============================== b00.txt ==============================

l2p17.05.00 pre-build - 26 May 05
update to final p17.04.00 versions

== Build system
SRT_D0                    p17-br-04 <- p17-br-03
  - Update root to v4_02_00a_eh -q GCC_3_4_3:opt
  - Switch from lund v6_223 to pythia v6_319
ctbuild                   p17-br-04 <- p17-br-03
  - Change from lund to pythia for v6_319
  - Fix some problems with shared libraries.
  - Add debugging printouts (commented out, tagged with AMJ) to find why
    a library is added to the link line.
  - add a new and delete found in Coin to the list of symbols to ignore.

ctbuild                   p17-br-05 <- p17-br-04
  Fixes rules so that the correct .o files get onto the link line under
  all circumstances. There were odd cases where that wasn't true for
  SRT packages before.

== External products
root v4_00_08g_eh -q +GCC_3_4_2:+opt -> root v4_02_00a_eh -q +GCC_3_4_3:+opt
lund v6_223 -> pythia v6_319 -q +GCC3_4_3
herwig v6_506 -q +GCC3_3_1 -> herwig v6_507 -q +GCC3_4_3
db_server_gen v5_1 -> delete

============================== b01.txt ==============================

l2p17.05.00 build 1 - 26 May 05

== Level 2: abaz
l2base                    p17-br-03 <- p17-br-02
  add ostream operator for maddr datatype to fix online gcc3.4 build
l2cttcftworker            p17-br-12 <- p17-br-10gcc
  - major one for l2io and l2trackerbase: version of the STT input
    and output objects changed from 0 to 1 because a new variables
    were introduced: input from STT "twopass" in L2STTTrack and
    "sttTwoPass" in GlobalTrack
  - minor one for l2cttcft and l2sttctt workers: fix to their test
    files. Until now, they were actually testing l2cpsworker.
l2errorlogger             l2-00-01-24 <- v00-01-23
  code fixes for p17
l2gbljet                  p17-br-03 <- p17-br-01
  Changes needed in order to have the CalCalibFilter (implementation).
l2gblmuon                 p17-br-04 <- p17-br-01
  - Fixed the tool to enhance kinematic pt changing the pt to the track
    pt if there is a matched track and changed the filter to cut if
    both the pt and the toroid pt are below the pt threshold
  - Bug fix of if((*muon)->pt() && (*muon)->ToroidPt()<_minET) {continue;}
    to if((*muon)->pt()<_minET && (*muon)->ToroidPt()<_minET) {continue;}
l2gblworker               p17-br-30 <- p17-br-21
  Changes needed in order to have the CalCalibFilter (implementation).
l2io                      p17-br-11 <- p17-br-03
  - SimInputBuffer (offline - trigsim): removed blockOffset() method
    which speeds up the code and then modified it so the necessary
    bit-shift from this method is done directly in the packer() method
    correctly.
  - InputBuffer (online): removed blockOffset() method which speeds up
    the code and modified the code and then modified it so the necessary
    bit-shift from this method is done directly in the packer() method
    correctly. Plus, Roger has implemented new way of checking for zeroes
    after the 1st data trailer (Alphas aligned data to 32 bit word, Betas
    equipped with PIII do not do it).
l2mbus                    l2-00-05-15 <- l2-00-05-14
  error message updates
l2online                  l2-01-00-04 <- v01-00-03
  code fixes for p17
l2sttcttworker            p17-br-13 <- p17-br-10gcc
  Change in output.evigen
  - Adjusted the comment only.
l2sttipworker             p17-br-12 <- p17-br-10gcc
  Change in output.evigen
  - Changed "name" and "l2output" from "sttiptrack" to "globaltrack",
    and also changed the "blockID" from STTPT_STD_OUTPUT to 
    STTIP_STD_OUTPUT. Adjusted the comment too.
l2trackerbase             p17-br-13 <- p17-br-10gcc
  l2trackerbase/src/GlobalTrack.cpp - put in a safety net - check the
  version number, must be >1, before sttTwoPass() is filled.
l2trackworker             p17-br-11 <- p17-br-10gcc
  - change the number of tracks out from PP workers to 200 
    (not to exceed limit on the DMA beta engine).
  - change max. number of tracks to 552 (2x6x46 from STT).
  ..these changes to reflect the similar ones done already in p16.
    And only now everything made to p17 to.
  - filter STT tracks with "fitFailed()=1". These tracks are not added
    into the output track list from L2CTT (STTpT/STTIP tracks)
  ..this should prevent L2CTT going up to ~75% (or more) in its
    processing time when STT runs in the extended mode.


== Build systems: jonckheere
SRT_D0                    p17-br-05 <- p17-br-04
  Add new arch_specs for pythia v6_319 and lhapdf v3_0
    arch_spec_cern_nopdflib.mk
    arch_spec_lhapdf_fnkits.mk
    arch_spec_pythia_nopydata.mk
D0-glpkgs
  lhapdf v3_0 <- v1-2
    Update per Thomas Nunnemann