about summary refs log tree commit diff
diff options
context:
space:
mode:
authorAbseil Team <absl-team@google.com>2019-04-16T19·11-0700
committerAlex Strelnikov <strel@google.com>2019-04-16T19·40-0400
commitd902eb869bcfacc1bad14933ed9af4bed006d481 (patch)
tree2c6b4c121bb5a696657d63db03a166404b3592b8
parenta02f62f456f2c4a7ecf2be3104fe0c6e16fbad9a (diff)
Export of internal Abseil changes.
--
babbb6421068af3831870fd5995444437ace6769 by Derek Mauro <dmauro@google.com>:

Rollback of:

Make raw_hash_set_test less flaky.

Split the timing loop into chunks so that we are less suceptible to
antogantistic processes.

PiperOrigin-RevId: 243854490

--
a2711f17a712f6d09799bf32363d67526737b486 by CJ Johnson <johnsoncj@google.com>:

Relocates IsAtLeastForwardIterator to internal/inlined_vector.h

PiperOrigin-RevId: 243846090

--
6c14cdbeb9a61022c27f8957654f930d8abf2fc1 by Matt Kulukundis <kfm@google.com>:

Make raw_hash_set_test less flaky.

Split the timing loop into chunks so that we are less suceptible to
antogantistic processes.

PiperOrigin-RevId: 243824289

--
ee6072a6b6e0ac653622524ceb09db3b9e870f96 by Samuel Benzaquen <sbenza@google.com>:

Improve format parser performance.
Replace the main switch with a lookup in the existing tag table.
Improve the ABI of ConsumeUnboundConversion a little.

PiperOrigin-RevId: 243824112

--
24b9e6476dfa4be8d644359eab8ac6816492f187 by Abseil Team <absl-team@google.com>:

Fix DR numbers: 3800 ? 3080, 3801 ? 3081.

PiperOrigin-RevId: 243804213

--
0660404074707e197684f07cc0bffe4a9c35cd2f by Abseil Team <absl-team@google.com>:

Internal change.

PiperOrigin-RevId: 243757359

--
ba0f5bb9b8584d75c4ffc44ff3cb8c691796ffc6 by Xiaoyi Zhang <zhangxy@google.com>:

Consolidate ABSL_INTERNAL_UNALIGNED_* implementation into memcpy.
The compiler should be good enough to optimize these operations.
See https://github.com/abseil/abseil-cpp/issues/269 for background.

PiperOrigin-RevId: 243323941

--
00853a8756548df7217513c562d604b4ee5c6ab9 by Eric Fiselier <ericwf@google.com>:

Reexport memory.h from optional.h for compatibility between libc++ and
libstdc++.

PiperOrigin-RevId: 243313425
GitOrigin-RevId: babbb6421068af3831870fd5995444437ace6769
Change-Id: Ic53c127ad857a431ad60c98b27cc585fed50a3e3
-rw-r--r--absl/base/internal/unaligned_access.h165
-rw-r--r--absl/container/inlined_vector.h25
-rw-r--r--absl/container/internal/inlined_vector.h9
-rw-r--r--absl/hash/BUILD.bazel2
-rw-r--r--absl/strings/charconv.h4
-rw-r--r--absl/strings/internal/str_format/bind_test.cc9
-rw-r--r--absl/strings/internal/str_format/parser.cc143
-rw-r--r--absl/strings/internal/str_format/parser.h128
-rw-r--r--absl/strings/internal/str_format/parser_test.cc25
-rw-r--r--absl/strings/numbers.cc4
10 files changed, 186 insertions, 328 deletions
diff --git a/absl/base/internal/unaligned_access.h b/absl/base/internal/unaligned_access.h
index 2d6673770afc..2cf7c1d45b3f 100644
--- a/absl/base/internal/unaligned_access.h
+++ b/absl/base/internal/unaligned_access.h
@@ -25,15 +25,6 @@
 // unaligned APIs
 
 // Portable handling of unaligned loads, stores, and copies.
-// On some platforms, like ARM, the copy functions can be more efficient
-// then a load and a store.
-//
-// It is possible to implement all of these these using constant-length memcpy
-// calls, which is portable and will usually be inlined into simple loads and
-// stores if the architecture supports it. However, such inlining usually
-// happens in a pass that's quite late in compilation, which means the resulting
-// loads and stores cannot participate in many other optimizations, leading to
-// overall worse code.
 
 // The unaligned API is C++ only.  The declarations use C++ features
 // (namespaces, inline) which are absent or incompatible in C.
@@ -108,164 +99,8 @@ inline void UnalignedStore64(void *p, uint64_t v) {
 #define ABSL_INTERNAL_UNALIGNED_STORE64(_p, _val) \
   (absl::base_internal::UnalignedStore64(_p, _val))
 
