diff mbox series

[v3,iwl-next] ice: allow hot-swapping XDP programs

Message ID 20230615113326.347770-1-maciej.fijalkowski@intel.com (mailing list archive)
State Awaiting Upstream
Delegated to: Netdev Maintainers
Headers show
Series [v3,iwl-next] ice: allow hot-swapping XDP programs | expand

Checks

Context Check Description
netdev/series_format warning Single patches do not need cover letters; Target tree name not specified in the subject
netdev/tree_selection success Guessed tree name to be net-next
netdev/fixes_present success Fixes tag not required for -next series
netdev/header_inline success No static functions without inline keyword in header files
netdev/build_32bit success Errors and warnings before: 8 this patch: 8
netdev/cc_maintainers warning 10 maintainers not CCed: kuba@kernel.org hawk@kernel.org daniel@iogearbox.net jesse.brandeburg@intel.com john.fastabend@gmail.com bpf@vger.kernel.org davem@davemloft.net pabeni@redhat.com ast@kernel.org edumazet@google.com
netdev/build_clang success Errors and warnings before: 8 this patch: 8
netdev/verify_signedoff success Signed-off-by tag matches author and committer
netdev/deprecated_api success None detected
netdev/check_selftest success No net selftest shell script
netdev/verify_fixes success No Fixes tag
netdev/build_allmodconfig_warn success Errors and warnings before: 8 this patch: 8
netdev/checkpatch success total: 0 errors, 0 warnings, 0 checks, 39 lines checked
netdev/kdoc success Errors and warnings before: 0 this patch: 0
netdev/source_inline success Was 0 now: 0

Commit Message

Fijalkowski, Maciej June 15, 2023, 11:33 a.m. UTC
Currently ice driver's .ndo_bpf callback brings interface down and up
independently of XDP resources' presence. This is only needed when
either these resources have to be configured or removed. It means that
if one is switching XDP programs on-the-fly with running traffic,
packets will be dropped.

To avoid this, compare early on ice_xdp_setup_prog() state of incoming
bpf_prog pointer vs the bpf_prog pointer that is already assigned to
VSI. Do the swap in case VSI has bpf_prog and incoming one are non-NULL.

Lastly, while at it, put old bpf_prog *after* the update of Rx ring's
bpf_prog pointer. In theory previous code could expose us to a state
where Rx ring's bpf_prog would still be referring to old_prog that got
released with earlier bpf_prog_put().

Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
---

v2->v3:
- move bpf_prog_put() after ice_rx_ring::xdp_prog update [Toke, Olek]
v1->v2:
- fix missing brace (sigh)

 drivers/net/ethernet/intel/ice/ice_main.c | 19 +++++++++----------
 1 file changed, 9 insertions(+), 10 deletions(-)

Comments

Toke Høiland-Jørgensen June 15, 2023, 12:29 p.m. UTC | #1
Maciej Fijalkowski <maciej.fijalkowski@intel.com> writes:

> Currently ice driver's .ndo_bpf callback brings interface down and up
> independently of XDP resources' presence. This is only needed when
> either these resources have to be configured or removed. It means that
> if one is switching XDP programs on-the-fly with running traffic,
> packets will be dropped.
>
> To avoid this, compare early on ice_xdp_setup_prog() state of incoming
> bpf_prog pointer vs the bpf_prog pointer that is already assigned to
> VSI. Do the swap in case VSI has bpf_prog and incoming one are non-NULL.
>
> Lastly, while at it, put old bpf_prog *after* the update of Rx ring's
> bpf_prog pointer. In theory previous code could expose us to a state
> where Rx ring's bpf_prog would still be referring to old_prog that got
> released with earlier bpf_prog_put().
>
> Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>

Acked-by: Toke Høiland-Jørgensen <toke@redhat.com>
Alexander Lobakin June 16, 2023, 2:05 p.m. UTC | #2
From: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
Date: Thu, 15 Jun 2023 13:33:26 +0200

> Currently ice driver's .ndo_bpf callback brings interface down and up
> independently of XDP resources' presence. This is only needed when
> either these resources have to be configured or removed. It means that
> if one is switching XDP programs on-the-fly with running traffic,
> packets will be dropped.
> 
> To avoid this, compare early on ice_xdp_setup_prog() state of incoming
> bpf_prog pointer vs the bpf_prog pointer that is already assigned to
> VSI. Do the swap in case VSI has bpf_prog and incoming one are non-NULL.
> 
> Lastly, while at it, put old bpf_prog *after* the update of Rx ring's
> bpf_prog pointer. In theory previous code could expose us to a state
> where Rx ring's bpf_prog would still be referring to old_prog that got
> released with earlier bpf_prog_put().
> 
> Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>

Reviewed-by: Alexander Lobakin <aleksander.lobakin@intel.com>

Good stuff, much missing previously.

[...]

Thanks,
Olek
Rout, ChandanX June 21, 2023, 12:16 p.m. UTC | #3
>-----Original Message-----
>From: Intel-wired-lan <intel-wired-lan-bounces@osuosl.org> On Behalf Of
>Fijalkowski, Maciej
>Sent: 15 June 2023 17:03
>To: intel-wired-lan@lists.osuosl.org
>Cc: toke@kernel.org; netdev@vger.kernel.org; Nguyen, Anthony L
><anthony.l.nguyen@intel.com>; Karlsson, Magnus
><magnus.karlsson@intel.com>; fred@cloudflare.com
>Subject: [Intel-wired-lan] [PATCH v3 iwl-next] ice: allow hot-swapping XDP
>programs
>
>Currently ice driver's .ndo_bpf callback brings interface down and up
>independently of XDP resources' presence. This is only needed when either
>these resources have to be configured or removed. It means that if one is
>switching XDP programs on-the-fly with running traffic, packets will be
>dropped.
>
>To avoid this, compare early on ice_xdp_setup_prog() state of incoming
>bpf_prog pointer vs the bpf_prog pointer that is already assigned to VSI. Do
>the swap in case VSI has bpf_prog and incoming one are non-NULL.
>
>Lastly, while at it, put old bpf_prog *after* the update of Rx ring's bpf_prog
>pointer. In theory previous code could expose us to a state where Rx ring's
>bpf_prog would still be referring to old_prog that got released with earlier
>bpf_prog_put().
>
>Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@intel.com>
>---
>
>v2->v3:
>- move bpf_prog_put() after ice_rx_ring::xdp_prog update [Toke, Olek]
>v1->v2:
>- fix missing brace (sigh)
>
> drivers/net/ethernet/intel/ice/ice_main.c | 19 +++++++++----------
> 1 file changed, 9 insertions(+), 10 deletions(-)
>

Tested-by: Chandan Kumar Rout <chandanx.rout@intel.com> (A Contingent Worker at Intel)
diff mbox series

Patch

diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index 62e91512aeab..a7c76fded603 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -2633,11 +2633,11 @@  static void ice_vsi_assign_bpf_prog(struct ice_vsi *vsi, struct bpf_prog *prog)
 	int i;
 
 	old_prog = xchg(&vsi->xdp_prog, prog);
-	if (old_prog)
-		bpf_prog_put(old_prog);
-
 	ice_for_each_rxq(vsi, i)
 		WRITE_ONCE(vsi->rx_rings[i]->xdp_prog, vsi->xdp_prog);
+
+	if (old_prog)
+		bpf_prog_put(old_prog);
 }
 
 /**
@@ -2922,6 +2922,12 @@  ice_xdp_setup_prog(struct ice_vsi *vsi, struct bpf_prog *prog,
 		}
 	}
 
+	/* hot swap progs and avoid toggling link */
+	if (ice_is_xdp_ena_vsi(vsi) == !!prog) {
+		ice_vsi_assign_bpf_prog(vsi, prog);
+		return 0;
+	}
+
 	/* need to stop netdev while setting up the program for Rx rings */
 	if (if_running && !test_and_set_bit(ICE_VSI_DOWN, vsi->state)) {
 		ret = ice_down(vsi);
@@ -2954,13 +2960,6 @@  ice_xdp_setup_prog(struct ice_vsi *vsi, struct bpf_prog *prog,
 		xdp_ring_err = ice_realloc_zc_buf(vsi, false);
 		if (xdp_ring_err)
 			NL_SET_ERR_MSG_MOD(extack, "Freeing XDP Rx resources failed");
-	} else {
-		/* safe to call even when prog == vsi->xdp_prog as
-		 * dev_xdp_install in net/core/dev.c incremented prog's
-		 * refcount so corresponding bpf_prog_put won't cause
-		 * underflow
-		 */
-		ice_vsi_assign_bpf_prog(vsi, prog);
 	}
 
 	if (if_running)