Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.


Topics - FallenTree

Pages: 1 2 [3] 4 5
31
Drivers / fluffos 3.0-alpha3
« on: May 06, 2013, 04:23:38 AM »
Download at  https://github.com/fluffos/fluffos/archive/fluffos-3.0-alpha3.tar.gz

Code: [Select]
git clone https://github.com/fluffos/fluffos
cd fluffos
git checkout fluffos-3.0-alpha3

Please submit issue to https://github.com/fluffos/fluffos

Changelog

https://github.com/fluffos/fluffos/blob/next-3.0/ChangeLog.fluffos-3.x

FluffOS 3.0-alpha3
  FluffOS has switched to c++ language.
  use try/catch instead of longjmp. (wodan)
  Code quality improvement.
  Fix using DEBUG without DEBUGMALLOC_EXTENSIONS cause memory corruption.

32
Drivers / fluffos 3.0-alpha2
« on: April 29, 2013, 05:42:19 AM »
My previous 2.27-sunyc-rc1 is renamed to 3.0-alpha1 per decision by wodan, from now on I will be releasing 3.x version for now.

Download at  https://github.com/fluffos/fluffos/archive/fluffos-3.0-alpha2.tar.gz
Code: [Select]
git clone https://github.com/fluffos/fluffos
cd fluffos
git checkout fluffos-3.0-alpha2

Please submit issue to https://github.com/fluffos/fluffos

Feature Status:
  64bit runtime on all platform (done)
  switch to clang/llvm c++ compiler
  switch to autoconf
  switch to TCMalloc
  get rid of edit_source
  get rid of addr_server
  LPC JIT compiler
  multi-threading


FluffOS 3.0-alpha2

General:
  Rebased to 2.27 released by wodan.
  Build will fail early when local_options is missing.
  Enforce source format using astyle.

BugFix:
  command() efun will correctly return eval cost.
  Crasher 14, crash when returning array types.

Tests:
  Tests are now executed randomly.
  debugmalloc will fill memory with magic values.

FluffOS 3.0-alpha1

New compile options:
  POSIX_TIMERS: better time preceision tracking for eval cost. (voltara@lpmuds.net)
  CALLOUT_LOOP_PROTECTION: protect call_out(0) loops. (voltara@lpmuds.net)
  SANE_SORTING: Use faster sorting implementation for "sort_array()", but requires
                LPC code to return conforming results.

General:
  Build script improvement and compile/warning fixes.
  Build under 32bit environment is now supported.
  Build under CYGWIN is now supported.
  Multiple crasher/memory leaks is fixed.
  Documentation has been moved to root directory.
  Enable Travis CI to automate test/build for each commit.
  Auto print backtrace dump when driver crash.
  Print warning on startup if core dump limit is 0.
  Compile fix for db.c without PACKAGE_ASYNC. (mactorg@lpmuds.net)
  General code quality improvements.

Packages:
  PACKAGE_CRYPTO: build fixes and enhancements. (voltara@lpmuds.net)
  PACKAGE_SHA1: Fix incorrect sha1() hash generation, verified with tests.
                (voltara@lpmuds.net)

Test:
  "make test" will launch testsuite and report any problems.
  DEBUGMALLOC, DEBUGMALLOC_EXTENSIONS and CHECK_MEMORY is working now.
  Extensive 64bit runtime tests is added.
  Switch operator tests.
  Benchmarker and auto-crasher improvments.

LPC:
  LPC runtime is strictly 64bit now and everything should conform.
  MIN_INT, MAX_FLOAT, MIN_FLOAT predefines.
  Maximum number of EFUNs has been raised to 65535 from 256.

Known Issues:
  "-MAX_INT" is not parsed correctly in LPC(old bug), see
    src/testsuite/single/tests/64bit.c for details.
  unique_mapping() EFUN will leak memory.
  crasher in testsuite needs improvements.

33
Drivers / fluffos 2.27-sunyc-rc1
« on: April 22, 2013, 04:51:09 AM »
You can get it from here: https://github.com/fluffos/fluffos/archive/fluffos-2.27-sunyc-rc1.tar.gz

Compat:
  sha1() will now return correct hash, if your were using broken one,you need to figure out way to update the hashes.
  LPC runtime is strictly 64bit, large int /float will be scannf ,printf correctly, if you were relying on broken behavior before, you need to take care.
  SANE_SORTING, see change log for details.

