The profiles for memory allocations, sync.Mutex contention, and general
blocking store their data in a shared hash table. The bookkeeping work
at the end of a garbage collection cycle involves maintenance on each
memory allocation record. Previously, a single lock guarded access to
the hash table and the contents of all records. When a program has
allocated memory at a large number of unique call stacks, the
maintenance following every garbage collection can hold that lock for
several milliseconds. That can prevent progress on all other goroutines
by delaying acquirep's call to mcache.prepareForSweep, which needs the
lock in mProf_Free to report when a profiled allocation is no longer in
use. With no user goroutines making progress, it is in effect a
multi-millisecond GC-related stop-the-world pause.
Split the lock so the call to mProf_Flush no longer delays each P's call
to mProf_Free: mProf_Free uses a lock on the memory records' N+1 cycle,
and mProf_Flush uses locks on the memory records' accumulator and their
N cycle. mProf_Malloc also no longer competes with mProf_Flush, as it
uses a lock on the memory records' N+2 cycle. The profiles for
sync.Mutex contention and general blocking now share a separate lock,
and another lock guards insertions to the shared hash table (uncommon in
the steady-state). Consumers of each type of profile take the matching
accumulator lock, so will observe consistent count and magnitude values
for each record.
For #45894
Change-Id: I615ff80618d10e71025423daa64b0b7f9dc57daa
Reviewed-on: https://go-review.googlesource.com/c/go/+/399956
Reviewed-by: Carlos Amedee <carlos@golang.org>
Run-TryBot: Rhys Hiltner <rhys@justin.tv>
Reviewed-by: Michael Knyszek <mknyszek@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Fundamentally, all of these memstats exist to serve the runtime in
managing memory. For the sake of simpler testing, couple these stats
more tightly with the GC.
This CL was mostly done automatically. The fields had to be moved
manually, but the references to the fields were updated via
gofmt -w -r 'memstats.<field> -> gcController.<field>' *.go
For #48409.
Change-Id: Ic036e875c98138d9a11e1c35f8c61b784c376134
Reviewed-on: https://go-review.googlesource.com/c/go/+/397678
Reviewed-by: Michael Pratt <mpratt@google.com>
Run-TryBot: Michael Knyszek <mknyszek@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
The inconsistent heaps stats in memstats are a bit messy. Primarily,
heap_sys is non-orthogonal with heap_released and heap_inuse. In later
CLs, we're going to want heap_sys-heap_released-heap_inuse, so clean
this up by replacing heap_sys with an orthogonal metric: heapFree.
heapFree represents page heap memory that is free but not released.
I think this change also simplifies a lot of reasoning about these
stats; it's much clearer what they mean, and to obtain HeapSys for
memstats, we no longer need to do the strange subtraction from heap_sys
when allocating specifically non-heap memory from the page heap.
Because we're removing heap_sys, we need to replace it with a sysMemStat
for mem.go functions. In this case, heap_released is the most
appropriate because we increase it anyway (again, non-orthogonality). In
which case, it makes sense for heap_inuse, heap_released, and heapFree
to become more uniform, and to just represent them all as sysMemStats.
While we're here and messing with the types of heap_inuse and
heap_released, let's also fix their names (and last_heap_inuse's name)
up to the more modern Go convention of camelCase.
For #48409.
Change-Id: I87fcbf143b3e36b065c7faf9aa888d86bd11710b
Reviewed-on: https://go-review.googlesource.com/c/go/+/397677
Run-TryBot: Michael Knyszek <mknyszek@google.com>
Reviewed-by: Michael Pratt <mpratt@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
This change adds a field to memstats called mappedReady that tracks how
much memory is in the Ready state at any given time. In essence, it's
the total memory usage by the Go runtime (with one exception which is
documented). Essentially, all memory mapped read/write that has either
been paged in or will soon.
To make tracking this not involve the many different stats that track
mapped memory, we track this statistic at a very low level. The downside
of tracking this statistic at such a low level is that it managed to
catch lots of situations where the runtime wasn't fully accounting for
memory. This change rectifies these situations by always accounting for
memory that's mapped in some way (i.e. always passing a sysMemStat to a
mem.go function), with *two* exceptions.
Rectifying these situations means also having the memory mapped during
testing being accounted for, so that tests (i.e. ReadMemStats) that
ultimately check mappedReady continue to work correctly without special
exceptions. We choose to simply account for this memory in other_sys.
Let's talk about the exceptions. The first is the arenas array for
finding heap arena metadata from an address is mapped as read/write in
one large chunk. It's tens of MiB in size. On systems with demand
paging, we assume that the whole thing isn't paged in at once (after
all, it maps to the whole address space, and it's exceedingly difficult
with today's technology to even broach having as much physical memory as
the total address space). On systems where we have to commit memory
manually, we use a two-level structure.
Now, the reason why this is an exception is because we have no mechanism
to track what memory is paged in, and we can't just account for the
entire thing, because that would *look* like an enormous overhead.
Furthermore, this structure is on a few really, really critical paths in
the runtime, so doing more explicit tracking isn't really an option. So,
we explicitly don't and call sysAllocOS to map this memory.
The second exception is that we call sysFree with no accounting to clean
up address space reservations, or otherwise to throw out mappings we
don't care about. In this case, also drop down to a lower level and call
sysFreeOS to explicitly avoid accounting.
