Summary:
Also, made test_mi10.s formatting consistent with the majority of the
MC tests.
Reviewers: vkalintiris
Subscribers: dsanders, llvm-commits
Differential Revision: http://reviews.llvm.org/D18435
llvm-svn: 265014
Change isConsecutiveLoads to check that loads are non-volatile as this
is a requirement for any load merges. Propagate change to two callers.
Reviewers: RKSimon
Subscribers: llvm-commits
Differential Revision: http://reviews.llvm.org/D18546
llvm-svn: 265013
Summary:
The bug was that microMIPS's [ls]w[lr]e instructions claimed to support a
12-bit offset when it is only 9-bit.
Reviewers: vkalintiris
Subscribers: llvm-commits, dsanders
Differential Revision: http://reviews.llvm.org/D18434
llvm-svn: 265010
When dealing with complex<float>, and similar structures with two
single-precision floating-point numbers, especially when such things are being
passed around by value, we'll sometimes end up loading both float values by
extracting them from one 64-bit integer load. It looks like this:
t13: i64,ch = load<LD8[%ref.tmp]> t0, t6, undef:i64
t16: i64 = srl t13, Constant:i32<32>
t17: i32 = truncate t16
t18: f32 = bitcast t17
t19: i32 = truncate t13
t20: f32 = bitcast t19
The problem, especially before the P8 where those bitcasts aren't legal (and
get expanded via the stack), is that it would have been better to use two
floating-point loads directly. Here we add a target-specific DAGCombine to do
just that. In short, we turn:
ld 3, 0(5)
stw 3, -8(1)
rldicl 3, 3, 32, 32
stw 3, -4(1)
lfs 3, -4(1)
lfs 0, -8(1)
into:
lfs 3, 4(5)
lfs 0, 0(5)
llvm-svn: 264988
The size savings are significant, and from what I can tell, both ICC and GCC do this.
Differential Revision: http://reviews.llvm.org/D18573
llvm-svn: 264966
Fix for issue introduced D17297, where we were breaking early from the loop detecting consecutive loads which could leave us thinking a consecutive load with zeros was possible.
llvm-svn: 264922
Summary:
Currently it's a module pass. Make it a function pass so that we can
move it to PassManagerBuilder's EP_EarlyAsPossible extension point,
which only accepts function passes.
Reviewers: rnk
Subscribers: tra, llvm-commits, jholewinski
Differential Revision: http://reviews.llvm.org/D18615
llvm-svn: 264919
Summary:
This results in higher register usage, but should make it easier for
the compiler to hide latency.
This pass is a prerequisite for some more scheduler improvements, and I
think the increase register usage with this patch is acceptable, because
when combined with the scheduler improvements, the total register usage
will decrease.
shader-db stats:
2382 shaders in 478 tests
Totals:
SGPRS: 48672 -> 49088 (0.85 %)
VGPRS: 34148 -> 34847 (2.05 %)
Code Size: 1285816 -> 1289128 (0.26 %) bytes
LDS: 28 -> 28 (0.00 %) blocks
Scratch: 492544 -> 573440 (16.42 %) bytes per wave
Max Waves: 6856 -> 6846 (-0.15 %)
Wait states: 0 -> 0 (0.00 %)
Depends on D18451
Reviewers: nhaehnle, arsenm
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D18452
llvm-svn: 264876
For compatability with GAS, nop and nopr are recognized as alises for
bc and bcr, respectively. A mask of 0 turns these instructions
effectively into no-operations.
Reviewed by Ulrich Weigand.
llvm-svn: 264875
These checks are redundant and can be removed
Reviewers: hans
Subscribers: llvm-commits, mzolotukhin
Differential Revision: http://reviews.llvm.org/D18564
llvm-svn: 264872
XOP's VPPERM has some great 'permute operations' that it can do as well as part of shuffling the bytes of a 128-bit vector - in this case we use it to perform BITREVERSE in a single instruction.
llvm-svn: 264870
operations.
Specifically, we had code that tried to badly approximate reconstructing
all of the possible variations on addressing modes in two x86
instructions based on those in one pseudo instruction. This is not the
first bug uncovered with doing this, so stop doing it altogether.
Instead generically and pedantically copy every operand from the address
over to both new instructions, and strip kill flags from any register
operands.
This fixes a subtle bug seen in the wild where we would mysteriously
drop parts of the addressing mode, causing for example the index
argument in the added test case to just be completely ignored.
Hypothetically, this was an extremely bad miscompile because it actually
caused a predictable and leveragable write of a 64bit quantity to an
unintended offset (the first element of the array intead of whatever
other element was intended). As a consequence, in theory this could even
have introduced security vulnerabilities.
However, this was only something that could happen with an atomic
floating point add. No other operation could trigger this bug, so it
seems extremely unlikely to have occured widely in the wild.
But it did in fact occur, and frequently in scientific applications
which were using relaxed atomic updates of a floating point value after
adding a delta. Those would end up being quite badly miscompiled by
LLVM, which is how we found this. Of course, this often looks like
a race condition in the code, but it was actually a miscompile.
I suspect that this whole RELEASE_FADD thing was a complete mistake.
There is no such operation, and I worry that anything other than add
will get remarkably worse codegeneration. But that's not for this
change....
llvm-svn: 264845
This is effectively NFC, minus the renaming of the options
(-cyclone-prefetch-distance -> -prefetch-distance).
