In file included from ../drivers/net/virtio/virtio_rxtx_packed.c:22:0: ../drivers/net/virtio/virtio_rxtx_packed_neon.h: In function ‘virtqueue_enqueue_batch_packed_vec’: ../drivers/net/virtio/virtio_rxtx_packed_neon.h:74:2: warning: implicit declaration of function ‘vreinterpretq_p128_u32’ [-Wimplicit-function-declaration] poly128_t cmp1 = vreinterpretq_p128_u32(~vceqq_u32(ref_seg, def_ref_seg)); ^ The message shows ‘vreinterpretq_p128_u32’ instrisic is not supported because an old version gcc (gcc 4.8.5) was used. So fix the issue with implementing the logic with other intrinsics. Bugzilla ID: 621 Fixes: 530887469350 ("net/virtio: add vectorized packed ring NEON Tx") Fixes: 5971ce5e2a59 ("net/virtio: add vectorized packed ring NEON Rx") Cc: stable@dpdk.org Signed-off-by: Joyce Kong <joyce.kong@arm.com> --- drivers/net/virtio/virtio_rxtx_packed_neon.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/virtio/virtio_rxtx_packed_neon.h b/drivers/net/virtio/virtio_rxtx_packed_neon.h index 01c77b712..00dd04277 100644 --- a/drivers/net/virtio/virtio_rxtx_packed_neon.h +++ b/drivers/net/virtio/virtio_rxtx_packed_neon.h @@ -71,8 +71,8 @@ virtqueue_enqueue_batch_packed_vec(struct virtnet_tx *txvq, uint32x4_t def_ref_seg = vdupq_n_u32(0x10001); /* Check refcnt and nb_segs. */ uint32x4_t ref_seg = vreinterpretq_u32_u8(vqtbl2q_u8(mbuf, ref_seg_msk)); - poly128_t cmp1 = vreinterpretq_p128_u32(~vceqq_u32(ref_seg, def_ref_seg)); - if (unlikely(cmp1)) + uint64x2_t cmp1 = vreinterpretq_u64_u32(~vceqq_u32(ref_seg, def_ref_seg)); + if (unlikely(vgetq_lane_u64(cmp1, 0) || vgetq_lane_u64(cmp1, 1))) return -1; /* Check headroom is enough. */ @@ -225,10 +225,10 @@ virtqueue_dequeue_batch_packed_vec(struct virtnet_rx *rxvq, if (vq->vq_packed.used_wrap_counter) v_used_flag = vdupq_n_u32(PACKED_FLAGS_MASK); - poly128_t desc_stats = vreinterpretq_p128_u32(~vceqq_u32(v_flag, v_used_flag)); + uint64x2_t desc_stats = vreinterpretq_u64_u32(~vceqq_u32(v_flag, v_used_flag)); /* Check all descs are used. */ - if (desc_stats) + if (unlikely(vgetq_lane_u64(desc_stats, 0) || vgetq_lane_u64(desc_stats, 1))) return -1; /* Load 2 mbuf pointers per time. */ -- 2.30.0
On Tue, Jan 26, 2021 at 10:20 AM Joyce Kong <joyce.kong@arm.com> wrote: > > In file included from ../drivers/net/virtio/virtio_rxtx_packed.c:22:0: > ../drivers/net/virtio/virtio_rxtx_packed_neon.h: In function > ‘virtqueue_enqueue_batch_packed_vec’: > ../drivers/net/virtio/virtio_rxtx_packed_neon.h:74:2: warning: > implicit declaration of function ‘vreinterpretq_p128_u32’ > [-Wimplicit-function-declaration] > poly128_t cmp1 = vreinterpretq_p128_u32(~vceqq_u32(ref_seg, def_ref_seg)); > ^ > > The message shows ‘vreinterpretq_p128_u32’ instrisic is not supported intrinsic* > because an old version gcc (gcc 4.8.5) was used. So fix the issue with > implementing the logic with other intrinsics. > > Bugzilla ID: 621 > Fixes: 530887469350 ("net/virtio: add vectorized packed ring NEON Tx") > Fixes: 5971ce5e2a59 ("net/virtio: add vectorized packed ring NEON Rx") > Cc: stable@dpdk.org No reason to copy stable@dpdk.org. Cc: reporter. -- David Marchand
