You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Mar 20, 2024. It is now read-only.
VSUXEI explicitly includes Unordered in its name to stipulate that the net result of the operation allows multiple outcomes visible to the local hart; essentially depending upon which element is last written to overlapping memory locations.
Explicitly include Ordered in the complimentary VSOXEI to stipulate that the data written to memory will be as if the active elements were first copied to an XLEN length buffer at the calculated offset processing the elements in the order from 0 to vl, and then only the affected bytes written to memory in whatever order RVVWMO allows.
The behaviour of VSUXEI should also be similarly stipulated: that the data written to memory will be as if the active elements were first copied to an XLEN length buffer at the calculated offset, processing each elements in any order, and then only the affected bytes written to memory in whatever order RVVWMO allows.e
The ideas are that
rather than suggesting VSUXEI is the exception to the rule, it indicates that the ordered VSOXEI is a more constrained alternative to the default processor model that allows concurrent execution wherever possible and
VSOXEI makes no guarantee on what is globally visible than what the RVVWMO model allows
VSUXEI is still constrained by the RVVWMO rules.
The text was updated successfully, but these errors were encountered:
VSUXEI explicitly includes Unordered in its name to stipulate that the net result of the operation allows multiple outcomes visible to the local hart; essentially depending upon which element is last written to overlapping memory locations.
Explicitly include Ordered in the complimentary VSOXEI to stipulate that the data written to memory will be as if the active elements were first copied to an XLEN length buffer at the calculated offset processing the elements in the order from 0 to vl, and then only the affected bytes written to memory in whatever order RVVWMO allows.
The behaviour of VSUXEI should also be similarly stipulated: that the data written to memory will be as if the active elements were first copied to an XLEN length buffer at the calculated offset, processing each elements in any order, and then only the affected bytes written to memory in whatever order RVVWMO allows.e
The ideas are that
The text was updated successfully, but these errors were encountered: