| Message ID | 20240515160246.5181-3-larysa.zaremba@intel.com |
|---|---|
| State | Accepted |
| Delegated to: | Anthony Nguyen |
| Headers | show |
| Series | Fix AF_XDP problems after changing queue number | expand |
On Wed, May 15, 2024 at 06:02:15PM +0200, Larysa Zaremba wrote: > Commit 6624e780a577 ("ice: split ice_vsi_setup into smaller functions") > has placed ice_vsi_free_q_vectors() after ice_destroy_xdp_rings() in > the rebuild process. The behaviour of the XDP rings config functions is > context-dependent, so the change of order has led to > ice_destroy_xdp_rings() doing additional work and removing XDP prog, when > it was supposed to be preserved. > > Also, dependency on the PF state reset flags creates an additional, > fortunately less common problem: > > * PFR is requested e.g. by tx_timeout handler > * .ndo_bpf() is asked to delete the program, calls ice_destroy_xdp_rings(), > but reset flag is set, so rings are destroyed without deleting the > program > * ice_vsi_rebuild tries to delete non-existent XDP rings, because the > program is still on the VSI > * system crashes > > With a similar race, when requested to attach a program, > ice_prepare_xdp_rings() can actually skip setting the program in the VSI > and nevertheless report success. > > Instead of reverting to the old order of function calls, add an enum > argument to both ice_prepare_xdp_rings() and ice_destroy_xdp_rings() in > order to distinguish between calls from rebuild and .ndo_bpf(). > > Fixes: efc2214b6047 ("ice: Add support for XDP") > Reviewed-by: Igor Bagnucki <igor.bagnucki@intel.com> > Signed-off-by: Larysa Zaremba <larysa.zaremba@intel.com> Reviewed-by: Simon Horman <horms@kernel.org>
This patch also fixes an issue when XDP programs become detached from the RX rings on channel number reconfiguration Regards, Sergey
>-----Original Message----- >From: Intel-wired-lan <intel-wired-lan-bounces@osuosl.org> On Behalf Of >Zaremba, Larysa >Sent: Wednesday, May 15, 2024 9:32 PM >To: intel-wired-lan@lists.osuosl.org; Keller, Jacob E <jacob.e.keller@intel.com> >Cc: Fijalkowski, Maciej <maciej.fijalkowski@intel.com>; Jesper Dangaard Brouer ><hawk@kernel.org>; Daniel Borkmann <daniel@iogearbox.net>; Zaremba, >Larysa <larysa.zaremba@intel.com>; Kitszel, Przemyslaw ><przemyslaw.kitszel@intel.com>; John Fastabend ><john.fastabend@gmail.com>; Alexei Starovoitov <ast@kernel.org>; David S. >Miller <davem@davemloft.net>; Eric Dumazet <edumazet@google.com>; >netdev@vger.kernel.org; Jakub Kicinski <kuba@kernel.org>; >bpf@vger.kernel.org; Paolo Abeni <pabeni@redhat.com>; Magnus Karlsson ><magnus.karlsson@gmail.com>; Bagnucki, Igor <igor.bagnucki@intel.com>; >linux-kernel@vger.kernel.org >Subject: [Intel-wired-lan] [PATCH iwl-net 2/3] ice: add flag to distinguish reset >from .ndo_bpf in XDP rings config > >Commit 6624e780a577 ("ice: split ice_vsi_setup into smaller functions") has >placed ice_vsi_free_q_vectors() after ice_destroy_xdp_rings() in the rebuild >process. The behaviour of the XDP rings config functions is context-dependent, >so the change of order has led to >ice_destroy_xdp_rings() doing additional work and removing XDP prog, when it >was supposed to be preserved. > >Also, dependency on the PF state reset flags creates an additional, fortunately >less common problem: > >* PFR is requested e.g. by tx_timeout handler >* .ndo_bpf() is asked to delete the program, calls ice_destroy_xdp_rings(), > but reset flag is set, so rings are destroyed without deleting the > program >* ice_vsi_rebuild tries to delete non-existent XDP rings, because the > program is still on the VSI >* system crashes > >With a similar race, when requested to attach a program, >ice_prepare_xdp_rings() can