Message ID | 009501d8a1be$b6199e20$224cda60$@nextmovesoftware.com |
---|---|
State | New, archived |
Headers |
Return-Path: <gcc-patches-bounces+ouuuleilei=gmail.com@gcc.gnu.org> Delivered-To: ouuuleilei@gmail.com Received: by 2002:a05:6a10:b5d6:b0:2b9:3548:2db5 with SMTP id v22csp415245pxt; Wed, 27 Jul 2022 06:43:46 -0700 (PDT) X-Google-Smtp-Source: AGRyM1v5XuD+gBxzKDhND1r4Z+nE1HZ6CGdDB5SYuR5t3KFVRPbJ25SOJnQXZHuOWWlXQzVKEQyX X-Received: by 2002:a05:6402:f22:b0:43b:c5e6:f60 with SMTP id i34-20020a0564020f2200b0043bc5e60f60mr23260397eda.136.1658929426264; Wed, 27 Jul 2022 06:43:46 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1658929426; cv=none; d=google.com; s=arc-20160816; b=IjS9uHVNXDE8jSSc6CWkCMxkS1Qle7mzwy6wAeG55Sk3vr6Ev97V8d+Dvl9buxZ1Gt uJF8IPRfaajurzeisCd08ziiAPhsAF82R8Zo0BG8UULOl08GDgHHwmdNdX1WAI8HgvDF k2Qpj7o4XrMehtALgASyh7HWjio7W+rR/LlvJCTw40I74vlZtOm8/pAcF3McI+k0TT4V LzWZ8xq7M8F0to+JRW9R9/p15c0hqdp8nUgW3FeVr6I9Bnt9Z89Cx2mFvQ7glZ4dj2bu njLUWQo9p53Fmuf8bcolBXGH0iEcE55hKL/7Tl0Z6IUMU0pRU5m+Zw1XeqDCn0J0+nwe JvJQ== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=sender:errors-to:cc:list-subscribe:list-help:list-post:list-archive :list-unsubscribe:list-id:precedence:content-language:thread-index :mime-version:message-id:date:subject:to:from:dkim-signature :dmarc-filter:delivered-to; bh=4OmK59TmtSwy4ytNbNZVSGCF+sl0tkJM1bolq/K+s0I=; b=N+W/CpjwGbYLyROUNvZlFInyQNVhQuNxPUUlOpxisjDZ4OJUGVGkjbDrC9SAFuSlFQ YxHRCl4RoTlfcwXJd4NBUuuNZhIfJp2obQtCC5dHMMBXvFFVo6B+iBeCm/543/zH3RM3 FuDdmVWTokvzN699r0FlyBuLevHt/AOPD8Va7HMhENknibcEJqOWY1oB0OEsIPQ2x7nI OCnU2RBpvWDyYO/ZIoV8kBQuS6T9wMrTkmaPzhI6twCIpGF2NsMW3jRaeLgQhJuXWCFA 7mS7xybKZN0ZbI8c4uGqKrEYhaibQZjarACKa4A7t4zHUoWrv+gKHKGIgQAQKPUe04w/ 42UA== ARC-Authentication-Results: i=1; mx.google.com; dkim=fail header.i=@nextmovesoftware.com header.s=default header.b="pFT0aq7/"; spf=pass (google.com: domain of gcc-patches-bounces+ouuuleilei=gmail.com@gcc.gnu.org designates 2620:52:3:1:0:246e:9693:128c as permitted sender) smtp.mailfrom="gcc-patches-bounces+ouuuleilei=gmail.com@gcc.gnu.org" Received: from sourceware.org (server2.sourceware.org. [2620:52:3:1:0:246e:9693:128c]) by mx.google.com with ESMTPS id hc16-20020a170907169000b0072b6ff08620si20370531ejc.1009.2022.07.27.06.43.45 for <ouuuleilei@gmail.com> (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Wed, 27 Jul 2022 06:43:46 -0700 (PDT) Received-SPF: pass (google.com: domain of gcc-patches-bounces+ouuuleilei=gmail.com@gcc.gnu.org designates 2620:52:3:1:0:246e:9693:128c as permitted sender) client-ip=2620:52:3:1:0:246e:9693:128c; Authentication-Results: mx.google.com; dkim=fail header.i=@nextmovesoftware.com header.s=default header.b="pFT0aq7/"; spf=pass (google.com: domain of gcc-patches-bounces+ouuuleilei=gmail.com@gcc.gnu.org designates 2620:52:3:1:0:246e:9693:128c as permitted sender) smtp.mailfrom="gcc-patches-bounces+ouuuleilei=gmail.com@gcc.gnu.org" Received: from server2.sourceware.org (localhost [IPv6:::1]) by sourceware.org (Postfix) with ESMTP id 123253856942 for <ouuuleilei@gmail.com>; Wed, 27 Jul 2022 13:42:54 +0000 (GMT) X-Original-To: gcc-patches@gcc.gnu.org Delivered-To: gcc-patches@gcc.gnu.org Received: from server.nextmovesoftware.com (server.nextmovesoftware.com [162.254.253.69]) by sourceware.org (Postfix) with ESMTPS id 925583856DF9 for <gcc-patches@gcc.gnu.org>; Wed, 27 Jul 2022 13:42:28 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.1 sourceware.org 925583856DF9 Authentication-Results: sourceware.org; dmarc=none (p=none dis=none) header.from=nextmovesoftware.com Authentication-Results: sourceware.org; spf=pass smtp.mailfrom=nextmovesoftware.com DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=nextmovesoftware.com; s=default; h=Content-Type:MIME-Version:Message-ID: Date:Subject:Cc:To:From:Sender:Reply-To:Content-Transfer-Encoding:Content-ID: Content-Description:Resent-Date:Resent-From:Resent-Sender:Resent-To:Resent-Cc :Resent-Message-ID:In-Reply-To:References:List-Id:List-Help:List-Unsubscribe: List-Subscribe:List-Post:List-Owner:List-Archive; bh=4OmK59TmtSwy4ytNbNZVSGCF+sl0tkJM1bolq/K+s0I=; b=pFT0aq7/uO4O/xYhZLn32pdhFY eyF5U9C4PPL4OKJt98tyurNTkZrsMK+hKpAAGl6bsrN85B2MOgySRn3tUIgYBc0xRpERXOHFlq3h1 iPBuwLrD8dHjd4I2r5qf3mw+CiTDndIZy4F9tic+PVywAKbn7251+re4F/vMvlylU0aCwoYasl1Df A6+4mBbxAEOVapYUQomA8mgvnpDl9sknISqct+ZDkoru5xJktQrVPaW4OK0vG0sVL6x/C+vEz81H5 QeXTThLemKdaYMCNRz0xDp2bKNjUU3hswndgrhFB1vMlNviabgp+10F65GT+ZaHXlEHxC/DZxc0qy zxoflaSQ==; Received: from host81-156-58-95.range81-156.btcentralplus.com ([81.156.58.95]:56901 helo=Dell) by server.nextmovesoftware.com with esmtpsa (TLS1.2) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.94.2) (envelope-from <roger@nextmovesoftware.com>) id 1oGhIx-00020r-NE; Wed, 27 Jul 2022 09:42:27 -0400 From: "Roger Sayle" <roger@nextmovesoftware.com> To: <gcc-patches@gcc.gnu.org> Subject: [PATCH] Some additional zero-extension related optimizations in simplify-rtx. Date: Wed, 27 Jul 2022 14:42:25 +0100 Message-ID: <009501d8a1be$b6199e20$224cda60$@nextmovesoftware.com> MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="----=_NextPart_000_0096_01D8A1C7.17E09E30" X-Mailer: Microsoft Outlook 16.0 Thread-Index: AdihuyxNCA0hzpRnTCiLNk5hVQZHeQ== Content-Language: en-gb X-AntiAbuse: This header was added to track abuse, please include it with any abuse report X-AntiAbuse: Primary Hostname - server.nextmovesoftware.com X-AntiAbuse: Original Domain - gcc.gnu.org X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12] X-AntiAbuse: Sender Address Domain - nextmovesoftware.com X-Get-Message-Sender-Via: server.nextmovesoftware.com: authenticated_id: roger@nextmovesoftware.com X-Authenticated-Sender: server.nextmovesoftware.com: roger@nextmovesoftware.com X-Source: X-Source-Args: X-Source-Dir: X-Spam-Status: No, score=-12.1 required=5.0 tests=BAYES_00, DKIM_SIGNED, DKIM_VALID, DKIM_VALID_AU, DKIM_VALID_EF, GIT_PATCH_0, SPF_HELO_NONE, SPF_PASS, TXREP autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on server2.sourceware.org X-BeenThere: gcc-patches@gcc.gnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Gcc-patches mailing list <gcc-patches.gcc.gnu.org> List-Unsubscribe: <https://gcc.gnu.org/mailman/options/gcc-patches>, <mailto:gcc-patches-request@gcc.gnu.org?subject=unsubscribe> List-Archive: <https://gcc.gnu.org/pipermail/gcc-patches/> List-Post: <mailto:gcc-patches@gcc.gnu.org> List-Help: <mailto:gcc-patches-request@gcc.gnu.org?subject=help> List-Subscribe: <https://gcc.gnu.org/mailman/listinfo/gcc-patches>, <mailto:gcc-patches-request@gcc.gnu.org?subject=subscribe> Cc: 'Segher Boessenkool' <segher@kernel.crashing.org> Errors-To: gcc-patches-bounces+ouuuleilei=gmail.com@gcc.gnu.org Sender: "Gcc-patches" <gcc-patches-bounces+ouuuleilei=gmail.com@gcc.gnu.org> X-getmail-retrieved-from-mailbox: =?utf-8?q?INBOX?= X-GMAIL-THRID: =?utf-8?q?1739513582239946658?= X-GMAIL-MSGID: =?utf-8?q?1739513582239946658?= |
Series |
Some additional zero-extension related optimizations in simplify-rtx.
|
|
Commit Message
Roger Sayle
July 27, 2022, 1:42 p.m. UTC
