[Docs, Chore] (docs, tools, MobileGL, scripts, CONTRIBUTING.md, .gitignore): translate the documentation and use US spelling

This commit is contained in:
BZLZHH
2026-10-31 19:58:19 +08:00
parent 47892afddc
commit 4f50bedde0
512 changed files with 2955 additions and 2945 deletions
+2 -2
View File
@@ -20,7 +20,7 @@ exemptions; the first is gone from the tree, and the second is compiled into the
embeds (the push monolith without MG_Remote), so it is shipped code and gets the same rule.
The rule is now "no frontend-pointer KEY in any named memo family, in any build", and nothing
else; the one allow-list entry below is a named, dated carve-out.
It does not evaluate the preprocessor - it recognises the conditional spellings the
It does not evaluate the preprocessor - it recognizes the conditional spellings the
tree actually uses - and it does not look at DirectVulkan, whose own key inventory is
P7 wave 2's.
@@ -107,7 +107,7 @@ def legacy_lines(code: str) -> set[int]:
def _retired_legacy_lines(code: str) -> set[int]:
"""The pre-W3b exemption, kept only for the self-test's record of what changed.
NOT a preprocessor. It recognises `#if MOBILEGL_PIPE_LEGACY_MEMOS`,
NOT a preprocessor. It recognizes `#if MOBILEGL_PIPE_LEGACY_MEMOS`,
`#if !MOBILEGL_BUILD_DISAGGREGATED`, and the `#else` of
`#if MOBILEGL_BUILD_DISAGGREGATED` - which is every spelling these two files use -
and treats every other conditional as transparent."""
+1 -1
View File
@@ -41,7 +41,7 @@ and integration harnesses drive the library rather than being it.
THE SITES OF THE WIDENED PERIMETER, TRIAGED. The widening found 83; the same package routed
five of them through Session::Fail and added one funnel of its own, so the floor the baseline
records is 79 over 20 files. The triage is the argument for why the gate can be green with
seventy-nine deaths inside it. The ENFORCEABLE artefact is the baseline's per-file map - this
seventy-nine deaths inside it. The ENFORCEABLE artifact is the baseline's per-file map - this
prose is why that map reads the way it does.
(i) INSIDE A SANCTIONED FUNNEL - 5 sites, and the only ones allowed to carry no marker. The
+2 -2
View File
@@ -73,7 +73,7 @@ def _list_property(test, name):
def tcp_lock_violations(document):
"""Entries of a ctest --show-only=json-v1 document that reach the loopback TCP supervisor -
FIXTURES_REQUIRED names mobilegl-tcp, or the entry is labelled exactly integration-tcp - and do
FIXTURES_REQUIRED names mobilegl-tcp, or the entry is labeled exactly integration-tcp - and do
not hold RESOURCE_LOCK mobilegl-tcp. Sorted names; empty when the registration is sound."""
bad = []
for test in document.get('tests', []):
@@ -121,7 +121,7 @@ def require_entries_passed(path, names):
`TcpLane.SupervisorProtocolControls` is registered with SKIP_RETURN_CODE 77 because the
script it runs needs flatc, which is deliberately not in the build graph. That is the right
behaviour for a developer who has not built one; it is the wrong behaviour for CI, where the
behavior for a developer who has not built one; it is the wrong behavior for CI, where the
lane's whole job is to run the supervisor's protocol controls and "skipped" is
indistinguishable from the three commits the entry did not exist. The workflow builds flatc
and then names the entry here, so a CI run in which the tool went missing reds by name
+1 -1
View File
@@ -53,7 +53,7 @@ FILES = {
}
# (file, a substring of the site's line) -> why no peer byte reaches it. The site is converted
# anyway - it is on the decode path and returns like its neighbours - and the reason names the
# anyway - it is on the decode path and returns like its neighbors - and the reason names the
# earlier check that refuses the same bytes first.
UNREACHABLE = [
("PipeWireCodec.cpp", '"segment-resolve"',
+1 -1
View File
@@ -18,7 +18,7 @@ surface as a silently absent field - the exact failure the revision was added to
This script pins the two together. protocol_revision_pins.json beside it is a table
{revision: sha256(protocol.fbs)} of the schema each revision was minted for. The check reads the
header's current revision, hashes protocol.fbs (line endings normalised, so a CRLF checkout hashes
header's current revision, hashes protocol.fbs (line endings normalized, so a CRLF checkout hashes
the same bytes as an LF one) and refuses when the current revision has no row or its row is not
the file's digest. Both directions are covered: a schema edit that forgot the bump is red (stale
row), and a bump that forgot the table is red (no row), so the table cannot fall behind the header.
+1 -1
View File
@@ -25,7 +25,7 @@ runs", not "runs inproc"), and that arm's scenario is monolith-only by construct
have a spawn counterpart, so it is named as an exception rather than allowed to widen the
tolerance for everything else.
P7 PACKAGE L ADDS THE SAME QUESTION FOR MAGMA, IN TWO TIERS, AND THE NORMALISATION IS
P7 PACKAGE L ADDS THE SAME QUESTION FOR MAGMA, IN TWO TIERS, AND THE NORMALIZATION IS
DIFFERENT ON PURPOSE. The Espryt half above strips the sub-lane tails `F1.` and `Ct.` because
they name neither the case nor the arm. The Magma half must NOT: `RunAhead.Credit1.` and
`RunAhead.Credit3.` are two entries of the same case, differing in the credit the lane pins, and
+3 -3
View File
@@ -19,7 +19,7 @@ asked about - and nothing about a green split lane says so. This asserts, per ba
after the arm and tail segments come off both sides: an entry name is `<Backend>.[<Arm>.][<tail>.]
<Suite>.<Case>`, and only the last two segments (the gtest name) are compared. `Split.F1.`,
`Spawn.ForcedDs.`, `Tcp.MultiDrawTierBaseVertex.` and the monolith's
`ForcedDepthStencilEmulation.` all normalise away.
`ForcedDepthStencilEmulation.` all normalize away.
WHAT COUNTS AS A SPLIT CASE differs by backend, on purpose:
* DirectGLES - an entry with an arm segment (Split/Spawn/Tcp/TcpDevice) that carries a GATED
@@ -33,7 +33,7 @@ WHAT COUNTS AS A SPLIT CASE differs by backend, on purpose:
WHAT COUNTS AS A MONOLITH CASE: an entry with no arm segment carrying `integration-gpu` - the label
the monolith registrations and the monolith-control mechanism entries share. That puts
`DirectGLES.PushMonolithArm.<case>` (labelled integration-split, but no arm) on the monolith side,
`DirectGLES.PushMonolithArm.<case>` (labeled integration-split, but no arm) on the monolith side,
where it belongs, and leaves build-verify's `DirectGLES.VerifySplit.<case>` (no integration-gpu)
out of both sides.
@@ -254,7 +254,7 @@ def self_test():
if got != rc or any(n not in text for n in needles) or any(n in text for n in absent):
failures.append(f"{label}: rc {got} (want {rc}); output:\n{text}")
expect("tails normalise on both sides", _doc(*BASE), "", 0,
expect("tails normalize on both sides", _doc(*BASE), "", 0,
"DirectGLES: monolith 3, on a gated split arm 3", "missing 0")
expect("a monolith case on no arm reds BY NAME",
_doc(*BASE, _entry("DirectGLES.CScenario.Five", "integration-gpu")), "", 1,
+1 -1
View File
@@ -189,7 +189,7 @@ class Accounting(unittest.TestCase):
`ctest -L integration-gpu -j 4` that entry and another Tcp case killed each other.
The stub mirrors the real json-v1 shapes: the fixture's own setup entry, a sound Tcp
case, the violator as it was registered, a Magma tcp entry that reaches the fixture
under another label, a labelled control with no fixture, and the device lane, whose
under another label, a labeled control with no fixture, and the device lane, whose
lock is its own."""
tally = module('junit_tally')
parity = module('spawn_lane_parity')