Commit fee4a7e4 authored by Kirill Smelkov's avatar Kirill Smelkov

.

parent 06ed10ee
...@@ -273,6 +273,14 @@ package main ...@@ -273,6 +273,14 @@ package main
// won't be served from OS file cache and instead will trigger a FUSE read // won't be served from OS file cache and instead will trigger a FUSE read
// request to wcfs. // request to wcfs.
// //
// - for file δFtail invalidation info is maintained:
//
// - tail of [](rev↑, []#blk)
// - {} #blk -> []rev↑ in tail XXX needed?
//
// δFtail describes changes to file we learned from ZODB invalidation.
// XXX and allows to quick #blk -> ... ?
//
// - XXX δZtail of invalidation info is maintained. // - XXX δZtail of invalidation info is maintained.
// //
// - tail of [](tid↑, []oid) // - tail of [](tid↑, []oid)
...@@ -365,7 +373,7 @@ package main ...@@ -365,7 +373,7 @@ package main
// δ(BTree) in wcfs context: // δ(BTree) in wcfs context:
// //
// . -k(blk) -> invalidata #blk // . -k(blk) -> invalidata #blk
// . +k(blk) -> ignore (no need to invalidate) // . +k(blk) -> invalidate #blk (e.g. if blk was previously read as hold)
// //
// //
// //
...@@ -374,7 +382,8 @@ package main ...@@ -374,7 +382,8 @@ package main
// XXX serving read from @<rev>/data // XXX serving read from @<rev>/data
// //
// //
// // XXX(integrate place=?) ZData - ne need to keep track -> ZBlk1 is always
// marked as changed on blk data change.
// //
// //
// - XXX(kill) 1 ZODB connection per 1 bigfile (each bigfile can be at its different @at, // - XXX(kill) 1 ZODB connection per 1 bigfile (each bigfile can be at its different @at,
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment