xo-alloc2/src/alloc2/DArena.cpp

594 lines
20 KiB
C++

/** @file DArena.cpp
*
* @author Roland Conybeare, Dec 2025
**/
#include "alloc/AAllocator.hpp"
#include "arena/DArena.hpp"
#include "arena/DArenaIterator.hpp"
#include "xo/alloc2/padding.hpp"
#include "xo/indentlog/scope.hpp"
#include "xo/indentlog/print/tag.hpp"
#include <cassert>
#include <sys/mman.h> // for ::munmap()
#include <unistd.h> // for ::getpagesize()
#include <string.h> // for ::memset()
namespace xo {
using xo::facet::typeseq;
using std::byte;
using std::size_t;
namespace mm {
auto
DArena::map_aligned_range(size_t req_z,
size_t align_z,
bool enable_hugepage_flag) -> range_type
{
scope log(XO_DEBUG(true),
xtag("req_z", req_z), xtag("align_z", align_z));
// 1. round up to multiple of align_z
size_t target_z = padding::with_padding(req_z, align_z); // 4.
// 2. mmap() will give us page-aligned memory,
// but not hugepage-aligned.
//
// Over-request by align_z to ensure
// aligned subrange of size target_z
//
byte * base = (byte *)(::mmap(nullptr,
target_z + align_z,
PROT_NONE,
MAP_PRIVATE | MAP_ANONYMOUS,
-1, 0));
// on mmap success: upper limit of mapped address range
byte * hi = base + (target_z + align_z);
// lowest hugepage-aligned address in [base, hi)
byte * aligned_base = (byte *)(padding::with_padding((size_t)base, align_z));
// end of hugeppage-aligned range starting at aligned_base
byte * aligned_hi = aligned_base + target_z;
log && log("acquired memory [lo,hi) using mmap",
xtag("lo", base),
xtag("aligned_lo", aligned_base),
xtag("req_z", req_z),
xtag("target_z", target_z),
xtag("aligned_hi", aligned_hi),
xtag("hi", hi));
// 3. assess mmap success
{
if (base == MAP_FAILED) {
throw std::runtime_error(tostr("ArenaAlloc: uncommitted allocation failed",
xtag("size", req_z)));
}
assert((size_t)aligned_base % align_z == 0);
assert(aligned_base >= base);
assert(aligned_base < base + align_z);
}
// 4. release unaligned prefix
if (base < aligned_base) {
size_t ua_prefix = aligned_base - base;
::munmap(base, ua_prefix);
}
// 5. release unaligned suffix
if (aligned_hi < hi) {
size_t suffix = hi - aligned_hi;
::munmap(aligned_hi, suffix);
}
if (enable_hugepage_flag) {
#ifdef __linux__
/** linux:
* opt-in to transparent huge pages (THP)
* provided OS configured to support them.
* otherwise fallback gracefully.
*
* Huge pages -> use fewer TLB entries + faster
* shorter path through page table.
*
* When we commit (i.e. obtain physical memory on page fault),
* typically expect to pay ~1us per superpage.
* Much better than ~500us to commit 512 4k VM pages.
*
* But wasted if we don't use the memory.
*
* Page table has a handful of levels
**/
::madvise(aligned_base, target_z, MADV_HUGEPAGE); // 8.
#endif
}
return std::make_pair(aligned_base, aligned_hi);
}
DArena
DArena::map(const ArenaConfig & cfg)
{
scope log(XO_DEBUG(true));
/* vm page size. 4KB, probably */
size_t page_z = getpagesize();
bool enable_hugepage_flag = (cfg.size_ >= cfg.hugepage_z_);
/* Align start of arena memory on this boundary.
