[m-rev.] for review: conditional structure reuse for LLDS backends

[Peter Wang]

On 2008-02-12, Julien Fischer <juliensf at csse.unimelb.edu.au> wrote:
>
> On Tue, 12 Feb 2008, Peter Wang wrote:
>
>> On 2008-02-11, Peter Wang <novalazy at gmail.com> wrote:
>>> On 2008-02-11, Julien Fischer <juliensf at csse.unimelb.edu.au> wrote:
>>>>
>>>> On Wed, 30 Jan 2008, Peter Wang wrote:
>>>>
>>>>> [Pending bootchecks in more grades.]
>>>>
>>>> Could you please post some benchmarks between conditional structure
>>>> reuse / structure reuse without static data / no structure reuse,
>>>> at some point.
>>>
>>> Will do.
>>
>> Here are some speed_test results from neptune.  The compiler built with
>> ctgc and intermodule optimisation doesn't work, so I doubt much reuse is
>> actually happening.
>
> What happens with the compiler built with ctgc and intermodule
> optimisation?

Still figuring it out. The symptom is:

...
% Optimizing jumps for module_defines_label_with_layout/2 mode 0
Uncaught Mercury exception:
Software Error: map.det_insert: key already present
        Key Type: ll_backend.llds.label
	Key Value: internal_label(2,
ordinary_proc_label(qualified(unqualified("ll_backend"), "llds_out"),
pf_predicate, qualified(unqualified("ll_backend"), "llds_out"),
"module_defines_label_with_layout", 2, 0))
        Value Type: maybe.maybe(ll_backend.llds.instruction)
Stack dump not available in this grade.

Peter

--------------------------------------------------------------------------
mercury-reviews mailing list
Post messages to:       mercury-reviews at csse.unimelb.edu.au
Administrative Queries: owner-mercury-reviews at csse.unimelb.edu.au
Subscriptions:          mercury-reviews-request at csse.unimelb.edu.au
--------------------------------------------------------------------------