i40e: Add an i40e_napi_poll tracepoint
Subsystems (3)
v2 cover i40e: Add an i40e_napi_poll tracepoint
Greetings:
Debugging and tuning the NAPI and i40e NIC parameters can be a bit tricky as there are many different options to test.
This change adds a tracepoint to i40e_napi_poll which exposes a lot of helpful debug information for users who'd like to get a better understanding of how their NIC is performing as they adjust various parameters and tuning knobs.
With this series applied, you can use the tracepoint with perf by running:
$ sudo perf trace -e i40e:i40e_napi_poll -a --call-graph=fp --libtraceevent_print
388.258 :0/0 i40e:i40e_napi_poll(i40e_napi_poll on dev eth2 q i40e-eth2-TxRx-9 irq 346 irq_mask 00000000,00000000,00000000,00000000,00000000,00800000 curr_cpu 23 budget 64 bpr 64 rx_cleaned 28 tx_cleaned 0 rx_clean_complete 1 tx_clean_complete 1)
i40e_napi_poll ([i40e]) i40e_napi_poll ([i40e]) __napi_poll ([kernel.kallsyms]) net_rx_action ([kernel.kallsyms]) __do_softirq ([kernel.kallsyms]) common_interrupt ([kernel.kallsyms]) asm_common_interrupt ([kernel.kallsyms]) intel_idle_irq ([kernel.kallsyms]) cpuidle_enter_state ([kernel.kallsyms]) cpuidle_enter ([kernel.kallsyms]) do_idle ([kernel.kallsyms]) cpu_startup_entry ([kernel.kallsyms]) [0x243fd8] ([kernel.kallsyms]) secondary_startup_64_no_verify ([kernel.kallsyms])
The output is verbose, but is helpful when trying to determine the impact of various turning parameters.
Thanks,
Joe
v1 -> v2:
- TX path modified to push an out parameter through the function call chain instead of modifying control flow.
- RX path modified to also use an out parameter to track the number of packets processed.
- Naming of tracepoint struct members and format string modified to be more readable.
Joe Damato (4):
i40e: Store the irq number in i40e_q_vector
i40e: Record number TXes cleaned during NAPI
i40e: Record number of RXes cleaned during NAPI
i40e: Add i40e_napi_poll tracepoint
drivers/net/ethernet/intel/i40e/i40e.h | 1 + drivers/net/ethernet/intel/i40e/i40e_main.c | 1 + drivers/net/ethernet/intel/i40e/i40e_trace.h | 49 ++++++++++++++++++++++++++++ drivers/net/ethernet/intel/i40e/i40e_txrx.c | 30 ++++++++++++----- drivers/net/ethernet/intel/i40e/i40e_xsk.c | 21 +++++++++--- drivers/net/ethernet/intel/i40e/i40e_xsk.h | 6 ++-- 6 files changed, 93 insertions(+), 15 deletions(-)
Signature · 1 line
2.7.4
PATCH v2 1/4 i40e: Store the irq number in i40e_q_vector
Make it easy to figure out the IRQ number for a particular i40e_q_vector by storing the assigned IRQ in the structure itself.
--- drivers/net/ethernet/intel/i40e/i40e.h | 1 + drivers/net/ethernet/intel/i40e/i40e_main.c | 1 + 2 files changed, 2 insertions(+)
drivers/net/ethernet/intel/i40e/i40e.h +1 -0
@@ -992,6 +992,7 @@ struct i40e_q_vector {struct rcu_head rcu;/* to avoid race with update stats on free */ char name[I40E_INT_NAME_STR_LEN ];bool arm_wb_state; +int irq_num;/* IRQ assigned to this q_vector */ }____cacheline_internodealigned_in_smp ;/* lan device */
drivers/net/ethernet/intel/i40e/i40e_main.c +1 -0
@@ -4123,6 +4123,7 @@ static int i40e_vsi_request_irq_msix(struct i40e_vsi *vsi, char *basename) }/* register for affinity change notifications */ + q_vector->irq_num = irq_num; q_vector->affinity_notify.notify = i40e_irq_affinity_notify; q_vector->affinity_notify.release = i40e_irq_affinity_release;irq_set_affinity_notifier (irq_num, &q_vector->affinity_notify);
Signature · 1 line
2.7.4
PATCH v2 2/4 i40e: Record number TXes cleaned during NAPI
In reply to Joe DamatoUpdate i40e_clean_tx_irq to take an out parameter (tx_cleaned) which stores the number TXs cleaned.
Likewise, update i40e_clean_xdp_tx_irq and i40e_xmit_zc to do the same.
Care has been taken to avoid changing the control flow of any functions involved.
--- drivers/net/ethernet/intel/i40e/i40e_txrx.c | 16 +++++++++++----- drivers/net/ethernet/intel/i40e/i40e_xsk.c | 15 +++++++++++---- drivers/net/ethernet/intel/i40e/i40e_xsk.h | 3 ++- 3 files changed, 24 insertions(+), 10 deletions(-)
drivers/net/ethernet/intel/i40e/i40e_txrx.c +11 -5
@@ -923,11 +923,13 @@ void i40e_detect_recover_hung(struct i40e_vsi *vsi)* @vsi: the VSI we care about * @tx_ring: Tx ring to clean * @napi_budget: Used to determine if we are in netpoll +* @tx_cleaned: Out parameter set to the number of TXes cleaned * * Returns true if there's any budget left (e.g. the clean is finished) **/ static bool i40e_clean_tx_irq (struct i40e_vsi *vsi, -struct i40e_ring *tx_ring,int napi_budget) +struct i40e_ring *tx_ring,int napi_budget, +unsigned int *tx_cleaned) {int i = tx_ring->next_to_clean;struct i40e_tx_buffer *tx_buf;
@@ -1026,7 +1028,7 @@ static bool i40e_clean_tx_irq(struct i40e_vsi *vsi,i40e_arm_wb (tx_ring, vsi, budget);if (ring_is_xdp (tx_ring)) -return !!budget; + goto out;/* notify netdev of completed buffers */ netdev_tx_completed_queue(txring_txq(tx_ring),
@@ -1048,6 +1050,8 @@ static bool i40e_clean_tx_irq(struct i40e_vsi *vsi, } } +out: + *tx_cleaned = total_packets;return !!budget; }
@@ -2689,10 +2693,12 @@ int i40e_napi_poll(struct napi_struct *napi, int budget) container_of(napi, struct i40e_q_vector, napi);struct i40e_vsi *vsi = q_vector->vsi;struct i40e_ring *ring; +bool tx_clean_complete = true;bool clean_complete = true;bool arm_wb = false;int budget_per_ring;int work_done =0 ; +unsigned int tx_cleaned =0 ;if (test_bit (__I40E_VSI_DOWN, vsi->state)) {napi_complete (napi);
@@ -2704,11 +2710,11 @@ int i40e_napi_poll(struct napi_struct *napi, int budget)*/ i40e_for_each_ring (ring, q_vector->tx) {bool wd = ring->xsk_pool ? -i40e_clean_xdp_tx_irq (vsi, ring) : -i40e_clean_tx_irq (vsi, ring, budget); +i40e_clean_xdp_tx_irq (vsi, ring, &tx_cleaned) : +i40e_clean_tx_irq (vsi, ring, budget, &tx_cleaned);if (!wd) { - clean_complete = false; + clean_complete = tx_clean_complete = false;continue ; } arm_wb |= ring->arm_wb;
drivers/net/ethernet/intel/i40e/i40e_xsk.c +11 -4
@@ -530,18 +530,22 @@ static void i40e_set_rs_bit(struct i40e_ring *xdp_ring)* i40e_xmit_zc - Performs zero-copy Tx AF_XDP * @xdp_ring: XDP Tx ring * @budget: NAPI budget +* @tx_cleaned: Out parameter of the TX packets processed * * Returns true if the work is finished. **/ -static bool i40e_xmit_zc (struct i40e_ring *xdp_ring,unsigned int budget) +static bool i40e_xmit_zc (struct i40e_ring *xdp_ring,unsigned int budget, +unsigned int *tx_cleaned) {struct xdp_desc *descs = xdp_ring->xsk_pool->tx_descs;u32 nb_pkts, nb_processed =0 ;unsigned int total_bytes =0 ; nb_pkts =xsk_tx_peek_release_desc_batch (xdp_ring->xsk_pool, budget); -if (!nb_pkts) +if (!nb_pkts) { + *tx_cleaned =0 ;return true; + }if (xdp_ring->next_to_use + nb_pkts >= xdp_ring->count) { nb_processed = xdp_ring->count - xdp_ring->next_to_use;
@@ -558,6 +562,7 @@ static bool i40e_xmit_zc(struct i40e_ring *xdp_ring, unsigned int budget)i40e_update_tx_stats (xdp_ring, nb_pkts, total_bytes); + *tx_cleaned = nb_pkts;return nb_pkts < budget; }
@@ -581,10 +586,12 @@ static void i40e_clean_xdp_tx_buffer(struct i40e_ring *tx_ring,* i40e_clean_xdp_tx_irq - Completes AF_XDP entries, and cleans XDP entries * @vsi: Current VSI * @tx_ring: XDP Tx ring +* @tx_cleaned: out parameter of number of TXes cleaned * * Returns true if cleanup/tranmission is done. **/ -bool i40e_clean_xdp_tx_irq (struct i40e_vsi *vsi,struct i40e_ring *tx_ring) +bool i40e_clean_xdp_tx_irq (struct i40e_vsi *vsi,struct i40e_ring *tx_ring, +unsigned int *tx_cleaned) {struct xsk_buff_pool *bp = tx_ring->xsk_pool;u32 i, completed_frames, xsk_frames =0 ;
@@ -634,7 +641,7 @@ bool i40e_clean_xdp_tx_irq(struct i40e_vsi *vsi, struct i40e_ring *tx_ring)if (xsk_uses_need_wakeup (tx_ring->xsk_pool))xsk_set_tx_need_wakeup (tx_ring->xsk_pool); -return i40e_xmit_zc (tx_ring,I40E_DESC_UNUSED (tx_ring)); +return i40e_xmit_zc (tx_ring,I40E_DESC_UNUSED (tx_ring), tx_cleaned); }/**
drivers/net/ethernet/intel/i40e/i40e_xsk.h +2 -1
@@ -30,7 +30,8 @@ int i40e_xsk_pool_setup(struct i40e_vsi *vsi, struct xsk_buff_pool *pool,bool i40e_alloc_rx_buffers_zc (struct i40e_ring *rx_ring,u16 cleaned_count);int i40e_clean_rx_irq_zc (struct i40e_ring *rx_ring,int budget); -bool i40e_clean_xdp_tx_irq (struct i40e_vsi *vsi,struct i40e_ring *tx_ring); +bool i40e_clean_xdp_tx_irq (struct i40e_vsi *vsi,struct i40e_ring *tx_ring, +unsigned int *tx_cleaned);int i40e_xsk_wakeup (struct net_device *dev,u32 queue_id,u32 flags);int i40e_realloc_rx_bi_zc (struct i40e_vsi *vsi,bool zc);void i40e_clear_rx_bi_zc (struct i40e_ring *rx_ring);
Signature · 1 line
2.7.4
Re: [next-queue v2 2/4] i40e: Record number TXes cleaned during NAPI
Quoted 110 lines · Joe Damato
Update
i40e_clean_tx_irqto take an out parameter (tx_cleaned) which stores the number TXs cleaned.Likewise, update
i40e_clean_xdp_tx_irqandi40e_xmit_zcto do the same.Care has been taken to avoid changing the control flow of any functions involved.
