Re: My take on the diff-optimizations-bytes branch

2011-01-26 Thread Stefan Fuhrmann
On 26.01.2011 03:09, Johan Corveleyn wrote: On Tue, Jan 25, 2011 at 1:37 AM, Stefan Fuhrmann wrote: [ ... snip ...] And, as promised, here some ideas how to get more speed from the generic code. Your latest commit: +#if SVN_UNALIGNED_ACCESS_IS_OK + + /* Skip quickly over the stuff between

Re: My take on the diff-optimizations-bytes branch

2011-01-25 Thread Johan Corveleyn
On Tue, Jan 25, 2011 at 1:37 AM, Stefan Fuhrmann wrote: [ ... snip ...] > And, as promised, here some ideas how to get more > speed from the generic code. Your latest commit: > > +#if SVN_UNALIGNED_ACCESS_IS_OK > + > +      /* Skip quickly over the stuff between EOLs. */ > +      for (i = 0, can_r

Re: My take on the diff-optimizations-bytes branch

2011-01-24 Thread Stefan Fuhrmann
On 23.01.2011 22:46, Johan Corveleyn wrote: On Sat, Jan 8, 2011 at 6:50 AM, Stefan Fuhrmann wrote: On 03.01.2011 02:14, Johan Corveleyn wrote: It would be interesting to see where the biggest gains are coming from (I'm guessing from the "per-machine-word" reading/comparing; I'd like to try th

Re: My take on the diff-optimizations-bytes branch

2011-01-23 Thread Johan Corveleyn
On Sun, Jan 23, 2011 at 10:46 PM, Johan Corveleyn wrote: > On Sat, Jan 8, 2011 at 6:50 AM, Stefan Fuhrmann > wrote: >> On 03.01.2011 02:14, Johan Corveleyn wrote: >>> It would be interesting to see where the biggest gains >>> are coming from (I'm guessing from the "per-machine-word" >>> reading/c

Re: My take on the diff-optimizations-bytes branch

2011-01-23 Thread Johan Corveleyn
On Sat, Jan 8, 2011 at 6:50 AM, Stefan Fuhrmann wrote: > On 03.01.2011 02:14, Johan Corveleyn wrote: >> It would be interesting to see where the biggest gains >> are coming from (I'm guessing from the "per-machine-word" >> reading/comparing; I'd like to try that first, maybe together with the >> a

Re: My take on the diff-optimizations-bytes branch

2011-01-22 Thread Johan Corveleyn
On Wed, Jan 19, 2011 at 1:19 AM, Stefan Fuhrmann wrote: > On 18.01.2011 12:56, Johan Corveleyn wrote: >> >> On Mon, Jan 17, 2011 at 3:12 AM, Johan Corveleyn >>  wrote: >>> >>> On Sat, Jan 8, 2011 at 6:50 AM, Stefan Fuhrmann >>>  wrote: >> >> [ ... snip ... ] >> But I think, the stack variabl

Re: My take on the diff-optimizations-bytes branch

2011-01-18 Thread Stefan Fuhrmann
On 18.01.2011 12:56, Johan Corveleyn wrote: On Mon, Jan 17, 2011 at 3:12 AM, Johan Corveleyn wrote: On Sat, Jan 8, 2011 at 6:50 AM, Stefan Fuhrmann wrote: [ ... snip ... ] But I think, the stack variable is certainly helpful and easy to do. Ok, I've done this (locally, still have to clean

Re: My take on the diff-optimizations-bytes branch

2011-01-18 Thread Johan Corveleyn
On Mon, Jan 17, 2011 at 3:12 AM, Johan Corveleyn wrote: > On Sat, Jan 8, 2011 at 6:50 AM, Stefan Fuhrmann > wrote: [ ... snip ... ] >> But I think, the stack variable is certainly helpful >> and easy to do. Ok, I've done this (locally, still have to clean up a little and then commit). It gives

Re: My take on the diff-optimizations-bytes branch

2011-01-16 Thread Johan Corveleyn
On Sat, Jan 8, 2011 at 6:50 AM, Stefan Fuhrmann wrote: > On 03.01.2011 02:14, Johan Corveleyn wrote: >> >> On Mon, Jan 3, 2011 at 12:13 AM, Johan Corveleyn >>  wrote: >>> >>> On Sun, Jan 2, 2011 at 10:04 PM, Johan Corveleyn >>>  wrote: >> For now, some feedback on the rest of the patch: >> >> [[[

Re: My take on the diff-optimizations-bytes branch

2011-01-07 Thread Stefan Fuhrmann
On 03.01.2011 02:14, Johan Corveleyn wrote: On Mon, Jan 3, 2011 at 12:13 AM, Johan Corveleyn wrote: On Sun, Jan 2, 2011 at 10:04 PM, Johan Corveleyn wrote: On Sat, Jan 1, 2011 at 11:25 PM, Stefan Fuhrmann wrote: Hi Johan, Thursday night I did something stupid and had a look at how svn bl

Re: My take on the diff-optimizations-bytes branch

2011-01-07 Thread Stefan Fuhrmann
On 03.01.2011 13:16, Johan Corveleyn wrote: On Mon, Jan 3, 2011 at 12:03 PM, Stefan Fuhrmann wrote: On 03.01.2011 02:14, Johan Corveleyn wrote: On Mon, Jan 3, 2011 at 12:13 AM, Johan Corveleyn wrote: [... snip ...] And it's fast too! It's taking only 58 seconds in "diff", vs. 72 for the

Re: My take on the diff-optimizations-bytes branch

2011-01-03 Thread Johan Corveleyn
On Mon, Jan 3, 2011 at 12:03 PM, Stefan Fuhrmann wrote: > On 03.01.2011 02:14, Johan Corveleyn wrote: >> >> On Mon, Jan 3, 2011 at 12:13 AM, Johan Corveleyn >>  wrote: >>> [... snip ...] >>> And it's fast too! It's taking only 58 seconds in "diff", vs. 72 for >>> the normal version. Splitting tha

Re: My take on the diff-optimizations-bytes branch

2011-01-03 Thread Stefan Fuhrmann
On 03.01.2011 02:14, Johan Corveleyn wrote: On Mon, Jan 3, 2011 at 12:13 AM, Johan Corveleyn wrote: On Sun, Jan 2, 2011 at 10:04 PM, Johan Corveleyn wrote: On Sat, Jan 1, 2011 at 11:25 PM, Stefan Fuhrmann wrote: Hi Johan, Thursday night I did something stupid and had a look at how svn bl

Re: My take on the diff-optimizations-bytes branch

2011-01-02 Thread Johan Corveleyn
On Mon, Jan 3, 2011 at 12:13 AM, Johan Corveleyn wrote: > On Sun, Jan 2, 2011 at 10:04 PM, Johan Corveleyn wrote: >> On Sat, Jan 1, 2011 at 11:25 PM, Stefan Fuhrmann >> wrote: >>> Hi Johan, >>> >>> Thursday night I did something stupid and had a look at  how >>> svn blame could be made faster ba

Re: My take on the diff-optimizations-bytes branch

2011-01-02 Thread Johan Corveleyn
On Sun, Jan 2, 2011 at 10:04 PM, Johan Corveleyn wrote: > On Sat, Jan 1, 2011 at 11:25 PM, Stefan Fuhrmann > wrote: >> Hi Johan, >> >> Thursday night I did something stupid and had a look at  how >> svn blame could be made faster based on the HEAD code in >> your branch. >> >> One night and most

Re: My take on the diff-optimizations-bytes branch

2011-01-02 Thread Johan Corveleyn
On Sat, Jan 1, 2011 at 11:25 PM, Stefan Fuhrmann wrote: > Hi Johan, > > Thursday night I did something stupid and had a look at  how > svn blame could be made faster based on the HEAD code in > your branch. > > One night and most of the following day later, I think I made it > a few percent faster