2014-08-25 11:25 GMT-04:00 Óscar Fuentes :
>
> On a CPU-intensive application, the 64 bit exe runs 40% slower than
> its 32 bit counterpart. It's a C++ app with a workload typical of a
> compiler.
>
> Are there known slow spots on the 64 bit runtimes?
You mean, other than the fact that all your po
Hey,
2014-08-25 17:25 GMT+02:00 Óscar Fuentes :
> On a CPU-intensive application, the 64 bit exe runs 40% slower than
> its 32 bit counterpart. It's a C++ app with a workload typical of a
> compiler.
>
> Are there known slow spots on the 64 bit runtimes?
Well, there might be some. I can't actual
Ok for apply?
Regards,
Kai
From 14ad36a7c55306745d50963918285ef8d0c9f7d6 Mon Sep 17 00:00:00 2001
From: Kai Tietz
Date: Mon, 25 Aug 2014 17:25:09 +0200
Subject: Add winapifamily-check, set default to RICHEDIT-version 8.0, add
support for version 7 and 8 of richedit
---
mingw-w64-headers/inclu
On a CPU-intensive application, the 64 bit exe runs 40% slower than
its 32 bit counterpart. It's a C++ app with a workload typical of a
compiler.
Are there known slow spots on the 64 bit runtimes?
I'm using latest gcc/runtime as shipped by MSYS2. Tried to profile the
executables but gprof doesn't
Patch ok to apply?
Regards,
Kai
From 801f53b8d222beb11a3316412c336040f72ca4c2 Mon Sep 17 00:00:00 2001
From: Kai Tietz
Date: Mon, 25 Aug 2014 16:55:18 +0200
Subject: Update for missing defines, and correct interface for C++ case
---
mingw-w64-headers/include/richole.h | 88 +++
On 08/25/14 15:01, Kai Tietz wrote:
> Ok for apply?
Looks good to me.
Thanks,
Jacek
--
Slashdot TV.
Video for Nerds. Stuff that matters.
http://tv.slashdot.org/
___
Mingw-w6
Ok for apply?
Regards,
Kai
From b933174e80f83a880439481938f43f020d94005b Mon Sep 17 00:00:00 2001
From: Kai Tietz
Date: Mon, 25 Aug 2014 14:57:52 +0200
Subject: Add winapifamily-check, correct enumerator values, add missing
types and API.
---
mingw-w64-headers/include/restartmanager.h | 260
Kai Tietz 2014-08-25 15:40:
Hi,
> What do you think about this?
My thoughts on the goals.
In my opinion, I think the priorities should be:
1. Ability to output PDB files.
2. Many precompiled libraries and usable GUI installer/package_manager
for it.
3. SEH for 32 bit.
About first and second, I
Hi,
2014-08-22 15:48 GMT+02:00 Adrien Nader :
> On Fri, Aug 22, 2014, JonY wrote:
>> On 8/22/2014 20:57, Adrien Nader wrote:
>> > On Fri, Aug 22, 2014, Adrien Nader wrote:
>> >> Feel free to send your ideas. A few on top of my head:
>> >> - SEH for 32bits
>> >> - Thorough analysis of the status of
JonY writes:
>> Could the compiler enforce "If you link to static libgcc, you should also
>> link libstdc++ statically" or at least warn about it?
>> It would save users such mysterious bugs resulting from having two copies
>> of ligcc around.
>
> Unfortunately, no, binutils doesn't have any noti
Ok to apply?
Regards,
Kai
From 69984b415eb0212ac02e85c4a057e5899347f7c0 Mon Sep 17 00:00:00 2001
From: Kai Tietz
Date: Mon, 25 Aug 2014 11:52:58 +0200
Subject: Add winapifamily-check, correct Win Vista SP1 check, add missing
types ...
---
mingw-w64-headers/include/mxdc.h | 170 +++
On 08/23/14 21:29, Kai Tietz wrote:
> Hi,
>
> this patch resolves fix-mes, adds win 7/8 APIs, fixes alignement issue
> of structures, and adds winapifamily check.
>
> Ok for apply?
Looks good to me.
Thanks,
Jacek
--
Slas
On 8/25/2014 16:50, Yaron Keren wrote:
> Hi Ruben,
>
> Thanks for pointing this out, the problem is indeed that libstdc++ is
> linked to the dynamic ligcc while the program (as above) links to the
> static version of libgcc. There are two seperate copies of libgcc that do
> not share exception inf
On 08/23/14 21:29, Kai Tietz wrote:
> Hi,
>
> this patch resolves fix-mes, adds win 7/8 APIs, fixes alignement issue
> of structures, and adds winapifamily check.
>
> Ok for apply?
Looks good to me.
Thanks,
Jacek
--
Slas
Hi Ruben,
Thanks for pointing this out, the problem is indeed that libstdc++ is
linked to the dynamic ligcc while the program (as above) links to the
static version of libgcc. There are two seperate copies of libgcc that do
not share exception information resulting in uncaught exception.
If -stat
2014-08-24 23:48 GMT+02:00 Yaron Keren :
> Hi Tai,
>
> Thanks for looking into this. You are correct the right frontend should be
> used, however this isn't the root of the problem.
> It works since g++ defaults to -shared-libgcc while gcc defaults to
> -static-libgcc.
> If you force g++ to static
16 matches
Mail list logo
