents;
- unsigned short stop_offset = 0;
+ struct ulptx_sgl *ulptx;
+ unsigned int transhdr_len;
+ unsigned int dst_size = 0, temp;
+ unsigned int kctx_len = 0, dnents;
unsigned int assoclen = req->assoclen;
unsigned int authsize = crypto_aead_authsize(tfm);
- int error = -EINVAL, src_nent;
+ int error = -EINVAL;
int null = 0;
gfp_t flags = req->base.flags & CRYPTO_TFM_REQ_MAY_SLEEP ? GFP_KERNEL :
GFP_ATOMIC;
- struct adapter *adap = padap(ctx->dev);
+ struct adapter *adap = padap(a_ctx(tfm)->dev);
- dst_size = req->assoclen + req->cryptlen + (op_type ? -authsize :
- authsize);
- if (aeadctx->enckey_len == 0 || (req->cryptlen <= 0))
- goto err;
+ if (req->cryptlen == 0)
+ return NULL;
- if (op_type && req->cryptlen < crypto_aead_authsize(tfm))
- goto err;
- src_nent = sg_nents_for_len(req->src, req->assoclen + req->cryptlen);
- if (src_nent < 0)
- goto err;
- src = scatterwalk_ffwd(reqctx->srcffwd, req->src, req->assoclen);
- reqctx->dst = src;
- if (req->src != req->dst) {
- error = chcr_copy_assoc(req, aeadctx);
- if (error)
- return ERR_PTR(error);
- reqctx->dst = scatterwalk_ffwd(reqctx->dstffwd, req->dst,
- req->assoclen);
- }
+ reqctx->b0_dma = 0;
if (get_aead_subtype(tfm) == CRYPTO_ALG_SUB_TYPE_AEAD_NULL) {
null = 1;
assoclen = 0;
}
- reqctx->dst_nents = sg_nents_for_len(reqctx->dst, req->cryptlen +
- (op_type ? -authsize : authsize));
- if (reqctx->dst_nents < 0) {
- pr_err("AUTHENC:Invalid Destination sg entries\n");
- error = -EINVAL;
- goto err;
+ dst_size = assoclen + req->cryptlen + (op_type ? -authsize :
+ authsize);
+ error = chcr_aead_common_init(req, op_type);
+ if (error)
+ return ERR_PTR(error);
+ if (dst_size) {
+ dnents = sg_nents_xlen(req->dst, assoclen, CHCR_DST_SG_SIZE, 0);
+ dnents += sg_nents_xlen(req->dst, req->cryptlen +
+ (op_type ? -authsize : authsize), CHCR_DST_SG_SIZE,
+ req->assoclen);
+ dnents += MIN_AUTH_SG; // For IV
+ } else {
+ dnents = 0;
}
- nents = dst_size ? dstsg_2k(reqctx->dst, req->cryptlen +
- (op_type ? -authsize : authsize)) : 0;
- dst_size = get_space_for_phys_dsgl(nents);
+
+ dst_size = get_space_for_phys_dsgl(dnents);
kctx_len = (ntohl(KEY_CONTEXT_CTX_LEN_V(aeadctx->key_ctx_hdr)) << 4)
- sizeof(chcr_req->key_ctx);
transhdr_len = CIPHER_TRANSHDR_SIZE(kctx_len, dst_size);
- if (chcr_aead_need_fallback(req, src_nent + MIN_AUTH_SG,
- T6_MAX_AAD_SIZE,
- transhdr_len + (sgl_len(src_nent + MIN_AUTH_SG) * 8),
- op_type)) {
+ reqctx->imm = (transhdr_len + assoclen + IV + req->cryptlen) <
+ SGE_MAX_WR_LEN;
+ temp = reqctx->imm ? (DIV_ROUND_UP((assoclen + IV + req->cryptlen), 16)
+ * 16) : (sgl_len(reqctx->src_nents + reqctx->aad_nents
+ + MIN_GCM_SG) * 8);
+ transhdr_len += temp;
+ transhdr_len = DIV_ROUND_UP(transhdr_len, 16) * 16;
+
+ if (chcr_aead_need_fallback(req, dnents, T6_MAX_AAD_SIZE,
+ transhdr_len, op_type)) {
atomic_inc(&adap->chcr_stats.fallback);
+ chcr_aead_dma_unmap(&ULD_CTX(a_ctx(tfm))->lldi.pdev->dev, req,
+ op_type);
return ERR_PTR(chcr_aead_fallback(req, op_type));
}
- skb = alloc_skb((transhdr_len + sizeof(struct sge_opaque_hdr)), flags);
+ skb = alloc_skb(SGE_MAX_WR_LEN, flags);
if (!skb) {
error = -ENOMEM;
goto err;
}
- /* LLD is going to write the sge hdr. */
- skb_reserve(skb, sizeof(struct sge_opaque_hdr));
-
- /* Write WR */
chcr_req = __skb_put_zero(skb, transhdr_len);
- stop_offset = (op_type == CHCR_ENCRYPT_OP) ? 0 : authsize;
+ temp = (op_type == CHCR_ENCRYPT_OP) ? 0 : authsize;
/*
* Input order is AAD,IV and Payload. where IV should be included as
@@ -1971,24 +2153,24 @@ static struct sk_buff *create_authenc_wr(struct aead_request *req,
* to the hardware spec
*/
chcr_req->sec_cpl.op_ivinsrtofst =
- FILL_SEC_CPL_OP_IVINSR(ctx->dev->rx_channel_id, 2,
- (ivsize ? (assoclen + 1) : 0));
- chcr_req->sec_cpl.pldlen = htonl(assoclen + ivsize + req->cryptlen);
+ FILL_SEC_CPL_OP_IVINSR(a_ctx(tfm)->dev->rx_channel_id, 2,
+ assoclen + 1);
+ chcr_req->sec_cpl.pldlen = htonl(assoclen + IV + req->cryptlen);
chcr_req->sec_cpl.aadstart_cipherstop_hi = FILL_SEC_CPL_CIPHERSTOP_HI(
assoclen ? 1 : 0, assoclen,
- assoclen + ivsize + 1,
- (stop_offset & 0x1F0) >> 4);
+ assoclen + IV + 1,
+ (temp & 0x1F0) >> 4);
chcr_req->sec_cpl.cipherstop_lo_authinsert = FILL_SEC_CPL_AUTHINSERT(
- stop_offset & 0xF,
- null ? 0 : assoclen + ivsize + 1,
- stop_offset, stop_offset);
+ temp & 0xF,
+ null ? 0 : assoclen + IV + 1,
+ temp, temp);
chcr_req->sec_cpl.seqno_numivs = FILL_SEC_CPL_SCMD0_SEQNO(op_type,
(op_type == CHCR_ENCRYPT_OP) ? 1 : 0,
CHCR_SCMD_CIPHER_MODE_AES_CBC,
actx->auth_mode, aeadctx->hmac_ctrl,
- ivsize >> 1);
+ IV >> 1);
chcr_req->sec_cpl.ivgen_hdrlen = FILL_SEC_CPL_IVGEN_HDRLEN(0, 0, 1,
- 0, 1, dst_size);
+ 0, 0, dst_size);
chcr_req->key_ctx.ctx_hdr = aeadctx->key_ctx_hdr;
if (op_type == CHCR_ENCRYPT_OP)
@@ -2001,39 +2183,312 @@ static struct sk_buff *create_authenc_wr(struct aead_request *req,
memcpy(chcr_req->key_ctx.key + (DIV_ROUND_UP(aeadctx->enckey_len, 16) <<
4), actx->h_iopad, kctx_len -
(DIV_ROUND_UP(aeadctx->enckey_len, 16) << 4));
-
+ memcpy(reqctx->iv, req->iv, IV);
phys_cpl = (struct cpl_rx_phys_dsgl *)((u8 *)(chcr_req + 1) + kctx_len);
- sg_param.nents = reqctx->dst_nents;
- sg_param.obsize = req->cryptlen + (op_type ? -authsize : authsize);
- sg_param.qid = qid;
- error = map_writesg_phys_cpl(&u_ctx->lldi.pdev->dev, phys_cpl,
- reqctx->dst, &sg_param);
- if (error)
- goto dstmap_fail;
-
- skb_set_transport_header(skb, transhdr_len);
-
- if (assoclen) {
- /* AAD buffer in */
- write_sg_to_skb(skb, &frags, req->src, assoclen);
-
- }
- write_buffer_to_skb(skb, &frags, req->iv, ivsize);
- write_sg_to_skb(skb, &frags, src, req->cryptlen);
+ ulptx = (struct ulptx_sgl *)((u8 *)(phys_cpl + 1) + dst_size);
+ chcr_add_aead_dst_ent(req, phys_cpl, assoclen, op_type, qid);
+ chcr_add_aead_src_ent(req, ulptx, assoclen, op_type);
atomic_inc(&adap->chcr_stats.cipher_rqst);
- create_wreq(ctx, chcr_req, &req->base, skb, size,
- sizeof(struct cpl_rx_phys_dsgl) + dst_size + kctx_len, 0);
+ temp = sizeof(struct cpl_rx_phys_dsgl) + dst_size +
+ kctx_len + (reqctx->imm ? (assoclen + IV + req->cryptlen) : 0);
+ create_wreq(a_ctx(tfm), chcr_req, &req->base, reqctx->imm, size,
+ transhdr_len, temp, 0);
reqctx->skb = skb;
- skb_get(skb);
+ reqctx->op = op_type;
return skb;
-dstmap_fail:
- /* ivmap_fail: */
- kfree_skb(skb);
err:
+ chcr_aead_dma_unmap(&ULD_CTX(a_ctx(tfm))->lldi.pdev->dev, req,
+ op_type);
+
return ERR_PTR(error);
}
+static int chcr_aead_dma_map(struct device *dev,
+ struct aead_request *req,
+ unsigned short op_type)
+{
+ int error;
+ struct chcr_aead_reqctx *reqctx = aead_request_ctx(req);
+ struct crypto_aead *tfm = crypto_aead_reqtfm(req);
+ unsigned int authsize = crypto_aead_authsize(tfm);
+ int dst_size;
+
+ dst_size = req->assoclen + req->cryptlen + (op_type ?
+ -authsize : authsize);
+ if (!req->cryptlen || !dst_size)
+ return 0;
+ reqctx->iv_dma = dma_map_single(dev, reqctx->iv, IV,
+ DMA_BIDIRECTIONAL);
+ if (dma_mapping_error(dev, reqctx->iv_dma))
+ return -ENOMEM;
+
+ if (req->src == req->dst) {
+ error = dma_map_sg(dev, req->src, sg_nents(req->src),
+ DMA_BIDIRECTIONAL);
+ if (!error)
+ goto err;
+ } else {
+ error = dma_map_sg(dev, req->src, sg_nents(req->src),
+ DMA_TO_DEVICE);
+ if (!error)
+ goto err;
+ error = dma_map_sg(dev, req->dst, sg_nents(req->dst),
+ DMA_FROM_DEVICE);
+ if (!error) {
+ dma_unmap_sg(dev, req->src, sg_nents(req->src),
+ DMA_TO_DEVICE);
+ goto err;
+ }
+ }
+
+ return 0;
+err:
+ dma_unmap_single(dev, reqctx->iv_dma, IV, DMA_BIDIRECTIONAL);
+ return -ENOMEM;
+}
+
+static void chcr_aead_dma_unmap(struct device *dev,
+ struct aead_request *req,
+ unsigned short op_type)
+{
+ struct chcr_aead_reqctx *reqctx = aead_request_ctx(req);
+ struct crypto_aead *tfm = crypto_aead_reqtfm(req);
+ unsigned int authsize = crypto_aead_authsize(tfm);
+ int dst_size;
+
+ dst_size = req->assoclen + req->cryptlen + (op_type ?
