Kernel: Add fast-path for sys$gettid()

The userspace locks are very aggressively calling sys$gettid() to find
out which thread ID they have.

Since syscalls are quite heavy, this can get very expensive for some
programs. This patch adds a fast-path for sys$gettid(), which makes it
skip all of the usual syscall validation and just return the thread ID
right away.

This cuts Kernel/Process.cpp compile time by ~18%, from ~29 to ~24 sec.

1 files changed, 8 insertions, 0 deletions

diff --git a/Kernel/Syscall.cpp b/Kernel/Syscall.cpp
index a0e73b9620..a0d297e96e 100644
--- a/Kernel/Syscall.cpp
+++ b/Kernel/Syscall.cpp
@@ -119,6 +119,14 @@ int handle(RegisterDump& regs, u32 function, u32 arg1, u32 arg2, u32 arg3)
 
 void syscall_handler(RegisterDump regs)
 {
+    // Special handling of the "gettid" syscall since it's extremely hot.
+    // FIXME: Remove this hack once userspace locks stop calling it so damn much.
+    if (regs.eax == SC_gettid) {
+        regs.eax = current->process().sys$gettid();
+        current->did_syscall();
+        return;
+    }
+
     // Make sure SMAP protection is enabled on syscall entry.
     clac();