Skip to content

V
V8
Switch branch/tag
  1. 26 Sep, 2019 1 commit
  3. 19 Sep, 2019 1 commit
  5. 23 Aug, 2019 1 commit
  7. 20 Aug, 2019 1 commit
  9. 13 Aug, 2019 1 commit
    • Jakob Gruber's avatar
      [isolate-data] Move hot fields closer to isolate_root · fb698cec
      Jakob Gruber authored 
      In generated code, we access fields inside IsolateData through the
root-register. On some platforms it is significantly cheaper to access
things that are close to the root-register value than things that are
located far away. The motivation for this CL was a 5% difference in
Octane/Mandreel scores between

// Part of the stack check.
cmpq rsp,[r13+0x9ea8]

and

cmpq rsp,[r13-0x30]  // Mandreel score improved by 5%.

This moves the StackGuard up to fix Mandreel. As a drive-by, also move
two more fields up that are accessed by each CallCFunction.

Tbr: yangguo@chromium.org
Bug: v8:9534,chromium:993264
Change-Id: I5418b63d40274a138e285fa3c99b96e33a814fb1
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1751345Reviewed-by: 's avatarJakob Gruber <jgruber@chromium.org>
Reviewed-by: 's avatarYang Guo <yangguo@chromium.org>
Auto-Submit: Jakob Gruber <jgruber@chromium.org>
Commit-Queue: Yang Guo <yangguo@chromium.org>
Cr-Commit-Position: refs/heads/master@{#63187}
      fb698cec
  11. 10 May, 2019 1 commit
  13. 24 Apr, 2019 1 commit
  15. 23 Apr, 2019 2 commits
  17. 18 Apr, 2019 1 commit
  19. 19 Mar, 2019 2 commits
    • Igor Sheludko's avatar
      [ptr-compr][ubsan] Use [Read/Write]UnalignedValue for unaligned fields · 0188634e
      Igor Sheludko authored 
      When pointer compression is enabled the [u]intptr_t and double fields are
only kTaggedSize aligned so in order to avoid undefined behavior in C++ code
we have to access these values in an unaligned pointer friendly way although
both x64 and arm64 architectures (where pointer compression is supported)
allow unaligned access.

These changes will be removed once v8:8875 is fixed and all the
kSystemPointerSize fields are properly aligned.

Bug: v8:7703
Change-Id: I4df477cbdeab806303bb4f675d52b61c06342c8e
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1528996
Commit-Queue: Igor Sheludko <ishell@chromium.org>
Reviewed-by: 's avatarUlan Degenbaev <ulan@chromium.org>
Reviewed-by: 's avatarJakob Gruber <jgruber@chromium.org>
Reviewed-by: 's avatarClemens Hammacher <clemensh@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60321}
      0188634e
    • Benedikt Meurer's avatar
      [turbofan] Significantly improve ConsString creation performance. · d6a60a0e
      Benedikt Meurer authored 
      This change significantly improves the performance of string
concatenation in optimized code for the case where the resulting string
is represented as a ConsString. On the relevant test cases we go from

  serializeNaive: 10762 ms.
  serializeClever: 7813 ms.
  serializeConcat: 10271 ms.

to

  serializeNaive: 10278 ms.
  serializeClever: 5533 ms.
  serializeConcat: 10310 ms.

which represents a 30% improvement on the "clever" benchmark, which
tests specifically the ConsString creation performance.

This was accomplished via a couple of different steps, which are briefly
outlined here:

  1. The empty_string gets its own map, so that we can easily recognize
     and handle it appropriately in the TurboFan type system. This
     allows us to express (and assert) that the inputs to NewConsString
     are non-empty strings, making sure that TurboFan no longer creates
     "crippled ConsStrings" with empty left or right hand sides.
  2. Further split the existing String types in TurboFan to be able to
     distinguish between OneByte and TwoByte strings on the type system
     level. This allows us to avoid having to dynamically lookup the
     resulting ConsString map in case of ConsString creation (i.e. when
     we know that both input strings are OneByte strings or at least
     one of the input strings is TwoByte).
  3. We also introduced more finegrained feedback for the Add bytecode
     in the interpreter, having it collect feedback about ConsStrings,
     specifically ConsOneByteString and ConsTwoByteString. This feedback
     can be used by TurboFan to only inline the relevant code for what
     was seen so far. This allows us to remove the Octane/Splay specific
     magic in JSTypedLowering to detect ConsString creation, and instead
     purely rely on the feedback of what was seen so far (also making it
     possible to change the semantics of NewConsString to be a low-level
     operator, which is only introduced in SimplifiedLowering by looking
     at the input types of StringConcat).
  4. On top of the before mentioned type and interpreter changes we added
     new operators CheckNonEmptyString, CheckNonEmptyOneByteString, and
     CheckNonEmptyTwoByteString, which perform the appropriate (dynamic)
     checks.