* Will use THP (transparent huge pages) if available
* and arena size is at least as large as hugepage size (2MB, probably)
*/
size_t align_z = (enable_hugepage_flag ? cfg.hugepage_z_ : page_z);
log && log(xtag("page_z", page_z),
xtag("align_z", align_z));
auto [lo, hi] = map_aligned_range(cfg.size_,
align_z,
enable_hugepage_flag);
if (!lo) {
// control here implies mmap() failed silently
throw std::runtime_error(tostr("ArenaAlloc: allocation failed",
xtag("size", cfg.size_)));
}
#ifdef NOPE
log && log(xtag("lo", (void*)lo_),
xtag("page_z", page_z_),
xtag("hugepage_z", hugepage_z_));
#endif
return DArena(cfg, page_z, align_z, lo, hi);
} /*map*/
DArena::DArena(const ArenaConfig & cfg,
size_type page_z,
size_type arena_align_z,
byte * lo,
byte * hi) : config_{cfg},
page_z_{page_z},
arena_align_z_{arena_align_z},
lo_{lo},
committed_z_{0},
free_{lo},
limit_{lo},
hi_{hi},
error_count_{0},
last_error_{}
{
//retval.checkpoint_ = lo_;
/** make sure guard size is aligned **/
config_.header_.guard_z_
= padding::with_padding(config_.header_.guard_z_);
}
DArena::DArena(DArena && other) {
config_ = other.config_;
page_z_ = other.page_z_;
arena_align_z_ = other.arena_align_z_;
lo_ = other.lo_;
committed_z_ = other.committed_z_;
free_ = other.free_;
limit_ = other.limit_;
hi_ = other.hi_;
error_count_ = other.error_count_;
last_error_ = other.last_error_;
other.config_ = ArenaConfig();
other.lo_ = nullptr;
other.committed_z_ = 0;
other.free_ = nullptr;
other.limit_ = nullptr;
other.hi_ = nullptr;
other.error_count_ = 0;
other.last_error_ = AllocError();
}
DArena &
DArena::operator=(DArena && other)
{
config_ = other.config_;
page_z_ = other.page_z_;
arena_align_z_ = other.arena_align_z_;
lo_ = other.lo_;
committed_z_ = other.committed_z_;
free_ = other.free_;
limit_ = other.limit_;
hi_ = other.hi_;
error_count_ = other.error_count_;
last_error_ = other.last_error_;
other.config_ = ArenaConfig();
other.lo_ = nullptr;
other.committed_z_ = 0;
other.free_ = nullptr;
other.limit_ = nullptr;
other.hi_ = nullptr;
other.error_count_ = 0;
other.last_error_ = AllocError();
return *this;
}
DArena::~DArena()
{
if (lo_) {
//log && log("unmap [lo,hi)",
// xtag("lo", lo_),
// xtag("z", hi_ - lo_),
// xtag("hi", hi_));
::munmap(lo_, hi_ - lo_);
}
// hygiene
lo_ = nullptr;
committed_z_ = 0;
// checkpoint_ = nullptr;
free_ = nullptr;
limit_ = nullptr;
hi_ = nullptr;
error_count_ = 0;
last_error_ = AllocError();
}
DArena::header_type *
DArena::obj2hdr(void * obj) noexcept
{
assert(config_.store_header_flag_);
return (header_type *)((byte *)obj - sizeof(header_type));
}
AllocInfo
DArena::alloc_info(value_type mem) const noexcept
{
if (!config_.store_header_flag_) [[unlikely]] {
this->capture_error(error::alloc_info_disabled);
return AllocInfo::error_not_configured(&config_.header_);
}
byte * header_mem = mem - sizeof(AllocHeader);
if (!this->contains(header_mem)) {
this->capture_error(error::alloc_info_address);
}
AllocHeader * header = (AllocHeader *)header_mem;
const byte * guard_lo
= header_mem - config_.header_.guard_z_;
const byte * guard_hi
= mem + config_.header_.size(*header);
return AllocInfo(&config_.header_,
guard_lo,
(AllocHeader *)header_mem,
guard_hi);
}
DArenaIterator
DArena::begin() const noexcept
{
return DArenaIterator::begin(this);
}
DArenaIterator
DArena::end() const noexcept
{
return DArenaIterator::end(this);
}
AllocHeader *
DArena::begin_header() const noexcept
{
if (config_.store_header_flag_ == false) {
this->capture_error(error::alloc_iterator_not_supported);
return nullptr;
}
return (AllocHeader *)(lo_ + config_.header_.guard_z_);
}
AllocHeader *
DArena::end_header() const noexcept
{
if (config_.store_header_flag_ == false) {
this->capture_error(error::alloc_iterator_not_supported);
return nullptr;
}
return (AllocHeader *)free_;
}
std::byte *
DArena::alloc(typeseq t, std::size_t req_z)
{
/* - primary allocation path:
* exactly 1 header per alloc() call.
* - store_header_flag follows configuration
*/
(void)t;
return _alloc(req_z, alloc_mode::standard);
}
std::byte *
DArena::super_alloc(typeseq t, std::size_t req_z)
{
/* - (uncommon) pattern for parent alloc immediately followed by
* zero-or-more susidiary allocs, all sharing a single header.