In file included from ../drivers/net/virtio/virtio_rxtx_packed.c:22:0: ../drivers/net/virtio/virtio_rxtx_packed_neon.h: In function ‘virtqueue_enqueue_batch_packed_vec’: ../drivers/net/virtio/virtio_rxtx_packed_neon.h:74:2: warning: implicit declaration of function ‘vreinterpretq_p128_u32’ [-Wimplicit-function-declaration] poly128_t cmp1 = vreinterpretq_p128_u32(~vceqq_u32(ref_seg, def_ref_seg)); ^ The message shows ‘vreinterpretq_p128_u32’ intrinsic is not supported because an old version gcc (gcc 4.8.5) is used. So fix the issue with implementing the logic with other intrinsics. Bugzilla ID: 621 Fixes: 530887469350 ("net/virtio: add vectorized packed ring NEON Tx") Fixes: 5971ce5e2a59 ("net/virtio: add vectorized packed ring NEON Rx") Reported-by: Subhi Masri <subhim@nvidia.com> Signed-off-by: Joyce Kong <joyce.kong@arm.com> --- drivers/net/virtio/virtio_rxtx_packed_neon.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/drivers/net/virtio/virtio_rxtx_packed_neon.h b/drivers/net/virtio/virtio_rxtx_packed_neon.h index 01c77b712..00dd04277 100644 --- a/drivers/net/virtio/virtio_rxtx_packed_neon.h +++ b/drivers/net/virtio/virtio_rxtx_packed_neon.h @@ -71,8 +71,8 @@ virtqueue_enqueue_batch_packed_vec(struct virtnet_tx *txvq, uint32x4_t def_ref_seg = vdupq_n_u32(0x10001); /* Check refcnt and nb_segs. */ uint32x4_t ref_seg = vreinterpretq_u32_u8(vqtbl2q_u8(mbuf, ref_seg_msk)); - poly128_t cmp1 = vreinterpretq_p128_u32(~vceqq_u32(ref_seg, def_ref_seg)); - if (unlikely(cmp1)) + uint64x2_t cmp1 = vreinterpretq_u64_u32(~vceqq_u32(ref_seg, def_ref_seg)); + if (unlikely(vgetq_lane_u64(cmp1, 0) || vgetq_lane_u64(cmp1, 1))) return -1; /* Check headroom is enough. */ @@ -225,10 +225,10 @@ virtqueue_dequeue_batch_packed_vec(struct virtnet_rx *rxvq, if (vq->vq_packed.used_wrap_counter) v_used_flag = vdupq_n_u32(PACKED_FLAGS_MASK); - poly128_t desc_stats = vreinterpretq_p128_u32(~vceqq_u32(v_flag, v_used_flag)); + uint64x2_t desc_stats = vreinterpretq_u64_u32(~vceqq_u32(v_flag, v_used_flag)); /* Check all descs are used. */ - if (desc_stats) + if (unlikely(vgetq_lane_u64(desc_stats, 0) || vgetq_lane_u64(desc_stats, 1))) return -1; /* Load 2 mbuf pointers per time. */ -- 2.30.0
> -----Original Message-----
> From: Joyce Kong <joyce.kong@arm.com>
> Sent: Tuesday, January 26, 2021 5:58 PM
> To: maxime.coquelin@redhat.com; david.marchand@redhat.com;
> i.maximets@ovn.org; Ruifeng Wang <Ruifeng.Wang@arm.com>; Honnappa
> Nagarahalli <Honnappa.Nagarahalli@arm.com>
> Cc: dev@dpdk.org; nd <nd@arm.com>; Subhi Masri <subhim@nvidia.com>
> Subject: [PATCH v2] net/virtio: fix compiling issue for vectorized NEON path
>
> In file included from ../drivers/net/virtio/virtio_rxtx_packed.c:22:0:
> ../drivers/net/virtio/virtio_rxtx_packed_neon.h: In function
> ‘virtqueue_enqueue_batch_packed_vec’:
> ../drivers/net/virtio/virtio_rxtx_packed_neon.h:74:2: warning:
> implicit declaration of function ‘vreinterpretq_p128_u32’
> [-Wimplicit-function-declaration]
> poly128_t cmp1 = vreinterpretq_p128_u32(~vceqq_u32(ref_seg,
> def_ref_seg)); ^
>
> The message shows ‘vreinterpretq_p128_u32’ intrinsic is not supported
> because an old version gcc (gcc 4.8.5) is used. So fix the issue with
> implementing the logic with other intrinsics.