-#elif defined(UNDEFINED_BEHAVIOR_SANITIZER)
-
-namespace absl {
-namespace base_internal {
-
-inline uint16_t UnalignedLoad16(const void *p) {
-  uint16_t t;
-  memcpy(&t, p, sizeof t);
-  return t;
-}
-
-inline uint32_t UnalignedLoad32(const void *p) {
-  uint32_t t;
-  memcpy(&t, p, sizeof t);
-  return t;
-}
-
-inline uint64_t UnalignedLoad64(const void *p) {
-  uint64_t t;
-  memcpy(&t, p, sizeof t);
-  return t;
-}
-
-inline void UnalignedStore16(void *p, uint16_t v) { memcpy(p, &v, sizeof v); }
-
-inline void UnalignedStore32(void *p, uint32_t v) { memcpy(p, &v, sizeof v); }
-
-inline void UnalignedStore64(void *p, uint64_t v) { memcpy(p, &v, sizeof v); }
-
-}  // namespace base_internal
-}  // namespace absl
-
-#define ABSL_INTERNAL_UNALIGNED_LOAD16(_p) \
-  (absl::base_internal::UnalignedLoad16(_p))
-#define ABSL_INTERNAL_UNALIGNED_LOAD32(_p) \
-  (absl::base_internal::UnalignedLoad32(_p))
-#define ABSL_INTERNAL_UNALIGNED_LOAD64(_p) \
-  (absl::base_internal::UnalignedLoad64(_p))
-
-#define ABSL_INTERNAL_UNALIGNED_STORE16(_p, _val) \
-  (absl::base_internal::UnalignedStore16(_p, _val))
-#define ABSL_INTERNAL_UNALIGNED_STORE32(_p, _val) \
-  (absl::base_internal::UnalignedStore32(_p, _val))
-#define ABSL_INTERNAL_UNALIGNED_STORE64(_p, _val) \
-  (absl::base_internal::UnalignedStore64(_p, _val))
-
-#elif defined(__x86_64__) || defined(_M_X64) || defined(__i386) || \
-    defined(_M_IX86) || defined(__ppc__) || defined(__PPC__) ||    \
-    defined(__ppc64__) || defined(__PPC64__)
-
-// x86 and x86-64 can perform unaligned loads/stores directly;
-// modern PowerPC hardware can also do unaligned integer loads and stores;
-// but note: the FPU still sends unaligned loads and stores to a trap handler!
-
-#define ABSL_INTERNAL_UNALIGNED_LOAD16(_p) \
-  (*reinterpret_cast<const uint16_t *>(_p))
-#define ABSL_INTERNAL_UNALIGNED_LOAD32(_p) \
-  (*reinterpret_cast<const uint32_t *>(_p))
-#define ABSL_INTERNAL_UNALIGNED_LOAD64(_p) \
-  (*reinterpret_cast<const uint64_t *>(_p))
-
-#define ABSL_INTERNAL_UNALIGNED_STORE16(_p, _val) \
-  (*reinterpret_cast<uint16_t *>(_p) = (_val))
-#define ABSL_INTERNAL_UNALIGNED_STORE32(_p, _val) \
-  (*reinterpret_cast<uint32_t *>(_p) = (_val))
-#define ABSL_INTERNAL_UNALIGNED_STORE64(_p, _val) \
-  (*reinterpret_cast<uint64_t *>(_p) = (_val))
-
-#elif defined(__arm__) && \
-      !defined(__ARM_ARCH_5__) && \
-      !defined(__ARM_ARCH_5T__) && \
-      !defined(__ARM_ARCH_5TE__) && \
-      !defined(__ARM_ARCH_5TEJ__) && \
-      !defined(__ARM_ARCH_6__) && \
-      !defined(__ARM_ARCH_6J__) && \
-      !defined(__ARM_ARCH_6K__) && \
-      !defined(__ARM_ARCH_6Z__) && \
-      !defined(__ARM_ARCH_6ZK__) && \
-      !defined(__ARM_ARCH_6T2__)
-
-
-// ARMv7 and newer support native unaligned accesses, but only of 16-bit
-// and 32-bit values (not 64-bit); older versions either raise a fatal signal,
-// do an unaligned read and rotate the words around a bit, or do the reads very
-// slowly (trip through kernel mode). There's no simple #define that says just
-// "ARMv7 or higher", so we have to filter away all ARMv5 and ARMv6
-// sub-architectures. Newer gcc (>= 4.6) set an __ARM_FEATURE_ALIGNED #define,
-// so in time, maybe we can move on to that.
-//
-// This is a mess, but there's not much we can do about it.
-//
-// To further complicate matters, only LDR instructions (single reads) are
-// allowed to be unaligned, not LDRD (two reads) or LDM (many reads). Unless we
-// explicitly tell the compiler that these accesses can be unaligned, it can and
-// will combine accesses. On armcc, the way to signal this is done by accessing
-// through the type (uint32_t __packed *), but GCC has no such attribute
-// (it ignores __attribute__((packed)) on individual variables). However,
-// we can tell it that a _struct_ is unaligned, which has the same effect,
-// so we do that.
-
-namespace absl {
-namespace base_internal {
-
-struct Unaligned16Struct {
-  uint16_t value;
-  uint8_t dummy;  // To make the size non-power-of-two.
-} ABSL_ATTRIBUTE_PACKED;
-
-struct Unaligned32Struct {
-  uint32_t value;
-  uint8_t dummy;  // To make the size non-power-of-two.
-} ABSL_ATTRIBUTE_PACKED;
-
-}  // namespace base_internal
-}  // namespace absl
-
-#define ABSL_INTERNAL_UNALIGNED_LOAD16(_p)                                  \
-  ((reinterpret_cast<const ::absl::base_internal::Unaligned16Struct *>(_p)) \
-       ->value)
-#define ABSL_INTERNAL_UNALIGNED_LOAD32(_p)                                  \
-  ((reinterpret_cast<const ::absl::base_internal::Unaligned32Struct *>(_p)) \
-       ->value)
-
-#define ABSL_INTERNAL_UNALIGNED_STORE16(_p, _val)                      \
-  ((reinterpret_cast< ::absl::base_internal::Unaligned16Struct *>(_p)) \
-       ->value = (_val))
-#define ABSL_INTERNAL_UNALIGNED_STORE32(_p, _val)                      \
-  ((reinterpret_cast< ::absl::base_internal::Unaligned32Struct *>(_p)) \
-       ->value = (_val))
-
-namespace absl {
-namespace base_internal {
-
-inline uint64_t UnalignedLoad64(const void *p) {
-  uint64_t t;
-  memcpy(&t, p, sizeof t);
-  return t;
-}
-
-inline void UnalignedStore64(void *p, uint64_t v) { memcpy(p, &v, sizeof v); }
-
-}  // namespace base_internal
-}  // namespace absl
-
-#define ABSL_INTERNAL_UNALIGNED_LOAD64(_p) \
-  (absl::base_internal::UnalignedLoad64(_p))
-#define ABSL_INTERNAL_UNALIGNED_STORE64(_p, _val) \
-  (absl::base_internal::UnalignedStore64(_p, _val))
-
 #else
 
