replica
/
hanchenye-llvm-project
2
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Improve performance TokenizeWindowsCommandLine 

Patcy by Takuto Ikuta.

This patch reduces lld link time of chromium's blink_core.dll in
component build.

Total size of input argument in .directives become nearly 300MB in the
build and calling many strchr and assert becomes bottleneck.

On my desktop machine, 4 times stats of the link time are like below.
Improved around 10%.

This patch
TotalSeconds : 13.4918885
TotalSeconds : 13.9474257
TotalSeconds : 13.4941082
TotalSeconds : 13.6077962
Avg : 13.63530465

master
TotalSeconds : 15.6938531
TotalSeconds : 15.7022508
TotalSeconds : 15.9567202
TotalSeconds : 14.5851505
Avg : 15.48449365

Differential Revision: https://reviews.llvm.org/D41590

llvm-svn: 321479
This commit is contained in:
Rui Ueyama 2017-12-27 08:59:52 +00:00
parent fb2fd20f50
commit 6ec880d9b5
1 changed files with 26 additions and 20 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

46
llvm/lib/Support/CommandLine.cpp
View File

@ -688,7 +688,9 @@ static bool EatsUnboundedNumberOfValues(const Option *O) {
O->getNumOccurrencesFlag() == cl::OneOrMore;
}
static bool isWhitespace(char C) { return strchr(" \t\n\r\f\v", C); }
static bool isWhitespace(char C) {
return C == ' ' || C == '\t' || C == '\r' || C == '\n';
}
static bool isQuote(char C) { return C == '\"' || C == '\''; }
@ -709,17 +711,19 @@ void cl::TokenizeGNUCommandLine(StringRef Src, StringSaver &Saver,
break;
}
char C = Src[I];
// Backslash escapes the next character.
if (I + 1 < E && Src[I] == '\\') {
if (I + 1 < E && C == '\\') {
++I; // Skip the escape.
Token.push_back(Src[I]);
continue;
}
// Consume a quoted string.
if (isQuote(Src[I])) {
char Quote = Src[I++];
while (I != E && Src[I] != Quote) {
if (isQuote(C)) {
++I;
while (I != E && Src[I] != C) {
// Backslash escapes the next character.
if (Src[I] == '\\' && I + 1 != E)
++I;
@ -732,7 +736,7 @@ void cl::TokenizeGNUCommandLine(StringRef Src, StringSaver &Saver,
}
// End the token if this is whitespace.
if (isWhitespace(Src[I])) {
if (isWhitespace(C)) {
if (!Token.empty())
NewArgv.push_back(Saver.save(StringRef(Token)).data());
Token.clear();
@ -740,7 +744,7 @@ void cl::TokenizeGNUCommandLine(StringRef Src, StringSaver &Saver,
}
// This is a normal character. Append it.
Token.push_back(Src[I]);
Token.push_back(C);
}
// Append the last token after hitting EOF with no whitespace.
@ -798,25 +802,27 @@ void cl::TokenizeWindowsCommandLine(StringRef Src, StringSaver &Saver,
// end of the source string.
enum { INIT, UNQUOTED, QUOTED } State = INIT;
for (size_t I = 0, E = Src.size(); I != E; ++I) {
char C = Src[I];
// INIT state indicates that the current input index is at the start of
// the string or between tokens.
if (State == INIT) {
if (isWhitespace(Src[I])) {
if (isWhitespace(C)) {
// Mark the end of lines in response files
if (MarkEOLs && Src[I] == '\n')
if (MarkEOLs && C == '\n')
NewArgv.push_back(nullptr);
continue;
}
if (Src[I] == '"') {
if (C == '"') {
State = QUOTED;
continue;
}
if (Src[I] == '\\') {
if (C == '\\') {
I = parseBackslash(Src, I, Token);
State = UNQUOTED;
continue;
}
Token.push_back(Src[I]);
Token.push_back(C);
State = UNQUOTED;
continue;
}
@ -825,38 +831,38 @@ void cl::TokenizeWindowsCommandLine(StringRef Src, StringSaver &Saver,
// quotes.
if (State == UNQUOTED) {
// Whitespace means the end of the token.
if (isWhitespace(Src[I])) {
if (isWhitespace(C)) {
NewArgv.push_back(Saver.save(StringRef(Token)).data());
Token.clear();
State = INIT;
// Mark the end of lines in response files
if (MarkEOLs && Src[I] == '\n')
if (MarkEOLs && C == '\n')
NewArgv.push_back(nullptr);
continue;
}
if (Src[I] == '"') {
if (C == '"') {
State = QUOTED;
continue;
}
if (Src[I] == '\\') {
if (C == '\\') {
I = parseBackslash(Src, I, Token);
continue;
}
Token.push_back(Src[I]);
Token.push_back(C);
continue;
}
// QUOTED state means that it's reading a token quoted by double quotes.
if (State == QUOTED) {
if (Src[I] == '"') {
if (C == '"') {
State = UNQUOTED;
continue;
}
if (Src[I] == '\\') {
if (C == '\\') {
I = parseBackslash(Src, I, Token);
continue;
}
Token.push_back(Src[I]);
Token.push_back(C);
}
}
// Append the last token after hitting EOF with no whitespace.