Summary:
Add a target option AllowRegisterRenaming that is used to opt in to
post-register-allocation renaming of registers. This is set to 0 by
default, which causes the hasExtraSrcRegAllocReq/hasExtraDstRegAllocReq
fields of all opcodes to be set to 1, causing
MachineOperand::isRenamable to always return false.
Set the AllowRegisterRenaming flag to 1 for all in-tree targets that
have lit tests that were effected by enabling COPY forwarding in
MachineCopyPropagation (AArch64, AMDGPU, ARM, Hexagon, Mips, PowerPC,
RISCV, Sparc, SystemZ and X86).
Add some more comments describing the semantics of the
MachineOperand::isRenamable function and how it is set and maintained.
Change isRenamable to check the operand's opcode
hasExtraSrcRegAllocReq/hasExtraDstRegAllocReq bit directly instead of
relying on it being consistently reflected in the IsRenamable bit
setting.
Clear the IsRenamable bit when changing an operand's register value.
Remove target code that was clearing the IsRenamable bit when changing
registers/opcodes now that this is done conservatively by default.
Change setting of hasExtraSrcRegAllocReq in AMDGPU target to be done in
one place covering all opcodes that have constant pipe read limit
restrictions.
Reviewers: qcolombet, MatzeB
Subscribers: aemerson, arsenm, jyknight, mcrosier, sdardis, nhaehnle, javed.absar, tpr, arichardson, kristof.beyls, kbarton, fedor.sergeev, asb, rbar, johnrusso, simoncook, jordy.potman.lists, apazos, sabuasal, niosHD, escha, nemanjai, llvm-commits
Differential Revision: https://reviews.llvm.org/D43042
llvm-svn: 325931
Cannon Lake does not support CLWB, therefore it
does not include all features listed under SKX anymore.
Instead, enumerate all SKX features with the exception of CLWB.
Patch by Gabor Buella
Differential Revision: https://reviews.llvm.org/D43380
llvm-svn: 325654
We currently emit up to 15-byte NOPs on all targets (apart from Silvermont), which stalls performance on some targets with decoders that struggle with 2 or 3 more '66' prefixes.
This patch flags recent AMD targets (btver1/znver1) to still emit 15-byte NOPs and bdver* targets to emit 11-byte NOPs. All other targets now emit 10-byte NOPs apart from SilverMont CPUs which still emit 7-byte NOPS.
Differential Revision: https://reviews.llvm.org/D42616
llvm-svn: 323693
Summary:
First, we need to explain the core of the vulnerability. Note that this
is a very incomplete description, please see the Project Zero blog post
for details:
https://googleprojectzero.blogspot.com/2018/01/reading-privileged-memory-with-side.html
The basis for branch target injection is to direct speculative execution
of the processor to some "gadget" of executable code by poisoning the
prediction of indirect branches with the address of that gadget. The
gadget in turn contains an operation that provides a side channel for
reading data. Most commonly, this will look like a load of secret data
followed by a branch on the loaded value and then a load of some
predictable cache line. The attacker then uses timing of the processors
cache to determine which direction the branch took *in the speculative
execution*, and in turn what one bit of the loaded value was. Due to the
nature of these timing side channels and the branch predictor on Intel
processors, this allows an attacker to leak data only accessible to
a privileged domain (like the kernel) back into an unprivileged domain.
The goal is simple: avoid generating code which contains an indirect
branch that could have its prediction poisoned by an attacker. In many
cases, the compiler can simply use directed conditional branches and
a small search tree. LLVM already has support for lowering switches in
this way and the first step of this patch is to disable jump-table
lowering of switches and introduce a pass to rewrite explicit indirectbr
sequences into a switch over integers.
However, there is no fully general alternative to indirect calls. We
introduce a new construct we call a "retpoline" to implement indirect
calls in a non-speculatable way. It can be thought of loosely as
a trampoline for indirect calls which uses the RET instruction on x86.
Further, we arrange for a specific call->ret sequence which ensures the
processor predicts the return to go to a controlled, known location. The
retpoline then "smashes" the return address pushed onto the stack by the
call with the desired target of the original indirect call. The result
is a predicted return to the next instruction after a call (which can be
used to trap speculative execution within an infinite loop) and an
actual indirect branch to an arbitrary address.
