Integrate measureme's hardware performance counter support.
*Note: this is a companion to https://github.com/rust-lang/measureme/pull/143, and duplicates some information with it for convenience*
**(much later) EDIT**: take any numbers with a grain of salt, they may have changed since initial PR open.
## Credits
I'd like to start by thanking `@alyssais,` `@cuviper,` `@edef1c,` `@glandium,` `@jix,` `@Mark-Simulacrum,` `@m-ou-se,` `@mystor,` `@nagisa,` `@puckipedia,` and `@yorickvP,` for all of their help with testing, and valuable insight and suggestions.
Getting here wouldn't have been possible without you!
(If I've forgotten anyone please let me know, I'm going off memory here, plus some discussion logs)
## Summary
This PR adds support to `-Z self-profile` for counting hardware events such as "instructions retired" (as opposed to being limited to time measurements), using the `rdpmc` instruction on `x86_64` Linux.
While other OSes may eventually be supported, preliminary research suggests some kind of kernel extension/driver is required to enable this, whereas on Linux any user can profile (at least) their own threads.
Supporting Linux on architectures other than x86_64 should be much easier (provided the hardware supports such performance counters), and was mostly not done due to a lack of readily available test hardware.
That said, 32-bit `x86` (aka `i686`) would be almost trivial to add and test once we land the initial `x86_64` version (as all the CPU detection code can be reused).
A new flag `-Z self-profile-counter` was added, to control which of the named `measureme` counters is used, and which defaults to `wall-time`, in order to keep `-Z self-profile`'s current functionality unchanged (at least for now).
The named counters so far are:
* `wall-time`: the existing time measurement
* name chosen for consistency with `perf.rust-lang.org`
* continues to use `std::time::Instant` for a nanosecond-precision "monotonic clock"
* `instructions:u`: the hardware performance counter usually referred to as "Instructions retired"
* here "retired" (roughly) means "fully executed"
* the `:u` suffix is from the Linux `perf` tool and indicates the counter only runs while userspace code is executing, and therefore counts no kernel instructions
* *see [Caveats/Subtracting IRQs](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#Subtracting-IRQs) for why this isn't entirely true and why `instructions-minus-irqs:u` should be preferred instead*
* `instructions-minus-irqs:u`: same as `instructions:u`, except the count of hardware interrupts ("IRQs" here for brevity) is subtracted
* *see [Caveats/Subtracting IRQs](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#Subtracting-IRQs) for why this should be preferred over `instructions:u`*
* `instructions-minus-r0420:u`: experimental counter, same as `instructions-minus-irqs:u` but subtracting an undocumented counter (`r0420:u`) instead of IRQs
* the `rXXXX` notation is again from Linux `perf`, and indicates a "raw" counter, with a hex representation of the low-level counter configuration - this was picked because we still don't *really* know what it is
* this only exists for (future) testing and isn't included/used in any comparisons/data we've put together so far
* *see [Challenges/Zen's undocumented 420 counter](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#Epilogue-Zen’s-undocumented-420-counter) for details on how this counter was found and what it does*
---
There are also some additional commits:
* ~~see [Challenges/Rebasing *shouldn't* affect the results, right?](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#Rebasing-*shouldn’t*-affect-the-results,-right) for details on the changes to `rustc_parse` and `rustc_trait_section` (the latter far more dubious, and probably shouldn't be merged, or not as-is)~~
* **EDIT**: the effects of these are no long quantifiable, the PR includes reverts for them
* ~~see [Challenges/`jemalloc`: purging will commence in ten seconds](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#jemalloc-purging-will-commence-in-ten-seconds) for details on the `jemalloc` change~~
* this is also separately found in #77162, and we probably want to avoid doing it by default, ideally we'd use the runtime control API `jemalloc` offers (assuming that can stop the timer that's already running, which I'm not sure about)
* **EDIT**: until we can do this based on `-Z` flags, this commit has also been reverted
* the `proc_macro` change was to avoid randomized hashing and therefore ASLR-like effects
---
**(much later) EDIT**: take any numbers with a grain of salt, they may have changed since initial PR open.
#### Write-up / report
Because of how extensive the full report ended up being, I've kept most of it [on `hackmd.io`](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view), but for convenient access, here are all the sections (with individual links):
<sup>(someone suggested I'd make a backup, so [here it is on the wayback machine](http://web.archive.org/web/20201127164748/https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view) - I'll need to remember to update that if I have to edit the write-up)</sup>
* [**Motivation**](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#Motivation)
* [**Results**](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#Results)
* [**Overhead**](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#Overhead)
*Preview (see the report itself for more details):*
|Counter|Total<br>`instructions-minus-irqs:u`|Overhead from "Baseline"<br>(for all 1903881<br>counter reads)|Overhead from "Baseline"<br>(per each counter read)|
|-|-|-|-|
|Baseline|63637621286 ±6||
|`instructions:u`|63658815885 ±2| +21194599 ±8| +11|
|`instructions-minus-irqs:u`|63680307361 ±13| +42686075 ±19| +22|
|`wall-time`|63951958376 ±10275|+314337090 ±10281|+165|
* [**"Macro" noise (self time)**](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#“Macro”-noise-(self-time))
*Preview (see the report itself for more details):*
|| `wall-time` (ns) | `instructions:u` | `instructions-minus-irqs:u`
-: | -: | -: | -:
`typeck` | 5478261360 ±283933373 (±~5.2%) | 17350144522 ±6392 (±~0.00004%) | 17351035832.5 ±4.5 (±~0.00000003%)
`expand_crate` | 2342096719 ±110465856 (±~4.7%) | 8263777916 ±2937 (±~0.00004%) | 8263708389 ±0 (±~0%)
`mir_borrowck` | 2216149671 ±119458444 (±~5.4%) | 8340920100 ±2794 (±~0.00003%) | 8341613983.5 ±2.5 (±~0.00000003%)
`mir_built` | 1269059734 ±91514604 (±~7.2%) | 4454959122 ±1618 (±~0.00004%) | 4455303811 ±1 (±~0.00000002%)
`resolve_crate` | 942154987.5 ±53068423.5 (±~5.6%) | 3951197709 ±39 (±~0.000001%) | 3951196865 ±0 (±~0%)
* [**"Micro" noise (individual sampling intervals)**](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#“Micro”-noise-(individual-sampling-intervals))
* [**Caveats**](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#Caveats)
* [**Disabling ASLR**](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#Disabling-ASLR)
* [**Non-deterministic proc macros**](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#Non-deterministic-proc-macros)
* [**Subtracting IRQs**](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#Subtracting-IRQs)
* [**Lack of support for multiple threads**](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#Lack-of-support-for-multiple-threads)
* [**Challenges**](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#Challenges)
* [**How do we even read hardware performance counters?**](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#How-do-we-even-read-hardware-performance-counters)
* [**ASLR: it's free entropy**](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#ASLR-it’s-free-entropy)
* [**The serializing instruction**](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#The-serializing-instruction)
* [**Getting constantly interrupted**](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#Getting-constantly-interrupted)
* [**AMD patented time-travel and dubbed it `SpecLockMap`<br><sup> or: "how we accidentally unlocked `rr` on AMD Zen"</sup>**](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#AMD-patented-time-travel-and-dubbed-it-SpecLockMapnbspnbspnbspnbspnbspnbspnbspnbspor-“how-we-accidentally-unlocked-rr-on-AMD-Zen”)
* [**`jemalloc`: purging will commence in ten seconds**](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#jemalloc-purging-will-commence-in-ten-seconds)
* [**Rebasing *shouldn't* affect the results, right?**](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#Rebasing-*shouldn’t*-affect-the-results,-right)
* [**Epilogue: Zen's undocumented 420 counter**](https://hackmd.io/sH315lO2RuicY-SEt7ynGA?view#Epilogue-Zen’s-undocumented-420-counter)
This adds the typeid and `vcall_visibility` metadata to vtables when the
-Cvirtual-function-elimination flag is set.
The typeid is generated in the same way as for the
`llvm.type.checked.load` intrinsic from the trait_ref.
The offset that is added to the typeid is always 0. This is because LLVM
assumes that vtables are constructed according to the definition in the
Itanium ABI. This includes an "address point" of the vtable. In C++ this
is the offset in the vtable where information for RTTI is placed. Since
there is no RTTI information in Rust's vtables, this "address point" is
always 0. This "address point" in combination with the offset passed to
the `llvm.type.checked.load` intrinsic determines the final function
that should be loaded from the vtable in the
`WholeProgramDevirtualization` pass in LLVM. That's why the
`llvm.type.checked.load` intrinsics are generated with the typeid of the
trait, rather than with that of the function that is called. This
matches what `clang` does for C++.