This patch implements some additional zero-extension and sign-extension related optimizations in simplify-rtx.cc. The original motivation comes from PR rtl-optimization/71775, where in comment #2 Andrew Pinski sees: Failed to match this instruction: (set (reg:DI 88 [ _1 ]) (sign_extend:DI (subreg:SI (ctz:DI (reg/v:DI 86 [ x ])) 0))) On many platforms the result of DImode CTZ is constrained to be a small unsigned integer (between 0 and 64), hence the truncation to 32-bits (using a SUBREG) and the following sign extension back to 64-bits are effectively a no-op, so the above should ideally (often) be simplified to "(set (reg:DI 88) (ctz:DI (reg/v:DI 86 [ x ]))". To implement this, and some closely related transformations, we build upon the existing val_signbit_known_clear_p predicate. In the first chunk, nonzero_bits knows that FFS and ABS can't leave the sign-bit bit set, so the simplification of of ABS (ABS (x)) and ABS (FFS (x)) can itself be simplified. The second transformation is that we can canonicalized SIGN_EXTEND to ZERO_EXTEND (as in the PR 71775 case above) when the operand's sign-bit is known to be clear. The final two chunks are for SIGN_EXTEND of a truncating SUBREG, and ZERO_EXTEND of a truncating SUBREG respectively. The nonzero_bits of a truncating SUBREG pessimistically thinks that the upper bits may have an arbitrary value (by taking the SUBREG), so we need look deeper at the SUBREG's operand to confirm that the high bits are known to be zero. Unfortunately, for PR rtl-optimization/71775, ctz:DI on x86_64 with default architecture options is undefined at zero, so we can't be sure the upper bits of reg:DI 88 will be sign extended (all zeros or all ones). nonzero_bits knows this, so the above transformations don't trigger, but the transformations themselves are perfectly valid for other operations such as FFS, POPCOUNT and PARITY, and on other targets/-march settings where CTZ is defined at zero. This patch has been tested on x86_64-pc-linux-gnu with make bootstrap and make -k check, both with and without --target_board=unix{-m32}, with no new failures. Testing with CSiBE shows these transformations trigger on several source files (and with -Os reduces the size of the code). Ok for mainline? 2022-07-27 Roger Sayle <roger@nextmovesoftware.com> gcc/ChangeLog * simplify_rtx.cc (simplify_unary_operation_1) <ABS>: Simplify test as both FFS and ABS result in nonzero_bits returning a mask that satisfies val_signbit_known_clear_p. <SIGN_EXTEND>: Canonicalize SIGN_EXTEND to ZERO_EXTEND when val_signbit_known_clear_p is true of the operand. Simplify sign extensions of SUBREG truncations of operands that are already suitably (zero) extended. <ZERO_EXTEND>: Simplify zero extensions of SUBREG truncations of operands that are already suitably zero extended. Thanks in advance, Roger --
Comments
Hi! On Wed, Jul 27, 2022 at 02:42:25PM +0100, Roger Sayle wrote: > This patch implements some additional zero-extension and sign-extension > related optimizations in simplify-rtx.cc. The original motivation comes > from PR rtl-optimization/71775, where in comment #2 Andrew Pinski sees: > > Failed to match this instruction: > (set (reg:DI 88 [ _1 ]) > (sign_extend:DI (subreg:SI (ctz:DI (reg/v:DI 86 [ x ])) 0))) > > On many platforms the result of DImode CTZ is constrained to be a > small unsigned integer (between 0 and 64), hence the truncation to > 32-bits (using a SUBREG) and the following sign extension back to > 64-bits are effectively a no-op, so the above should ideally (often) > be simplified to "(set (reg:DI 88) (ctz:DI (reg/v:DI 86 [ x ]))". And you can also do that if ctz is undefined for a zero argument! > To implement this, and some closely related transformations, we build > upon the existing val_signbit_known_clear_p predicate. In the first > chunk, nonzero_bits knows that FFS and ABS can't leave the sign-bit > bit