- 09 Feb, 2018 5 commits
-
-
Kirill Smelkov authored
-
Kirill Smelkov authored
-
Kirill Smelkov authored
-
Kirill Smelkov authored
-
Kirill Smelkov authored
-
- 08 Feb, 2018 3 commits
-
-
Kirill Smelkov authored
-
Kirill Smelkov authored
-
Kirill Smelkov authored
-
- 07 Feb, 2018 1 commit
-
-
Kirill Smelkov authored
-
- 06 Feb, 2018 5 commits
-
-
Kirill Smelkov authored
-
Kirill Smelkov authored
Don't try to play with file.readline() - that seems to be working for usual cases, but: 1. it does not work for cases when e.g. a dot is printed for every test run on the same line (e.g. neo/py tests, pytest, etc...) 2. in case when there is a lot of output available ready to be read it would be a waste of resources to read/flush it line by line. -> use raw OS read with sane semantics.
-
Kirill Smelkov authored
(fix syntax error there)
-
Kirill Smelkov authored
We now don't buffer anything inside runTestSuite, but the generating process itself could buffer e.g. stdout. Tell it not to do so.
-
Kirill Smelkov authored
if so `hostname --fqdn` errors with hostname: Name or service not known Fall back to just small hostname if there is no fqdn one.
-
- 05 Feb, 2018 3 commits
-
-
Kirill Smelkov authored
stderr is unbuffered, but by default stdout is buffered, so if stdout in turn is redirected to be read by some other logging program, due to buffering output could be delayed. Don't delay the output.
-
Kirill Smelkov authored
We send output from tested process to master. We also print it to stdout,stderr so it appears in testnode logs. However till now it was like, whole output first read, and only then emitted to log as a whole, thus not allowing to oversee current test progress by watching testnode log tail. Fix it by implementing the teeing process manually and using only line-buffered mode inside.
-
Kirill Smelkov authored
-
- 02 Feb, 2018 5 commits
-
-
Kirill Smelkov authored
-
Kirill Smelkov authored
Extracted from nexedi/slapos!282
-
Kirill Smelkov authored
killall is optional (comes from psmisc package) and besides `killall runzeo` is not safe as it can be killing not only runzeo spawn from under neotest, but also from other processes. -> Switch to explicit kill with runzeo pid.
-
Kirill Smelkov authored
This is useful for development and just checking ZODB part of the benchmarks without going first through cpu and disk parts. run-client renamed to zbench-client for consistency.
-
Kirill Smelkov authored
-
- 01 Feb, 2018 2 commits
-
-
Kirill Smelkov authored
-
Kirill Smelkov authored
-
- 25 Jan, 2018 2 commits
-
-
Kirill Smelkov authored
Amends c884bfd5 (X neo/protogen: Catch length checks overflows on decode)
-
Kirill Smelkov authored
-
- 24 Jan, 2018 4 commits
-
-
Kirill Smelkov authored
See https://github.com/golang/go/issues/19126#issuecomment-358743715. Examples: before after TEXT ·(*GetObject).neoMsgDecode(SB), NOSPLIT, $8-56 // zproto-marshal.go:1914 │TEXT ·(*GetObject).neoMsgDecode(SB), NOSPLIT, $0-56 // zproto-marshal.go:1914 // SUBQ $8, SP │ NO_LOCAL_POINTERS // MOVQ BP, (SP) (BP save) │ // FUNCDATA $0, gclocals·846769608458630ae82546dab39e913e(SB) (args) // LEAQ (SP), BP (BP init) │ // FUNCDATA $1, gclocals·33cdeccccebe80329f1fdbee7f5874cb(SB) (no locals) // FUNCDATA $0, gclocals·21e863e2261befa92f8534560680bbb6(SB) (args) │ MOVQ data+24(SP), AX FUNCDATA $1, gclocals·69c1753bd5f81501d95132d08af04464(SB) (locals) │ CMPQ AX, $24 // zproto-marshal.go:1915 MOVQ data+32(SP), AX │ JGE pc45 CMPQ AX, $24 // zproto-marshal.go:1915 │ MOVQ ·ErrDecodeOverflow+8(SB), AX // zproto-marshal.go:1924 JLT pc163 │ MOVQ ·ErrDecodeOverflow(SB), CX MOVQ data+24(SP), CX │ MOVQ $0, _r1+40(SP) MOVQ (CX), DX // zproto-marshal.go:1918 │ MOVQ CX, _r2+48(SP) BSWAPQ DX │ MOVQ AX, _r2+56(SP) MOVQ p+16(SP), BX │ RET MOVQ DX, (BX) │pc45: LEAQ -8(AX), DX // zproto-marshal.go:1919 │ MOVQ data+16(SP), AX MOVQ data+40(SP), SI │ MOVQ (AX), CX // zproto-marshal.go:1918 LEAQ -8(SI), DI │ BSWAPQ CX NEGQ DI │ MOVQ p+8(SP), DX SARQ $63, DI │ MOVQ CX, (DX) ANDQ $8, DI │ MOVQ 8(AX), CX // zproto-marshal.go:1919 CMPQ DX, $7 │ BSWAPQ CX JLS pc212 │ MOVQ CX, 8(DX) MOVQ (CX)(DI*1), DX │ MOVQ 16(AX), AX // zproto-marshal.go:1920 BSWAPQ DX │ BSWAPQ AX MOVQ DX, 8(BX) │ MOVQ AX, 16(DX) LEAQ -16(SI), DX // zproto-marshal.go:1920 │ MOVQ $24, _r1+40(SP) // zproto-marshal.go:1921 NEGQ DX │ MOVQ $0, _r2+48(SP) ADDQ $-16, AX │ MOVQ $0, _r2+56(SP) SARQ $63, DX │ RET ANDQ $16, DX │ CMPQ AX, $7 │ JLS pc205 │ MOVQ (CX)(DX*1), AX │ BSWAPQ AX │ MOVQ AX, 16(BX) │ MOVQ $24, _r1+48(SP) // zproto-marshal.go:1921 │ MOVQ $0, _r2+56(SP) │ MOVQ $0, _r2+64(SP) │ // MOVQ (SP), BP (BP restore) │ // ADDQ $8, SP (SP restore) │ RET │ pc163: │ MOVQ ·ErrDecodeOverflow(SB), AX // zproto-marshal.go:1924 │ MOVQ ·ErrDecodeOverflow+8(SB), CX │ MOVQ $0, _r1+48(SP) │ MOVQ AX, _r2+56(SP) │ MOVQ CX, _r2+64(SP) │ // MOVQ (SP), BP (BP restore) │ // ADDQ $8, SP (SP restore) │ RET │ pc205: │ PCDATA $0, $1 // zproto-marshal.go:1920 │ CALL runtime.panicindex(SB) │ UNDEF │ pc212: │ PCDATA $0, $1 // zproto-marshal.go:1919 │ CALL runtime.panicindex(SB) │ UNDEF │
-
Kirill Smelkov authored
Improves signal/noise ratio in generated decoder.
-
Kirill Smelkov authored
For example a list is encoded as l u32 [l]item itemType on decode len is read from data stream and for rest of data len(data) is checked to be < l*sizeof(item). However since l is u32 and sizeof(item) is just number the result of `l * sizeof(item)` has also u32 type. However it could overflow e.g. for l = 0x20000000 sizeof(item) = 8 with the l*sizeof(item) being = u32(0) (exactly zero) -> oops. Avoid the problem by doing all checking arithmetics with u64 ints.
-
Kirill Smelkov authored
-
- 23 Jan, 2018 2 commits
-
-
Kirill Smelkov authored
-
Kirill Smelkov authored
* origin/master: client: kill .supportsTransactionalUndo() client: for read accesses, pick a random good node, connected or not storage: optimize storage layout of raw data for replication sqlite: remove useless AUTOINCREMENT for data.id (reuse of deleted ids is fine) storage: speed up reads by indexing 'obj' primarily by 'oid' (instead of 'tid') storage: pass schema of tables to migration methods storage: update backend version between each migration step
-
- 17 Jan, 2018 1 commit
-
-
Kirill Smelkov authored
Usage of supportsTransactionalUndo() was removed from ZODB in 2007 - see e.g. the following commits: https://github.com/zopefoundation/ZODB/commit/a06bfc03 https://github.com/zopefoundation/ZODB/commit/e667b022 https://github.com/zopefoundation/ZODB/commit/f595f7e7 ... /reviewed-by @vpelletier /reviewed-on !8
-
- 16 Jan, 2018 7 commits
-
-
Kirill Smelkov authored
-
Kirill Smelkov authored
-
Kirill Smelkov authored
-
Kirill Smelkov authored
-
Kirill Smelkov authored
-
Kirill Smelkov authored
-
Kirill Smelkov authored
* y/go: (25 commits) go/zodb/zodbtools: TODO (cmp, analyze) go/zodb/zodbtools: Catobj go/zodb/zodbtools: Info go/zodb/zodbtools: Dump go/zodb: Start of zodbtools - tools for managing ZODB databases go/zodb/fs1tools: Notes about other possible useful commands currently being there on ZODB/py side go/zodb/fs1tools: Reindex, Verify-index go/zodb/fs1tools: Dump go/zodb/fs1: Start fs1tools - tools for managing and maintaining ZODB FileStorage v1 databases go/zodb/fs1: My notes on I/O go/zodb/fs1: Register FileStorage to zodb & wks go/zodb/fs1: Actual FileStorage ZODB driver go/zodb/fs1: Add routines to (re)build and verify index from/wrt original FileStorage data go/zodb/fs1: Index save/load go/zodb/fs1: BTree specialized with KEY=zodb.Oid, VALUE=int64 go/zodb: Start of FileStorage support go/zodb: Way for storage-drivers to be registered and for clients to open them by URL zodb/go: In-RAM client cache go/zodb: Minimal serialization compatibility with ZODB/py go/zodb: Tid connection with time ...
-