The vcall_visibility metadata depends on three factors:
1. LTO level: Currently this is always fat LTO, because LLVM only
supports this optimization with fat LTO.
2. Visibility of the trait: If the trait is publicly visible, VFE
can only act on its vtables after linking.
3. Number of CGUs: if there is more than one CGU, also vtables with
restricted visibility could be seen outside of the CGU, so VFE can
only act on them after linking.
To reflect this, there are three visibility levels: Public, LinkageUnit,
and TranslationUnit.
Add the intrinsic
declare {i8*, i1} @llvm.type.checked.load(i8* %ptr, i32 %offset, metadata %type)
This is used in the VFE optimization when lowering loading functions
from vtables to LLVM IR. The `metadata` is used to map the function to
all vtables this function could belong to. This ensures that functions
from vtables that might be used somewhere won't get removed.
This function computes a Itanium-like typeid for a trait_ref. This is
required for the VFE optimization in LLVM. It is used to map
`llvm.type.checked.load` invocations, that is loading the function from
a vtable, to the vtables this function could be from.
It is important to note that `typeid`s are not unique. So multiple
vtables of the same trait can share `typeid`s.
To apply the optimization the `Virtual Function Elim` module flag has to
be set. To apply this optimization post-link the `LTOPostLink` module
flag has to be set.
Adds the virtual-function-elimination unstable compiler flag and a check
that this flag is only used in combination with -Clto. LLVM can only
apply this optimization with fat LTO.
Rework `branches_sharing_code`
fixes#7378
This changes the lint from checking pairs of blocks, to checking all the blocks at the same time. As such there's almost none of the original code left.
changelog: Don't lint `branches_sharing_code` when using different binding names
rustdoc: remove link on slice brackets
This is #91778, take two.
Fixes#91173
The reason I'm reevaluating this change is #97668, which makes fully-generic slices link to the slice docs page. This fixes some downsides in the original PR, where `Box<[T]>`, for example, was not linked to the primitive.slice.html page. In this PR, the `[T]` inside is still a link.
The other major reason for wanting to reevaluate this is the changed color scheme. When this feature was first introduced in rustdoc, primitives were a different color from structs and enums. This way, eagle-eyed users could figure out that the square brackets were separate links from the structs inside. Now, all types have the same color, so a significant fraction of users won't even know the links are there unless they pay close attention to the status bar or use an accessibility tool that lists all links on the page.
lint: add diagnostic translation migration lints
Introduce allow-by-default lints for checking whether diagnostics are written in
`SessionDiagnostic` or `AddSubdiagnostic` impls and whether diagnostics are translatable. These lints can be denied for modules once they are fully migrated to impls and translation.
These lints are intended to be temporary - once all diagnostics have been changed then we can just change the APIs we have and that will enforce these constraints thereafter.
r? `````@oli-obk`````
Rename the `ConstS::val` field as `kind`.
And likewise for the `Const::val` method.
Because its type is called `ConstKind`. Also `val` is a confusing name
because `ConstKind` is an enum with seven variants, one of which is
called `Value`. Also, this gives consistency with `TyS` and `PredicateS`
which have `kind` fields.
The commit also renames a few `Const` variables from `val` to `c`, to
avoid confusion with the `ConstKind::Value` variant.
r? `@BoxyUwU`
Remove RegionckMode in favor of calling new skip_region_resolution
Simple cleanup. We can skip a bunch of stuff for places where NLL does the region checking, so skip earlier.
r? rust-lang/types
This avoids the name clash with `rustc_serialize::Encoder` (a trait),
and allows lots qualifiers to be removed and imports to be simplified
(e.g. fewer `as` imports).
(This was previously merged as commit 5 in #94732 and then was reverted
in #97905 because of a perf regression caused by commit 4 in #94732.)
Our condvar doesn't support setting attributes, like
pthread_condattr_setclock, which the current wait_timeout expects to
have configured.
Switch to a different implementation, following espidf.