Signed-off-by Joe Damato <[email protected]>--- drivers/net/ethernet/intel/i40e/i40e_txrx.c | 16 +++++++++++----- drivers/net/ethernet/intel/i40e/i40e_xsk.c | 15 +++++++++++---- drivers/net/ethernet/intel/i40e/i40e_xsk.h | 3 ++- 3 files changed, 24 insertions(+), 10 deletions(-)drivers/net/ethernet/intel/i40e/i40e_txrx.c +11
-5@@ -923,11 +923,13 @@ void i40e_detect_recover_hung(struct i40e_vsi *vsi)* @vsi: the VSI we care about * @tx_ring: Tx ring to clean * @napi_budget: Used to determine if we are in netpoll +* @tx_cleaned: Out parameter set to the number of TXes cleaned * * Returns true if there's any budget left (e.g. the clean is finished) **/ static bool i40e_clean_tx_irq (struct i40e_vsi *vsi, -struct i40e_ring *tx_ring,int napi_budget) +struct i40e_ring *tx_ring,int napi_budget, +unsigned int *tx_cleaned) {int i = tx_ring->next_to_clean;struct i40e_tx_buffer *tx_buf;@@ -1026,7 +1028,7 @@ static bool i40e_clean_tx_irq(struct i40e_vsi *vsi,i40e_arm_wb (tx_ring, vsi, budget);if (ring_is_xdp (tx_ring)) -return !!budget; + goto out;/* notify netdev of completed buffers */ netdev_tx_completed_queue(txring_txq(tx_ring),@@ -1048,6 +1050,8 @@ static bool i40e_clean_tx_irq(struct i40e_vsi *vsi, } } +out: + *tx_cleaned = total_packets;return !!budget; }@@ -2689,10 +2693,12 @@ int i40e_napi_poll(struct napi_struct *napi, int budget) container_of(napi, struct i40e_q_vector, napi);struct i40e_vsi *vsi = q_vector->vsi;struct i40e_ring *ring; +bool tx_clean_complete = true;bool clean_complete = true;bool arm_wb = false;int budget_per_ring;int work_done =0 ; +unsigned int tx_cleaned =0 ;if (test_bit (__I40E_VSI_DOWN, vsi->state)) {napi_complete (napi);@@ -2704,11 +2710,11 @@ int i40e_napi_poll(struct napi_struct *napi, int budget)*/ i40e_for_each_ring (ring, q_vector->tx) {bool wd = ring->xsk_pool ? -i40e_clean_xdp_tx_irq (vsi, ring) : -i40e_clean_tx_irq (vsi, ring, budget); +i40e_clean_xdp_tx_irq (vsi, ring, &tx_cleaned) : +i40e_clean_tx_irq (vsi, ring, budget, &tx_cleaned);if (!wd) { - clean_complete = false; + clean_complete = tx_clean_complete = false;continue ; } arm_wb |= ring->arm_wb;drivers/net/ethernet/intel/i40e/i40e_xsk.c +6
-2@@ -530,18 +530,22 @@ static void i40e_set_rs_bit(struct i40e_ring *xdp_ring)* i40e_xmit_zc - Performs zero-copy Tx AF_XDP * @xdp_ring: XDP Tx ring * @budget: NAPI budget +* @tx_cleaned: Out parameter of the TX packets processed * * Returns true if the work is finished. **/ -static bool i40e_xmit_zc (struct i40e_ring *xdp_ring,unsigned int budget) +static bool i40e_xmit_zc (struct i40e_ring *xdp_ring,unsigned int budget, +unsigned int *tx_cleaned) {struct xdp_desc *descs = xdp_ring->xsk_pool->tx_descs;u32 nb_pkts, nb_processed =0 ;unsigned int total_bytes =0 ; nb_pkts =xsk_tx_peek_release_desc_batch (xdp_ring->xsk_pool, budget); -if (!nb_pkts) +if (!nb_pkts) { + *tx_cleaned =0 ;return true; + }if (xdp_ring->next_to_use + nb_pkts >= xdp_ring->count) { nb_processed = xdp_ring->count - xdp_ring->next_to_use;
drivers/net/ethernet/intel/i40e/i40e_xsk.c static bool i40e_xmit_zc(struct i40e_ring *xdp_ring, unsigned int budget) · line 565
Quoted 4 lines
drivers/net/ethernet/intel/i40e/i40e_xsk.c +1
-0@@ -558,6 +562,7 @@ static bool i40e_xmit_zc(struct i40e_ring *xdp_ring, unsigned int budget)i40e_update_tx_stats (xdp_ring, nb_pkts, total_bytes); + *tx_cleaned = nb_pkts;
With XDP, I don't think we should count these as tx_cleaned packets. These are transmitted packets. The tx_cleaned would be the xsk_frames counter in i40e_clean_xdp_tx_irq May be we need 2 counters for xdp.
Quoted 31 lines
drivers/net/ethernet/intel/i40e/i40e_xsk.c +4
-2return nb_pkts < budget; }@@ -581,10 +586,12 @@ static void i40e_clean_xdp_tx_buffer(struct i40e_ring *tx_ring,* i40e_clean_xdp_tx_irq - Completes AF_XDP entries, and cleans XDP entries * @vsi: Current VSI * @tx_ring: XDP Tx ring +* @tx_cleaned: out parameter of number of TXes cleaned * * Returns true if cleanup/tranmission is done. **/ -bool i40e_clean_xdp_tx_irq (struct i40e_vsi *vsi,struct i40e_ring *tx_ring) +bool i40e_clean_xdp_tx_irq (struct i40e_vsi *vsi,struct i40e_ring *tx_ring, +unsigned int *tx_cleaned) {struct xsk_buff_pool *bp = tx_ring->xsk_pool;u32 i, completed_frames, xsk_frames =0 ;@@ -634,7 +641,7 @@ bool i40e_clean_xdp_tx_irq(struct i40e_vsi *vsi, struct i40e_ring *tx_ring)if (xsk_uses_need_wakeup (tx_ring->xsk_pool))xsk_set_tx_need_wakeup (tx_ring->xsk_pool); -return i40e_xmit_zc (tx_ring,I40E_DESC_UNUSED (tx_ring)); +return i40e_xmit_zc (tx_ring,I40E_DESC_UNUSED (tx_ring), tx_cleaned); }/** drivers/net/ethernet/intel/i40e/i40e_xsk.h +0
-0
drivers/net/ethernet/intel/i40e/i40e_xsk.h int i40e_xsk_pool_setup(struct i40e_vsi *vsi, struct xsk_buff_pool *pool, · line 37
Quoted 9 lines
drivers/net/ethernet/intel/i40e/i40e_xsk.h +2
-1@@ -30,7 +30,8 @@ int i40e_xsk_pool_setup(struct i40e_vsi *vsi, struct xsk_buff_pool *pool,bool i40e_alloc_rx_buffers_zc (struct i40e_ring *rx_ring,u16 cleaned_count);int i40e_clean_rx_irq_zc (struct i40e_ring *rx_ring,int budget); -bool i40e_clean_xdp_tx_irq (struct i40e_vsi *vsi,struct i40e_ring *tx_ring); +bool i40e_clean_xdp_tx_irq (struct i40e_vsi *vsi,struct i40e_ring *tx_ring, +unsigned int *tx_cleaned);int i40e_xsk_wakeup (struct net_device *dev,u32 queue_id,u32 flags);int i40e_realloc_rx_bi_zc (struct i40e_vsi *vsi,bool zc);void i40e_clear_rx_bi_zc (struct i40e_ring *rx_ring);
_______________________________________________
Intel-wired-lan mailing list
[email protected]
https://lists.osuosl.org/mailman/listinfo/intel-wired-lan
Re: [next-queue v2 2/4] i40e: Record number TXes cleaned during NAPI
Quoted 101 lines · Sridhar
Quoted 100 lines · Joe Damato
Update
i40e_clean_tx_irqto take an out parameter (tx_cleaned) which stores the number TXs cleaned.Likewise, update
i40e_clean_xdp_tx_irqandi40e_xmit_zcto do the same.Care has been taken to avoid changing the control flow of any functions involved.
