wangtao
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
14,588 Commits 29 Branches 0 Tags 2.2 GiB
Commit Graph

86 Commits

Author SHA1 Message Date
Nate Begeman fac8529df8 Implement floating point select for lt, gt, le, ge using the powerpc fsel 
instruction.

Now, rather than emitting the following loop out of bisect:
.LBB_main_19:	; no_exit.0.i
	rlwinm r3, r2, 3, 0, 28
	lfdx f1, r3, r27
	addis r3, r30, ha16(.CPI_main_1-"L00000$pb")
	lfd f2, lo16(.CPI_main_1-"L00000$pb")(r3)
	fsub f2, f2, f1
	addis r3, r30, ha16(.CPI_main_1-"L00000$pb")
	lfd f4, lo16(.CPI_main_1-"L00000$pb")(r3)
	fcmpu cr0, f1, f4
	bge .LBB_main_64	; no_exit.0.i
.LBB_main_63:	; no_exit.0.i
	b .LBB_main_65	; no_exit.0.i
.LBB_main_64:	; no_exit.0.i
	fmr f2, f1
.LBB_main_65:	; no_exit.0.i
	addi r3, r2, 1
	rlwinm r3, r3, 3, 0, 28
	lfdx f1, r3, r27
	addis r3, r30, ha16(.CPI_main_1-"L00000$pb")
	lfd f4, lo16(.CPI_main_1-"L00000$pb")(r3)
	fsub f4, f4, f1
	addis r3, r30, ha16(.CPI_main_1-"L00000$pb")
	lfd f5, lo16(.CPI_main_1-"L00000$pb")(r3)
	fcmpu cr0, f1, f5
	bge .LBB_main_67	; no_exit.0.i
.LBB_main_66:	; no_exit.0.i
	b .LBB_main_68	; no_exit.0.i
.LBB_main_67:	; no_exit.0.i
	fmr f4, f1
.LBB_main_68:	; no_exit.0.i
	fadd f1, f2, f4
	addis r3, r30, ha16(.CPI_main_2-"L00000$pb")
	lfd f2, lo16(.CPI_main_2-"L00000$pb")(r3)
	fmul f1, f1, f2
	rlwinm r3, r2, 3, 0, 28
	lfdx f2, r3, r28
	fadd f4, f2, f1
	fcmpu cr0, f4, f0
	bgt .LBB_main_70	; no_exit.0.i
.LBB_main_69:	; no_exit.0.i
	b .LBB_main_71	; no_exit.0.i
.LBB_main_70:	; no_exit.0.i
	fmr f0, f4
.LBB_main_71:	; no_exit.0.i
	fsub f1, f2, f1
	addi r2, r2, -1
	fcmpu cr0, f1, f3
	blt .LBB_main_73	; no_exit.0.i
.LBB_main_72:	; no_exit.0.i
	b .LBB_main_74	; no_exit.0.i
.LBB_main_73:	; no_exit.0.i
	fmr f3, f1
.LBB_main_74:	; no_exit.0.i
	cmpwi cr0, r2, -1
	fmr f16, f0
	fmr f17, f3
	bgt .LBB_main_19	; no_exit.0.i

We emit this instead:
.LBB_main_19:	; no_exit.0.i
	rlwinm r3, r2, 3, 0, 28
	lfdx f1, r3, r27
	addis r3, r30, ha16(.CPI_main_1-"L00000$pb")
	lfd f2, lo16(.CPI_main_1-"L00000$pb")(r3)
	fsub f2, f2, f1
	fsel f1, f1, f1, f2
	addi r3, r2, 1
	rlwinm r3, r3, 3, 0, 28
	lfdx f2, r3, r27
	addis r3, r30, ha16(.CPI_main_1-"L00000$pb")
	lfd f4, lo16(.CPI_main_1-"L00000$pb")(r3)
	fsub f4, f4, f2
	fsel f2, f2, f2, f4
	fadd f1, f1, f2
	addis r3, r30, ha16(.CPI_main_2-"L00000$pb")
	lfd f2, lo16(.CPI_main_2-"L00000$pb")(r3)
	fmul f1, f1, f2
	rlwinm r3, r2, 3, 0, 28
	lfdx f2, r3, r28
	fadd f4, f2, f1
	fsub f5, f0, f4
	fsel f0, f5, f0, f4
	fsub f1, f2, f1
	addi r2, r2, -1
	fsub f2, f1, f3
	fsel f3, f2, f3, f1
	cmpwi cr0, r2, -1
	fmr f16, f0
	fmr f17, f3
	bgt .LBB_main_19	; no_exit.0.i

llvm-svn: 16764
 2004-10-06 09:53:04 +00:00
Nate Begeman 2f1d0ae95e Generate better code by being far less clever when it comes to the select instruction. Don't create overlapping register lifetimes 
llvm-svn: 16580
 2004-09-29 05:00:31 +00:00
Nate Begeman 7b6df6def2 improve Type::BoolTy codegen by eliminating unnecessary clears and sign extends 
llvm-svn: 16578
 2004-09-29 03:45:33 +00:00
Nate Begeman 26566f0b68 To go along with sabre's improved InstCombining, improve recognition of 
integers that we can use as immediate values in instructions.

Example from yacr2:
-       lis r10, -1
-       ori r10, r10, 65535
-       add r28, r28, r10
+       addi r28, r28, -1
        addi r7, r7, 1
        addi r9, r9, 1
        b .LBB_main_9   ; loopentry.1.i214

llvm-svn: 16566
 2004-09-29 02:35:05 +00:00
Nate Begeman 8656a156cf Correct some BuildMI arguments for the upcoming simple scheduler 
llvm-svn: 16519
 2004-09-27 05:08:17 +00:00
Nate Begeman 49cf74b26c Fix the last of the major PPC GEP folding deficiencies. This will allow 
the ISel to use indexed and non-zero immediate offsets for GEPs that have
more than one use.  This is common for instruction sequences such as a load
followed by a modify and store to the same address.

llvm-svn: 16493
 2004-09-23 05:31:33 +00:00
Nate Begeman 033b816171 add optimized code sequences for setcc x, 0 
llvm-svn: 16478
 2004-09-22 04:40:25 +00:00
Misha Brukman 87201ce8f9 s/ISel/PPC32ISel/ to have unique class names for debugging via gdb because the 
C++ front-end in gcc does not mangle classes in anonymous namespaces correctly.

llvm-svn: 16470
 2004-09-21 18:22:19 +00:00
Nate Begeman 4bfceb1ed5 All PPC instructions are now auto-printed 
32 and 64 bit AsmWriters unified
Darwin and AIX specific features of AsmWriter split out

llvm-svn: 16163
 2004-09-04 05:00:00 +00:00
Nate Begeman 6173878304 Convert remaining X-Form and Pseudo instructions over to asm writer 
llvm-svn: 16142
 2004-09-02 08:13:00 +00:00
Reid Spencer 7c16caa336 Changes For Bug 352 
Move include/Config and include/Support into include/llvm/Config,
include/llvm/ADT and include/llvm/Support. From here on out, all LLVM
public header files must be under include/llvm/.

llvm-svn: 16137
 2004-09-01 22:55:40 +00:00
Nate Begeman 4483df8b63 Implement the following missing functionality in the PPC backend: 
cast fp->bool
cast ulong->fp
algebraic right shift long by non-constant value
These changes tested across most of the test suite.  Fixes Regression/casts

llvm-svn: 16081
 2004-08-29 08:19:32 +00:00
Nate Begeman 1c57b4fa32 Kill a majority of unnecessary sign extensions for byte loads 
llvm-svn: 15991
 2004-08-22 08:10:15 +00:00
Nate Begeman 45b0b7cd7c Back out branchless SetCC code. While it helped a lot in some cases, it 
hurt a lot in others.  Instead, improve branching version of SetCC and
Select instructions.  The old code will be in CVS should we ever need to
dig it up again.

llvm-svn: 15979
 2004-08-21 20:42:14 +00:00
Nate Begeman 1b1a784afa Implement code to convert SetCC into straight line code where appropriate. Add necessary instructions for this transformation to the .td file. 
llvm-svn: 15952
 2004-08-20 09:56:22 +00:00
Misha Brukman 75e987d0b8 This PHI has 4 additional operands, not 2. 
llvm-svn: 15926
 2004-08-19 21:00:12 +00:00
Nate Begeman d5c6380015 Convert casts that will have no effect into move instructions. 
llvm-svn: 15914
 2004-08-19 08:07:50 +00:00
Nate Begeman e4e6d92d1d Clean up floating point instruction selection. 
Change int->float cast code to put conversion constants in constant pool.
Shorten code sequence for constant pool fp loads.
Remove LOADLoDirect/LOADLoIndirect psuedo instructions and tweak asmwriter

