From cc14bcbd7a015806c3bcd15ee725ab78ba9aafde Mon Sep 17 00:00:00 2001 From: lkcl Date: Fri, 25 Dec 2020 18:02:33 +0000 Subject: [PATCH] --- openpower/sv/overview.mdwn | 16 ++++++++++++++++ 1 file changed, 16 insertions(+) diff --git a/openpower/sv/overview.mdwn b/openpower/sv/overview.mdwn index 92a7b7679..bbef5ae5b 100644 --- a/openpower/sv/overview.mdwn +++ b/openpower/sv/overview.mdwn @@ -505,3 +505,19 @@ With VL truncated the analysis looking for the zero byte and the subsequent STOR SIMD strncpy hand-written assembly routines are, to be blunt about it, a total nightmare. 240 instructions is not uncommon, and the worst thing about them is that they are unable to cope with detection of a page fault condition. +# Data-dependent fail-first + +This is a minor variant on the CR-based predicate-result mode. Where ored-result continues with independent element testing, data-dependent fail-first *stops* at the first failure: + + for i in range(VL): + # predication test, skip all masked out elements. + if predicate_masked_out(i): continue # skip + result = op(iregs[RA+i], iregs[RB+i]) + CRnew = analyse(result) # calculates eq/lt/gt + # now test CR, similar to branch + if CRnew[BO[0:1]] != BO[2]: + VL = i # truncate: only successes allowed + break + # test passed: store result (and CR?) + iregs[RT+i] = result + if Rc=1: crregs[offs+i] = CRnew -- 2.30.2