Re: [Firebird-devel] march switch in posix build for x86 architecture

2015-12-30 Thread James Starkey
Putting aside the questions of when these instructions were added, under what circumstances could / would a compiler optimize something with a conditional move (cmov)? The new operations are supported by compiler intrinsics at best, but often require inline assembler. Even now, the cpuid instruct

Re: [Firebird-devel] march switch in posix build for x86 architecture

2015-12-30 Thread Dimitry Sibiryakov
30.12.2015 10:26, Dmitry Yemanov wrote: > Maybe I'm missing something, but I see in both v2.5 and v3.0: > > OPTIMIZE_FLAGS=-O3 -march=i586 -mtune=i686 -fno-omit-frame-pointer You are right. I'm sorry. -- WBR, SD.

Re: [Firebird-devel] march switch in posix build for x86 architecture

2015-12-30 Thread Dmitry Yemanov
29.12.2015 19:56, Dimitry Sibiryakov wrote: > > Isn't it a good time to drop support of 20 years old processors and raise > -march value > from i386 to i686? Maybe I'm missing something, but I see in both v2.5 and v3.0: OPTIMIZE_FLAGS=-O3 -march=i586 -mtune=i686 -fno-omit-frame-pointer Dmitry

Re: [Firebird-devel] march switch in posix build for x86 architecture

2015-12-30 Thread Mark Rotteveel
On Tue, 29 Dec 2015 19:43:37 -0500, James Starkey wrote: > Could you explain to us what optimization a are possible on 686 that are > not on the 386? Look at https://en.wikipedia.org/wiki/X86_instruction_listings#Added_with_80486 and go down. Instructions added (not all relevant for optimization

Re: [Firebird-devel] march switch in posix build for x86 architecture

2015-12-30 Thread Mark Rotteveel
On Tue, 29 Dec 2015 22:04:52 -0200, "Carlos H. Cantu" wrote: > Let's try to see the case from another angle: > > On Windows, Firebird 3.0 uses VC 2010 runtime. From MS site, the > minimum requirements of this runtime are: > > Windows XP SP3 (SP3 was released in 2008) > Computer with 900 MHz or

Re: [Firebird-devel] march switch in posix build for x86 architecture

2015-12-30 Thread Michal Kubecek
On Tue, Dec 29, 2015 at 10:13:26PM -0500, James Starkey wrote: > On Tuesday, December 29, 2015, Michal Kubecek wrote: > > > On Tue, Dec 29, 2015 at 07:43:37PM -0500, James Starkey wrote: > > > Could you explain to us what optimization a are possible on 686 that > > > are not on the 386? > > >From

Re: [Firebird-devel] march switch in posix build for x86 architecture

2015-12-29 Thread Geoff Worboys
Carlos H. Cantu wrote: > Anyway, I think the real question is: how much performance > increase this would bring, in real world environment? Do you > have this number? If you tell me +25%, I would vote to drop > i386 support :) Yep. This is it exactly. Like all optimisations: Work out what it bu

Re: [Firebird-devel] march switch in posix build for x86 architecture

2015-12-29 Thread James Starkey
Nope, interlocked compare and exchange were part of the 386 instruction set. On Tuesday, December 29, 2015, Michal Kubecek wrote: > On Tue, Dec 29, 2015 at 07:43:37PM -0500, James Starkey wrote: > > Could you explain to us what optimization a are possible on 686 that > > are not on the 386? > >

Re: [Firebird-devel] march switch in posix build for x86 architecture

2015-12-29 Thread Michal Kubecek
On Tue, Dec 29, 2015 at 10:04:52PM -0200, Carlos H. Cantu wrote: > > Let's try to see the case from another angle: > > On Windows, Firebird 3.0 uses VC 2010 runtime. From MS site, the > minimum requirements of this runtime are: > > Windows XP SP3 (SP3 was released in 2008) > Computer with 900 M

Re: [Firebird-devel] march switch in posix build for x86 architecture

2015-12-29 Thread Michal Kubecek
On Tue, Dec 29, 2015 at 07:43:37PM -0500, James Starkey wrote: > Could you explain to us what optimization a are possible on 686 that > are not on the 386? >From the top of my head, 386 doesn't even have any atomic cmpxchg or xadd instruction which rather complicates any lock or semaphore implemen

Re: [Firebird-devel] march switch in posix build for x86 architecture

2015-12-29 Thread James Starkey
Could you explain to us what optimization a are possible on 686 that are not on the 386? On Tuesday, December 29, 2015, Dimitry Sibiryakov wrote: > 29.12.2015 21:37, James Starkey wrote: > > Why? For all practical purposes they're the same architecture. There > is no upside to > > the project

Re: [Firebird-devel] march switch in posix build for x86 architecture

2015-12-29 Thread Carlos H. Cantu
DS> 29.12.2015 21:37, James Starkey wrote: >> Why? For all practical purposes they're the same architecture. There is >> no upside to >> the project and only down side for users. DS>Using of i386 command set limits optimization possibilities for compilers. DS>And, frankly, can you ima

Re: [Firebird-devel] march switch in posix build for x86 architecture

2015-12-29 Thread Dimitry Sibiryakov
29.12.2015 21:37, James Starkey wrote: > Why? For all practical purposes they're the same architecture. There is no > upside to > the project and only down side for users. Using of i386 command set limits optimization possibilities for compilers. And, frankly, can you imagine that nowada

Re: [Firebird-devel] march switch in posix build for x86 architecture

2015-12-29 Thread James Starkey
Why? For all practical purposes they're the same architecture. There is no upside to the project and only down side for users. Go ahead and drop Apollo/Domain, with my blessings. On Tuesday, December 29, 2015, Dimitry Sibiryakov wrote: >Hello, All. > >Isn't it a good time to drop sup

[Firebird-devel] march switch in posix build for x86 architecture

2015-12-29 Thread Dimitry Sibiryakov
Hello, All. Isn't it a good time to drop support of 20 years old processors and raise -march value from i386 to i686? -- WBR, SD. -- Firebird-Devel mailing list, web interface at https://lists.sourceforge.ne