The third exception is debuglog allocations. That is purely a debugging
facility and ideally we want it to have as small an impact on the
runtime as possible. If we include it in mappedReady calculations, it
could cause GC pacing shifts in future CLs, especailly if one increases
the debuglog buffer sizes as a one-off.
As of this CL, these are the only three places in the runtime that would
pass nil for a stat to any of the functions in mem.go. As a result, this
CL makes sysMemStats mandatory to facilitate better accounting in the
future. It's now much easier to grep and find out where accounting is
explicitly elided, because one doesn't have to follow the trail of
sysMemStat nil pointer values, and can just look at the function name.
For #48409.
Change-Id: I274eb467fc2603881717482214fddc47c9eaf218
Reviewed-on: https://go-review.googlesource.com/c/go/+/393402
Reviewed-by: Michael Pratt <mpratt@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Run-TryBot: Michael Knyszek <mknyszek@google.com>
A future change to gofmt will rewrite
// Doc comment.
//go:foo
to
// Doc comment.
//
//go:foo
Apply that change preemptively to all comments (not necessarily just doc comments).
For #51082.
Change-Id: Iffe0285418d1e79d34526af3520b415a12203ca9
Reviewed-on: https://go-review.googlesource.com/c/go/+/384260
Trust: Russ Cox <rsc@golang.org>
Run-TryBot: Russ Cox <rsc@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
This change lifts all non-platform-specific code out of sys* functions
for each platform up into wrappers, and moves documentation about the OS
virtual memory abstraction layer from malloc.go to mem.go, which
contains those wrappers.
Change-Id: Ie803e4447403eaafc508b34b53a1a47d6cee9388
Reviewed-on: https://go-review.googlesource.com/c/go/+/393398
Reviewed-by: Austin Clements <austin@google.com>
Reviewed-by: Michael Pratt <mpratt@google.com>
Trust: Michael Knyszek <mknyszek@google.com>
Currently, the ReadMemStats (really this is all happening in
readmemstats_m, but that's just a direct call from ReadMemStats) call
chain first populates some fields in memstats, then copies those into
the final MemStats location. This used to make a lot of sense when
memstats' structure aligned with MemStats, and the values were just
copied from one to other. Sometime in the last few releases, we switched
to populating the MemStats manually because a lot of fields had diverged
from their internal representation. Now, we're left with a lot of fields
in memstats that pollute the structure: they only exist to be updated
for the sake of ReadMemStats. Since we're going to be adding more fields
to memstats in further CLs, this is a good opportunity to clean up.
As a result of this change, updatememstats, which used to just update
the aforementioned intermediate fields in memstats, is no longer
necessary, so it is removed.
Change-Id: Ifabfb3ac3002641105af62e9509a6351165dcd87
Reviewed-on: https://go-review.googlesource.com/c/go/+/393397
Trust: Michael Knyszek <mknyszek@google.com>
Reviewed-by: Austin Clements <austin@google.com>
Reviewed-by: Michael Pratt <mpratt@google.com>
In iOS <14, the address space is strictly limited to 8 GiB, or 33 bits.
As a result, the page allocator also assumes all heap memory lives in
this region. This is especially necessary because the page allocator has
a PROT_NONE mapping proportional to the size of the usable address
space, so this keeps that mapping very small.
However starting with iOS 14, this restriction is relaxed, and mmap may
start returning addresses outside of the <14 range. Today this means
that in iOS 14 and later, users experience an error in the page
allocator when a heap arena is mapped outside of the old range.
This change increases the ios/arm64 heapAddrBits to 40 while
simultaneously making ios/arm64 use the 64-bit pagealloc implementation
(with reservations and incremental mapping) to accommodate both iOS
versions <14 and 14+.
Once iOS <14 is deprecated, we can remove these exceptions and treat
ios/arm64 like any other arm64 platform.
This change also makes the BaseChunkIdx expression a little bit easier
to read, while we're here.
Fixes#46860.
Change-Id: I13865f799777739109585f14f1cc49d6d57e096b
Reviewed-on: https://go-review.googlesource.com/c/go/+/344401
Trust: Michael Knyszek <mknyszek@google.com>
Run-TryBot: Michael Knyszek <mknyszek@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Cherry Mui <cherryyz@google.com>
Reviewed-by: Austin Clements <austin@google.com>
Add explicit address sanitizer instrumentation to the runtime and
syscall packages. The compiler does not instrument the runtime
package. It does instrument the syscall package, but we need to add
a couple of cases that it can't see.
Refer to the implementation of the asan malloc runtime library,
this patch also allocates extra memory as the redzone, around the
returned memory region, and marks the redzone as unaddressable to
detect the overflows or underflows.
Updates #44853.
Change-Id: I2753d1cc1296935a66bf521e31ce91e35fcdf798
Reviewed-on: https://go-review.googlesource.com/c/go/+/298614
Run-TryBot: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Trust: fannie zhang <Fannie.Zhang@arm.com>
Currently, the runtime zeroes allocations in several ways. First, small
object spans are always zeroed if they come from mheap, and their slots
are zeroed later in mallocgc if needed. Second, large object spans
(objects that have their own spans) plumb the need for zeroing down into
mheap. Thirdly, large objects that have no pointers have their zeroing
delayed until after preemption is reenabled, but before returning in
mallocgc.