On 64-bit x86 ABIs, this is especially easily done in the compiler by
using a guaranteed scratch register to pass the target into this device.
For 32-bit ABIs there isn't a guaranteed scratch register and so several
different retpoline variants are introduced to use a scratch register if
one is available in the calling convention and to otherwise use direct
stack push/pop sequences to pass the target address.
This "retpoline" mitigation is fully described in the following blog
post: https://support.google.com/faqs/answer/7625886
We also support a target feature that disables emission of the retpoline
thunk by the compiler to allow for custom thunks if users want them.
These are particularly useful in environments like kernels that
routinely do hot-patching on boot and want to hot-patch their thunk to
different code sequences. They can write this custom thunk and use
`-mretpoline-external-thunk` *in addition* to `-mretpoline`. In this
case, on x86-64 thu thunk names must be:
```
__llvm_external_retpoline_r11
```
or on 32-bit:
```
__llvm_external_retpoline_eax
__llvm_external_retpoline_ecx
__llvm_external_retpoline_edx
__llvm_external_retpoline_push
```
And the target of the retpoline is passed in the named register, or in
the case of the `push` suffix on the top of the stack via a `pushl`
instruction.
There is one other important source of indirect branches in x86 ELF
binaries: the PLT. These patches also include support for LLD to
generate PLT entries that perform a retpoline-style indirection.
The only other indirect branches remaining that we are aware of are from
precompiled runtimes (such as crt0.o and similar). The ones we have
found are not really attackable, and so we have not focused on them
here, but eventually these runtimes should also be replicated for
retpoline-ed configurations for completeness.
For kernels or other freestanding or fully static executables, the
compiler switch `-mretpoline` is sufficient to fully mitigate this
particular attack. For dynamic executables, you must compile *all*
libraries with `-mretpoline` and additionally link the dynamic
executable and all shared libraries with LLD and pass `-z retpolineplt`
(or use similar functionality from some other linker). We strongly
recommend also using `-z now` as non-lazy binding allows the
retpoline-mitigated PLT to be substantially smaller.
When manually apply similar transformations to `-mretpoline` to the
Linux kernel we observed very small performance hits to applications
running typical workloads, and relatively minor hits (approximately 2%)
even for extremely syscall-heavy applications. This is largely due to
the small number of indirect branches that occur in performance
sensitive paths of the kernel.
When using these patches on statically linked applications, especially
C++ applications, you should expect to see a much more dramatic
performance hit. For microbenchmarks that are switch, indirect-, or
virtual-call heavy we have seen overheads ranging from 10% to 50%.
However, real-world workloads exhibit substantially lower performance
impact. Notably, techniques such as PGO and ThinLTO dramatically reduce
the impact of hot indirect calls (by speculatively promoting them to
direct calls) and allow optimized search trees to be used to lower
switches. If you need to deploy these techniques in C++ applications, we
*strongly* recommend that you ensure all hot call targets are statically
linked (avoiding PLT indirection) and use both PGO and ThinLTO. Well
tuned servers using all of these techniques saw 5% - 10% overhead from
the use of retpoline.
We will add detailed documentation covering these components in
subsequent patches, but wanted to make the core functionality available
as soon as possible. Happy for more code review, but we'd really like to
get these patches landed and backported ASAP for obvious reasons. We're
planning to backport this to both 6.0 and 5.0 release streams and get
a 5.0 release with just this cherry picked ASAP for distros and vendors.
This patch is the work of a number of people over the past month: Eric, Reid,
Rui, and myself. I'm mailing it out as a single commit due to the time
sensitive nature of landing this and the need to backport it. Huge thanks to
everyone who helped out here, and everyone at Intel who helped out in
discussions about how to craft this. Also, credit goes to Paul Turner (at
Google, but not an LLVM contributor) for much of the underlying retpoline
design.
Reviewers: echristo, rnk, ruiu, craig.topper, DavidKreitzer
Subscribers: sanjoy, emaste, mcrosier, mgorny, mehdi_amini, hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D41723
llvm-svn: 323155
This will cause the vectorizers to do some limiting of the vector widths they create. This is not a strict limit. There are reasons I know of that the loop vectorizer will generate larger vectors for.
