mirror of
https://github.com/jart/cosmopolitan.git
synced 2025-07-03 09:48:29 +00:00
ctl::string small-string optimization (#1199)
A small-string optimization is a way of reusing inline storage space for sufficiently small strings, rather than allocating them on the heap. The current approach takes after an old Facebook string class: it reuses the highest-order byte for flags and small-string size, in such a way that a maximally-sized small string will have its last byte zeroed, making it a null terminator for the C string. The only flag we have is in the highest-order bit, that says whether the string is big (set) or small (cleared.) Most of the logic switches based on the value of this bit; e.g. data() returns big()->p if it's set, else small()->buf if it's cleared. For a small string, the capacity is always fixed at sizeof(string) - 1 bytes; we store the length in the last byte, but we store it as the number of remaining bytes of capacity, so that at max size, the last byte will read zero and serve as our null terminator. Morally speaking, our class's storage is a union over two POD C structs. For now I gravitated towards a slightly more obtuse approach: the string class itself contains a blob of the right size, and we alias that blob's pointer for the two structs, taking some care not to run afoul of object lifetime rules in C++. If anyone wants to improve on this, contributions are welcome. This commit also introduces the `ctl::__` namespace. It can't be legally spelled by library users, and serves as our version of boost's "detail". We introduced a string::swap function, and we now use that in operator=. operator= now takes its argument by value, so we never need to check for the case where the pointers are equal and can just swap the entire store of the argument with our own, leaving the C++ destructor to free our old storage afterwards. There are probably still a few places where our capacity is slightly off and we grow too fast, although there don't appear to be any where we are too slow. I will leave these to be fixed in future changes.
This commit is contained in:
Jōshin
GitHub
parent
df6b384e31
commit
8b3e368e9a
4 changed files with 333 additions and 139 deletions
247
ctl/string.cc
247
ctl/string.cc
|
|
@ -23,7 +23,9 @@
|
|||
|
||||
namespace ctl {
|
||||
|
||||
string::~string() noexcept
|
||||
namespace __ {
|
||||
|
||||
big_string::~big_string() /* noexcept */
|
||||
{
|
||||
if (n) {
|
||||
if (n >= c)
|
||||
|
|
@ -36,27 +38,36 @@ string::~string() noexcept
|
|||
free(p);
|
||||
}
|
||||
|
||||
string::string(const char* s) noexcept
|
||||
} // namespace __
|
||||
|
||||
string::~string() /* noexcept */
|
||||
{
|
||||
if (isbig()) {
|
||||
big()->~big_string();
|
||||
}
|
||||
}
|
||||
|
||||
string::string(const char* s) noexcept : string()
|
||||
{
|
||||
append(s, strlen(s));
|
||||
}
|
||||
|
||||
string::string(const string& s) noexcept
|
||||
string::string(const string& s) noexcept : string()
|
||||
{
|
||||
append(s.data(), s.size());
|
||||
}
|
||||
|
||||
string::string(const string_view s) noexcept : string()
|
||||
{
|
||||
append(s.p, s.n);
|
||||
}
|
||||
|
||||
string::string(const string_view s) noexcept
|
||||
{
|
||||
append(s.p, s.n);
|
||||
}
|
||||
|
||||
string::string(size_t size, char ch) noexcept
|
||||
string::string(size_t size, char ch) noexcept : string()
|
||||
{
|
||||
resize(size, ch);
