- Oct 13, 2009
-
-
Bob Wilson authored
llvm-svn: 83973
-
- Oct 10, 2009
-
-
Dan Gohman authored
when loading from an invariant memory location. llvm-svn: 83688
-
- Oct 06, 2009
-
-
Bob Wilson authored
Patch by Johnny Chen. llvm-svn: 83407
-
- Oct 01, 2009
-
-
Evan Cheng authored
ld / st pairs, etc. llvm-svn: 83197
-
Evan Cheng authored
Change ld/st multiples to explicitly model the writeback to base register. This fixes most of the -ldstopti-before-sched2 regressions. llvm-svn: 83191
-
- Sep 30, 2009
-
-
Jim Grosbach authored
Patch by Sylvere Teissier. llvm-svn: 83135
-
- Sep 28, 2009
-
-
Evan Cheng authored
instruction. This makes it re-materializable. Thumb2 will split it back out into two instructions so IT pass will generate the right mask. Also, this expose opportunies to optimize the movw to a 16-bit move. llvm-svn: 82982
-
Anton Korobeynikov authored
Disable rematting of it for now. llvm-svn: 82975
-
Anton Korobeynikov authored
This should be better than single load from constpool. llvm-svn: 82948
-
- Sep 09, 2009
-
-
Evan Cheng authored
llvm-svn: 81310
-
- Sep 04, 2009
-
-
David Goodwin authored
llvm-svn: 80956
-
- Sep 01, 2009
-
-
David Goodwin authored
llvm-svn: 80699
-
- Aug 28, 2009
-
-
Evan Cheng authored
llvm-svn: 80350
-
- Aug 27, 2009
-
-
Misha Brukman authored
See http://llvm.org/PR4687 for more info and links. llvm-svn: 80244
-
- Aug 21, 2009
-
-
Bob Wilson authored
several things other than Neon vector lane numbers. For inline assembly operands with a "c" print code, check that they really are immediates. llvm-svn: 79676
-
- Aug 19, 2009
-
-
David Goodwin authored
llvm-svn: 79436
-
- Aug 13, 2009
-
-
Jim Grosbach authored
llvm-svn: 78918
-
David Goodwin authored
llvm-svn: 78908
-
Jim Grosbach authored
llvm-svn: 78905
-
Jim Grosbach authored
llvm-svn: 78904
-
- Aug 12, 2009
-
-
David Goodwin authored
Enhance the InstrStage object to enable the specification of an Itinerary with overlapping stages. The default is to maintain the current behavior that the "next" stage immediately follows the previous one. llvm-svn: 78827
-
Jim Grosbach authored
llvm-svn: 78806
-
- Aug 11, 2009
-
-
Owen Anderson authored
the latter is capable of representing either a primitive or an extended type. llvm-svn: 78713
-
Jim Grosbach authored
and short. Well, it's kinda short. Definitely nasty and brutish. The front-end generates the register/unregister calls into the SjLj runtime, call-site indices and landing pad dispatch. The back end fills in the LSDA with the call-site information provided by the front end. Catch blocks are not yet implemented. Built on Darwin and verified no llvm-core "make check" regressions. llvm-svn: 78625
-
Owen Anderson authored
llvm-svn: 78610
-
- Aug 09, 2009
-
-
Anton Korobeynikov authored
'no_hash' modifier. Hopefully this will make Daniel happy :) llvm-svn: 78514
-
- Aug 08, 2009
-
-
Anton Korobeynikov authored
llvm-svn: 78468
-
- Aug 06, 2009
-
-
David Goodwin authored
Add parameter to pattern classes to enable an itinerary to be specified for instructions. For now just use the existing itineraries or NoItinerary. llvm-svn: 78321
-
- Aug 04, 2009
-
-
David Goodwin authored
Initial support for single-precision FP using NEON. Added "neonfp" attribute to enable. Added patterns for some binary FP operations. llvm-svn: 78081
-
- Jul 29, 2009
-
-
Evan Cheng authored
llvm-svn: 77507
-
Evan Cheng authored
- Darwin Thumb2 call clobbers r9. llvm-svn: 77500
-
Evan Cheng authored
llvm-svn: 77422
-
- Jul 28, 2009
-
-
Evan Cheng authored
In thumb2 mode, add pc is unpredictable. Use add + mov pc instead (that is until more optimization goes in). llvm-svn: 77364
-
- Jul 25, 2009
-
-
Evan Cheng authored
Before: adr r12, #LJTI3_0_0 ldr pc, [r12, +r0, lsl #2] LJTI3_0_0: .long LBB3_24 .long LBB3_30 .long LBB3_31 .long LBB3_32 After: adr r12, #LJTI3_0_0 add pc, r12, +r0, lsl #2 LJTI3_0_0: b.w LBB3_24 b.w LBB3_30 b.w LBB3_31 b.w LBB3_32 This has several advantages. 1. This will make it easier to optimize this to a TBB / TBH instruction + (smaller) table. 2. This eliminate the need for ugly asm printer hack to force the address into thumb addresses (bit 0 is one). 3. Same codegen for pic and non-pic. 4. This eliminate the need to align the table so constantpool island pass won't have to over-estimate the size. Based on my calculation, the later is probably slightly faster as well since ldr pc with shifter address is very slow. That is, it should be a win as long as the HW implementation can do a reasonable job of branch predict the second branch. llvm-svn: 77024
-
- Jul 23, 2009
-
-
Evan Cheng authored
llvm-svn: 76803
-
- Jul 22, 2009
-
-
Evan Cheng authored
llvm-svn: 76729
-
- Jul 14, 2009
-
-
Evan Cheng authored
2. BX does not "use" the link register, it defines it. 3. Fix a couple more places in thumb td file that still uses pre-UAL syntax. llvm-svn: 75585
-
David Goodwin authored
llvm-svn: 75576
-
- Jul 11, 2009
-
-
Evan Cheng authored
Major changes to Thumb (not Thumb2). Many 16-bit instructions either modifies CPSR when they are outside the IT blocks, or they can predicated when in Thumb2. Move the implicit def of CPSR to an optional def which defaults CPSR. This allows the 's' bit to be toggled dynamically. A side-effect of this change is asm printer is now using unified assembly. There are some minor clean ups and fixes as well. llvm-svn: 75359
-
- Jul 10, 2009
-
-
Evan Cheng authored
llvm-svn: 75187
-