I've written this in such a way that the interface will only return a properly supported width(0/128/256/512) even if the attribute says something funny like 384 or 10.
This has been split from D41895 with the remainder in a follow up commit.
llvm-svn: 323015
This adds a new instrinsic to support the rdpid instruction. The implementation is a bit weird because the intrinsic is defined as always returning 32-bits, but the assembler support thinks the instruction produces a 64-bit register in 64-bit mode. But really it zeros the upper 32 bits. So I had to add separate patterns where 64-bit mode uses an extract_subreg.
Differential Revision: https://reviews.llvm.org/D42205
llvm-svn: 322910
After D41349, we can no get a MCSubtargetInfo into the MCAsmBackend constructor. This allows us to get NOPL from a subtarget feature rather than a CPU name blacklist.
Differential Revision: https://reviews.llvm.org/D41721
llvm-svn: 322227
Previously prefetch was only considered legal if sse was enabled, but it should be supported with 3dnow as well.
The prfchw flag now imply at least some form of prefetch without the write hint is available, either the sse or 3dnow version. This is true even if 3dnow and sse are explicitly disabled.
Similarly prefetchwt1 feature implies availability of prefetchw and the the prefetcht0/1/2/nta instructions. This way we can support _MM_HINT_ET0 using prefetchw and _MM_HINT_ET1 with prefetchwt1. And its assumed that if we have levels for the write hint we would have levels for the non-write hint, thus why we enable the sse prefetch instructions.
I believe this behavior is consistent with gcc. I've updated the prefetch.ll to test all of these combinations.
llvm-svn: 321335
As mentioned in D38318 and D40865, modern Intel processors prefer to combine multiple shuffles to a variable shuffle mask (PSHUFB/VPERMPS etc.) instead of having multiple stage 'fixed' shuffles which put more pressure on Port 5 (at the expense of extra shuffle mask loads).
This patch provides a FeatureFastVariableShuffle target flag for Haswell+ CPUs that prefers combining 2 or more fixed shuffles to a single variable shuffle (default is 3 shuffles).
The long term aim is to drive more of this from schedule data (probably via the MC) but we're not close to being ready for that yet.
Differential Revision: https://reviews.llvm.org/D41323
llvm-svn: 321074
This separates the CPU specific scheduler model includes to occur after the instructions. Moves the instruction includes between the basic scheduler information and the CPU specific scheduler models.
llvm-svn: 320313
Shadow stack solution introduces a new stack for return addresses only.
The HW has a Shadow Stack Pointer (SSP) that points to the next return address.
If we return to a different address, an exception is triggered.
The shadow stack is managed using a series of intrinsics that are introduced in this patch as well as the new register (SSP).
The intrinsics are mapped to new instruction set that implements CET mechanism.
The patch also includes initial infrastructure support for IBT.
For more information, please see the following:
https://software.intel.com/sites/default/files/managed/4d/2a/control-flow-enforcement-technology-preview.pdf
Differential Revision: https://reviews.llvm.org/D40223
Change-Id: I4daa1f27e88176be79a4ac3b4cd26a459e88fed4
llvm-svn: 318996
Summary:
This adds a new fast gather feature bit to cover all CPUs that support fast gather that we can use independent of whether the AVX512 feature is enabled. I'm only using this new bit to qualify AVX2 codegen. AVX512 is still implicitly assuming fast gather to keep tests working and to match the scatter behavior.
Test command lines have been added for these two cases.
Reviewers: magabari, delena, RKSimon, zvi
Reviewed By: RKSimon
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D40282
llvm-svn: 318983
This is based on table 1-1 of the October 2017 revision of Intel® Architecture Instruction Set Extensions and Future Features Programming Reference
llvm-svn: 318799
The EVEX to VEX pass is already assuming this is true under AVX512VL. We had special patterns to use zmm instructions if VLX and F16C weren't available.
Instead just make AVX512 imply F16C to make the EVEX to VEX behavior explicitly legal and remove the extra patterns.
