This flag enables feedback allocation heuristics to be based on the
function size. The threshold for feedback allocation is set to
12 * bytecode size to roughly mimic the allocation after 12 invocations.
We tried 4 * bytecode size earlier and there were few memory regressions
on real world pages. Speedometer improved by ~1% [1, 2]. This time
trying with a higher factor this time to see if we still have
speedometer improvements without any memory regressions.
[1] https://bugs.chromium.org/p/chromium/issues/detail?id=1177124
[2] https://bugs.chromium.org/p/chromium/issues/detail?id=1177241
Change-Id: I39c7d4aaf90b948b07419e4598e2193b8355c067
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2752150
Commit-Queue: Mythri Alle <mythria@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Toon Verwaest <verwaest@chromium.org>
Cr-Commit-Position: refs/heads/master@{#73369}
A recent unrelated change caused these tests to get unlucky in
GC stress mode. Their "assertOptimized" expectations rely on
certain type feedback data not getting flushed at the wrong time.
Bug: v8:10846
Change-Id: I86d0b0c049539e4a69aa764cc6ec92465ca12beb
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/2381458
Reviewed-by: Maya Lekova <mslekova@chromium.org>
Commit-Queue: Maya Lekova <mslekova@chromium.org>
Auto-Submit: Jakob Kummerow <jkummerow@chromium.org>
Cr-Commit-Position: refs/heads/master@{#69617}
Updates CSA::TryToIntptr to handle array indices that are less than
INT_MAX which allows to handle string keys in the ICs.
Updates ICs to go monomorphic for string keys that are array indices.
Updates Turbofan to handle array indices when lowering element access.
Change-Id: Ibdde20130e075d0d645ab4a8266a968335eaad84
Bug: v8:9449
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1813018
Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
Reviewed-by: Toon Verwaest <verwaest@chromium.org>
Reviewed-by: Georg Neis <neis@chromium.org>
Commit-Queue: Sathya Gunasekaran <gsathya@chromium.org>
Cr-Commit-Position: refs/heads/master@{#64320}
With bytecode flushing and lazy feedback allocation, we need to call
%PrepareForOptimization before we call %OptimizeFunctionOnNextCall
Bug: v8:8801, v8:8394
Change-Id: I81918f174b2f97cbaa8b8ef2e459080c2581f535
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1588415
Commit-Queue: Mythri Alle <mythria@chromium.org>
Commit-Queue: Ross McIlroy <rmcilroy@chromium.org>
Reviewed-by: Ross McIlroy <rmcilroy@chromium.org>
Cr-Commit-Position: refs/heads/master@{#61122}
The optimization was replacing
String.fromCharCode(x) == "y"
with x == y instead of (x & 0xFFFF) == y if x was outside
of uint16 range.
Bug: v8:7340, v8:7531
Change-Id: I967306cc2e05c28de82e16cf1b2312fe47396a7d
Reviewed-on: https://chromium-review.googlesource.com/979808
Reviewed-by: Georg Neis <neis@chromium.org>
Commit-Queue: Sigurd Schneider <sigurds@chromium.org>
Cr-Commit-Position: refs/heads/master@{#52214}
With the new builtin optimization guard we can just speculatively assume
that the index passed to String#charAt and String#charCodeAt (in
optimized
code) is going to be within the valid range for the receiver. This is
what Crankshaft used to do, and it avoids Smi checks on the result for
String#charCodeAt, since it can no longer return NaN.
This gives rise to further optimizations of these builtins (i.e. to
completely avoid the tagging of char codes), and by itself already
improves the regression test originally reported from 650ms to
610ms.
Bug: v8:7127, v8:7326
Change-Id: I6c160540a1e002a37e44fa7f920e5e8f8c2c4210
Cq-Include-Trybots: master.tryserver.chromium.linux:linux_chromium_rel_ng
Reviewed-on: https://chromium-review.googlesource.com/873382
Commit-Queue: Sigurd Schneider <sigurds@chromium.org>
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50888}
For M65 I added a fast-path for String.p.charCodeAt, which behaves
incorrectly for wide characters. This patch disables the incorrect
fast-path. This is likely no performance regression against M64,
as M64 did not have a fast path for String.p.charCodeAt.
Bug: v8:7371
Change-Id: I8ea9c5da5a583138d87ccb8ffe6d61eba16630b9
Reviewed-on: https://chromium-review.googlesource.com/886782
Commit-Queue: Sigurd Schneider <sigurds@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50873}
The fast-path of S.p.charAt/charCodeAt wrongly truncates the index,
i.e. charAt(k + 4294967295) yields the same as charAt(k-1). This CL
fixes this behaviour, at the cost of not providing a fast-path for
charAt(1.1), i.e. if charAt/charCodeAt is called with a Number.
Bug: chromium:800594
Change-Id: Ic8e749380d3118f0c9469eb626e81bf72cf09fec
Reviewed-on: https://chromium-review.googlesource.com/860003
Reviewed-by: Jaroslav Sevcik <jarin@chromium.org>
Commit-Queue: Sigurd Schneider <sigurds@chromium.org>
Cr-Commit-Position: refs/heads/master@{#50497}
1. Replaces --crankshaft with --opt in tests.
2. Also fixes presubmit to check for --opt flag when
assertOptimized is used.
3. Updates testrunner/local/variants.py and
v8_foozie.py to use --opt flag.
This would mean, nooptimize variant means there are
no optimizations. Not even with %OptimizeFunctionOnNextCall.
Bug:v8:6325
Change-Id: I638e743d0773a6729c6b9749e2ca1e2537f12ce6
Reviewed-on: https://chromium-review.googlesource.com/490206
Commit-Queue: Mythri Alle <mythria@chromium.org>
Reviewed-by: Michael Achenbach <machenbach@chromium.org>
Reviewed-by: Michael Starzinger <mstarzinger@chromium.org>
Cr-Commit-Position: refs/heads/master@{#44985}
This CL adds --crankshaft and --no-always-opt flags to the tests that use
assertOptimized() and assertUnoptimized() respectively.
This CL also adds presubmit checks that ensure that tests have the proper
flags set.
BUG=v8:5890
Review-Url: https://codereview.chromium.org/2653753007
Cr-Commit-Position: refs/heads/master@{#42709}
Instead of CanRetainOtherContext, we now manually blacklist all access-checked objects.
BUG=
Review URL: https://codereview.chromium.org/1020803004
Cr-Commit-Position: refs/heads/master@{#27473}
