Re: The Value of Thumb-2

2011-10-20 Thread Mans Rullgard
On 20 October 2011 23:22, Michael Hope wrote: > On Fri, Oct 21, 2011 at 11:16 AM, Mans Rullgard > wrote: >> On 20 October 2011 23:07, Michael Hope wrote: >>> On Fri, Oct 21, 2011 at 7:48 AM, James Tunnicliffe >>> wrote: This isn't exactly overflowing with up to date numbers, but... >>

Re: The Value of Thumb-2

2011-10-20 Thread Mans Rullgard
On 20 October 2011 23:22, Michael Hope wrote: > On Fri, Oct 21, 2011 at 11:16 AM, Mans Rullgard > wrote: >> On 20 October 2011 23:07, Michael Hope wrote: >>> On Fri, Oct 21, 2011 at 7:48 AM, James Tunnicliffe >>> wrote: This isn't exactly overflowing with up to date numbers, but... >>

Re: The Value of Thumb-2

2011-10-20 Thread Michael Hope
On Fri, Oct 21, 2011 at 11:16 AM, Mans Rullgard wrote: > On 20 October 2011 23:07, Michael Hope wrote: >> On Fri, Oct 21, 2011 at 7:48 AM, James Tunnicliffe >> wrote: >>> This isn't exactly overflowing with up to date numbers, but... >>> >>> http://elinux.org/images/8/8a/Experiment_with_Linux_an

Re: The Value of Thumb-2

2011-10-20 Thread Mans Rullgard
On 20 October 2011 23:07, Michael Hope wrote: > On Fri, Oct 21, 2011 at 7:48 AM, James Tunnicliffe > wrote: >> This isn't exactly overflowing with up to date numbers, but... >> >> http://elinux.org/images/8/8a/Experiment_with_Linux_and_ARM_Thumb-2_ISA.pdf >> >> Slides 14 and 15 say that across EE

Re: The Value of Thumb-2

2011-10-20 Thread Michael Hope
On Fri, Oct 21, 2011 at 7:48 AM, James Tunnicliffe wrote: > This isn't exactly overflowing with up to date numbers, but... > > http://elinux.org/images/8/8a/Experiment_with_Linux_and_ARM_Thumb-2_ISA.pdf > > Slides 14 and 15 say that across EEMBC Thumb-2 gives 98% of the > performance of ARM 32 bit

Re: The Value of Thumb-2

2011-10-20 Thread Mans Rullgard
On 20 October 2011 18:27, Christian Robottom Reis wrote: >    - Do we know how much better Thumb-2 actually is, in practice? It's >      easy for us to confirm this on Android; what do the numbers and >      feel of the system tell us? I did some tests comparing Libav built for ARM and Thumb-2.

Re: The Value of Thumb-2

2011-10-20 Thread James Tunnicliffe
This isn't exactly overflowing with up to date numbers, but... http://elinux.org/images/8/8a/Experiment_with_Linux_and_ARM_Thumb-2_ISA.pdf Slides 14 and 15 say that across EEMBC Thumb-2 gives 98% of the performance of ARM 32 bit instructions (assume performance optimised) and binaries are 26% sma

Re: The Value of Thumb-2

2011-10-20 Thread Roger Teague
Hi Kiko, These are all excellent questions and I for one would be keen to see a position on this. How would it be captured and the output published? Ta Sent from yet another ARM powered device On 20 Oct 2011, at 18:30, "Christian Robottom Reis" wrote: > Hello there, > >Coming back from