Re: [Mono-devel-list] DataView.Sort behaves different on mono than Windows
Hey,
This scenario perfectly works fine for me here. Find the attached
sample code. it works as expected. I use mono 1.1.7
Let me know whether this works for you too. If not , I suggest to
upgrade ;-). If you have any other specific situation, please post a
simple code so that I can look into it.
suresh.
On Thu, 2005-04-14 at 16:12 -0700, Joe Audette wrote:
> Hi,
>
> In my new RSS Feed Aggregator for mojoportal, I have a
> DataView named entries that I sort using:
> entries.Sort = "PubDate DESC";
>
> On Windows it sorts by date desc as expected, but
> under mono 1.1.5 from svn on Suse 9.2 it sorts
> descending on date but keeps things grouped together
> by author so newer dated posts are not always on top.
> The feed with the newest post shows first but then all
> of its posts are shown next in date descending order
> no matter if the other feeds have posts that are
> newer, then the feed with the next newest post and so
> on.
>
> I wonder if it has anything to do with the sequence of
> columns in the DataTable underlying the DataView. In
> my DataTable the PubDate field is the last column. I
> may try some more experimentation and see if I can
> find a workaround but I believe this is a bug and
> thought I should mention it.
>
> Thanks,
>
> Joe
>
> joe_audette [at] yahoo dotcom
> http://www.joeaudette.com
> http://www.mojoportal.com
> ___
> Mono-devel-list mailing list
> Mono-devel-list@lists.ximian.com
> http://lists.ximian.com/mailman/listinfo/mono-devel-list
using System;
using System.Data;
class MainClass
{
public static void Main (string [] args)
{
DataTable dt = new DataTable ("test");
dt.Columns.Add ("author", typeof (string));
dt.Columns.Add ("id", typeof (int));
dt.Columns.Add ("pubdate", typeof (DateTime));
dt.Rows.Add (new object [] {"suresh", 5, new DateTime (2005,3,3)}) ;
dt.Rows.Add (new object [] {"suresh", 10, new DateTime (2005,3,6)}) ;
dt.Rows.Add (new object [] {"ramesh", 6, new DateTime (2005,3,2)}) ;
dt.Rows.Add (new object [] {"ramesh",12, new DateTime (2005,1,2)}) ;
dt.Rows.Add (new object [] {"kamesh", 3, new DateTime (2005,3,5)}) ;
dt.Rows.Add (new object [] {"raju", 2, new DateTime (2005,2,5)}) ;
Console.WriteLine ("DataTable : ");
Print (dt);
DataView dv = new DataView (dt, "author like '%sh'", "pubdate", DataViewRowState.CurrentRows);
Console.WriteLine ("\nfiltered sorted view :");
Print (dv);
dv.Sort = "pubdate desc";
Console.WriteLine ("\nsorting dec:");
Print (dv);
}
public static void Print (DataTable dt)
{
for (int i = 0 ; i < dt.Columns.Count ; i++) {
Console.Write ("{0,20}", dt.Columns [i].ColumnName);
}
Console.WriteLine ();
foreach (DataRow r in dt.Rows) {
for (int i =0 ; i < dt.Columns.Count; i++)
Console.Write ("{0,20}", r [i].ToString ());
Console.WriteLine ();
}
}
public static void Print (DataView dv)
{
for (int i = 0 ; i < dv.Table.Columns.Count ; i++) {
Console.Write ("{0,20}", dv.Table.Columns [i].ColumnName);
}
Console.WriteLine ();
foreach (DataRowView r in dv) {
for (int i =0 ; i < dv.Table.Columns.Count; i++)
Console.Write ("{0,20}", r [i].ToString ());
Console.WriteLine ();
}
}
}
[Mono-devel-list] I have some TAO-OpenGl/mono/linux examples
Hi list!!! I have some examples of TAO-OpenGl on linux with mono, some examples are very simple (a triangle in a GLU-window) and other a little more complex, I don't know if there is somebody interested, and may be somebody can help me to debug some examples, becouse I transferring my VB.NET-TAO examples and of course I writing now in C#. Bye. Mauricio. ___ Mono-devel-list mailing list Mono-devel-list@lists.ximian.com http://lists.ximian.com/mailman/listinfo/mono-devel-list
[Mono-devel-list] DataView.Sort behaves different on mono than Windows