+ -authsize : authsize);
+ if (!req->cryptlen || !dst_size)
+ return;
+
+ dma_unmap_single(dev, reqctx->iv_dma, IV,
+ DMA_BIDIRECTIONAL);
+ if (req->src == req->dst) {
+ dma_unmap_sg(dev, req->src, sg_nents(req->src),
+ DMA_BIDIRECTIONAL);
+ } else {
+ dma_unmap_sg(dev, req->src, sg_nents(req->src),
+ DMA_TO_DEVICE);
+ dma_unmap_sg(dev, req->dst, sg_nents(req->dst),
+ DMA_FROM_DEVICE);
+ }
+}
+
+static inline void chcr_add_aead_src_ent(struct aead_request *req,
+ struct ulptx_sgl *ulptx,
+ unsigned int assoclen,
+ unsigned short op_type)
+{
+ struct ulptx_walk ulp_walk;
+ struct chcr_aead_reqctx *reqctx = aead_request_ctx(req);
+
+ if (reqctx->imm) {
+ u8 *buf = (u8 *)ulptx;
+
+ if (reqctx->b0_dma) {
+ memcpy(buf, reqctx->scratch_pad, reqctx->b0_len);
+ buf += reqctx->b0_len;
+ }
+ sg_pcopy_to_buffer(req->src, sg_nents(req->src),
+ buf, assoclen, 0);
+ buf += assoclen;
+ memcpy(buf, reqctx->iv, IV);
+ buf += IV;
+ sg_pcopy_to_buffer(req->src, sg_nents(req->src),
+ buf, req->cryptlen, req->assoclen);
+ } else {
+ ulptx_walk_init(&ulp_walk, ulptx);
+ if (reqctx->b0_dma)
+ ulptx_walk_add_page(&ulp_walk, reqctx->b0_len,
+ &reqctx->b0_dma);
+ ulptx_walk_add_sg(&ulp_walk, req->src, assoclen, 0);
+ ulptx_walk_add_page(&ulp_walk, IV, &reqctx->iv_dma);
+ ulptx_walk_add_sg(&ulp_walk, req->src, req->cryptlen,
+ req->assoclen);
+ ulptx_walk_end(&ulp_walk);
+ }
+}
+
+static inline void chcr_add_aead_dst_ent(struct aead_request *req,
+ struct cpl_rx_phys_dsgl *phys_cpl,
+ unsigned int assoclen,
+ unsigned short op_type,
+ unsigned short qid)
+{
+ struct chcr_aead_reqctx *reqctx = aead_request_ctx(req);
+ struct crypto_aead *tfm = crypto_aead_reqtfm(req);
+ struct dsgl_walk dsgl_walk;
+ unsigned int authsize = crypto_aead_authsize(tfm);
+ u32 temp;
+
+ dsgl_walk_init(&dsgl_walk, phys_cpl);
+ if (reqctx->b0_dma)
+ dsgl_walk_add_page(&dsgl_walk, reqctx->b0_len, &reqctx->b0_dma);
+ dsgl_walk_add_sg(&dsgl_walk, req->dst, assoclen, 0);
+ dsgl_walk_add_page(&dsgl_walk, IV, &reqctx->iv_dma);
+ temp = req->cryptlen + (op_type ? -authsize : authsize);
+ dsgl_walk_add_sg(&dsgl_walk, req->dst, temp, req->assoclen);
+ dsgl_walk_end(&dsgl_walk, qid);
+}
+
+static inline void chcr_add_cipher_src_ent(struct ablkcipher_request *req,
+ struct ulptx_sgl *ulptx,
+ struct cipher_wr_param *wrparam)
+{
+ struct ulptx_walk ulp_walk;
+ struct chcr_blkcipher_req_ctx *reqctx = ablkcipher_request_ctx(req);
+
+ if (reqctx->imm) {
+ u8 *buf = (u8 *)ulptx;
+
+ memcpy(buf, reqctx->iv, IV);
+ buf += IV;
+ sg_pcopy_to_buffer(req->src, sg_nents(req->src),
+ buf, wrparam->bytes, reqctx->processed);
+ } else {
+ ulptx_walk_init(&ulp_walk, ulptx);
+ ulptx_walk_add_page(&ulp_walk, IV, &reqctx->iv_dma);
+ ulptx_walk_add_sg(&ulp_walk, reqctx->srcsg, wrparam->bytes,
+ reqctx->src_ofst);
+ reqctx->srcsg = ulp_walk.last_sg;
+ reqctx->src_ofst = ulp_walk.last_sg_len;
+ ulptx_walk_end(&ulp_walk);
+ }
+}
+
+static inline void chcr_add_cipher_dst_ent(struct ablkcipher_request *req,
+ struct cpl_rx_phys_dsgl *phys_cpl,
+ struct cipher_wr_param *wrparam,
+ unsigned short qid)
+{
+ struct chcr_blkcipher_req_ctx *reqctx = ablkcipher_request_ctx(req);
+ struct dsgl_walk dsgl_walk;
+
+ dsgl_walk_init(&dsgl_walk, phys_cpl);
+ dsgl_walk_add_page(&dsgl_walk, IV, &reqctx->iv_dma);
+ dsgl_walk_add_sg(&dsgl_walk, reqctx->dstsg, wrparam->bytes,
+ reqctx->dst_ofst);
+ reqctx->dstsg = dsgl_walk.last_sg;
+ reqctx->dst_ofst = dsgl_walk.last_sg_len;
+
+ dsgl_walk_end(&dsgl_walk, qid);
+}
+
+static inline void chcr_add_hash_src_ent(struct ahash_request *req,
+ struct ulptx_sgl *ulptx,
+ struct hash_wr_param *param)
+{
+ struct ulptx_walk ulp_walk;
+ struct chcr_ahash_req_ctx *reqctx = ahash_request_ctx(req);
+
+ if (reqctx->imm) {
+ u8 *buf = (u8 *)ulptx;
+
+ if (param->bfr_len) {
+ memcpy(buf, reqctx->reqbfr, param->bfr_len);
+ buf += param->bfr_len;
+ }
+ sg_pcopy_to_buffer(req->src, sg_nents(req->src),
+ buf, param->sg_len, 0);
+ } else {
+ ulptx_walk_init(&ulp_walk, ulptx);
+ if (param->bfr_len)
+ ulptx_walk_add_page(&ulp_walk, param->bfr_len,
+ &reqctx->dma_addr);
+ ulptx_walk_add_sg(&ulp_walk, req->src, param->sg_len,
+ 0);
+// reqctx->srcsg = ulp_walk.last_sg;
+// reqctx->src_ofst = ulp_walk.last_sg_len;
+ ulptx_walk_end(&ulp_walk);
+ }
+}
+
+
+static inline int chcr_hash_dma_map(struct device *dev,
+ struct ahash_request *req)
+{
+ struct chcr_ahash_req_ctx *req_ctx = ahash_request_ctx(req);
+ int error = 0;
+
+ if (!req->nbytes)
+ return 0;
+ error = dma_map_sg(dev, req->src, sg_nents(req->src),
+ DMA_TO_DEVICE);
+ if (!error)
+ return error;
+ req_ctx->is_sg_map = 1;
+ return 0;
+}
+
+static inline void chcr_hash_dma_unmap(struct device *dev,
+ struct ahash_request *req)
+{
+ struct chcr_ahash_req_ctx *req_ctx = ahash_request_ctx(req);
+
+ if (!req->nbytes)
+ return;
+
+ dma_unmap_sg(dev, req->src, sg_nents(req->src),
+ DMA_TO_DEVICE);
+ req_ctx->is_sg_map = 0;
+
+}
+
+
+static int chcr_cipher_dma_map(struct device *dev,
+ struct ablkcipher_request *req)
+{
+ int error;
+ struct chcr_blkcipher_req_ctx *reqctx = ablkcipher_request_ctx(req);
+
+ reqctx->iv_dma = dma_map_single(dev, reqctx->iv, IV,
+ DMA_BIDIRECTIONAL);
+ if (dma_mapping_error(dev, reqctx->iv_dma))
+ return -ENOMEM;
+
+ if (req->src == req->dst) {
+ error = dma_map_sg(dev, req->src, sg_nents(req->src),
+ DMA_BIDIRECTIONAL);
+ if (!error)
+ goto err;
+ } else {
+ error = dma_map_sg(dev, req->src, sg_nents(req->src),
+ DMA_TO_DEVICE);
+ if (!error)
+ goto err;
+ error = dma_map_sg(dev, req->dst, sg_nents(req->dst),
+ DMA_FROM_DEVICE);
+ if (!error) {
+ dma_unmap_sg(dev, req->src, sg_nents(req->src),
+ DMA_TO_DEVICE);
+ goto err;
+ }
+ }
+
+ return 0;
+err:
+ dma_unmap_single(dev, reqctx->iv_dma, IV, DMA_BIDIRECTIONAL);
+ return -ENOMEM;
+}
+static void chcr_cipher_dma_unmap(struct device *dev,
+ struct ablkcipher_request *req)
+{
+ struct chcr_blkcipher_req_ctx *reqctx = ablkcipher_request_ctx(req);
+
+ dma_unmap_single(dev, reqctx->iv_dma, IV,
+ DMA_BIDIRECTIONAL);
+ if (req->src == req->dst) {
+ dma_unmap_sg(dev, req->src, sg_nents(req->src),
+ DMA_BIDIRECTIONAL);
+ } else {
+ dma_unmap_sg(dev, req->src, sg_nents(req->src),
+ DMA_TO_DEVICE);
+ dma_unmap_sg(dev, req->dst, sg_nents(req->dst),
+ DMA_FROM_DEVICE);
+ }
+}
+
static int set_msg_len(u8 *block, unsigned int msglen, int csize)
{
__be32 data;
@@ -2118,15 +2573,13 @@ static int ccm_format_packet(struct aead_request *req,
static void fill_sec_cpl_for_aead(struct cpl_tx_sec_pdu *sec_cpl,
unsigned int dst_size,
struct aead_request *req,
- unsigned short op_type,
- struct chcr_context *chcrctx)
+ unsigned short op_type)
{
struct crypto_aead *tfm = crypto_aead_reqtfm(req);
- struct chcr_aead_ctx *aeadctx = AEAD_CTX(crypto_aead_ctx(tfm));
- unsigned int ivsize = AES_BLOCK_SIZE;
+ struct chcr_aead_ctx *aeadctx = AEAD_CTX(a_ctx(tfm));
unsigned int cipher_mode = CHCR_SCMD_CIPHER_MODE_AES_CCM;
unsigned int mac_mode = CHCR_SCMD_AUTH_MODE_CBCMAC;
- unsigned int c_id = chcrctx->dev->rx_channel_id;
+ unsigned int c_id = a_ctx(tfm)->dev->rx_channel_id;
unsigned int ccm_xtra;
unsigned char tag_offset = 0, auth_offset = 0;
unsigned int assoclen;
@@ -2139,7 +2592,7 @@ static void fill_sec_cpl_for_aead(struct cpl_tx_sec_pdu *sec_cpl,
((assoclen) ? CCM_AAD_FIELD_SIZE : 0);
auth_offset = req->cryptlen ?