The change was requested by Tim in D17943.
llvm-svn: 264806
They were previously expanded to CAS loops in a custom isel expansion,
but AtomicExpandPass knows how to do that generically.
Testing is covered by the existing sparc atomics.ll testcases.
llvm-svn: 264771
Summary:
However, this has no effect at this time because the instructions affected
are marked 'isCodeGenOnly=1' and have no alternative for the MC layer.
Reviewers: vkalintiris
Subscribers: llvm-commits, dsanders
Differential Revision: http://reviews.llvm.org/D18179
llvm-svn: 264712
Fixed fp_to_uint instruction selection on KNL.
One pattern was missing for <4 x double> to <4 x i32>
Differential Revision: http://reviews.llvm.org/D18512
llvm-svn: 264701
Instead of using two feature bits, one to indicate the availability of the
popcnt[dw] instructions, and another to indicate whether or not they're fast,
use a single enum. This allows more consistent control via target attribute
strings, and via Clang's command line.
llvm-svn: 264690
If all a BUILD_VECTOR's source elements are the same bit (AND/XOR/OR) operation type and each has one constant operand, lower to a pair of BUILD_VECTOR and just apply the bit operation to the vectors.
The constant operands will form a constant vector meaning that we still only have a single BUILD_VECTOR to lower and we will have replaced all the scalarized operations with a single SSE equivalent.
Its not in our interest to start make a general purpose vectorizer from this, but I'm seeing enough of these scalar bit operations from the later legalization/scalarization stages to support them at least.
Differential Revision: http://reviews.llvm.org/D18492
llvm-svn: 264666
The A2 cores support the popcntw/popcntd instructions, but they're microcoded,
and slower than our default software emulation. Specifically, popcnt[dw] take
approximately 74 cycles, whereas our software emulation takes only 24-28
cycles.
I've added a new target feature to indicate a slow popcnt[dw], instead of just
removing the existing target feature from the a2/a2q processor models, because:
1. This allows us to return more accurate information via the TTI interface
(I recognize that this currently makes no practical difference)
2. Is hopefully easier to understand (it allows the core's features to match
its manual while still having the desired effect).
llvm-svn: 264600
MachineFunctionProperties represents a set of properties that a MachineFunction
can have at particular points in time. Existing examples of this idea are
MachineRegisterInfo::isSSA() and MachineRegisterInfo::tracksLiveness() which
will eventually be switched to use this mechanism.
This change introduces the AllVRegsAllocated property; i.e. the property that
all virtual registers have been allocated and there are no VReg operands
left.
With this mechanism, passes can declare that they require a particular property
to be set, or that they set or clear properties by implementing e.g.
MachineFunctionPass::getRequiredProperties(). The MachineFunctionPass base class
verifies that the requirements are met, and handles the setting and clearing
based on the delcarations. Passes can also directly query and update the current
properties of the MF if they want to have conditional behavior.
This change annotates the target-independent post-regalloc passes; future
changes will also annotate target-specific ones.
Reviewers: qcolombet, hfinkel
Differential Revision: http://reviews.llvm.org/D18421
llvm-svn: 264593
Summary:
This helps prevent load clustering from drastically increasing register
pressure by trying to cluster 4 SMRDx8 loads together. The limit of 16
bytes was chosen, because it seems like that was the original intent
of setting the limit to 4 instructions, but more analysis could show
that a different limit is better.
This fixes yields small decreases in register usage with shader-db, but
also helps avoid a large increase in register usage when lane mask
tracking is enabled in the machine scheduler, because lane mask tracking
enables more opportunities for load clustering.
shader-db stats:
2379 shaders in 477 tests
Totals:
SGPRS: 49744 -> 48600 (-2.30 %)
VGPRS: 34120 -> 34076 (-0.13 %)
Code Size: 1282888 -> 1283184 (0.02 %) bytes
LDS: 28 -> 28 (0.00 %) blocks
Scratch: 495616 -> 492544 (-0.62 %) bytes per wave
Max Waves: 6843 -> 6853 (0.15 %)
Wait states: 0 -> 0 (0.00 %)
Reviewers: nhaehnle, arsenm
Subscribers: arsenm, llvm-commits
Differential Revision: http://reviews.llvm.org/D18451
llvm-svn: 264589
- Do not optimize stack slots in optnone functions.
- Get aligned-base register from HexagonMachineFunctionInfo instead of
looking for ALIGNA instruction in the function's body.
llvm-svn: 264580
ICMP instruction selection fails on SKX and KNL for i1 operand.
I use XOR to resolve:
(A == B) is equivalent to (A xor B) == 0
Differential Revision: http://reviews.llvm.org/D18511
llvm-svn: 264566
Intrinsic::maxnum and Intrinsic::minnum, along with the associated libc
function calls (fmax[f], etc.) generally map to function calls after lowering.
For some vector types with QPX at least, however, we can legally lower these,
and we don't need to prohibit CTR-based loops on their account.
It turned out, however, that the logic that checked the opcodes associated with
intrinsics was broken (it would set the Opcode variable, but that variable was
later checked only if set for some otherwise-external function call.
This fixes the latter problem and adds the FMAX/MINNUM mappings.
llvm-svn: 264532
Daniel Sanders