Our LoadIndexedInstr/StoreIndexedInstr can operate on an untagged array,
for which we assume it can alias with anything (despite constant
index).
Optimized dart:ffi code will create LoadIndexedInstr/StoreIndexedInstr
that operate on IntConverterInstr. Those can be views into arbitrary
memory so we should assume it can alias with anything (despite constant
index).
Right now it happens to be not a problem (and we cannot have a Dart test
to trigger it), because every LoadIndexedInstr/StoreIndexedInstr that
dart:ffi code generates happens to use a 0 index: It will load the
address of the pointer, adds the scaled index to it and then converts
the result of this arithmetic to an untagged pointer, used by the
load/store with index 0.
Change-Id: I94268486603b78e4d21c3dafaccd133da35193f3
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/121843
Commit-Queue: Martin Kustermann <kustermann@google.com>
Reviewed-by: Vyacheslav Egorov <vegorov@google.com>
Fixes an assumption that CompressedStackMapsIterator::Find() is never
passed a PC offset of 0. Adds back an ASSERT that was dropped which checks
for specific cases where a given PC offset does not have a stack map entry.
Original commit message:
Lifting the PC offset in a2bb730 was a small, lightweight change that
gave us big gains, at least on 32-bit architectures. Here, we make
much more invasive changes that will improve the amount of memory used
by the information previously stored in StackMap objects.
Instead of allocating separate objects for StackMaps, we instead compress
all StackMap information for a given Code object into a single object
(CompressedStackMaps, or CSM for short). This replaces the Array used to
store PC offsets (as Smis) and the individual StackMap objects.
While we lose all canonicalization for individual StackMap entries, the
drop in space required to store stack map information more than offsets that.
-----
The impact on AOT snapshot size when compiling the Flutter Gallery
in release mode:
armv7: Total size -2.58% (Isolate RO: +14.46%, Isolate snapshot: -22.93%)
armv8: Total size -1.85% (Isolate RO: +15.69%, Isolate snapshot: -22.97%)
The impact on in-memory, not on-disk, size for the Flutter Gallery as seen
in the Observatory while running a profile (not release) build:
armv7: Drops from 7.1 MB to 6.2MB (-0.9 MB)
armv8: Drops from 13.5MB to 11.7MB (-1.8 MB)
-----
Bug: https://github.com/dart-lang/sdk/issues/35274, https://github.com/dart-lang/sdk/issues/38873
Cq-Include-Trybots: luci.dart.try:vm-kernel-precomp-linux-debug-x64-try,vm-kernel-precomp-linux-debug-simarm_x64-try,vm-kernel-mac-debug-simdbc64-try
Change-Id: I111b129b0ed64f03184370bceb7cda69d5d4b3c9
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/121700
Commit-Queue: Teagan Strickland <sstrickl@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
Removes deprecated `load`, `store`, `allocate`, `free`, and `asExternalTypedData` methods from Pointer because these have been replace by extension methods and `package:ffi`.
Closes: https://github.com/dart-lang/sdk/issues/38860
Change-Id: I76d21bcfd2fdd6ee504a51db821345125beb92d2
Cq-Include-Trybots: luci.dart.try:vm-ffi-android-debug-arm-try,vm-ffi-android-debug-arm64-try,app-kernel-linux-debug-x64-try,vm-kernel-linux-debug-ia32-try,vm-dartkb-linux-debug-simarm64-try,vm-kernel-win-debug-x64-try,vm-kernel-win-debug-ia32-try,vm-dartkb-linux-debug-x64-try,vm-kernel-precomp-linux-debug-x64-try,vm-dartkb-linux-release-x64-abi-try,vm-kernel-precomp-android-release-arm64-try,vm-kernel-asan-linux-release-x64-try,vm-kernel-linux-release-simarm-try,vm-kernel-linux-release-simarm64-try,vm-kernel-precomp-android-release-arm_x64-try,vm-kernel-reload-mac-release-simdbc64-try,vm-kernel-precomp-obfuscate-linux-release-x64-try,vm-kernel-reload-rollback-linux-debug-x64-try,vm-kernel-precomp-mac-release-simarm_x64-try,dart-sdk-linux-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/121708
Commit-Queue: Daco Harkes <dacoharkes@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
Support unaligned memory accesses for regex compiler for arm64
target. Also fix a bug in LoadCodeUnitsInstr IL visitor where
Address::OperandSizeFor was used to get access size (which
returned size of one element independently of actual element
count).