All known CPUs with AVX512 have F16C so this should safe for now.
llvm-svn: 317521
Previously our VEX patterns were checking Subtarget.hasFMA() which checked FMA || AVX512. So we were behaving as if AVX512 implied it anyway. Which means we'd allow VEX encoded 128/256 FMA when AVX512F was enabled but AVX512VL is off. Regardless of the FMA flag.
EVEX to VEX also transforms scalar EVEX FMA instructions to their VEX versions even without the FMA flag. Similarly for 128/256 under AVX512VL.
So this makes AVX512 imply FeatureFMA to make our current behavior explicit.
All known CPUs that support AVX512 have VEX FMA instructions.
llvm-svn: 317520
As indicated by Table 1-1 in Intel Architecture Instruction Set Extensions and Future Features Programming Reference from October 2017.
llvm-svn: 316592
Adding the scheduling information for the Browadwell (BDW) CPU target.
This patch adds the instruction scheduling information for the Broadwell (BDW) architecture target by adding the file X86SchedBroadwell.td located under the X86 Target.
We used the scheduling information retrieved from the Broadwell architects in order to create the file.
The scheduling information includes latency, number of micro-Ops and used ports by each BDW instruction.
The patch continues the scheduling replacement and insertion effort started with the SandyBridge (SNB) target in r310792, the Haswell (HSW) target in r311879, the SkylakeClient (SKL) target in rL313613 + rL315978 and the SkylakeServer (SKX) in rL315175.
Performance fluctuations may be expected due to code alignment effects.
Reviewers: zvi, RKSimon, craig.topper
Differential Revision: https://reviews.llvm.org/D39054
Change-Id: If6f799e5ff60e1091c8d43b05ea78c53581bae01
llvm-svn: 316492
Turns out we have no patterns on the instructions that were using this feature flag for other reasons. These instructions are slow on all modern CPUs so it seems unlikely that we will spend any effort supporting these instructions going forward. So we might as well just kill of the feature flag and just fix up the comments.
llvm-svn: 315862
Summary: I see nothing in Agner Fog's tables to indicate that this improved between Ivy Bridge and Haswell. It's also set for all Atom CPUs so I assume KNL should have it too.
Reviewers: RKSimon, zvi, gadi.haber
Reviewed By: gadi.haber
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D38890
llvm-svn: 315859
This adds Intel's Knights Mill CPU to valid CPU names for the backend. For now its an alias of "knl", but ultimately we need to support AVX5124FMAPS and AVX5124VNNIW instruction sets for it.
Differential Revision: https://reviews.llvm.org/D38811
llvm-svn: 315722
Adding the scheduling information for the SkylakeServer (SKX) target.
This patch adds the instruction scheduling information for the SkylakeServer (SKX) architecture target by adding the file X86SchedSkylakeServer.td located under the X86 Target.
We used the scheduling information retrieved from the Skylake architects in order to create the file.
The scheduling information includes latency, number of micro-Ops and used ports by each SKL instruction.
The patch continues the scheduling replacement and insertion effort started with the SNB target in r310792, the HSW target in r311879 and the SkylakeClient (SKL) target in rL313613.
Please expect some performance fluctuations due to code alignment effects.
Reviewers: zvi, RKSimon, craig.topper, chandlerc, aymanmu
Differential Revision: https://reviews.llvm.org/D38443
Change-Id: I5c228fcc09e9e5a99b6116e62b356c4f9b971185
llvm-svn: 315175
This patch adds the instruction scheduling information for the SkylakeClient (SKL) architecture target by adding the file X86SchedSkylakeClient.td located under the X86 Target.
We used the scheduling information retrieved from the Skylake architects in order to create the file.
The scheduling information includes latency, number of micro-Ops and used ports by each SKL instruction.
The patch continues the scheduling replacement and insertion effort started with the SNB target in r307529 and r310792 and for HSW in r311879.
Please expect some performance fluctuations due to code alignment effects.