There are several more improvements that are possible based on this, but
since the change was already quite big, we decided not to put everything
into the first change, but do some follow up tweaks to the type system,
and builtin optimizations later.

Tbr: mstarzinger@chromium.org
Bug: v8:8834, v8:8931, v8:8939, v8:8951
Change-Id: Ia24e17c6048bf2b04df966d3cd441f0edda05c93
Cq-Include-Trybots: luci.chromium.try:linux-blink-rel
Doc: https://bit.ly/fast-string-concatenation-in-javascript
Reviewed-on: https://chromium-review.googlesource.com/c/v8/v8/+/1499497
Commit-Queue: Michael Achenbach <machenbach@chromium.org>
Reviewed-by: 's avatarYang Guo <yangguo@chromium.org>
Reviewed-by: 's avatarJaroslav Sevcik <jarin@chromium.org>
Reviewed-by: 's avatarMythri Alle <mythria@chromium.org>
Reviewed-by: 's avatarBenedikt Meurer <bmeurer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60318}
      d6a60a0e
  21. 18 Mar, 2019 1 commit
  23. 06 Mar, 2019 1 commit
  25. 04 Mar, 2019 2 commits
    • Igor Sheludko's avatar
      [ptr-compr] Prepare for changing kTaggedSize, pt.3 · 17448030
      Igor Sheludko authored 
      This CL also gives up trying to maintain double and system word
fields at aligned addresses because currently it's not always
maintained (v8:8875) and Torque object definitions do not support
padding fields (v8:8863).

Given that both platforms where pointer compression is going to be
enabled (x64 and arm64) support loading of doubles and full words
from 4-byte aligned addresses we are fine.

Bug: v8:7703
Change-Id: I99fc6da5a0927f4db9b8fb24c7cc0bfc416523bc
Reviewed-on: https://chromium-review.googlesource.com/c/1496974
Auto-Submit: Igor Sheludko <ishell@chromium.org>
Commit-Queue: Ulan Degenbaev <ulan@chromium.org>
Reviewed-by: 's avatarToon Verwaest <verwaest@chromium.org>
Reviewed-by: 's avatarUlan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60013}
      17448030
    • Benedikt Meurer's avatar
      [cleanup] Remove obsolete "one byte data hint" for strings. · 683cf6f4
      Benedikt Meurer authored 
      In the early days of Chrome when we used WebKit there was no support for
ASCII strings on the C++ side, so we put a hint onto these two-byte
strings that said "string only contains one byte data", such that
internally in V8 when these were involved in string operations, we could
instead create the *cheaper* one byte strings.

Nowadays Blink properly supports one-byte string representations and
this additional hint only comes with overhead, since we check it in
quite a few places (i.e. on the hot path for string concatenation), plus
we end up consuming more memory due to the additional string maps.
Removing the hint also frees one bit in the InstanceType zoo for
strings.

This alone improves performance on the `bench-dom-serialize.js` test case
by around **3%**.

Tbr: mstarzinger@chromium.org
Bug: v8:6622, v8:8834, v8:8939
Cq-Include-Trybots: luci.chromium.try:linux-blink-rel
Change-Id: I0753f2859cee7b5a37b6f0da64d8ec39fcb044ff
Doc: https://bit.ly/fast-string-concatenation-in-javascript
Reviewed-on: https://chromium-review.googlesource.com/c/1498478
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: 's avatarYang Guo <yangguo@chromium.org>
Cr-Commit-Position: refs/heads/master@{#60006}
      683cf6f4
  27. 08 Feb, 2019 1 commit
  29. 08 Jan, 2019 1 commit
  31. 27 Dec, 2018 1 commit
  33. 13 Dec, 2018 1 commit
  35. 12 Dec, 2018 1 commit
  37. 11 Dec, 2018 1 commit
  39. 30 Nov, 2018 1 commit
  41. 22 Nov, 2018 1 commit
  43. 21 Nov, 2018 2 commits
  45. 20 Nov, 2018 2 commits
  47. 05 Nov, 2018 1 commit
  49. 30 Oct, 2018 1 commit
  51. 26 Oct, 2018 1 commit
  53. 24 Oct, 2018 1 commit
  55. 16 Oct, 2018 1 commit
  57. 13 Sep, 2018 1 commit
    • Benedikt Meurer's avatar
      [objects] Change String::length field to uint32_t. · c7a0049e
      Benedikt Meurer authored 
      This changes the Name::hash_field and Symbol::flags to uint32_t as
well, so that both Symbols and Strings consume one fewer word on 64-bit
architectures now. More importantly the access to String::length is
always a 32-bit field load now, even with 31-bit Smis (i.e. on ARM or
on 64-bit with pointer compression), so the access should be faster.

