Piotr Balcer comments

Results 193 comments of


                                            Piotr Balcer

libpmem-issue: unchangeability of relative address:: pmem_map_file

Yes. If you are mapping the same part of the file, you will always get the same data.

obj_msync_nofail fails when msync is interrupted by signal

Let's add an ENV variable to control this behavior, and by default let's repeat 100 times on EINTR.

Tests `pmempool_sync_remote` are failing

Looks like poolsets are not opening/parsing correctly: ``` error: opening poolset failed /dev/shm/shm0/dir1/test_pmempool_sync_remote/data/pool.set: Invalid argument error: opening poolset failed /dev/shm/shm0/dir1/test_pmempool_sync_remote/data/pool.set: Invalid argument error: opening poolset failed /dev/shm/shm0/dir1/test_pmempool_sync_remote/data/pool.set: Invalid argument error:...

Tests `pmempool_sync_remote` are failing

@draghan is this still an issue?

FEAT: Fragmentation and long-term stability focused benchmarks

@Greg091 What's the status of this task?

Create obj on /dev/daxX.Y with no badblock failed on checking badblock

libpmemobj checks bad blocks and shutdown state by default. AFAIR one of the problems of the existing ndctl interface is that there's no way to distinguish between "bad blocks not...

pmreorder treats c++ operators as markers

@igchor is this fixed? if so, please close the issue.

libpmemobj: Allocated objects not aligned to cache boundaries

The allocated raw objects are technically cache line aligned, but the effective alignment of user data is offset by the object's header. This "soft alignment" allows you to reason about...

libpmemobj: Allocated objects not aligned to cache boundaries

I'm afraid not - the default allocation classes all use the compact header and do not use alignment. The first implementation of the allocation classes interface did allow the application...

Stable 1.12

@lplewa please rebase and resolve the conflict.