Re: [PATCH 1/4] x86: entry.S: tidy up several suboptimal insns

[Borislav Petkov <bp@alien8.de>]

On Tue, Feb 24, 2015 at 07:51:32PM +0100, Denys Vlasenko wrote:
> In all three 32-bit entry points, %eax is zero-extended to %rax.
> It is safe to do 32-bit compare when checking that syscall#
> is not too large.
> 
> The last instance of "mysterious" SS+8 constant is replaced by SIZEOF_PTREGS.
> 
> The $AUDIT_ARCH_X86_64 parameter to syscall_trace_enter_phase1/2
> is a 32-bit constant, loading it with 64-bit MOV produces 10-byte insn
> instead of 5-byte one.
> 
> After TEST insn, JE anctually means "jump of zero",
> let's use JZ mnemonic instead.

Actually, JE == LZ as that's the same opcode for testing ZF=1.

And I have to object:

        testl $3,CS(%rsp)
	je retint_kernel

is much more understandable than

        testl $3,CS(%rsp)
	jz retint_kernel

It basically says are $3 and CS(%rsp) equal.

JZ, on the other hand, not so clear: the TEST ANDed the two operands and
set flags accordingly, so JZ is testing the ZF. This means, you actually
know what TEST does and you haven't forgotten.

-- 
Regards/Gruss,
    Boris.

ECO tip #101: Trim your mails when you reply.
--