summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2006-06-01move test for NO_STRING_ADDRESS_IN_DESCRIPTOR after include of cgport.hJohn van Groningen1-3/+4
2006-06-01fix ccall with a real as normal argument and an integer as state argumentJohn van Groningen1-4/+2
(after second : )
2006-05-31pass floating point parameters in registers in ccall for AI64John van Groningen1-24/+36
2006-05-31fix printing of integers of more than 32 bitsJohn van Groningen1-1/+1
2006-05-29use offset -24 for apply update entry if profiling is usedJohn van Groningen2-13/+18
2006-05-29use string with 4 byte length for .string directive for 64 bit architecturesJohn van Groningen1-12/+23
2006-05-11fix ccall on 64 bit platforms with Real resultJohn van Groningen1-1/+1
2006-05-11allow shifts up to 63 bits instead of 31John van Groningen2-4/+4
2006-05-10add shift_s instructions, use RELA for ELF, port to linuxJohn van Groningen1-145/+478
2006-05-10add shift_s instructions, port printf calls to linuxJohn van Groningen1-2/+88
2006-05-01add ccall for 64 bit linuxJohn van Groningen2-5/+366
2006-05-01use int_64 instead of __int64, define int_64, include stdint.h on 64 bits linuxJohn van Groningen12-30/+73
2006-04-25enable NO_STRING_ADDRESS_IN_DESCRIPTOR for sparcJohn van Groningen1-5/+7
2006-04-25optimize apply with >1 argumentsJohn van Groningen2-2/+68
2006-04-25enable NEW_DESCRIPTORS for sparcJohn van Groningen1-1/+1
2006-04-18compute register uses for GFROMF and GTOFJohn van Groningen1-2/+18
2006-04-18fix previous revision by using IBLTU instead of IBLT for unsigned < whenJohn van Groningen1-5/+5
generating code for floating point branch
2006-03-28intel_asm is only defined on IA32 and AMD64John van Groningen1-0/+4
2006-03-24remove use of scratch register on IA32 for shift instructions (by addingJohn van Groningen11-432/+1089
instruction with extra register), IScc, IFScc, IFBcc (by adding IFCcc instructions with extra register), ICMPW (not used anymore) prevent generating FLD and FSTP instructions between FCOMP and FNSTSW instruction
2006-03-24optimize shift n bitand 31 or 64 for IA32 and AMD64John van Groningen1-0/+37
2006-03-24fix entier for 64 bit processorsJohn van Groningen1-4/+8
2006-03-14improve computation of used registers for reading from arrays, preventsJohn van Groningen1-2/+4
incrementing counters too soon in many small loops using one array
2006-02-24add generating instruction for converting real to intJohn van Groningen2-38/+64
2006-01-20fix bugs in code generation for floating point computations for amd64John van Groningen5-7/+56
2006-01-20optimize cmp $0,r , generate test r,r insteadJohn van Groningen4-15/+50
2006-01-20fix incorrect code generation for return with > 7 floating point parametersJohn van Groningen3-0/+43
in registers and < 2 integer parameters in registers for IA32
2006-01-11generate code for amd64John van Groningen4-0/+9953
2006-01-10fiix bug introduced in previous revision because of incorrect merging of ↵John van Groningen1-0/+1
source code
2006-01-10port to 64 bit windowsJohn van Groningen17-553/+1911
2005-12-06add pushLc instruction for pushing the value of a label, withJohn van Groningen3-0/+31
a '_' or '.' added at the beginning if this is specified in the c ABI for this platform
2005-11-24add passing of float argument in ccall for the PowerPCJohn van Groningen1-1/+13
2005-11-21use new descriptors on the PowerPCJohn van Groningen1-3/+4
2005-11-21change descriptors for apply with >1 argument and selectorsJohn van Groningen2-141/+223
2005-11-11add .descs directive for selectorsJohn van Groningen4-8/+94
2005-11-11improve printing of stringsJohn van Groningen2-1/+5
2005-10-10use yet_args_neededed_n labels for n>4 for new descriptorsJohn van Groningen1-1/+5
2005-10-06use new apply and new descriptors for IA32,John van Groningen11-46/+331
add addLU and subLU instructions for IA32
2005-05-18test for GNEG in mark_graph_1 for all platformsJohn van Groningen1-1/+1
2005-05-11fix bug in mulUUL instructionJohn van Groningen1-1/+1
2005-04-14add some optimizations for arrays of unboxed records:John van Groningen3-87/+165
on the PowerPC and Sparc if the offset!=0 and the index!=NULL the index now contains the address of array + the offset of the element, the offset computation of r_select can be reused by r_update, r_update tries to avoid unnecessary updates, r_replace recognizes floating point stores, and sometimes floating point loads
2005-04-14use fmadd also when add uses indirect loadJohn van Groningen2-83/+158
2005-04-01add Makefiles for MacOS XJohn van Groningen3-0/+138
2005-03-16add some tests for MAIN_CLMJohn van Groningen1-3/+3
2005-03-16don't always define G_MACH_O for GNU_CJohn van Groningen1-1/+1
2005-03-16remove test for PROJECT_BUILDERJohn van Groningen1-1/+1
2005-01-24codegenerator makefile for cygwinJohn van Groningen1-0/+14
2005-01-19implement mulUUL and divLU instructions for IA32,John van Groningen11-370/+1406
define acos, asin, exp, ln and log10 instructons only for M68000 platform, add code for jsr_ap and jmp_ap instructions (not yet enabled)
2005-01-13add divU for the PowerPCJohn van Groningen10-17/+56
2005-01-10use r_to_i_buffer only for PowerPC and not for IA32John van Groningen1-4/+5
2005-01-10fix 4 byte alignment after replacing jmp with a compare, branch and jump toJohn van Groningen1-0/+6
next block, and then remove the latter jump
generated by cgit v1.2.3 (git 2.39.1) at 2025-03-15 15:47:52 +0000