Reviewers: craig.topper, zvi, chandlerc, igorb, aymanmus, RKSimon, delena
Differential Revision: https://reviews.llvm.org/D37294
llvm-svn: 313613
Adding x86 Processor families to initialize several uArch properties (based on the family)
This patch shows how gather cost can be initialized based on the proc. family
Differential Revision: https://reviews.llvm.org/D35348
llvm-svn: 313132
Currently we start applying this on Haswell and newer. I don't believe anything changed in the Haswell architecture to make this the right cutoff point. The partial flag handling around this has been roughly the same since Sandybridge.
Differential Revision: https://reviews.llvm.org/D37250
llvm-svn: 312099
Summary:
Currently we determine if macro fusion is supported based on the AVX flag as a proxy for the processor being Sandy Bridge".
This is really strange as now AMD supports AVX. It also means if user explicitly disables AVX we disable macro fusion.
This patch adds an explicit macro fusion feature. I've also enabled for the generic 64-bit CPU (which doesn't have AVX)
This is probably another candidate for being in the MI layer, but for now I at least wanted to correct the overloading of the AVX feature.
Reviewers: spatel, chandlerc, RKSimon, zvi
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D37280
llvm-svn: 312097
Summary: Knights Landing, because it is Atom derived, has slow two memory operand instructions. Mark the Knights Landing CPU model accordingly.
Patch by David Zarzycki.
Reviewers: craig.topper
Reviewed By: craig.topper
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D37224
llvm-svn: 311979
FeatureSlowUAMem32.
The idea was to mark things that are slow on widely available processors
as slow in the generic CPU so that the code generated for that CPU would
be fast across those processors. However, for this feature that doesn't
work out very well at all.
The problem here is that you can very easily enable AVX or AVX2 on top
of this generic CPU. For example, this can happen just by using AVX2
intrinsics from Clang within a region of code guarded by a dynamic CPU
feature test. When you do that, the generated code with SlowUAMem32 set
is ... amazingly slower. The problem is that there really aren't very
good alternatives to the unaligned loads, and so our vector codegen
regresses significantly.
The other issue is that there are plenty of AMD CPUs with AVX1 that
don't set FeatureSlowUAMem32 and so we shouldn't just check for AVX2
instead of this special feature. =/
It would be nice to have the target attriute logic be able to
enable/disable more than just one feature at a time and control this in
a more fine grained and useful way, but that doesn't seem easy. Given
that it is only Sandybridge and Ivybridge that set this feature, for now
I'm just backing it out of the generic CPU. That has the additional
advantage of going back to the previous state that people seemed vaguely
happy with.
llvm-svn: 311740
widely used processors.
This occured to me when I saw that we were generating 'inc' and 'dec'
when for Haswell and newer we shouldn't. However, there were a few "X is
slow" things that we should probably just set.
I've avoided any of the "X is fast" features because most of those would
be pretty serious regressions on processors where X isn't actually fast.
The slow things are likely to be negligible costs on processors where
these aren't slow and a significant win when they are slow.
In retrospect this seems somewhat obvious. Not sure why we didn't do
this a long time ago.
Differential Revision: https://reviews.llvm.org/D36947
llvm-svn: 311318
Summary:
This patch adds the following
1. Adds a skeleton scheduler model for AMD Znver1.
2. Introduces the znver1 execution units and pipes.
3. Caters the instructions based on the generic scheduler classes.
4. Further additions to the scheduler model with instruction itineraries will be carried out incrementally based on
a. Instructions types
b. Registers used
5. Since itineraries are not added based on instructions, throughput information are bound to change when incremental changes are added.
6. Scheduler testcases are modified accordingly to suit the new model.
Patch by Ganesh Gopalasubramanian. With minor formatting tweaks from me.
Reviewers: craig.topper, RKSimon
Subscribers: javed.absar, shivaram, ddibyend, vprasad
Differential Revision: https://reviews.llvm.org/D35293
llvm-svn: 308411
Summary: I believe this should be supported on GLM since RDSEED is.
Reviewers: m_zuckerman, zvi, RKSimon
Reviewed By: RKSimon
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D34828
llvm-svn: 307060
AVX512_VPOPCNTDQ is a new feature set that was published by Intel.
The patch represents the LLVM side of the addition of two new intrinsic based instructions (vpopcntd and vpopcntq).
Differential Revision: https://reviews.llvm.org/D33169
llvm-svn: 303858
Geoff Berry