| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
There are a couple of reasons to split the temporaries off from the
arguments and locals:
Temporaries are invisible, and changes to them cannot be observed.
On the other hand, arguments and locals are visible, and writes to them
can be seen from other places (nested functions), or by using the
arguments array. So, in practice these correspond to memory locations.
(One could argue that if neither nested functions, nor eval(), nor
arguments[] is used, the loads/stores are invisible too. But that's an
optimization, and changing locals/arguments to temporaries can be done
in a separate pass.)
Because of the "volatile" nature of arguments and locals, their usage
cannot be optimized. All optimizations (SSA construction, register
allocation, copy elimination, etc.) work on temporaries. Being able to
easily ignore all non-temporaries has the benefit that optimizations can
be faster.
Previously, Temps were not uniquely numbered: argument 1, local 1, and
temporary 1 all had the same number and were distinguishable by their
type. So, for any mapping from Temp to something else, a QHash was used.
Now that Temps only hold proper temporaries, the indexes do uniquely
identify them. Add to that the fact that after transforming to SSA form
all temporaries are renumbered starting from 0 and without any holes in
the numbering, many of those datastructures can be changed to simple
vectors. That change gives a noticeable performance improvement.
One implication of this change is that a number of functions that took
a Temp as their argument, now need to take Temp-or-ArgLocal, so Expr.
However, it turns out that there are very few places where that applies,
as many of those places also need to take constants or names. However,
explicitly separating memory loads/stores for arguments/locals from
temporaries adds the benefit that it's now easier to do a peep-hole
optimizer for those load/store operations in the future: when a load is
directly preceded by a store, it can be eliminated if the value is
still available in a temporary.
Change-Id: I4114006b076795d9ea9fe3649cdb3b9d7b7508f0
Reviewed-by: Simon Hausmann <[email protected]>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Do not generate constant shifts of 0. We do not use the flags, so it's a
move. On ARM it's actually important not to do this, because lsr/asr
with imm=0 is a special case (shift of 32 bits).
When in the area, also skip generating an and of the second operand with
0x1f. For Intel this is done on the CPU, and for ARM the JSC assembler
will generate it for us.
This patch also updates the ARM disassembler to print the right
immediate values for the shifts.
Change-Id: I7c92c8d899352712c84e5534c48392d75466be0e
Reviewed-by: Simon Hausmann <[email protected]>
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
When doing of integers where we use a three argument variant of masm
(lhs/rhs/target), we need three general purpose registers. If the target temp
of the binop is a register, we use that as a target, otherwise fall back to
ReturnValueRegister (scratch). In that case we don't need to move from RVR to
target register. Additionally we need to load lhs and rhs into registers, and
for the lhs we use the target register and for the rhs the scratch register.
So we start by loading the lhs into the target register and the rhs into
the scratch register. However if the rhs is already assigned to a register
and that register happens to be the target register, then the earlier load
of the lhs into the target register overwrote our rhs!
This is fixed by being more careful in the choice of the target temp's assigned
register as "scratch" register for the lhs, i.e. don't use it if the target
temp is also assigned to the same register as the rhs.
Task-number: QTBUG-38097
Change-Id: I2ffec55cb98818fa9ebb5a76a32b6dca72175893
Reviewed-by: Lars Knoll <[email protected]>
|
| |
|
|
|
| |
Change-Id: I6185b59a7dfd6977ce82581ab4385e07d78f13f6
Reviewed-by: Simon Hausmann <[email protected]>
|
| |
|
|
|
|
|
|
|
| |
Move all our runtime methods into the QV4::Runtime
struct and give them nicer names without underscores.
Sort them logically and remove a few unused methods.
Change-Id: Ib69b71764ff194d0ba211aac581f9a99734d8180
Reviewed-by: Simon Hausmann <[email protected]>
|
| |
|
|
|
| |
Change-Id: I8f96b8d570dd4c0139b0a2e595055b3b2c6dae70
Reviewed-by: Simon Hausmann <[email protected]>
|
| |
|
|
|
|
|
|
| |
QQmlJS::MASM -> QV4::JIT
QQmlJS::V4IR -> QV4::IR
Change-Id: I707e8990459114a699c200fe3c22cec3c8df1afc
Reviewed-by: Simon Hausmann <[email protected]>
|
|
|
Move the binop handling into qv4binop* to clean up the
code and ease maintenance.
Change-Id: I0053380be7f326a2100302a63e921698a5b28c2a
Reviewed-by: Simon Hausmann <[email protected]>
|
