RFR: 8226297: Dual-pivot quicksort improvements
brent.christian at oracle.com
Tue Aug 6 23:33:13 UTC 2019
On 8/6/19 12:47 PM, Vladimir Yaroslavskiy wrote:
> I moved Object sorting related stuff after primitives sorting methods
> to separate functionality logically.
Sure, fine to keep that all together. I can move that back:
> The order of methods in my version is:
> 1. primitives (sequential sorting)
> - int
> - long
> - byte
> - char
> - short
> - float
> - double
The order for sequential sorting of primitives in Arrays.java checked
into the JDK is:
It simplifies the webrev for reviewing to keep that ordering, so that's
what I've done.
> Вторник, 6 августа 2019, 21:35 +03:00 от Brent Christian
> <brent.christian at oracle.com>:
> Hi, Laurent
> I'm not sure what exactly is causing the problem, but here's my hunch:
> In Vladimir's version of Arrays.java:
> got moved around, but were unchanged.
> In the interest of keeping the change as simple as possible, I restored
> these to their original location, so they don't show up in my changes.
> That could confuse things when comparing diffs.
> One idea would be to restore those elements back in their original
> locations in your version, and re-generate your patch. I don't know if
> that would be less work than just comparing raw files.
> Alternatively, if it would be easiest for those familiar with the
> evolution of this fix to leave things where Vladimir had them, I can do
> On 8/6/19 6:32 AM, Laurent Bourgès wrote:
> > Hi Brent,
> > Thank you for sponsoring this patch.
> > I tried to compare your webrev against my latest (diff patch
> files) but
> > it gives me too many changes lines.
> > Do you have another idea to see incremental changes only ?
> > (anyway I can compare raw files)
> > Thanks,
> > Laurent
> > Le lun. 5 août 2019 à 23:43, Brent Christian
> <brent.christian at oracle.com <mailto:brent.christian at oracle.com>
> > <mailto:brent.christian at oracle.com>> a écrit :
> > Hi,
> > Please review Vladimir Yaroslavskiy's changes to DualPivotQuickSort
> > (seen earlier on this alias). I will be sponsoring this change.
> > I have a webrev against jdk-jdk here:
> > http://cr.openjdk.java.net/~bchristi/8226297/webrev03-rfr/
> > (Note that I did a little re-ordering, and removed some superfluous
> > spacing changes, in order to simplify the webrev. I've also included
> > Vladimir's FailedFloat test case.)
> > Information about benchmarking the changes was posted recently.
> > An automated test run passes cleanly.
> > Thanks!
> > -Brent
> > --
> > 1.
> > 2.
> > 3.
> Vladimir Yaroslavskiy
More information about the core-libs-dev