Signed-off-by Joe Damato <[email protected]>--- drivers/net/ethernet/intel/i40e/i40e_txrx.c | 16 +++++++++++----- drivers/net/ethernet/intel/i40e/i40e_xsk.c | 15 +++++++++++---- drivers/net/ethernet/intel/i40e/i40e_xsk.h | 3 ++- 3 files changed, 24 insertions(+), 10 deletions(-)drivers/net/ethernet/intel/i40e/i40e_txrx.c +11
-5@@ -923,11 +923,13 @@ void i40e_detect_recover_hung(struct i40e_vsi *vsi)* @vsi: the VSI we care about * @tx_ring: Tx ring to clean * @napi_budget: Used to determine if we are in netpoll +* @tx_cleaned: Out parameter set to the number of TXes cleaned * * Returns true if there's any budget left (e.g. the clean is finished) **/ static bool i40e_clean_tx_irq (struct i40e_vsi *vsi, -struct i40e_ring *tx_ring,int napi_budget) +struct i40e_ring *tx_ring,int napi_budget, +unsigned int *tx_cleaned) {int i = tx_ring->next_to_clean;struct i40e_tx_buffer *tx_buf;@@ -1026,7 +1028,7 @@ static bool i40e_clean_tx_irq(struct i40e_vsi *vsi,i40e_arm_wb (tx_ring, vsi, budget);if (ring_is_xdp (tx_ring)) -return !!budget; + goto out;/* notify netdev of completed buffers */ netdev_tx_completed_queue(txring_txq(tx_ring),@@ -1048,6 +1050,8 @@ static bool i40e_clean_tx_irq(struct i40e_vsi *vsi, } } +out: + *tx_cleaned = total_packets;return !!budget; }@@ -2689,10 +2693,12 @@ int i40e_napi_poll(struct napi_struct *napi, int budget) container_of(napi, struct i40e_q_vector, napi);struct i40e_vsi *vsi = q_vector->vsi;struct i40e_ring *ring; +bool tx_clean_complete = true;bool clean_complete = true;bool arm_wb = false;int budget_per_ring;int work_done =0 ; +unsigned int tx_cleaned =0 ;if (test_bit (__I40E_VSI_DOWN, vsi->state)) {napi_complete (napi);@@ -2704,11 +2710,11 @@ int i40e_napi_poll(struct napi_struct *napi, int budget)*/ i40e_for_each_ring (ring, q_vector->tx) {bool wd = ring->xsk_pool ? -i40e_clean_xdp_tx_irq (vsi, ring) : -i40e_clean_tx_irq (vsi, ring, budget); +i40e_clean_xdp_tx_irq (vsi, ring, &tx_cleaned) : +i40e_clean_tx_irq (vsi, ring, budget, &tx_cleaned);if (!wd) { - clean_complete = false; + clean_complete = tx_clean_complete = false;continue ; } arm_wb |= ring->arm_wb;drivers/net/ethernet/intel/i40e/i40e_xsk.c +6
-2@@ -530,18 +530,22 @@ static void i40e_set_rs_bit(struct i40e_ring *xdp_ring)* i40e_xmit_zc - Performs zero-copy Tx AF_XDP * @xdp_ring: XDP Tx ring * @budget: NAPI budget +* @tx_cleaned: Out parameter of the TX packets processed * * Returns true if the work is finished. **/ -static bool i40e_xmit_zc (struct i40e_ring *xdp_ring,unsigned int budget) +static bool i40e_xmit_zc (struct i40e_ring *xdp_ring,unsigned int budget, +unsigned int *tx_cleaned) {struct xdp_desc *descs = xdp_ring->xsk_pool->tx_descs;u32 nb_pkts, nb_processed =0 ;unsigned int total_bytes =0 ; nb_pkts =xsk_tx_peek_release_desc_batch (xdp_ring->xsk_pool, budget); -if (!nb_pkts) +if (!nb_pkts) { + *tx_cleaned =0 ;return true; + }if (xdp_ring->next_to_use + nb_pkts >= xdp_ring->count) {
Quoted 8 lines
drivers/net/ethernet/intel/i40e/i40e_xsk.c static bool i40e_xmit_zc(struct i40e_ring *xdp_ring, unsigned int budget) · line 563
Quoted 4 lines
drivers/net/ethernet/intel/i40e/i40e_xsk.c +1
-0nb_processed = xdp_ring->count - xdp_ring->next_to_use;@@ -558,6 +562,7 @@ static bool i40e_xmit_zc(struct i40e_ring *xdp_ring, unsigned int budget)i40e_update_tx_stats (xdp_ring, nb_pkts, total_bytes); + *tx_cleaned = nb_pkts;With XDP, I don't think we should count these as
tx_cleanedpackets. These are transmitted packets. Thetx_cleanedwould be thexsk_framescounter ini40e_clean_xdp_tx_irqMay be we need 2 counters for xdp.
I think there's two issues you are describing, which are separate in my mind.
1.) The name "tx_cleaned", and
2.) Whether nb_pkts is the right thing to write as the out param.
For #1: I'm OK to change the name if that's the blocker here; please suggest a suitable alternative that you'll accept.
For #2: nb_pkts is, IMO, the right value to bubble up to the tracepoint because nb_pkts affects clean_complete in i40e_napi_poll which in turn determines whether or not polling mode is entered.
The purpose of the tracepoint is to determine when/why/how you are entering polling mode, so if nb_pkts plays a role in that calculation, it's the right number to output.
Quoted 37 lines
Quoted 36 lines
drivers/net/ethernet/intel/i40e/i40e_xsk.c +4
-2return nb_pkts < budget; }@@ -581,10 +586,12 @@ static void i40e_clean_xdp_tx_buffer(struct i40e_ring *tx_ring,* i40e_clean_xdp_tx_irq - Completes AF_XDP entries, and cleans XDP entries * @vsi: Current VSI * @tx_ring: XDP Tx ring +* @tx_cleaned: out parameter of number of TXes cleaned * * Returns true if cleanup/tranmission is done. **/ -bool i40e_clean_xdp_tx_irq (struct i40e_vsi *vsi,struct i40e_ring *tx_ring) +bool i40e_clean_xdp_tx_irq (struct i40e_vsi *vsi,struct i40e_ring *tx_ring, +unsigned int *tx_cleaned) {struct xsk_buff_pool *bp = tx_ring->xsk_pool;u32 i, completed_frames, xsk_frames =0 ;@@ -634,7 +641,7 @@ bool i40e_clean_xdp_tx_irq(struct i40e_vsi *vsi, struct i40e_ring *tx_ring)if (xsk_uses_need_wakeup (tx_ring->xsk_pool))xsk_set_tx_need_wakeup (tx_ring->xsk_pool); -return i40e_xmit_zc (tx_ring,I40E_DESC_UNUSED (tx_ring)); +return i40e_xmit_zc (tx_ring,I40E_DESC_UNUSED (tx_ring), tx_cleaned); }/** drivers/net/ethernet/intel/i40e/i40e_xsk.h +2
-1@@ -30,7 +30,8 @@ int i40e_xsk_pool_setup(struct i40e_vsi *vsi, struct xsk_buff_pool *pool,bool i40e_alloc_rx_buffers_zc (struct i40e_ring *rx_ring,u16 cleaned_count);int i40e_clean_rx_irq_zc (struct i40e_ring *rx_ring,int budget); -bool i40e_clean_xdp_tx_irq (struct i40e_vsi *vsi,struct i40e_ring *tx_ring); +bool i40e_clean_xdp_tx_irq (struct i40e_vsi *vsi,struct i40e_ring *tx_ring, +unsigned int *tx_cleaned);int i40e_xsk_wakeup (struct net_device *dev,u32 queue_id,u32 flags);int i40e_realloc_rx_bi_zc (struct i40e_vsi *vsi,bool zc);void i40e_clear_rx_bi_zc (struct i40e_ring *rx_ring);
Re: [next-queue v2 2/4] i40e: Record number TXes cleaned during NAPI
Quoted 120 lines · Joe Damato
Quoted 101 lines · Sridhar
Quoted 100 lines · Joe Damato
Update
i40e_clean_tx_irqto take an out parameter (tx_cleaned) which stores the number TXs cleaned.Likewise, update
i40e_clean_xdp_tx_irqandi40e_xmit_zcto do the same.Care has been taken to avoid changing the control flow of any functions involved.
