[PATCH] D92180: Speedup some unicode rendering

[serge via Phabricator via llvm-commits]

serge-sans-paille created this revision.
serge-sans-paille added reviewers: aguinet, alexfh.
Herald added subscribers: llvm-commits, dexonsmith, hiraditya.
Herald added a project: LLVM.
serge-sans-paille requested review of this revision.

Use a short path for column width computation for ascii characters. Especially relevant for llvm-objdump.

      

before:

  
  % time ./bin/llvm-objdump -D  -j .text /lib/libc.so.6 >/dev/null
  ./bin/llvm-objdump -D -j .text /lib/libc.so.6 > /dev/null  0.75s user 0.01s system 99% cpu 0.757 total


after:

  
  % time ./bin/llvm-objdump -D  -j .text /lib/libc.so.6 >/dev/null
  ./bin/llvm-objdump -D -j .text /lib/libc.so.6 > /dev/null  0.37s user 0.01s system 99% cpu 0.378 total


https://reviews.llvm.org/D92180

Files:
  llvm/lib/Support/Unicode.cpp
  llvm/unittests/Support/UnicodeTest.cpp


Index: llvm/unittests/Support/UnicodeTest.cpp
===================================================================
--- llvm/unittests/Support/UnicodeTest.cpp
+++ llvm/unittests/Support/UnicodeTest.cpp
@@ -7,6 +7,7 @@
 //===----------------------------------------------------------------------===//
 
 #include "llvm/Support/Unicode.h"
+#include "llvm/Support/ConvertUTF.h"
 #include "gtest/gtest.h"
 
 namespace llvm {
@@ -23,6 +24,7 @@
   EXPECT_EQ(6, columnWidthUTF8("abcdef"));
 
   EXPECT_EQ(-1, columnWidthUTF8("\x01"));
+  EXPECT_EQ(-1, columnWidthUTF8("\t"));
   EXPECT_EQ(-1, columnWidthUTF8("aaaaaaaaaa\x01"));
   EXPECT_EQ(-1, columnWidthUTF8("\342\200\213")); // 200B ZERO WIDTH SPACE
 
@@ -84,6 +86,18 @@
   EXPECT_TRUE(isPrintable(0x20000));  // CJK UNIFIED IDEOGRAPH-20000
 
   EXPECT_FALSE(isPrintable(0x10FFFF)); // noncharacter
+
+  // test the validity of a fast path in columnWidthUTF8
+  for (unsigned char c = 0; c < 128; ++c) {
+    const UTF8 buf8[1] = {c};
+    const UTF8 *Target8 = &buf8[0];
+    UTF32 buf32[1];
+    UTF32 *Target32 = &buf32[0];
+    auto status = ConvertUTF8toUTF32(&Target8, Target8 + 1, &Target32,
+                                     Target32 + 1, strictConversion);
+    EXPECT_TRUE(status == conversionOK);
+    EXPECT_TRUE((bool)isprint(c) == (bool)isPrintable(buf32[0]));
+  }
 }
 
 } // namespace
Index: llvm/lib/Support/Unicode.cpp
===================================================================
--- llvm/lib/Support/Unicode.cpp
+++ llvm/lib/Support/Unicode.cpp
@@ -353,6 +353,15 @@
   unsigned Length;
   for (size_t i = 0, e = Text.size(); i < e; i += Length) {
     Length = getNumBytesForUTF8(Text[i]);
+
+    // fast path for ASCII characters
+    if (Length == 1) {
+      if (!isprint(Text[i]))
+        return ErrorNonPrintableCharacter;
+      ColumnWidth += 1;
+      continue;
+    }
+
     if (Length <= 0 || i + Length > Text.size())
       return ErrorInvalidUTF8;
     UTF32 buf[1];


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D92180.307865.patch
Type: text/x-patch
Size: 1957 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20201126/cb2396ca/attachment.bin>