- (assoclen + ivsize + 1 + ccm_xtra) : 0;
+ (assoclen + IV + 1 + ccm_xtra) : 0;
if (op_type == CHCR_DECRYPT_OP) {
if (crypto_aead_authsize(tfm) != req->cryptlen)
tag_offset = crypto_aead_authsize(tfm);
@@ -2149,14 +2602,13 @@ static void fill_sec_cpl_for_aead(struct cpl_tx_sec_pdu *sec_cpl,
sec_cpl->op_ivinsrtofst = FILL_SEC_CPL_OP_IVINSR(c_id,
- 2, (ivsize ? (assoclen + 1) : 0) +
- ccm_xtra);
+ 2, assoclen + 1 + ccm_xtra);
sec_cpl->pldlen =
- htonl(assoclen + ivsize + req->cryptlen + ccm_xtra);
+ htonl(assoclen + IV + req->cryptlen + ccm_xtra);
/* For CCM there wil be b0 always. So AAD start will be 1 always */
sec_cpl->aadstart_cipherstop_hi = FILL_SEC_CPL_CIPHERSTOP_HI(
1, assoclen + ccm_xtra, assoclen
- + ivsize + 1 + ccm_xtra, 0);
+ + IV + 1 + ccm_xtra, 0);
sec_cpl->cipherstop_lo_authinsert = FILL_SEC_CPL_AUTHINSERT(0,
auth_offset, tag_offset,
@@ -2165,10 +2617,10 @@ static void fill_sec_cpl_for_aead(struct cpl_tx_sec_pdu *sec_cpl,
sec_cpl->seqno_numivs = FILL_SEC_CPL_SCMD0_SEQNO(op_type,
(op_type == CHCR_ENCRYPT_OP) ? 0 : 1,
cipher_mode, mac_mode,
- aeadctx->hmac_ctrl, ivsize >> 1);
+ aeadctx->hmac_ctrl, IV >> 1);
sec_cpl->ivgen_hdrlen = FILL_SEC_CPL_IVGEN_HDRLEN(0, 0, 1, 0,
- 1, dst_size);
+ 0, dst_size);
}
int aead_ccm_validate_input(unsigned short op_type,
@@ -2188,119 +2640,83 @@ int aead_ccm_validate_input(unsigned short op_type,
return -EINVAL;
}
}
- if (aeadctx->enckey_len == 0) {
- pr_err("CCM: Encryption key not set\n");
- return -EINVAL;
- }
return 0;
}
-unsigned int fill_aead_req_fields(struct sk_buff *skb,
- struct aead_request *req,
- struct scatterlist *src,
- unsigned int ivsize,
- struct chcr_aead_ctx *aeadctx)
-{
- unsigned int frags = 0;
- struct crypto_aead *tfm = crypto_aead_reqtfm(req);
- struct chcr_aead_reqctx *reqctx = aead_request_ctx(req);
- /* b0 and aad length(if available) */
-
- write_buffer_to_skb(skb, &frags, reqctx->scratch_pad, CCM_B0_SIZE +
- (req->assoclen ? CCM_AAD_FIELD_SIZE : 0));
- if (req->assoclen) {
- if (get_aead_subtype(tfm) == CRYPTO_ALG_SUB_TYPE_AEAD_RFC4309)
- write_sg_to_skb(skb, &frags, req->src,
- req->assoclen - 8);
- else
- write_sg_to_skb(skb, &frags, req->src, req->assoclen);
- }
- write_buffer_to_skb(skb, &frags, reqctx->iv, ivsize);
- if (req->cryptlen)
- write_sg_to_skb(skb, &frags, src, req->cryptlen);
-
- return frags;
-}
-
static struct sk_buff *create_aead_ccm_wr(struct aead_request *req,
unsigned short qid,
int size,
unsigned short op_type)
{
struct crypto_aead *tfm = crypto_aead_reqtfm(req);
- struct chcr_context *ctx = crypto_aead_ctx(tfm);
- struct uld_ctx *u_ctx = ULD_CTX(ctx);
- struct chcr_aead_ctx *aeadctx = AEAD_CTX(ctx);
+ struct chcr_aead_ctx *aeadctx = AEAD_CTX(a_ctx(tfm));
struct chcr_aead_reqctx *reqctx = aead_request_ctx(req);
struct sk_buff *skb = NULL;
struct chcr_wr *chcr_req;
struct cpl_rx_phys_dsgl *phys_cpl;
- struct phys_sge_parm sg_param;
- struct scatterlist *src;
- unsigned int frags = 0, transhdr_len, ivsize = AES_BLOCK_SIZE;
- unsigned int dst_size = 0, kctx_len, nents;
- unsigned int sub_type;
+ struct ulptx_sgl *ulptx;
+ unsigned int transhdr_len;
+ unsigned int dst_size = 0, kctx_len, dnents, temp;
+ unsigned int sub_type, assoclen = req->assoclen;
unsigned int authsize = crypto_aead_authsize(tfm);
- int error = -EINVAL, src_nent;
+ int error = -EINVAL;
gfp_t flags = req->base.flags & CRYPTO_TFM_REQ_MAY_SLEEP ? GFP_KERNEL :
GFP_ATOMIC;
- struct adapter *adap = padap(ctx->dev);
+ struct adapter *adap = padap(a_ctx(tfm)->dev);
- dst_size = req->assoclen + req->cryptlen + (op_type ? -authsize :
+ reqctx->b0_dma = 0;
+ sub_type = get_aead_subtype(tfm);
+ if (sub_type == CRYPTO_ALG_SUB_TYPE_AEAD_RFC4309)
+ assoclen -= 8;
+ dst_size = assoclen + req->cryptlen + (op_type ? -authsize :
authsize);
- if (op_type && req->cryptlen < crypto_aead_authsize(tfm))
- goto err;
- src_nent = sg_nents_for_len(req->src, req->assoclen + req->cryptlen);
- if (src_nent < 0)
- goto err;
+ error = chcr_aead_common_init(req, op_type);
+ if (error)
+ return ERR_PTR(error);
- sub_type = get_aead_subtype(tfm);
- src = scatterwalk_ffwd(reqctx->srcffwd, req->src, req->assoclen);
- reqctx->dst = src;
- if (req->src != req->dst) {
- error = chcr_copy_assoc(req, aeadctx);
- if (error) {
- pr_err("AAD copy to destination buffer fails\n");
- return ERR_PTR(error);
- }
- reqctx->dst = scatterwalk_ffwd(reqctx->dstffwd, req->dst,
- req->assoclen);
- }
- reqctx->dst_nents = sg_nents_for_len(reqctx->dst, req->cryptlen +
- (op_type ? -authsize : authsize));
- if (reqctx->dst_nents < 0) {
- pr_err("CCM:Invalid Destination sg entries\n");
- error = -EINVAL;
- goto err;
- }
+
+ reqctx->b0_len = CCM_B0_SIZE + (assoclen ? CCM_AAD_FIELD_SIZE : 0);
error = aead_ccm_validate_input(op_type, req, aeadctx, sub_type);
if (error)
goto err;
- nents = dst_size ? dstsg_2k(reqctx->dst, req->cryptlen +
- (op_type ? -authsize : authsize)) : 0;
- dst_size = get_space_for_phys_dsgl(nents);
+ if (dst_size) {
+ dnents = sg_nents_xlen(req->dst, assoclen, CHCR_DST_SG_SIZE, 0);
+ dnents += sg_nents_xlen(req->dst, req->cryptlen
+ + (op_type ? -authsize : authsize),
+ CHCR_DST_SG_SIZE, req->assoclen);
+ dnents += MIN_CCM_SG; // For IV and B0
+ } else {
+ dnents = 0;
+ }
+ dst_size = get_space_for_phys_dsgl(dnents);
kctx_len = ((DIV_ROUND_UP(aeadctx->enckey_len, 16)) << 4) * 2;
transhdr_len = CIPHER_TRANSHDR_SIZE(kctx_len, dst_size);
- if (chcr_aead_need_fallback(req, src_nent + MIN_CCM_SG,
- T6_MAX_AAD_SIZE - 18,
- transhdr_len + (sgl_len(src_nent + MIN_CCM_SG) * 8),
- op_type)) {
+ reqctx->imm = (transhdr_len + assoclen + IV + req->cryptlen +
+ reqctx->b0_len) <= SGE_MAX_WR_LEN;
+ temp = reqctx->imm ? (DIV_ROUND_UP((assoclen + IV + req->cryptlen +
+ reqctx->b0_len), 16) * 16) :
+ (sgl_len(reqctx->src_nents + reqctx->aad_nents +
+ MIN_CCM_SG) * 8);
+ transhdr_len += temp;
+ transhdr_len = DIV_ROUND_UP(transhdr_len, 16) * 16;
+
+ if (chcr_aead_need_fallback(req, dnents, T6_MAX_AAD_SIZE -
+ reqctx->b0_len, transhdr_len, op_type)) {
atomic_inc(&adap->chcr_stats.fallback);
+ chcr_aead_dma_unmap(&ULD_CTX(a_ctx(tfm))->lldi.pdev->dev, req,
+ op_type);
return ERR_PTR(chcr_aead_fallback(req, op_type));
}
-
- skb = alloc_skb((transhdr_len + sizeof(struct sge_opaque_hdr)), flags);
+ skb = alloc_skb(SGE_MAX_WR_LEN, flags);
if (!skb) {
error = -ENOMEM;
goto err;
}
- skb_reserve(skb, sizeof(struct sge_opaque_hdr));
+ chcr_req = (struct chcr_wr *) __skb_put_zero(skb, transhdr_len);
- chcr_req = __skb_put_zero(skb, transhdr_len);
-
- fill_sec_cpl_for_aead(&chcr_req->sec_cpl, dst_size, req, op_type, ctx);
+ fill_sec_cpl_for_aead(&chcr_req->sec_cpl, dst_size, req, op_type);
chcr_req->key_ctx.ctx_hdr = aeadctx->key_ctx_hdr;
memcpy(chcr_req->key_ctx.key, aeadctx->key, aeadctx->enckey_len);
@@ -2308,29 +2724,37 @@ static struct sk_buff *create_aead_ccm_wr(struct aead_request *req,
16), aeadctx->key, aeadctx->enckey_len);
phys_cpl = (struct cpl_rx_phys_dsgl *)((u8 *)(chcr_req + 1) + kctx_len);
+ ulptx = (struct ulptx_sgl *)((u8 *)(phys_cpl + 1) + dst_size);
error = ccm_format_packet(req, aeadctx, sub_type, op_type);
if (error)
goto dstmap_fail;
- sg_param.nents = reqctx->dst_nents;
- sg_param.obsize = req->cryptlen + (op_type ? -authsize : authsize);
- sg_param.qid = qid;
- error = map_writesg_phys_cpl(&u_ctx->lldi.pdev->dev, phys_cpl,
- reqctx->dst, &sg_param);
- if (error)
+ reqctx->b0_dma = dma_map_single(&ULD_CTX(a_ctx(tfm))->lldi.pdev->dev,
+ &reqctx->scratch_pad, reqctx->b0_len,
+ DMA_BIDIRECTIONAL);
+ if (dma_mapping_error(&ULD_CTX(a_ctx(tfm))->lldi.pdev->dev,
+ reqctx->b0_dma)) {
+ error = -ENOMEM;
goto dstmap_fail;
+ }
+
+ chcr_add_aead_dst_ent(req, phys_cpl, assoclen, op_type, qid);
+ chcr_add_aead_src_ent(req, ulptx, assoclen, op_type);
- skb_set_transport_header(skb, transhdr_len);
- frags = fill_aead_req_fields(skb, req, src, ivsize, aeadctx);
atomic_inc(&adap->chcr_stats.aead_rqst);
- create_wreq(ctx, chcr_req, &req->base, skb, 0,
- sizeof(struct cpl_rx_phys_dsgl) + dst_size + kctx_len, 0);
+ temp = sizeof(struct cpl_rx_phys_dsgl) + dst_size +
+ kctx_len + (reqctx->imm ? (assoclen + IV + req->cryptlen +
+ reqctx->b0_len) : 0);
+ create_wreq(a_ctx(tfm), chcr_req, &req->base, reqctx->imm, 0,
+ transhdr_len, temp, 0);
reqctx->skb = skb;
- skb_get(skb);
+ reqctx->op = op_type;
+
return skb;
dstmap_fail:
kfree_skb(skb);
err:
+ chcr_aead_dma_unmap(&ULD_CTX(a_ctx(tfm))->lldi.pdev->dev, req, op_type);
return ERR_PTR(error);
}
@@ -2340,101 +2764,84 @@ static struct sk_buff *create_gcm_wr(struct aead_request *req,
unsigned short op_type)
{
struct crypto_aead *tfm = crypto_aead_reqtfm(req);
- struct chcr_context *ctx = crypto_aead_ctx(tfm);
- struct uld_ctx *u_ctx = ULD_CTX(ctx);
- struct chcr_aead_ctx *aeadctx = AEAD_CTX(ctx);
+ struct chcr_aead_ctx *aeadctx = AEAD_CTX(a_ctx(tfm));
struct chcr_aead_reqctx *reqctx = aead_request_ctx(req);
struct sk_buff *skb = NULL;
struct chcr_wr *chcr_req;
struct cpl_rx_phys_dsgl *phys_cpl;
