* copyright (C) 1984-2019 merrill consultants dallas texas usa
Yüklə 28,67 Mb.
|
|
R748LFLF R748LFLY R748LFLS R748LFPQ R748LFIT R748LFCR R748LFR1 R748LFR2 R748LFIF R748LFOD R748LFOA R748LFDF R748LFIO R748LFTC R748LFBC -TYPE748A New variables: R748AAS0 R748AAS2 R748AAS3 R748AAS4 -TYPE748R New variables: R748RTQ -TYPE748X New variables: R748XPTQ -TYPE89 New variables: SMF89CHC SMF89ACR SMF89NCR SMF89SCF SMF89CAI SMF89CCR SMF89PCF -TYPE9033 New dataset with variables SMF9033T SMF9033N SMF9033S -TYPE9034 New dataset with variables SMF9034T SMF9034F SMF9034A SMF9034N SMF9034S SMF9034I SMF9034R SMF9034F -TYPE113 New variables, but see also Change 28.166. SM113CPT -TYPE119 New Datasets: T11904 TYP11904 TCP PROFILE INFORMATION FOR STACK T11932 TYP11932 DVIPA STATUS CHANGE T11933 TYP11933 DVIPA REMOVED T11934 TYP11934 DVIPA TARGET ADDED T11935 TYP11935 DVIPA TARGET REMOVED T11936 TYP11936 DVIPA TARGET SERVER STARTED T11937 TYP11937 DVIPA TARGET SERVER ENDED T11948 TYP11948 CSSMTP CONFIGURATION DATA T11949 TYP11949 CSSMTP TARGET SERVER CONNECTION T11950 TYP11950 CSSMTP MAIL T11951 TYP11951 CSSMTP SPOOL T11952 TYP11952 CSSMTP STATISTICS T11973 TYP11973 IPSEC IKE TUNNEL ACTIVATE T11974 TYP11974 IPSEC IKE TUNNEL DEACTIVATE T11975 TYP11975 IPSEC DYNAMIC TUNNEL ACTIVATE T11976 TYP11976 IPSEC DYNAMIC TUNNEL DEACTIVATE T11977 TYP11977 IPSEC DYNAMIC TUNNEL ADDED T11978 TYP11978 IPSEC DYNAMIC TUNNEL REMOVED T11979 TYP11979 IPSEC MANUAL TUNNEL ACTIVATE T11980 TYP11980 IPSEC MANUAL TUNNEL DEACTIVATE -TYP11904 dataset is NOT decoded, it has 21 sections and I hope NO ONE EVERY wants it decoded, but if you REALLY do, then please send an example SMF record. -TYP11948-TYP11952 CSSMTP common segments are decoded but not all subtype-specific segments are, since I have no data records. -TYP11948-TYP11952 IPSEC common segments are decoded but not all subtype-specific segments are, since I have no data records. -Many of the new 119 datasets contain IPV6 IP addresses, which are a $CHAR16. field with hexadecimal values that are expanded to a 39-byte text field with colons. This 16-byte hex value in the SMF record '0123456789ABCDEF0123456789abcdef'X '7CC3C1E2000000050000000000003000'X is printed in ftp connection messages as 0123 4567 89ABCDEF cdef 7CC3:c1e2:00000005::3000 is expanded to this 39-character text field in MXG: 0123:4567:89AB:CDEF:0123:4567:89AB:CDEF This code creates the 39-byte recognizable text value: INPUT CHAR16IP $CHAR16. @; IPADR6=PUT(INPUT(SUBSTR(CHAR16IP,1,2),$CHAR2.),$HEX4.) !!':'!!
PUT(INPUT(SUBSTR(CHAR16IP,3,2),$CHAR2.),$HEX4.) !!':'!!
PUT(INPUT(SUBSTR(CHAR16IP,5,2),$CHAR2.),$HEX4.) !!':'!!
PUT(INPUT(SUBSTR(CHAR16IP,7,2),$CHAR2.),$HEX4.) !!':'!!
PUT(INPUT(SUBSTR(CHAR16IP,9,2),$CHAR2.),$HEX4.) !!':'!!
PUT(INPUT(SUBSTR(CHAR16IP,11,2),$CHAR2.),$HEX4.) !!':'!!
PUT(INPUT(SUBSTR(CHAR16IP,13,2),$CHAR2.),$HEX4.) !!':'!!