|
||||
}
|
||||
|
||||
string::string(const char* s, size_t size) noexcept
|
||||
string::string(const char* s, size_t size) noexcept : string()
|
||||
{
|
||||
append(s, size);
|
||||
}
|
||||
|
|
@ -64,29 +75,39 @@ string::string(const char* s, size_t size) noexcept
|
|||
const char*
|
||||
string::c_str() const noexcept
|
||||
{
|
||||
if (!n)
|
||||
if (!size())
|
||||
return "";
|
||||
if (n >= c)
|
||||
if (size() >= capacity())
|
||||
__builtin_trap();
|
||||
if (p[n])
|
||||
if (data()[size()])
|
||||
__builtin_trap();
|
||||
return p;
|
||||
return data();
|
||||
}
|
||||
|
||||
void
|
||||
string::reserve(size_t c2) noexcept
|
||||
{
|
||||
char* p2;
|
||||
if (c2 < n)
|
||||
c2 = n;
|
||||
size_t n = size();
|
||||
if (c2 < n + 1)
|
||||
c2 = n + 1;
|
||||
if (c2 <= __::string_size)
|
||||
return;
|
||||
if (ckd_add(&c2, c2, 15))
|
||||
__builtin_trap();
|
||||
c2 &= -16;
|
||||
if (!(p2 = (char*)realloc(p, c2)))
|
||||
__builtin_trap();
|
||||
if (!isbig()) {
|
||||
if (!(p2 = (char*)malloc(c2)))
|
||||
__builtin_trap();
|
||||
__builtin_memcpy(p2, data(), size() + 1);
|
||||
} else {
|
||||
if (!(p2 = (char*)realloc(big()->p, c2)))
|
||||
__builtin_trap();
|
||||
}
|
||||
std::atomic_signal_fence(std::memory_order_seq_cst);
|
||||
c = c2;
|
||||
p = p2;
|
||||
set_big_capacity(c2);
|
||||
big()->n = n;
|
||||
big()->p = p2;
|
||||
}
|
||||
|
||||
void
|
||||
|
|
@ -96,41 +117,54 @@ string::resize(size_t n2, char ch) noexcept
|
|||
if (ckd_add(&c2, n2, 1))
|
||||
__builtin_trap();
|
||||
reserve(c2);
|
||||
if (n2 > n)
|
||||
memset(p + n, ch, n2 - n);
|
||||
p[n = n2] = 0;
|
||||
if (n2 > size())
|
||||
memset(data() + size(), ch, n2 - size());
|
||||
if (isbig()) {
|
||||
big()->p[big()->n = n2] = 0;
|
||||
} else {
|
||||
set_small_size(n2);
|
||||
data()[size()] = 0;
|
||||
}
|
||||
}
|
||||
|
||||
void
|
||||
string::append(char ch) noexcept
|
||||
{
|
||||
if (n + 2 > c) {
|
||||
size_t c2 = c + 2;
|
||||
c2 += c2 >> 1;
|
||||
size_t n2;
|
||||
if (ckd_add(&n2, size(), 2))
|
||||
__builtin_trap();
|
||||
if (n2 > capacity()) {
|
||||
size_t c2 = capacity() + 2;
|
||||
if (ckd_add(&c2, c2, c2 >> 1))
|
||||
__builtin_trap();
|
||||
reserve(c2);
|
||||
}
|
||||
p[n++] = ch;
|
||||
p[n] = 0;
|
||||
// XXX do we care to fence this?
|
||||
data()[size()] = ch;
|
||||
data()[size() + 1] = 0;
|
||||
if (isbig()) {
|
||||
++big()->n;
|
||||
} else {
|
||||
--small()->rem;
|
||||
}
|
||||
}
|
||||
|
||||
void
|
||||
string::grow(size_t size) noexcept
|
||||
{
|
||||
size_t need;
|
||||
if (ckd_add(&need, n, size))
|
||||
if (ckd_add(&need, this->size(), size))
|
||||
__builtin_trap();
|
||||
if (ckd_add(&need, need, 1))
|
||||
__builtin_trap();
|
||||
if (need <= c)
|
||||
if (need <= capacity())
|
||||
return;
|
||||
size_t c2 = c;
|
||||
if (!c2) {
|
||||
c2 = need;
|
||||
} else {
|
||||
while (c2 < need)
|
||||
if (ckd_add(&c2, c2, c2 >> 1))
|
||||
__builtin_trap();
|
||||
}
|
||||
size_t c2 = capacity();
|
||||
if (!c2)
|
||||
__builtin_trap();
|
||||
while (c2 < need)
|
||||
if (ckd_add(&c2, c2, c2 >> 1))
|
||||
__builtin_trap();
|
||||
reserve(c2);
|
||||
}
|
||||
|
||||
|
|
@ -139,8 +173,14 @@ string::append(char ch, size_t size) noexcept
|
|||
{
|
||||
grow(size);
|
||||
if (size)
|
||||
memset(p + n, ch, size);
|
||||
p[n += size] = 0;
|
||||
memset(data() + this->size(), ch, size);
|
||||
// XXX fence?