- struct phys_sge_parm sg_param;
- struct scatterlist *src;
- unsigned int frags = 0, transhdr_len;
- unsigned int ivsize = AES_BLOCK_SIZE;
- unsigned int dst_size = 0, kctx_len, nents, assoclen = req->assoclen;
- unsigned char tag_offset = 0;
+ struct ulptx_sgl *ulptx;
+ unsigned int transhdr_len, dnents = 0;
+ unsigned int dst_size = 0, temp = 0, kctx_len, assoclen = req->assoclen;
unsigned int authsize = crypto_aead_authsize(tfm);
- int error = -EINVAL, src_nent;
+ int error = -EINVAL;
gfp_t flags = req->base.flags & CRYPTO_TFM_REQ_MAY_SLEEP ? GFP_KERNEL :
GFP_ATOMIC;
- struct adapter *adap = padap(ctx->dev);
-
- dst_size = assoclen + req->cryptlen + (op_type ? -authsize :
- authsize);
- /* validate key size */
- if (aeadctx->enckey_len == 0)
- goto err;
+ struct adapter *adap = padap(a_ctx(tfm)->dev);
- if (op_type && req->cryptlen < crypto_aead_authsize(tfm))
- goto err;
- src_nent = sg_nents_for_len(req->src, assoclen + req->cryptlen);
- if (src_nent < 0)
- goto err;
+ if (get_aead_subtype(tfm) == CRYPTO_ALG_SUB_TYPE_AEAD_RFC4106)
+ assoclen = req->assoclen - 8;
- src = scatterwalk_ffwd(reqctx->srcffwd, req->src, assoclen);
- reqctx->dst = src;
- if (req->src != req->dst) {
- error = chcr_copy_assoc(req, aeadctx);
+ reqctx->b0_dma = 0;
+ dst_size = assoclen + req->cryptlen + (op_type ? -authsize : authsize);
+ error = chcr_aead_common_init(req, op_type);
if (error)
return ERR_PTR(error);
- reqctx->dst = scatterwalk_ffwd(reqctx->dstffwd, req->dst,
- req->assoclen);
- }
- reqctx->dst_nents = sg_nents_for_len(reqctx->dst, req->cryptlen +
- (op_type ? -authsize : authsize));
- if (reqctx->dst_nents < 0) {
- pr_err("GCM:Invalid Destination sg entries\n");
- error = -EINVAL;
- goto err;
+ if (dst_size) {
+ dnents = sg_nents_xlen(req->dst, assoclen, CHCR_DST_SG_SIZE, 0);
+ dnents += sg_nents_xlen(req->dst,
+ req->cryptlen + (op_type ? -authsize : authsize),
+ CHCR_DST_SG_SIZE, req->assoclen);
+ dnents += MIN_GCM_SG; // For IV
+ } else {
+ dnents = 0;
}
-
- nents = dst_size ? dstsg_2k(reqctx->dst, req->cryptlen +
- (op_type ? -authsize : authsize)) : 0;
- dst_size = get_space_for_phys_dsgl(nents);
+ dst_size = get_space_for_phys_dsgl(dnents);
kctx_len = ((DIV_ROUND_UP(aeadctx->enckey_len, 16)) << 4) +
AEAD_H_SIZE;
transhdr_len = CIPHER_TRANSHDR_SIZE(kctx_len, dst_size);
- if (chcr_aead_need_fallback(req, src_nent + MIN_GCM_SG,
- T6_MAX_AAD_SIZE,
- transhdr_len + (sgl_len(src_nent + MIN_GCM_SG) * 8),
- op_type)) {
+ reqctx->imm = (transhdr_len + assoclen + IV + req->cryptlen) <=
+ SGE_MAX_WR_LEN;
+ temp = reqctx->imm ? (DIV_ROUND_UP((assoclen + IV +
+ req->cryptlen), 16) * 16) : (sgl_len(reqctx->src_nents +
+ reqctx->aad_nents + MIN_GCM_SG) * 8);
+ transhdr_len += temp;
+ transhdr_len = DIV_ROUND_UP(transhdr_len, 16) * 16;
+ if (chcr_aead_need_fallback(req, dnents, T6_MAX_AAD_SIZE,
+ transhdr_len, op_type)) {
atomic_inc(&adap->chcr_stats.fallback);
+ chcr_aead_dma_unmap(&ULD_CTX(a_ctx(tfm))->lldi.pdev->dev, req,
+ op_type);
return ERR_PTR(chcr_aead_fallback(req, op_type));
}
- skb = alloc_skb((transhdr_len + sizeof(struct sge_opaque_hdr)), flags);
+ skb = alloc_skb(SGE_MAX_WR_LEN, flags);
if (!skb) {
error = -ENOMEM;
goto err;
}
- /* NIC driver is going to write the sge hdr. */
- skb_reserve(skb, sizeof(struct sge_opaque_hdr));
-
chcr_req = __skb_put_zero(skb, transhdr_len);
- if (get_aead_subtype(tfm) == CRYPTO_ALG_SUB_TYPE_AEAD_RFC4106)
- assoclen = req->assoclen - 8;
-
- tag_offset = (op_type == CHCR_ENCRYPT_OP) ? 0 : authsize;
+ //Offset of tag from end
+ temp = (op_type == CHCR_ENCRYPT_OP) ? 0 : authsize;
chcr_req->sec_cpl.op_ivinsrtofst = FILL_SEC_CPL_OP_IVINSR(
- ctx->dev->rx_channel_id, 2, (ivsize ?
- (assoclen + 1) : 0));
+ a_ctx(tfm)->dev->rx_channel_id, 2,
+ (assoclen + 1));
chcr_req->sec_cpl.pldlen =
- htonl(assoclen + ivsize + req->cryptlen);
+ htonl(assoclen + IV + req->cryptlen);
chcr_req->sec_cpl.aadstart_cipherstop_hi = FILL_SEC_CPL_CIPHERSTOP_HI(
assoclen ? 1 : 0, assoclen,
- assoclen + ivsize + 1, 0);
+ assoclen + IV + 1, 0);
chcr_req->sec_cpl.cipherstop_lo_authinsert =
- FILL_SEC_CPL_AUTHINSERT(0, assoclen + ivsize + 1,
- tag_offset, tag_offset);
+ FILL_SEC_CPL_AUTHINSERT(0, assoclen + IV + 1,
+ temp, temp);
chcr_req->sec_cpl.seqno_numivs =
FILL_SEC_CPL_SCMD0_SEQNO(op_type, (op_type ==
CHCR_ENCRYPT_OP) ? 1 : 0,
CHCR_SCMD_CIPHER_MODE_AES_GCM,
CHCR_SCMD_AUTH_MODE_GHASH,
- aeadctx->hmac_ctrl, ivsize >> 1);
+ aeadctx->hmac_ctrl, IV >> 1);
chcr_req->sec_cpl.ivgen_hdrlen = FILL_SEC_CPL_IVGEN_HDRLEN(0, 0, 1,
- 0, 1, dst_size);
+ 0, 0, dst_size);
chcr_req->key_ctx.ctx_hdr = aeadctx->key_ctx_hdr;
memcpy(chcr_req->key_ctx.key, aeadctx->key, aeadctx->enckey_len);
memcpy(chcr_req->key_ctx.key + (DIV_ROUND_UP(aeadctx->enckey_len, 16) *
@@ -2452,30 +2859,21 @@ static struct sk_buff *create_gcm_wr(struct aead_request *req,
*((unsigned int *)(reqctx->iv + 12)) = htonl(0x01);
phys_cpl = (struct cpl_rx_phys_dsgl *)((u8 *)(chcr_req + 1) + kctx_len);
- sg_param.nents = reqctx->dst_nents;
- sg_param.obsize = req->cryptlen + (op_type ? -authsize : authsize);
- sg_param.qid = qid;
- error = map_writesg_phys_cpl(&u_ctx->lldi.pdev->dev, phys_cpl,
- reqctx->dst, &sg_param);
- if (error)
- goto dstmap_fail;
+ ulptx = (struct ulptx_sgl *)((u8 *)(phys_cpl + 1) + dst_size);
- skb_set_transport_header(skb, transhdr_len);
- write_sg_to_skb(skb, &frags, req->src, assoclen);
- write_buffer_to_skb(skb, &frags, reqctx->iv, ivsize);
- write_sg_to_skb(skb, &frags, src, req->cryptlen);
+ chcr_add_aead_dst_ent(req, phys_cpl, assoclen, op_type, qid);
+ chcr_add_aead_src_ent(req, ulptx, assoclen, op_type);
atomic_inc(&adap->chcr_stats.aead_rqst);
- create_wreq(ctx, chcr_req, &req->base, skb, size,
- sizeof(struct cpl_rx_phys_dsgl) + dst_size + kctx_len,
- reqctx->verify);
+ temp = sizeof(struct cpl_rx_phys_dsgl) + dst_size +
+ kctx_len + (reqctx->imm ? (assoclen + IV + req->cryptlen) : 0);
+ create_wreq(a_ctx(tfm), chcr_req, &req->base, reqctx->imm, size,
+ transhdr_len, temp, reqctx->verify);
reqctx->skb = skb;
- skb_get(skb);
+ reqctx->op = op_type;
return skb;
-dstmap_fail:
- /* ivmap_fail: */
- kfree_skb(skb);
err:
+ chcr_aead_dma_unmap(&ULD_CTX(a_ctx(tfm))->lldi.pdev->dev, req, op_type);
return ERR_PTR(error);
}
@@ -2483,8 +2881,7 @@ static struct sk_buff *create_gcm_wr(struct aead_request *req,
static int chcr_aead_cra_init(struct crypto_aead *tfm)
{
- struct chcr_context *ctx = crypto_aead_ctx(tfm);
- struct chcr_aead_ctx *aeadctx = AEAD_CTX(ctx);
+ struct chcr_aead_ctx *aeadctx = AEAD_CTX(a_ctx(tfm));
struct aead_alg *alg = crypto_aead_alg(tfm);
aeadctx->sw_cipher = crypto_alloc_aead(alg->base.cra_name, 0,
@@ -2495,25 +2892,20 @@ static int chcr_aead_cra_init(struct crypto_aead *tfm)
crypto_aead_set_reqsize(tfm, max(sizeof(struct chcr_aead_reqctx),
sizeof(struct aead_request) +
crypto_aead_reqsize(aeadctx->sw_cipher)));
- aeadctx->null = crypto_get_default_null_skcipher();
- if (IS_ERR(aeadctx->null))
- return PTR_ERR(aeadctx->null);
- return chcr_device_init(ctx);
+ return chcr_device_init(a_ctx(tfm));
}
static void chcr_aead_cra_exit(struct crypto_aead *tfm)
{
- struct chcr_context *ctx = crypto_aead_ctx(tfm);
- struct chcr_aead_ctx *aeadctx = AEAD_CTX(ctx);
+ struct chcr_aead_ctx *aeadctx = AEAD_CTX(a_ctx(tfm));
- crypto_put_default_null_skcipher();
crypto_free_aead(aeadctx->sw_cipher);
}
static int chcr_authenc_null_setauthsize(struct crypto_aead *tfm,
unsigned int authsize)
{
- struct chcr_aead_ctx *aeadctx = AEAD_CTX(crypto_aead_ctx(tfm));
+ struct chcr_aead_ctx *aeadctx = AEAD_CTX(a_ctx(tfm));
aeadctx->hmac_ctrl = CHCR_SCMD_HMAC_CTRL_NOP;
aeadctx->mayverify = VERIFY_HW;
@@ -2522,7 +2914,7 @@ static int chcr_authenc_null_setauthsize(struct crypto_aead *tfm,
static int chcr_authenc_setauthsize(struct crypto_aead *tfm,
unsigned int authsize)
{
- struct chcr_aead_ctx *aeadctx = AEAD_CTX(crypto_aead_ctx(tfm));
+ struct chcr_aead_ctx *aeadctx = AEAD_CTX(a_ctx(tfm));
u32 maxauth = crypto_aead_maxauthsize(tfm);
/*SHA1 authsize in ipsec is 12 instead of 10 i.e maxauthsize / 2 is not
@@ -2560,7 +2952,7 @@ static int chcr_authenc_setauthsize(struct crypto_aead *tfm,
static int chcr_gcm_setauthsize(struct crypto_aead *tfm, unsigned int authsize)
{
- struct chcr_aead_ctx *aeadctx = AEAD_CTX(crypto_aead_ctx(tfm));
+ struct chcr_aead_ctx *aeadctx = AEAD_CTX(a_ctx(tfm));
switch (authsize) {
case ICV_4:
@@ -2600,7 +2992,7 @@ static int chcr_gcm_setauthsize(struct crypto_aead *tfm, unsigned int authsize)
static int chcr_4106_4309_setauthsize(struct crypto_aead *tfm,
unsigned int authsize)
{
- struct chcr_aead_ctx *aeadctx = AEAD_CTX(crypto_aead_ctx(tfm));
+ struct chcr_aead_ctx *aeadctx = AEAD_CTX(a_ctx(tfm));
switch (authsize) {
case ICV_8:
@@ -2626,7 +3018,7 @@ static int chcr_4106_4309_setauthsize(struct crypto_aead *tfm,
static int chcr_ccm_setauthsize(struct crypto_aead *tfm,
unsigned int authsize)
{
- struct chcr_aead_ctx *aeadctx = AEAD_CTX(crypto_aead_ctx(tfm));