Change-Id: I77c449c8597f8cfd500b8d701f72c39c46a443b7
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/121424
Commit-Queue: Vyacheslav Egorov <vegorov@google.com>
Reviewed-by: Vyacheslav Egorov <vegorov@google.com>
Speeds up the FfiMemory benchmarks by a factor of 3 in AoT.
Issue: https://github.com/dart-lang/sdk/issues/38172
Change-Id: Ic6d920e27ff905a7e08c47f4afb229d8620068ec
Cq-Include-Trybots: luci.dart.try:vm-ffi-android-debug-arm-try,vm-ffi-android-debug-arm64-try,app-kernel-linux-debug-x64-try,vm-kernel-linux-debug-ia32-try,vm-dartkb-linux-debug-simarm64-try,vm-kernel-win-debug-x64-try,vm-kernel-win-debug-ia32-try,vm-dartkb-linux-debug-x64-try,vm-kernel-precomp-linux-debug-x64-try,vm-dartkb-linux-release-x64-abi-try,vm-kernel-precomp-android-release-arm64-try,vm-kernel-asan-linux-release-x64-try,vm-kernel-linux-release-simarm-try,vm-kernel-linux-release-simarm64-try,vm-kernel-precomp-android-release-arm_x64-try,vm-kernel-reload-mac-release-simdbc64-try,vm-kernel-precomp-obfuscate-linux-release-x64-try,vm-kernel-reload-rollback-linux-debug-x64-try,vm-kernel-precomp-mac-release-simarm_x64-try,dart-sdk-linux-try,flutter-engine-linux-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/121580
Commit-Queue: Daco Harkes <dacoharkes@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
This reverts commit d77f4913a1.
Reason for revert: large number of flaky crashes on flakiness dashboard (https://github.com/dart-lang/sdk/issues/38873); crashes on vm-kernel-mac-debug-simdbc64 bot.
Original change's description:
> [vm/compiler] Further compress the information previously in StackMaps.
>
> Lifting the PC offset in a2bb730 was a small, lightweight change that
> gave us big gains, at least on 32-bit architectures. Here, we make
> much more invasive changes that will improve the amount of memory used
> by the information previously stored in StackMap objects.
>
> Instead of allocating separate objects for StackMaps, we instead compress
> all StackMap information for a given Code object into a single object
> (CompressedStackMaps, or CSM for short). This replaces the Array used to
> store PC offsets (as Smis) and the individual StackMap objects.
>
> While we lose all canonicalization for individual StackMap entries, the
> drop in space required to store stack map information more than offsets that.
>
> -----
>
> The impact on AOT snapshot size when compiling the Flutter Gallery
> in release mode:
>
> armv7: Total size -2.58% (Isolate RO: +14.46%, Isolate snapshot: -22.93%)
> armv8: Total size -1.85% (Isolate RO: +15.69%, Isolate snapshot: -22.97%)
>
> The impact on in-memory, not on-disk, size for the Flutter Gallery as seen
> in the Observatory while running a profile (not release) build:
>
> armv7: Drops from 7.1 MB to 6.2MB (-0.9 MB)
> armv8: Drops from 13.5MB to 11.7MB (-1.8 MB)
>
> -----
>
> Bug: https://github.com/dart-lang/sdk/issues/35274
> Cq-Include-Trybots: luci.dart.try:vm-kernel-precomp-linux-debug-x64-try,vm-kernel-precomp-linux-debug-simarm_x64-try
> Change-Id: Ie3bb898d557215146260a560423f5fa27bdff512
> Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/119640
> Commit-Queue: Teagan Strickland <sstrickl@google.com>
> Reviewed-by: Martin Kustermann <kustermann@google.com>
> Reviewed-by: Ryan Macnak <rmacnak@google.com>
TBR=kustermann@google.com,rmacnak@google.com,sstrickl@google.com
# Not skipping CQ checks because original CL landed > 1 day ago.
Bug: https://github.com/dart-lang/sdk/issues/35274
Change-Id: Ib844b4d52e3aa4aca62a896cfe349c680bda7461
Cq-Include-Trybots: luci.dart.try:vm-kernel-precomp-linux-debug-x64-try, vm-kernel-precomp-linux-debug-simarm_x64-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/121642
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Commit-Queue: Alexander Markov <alexmarkov@google.com>
In order to support obfuscation, string constants are decoupled
from public names in bytecode (names are obfuscated, but string
constants aren't). List of protected names is written in a separate
section in bytecode component.