>
> Bugzilla ID: 621
> Fixes: 530887469350 ("net/virtio: add vectorized packed ring NEON Tx")
> Fixes: 5971ce5e2a59 ("net/virtio: add vectorized packed ring NEON Rx")
>
> Reported-by: Subhi Masri <subhim@nvidia.com>
> Signed-off-by: Joyce Kong <joyce.kong@arm.com>
> ---
> drivers/net/virtio/virtio_rxtx_packed_neon.h | 8 ++++----
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/net/virtio/virtio_rxtx_packed_neon.h
> b/drivers/net/virtio/virtio_rxtx_packed_neon.h
> index 01c77b712..00dd04277 100644
> --- a/drivers/net/virtio/virtio_rxtx_packed_neon.h
> +++ b/drivers/net/virtio/virtio_rxtx_packed_neon.h
> @@ -71,8 +71,8 @@ virtqueue_enqueue_batch_packed_vec(struct
> virtnet_tx *txvq,
> uint32x4_t def_ref_seg = vdupq_n_u32(0x10001);
> /* Check refcnt and nb_segs. */
> uint32x4_t ref_seg = vreinterpretq_u32_u8(vqtbl2q_u8(mbuf,
> ref_seg_msk));
> - poly128_t cmp1 = vreinterpretq_p128_u32(~vceqq_u32(ref_seg,
> def_ref_seg));
> - if (unlikely(cmp1))
> + uint64x2_t cmp1 = vreinterpretq_u64_u32(~vceqq_u32(ref_seg,
> def_ref_seg));
> + if (unlikely(vgetq_lane_u64(cmp1, 0) || vgetq_lane_u64(cmp1, 1)))
> return -1;
>
> /* Check headroom is enough. */
> @@ -225,10 +225,10 @@ virtqueue_dequeue_batch_packed_vec(struct
> virtnet_rx *rxvq,
> if (vq->vq_packed.used_wrap_counter)
> v_used_flag = vdupq_n_u32(PACKED_FLAGS_MASK);
>
> - poly128_t desc_stats = vreinterpretq_p128_u32(~vceqq_u32(v_flag,
> v_used_flag));
> + uint64x2_t desc_stats = vreinterpretq_u64_u32(~vceqq_u32(v_flag,
> +v_used_flag));
>
> /* Check all descs are used. */
> - if (desc_stats)
> + if (unlikely(vgetq_lane_u64(desc_stats, 0) ||
> +vgetq_lane_u64(desc_stats, 1)))
> return -1;
>
> /* Load 2 mbuf pointers per time. */
> --
> 2.30.0
Reviewed-by: Ruifeng Wang <ruifeng.wang@arm.com>
Hi Subhi,
On 1/26/21 10:57 AM, Joyce Kong wrote:
> In file included from ../drivers/net/virtio/virtio_rxtx_packed.c:22:0:
> ../drivers/net/virtio/virtio_rxtx_packed_neon.h: In function
> ‘virtqueue_enqueue_batch_packed_vec’:
> ../drivers/net/virtio/virtio_rxtx_packed_neon.h:74:2: warning:
> implicit declaration of function ‘vreinterpretq_p128_u32’
> [-Wimplicit-function-declaration]
> poly128_t cmp1 = vreinterpretq_p128_u32(~vceqq_u32(ref_seg, def_ref_seg));
> ^
>
> The message shows ‘vreinterpretq_p128_u32’ intrinsic is not supported
> because an old version gcc (gcc 4.8.5) is used. So fix the issue with
> implementing the logic with other intrinsics.
