summaryrefslogtreecommitdiffhomepage
path: root/ir/ana
Commit message (Collapse)AuthorAge
* execfreq: Only free DFS when we abort due to too large function.Andreas Fried2019-12-13
| | | | By this point, the IR resources have not been allocated yet.
* store the index of the first variadic parameter in method typesJohannes Bucher2019-10-18
|
* irdom: Make `get_Block_*dom_depth` publicSebastian Graf2019-01-16
|
* Use Boolean operator for Boolean valuesSebastian Buchwald2018-11-23
|
* Added an upper bound on graph size to execfreq.cSebastian Graf2018-08-28
|
* Add construct_confirms_only function.Andreas Fried2018-07-05
| | | | | This function only inserts Confirms into a graph without performing any further optimizations.
* Fix typos in commentsChristoph Erhardt2017-12-15
|
* Unify indentation of function parametersSebastian Buchwald2017-03-31
|
* Reorganize include directivesMatthias Braun2017-02-17
|
* ir: Factor out 'is_Const(x) && is_Const_null(x)' into the new function ↵Christoph Mallon2016-11-04
| | | | 'is_irn_null()'.
* Remove tuples before determining entity usage.Manuel Mohr2016-10-20
| | | | | | | Some optimizations, e.g., inlining, can create tuples with tuples as operands. The current code in determine_entity_usage cannot deal with this situation and computes wrong usage information. For now, we just remove tuples before analysis.
* Use new irg property framework.Manuel Mohr2016-10-20
|
* Assure that there are no tuples when calculating post-dominators.Christoph Mallon2016-10-08
| | | | | The dominator calculation assures the same. This fixes the optimize build for several test cases.
* Assert (post)dominance property for all API functions.Sebastian Buchwald2016-10-07
| | | | Also fix some missing or wrong uses of (post)dominance property methods.
* Set the additional properties right when construction a method type.Christoph Mallon2016-07-11
|
* Rename pdeq_new.h => pdeq.hMatthias Braun2016-06-27
|
* Use pdeq_new instead of pdeqMatthias Braun2016-06-27
|
* Remove duplicated code blockRaphael von der Grün2016-06-14
|
* Fix typos and language in some commentsRaphael von der Grün2016-06-14
|
* Remove the functions for 'atomic entities'.Christoph Mallon2016-06-12
| | | | | | | | | | * 'get_atomic_ent_value()' is a wrapper, which tries to unify the different kind of initalizers by wastefully creating nodes on demand. It had only a single user, which tried to use it for method entities, which does not work anymore (see 3rd bullet point). * 'set_atomic_ent_value()' is a thin and unnecessary wrapper to set a const node initializer. * 'is_atomic_entity()' classifies entities having (among primitive and pointer types) method type as having an atomic initializer. They do not for a long time (was an address of itself). Except for the two functions above, this function is unused. * 'Atomic' is a misnomer and should rather be 'scalar'.
* tr: Set variadicity directly when creating method types.Christoph Mallon2016-05-22
|
* irmemory: Factorise common code.Christoph Mallon2016-05-22
|
* Use assertion instead of "this should not happen" commentSebastian Buchwald2016-05-18
|
* Fix indentationSebastian Buchwald2016-05-12
|
* Rename get_type_{size|alignment}_bytes() to get_type{size|alignment}()Matthias Braun2016-02-14
|
* Rename is_po2() to is_po2_or_zero()Matthias Braun2016-02-13
| | | | | The function also returns true for zero even though that is not a power of two.
* constbits: Make debug info a bit more verbose.Christoph Mallon2016-02-13
|
* Do not use computed_value() in constbits analysis.Sebastian Buchwald2016-02-13
| | | | | | | | computed_value() uses some non-trivial patterns (Confirm nodes etc.) that break the assumption that the transfer function only looks at its direct operands. This fixes opt/fehler313.c.
* Print debug info at end of transfer function.Sebastian Buchwald2016-02-13
|
* Print start of verification in debug output.Sebastian Buchwald2016-02-13
|
* Add explicit goto.Sebastian Buchwald2016-02-13
|
* dca: Delete outdated comment.Andreas Seltenreich2016-02-10
|
* irmemory: Handle Sub in get_address_info().Christoph Mallon2016-01-28
| | | | This fixes opt/loadstore9.c
* irmemory: Factorise two identical loops to climb pointer calculations.Christoph Mallon2016-01-28
|
* irmemory: Always (try to) climb to the base address in _get_alias_relation().Christoph Mallon2016-01-28
| | | | | | This fixes opt/loadstore8.c. Though it does not correct all cases. More work is needed here.
* Do not set a node as its own operand.Sebastian Buchwald2016-01-09
| | | | This fixes opt/fehler314.c.
* Remove superfluous computations.Sebastian Buchwald2016-01-09
|
* Fix typos: datastructure -> data structure.Sebastian Buchwald2015-12-21
|
* Replace get_{struct|union|class}_XXX with get_compound_XXX where possibleMatthias Braun2015-12-07
|
* Cleanup: match declarationMatthias Braun2015-12-06
|
* Simplify code to check pointer classes.Sebastian Buchwald2015-12-02
|
* Constbits: Repeat evaluation of Phi operands, if necessary.Sebastian Buchwald2015-11-06
| | | | This fixes opt/fehler308.c.
* Remove superfluous if statement.Sebastian Buchwald2015-11-06
|
* Remove wrong break to reach a global fix point.Sebastian Buchwald2015-10-23
|
* Adapt debugging switch.Sebastian Buchwald2015-10-23
|
* tv: Remove mode parameter from tarval_subMatthias Braun2015-10-18
| | | | | We can get a proper mode for pointer-pointer from get_reference_offset_mode().
* constbits: Add fixpoint verifier.Christoph Mallon2015-10-12
|
* Reformat comments.Sebastian Buchwald2015-09-30
|
* Always include irop_t.h inside libfirmMatthias Braun2015-09-10
|
* Always include irnode_t.h from inside libfirm (or firm_types.h)Matthias Braun2015-09-10
|