This patch adds a feature flag to avoid unaligned 32-byte load/store AVX codegen
for Sandy Bridge and Ivy Bridge. There is no functionality change intended for
those chips. Previously, the absence of AVX2 was being used as a proxy to detect
this feature. But that hindered codegen for AVX-enabled AMD chips such as btver2
that do not have the 32-byte unaligned access slowdown.
Performance measurements are included in PR21541 ( http://llvm.org/bugs/show_bug.cgi?id=21541 ).
Differential Revision: http://reviews.llvm.org/D6355
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@222544 91177308-0d34-0410-b5e6-96231b3b80d8
This is a first step for generating SSE rcp instructions for reciprocal
calcs when fast-math allows it. This is very similar to the rsqrt optimization
enabled in D5658 ( http://reviews.llvm.org/rL220570 ).
For now, be conservative and only enable this for AMD btver2 where performance
improves significantly both in terms of latency and throughput.
We may never enable this codegen for Intel Core* chips because the divider circuits
are just too fast. On SandyBridge, divss can be as fast as 10 cycles versus the 21
cycle critical path for the rcp + mul + sub + mul + add estimate.
Follow-on patches may allow configuration of the number of Newton-Raphson refinement
steps, add AVX512 support, and enable the optimization for more chips.
More background here: http://llvm.org/bugs/show_bug.cgi?id=21385
Differential Revision: http://reviews.llvm.org/D6175
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221706 91177308-0d34-0410-b5e6-96231b3b80d8
This patch adds 'FeatureSlowSHLD' to 'bdver3'.
According to the official AMD optimization guide for amdfam15: "Using
alternative code in place of SHLD achieves lower overall latency and
requires fewer execution resources. The 32-bit and 64-bit forms of
ADD, ADC, SHR, and LEA (except 16-bit form) are DirectPath
instructions, while SHLD is a VectorPath instruction."
This patch also explicitly sets feature AVX and SSE4A for all the bdver*
cpus. This part of the patch is a non-functional change and it is mainly
done for clarity reasons (Both XOP and FMA4 already imply AVX and SSE4A).
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@221296 91177308-0d34-0410-b5e6-96231b3b80d8
This is a first step for generating SSE rsqrt instructions for
reciprocal square root calcs when fast-math is allowed.
For now, be conservative and only enable this for AMD btver2
where performance improves significantly - for example, 29%
on llvm/projects/test-suite/SingleSource/Benchmarks/BenchmarkGame/n-body.c
(if we convert the data type to single-precision float).
This patch adds a two constant version of the Newton-Raphson
refinement algorithm to DAGCombiner that can be selected by any target
via a parameter returned by getRsqrtEstimate()..
See PR20900 for more details:
http://llvm.org/bugs/show_bug.cgi?id=20900
Differential Revision: http://reviews.llvm.org/D5658
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@220570 91177308-0d34-0410-b5e6-96231b3b80d8
This is a first pass at a scheduling model for Jaguar.
It's structured largely on the existing SandyBridge and SLM sched models.
Using this model, in addition to turning on the PostRA scheduler, results in
some perf wins on internal and 3rd party benchmarks. There's not much difference
in LLVM's test-suite benchmarking subset of tests.
Differential Revision: http://reviews.llvm.org/D5229
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@217457 91177308-0d34-0410-b5e6-96231b3b80d8
This allows assembling the two new instructions, encls and enclu for the
SKX processor model.
Note the diffs are a bigger than what might think, but to fit the new
MRM_CF and MRM_D7 in things in the right places things had to be
renumbered and shuffled down causing a bit more diffs.
rdar://16228228
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@214460 91177308-0d34-0410-b5e6-96231b3b80d8
After Alexey Volkov, I'm adding the same property for KNL, that prefers ADD/SUB instead of INC/DEC.
Added a test.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@212178 91177308-0d34-0410-b5e6-96231b3b80d8
According to Intel Software Optimization Manual
on Silvermont INC or DEC instructions require
an additional uop to merge the flags.
As a result, a branch instruction depending
on an INC or a DEC instruction incurs a 1 cycle penalty.
Differential Revision: http://reviews.llvm.org/D3990
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@210466 91177308-0d34-0410-b5e6-96231b3b80d8
According to Intel Software Optimization Manual on Silvermont in some cases LEA
is better to be replaced with ADD instructions:
"The rule of thumb for ADDs and LEAs is that it is justified to use LEA
with a valid index and/or displacement for non-destructive destination purposes
(especially useful for stack offset cases), or to use a SCALE.
Otherwise, ADD(s) are preferable."
Differential Revision: http://reviews.llvm.org/D3826
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@209198 91177308-0d34-0410-b5e6-96231b3b80d8
default architecture for reasonable modern x86 processors, actually be
modern. This processor model should essentially be "tuned" for modern
x86 chips as much as possible without undue penalties on any specific
architecture. Previously we weren't even using the nice scheduling
models. There are a few other tweaks needed here, but this change at
least I have benchmarked across a decent swatch of chips (intel's
clovertown, westmere, and sandybridge; amd's istanbul) and seen no
significant regressions.
If anyone has suggested ways to test this, just let me know. Somewhat
alarmingly, no existing tests failed.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@208230 91177308-0d34-0410-b5e6-96231b3b80d8
This is not really expected to work right yet. Mostly because we will
still emit the OpSize (0x66) prefix in all the wrong places, along with
a number of other corner cases. Those will all be fixed in the subsequent
commits.
