gem5 - gem5

Age	Commit message (Collapse)	Author
2009-07-06	ruby: replace strings that were missed in original ruby import.	Nathan Binkert

2009-07-04	inorder: Fix up some reference stats.	Gabe Black

2009-07-02	ARM: Fix how address mode bits are handled.	Jack Whitham

2009-07-02	ARM: Fix the code snippet for mla.	Jack Whitham

2009-07-02	typo: correct spelling	Nathan Binkert

2009-07-02	attrdict: correct delattr	Nathan Binkert

2009-07-02	stats: fix usage of jobfile	Nathan Binkert

2009-07-01	ARM: Make DataOps select from a set of ways to set the c and v flags.	Gabe Black

2009-07-01	ARM: Get rid of some bitfields that aren't used. A few may need to be readded.	Gabe Black

2009-07-01	ARM: Add a findLsbSet function and use it to implement clz.	Gabe Black

2009-07-01	ARM: Add defaults for DataOp flag code.	Gabe Black

2009-07-01	ARM: Get rid of the val2 variable.	Gabe Black

2009-07-01	ARM: Centralize the declaration of resTemp.	Gabe Black

2009-07-01	ARM: Add a DataImmOp format similar to DataOp.	Gabe Black

2009-07-01	ARM: Decode some media instructions. These are untested.	Gabe Black

2009-07-01	ARM: Use the new DataOp format to simplify the decoder.	Gabe Black

2009-07-01	ARM: Add in some new artificial fields that make decoding a little easier.	Gabe Black

2009-07-01	ARM: Recognize the IntRegs trace flag.	Gabe Black

2009-07-01	ARM: Add a DataOp format so data op definitions can be aggregated.	Gabe Black

2009-06-27	ARM: Show more information when disassembling data processing intstructions.	Gabe Black
	This will need more work, but it should be a lot closer.
2009-06-27	ARM: Show branch targets relative to the nearest symbol.	Gabe Black

2009-06-27	ARM: Write a function for printing mnemonics and predicates.	Gabe Black

2009-06-26	ARM: Fill out the printReg function.	Gabe Black

2009-06-24	ARM: Fix signed multiply long and add some unimplemented loads.	Jack Whitman

2009-06-24	ARM: Link register is trashed by non-executed branch and link operations.	Jack Whitman

2009-06-23	ARM: Added unimplemented load/store multiple instructions.	Jack Whitman

2009-06-23	makerelease: Don't hide x86 any more.	Gabe Black

2009-06-23	Regressions: Build ARM_SE regressions by default.	Gabe Black

2009-06-21	ARM: Simplify some utility functions.	Gabe Black

2009-06-21	ARM: Move util functions out of the isa desc.	Gabe Black

2009-06-21	ARM: Simplify the ISA desc by pulling some classes out of it.	Gabe Black

2009-06-21	ARM: Remove the currently unecessary FPAOp class.	Gabe Black

2009-06-21	ARM: Make inst bitfields accessible outside of the isa desc.	Gabe Black

2009-06-21	ARM: Don't downconvert ExtMachInsts to MachInsts.	Gabe Black

2009-06-21	BitUnion: Add more constiness.	Gabe Black

2009-06-21	ARM: Get rid of a few more unused operands.	Gabe Black

2009-06-21	ARM: Get rid of unnecessary Re operand.	Gabe Black

2009-06-21	ARM: Clear out some inherited hangers on in util.isa and utility.hh.	Gabe Black

2009-06-21	ARM: Get rid of unnecessary fp_enable_checks.	Gabe Black

2009-06-21	ARM: Adjust simplify rotate_imm slightly.	Gabe Black

2009-06-21	ARM: Make the isa parser aware that CPSR is being used.	Gabe Black

2009-06-21	ARM: Pull some static code out of the isa desc and create miscregs.hh.	Gabe Black

2009-06-21	ARM: Get rid of unused postacc_code.	Gabe Black

2009-06-12	scons: Make shared library builds work again	Nathan Binkert
	Compile gzstream as position independent code use the PIC version of date for shared libs...oops
2009-06-10	copyright: I missed some copyrights during ruby integration	Nathan Binkert

2009-06-09	ARM: Add a hello world regression.	Gabe Black

2009-06-09	ARM: Add a hello world binary.	Gabe Black

2009-06-09	ARM: Hook in the mmap2 system call. Make ArmLinuxProcess handle 5,6 syscall ↵	Gabe Black
	params.
2009-06-09	ARM: Add a memory_barrier function to the "comm page".	Gabe Black
	This function doesn't actually provide a memory barrier (I don't think they're implemented) and instead just returns.
2009-06-09	ARM: Add a cmpxchg implementation to the "comm page".	Gabe Black
	This implementation does what it's supposed to (I think), but it's not atomic and doesn't have memory barriers like the kernel's version.