Re: [dpdk-dev] [PATCH v6 0/3] Improve zero-length memzone allocation

2018-07-13 Thread Thomas Monjalon
31/05/2018 11:50, Anatoly Burakov: > This patchset does two things. First, it enables reserving > memzones of zero-length that are IOVA-contiguous. Second, > it fixes a long-standing race condition in reserving > zero-length memzones, where malloc heap is not locked between > stats collection and r

[dpdk-dev] [PATCH v6 0/3] Improve zero-length memzone allocation

2018-05-31 Thread Anatoly Burakov
This patchset does two things. First, it enables reserving memzones of zero-length that are IOVA-contiguous. Second, it fixes a long-standing race condition in reserving zero-length memzones, where malloc heap is not locked between stats collection and reservation, and will instead allocate biggest