actually skip setting the program in the VSI and >nevertheless report success. > >Instead of reverting to the old order of function calls, add an enum argument >to both ice_prepare_xdp_rings() and ice_destroy_xdp_rings() in order to >distinguish between calls from rebuild and .ndo_bpf(). > >Fixes: efc2214b6047 ("ice: Add support for XDP") >Reviewed-by: Igor Bagnucki <igor.bagnucki@intel.com> >Signed-off-by: Larysa Zaremba <larysa.zaremba@intel.com> >--- > drivers/net/ethernet/intel/ice/ice.h | 11 +++++++++-- > drivers/net/ethernet/intel/ice/ice_lib.c | 5 +++-- >drivers/net/ethernet/intel/ice/ice_main.c | 22 ++++++++++++---------- > 3 files changed, 24 insertions(+), 14 deletions(-) > Tested-by: Chandan Kumar Rout <chandanx.rout@intel.com> (A Contingent Worker at Intel)
diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h index d4d840729bda..b91b2594b29d 100644 --- a/drivers/net/ethernet/intel/ice/ice.h +++ b/drivers/net/ethernet/intel/ice/ice.h @@ -930,9 +930,16 @@ int ice_down(struct ice_vsi *vsi); int ice_down_up(struct ice_vsi *vsi); int ice_vsi_cfg_lan(struct ice_vsi *vsi); struct ice_vsi *ice_lb_vsi_setup(struct ice_pf *pf, struct ice_port_info *pi); + +enum ice_xdp_cfg { + ICE_XDP_CFG_FULL, /* Fully apply new config in .ndo_bpf() */ + ICE_XDP_CFG_PART, /* Save/use part of config in VSI rebuild */ +}; + int ice_vsi_determine_xdp_res(struct ice_vsi *vsi); -int ice_prepare_xdp_rings(struct ice_vsi *vsi, struct bpf_prog *prog); -int ice_destroy_xdp_rings(struct ice_vsi *vsi); +int ice_prepare_xdp_rings(struct ice_vsi *vsi, struct bpf_prog *prog, + enum ice_xdp_cfg cfg_type); +int ice_destroy_xdp_rings(struct ice_vsi *vsi, enum ice_xdp_cfg cfg_type); int ice_xdp_xmit(struct net_device *dev, int n, struct xdp_frame **frames, u32 flags); diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c index c0a7ff6c7e87..dd8b374823ee 100644 --- a/drivers/net/ethernet/intel/ice/ice_lib.c +++ b/drivers/net/ethernet/intel/ice/ice_lib.c @@ -2285,7 +2285,8 @@ static int ice_vsi_cfg_def(struct ice_vsi *vsi) ret = ice_vsi_determine_xdp_res(vsi); if (ret) goto unroll_vector_base; - ret = ice_prepare_xdp_rings(vsi, vsi->xdp_prog); + ret = ice_prepare_xdp_rings(vsi, vsi->xdp_prog, + ICE_XDP_CFG_PART); if (ret) goto unroll_vector_base; } @@ -2429,7 +2430,7 @@ void ice_vsi_decfg(struct ice_vsi *vsi) /* return value check can be skipped here, it always returns * 0 if reset is in progress */ - ice_destroy_xdp_rings(vsi); + ice_destroy_xdp_rings(vsi, ICE_XDP_CFG_PART); ice_vsi_clear_rings(vsi); ice_vsi_free_q_vectors(vsi); diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c index f60c022f7960..2a270aacd24a 100644 --- a/drivers/net/ethernet/intel/ice/ice_main.c +++ b/drivers/net/ethernet/intel/ice/ice_main.c @@ -2711,10 +2711,12 @@ static void ice_vsi_assign_bpf_prog(struct ice_vsi *vsi, struct bpf_prog *prog) * ice_prepare_xdp_rings - Allocate, configure and setup Tx rings for XDP * @vsi: VSI to bring up Tx rings used by XDP * @prog: bpf program that will be assigned to VSI + * @cfg_type: create from scratch or restore the existing configuration * * Return 0 on success and negative value on error */ -int ice_prepare_xdp_rings(struct ice_vsi *vsi, struct bpf_prog *prog) +int ice_prepare_xdp_rings(struct ice_vsi *vsi, struct bpf_prog *prog, + enum ice_xdp_cfg cfg_type) { u16 max_txqs[ICE_MAX_TRAFFIC_CLASS] = { 0 }; int xdp_rings_rem = vsi->num_xdp_txq; @@ -2790,7 +2792,7 @@ int ice_prepare_xdp_rings(struct ice_vsi *vsi, struct bpf_prog *prog) * taken into account at the end of ice_vsi_rebuild, where * ice_cfg_vsi_lan is being called */ - if (ice_is_reset_in_progress(pf->state)) + if (cfg_type == ICE_XDP_CFG_PART) return 0; /* tell the Tx scheduler that right now we have @@ -2842,22 +2844,21 @@ int ice_prepare_xdp_rings(struct ice_vsi *vsi, struct bpf_prog *prog) /** * ice_destroy_xdp_rings - undo the configuration made by ice_prepare_xdp_rings * @vsi: VSI to remove XDP rings + * @cfg_type: disable XDP permanently or allow it to be restored later * * Detach XDP rings from irq vectors, clean up the PF bitmap and free * resources */ -int ice_destroy_xdp_rings(struct ice_vsi *vsi) +int ice_destroy_xdp_rings(struct ice_vsi *vsi, enum ice_xdp_cfg cfg_type) { u16 max_txqs[ICE_MAX_TRAFFIC_CLASS] = { 0 }; struct ice_pf *pf = vsi->back; int i, v_idx; /* q_vectors are freed in reset path so there's no point in detaching - * rings; in case of rebuild being triggered not from reset bits - * in pf->state won't be set, so additionally check first q_vector - * against NULL + * rings */ - if (ice_is_reset_in_progress(pf->state) || !vsi->q_vectors[0]) + if (cfg_type == ICE_XDP_CFG_PART) goto free_qmap; ice_for_each_q_vector(vsi, v_idx) { @@ -2898,7 +2899,7 @@ int ice_destroy_xdp_rings(struct ice_vsi *vsi) if (static_key_enabled(&ice_xdp_locking_key)) static_branch_dec(&ice_xdp_locking_key); - if (ice_is_reset_in_progress(pf->state) || !vsi->q_vectors[0]) + if (cfg_type == ICE_XDP_CFG_PART) return 0; ice_vsi_assign_bpf_prog(vsi, NULL); @@ -3009,7 +3010,8 @@ ice_xdp_setup_prog(struct ice_vsi *vsi, struct bpf_prog *prog, if (xdp_ring_err) { NL_SET_ERR_MSG_MOD(extack, "Not enough Tx resources for XDP"); } else { - xdp_ring_err = ice_prepare_xdp_rings(vsi, prog); + xdp_ring_err = ice_prepare_xdp_rings(vsi, prog, + ICE_XDP_CFG_FULL); if (xdp_ring_err) NL_SET_ERR_MSG_MOD(extack, "Setting up XDP Tx resources failed"); } @@ -3020,7 +3022,7 @@ ice_xdp_setup_prog(struct ice_vsi *vsi, struct bpf_prog *prog, NL_SET_ERR_MSG_MOD(extack, "Setting up XDP Rx resources failed"); } else if (ice_is_xdp_ena_vsi(vsi) && !prog) { xdp_features_clear_redirect_target(vsi->netdev); - xdp_ring_err = ice_destroy_xdp_rings(vsi); + xdp_ring_err = ice_destroy_xdp_rings(vsi, ICE_XDP_CFG_FULL); if (xdp_ring_err) NL_SET_ERR_MSG_MOD(extack, "Freeing XDP Tx resources failed"); /* reallocate Rx queues that were used for zero-copy */
Commit 6624e780a577 ("ice: split ice_vsi_setup into smaller functions") has placed ice_vsi_free_q_vectors() after ice_destroy_xdp_rings() in the rebuild process. The behaviour of the XDP rings config functions is context-dependent, so the change of order has led to ice_destroy_xdp_rings() doing additional work and removing XDP prog, when it was supposed to be preserved. Also, dependency on the PF state reset flags creates an additional, fortunately less common problem: * PFR is requested e.g. by tx_timeout handler * .ndo_bpf() is asked to delete the program, calls ice_destroy_xdp_rings(), but reset flag is set, so rings are destroyed without deleting the program * ice_vsi_rebuild tries to delete non-existent XDP rings, because the program is still on the VSI * system crashes With a similar race, when requested to attach a program, ice_prepare_xdp_rings() can actually skip setting the program in the VSI and nevertheless report success. Instead of reverting to the old order of function calls, add an enum argument to both ice_prepare_xdp_rings() and ice_destroy_xdp_rings() in order to distinguish between calls from rebuild and .ndo_bpf(). Fixes: efc2214b6047 ("ice: Add support for XDP") Reviewed-by: Igor Bagnucki <igor.bagnucki@intel.com> Signed-off-by: Larysa Zaremba <larysa.zaremba@intel.com> --- drivers/net/ethernet/intel/ice/ice.h | 11 +++++++++-- drivers/net/ethernet/intel/ice/ice_lib.c | 5 +++-- drivers/net/ethernet/intel/ice/ice_main.c | 22 ++++++++++++---------- 3 files changed, 24 insertions(+), 14 deletions(-)