Obfuscator is extended to support getter and setter
names coming from bytecode.
Change-Id: I8e8d820d1a8b97e32e8ad1b064b827bdb017430b
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/121261
Commit-Queue: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Régis Crelier <regis@google.com>
Reviewed-by: Ryan Macnak <rmacnak@google.com>
tcmalloc and jemalloc have both been observed to hold onto lots of free'd zone segments (jemalloc to the point of causing OOM), so instead of using malloc to allocate segments, we allocate directly from mmap/zx_vmo_create/VirtualAlloc, and cache a small number of the normal sized segments.
flutter doctor at exit ("idle")
Process::MaxRSS() 261009408 -> 258756608 (-0.86%)
Process::CurrentRSS() 252932096 -> 205881344 (-18.6%)
Bug: https://github.com/dart-lang/sdk/issues/38820
Change-Id: I319069678b9d44383a9921ec03a963517991dd11
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/121264
Commit-Queue: Ryan Macnak <rmacnak@google.com>
Reviewed-by: Siva Annamalai <asiva@google.com>
Reviewed-by: Zach Anderson <zra@google.com>
Follow up of https://dart-review.googlesource.com/c/sdk/+/117547
This gets rid of unnecessary allocations in hot loops with indexed loads and stores.
Issue: https://github.com/dart-lang/sdk/issues/38172
Change-Id: I37a4b1aba00084e465d47cce79bb9963e1afc104
Cq-Include-Trybots: luci.dart.try:vm-ffi-android-debug-arm-try,vm-ffi-android-debug-arm64-try,app-kernel-linux-debug-x64-try,vm-kernel-linux-debug-ia32-try,vm-dartkb-linux-debug-simarm64-try,vm-kernel-win-debug-x64-try,vm-kernel-win-debug-ia32-try,vm-dartkb-linux-debug-x64-try,vm-kernel-precomp-linux-debug-x64-try,vm-dartkb-linux-release-x64-abi-try,vm-kernel-precomp-android-release-arm64-try,vm-kernel-asan-linux-release-x64-try,vm-kernel-linux-release-simarm-try,vm-kernel-linux-release-simarm64-try,vm-kernel-mac-debug-simdbc64-try,vm-kernel-precomp-android-release-arm_x64-try,vm-kernel-reload-mac-release-simdbc64-try,vm-kernel-precomp-obfuscate-linux-release-x64-try,vm-kernel-reload-rollback-linux-debug-x64-try,vm-kernel-precomp-mac-release-simarm_x64-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/119645
Reviewed-by: Martin Kustermann <kustermann@google.com>
Lifting the PC offset in a2bb730 was a small, lightweight change that
gave us big gains, at least on 32-bit architectures. Here, we make
much more invasive changes that will improve the amount of memory used
by the information previously stored in StackMap objects.
Instead of allocating separate objects for StackMaps, we instead compress
all StackMap information for a given Code object into a single object
(CompressedStackMaps, or CSM for short). This replaces the Array used to
store PC offsets (as Smis) and the individual StackMap objects.
While we lose all canonicalization for individual StackMap entries, the
drop in space required to store stack map information more than offsets that.
-----
The impact on AOT snapshot size when compiling the Flutter Gallery
in release mode:
armv7: Total size -2.58% (Isolate RO: +14.46%, Isolate snapshot: -22.93%)
armv8: Total size -1.85% (Isolate RO: +15.69%, Isolate snapshot: -22.97%)
The impact on in-memory, not on-disk, size for the Flutter Gallery as seen
in the Observatory while running a profile (not release) build:
armv7: Drops from 7.1 MB to 6.2MB (-0.9 MB)
armv8: Drops from 13.5MB to 11.7MB (-1.8 MB)
-----
Bug: https://github.com/dart-lang/sdk/issues/35274
Cq-Include-Trybots: luci.dart.try:vm-kernel-precomp-linux-debug-x64-try,vm-kernel-precomp-linux-debug-simarm_x64-try
Change-Id: Ie3bb898d557215146260a560423f5fa27bdff512
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/119640
Commit-Queue: Teagan Strickland <sstrickl@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
Reviewed-by: Ryan Macnak <rmacnak@google.com>
When creating a place for an indirect indexed access into typed data
list (one which uses LoadUntagged to load base pointer), don't consider
this place to be constant indexed even if index itself is a constant.