* - collapses into alloc() behavior when
* ArenaConfig.store_header_flag_ disabled
*/
(void)t;
return _alloc(req_z,
alloc_mode::super);
}
std::byte *
DArena::sub_alloc(std::size_t req_z,
bool complete_flag)
{
/* - (uncommon) pattern for subsidiary allocs:
* that piggyback onto preceding super_alloc()
* - collapses into alloc() behavior when
* ArenaConfig.store_header_flag_ disabled
*/
return _alloc(req_z,
(complete_flag
? alloc_mode::sub_complete
: alloc_mode::sub_incomplete));
}
void
DArena::capture_error(error err,
size_type target_z) const
{
DArena * self = const_cast<DArena *>(this);
++(self->error_count_);
self->last_error_ = AllocError(err,
error_count_,
target_z,
committed_z_,
reserved());
}
byte *
DArena::_alloc(std::size_t req_z, alloc_mode mode)
{
scope log(XO_DEBUG(config_.debug_flag_));
/*
* sub_complete
* sub_incomplete |
* standard super | |
* v v v v
*/
std::array<bool, 4> store_header_v = {{ true, true, false, false }};
std::array<bool, 4> retain_header_v = {{ false, true, false, false }};
std::array<bool, 4> store_guard_v = {{ true, false, false, true }};
/* -> write header at free_ */
bool store_header_flag = false;
/* -> stash last_header_*/
bool retain_header_flag = false;
/* -> write guard bytes */
bool store_guard = false;
if (config_.store_header_flag_) {
store_header_flag = store_header_v[(int)mode];
retain_header_flag = retain_header_v[(int)mode];
store_guard = store_guard_v[(int)mode];
}
assert(padding::is_aligned((size_t)free_));
/*
* free_(pre)
* v
*
* <-------------z1--------------->
* < guard >< hz >< req_z >< dz >< guard >
*
* used <== +++++++++0000zzzz@@@@@@@@@@@@@@@@@ppppppp+++++++++ ==> avail
*
* ^ ^ ^
* header mem |
* ^ |
* last_header_ free_(post)
*
* [+] guard after each allocation, for simple sanitize checks
* [0] unused header bits (avail to application)
* [z] record allocation size
* [@] new allocated memory
* [p] padding (to uintptr_t alignment)
*/
/* non-zero if header feature enabled */
size_t hz = 0;
/* dz: pad req_z to alignment size (multiple of 8 bytes, probably) */
size_t dz = padding::alloc_padding(req_z);
size_t z0 = req_z + dz;
/* if non-zero:
* will store padded alloc size at the beginning of each allocation
* reminder:
* important to store padded size for correct arena iteration
*/
uint64_t header = req_z + dz;
if (store_header_flag)
{
if (config_.header_.is_size_enabled()) [[likely]] {
hz = sizeof(header);
} else {
/* req_z doesn't fit in configured header_size_mask bits */
capture_error(error::header_size_mask);
return nullptr;
}
}
size_t z1 = hz + z0;
assert(padding::is_aligned(z1));
if (!this->expand(this->allocated() + z1)) [[unlikely]] {
/* (error state already captured) */
return nullptr;
}
if (store_header_flag) {
/* capturing header */
*(uint64_t *)free_ = header;
if (retain_header_flag) {
/* and rembering for subsequent
* sub_alloc()
*/
last_header_ = (AllocHeader *)free_;
}
}
byte * mem = free_ + hz;
this->free_ += z1;
if (store_guard) {
/* write guard bytes for overrun detection */
::memset(free_,
config_.header_.guard_byte_,
config_.header_.guard_z_);
this->free_ += config_.header_.guard_z_;
}
log && log(xtag("self", config_.name_),
xtag("hz", hz),
xtag("z0", req_z),
xtag("+pad", dz),
xtag("z1", z1),
xtag("size", this->committed()),
xtag("avail", this->available()));
log && log(xtag("mem", mem),
xtag("free", free_));
return mem;
}
void
DArena::establish_initial_guard() noexcept
{
assert(free_ == lo_);
::memset(this->free_,
config_.header_.guard_byte_,
config_.header_.guard_z_);
this->free_ += config_.header_.guard_z_;
}
bool
DArena::expand(size_t target_z) noexcept
{
scope log(XO_DEBUG(config_.debug_flag_),
xtag("target_z", target_z),
xtag("committed_z", committed_z_));
if (target_z <= committed_z_) [[likely]] {
log && log("trivial success, offset within committed range",
xtag("target_z", target_z),
xtag("committed_z", committed_z_));
return true;
}
if (lo_ + target_z > hi_) [[unlikely]] {
this->capture_error(error::reserve_exhausted, target_z);
return false;
}
/*
* pre:
*
* _______________...................................
* ^ ^ ^
* lo limit hi
*
* < committed_z >
* <----------target_z----------->
* > <- z: 0 <= z < hugepage_z
* <---------aligned_target_z--------->
* <--- add_commit_z -->
*
* post:
* ____________________________________..............
* ^ ^ ^
* lo limit hi
*
*/
std::size_t aligned_target_z = padding::with_padding(target_z, arena_align_z_);
std::byte * commit_start = limit_; // = lo_ + committed_z_;
std::size_t add_commit_z = aligned_target_z - committed_z_;
assert(limit_ == lo_ + committed_z_);
if (::mprotect(commit_start,
add_commit_z,
PROT_READ | PROT_WRITE) != 0) [[unlikely]]
{
if (log) {
log("commit failed!");
log(xtag("aligned_target_z", aligned_target_z),
xtag("commit_start", commit_start),
xtag("add_commit_z", add_commit_z),
xtag("commit_end", commit_start + add_commit_z)
);
}
capture_error(error::commit_failed, add_commit_z);
return false;
}
committed_z_ = aligned_target_z;
limit_ = lo_ + committed_z_;
if (commit_start == lo_) [[unlikely]] {
/* first expand() for this allocator - start with guard_z_ bytes */
this->establish_initial_guard();
}
assert(committed_z_ % arena_align_z_ == 0);
assert(reinterpret_cast<size_t>(limit_) % arena_align_z_ == 0);
return true;
} /*expand*/
void
DArena::clear() noexcept
{
this->free_ = lo_;
this->establish_initial_guard();
}
}
} /*namespace xo*/
/* end DArena.cpp */