From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-it0-x22a.google.com (mail-it0-x22a.google.com [IPv6:2607:f8b0:4001:c0b::22a]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by ml01.01.org (Postfix) with ESMTPS id B64DC1A1E02 for ; Fri, 26 Aug 2016 10:14:11 -0700 (PDT) Received: by mail-it0-x22a.google.com with SMTP id g62so1849813ith.1 for ; Fri, 26 Aug 2016 10:14:11 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linaro.org; s=google; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc; bh=rlHY/umU4cswmI/wcYXrYZoiJ2nsmuN33dNqSwJ99gY=; b=TMk94PmUHS1xwOf2upJnPae2bELIFDm4ps9Fnzl1daOOaa6l7UvX8M8i9/i/0cf5VW TwWuDjnSIu4l4IpEq13ZtqO3oa+5ImvZowkPHhX7FjVWZiuHrV2Ft0mgq8t5Ue0/gCcO C2bWBgo9VbZGRjBeWr56miq4v6rdH+ZneO2Zg= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc; bh=rlHY/umU4cswmI/wcYXrYZoiJ2nsmuN33dNqSwJ99gY=; b=avIzDlMJZ4uuTNn647i4kNaF6GQqUZbe9CSlpfERYG4hN/3ayo4dfOazw7UGfBJjMs d7b3WMIUuzgwczuKdNLW097TKL4ONMyVDeoDClSpKNLs72M8sBKn00PR0MISaxspyVOk EbEV2GM60C9mCtG4i8N4F32h3S9KpxLmOYwGqdMRrdiGYGTJuqSejE9ZU+Ivow2w+Lmn zHnUDm2HGF9mHlesBXwrCnPTywtaIuz/uu7RF0qmdkxpFPqba/+yfJToP9ZktVdFdgf7 /t2JYgQFZHErVr+SFXyXK+Dy1J+Drbt3wsQnSdWRbOafYQR6d7bSKzfOiy4PJUfzs/EG UDFA== X-Gm-Message-State: AEkoousOPiakz7BlNItJJxqMOakzcAHMCafTwuBrvao4duGFhvnI3JAVLnJtHtLRWwGP9QkTkd+Kln9eNVM9jkLe X-Received: by 10.36.57.215 with SMTP id l206mr14092141ita.5.1472231650878; Fri, 26 Aug 2016 10:14:10 -0700 (PDT) MIME-Version: 1.0 Received: by 10.36.204.195 with HTTP; Fri, 26 Aug 2016 10:14:10 -0700 (PDT) In-Reply-To: <20160826125439.GF4715@bivouac.eciton.net> References: <1471445945-19239-1-git-send-email-ard.biesheuvel@linaro.org> <1471445945-19239-4-git-send-email-ard.biesheuvel@linaro.org> <20160826125439.GF4715@bivouac.eciton.net> From: Ard Biesheuvel Date: Fri, 26 Aug 2016 18:14:10 +0100 Message-ID: To: Leif Lindholm Cc: edk2-devel-01 , Jeff Brasen , "Tian, Feng" , "Zeng, Star" , Daniil Egranov Subject: Re: [PATCH v3 3/4] MdeModulePkg/EbxDxe AARCH64: use tail call for EBC to native thunk X-BeenThere: edk2-devel@lists.01.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: EDK II Development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 26 Aug 2016 17:14:12 -0000 Content-Type: text/plain; charset=UTF-8 On 26 August 2016 at 13:54, Leif Lindholm wrote: > On Wed, Aug 17, 2016 at 04:59:04PM +0200, Ard Biesheuvel wrote: >> Instead of pessimistically copying at least 64 bytes from the VM stack >> to the native stack, and popping off the register arguments again >> before doing the native call, try to avoid touching the stack completely >> if the VM stack frame is < 64 bytes. Also, if the stack frame does exceed > > Should this say "does not"? > No, if it exceeds 64 bytes, we may have arguments that should be passed on the stack (but we're not sure, unfortunately, since the VM stack frame is not annotated, i.e., it could simply be local vars in the caller) In this case, there is no need to push all arguments onto the native stack, and pop off the first 8 into registers again, which is what the original code was doing. >> 64 bytes, there is no need to copy the first 64 bytes, since we are passing >> those in registers anyway. >> >> Contributed-under: TianoCore Contribution Agreement 1.0 >> Signed-off-by: Ard Biesheuvel >> --- >> MdeModulePkg/Universal/EbcDxe/AArch64/EbcLowLevel.S | 73 +++++++++++++++----- >> 1 file changed, 55 insertions(+), 18 deletions(-) >> >> diff --git a/MdeModulePkg/Universal/EbcDxe/AArch64/EbcLowLevel.S b/MdeModulePkg/Universal/EbcDxe/AArch64/EbcLowLevel.S >> index cb7a70b5a4f8..d95713e82b0f 100644 >> --- a/MdeModulePkg/Universal/EbcDxe/AArch64/EbcLowLevel.S >> +++ b/MdeModulePkg/Universal/EbcDxe/AArch64/EbcLowLevel.S >> @@ -35,30 +35,67 @@ ASM_GLOBAL ASM_PFX(mEbcInstructionBufferTemplate) >> //**************************************************************************** >> // UINTN EbcLLCALLEXNative(UINTN FuncAddr, UINTN NewStackPointer, VOID *FramePtr) >> ASM_PFX(EbcLLCALLEXNative): >> - stp x19, x20, [sp, #-16]! >> - stp x29, x30, [sp, #-16]! >> + mov x8, x0 // Preserve x0 >> + mov x9, x1 // Preserve x1 >> >> - mov x19, x0 >> - mov x20, sp >> - sub x2, x2, x1 // Length = NewStackPointer-FramePtr >> - sub sp, sp, x2 >> - sub sp, sp, #64 // Make sure there is room for at least 8 args in the new stack >> - mov x0, sp >> + // >> + // If the EBC stack frame is smaller than or equal to 64 bytes, we know there >> + // are no stacked arguments #9 and beyond that we need to copy to the native >> + // stack. In this case, we can perform a tail call which is much more >> + // efficient, since there is no need to touch the native stack at all. >> + // >> + sub x3, x2, x1 // Length = NewStackPointer - FramePtr >> + cmp x3, #64 >> + b.gt 1f >> >> - bl CopyMem // Sp, NewStackPointer, Length >> + adr x0, 0f >> + sub x0, x0, x3, lsr #1 >> + br x0 >> >> - ldp x0, x1, [sp], #16 >> - ldp x2, x3, [sp], #16 >> - ldp x4, x5, [sp], #16 >> - ldp x6, x7, [sp], #16 >> + ldr x7, [x9, #56] >> + ldr x6, [x9, #48] >> + ldr x5, [x9, #40] >> + ldr x4, [x9, #32] >> + ldr x3, [x9, #24] >> + ldr x2, [x9, #16] >> + ldr x1, [x9, #8] >> + ldr x0, [x9] > > Why not keep using ldp, but with x9? > This is a computed jump, depending on the size of the stack frame, hence the inverted order, and the LDRs. While probably harmless in most cases, it is arguably incorrect to copy random data from memory into these registers (and could even be a security issue) >> >> - blr x19 >> +0: br x8 >> >> - mov sp, x20 >> - ldp x29, x30, [sp], #16 >> - ldp x19, x20, [sp], #16 >> + // >> + // More than 64 bytes: we need to build the full native stack frame and copy >> + // the part of the VM stack exceeding 64 bytes (which may contain stacked >> + // arguments) to the native stack >> + // >> +1: stp x29, x30, [sp, #-16]! >> + mov x29, sp >> >> - ret >> + // >> + // Ensure that the stack pointer remains 16 byte aligned, >> + // even if the size of the VM stack frame is not a multiple of 16 >> + // >> + add x1, x1, #64 // Skip over [potential] reg params >> + tbz x3, #3, 2f // Multiple of 16? >> + ldr x4, [x2, #-8]! // No? Then push one word >> + str x4, [sp, #-16]! // ... but use two slots >> + b 3f >> + >> +2: ldp x4, x5, [x2, #-16]! >> + stp x4, x5, [sp, #-16]! >> +3: cmp x2, x1 >> + b.gt 2b >> + >> + ldp x0, x1, [x9] >> + ldp x2, x3, [x9, #16] >> + ldp x4, x5, [x9, #32] >> + ldp x6, x7, [x9, #48] >> + >> + blr x8 >> + >> + mov sp, x29 >> + ldp x29, x30, [sp], #16 >> + ret >> >> //**************************************************************************** >> // EbcLLEbcInterpret >> -- >> 2.7.4 >>