Hi, In my new RSS Feed Aggregator for mojoportal, I have a DataView named entries that I sort using: entries.Sort = "PubDate DESC"; On Windows it sorts by date desc as expected, but under mono 1.1.5 from svn on Suse 9.2 it sorts descending on date but keeps things grouped together by author so newer dated posts are not always on top. The feed with the newest post shows first but then all of its posts are shown next in date descending order no matter if the other feeds have posts that are newer, then the feed with the next newest post and so on. I wonder if it has anything to do with the sequence of columns in the DataTable underlying the DataView. In my DataTable the PubDate field is the last column. I may try some more experimentation and see if I can find a workaround but I believe this is a bug and thought I should mention it. Thanks, Joe joe_audette [at] yahoo dotcom http://www.joeaudette.com http://www.mojoportal.com ___ Mono-devel-list mailing list Mono-devel-list@lists.ximian.com http://lists.ximian.com/mailman/listinfo/mono-devel-list
[Mono-devel-list] Limiting Memory Allocation
I'm currently working on embeding Mono in a virtual world to support user scripting and would like to limit the ammount of memory each script object can allocate. I'd like to be able to intercept each heap allocation and decide whether to allow it on a per-script (object) basis. Is this possible? What's the best approach? Cheers, Jim. Send instant messages to your online friends http://uk.messenger.yahoo.com ___ Mono-devel-list mailing list Mono-devel-list@lists.ximian.com http://lists.ximian.com/mailman/listinfo/mono-devel-list
Re: [Mono-devel-list] Problems with Array.AsReadOnly
It's working now smoothly :-) Carlos. > Hi Carlos, > > I think this is already fixed. I'm running your sample from svn of > 2005-04-13 23:55, localtime GMT-3 and I get the expected result: 10 > > Please, update your svn and give it a try. > > Thanks for reporting. > > > - -- > Regards, > > Francisco Figueiredo Jr. > Npgsql Lead Developer > http://gborg.postgresql.org/project/npgsql > MonoBrasil Project Founder Member > http://monobrasil.softwarelivre.org > > > - - > "Science without religion is lame; > religion without science is blind." > > ~ Albert Einstein > -BEGIN PGP SIGNATURE- > Version: GnuPG v1.2.6 (GNU/Linux) > Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org > > iQEVAwUBQl3fHf7iFmsNzeXfAQI1pgf/QmdAna8YNHC5Z4JjBOjjscQa/iqkVnnx > 1hTNZoxvNEaAyUU0T9l2AmWtrzcIfBay4pACpQ/PYPvQS8t0rVZQUsZVQCF/K4US > 2hl4xyAK72znxvgXxf/t6fDMqhNuxY9y4ItjxZyXzhDkGSxhekUcphpa53YKWqHu > T74YRF/ntZkWMBPmj6RHCMHyezMyYBR0iYfToc3j3SMVZeUt40vGfcBVtVNPER4y > +77ntCn5LGwTPyMDRN5msOOlioYIAbyeuLiX4mKNP47qvtQUBOsQLGvlwqSboOM5 > kVtfhSZZ7/GTr4S7pXPbhJOaAW4fXqjht6W2IdAxZj9CyYv31Br0rg== > =TKr7 > -END PGP SIGNATURE- > > ___ Mono-devel-list mailing list Mono-devel-list@lists.ximian.com http://lists.ximian.com/mailman/listinfo/mono-devel-list
[Mono-devel-list] CodeGenerator doesn't support C Unions
I try to know why there's a problem with Gda.Value; and searching I
found that some one change the code made from CodeGenerator to use
FieldOffset in cero for each element in the union that exist in
GdaValue struct:
typedef struct GdaValue {
GdaValueType type;
union {
gint64 v_bigint;
guint64 v_biguint;
gpointer v_binary;
GdaBlob v_blob;
gboolean v_boolean;
GdaDate v_date;
gdouble v_double;
GdaGeometricPoint v_point;
GObject *v_gobj;
gint v_integer;
GdaValueList *v_list;
GdaMoney v_money;
GdaNumeric v_numeric;
gfloat v_single;
gshort v_smallint;
gushort v_smalluint;
gchar *v_string;
GdaTime v_time;
GdaTimestamp v_timestamp;
gchar v_tinyint;
guchar v_tinyuint;
GdaValueType v_type;
guint v_uinteger;
} value;
}
There's no code to generate a union like in C# using CodeGenerator.
The StructBase object write the code but define:
[StructLayout(LayoutKind.Sequential)]
And in the case of union mus be:
[StructLayout(LayoutKind.Explicit)]
Then may need to create a UnionBase class to write the correct code,
or modify the StructBase class to admit in the constructor an argument
telling him the type of object (struct or union) and write the correct
sentence.
But now, the Field class, must have a metod to write the FiledOffset
attribute (allways in cero) to have a C union in C#.
___
Mono-devel-list mailing list
Mono-devel-list@lists.ximian.com
http://lists.ximian.com/mailman/listinfo/mono-devel-list
Re: [Mono-devel-list] more ppc atomic.h cleanups
I actually do not know which of the Interlocked* API calls are expected to have this behavior, so I did not add it. In particular, I know the InterlockedCompareExchange call has a memory barrier semantic on x86, but that behavior does not seem to be defined for any other platform. - David Waite On 4/14/05, Eyal Alaluf <[EMAIL PROTECTED]> wrote: > Hi, all. > > Shouldn't the interlocked API's implementation use sync & isync for the memory > barriers? > At least in C++, we have seen many cases where people implemented their own > mutexes using interlocked APIs. Now on multiple CPU system and PPC in > particular a mutex needs to flush the memory barriers so two diferent threads > sharing the same data will be synchronized if they use the mutex properly. > > Eyal. > > On Wed, 13 Apr 2005, David Waite wrote: > > > Date: Wed, 13 Apr 2005 12:31:50 -0600 > > From: David Waite <[EMAIL PROTECTED]> > > To: Mono Development List > > Subject: [Mono-devel-list] more ppc atomic.h cleanups > > > > Attached is a patch that performs cleanups of atomic.h for the ppc > > architecture: > > > > * InterlockedIncrement, InterlockedDecrement: use one fewer register, > > avoid initialization and remove extra addition/subtraction at the end > > * InterlockedExchange: remove tmp initialization, rename tmp to result > > to be consistant with other methods > > > > -David Waite > > > > > ___ Mono-devel-list mailing list Mono-devel-list@lists.ximian.com http://lists.ximian.com/mailman/listinfo/mono-devel-list
Re: [Mono-devel-list] Re: [PATCH] Bug 74586 (and other oddities with the beta)
Woops. I thought I had made a successful switch back to mainline from your branch, but checking out a new copy of mainline shows it's fixed. My bad. Nice layer, btw. Bill On 4/14/05, Bill Middleton <[EMAIL PROTECTED]> wrote: > Still broken in the mainline. :| > > Bill > > On 4/14/05, Dick Porter <[EMAIL PROTECTED]> wrote: > > On Thu, 2005-04-14 at 10:10 +0200, Bill Middleton wrote: > > > Here's a better patch. I hope I'm getting closer here... :) > > > > I fixed this yesterday :) > > > > - Dick > > > > ___ > > Mono-devel-list mailing list > > Mono-devel-list@lists.ximian.com > > http://lists.ximian.com/mailman/listinfo/mono-devel-list > > > ___ Mono-devel-list mailing list Mono-devel-list@lists.ximian.com http://lists.ximian.com/mailman/listinfo/mono-devel-list
Re: [Mono-devel-list] Re: [PATCH] Bug 74586 (and other oddities with the beta)
Still broken in the mainline. :| Bill On 4/14/05, Dick Porter <[EMAIL PROTECTED]> wrote: > On Thu, 2005-04-14 at 10:10 +0200, Bill Middleton wrote: > > Here's a better patch. I hope I'm getting closer here... :) > > I fixed this yesterday :) > > - Dick > > ___ > Mono-devel-list mailing list > Mono-devel-list@lists.ximian.com > http://lists.ximian.com/mailman/listinfo/mono-devel-list > ___ Mono-devel-list mailing list Mono-devel-list@lists.ximian.com http://lists.ximian.com/mailman/listinfo/mono-devel-list
Re: [Mono-devel-list] Mono.Data.SqliteClient status
> > there is also another project doing an ado.net provider for 3.x version of > sqlite at sf.net: > Thanks for the pointer! > it is using the bsd license. i wanted to try using it - i got it to compile > with > mono - but have not tested it (with success) yet, it has some capitalization > issues w.r.t. class names, etc which make it not possible to just drop in and > switch the namespaces. and i was too lazy to change my code to make it work, > but Hmmm, I do not see at all the point of dumping the work already done. I mean that there are some gaps in the implementation, but it is a good work indeed. And the C/C++ API is pretty straightforward. But maybe there are some version 3 features that would imply a major code reworks, aren't there? I have been using version 2 both from C++ and Python for quite a few small projects, but I have never tried version 3... Thanks for your prompt reply, Miguel. ___ Mono-devel-list mailing list Mono-devel-list@lists.ximian.com http://lists.ximian.com/mailman/listinfo/mono-devel-list
Re: [Mono-devel-list] Mono.Data.SqliteClient status
