Hi, Stephen wrote: > Fixing this in raid seems far, far preferable to fixing it in the > filesystems. The filesystem should be allowed to use the buffer cache > for metadata and should be able to assume that there is a way to prevent > those buffers from being written to disk until it is ready. What about doing it in the page cache: i.e. reserve pages for journaling and let them hit the buffer cache only when the transaction allows it? This may be a naive suggestion, but it looks logical. - Peter -
- Re: Raid resync changes buffer ca... Stephen C. Tweedie
- Re: Raid resync changes buffe... Theodore Y. Ts'o
- Re: Raid resync changes buffe... Ingo Molnar
- Re: (reiserfs) Re: Raid resyn... Stephen C. Tweedie
- Re: (reiserfs) Re: Raid resyn... Ingo Molnar
- Re: (reiserfs) Re: Raid resyn... Rik van Riel
- Re: (reiserfs) Re: Raid resyn... Stephen C. Tweedie
- Re: (reiserfs) Re: Raid resyn... Stephen C. Tweedie
- Buffer and page cache braam
- Re: Buffer and page cache Stephen C. Tweedie
- Re: Raid resync changes buffer cache semantics --- not ... braam
- Re: Raid resync changes buffer cache semantics ---... Stephen C. Tweedie
- Re: Raid resync changes buffer cache semantics --- not ... Matt Zinkevicius
- Re: Raid resync changes buffer cache semantics ---... Daniel Veillard