hi guys im building all software and library with this flags on my cyrus+
> mcpu=e5500 -mno-altivec -mtune=e5500
and can say im facing little increasing of performance here.
been build :
mesa 11.3 dev with llvm 38
lib sdl2
qemu
host ubuntu mate 16.04
tested qemu-system-i386 with sdl2 Aros guest an
On Fri, Apr 01, 2016 at 07:59:50AM +0900, Hiroyuki Yamamoto wrote:
> Hi,
>
> wbon...@theitmakers.com wrote:
> > The current target is e5500, which means i rebuild packages with the
> > following additionnal flags : -mcpu=e5500 -mno-altivec -mtune=e5500
> >
> > Is there some more generic flags i c
Hi,
wbon...@theitmakers.com wrote:
> The current target is e5500, which means i rebuild packages with the
> following additionnal flags : -mcpu=e5500 -mno-altivec -mtune=e5500
>
> Is there some more generic flags i could use ?
According to binutils, opcodes of e5500 is added to ones of Book E, a
On Mar 31, 2016, at 8:34 AM, Lennart Sorensen
wrote:
> On Thu, Mar 31, 2016 at 01:18:53AM -0700, Rick Thomas wrote:
>> The altivec instructions and other optimizations speed-up significantly
>> certain kinds of programs on CPUs that can use them. It would be a shame to
>> loose the optimized
Hi,
> That's actually a good question. I am guessing you could contact the
> DSA team directly:
>
> https://www.debian.org/doc/manuals/developers-reference/resources.html#server-machines
>
> You still have the option to maintain this machine by yourself
> (non-DSA machines are just fine: https://d
On Thu, Mar 31, 2016 at 01:18:53AM -0700, Rick Thomas wrote:
> The altivec instructions and other optimizations speed-up significantly
> certain kinds of programs on CPUs that can use them. It would be a shame to
> loose the optimized versions by dropping back to the lowest common
> denominator
On Mar 30, 2016, at 10:13 AM, wbon...@theitmakers.com wrote:
> Hi,
>
> Le 30.03.2016 17:09, Breno Leitao a écrit :
>> Hi Mathieu,
>> On 03/30/2016 05:40 AM, Mathieu Malaterre wrote:
So i would like to know what is the target, and if it is
possible to consider supporting the "most gener
On Wed, Mar 30, 2016 at 6:36 PM, wrote:
> Hi Hiroyuki,
>
> Thanks for your answer (and for all answers that followed). These are very
> good news for me.
>
>> So, I'm planning that almost all ppc64 binary packages will be rebuilt
>> as binNMU.
>> I have already given binNMU command for only ppc64
Hi Hiroyuki,
Thanks for your answer (and for all answers that followed). These are
very good news for me.
So, I'm planning that almost all ppc64 binary packages will be rebuilt
as binNMU.
I have already given binNMU command for only ppc64 port to wanna-build
DB at the last Friday,
but there a
Hi,
On 03/30/2016 02:13 PM, wbon...@theitmakers.com wrote:
>> Do you mean setting "-mno-altivec" as a generic compiler flag for ppc64?
>
> Yes, that's what i mean. "-mno-altivec" and maybe some other "-mno-something"
> flags, so it could run on any PPC64 cpu, even if it does not have
> altivec/v
On Wed, Mar 30, 2016 at 07:13:49PM +0200, wbon...@theitmakers.com wrote:
> Yes, that's what i mean. "-mno-altivec" and maybe some other
> "-mno-something" flags, so it could run on any PPC64 cpu, even if it does
> not have altivec/vmx, or if it does not have the optional opcodes of the
> PPC64 stan
On Wed, Mar 30, 2016 at 12:06:36PM -0300, Breno Leitao wrote:
> I am wondering if you cannot try to mimic this non-VMX CPU on QEMU using
> -cpu.
>
> Running "$ qemu-system-ppc -cpu help" shows a lot of CPUs that are supported
> in
> QEMU. I am still not sure if we can exploit KVM if we use a di
Hi,
Le 30.03.2016 17:09, Breno Leitao a écrit :
Hi Mathieu,
On 03/30/2016 05:40 AM, Mathieu Malaterre wrote:
So i would like to know what is the target, and if it is
possible to consider supporting the "most generic PPC64 target", just
like it is done for x86 archs ?