>
> Bugzilla ID: 621
> Fixes: 530887469350 ("net/virtio: add vectorized packed ring NEON Tx")
> Fixes: 5971ce5e2a59 ("net/virtio: add vectorized packed ring NEON Rx")
>
> Reported-by: Subhi Masri <subhim@nvidia.com>
> Signed-off-by: Joyce Kong <joyce.kong@arm.com>
> ---
> drivers/net/virtio/virtio_rxtx_packed_neon.h | 8 ++++----
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/net/virtio/virtio_rxtx_packed_neon.h b/drivers/net/virtio/virtio_rxtx_packed_neon.h
> index 01c77b712..00dd04277 100644
> --- a/drivers/net/virtio/virtio_rxtx_packed_neon.h
> +++ b/drivers/net/virtio/virtio_rxtx_packed_neon.h
> @@ -71,8 +71,8 @@ virtqueue_enqueue_batch_packed_vec(struct virtnet_tx *txvq,
> uint32x4_t def_ref_seg = vdupq_n_u32(0x10001);
> /* Check refcnt and nb_segs. */
> uint32x4_t ref_seg = vreinterpretq_u32_u8(vqtbl2q_u8(mbuf, ref_seg_msk));
> - poly128_t cmp1 = vreinterpretq_p128_u32(~vceqq_u32(ref_seg, def_ref_seg));
> - if (unlikely(cmp1))
> + uint64x2_t cmp1 = vreinterpretq_u64_u32(~vceqq_u32(ref_seg, def_ref_seg));
> + if (unlikely(vgetq_lane_u64(cmp1, 0) || vgetq_lane_u64(cmp1, 1)))
> return -1;
>
> /* Check headroom is enough. */
> @@ -225,10 +225,10 @@ virtqueue_dequeue_batch_packed_vec(struct virtnet_rx *rxvq,
> if (vq->vq_packed.used_wrap_counter)
> v_used_flag = vdupq_n_u32(PACKED_FLAGS_MASK);
>
> - poly128_t desc_stats = vreinterpretq_p128_u32(~vceqq_u32(v_flag, v_used_flag));
> + uint64x2_t desc_stats = vreinterpretq_u64_u32(~vceqq_u32(v_flag, v_used_flag));
>
> /* Check all descs are used. */
> - if (desc_stats)
> + if (unlikely(vgetq_lane_u64(desc_stats, 0) || vgetq_lane_u64(desc_stats, 1)))
> return -1;
>
> /* Load 2 mbuf pointers per time. */
>
Can you confirm the patch works for you?
Thanks,
Maxime
Hi, > -----Original Message----- > From: Maxime Coquelin <maxime.coquelin@redhat.com> > Sent: Wednesday, January 27, 2021 10:44 AM > To: Subhi Masri <subhim@nvidia.com> > Cc: dev@dpdk.org; nd@arm.com; Joyce Kong <joyce.kong@arm.com>; > david.marchand@redhat.com; honnappa.nagarahalli@arm.com; > ruifeng.wang@arm.com; i.maximets@ovn.org > Subject: Re: [PATCH v2] net/virtio: fix compiling issue for vectorized NEON > path > > External email: Use caution opening links or attachments > > > Hi Subhi, > > On 1/26/21 10:57 AM, Joyce Kong wrote: > > In file included from ../drivers/net/virtio/virtio_rxtx_packed.c:22:0: > > ../drivers/net/virtio/virtio_rxtx_packed_neon.h: In function > > ‘virtqueue_enqueue_batch_packed_vec’: > > ../drivers/net/virtio/virtio_rxtx_packed_neon.h:74:2: warning: > > implicit declaration of function ‘vreinterpretq_p128_u32’ > > [-Wimplicit-function-declaration] > > poly128_t cmp1 = vreinterpretq_p128_u32(~vceqq_u32(ref_seg, > > def_ref_seg)); ^ > > > > The message shows ‘vreinterpretq_p128_u32’ intrinsic