diff mbox series

[v3,2/2] net/mlx5: fix Rx descriptor status returned value

Message ID 20201116170258.18583-3-maxime.leroy@6wind.com (mailing list archive)
State Superseded, archived
Delegated to: Raslan Darawsheh
Headers show
Series net/mlx5: fixes for rx queue count calculation | expand

Checks

Context Check Description
ci/travis-robot success Travis build: passed
ci/iol-mellanox-Performance success Performance Testing PASS
ci/iol-intel-Performance success Performance Testing PASS
ci/iol-intel-Functional fail Functional Testing issues
ci/iol-testing success Testing PASS
ci/iol-broadcom-Functional success Functional Testing PASS
ci/iol-broadcom-Performance success Performance Testing PASS
ci/Intel-compilation success Compilation OK
ci/checkpatch warning coding style issues

Commit Message

Maxime Leroy Nov. 16, 2020, 5:02 p.m. UTC
From: Didier Pallard <didier.pallard@6wind.com>

Three bugs in rx_queue_count function:
- One entry may contain several segments, so 'used' must be multiplied
  by number of segments per entry to properly reflect the queue usage.
- The number of cqes is equals to (1U << rxq->elts_n) - 1 in non mqrt
  mode. The range returned by rx_queue_count should be the number of
  entries used in queue, so it ranges from 0 to max number of entries
  in queue, not this number minus one.
- For MQRT mode, we need to take acount of the number of strd.

Fixes: 8788fec1f269 ("net/mlx5: implement descriptor status API")
Signed-off-by: Didier Pallard <didier.pallard@6wind.com>
Signed-off-by: Maxime Leroy <maxime.leroy@6wind.com>
---
 drivers/net/mlx5/mlx5_rxtx.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

Comments

Viacheslav Ovsiienko Nov. 16, 2020, 8:03 p.m. UTC | #1
Hi, Maxime

It seems there are some typos (see below).
Beside these ones - ACKed.

> -----Original Message-----
> From: Maxime Leroy <maxime.leroy@6wind.com>
> Sent: Monday, November 16, 2020 19:03
> To: Matan Azrad <matan@nvidia.com>; Shahaf Shuler <shahafs@nvidia.com>;
> Slava Ovsiienko <viacheslavo@nvidia.com>; Olivier Matz
> <olivier.matz@6wind.com>
> Cc: dev@dpdk.org; Didier Pallard <didier.pallard@6wind.com>
> Subject: [PATCH v3 2/2] net/mlx5: fix Rx descriptor status returned value
> 
> From: Didier Pallard <didier.pallard@6wind.com>
> 
> Three bugs in rx_queue_count function:
> - One entry may contain several segments, so 'used' must be multiplied
>   by number of segments per entry to properly reflect the queue usage.
> - The number of cqes is equals to (1U << rxq->elts_n) - 1 in non mqrt
>   mode. The range returned by rx_queue_count should be the number of

Should it be "bullet number 3" ?

>   entries used in queue, so it ranges from 0 to max number of entries
>   in queue, not this number minus one.

WARNING:TYPO_SPELLING: 'acount' may be misspelled - perhaps 'account'?
Should it be - "to take into account?"
Should it be "MPRQ" ?

> - For MQRT mode, we need to take acount of the number of strd.
> 
> Fixes: 8788fec1f269 ("net/mlx5: implement descriptor status API")
> Signed-off-by: Didier Pallard <didier.pallard@6wind.com>
> Signed-off-by: Maxime Leroy <maxime.leroy@6wind.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
Maxime Leroy Nov. 17, 2020, 8:52 a.m. UTC | #2
Hi Slava,

On Mon, Nov 16, 2020 at 9:03 PM Slava Ovsiienko <viacheslavo@nvidia.com> wrote:
>
> Hi, Maxime
>
> It seems there are some typos (see below).
> Beside these ones - ACKed.
>
> > -----Original Message-----
> > From: Maxime Leroy <maxime.leroy@6wind.com>
> > Sent: Monday, November 16, 2020 19:03
> > To: Matan Azrad <matan@nvidia.com>; Shahaf Shuler <shahafs@nvidia.com>;
> > Slava Ovsiienko <viacheslavo@nvidia.com>; Olivier Matz
> > <olivier.matz@6wind.com>
> > Cc: dev@dpdk.org; Didier Pallard <didier.pallard@6wind.com>
> > Subject: [PATCH v3 2/2] net/mlx5: fix Rx descriptor status returned value
> >
> > From: Didier Pallard <didier.pallard@6wind.com>
> >
> > Three bugs in rx_queue_count function:
> > - One entry may contain several segments, so 'used' must be multiplied
> >   by number of segments per entry to properly reflect the queue usage.
> > - The number of cqes is equals to (1U << rxq->elts_n) - 1 in non mqrt
> >   mode. The range returned by rx_queue_count should be the number of
>
> Should it be "bullet number 3" ?

This bullet is for SPRQ mode. The bullet number 3 is for MPRQ mode.
Not sure to understand your point.