Signed-off-by Joe Damato <[email protected]>--- drivers/net/ethernet/intel/i40e/i40e_txrx.c | 16 +++++++++++----- drivers/net/ethernet/intel/i40e/i40e_xsk.c | 15 +++++++++++---- drivers/net/ethernet/intel/i40e/i40e_xsk.h | 3 ++- 3 files changed, 24 insertions(+), 10 deletions(-)drivers/net/ethernet/intel/i40e/i40e_txrx.c +11
-5@@ -923,11 +923,13 @@ void i40e_detect_recover_hung(struct i40e_vsi *vsi)* @vsi: the VSI we care about * @tx_ring: Tx ring to clean * @napi_budget: Used to determine if we are in netpoll +* @tx_cleaned: Out parameter set to the number of TXes cleaned * * Returns true if there's any budget left (e.g. the clean is finished) **/ static bool i40e_clean_tx_irq (struct i40e_vsi *vsi, -struct i40e_ring *tx_ring,int napi_budget) +struct i40e_ring *tx_ring,int napi_budget, +unsigned int *tx_cleaned) {int i = tx_ring->next_to_clean;struct i40e_tx_buffer *tx_buf;@@ -1026,7 +1028,7 @@ static bool i40e_clean_tx_irq(struct i40e_vsi *vsi,i40e_arm_wb (tx_ring, vsi, budget);if (ring_is_xdp (tx_ring)) -return !!budget; + goto out;/* notify netdev of completed buffers */ netdev_tx_completed_queue(txring_txq(tx_ring),@@ -1048,6 +1050,8 @@ static bool i40e_clean_tx_irq(struct i40e_vsi *vsi, } } +out: + *tx_cleaned = total_packets;return !!budget; }@@ -2689,10 +2693,12 @@ int i40e_napi_poll(struct napi_struct *napi, int budget) container_of(napi, struct i40e_q_vector, napi);struct i40e_vsi *vsi = q_vector->vsi;struct i40e_ring *ring; +bool tx_clean_complete = true;bool clean_complete = true;bool arm_wb = false;int budget_per_ring;int work_done =0 ; +unsigned int tx_cleaned =0 ;if (test_bit (__I40E_VSI_DOWN, vsi->state)) {napi_complete (napi);@@ -2704,11 +2710,11 @@ int i40e_napi_poll(struct napi_struct *napi, int budget)*/ i40e_for_each_ring (ring, q_vector->tx) {bool wd = ring->xsk_pool ? -i40e_clean_xdp_tx_irq (vsi, ring) : -i40e_clean_tx_irq (vsi, ring, budget); +i40e_clean_xdp_tx_irq (vsi, ring, &tx_cleaned) : +i40e_clean_tx_irq (vsi, ring, budget, &tx_cleaned);if (!wd) { - clean_complete = false; + clean_complete = tx_clean_complete = false;continue ; } arm_wb |= ring->arm_wb;drivers/net/ethernet/intel/i40e/i40e_xsk.c +6
-2@@ -530,18 +530,22 @@ static void i40e_set_rs_bit(struct i40e_ring *xdp_ring)* i40e_xmit_zc - Performs zero-copy Tx AF_XDP * @xdp_ring: XDP Tx ring * @budget: NAPI budget +* @tx_cleaned: Out parameter of the TX packets processed * * Returns true if the work is finished. **/ -static bool i40e_xmit_zc (struct i40e_ring *xdp_ring,unsigned int budget) +static bool i40e_xmit_zc (struct i40e_ring *xdp_ring,unsigned int budget, +unsigned int *tx_cleaned) {struct xdp_desc *descs = xdp_ring->xsk_pool->tx_descs;u32 nb_pkts, nb_processed =0 ;unsigned int total_bytes =0 ; nb_pkts =xsk_tx_peek_release_desc_batch (xdp_ring->xsk_pool, budget); -if (!nb_pkts) +if (!nb_pkts) { + *tx_cleaned =0 ;return true; + }if (xdp_ring->next_to_use + nb_pkts >= xdp_ring->count) {Quoted 8 lines
drivers/net/ethernet/intel/i40e/i40e_xsk.c static bool i40e_xmit_zc(struct i40e_ring *xdp_ring, unsigned int budget) · line 563
Quoted 4 lines
drivers/net/ethernet/intel/i40e/i40e_xsk.c +1
-0nb_processed = xdp_ring->count - xdp_ring->next_to_use;@@ -558,6 +562,7 @@ static bool i40e_xmit_zc(struct i40e_ring *xdp_ring, unsigned int budget)i40e_update_tx_stats (xdp_ring, nb_pkts, total_bytes); + *tx_cleaned = nb_pkts;With XDP, I don't think we should count these as
tx_cleanedpackets. These are transmitted packets. Thetx_cleanedwould be thexsk_framescounter ini40e_clean_xdp_tx_irqMay be we need 2 counters for xdp.I think there's two issues you are describing, which are separate in my mind.
1.) The name "
tx_cleaned", and
2.) Whethernb_pktsis the right thing to write as the out param.For #1: I'm OK to change the name if that's the blocker here; please suggest a suitable alternative that you'll accept.
Quoted 7 lines
For #2:
nb_pktsis, IMO, the right value to bubble up to the tracepoint becausenb_pktsaffectsclean_completeini40e_napi_pollwhich in turn determines whether or not polling mode is entered.The purpose of the tracepoint is to determine when/why/how you are entering polling mode, so if
nb_pktsplays a role in that calculation, it's the right number to output.
I suppose the alternative is to only fire the tracepoint when *not* in XDP. Then the changes to the XDP stuff can be dropped and a separate set of tracepoints for XDP can be created in the future.
That might reduce the complexity a bit, and will probably still be pretty useful for people tuning their non-XDP workloads.
_______________________________________________
Intel-wired-lan mailing list
[email protected]
https://lists.osuosl.org/mailman/listinfo/intel-wired-lan
Re: [next-queue v2 2/4] i40e: Record number TXes cleaned during NAPI
Quoted 125 lines · Joe Damato
Quoted 124 lines · Joe Damato
Quoted 101 lines · Sridhar
Quoted 100 lines · Joe Damato
Update
i40e_clean_tx_irqto take an out parameter (tx_cleaned) which stores the number TXs cleaned.Likewise, update
i40e_clean_xdp_tx_irqandi40e_xmit_zcto do the same.Care has been taken to avoid changing the control flow of any functions involved.