Changes Log: https://github.com/fluffos/fluffos/blob/fluffos-2.27-sunyc-rc1/ChangeLog.fluffos-2.x

New compile options:
  POSIX_TIMERS: better time preceision tracking for eval cost. (voltara@lpmuds.net)
  CALLOUT_LOOP_PROTECTION: protect call_out(0) loops. (voltara@lpmuds.net)
  SANE_SORTING: Use faster sorting implementation for "sort_array()", but requires
                LPC code to return conforming results.

General:
  Build script improvement and compile/warning fixes.
  Build under 32bit environment is now supported.
  Build under CYGWIN is now supported.
  Multiple crasher/memory leaks is fixed.
  Documentation has been moved to root directory.
  Enable Travis CI to automate test/build for each commit.
  Auto print backtrace dump when driver crash.
  Print warning on startup if core dump limit is 0.
  Compile fix for db.c without PACKAGE_ASYNC. (mactorg@lpmuds.net)
  General code quality improvements.

Packages:
  PACKAGE_CRYPTO: build fixes and enhancements. (voltara@lpmuds.net)
  PACKAGE_SHA1: Fix incorrect sha1() hash generation, verified with tests.
                (voltara@lpmuds.net)

Test:
  "make test" will launch testsuite and report any problems.
  DEBUGMALLOC, DEBUGMALLOC_EXTENSIONS and CHECK_MEMORY is working now.
  Extensive 64bit runtime tests is added.
  Switch operator tests.
  Benchmarker and auto-crasher improvments.

LPC:
  LPC runtime is strictly 64bit now and everything should conform.
  MIN_INT, MAX_FLOAT, MIN_FLOAT predefines.
  Maximum number of EFUNs has been raised to 65535 from 256.

Known Issues:
  "-MAX_INT" still not being parsed correctly in LPC, see
    src/testsuite/single/tests/64bit.c for details.
  unique_mapping() EFUN still leaks memory.
  crasher in testsuite needs improvements.

34
Drivers / Thoughts on licensing issue
« on: April 01, 2013, 03:09:02 AM »
Hi,

I want to know what you guys thinks about the license issue around the driver. I know there has been effort to make it GPL by convincing the owners but failed.

Do you guys think licensing is what blocks future development? The real restrictions in current driver license is that you can't use it for profit. Which is arguably a good driving fact that many open source project receive support and survive.

I had a thought about converting driver code to either MIT or GPL by incrementally re-write the old pieces (it needs a rewrite anyway). What do you guys think? This will be a good move? This will draw more people to work on driver?

Cheers.

35
Drivers / 64bitmalloc : cannot allocate memory
« on: March 25, 2013, 07:39:05 AM »
Wodan / anyone else running 64bitmalloc

Have you met this problem in 2.26?

mmap failed! (55): Cannot allocate memory
where = 117f8193000
mmap failed! (55): Cannot allocate memory
where = 117f8193000
mmap failed! (55): Cannot allocate memory
where = 117f8193000
mmap failed! (55): Cannot allocate memory
where = 117f8193000
.....
Then
malloc: : Cannot allocate memory

I checked ulimit -a which allow unlimited virtual memory, no idead why it start fail to allocate memory.

36
Drivers / Add your EFUN package.
« on: March 18, 2013, 12:04:08 AM »
Now that 256 EFUNs limits has been removed, and automatic build test is in place. I'm happy to include any self-maintained efuns as packages into the next-2.27 tree.

This is of course my attempt to enhance collaboration and reduce the number of self-maintained patches, complexity of future upgrades.   

You can send me patches, url,  files, or send pull-request directly at github  https://github.com/fluffos/fluffos

The only requirement I have is that it must be able to build and pass existing tests. Even better to add more tests into testsuite!

Cheers!

37
Code Vault / Is there a vim or equivalent for mud use?
« on: March 17, 2013, 08:26:55 PM »
I found it astonishing to still have to use "ED" to do any sort of in game editing.

38
Drivers / 2.27 ?
« on: March 14, 2013, 05:10:17 PM »
Hey wodan, while the patches are accumulating on 2.27  , what's your plan ? do you have changes now , do you still want us to generate patches for you to release and merge ?  or do you rather just include your patches against 2.26.1 towards 2.27 altogether? no hurry on release though, just want to get ready when time comes.

Thanks.

