Arm Development Studio forum Neon vldx_lane_y compilation efficiency

Locked Locked
Replies 3 replies
Subscribers 119 subscribers
Views 3283 views
Users 0 members are here

Options

Related

How was your experience today?

This discussion has been locked.

You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Neon vldx_lane_y compilation efficiency

Pierre ELINE over 12 years ago

Parents

Pierre ELINE over 12 years ago

Note: This was originally posted on 16th November 2011 at http://forums.arm.com

Thx for your reply.

I hardly read the cycle table, but according to this simulator there is no difference in term of cycle between (http://pulsar.websha...sample-96ee929b):

vld2.8 {d0[0],d1[0]}, [r0] vld2.8 {d0[1],d1[1]}, [r0] vld2.8 {d0[2],d1[2]}, [r0] vld2.8 {d0[3],d1[3]}, [r0]

and (http://pulsar.websha...sample-469dc426):

vld2.8 {d0[0],d1[0]}, [r0] vld2.8 {d2[1],d3[1]}, [r0] vld2.8 {d4[2],d5[2]}, [r0] vld2.8 {d6[3],d7[3]}, [r0]
Cancel
Vote up 0 Vote down

Cancel

Reply

Pierre ELINE over 12 years ago

Note: This was originally posted on 16th November 2011 at http://forums.arm.com

Thx for your reply.

I hardly read the cycle table, but according to this simulator there is no difference in term of cycle between (http://pulsar.websha...sample-96ee929b):

vld2.8 {d0[0],d1[0]}, [r0] vld2.8 {d0[1],d1[1]}, [r0] vld2.8 {d0[2],d1[2]}, [r0] vld2.8 {d0[3],d1[3]}, [r0]

and (http://pulsar.websha...sample-469dc426):

vld2.8 {d0[0],d1[0]}, [r0] vld2.8 {d2[1],d3[1]}, [r0] vld2.8 {d4[2],d5[2]}, [r0] vld2.8 {d6[3],d7[3]}, [r0]
Cancel
Vote up 0 Vote down

Cancel

Children

No data