set, Is that guaranteed in all cases? Also at -O0, also for args bigger than 64 bits? > Unfortunately, for PR rtl-optimization/71775, ctz:DI on x86_64 with > default architecture options is undefined at zero, so we can't be sure > the upper bits of reg:DI 88 will be sign extended (all zeros or all ones). > nonzero_bits knows this, so the above transformations don't trigger, > but the transformations themselves are perfectly valid for other > operations such as FFS, POPCOUNT and PARITY, and on other targets/-march > settings where CTZ is defined at zero. It is valid to do whatever you want if the result of CTZ or CLZ is undefined, so the sign_extend of c[lt]z is a nop. However if C[LT]Z_DEFINED_VALUE_AT_ZERO is non-zero you have to check if the returned value (the macro's second arg) survives sign-extending. > /* If operand is something known to be positive, ignore the ABS. */ > - if (GET_CODE (op) == FFS || GET_CODE (op) == ABS > - || val_signbit_known_clear_p (GET_MODE (op), > - nonzero_bits (op, GET_MODE (op)))) > + if (val_signbit_known_clear_p (GET_MODE (op), > + nonzero_bits (op, GET_MODE (op)))) > return op; I don't think val_signbit_known_clear_p is true in all cases, as I said above, but we do not care about generated code quality for these cases. OK. > + /* We can canonicalize SIGN_EXTEND (op) as ZERO_EXTEND (op) when > + we know the sign bit of OP must be clear. */ > + if (val_signbit_known_clear_p (GET_MODE (op), > + nonzero_bits (op, GET_MODE (op)))) > + return simplify_gen_unary (ZERO_EXTEND, mode, op, GET_MODE (op)); OK. > + /* (sign_extend:DI (subreg:SI (ctz:DI ...))) is (ctz:DI ...). */ > + if (GET_CODE (op) == SUBREG > + && subreg_lowpart_p (op) > + && GET_MODE (SUBREG_REG (op)) == mode > + && is_a <scalar_int_mode> (mode, &int_mode) > + && is_a <scalar_int_mode> (GET_MODE (op), &op_mode) > + && GET_MODE_PRECISION (int_mode) <= HOST_BITS_PER_WIDE_INT > + && GET_MODE_PRECISION (op_mode) < GET_MODE_PRECISION (int_mode) > + && (nonzero_bits (SUBREG_REG (op), mode) > + & ~(GET_MODE_MASK (op_mode)>>1)) == 0) (spaces around >> please) Please use val_signbit_known_{set,clear}_p? > + return SUBREG_REG (op); Also, this is not correct for C[LT]Z_DEFINED_VALUE_AT_ZERO non-zero if the value it returns in its second arg does not survive sign extending unmodified (if it is 0xffffffff for an extend from SI to DI for exxample). > + /* (zero_extend:DI (subreg:SI (ctz:DI ...))) is (ctz:DI ...). */ > + if (GET_CODE (op) == SUBREG > + && subreg_lowpart_p (op) > + && GET_MODE (SUBREG_REG (op)) == mode > + && is_a <scalar_int_mode> (mode, &int_mode) > + && is_a <scalar_int_mode> (GET_MODE (op), &op_mode) > + && GET_MODE_PRECISION (int_mode) <= HOST_BITS_PER_WIDE_INT > + && GET_MODE_PRECISION (op_mode) < GET_MODE_PRECISION (int_mode) > + && (nonzero_bits (SUBREG_REG (op), mode) > + & ~GET_MODE_MASK (op_mode)) == 0) > + return SUBREG_REG (op); This has that same problem. Segher
Hi Segher, > On Wed, Jul 27, 2022 at 02:42:25PM +0100, Roger Sayle wrote: > > This patch implements some additional zero-extension and > > sign-extension related optimizations in simplify-rtx.cc. The original > > motivation comes from PR rtl-optimization/71775, where in comment #2 > Andrew Pinski sees: > > > > Failed to match this instruction: > > (set (reg:DI 88 [ _1 ]) > > (sign_extend:DI (subreg:SI (ctz:DI (reg/v:DI 86 [ x ])) 0))) > > > > On many platforms the result of DImode CTZ is constrained to be a > > small unsigned integer (between 0 and 64), hence the truncation to > > 32-bits (using a SUBREG) and the following sign extension back to > > 64-bits are effectively a no-op, so the above should ideally (often) > > be simplified to "(set (reg:DI 88) (ctz:DI (reg/v:DI 86 [ x ]))". > > And you can also do that if ctz is undefined for a zero argument! Forgive my perhaps poor use of terminology. The case of ctz 0 on x64_64 isn't "undefined behaviour" (UB) in the C/C++ sense that would allow us to do anything, but implementation defined (which Intel calls "undefined" in their documentation). Hence, we don't know which DI value is placed in the result register. In this case, truncating to SI mode, then sign extending the result is not a no-op, as the top bits will/must now all be the same [though admittedly to an unknown undefined signbit]. Hence the above optimization would be invalid, as it doesn't guarantee the result would be sign-extended. > > To implement this, and some closely related transformations, we build > > upon the existing val_signbit_known_clear_p predicate. In the first > > chunk, nonzero_bits knows that FFS and ABS can't leave the sign-bit > > bit set, > > Is that guaranteed in all cases? Also at -O0, also for args bigger than > 64 bits? val_signbit_known_clear_p should work for any size/precision arg. I'm not sure if the results are affected by -O0, but even if they are, this will not affect correctness only whether these optimizations are performed, which is precisely what -O0 controls. > > + /* (sign_extend:DI (subreg:SI (ctz:DI ...))) is (ctz:DI ...). */ > > + if (GET_CODE (op) == SUBREG > > + && subreg_lowpart_p (op) > > + && GET_MODE (SUBREG_REG (op)) == mode > > + && is_a <scalar_int_mode> (mode, &int_mode) > > + && is_a <scalar_int_mode> (GET_MODE (op), &op_mode) > > + && GET_MODE_PRECISION (int_mode) <= HOST_BITS_PER_WIDE_INT > > + && GET_MODE_PRECISION (op_mode) < GET_MODE_PRECISION > (int_mode) > > + && (nonzero_bits (SUBREG_REG (op), mode) > > + & ~(GET_MODE_MASK (op_mode)>>1)) == 0) > > (spaces around >> please) Doh! Good catch, thanks. > Please use val_signbit_known_{set,clear}_p? Alas, it's not just the SI mode's signbit that we care about, but all of the bits above it in the DImode operand/result. These all need to be zero, for the operand to already be zero-extended/sign_extended. > > + return SUBREG_REG (op); > > Also, this is not correct for C[LT]Z_DEFINED_VALUE_AT_ZERO non-zero if the > value it returns in its second arg does not survive sign extending unmodified (if it > is 0xffffffff for an extend from SI to DI for example). Fortunately, C[LT]Z_DEFINED_VALUE_AT_ZERO being defined to return a negative result, such as -1 is already handled (accounted for) in nonzero_bits. The relevant code in rtlanal.cc's nonzero_bits1 is: case CTZ: /* If CTZ has a known value at zero, then the nonzero bits are that value, plus the number of bits in the mode minus one. */ if (CTZ_DEFINED_VALUE_AT_ZERO (mode, nonzero)) nonzero |= (HOST_WIDE_INT_1U << (floor_log2 (mode_width))) - 1; else nonzero = -1; break; Hence, any bits set by the constant returned by the target's DEFINED_VALUE_AT_ZERO will be set in the result of nonzero_bits. So if this is negative, say -1, then val_signbit_known_clear_p (or the more complex tests above) will return false. I'm currently bootstrapping and regression testing the whitespace change/correction suggested above. Thanks, Roger --
Hi Segher, > > > To implement this, and some closely related transformations, we > > > build upon the existing val_signbit_known_clear_p predicate. In the > > > first chunk, nonzero_bits knows that FFS and ABS can't leave the > > > sign-bit bit set, > > > > Is that guaranteed in all cases? Also at -O0, also for args bigger > > than 64 bits? > > val_signbit_known_clear_p should work for any size/precision arg. No, you're right! Please forgive/excuse me. Neither val_signbit_p nor nonzero_bits have yet been updated to use "wide_int", so don't work for TImode or wider modes. Doh! I'm shocked. Roger --
"Roger Sayle" <roger@nextmovesoftware.com> writes: > This patch implements some additional zero-extension and sign-extension > related optimizations in simplify-rtx.cc. The original motivation comes > from PR rtl-optimization/71775, where in comment #2 Andrew Pinski sees: > > Failed to match this instruction: > (set (reg:DI 88 [ _1 ]) > (sign_extend:DI (subreg:SI (ctz:DI (reg/v:DI 86 [ x ])) 0))) > > On many platforms the result of DImode CTZ is constrained to be a > small unsigned integer (between 0 and 64), hence the truncation to > 32-bits (using a SUBREG) and the following sign extension back to > 64-bits are effectively a no-op, so the above should ideally (often) > be simplified to "(set (reg:DI 88) (ctz:DI (reg/v:DI 86 [ x ]))". > > To implement this, and some closely related transformations, we build > upon the existing val_signbit_known_clear_p predicate. In the first > chunk, nonzero_bits knows that FFS and ABS can't leave the sign-bit > bit set, so the simplification of of ABS (ABS (x)) and ABS (FFS (x)) > can itself be simplified. I think I misunderstood, but just in case: RTL ABS is well-defined for the minimum integer (giving back the minimum integer), so we can't assume that ABS leaves the sign bit clear. Thanks, Richard > The second transformation is that we can > canonicalized SIGN_EXTEND to ZERO_EXTEND (as in the PR 71775 case above) > when the operand's sign-bit is known to be clear. The final two chunks > are for SIGN_EXTEND of a truncating SUBREG, and ZERO_EXTEND of a > truncating SUBREG respectively. The nonzero_bits of a truncating > SUBREG pessimistically thinks that the upper bits may have an > arbitrary value (by taking the SUBREG), so we need look deeper at the > SUBREG's operand to confirm that the high bits are known to be zero. > > Unfortunately, for PR rtl-optimization/71775, ctz:DI on x86_64 with > default architecture options is undefined at zero, so we can't be sure > the upper bits of reg:DI 88 will be sign extended (all zeros or all ones). > nonzero_bits knows this, so the above transformations don't trigger, > but the transformations themselves are perfectly valid for other > operations such as FFS, POPCOUNT and PARITY, and on other targets/-march > settings where CTZ is defined at zero. > > This patch has been tested on x86_64-pc-linux-gnu with make bootstrap > and make -k check, both with and without --target_board=unix{-m32}, > with no new failures. Testing with CSiBE shows these transformations > trigger on several source files (and with -Os reduces the size of the > code). Ok for mainline? > > > 2022-07-27 Roger Sayle <roger@nextmovesoftware.com> > > gcc/ChangeLog > * simplify_rtx.cc (simplify_unary_operation_1) <ABS>: Simplify > test as both FFS and ABS result in nonzero_bits returning a > mask that satisfies val_signbit_known_clear_p. > <SIGN_EXTEND>: Canonicalize SIGN_EXTEND to ZERO_EXTEND when > val_signbit_known_clear_p is true of the operand. > Simplify sign extensions of SUBREG truncations of operands > that are already suitably (zero) extended. > <ZERO_EXTEND>: Simplify zero extensions of SUBREG truncations > of operands that are already suitably zero extended. > > > Thanks in advance, > Roger > -- > > diff --git a/gcc/simplify-rtx.cc b/gcc/simplify-rtx.cc > index fa20665..e62bf56 100644 > --- a/gcc/simplify-rtx.cc > +++ b/gcc/simplify-rtx.cc > @@ -1366,9 +1366,8 @@ simplify_context::simplify_unary_operation_1 (rtx_code code, machine_mode mode, > break; > > /* If operand is something known to be positive, ignore the ABS. */ > - if (GET_CODE (op) == FFS || GET_CODE (op) == ABS > - || val_signbit_known_clear_p (GET_MODE (op), > - nonzero_bits (op, GET_MODE (op)))) > + if (val_signbit_known_clear_p (GET_MODE (op), > + nonzero_bits (op, GET_MODE (op)))) > return op; > > /* If operand is known to be only -1 or 0, convert ABS to NEG. */ > @@ -1615,6 +1614,24 @@ simplify_context::simplify_unary_operation_1 (rtx_code code, machine_mode mode, > } > } > > + /* We can canonicalize SIGN_EXTEND (op) as ZERO_EXTEND (op) when > + we know the sign bit of OP must be clear. */ > + if (val_signbit_known_clear_p (GET_MODE (op), > + nonzero_bits (op, GET_MODE (op)))) > + return simplify_gen_unary (ZERO_EXTEND, mode, op, GET_MODE (op)); > + > + /* (sign_extend:DI (subreg:SI (ctz:DI ...))) is (ctz:DI ...). */ > + if (GET_CODE (op) == SUBREG > + && subreg_lowpart_p (op) > + && GET_MODE (SUBREG_REG (op)) == mode > + && is_a <scalar_int_mode> (mode, &int_mode) > + && is_a <scalar_int_mode> (GET_MODE (op), &op_mode) > + && GET_MODE_PRECISION (int_mode) <= HOST_BITS_PER_WIDE_INT > + && GET_MODE_PRECISION (op_mode) < GET_MODE_PRECISION (int_mode) > + && (nonzero_bits (SUBREG_REG (op), mode) > + & ~(GET_MODE_MASK (op_mode)>>1)) == 0) > + return SUBREG_REG (op); > + > #if defined(POINTERS_EXTEND_UNSIGNED) > /* As we do not know which address space the pointer is referring to, > we can do this only if the target does not support different pointer > @@ -1765,6 +1782,18 @@ simplify_context::simplify_unary_operation_1 (rtx_code code, machine_mode mode, > op0_mode); > } > > + /* (zero_extend:DI (subreg:SI (ctz:DI ...))) is (ctz:DI ...). */ > + if (GET_CODE (op) == SUBREG > + && subreg_lowpart_p (op) > + && GET_MODE (SUBREG_REG (op)) == mode > + && is_a <scalar_int_mode> (mode, &int_mode) > + && is_a <scalar_int_mode> (GET_MODE (op), &op_mode) > + && GET_MODE_PRECISION (int_mode) <= HOST_BITS_PER_WIDE_INT > + && GET_MODE_PRECISION (op_mode) < GET_MODE_PRECISION (int_mode) > + && (nonzero_bits (SUBREG_REG (op), mode) > + & ~GET_MODE_MASK (op_mode)) == 0) > + return SUBREG_REG (op); > + > #if defined(POINTERS_EXTEND_UNSIGNED) > /* As we do not know which address space the pointer is referring to, > we can do this only if the target does not support different pointer
Hi! On Fri, Jul 29, 2022 at 07:57:51AM +0100, Roger Sayle wrote: > > On Wed, Jul 27, 2022 at 02:42:25PM +0100, Roger Sayle wrote: > > > This patch implements some additional zero-extension and > > > sign-extension related optimizations in simplify-rtx.cc. The original > > > motivation comes from PR rtl-optimization/71775, where in comment #2 > > Andrew Pinski sees: > > > > > > Failed to match this instruction: > > > (set (reg:DI 88 [ _1 ]) > > > (sign_extend:DI (subreg:SI (ctz:DI (reg/v:DI 86 [ x ])) 0))) > > > > > > On many platforms the result of DImode CTZ is constrained to be a > > > small unsigned integer (between 0 and 64), hence the truncation to > > > 32-bits (using a SUBREG) and the following sign extension back to > > > 64-bits are effectively a no-op, so the above should ideally (often) > > > be simplified to "(set (reg:DI 88) (ctz:DI (reg/v:DI 86 [ x ]))". > > > > And you can also do that if ctz is undefined for a zero argument! > > Forgive my perhaps poor use of terminology. The case of ctz 0 on > x64_64 isn't "undefined behaviour" (UB) in the C/C++ sense that > would allow us to do anything, but implementation defined (which > Intel calls "undefined" in their documentation). This is about CTZ in RTL, in GCC. CTZ_DEFINED_VALUE_AT_ZERO is 0 here, which means a zero argument gives an undefined result. > Hence, we don't > know which DI value is placed in the result register. In this case, > truncating to SI mode, then sign extending the result is not a no-op, > as the top bits will/must now all be the same [though admittedly to an > unknown undefined signbit]. And any value is valid. > Hence the above optimization would > be invalid, as it doesn't guarantee the result would be sign-extended. It does not have to be! Truncating an undefined DImode value to SIMode gives an undefined SImode value. On most architectures (including x86 afaik) you do not need to do any machine insn for that (the top 32 bits in the register are just ignored for a SImode value). > > Also, this is not correct for C[LT]Z_DEFINED_VALUE_AT_ZERO non-zero if the > > value it returns in its second arg does not survive sign extending > unmodified (if it > > is 0xffffffff for an extend from SI to DI for example). > > Fortunately, C[LT]Z_DEFINED_VALUE_AT_ZERO being defined to return a negative > result, such as -1 is already handled (accounted for) in nonzero_bits. The > relevant > code in rtlanal.cc's nonzero_bits1 is: A negative result, yes. But that was not my example. Segher
diff --git a/gcc/simplify-rtx.cc b/gcc/simplify-rtx.cc index fa20665..e62bf56 100644 --- a/gcc/simplify-rtx.cc +++ b/gcc/simplify-rtx.cc @@ -1366,9 +1366,8 @@ simplify_context::simplify_unary_operation_1 (rtx_code code, machine_mode mode, break; /* If operand is something known to be positive, ignore the ABS. */ - if (GET_CODE (op) == FFS || GET_CODE (op) == ABS - || val_signbit_known_clear_p (GET_MODE (op), - nonzero_bits (op, GET_MODE (op)))) + if (val_signbit_known_clear_p (GET_MODE (op), + nonzero_bits (op, GET_MODE (op)))) return op; /* If operand is known to be only -1 or 0, convert ABS to NEG. */ @@ -1615,6 +1614,24 @@ simplify_context::simplify_unary_operation_1 (rtx_code code, machine_mode mode, } } + /* We can canonicalize SIGN_EXTEND (op) as ZERO_EXTEND (op) when + we know the sign bit of OP must be clear. */ + if (val_signbit_known_clear_p (GET_MODE (op), + nonzero_bits (op, GET_MODE (op)))) + return simplify_gen_unary (ZERO_EXTEND, mode, op, GET_MODE (op)); + + /* (sign_extend:DI (subreg:SI (ctz:DI ...))) is (ctz:DI ...). */ + if (GET_CODE (op) == SUBREG + && subreg_lowpart_p (op) + && GET_MODE (SUBREG_REG (op)) == mode + && is_a <scalar_int_mode> (mode, &int_mode) + && is_a <scalar_int_mode> (GET_MODE (op), &op_mode) + && GET_MODE_PRECISION (int_mode) <= HOST_BITS_PER_WIDE_INT + && GET_MODE_PRECISION (op_mode) < GET_MODE_PRECISION (int_mode) + && (nonzero_bits (SUBREG_REG (op), mode) + & ~(GET_MODE_MASK (op_mode)>>1)) == 0) + return SUBREG_REG (op); + #if defined(POINTERS_EXTEND_UNSIGNED) /* As we do not know which address space the pointer is referring to, we can do this only if the target does not support different pointer @@ -1765,6 +1782,18 @@ simplify_context::simplify_unary_operation_1 (rtx_code code, machine_mode mode, op0_mode); } + /* (zero_extend:DI (subreg:SI (ctz:DI ...))) is (ctz:DI ...). */ + if (GET_CODE (op) == SUBREG + && subreg_lowpart_p (op) + && GET_MODE (SUBREG_REG (op)) == mode + && is_a <scalar_int_mode> (mode, &int_mode) + && is_a <scalar_int_mode> (GET_MODE (op), &op_mode) + && GET_MODE_PRECISION (int_mode) <= HOST_BITS_PER_WIDE_INT + && GET_MODE_PRECISION (op_mode) < GET_MODE_PRECISION (int_mode) + && (nonzero_bits (SUBREG_REG (op), mode) + & ~GET_MODE_MASK (op_mode)) == 0) + return SUBREG_REG (op); + #if defined(POINTERS_EXTEND_UNSIGNED) /* As we do not know which address space the pointer is referring to, we can do this only if the target does not support different pointer