39
Drivers / compile driver in C++ mode
« on: March 06, 2013, 03:45:14 PM »
Is there concerns about switch driver to compile in C++ mode?  then it's possible to use things like STL, boost or existing libraries to replace the outdated part in the driver.  My preliminary testing of compile use g++ works perfectly fine. I'm not sure whether people would be comfortable with that.

40
Drivers / proposal for multi-threaded comm changes.
« on: March 04, 2013, 05:35:16 PM »
Hi,

there has been some effort in Chinese fluffos forks to utilize epoll instead of select, where I think it's not solving the root problem. The root problem is that all communicates are handled in main thread.  takes precious time for your main mud loop have to finish each heartbeat.

I've been thinking of a way to separate them. so far I think what can be done is:

1) run a libevent dispatch loop in separate thread for all socket communicates and compression etc.  The libevent loop would send out/receive stuff as needed and saved in its own datastructure waiting for main loop to get it.  It's like a messaging queue but actually can be just implemented as lock-protected linked-list, or something like that.

2) main loop still calls proceessio() routinely,but it is merely memcpy back-and-forth the data main loop (interactive_t) and data structure in the communication thread. This can be implemented with locks, but I rather look for a lock-free solution.

The idea is to make minimal changes to current code to first move communication stuff to separate thread first, and no behavior changes.  then when that is done and tested and trusted,  try to unify the two data structures so that no memcpying is needed.

I would like to hear your opinions.

41
Drivers / Merge branch open for fluffos 2.27
« on: March 03, 2013, 11:53:57 PM »
Please send your pull request to

https://github.com/fluffos/fluffos/tree/next-2.27

Issues are welcomed there too.

If you would rather just send patch file, please reply here and I will create a pull request for you.

All submission subject to wodan's acceptance :-p

42
Drivers / https://github.com/fluffos/fluffos
« on: February 27, 2013, 04:59:08 AM »
actually I've taken the liberty of creating

https://github.com/fluffos/fluffos

let's just use that as the official repo, currently at 2.24.  there is a few cosmetic changes I want to do like updating the README,  use github's wiki feature to show all efuns's doc etc etc.  let's wait for wodan's 2.26 release.

If someone else want access, let me know.  (that is for push, you can always just send pull-request).

43
Drivers / one more patch before next release
« on: February 27, 2013, 04:44:49 AM »
From 9aa7fc9f857dfe822579591c0136159b34b0fdf1 Mon Sep 17 00:00:00 2001
From: Yucong Sun <sunyucong@gmail.com>
Date: Thu, 21 Feb 2013 22:30:01 -0800
Subject: [PATCH] Adding class stats default in options.h

---
 src/options.h |    1 +
 1 file changed, 1 insertion(+)

diff --git a/src/options.h b/src/options.h
index d846c82..0704b41 100644
--- a/src/options.h
+++ b/src/options.h
@@ -317,6 +317,7 @@
  * Similarly for arrays ...
  */
 #define ARRAY_STATS
+#define CLASS_STATS

 /* LOG_CATCHES: define this to cause errors that are catch()'d to be
  *   sent to the debug log anyway.
--
1.7.9.5

44
Drivers / ETA on 2.25
« on: February 26, 2013, 03:21:27 AM »
wodan: do you have a ETA on releasing 2.25?  I think I would hold back large work for now until your new release.

45
Drivers / fluffos.org
« on: February 20, 2013, 06:12:38 PM »
So.. any one know who hold fluffos.org? any plan to make use of it? If possible I would like to volunteer to use it to host doxgen . maybe wiki of efuns etc etc.  and I have the server to host it. well, I guess I can alwasy start on any domain.

Domain Name:FLUFFOS.ORG
Created On:31-Oct-2007 13:02:59 UTC
Last Updated On:01-Nov-2012 01:27:12 UTC
Expiration Date:31-Oct-2013 13:02:59 UTC

Registrant ID:ACT24-402153
Registrant Name:Bjorn Andersson
Registrant Organization:Bjorn Andersson
Registrant Street1:Vastmannagatan 20
Registrant Street2:
Registrant Street3:
Registrant City:Orebro
Registrant State/Province:
Registrant Postal Code:703 42
Registrant Country:SE
Registrant Phone:+46.196117649
Registrant Phone Ext.:
Registrant FAX:
Registrant FAX Ext.:
Registrant Email:viking@duds.org

Pages: 1 2 [3] 4 5