>
> >   entries used in queue, so it ranges from 0 to max number of entries
> >   in queue, not this number minus one.
>
> WARNING:TYPO_SPELLING: 'acount' may be misspelled - perhaps 'account'?
> Should it be - "to take into account?"
> Should it be "MPRQ" ?

I will fix it in V4.

Thanks for the review.

>
> > - For MQRT mode, we need to take acount of the number of strd.
> >
> > Fixes: 8788fec1f269 ("net/mlx5: implement descriptor status API")
> > Signed-off-by: Didier Pallard <didier.pallard@6wind.com>
> > Signed-off-by: Maxime Leroy <maxime.leroy@6wind.com>
> Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
>
Viacheslav Ovsiienko Nov. 17, 2020, 9:17 a.m. UTC | #3
Hi, Maxime

> -----Original Message-----
> From: Maxime Leroy <maxime.leroy@6wind.com>
> Sent: Tuesday, November 17, 2020 10:52
> To: Slava Ovsiienko <viacheslavo@nvidia.com>
> Cc: Matan Azrad <matan@nvidia.com>; Shahaf Shuler <shahafs@nvidia.com>;
> Olivier Matz <olivier.matz@6wind.com>; dev@dpdk.org; Didier Pallard
> <didier.pallard@6wind.com>
> Subject: Re: [PATCH v3 2/2] net/mlx5: fix Rx descriptor status returned value
> 
> Hi Slava,
> 
> On Mon, Nov 16, 2020 at 9:03 PM Slava Ovsiienko <viacheslavo@nvidia.com>
> wrote:
> >
> > Hi, Maxime
> >
> > It seems there are some typos (see below).
> > Beside these ones - ACKed.
> >
> > > -----Original Message-----
> > > From: Maxime Leroy <maxime.leroy@6wind.com>
> > > Sent: Monday, November 16, 2020 19:03
> > > To: Matan Azrad <matan@nvidia.com>; Shahaf Shuler
> > > <shahafs@nvidia.com>; Slava Ovsiienko <viacheslavo@nvidia.com>;
> > > Olivier Matz <olivier.matz@6wind.com>
> > > Cc: dev@dpdk.org; Didier Pallard <didier.pallard@6wind.com>
> > > Subject: [PATCH v3 2/2] net/mlx5: fix Rx descriptor status returned
> > > value
> > >
> > > From: Didier Pallard <didier.pallard@6wind.com>
> > >
> > > Three bugs in rx_queue_count function:
> > > - One entry may contain several segments, so 'used' must be multiplied
> > >   by number of segments per entry to properly reflect the queue usage.
> > > - The number of cqes is equals to (1U << rxq->elts_n) - 1 in non mqrt
> > >   mode. The range returned by rx_queue_count should be the number of
> >
> > Should it be "bullet number 3" ?
> 
> This bullet is for SPRQ mode. The bullet number 3 is for MPRQ mode.
> Not sure to understand your point.
Sorry, my bad - Now I see the third bullet at the MPRQ, so - please, disregard this my point:
> > Should it be "bullet number 3" ?

With best regards, Slava 
> >
> > >   entries used in queue, so it ranges from 0 to max number of entries
> > >   in queue, not this number minus one.
> >
> > WARNING:TYPO_SPELLING: 'acount' may be misspelled - perhaps 'account'?
> > Should it be - "to take into account?"
> > Should it be "MPRQ" ?
> 
> I will fix it in V4.
> 
> Thanks for the review.
> 
> >
> > > - For MQRT mode, we need to take acount of the number of strd.
> > >
> > > Fixes: 8788fec1f269 ("net/mlx5: implement descriptor status API")
> > > Signed-off-by: Didier Pallard <didier.pallard@6wind.com>
> > > Signed-off-by: Maxime Leroy <maxime.leroy@6wind.com>
> > Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
> >
diff mbox series

Patch

diff --git a/drivers/net/mlx5/mlx5_rxtx.c b/drivers/net/mlx5/mlx5_rxtx.c
index 2733dcd3..2ecf901f 100644
--- a/drivers/net/mlx5/mlx5_rxtx.c
+++ b/drivers/net/mlx5/mlx5_rxtx.c
@@ -463,6 +463,9 @@  rx_queue_count(struct mlx5_rxq_data *rxq)
 	struct rxq_zip *zip = &rxq->zip;
 	volatile struct mlx5_cqe *cqe;
 	const unsigned int cqe_n = (1 << rxq->cqe_n);
+	const unsigned int sges_n = (1 << rxq->sges_n);
+	const unsigned int elts_n = (1 << rxq->elts_n);
+	const unsigned int strd_n = (1 << rxq->strd_num_n);
 	const unsigned int cqe_cnt = cqe_n - 1;
 	unsigned int cq_ci, used;
 
@@ -488,7 +491,7 @@  rx_queue_count(struct mlx5_rxq_data *rxq)
 		used += n;
 		cqe = &(*rxq->cqes)[cq_ci & cqe_cnt];
 	}
-	used = RTE_MIN(used, cqe_n);
+	used = RTE_MIN(used * sges_n, elts_n * strd_n);
 	return used;
 }