59bec00ace
Introduce x86_64 %rip-relative addressing to the PER_CPU_VAR() macro. Instructions using %rip-relative address operand are one byte shorter than their absolute address counterparts and are also compatible with position independent executable (-fpie) builds. The patch reduces code size of a test kernel build by 150 bytes. The PER_CPU_VAR() macro is intended to be applied to a symbol and should not be used with register operands. Introduce the new __percpu macro and use it in cmpxchg{8,16}b_emu.S instead. Also add a missing function comment to this_cpu_cmpxchg8b_emu(). No functional changes intended. Signed-off-by: Uros Bizjak <ubizjak@gmail.com> Signed-off-by: Ingo Molnar <mingo@kernel.org> Cc: linux-kernel@vger.kernel.org Cc: Brian Gerst <brgerst@gmail.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Sean Christopherson <seanjc@google.com> |
||
---|---|---|
.. | ||
alpha | ||
arc | ||
arm | ||
arm64 | ||
csky | ||
hexagon | ||
ia64 | ||
loongarch | ||
m68k | ||
microblaze | ||
mips | ||
nios2 | ||
openrisc | ||
parisc | ||
powerpc | ||
riscv | ||
s390 | ||
sh | ||
sparc | ||
um | ||
x86 | ||
xtensa | ||
.gitignore | ||
Kconfig |