-// ABSL_INTERNAL_NEED_ALIGNED_LOADS is defined when the underlying platform
-// doesn't support unaligned access.
-#define ABSL_INTERNAL_NEED_ALIGNED_LOADS
-
-// These functions are provided for architectures that don't support
-// unaligned loads and stores.
-
 namespace absl {
 namespace base_internal {
 
diff --git a/absl/container/inlined_vector.h b/absl/container/inlined_vector.h
index 1ab4b2f44741..978d5033de93 100644
--- a/absl/container/inlined_vector.h
+++ b/absl/container/inlined_vector.h
@@ -73,17 +73,12 @@ class InlinedVector {
   using AllocatorTraits = typename Storage::AllocatorTraits;
 
   template <typename Iterator>
-  using IsAtLeastForwardIterator = std::is_convertible<
-      typename std::iterator_traits<Iterator>::iterator_category,
-      std::forward_iterator_tag>;
+  using EnableIfAtLeastForwardIterator = absl::enable_if_t<
+      inlined_vector_internal::IsAtLeastForwardIterator<Iterator>::value>;
 
   template <typename Iterator>
-  using EnableIfAtLeastForwardIterator =
-      absl::enable_if_t<IsAtLeastForwardIterator<Iterator>::value>;
-
-  template <typename Iterator>
-  using DisableIfAtLeastForwardIterator =
-      absl::enable_if_t<!IsAtLeastForwardIterator<Iterator>::value>;
+  using DisableIfAtLeastForwardIterator = absl::enable_if_t<
+      !inlined_vector_internal::IsAtLeastForwardIterator<Iterator>::value>;
 
   using rvalue_reference = typename Storage::rvalue_reference;
 
@@ -1060,7 +1055,9 @@ class InlinedVector {
 
   template <typename ForwardIt>
   void AssignForwardRange(ForwardIt first, ForwardIt last) {
-    static_assert(IsAtLeastForwardIterator<ForwardIt>::value, "");
+    static_assert(absl::inlined_vector_internal::IsAtLeastForwardIterator<
+                      ForwardIt>::value,
+                  "");
 
     auto length = std::distance(first, last);
 
@@ -1084,7 +1081,9 @@ class InlinedVector {
 
   template <typename ForwardIt>
   void AppendForwardRange(ForwardIt first, ForwardIt last) {
-    static_assert(IsAtLeastForwardIterator<ForwardIt>::value, "");
+    static_assert(absl::inlined_vector_internal::IsAtLeastForwardIterator<
+                      ForwardIt>::value,
+                  "");
 
     auto length = std::distance(first, last);
     reserve(size() + length);
@@ -1113,7 +1112,9 @@ class InlinedVector {
   template <typename ForwardIt>
   iterator InsertWithForwardRange(const_iterator position, ForwardIt first,
                                   ForwardIt last) {
-    static_assert(IsAtLeastForwardIterator<ForwardIt>::value, "");
+    static_assert(absl::inlined_vector_internal::IsAtLeastForwardIterator<
+                      ForwardIt>::value,
+                  "");
     assert(position >= begin() && position <= end());
 
     if (ABSL_PREDICT_FALSE(first == last))
diff --git a/absl/container/internal/inlined_vector.h b/absl/container/internal/inlined_vector.h
index 168d5b4e1d34..b8b4f4ce3b39 100644
--- a/absl/container/internal/inlined_vector.h
+++ b/absl/container/internal/inlined_vector.h
@@ -26,6 +26,11 @@
 namespace absl {
 namespace inlined_vector_internal {
 
+template <typename Iterator>
+using IsAtLeastForwardIterator = std::is_convertible<
+    typename std::iterator_traits<Iterator>::iterator_category,
+    std::forward_iterator_tag>;
+
 template <typename InlinedVector>
 class Storage;
 