llvm-svn: 15913
 2004-08-19 05:20:54 +00:00
Nate Begeman 0818541631 Re-fix hiding the Frame Pointer from the register allocator in functions 
that have a frame pointer.  This change fixes Burg.  In addition, make
the necessary changes to floating point code gen and constant loading after
Chris Lattner's fixes to the asm writer.  These changes fix MallocBench/gs

llvm-svn: 15873
 2004-08-17 07:17:44 +00:00
Misha Brukman 116f9277f6 PowerPC 32-/64-bit split: Part I, PPC32* bit files, adapted from former PowerPC* 
llvm-svn: 15850
 2004-08-17 04:55:41 +00:00
Nate Begeman 8b44a07246 Fix mismatched adjust down/up of SP in functions that contain variable 
sized allocas.

llvm-svn: 15806
 2004-08-16 01:50:22 +00:00
Nate Begeman 373744c6dc Fix float to int codepath by always allocating 8 bytes for the target of a double store; optimize cmplwi generation. 
llvm-svn: 15759
 2004-08-15 06:42:28 +00:00
Nate Begeman caeb78e720 Fix handling of FP constants with single precision, and loading of internal linkage function addresses 
llvm-svn: 15742
 2004-08-14 22:11:38 +00:00
Nate Begeman 5bf9bfe398 Fix siod by switching BoolTy to byte rather than int until CFE changes for 
Darwin.  Also, change asm printer to output proper stubs for external
functions whose address is passed as an argument to aid in bugpointing.

llvm-svn: 15721
 2004-08-13 09:32:01 +00:00
Nate Begeman 420213f3c5 Fix 177.mesa compilation, don't use floating point regs for base addresses! 
llvm-svn: 15720
 2004-08-13 04:45:14 +00:00
Nate Begeman 2f1d849271 Fix llc crasher compiling siod by giving BuildMI the correct number of arguments 
llvm-svn: 15719
 2004-08-13 03:56:49 +00:00
Nate Begeman f17ea0f7b7 Clean up 32/64bit and Darwin/AIX split. Next steps: 64 bit ISel, AIX asm printer. 
llvm-svn: 15662
 2004-08-11 07:40:04 +00:00
Chris Lattner 6f0291792e Fix a case where constantexprs could leak into the PPC isel. 
llvm-svn: 15661
 2004-08-11 07:34:50 +00:00
Nate Begeman 7526da6ead Fix 255.vortex by using getClassB instead of getClass 
llvm-svn: 15648
 2004-08-11 03:30:55 +00:00
Misha Brukman d022a5ac5a Breaking up the PowerPC target into 32- and 64-bit subparts, Part I: 32-bit. 
llvm-svn: 15634
 2004-08-11 00:09:42 +00:00
Misha Brukman dad438bfb9 Renamed PPC32 (namespace for regs, opcodes) to PPC to include 64-bit targets 
llvm-svn: 15631
 2004-08-10 22:47:03 +00:00
Nate Begeman 63be70d8f2 Fix casts of float to unsigned long 
Replace STDX (store 64 bit int indexed) with STFDX (store double indexed)
Fix latent bug in indexed load generation
Generate indexed loads and stores in many more cases

llvm-svn: 15626
 2004-08-10 20:42:36 +00:00
Chris Lattner a8dcf2423e Changes commited for Nate Begeman: 
Use a PowerPC specific prolog epilog inserter to control where spilled
callee save regs are placed on the stack.
Get rid of implicit return address stack slot, save return address reg
(LR) in appropriate slot
Improve code generated for functions that don't have calls or access
globals


Note from Chris: PowerPCPEI will eventually be eliminated, once the
functionality is merged into CodeGen/PrologEpilogInserter.cpp

llvm-svn: 15536
 2004-08-06 06:58:50 +00:00
Misha Brukman 862bb562cc Simplify loading (un)signed constants to registers, patch by Nate Begeman. 
llvm-svn: 15306
 2004-07-28 19:13:49 +00:00
Misha Brukman f7037385e2 LI can only take signed values, so values > 32767 can only be loaded with ORI 
llvm-svn: 15299
 2004-07-28 00:56:04 +00:00
Misha Brukman 5b092c15c6 Build COND_BRANCHes which may become long or short, decided by a later pass. 
Patch by Nate Begeman.

llvm-svn: 15282
 2004-07-27 18:35:23 +00:00
Misha Brukman 640ec1e5e6 Add IMPLICIT_DEF of LR for branch-and-link instrs (calls and global accesses) 
llvm-svn: 15270
 2004-07-27 17:13:58 +00:00
Misha Brukman 74e3a17958 * Rewrote casts 
* Implemented GEP folding
* Dynamically output global address stuff once per function
* Fix casting fp<->short/byte

Patch contributed by Nate Begeman.

llvm-svn: 15237
 2004-07-26 18:13:24 +00:00
Misha Brukman c150bdb52e Eliminate spurious empty space; make code easier to page through. 
llvm-svn: 15146
 2004-07-23 22:35:49 +00:00
Misha Brukman 26a04b79cb Implement casting a floating point to 32-bit unsigned value 
llvm-svn: 15143
 2004-07-23 20:32:59 +00:00
Misha Brukman 7dba17d1ce * Codegen of GEPs dramatically improved by folding multiplies and adds 
* Function pointers implemented correctly using appropriate stubs

Contributed by Nate Begeman.

llvm-svn: 15133
 2004-07-23 16:08:20 +00:00
Misha Brukman db72025d1d * Change class of BoolTy back to cInt 
* Fix indentation back to 2 spaces

llvm-svn: 15121
 2004-07-23 01:11:19 +00:00
Misha Brukman 333f58836c * Change bool from cInt to cByte (for now) 
* Don't allow negative immediates to users of unsigned immediates
* Fix long compares
* Support <const int>, op as a potential immediate candidate
* Fix sign extension of short and byte loads
* Fix and improve integer casts
* Fix passing of doubles as vararg functions

Patch contributed by Nate Begeman.

llvm-svn: 15109
 2004-07-22 15:58:04 +00:00
Misha Brukman dbd14bc3e5 * Add the lost fix to define the second reg of a 2-reg representation of longs 
* Fix opcode RLWNM -> RLWINM since it uses an immediate const shift value

llvm-svn: 15087
 2004-07-21 20:30:18 +00:00
Misha Brukman 87e07f8596 * Speed up canUseAsImmediateForOpcode() by comparing Operand before 
dyn_cast<>ing and checking Constant's value
* Convert tabs to spaces

llvm-svn: 15086
 2004-07-21 20:22:06 +00:00
Misha Brukman 072af37088 * Fix printing of signed immediate values 
* Generation of opcodes that take 16 bit immediates
* Rewrote multiply to be correct for 64 bit values
* Rewrote all the long handling to be correct for PowerPC
* Fix visitSelectInst() to define the upper register of the pair of regs
  representing a long value

Patch contributed by Nate Begeman.

llvm-svn: 15083
 2004-07-21 20:09:08 +00:00
Misha Brukman 8e36261aa8 Treat external variables similarly to those with weak linkage: load indirect. 
llvm-svn: 15047
 2004-07-20 20:43:05 +00:00
Misha Brukman 684ef6ca20 * Differentiate between global and weak symbol loads 
* Fix functions that take more than 32 bytes of args
* Alignment of doubles in structs is 4 bytes, not 8
* Fix passing long args: rN = hi, rN+1 = lo
* Rewrite signed divide
* Rewrite Intrinsic::returnaddress

Patch courtesy of Nate Begeman.

llvm-svn: 15036
 2004-07-20 15:51:37 +00:00
Misha Brukman 5e744fa1a3 Move handing of GlobalValues from getReg() to copyConstantToRegister(), this 
will avoid extra register-to-register copies.  Thanks to Chris for the idea.

llvm-svn: 15019
 2004-07-20 00:59:38 +00:00
Misha Brukman 818a9dc317 * cFP class split into cFP32 and cFP64 
* Fn args passed in registers are now recorded as used by the call instruction
`-> asm printer updated to not print out those registers with the call instr
* Stack frame layout in prolog/epilog fixed, spills and vararg fns now work
* float/double to signed int codegen now correct
* various single precision float codegen bugs fixed
* const integer multiply codegen fixed
* select and setcc blocks inserted into the correct place in machine CFG
* load of integer constant code optimized

All of Shootout tests now work.  Great thanks to Nate Begeman for the patch!

llvm-svn: 15014
 2004-07-20 00:41:46 +00:00