| Commit message (Collapse) | Author | Age | Files | Lines |
| | |
|
| |
|
|
| |
restructured (nw)
|
| | |
|
| | |
|
| |
|
|
| |
Galibert]
|
| | |
|
| | |
|
| | |
|
| | |
|
| |\ |
|
| | | |
|
| |/ |
|
| | |
|
| |
|
|
|
|
| |
Don't remove and reinsert nodes that correspond to a duplicate memory
access, because tree rebalancing kills performances. Update the node's
element content instead.
|
| |
|
|
|
|
| |
The curpc can be in the middle of an instruction being decoded, so
it is not pertinent to recompute the disassembly starting from
this address.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Fixed an annoying inconsistency between memory_share and memory_region:
the width() method of the former returned the width in bits (8, 16, 32 or 64)
while the width() method of the latter returned the width in bytes
(1, 2, 4 or 8). Now both classes have a bitwidth() method and a bytewidth()
method. Updated all callers to use whichever one was more appropriate.
Removed the implicit-cast-to-any-integer-pointer ability of memory_regions,
which was rather unsafe (if you weren't careful with your * operators and
casts it was easy to accidentally get a pointer to the memory_region object
itself instead of to the data, with no warning from the compiler... or at
least I kept doing it) Updated all devices and drivers that were accessing
regions that way to use a region_ptr_finder when possible, and otherwise to
call base() explicitly.
|
| | |
|
| |
|
|
| |
warnings of clang 3.5 when compiling MESS (nw)
|
| | |
|
| | |
|
| |
|
|
| |
watchpoints continue to work as expected [Alex Jackson]
|
| |
|
|
|
|
| |
left same as before [Miodrag Milanovic]
-Created osd_options as base option class for non-system specific options
|
| | |
|
| |
|
|
|
| |
Some minor enhancements to dynamic_array, including clearing to specific values
and expanding and clearing newly allocated values.
|
| |
|
|
| |
to new license tagged form.
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
| |
It now shows breakpoints for all CPUs and lets you sort by
each field.
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
pcatmem(p|d|i). [Andrew Gardner]
Fixes left-click selection bug in the memory window. [Andrew Gardner]
Explanation:
------------
Call trackmem to start tracking which PC writes to which address in memory and
pcatmem(p|d|i) to query a memory region for which PC wrote to it. Users of
the QT debugger can also right click on a memory address in the memory window
to make a popup message appear with the results - right-clicking also
automatically copies the resultant PC onto the clipboard. (I'll attach an
image of this behavior in a follow-up mail).
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
program counter has visited in the dasm windows.
Run "help trackpc" in the debugger to see the options. [Andrew Gardner]
Out of whatsnew:
This isn't enabled by default because of how sloooow it is to disassemble each
opcode when you want to compute its crc32. That can be sped up with lookup
tables and the like.
There's a good chance I should pull the 'clear tracks' argument into its own
command, but it functions as-is.
This can be added to the windows debugger with a simple change to the
osd display code.
|
| |
|
|
|
| |
[Andrew Gardner]
|
| | |
|
| |
|
|
|
|
|
| |
Fixed "ignore" command reporting incorrect invalid CPUs.
Fixed crazy code responsible for opcodes' crc32s in the comments system
(what was I thinking back then?). [Andrew Gardner]
|
| |
|
|
|
|
| |
- Added support for registerpoints.
- Added 'exit' as a synonym for 'quit'.
|
| |
|
|
|
|
| |
amount of CPU cycles taken by the last instruction. This is useful to people
(me) who are rewriting CPU cores, specifically with regression testing.
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
| |
device_memory_interface::space() assert against NULL and
return a reference, and pushed references throughout all
address space usage in the system. Added a has_space()
method to check for those rare case when it is ambiguous.
[Aaron Giles]
Also reinstated the generic space and added fatal error
handlers if anyone tries to actually read/write from it.
|
| |
|
|
|
|
|
| |
read/write handlers to take an address_space & instead
of an address_space *. Also update pretty much all other
functions to take a reference where appropriate.
[Aaron Giles]
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
and set_state() to set_state_int() for consistency.
Update all callers. Also add set_pc() helper and updated
all callers to use that instead of set_state_int(STATE_GENPC)
[Aaron Giles]
Added device_t::state() method to get the state interface.
Added redundant device_state_interface::state() method to
catch redundant use of it. [Aaron Giles]
Removed cpu_get_reg() and cpu_set_reg() macros in favor of
using the above methods. [Aaron Giles]
|
| | |
|
| |
|
|
| |
termination (no whatsnew)
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
memory_region management into the memory manager instead
of directly in the machine. Hid the global region method;
now all regions must be looked up relative to a device.
If you're a member function, you can just use memregion("tag")
directly. If you're a global function or a device referencing
global regions, use machine().root_device().memregion("tag")
to look up regions relative to the root.
S&R to convert all references:
machine([()]*)\.region
machine\1\.root_device\(\).subregion
Then remove redundant machine().root_device() within src/mame:
([ \t])machine\(\)\.root_device\(\)\.
\1
And use state->memregion() if we have a state variable present:
(state *= *[^;]+driver_data[^}]+)([^ \t]*)machine[()]*\.root_device\(\)\.
\1state->
Finally some cleanup:
screen.state->
state->
device->state->
state->
space->state->
state->
And a few hand-tweaks.
|
| |
|
|
| |
device search.
|
| | |
|