Re: [sqlite] PRAGMA locking_mode = EXCLUSIVE

2007-03-27 Thread Joe Wilson
--- [EMAIL PROTECTED] wrote: > There is a plot of runtime of CVSHEAD versus version 3.3.13 > at > >http://www.sqlite.org/relspeed-20070326-ephemeral.gif > > Any value less than > 1.0 means that performance has improved. The test script is > >

Re: [sqlite] PRAGMA locking_mode = EXCLUSIVE

2007-03-27 Thread Iulian Musat
Really interesting. Is there an easy way of producing the big "sqlite3.c" from CVS head? I'm curios to run some of my tests this way. -i [EMAIL PROTECTED] wrote: Joe Wilson <[EMAIL PROTECTED]> wrote: What operations/work patterns will benefit most from PRAGMA locking_mode = EXCLUSIVE?

Re: [sqlite] PRAGMA locking_mode = EXCLUSIVE

2007-03-26 Thread Joe Wilson
Congratulations. This is a big milestone. The single "sqlite3.c" source file release will certainly simplify embeddeding. I guess the single translation unit thing is why GCC is busy working on link-time optimization (and why LLVM is faster in some cases). --- [EMAIL PROTECTED] wrote: > When

Re: [sqlite] PRAGMA locking_mode = EXCLUSIVE

2007-03-26 Thread drh
Joe Wilson <[EMAIL PROTECTED]> wrote: > What operations/work patterns will benefit most from > PRAGMA locking_mode = EXCLUSIVE? > > Can you roughly quantify the speedups in such cases? > When you do "PRAGMA locking_mode=EXCLUSIVE", it means that SQLite will not release its EXCLUSIVE lock after

[sqlite] PRAGMA locking_mode = EXCLUSIVE

2007-03-26 Thread Joe Wilson
What operations/work patterns will benefit most from PRAGMA locking_mode = EXCLUSIVE? Can you roughly quantify the speedups in such cases? TV dinner still cooling? Check out "Tonight's Picks" on Yahoo! TV.