bcm5719-llvm/llvm/test/Transforms/LoopUnroll/ARM, branch meklort-10.0.1Project Ortega BCM5719 LLVMhttps://git.raptorcs.com/git/bcm5719-llvm/atom?h=meklort-10.0.12020-01-10T09:41:41+00:00Follow up of 67bf9a6154d4b82c, minor fix in test case, removed duplicate option2020-01-10T09:41:41+00:00Sjoerd Meijersjoerd.meijer@arm.com2020-01-10T09:41:41+00:00urn:sha1:356685a1d8972180f472c1333e8e89dbcc704c1d[SVEV] Recognise hardware-loop intrinsic loop.decrement.reg2020-01-10T09:35:00+00:00Sjoerd Meijersjoerd.meijer@arm.com2020-01-10T09:30:02+00:00urn:sha1:67bf9a6154d4b82c6c01aad01141bf08c1bbd0f6
Teach SCEV about the @loop.decrement.reg intrinsic, which has exactly the same
semantics as a sub expression. This allows us to query hardware-loops, which
contain this @loop.decrement.reg intrinsic, so that we can calculate iteration
counts, exit values, etc. of hardwareloops.
This "int_loop_decrement_reg" intrinsic is defined as "IntrNoDuplicate". Thus,
while hardware-loops and tripcounts now become analysable by SCEV, this
prevents the usual loop transformations from applying transformations on
hardware-loops, which is what we want at this point, for which I have added
test cases for loopunrolling and IndVarSimplify and LFTR.
Differential Revision: https://reviews.llvm.org/D71563
[ARM][MVE] Don't unroll intrinsic loops.2020-01-09T11:57:34+00:00Sam Parkersam.parker@arm.com2020-01-09T11:57:04+00:00urn:sha1:15c7fa4d11eeb50095ae571c645427b9a267bdee
We don't unroll vector loops for MVE targets, but we miss the case
when loops only contain intrinsic calls. So just move the logic a
bit to catch this case.
Differential Revision: https://reviews.llvm.org/D72440
[MVE] Don't try to unroll vectorised MVE loops2019-08-11T08:53:18+00:00David Greendavid.green@arm.com2019-08-11T08:53:18+00:00urn:sha1:11c4602fce16c22858de764bff21be4160d323af
Due to the nature of the beat system in the MVE architecture, along with tail
predication and low-overhead loops, unrolling has less benefit compared to
normal loops. You can not, for example, hide the latency of a load with other
instructions as you can for scalar code. Preventing unrolling also makes the
code easier to read and reason about.
So if a loop contains vector code, don't enable the runtime unrolling. At least
for the time being.
Differential Revision: https://reviews.llvm.org/D65803
llvm-svn: 368530
[lit] Delete empty lines at the end of lit.local.cfg NFC2019-06-17T09:51:07+00:00Fangrui Songmaskray@google.com2019-06-17T09:51:07+00:00urn:sha1:ac14f7b10cffe2be548607269e036244cd16acc3
llvm-svn: 363538
[ARM] Enable Unroll UpperBound2019-06-10T10:22:14+00:00David Greendavid.green@arm.com2019-06-10T10:22:14+00:00urn:sha1:d847aa573b655fe71aad323c7f579c6302c702aa
This option allows loops with small max trip counts to be fully unrolled. This
can help with code like the remainder loops from manually unrolled loops like
those that appear in the cmsis dsp library. We would apparently previously
runtime unroll them with the default unroll count (4).
Differential Revision: https://reviews.llvm.org/D63064
llvm-svn: 362928
Revert "Temporarily Revert "Add basic loop fusion pass.""2019-04-17T04:52:47+00:00Eric Christopherechristo@gmail.com2019-04-17T04:52:47+00:00urn:sha1:cee313d288a4faf0355d76fb6e0e927e211d08a5
The reversion apparently deleted the test/Transforms directory.
Will be re-reverting again.
llvm-svn: 358552
Temporarily Revert "Add basic loop fusion pass."2019-04-17T02:12:23+00:00Eric Christopherechristo@gmail.com2019-04-17T02:12:23+00:00urn:sha1:a86343512845c9c1fdbac865fea88aa5fce7142a
As it's causing some bot failures (and per request from kbarton).
This reverts commit r358543/ab70da07286e618016e78247e4a24fcb84077fda.
llvm-svn: 358546
[ARM] Use Cortex-A57 sched model for Cortex-A722018-10-25T15:08:29+00:00Sam Parkersam.parker@arm.com2018-10-25T15:08:29+00:00urn:sha1:a16667e79be9164943dd76235d5406ae519b9042
This mirrors what we already do for AArch64 as the cores are similar.
As discussed in the review, enabling the machine scheduler causes
more variations in performance changes so it is not enabled for now.
This patch improves LNT scores by a geomean of 1.57% at -O3.
Differential Revision: https://reviews.llvm.org/D53562
llvm-svn: 345272
[ARM] Allow unrolling of multi-block loops.2017-10-23T08:05:14+00:00Sam Parkersam.parker@arm.com2017-10-23T08:05:14+00:00urn:sha1:487ab869422b1eb7be932eb70445d9d0541fdce1
Before, loop unrolling was only enabled for loops with a single
block. This restriction has been removed and replaced by:
- allow a maximum of two exiting blocks,
- a four basic block limit for cores with a branch predictor.
Differential Revision: https://reviews.llvm.org/D38952
llvm-svn: 316313