Signed-off-by Joe Damato <[email protected]>--- drivers/net/ethernet/intel/i40e/i40e_txrx.c | 16 +++++++++++----- drivers/net/ethernet/intel/i40e/i40e_xsk.c | 15 +++++++++++---- drivers/net/ethernet/intel/i40e/i40e_xsk.h | 3 ++- 3 files changed, 24 insertions(+), 10 deletions(-)drivers/net/ethernet/intel/i40e/i40e_txrx.c +11
-5@@ -923,11 +923,13 @@ void i40e_detect_recover_hung(struct i40e_vsi *vsi)* @vsi: the VSI we care about * @tx_ring: Tx ring to clean * @napi_budget: Used to determine if we are in netpoll +* @tx_cleaned: Out parameter set to the number of TXes cleaned * * Returns true if there's any budget left (e.g. the clean is finished) **/ static bool i40e_clean_tx_irq (struct i40e_vsi *vsi, -struct i40e_ring *tx_ring,int napi_budget) +struct i40e_ring *tx_ring,int napi_budget, +unsigned int *tx_cleaned) {int i = tx_ring->next_to_clean;struct i40e_tx_buffer *tx_buf;@@ -1026,7 +1028,7 @@ static bool i40e_clean_tx_irq(struct i40e_vsi *vsi,i40e_arm_wb (tx_ring, vsi, budget);if (ring_is_xdp (tx_ring)) -return !!budget; + goto out;/* notify netdev of completed buffers */ netdev_tx_completed_queue(txring_txq(tx_ring),@@ -1048,6 +1050,8 @@ static bool i40e_clean_tx_irq(struct i40e_vsi *vsi, } } +out: + *tx_cleaned = total_packets;return !!budget; }@@ -2689,10 +2693,12 @@ int i40e_napi_poll(struct napi_struct *napi, int budget) container_of(napi, struct i40e_q_vector, napi);struct i40e_vsi *vsi = q_vector->vsi;struct i40e_ring *ring; +bool tx_clean_complete = true;bool clean_complete = true;bool arm_wb = false;int budget_per_ring;int work_done =0 ; +unsigned int tx_cleaned =0 ;if (test_bit (__I40E_VSI_DOWN, vsi->state)) {napi_complete (napi);@@ -2704,11 +2710,11 @@ int i40e_napi_poll(struct napi_struct *napi, int budget)*/ i40e_for_each_ring (ring, q_vector->tx) {bool wd = ring->xsk_pool ? -i40e_clean_xdp_tx_irq (vsi, ring) : -i40e_clean_tx_irq (vsi, ring, budget); +i40e_clean_xdp_tx_irq (vsi, ring, &tx_cleaned) : +i40e_clean_tx_irq (vsi, ring, budget, &tx_cleaned);if (!wd) { - clean_complete = false; + clean_complete = tx_clean_complete = false;continue ; } arm_wb |= ring->arm_wb;drivers/net/ethernet/intel/i40e/i40e_xsk.c +6
-2@@ -530,18 +530,22 @@ static void i40e_set_rs_bit(struct i40e_ring *xdp_ring)* i40e_xmit_zc - Performs zero-copy Tx AF_XDP * @xdp_ring: XDP Tx ring * @budget: NAPI budget +* @tx_cleaned: Out parameter of the TX packets processed * * Returns true if the work is finished. **/ -static bool i40e_xmit_zc (struct i40e_ring *xdp_ring,unsigned int budget) +static bool i40e_xmit_zc (struct i40e_ring *xdp_ring,unsigned int budget, +unsigned int *tx_cleaned) {struct xdp_desc *descs = xdp_ring->xsk_pool->tx_descs;u32 nb_pkts, nb_processed =0 ;unsigned int total_bytes =0 ; nb_pkts =xsk_tx_peek_release_desc_batch (xdp_ring->xsk_pool, budget); -if (!nb_pkts) +if (!nb_pkts) { + *tx_cleaned =0 ;return true; + }if (xdp_ring->next_to_use + nb_pkts >= xdp_ring->count) {Quoted 8 lines
drivers/net/ethernet/intel/i40e/i40e_xsk.c static bool i40e_xmit_zc(struct i40e_ring *xdp_ring, unsigned int budget) · line 563
Quoted 4 lines
drivers/net/ethernet/intel/i40e/i40e_xsk.c +1
-0nb_processed = xdp_ring->count - xdp_ring->next_to_use;@@ -558,6 +562,7 @@ static bool i40e_xmit_zc(struct i40e_ring *xdp_ring, unsigned int budget)i40e_update_tx_stats (xdp_ring, nb_pkts, total_bytes); + *tx_cleaned = nb_pkts;With XDP, I don't think we should count these as
tx_cleanedpackets. These are transmitted packets. Thetx_cleanedwould be thexsk_framescounter ini40e_clean_xdp_tx_irqMay be we need 2 counters for xdp.I think there's two issues you are describing, which are separate in my mind.
1.) The name "
tx_cleaned", and
2.) Whethernb_pktsis the right thing to write as the out param.For #1: I'm OK to change the name if that's the blocker here; please suggest a suitable alternative that you'll accept.
For #2:
nb_pktsis, IMO, the right value to bubble up to the tracepoint becausenb_pktsaffectsclean_completeini40e_napi_pollwhich in turn determines whether or not polling mode is entered.
Quoted 7 lines
Quoted 3 lines
The purpose of the tracepoint is to determine when/why/how you are entering polling mode, so if
nb_pktsplays a role in that calculation, it's the right number to output.I suppose the alternative is to only fire the tracepoint when *not* in XDP. Then the changes to the XDP stuff can be dropped and a separate set of tracepoints for XDP can be created in the future.
Let's be clear that it's the AF_XDP quirk that we have in here that actual xmit happens within NAPI polling routine.
Sridhar is right with having xsk_frames as tx_cleaned but you're also right that nb_pkts affects napi polling. But then if you look at Rx side there is an analogous case with buffer allocation affecting napi polling.
Quoted 3 lines
That might reduce the complexity a bit, and will probably still be pretty useful for people tuning their non-XDP workloads.
_______________________________________________
Intel-wired-lan mailing list
[email protected]
https://lists.osuosl.org/mailman/listinfo/intel-wired-lan
Re: [next-queue v2 2/4] i40e: Record number TXes cleaned during NAPI
Quoted 130 lines · Maciej Fijalkowski
Quoted 123 lines · Joe Damato
Quoted 122 lines · Joe Damato
Quoted 100 lines · Sridhar
Quoted 99 lines · Joe Damato
Update
i40e_clean_tx_irqto take an out parameter (tx_cleaned) which stores the number TXs cleaned.Likewise, update
i40e_clean_xdp_tx_irqandi40e_xmit_zcto do the same.Care has been taken to avoid changing the control flow of any functions involved.
Signed-off-by Joe Damato <[email protected]>--- drivers/net/ethernet/intel/i40e/i40e_txrx.c | 16 +++++++++++----- drivers/net/ethernet/intel/i40e/i40e_xsk.c | 15 +++++++++++---- drivers/net/ethernet/intel/i40e/i40e_xsk.h | 3 ++- 3 files changed, 24 insertions(+), 10 deletions(-)drivers/net/ethernet/intel/i40e/i40e_txrx.c +11
-5@@ -923,11 +923,13 @@ void i40e_detect_recover_hung(struct i40e_vsi *vsi)* @vsi: the VSI we care about * @tx_ring: Tx ring to clean * @napi_budget: Used to determine if we are in netpoll +* @tx_cleaned: Out parameter set to the number of TXes cleaned * * Returns true if there's any budget left (e.g. the clean is finished) **/ static bool i40e_clean_tx_irq (struct i40e_vsi *vsi, -struct i40e_ring *tx_ring,int napi_budget) +struct i40e_ring *tx_ring,int napi_budget, +unsigned int *tx_cleaned) {int i = tx_ring->next_to_clean;struct i40e_tx_buffer *tx_buf;@@ -1026,7 +1028,7 @@ static bool i40e_clean_tx_irq(struct i40e_vsi *vsi,i40e_arm_wb (tx_ring, vsi, budget);if (ring_is_xdp (tx_ring)) -return !!budget; + goto out;/* notify netdev of completed buffers */ netdev_tx_completed_queue(txring_txq(tx_ring),@@ -1048,6 +1050,8 @@ static bool i40e_clean_tx_irq(struct i40e_vsi *vsi, } } +out: + *tx_cleaned = total_packets;return !!budget; }@@ -2689,10 +2693,12 @@ int i40e_napi_poll(struct napi_struct *napi, int budget) container_of(napi, struct i40e_q_vector, napi);struct i40e_vsi *vsi = q_vector->vsi;struct i40e_ring *ring; +bool tx_clean_complete = true;bool clean_complete = true;bool arm_wb = false;int budget_per_ring;int work_done =0 ; +unsigned int tx_cleaned =0 ;if (test_bit (__I40E_VSI_DOWN, vsi->state)) {napi_complete (napi);@@ -2704,11 +2710,11 @@ int i40e_napi_poll(struct napi_struct *napi, int budget)*/ i40e_for_each_ring (ring, q_vector->tx) {bool wd = ring->xsk_pool ? -i40e_clean_xdp_tx_irq (vsi, ring) : -i40e_clean_tx_irq (vsi, ring, budget); +i40e_clean_xdp_tx_irq (vsi, ring, &tx_cleaned) : +i40e_clean_tx_irq (vsi, ring, budget, &tx_cleaned);if (!wd) { - clean_complete = false; + clean_complete = tx_clean_complete = false;continue ; } arm_wb |= ring->arm_wb;drivers/net/ethernet/intel/i40e/i40e_xsk.c +6
-2@@ -530,18 +530,22 @@ static void i40e_set_rs_bit(struct i40e_ring *xdp_ring)* i40e_xmit_zc - Performs zero-copy Tx AF_XDP * @xdp_ring: XDP Tx ring * @budget: NAPI budget +* @tx_cleaned: Out parameter of the TX packets processed * * Returns true if the work is finished. **/ -static bool i40e_xmit_zc (struct i40e_ring *xdp_ring,unsigned int budget) +static bool i40e_xmit_zc (struct i40e_ring *xdp_ring,unsigned int budget, +unsigned int *tx_cleaned) {struct xdp_desc *descs = xdp_ring->xsk_pool->tx_descs;u32 nb_pkts, nb_processed =0 ;unsigned int total_bytes =0 ; nb_pkts =xsk_tx_peek_release_desc_batch (xdp_ring->xsk_pool, budget); -if (!nb_pkts) +if (!nb_pkts) { + *tx_cleaned =0 ;return true; + }Quoted 8 lines
drivers/net/ethernet/intel/i40e/i40e_xsk.c static bool i40e_xmit_zc(struct i40e_ring *xdp_ring, unsigned int budget) · line 563
Quoted 5 lines
drivers/net/ethernet/intel/i40e/i40e_xsk.c +1
-0if (xdp_ring->next_to_use + nb_pkts >= xdp_ring->count) { nb_processed = xdp_ring->count - xdp_ring->next_to_use;@@ -558,6 +562,7 @@ static bool i40e_xmit_zc(struct i40e_ring *xdp_ring, unsigned int budget)i40e_update_tx_stats (xdp_ring, nb_pkts, total_bytes); + *tx_cleaned = nb_pkts;With XDP, I don't think we should count these as
tx_cleanedpackets. These are transmitted packets. Thetx_cleanedwould be thexsk_framescounter ini40e_clean_xdp_tx_irqMay be we need 2 counters for xdp.I think there's two issues you are describing, which are separate in my mind.