All of this has two consequences:
1. Spans for small objects that come from mheap are sometimes
unnecessarily zeroed, even if the mallocgc call that created them
doesn't need the object slot to be zeroed.
2. This is all messy and difficult to reason about.
This CL simplifies this code, resolving both (1) and (2). First, it
recognizes that zeroing in mheap is unnecessary for small object spans;
mallocgc and its callees in mcache and mcentral, by design, are *always*
able to deal with non-zeroed spans. They must, for they deal with
recycled spans all the time. Once this fact is made clear, the only
remaining use of zeroing in mheap is for large objects.
As a result, this CL lifts mheap zeroing for large objects into
mallocgc, to parallel all the other codepaths in mallocgc. This is makes
the large object allocation code less surprising.
Next, this CL sets the flag for the delayed zeroing explicitly in the one
case where it matters, and inverts and renames the flag from isZeroed to
delayZeroing.
Finally, it adds a check to make sure that only pointer-free allocations
take the delayed zeroing codepath, as an extra safety measure.
Benchmark results: https://perf.golang.org/search?q=upload:20211028.8
Inspired by tapir.liu@gmail.com's CL 343470.
Change-Id: I7e1296adc19ce8a02c8d93a0a5082aefb2673e8f
Reviewed-on: https://go-review.googlesource.com/c/go/+/359477
Trust: Michael Knyszek <mknyszek@google.com>
Reviewed-by: David Chase <drchase@google.com>
Since all callers of getMCache appear to have mp available,
we pass the mp to getMCache, and reduce one call to getg.
And after modification, getMCache is also inlined.
Change-Id: Ib7880c118336acc026ecd7c60c5a88722c3ddfc7
Reviewed-on: https://go-review.googlesource.com/c/go/+/349329
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Michael Knyszek <mknyszek@google.com>
Trust: Michael Knyszek <mknyszek@google.com>
Trust: Carlos Amedee <carlos@golang.org>
It was not safe to do mcache profiling updates outside the critical
section, but we got lucky because the runtime was not preemptible.
Adding chunked memory clearing (CL 270943) created preemption
opportunities, which led to corruption of runtime data structures.
Fixes#47304.
Fixes#47302.
Change-Id: I461615470d62328a83ccbac537fbdc6dcde81c85
Reviewed-on: https://go-review.googlesource.com/c/go/+/336449
Trust: David Chase <drchase@google.com>
Run-TryBot: David Chase <drchase@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Austin Clements <austin@google.com>
Reviewed-by: Keith Randall <khr@golang.org>
Now that deferred functions are always argumentless and defer
records are no longer with arguments, defer record can be fixed
size (just the _defer struct). This allows us to simplify the
allocation of defer records, specifically, remove the defer
classes and the pools of different sized defers.
Change-Id: Icc4b16afc23b38262ca9dd1f7369ad40874cf701
Reviewed-on: https://go-review.googlesource.com/c/go/+/326062
Trust: Cherry Mui <cherryyz@google.com>
Run-TryBot: Cherry Mui <cherryyz@google.com>
Reviewed-by: Michael Knyszek <mknyszek@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
If something "huge" is allocated, and the zeroing is trivial (no pointers
involved) then zero it by chunks in a loop so that preemption can occur,
not all in a single non-preemptible call.
Benchmarking suggests that 256K is the best chunk size.
Updates #42642.
Change-Id: I94015e467eaa098c59870e479d6d83bc88efbfb4
Reviewed-on: https://go-review.googlesource.com/c/go/+/270943
Trust: David Chase <drchase@google.com>
Run-TryBot: David Chase <drchase@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Michael Knyszek <mknyszek@google.com>
Top align allocations in tinyalloc buckets when in race mode.
This will make checkptr checks more reliable, because any code
that modifies a pointer past the end of the object will trigger
a checkptr error.
No test, because we need -race for this to actually kick in. We could
add it to the race detector tests, but the race detector tests are all
geared towards race detector reports, not checkptr reports. Mucking
with parsing reports is more than a test is worth.
Fixes#38872
Change-Id: Ie56f0fbd1a9385539f6631fd1ac40c3de5600154
Reviewed-on: https://go-review.googlesource.com/c/go/+/315029
Trust: Keith Randall <khr@golang.org>
Run-TryBot: Keith Randall <khr@golang.org>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
iOS arm64 is a 64-bit platform but with a strictly 32-bit address space
(technically 33 bits, but the bottom half is unavailable to the
application). Since address space is limited, use 4 MiB arenas instead
of 64 MiB arenas. No changes are needed to the arena index because it's
still relatively small; this change just brings iOS more in line with
32-bit platforms.
Change-Id: I484e2d273d896fd0a57cd5c25012df0aef160290
Reviewed-on: https://go-review.googlesource.com/c/go/+/270538
Trust: Michael Knyszek <mknyszek@google.com>
Trust: Emmanuel Odeke <emmanuel@orijtech.com>
Trust: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Michael Knyszek <mknyszek@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Austin Clements <austin@google.com>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
This change moves the call of sysMap from (*mheap).sysAlloc into
(*mheap).grow, so we only sysMap what we're going to use in the near
future (thanks to the curArena mechanism). The purpose of this change is
to better support systems with strict overcommit rules which generally
accept reserved memory but not prepared memory (see malloc.go for exact
descriptions of these states).
This move requires changing linearAlloc to only optionally map memory.
In one case, with mheap.heapArenaAlloc, we do want it to map memory. But
now in the other case, with mheap.arena, we don't, because we want grow
to take care of it.
The risk with this change is we may make more syscalls than before on
systems with 64 MiB arenas, but because heap growth is relatively rare
this is unlikely to be a noticable issue. We also bound the amount of
syscalls made by only extending curArena (and thus mapping) by
pallocChunkPages*pageSize which is 4 MiB.
Fixes#42612.
Change-Id: I736df696afe78ddb1a747a896caa0db8726027e5
Reviewed-on: https://go-review.googlesource.com/c/go/+/270537
Trust: Michael Knyszek <mknyszek@google.com>
Run-TryBot: Michael Knyszek <mknyszek@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Michael Pratt <mpratt@google.com>
This change moves the responsibility of throwing if an mcache is not
available to the caller, because the inlining cost of throw is set very
high in the compiler. Even if it was reduced down to the cost of a usual
function call, it would still be too expensive, so just move it out.
This choice also makes sense in the context of #42339 since we're going
to have to handle the case where we don't have an mcache to update stats
in a few contexts anyhow.
Also, add getMCache to the list of functions that should be inlined to
prevent future regressions.
getMCache is called on the allocation fast path and because its not
inlined actually causes a significant regression (~10%) in some
microbenchmarks.
Fixes#42305.
Change-Id: I64ac5e4f26b730bd4435ea1069a4a50f55411ced
Reviewed-on: https://go-review.googlesource.com/c/go/+/267157
Trust: Michael Knyszek <mknyszek@google.com>
Run-TryBot: Michael Knyszek <mknyszek@google.com>
Reviewed-by: Michael Pratt <mpratt@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Some functions that required holding the heap lock _or_ world stop have
been simplified to simply requiring the heap lock. This is conceptually
simpler and taking the heap lock during world stop is guaranteed to not
contend. This was only done on functions already called on the
systemstack to avoid too many extra systemstack calls in GC.
Updates #40677
Change-Id: I15aa1dadcdd1a81aac3d2a9ecad6e7d0377befdc
Reviewed-on: https://go-review.googlesource.com/c/go/+/250262
Run-TryBot: Michael Pratt <mpratt@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Austin Clements <austin@google.com>
Trust: Michael Pratt <mpratt@google.com>
Currently, on all supported platforms, the race detector (LLVM
TSAN) expects the Go heap is at 0xc000000000 - 0xe000000000.
Move the raceenabled condition first, so we always allocate
there.
This means on Linux/ARM64 when race detector is on we will
allocate to 0xc000000000 - 0xe000000000, instead of 0x4000000000.
The old address is meant for 39-bit VMA. But the race detector
only supports 48-bit VMA anyway. So this is fine.
Change-Id: I51ac8eff68297b37c8c651a93145cc94f83a939d
Reviewed-on: https://go-review.googlesource.com/c/go/+/266372
Trust: Cherry Zhang <cherryyz@google.com>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
This change moves the mcache-local malloc stats into the
consistentHeapStats structure so the malloc stats can be managed
consistently with the memory stats. The one exception here is
tinyAllocs for which moving that into the global stats would incur
several atomic writes on the fast path. Microbenchmarks for just one CPU
core have shown a 50% loss in throughput. Since tiny allocation counnt
isn't exposed anyway and is always blindly added to both allocs and
frees, let that stay inconsistent and flush the tiny allocation count
every so often.
Change-Id: I2a4b75f209c0e659b9c0db081a3287bf227c10ca
Reviewed-on: https://go-review.googlesource.com/c/go/+/247039
Run-TryBot: Michael Knyszek <mknyszek@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Michael Knyszek <mknyszek@google.com>
Reviewed-by: Michael Pratt <mpratt@google.com>
This change adds a function getMCache which returns the current P's
mcache if it's available, and otherwise tries to get mcache0 if we're
bootstrapping. This function will come in handy as we need to replicate
this behavior in multiple places in future changes.
Change-Id: I536073d6f6dc6c6390269e613ead9f8bcb6e7f98
Reviewed-on: https://go-review.googlesource.com/c/go/+/246976
Run-TryBot: Michael Knyszek <mknyszek@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Michael Knyszek <mknyszek@google.com>
Reviewed-by: Michael Pratt <mpratt@google.com>
This change breaks apart gc_sys into three distinct pieces. Two of those
pieces are pieces which come from heap_sys since they're allocated from
the page heap. The rest comes from memory mapped from e.g.
persistentalloc which better fits the purpose of a sysMemStat. Also,
rename gc_sys to gcMiscSys.
Change-Id: I098789170052511e7b31edbcdc9a53e5c24573f7
Reviewed-on: https://go-review.googlesource.com/c/go/+/246973
Run-TryBot: Michael Knyszek <mknyszek@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Michael Knyszek <mknyszek@google.com>
Reviewed-by: Michael Pratt <mpratt@google.com>
This change modifies the type of several mstats fields to be a new type:
sysMemStat. This type has the same structure as the fields used to have.
The purpose of this change is to make it very clear which stats may be
used in various functions for accounting (usually the platform-specific
sys* functions, but there are others). Currently there's an implicit
understanding that the *uint64 value passed to these functions is some
kind of statistic whose value is atomically managed. This understanding
isn't inherently problematic, but we're about to change how some stats
(which currently use mSysStatInc and mSysStatDec) work, so we want to
make it very clear what the various requirements are around "sysStat".
This change also removes mSysStatInc and mSysStatDec in favor of a
method on sysMemStat. Note that those two functions were originally
written the way they were because atomic 64-bit adds required a valid G
on ARM, but this hasn't been the case for a very long time (since
golang.org/cl/14204, but even before then it wasn't clear if mutexes
required a valid G anymore). Today we implement 64-bit adds on ARM with
a spinlock table.
Change-Id: I4e9b37cf14afc2ae20cf736e874eb0064af086d7
Reviewed-on: https://go-review.googlesource.com/c/go/+/246971
Run-TryBot: Michael Knyszek <mknyszek@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Michael Knyszek <mknyszek@google.com>
Reviewed-by: Michael Pratt <mpratt@google.com>
This change renames a bunch of malloc statistics stored in the mcache
that are all named with the "local_" prefix. It also renames largeAlloc
to allocLarge to prevent a naming conflict, and next_sample because it
would be the last mcache field with the old C naming style.
Change-Id: I29695cb83b397a435ede7e9ad5c3c9be72767ea3
Reviewed-on: https://go-review.googlesource.com/c/go/+/246969
Trust: Michael Knyszek <mknyszek@google.com>
Run-TryBot: Michael Knyszek <mknyszek@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Michael Pratt <mpratt@google.com>
This change makes nlargealloc and largealloc into mcache fields just
like nlargefree and largefree. These local fields become the new
source-of-truth. This change also moves the accounting for these fields
out of allocSpan (which is an inappropriate place for it -- this
accounting generally happens much closer to the point of allocation) and
into largeAlloc. This move is partially possible now that we can call
gcController.revise at that point.
Furthermore, this change moves largeAlloc into mcache.go and makes it a
method of mcache. While there's a little bit of a mismatch here because
largeAlloc barely interacts with the mcache, it helps solidify the
mcache as the first allocation layer and provides a clear place to
aggregate and manage statistics.
Change-Id: I37b5e648710733bb4c04430b71e96700e438587a
Reviewed-on: https://go-review.googlesource.com/c/go/+/246965
Trust: Michael Knyszek <mknyszek@google.com>
Run-TryBot: Michael Knyszek <mknyszek@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Michael Pratt <mpratt@google.com>
Setting inittrace=1 causes the runtime to emit a single line to standard error for
each package with init work, summarizing the execution time and memory allocation.
The emitted debug information for init functions can be used to find bottlenecks
or regressions in Go startup performance.
Packages with no init function work (user defined or compiler generated) are omitted.
Tracing plugin inits is not supported as they can execute concurrently. This would
make the implementation of tracing more complex while adding support for a very rare
use case. Plugin inits can be traced separately by testing a main package importing
the plugins package imports explicitly.
$ GODEBUG=inittrace=1 go test
init internal/bytealg @0.008 ms, 0 ms clock, 0 bytes, 0 allocs
init runtime @0.059 ms, 0.026 ms clock, 0 bytes, 0 allocs
init math @0.19 ms, 0.001 ms clock, 0 bytes, 0 allocs
init errors @0.22 ms, 0.004 ms clock, 0 bytes, 0 allocs
init strconv @0.24 ms, 0.002 ms clock, 32 bytes, 2 allocs
init sync @0.28 ms, 0.003 ms clock, 16 bytes, 1 allocs
init unicode @0.44 ms, 0.11 ms clock, 23328 bytes, 24 allocs
...
Inspired by stapelberg@google.com who instrumented doInit
in a prototype to measure init times with GDB.
Fixes#41378
Change-Id: Ic37c6a0cfc95488de9e737f5e346b8dbb39174e1
Reviewed-on: https://go-review.googlesource.com/c/go/+/254659
Trust: Martin Möhrmann <moehrmann@google.com>
Run-TryBot: Martin Möhrmann <moehrmann@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
Apparently macOS/ARM64 has 47-bit addresses, instead of 33-bit as
on ios/ARM64. Enable more address bits.
Updates #38485.
Change-Id: I8aa64ba22a3933e3d9c4fffd17d902b5f31c30e3
Reviewed-on: https://go-review.googlesource.com/c/go/+/256918
Trust: Cherry Zhang <cherryyz@google.com>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Michael Knyszek <mknyszek@google.com>
Currently on 32-bit systems 8-byte fields in a struct have an alignment
of 4 bytes, which means that atomic instructions may fault. This issue
is tracked in #36606.
Our current workaround is to allocate memory and put any such atomically
accessed fields at the beginning of the object. This workaround fails
because the tiny allocator might not align the object right. This case
specifically only happens with 12-byte objects because a type's size is
rounded up to its alignment. So if e.g. we have a type like:
type obj struct {
a uint64
b byte
}
then its size will be 12 bytes, because "a" will require a 4 byte
alignment. This argument may be extended to all objects of size 9-15
bytes.
So, make this workaround work by specifically aligning such objects to 8
bytes on 32-bit systems. This change leaves a TODO to remove the code
once #36606 gets resolved. It also adds a test which will presumably no
longer be necessary (the compiler should enforce the right alignment)
when it gets resolved as well.
Fixes#37262.