Bug: v8:7065, v8:8171
Change-Id: I1a38f4470d62fbeba2b3bc5fcf4ecdbada7d6b8a
Tbr: ulan@chromium.org, yangguo@chromium.org, ishell@chromium.org
Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;luci.v8.try:v8_linux_noi18n_rel_ng
Reviewed-on: https://chromium-review.googlesource.com/1224432Reviewed-by: 's avatarBenedikt Meurer <bmeurer@chromium.org>
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55861}
      c7a0049e
  59. 12 Sep, 2018 6 commits
    • Sathya Gunasekaran's avatar
      Revert "Reland "[objects] Change String::length field to uint32_t."" · 350dfb62
      Sathya Gunasekaran authored 
      This reverts commit a03cec2c.

Reason for revert: https://ci.chromium.org/p/v8/builders/luci.v8.ci/V8%20Linux64%20GC%20Stress%20-%20custom%20snapshot/21320

Original change's description:
> Reland "[objects] Change String::length field to uint32_t."
> 
> This is a reland of 1f1eb625, the
> breakage on the GCStress bot seems to be unrelated (maybe flushed
> out by this change). We decided to reland to figure out whether it's
> a random flake or really triggered by this particular change.
> 
> Original change's description:
> > [objects] Change String::length field to uint32_t.
> >
> > This changes the Name::hash_field and Symbol::flags to uint32_t as
> > well, so that both Symbols and Strings consume one fewer word on 64-bit
> > architectures now. More importantly the access to String::length is
> > always a 32-bit field load now, even with 31-bit Smis (i.e. on ARM or
> > on 64-bit with pointer compression), so the access should be faster.
> >
> > Bug: v8:7065, v8:8171
> > Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;luci.v8.try:v8_linux_noi18n_rel_ng
> > Change-Id: I5523deb1f84ece91fa2fea775d50318bd1300493
> > Reviewed-on: https://chromium-review.googlesource.com/1221288
> > Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
> > Reviewed-by: Yang Guo <yangguo@chromium.org>
> > Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
> > Reviewed-by: Igor Sheludko <ishell@chromium.org>
> > Cr-Commit-Position: refs/heads/master@{#55825}
> 
> Bug: v8:7065, v8:8171
> Tbr: tebbi@chromium.org, yangguo@chromium.org, ishell@chromium.org, ulan@chromium.org
> Change-Id: I2be24ac018591c04c826e7e8db82e007b738d156
> Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;luci.v8.try:v8_linux_noi18n_rel_ng
> Reviewed-on: https://chromium-review.googlesource.com/1222308
> Reviewed-by: Benedikt Meurer <bmeurer@chromium.org>
> Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#55838}

TBR=yangguo@chromium.org,tebbi@chromium.org,ishell@chromium.org,bmeurer@chromium.org

Change-Id: Ic741c3d407d4257a8c86b3082b9a19e33dc89215
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: v8:7065, v8:8171
Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;luci.v8.try:v8_linux_noi18n_rel_ng
Reviewed-on: https://chromium-review.googlesource.com/1222368Reviewed-by: 's avatarSathya Gunasekaran <gsathya@chromium.org>
Commit-Queue: Sathya Gunasekaran <gsathya@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55839}
      350dfb62
    • Benedikt Meurer's avatar
      Reland "[objects] Change String::length field to uint32_t." · a03cec2c
      Benedikt Meurer authored 
      This is a reland of 1f1eb625, the
breakage on the GCStress bot seems to be unrelated (maybe flushed
out by this change). We decided to reland to figure out whether it's
a random flake or really triggered by this particular change.