I emailed the maintainers/copyright holders not too long ago (and got responses) with the intention of working further on the SqliteClient and have been too busy tending my own code to get started. And yes, in my opinion the SqliteClient has several issues (don't look inside SqliteDataReader, or SqliteCommand :) and when I get the time I will be submitting patches to them and bugzilla. If you are going to keep doing work on this, lets talk off list and split the tasks so we don't end up duplicating each others efforts. On 4/14/05, Miquel Ramírez <[EMAIL PROTECTED]> wrote: > Hi, > > I am developing an application under mono-1.1.5 that relies heavily on > Sqlite and I am finding several unexpected problems. For instance, > ADO.Net parameters are not supported at all, although a patch > implementing that was submitted several months ago: > > http://lists.ximian.com/archives/public/mono-devel-list/2004-June/006314.html > > I have just checked out the class libraries out of svn and the code is > exactly the same... and the question that comes up to my mind is > whether Mono.Data.SqliteClient is currently being actively maintained. > However, that code features several bugs - specially in the parameter > replacement. > > I also found a trivial - and pretty puzzling "Hey, why the heck it is > not rolling back changes!!!" :D - bug in SqliteTransaction.cs, as the > accompanying diff I hope shows. > > Finally, I am having other weird problems, never seen before that I am > not being able at all to replicate - DML statements affecting too much > rows and less columns as any sane person would expect... however I am > unsure this is not entirely my fault. This is putting me in a frame of > mind that I suspect even of my speaker's shadow on the table :S > > Miguel. > > > -- Nick -- ___ Mono-devel-list mailing list Mono-devel-list@lists.ximian.com http://lists.ximian.com/mailman/listinfo/mono-devel-list
Re: [Mono-devel-list] Re: [PATCH] Bug 74586 (and other oddities with the beta)
On Thu, 2005-04-14 at 10:10 +0200, Bill Middleton wrote: > Here's a better patch. I hope I'm getting closer here... :) I fixed this yesterday :) - Dick ___ Mono-devel-list mailing list Mono-devel-list@lists.ximian.com http://lists.ximian.com/mailman/listinfo/mono-devel-list
Re: [Mono-devel-list] Mono.Data.SqliteClient status
hi, there is also another project doing an ado.net provider for 3.x version of sqlite at sf.net: project: http://sourceforge.net/projects/adodotnetsqlite article: http://www.eggheadcafe.com/articles/20040714.asp it is using the bsd license. i wanted to try using it - i got it to compile with mono - but have not tested it (with success) yet, it has some capitalization issues w.r.t. class names, etc which make it not possible to just drop in and switch the namespaces. and i was too lazy to change my code to make it work, but looks interesting as well... regards, --chris On Thu, 14 Apr 2005, Miquel Ramírez wrote: > Hi, > > I am developing an application under mono-1.1.5 that relies heavily on > Sqlite and I am finding several unexpected problems. For instance, > ADO.Net parameters are not supported at all, although a patch > implementing that was submitted several months ago: > > http://lists.ximian.com/archives/public/mono-devel-list/2004-June/006314.html > > I have just checked out the class libraries out of svn and the code is > exactly the same... and the question that comes up to my mind is > whether Mono.Data.SqliteClient is currently being actively maintained. > However, that code features several bugs - specially in the parameter > replacement. > > I also found a trivial - and pretty puzzling "Hey, why the heck it is > not rolling back changes!!!" :D - bug in SqliteTransaction.cs, as the > accompanying diff I hope shows. > > Finally, I am having other weird problems, never seen before that I am > not being able at all to replicate - DML statements affecting too much > rows and less columns as any sane person would expect... however I am > unsure this is not entirely my fault. This is putting me in a frame of > mind that I suspect even of my speaker's shadow on the table :S > > Miguel. > ___ Mono-devel-list mailing list Mono-devel-list@lists.ximian.com http://lists.ximian.com/mailman/listinfo/mono-devel-list
[Mono-devel-list] Mono.Data.SqliteClient status
Hi,
I am developing an application under mono-1.1.5 that relies heavily on
Sqlite and I am finding several unexpected problems. For instance,
ADO.Net parameters are not supported at all, although a patch
implementing that was submitted several months ago:
http://lists.ximian.com/archives/public/mono-devel-list/2004-June/006314.html
I have just checked out the class libraries out of svn and the code is
exactly the same... and the question that comes up to my mind is
whether Mono.Data.SqliteClient is currently being actively maintained.