Change-Id: I3a34e5b014b3c37ed2e5e75e62d71d8640aa42bc
Reviewed-on: https://go-review.googlesource.com/c/go/+/254057
Reviewed-by: Cherry Zhang <cherryyz@google.com>
Reviewed-by: Austin Clements <austin@google.com>
Run-TryBot: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Michael Knyszek <mknyszek@google.com>
Introduce GOOS=ios for iOS systems. GOOS=ios matches "darwin"
build tag, like GOOS=android matches "linux" and GOOS=illumos
matches "solaris". Only ios/arm64 is supported (ios/amd64 is
not).
GOOS=ios and GOOS=darwin remain essentially the same at this
point. They will diverge at later time, to differentiate macOS
and iOS.
Uses of GOOS=="darwin" are changed to (GOOS=="darwin" || GOOS=="ios"),
except if it clearly means macOS (e.g. GOOS=="darwin" && GOARCH=="amd64"),
it remains GOOS=="darwin".
Updates #38485.
Change-Id: I4faacdc1008f42434599efb3c3ad90763a83b67c
Reviewed-on: https://go-review.googlesource.com/c/go/+/254740
Trust: Cherry Zhang <cherryyz@google.com>
Run-TryBot: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Reviewed-by: Austin Clements <austin@google.com>
This change deletes the old mcentral implementation from the code base
and the newMCentralImpl feature flag along with it.
Updates #37487.
Change-Id: Ibca8f722665f0865051f649ffe699cbdbfdcfcf2
Reviewed-on: https://go-review.googlesource.com/c/go/+/221184
Run-TryBot: Michael Knyszek <mknyszek@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Austin Clements <austin@google.com>
Reviewed-by: Michael Pratt <mpratt@google.com>
This constant does not make it into DWARF because it is an ideal
constant larger than maxint (1<<63-1). DWARF has no way to represent
signed values that large. Define a different typed constant that
is unsigned and so can represent this constant properly.
Viewcore needs this constant to interrogate the heap data structures.
In addition, the sign of arenaBaseOffset changed in 1.15, and providing
a new name lets viewcore detect the sign change easily.
Change-Id: I4274a2f6e79ebbf1411e85d64758fac1672fb96b
Reviewed-on: https://go-review.googlesource.com/c/go/+/240198
Run-TryBot: Keith Randall <khr@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Michael Knyszek <mknyszek@google.com>
The page sweeper depends on spans being marked if any object in the
span is marked, but currently only greyobject does this.
gcmarknewobject and wbBufFlush1 also mark objects, but neither set
span marks. As a result, if there are live objects on a span, but
they're all marked via allocation or write barriers, then the span
itself won't be marked and the page reclaimer will free the span,
ultimately leading to memory corruption when the memory for those live
allocations gets reused.
Fix this by making gcmarknewobject and wbBufFlush1 also mark pages.
No test because I have no idea how to reliably (or even unreliably)
trigger this.
Fixes#39432.
Performance is a wash or very slightly worse. I benchmarked the
gcmarknewobject and wbBufFlush1 changes independently and both showed
a slight performance improvement, so I'm going to call this noise.
name old time/op new time/op delta
BiogoIgor 15.9s ± 2% 15.9s ± 2% ~ (p=0.758 n=25+25)
BiogoKrishna 15.7s ± 3% 15.7s ± 3% ~ (p=0.382 n=21+21)
BleveIndexBatch100 4.94s ± 3% 5.07s ± 4% +2.63% (p=0.000 n=25+25)
CompileTemplate 204ms ± 1% 205ms ± 1% +0.43% (p=0.000 n=21+23)
CompileUnicode 77.8ms ± 1% 78.1ms ± 1% ~ (p=0.130 n=23+23)
CompileGoTypes 731ms ± 1% 733ms ± 1% +0.30% (p=0.006 n=22+22)
CompileCompiler 3.64s ± 2% 3.65s ± 3% ~ (p=0.179 n=24+25)
CompileSSA 8.44s ± 1% 8.46s ± 1% +0.30% (p=0.003 n=22+23)
CompileFlate 132ms ± 1% 133ms ± 1% ~ (p=0.098 n=22+22)
CompileGoParser 164ms ± 1% 164ms ± 1% +0.37% (p=0.000 n=21+23)
CompileReflect 455ms ± 1% 457ms ± 2% +0.50% (p=0.002 n=20+22)
CompileTar 182ms ± 2% 182ms ± 1% ~ (p=0.382 n=22+22)
CompileXML 245ms ± 3% 245ms ± 1% ~ (p=0.070 n=21+23)
CompileStdCmd 16.5s ± 2% 16.5s ± 3% ~ (p=0.486 n=23+23)
FoglemanFauxGLRenderRotateBoat 12.9s ± 1% 13.0s ± 1% +0.97% (p=0.000 n=21+24)
FoglemanPathTraceRenderGopherIter1 18.6s ± 1% 18.7s ± 0% ~ (p=0.083 n=23+24)
GopherLuaKNucleotide 28.4s ± 1% 29.3s ± 1% +2.84% (p=0.000 n=25+25)
MarkdownRenderXHTML 252ms ± 0% 251ms ± 1% -0.50% (p=0.000 n=23+24)
Tile38WithinCircle100kmRequest 516µs ± 2% 516µs ± 2% ~ (p=0.763 n=24+25)
Tile38IntersectsCircle100kmRequest 689µs ± 2% 689µs ± 2% ~ (p=0.617 n=24+24)
Tile38KNearestLimit100Request 608µs ± 1% 606µs ± 2% -0.35% (p=0.030 n=19+22)
[Geo mean] 522ms 524ms +0.41%
https://perf.golang.org/search?q=upload:20200606.4
Change-Id: I8b331f310dbfaba0468035f207467c8403005bf5
Reviewed-on: https://go-review.googlesource.com/c/go/+/236817
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Michael Knyszek <mknyszek@google.com>
Currently maxOffAddr is defined in terms of the whole 64-bit address
space, assuming that it's all supported, by using ^uintptr(0) as the
maximal address in the offset space. In reality, the maximal address in
the offset space is (1<<heapAddrBits)-1 because we don't have more than
that actually available to us on a given platform.