Base pointer returned by LoadUntagged can be pointing into some other
typed array - meaning that the actual index is C plus an unknown
offset.
Fixes https://github.com/dart-lang/sdk/issues/38467
Change-Id: I1fb50c2d265537d2b6dd185c10ddc74dc06147b2
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/121123
Commit-Queue: Vyacheslav Egorov <vegorov@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
Functions which have optional parameters or which are generic need the
arguments descriptor to be passed. Since switchable call-sites do not
populate the arguments descriptor, such functions are always called via
a stub (i.e. the switchable calls transition never to monomorphic
state).
There is therefore no reason to generate the monomorphic prologues for
those functions.
Flutter Gallery size impact:
armv8: -0.31% RX
armv7: -0.226% RX
Change-Id: I7d6b554cbcdc90b85a278487c5541fed4e64b5bc
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/120961
Reviewed-by: Ryan Macnak <rmacnak@google.com>
Commit-Queue: Martin Kustermann <kustermann@google.com>
In order to support causal-async-stacks, the following synthetic calls
are inserted: _asyncStackTraceHelper, _setAsyncThreadStackTrace and
_clearAsyncThreadStackTrace.
_asyncStackTraceHelper is always generated in the async transformation.
_setAsyncThreadStackTrace and _clearAsyncThreadStackTrace are inserted
during bytecode generation (or in the flow graph builder in case of AST
pipeline). If bytecode generation options are set inconsistently with
--causal-async-stacks VM option, then it was possible that
_asyncStackTraceHelper was generated but other calls were not generated,
causing incorrect (full, non-truncated) async stack traces. This also
causes performance regression as collecting full stack traces takes
much more time.
This change makes generation of all these calls more consistent: if
causal async stacks are disabled during bytecode generation, then all
three calls are omitted. Also, in case these calls were generated and
present in bytecode, but VM option --causal-async-stacks is disabled,
then bytecode flow graph builder turns all these three calls into no-ops.
Change-Id: I93eb1d83c675ee093799bb8e37ca3d60a3c5c19d
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/120927
Commit-Queue: Alexander Markov <alexmarkov@google.com>
Reviewed-by: Ryan Macnak <rmacnak@google.com>
Reviewed-by: Régis Crelier <regis@google.com>
While emitting code for a ParallelMove on ARM32, the TMP register was
being used as a dest/src. When the stack slot offset is large,
StoreToOffset creates a new base in TMP in order to use a smaller
offset. Since the ParallelMove implementation was hardcoding TMP as the
dest/src register, the assert that the register != TMP failed with large
offsets. Now we instead perform a check to see if the offset will fit to
determine whether or not we can use TMP or if we need to fall back and
use LR as our temporary (we can do this since we're in the middle of a
ParallelMove).
Fixes https://github.com/dart-lang/sdk/issues/38654
Change-Id: I4d8cc8b54f6090d86992aabb8ff44d088bc02d3e
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/119584
Reviewed-by: Ryan Macnak <rmacnak@google.com>
Commit-Queue: Ben Konyi <bkonyi@google.com>
Original CL in patchset 1.
Fix for simdbc in patchset 4.
Fix for arm32 precompiled in: https://dart-review.googlesource.com/c/sdk/+/120660/
This CL optimizes Pointer operations in hot loops for Pointer<NativeInteger/NativeDouble/Pointer> (not for structs).
Design: go/dart-ffi-pointers-il
It provides roughly a 100x speedup for the FfiMemory benchmark. The next 5x speedup is to get rid of allocations due to `load` and `store` not being inlined.
FFI API is changed to enable optimizations:
* Disable dynamic invocations of Pointer.load / Pointer.store.
* Disallow implicit downcast of argument passed to Pointer.store.
* Stop zeroing out Pointer.address on Pointer.free().
Issue: https://github.com/dart-lang/sdk/issues/38172
Related issues:
Closes: https://github.com/dart-lang/sdk/issues/35902 (Disallowing dynamic invocations of Pointer ops.)
Closes: https://github.com/dart-lang/sdk/issues/37385 (Function variance checking)
Change-Id: I3921a595fd05026d6ca565ace496771d7c1d877b
Cq-Include-Trybots: luci.dart.try:vm-ffi-android-debug-arm-try,vm-ffi-android-debug-arm64-try,app-kernel-linux-debug-x64-try,vm-kernel-linux-debug-ia32-try,vm-dartkb-linux-debug-simarm64-try,vm-kernel-win-debug-x64-try,vm-kernel-win-debug-ia32-try,vm-dartkb-linux-debug-x64-try,vm-kernel-precomp-linux-debug-x64-try,vm-dartkb-linux-release-x64-abi-try,vm-kernel-precomp-android-release-arm64-try,vm-kernel-asan-linux-release-x64-try,vm-kernel-linux-release-simarm-try,vm-kernel-linux-release-simarm64-try,vm-kernel-mac-debug-simdbc64-try,vm-kernel-precomp-android-release-arm_x64-try,vm-kernel-reload-mac-release-simdbc64-try,vm-kernel-precomp-obfuscate-linux-release-x64-try,vm-kernel-reload-rollback-linux-debug-x64-try,vm-kernel-precomp-mac-release-simarm_x64-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/120661
Reviewed-by: Martin Kustermann <kustermann@google.com>
Commit-Queue: Daco Harkes <dacoharkes@google.com>
Consume the extension flag setting for functions and the extension flag and is late
flags for fields when reading kernel file and set bit in Function/Field objects.
Change-Id: I661d30d53e817d968151d2b95474a6daf918de13
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/118000
Reviewed-by: Alexander Markov <alexmarkov@google.com>
Commit-Queue: Siva Annamalai <asiva@google.com>
This CL optimizes Pointer operations in hot loops for Pointer<NativeInteger/NativeDouble/Pointer> (not for structs).
Design: go/dart-ffi-pointers-il
It provides roughly a 100x speedup for the FfiMemory benchmark. The next 5x speedup is to get rid of allocations due to `load` and `store` not being inlined.
FFI API is changed to enable optimizations:
* Disable dynamic invocations of Pointer.load / Pointer.store.
* Disallow implicit downcast of argument passed to Pointer.store.
* Stop zeroing out Pointer.address on Pointer.free().
Issue: https://github.com/dart-lang/sdk/issues/38172
Related issues:
Closes: https://github.com/dart-lang/sdk/issues/35902 (Disallowing dynamic invocations of Pointer ops.)
Closes: https://github.com/dart-lang/sdk/issues/37385 (Function variance checking.)
Change-Id: I96058d8b5b49052eb6999f084372e6f08b4f6f17
Cq-Include-Trybots: luci.dart.try:vm-ffi-android-debug-arm-try,vm-ffi-android-debug-arm64-try,app-kernel-linux-debug-x64-try,vm-kernel-linux-debug-ia32-try,vm-dartkb-linux-debug-simarm64-try,vm-kernel-win-debug-x64-try,vm-kernel-win-debug-ia32-try,vm-dartkb-linux-debug-x64-try,vm-kernel-precomp-linux-debug-x64-try,vm-dartkb-linux-release-x64-abi-try,vm-kernel-precomp-android-release-arm64-try
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/117547
Commit-Queue: Daco Harkes <dacoharkes@google.com>
Reviewed-by: Martin Kustermann <kustermann@google.com>
Allocate space for ForwardingPages when allocating a new HeapPages that is subject to compaction, instead of allocating it at compaction time, which is when we're already low on memory.
This has a 4KB/256KB = 1.5% overhead for ordinary pages, no overhead for large and code pages.
Bug: https://github.com/dart-lang/sdk/issues/37772
Change-Id: I9ce4d0791093df8ee98d9bda17f0a1cc33788e28
Reviewed-on: https://dart-review.googlesource.com/c/sdk/+/114020
Commit-Queue: Ryan Macnak <rmacnak@google.com>
Reviewed-by: Siva Annamalai <asiva@google.com>
Vyacheslav Egorov