PUT(INPUT(SUBSTR(CHAR16IP,15,2),$CHAR2.),$HEX4.); -TYPE119 New Variables: TYP11902: TTTTLSFP TYP11903: FCCCONID FCDCONID -DCOLLECT: Added to DCOLDSET: DCDDS9F1 DCDJBNMC DSCSTNMC DCDTIMEC -DCOLLECT: Added to DCOLDC: DDCFCT DDCBLMT DDCCFS DDCDVCS DDCFSCAL DDCFRLOG DDCFRLGS DDCFEXTC DDCFA2GB DDCFPSEG DDCFKYL1 DDCFKYC1 DDCFKYL2 DDCFKYC2 DDCFVSP DDCFSDB DDCFVORD DDCFCAR DDCCT DDCDSCF DDCRBYTE DDCBLKLM DDCBSZLM DDCTAPE1 DDCPSCA DDCPSEG DDCVSP DDCVSPV DDCKLBL1 DDCKLBN1 DDCKYCD1 DDCKLBL2 DDCKLBN2 DDCKYCD2 -Internally, in VMAC7072, thirty-two sets of variables ARE created for engines 64-95 and used to create the TYPE70 measurements, with these variable name suffixes: Variables CPUID SUFFIX CPUID SUFFIX CPUID SUFFIX (All other) 00-09 D0-D9 10-21 DA-DL 22-34 DN-DZ 35-60 ZA-ZZ 61-86 YA-YZ 87-95 VA-VI Variables PCTxxxBss 00-09 Y0-Y9 10-21 YA-YL 22-34 YN-YZ where xxx= CPB CIB IFB ZIB (all other PCT are above). And four new formats are created that map these pairs of suffix and CPU numbers (but probably will never be used): $MG070SC maps suffix to "all other" variable CPU Number $MG070CS maps CPU number to "all other" suffix $MG070SP maps suffix to "four PCT" variable CPU Number $MG070PS maps CPU Number to "four PCT variable suffix
IMACZDAT ZEE OBS WAS CREATED" and it is used as both an audit date Aug 6, 2010 and for selection of which observations are kept in MONTH PDB's in MONTHBLD and BLDSMPDB. If you have to rebuild a PDB library, you must change the ZDATE to have the correct date, or data can be lost when in the MONTH PDB. Previously, you had to EDIT the IMACZDAT member, but you can instead use this DATA step to set the ZDATE value: DATA _NULL_; DATE='01JUL2010'D; TIME=PUT(DHMS(DATE,0,0,0),32.); DATEC=PUT(DATE,16.); CALL SYMPUT('ZDATE',DATEC); CALL SYMPUT('ZTIME',TIME); RUN; %LET MXGZDATE=%QUOTE( ZDATE=&ZDATE; ZTIME=&ZTIME' ); %INCLUDE SOURCLIB(BUILDPDB); Thanks to Chuck Hopf, Independent Consultant, USA. Change 28.173 Variable ZIPTM was used in GRAFWRKX to carry the zIIP GRAFWRKX time, but it already existed in RMFINTRV as the total of Aug 5, 2010 the TYPE72GO zIIP time, it was included as uncaptured zIIP time on the bar charts. Now the uncaptured zIIP and IFA/zAAP times are correctly calculated for the charts.
IMACCADI the insertion of fields, causing INVALID JESNR messages. VMAC6 There is no "version" field in CA's record, so a test for Aug 5, 2010 LENGTH=347 AND SMF6LN1=287 is used to identify V11 data. These new variables are created: CADIATYP CADIDJDC CADIDPLX CADIEFRM CADIJDEI CADIJDLI CADIOLVR CADIPASS CADIPCR CADIPRTR CADIXFRM Thanks to Glenn Bowman, Wakefern, USA. Change 28.171 IFLs only, MXG 28.01+, STARTIME was a missing value in VMXG70PR the PDB.ASUM70LP dataset. Aug 3, 2010 Thanks to William Edwards, Blue Cross Blue Shield of Florida, USA. Thanks to Bob Maple, Blue Cross Blue Shield of Florida, USA.