1.) The name "
tx_cleaned", and
2.) Whethernb_pktsis the right thing to write as the out param.For #1: I'm OK to change the name if that's the blocker here; please suggest a suitable alternative that you'll accept.
For #2:
nb_pktsis, IMO, the right value to bubble up to the tracepoint becausenb_pktsaffectsclean_completeini40e_napi_pollwhich in turn determines whether or not polling mode is entered.Quoted 6 lines
Quoted 3 lines
The purpose of the tracepoint is to determine when/why/how you are entering polling mode, so if
nb_pktsplays a role in that calculation, it's the right number to output.I suppose the alternative is to only fire the tracepoint when *not* in XDP. Then the changes to the XDP stuff can be dropped and a separate set of tracepoints for XDP can be created in the future.
Quoted 6 lines
Let's be clear that it's the
AF_XDPquirk that we have in here that actual xmit happens within NAPI polling routine.Sridhar is right with having
xsk_framesastx_cleanedbut you're also right thatnb_pktsaffects napi polling. But then if you look at Rx side there is an analogous case with buffer allocation affecting napi polling.
To be correct, I would suggest 2 out parameters to i40e_clean_xdp_tx_irq() tx_cleaned and xdp_transmitted. tx_cleaned should be filled in with xsk_frames. Add xdp_transmitted as an out parameter to i40e_xmit_zc() and fill it with nb_pkts.
I am not completely clear on the reasoning behind setting clean_complete based on number of packets transmitted in case of XDP.
Quoted 3 lines
Quoted 3 lines
That might reduce the complexity a bit, and will probably still be pretty useful for people tuning their non-XDP workloads.
This option is fine too.
Re: [next-queue v2 2/4] i40e: Record number TXes cleaned during NAPI
Quoted 134 lines · Sridhar
Quoted 133 lines · Maciej Fijalkowski
Quoted 123 lines · Joe Damato
Quoted 122 lines · Joe Damato
Quoted 100 lines · Sridhar
Quoted 99 lines · Joe Damato
Update
i40e_clean_tx_irqto take an out parameter (tx_cleaned) which stores the number TXs cleaned.Likewise, update
i40e_clean_xdp_tx_irqandi40e_xmit_zcto do the same.Care has been taken to avoid changing the control flow of any functions involved.
Signed-off-by Joe Damato <[email protected]>--- drivers/net/ethernet/intel/i40e/i40e_txrx.c | 16 +++++++++++----- drivers/net/ethernet/intel/i40e/i40e_xsk.c | 15 +++++++++++---- drivers/net/ethernet/intel/i40e/i40e_xsk.h | 3 ++- 3 files changed, 24 insertions(+), 10 deletions(-)drivers/net/ethernet/intel/i40e/i40e_txrx.c +11
-5@@ -923,11 +923,13 @@ void i40e_detect_recover_hung(struct i40e_vsi *vsi)* @vsi: the VSI we care about * @tx_ring: Tx ring to clean * @napi_budget: Used to determine if we are in netpoll +* @tx_cleaned: Out parameter set to the number of TXes cleaned * * Returns true if there's any budget left (e.g. the clean is finished) **/ static bool i40e_clean_tx_irq (struct i40e_vsi *vsi, -struct i40e_ring *tx_ring,int napi_budget) +struct i40e_ring *tx_ring,int napi_budget, +unsigned int *tx_cleaned) {int i = tx_ring->next_to_clean;struct i40e_tx_buffer *tx_buf;@@ -1026,7 +1028,7 @@ static bool i40e_clean_tx_irq(struct i40e_vsi *vsi,i40e_arm_wb (tx_ring, vsi, budget);if (ring_is_xdp (tx_ring)) -return !!budget; + goto out;/* notify netdev of completed buffers */ netdev_tx_completed_queue(txring_txq(tx_ring),@@ -1048,6 +1050,8 @@ static bool i40e_clean_tx_irq(struct i40e_vsi *vsi, } } +out: + *tx_cleaned = total_packets;return !!budget; }@@ -2689,10 +2693,12 @@ int i40e_napi_poll(struct napi_struct *napi, int budget) container_of(napi, struct i40e_q_vector, napi);struct i40e_vsi *vsi = q_vector->vsi;struct i40e_ring *ring; +bool tx_clean_complete = true;bool clean_complete = true;bool arm_wb = false;int budget_per_ring;int work_done =0 ; +unsigned int tx_cleaned =0 ;if (test_bit (__I40E_VSI_DOWN, vsi->state)) {napi_complete (napi);@@ -2704,11 +2710,11 @@ int i40e_napi_poll(struct napi_struct *napi, int budget)*/ i40e_for_each_ring (ring, q_vector->tx) {bool wd = ring->xsk_pool ? -i40e_clean_xdp_tx_irq (vsi, ring) : -i40e_clean_tx_irq (vsi, ring, budget); +i40e_clean_xdp_tx_irq (vsi, ring, &tx_cleaned) : +i40e_clean_tx_irq (vsi, ring, budget, &tx_cleaned);if (!wd) { - clean_complete = false; + clean_complete = tx_clean_complete = false;continue ; } arm_wb |= ring->arm_wb;drivers/net/ethernet/intel/i40e/i40e_xsk.c +6
-2@@ -530,18 +530,22 @@ static void i40e_set_rs_bit(struct i40e_ring *xdp_ring)* i40e_xmit_zc - Performs zero-copy Tx AF_XDP * @xdp_ring: XDP Tx ring * @budget: NAPI budget +* @tx_cleaned: Out parameter of the TX packets processed * * Returns true if the work is finished. **/ -static bool i40e_xmit_zc (struct i40e_ring *xdp_ring,unsigned int budget) +static bool i40e_xmit_zc (struct i40e_ring *xdp_ring,unsigned int budget, +unsigned int *tx_cleaned) {struct xdp_desc *descs = xdp_ring->xsk_pool->tx_descs;u32 nb_pkts, nb_processed =0 ;unsigned int total_bytes =0 ; nb_pkts =xsk_tx_peek_release_desc_batch (xdp_ring->xsk_pool, budget); -if (!nb_pkts) +if (!nb_pkts) { + *tx_cleaned =0 ;return true; + }Quoted 8 lines
drivers/net/ethernet/intel/i40e/i40e_xsk.c static bool i40e_xmit_zc(struct i40e_ring *xdp_ring, unsigned int budget) · line 563
Quoted 5 lines
drivers/net/ethernet/intel/i40e/i40e_xsk.c +1
-0if (xdp_ring->next_to_use + nb_pkts >= xdp_ring->count) { nb_processed = xdp_ring->count - xdp_ring->next_to_use;@@ -558,6 +562,7 @@ static bool i40e_xmit_zc(struct i40e_ring *xdp_ring, unsigned int budget)i40e_update_tx_stats (xdp_ring, nb_pkts, total_bytes); + *tx_cleaned = nb_pkts;With XDP, I don't think we should count these as
tx_cleanedpackets. These are transmitted packets. Thetx_cleanedwould be thexsk_framescounter ini40e_clean_xdp_tx_irqMay be we need 2 counters for xdp.I think there's two issues you are describing, which are separate in my mind.
1.) The name "
tx_cleaned", and
2.) Whethernb_pktsis the right thing to write as the out param.For #1: I'm OK to change the name if that's the blocker here; please suggest a suitable alternative that you'll accept.
For #2:
nb_pktsis, IMO, the right value to bubble up to the tracepoint becausenb_pktsaffectsclean_completeini40e_napi_pollwhich in turn determines whether or not polling mode is entered.Quoted 6 lines
Quoted 3 lines
The purpose of the tracepoint is to determine when/why/how you are entering polling mode, so if
nb_pktsplays a role in that calculation, it's the right number to output.I suppose the alternative is to only fire the tracepoint when *not* in XDP. Then the changes to the XDP stuff can be dropped and a separate set of tracepoints for XDP can be created in the future.
Let's be clear that it's the
AF_XDPquirk that we have in here that actual xmit happens within NAPI polling routine.
Quoted 8 lines
Quoted 3 lines
Sridhar is right with having
xsk_framesastx_cleanedbut you're also right thatnb_pktsaffects napi polling. But then if you look at Rx side there is an analogous case with buffer allocation affecting napi polling.To be correct, I would suggest 2 out parameters to
i40e_clean_xdp_tx_irq()tx_cleanedand xdp_transmitted.tx_cleanedshould be filled in withxsk_frames. Addxdp_transmittedas an out parameter toi40e_xmit_zc()and fill it withnb_pkts.
Sorry, but I don't see the value in the second param. NAPI decides what to do based on nb_pkts. That's the only parameter that matters for the purpose of NAPI going into poll mode or not, right?
If so: I don't see any reason why a second parameter is necessary.
As I mentioned earlier: if it's just that the name of the parameter isn't right (e.g., you want it to be 'tx_processed' instead of 'tx_cleaned') then that's an easy fix; I'll just change the name.
It doesn't seem helpful to have xsk_frames as an out parameter for i40e_napi_poll tracepoint; that value is not used to determine anything about i40e's NAPI.
Quoted 7 lines
I am not completely clear on the reasoning behind setting
clean_completebased on number of packets transmitted in case of XDP.Quoted 3 lines
Quoted 3 lines
That might reduce the complexity a bit, and will probably still be pretty useful for people tuning their non-XDP workloads.
This option is fine too.
I'll give Jesse a chance to weigh in before I proceed with spinning a v3.
Re: [next-queue v2 2/4] i40e: Record number TXes cleaned during NAPI
Quoted 5 lines · Joe Damato
Sorry, but I don't see the value in the second param. NAPI decides what to do based on
nb_pkts. That's the only parameter that matters for the purpose of NAPI going into poll mode or not, right?If so: I don't see any reason why a second parameter is necessary.
Sridhar and I talked about this offline. We agree now that you can just proceed with the single parameter.
Quoted 4 lines
As I mentioned earlier: if it's just that the name of the parameter isn't right (e.g., you want it to be '
tx_processed' instead of 'tx_cleaned') then that's an easy fix; I'll just change the name.
I think the name change isn't necessary, since we're not going to extend this patch with full XDP events printed (see below)
Quoted 8 lines
It doesn't seem helpful to have
xsk_framesas an out parameter fori40e_napi_polltracepoint; that value is not used to determine anything about i40e's NAPI.Quoted 3 lines
I am not completely clear on the reasoning behind setting
clean_completebased on number of packets transmitted in case of XDP.Quoted 1 line
Quoted 6 lines
Quoted 4 lines
Quoted 2 lines
Quoted 2 lines
That might reduce the complexity a bit, and will probably still be pretty useful for people tuning their non-XDP workloads.
This option is fine too.
I'll give Jesse a chance to weigh in before I proceed with spinning a v3.
I'm ok with the patch you have now, that shows nb_pkts because it's the input to the polling decision. We can add the detail about XDP transmits cleaned in a later series or patch that is by someone who wants the XDP details in the napi poll context.
_______________________________________________
Intel-wired-lan mailing list
[email protected]
https://lists.osuosl.org/mailman/listinfo/intel-wired-lan
Re: [next-queue v2 2/4] i40e: Record number TXes cleaned during NAPI
Quoted 9 lines · Jesse Brandeburg
Quoted 5 lines · Joe Damato
Sorry, but I don't see the value in the second param. NAPI decides what to do based on
nb_pkts. That's the only parameter that matters for the purpose of NAPI going into poll mode or not, right?If so: I don't see any reason why a second parameter is necessary.
Sridhar and I talked about this offline. We agree now that you can just proceed with the single parameter.
OK, thanks.
Quoted 19 lines
Quoted 4 lines
As I mentioned earlier: if it's just that the name of the parameter isn't right (e.g., you want it to be '
tx_processed' instead of 'tx_cleaned') then that's an easy fix; I'll just change the name.I think the name change isn't necessary, since we're not going to extend this patch with full XDP events printed (see below)
Quoted 11 lines
It doesn't seem helpful to have
xsk_framesas an out parameter fori40e_napi_polltracepoint; that value is not used to determine anything about i40e's NAPI.Quoted 6 lines
I am not completely clear on the reasoning behind setting
clean_completebased on number of packets transmitted in case of XDP.Quoted 3 lines
Quoted 3 lines
That might reduce the complexity a bit, and will probably still be pretty useful for people tuning their non-XDP workloads.
Quoted 8 lines
Quoted 3 lines
Quoted 1 line
This option is fine too.
I'll give Jesse a chance to weigh in before I proceed with spinning a v3.
I'm ok with the patch you have now, that shows
nb_pktsbecause it's the input to the polling decision. We can add the detail about XDP transmits cleaned in a later series or patch that is by someone who wants the XDP details in the napi poll context.
Thanks for the detailed and thoughtful feedback, it is much appreciated.
I'll leave this patch the way it is then and tweak the RX patch to include an rx_clean_complete boolean as I mentioned in my response to that patch and send out a v3.
FWIW, I had assumed that you would suggest dropping the XDP stuff so I pre-emptively spun a branch locally that dropped it... it is a much smaller change of course, but I suspect that this tracepoint might useful for XDP users, so I think the decision to leave it with nb_pkts makes sense.
Thanks again for the review. I'll send a v3 shortly.
Re: [next-queue v2 2/4] i40e: Record number TXes cleaned during NAPI
Quoted 6 lines · Joe Damato
Quoted 5 lines · Jesse Brandeburg
Quoted 4 lines · Joe Damato
Sorry, but I don't see the value in the second param. NAPI decides what to do based on
nb_pkts. That's the only parameter that matters for the purpose of NAPI going into poll mode or not, right?
Quoted 14 lines
Quoted 4 lines
Quoted 1 line
If so: I don't see any reason why a second parameter is necessary.
Sridhar and I talked about this offline. We agree now that you can just proceed with the single parameter.
OK, thanks.
Quoted 7 lines
Quoted 4 lines
As I mentioned earlier: if it's just that the name of the parameter isn't right (e.g., you want it to be '
tx_processed' instead of 'tx_cleaned') then that's an easy fix; I'll just change the name.I think the name change isn't necessary, since we're not going to extend this patch with full XDP events printed (see below)
So better to keep the twisted naming?
Quoted 12 lines
Quoted 12 lines
Quoted 12 lines
It doesn't seem helpful to have
xsk_framesas an out parameter fori40e_napi_polltracepoint; that value is not used to determine anything about i40e's NAPI.Quoted 6 lines
I am not completely clear on the reasoning behind setting
clean_completebased on number of packets transmitted in case of XDP.Quoted 3 lines
Quoted 3 lines
That might reduce the complexity a bit, and will probably still be pretty useful for people tuning their non-XDP workloads.
Quoted 8 lines
Quoted 8 lines
Quoted 3 lines
Quoted 1 line
This option is fine too.
I'll give Jesse a chance to weigh in before I proceed with spinning a v3.
I'm ok with the patch you have now, that shows
nb_pktsbecause it's the input to the polling decision. We can add the detail about XDP transmits cleaned in a later series or patch that is by someone who wants the XDP details in the napi poll context.
Please spell out whole AF_XDP instead of referring to XDP. Future readers might get confused. XDP is totally fine with what Joe is doing, I'm trying to bring up whole AF_XDP term and I feel like I'm being ignored.
number of produced packets to HW tx ring != number of produced packets to AF_XDP CQ ring.
Quoted 7 lines
Thanks for the detailed and thoughtful feedback, it is much appreciated.
I'll leave this patch the way it is then and tweak the RX patch to include an
rx_clean_completeboolean as I mentioned in my response to that patch and send out a v3.
Quoted 6 lines
FWIW, I had assumed that you would suggest dropping the XDP stuff so I pre-emptively spun a branch locally that dropped it... it is a much smaller change of course, but I suspect that this tracepoint might useful for XDP users, so I think the decision to leave it with
nb_pktsmakes sense.Thanks again for the review. I'll send a v3 shortly.
_______________________________________________
Intel-wired-lan mailing list
[email protected]
https://lists.osuosl.org/mailman/listinfo/intel-wired-lan
PATCH v2 3/4 i40e: Record number of RXes cleaned during NAPI
In reply to Joe DamatoAdjust i40e_clean_rx_irq and i40e_clean_rx_irq_zc to accept an out parameter which records the number of RX packets cleaned.
Care has been taken to avoid any changes in control flow.