However, that code features several bugs - specially in the parameter
replacement.
I also found a trivial - and pretty puzzling "Hey, why the heck it is
not rolling back changes!!!" :D - bug in SqliteTransaction.cs, as the
accompanying diff I hope shows.
Finally, I am having other weird problems, never seen before that I am
not being able at all to replicate - DML statements affecting too much
rows and less columns as any sane person would expect... however I am
unsure this is not entirely my fault. This is putting me in a frame of
mind that I suspect even of my speaker's shadow on the table :S
Miguel.
Index: SqliteTransaction.cs
===
--- SqliteTransaction.cs(revision 42995)
+++ SqliteTransaction.cs(working copy)
@@ -100,7 +100,7 @@
try
{
SqliteCommand cmd = _connection.CreateCommand();
- cmd.CommandText = "COMMIT";
+ cmd.CommandText = "ROLLBACK";
cmd.ExecuteNonQuery();
_open = false;
}
Re: [Mono-devel-list] Re: [Mono-patches] r42986 - trunk/mcs/class/System.XML/System.Xml.Serialization
Lluis Sanchez wrote: > In any case, the patch is not correct and I had to revert it. The > encoding of the element names should be done only once when creating the > maps in XmlReflectionImporter and SoapReflectionImporter. There should > be no need to encode names in other classes. No problem. But now the testcase fails again, so I opened a bug in bugzilla: http://bugzilla.ximian.com/show_bug.cgi?id=74606 Regards, Andrew Skiba. ___ Mono-devel-list mailing list Mono-devel-list@lists.ximian.com http://lists.ximian.com/mailman/listinfo/mono-devel-list
Re: [Mono-devel-list] Change font size.
Hi, If you want to do complex things, you will need to use Pango. If all you want is set a title in a bigger font and such, the easiest way is to use either pango with textmarkups (you can add a label with text = "mytext" or something like that and it will show bold) or use a resource file (look at the style and resource gtk classes.) I use the resource file approach. wkr, Anset Pablo Cardona wrote: > Hello, I'm developing an application with gtk#, and I'm trying to > change the fonts that is used in the application, to use diferent > sizes of letters. I didn't find any class or method that changes the > properties of the font used by the application. So I don't know if it > is possible to change the fonts of an application, and if it is > possible wich class can do it. ___ Mono-devel-list mailing list Mono-devel-list@lists.ximian.com http://lists.ximian.com/mailman/listinfo/mono-devel-list
Re: [Mono-devel-list] Re: [Mono-patches] r42986 - trunk/mcs/class/System.XML/System.Xml.Serialization
On dj, 2005-04-14 at 12:26 +0300, Andrew Skiba wrote: > Lluis Sanchez wrote: > > Hi, > > > > I did not aprove this patch, and I need to know why it is needed before > > aproving it (i.e. a test case). > > > > Lluis. > > > > Hi. > > On Sunday I posted a message with an explanation and the testcase to the > dev-list: > http://lists.ximian.com/archives/public/mono-devel-list/2005-April/011474.html I missed that message, probably because it didn't say anything about XmlSerializer in the subject :-) In any case, the patch is not correct and I had to revert it. The encoding of the element names should be done only once when creating the maps in XmlReflectionImporter and SoapReflectionImporter. There should be no need to encode names in other classes. Lluis. > > Andrew Skiba. > ___ > Mono-devel-list mailing list > Mono-devel-list@lists.ximian.com > http://lists.ximian.com/mailman/listinfo/mono-devel-list ___ Mono-devel-list mailing list Mono-devel-list@lists.ximian.com http://lists.ximian.com/mailman/listinfo/mono-devel-list
Re: [Mono-devel-list] Change font size.
My guess is that the gtk-sharp list might have your answers. Subscribe (or browse its archives) at http://mono-project.com/Mailing_Lists or even better: http://lists.ximian.com/mailman/listinfo Bill On 4/14/05, Pablo Cardona <[EMAIL PROTECTED]> wrote: > Hello, I'm developing an application with gtk#, and I'm trying to > change the fonts that is used in the application, to use diferent > sizes of letters. I didn't find any class or method that changes the > properties of the font used by the application. So I don't know if it > is possible to change the fonts of an application, and if it is > possible wich class can do it. > > Thanks for the help. > ___ > Mono-devel-list mailing list > Mono-devel-list@lists.ximian.com > http://lists.ximian.com/mailman/listinfo/mono-devel-list > ___ Mono-devel-list mailing list Mono-devel-list@lists.ximian.com http://lists.ximian.com/mailman/listinfo/mono-devel-list
[Mono-devel-list] Change font size.