is not supported > > because an old version gcc (gcc 4.8.5) is used. So fix the issue with > > implementing the logic with other intrinsics. > > > > Bugzilla ID: 621 > > Fixes: 530887469350 ("net/virtio: add vectorized packed ring NEON Tx") > > Fixes: 5971ce5e2a59 ("net/virtio: add vectorized packed ring NEON Rx") > > > > Reported-by: Subhi Masri <subhim@nvidia.com> > > Signed-off-by: Joyce Kong <joyce.kong@arm.com> > > --- > > drivers/net/virtio/virtio_rxtx_packed_neon.h | 8 ++++---- > > 1 file changed, 4 insertions(+), 4 deletions(-) > > > > diff --git a/drivers/net/virtio/virtio_rxtx_packed_neon.h > > b/drivers/net/virtio/virtio_rxtx_packed_neon.h > > index 01c77b712..00dd04277 100644 > > --- a/drivers/net/virtio/virtio_rxtx_packed_neon.h > > +++ b/drivers/net/virtio/virtio_rxtx_packed_neon.h > > @@ -71,8 +71,8 @@ virtqueue_enqueue_batch_packed_vec(struct > virtnet_tx *txvq, > > uint32x4_t def_ref_seg = vdupq_n_u32(0x10001); > > /* Check refcnt and nb_segs. */ > > uint32x4_t ref_seg = vreinterpretq_u32_u8(vqtbl2q_u8(mbuf, > ref_seg_msk)); > > - poly128_t cmp1 = vreinterpretq_p128_u32(~vceqq_u32(ref_seg, > def_ref_seg)); > > - if (unlikely(cmp1)) > > + uint64x2_t cmp1 = vreinterpretq_u64_u32(~vceqq_u32(ref_seg, > def_ref_seg)); > > + if (unlikely(vgetq_lane_u64(cmp1, 0) || vgetq_lane_u64(cmp1, > > + 1))) > > return -1; > > > > /* Check headroom is enough. */ > > @@ -225,10 +225,10 @@ virtqueue_dequeue_batch_packed_vec(struct > virtnet_rx *rxvq, > > if (vq->vq_packed.used_wrap_counter) > > v_used_flag = vdupq_n_u32(PACKED_FLAGS_MASK); > > > > - poly128_t desc_stats = vreinterpretq_p128_u32(~vceqq_u32(v_flag, > v_used_flag)); > > + uint64x2_t desc_stats = vreinterpretq_u64_u32(~vceqq_u32(v_flag, > > + v_used_flag)); > > > > /* Check all descs are used. */ > > - if (desc_stats) > > + if (unlikely(vgetq_lane_u64(desc_stats, 0) || > > + vgetq_lane_u64(desc_stats, 1))) > > return -1; > > > > /* Load 2 mbuf pointers per time. */ > > > > Can you confirm the patch works for you? > Yes it does , you can add my tag. Tested-by: Subhi Masri <subhim@nvidia.com> > Thanks, > Maxime Regards, Subhi Masri
On 1/26/21 10:57 AM, Joyce Kong wrote:
> In file included from ../drivers/net/virtio/virtio_rxtx_packed.c:22:0:
> ../drivers/net/virtio/virtio_rxtx_packed_neon.h: In function
> ‘virtqueue_enqueue_batch_packed_vec’:
> ../drivers/net/virtio/virtio_rxtx_packed_neon.h:74:2: warning:
> implicit declaration of function ‘vreinterpretq_p128_u32’
> [-Wimplicit-function-declaration]
> poly128_t cmp1 = vreinterpretq_p128_u32(~vceqq_u32(ref_seg, def_ref_seg));
> ^
>
> The message shows ‘vreinterpretq_p128_u32’ intrinsic is not supported
> because an old version gcc (gcc 4.8.5) is used. So fix the issue with
> implementing the logic with other intrinsics.