Patch from David Woodhouse.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@198584 91177308-0d34-0410-b5e6-96231b3b80d8
AMD's processors family K7, K8, K10, K12, K15 and K16 are known to have SHLD/SHRD instructions with very poor latency. Optimization guides for these processors recommend using an alternative sequence of instructions. For these AMD's processors, I disabled folding (or (x << c) | (y >> (64 - c))) when we are not optimizing for size.
It might be beneficial to disable this folding for some of the Intel's processors. However, since I couldn't find specific recommendations regarding using SHLD/SHRD instructions on Intel's processors, I haven't disabled this peephole for Intel.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@195383 91177308-0d34-0410-b5e6-96231b3b80d8
Implements Instruction scheduler latencies for Silvermont,
using latencies from the Intel Silvermont Optimization Guide.
Auto detects SLM.
Turns on post RA scheduler when generating code for SLM.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@190717 91177308-0d34-0410-b5e6-96231b3b80d8
Add basic assembly/disassembly support for the first Intel SHA
instruction 'sha1rnds4'. Also includes feature flag, and test cases.
Support for the remaining instructions will follow in a separate patch.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@190611 91177308-0d34-0410-b5e6-96231b3b80d8
latency for certain models of the Intel Atom family, by converting
instructions into their equivalent LEA instructions, when it is both
useful and possible to do so.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@180573 91177308-0d34-0410-b5e6-96231b3b80d8
variant/dialect. Addresses a FIXME in the emitMnemonicAliases function.
Use and test case to come shortly.
rdar://13688439 and part of PR13340.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@179804 91177308-0d34-0410-b5e6-96231b3b80d8
indirect through a memory address is to load the memory address into
a register and then call indirect through the register.
This patch implements this improvement by modifying SelectionDAG to
force a function address which is a memory reference to be loaded
into a virtual register.
Patch by Sriram Murali.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@178171 91177308-0d34-0410-b5e6-96231b3b80d8
All Intel CPUs since Yonah look a lot alike, at least at the granularity
of the scheduling models. We can add more accurate models for
processors that aren't Sandy Bridge if required. Haswell will probably
need its own.
The Atom processor and anything based on NetBurst is completely
different. So are the non-Intel chips.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@178080 91177308-0d34-0410-b5e6-96231b3b80d8
The current Intel Atom microarchitecture has a feature whereby
when a function returns early then it is slightly faster to execute
a sequence of NOP instructions to wait until the return address is ready,
as opposed to simply stalling on the ret instruction until
the return address is ready.
When compiling for X86 Atom only, this patch will run a pass,
called "X86PadShortFunction" which will add NOP instructions where less
than four cycles elapse between function entry and return.
It includes tests.
This patch has been updated to address Nadav's review comments
- Optimize only at >= O1 and don't do optimization if -Os is set
- Stores MachineBasicBlock* instead of BBNum
- Uses DenseMap instead of std::map
- Fixes placement of braces
Patch by Andy Zhang.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171879 91177308-0d34-0410-b5e6-96231b3b80d8
URL: http://llvm.org/viewvc/llvm-project?rev=171524&view=rev
Log:
The current Intel Atom microarchitecture has a feature whereby when a function
returns early then it is slightly faster to execute a sequence of NOP
instructions to wait until the return address is ready,
as opposed to simply stalling on the ret instruction
until the return address is ready.
When compiling for X86 Atom only, this patch will run a pass, called
"X86PadShortFunction" which will add NOP instructions where less than four
cycles elapse between function entry and return.
It includes tests.
Patch by Andy Zhang.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171603 91177308-0d34-0410-b5e6-96231b3b80d8
returns early then it is slightly faster to execute a sequence of NOP
instructions to wait until the return address is ready,
as opposed to simply stalling on the ret instruction
until the return address is ready.
When compiling for X86 Atom only, this patch will run a pass, called
"X86PadShortFunction" which will add NOP instructions where less than four
cycles elapse between function entry and return.
It includes tests.
Patch by Andy Zhang.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@171524 91177308-0d34-0410-b5e6-96231b3b80d8
Not all chips targeted by x86_64 have this feature, but a dramatically
increasing number do. Specifying a chip-specific tuning parameter will
continue to turn the feature on or off as appropriate for that
particular chip, but the generic flag should try to achieve the best
performance on the most widely available hardware. Today, the number of
chips with fast UA access dwarfs those without in the x86-64 space.
Note that this also brings LLVM's code generation for this '-march' flag
more in line with that of modern GCCs. Reviewed by Dan Gohman.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@170269 91177308-0d34-0410-b5e6-96231b3b80d8
Summary:
Not all chips targeted by x86_64 have this feature, but a dramatically
increasing number do. Specifying a chip-specific tuning parameter will
continue to turn the feature on or off as appropriate for that
particular chip, but the generic flag should try to achieve the best
performance on the most widely available hardware. Today, the number of
chips with fast UA access dwarfs those without in the x86-64 space.
Note that this also brings LLVM's code generation for this '-march' flag
more in line with that of modern GCCs.
CC: llvm-commits
Differential Revision: http://llvm-reviews.chandlerc.com/D195
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@169740 91177308-0d34-0410-b5e6-96231b3b80d8