On most platforms this is fine, because arenaBaseOffset is just
connecting two segments of address space, but on AIX we use it as an
actual offset for the starting address of the available address space,
which is limited. This means using ^uintptr(0) as the maximal address in
the offset address space causes wrap-around, especially when we just
want to represent a range approximately like [addr, infinity), which
today we do by using maxOffAddr.
To fix this, we define maxOffAddr more appropriately, in terms of
(1<<heapAddrBits)-1.
This change also redefines arenaBaseOffset to not be the negation of the
virtual address corresponding to address zero in the virtual address
space, but instead directly as the virtual address corresponding to
zero. This matches the existing documentation more closely and makes the
logic around arenaBaseOffset decidedly simpler, especially when trying
to reason about its use on AIX.
Fixes#38966.
Change-Id: I1336e5036a39de846f64cc2d253e8536dee57611
Reviewed-on: https://go-review.googlesource.com/c/go/+/233497
Run-TryBot: Michael Knyszek <mknyszek@google.com>
Reviewed-by: Austin Clements <austin@google.com>
Reviewed-by: Michael Pratt <mpratt@google.com>
Currently linearAlloc manages an exclusive "end" address for the top of
its reserved space. While unlikely for a linearAlloc to be allocated
with an "end" address hitting the top of the address space, it is
possible and could lead to overflow.
Avoid overflow by chopping off the last byte from the linearAlloc if
it's bumping up against the top of the address space defensively. In
practice, this means that if 32-bit platforms map the top of the address
space and use the linearAlloc to acquire arenas, the top arena will not
be usable.
Fixes#35954.
Change-Id: I512cddcd34fd1ab15cb6ca92bbf899fc1ef22ff6
Reviewed-on: https://go-review.googlesource.com/c/go/+/231338
Run-TryBot: Michael Knyszek <mknyszek@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Austin Clements <austin@google.com>
Currently mcentral is implemented as a couple of linked lists of spans
protected by a lock. Unfortunately this design leads to significant lock
contention.
The span ownership model is also confusing and complicated. In-use spans
jump between being owned by multiple sources, generally some combination
of a gcSweepBuf, a concurrent sweeper, an mcentral or an mcache.
So first to address contention, this change replaces those linked lists
with gcSweepBufs which have an atomic fast path. Then, we change up the
ownership model: a span may be simultaneously owned only by an mcentral
and the page reclaimer. Otherwise, an mcentral (which now consists of
sweep bufs), a sweeper, or an mcache are the sole owners of a span at
any given time. This dramatically simplifies reasoning about span
ownership in the runtime.
As a result of this new ownership model, sweeping is now driven by
walking over the mcentrals rather than having its own global list of
spans. Because we no longer have a global list and we traditionally
haven't used the mcentrals for large object spans, we no longer have
anywhere to put large objects. So, this change also makes it so that we
keep large object spans in the appropriate mcentral lists.
In terms of the static lock ranking, we add the spanSet spine locks in
pretty much the same place as the mcentral locks, since they have the
potential to be manipulated both on the allocation and sweep paths, like
the mcentral locks.
This new implementation is turned on by default via a feature flag
called go115NewMCentralImpl.
Benchmark results for 1 KiB allocation throughput (5 runs each):
name \ MiB/s go113 go114 gotip gotip+this-patch
AllocKiB-1 1.71k ± 1% 1.68k ± 1% 1.59k ± 2% 1.71k ± 1%
AllocKiB-2 2.46k ± 1% 2.51k ± 1% 2.54k ± 1% 2.93k ± 1%
AllocKiB-4 4.27k ± 1% 4.41k ± 2% 4.33k ± 1% 5.01k ± 2%
AllocKiB-8 4.38k ± 3% 5.24k ± 1% 5.46k ± 1% 8.23k ± 1%
AllocKiB-12 4.38k ± 3% 4.49k ± 1% 5.10k ± 1% 10.04k ± 0%
AllocKiB-16 4.31k ± 1% 4.14k ± 3% 4.22k ± 0% 10.42k ± 0%
AllocKiB-20 4.26k ± 1% 3.98k ± 1% 4.09k ± 1% 10.46k ± 3%
AllocKiB-24 4.20k ± 1% 3.97k ± 1% 4.06k ± 1% 10.74k ± 1%
AllocKiB-28 4.15k ± 0% 4.00k ± 0% 4.20k ± 0% 10.76k ± 1%
Fixes#37487.