>
> Bugzilla ID: 621
> Fixes: 530887469350 ("net/virtio: add vectorized packed ring NEON Tx")
> Fixes: 5971ce5e2a59 ("net/virtio: add vectorized packed ring NEON Rx")
>
> Reported-by: Subhi Masri <subhim@nvidia.com>
> Signed-off-by: Joyce Kong <joyce.kong@arm.com>
> ---
> drivers/net/virtio/virtio_rxtx_packed_neon.h | 8 ++++----
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/net/virtio/virtio_rxtx_packed_neon.h b/drivers/net/virtio/virtio_rxtx_packed_neon.h
> index 01c77b712..00dd04277 100644
> --- a/drivers/net/virtio/virtio_rxtx_packed_neon.h
> +++ b/drivers/net/virtio/virtio_rxtx_packed_neon.h
> @@ -71,8 +71,8 @@ virtqueue_enqueue_batch_packed_vec(struct virtnet_tx *txvq,
> uint32x4_t def_ref_seg = vdupq_n_u32(0x10001);
> /* Check refcnt and nb_segs. */
> uint32x4_t ref_seg = vreinterpretq_u32_u8(vqtbl2q_u8(mbuf, ref_seg_msk));
> - poly128_t cmp1 = vreinterpretq_p128_u32(~vceqq_u32(ref_seg, def_ref_seg));
> - if (unlikely(cmp1))
> + uint64x2_t cmp1 = vreinterpretq_u64_u32(~vceqq_u32(ref_seg, def_ref_seg));
> + if (unlikely(vgetq_lane_u64(cmp1, 0) || vgetq_lane_u64(cmp1, 1)))
> return -1;
>
> /* Check headroom is enough. */
> @@ -225,10 +225,10 @@ virtqueue_dequeue_batch_packed_vec(struct virtnet_rx *rxvq,
> if (vq->vq_packed.used_wrap_counter)
> v_used_flag = vdupq_n_u32(PACKED_FLAGS_MASK);
>
> - poly128_t desc_stats = vreinterpretq_p128_u32(~vceqq_u32(v_flag, v_used_flag));
> + uint64x2_t desc_stats = vreinterpretq_u64_u32(~vceqq_u32(v_flag, v_used_flag));
>
> /* Check all descs are used. */
> - if (desc_stats)
> + if (unlikely(vgetq_lane_u64(desc_stats, 0) || vgetq_lane_u64(desc_stats, 1)))
> return -1;
>
> /* Load 2 mbuf pointers per time. */
>
Reviewed-by: Maxime Coquelin <maxime.coquelin@redhat.com>
Thanks,
Maxime
On 1/27/21 11:37 AM, Subhi Masri wrote: > Hi, > >> -----Original Message----- >> From: Maxime Coquelin <maxime.coquelin@redhat.com> >> Sent: Wednesday, January 27, 2021 10:44 AM >> To: Subhi Masri <subhim@nvidia.com> >> Cc: dev@dpdk.org; nd@arm.com; Joyce Kong <joyce.kong@arm.com>; >> david.marchand@redhat.com; honnappa.nagarahalli@arm.com; >> ruifeng.wang@arm.com; i.maximets@ovn.org >> Subject: Re: [PATCH v2] net/virtio: fix compiling issue for vectorized NEON >> path >> >> External email: Use caution opening links or attachments >> >> >> Hi Subhi, >> >> On 1/26/21 10:57 AM, Joyce Kong wrote: >>> In file included from ../drivers/net/virtio/virtio_rxtx_packed.c:22:0: >>> ../drivers/net/virtio/virtio_rxtx_packed_neon.h: In function >>> ‘virtqueue_enqueue_batch_packed_vec’: >>> ../drivers/net/virtio/virtio_rxtx_packed_neon.h:74:2: warning: >>> implicit declaration of function ‘vreinterpretq_p128_u32’ >>> [-Wimplicit-function-declaration] >>> poly128_t cmp1 = vreinterpretq_p128_u32(~vceqq_u32(ref_seg, >>> def_ref_seg)); ^ >>> >>> The message shows ‘vreinterpretq_p128_u32’ intrinsic is not supported >>> because an old version gcc (gcc 4.8.5) is used. So fix the issue with >>> implementing the logic with other intrinsics. >>> >>> Bugzilla ID: 621 >>> Fixes: 530887469350 ("net/virtio: add vectorized packed ring NEON Tx") >>> Fixes: 5971ce5e2a59 ("net/virtio: add vectorized packed ring NEON Rx") >>> >>> Reported-by: Subhi Masri <subhim@nvidia.com> >>> Signed-off-by: Joyce Kong <joyce.kong@arm.com> >>> --- >>> drivers/net/virtio/virtio_rxtx_packed_neon.h | 8 ++++---- >>> 1 file changed, 4 insertions(+), 4 deletions(-) >>> >>> diff --git a/drivers/net/virtio/virtio_rxtx_packed_neon.h >>> b/drivers/net/virtio/virtio_rxtx_packed_neon.h >>> index 01c77b712..00dd04277 100644 >>> --- a/drivers/net/virtio/virtio_rxtx_packed_neon.h >>> +++ b/drivers/net/virtio/virtio_rxtx_packed_neon.h >>> @@ -71,8 +71,8 @@ virtqueue_enqueue_batch_packed_vec(struct >> virtnet_tx *txvq, >>> uint32x4_t def_ref_seg = vdupq_n_u32(0x10001); >>> /* Check refcnt and nb_segs. */ >>> uint32x4_t ref_seg = vreinterpretq_u32_u8(vqtbl2q_u8(mbuf, >> ref_seg_msk)); >>> - poly128_t cmp1 = vreinterpretq_p128_u32(~vceqq_u32(ref_seg, >> def_ref_seg)); >>> - if (unlikely(cmp1)) >>> + uint64x2_t cmp1 = vreinterpretq_u64_u32(~vceqq_u32(ref_seg, >> def_ref_seg)); >>> + if (unlikely(vgetq_lane_u64(cmp1, 0) || vgetq_lane_u64(cmp1, >>> + 1))) >>> return -1; >>> >>> /* Check headroom is enough. */ >>> @@ -225,10 +225,10 @@ virtqueue_dequeue_batch_packed_vec(struct >> virtnet_rx *rxvq, >>> if (vq->vq_packed.used_wrap_counter) >>> v_used_flag = vdupq_n_u32(PACKED_FLAGS_MASK); >>> >>> - poly128_t desc_stats = vreinterpretq_p128_u32(~vceqq_u32(v_flag, >> v_used_flag)); >>> + uint64x2_t desc_stats = vreinterpretq_u64_u32(~vceqq_u32(v_flag, >>> + v_used_flag)); >>> >>> /* Check all descs are used. */ >>> - if (desc_stats) >>> + if (unlikely(vgetq_lane_u64(desc_stats, 0) || >>> + vgetq_lane_u64(desc_stats, 1))) >>> return -1; >>> >>> /* Load 2 mbuf pointers per time. */ >>> >> >> Can you confirm the patch works for you? >> > Yes it does , you can add my tag. > Tested-by: Subhi Masri <subhim@nvidia.com> Thanks! Maxime >> Thanks, >> Maxime > > > Regards, > Subhi Masri >
On 1/26/21 10:57 AM, Joyce Kong wrote:
> In file included from ../drivers/net/virtio/virtio_rxtx_packed.c:22:0:
> ../drivers/net/virtio/virtio_rxtx_packed_neon.h: In function
> ‘virtqueue_enqueue_batch_packed_vec’:
> ../drivers/net/virtio/virtio_rxtx_packed_neon.h:74:2: warning:
> implicit declaration of function ‘vreinterpretq_p128_u32’
> [-Wimplicit-function-declaration]
> poly128_t cmp1 = vreinterpretq_p128_u32(~vceqq_u32(ref_seg, def_ref_seg));
> ^
>
> The message shows ‘vreinterpretq_p128_u32’ intrinsic is not supported
> because an old version gcc (gcc 4.8.5) is used. So fix the issue with
> implementing the logic with other intrinsics.
>
> Bugzilla ID: 621
> Fixes: 530887469350 ("net/virtio: add vectorized packed ring NEON Tx")
> Fixes: 5971ce5e2a59 ("net/virtio: add vectorized packed ring NEON Rx")
>
> Reported-by: Subhi Masri <subhim@nvidia.com>
> Signed-off-by: Joyce Kong <joyce.kong@arm.com>
> ---
> drivers/net/virtio/virtio_rxtx_packed_neon.h | 8 ++++----
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
Applied to dpdk-next-virtio/main.
Thanks,
Maxime