MemoryChunkLayout::MaxRegularCodeObjectSize() can be cached in a
global variable on process initialization. This should help to increase
code object allocation performance, since this method was called on
each code object allocation.
Bug: v8:11891
Change-Id: I870bd37202370aec89ef2db24264e363099bf8a0
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2966387
Commit-Queue: Dominik Inführ <dinfuehr@chromium.org>
Reviewed-by: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Cr-Commit-Position: refs/heads/master@{#75215}
Executable V8 pages include 3 reserved OS pages: one for the writable
header and two as guards. On systems with 64k OS pages, the amount of
allocatable space left for objects can then be quite smaller than the
page size, only 64k for each 256k page.
This means regular code objects cannot be larger than 64k, while the
maximum regular object size is fixed to 128k, half of the page size. As
a result code object never reach this limit and we can end up filling
regular pages with few large code objects.
To fix this, we change the maximum code object size to be runtime value,
set to half of the allocatable space per page. On systems with 64k OS
pages, the limit will be 32k.
Alternatively, we could increase the V8 page size to 512k on Arm64 linux
so we wouldn't waste code space. However, systems with 4k OS pages are
more common, and those with 64k pages tend to have more memory available
so we should be able to live with it.
Bug: v8:10808
Change-Id: I5d807e7a3df89f1e9c648899e9ba2f8e2648264c
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2460809
Reviewed-by: Igor Sheludko <ishell@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Pierre Langlois <pierre.langlois@arm.com>
Cr-Commit-Position: refs/heads/master@{#70569}
With the new Turbofan variants (NCI and Turboprop), we need a way to
distinguish between them both during and after compilation. We
initially introduced CompilationTarget to track the variant during
compilation, but decided to reuse the code kind as the canonical spot to
store this information instead.
Why? Because it is an established mechanism, already available in most
of the necessary spots (inside the pipeline, on Code objects, in
profiling traces).
This CL removes CompilationTarget and adds a new
NATIVE_CONTEXT_INDEPENDENT kind, plus helper functions to determine
various things about a given code kind (e.g.: does this code kind
deopt?).
As a (very large) drive-by, refactor both Code::Kind and
AbstractCode::Kind into a new CodeKind enum class.
Bug: v8:8888
Change-Id: Ie858b9a53311b0731630be35cf5cd108dee95b39
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2336793
Commit-Queue: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Clemens Backes <clemensb@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Dominik Inführ <dinfuehr@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69244}
Jitless mode does not allocate executable memory, so we can avoid reserving
memory in such configurations.
Bug: v8:10033
Change-Id: Ie6a943084e3bade85848e3219cb4d8779ed34830
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1981505
Commit-Queue: Michael Lippautz <mlippautz@chromium.org>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Cr-Commit-Position: refs/heads/master@{#66388}
This increases the kInstructionSize to give the nosnap Isolate
more space to initialize itself.
This is a reland of 2980961182
Original change's description:
> [heap] Add tests to CodeBuilder Build and TryBuild
>
> Bug: chromium:957934
> Change-Id: Ia1ce5f6150b7f862b38f4c135236b154a1b02d6c
> Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1594564
> Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> Commit-Queue: Maciej Goszczycki <goszczycki@google.com>
> Cr-Commit-Position: refs/heads/master@{#61391}
Bug: chromium:957934
Change-Id: Ib7458f7c18063dbc8bc685ddbb9f22f781c52aa5
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1609477
Commit-Queue: Maciej Goszczycki <goszczycki@google.com>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Dan Elphick <delphick@chromium.org>
Cr-Commit-Position: refs/heads/master@{#61454}
CodeBuilder was calling AllocateRawWithLightRetry when it should have been
calling AllocateRawWithRetryOrFail (and vice versa).
Also improved variable naming.
Bug: chromium:957934
Change-Id: I03a95165f6d5b44c1f47d08d338d48bcc37c6d04
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1590075
Commit-Queue: Maciej Goszczycki <goszczycki@google.com>
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: Dan Elphick <delphick@chromium.org>
Cr-Commit-Position: refs/heads/master@{#61210}
Switch all uses of NewCode and TryNewCode to CodeBuilder and remove these
methods.
NewCode and TryNewCode use a large number of default parameters, which makes
it difficult to use and add any new ones. Large chunks of code were also
duplicated across TryNewCode and NewCode. The previous CL
(https://chromium-review.googlesource.com/c/v8/v8/+/1585736) added a new
CodeBuilder class which allows much simpler building of Code objects.
Bug: v8:9183
Change-Id: I9f6884f35a3284cbd40746376f0f27e36f9051b5
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1585737
Commit-Queue: Maciej Goszczycki <goszczycki@google.com>
Reviewed-by: Jakob Gruber <jgruber@chromium.org>
Reviewed-by: Hannes Payer <hpayer@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Reviewed-by: Dan Elphick <delphick@chromium.org>
Cr-Commit-Position: refs/heads/master@{#61096}
If the size of a large code object is larger than
kMaxRegularHeapObjectSize, then it should be allocated in the large code
space. Currently if the size is > kMaxRegularHeapObjectSize but <
512000, then it can still be allocated in the normal code space.
Change-Id: I72dbd38803c3d5d414bae85e9e0b15482e50e1c2
Reviewed-on: https://chromium-review.googlesource.com/c/1363137
Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
Commit-Queue: Peter Marshall <petermarshall@chromium.org>
Cr-Commit-Position: refs/heads/master@{#58046}
