8 Commits (0cf57f220f8a9cee30cb714cba10e03e8750ccae)

Author	SHA1	Message	Date
Richard Fuchs	649e225bc9	MT#55283 fix alloc alignment ... We need at least 64 bits for atomic64 Change-Id: I08724d204f2a1603088c8737ca096a4b64314f6b	7 months ago
Richard Fuchs	675eb7bea6	MT#55283 fix int size issues ... Change-Id: Ibd7bdea16ae57a4a86076a05b22bb3ca64be8ce1	7 months ago
Richard Fuchs	69e634f94a	MT#55283 store metadata head pointer ... Put a pointer to the shard itself in the beginning of the buffer. This facilitates quick lookup. Change-Id: I0109d77e56afac3189775dd708970288f175dfe6	12 months ago
Richard Fuchs	f7fbe490b0	MT#55283 request aligned addresses from mmap ... Or rather, request double the amount, then use the contained aligned block. Change-Id: Iaf1da4b3df243a1d4391fc2ff46eba1fec3d738c	12 months ago
Richard Fuchs	9bd5670f89	MT#55283 use fixed allocation size ... Allocate all bufferpool shards of the same size, regardless of underlying allocator. This way increase memory usage a bit, but we're already quite heavy on that, so no big deal. Change-Id: Icbe09cd2f9b33bc58ab1efe7de293dea00236fec	12 months ago
Richard Fuchs	5f9e43b3e0	MT#55283 move alignment macros to header file ... Change-Id: I266cbefe8351f77399bb6e50b2122f9a45249529	12 months ago
Richard Fuchs	1a98502a1b	MT#55283 used aligned memory for bufferpool ... Use allocators that return memory blocks aligned to the requested size, instead of generic malloc. This makes it easier to play tricks with the memory blocks. Change-Id: Iad4b1127c76e48c2e9b4ad8489118d4883a24f6a	12 months ago
Richard Fuchs	872c06d1cc	MT#55283 add bufferpool implementation ... Manages a pool of bump allocators that allow re-use of buffers after they become empty Change-Id: Id677a2e18b39c2261b8ceae31056d4ad9125bf2c	2 years ago