Matthias Braun [Thu, 14 Apr 2011 13:40:08 +0000 (15:40 +0200)]
assume regex.h exists
Matthias Braun [Thu, 14 Apr 2011 13:22:48 +0000 (15:22 +0200)]
avoid zlib dependency
Matthias Braun [Thu, 14 Apr 2011 13:18:54 +0000 (15:18 +0200)]
assume we always have strings.h
Matthias Braun [Thu, 14 Apr 2011 12:51:51 +0000 (14:51 +0200)]
nobody needs alloca.h these days (hopefully) - using HAVE_s in public headers is wrong anyway
Matthias Braun [Thu, 14 Apr 2011 12:39:53 +0000 (14:39 +0200)]
remove configfile parsing stuff, this also eliminates the dependency on bison/flex
Matthias Braun [Thu, 14 Apr 2011 12:38:35 +0000 (14:38 +0200)]
simplify/refactor fltcalc code (avoid some macros)
Matthias Braun [Thu, 14 Apr 2011 10:37:45 +0000 (12:37 +0200)]
assume everyone except msvc has long double (it's C99 after all)
Matthias Braun [Thu, 14 Apr 2011 10:29:54 +0000 (12:29 +0200)]
minimize number of build variants: always enable statistics
Matthias Braun [Thu, 14 Apr 2011 09:24:41 +0000 (11:24 +0200)]
remove unused code
Matthias Braun [Wed, 13 Apr 2011 16:59:21 +0000 (18:59 +0200)]
NEWS update
Matthias Braun [Wed, 13 Apr 2011 13:54:04 +0000 (15:54 +0200)]
fix warning
Matthias Braun [Wed, 13 Apr 2011 13:51:18 +0000 (15:51 +0200)]
remove unused code
Christoph Mallon [Fri, 15 Apr 2011 13:52:18 +0000 (15:52 +0200)]
Change use of rdtsc so it works with AMD64, too.
Andreas Zwinkau [Fri, 15 Apr 2011 12:21:14 +0000 (14:21 +0200)]
Use get/set functions instead of direct access to is_matured
Christoph Mallon [Thu, 14 Apr 2011 17:36:39 +0000 (19:36 +0200)]
Try fixing if-conversion in presence of endless loops.
Merge the lower block into the upper block (instead of the other way round), so we do not turn the upper block, which might still be in the DFS stack, into and ID node.
Christoph Mallon [Thu, 14 Apr 2011 14:42:47 +0000 (16:42 +0200)]
Revert "Let the block walker enter endless loops only at kept blocks, not Phis."
It does not fix the if-conv problem (there can still be odd keeps into if diamonds) and it also breaks removal of unreachable code, which is only kept at PhiMs.
This reverts commit
351a807ef7c6225b11cc3ac88be60ea0c9db9b7f.
Christoph Mallon [Thu, 14 Apr 2011 12:22:20 +0000 (14:22 +0200)]
Remove the unused function fail_char().
Christoph Mallon [Thu, 14 Apr 2011 12:09:26 +0000 (14:09 +0200)]
Make debug output more portable.
Christoph Mallon [Thu, 14 Apr 2011 11:53:37 +0000 (13:53 +0200)]
Reduce code duplication in the (long)-double-int-union-hack.
Christoph Mallon [Thu, 14 Apr 2011 09:26:00 +0000 (11:26 +0200)]
Let the block walker enter endless loops only at kept blocks, not Phis.
- This is sufficient, because every endless loop must have one kept block.
- This also prevents that the if-conversion enters endless loops at strange points.
This lead to exchanging a block, which is currently being recursed, so when later returning to the exchanged block (which now is an Id-node) in the walker stuff blew up.
Matthias Braun [Wed, 13 Apr 2011 12:18:35 +0000 (14:18 +0200)]
adapt sparc backend to new prolog/epilog handling
Matthias Braun [Wed, 13 Apr 2011 13:01:29 +0000 (15:01 +0200)]
panic if TLS is found in sparc backend
Matthias Braun [Wed, 13 Apr 2011 12:12:53 +0000 (14:12 +0200)]
remove r_keep_alive and let keep_alive decide irg based on the node
Matthias Braun [Wed, 13 Apr 2011 11:06:37 +0000 (13:06 +0200)]
remove stray declaration
Michael Beck [Wed, 13 Apr 2011 11:32:36 +0000 (13:32 +0200)]
BugFix: fixed access to free'ed pointer.
