| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
none of these are actual textrels because of ld-time binding performed
by -Bsymbolic-functions, but I'm changing them with the goal of making
ld-time binding purely an optimization rather than relying on it for
semantic purposes.
in the case of memmove's call to memcpy, making it explicit that the
memmove asm is assuming the forward-copying behavior of the memcpy asm
is desirable anyway; in case memcpy is ever changed, the semantic
mismatch would be apparent while editing memmcpy.s.
|
|
on most cpu models, "rep stosl" has high overhead that makes it
undesirable for small memset sizes. the new code extends the
minimal-branch fast path for short memsets from size 15 up to size 62,
and shrink-wraps this code path. in addition, "rep stosl" is very
sensitive to misalignment. the cost varies with size and with cpu
model, but it has been observed performing 1.5 to 4 times slower when
the destination address is not aligned mod 16. the new code thus
ensures alignment mod 16, but also preserves any existing additional
alignment, in case there are cpu models where it is beneficial.
this version is based in part on changes to the x86_64 memset asm
proposed by Denys Vlasenko.
|
|
the concept of both versions is the same; they differ only in details.
for long runs, they use "rep movsl" or "rep movsq", and for small
runs, they use a trick, writing from both ends towards the middle,
that reduces the number of branches needed. in addition, if memset is
called multiple times with the same length, all branches will be
predicted; there are no loops.
for larger runs, there are likely faster approaches than "rep", at
least on some cpu models. for 32-bit, it's unlikely that there is any
faster approach that does not require non-baseline instructions; doing
anything fancier would require inspecting cpu capabilities. for
64-bit, there may very well be faster versions that work on all
models; further optimization could be explored in the future.
with these changes, memset is anywhere between 50% faster and 6 times
faster, depending on the cpu model and the length and alignment of the
destination buffer.
|
|
for the sake of simplicity, I've only used rep movsb rather than
breaking up the copy for using rep movsd/q. on all modern cpus, this
seems to be fine, but if there are performance problems, there might
be a need to go back and add support for rep movsd/q.
|
|
|