|
||||
if (isbig()) {
|
||||
big()->n += size;
|
||||
} else {
|
||||
small()->rem -= size;
|
||||
}
|
||||
data()[this->size()] = 0;
|
||||
}
|
||||
|
||||
void
|
||||
|
|
@ -148,55 +188,53 @@ string::append(const void* data, size_t size) noexcept
|
|||
{
|
||||
grow(size);
|
||||
if (size)
|
||||
memcpy(p + n, data, size);
|
||||
p[n += size] = 0;
|
||||
memcpy(this->data() + this->size(), data, size);
|
||||
if (isbig()) {
|
||||
big()->n += size;
|
||||
} else {
|
||||
small()->rem -= size;
|
||||
}
|
||||
this->data()[this->size()] = 0;
|
||||
}
|
||||
|
||||
void
|
||||
string::pop_back() noexcept
|
||||
{
|
||||
if (!n)
|
||||
if (!size())
|
||||
__builtin_trap();
|
||||
p[--n] = 0;
|
||||
if (isbig()) {
|
||||
--big()->n;
|
||||
} else {
|
||||
++small()->rem;
|
||||
}
|
||||
data()[size()] = 0;
|
||||
}
|
||||
|
||||
string&
|
||||
string::operator=(string&& s) noexcept
|
||||
string::operator=(string s) noexcept
|
||||
{
|
||||
if (p != s.p) {
|
||||
if (p) {
|
||||
clear();
|
||||
append(s.p, s.n);
|
||||
} else {
|
||||
p = s.p;
|
||||
n = s.n;
|
||||
c = s.c;
|
||||
s.p = nullptr;
|
||||
s.n = 0;
|
||||
s.c = 0;
|
||||
}
|
||||
}
|
||||
swap(s);
|
||||
return *this;
|
||||
}
|
||||
|
||||
bool
|
||||
string::operator==(const string_view s) const noexcept
|
||||
{
|
||||
if (n != s.n)
|
||||
if (size() != s.n)
|
||||
return false;
|
||||
if (!n)
|
||||
if (!s.n)
|
||||
return true;
|
||||
return !memcmp(p, s.p, n);
|
||||
return !memcmp(data(), s.p, s.n);
|
||||
}
|
||||
|
||||
bool
|
||||
string::operator!=(const string_view s) const noexcept
|
||||
{
|
||||
if (n != s.n)
|
||||
if (size() != s.n)
|
||||
return true;
|
||||
if (!n)
|
||||
if (!s.n)
|
||||
return false;
|
||||
return !!memcmp(p, s.p, n);
|
||||
return !!memcmp(data(), s.p, s.n);
|
||||
}
|
||||
|
||||
bool
|
||||
|
|
@ -204,35 +242,35 @@ string::contains(const string_view s) const noexcept
|
|||
{
|
||||
if (!s.n)
|
||||
return true;
|
||||
return !!memmem(p, n, s.p, s.n);
|
||||
return !!memmem(data(), size(), s.p, s.n);
|
||||
}
|
||||
|
||||
bool
|
||||
string::ends_with(const string_view s) const noexcept
|
||||
{
|
||||
if (n < s.n)
|
||||
if (size() < s.n)
|
||||
return false;
|
||||
if (!s.n)
|
||||
return true;
|
||||
return !memcmp(p + n - s.n, s.p, s.n);
|
||||
return !memcmp(data() + size() - s.n, s.p, s.n);
|
||||
}
|
||||
|
||||
bool
|
||||
string::starts_with(const string_view s) const noexcept
|
||||
{
|
||||
if (n < s.n)
|
||||
if (size() < s.n)
|
||||
return false;
|
||||
if (!s.n)
|
||||
return true;
|
||||
return !memcmp(p, s.p, s.n);
|
||||
return !memcmp(data(), s.p, s.n);
|
||||
}
|
||||
|
||||
size_t
|
||||
string::find(char ch, size_t pos) const noexcept
|
||||
{
|
||||
char* q;
|
||||
if ((q = (char*)memchr(p, ch, n)))
|
||||
return q - p;
|
||||
if ((q = (char*)memchr(data(), ch, size())))
|
||||
return q - data();
|
||||
return npos;
|
||||
}
|
||||
|
||||
|
|
@ -240,10 +278,10 @@ size_t
|
|||
string::find(const string_view s, size_t pos) const noexcept
|
||||
{
|
||||
char* q;
|
||||
if (pos > n)
|
||||
if (pos > size())
|
||||
__builtin_trap();
|
||||
if ((q = (char*)memmem(p + pos, n - pos, s.p, s.n)))
|
||||
return q - p;
|
||||
if ((q = (char*)memmem(data() + pos, size() - pos, s.p, s.n)))