Michael Beck [Wed, 13 Apr 2011 11:13:46 +0000 (13:13 +0200)]
C99 feature removed.
Matthias Braun [Wed, 13 Apr 2011 10:26:08 +0000 (12:26 +0200)]
fix incorrect and(Cmp(a,0),Cmp(b,0)) localopt
Matthias Braun [Tue, 12 Apr 2011 11:03:49 +0000 (13:03 +0200)]
add notes about const/pure properties and inifite loops
Matthias Braun [Tue, 12 Apr 2011 09:08:18 +0000 (11:08 +0200)]
avoid new_NoMem in favor or get_irg_no_mem
Matthias Braun [Mon, 11 Apr 2011 17:12:50 +0000 (19:12 +0200)]
fix some warnings by removing the strange abi_get_ignore_irn API
Matthias Braun [Mon, 11 Apr 2011 12:46:35 +0000 (14:46 +0200)]
remove old+unused code
Matthias Braun [Mon, 11 Apr 2011 12:45:34 +0000 (14:45 +0200)]
move private API into .c file
Matthias Braun [Mon, 11 Apr 2011 11:47:48 +0000 (13:47 +0200)]
add another out-edges hack to avoid finding dead code
Sebastian Buchwald [Tue, 12 Apr 2011 08:51:17 +0000 (10:51 +0200)]
Added missing proj nr.
Sebastian Buchwald [Mon, 11 Apr 2011 13:35:50 +0000 (15:35 +0200)]
Both memory projs are involved in the transformation.
This fixes opt/fehler183.c.
Sebastian Buchwald [Mon, 11 Apr 2011 09:08:15 +0000 (11:08 +0200)]
Fixed warning.
Sebastian Buchwald [Mon, 11 Apr 2011 09:10:51 +0000 (11:10 +0200)]
Fixed some whitespaces.
Andreas Zwinkau [Mon, 11 Apr 2011 08:56:38 +0000 (10:56 +0200)]
Simplify code
Michael Beck [Sat, 9 Apr 2011 01:04:25 +0000 (03:04 +0200)]
C99 features removed.
Michael Beck [Sat, 9 Apr 2011 00:59:05 +0000 (02:59 +0200)]
BugFix: fixed type name.
Andreas Zwinkau [Fri, 8 Apr 2011 14:21:07 +0000 (16:21 +0200)]
Remove superfluous asserts
Matthias Braun [Fri, 8 Apr 2011 13:34:21 +0000 (15:34 +0200)]
more aggressive node-killing to avoid dangling edges
Andreas Zwinkau [Fri, 8 Apr 2011 13:43:15 +0000 (15:43 +0200)]
Merge branch 'kaps-included' into master
Matthias Braun [Fri, 1 Apr 2011 08:17:03 +0000 (10:17 +0200)]
Another rewrite of prolog/epilog handling: Delay their creation until after register allocation
Andreas Zwinkau [Fri, 8 Apr 2011 12:44:11 +0000 (14:44 +0200)]
Remove ifdef FIRM_KAPS
Always make pbqp coloring available
Andreas Zwinkau [Fri, 8 Apr 2011 12:14:10 +0000 (14:14 +0200)]
Include kaps into build
Andreas Zwinkau [Fri, 8 Apr 2011 12:10:06 +0000 (14:10 +0200)]
merge kaps
Sebastian Buchwald [Tue, 22 Mar 2011 18:31:38 +0000 (19:31 +0100)]
Revived grgen dumper for TTC 2011.
Matthias Braun [Thu, 31 Mar 2011 17:47:52 +0000 (19:47 +0200)]
no conservative recalculation of everything in irlivechk
Matthias Braun [Tue, 5 Apr 2011 14:59:36 +0000 (16:59 +0200)]
all backends have decreasing stack
Matthias Braun [Thu, 31 Mar 2011 17:17:35 +0000 (19:17 +0200)]
change get_ignore_irn API to take irg instead of abi
Matthias Braun [Thu, 31 Mar 2011 11:40:12 +0000 (13:40 +0200)]
rename be_pos_XXX to n_be_XXX to be consistent with the architecture specific constants
Matthias Braun [Thu, 31 Mar 2011 07:57:30 +0000 (09:57 +0200)]
cleanup iredges API a bit
Matthias Braun [Wed, 30 Mar 2011 16:49:33 +0000 (18:49 +0200)]
identify bitfield types based on whether they have a base_type set (maybe should introduce a real type for bitfield types?)
Sebastian Buchwald [Thu, 17 Mar 2011 17:56:08 +0000 (18:56 +0100)]
Use a more scalable algorithm for SSA reconstruction.
The algorithm now processes a block at most once,
instead of |uses| times.
Sebastian Buchwald [Fri, 8 Apr 2011 09:03:17 +0000 (11:03 +0200)]
Revert "Prepared for diff."
This reverts commit
0a8187425d279eb16ba56f5f04465653a9dfb26a.
Sebastian Buchwald [Thu, 17 Mar 2011 17:56:08 +0000 (18:56 +0100)]
Prepared for diff.
Sebastian Buchwald [Wed, 9 Mar 2011 15:07:01 +0000 (16:07 +0100)]
Fixed some typos.
Christoph Mallon [Thu, 7 Apr 2011 11:43:00 +0000 (13:43 +0200)]
Remove stale ENUM_BITSET(pn_Cmp).
Michael Beck [Thu, 7 Apr 2011 07:28:36 +0000 (09:28 +0200)]
irg_verify() now runs on unpinned graphs without asserting.
Christoph Mallon [Wed, 6 Apr 2011 13:27:06 +0000 (15:27 +0200)]
Fix regression: Allow setting the current block to NULL.
Michael Beck [Sun, 3 Apr 2011 16:44:25 +0000 (18:44 +0200)]
Fixed tail recursion optimization.
- use get_r/set_r instead of get_/set_* functions, so we
can operate without current_ir_graph set
- set current_ir_graph, but only for debugging now
- fixed wrong usage of indexes for local SSA variables
- placed resorce checking at the right plaace
- removed #if 0'ed debug code and replace it by DB()
- fixed doxygen comments
Michael Beck [Sun, 3 Apr 2011 16:36:47 +0000 (18:36 +0200)]
Add assertions if set_cur_block() is tried for a block on a wrong irg.
Fixed typo.
Michael Beck [Sun, 3 Apr 2011 15:09:10 +0000 (17:09 +0200)]
C99 feature removed.
Matthias Braun [Wed, 30 Mar 2011 14:46:12 +0000 (16:46 +0200)]
Revert "merge gcse into place_code (a better solution would be having it always on until the backend scheduler)"
This reverts commit
bdddae273a9cd497dc79c2f90726482620d3c8d4.
Matthias Braun [Wed, 23 Mar 2011 13:20:34 +0000 (14:20 +0100)]
simplify Sel lowering code
Matthias Braun [Wed, 23 Mar 2011 13:19:33 +0000 (14:19 +0100)]
amend to tls_chagnes
Matthias Braun [Wed, 23 Mar 2011 13:01:42 +0000 (14:01 +0100)]
remove get_irg_tls() concept, simply use SymConst and let the backend figure it out
Matthias Braun [Thu, 17 Mar 2011 17:59:05 +0000 (18:59 +0100)]
remove the deprecated and unused construct of a value_res_base entities in method-types
Matthias Braun [Thu, 17 Mar 2011 15:30:47 +0000 (16:30 +0100)]
remove names from parameters of method_types - it is pure debug info and should already be contained in type_dbg_info if available
Matthias Braun [Thu, 17 Mar 2011 15:09:52 +0000 (16:09 +0100)]
remove pointless first_variadic_param attribute from method types
Matthias Braun [Thu, 17 Mar 2011 10:19:16 +0000 (11:19 +0100)]
rename Store_align/Load_align to XXX_unaligned and correctly put it into the spec file
Matthias Braun [Wed, 30 Mar 2011 12:48:42 +0000 (14:48 +0200)]
simplify opt_funccall interface
Matthias Braun [Wed, 30 Mar 2011 13:10:30 +0000 (15:10 +0200)]
merge gcse into place_code (a better solution would be having it always on until the backend scheduler)
Matthias Braun [Fri, 25 Mar 2011 16:31:51 +0000 (17:31 +0100)]
fix wrong usage of inverted vs negated condition codes
Matthias Braun [Fri, 25 Mar 2011 15:05:27 +0000 (16:05 +0100)]
fix ir_is_equality_cmp_0 returning true for signed_var > 0
Matthias Braun [Fri, 25 Mar 2011 15:05:07 +0000 (16:05 +0100)]
fix firm producing invalid code for return a > 0 with a being an int
Michael Beck [Thu, 24 Mar 2011 23:32:07 +0000 (00:32 +0100)]
Updated comment.
Michael Beck [Thu, 24 Mar 2011 23:30:05 +0000 (00:30 +0100)]
Fix typos in comments: s/wether/whether/ and related corrections.
Michael Beck [Thu, 24 Mar 2011 23:16:08 +0000 (00:16 +0100)]
C99 feature removed.
Michael Beck [Thu, 24 Mar 2011 23:01:28 +0000 (00:01 +0100)]
Use debug printer instead direct output, fix some size_t related warnings.
Matthias Braun [Thu, 24 Mar 2011 19:28:16 +0000 (20:28 +0100)]
fix and(cmp,cmp) optimisation rule, loosen verify to allow and/or/xor for reference modes
Matthias Braun [Thu, 24 Mar 2011 17:29:23 +0000 (18:29 +0100)]
implement some localopt rules for And(Cmp(a,0), Cmp(b,0))
Matthias Braun [Thu, 24 Mar 2011 16:48:24 +0000 (17:48 +0100)]
fix lower_d lowering Cmps too realy resulting in unnecessary 0/1 production
Matthias Braun [Tue, 22 Mar 2011 19:01:54 +0000 (20:01 +0100)]
pin prolog/epilog nodes
Matthias Braun [Tue, 22 Mar 2011 19:00:50 +0000 (20:00 +0100)]
add exc_attr to all be_nodes so you can dynamically pin/unpin them
Christoph Mallon [Thu, 17 Mar 2011 13:01:38 +0000 (14:01 +0100)]
Move macro constant into the only file, which uses it.
Christoph Mallon [Thu, 17 Mar 2011 13:01:05 +0000 (14:01 +0100)]
Remove dead assignment und subsequently unused macro constant.
Christoph Mallon [Thu, 17 Mar 2011 13:00:12 +0000 (14:00 +0100)]
s/keeped/kept/.
Matthias Braun [Thu, 17 Mar 2011 13:34:38 +0000 (14:34 +0100)]
ia32: get_eip and fpu_trunc_mode are not always NULL before transformation
Michael Beck [Thu, 17 Mar 2011 12:20:01 +0000 (13:20 +0100)]
Removed unused variables.
Matthias Braun [Wed, 16 Mar 2011 17:40:35 +0000 (18:40 +0100)]
load-after-load and load-after-store with different mode was implemented in a little-endian way, disable on big-endian (until we have a real implementation)
Matthias Braun [Wed, 16 Mar 2011 17:28:53 +0000 (18:28 +0100)]
remove unused cacheopt code
Matthias Braun [Wed, 16 Mar 2011 09:23:51 +0000 (10:23 +0100)]
sparc: abort when encountering (non-lowered) 64bit code
Matthias Braun [Tue, 15 Mar 2011 16:03:15 +0000 (17:03 +0100)]
emit global asms in all backends
Matthias Braun [Tue, 15 Mar 2011 15:58:53 +0000 (16:58 +0100)]
bestabs: move stabs but not backend specific text0: code into stabs
Matthias Braun [Tue, 15 Mar 2011 15:52:26 +0000 (16:52 +0100)]
bedbgout: improve function names, ensure that the end of compilation unit thing is actually called
Matthias Braun [Mon, 14 Mar 2011 17:56:59 +0000 (18:56 +0100)]
fix bug in switch lowerer (introduced when Cmp changed)
Matthias Braun [Mon, 7 Mar 2011 19:04:07 +0000 (20:04 +0100)]
sparc: to avoid trouble, move all nodes before sparc_Restore
Matthias Braun [Mon, 7 Mar 2011 19:01:17 +0000 (20:01 +0100)]
avoid verifier warnings in sparc by explicitely modeling RestoreZero using the frame pointer