Change-Id: I92d47355acacf9af2c41bf080c08a8c1638ba210
Reviewed-on: https://go-review.googlesource.com/c/go/+/221182
Run-TryBot: Michael Knyszek <mknyszek@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Austin Clements <austin@google.com>
Currently markrootSpans, the scanning routine which scans span specials
(particularly finalizers) as roots, uses sweepSpans to shard work and
find spans to mark.
However, as part of a future CL to change span ownership and how
mcentral works, we want to avoid having markrootSpans use the sweep bufs
to find specials, so in this change we introduce a new mechanism.
Much like for the page reclaimer, we set up a per-page bitmap where the
first page for a span is marked if the span contains any specials, and
unmarked if it has no specials. This bitmap is updated by addspecial,
removespecial, and during sweeping.
markrootSpans then shards this bitmap into mark work and markers iterate
over the bitmap looking for spans with specials to mark. Unlike the page
reclaimer, we don't need to use the pageInUse bits because having a
special implies that a span is in-use.
While in terms of computational complexity this design is technically
worse, because it needs to iterate over the mapped heap, in practice
this iteration is very fast (we can skip over large swathes of the heap
very quickly) and we only look at spans that have any specials at all,
rather than having to touch each span.
This new implementation of markrootSpans is behind a feature flag called
go115NewMarkrootSpans.
Updates #37487.
Change-Id: I8ea07b6c11059f6d412fe419e0ab512d989377b8
Reviewed-on: https://go-review.googlesource.com/c/go/+/221178
Run-TryBot: Michael Knyszek <mknyszek@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Austin Clements <austin@google.com>
A case that I missed in CL 205239: profilealloc can be called at
program startup if GOMAXPROCS is large enough.
Fixes#38474
Change-Id: I2f089fc6ec00c376680e1c0b8a2557b62789dd7f
Reviewed-on: https://go-review.googlesource.com/c/go/+/228420
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Michael Pratt <mpratt@google.com>
I took some of the infrastructure from Austin's lock logging CR
https://go-review.googlesource.com/c/go/+/192704 (with deadlock
detection from the logs), and developed a setup to give static lock
ranking for runtime locks.
Static lock ranking establishes a documented total ordering among locks,
and then reports an error if the total order is violated. This can
happen if a deadlock happens (by acquiring a sequence of locks in
different orders), or if just one side of a possible deadlock happens.
Lock ordering deadlocks cannot happen as long as the lock ordering is
followed.
Along the way, I found a deadlock involving the new timer code, which Ian fixed
via https://go-review.googlesource.com/c/go/+/207348, as well as two other
potential deadlocks.
See the constants at the top of runtime/lockrank.go to show the static
lock ranking that I ended up with, along with some comments. This is
great documentation of the current intended lock ordering when acquiring
multiple locks in the runtime.
I also added an array lockPartialOrder[] which shows and enforces the
current partial ordering among locks (which is embedded within the total
ordering). This is more specific about the dependencies among locks.
I don't try to check the ranking within a lock class with multiple locks
that can be acquired at the same time (i.e. check the ranking when
multiple hchan locks are acquired).
Currently, I am doing a lockInit() call to set the lock rank of most
locks. Any lock that is not otherwise initialized is assumed to be a
leaf lock (a very high rank lock), so that eliminates the need to do
anything for a bunch of locks (including all architecture-dependent
locks). For two locks, root.lock and notifyList.lock (only in the
runtime/sema.go file), it is not as easy to do lock initialization, so
instead, I am passing the lock rank with the lock calls.
For Windows compilation, I needed to increase the StackGuard size from
896 to 928 because of the new lock-rank checking functions.
Checking of the static lock ranking is enabled by setting
GOEXPERIMENT=staticlockranking before doing a run.
To make sure that the static lock ranking code has no overhead in memory
or CPU when not enabled by GOEXPERIMENT, I changed 'go build/install' so
that it defines a build tag (with the same name) whenever any experiment
has been baked into the toolchain (by checking Expstring()). This allows
me to avoid increasing the size of the 'mutex' type when static lock
ranking is not enabled.
Fixes#38029
Change-Id: I154217ff307c47051f8dae9c2a03b53081acd83a
Reviewed-on: https://go-review.googlesource.com/c/go/+/207619
Reviewed-by: Dan Scales <danscales@google.com>
Reviewed-by: Keith Randall <khr@golang.org>
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
There are a handful of places where the runtime wants to round up the
result of a division. We just introduced a helper to do this. This CL
replaces all of the hand-coded round-ups (that I could find) with this
helper.
Change-Id: I465d152157ff0f3cad40c0aa57491e4f2de510ad
Reviewed-on: https://go-review.googlesource.com/c/go/+/224385
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
Having an mcache field in both m and p is confusing, so remove it from m.
Always use mcache field from p. Use new variable mcache0 during bootstrap.
Change-Id: If2cba9f8bb131d911d512b61fd883a86cf62cc98
Reviewed-on: https://go-review.googlesource.com/c/go/+/205239
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Austin Clements <austin@google.com>
Rhys Hiltner
Michael Anthony Knyszek
Russ Cox
fanzha02
Meng Zhuo
Leonard Wang
Matthew Dempsky
David Chase
Cherry Mui
tyltr
Keith Randall
Lizzzcai
Brad Fitzpatrick
Michael Pratt
Martin Möhrmann
Austin Clements
Ian Lance Taylor
Dan Scales