 sergei
|
9b9ea15d68
+ m68k: initial implementation of do_spill_replace (supports only MOVE instructions for now). Tested with qemu for Coldfire target.
|
11 lat temu |
 Jonas Maebe
|
246caf3526
* never alias/coalesce cpu registers that are not usable by the register
|
11 lat temu |
 florian
|
23c724f885
* prevent a temp. register allocated during spilling being immediately spilled again, resolves #25164
|
11 lat temu |
|
florian
|
829764e96b
+ replace registers to spill in mov instructions and convert them into ldr/str
|
11 lat temu |
 Károly Balogh
|
587acf6452
fix infinite spilling on m68k, patch by Florian, additional IE workaround by me, based on a similar hack in the ARM cg...
|
11 lat temu |
|
Károly Balogh
|
70f9b098e0
no longer define a segment field for m68k
|
11 lat temu |
|
sergei
|
d2a9308181
+ SPARC: implemented register spill replacement.
|
11 lat temu |
 svenbarth
|
32a18199c5
Print aX or aregX for address registers if DEBUG_REGISTERLIFE is defined.
|
12 lat temu |
|
florian
|
dc567acc9a
* accumulate register weight instead of taking only the maximum
|
12 lat temu |
|
florian
|
c2f7ccb253
* coalesce more aggressive
|
12 lat temu |
|
sergei
|
8b8553991a
+ MIPS: prevent coalescing written-to registers with $sp,$fp,$zero and $at.
|
12 lat temu |
|
florian
|
23c84fd27c
* more fixes for segment register usage
|
12 lat temu |
|
florian
|
b85addb451
+ color segment register
|
12 lat temu |
|
svenbarth
|
d15304c25e
Also check for the type of the register when replacing it. In certain circumstances this can still lead to access to invalid memory resulting either in an access violation or invalid registers.
|
12 lat temu |
 pierre
|
322b793506
Try to do something for m68k integer/address registers; not working yet :!(
|
12 lat temu |
|
svenbarth
|
2ada9a528b
Fix a critical bug in the register allocator (at least for CPUs with seperate address
|
12 lat temu |
|
florian
|
95830d5a4a
+ DEBUG_REGISTERLIFE define
|
13 lat temu |
|
florian
|
f75e39e660
+ aliasing of registers, allows to split live ranges
|
13 lat temu |
|
florian
|
f9dd5bf5e6
+ live_start and live_end_properties
|
13 lat temu |
|
florian
|
c2d067940a
* get rid of hash chars, modern tools and IDEs can use normal comments as documentation if needed
|
13 lat temu |
|
florian
|
b78f9826fb
* call AllocMem instead of GetMem/FillChar
|
13 lat temu |
|
florian
|
ecdec2e431
* disable reg. allocator optimization introduced in r21812 for now because it caused some trouble, see #22405
|
13 lat temu |
 masta
|
f9cdf3d4ca
Be more specific in spilling debug messages
|
13 lat temu |
|
masta
|
aef81cad68
Add debug information about register spilling
|
13 lat temu |
|
florian
|
9833f642fe
* coalesce if either u or v are precolored, tested on arm and i386, it gives for both a small advantage
|
13 lat temu |
|
Jonas Maebe
|
0fc422f244
* moved definition of maxcpuregister and tcpuregisterset from cgbase to
|
13 lat temu |
|
Jonas Maebe
|
6f336a5d13
* the register allocator always knows the exact number of the first imaginary
|
13 lat temu |
|
Jonas Maebe
|
aee5380ae0
* merged trunk up to r20882
|
13 lat temu |
|
florian
|
6211a83865
* write variable locations of 64 bit integer values on 32 bit targets correctly
|
14 lat temu |
|
Jonas Maebe
|
00cb8f5725
* made maxregs related information protected instead of private in trgobj,
|
14 lat temu |