|
||||
return q - data();
|
||||
return npos;
|
||||
}
|
||||
|
||||
|
|
@ -251,15 +289,15 @@ string
|
|||
string::substr(size_t pos, size_t count) const noexcept
|
||||
{
|
||||
size_t last;
|
||||
if (pos > n)
|
||||
if (pos > size())
|
||||
__builtin_trap();
|
||||
if (count > n - pos)
|
||||
count = n - pos;
|
||||
if (count > size() - pos)
|
||||
count = size() - pos;
|
||||
if (ckd_add(&last, pos, count))
|
||||
last = n;
|
||||
if (last > n)
|
||||
last = size();
|
||||
if (last > size())
|
||||
__builtin_trap();
|
||||
return string(p + pos, count);
|
||||
return string(data() + pos, count);
|
||||
}
|
||||
|
||||
string&
|
||||
|
|
@ -268,12 +306,12 @@ string::replace(size_t pos, size_t count, const string_view& s) noexcept
|
|||
size_t last;
|
||||
if (ckd_add(&last, pos, count))
|
||||
__builtin_trap();
|
||||
if (last > n)
|
||||
if (last > size())
|
||||
__builtin_trap();
|
||||
size_t need;
|
||||
if (ckd_add(&need, pos, s.n))
|
||||
__builtin_trap();
|
||||
size_t extra = n - last;
|
||||
size_t extra = size() - last;
|
||||
if (ckd_add(&need, need, extra))
|
||||
__builtin_trap();
|
||||
size_t c2;
|
||||
|
|
@ -281,42 +319,57 @@ string::replace(size_t pos, size_t count, const string_view& s) noexcept
|
|||
__builtin_trap();
|
||||
reserve(c2);
|
||||
if (extra)
|
||||
memmove(p + pos + s.n, p + last, extra);
|
||||
memcpy(p + pos, s.p, s.n);
|
||||
p[n = need] = 0;
|
||||
memmove(data() + pos + s.n, data() + last, extra);
|
||||
memcpy(data() + pos, s.p, s.n);
|
||||
if (isbig()) {
|
||||
big()->p[big()->n = need] = 0;
|
||||
} else {
|
||||
set_small_size(need);
|
||||
data()[size()] = 0;
|
||||
}
|
||||
return *this;
|
||||
}
|
||||
|
||||
string&
|
||||
string::insert(size_t i, const string_view s) noexcept
|
||||
{
|
||||
if (i > n)
|
||||
if (i > size())
|
||||
__builtin_trap();
|
||||
size_t extra = n - i;
|
||||
size_t extra = size() - i;
|
||||
size_t need;
|
||||
if (ckd_add(&need, n, s.n))
|
||||
if (ckd_add(&need, size(), s.n))
|
||||
__builtin_trap();
|
||||
if (ckd_add(&need, need, 1))
|
||||
__builtin_trap();
|
||||
reserve(need);
|
||||
if (extra)
|
||||
memmove(p + i + s.n, p + i, extra);
|
||||
memcpy(p + i, s.p, s.n);
|
||||
p[n += s.n] = 0;
|
||||
memmove(data() + i + s.n, data() + i, extra);
|
||||
memcpy(data() + i, s.p, s.n);
|
||||
if (isbig()) {
|
||||
big()->n += s.n;
|
||||
} else {
|
||||
small()->rem -= s.n;
|
||||
}
|
||||
data()[size()] = 0;
|
||||
return *this;
|
||||
}
|
||||
|
||||
string&
|
||||
string::erase(size_t pos, size_t count) noexcept
|
||||
{
|
||||
if (pos > n)
|
||||
if (pos > size())
|
||||
__builtin_trap();
|
||||
if (count > n - pos)
|
||||
count = n - pos;
|
||||
size_t extra = n - (pos + count);
|
||||
if (count > size() - pos)
|
||||
count = size() - pos;
|
||||
size_t extra = size() - (pos + count);
|
||||
if (extra)
|
||||
memmove(p + pos, p + pos + count, extra);
|
||||
p[n = pos + extra] = 0;
|
||||
memmove(data() + pos, data() + pos + count, extra);
|
||||
if (isbig()) {
|
||||
big()->n = pos + extra;
|
||||
} else {
|
||||
set_small_size(pos + extra);
|
||||
}
|
||||
data()[size()] = 0;
|
||||
return *this;
|
||||
}
|
||||
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue