Re: RFR: 8168073: Speed up URI creation during module bootstrap

2016-10-18 Thread Chris Hegarty
On 17/10/2016 22:18, Claes Redestad wrote: > >> On 2016-10-17 21:35, Alan Bateman wrote: >>> >>> JavaNetHttpCookieAccess, JavaNetInetAddressAccess and >>> JavaNetSocketAccess are the other 3 that are used to get at non-public >>> types in java.net so I don't think JavaNetUriAccess would be out

Re: RFR: 8168073: Speed up URI creation during module bootstrap

2016-10-18 Thread Alan Bateman
On 17/10/2016 22:18, Claes Redestad wrote: On 2016-10-17 21:35, Alan Bateman wrote: JavaNetHttpCookieAccess, JavaNetInetAddressAccess and JavaNetSocketAccess are the other 3 that are used to get at non-public types in java.net so I don't think JavaNetUriAccess would be out of place. TIL!

Re: RFR: 8168073: Speed up URI creation during module bootstrap

2016-10-17 Thread Mandy Chung
> On Oct 17, 2016, at 2:18 PM, Claes Redestad wrote: > > http://cr.openjdk.java.net/~redestad/8168073/webrev.02/ +1 Mandy

Re: RFR: 8168073: Speed up URI creation during module bootstrap

2016-10-17 Thread Paul Sandoz
> On 17 Oct 2016, at 14:18, Claes Redestad wrote: > > > > On 2016-10-17 21:35, Alan Bateman wrote: >> >> >> On 17/10/2016 19:45, Claes Redestad wrote: >>> >>> Most other SharedSecrets classes seems to be per-package, so not sure >>> if this case is special enough

Re: RFR: 8168073: Speed up URI creation during module bootstrap

2016-10-17 Thread Claes Redestad
On 2016-10-17 21:35, Alan Bateman wrote: On 17/10/2016 19:45, Claes Redestad wrote: Most other SharedSecrets classes seems to be per-package, so not sure if this case is special enough to start making them per-class, other than the slightly more natural naming. JavaNetHttpCookieAccess,

Re: RFR: 8168073: Speed up URI creation during module bootstrap

2016-10-17 Thread Paul Sandoz
> On 17 Oct 2016, at 10:53, Claes Redestad wrote: > > Hi Paul, > > On 2016-10-17 19:39, Paul Sandoz wrote: >> Hi Claes, >> >> This looks good. > > Thanks! > >> >> Did you consider adding asserts to the package private constructor? > > No, might be reasonable. Do

Re: RFR: 8168073: Speed up URI creation during module bootstrap

2016-10-17 Thread Mandy Chung
> On Oct 17, 2016, at 11:07 AM, Alan Bateman wrote: > > > > On 17/10/2016 12:17, Claes Redestad wrote: >> Hi, >> >> one partial cause for startup regressions due to jigsaw is related to >> creating >> URIs for the location of each module. >> >> By providing a

Re: RFR: 8168073: Speed up URI creation during module bootstrap

2016-10-17 Thread Alan Bateman
On 17/10/2016 19:45, Claes Redestad wrote: Most other SharedSecrets classes seems to be per-package, so not sure if this case is special enough to start making them per-class, other than the slightly more natural naming. JavaNetHttpCookieAccess, JavaNetInetAddressAccess and

Re: RFR: 8168073: Speed up URI creation during module bootstrap

2016-10-17 Thread Claes Redestad
On 2016-10-17 20:07, Alan Bateman wrote: On 17/10/2016 12:17, Claes Redestad wrote: Hi, one partial cause for startup regressions due to jigsaw is related to creating URIs for the location of each module. By providing a package-private constructor we can avoid the time to scan and

Re: RFR: 8168073: Speed up URI creation during module bootstrap

2016-10-17 Thread Alan Bateman
On 17/10/2016 12:17, Claes Redestad wrote: Hi, one partial cause for startup regressions due to jigsaw is related to creating URIs for the location of each module. By providing a package-private constructor we can avoid the time to scan and validate the URI, which takes a little time

Re: RFR: 8168073: Speed up URI creation during module bootstrap

2016-10-17 Thread Claes Redestad
Hi Paul, On 2016-10-17 19:39, Paul Sandoz wrote: Hi Claes, This looks good. Thanks! Did you consider adding asserts to the package private constructor? No, might be reasonable. Do you insist? :-) /Claes Paul. On 17 Oct 2016, at 04:17, Claes Redestad

Re: RFR: 8168073: Speed up URI creation during module bootstrap

2016-10-17 Thread Claes Redestad
Hi Paul, On 2016-10-17 19:39, Paul Sandoz wrote: Hi Claes, This looks good. Thanks! Did you consider adding asserts to the package private constructor? No, might be reasonable. Do you insist? :-) /Claes Paul. On 17 Oct 2016, at 04:17, Claes Redestad

Re: RFR: 8168073: Speed up URI creation during module bootstrap

2016-10-17 Thread Paul Sandoz
Hi Claes, This looks good. Did you consider adding asserts to the package private constructor? Paul. > On 17 Oct 2016, at 04:17, Claes Redestad wrote: > > Hi, > > one partial cause for startup regressions due to jigsaw is related to creating > URIs for the

RFR: 8168073: Speed up URI creation during module bootstrap

2016-10-17 Thread Claes Redestad
Hi, one partial cause for startup regressions due to jigsaw is related to creating URIs for the location of each module. By providing a package-private constructor we can avoid the time to scan and validate the URI, which takes a little time (executes ~80K bytecodes) but also pushes