Do you mean setting "-mn
the best is without altivec at all.
here on P5020 many programs icewasell too crash because altivec
luigi
Inviato da iPad
> Il giorno 30 mar 2016, alle ore 17:27, Breno Leitao ha
> scritto:
>
> Hi Mathieu,
>
> On 03/30/2016 05:40 AM, Mathieu Malaterre wrote:
>>> So i would like to know what
Hi Mathieu,
On 03/30/2016 05:40 AM, Mathieu Malaterre wrote:
>> So i would like to know what is the target, and if it is
>> possible to consider supporting the "most generic PPC64 target", just
>> like it is done for x86 archs ?
Do you mean setting "-mno-altivec" as a generic compiler flag for pp
Hi Lennart,
On 03/30/2016 11:30 AM, Lennart Sorensen wrote:
> Certainly. And getting access to one of the ones that causes trouble
> by not having VMX seems even harder in fact.
I am wondering if you cannot try to mimic this non-VMX CPU on QEMU using -cpu.
Running "$ qemu-system-ppc -cpu help"
On Wed, Mar 30, 2016 at 04:41:40PM +0200, Mathieu Malaterre wrote:
> On Wed, Mar 30, 2016 at 4:30 PM, Lennart Sorensen
> wrote:
> > On Wed, Mar 30, 2016 at 04:26:59PM +0200, Mathieu Malaterre wrote:
> >> Let me rephrase that in: "Wow somebody (user)tagged specifically with
> >> ppc64 the nouveau b
On Wed, Mar 30, 2016 at 4:30 PM, Lennart Sorensen
wrote:
> On Wed, Mar 30, 2016 at 04:26:59PM +0200, Mathieu Malaterre wrote:
>> Let me rephrase that in: "Wow somebody (user)tagged specifically with
>> ppc64 the nouveau bugs" ... :)
>>
>> This was not clear. But the offer to ppc64 machine(s) was m
On Wed, Mar 30, 2016 at 04:26:59PM +0200, Mathieu Malaterre wrote:
> Let me rephrase that in: "Wow somebody (user)tagged specifically with
> ppc64 the nouveau bugs" ... :)
>
> This was not clear. But the offer to ppc64 machine(s) was made by the
> ppc64el people:
>
> https://lists.debian.org/debi
On Wed, Mar 30, 2016 at 4:18 PM, Lennart Sorensen
wrote:
> On Wed, Mar 30, 2016 at 10:40:52AM +0200, Mathieu Malaterre wrote:
>> Well that's odd. Anyway feel free to report the bug and assign it the
>> usertags to ppc64:
>>
>> https://udd.debian.org/cgi-bin/bts-usertags.cgi?tag=ppc64&user=debian-p
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA256
Hi, William,
William Bonnet wrote:
> ./ Is there a defined "minimal" target for the PPC64 port ? I have
> understood that VMX support is not enabled [1] (please tell me if i am
> wrong). But VMX is not the only optional feature from PowerPC64
> stan
On Wed, Mar 30, 2016 at 10:40:52AM +0200, Mathieu Malaterre wrote:
> Well that's odd. Anyway feel free to report the bug and assign it the
> usertags to ppc64:
>
> https://udd.debian.org/cgi-bin/bts-usertags.cgi?tag=ppc64&user=debian-powerpc%40lists.debian.org
Wow, 5 out of 11 bugs are to do with
Hi,
On Tue, Mar 29, 2016 at 8:54 PM, William Bonnet wrote:
> Hi all,
>
> I am currently working (office work) on a project using PPC64 based
> boards, each board is running a NXP e5500 processor. We plan to use
> Debian, and thus the PPC64 port. So, I have tested the version from
> debian-ports,
will be better ;-)
Luigi
> To: debian-powerpc@lists.debian.org
> From: will...@wbonnet.net
> Subject: PPC64 port status
> Date: Tue, 29 Mar 2016 20:54:37 +0200
>
> Hi all,
>
> I am currently working (office work) on a project using PPC64 based
> boards, each b
Hi all,
I am currently working (office work) on a project using PPC64 based
boards, each board is running a NXP e5500 processor. We plan to use
Debian, and thus the PPC64 port. So, I have tested the version from
debian-ports, it is almost working, but it seems that packages have been
compiled with
25 matches
Mail list logo