+ struct chcr_aead_ctx *aeadctx = AEAD_CTX(a_ctx(tfm));
switch (authsize) {
case ICV_4:
@@ -2669,8 +3061,7 @@ static int chcr_ccm_common_setkey(struct crypto_aead *aead,
const u8 *key,
unsigned int keylen)
{
- struct chcr_context *ctx = crypto_aead_ctx(aead);
- struct chcr_aead_ctx *aeadctx = AEAD_CTX(ctx);
+ struct chcr_aead_ctx *aeadctx = AEAD_CTX(a_ctx(aead));
unsigned char ck_size, mk_size;
int key_ctx_size = 0;
@@ -2703,8 +3094,7 @@ static int chcr_aead_ccm_setkey(struct crypto_aead *aead,
const u8 *key,
unsigned int keylen)
{
- struct chcr_context *ctx = crypto_aead_ctx(aead);
- struct chcr_aead_ctx *aeadctx = AEAD_CTX(ctx);
+ struct chcr_aead_ctx *aeadctx = AEAD_CTX(a_ctx(aead));
int error;
crypto_aead_clear_flags(aeadctx->sw_cipher, CRYPTO_TFM_REQ_MASK);
@@ -2722,8 +3112,7 @@ static int chcr_aead_ccm_setkey(struct crypto_aead *aead,
static int chcr_aead_rfc4309_setkey(struct crypto_aead *aead, const u8 *key,
unsigned int keylen)
{
- struct chcr_context *ctx = crypto_aead_ctx(aead);
- struct chcr_aead_ctx *aeadctx = AEAD_CTX(ctx);
+ struct chcr_aead_ctx *aeadctx = AEAD_CTX(a_ctx(aead));
int error;
if (keylen < 3) {
@@ -2749,8 +3138,7 @@ static int chcr_aead_rfc4309_setkey(struct crypto_aead *aead, const u8 *key,
static int chcr_gcm_setkey(struct crypto_aead *aead, const u8 *key,
unsigned int keylen)
{
- struct chcr_context *ctx = crypto_aead_ctx(aead);
- struct chcr_aead_ctx *aeadctx = AEAD_CTX(ctx);
+ struct chcr_aead_ctx *aeadctx = AEAD_CTX(a_ctx(aead));
struct chcr_gcm_ctx *gctx = GCM_CTX(aeadctx);
struct crypto_cipher *cipher;
unsigned int ck_size;
@@ -2822,8 +3210,7 @@ static int chcr_gcm_setkey(struct crypto_aead *aead, const u8 *key,
static int chcr_authenc_setkey(struct crypto_aead *authenc, const u8 *key,
unsigned int keylen)
{
- struct chcr_context *ctx = crypto_aead_ctx(authenc);
- struct chcr_aead_ctx *aeadctx = AEAD_CTX(ctx);
+ struct chcr_aead_ctx *aeadctx = AEAD_CTX(a_ctx(authenc));
struct chcr_authenc_ctx *actx = AUTHENC_CTX(aeadctx);
/* it contains auth and cipher key both*/
struct crypto_authenc_keys keys;
@@ -2943,8 +3330,7 @@ static int chcr_authenc_setkey(struct crypto_aead *authenc, const u8 *key,
static int chcr_aead_digest_null_setkey(struct crypto_aead *authenc,
const u8 *key, unsigned int keylen)
{
- struct chcr_context *ctx = crypto_aead_ctx(authenc);
- struct chcr_aead_ctx *aeadctx = AEAD_CTX(ctx);
+ struct chcr_aead_ctx *aeadctx = AEAD_CTX(a_ctx(authenc));
struct chcr_authenc_ctx *actx = AUTHENC_CTX(aeadctx);
struct crypto_authenc_keys keys;
int err;
@@ -3016,7 +3402,7 @@ static int chcr_aead_encrypt(struct aead_request *req)
static int chcr_aead_decrypt(struct aead_request *req)
{
struct crypto_aead *tfm = crypto_aead_reqtfm(req);
- struct chcr_aead_ctx *aeadctx = AEAD_CTX(crypto_aead_ctx(tfm));
+ struct chcr_aead_ctx *aeadctx = AEAD_CTX(a_ctx(tfm));
struct chcr_aead_reqctx *reqctx = aead_request_ctx(req);
int size;
@@ -3049,30 +3435,29 @@ static int chcr_aead_op(struct aead_request *req,
create_wr_t create_wr_fn)
{
struct crypto_aead *tfm = crypto_aead_reqtfm(req);
- struct chcr_context *ctx = crypto_aead_ctx(tfm);
struct uld_ctx *u_ctx;
struct sk_buff *skb;
- if (!ctx->dev) {
+ if (!a_ctx(tfm)->dev) {
pr_err("chcr : %s : No crypto device.\n", __func__);
return -ENXIO;
}
- u_ctx = ULD_CTX(ctx);
+ u_ctx = ULD_CTX(a_ctx(tfm));
if (cxgb4_is_crypto_q_full(u_ctx->lldi.ports[0],
- ctx->tx_qidx)) {
+ a_ctx(tfm)->tx_qidx)) {
if (!(req->base.flags & CRYPTO_TFM_REQ_MAY_BACKLOG))
return -EBUSY;
}
/* Form a WR from req */
- skb = create_wr_fn(req, u_ctx->lldi.rxq_ids[ctx->rx_qidx], size,
+ skb = create_wr_fn(req, u_ctx->lldi.rxq_ids[a_ctx(tfm)->rx_qidx], size,
op_type);
if (IS_ERR(skb) || !skb)
return PTR_ERR(skb);
skb->dev = u_ctx->lldi.ports[0];
- set_wr_txq(skb, CPL_PRIORITY_DATA, ctx->tx_qidx);
+ set_wr_txq(skb, CPL_PRIORITY_DATA, a_ctx(tfm)->tx_qidx);
chcr_send_wr(skb);
return -EINPROGRESS;
}
diff --git a/drivers/crypto/chelsio/chcr_algo.h b/drivers/crypto/chelsio/chcr_algo.h
index d1a787b..96c9335 100644
--- a/drivers/crypto/chelsio/chcr_algo.h
+++ b/drivers/crypto/chelsio/chcr_algo.h
@@ -214,27 +214,22 @@
calc_tx_flits_ofld(skb) * 8), 16)))
#define FILL_CMD_MORE(immdatalen) htonl(ULPTX_CMD_V(ULP_TX_SC_IMM) |\
- ULP_TX_SC_MORE_V((immdatalen) ? 0 : 1))
-
+ ULP_TX_SC_MORE_V((immdatalen)))
#define MAX_NK 8
-#define CRYPTO_MAX_IMM_TX_PKT_LEN 256
-#define MAX_WR_SIZE 512
#define ROUND_16(bytes) ((bytes) & 0xFFFFFFF0)
#define MAX_DSGL_ENT 32
-#define MAX_DIGEST_SKB_SGE (MAX_SKB_FRAGS - 1)
#define MIN_CIPHER_SG 1 /* IV */
-#define MIN_AUTH_SG 2 /*IV + AAD*/
-#define MIN_GCM_SG 2 /* IV + AAD*/
+#define MIN_AUTH_SG 1 /* IV */
+#define MIN_GCM_SG 1 /* IV */
#define MIN_DIGEST_SG 1 /*Partial Buffer*/
-#define MIN_CCM_SG 3 /*IV+AAD+B0*/
+#define MIN_CCM_SG 2 /*IV+B0*/
#define SPACE_LEFT(len) \
- ((MAX_WR_SIZE - WR_MIN_LEN - (len)))
+ ((SGE_MAX_WR_LEN - WR_MIN_LEN - (len)))
-unsigned int sgl_ent_len[] = {0, 0, 16, 24, 40,
- 48, 64, 72, 88,
- 96, 112, 120, 136,
- 144, 160, 168, 184,
- 192};
+unsigned int sgl_ent_len[] = {0, 0, 16, 24, 40, 48, 64, 72, 88,
+ 96, 112, 120, 136, 144, 160, 168, 184,
+ 192, 208, 216, 232, 240, 256, 264, 280,
+ 288, 304, 312, 328, 336, 352, 360, 376};
unsigned int dsgl_ent_len[] = {0, 32, 32, 48, 48, 64, 64, 80, 80,
112, 112, 128, 128, 144, 144, 160, 160,
192, 192, 208, 208, 224, 224, 240, 240,
@@ -258,7 +253,6 @@ struct hash_wr_param {
struct cipher_wr_param {
struct ablkcipher_request *req;
- struct scatterlist *srcsg;
char *iv;
int bytes;
unsigned short qid;
@@ -298,31 +292,11 @@ enum {
ICV_16 = 16
};
-struct hash_op_params {
- unsigned char mk_size;
- unsigned char pad_align;
- unsigned char auth_mode;
- char hash_name[MAX_HASH_NAME];
- unsigned short block_size;
- unsigned short word_size;
- unsigned short ipad_size;
-};
-
struct phys_sge_pairs {
__be16 len[8];
__be64 addr[8];
};
-struct phys_sge_parm {
- unsigned int nents;
- unsigned int obsize;
- unsigned short qid;
-};
-
-struct crypto_result {
- struct completion completion;
- int err;
-};
static const u32 sha1_init[SHA1_DIGEST_SIZE / 4] = {
SHA1_H0, SHA1_H1, SHA1_H2, SHA1_H3, SHA1_H4,
diff --git a/drivers/crypto/chelsio/chcr_crypto.h b/drivers/crypto/chelsio/chcr_crypto.h
index 8f436f8..94a87e3 100644
--- a/drivers/crypto/chelsio/chcr_crypto.h
+++ b/drivers/crypto/chelsio/chcr_crypto.h
@@ -149,9 +149,23 @@
#define CHCR_HASH_MAX_BLOCK_SIZE_64 64
#define CHCR_HASH_MAX_BLOCK_SIZE_128 128
-#define CHCR_SG_SIZE 2048
+#define CHCR_SRC_SG_SIZE (0x10000 - sizeof(int))
+#define CHCR_DST_SG_SIZE 2048
-/* Aligned to 128 bit boundary */
+static inline struct chcr_context *a_ctx(struct crypto_aead *tfm)
+{
+ return crypto_aead_ctx(tfm);
+}
+
+static inline struct chcr_context *c_ctx(struct crypto_ablkcipher *tfm)
+{
+ return crypto_ablkcipher_ctx(tfm);
+}
+
+static inline struct chcr_context *h_ctx(struct crypto_ahash *tfm)
+{
+ return crypto_tfm_ctx(crypto_ahash_tfm(tfm));
+}
struct ablk_ctx {
struct crypto_skcipher *sw_cipher;
@@ -165,15 +179,39 @@ struct ablk_ctx {
};
struct chcr_aead_reqctx {
struct sk_buff *skb;
- struct scatterlist *dst;
- struct scatterlist srcffwd[2];
- struct scatterlist dstffwd[2];
+ dma_addr_t iv_dma;
+ dma_addr_t b0_dma;
+ unsigned int b0_len;
+ unsigned int op;
+ short int aad_nents;
+ short int src_nents;
short int dst_nents;
+ u16 imm;
u16 verify;
u8 iv[CHCR_MAX_CRYPTO_IV_LEN];
unsigned char scratch_pad[MAX_SCRATCH_PAD_SIZE];
};
+struct ulptx_walk {
+ struct ulptx_sgl *sgl;
+ unsigned int nents;
+ unsigned int pair_idx;
+ unsigned int last_sg_len;
+ struct scatterlist *last_sg;
+ struct ulptx_sge_pair *pair;
+
+};
+
+struct dsgl_walk {
+ unsigned int nents;
+ unsigned int last_sg_len;
+ struct scatterlist *last_sg;
+ struct cpl_rx_phys_dsgl *dsgl;
+ struct phys_sge_pairs *to;
+};
+
+
+
struct chcr_gcm_ctx {
u8 ghash_h[AEAD_H_SIZE];
};
@@ -194,7 +232,6 @@ struct __aead_ctx {
struct chcr_aead_ctx {
__be32 key_ctx_hdr;
unsigned int enckey_len;
- struct crypto_skcipher *null;
struct crypto_aead *sw_cipher;
u8 salt[MAX_SALT];
u8 key[CHCR_AES_MAX_KEY_LEN];
@@ -230,8 +267,11 @@ struct chcr_ahash_req_ctx {
u8 bfr2[CHCR_HASH_MAX_BLOCK_SIZE_128];
u8 *reqbfr;
u8 *skbfr;
+ dma_addr_t dma_addr;
+ u32 dma_len;
u8 reqlen;
- /* DMA the partial hash in it */
+ u8 imm;
+ u8 is_sg_map;
u8 partial_hash[CHCR_HASH_MAX_DIGEST_SIZE];
u64 data_len; /* Data len till time */
/* SKB which is being sent to the hardware for processing */
@@ -240,14 +280,15 @@ struct chcr_ahash_req_ctx {
struct chcr_blkcipher_req_ctx {
struct sk_buff *skb;
- struct scatterlist srcffwd[2];
- struct scatterlist dstffwd[2];
struct scatterlist *dstsg;
- struct scatterlist *dst;
unsigned int processed;
unsigned int last_req_len;
+ struct scatterlist *srcsg;
+ unsigned int src_ofst;
+ unsigned int dst_ofst;
unsigned int op;
- short int dst_nents;
+ dma_addr_t iv_dma;
+ u16 imm;
u8 iv[CHCR_MAX_CRYPTO_IV_LEN];
};
@@ -261,24 +302,6 @@ struct chcr_alg_template {
} alg;
};
-struct chcr_req_ctx {
- union {
- struct ahash_request *ahash_req;
- struct aead_request *aead_req;
- struct ablkcipher_request *ablk_req;
- } req;
- union {
- struct chcr_ahash_req_ctx *ahash_ctx;
- struct chcr_aead_reqctx *reqctx;
- struct chcr_blkcipher_req_ctx *ablk_ctx;
- } ctx;
-};
-
-struct sge_opaque_hdr {
- void *dev;
- dma_addr_t addr[MAX_SKB_FRAGS + 1];
-};
-
typedef struct sk_buff *(*create_wr_t)(struct aead_request *req,
unsigned short qid,
int size,
@@ -291,4 +314,37 @@ static int chcr_aead_op(struct aead_request *req_base,
static inline int get_aead_subtype(struct crypto_aead *aead);
static int chcr_handle_cipher_resp(struct ablkcipher_request *req,
unsigned char *input, int err);
+static void chcr_verify_tag(struct aead_request *req, u8 *input, int *err);
+static int chcr_aead_dma_map(struct device *dev, struct aead_request *req,
+ unsigned short op_type);
+static void chcr_aead_dma_unmap(struct device *dev, struct aead_request
+ *req, unsigned short op_type);
+static inline void chcr_add_aead_dst_ent(struct aead_request *req,
+ struct cpl_rx_phys_dsgl *phys_cpl,
+ unsigned int assoclen,
+ unsigned short op_type,
+ unsigned short qid);
+static inline void chcr_add_aead_src_ent(struct aead_request *req,
+ struct ulptx_sgl *ulptx,
+ unsigned int assoclen,
+ unsigned short op_type);
+static inline void chcr_add_cipher_src_ent(struct ablkcipher_request *req,
+ struct ulptx_sgl *ulptx,
+ struct cipher_wr_param *wrparam);
+static int chcr_cipher_dma_map(struct device *dev,
+ struct ablkcipher_request *req);
+static void chcr_cipher_dma_unmap(struct device *dev,
+ struct ablkcipher_request *req);
+static inline void chcr_add_cipher_dst_ent(struct ablkcipher_request *req,
+ struct cpl_rx_phys_dsgl *phys_cpl,
+ struct cipher_wr_param *wrparam,
+ unsigned short qid);
+int sg_nents_len_skip(struct scatterlist *sg, u64 len, u64 skip);
+static inline void chcr_add_hash_src_ent(struct ahash_request *req,
+ struct ulptx_sgl *ulptx,
+ struct hash_wr_param *param);
+static inline int chcr_hash_dma_map(struct device *dev,
+ struct ahash_request *req);
+static inline void chcr_hash_dma_unmap(struct device *dev,
+ struct ahash_request *req);
#endif /* __CHCR_CRYPTO_H__ */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/sge.c b/drivers/net/ethernet/chelsio/cxgb4/sge.c
index ede1220..1c1cc48 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/sge.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/sge.c
@@ -1537,7 +1537,13 @@ int t4_mgmt_tx(struct adapter *adap, struct sk_buff *skb)
*/
static inline int is_ofld_imm(const struct sk_buff *skb)
{
- return skb->len <= MAX_IMM_TX_PKT_LEN;
+ struct work_request_hdr *req = (struct work_request_hdr *)skb->data;
+ unsigned long opcode = FW_WR_OP_G(ntohl(req->wr_hi));
+
+ if (opcode == FW_CRYPTO_LOOKASIDE_WR)
+ return skb->len <= SGE_MAX_WR_LEN;
+ else
+ return skb->len <= MAX_IMM_TX_PKT_LEN;
}
/**
--
2.1.4
^ permalink raw reply related
* [PATCH 7/7] crypto:chelsio: Fix memory leak
From: Harsh Jain @ 2017-10-03 6:46 UTC (permalink / raw)
To: herbert, linux-crypto, netdev; +Cc: Harsh Jain
In-Reply-To: <cover.1507010612.git.harsh@chelsio.com>
Fix memory leak when device does not support crypto.
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Harsh Jain <harsh@chelsio.com>
---
drivers/crypto/chelsio/chcr_core.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/drivers/crypto/chelsio/chcr_core.c b/drivers/crypto/chelsio/chcr_core.c
index b6dd9cb..4f677b3 100644
--- a/drivers/crypto/chelsio/chcr_core.c
+++ b/drivers/crypto/chelsio/chcr_core.c
@@ -154,15 +154,15 @@ static void *chcr_uld_add(const struct cxgb4_lld_info *lld)
struct uld_ctx *u_ctx;
/* Create the device and add it in the device list */
+ if (!(lld->ulp_crypto & ULP_CRYPTO_LOOKASIDE))
+ return ERR_PTR(-EOPNOTSUPP);
+
+ /* Create the device and add it in the device list */
u_ctx = kzalloc(sizeof(*u_ctx), GFP_KERNEL);
if (!u_ctx) {
u_ctx = ERR_PTR(-ENOMEM);
goto out;
}
- if (!(lld->ulp_crypto & ULP_CRYPTO_LOOKASIDE)) {
- u_ctx = ERR_PTR(-ENOMEM);
- goto out;
- }
u_ctx->lldi = *lld;
out:
return u_ctx;
--
2.1.4
^ permalink raw reply related
* [PATCH 5/7] crypto:chelsio:Remove allocation of sg list to implement 2K limit of dsgl header
From: Harsh Jain @ 2017-10-03 6:46 UTC (permalink / raw)
To: herbert, linux-crypto, netdev; +Cc: Harsh Jain
In-Reply-To: <cover.1507010612.git.harsh@chelsio.com>
Update DMA address index instead of allocating new sg list to impose 2k size limit for each entry.
Signed-off-by: Harsh Jain <harsh@chelsio.com>
---
drivers/crypto/chelsio/chcr_algo.c | 237 +++++++++++------------------------
drivers/crypto/chelsio/chcr_algo.h | 3 +-
drivers/crypto/chelsio/chcr_core.h | 2 +-
drivers/crypto/chelsio/chcr_crypto.h | 6 -
4 files changed, 76 insertions(+), 172 deletions(-)
diff --git a/drivers/crypto/chelsio/chcr_algo.c b/drivers/crypto/chelsio/chcr_algo.c
index e0ab34a..b13991d 100644
--- a/drivers/crypto/chelsio/chcr_algo.c
+++ b/drivers/crypto/chelsio/chcr_algo.c
@@ -117,6 +117,21 @@ static inline unsigned int sgl_len(unsigned int n)
return (3 * n) / 2 + (n & 1) + 2;
}
+static int dstsg_2k(struct scatterlist *sgl, unsigned int reqlen)
+{
+ int nents = 0;
+ unsigned int less;
+
+ while (sgl && reqlen) {
+ less = min(reqlen, sgl->length);
+ nents += DIV_ROUND_UP(less, CHCR_SG_SIZE);
+ reqlen -= less;
+ sgl = sg_next(sgl);
+ }
+
+ return nents;
+}
+
static void chcr_verify_tag(struct aead_request *req, u8 *input, int *err)
{
u8 temp[SHA512_DIGEST_SIZE];
@@ -166,8 +181,6 @@ int chcr_handle_resp(struct crypto_async_request *req, unsigned char *input,
kfree_skb(ctx_req.ctx.reqctx->skb);
ctx_req.ctx.reqctx->skb = NULL;
}
- free_new_sg(ctx_req.ctx.reqctx->newdstsg);
- ctx_req.ctx.reqctx->newdstsg = NULL;
if (ctx_req.ctx.reqctx->verify == VERIFY_SW) {
chcr_verify_tag(ctx_req.req.aead_req, input,
&err);
@@ -388,31 +401,41 @@ static void write_phys_cpl(struct cpl_rx_phys_dsgl *phys_cpl,
{
struct phys_sge_pairs *to;
unsigned int len = 0, left_size = sg_param->obsize;
- unsigned int nents = sg_param->nents, i, j = 0;
+ unsigned int j = 0;
+ int offset, ent_len;
phys_cpl->op_to_tid = htonl(CPL_RX_PHYS_DSGL_OPCODE_V(CPL_RX_PHYS_DSGL)
| CPL_RX_PHYS_DSGL_ISRDMA_V(0));
+ to = (struct phys_sge_pairs *)((unsigned char *)phys_cpl +
+ sizeof(struct cpl_rx_phys_dsgl));
+ while (left_size && sg) {
+ len = min_t(u32, left_size, sg_dma_len(sg));
+ offset = 0;
+ while (len) {
+ ent_len = min_t(u32, len, CHCR_SG_SIZE);
+ to->len[j % 8] = htons(ent_len);
+ to->addr[j % 8] = cpu_to_be64(sg_dma_address(sg) +
+ offset);
+ offset += ent_len;
+ len -= ent_len;
+ j++;
+ if ((j % 8) == 0)
+ to++;
+ }
+ left_size -= min(left_size, sg_dma_len(sg));
+ sg = sg_next(sg);
+ }
phys_cpl->pcirlxorder_to_noofsgentr =
htonl(CPL_RX_PHYS_DSGL_PCIRLXORDER_V(0) |
CPL_RX_PHYS_DSGL_PCINOSNOOP_V(0) |
CPL_RX_PHYS_DSGL_PCITPHNTENB_V(0) |
CPL_RX_PHYS_DSGL_PCITPHNT_V(0) |
CPL_RX_PHYS_DSGL_DCAID_V(0) |
- CPL_RX_PHYS_DSGL_NOOFSGENTR_V(nents));
+ CPL_RX_PHYS_DSGL_NOOFSGENTR_V(j));
phys_cpl->rss_hdr_int.opcode = CPL_RX_PHYS_ADDR;
phys_cpl->rss_hdr_int.qid = htons(sg_param->qid);
phys_cpl->rss_hdr_int.hash_val = 0;
- to = (struct phys_sge_pairs *)((unsigned char *)phys_cpl +
- sizeof(struct cpl_rx_phys_dsgl));
- for (i = 0; nents && left_size; to++) {
- for (j = 0; j < 8 && nents && left_size; j++, nents--) {
- len = min(left_size, sg_dma_len(sg));
- to->len[j] = htons(len);
- to->addr[j] = cpu_to_be64(sg_dma_address(sg));
- left_size -= len;
- sg = sg_next(sg);
- }
- }
+
}
static inline int map_writesg_phys_cpl(struct device *dev,
@@ -523,31 +546,33 @@ static int generate_copy_rrkey(struct ablk_ctx *ablkctx,
static int chcr_sg_ent_in_wr(struct scatterlist *src,
struct scatterlist *dst,
unsigned int minsg,
- unsigned int space,
- short int *sent,
- short int *dent)
+ unsigned int space)
{
int srclen = 0, dstlen = 0;
int srcsg = minsg, dstsg = 0;
+ int offset = 0, less;
- *sent = 0;
- *dent = 0;
while (src && dst && ((srcsg + 1) <= MAX_SKB_FRAGS) &&
space > (sgl_ent_len[srcsg + 1] + dsgl_ent_len[dstsg])) {
srclen += src->length;
srcsg++;
+ offset = 0;
while (dst && ((dstsg + 1) <= MAX_DSGL_ENT) &&
space > (sgl_ent_len[srcsg] + dsgl_ent_len[dstsg + 1])) {
if (srclen <= dstlen)
break;
- dstlen += dst->length;
- dst = sg_next(dst);
+ less = min_t(unsigned int, dst->length - offset,
+ CHCR_SG_SIZE);
+ dstlen += less;
+ offset += less;
+ if (offset == dst->length) {
+ dst = sg_next(dst);
+ offset = 0;
+ }
dstsg++;
}
src = sg_next(src);
}
- *sent = srcsg - minsg;
- *dent = dstsg;
return min(srclen, dstlen);
}
@@ -631,13 +656,15 @@ static struct sk_buff *create_cipher_wr(struct cipher_wr_param *wrparam)
struct phys_sge_parm sg_param;
unsigned int frags = 0, transhdr_len, phys_dsgl;
int error;
+ int nents;
unsigned int ivsize = AES_BLOCK_SIZE, kctx_len;
gfp_t flags = wrparam->req->base.flags & CRYPTO_TFM_REQ_MAY_SLEEP ?