@@ -89,9 +94,7 @@ class Storage<InlinedVector<T, N, A>> {
 
   void AddSize(size_type count) { GetSizeAndIsAllocated() += count << 1; }
 
-  void SetAllocatedData(pointer data) {
-    data_.allocated.allocated_data = data;
-  }
+  void SetAllocatedData(pointer data) { data_.allocated.allocated_data = data; }
 
   void SetAllocatedCapacity(size_type capacity) {
     data_.allocated.allocated_capacity = capacity;
diff --git a/absl/hash/BUILD.bazel b/absl/hash/BUILD.bazel
index 8c2daf703bcd..e1e6eae8516d 100644
--- a/absl/hash/BUILD.bazel
+++ b/absl/hash/BUILD.bazel
@@ -35,10 +35,10 @@ cc_library(
     copts = ABSL_DEFAULT_COPTS,
     linkopts = ABSL_DEFAULT_LINKOPTS,
     deps = [
-        ":city",
         "//absl/base:core_headers",
         "//absl/base:endian",
         "//absl/container:fixed_array",
+        "//absl/hash:city",
         "//absl/meta:type_traits",
         "//absl/numeric:int128",
         "//absl/strings",
diff --git a/absl/strings/charconv.h b/absl/strings/charconv.h
index 59f74bf09b93..3f5891ba7f3f 100644
--- a/absl/strings/charconv.h
+++ b/absl/strings/charconv.h
@@ -49,9 +49,9 @@ struct from_chars_result {
 // this only supports the `double` and `float` types.
 //
 // This interface incorporates the proposed resolutions for library issues
-// DR 3800 and DR 3801.  If these are adopted with different wording,
+// DR 3080 and DR 3081.  If these are adopted with different wording,
 // Abseil's behavior will change to match the standard.  (The behavior most
-// likely to change is for DR 3801, which says what `value` will be set to in
+// likely to change is for DR 3081, which says what `value` will be set to in
 // the case of overflow and underflow.  Code that wants to avoid possible
 // breaking changes in this area should not depend on `value` when the returned
 // from_chars_result indicates a range error.)
diff --git a/absl/strings/internal/str_format/bind_test.cc b/absl/strings/internal/str_format/bind_test.cc
index ba6470ec4a36..2574801a5c38 100644
--- a/absl/strings/internal/str_format/bind_test.cc
+++ b/absl/strings/internal/str_format/bind_test.cc
@@ -9,16 +9,11 @@ namespace absl {
 namespace str_format_internal {
 namespace {
 
-template <typename T, size_t N>
-size_t ArraySize(T (&)[N]) {
-  return N;
-}
-
 class FormatBindTest : public ::testing::Test {
  public:
   bool Extract(const char *s, UnboundConversion *props, int *next) const {
-    absl::string_view src = s;
-    return ConsumeUnboundConversion(&src, props, next) && src.empty();
+    return ConsumeUnboundConversion(s, s + strlen(s), props, next) ==
+           s + strlen(s);
   }
 };
 
diff --git a/absl/strings/internal/str_format/parser.cc b/absl/strings/internal/str_format/parser.cc
index 10487f23c855..9ef5615cc9ee 100644
--- a/absl/strings/internal/str_format/parser.cc
+++ b/absl/strings/internal/str_format/parser.cc
@@ -15,6 +15,44 @@
 
 namespace absl {
 namespace str_format_internal {
+
+using CC = ConversionChar::Id;
+using LM = LengthMod::Id;
+ABSL_CONST_INIT const ConvTag kTags[256] = {
+    {},    {},    {},    {},    {},    {},    {},    {},     // 00-07
+    {},    {},    {},    {},    {},    {},    {},    {},     // 08-0f
+    {},    {},    {},    {},    {},    {},    {},    {},     // 10-17
+    {},    {},    {},    {},    {},    {},    {},    {},     // 18-1f
+    {},    {},    {},    {},    {},    {},    {},    {},     // 20-27
+    {},    {},    {},    {},    {},    {},    {},    {},     // 28-2f
+    {},    {},    {},    {},    {},    {},    {},    {},     // 30-37
+    {},    {},    {},    {},    {},    {},    {},    {},     // 38-3f
+    {},    CC::A, {},    CC::C, {},    CC::E, CC::F, CC::G,  // @ABCDEFG
+    {},    {},    {},    {},    LM::L, {},    {},    {},     // HIJKLMNO
+    {},    {},    {},    CC::S, {},    {},    {},    {},     // PQRSTUVW
+    CC::X, {},    {},    {},    {},    {},    {},    {},     // XYZ[\]^_
+    {},    CC::a, {},    CC::c, CC::d, CC::e, CC::f, CC::g,  // `abcdefg
+    LM::h, CC::i, LM::j, {},    LM::l, {},    CC::n, CC::o,  // hijklmno
+    CC::p, LM::q, {},    CC::s, LM::t, CC::u, {},    {},     // pqrstuvw
+    CC::x, {},    LM::z, {},    {},    {},    {},    {},     // xyz{|}!
+    {},    {},    {},    {},    {},    {},    {},    {},     // 80-87
+    {},    {},    {},    {},    {},    {},    {},    {},     // 88-8f
+    {},    {},    {},    {},    {},    {},    {},    {},     // 90-97
+    {},    {},    {},    {},    {},    {},    {},    {},     // 98-9f
+    {},    {},    {},    {},    {},    {},    {},    {},     // a0-a7
+    {},    {},    {},    {},    {},    {},    {},    {},     // a8-af
+    {},    {},    {},    {},    {},    {},    {},    {},     // b0-b7
+    {},    {},    {},    {},    {},    {},    {},    {},     // b8-bf
+    {},    {},    {},    {},    {},    {},    {},    {},     // c0-c7
+    {},    {},    {},    {},    {},    {},    {},    {},     // c8-cf
+    {},    {},    {},    {},    {},    {},    {},    {},     // d0-d7
+    {},    {},    {},    {},    {},    {},    {},    {},     // d8-df
+    {},    {},    {},    {},    {},    {},    {},    {},     // e0-e7
+    {},    {},    {},    {},    {},    {},    {},    {},     // e8-ef
+    {},    {},    {},    {},    {},    {},    {},    {},     // f0-f7
+    {},    {},    {},    {},    {},    {},    {},    {},     // f8-ff
+};
+
 namespace {
 
