1
0
mirror of https://github.com/Zygo/bees.git synced 2025-05-17 21:35:45 +02:00

readahead: inject more sanity at the foundation of an insane architecture

This solves a third bad problem with bees reads:

3.  The architecture above the read operations will issue read requests
for the same physical blocks over and over in a short period of time.

Fixing that properly requires rewriting the upper-level code, but a
simple small table of recent read requests can reduce the effect of the
problem by orders of magnitude.

Signed-off-by: Zygo Blaxell <bees@furryterror.org>
This commit is contained in:
Zygo Blaxell 2024-11-18 14:22:26 -05:00
parent 8d08a3c06f
commit e78e05e212

View File

@ -214,10 +214,35 @@ BeesTooLong::operator=(const func_type &f)
return *this;
}
static
bool
bees_readahead_check(int const fd, off_t const offset, size_t const size)
{
// FIXME: the rest of the code calls this function more often than necessary,
// usually back-to-back calls on the same range in a loop.
// Simply discard requests that are identical to recent requests from the same thread.
const Stat stat_rv(fd);
auto tup = make_tuple(offset, size, stat_rv.st_dev, stat_rv.st_ino);
static mutex s_recent_mutex;
static set<decltype(tup)> s_recent;
unique_lock<mutex> lock(s_recent_mutex);
if (s_recent.size() > BEES_MAX_EXTENT_REF_COUNT) {
s_recent.clear();
BEESCOUNT(readahead_clear);
}
const auto rv = s_recent.insert(tup);
// If we recently did this readahead, we're done here
if (!rv.second) {
BEESCOUNT(readahead_skip);
}
return rv.second;
}
static
void
bees_readahead_nolock(int const fd, const off_t offset, const size_t size)
{
if (!bees_readahead_check(fd, size, offset)) return;
Timer readahead_timer;
BEESNOTE("readahead " << name_fd(fd) << " offset " << to_hex(offset) << " len " << pretty(size));
BEESTOOLONG("readahead " << name_fd(fd) << " offset " << to_hex(offset) << " len " << pretty(size));
@ -248,13 +273,15 @@ bees_readahead_nolock(int const fd, const off_t offset, const size_t size)
BEESCOUNTADD(readahead_ms, readahead_timer.age() * 1000);
}
static mutex s_only_one;
void
bees_readahead_pair(int fd, off_t offset, size_t size, int fd2, off_t offset2, size_t size2)
{
BEESNOTE("waiting to readahead " << name_fd(fd) << " offset " << to_hex(offset) << " len " << pretty(size)
<< ", " << name_fd(fd2) << " offset " << to_hex(offset2) << " len " << pretty(size2));
static mutex only_one;
unique_lock<mutex> m_lock(only_one);
if (!bees_readahead_check(fd, size, offset) && !bees_readahead_check(fd2, offset2, size2)) return;
BEESNOTE("waiting to readahead " << name_fd(fd) << " offset " << to_hex(offset) << " len " << pretty(size) << ","
<< "\n\t" << name_fd(fd2) << " offset " << to_hex(offset2) << " len " << pretty(size2));
unique_lock<mutex> m_lock(s_only_one);
bees_readahead_nolock(fd, offset, size);
bees_readahead_nolock(fd2, offset2, size2);
}
@ -262,9 +289,9 @@ bees_readahead_pair(int fd, off_t offset, size_t size, int fd2, off_t offset2, s
void
bees_readahead(int const fd, const off_t offset, const size_t size)
{
if (!bees_readahead_check(fd, size, offset)) return;
BEESNOTE("waiting to readahead " << name_fd(fd) << " offset " << to_hex(offset) << " len " << pretty(size));
static mutex only_one;
unique_lock<mutex> m_lock(only_one);
unique_lock<mutex> m_lock(s_only_one);
bees_readahead_nolock(fd, offset, size);
}