GFP_KERNEL : GFP_ATOMIC;
struct adapter *adap = padap(ctx->dev);
- phys_dsgl = get_space_for_phys_dsgl(reqctx->dst_nents);
-
+ reqctx->dst_nents = sg_nents_for_len(reqctx->dst, wrparam->bytes);
+ nents = dstsg_2k(reqctx->dst, wrparam->bytes);
+ phys_dsgl = get_space_for_phys_dsgl(nents);
kctx_len = (DIV_ROUND_UP(ablkctx->enckey_len, 16) * 16);
transhdr_len = CIPHER_TRANSHDR_SIZE(kctx_len, phys_dsgl);
skb = alloc_skb((transhdr_len + sizeof(struct sge_opaque_hdr)), flags);
@@ -1020,8 +1047,7 @@ static int chcr_handle_cipher_resp(struct ablkcipher_request *req,
goto complete;
}
bytes = chcr_sg_ent_in_wr(wrparam.srcsg, reqctx->dst, 1,
- SPACE_LEFT(ablkctx->enckey_len),
- &wrparam.snent, &reqctx->dst_nents);
+ SPACE_LEFT(ablkctx->enckey_len));
if ((bytes + reqctx->processed) >= req->nbytes)
bytes = req->nbytes - reqctx->processed;
else
@@ -1060,8 +1086,6 @@ static int chcr_handle_cipher_resp(struct ablkcipher_request *req,
chcr_send_wr(skb);
return 0;
complete:
- free_new_sg(reqctx->newdstsg);
- reqctx->newdstsg = NULL;
req->base.complete(&req->base, err);
return err;
}
@@ -1077,9 +1101,8 @@ static int process_cipher(struct ablkcipher_request *req,
struct chcr_context *ctx = crypto_ablkcipher_ctx(tfm);
struct ablk_ctx *ablkctx = ABLK_CTX(ctx);
struct cipher_wr_param wrparam;
- int bytes, nents, err = -EINVAL;
+ int bytes, err = -EINVAL;
- reqctx->newdstsg = NULL;
reqctx->processed = 0;
if (!req->info)
goto error;
@@ -1091,18 +1114,9 @@ static int process_cipher(struct ablkcipher_request *req,
goto error;
}
wrparam.srcsg = req->src;
- if (is_newsg(req->dst, &nents)) {
- reqctx->newdstsg = alloc_new_sg(req->dst, nents);
- if (IS_ERR(reqctx->newdstsg))
- return PTR_ERR(reqctx->newdstsg);
- reqctx->dstsg = reqctx->newdstsg;
- } else {
reqctx->dstsg = req->dst;
- }
bytes = chcr_sg_ent_in_wr(wrparam.srcsg, reqctx->dstsg, MIN_CIPHER_SG,
- SPACE_LEFT(ablkctx->enckey_len),
- &wrparam.snent,
- &reqctx->dst_nents);
+ SPACE_LEFT(ablkctx->enckey_len));
if ((bytes + reqctx->processed) >= req->nbytes)
bytes = req->nbytes - reqctx->processed;
else
@@ -1152,8 +1166,6 @@ static int process_cipher(struct ablkcipher_request *req,
return 0;
error:
- free_new_sg(reqctx->newdstsg);
- reqctx->newdstsg = NULL;
return err;
}
@@ -1824,63 +1836,6 @@ static void chcr_hmac_cra_exit(struct crypto_tfm *tfm)
}
}
-static int is_newsg(struct scatterlist *sgl, unsigned int *newents)
-{
- int nents = 0;
- int ret = 0;
-
- while (sgl) {
- if (sgl->length > CHCR_SG_SIZE)
- ret = 1;
- nents += DIV_ROUND_UP(sgl->length, CHCR_SG_SIZE);
- sgl = sg_next(sgl);
- }
- *newents = nents;
- return ret;
-}
-
-static inline void free_new_sg(struct scatterlist *sgl)
-{
- kfree(sgl);
-}
-
-static struct scatterlist *alloc_new_sg(struct scatterlist *sgl,
- unsigned int nents)
-{
- struct scatterlist *newsg, *sg;
- int i, len, processed = 0;
- struct page *spage;
- int offset;
-
- newsg = kmalloc_array(nents, sizeof(struct scatterlist), GFP_KERNEL);
- if (!newsg)
- return ERR_PTR(-ENOMEM);
- sg = newsg;
- sg_init_table(sg, nents);
- offset = sgl->offset;
- spage = sg_page(sgl);
- for (i = 0; i < nents; i++) {
- len = min_t(u32, sgl->length - processed, CHCR_SG_SIZE);
- sg_set_page(sg, spage, len, offset);
- processed += len;
- offset += len;
- if (offset >= PAGE_SIZE) {
- offset = offset % PAGE_SIZE;
- spage++;
- }
- if (processed == sgl->length) {
- processed = 0;
- sgl = sg_next(sgl);
- if (!sgl)
- break;
- spage = sg_page(sgl);
- offset = sgl->offset;
- }
- sg = sg_next(sg);
- }
- return newsg;
-}
-
static int chcr_copy_assoc(struct aead_request *req,
struct chcr_aead_ctx *ctx)
{
@@ -1953,7 +1908,6 @@ static struct sk_buff *create_authenc_wr(struct aead_request *req,
GFP_ATOMIC;
struct adapter *adap = padap(ctx->dev);
- reqctx->newdstsg = NULL;
dst_size = req->assoclen + req->cryptlen + (op_type ? -authsize :
authsize);
if (aeadctx->enckey_len == 0 || (req->cryptlen <= 0))
@@ -1965,24 +1919,13 @@ static struct sk_buff *create_authenc_wr(struct aead_request *req,
if (src_nent < 0)
goto err;
src = scatterwalk_ffwd(reqctx->srcffwd, req->src, req->assoclen);
-
+ reqctx->dst = src;
if (req->src != req->dst) {
error = chcr_copy_assoc(req, aeadctx);
if (error)
return ERR_PTR(error);
- }
- if (dst_size && is_newsg(req->dst, &nents)) {
- reqctx->newdstsg = alloc_new_sg(req->dst, nents);
- if (IS_ERR(reqctx->newdstsg))
- return ERR_CAST(reqctx->newdstsg);
- reqctx->dst = scatterwalk_ffwd(reqctx->dstffwd,
- reqctx->newdstsg, req->assoclen);
- } else {
- if (req->src == req->dst)
- reqctx->dst = src;
- else
- reqctx->dst = scatterwalk_ffwd(reqctx->dstffwd,
- req->dst, req->assoclen);
+ reqctx->dst = scatterwalk_ffwd(reqctx->dstffwd, req->dst,
+ req->assoclen);
}
if (get_aead_subtype(tfm) == CRYPTO_ALG_SUB_TYPE_AEAD_NULL) {
null = 1;
@@ -1995,7 +1938,9 @@ static struct sk_buff *create_authenc_wr(struct aead_request *req,
error = -EINVAL;
goto err;
}
- dst_size = get_space_for_phys_dsgl(reqctx->dst_nents);
+ nents = dst_size ? dstsg_2k(reqctx->dst, req->cryptlen +
+ (op_type ? -authsize : authsize)) : 0;
+ dst_size = get_space_for_phys_dsgl(nents);
kctx_len = (ntohl(KEY_CONTEXT_CTX_LEN_V(aeadctx->key_ctx_hdr)) << 4)
- sizeof(chcr_req->key_ctx);
transhdr_len = CIPHER_TRANSHDR_SIZE(kctx_len, dst_size);
@@ -2004,8 +1949,6 @@ static struct sk_buff *create_authenc_wr(struct aead_request *req,
transhdr_len + (sgl_len(src_nent + MIN_AUTH_SG) * 8),
op_type)) {
atomic_inc(&adap->chcr_stats.fallback);
- free_new_sg(reqctx->newdstsg);
- reqctx->newdstsg = NULL;
return ERR_PTR(chcr_aead_fallback(req, op_type));
}
skb = alloc_skb((transhdr_len + sizeof(struct sge_opaque_hdr)), flags);
@@ -2088,8 +2031,6 @@ static struct sk_buff *create_authenc_wr(struct aead_request *req,
/* ivmap_fail: */
kfree_skb(skb);
err:
- free_new_sg(reqctx->newdstsg);
- reqctx->newdstsg = NULL;
return ERR_PTR(error);
}
@@ -2307,7 +2248,6 @@ static struct sk_buff *create_aead_ccm_wr(struct aead_request *req,
dst_size = req->assoclen + req->cryptlen + (op_type ? -authsize :
authsize);
- reqctx->newdstsg = NULL;
if (op_type && req->cryptlen < crypto_aead_authsize(tfm))
goto err;
src_nent = sg_nents_for_len(req->src, req->assoclen + req->cryptlen);
@@ -2316,25 +2256,15 @@ static struct sk_buff *create_aead_ccm_wr(struct aead_request *req,
sub_type = get_aead_subtype(tfm);
src = scatterwalk_ffwd(reqctx->srcffwd, req->src, req->assoclen);
+ reqctx->dst = src;
if (req->src != req->dst) {
error = chcr_copy_assoc(req, aeadctx);
if (error) {
pr_err("AAD copy to destination buffer fails\n");
return ERR_PTR(error);
}
- }
- if (dst_size && is_newsg(req->dst, &nents)) {
- reqctx->newdstsg = alloc_new_sg(req->dst, nents);
- if (IS_ERR(reqctx->newdstsg))
- return ERR_CAST(reqctx->newdstsg);
- reqctx->dst = scatterwalk_ffwd(reqctx->dstffwd,
- reqctx->newdstsg, req->assoclen);
- } else {
- if (req->src == req->dst)
- reqctx->dst = src;
- else
- reqctx->dst = scatterwalk_ffwd(reqctx->dstffwd,
- req->dst, req->assoclen);
+ reqctx->dst = scatterwalk_ffwd(reqctx->dstffwd, req->dst,
+ req->assoclen);
}
reqctx->dst_nents = sg_nents_for_len(reqctx->dst, req->cryptlen +
(op_type ? -authsize : authsize));
@@ -2346,8 +2276,9 @@ static struct sk_buff *create_aead_ccm_wr(struct aead_request *req,
error = aead_ccm_validate_input(op_type, req, aeadctx, sub_type);
if (error)
goto err;
-
- dst_size = get_space_for_phys_dsgl(reqctx->dst_nents);
+ nents = dst_size ? dstsg_2k(reqctx->dst, req->cryptlen +
+ (op_type ? -authsize : authsize)) : 0;
+ dst_size = get_space_for_phys_dsgl(nents);
kctx_len = ((DIV_ROUND_UP(aeadctx->enckey_len, 16)) << 4) * 2;
transhdr_len = CIPHER_TRANSHDR_SIZE(kctx_len, dst_size);
if (chcr_aead_need_fallback(req, src_nent + MIN_CCM_SG,
@@ -2355,8 +2286,6 @@ static struct sk_buff *create_aead_ccm_wr(struct aead_request *req,
transhdr_len + (sgl_len(src_nent + MIN_CCM_SG) * 8),
op_type)) {
atomic_inc(&adap->chcr_stats.fallback);
- free_new_sg(reqctx->newdstsg);
- reqctx->newdstsg = NULL;
return ERR_PTR(chcr_aead_fallback(req, op_type));
}
@@ -2402,8 +2331,6 @@ static struct sk_buff *create_aead_ccm_wr(struct aead_request *req,
dstmap_fail:
kfree_skb(skb);
err:
- free_new_sg(reqctx->newdstsg);
- reqctx->newdstsg = NULL;
return ERR_PTR(error);
}
@@ -2432,7 +2359,6 @@ static struct sk_buff *create_gcm_wr(struct aead_request *req,
GFP_ATOMIC;
struct adapter *adap = padap(ctx->dev);
- reqctx->newdstsg = NULL;
dst_size = assoclen + req->cryptlen + (op_type ? -authsize :
authsize);
/* validate key size */
@@ -2446,26 +2372,14 @@ static struct sk_buff *create_gcm_wr(struct aead_request *req,
goto err;
src = scatterwalk_ffwd(reqctx->srcffwd, req->src, assoclen);
+ reqctx->dst = src;
if (req->src != req->dst) {
error = chcr_copy_assoc(req, aeadctx);
if (error)
return ERR_PTR(error);
+ reqctx->dst = scatterwalk_ffwd(reqctx->dstffwd, req->dst,
+ req->assoclen);
}
-
- if (dst_size && is_newsg(req->dst, &nents)) {
- reqctx->newdstsg = alloc_new_sg(req->dst, nents);
- if (IS_ERR(reqctx->newdstsg))
- return ERR_CAST(reqctx->newdstsg);
- reqctx->dst = scatterwalk_ffwd(reqctx->dstffwd,
- reqctx->newdstsg, assoclen);
- } else {
- if (req->src == req->dst)
- reqctx->dst = src;
- else
- reqctx->dst = scatterwalk_ffwd(reqctx->dstffwd,
- req->dst, assoclen);
- }
-
reqctx->dst_nents = sg_nents_for_len(reqctx->dst, req->cryptlen +
(op_type ? -authsize : authsize));
if (reqctx->dst_nents < 0) {
@@ -2474,8 +2388,9 @@ static struct sk_buff *create_gcm_wr(struct aead_request *req,
goto err;
}
-
- dst_size = get_space_for_phys_dsgl(reqctx->dst_nents);
+ nents = dst_size ? dstsg_2k(reqctx->dst, req->cryptlen +
+ (op_type ? -authsize : authsize)) : 0;
+ dst_size = get_space_for_phys_dsgl(nents);
kctx_len = ((DIV_ROUND_UP(aeadctx->enckey_len, 16)) << 4) +
AEAD_H_SIZE;
transhdr_len = CIPHER_TRANSHDR_SIZE(kctx_len, dst_size);
@@ -2484,8 +2399,6 @@ static struct sk_buff *create_gcm_wr(struct aead_request *req,
transhdr_len + (sgl_len(src_nent + MIN_GCM_SG) * 8),
op_type)) {
atomic_inc(&adap->chcr_stats.fallback);
- free_new_sg(reqctx->newdstsg);
- reqctx->newdstsg = NULL;
return ERR_PTR(chcr_aead_fallback(req, op_type));
}
skb = alloc_skb((transhdr_len + sizeof(struct sge_opaque_hdr)), flags);
@@ -2563,8 +2476,6 @@ static struct sk_buff *create_gcm_wr(struct aead_request *req,
/* ivmap_fail: */
kfree_skb(skb);
err:
- free_new_sg(reqctx->newdstsg);
- reqctx->newdstsg = NULL;
return ERR_PTR(error);
}
diff --git a/drivers/crypto/chelsio/chcr_algo.h b/drivers/crypto/chelsio/chcr_algo.h
index 54851ec..d1a787b 100644
--- a/drivers/crypto/chelsio/chcr_algo.h
+++ b/drivers/crypto/chelsio/chcr_algo.h
@@ -221,7 +221,7 @@
#define MAX_WR_SIZE 512
#define ROUND_16(bytes) ((bytes) & 0xFFFFFFF0)
#define MAX_DSGL_ENT 32
-#define MAX_DIGEST_SKB_SGE (MAX_SKB_FRAGS - 2)
+#define MAX_DIGEST_SKB_SGE (MAX_SKB_FRAGS - 1)
#define MIN_CIPHER_SG 1 /* IV */
#define MIN_AUTH_SG 2 /*IV + AAD*/
#define MIN_GCM_SG 2 /* IV + AAD*/
@@ -261,7 +261,6 @@ struct cipher_wr_param {
struct scatterlist *srcsg;
char *iv;
int bytes;
- short int snent;
unsigned short qid;
};
enum {
diff --git a/drivers/crypto/chelsio/chcr_core.h b/drivers/crypto/chelsio/chcr_core.h
index c9a19b2..94e7412 100644
--- a/drivers/crypto/chelsio/chcr_core.h
+++ b/drivers/crypto/chelsio/chcr_core.h
@@ -89,7 +89,7 @@ struct uld_ctx {
struct chcr_dev *dev;
};
-struct uld_ctx * assign_chcr_device(void);
+struct uld_ctx *assign_chcr_device(void);
int chcr_send_wr(struct sk_buff *skb);
int start_crypto(void);
int stop_crypto(void);
diff --git a/drivers/crypto/chelsio/chcr_crypto.h b/drivers/crypto/chelsio/chcr_crypto.h
index b3722b3..8f436f8 100644
--- a/drivers/crypto/chelsio/chcr_crypto.h
+++ b/drivers/crypto/chelsio/chcr_crypto.h
@@ -166,7 +166,6 @@ struct ablk_ctx {
struct chcr_aead_reqctx {
struct sk_buff *skb;
struct scatterlist *dst;
- struct scatterlist *newdstsg;
struct scatterlist srcffwd[2];
struct scatterlist dstffwd[2];
short int dst_nents;
@@ -245,7 +244,6 @@ struct chcr_blkcipher_req_ctx {
struct scatterlist dstffwd[2];
struct scatterlist *dstsg;
struct scatterlist *dst;
- struct scatterlist *newdstsg;
unsigned int processed;
unsigned int last_req_len;
unsigned int op;
@@ -291,10 +289,6 @@ static int chcr_aead_op(struct aead_request *req_base,
int size,
create_wr_t create_wr_fn);
static inline int get_aead_subtype(struct crypto_aead *aead);
-static int is_newsg(struct scatterlist *sgl, unsigned int *newents);
-static struct scatterlist *alloc_new_sg(struct scatterlist *sgl,
- unsigned int nents);
-static inline void free_new_sg(struct scatterlist *sgl);
static int chcr_handle_cipher_resp(struct ablkcipher_request *req,
unsigned char *input, int err);
#endif /* __CHCR_CRYPTO_H__ */
--
2.1.4
^ permalink raw reply related
* [PATCH 4/7] crypto:chelsio:Use x8_ble gf multiplication to calculate IV.
From: Harsh Jain @ 2017-10-03 6:46 UTC (permalink / raw)
To: herbert, linux-crypto, netdev; +Cc: Harsh Jain
In-Reply-To: <cover.1507010612.git.harsh@chelsio.com>
gf128mul_x8_ble() will reduce gf Multiplication iteration by 8.
Signed-off-by: Harsh Jain <harsh@chelsio.com>
---
drivers/crypto/chelsio/chcr_algo.c | 11 +++++++++--
drivers/crypto/chelsio/chcr_crypto.h | 1 +
2 files changed, 10 insertions(+), 2 deletions(-)
diff --git a/drivers/crypto/chelsio/chcr_algo.c b/drivers/crypto/chelsio/chcr_algo.c
index e4bf32d..e0ab34a 100644
--- a/drivers/crypto/chelsio/chcr_algo.c
+++ b/drivers/crypto/chelsio/chcr_algo.c
@@ -888,9 +888,11 @@ static int chcr_update_tweak(struct ablkcipher_request *req, u8 *iv)
int ret, i;
u8 *key;
unsigned int keylen;
+ int round = reqctx->last_req_len / AES_BLOCK_SIZE;
+ int round8 = round / 8;
cipher = ablkctx->aes_generic;
- memcpy(iv, req->info, AES_BLOCK_SIZE);
+ memcpy(iv, reqctx->iv, AES_BLOCK_SIZE);
keylen = ablkctx->enckey_len / 2;
key = ablkctx->key + keylen;
@@ -899,7 +901,10 @@ static int chcr_update_tweak(struct ablkcipher_request *req, u8 *iv)
goto out;
crypto_cipher_encrypt_one(cipher, iv, iv);
- for (i = 0; i < (reqctx->processed / AES_BLOCK_SIZE); i++)
+ for (i = 0; i < round8; i++)
+ gf128mul_x8_ble((le128 *)iv, (le128 *)iv);
+
+ for (i = 0; i < (round % 8); i++)
gf128mul_x_ble((le128 *)iv, (le128 *)iv);
crypto_cipher_decrypt_one(cipher, iv, iv);
@@ -1040,6 +1045,7 @@ static int chcr_handle_cipher_resp(struct ablkcipher_request *req,
CRYPTO_ALG_SUB_TYPE_CTR)
bytes = adjust_ctr_overflow(reqctx->iv, bytes);
reqctx->processed += bytes;
+ reqctx->last_req_len = bytes;
wrparam.qid = u_ctx->lldi.rxq_ids[ctx->rx_qidx];
wrparam.req = req;
wrparam.bytes = bytes;
@@ -1132,6 +1138,7 @@ static int process_cipher(struct ablkcipher_request *req,
goto error;
}
reqctx->processed = bytes;
+ reqctx->last_req_len = bytes;
reqctx->dst = reqctx->dstsg;
reqctx->op = op_type;
wrparam.qid = qid;
diff --git a/drivers/crypto/chelsio/chcr_crypto.h b/drivers/crypto/chelsio/chcr_crypto.h
index 30af1ee..b3722b3 100644
--- a/drivers/crypto/chelsio/chcr_crypto.h
+++ b/drivers/crypto/chelsio/chcr_crypto.h
@@ -247,6 +247,7 @@ struct chcr_blkcipher_req_ctx {
struct scatterlist *dst;
struct scatterlist *newdstsg;
unsigned int processed;
+ unsigned int last_req_len;
unsigned int op;
short int dst_nents;
u8 iv[CHCR_MAX_CRYPTO_IV_LEN];
--
2.1.4
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox