arch/alpha/isa/mem.isa:
Include methods that allow a memory operation to be split between the part that initiates the access, and the part that completes the access. In these functions the Mem variable is explicitly declared; in the default execute functions, the Mem variable is still handled through %(op_decl)s.
arch/isa_parser.py:
Include recording the type of the memory access variable so that it can be used if it needs to be explicitly declared in a template.
Have memory operands consider themselves neither a source nor a destination to avoid including themselves on the op_src_decl list or the op_dest_decl list.
Record op_src_decl and op_dest_decl lists to allow for declaring only source or destination operands. This is needed for the split memory access methods.
--HG--
extra : convert_revision : f674f7a2f747ae40ba8c3a0933b0337c87ee0b6c
instruction format functions in ISA description language.
Take advantage of these features to clean up memory
instruction definitions in Alpha.
arch/alpha/isa/decoder.isa:
arch/alpha/isa/mem.isa:
arch/alpha/isa/pal.isa:
Take advantage of new keyword parameters to disambiguate
instruction vs. memory-request flags, and to provide
a default EA calculation for memory ops (since 99% of them
are the same).
arch/isa_parser.py:
Add two new features to instruction format functions:
- Keyword parameters, a la Python.
- List-valued arguments.
Also export makeList() function to Python code blocks,
as this is handy for dealing with flags.
--HG--
extra : convert_revision : 99bbbaa2e765230aa96b6a06ed193793325f9fb0
Should enable implementation of split-phase timing loads
with new memory model.
May create slight timing differences under FullCPU, as I
believe we were not handling software prefetches correctly
before when the split MemAcc/Exec model was used. I haven't
looked into this in any detail though.
arch/alpha/isa/decoder.isa:
HwLoadStore format split into separate HwLoad and
HwStore formats.
Copy instructions now fall under MiscPrefetch format.
Mem_write_result is now just write_result in store
conditionals.
arch/alpha/isa/mem.isa:
Split MemAccExecute and LoadStoreExecute templates
into separate templates for loads and stores; now
that memory operands are handled differently from
registers, it's impossible to have a single template
serve both.
Also unified the handling of "regular" prefetches
(loads to r31) and "misc" prefetches (e.g., wh64)
under the new scheme. It looks like SW prefetches
were not handled correctly in FullCPU up til now,
since we generated an execute() method for the outer
instruction but didn't generate a proper method for
MemAcc::execute() (instead getting a default no-op
method for that).
arch/alpha/isa/pal.isa:
Split HwLoadStore into separate HwLoad and HwStore
formats to select proper template (see change to
mem.isa in this changeset).
arch/isa_parser.py:
Stop trying to treat memory operands like register
operands, since we never used them in a uniform way
anyway, and it made it impossible to do split-phase
loads as needed for the new CPU model. Now there's no
more 'op_mem_rd', 'op_nonmem_rd', etc.: 'op_rd' just does
register operands, and the template code is responsible
for formulating the call to the memory system. Right now
the only thing exported by InstObjParams is a new attribute
'mem_acc_size' which gives the memory access size in bits,
though more attributes can be added if needed.
Also moved code in findOperands() method to
OperandDescriptorList.__init__(), which is where it belongs.
--HG--
extra : convert_revision : 6d53d07e0c5e828455834ded4395fa40f9146a34
(thanks to Gabe's include feature!).
arch/alpha/isa/main.isa:
Split out into multiple .isa files.
--HG--
extra : convert_revision : 30d8edf74ea194d4a208febf1e66edc72a7dbd5d