 bool CheckFastPathSetting(const UnboundConversion& conv) {
@@ -36,60 +74,17 @@ bool CheckFastPathSetting(const UnboundConversion& conv) {
   return should_be_basic == conv.flags.basic;
 }
 
-// Keep a single table for all the conversion chars and length modifiers.
-// We invert the length modifiers to make them negative so that we can easily
-// test for them.
-// Everything else is `none`, which is a negative constant.
-using CC = ConversionChar::Id;
-using LM = LengthMod::Id;
-static constexpr std::int8_t none = -128;
-static constexpr std::int8_t kIds[] = {
-    none,   none,   none,   none,  none,   none,  none,  none,   // 00-07
-    none,   none,   none,   none,  none,   none,  none,  none,   // 08-0f
-    none,   none,   none,   none,  none,   none,  none,  none,   // 10-17
-    none,   none,   none,   none,  none,   none,  none,  none,   // 18-1f
-    none,   none,   none,   none,  none,   none,  none,  none,   // 20-27
-    none,   none,   none,   none,  none,   none,  none,  none,   // 28-2f
-    none,   none,   none,   none,  none,   none,  none,  none,   // 30-37
-    none,   none,   none,   none,  none,   none,  none,  none,   // 38-3f
-    none,   CC::A,  none,   CC::C, none,   CC::E, CC::F, CC::G,  // @ABCDEFG
-    none,   none,   none,   none,  ~LM::L, none,  none,  none,   // HIJKLMNO
-    none,   none,   none,   CC::S, none,   none,  none,  none,   // PQRSTUVW
-    CC::X,  none,   none,   none,  none,   none,  none,  none,   // XYZ[\]^_
-    none,   CC::a,  none,   CC::c, CC::d,  CC::e, CC::f, CC::g,  // `abcdefg
-    ~LM::h, CC::i,  ~LM::j, none,  ~LM::l, none,  CC::n, CC::o,  // hijklmno
-    CC::p,  ~LM::q, none,   CC::s, ~LM::t, CC::u, none,  none,   // pqrstuvw
-    CC::x,  none,   ~LM::z, none,  none,   none,  none,  none,   // xyz{|}~!
-    none,   none,   none,   none,  none,   none,  none,  none,   // 80-87
-    none,   none,   none,   none,  none,   none,  none,  none,   // 88-8f
-    none,   none,   none,   none,  none,   none,  none,  none,   // 90-97
-    none,   none,   none,   none,  none,   none,  none,  none,   // 98-9f
-    none,   none,   none,   none,  none,   none,  none,  none,   // a0-a7
-    none,   none,   none,   none,  none,   none,  none,  none,   // a8-af
-    none,   none,   none,   none,  none,   none,  none,  none,   // b0-b7
-    none,   none,   none,   none,  none,   none,  none,  none,   // b8-bf
-    none,   none,   none,   none,  none,   none,  none,  none,   // c0-c7
-    none,   none,   none,   none,  none,   none,  none,  none,   // c8-cf
-    none,   none,   none,   none,  none,   none,  none,  none,   // d0-d7
-    none,   none,   none,   none,  none,   none,  none,  none,   // d8-df
-    none,   none,   none,   none,  none,   none,  none,  none,   // e0-e7
-    none,   none,   none,   none,  none,   none,  none,  none,   // e8-ef
-    none,   none,   none,   none,  none,   none,  none,  none,   // f0-f7
-    none,   none,   none,   none,  none,   none,  none,  none,   // f8-ff
-};
-
 template <bool is_positional>
-bool ConsumeConversion(string_view *src, UnboundConversion *conv,
-                       int *next_arg) {
-  const char *pos = src->data();
-  const char *const end = pos + src->size();
+const char *ConsumeConversion(const char *pos, const char *const end,
+                              UnboundConversion *conv, int *next_arg) {
+  const char* const original_pos = pos;
   char c;
   // Read the next char into `c` and update `pos`. Returns false if there are
   // no more chars to read.
-#define ABSL_FORMAT_PARSER_INTERNAL_GET_CHAR()        \
-  do {                                                \
-    if (ABSL_PREDICT_FALSE(pos == end)) return false; \
-    c = *pos++;                                       \
+#define ABSL_FORMAT_PARSER_INTERNAL_GET_CHAR()          \
+  do {                                                  \
+    if (ABSL_PREDICT_FALSE(pos == end)) return nullptr; \
+    c = *pos++;                                         \
   } while (0)
 
   const auto parse_digits = [&] {
@@ -111,10 +106,10 @@ bool ConsumeConversion(string_view *src, UnboundConversion *conv,
 
   if (is_positional) {
     ABSL_FORMAT_PARSER_INTERNAL_GET_CHAR();
-    if (ABSL_PREDICT_FALSE(c < '1' || c > '9')) return false;
+    if (ABSL_PREDICT_FALSE(c < '1' || c > '9')) return nullptr;
     conv->arg_position = parse_digits();
     assert(conv->arg_position > 0);
-    if (ABSL_PREDICT_FALSE(c != '$')) return false;
+    if (ABSL_PREDICT_FALSE(c != '$')) return nullptr;
   }
 
