From: Uros Bizjak Date: Fri, 8 May 2020 09:22:47 +0000 (+0200) Subject: x86/cpu: Use INVPCID mnemonic in invpcid.h X-Git-Tag: v5.10.7~2522^2~1 X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=7e32a9dac9926241d56851e1517c9391d39fb48e;p=platform%2Fkernel%2Flinux-rpi.git x86/cpu: Use INVPCID mnemonic in invpcid.h The current minimum required version of binutils is 2.23, which supports the INVPCID instruction mnemonic. Replace the byte-wise specification of INVPCID with the proper mnemonic. [ bp: Add symbolic operand names for increased readability and flip their order like the insn expects them for the AT&T syntax. ] Signed-off-by: Uros Bizjak Signed-off-by: Borislav Petkov Reviewed-by: H. Peter Anvin (Intel) Acked-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/20200508092247.132147-1-ubizjak@gmail.com Signed-off-by: Borislav Petkov --- diff --git a/arch/x86/include/asm/invpcid.h b/arch/x86/include/asm/invpcid.h index 989cfa8..734482a 100644 --- a/arch/x86/include/asm/invpcid.h +++ b/arch/x86/include/asm/invpcid.h @@ -12,12 +12,9 @@ static inline void __invpcid(unsigned long pcid, unsigned long addr, * stale TLB entries and, especially if we're flushing global * mappings, we don't want the compiler to reorder any subsequent * memory accesses before the TLB flush. - * - * The hex opcode is invpcid (%ecx), %eax in 32-bit mode and - * invpcid (%rcx), %rax in long mode. */ - asm volatile (".byte 0x66, 0x0f, 0x38, 0x82, 0x01" - : : "m" (desc), "a" (type), "c" (&desc) : "memory"); + asm volatile("invpcid %[desc], %[type]" + :: [desc] "m" (desc), [type] "r" (type) : "memory"); } #define INVPCID_TYPE_INDIV_ADDR 0