Gpc January 2010

gpc@gnu.de

1 participants
3 discussions

Inefficiency of `mod'
by Emil Jerabek 25 Oct '10

25 Oct '10

Hi, during some debugging I noticed that GPC generates very hairy code for the `mod' operator with signed arguments. A closer look revealed that the code contains a lot of branching on conditions which can be determined on compile time, and most surprisingly, the value of X mod Y is recomputed _twice_, if X is negative. (I've attached a commented assembler dump.) This may be partially a back-end problem (CSE?), but I feel there's something wrong in the relevant code in gpc-common.c (build_pascal_binary_op) as well. `exp1 mod exp2' gets translated into something like exp2 <= 0 ? error : exp1 >= 0 ? exp1 mod exp2 : ((-exp1) mod exp2) == 0 ? 0 : exp2 - (-exp1) mod exp2 This arranges that both arguments supplied to the "internal mod" are non-negative, but this information is not passed on -- the arguments are still declared as signed types, which apparently makes the back-end to produce the mess it does. I've done a little test: the following replacement for `mod', which differs from the built-in one essentially just by the type-casts to `Cardinal', inline operator xmod (X, Y: Integer) = Z: Integer; begin if Y <= 0 then RuntimeError (714); if X >= 0 then Z := Cardinal (X) mod Y else begin Z := Cardinal (-X) mod Y; if Z <> 0 then Z := Y - Z end end; indeed generates a better code, it measures about 15% faster for positive X and more than twice faster for negative X. YMMV. If this is possible in user code, it should work in the compiler internals as well :) The result type of the whole `mod' expression looks dubious too. IIUIC, `exp1 mod exp2' inherits the type of exp1. Now, the result of `mod' is guaranteed to be nonnegative and to fit in the type of exp2 (or more precisely, it is in 0 .. High (type of (exp2)) - 1), whereas it needn't have anything to do with the type of exp1: consider var X, Z: Integer; Y: LongInt; begin X := -8; Y := High (LongInt); Z := X mod Y { -> overflow } end Or am I missing something? Emil (BTW, where is fjf434c.pas?)

10 16

GPC support for AIX 6?
by Gert Doering 16 Feb '10

16 Feb '10

Hi Gnu Pascal list, we're currently using IBM's xpc pascal compiler on IBM AIX 5.3. New machines are being shipped with AIX 6, and IBM's xpc is no longer supported there - for the time being, it actuall works (the xpc frontend generated intermediate code, and the second stage compiler still groks it). Management is not happy about this, as it might break any day, and I have been asked to "provide GPC" on AIX 6. Googling seems to suggest that the earliest release of GCC to work on AIX 6 is gcc 4.2.4 This mailing list's archive suggests that GPC and GCC 4.2 do not yet work together, due to larger efforts required due to backend changes (Waldek's mail from March 12, 2008). I reported this to management ("there is no GPC on AIX 6"), and they came back with the question "if we pay one of the GPC developers to do a GPC port to a recent GCC, to support AIX 6, how much would that cost?". Here we go - I'm posing this question to you. Is one of you capable and willing to make GPC work on AIX 6? And if yes, what would your conditions be (estimated time to do this, estimated costs)? (I can't promise anything, but will report back to management and they will decide...) thanks, gert -- Gert Doering gd(a)medat.de MEDAT Muenchen Tel: 089/126808-0 Fax: -50

7 13

CWSDPMI r7 - 4GB support (Beta Test)
by Rugxulo 14 Jan '10

14 Jan '10

Charles Sandmann writes: "I would like to announce the beta for CWSDPMI release 7. The purpose of this release is to allow high performance access to large amounts of physical memory (up to 4GB). Memory above 32-bit address space (common on machines over 3GB physical memory) will be mapped to fill the address hole to provide a full 4GB memory. 4MB pages will be automatically used for large memory requests. Release 7 also includes the bug fixes included in the release 5 2008 refresh, and improved 4MB page support included in the never completed r6. Current DJGPP releases support large memory allocation requests and with CWSDPMI release 7 allow a contiguous memory block over 4091MB. This CWSDPMI release is dedicated to those few dozen people who want to prove their 4GB+ computers can actually run 4GB 32-bit applications. For those who need more than 4GB, I'll provide some code that allows you to map 5GB+ memory (up to 64GB) into the address space on request. The next release will probably be when I get several people bugging me that 64GB isn't enough under DOS :-) Binary Distribution: http://clio.rice.edu/djgpp/beta/csdpmi7b.zip Source Distribution: http://clio.rice.edu/djgpp/beta/csdpmi7s.zip Please send me any comments or bug reports. Have fun, and thanks for testing !"

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

Gpc January 2010