- 20 Sep, 2011 4 commits
-
-
Matthias Braun authored
-
Matthias Braun authored
-
Matthias Braun authored
-
Matthias Braun authored
-
- 16 Sep, 2011 1 commit
-
-
Christoph Mallon authored
-
- 15 Sep, 2011 1 commit
-
-
Matthias Braun authored
- Make API private as it should only be called by backends - Let ia32 backend create special ia32_Set nodes instead of relying on muxes which must not be touched anymore - Does not create ConvB nodes anymore but instead produces the Cmp directly. (All backends did this anyway during code-selection so I was able to leave this case out code-selection) - First collect nodes to lower, then lower them. This avoids robustness problems when transforming the graph while at the same time walking it.
-
- 14 Sep, 2011 2 commits
-
-
Andreas Zwinkau authored
-
Andreas Zwinkau authored
-
- 18 Aug, 2011 1 commit
-
-
yb9976 authored
-
- 17 Aug, 2011 1 commit
-
-
yb9976 authored
We have to convert the values the same way the program does.
-
- 27 Jul, 2011 1 commit
-
-
Christoph Mallon authored
-
- 26 Jul, 2011 1 commit
-
-
yb9976 authored
This fixes opt/fehler194.c.
-
- 19 Jul, 2011 2 commits
- 15 Jul, 2011 2 commits
- 21 Jun, 2011 1 commit
-
-
Matthias Braun authored
-
- 09 Jun, 2011 1 commit
-
-
yb9976 authored
-
- 01 Jun, 2011 3 commits
-
-
Matthias Braun authored
-
Matthias Braun authored
-
Matthias Braun authored
-
- 31 May, 2011 5 commits
-
-
Matthias Braun authored
-
Matthias Braun authored
-
Matthias Braun authored
There are cases where we can normalize in 2 directions and both have their merits. Just do both in this case. The frontend should at some point set IRG_GRAPH_STATE_NORMALISATION2 to trigger the 2nd direction.
-
Matthias Braun authored
-
Matthias Braun authored
-
- 30 May, 2011 1 commit
-
-
Matthias Braun authored
-
- 28 May, 2011 1 commit
-
-
Matthias Braun authored
-
- 27 May, 2011 3 commits
-
-
Matthias Braun authored
-
Matthias Braun authored
-
yb9976 authored
-
- 26 May, 2011 1 commit
-
-
Matthias Braun authored
-
- 25 May, 2011 1 commit
-
-
Andreas Zwinkau authored
Fixes opt/localopts2 Because of Phis blocks cannot be equivalent. The cfopt must do all block merging. For example, consider the case where block A has one predecessor block B, which contains a Jmp. Naively, we might exchange(A,B). However, A might contain a Phi with one operand, then the graph is broken afterwards, since we move the Phi into B as well.
-
- 23 May, 2011 1 commit
-
-
Matthias Braun authored
-
- 17 May, 2011 1 commit
-
-
Andreas Zwinkau authored
GCSEing nodes into unreachable blocks is ok now. The verifier does not check the cfg, when irg is floating. Code placement fixes and pins the irg after gcse. Nodes in unreachable blocks are moved into the first block after the start block. Reachability checks in place_late are converted into asserts, because after place_early, there are no unreachable blocks anymore. reverts c64fe299 and 318579c4 but also fixes opt/unreachable_block
-
- 16 May, 2011 2 commits
-
-
Andreas Zwinkau authored
-
Andreas Zwinkau authored
During global CSE we now ignore nodes in unreachable blocks, otherwise cse could pick the wrong representative node and use unreachable nodes in reachable code.
-
- 10 May, 2011 1 commit
-
-
Matthias Braun authored
only set block/phi inputs to bad to kill unreachable code, all other code will be dead and removed anyway
-
- 09 May, 2011 1 commit
-
-
Andreas Zwinkau authored
-
- 06 May, 2011 1 commit
-
-
Matthias Braun authored
-