   ABSL_FORMAT_PARSER_INTERNAL_GET_CHAR();
@@ -129,10 +124,9 @@ bool ConsumeConversion(string_view *src, UnboundConversion *conv,
     conv->flags.basic = false;
 
     for (; c <= '0';) {
-      // FIXME: We might be able to speed this up reusing the kIds lookup table
-      // from above.
-      // It might require changing Flags to be a plain integer where we can |= a
-      // value.
+      // FIXME: We might be able to speed this up reusing the lookup table from
+      // above. It might require changing Flags to be a plain integer where we
+      // can |= a value.
       switch (c) {
         case '-':
           conv->flags.left = true;
@@ -160,20 +154,20 @@ flags_done:
       if (c >= '0') {
         int maybe_width = parse_digits();
         if (!is_positional && c == '$') {
-          if (ABSL_PREDICT_FALSE(*next_arg != 0)) return false;
+          if (ABSL_PREDICT_FALSE(*next_arg != 0)) return nullptr;
           // Positional conversion.
           *next_arg = -1;
           conv->flags = Flags();
           conv->flags.basic = true;
-          return ConsumeConversion<true>(src, conv, next_arg);
+          return ConsumeConversion<true>(original_pos, end, conv, next_arg);
         }
         conv->width.set_value(maybe_width);
       } else if (c == '*') {
         ABSL_FORMAT_PARSER_INTERNAL_GET_CHAR();
         if (is_positional) {
-          if (ABSL_PREDICT_FALSE(c < '1' || c > '9')) return false;
+          if (ABSL_PREDICT_FALSE(c < '1' || c > '9')) return nullptr;
           conv->width.set_from_arg(parse_digits());
-          if (ABSL_PREDICT_FALSE(c != '$')) return false;
+          if (ABSL_PREDICT_FALSE(c != '$')) return nullptr;
           ABSL_FORMAT_PARSER_INTERNAL_GET_CHAR();
         } else {
           conv->width.set_from_arg(++*next_arg);
@@ -188,9 +182,9 @@ flags_done:
       } else if (c == '*') {
         ABSL_FORMAT_PARSER_INTERNAL_GET_CHAR();
         if (is_positional) {
-          if (ABSL_PREDICT_FALSE(c < '1' || c > '9')) return false;
+          if (ABSL_PREDICT_FALSE(c < '1' || c > '9')) return nullptr;
           conv->precision.set_from_arg(parse_digits());
-          if (c != '$') return false;
+          if (c != '$') return nullptr;
           ABSL_FORMAT_PARSER_INTERNAL_GET_CHAR();
         } else {
           conv->precision.set_from_arg(++*next_arg);
@@ -201,14 +195,14 @@ flags_done:
     }
   }
 
-  std::int8_t id = kIds[static_cast<unsigned char>(c)];
+  auto tag = GetTagForChar(c);
 
-  if (id < 0) {
-    if (ABSL_PREDICT_FALSE(id == none)) return false;
+  if (ABSL_PREDICT_FALSE(!tag.is_conv())) {
+    if (ABSL_PREDICT_FALSE(!tag.is_length())) return nullptr;
 
     // It is a length modifier.
     using str_format_internal::LengthMod;
-    LengthMod length_mod = LengthMod::FromId(static_cast<LM>(~id));
+    LengthMod length_mod = tag.as_length();
     ABSL_FORMAT_PARSER_INTERNAL_GET_CHAR();
     if (c == 'h' && length_mod.id() == LengthMod::h) {
       conv->length_mod = LengthMod::FromId(LengthMod::hh);
@@ -219,25 +213,24 @@ flags_done:
     } else {
       conv->length_mod = length_mod;
     }
-    id = kIds[static_cast<unsigned char>(c)];
-    if (ABSL_PREDICT_FALSE(id < 0)) return false;
+    tag = GetTagForChar(c);
+    if (ABSL_PREDICT_FALSE(!tag.is_conv())) return nullptr;
   }
 
   assert(CheckFastPathSetting(*conv));
   (void)(&CheckFastPathSetting);
 
-  conv->conv = ConversionChar::FromId(static_cast<CC>(id));
+  conv->conv = tag.as_conv();
   if (!is_positional) conv->arg_position = ++*next_arg;
-  *src = string_view(pos, end - pos);
-  return true;
+  return pos;
 }
 