--- drivers/net/ethernet/intel/i40e/i40e_txrx.c | 11 ++++++++--- drivers/net/ethernet/intel/i40e/i40e_xsk.c | 6 +++++- drivers/net/ethernet/intel/i40e/i40e_xsk.h | 3 ++- 3 files changed, 15 insertions(+), 5 deletions(-)
drivers/net/ethernet/intel/i40e/i40e_txrx.c +8 -3
@@ -2426,6 +2426,7 @@ static void i40e_inc_ntc(struct i40e_ring *rx_ring)* i40e_clean_rx_irq - Clean completed descriptors from Rx ring - bounce buf * @rx_ring: rx descriptor ring to transact packets on * @budget: Total limit on number of packets to process +* @rx_cleaned: Out parameter of the number of packets processed * * This function provides a "bounce buffer" approach to Rx interrupt * processing. The advantage to this is that on systems that have
@@ -2434,7 +2435,8 @@ static void i40e_inc_ntc(struct i40e_ring *rx_ring)* * Returns amount of work completed **/ -static int i40e_clean_rx_irq (struct i40e_ring *rx_ring,int budget) +static int i40e_clean_rx_irq (struct i40e_ring *rx_ring,int budget, +unsigned int *rx_cleaned) {unsigned int total_rx_bytes =0 , total_rx_packets =0 , frame_sz =0 ;u16 cleaned_count =I40E_DESC_UNUSED (rx_ring);
@@ -2571,6 +2573,8 @@ static int i40e_clean_rx_irq(struct i40e_ring *rx_ring, int budget)i40e_update_rx_stats (rx_ring, total_rx_bytes, total_rx_packets); + *rx_cleaned = total_rx_packets; +/* guarantee a trip back through this routine if there was a failure */ return failure ? budget : (int )total_rx_packets; }
@@ -2699,6 +2703,7 @@ int i40e_napi_poll(struct napi_struct *napi, int budget)int budget_per_ring;int work_done =0 ;unsigned int tx_cleaned =0 ; +unsigned int rx_cleaned =0 ;if (test_bit (__I40E_VSI_DOWN, vsi->state)) {napi_complete (napi);
@@ -2738,8 +2743,8 @@ int i40e_napi_poll(struct napi_struct *napi, int budget)i40e_for_each_ring (ring, q_vector->rx) {int cleaned = ring->xsk_pool ? -i40e_clean_rx_irq_zc (ring, budget_per_ring) : -i40e_clean_rx_irq (ring, budget_per_ring); +i40e_clean_rx_irq_zc (ring, budget_per_ring, &rx_cleaned) : +i40e_clean_rx_irq (ring, budget_per_ring, &rx_cleaned); work_done += cleaned;/* if we clean as many as budgeted, we must not be done */
drivers/net/ethernet/intel/i40e/i40e_xsk.c +5 -1
@@ -378,10 +378,12 @@ static void i40e_handle_xdp_result_zc(struct i40e_ring *rx_ring,* i40e_clean_rx_irq_zc - Consumes Rx packets from the hardware ring * @rx_ring: Rx ring * @budget: NAPI budget +* @rx_cleaned: out parameter of the packets processed * * Returns amount of work completed **/ -int i40e_clean_rx_irq_zc (struct i40e_ring *rx_ring,int budget) +int i40e_clean_rx_irq_zc (struct i40e_ring *rx_ring,int budget, +unsigned int *rx_cleaned) {unsigned int total_rx_bytes =0 , total_rx_packets =0 ;u16 next_to_clean = rx_ring->next_to_clean;
@@ -452,6 +454,8 @@ int i40e_clean_rx_irq_zc(struct i40e_ring *rx_ring, int budget)i40e_finalize_xdp_rx (rx_ring, xdp_xmit);i40e_update_rx_stats (rx_ring, total_rx_bytes, total_rx_packets); + *rx_cleaned = total_rx_packets; +if (xsk_uses_need_wakeup (rx_ring->xsk_pool)) {if (failure || next_to_clean == rx_ring->next_to_use)xsk_set_rx_need_wakeup (rx_ring->xsk_pool);
drivers/net/ethernet/intel/i40e/i40e_xsk.h +2 -1
@@ -28,7 +28,8 @@ int i40e_queue_pair_enable(struct i40e_vsi *vsi, int queue_pair);int i40e_xsk_pool_setup (struct i40e_vsi *vsi,struct xsk_buff_pool *pool,u16 qid);bool i40e_alloc_rx_buffers_zc (struct i40e_ring *rx_ring,u16 cleaned_count); -int i40e_clean_rx_irq_zc (struct i40e_ring *rx_ring,int budget); +int i40e_clean_rx_irq_zc (struct i40e_ring *rx_ring,int budget, +unsigned int *rx_cleaned);bool i40e_clean_xdp_tx_irq (struct i40e_vsi *vsi,struct i40e_ring *tx_ring,unsigned int *tx_cleaned);
Signature · 1 line
2.7.4
Re: [next-queue v2 3/4] i40e: Record number of RXes cleaned during NAPI
Quoted 2 lines · Joe Damato
Adjust
i40e_clean_rx_irqandi40e_clean_rx_irq_zcto accept an out parameter which records the number of RX packets cleaned.
I just realized that this change probably also needs to include an "rx_clean_complete" as was added in the previous patch for the TX case so that when the tracepoint is hit it will be more clear which of the two (RX or TX) triggered clean_complete = false.
I think the tracepoint should have separate bool flags for each of these cases (but neither will be used to modify control flow as Jesse asked earlier).
I'll leave that fix for the v3, in addition to addressing any other feedback on the rest of the changes.
_______________________________________________
Intel-wired-lan mailing list
[email protected]
https://lists.osuosl.org/mailman/listinfo/intel-wired-lan
PATCH v2 4/4 i40e: Add i40e_napi_poll tracepoint
In reply to Joe DamatoAdd a tracepoint for i40e_napi_poll that allows users to get detailed information about the amount of work done. This information can help users better tune the correct NAPI parameters (like weight and budget), as well as debug NIC settings like rx-usecs and tx-usecs, etc.
An example of the output from this tracepoint:
$ sudo perf trace -e i40e:i40e_napi_poll -a --call-graph=fp --libtraceevent_print
[..snip..]
388.258 :0/0 i40e:i40e_napi_poll(i40e_napi_poll on dev eth2 q i40e-eth2-TxRx-9 irq 346 irq_mask 00000000,00000000,00000000,00000000,00000000,00800000 curr_cpu 23 budget 64 bpr 64 rx_cleaned 28 tx_cleaned 0 rx_clean_complete 1 tx_clean_complete 1)
i40e_napi_poll ([i40e]) i40e_napi_poll ([i40e]) __napi_poll ([kernel.kallsyms]) net_rx_action ([kernel.kallsyms]) __do_softirq ([kernel.kallsyms]) common_interrupt ([kernel.kallsyms]) asm_common_interrupt ([kernel.kallsyms]) intel_idle_irq ([kernel.kallsyms]) cpuidle_enter_state ([kernel.kallsyms]) cpuidle_enter ([kernel.kallsyms]) do_idle ([kernel.kallsyms]) cpu_startup_entry ([kernel.kallsyms]) [0x243fd8] ([kernel.kallsyms]) secondary_startup_64_no_verify ([kernel.kallsyms])
--- drivers/net/ethernet/intel/i40e/i40e_trace.h | 49 ++++++++++++++++++++++++++++ drivers/net/ethernet/intel/i40e/i40e_txrx.c | 3 ++ 2 files changed, 52 insertions(+)
drivers/net/ethernet/intel/i40e/i40e_trace.h +49 -0
@@ -55,6 +55,55 @@* being built from shared code. */ +#define NO_DEV "(i40e no_device)" + +TRACE_EVENT (i40e_napi_poll, + + TP_PROTO(struct napi_struct *napi, struct i40e_q_vector *q, int budget, + int budget_per_ring, unsigned int rx_cleaned, unsigned int tx_cleaned, + bool rx_clean_complete, bool tx_clean_complete), + +TP_ARGS (napi, q, budget, budget_per_ring, rx_cleaned, tx_cleaned, + rx_clean_complete, tx_clean_complete), + +TP_STRUCT__entry ( + __field(int, budget) + __field(int, budget_per_ring) + __field(unsigned int, rx_cleaned) + __field(unsigned int, tx_cleaned) + __field(int, rx_clean_complete) + __field(int, tx_clean_complete) + __field(int, irq_num) + __field(int, curr_cpu) + __string(qname, q->name) + __string(dev_name, napi->dev ? napi->dev->name : NO_DEV) +__bitmask (irq_affinity, nr_cpumask_bits) + ), + +TP_fast_assign ( + __entry->budget = budget; + __entry->budget_per_ring = budget_per_ring; + __entry->rx_cleaned = rx_cleaned; + __entry->tx_cleaned = tx_cleaned; + __entry->rx_clean_complete = rx_clean_complete; + __entry->tx_clean_complete = tx_clean_complete; + __entry->irq_num = q->irq_num; + __entry->curr_cpu = get_cpu(); + __assign_str(qname, q->name); + __assign_str(dev_name, napi->dev ? napi->dev->name : NO_DEV); +__assign_bitmask (irq_affinity,cpumask_bits (&q->affinity_mask), + nr_cpumask_bits); + ), + +TP_printk ("i40e_napi_poll on dev %s q %s irq %d irq_mask %s curr_cpu %d " +"budget %d bpr %d rx_cleaned %lu tx_cleaned %lu " +"rx_clean_complete %d tx_clean_complete %d" , +__get_str (dev_name),__get_str (qname), __entry->irq_num, +__get_bitmask (irq_affinity), __entry->curr_cpu, __entry->budget, + __entry->budget_per_ring, __entry->rx_cleaned, __entry->tx_cleaned, + __entry->rx_clean_complete, __entry->tx_clean_complete) +); +/* Events related to a vsi & ring */ DECLARE_EVENT_CLASS( i40e_tx_template,
drivers/net/ethernet/intel/i40e/i40e_txrx.c +3 -0
@@ -2752,6 +2752,9 @@ int i40e_napi_poll(struct napi_struct *napi, int budget) clean_complete = false; } +trace_i40e_napi_poll (napi, q_vector, budget, budget_per_ring, rx_cleaned, + tx_cleaned, clean_complete, tx_clean_complete); +/* If work not completed, return budget and polling will return */ if (!clean_complete) {int cpu_id =smp_processor_id ();
Signature · 1 line
2.7.4