VMXGFIND suppress unneeded/unwanted log messages, and MXGNOTE Jul 29, 2010 added at the end to report how many datasets were found with matching values. Change 28.169 -New ANALCAPD analysis will estimate the impact of setting ANALCAPD a Defined Capacity limit, at the CEC level, with both a Jul 29, 2010 text report and a graph (if SAS/GRAPH is found) or plot Aug 14, 2010 (if no SAS/GRAPH) of the intervals which might be limited by a cap, i.e., intervals where the rolling four hour average exceeded the Cap value that you specified. Arguments: PDB=PDB Libname where the ASUMCEC dataset is found DEFCAP=130 MSU Value of the CAP you want to examine INTERVAL=HOUR Must match the ASUMCEC interval Thanks to Chuck Hopf, Independent Consultant, USA.
MULTIPDB with ALL of the datasets that exist in the list of input VGETSORT data libraries to be read. If a dataset is sorted in all VGETLIBS of the input data libraries, the output dataset will also Jul 27, 2010 be sorted. Options for KEEPing or DROPing datasets by Aug 9, 2010 name is supported, and the input list can use a colon as a wildcard to read all DDs starting with that string. // EXEC MXGSAS //PDB1 DD DSN=PDB(0),DISP=SHR //PDB2 DD DSN=PDB(-1),DISP=SHR //PDB3 DD DSN=PDB(-2),DISP=SHR //PDB4 DD DSN=PDB(-3),DISP=SHR //MON DD DSN=PDB.MON,DISP=SHR //TUE DD DSN=PDB.TUE,DISP=SHR //NEW DD DSN=PDB.COMBINED,DISP=(,CATLG)...... %MULTIPDB(INLIBS=PDB: MON TUE,OUTLIB=NEW); will create all datasets found in all of the INLIBS= into the //NEW data library. -VGETSORT added parameters for efficiency and to reduce the number of messages written to the SAS log. -New VGETLIBS %MACRO will take a list of LIBNAMEs and see if they exist and what engine they were created by. The list of libnames and engines are returned, and used by MULTIPDB to determine if the INLIBS= libnames exist. If none of the LIBNAMES in INLIBS= are found, MULTIPDB terminates. Change 28.167 Variables NDMCERT and NDMCERI are added to NDMCT dataset. VMACNDM
Jul 27, 2010 Change 28.166 Revised, enhanced support for TYPE113 SMF records, based ASUM113 on John Burg's SHARE Boston 2010 Presentation. FORMATS -PDB.TYPE113 dataset contains the complete detail level, VMAC113 with an observation for each CPU Address, while the new TYPS113 PDB.ASUM113 dataset, created by revised ASUM113, has all TYPS113E of the counters summed for each LPAR for each interval; Jul 27, 2010 This PDB.ASUM113-level is used to calculate the RNI for Aug 12, 2010 each LPAR for zPCR; while all the new variables are also created in TYPE113 dataset, IBM recommends that the LPAR summary PDB.ASUM113 data be normally used. -The ASUM113 program will look for PDB.TYPE70PR dataset in the same PDB library as TYPE113, and if found, it is used to add variables CPUTYPE CECSER LPARWLMG and SMF70HDM in both PDB.TYPE113 and PDB.ASUM113. -The TYPS113E "Enhanced" standalone program reads 70 & 113 SMF to create both PDB.TYPE113 and PDB.ASUM113 with the TYPE70PR variables added. - New RNI metric is created in TYPE 113, additional metrics are calculated, and SM113VN2 is formatted to identify if the processor type is z10 or a new z196. -SM113VN2=1 for z10, SM113VN2=2 for z196; new MG113VN format will print z10 or z196. -Relative Nest Intensity calculations for the z10 and z196 are added to dataset TYPE113 using these two algorithms, but only the new variable RNI is actually created: Z10RNI =(1.0*L2LP+2.4*L2RP+7.5*MEMP)/100. Z196RNI=1.6*(0.4*L3P+1.0*L4LP+2.4*L4RP+7.5*MEMP)/100 -Users will need to know the RNI to determine the correct LSPR workload for the latest LSPRs, which is documented in the new LSPR Manual on ResourceLink: https://www-304.ibm.com/servers/resourcelink/lib03060.nsf /pages/lsprindexpdf/$file/SC28118714_20100714.pdf -New variables are created in TYPE113: z196 only: L2P ='PERCENT*SOURCED*FROM*L2*CACHE' L3P ='PERCENT*SOURCED*FROM*L3*CACHE' L4LP='PERCENT*SOURCED*FROM*L4*SAME BOOK' L4RP='PERCENT*SOURCED*FROM*L4*DIFFERENT*BOOK' z10 and z196, but different equations for each CEC: ESTICCPI='ESTIMATED*INSTRUCTION*COMPLEXITY*CPI' ESTFINCP='ESTIMATED*CPI FROM*FINITE*CACHE/MEM' ESTSCP1M='ESTIMATED*SOURCING*CYCLES*PER L1 MISS' RNI ='RELATIVE*NEST*INTENSITY' EFFGHZ ='EFFECTIVE*GIGAHERTZ*CYCLES*PER NANO' TLB1MISS='TLB*CPU MISS*PERCENT OF*TOTAL CPU' TLB1CYCL='CYCLES*PER*TLB*MISS' PTEPCTMI='PAGETABLE*ENTRY*PCT OF TLB*MISSES' -These z10-only variables have missing values (i.e., not calculated for z196): L15P L2LP L2RP. (Added 26Aug2011). -SM113CPT was not kept in the TYPE113 dataset. -If a counter reset is found in one counter, that entire interval from that SYSTEM/LPAR is deleted. -The original logic attempted to correct when counters wrapped, by creating the value of a full 8-bytes 'FF'x, to add when wrap was detected: IF EIGHTFFF=. THEN DO; CHAR8FFF='FFFFFFFFFFFFFFFF'X; EIGHTFFF=INPUT(CHAR8FFF,&PIB.8.); RETAIN EIGHTFFF; DROP CHAR8FFF; END; and that worked fine on PC SAS. But on z/OS, the value of EIGHTFFF is ZERO, instead of the expected 1.8E+19 value. SAS Support has recognized this as a defect (that has been that way since SAS V6 on z/OS), so this iteration will just delete intervals if the counter wrapped. -John's SHARE paper had some missing counters for the z196 equations that are included in this MXG update and will be in his final version of that paper: L4LP adds EXTND152 and EXTND155 L4RP adds EXTND134 and EXTND143 MEMP subtracts EXTND134 and EXTND143 -The ANAL113 program print's the report on page 26 of his paper, and also prints the counters that are used to create that report for each LPAR for each interval, using the PDB.ASUM113 dataset as input. -I will update MXG ANALZPCR to use RNI in a later change. -APAR OA30486 became available in Oct, 2010, back to z/OS 1.10, adding SMFINTVAL=SYNC so that SMF 113 records can be synchronized with your SMF and RMF interval data. Thanks to Cheryl Watson, Watson & Walker, USA. Thanks to Scott Barry, SBBWorks, USA. Thanks to John Burg, IBM, USA. Thanks to Tony Hirst, Wells Fargo, USA. Thanks to Richard Schwartz, State Street Bank, USA.
VMACIMS caused INPUT STATEMENT EXCEEDED RECORD LENGTH. The DO Jul 26, 2010 group for the '09'x subtype was incorrectly located. Thanks to David Schumann, Blue Cross Blue Shield of Minnesota, USA. Change 28.164 Support for SMF 119 new subtypes 32-37 and 48-52. EXT11932 Nineteen new datasets are created (but not yet data'd). EXT11933 EXT11934 dddddd Dataset Description EXT11935 T11932 TYP11932 DVIPA STATUS CHANGE EXT11936 T11933 TYP11933 DVIPA REMOVED EXT11937 T11934 TYP11934 DVIPA TARGET ADDED EXT11948 T11935 TYP11935 DVIPA TARGET REMOVED EXT11949 T11936 TYP11936 DVIPA TARGET SERVER STARTED EXT11950 T11937 TYP11937 DVIPA TARGET SERVER ENDED EXT11951 T11948 TYP11948 CSSMTP CONFIGURATION DATA EXT11952 T11949 TYP11949 CSSMTP TARGET SERVER CONNECTION EXT11973 T11950 TYP11950 CSSMTP MAIL EXT11974 T11951 TYP11951 CSSMTP SPOOL EXT11975 T11952 TYP11952 CSSMTP STATISTICS EXT11976 T11973 TYP11973 IPSEC IKE TUNNEL ACTIVATE EXT11977 T11974 TYP11974 IPSEC IKE TUNNEL DEACTIVATE EXT11978 T11975 TYP11975 IPSEC DYNAMIC TUNNEL ACTIVATE EXT11979 T11976 TYP11976 IPSEC DYNAMIC TUNNEL DEACTIVATE EXT11980 T11977 TYP11977 IPSEC DYNAMIC TUNNEL ADDED VMAC119 T11978 TYP11978 IPSEC DYNAMIC TUNNEL REMOVED VMXGINIT T11979 TYP11979 IPSEC MANUAL TUNNEL ACTIVATE Jul 25, 2010 T11980 TYP11980 IPSEC MANUAL TUNNEL DEACTIVATE
VMACTMDB structure; the original code did not define the _Vdddddd Jul 23, 2010 macros, and the _Sdddddd dataset sort macros were DATA steps rather than PROC SORTS. Thanks to Ernie Amador, UC Davis Health System, USA.
VMACITRF datasets to support ITRF V420 IF2. Jul 22, 2010
TIMEBILD the %TIMEBILD invocations. The execution just hung with Jul 21, 2010 no error message nor clue, except for an astute user. Thanks to Graham Cornwall, Donovan Data Systems, ENGLAND. Change 28.160 -CleverView for TCP/IP TN3270 variable CTCPIPAD was wrong, VMACCTCP with constant value of 64.64.64.64, now corrected. Jul 21, 2010 -The CVTTZ GMT Offset field was added to the SMF record, and when it exists is used to correct the CTCPTIME STCK value to the local clock. CTCPTIME is the Start Time of the session and will be constant for all intervals for that session. -The interval records cannot be SYNC'd with SMF intervals. -The CTCPHOST (HOST APPLICATION NAME) can be blank, when the user has not yet logged on. Thanks to John Howard, Florida Northwood Shared Center, USA.
GDGDAILY did not pick up yesterday's SPIN data library. Jul 20, 2010
VMAC42 many retries were performed to obtain buffer latches to Jul 19, 2010 satisfy record management requests. From APAR text: -If a client has a serious application contention issue while accessing buffers to read data, there is no external indication or measurement data to show this reason for delay. -While primary serialization in RLS is at the record level, there is still a need for SMSVSAM to perform very short term serialization of the local buffer containing the CI in order to maintain integrity of the local buffer pool. -When contention is excessive, a single record management request may need to retry an obtain for the buffer latch many times. A small amount of contention is normal, so displaying latches at a given point in time does not give an indicator of how long a request may have been delayed. Variables SMF42GRW and SMFA2GRW are added to datasets TYPE42D1 and TYPE42D3 respectively. CHANGE 28.157 Support for VM ACCOUNT ID='09' for ISFC (Inter-System EXVMISFA For Communications) record creates five datasets: EXVMISFE EXVMISFI dddddd Dataset Description EXVMISFL VMISFA VMISFA ISFC ACTIVE CONVERSATION EXVMISFS VMISFE VMISFE ISFC END CONVERSATION FORMATS VMISFI VMISFI ISFC INITIATE CONVERSATION VMXGINIT VMISFL VMISFL ISFC LINK STATISTICS TYPEVM VMISFS VMISFS ISFC START CONVERSATION Jul 19, 2010 Thanks to Greg Reed, Travelport GDS, USA. Thanks to Rick Stuchell, Travelport GDS, USA. CHANGE 28.156 Support for IFCID=27 decodes QW0027SP and QW0027NR and FORMATS decodes and formats QW0027OZ. VMAC102 Jul 18, 2010 Thanks to Victoria Lepak, Aetna, USA. CHANGE 28.155 MXG 28.04, only if _N7072 is used. The output of the new VMAC7072 temporary TYPE70EC and TYPE70EL datasets now names their Jul 17, 2010 _WTY70EC and _WTY70EL tokens so that _N7072 can be used. Thanks to Kim Westcott, State of New York, USA. CHANGE 28.154 -MXG detection of inserted or excluded fields in CICS 110 VMAC110 is enhanced by test for CPUTM GT 2*ELAPSTM that prints Jul 18, 2010 error messages if true. Previous IF TASKNR=. tested the Jul 21, 2010 only Packed Decimal field, which is a missing value if MXG was out of alignment, but that could overlook new inserted fields after the location of Task Number. -Spurious message that STID=110 had skipped data removed; the SKIP=SKIP-604 should have been SKIP=SKIP-652, but the CICW2R dataset was correct in spite of that message. -Variable PGDPGM, Program Name, was not kept in dataset CICPGD, nor was it labeled, a major oversight as all of the information in CICPGD Statistics are for the program! Thanks to Victoria Lepak, Aetna, USA. Thanks to Gerard Bosker, Rabobank Nederland, THE NETHERLANDS. Thanks to Don Deese, (CPExpert), Computer Management Sciences, USA.
VMACXAM
Aug 15, 2010 ====== Changes thru 28.152 were in MXG 28.04 dated Jul 5, 2010======== CHANGE 28.152 Support for zTPFC, TPF Continuous Monitoring has a few EXTPFC92 fields added to existing datasets and two new datasets EXTPFC98 added by this change IMACTPFC
VMACTPFC DDDDDD DATASET Description Jul 2, 2010 Jul 21, 2010 TPFC92 TPFC92 LPAR UTILIZATION TPFC98 TPFC98 DASD SERVICE TIME Thanks to Bob Wilcox, HP, USA.
EXZCO01 record, which replaces the TDSL product with these three EXZCO02 new datasets: EXZCO03
IMACZCOS dddddd dataset description TYPEZCOS
TYPSZCOS ZCO01 ZCOS01 ZCOST CPC VMACZCOS ZCO02 ZCOS02 ZCOST LPAR VMXGINIT ZCO03 ZCOS03 ZCOST SYSPLEX/CPC Jul 2, 2010 Thanks to Alan Delaroche, zCostServices, FRANCE.
VMAC42 addition of variables SMF42FAI and SMF42FAJ caused the Jul 1, 2010 second and subsequent segments to be misaligned with bad data values resulting in TYPE42S1 dataset. Thanks to Michael Friske, FMR, USA.
ANAL119
Jun 30, 2010 CHANGE 28.148 TMS DEVTYPE was blank for TRTCH 'F0'X thru 'FF'x, which VMACTMS5 are now decoded as 3592 or Titanium devices. Jun 28, 2010 Thanks to Scott Barry, SBBWorks, USA. CHANGE 28.147 Another kewl tool. Will search data libraries for all VMXGSRCH variables that match a specific character string. Jul 4, 2010 You can: -PRINT X OBS of every dataset where any variable has a match -PRINT X OBS of every dataset where any variable has a match and copy all of the OBS to some other DD -or just get a count of the number of variables and datasets and obs that have a match.
the value of any variable is CHUCK. %VMXGSRCH(LIBNAME=_ALL_,RESULTS=PRINT,VALUE=CHUCK);
the value of any variable is CHUCK. %VMXGSRCH(LIBNAME=_ALL_,RESULTS=PRINT,VALUE=CHUCK, COPYTO=WORK); Example 3: PRINT/copy ALL OBS IN ALL DATASETS WHERE the value of any variable is CHUCK, but suppress the NOTES on the log (and there are LOTS of them). %VMXGSRCH(LIBNAME=_ALL_,RESULTS=PRINT,VALUE=CHUCK, COPYTO=WORK,LOG=NO);
in how many datasets and obs have a match. %VMXGSRCH(LIBNAME=_ALL_,RESULTS=COUNT,VALUE=CHUCK, LOG=NO);
Thanks to Chuck Hopf, Independent Consultant, USA. Thanks to Scott Barry, SBBWorks, USA. Change 28.146 -Major enhancement to DB2PM-like reporting in ANALDB2R ANALDB2R which has been upgraded to support DB2 V9 data, and is ASUMDBAA internally completely revised to make maintenance simple EXDB2ACR in the future. FIXDB2A -And, a MAJOR change in PDB.DB2ACCT Buffer Pool data: TRNDDBAA Four sets of DB2 Buffer Pool variables QBnCxxx in DB2ACCT VGETOBS DB2ACCTP and DB2STATS are redefined and contents changed. VMACDB2 QB1Cxxx variables contain the 4K Buffer Size counters, VMACDB2H QB2Cxxx variables contain the 8K Buffer Size counters, VMXGDBAA QB3Cxxx variables contain the 16K Buffer Size counters VMXGDBSS and QB4Cxxx variables now contain 32K Buffer counters. Jul 4, 2010 Originally, the four sets of counters were for BP0, BP1, Aug 14, 2010 all 4K, and all 32K, but then 8K and 16K counters were added into the QB4Cxxx variables, making a real mess. Yüklə 28,67 Mb. Dostları ilə paylaş:
|