 }  // namespace
 
-bool ConsumeUnboundConversion(string_view *src, UnboundConversion *conv,
-                              int *next_arg) {
-  if (*next_arg < 0) return ConsumeConversion<true>(src, conv, next_arg);
-  return ConsumeConversion<false>(src, conv, next_arg);
+const char *ConsumeUnboundConversion(const char *p, const char *end,
+                                     UnboundConversion *conv, int *next_arg) {
+  if (*next_arg < 0) return ConsumeConversion<true>(p, end, conv, next_arg);
+  return ConsumeConversion<false>(p, end, conv, next_arg);
 }
 
 struct ParsedFormatBase::ParsedFormatConsumer {
diff --git a/absl/strings/internal/str_format/parser.h b/absl/strings/internal/str_format/parser.h
index 1022f06297a7..b7614a09fa4e 100644
--- a/absl/strings/internal/str_format/parser.h
+++ b/absl/strings/internal/str_format/parser.h
@@ -63,17 +63,45 @@ struct UnboundConversion {
   ConversionChar conv;
 };
 
-// Consume conversion spec prefix (not including '%') of '*src' if valid.
+// Consume conversion spec prefix (not including '%') of [p, end) if valid.
 // Examples of valid specs would be e.g.: "s", "d", "-12.6f".
-// If valid, the front of src is advanced such that src becomes the
-// part following the conversion spec, and the spec part is broken down and
-// returned in 'conv'.
-// If invalid, returns false and leaves 'src' unmodified.
-// For example:
-//   Given "d9", returns "d", and leaves src="9",
-//   Given "!", returns "" and leaves src="!".
-bool ConsumeUnboundConversion(string_view* src, UnboundConversion* conv,
-                              int* next_arg);
+// If valid, it returns the first character following the conversion spec,
+// and the spec part is broken down and returned in 'conv'.
+// If invalid, returns nullptr.
+const char* ConsumeUnboundConversion(const char* p, const char* end,
+                                     UnboundConversion* conv, int* next_arg);
+
+// Helper tag class for the table below.
+// It allows fast `char -> ConversionChar/LengthMod` checking and conversions.
+class ConvTag {
+ public:
+  constexpr ConvTag(ConversionChar::Id id) : tag_(id) {}  // NOLINT
+  // We invert the length modifiers to make them negative so that we can easily
+  // test for them.
+  constexpr ConvTag(LengthMod::Id id) : tag_(~id) {}  // NOLINT
+  // Everything else is -128, which is negative to make is_conv() simpler.
+  constexpr ConvTag() : tag_(-128) {}
+
+  bool is_conv() const { return tag_ >= 0; }
+  bool is_length() const { return tag_ < 0 && tag_ != -128; }
+  ConversionChar as_conv() const {
+    assert(is_conv());
+    return ConversionChar::FromId(static_cast<ConversionChar::Id>(tag_));
+  }
+  LengthMod as_length() const {
+    assert(is_length());
+    return LengthMod::FromId(static_cast<LengthMod::Id>(~tag_));
+  }
+
+ private:
+  std::int8_t tag_;
+};
+
+extern const ConvTag kTags[256];
+// Keep a single table for all the conversion chars and length modifiers.
+inline ConvTag GetTagForChar(char c) {
+  return kTags[static_cast<unsigned char>(c)];
+}
 
 // Parse the format string provided in 'src' and pass the identified items into
 // 'consumer'.
@@ -88,51 +116,53 @@ bool ConsumeUnboundConversion(string_view* src, UnboundConversion* conv,
 template <typename Consumer>
 bool ParseFormatString(string_view src, Consumer consumer) {
   int next_arg = 0;
-  while (!src.empty()) {
-    const char* percent =
-        static_cast<const char*>(memchr(src.data(), '%', src.size()));
+  const char* p = src.data();
+  const char* const end = p + src.size();
+  while (p != end) {
+    const char* percent = static_cast<const char*>(memchr(p, '%', end - p));
     if (!percent) {
       // We found the last substring.
-      return consumer.Append(src);
+      return consumer.Append(string_view(p, end - p));
     }
     // We found a percent, so push the text run then process the percent.
-    size_t percent_loc = percent - src.data();
-    if (!consumer.Append(string_view(src.data(), percent_loc))) return false;
-    if (percent + 1 >= src.data() + src.size()) return false;
-
-    UnboundConversion conv;
-
-    switch (percent[1]) {
-      case '%':
-        if (!consumer.Append("%")) return false;
-        src.remove_prefix(percent_loc + 2);
-        continue;
-
-#define PARSER_CASE(ch)                                     \
-  case #ch[0]:                                              \
-    src.remove_prefix(percent_loc + 2);                     \
-    conv.conv = ConversionChar::FromId(ConversionChar::ch); \
-    conv.arg_position = ++next_arg;                         \
-    break;
-        ABSL_CONVERSION_CHARS_EXPAND_(PARSER_CASE, );
-#undef PARSER_CASE
-
-      default:
-        src.remove_prefix(percent_loc + 1);
-        if (!ConsumeUnboundConversion(&src, &conv, &next_arg)) return false;
-        break;
-    }
-    if (next_arg == 0) {
-      // This indicates an error in the format std::string.
-      // The only way to get next_arg == 0 is to have a positional argument
-      // first which sets next_arg to -1 and then a non-positional argument
-      // which does ++next_arg.
-      // Checking here seems to be the cheapeast place to do it.
+    if (ABSL_PREDICT_FALSE(!consumer.Append(string_view(p, percent - p)))) {
       return false;
     }
-    if (!consumer.ConvertOne(
-            conv, string_view(percent + 1, src.data() - (percent + 1)))) {
-      return false;
+    if (ABSL_PREDICT_FALSE(percent + 1 >= end)) return false;
+
+    auto tag = GetTagForChar(percent[1]);
+    if (tag.is_conv()) {
+      if (ABSL_PREDICT_FALSE(next_arg < 0)) {
+        // This indicates an error in the format std::string.
+        // The only way to get `next_arg < 0` here is to have a positional
+        // argument first which sets next_arg to -1 and then a non-positional
+        // argument.
+        return false;
+      }
+      p = percent + 2;
+
+      // Keep this case separate from the one below.
+      // ConvertOne is more efficient when the compiler can see that the `basic`
+      // flag is set.
+      UnboundConversion conv;
+      conv.conv = tag.as_conv();
+      conv.arg_position = ++next_arg;
+      if (ABSL_PREDICT_FALSE(
+              !consumer.ConvertOne(conv, string_view(percent + 1, 1)))) {
+        return false;
+      }
+    } else if (percent[1] != '%') {
+      UnboundConversion conv;
+      p = ConsumeUnboundConversion(percent + 1, end, &conv, &next_arg);
+      if (ABSL_PREDICT_FALSE(p == nullptr)) return false;
+      if (ABSL_PREDICT_FALSE(!consumer.ConvertOne(
+          conv, string_view(percent + 1, p - (percent + 1))))) {
+        return false;
+      }
+    } else {
+      if (ABSL_PREDICT_FALSE(!consumer.Append("%"))) return false;
+      p = percent + 2;
+      continue;
     }
   }
   return true;
diff --git a/absl/strings/internal/str_format/parser_test.cc b/absl/strings/internal/str_format/parser_test.cc
index ff7057534f43..6d356093ce60 100644
--- a/absl/strings/internal/str_format/parser_test.cc
+++ b/absl/strings/internal/str_format/parser_test.cc
@@ -1,6 +1,8 @@
 #include "absl/strings/internal/str_format/parser.h"
 
