diff options
| author | Santi Galan <santi.galan@metempsy.com> | 2017-04-18 11:08:46 -0500 |
|---|---|---|
| committer | Jason Lowe-Power <jason@lowepower.com> | 2017-04-18 11:12:35 -0500 |
| commit | 3cc6df46a3cd0aa97cd960f7f284d21c36a05181 (patch) | |
| tree | 268f087ae4a55dad26c711bede3b9e1d9c381093 /src | |
| parent | efe3bfcd0c6e1161bea621a6e512378fe228fed4 (diff) | |
| download | gem5-3cc6df46a3cd0aa97cd960f7f284d21c36a05181.tar.xz | |
x86: fixed branching() computation for branch uops
When a branch micro-op belongs to a flow and the micro-op does not change
the nPC and just updates the nuPC (like a 'rep movs' flow), branching()
function always returns not-taken no matter actual micro-branch outcome.
Provided fix adds to the equation nuPC attribute checking since these kind
of branch micro-op only updates that pointer.
This issue has been found while debugging the performance of a copy-loop
implemented with memcopy function. Without the fix, 'rep movss' internal
micro-branch was always predicted as not-taken causing an squash event
after every branch micro-branch execution.
Using the provided test, branch mispredition went from 1922 without the fix
to 7.
Change-Id: I1bcbefae26aef47e3135817ef99b53d0ea0a98fa
Diffstat (limited to 'src')
| -rw-r--r-- | src/arch/x86/types.hh | 3 |
1 files changed, 2 insertions, 1 deletions
diff --git a/src/arch/x86/types.hh b/src/arch/x86/types.hh index 954f9f16e..8d47b7efb 100644 --- a/src/arch/x86/types.hh +++ b/src/arch/x86/types.hh @@ -318,7 +318,8 @@ namespace X86ISA bool branching() const { - return this->npc() != this->pc() + size(); + return (this->npc() != this->pc() + size()) || + (this->nupc() != this->upc() + 1); } void |