Hello, I'm developing an application with gtk#, and I'm trying to change the fonts that is used in the application, to use diferent sizes of letters. I didn't find any class or method that changes the properties of the font used by the application. So I don't know if it is possible to change the fonts of an application, and if it is possible wich class can do it. Thanks for the help. ___ Mono-devel-list mailing list Mono-devel-list@lists.ximian.com http://lists.ximian.com/mailman/listinfo/mono-devel-list
[Mono-devel-list] Re: [Mono-patches] r42986 - trunk/mcs/class/System.XML/System.Xml.Serialization
Lluis Sanchez wrote: Hi, I did not aprove this patch, and I need to know why it is needed before aproving it (i.e. a test case). Lluis. Hi. On Sunday I posted a message with an explanation and the testcase to the dev-list: http://lists.ximian.com/archives/public/mono-devel-list/2005-April/011474.html As nobody replied, 2 days ago I replied to the original message, and wrote, that I'm going to commit. As I see now, I only sent the second message to myself, as I forgot to CC the devlist. So I'm sorry for committing without your approval, it was not on purpose, really. Andrew Skiba. ___ Mono-devel-list mailing list Mono-devel-list@lists.ximian.com http://lists.ximian.com/mailman/listinfo/mono-devel-list
[Mono-devel-list] unref on 1912 called when ref was already 0
Hi, I'm getting messages such as the following: ** (../../../test/Ice/operations/server.exe:25872): WARNING **: : unref on 1912 called when ref was already 0 These happen sporadically, and only on a dual-CPU machine, which makes me think that some kind of race condition is involved. I only see those messages when a process is about to shut down. I have no idea what the messages mean -- can anyone suggest a possible cause? Thanks, Michi. ___ Mono-devel-list mailing list Mono-devel-list@lists.ximian.com http://lists.ximian.com/mailman/listinfo/mono-devel-list
[Mono-devel-list] Re: [PATCH] Bug 74586 (and other oddities with the beta)
I found a much better solution to this bug by looking at the 1.1.16
source. The problem turns out to be that FindFirstFile() doesn't
properly return INVALID_FILE_HANDLE when scanning an empty directory
or pattern that has no matches.
Here's a better patch. I hope I'm getting closer here... :)
Bill
Index: mono/io-layer/io.c
===
--- mono/io-layer/io.c (revision 42986)
+++ mono/io-layer/io.c (working copy)
@@ -2745,7 +2745,7 @@
!FindNextFile (handle, find_data)) {
FindClose (handle);
SetLastError (ERROR_NO_MORE_FILES);
- find_ret = INVALID_HANDLE_VALUE;
+ return(INVALID_HANDLE_VALUE);
}
return (handle);
On 4/14/05, Bill Middleton <[EMAIL PROTECTED]> wrote:
> On 4/13/05, Bill Middleton <[EMAIL PROTECTED]> wrote:
[Gibberish]
___
Mono-devel-list mailing list
Mono-devel-list@lists.ximian.com
http://lists.ximian.com/mailman/listinfo/mono-devel-list
Re: [Mono-devel-list] more ppc atomic.h cleanups
Hi, all. Shouldn't the interlocked API's implementation use sync & isync for the memory barriers? At least in C++, we have seen many cases where people implemented their own mutexes using interlocked APIs. Now on multiple CPU system and PPC in particular a mutex needs to flush the memory barriers so two diferent threads sharing the same data will be synchronized if they use the mutex properly. Eyal. On Wed, 13 Apr 2005, David Waite wrote: > Date: Wed, 13 Apr 2005 12:31:50 -0600 > From: David Waite <[EMAIL PROTECTED]> > To: Mono Development List > Subject: [Mono-devel-list] more ppc atomic.h cleanups > > Attached is a patch that performs cleanups of atomic.h for the ppc > architecture: > > * InterlockedIncrement, InterlockedDecrement: use one fewer register, > avoid initialization and remove extra addition/subtraction at the end > * InterlockedExchange: remove tmp initialization, rename tmp to result > to be consistant with other methods > > -David Waite > atomic-ppc-cleanups.diff Description: Binary data
