Unordered container insertion hints

[François Dumont]

Patch applied, I also needed to update some dg-error line numbers, it 
was not part of the original patch.

The only remaining thing will be to update the doxygen links on the 
unordered containers to point to the new hint doc.

2013-06-29  François Dumont  <fdumont@gcc.gnu.org>

     * include/bits/hashtable_policy.h (_Insert_base): Consider hint in
     insert methods.
     * include/bits/hashtable.h: Likewise.
     * testsuite/23_containers/unordered_multimap/insert/hint.cc: New.
     * 
testsuite/performance/23_containers/insert/unordered_multiset_hint.cc:
     New.
     * testsuite/23_containers/unordered_set/instantiation_neg.cc:
     Adjust dg-error line number.
     * testsuite/23_containers/unordered_set/
     not_default_constructible_hash_neg.cc: Likewise.
     * doc/xml/manual/containers.xml: Document hinting in unordered
     containers.

François


On 06/28/2013 01:46 PM, Jonathan Wakely wrote:
> On 23 June 2013 13:51, Jonathan Wakely wrote:
>> On 19 June 2013 20:56, François Dumont wrote:
>>>      Still no chance to have a look ?
>> I'll try to finish reviewing it today, thanks for the reminder.
> Sorry for the delay.  The patch is OK, with a few small changes to the new docs:
>
> Please change "rational" to "rationale"
>
> "can require to update the bucket" should be "can require updating the bucket"
>
> "need to compute following node hash code." should be "need to compute
> the following node's hash code."
>
> "won't compute next element hash code" should be "won't compute the
> next element's hash code"
>
> "bench" should be "benchmark"
>
> I would say "more harm than good" instead of "more bad than good"
>
> Thanks!
>

-------------- next part --------------
A non-text attachment was scrubbed...
Name: hashtable_hint.patch
Type: text/x-patch
Size: 27467 bytes
Desc: not available
URL: <http://gcc.gnu.org/pipermail/libstdc++/attachments/20130629/a70f35bd/attachment.bin>