 #include <string.h>
+
+#include "gmock/gmock.h"
 #include "gtest/gtest.h"
 #include "absl/base/macros.h"
 
@@ -9,6 +11,8 @@ namespace str_format_internal {
 
 namespace {
 
+using testing::Pair;
+
 TEST(LengthModTest, Names) {
   struct Expectation {
     int line;
@@ -63,20 +67,21 @@ TEST(ConversionCharTest, Names) {
 
 class ConsumeUnboundConversionTest : public ::testing::Test {
  public:
-  typedef UnboundConversion Props;
-  string_view Consume(string_view* src) {
+  std::pair<string_view, string_view> Consume(string_view src) {
     int next = 0;
-    const char* prev_begin = src->data();
     o = UnboundConversion();  // refresh
-    ConsumeUnboundConversion(src, &o, &next);
-    return {prev_begin, static_cast<size_t>(src->data() - prev_begin)};
+    const char* p = ConsumeUnboundConversion(
+        src.data(), src.data() + src.size(), &o, &next);
+    if (!p) return {{}, src};
+    return {string_view(src.data(), p - src.data()),
+            string_view(p, src.data() + src.size() - p)};
   }
 
   bool Run(const char *fmt, bool force_positional = false) {
-    string_view src = fmt;
     int next = force_positional ? -1 : 0;
     o = UnboundConversion();  // refresh
-    return ConsumeUnboundConversion(&src, &o, &next) && src.empty();
+    return ConsumeUnboundConversion(fmt, fmt + strlen(fmt), &o, &next) ==
+           fmt + strlen(fmt);
   }
   UnboundConversion o;
 };
@@ -104,11 +109,7 @@ TEST_F(ConsumeUnboundConversionTest, ConsumeSpecification) {
   };
   for (const auto& e : kExpect) {
     SCOPED_TRACE(e.line);
-    string_view src = e.src;
-    EXPECT_EQ(e.src, src);
-    string_view out = Consume(&src);
-    EXPECT_EQ(e.out, out);
-    EXPECT_EQ(e.src_post, src);
+    EXPECT_THAT(Consume(e.src), Pair(e.out, e.src_post));
   }
 }
 
diff --git a/absl/strings/numbers.cc b/absl/strings/numbers.cc
index 558c3396681d..38d148690e2e 100644
--- a/absl/strings/numbers.cc
+++ b/absl/strings/numbers.cc
@@ -54,7 +54,7 @@ bool SimpleAtof(absl::string_view str, float* out) {
     // not all non-whitespace characters consumed
     return false;
   }
-  // from_chars() with DR 3801's current wording will return max() on
+  // from_chars() with DR 3081's current wording will return max() on
   // overflow.  SimpleAtof returns infinity instead.
   if (result.ec == std::errc::result_out_of_range) {
     if (*out > 1.0) {
@@ -80,7 +80,7 @@ bool SimpleAtod(absl::string_view str, double* out) {
     // not all non-whitespace characters consumed
     return false;
   }
-  // from_chars() with DR 3801's current wording will return max() on
+  // from_chars() with DR 3081's current wording will return max() on
   // overflow.  SimpleAtod returns infinity instead.
   if (result.ec == std::errc::result_out_of_range) {
     if (*out > 1.0) {