Original change's description:
> [objects] Change String::length field to uint32_t.
>
> This changes the Name::hash_field and Symbol::flags to uint32_t as
> well, so that both Symbols and Strings consume one fewer word on 64-bit
> architectures now. More importantly the access to String::length is
> always a 32-bit field load now, even with 31-bit Smis (i.e. on ARM or
> on 64-bit with pointer compression), so the access should be faster.
>
> Bug: v8:7065, v8:8171
> Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;luci.v8.try:v8_linux_noi18n_rel_ng
> Change-Id: I5523deb1f84ece91fa2fea775d50318bd1300493
> Reviewed-on: https://chromium-review.googlesource.com/1221288
> Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
> Reviewed-by: Yang Guo <yangguo@chromium.org>
> Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
> Reviewed-by: Igor Sheludko <ishell@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#55825}

Bug: v8:7065, v8:8171
Tbr: tebbi@chromium.org, yangguo@chromium.org, ishell@chromium.org, ulan@chromium.org
Change-Id: I2be24ac018591c04c826e7e8db82e007b738d156
Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;luci.v8.try:v8_linux_noi18n_rel_ng
Reviewed-on: https://chromium-review.googlesource.com/1222308Reviewed-by: 's avatarBenedikt Meurer <bmeurer@chromium.org>
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55838}
      a03cec2c
    • Benedikt Meurer's avatar
      Revert "Reland "[objects] Change String::length field to uint32_t."" · bd69d64d
      Benedikt Meurer authored 
      This reverts commit df6157ae.

Reason for revert: trybots didn't rerun :-/

Original change's description:
> Reland "[objects] Change String::length field to uint32_t."
> 
> This is a reland of 1f1eb625, the
> breakage on the GCStress bot seems to be unrelated (maybe flushed
> out by this change). We decided to reland to figure out whether it's
> a random flake or really triggered by this particular change.
> 
> Original change's description:
> > [objects] Change String::length field to uint32_t.
> >
> > This changes the Name::hash_field and Symbol::flags to uint32_t as
> > well, so that both Symbols and Strings consume one fewer word on 64-bit
> > architectures now. More importantly the access to String::length is
> > always a 32-bit field load now, even with 31-bit Smis (i.e. on ARM or
> > on 64-bit with pointer compression), so the access should be faster.
> >
> > Bug: v8:7065, v8:8171
> > Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;luci.v8.try:v8_linux_noi18n_rel_ng
> > Change-Id: I5523deb1f84ece91fa2fea775d50318bd1300493
> > Reviewed-on: https://chromium-review.googlesource.com/1221288
> > Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
> > Reviewed-by: Yang Guo <yangguo@chromium.org>
> > Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
> > Reviewed-by: Igor Sheludko <ishell@chromium.org>
> > Cr-Commit-Position: refs/heads/master@{#55825}
> 
> Tbr: tebbi@chromium.org, yangguo@chromium.org, ishell@chromium.org
> Bug: v8:7065, v8:8171
> Change-Id: I3c7d0b00abb15fa98ab622f9ecd8602fc798cbc3
> Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;luci.v8.try:v8_linux_noi18n_rel_ng
> Reviewed-on: https://chromium-review.googlesource.com/1221290
> Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
> Reviewed-by: Ulan Degenbaev <ulan@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#55836}

TBR=ulan@chromium.org,yangguo@chromium.org,tebbi@chromium.org,ishell@chromium.org,bmeurer@chromium.org

Change-Id: Ieaf3be31166abb02e37370ad846c38fa3d114693
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: v8:7065, v8:8171
Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;luci.v8.try:v8_linux_noi18n_rel_ng
Reviewed-on: https://chromium-review.googlesource.com/1222306Reviewed-by: 's avatarBenedikt Meurer <bmeurer@chromium.org>
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55837}
      bd69d64d
    • Benedikt Meurer's avatar
      Reland "[objects] Change String::length field to uint32_t." · df6157ae
      Benedikt Meurer authored 
      This is a reland of 1f1eb625, the
breakage on the GCStress bot seems to be unrelated (maybe flushed
out by this change). We decided to reland to figure out whether it's
a random flake or really triggered by this particular change.

