home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.amiga.programmer
- Path: sparky!uunet!gatech!concert!sas!mozart.unx.sas.com!walker
- From: walker@twix.unx.sas.com (Doug Walker)
- Subject: Re: Unusual SAS6.0 results
- Originator: walker@twix.unx.sas.com
- Sender: news@unx.sas.com (Noter of Newsworthy Events)
- Message-ID: <BxtBBu.L5w@unx.sas.com>
- Date: Mon, 16 Nov 1992 14:09:30 GMT
- References: <BxrwC8.3AG@fc.hp.com>
- Nntp-Posting-Host: twix.unx.sas.com
- Organization: SAS Institute Inc.
- Lines: 72
-
-
- In article <BxrwC8.3AG@fc.hp.com>, koren@fc.hp.com (Steve Koren) writes:
- |> 2. GENPROTO mode crashes the machine a lot. I get consistant crashes
- |> on 2 out of 30 source files. No big deal though; I can generate the
- |> prototypes by hand.
-
- We fixed a bug in GENPROTO for the upcoming 6.1 patch that may fix
- your problem, but if you want to make sure it gets fixed, contact
- technical support with a test case. You can use the EMITS facility,
- described in the "Getting Help" chapter of the User's Guide, to
- report the problem via email.
-
- |> with the same speed. However, the medium sized executable is not
- |> so happy. It runs as before, but it is *much* slower. I'm talking
- |> about 20X here. It is getting no enforcer hits; I've checked that. Since
-
- No idea what is going on without source code or a runnable program to
- look at. There is nothing in 6.0 that should have caused this.
- For such a dramatic difference, there has to be something different
- in the way the program is linked - different library functions,
- different startup, something.
-
- |> I really wanted to get it back to its old execution speed, I did a
- |> lprof and lstat on it. I'm getting _really_ wierd results. Here is
- |> the 1st few lines from the lstat output:
- |>
- |> 95.4023% 95.4023% f97c 249 _strncpy <- I DON'T CALL THIS!
- |> 0.7662% 96.1685% 7c58 2 @Read_String [lines 290-327 in Lexer.c]
- |> 0.7662% 96.9349% 92d4 2 _yyparse [lines 966-1103 in Parser.y.C]
-
- First, strncpy may be called by a library function that you call.
- Use the MAP and MAPXREF options to generate a linker cross-reference
- and see who is calling strncpy. You'll find some lines in the cross-
- reference like the following:
-
- symbol (Def) module Off (Ref) Module Off (Ref) Module Off (Ref) Module
- _strncpy strncpy ### name1 ### name2 ### name3
- ### name4 ...
-
- The name1...name<n> are the places that refer to strncpy. Somebody
- refers to it or it wouldn't be linked in.
-
- If this doesn't fix the problem, it's possible that this is an lstat
- bug. Again, contact technical support.
-
- |> BTW, the program which is running 20X slower basically does this: some
- |> FGets() calls, a bit of processing to build a parse tree out of that
- |> data, and an execution of that parse tree. The rough times I see are:
- |>
- |> Old 5.10b executable: 1 second
- |> New 6.0 execeutable: 15 to 20 seconds
- |>
- |> - steve "confused" k
-
- There is one possibility that showed up with the Hack Lite source while
- we were testing. 6.0 uses shared libraries for speed. Sometimes
- what happens on an A3000 or some other machine with a 68020 or
- higher is that all FAST memory gets used up by the compiler libraries,
- but since CHIP memory is still available, the system doesn't flush
- the libraries. Try doing an AVAIL FLUSH command before running
- your program. If it returns to its normal speed, you were out of
- FAST ram.
-
- --
- *****
- =*|_o_o|\\=====Doug Walker, Software Distiller====== BBS: (919)460-7430 =
- *|. o.| || 1200/2400/9600 Dual
- | o |// For all you do, this bug's for you!
- ======
- usenet: walker@unx.sas.com bix: djwalker
- Any opinions expressed are mine, not those of SAS Institute, Inc.
-
-