Original change's description:
> [objects] Change String::length field to uint32_t.
>
> This changes the Name::hash_field and Symbol::flags to uint32_t as
> well, so that both Symbols and Strings consume one fewer word on 64-bit
> architectures now. More importantly the access to String::length is
> always a 32-bit field load now, even with 31-bit Smis (i.e. on ARM or
> on 64-bit with pointer compression), so the access should be faster.
>
> Bug: v8:7065, v8:8171
> Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;luci.v8.try:v8_linux_noi18n_rel_ng
> Change-Id: I5523deb1f84ece91fa2fea775d50318bd1300493
> Reviewed-on: https://chromium-review.googlesource.com/1221288
> Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
> Reviewed-by: Yang Guo <yangguo@chromium.org>
> Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
> Reviewed-by: Igor Sheludko <ishell@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#55825}

Tbr: tebbi@chromium.org, yangguo@chromium.org, ishell@chromium.org
Bug: v8:7065, v8:8171
Change-Id: I3c7d0b00abb15fa98ab622f9ecd8602fc798cbc3
Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;luci.v8.try:v8_linux_noi18n_rel_ng
Reviewed-on: https://chromium-review.googlesource.com/1221290
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: 's avatarUlan Degenbaev <ulan@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55836}
      df6157ae
    • Leszek Swirski's avatar
      Revert "[objects] Change String::length field to uint32_t." · 4bbb7c4e
      Leszek Swirski authored 
      This reverts commit 1f1eb625.

Reason for revert: GC Stress failure (https://ci.chromium.org/p/v8/builders/luci.v8.ci/V8%20Linux64%20GC%20Stress%20-%20custom%20snapshot/21311) 

Original change's description:
> [objects] Change String::length field to uint32_t.
> 
> This changes the Name::hash_field and Symbol::flags to uint32_t as
> well, so that both Symbols and Strings consume one fewer word on 64-bit
> architectures now. More importantly the access to String::length is
> always a 32-bit field load now, even with 31-bit Smis (i.e. on ARM or
> on 64-bit with pointer compression), so the access should be faster.
> 
> Bug: v8:7065, v8:8171
> Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;luci.v8.try:v8_linux_noi18n_rel_ng
> Change-Id: I5523deb1f84ece91fa2fea775d50318bd1300493
> Reviewed-on: https://chromium-review.googlesource.com/1221288
> Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
> Reviewed-by: Yang Guo <yangguo@chromium.org>
> Reviewed-by: Tobias Tebbi <tebbi@chromium.org>
> Reviewed-by: Igor Sheludko <ishell@chromium.org>
> Cr-Commit-Position: refs/heads/master@{#55825}

TBR=yangguo@chromium.org,tebbi@chromium.org,ishell@chromium.org,bmeurer@chromium.org

Change-Id: I73f3200902f9d52e5664d48c938e37d9dfb7bce7
No-Presubmit: true
No-Tree-Checks: true
No-Try: true
Bug: v8:7065, v8:8171
Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;luci.v8.try:v8_linux_noi18n_rel_ng
Reviewed-on: https://chromium-review.googlesource.com/1221706Reviewed-by: 's avatarLeszek Swirski <leszeks@chromium.org>
Commit-Queue: Leszek Swirski <leszeks@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55826}
      4bbb7c4e
    • Benedikt Meurer's avatar
      [objects] Change String::length field to uint32_t. · 1f1eb625
      Benedikt Meurer authored 
      This changes the Name::hash_field and Symbol::flags to uint32_t as
well, so that both Symbols and Strings consume one fewer word on 64-bit
architectures now. More importantly the access to String::length is
always a 32-bit field load now, even with 31-bit Smis (i.e. on ARM or
on 64-bit with pointer compression), so the access should be faster.

Bug: v8:7065, v8:8171
Cq-Include-Trybots: luci.chromium.try:linux_chromium_rel_ng;luci.v8.try:v8_linux_noi18n_rel_ng
Change-Id: I5523deb1f84ece91fa2fea775d50318bd1300493
Reviewed-on: https://chromium-review.googlesource.com/1221288
Commit-Queue: Benedikt Meurer <bmeurer@chromium.org>
Reviewed-by: 's avatarYang Guo <yangguo@chromium.org>
Reviewed-by: 's avatarTobias Tebbi <tebbi@chromium.org>
Reviewed-by: 's avatarIgor Sheludko <ishell@chromium.org>
Cr-Commit-Position: refs/heads/master@{#55825}
      1f1eb625