From patchwork Fri Feb 3 21:21:11 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Michael Meissner X-Patchwork-Id: 64278 Return-Path: X-Original-To: patchwork@sourceware.org Delivered-To: patchwork@sourceware.org Received: from server2.sourceware.org (localhost [IPv6:::1]) by sourceware.org (Postfix) with ESMTP id 6342F385843A for ; Fri, 3 Feb 2023 21:21:52 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 6342F385843A DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org; s=default; t=1675459312; bh=QtK0VIFtRPiU9gDiNi4/j97hdruD0glOwGlI0CsrdU4=; h=Date:To:Subject:References:In-Reply-To:List-Id:List-Unsubscribe: List-Archive:List-Post:List-Help:List-Subscribe:From:Reply-To: From; b=sqT7CZPNP2fBYlxDXX9oXX/HAO5nWY93R7T0SeNyAKoXLopqB2B36S15gl0kQbUAl MvdkQBTwF8cuX7kqcy6gYdUu0gGFzMojkzrpkQIBifRmLN0CDCGswURaYMrNi4WVoH TQxc+dXpNTWjVbhHGVNAXLmKuOxd8IpNvTswTxs0= X-Original-To: gcc-patches@gcc.gnu.org Delivered-To: gcc-patches@gcc.gnu.org Received: from mx0a-001b2d01.pphosted.com (mx0a-001b2d01.pphosted.com [148.163.156.1]) by sourceware.org (Postfix) with ESMTPS id 4A8F13858C52 for ; Fri, 3 Feb 2023 21:21:19 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.2 sourceware.org 4A8F13858C52 Received: from pps.filterd (m0098399.ppops.net [127.0.0.1]) by mx0a-001b2d01.pphosted.com (8.17.1.19/8.17.1.19) with ESMTP id 313L9SZD012763; Fri, 3 Feb 2023 21:21:18 GMT Received: from pps.reinject (localhost [127.0.0.1]) by mx0a-001b2d01.pphosted.com (PPS) with ESMTPS id 3nh7b340qs-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Fri, 03 Feb 2023 21:21:17 +0000 Received: from m0098399.ppops.net (m0098399.ppops.net [127.0.0.1]) by pps.reinject (8.17.1.5/8.17.1.5) with ESMTP id 313KUfpW022411; Fri, 3 Feb 2023 21:21:17 GMT Received: from ppma04dal.us.ibm.com (7a.29.35a9.ip4.static.sl-reverse.com [169.53.41.122]) by mx0a-001b2d01.pphosted.com (PPS) with ESMTPS id 3nh7b340qb-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Fri, 03 Feb 2023 21:21:17 +0000 Received: from pps.filterd (ppma04dal.us.ibm.com [127.0.0.1]) by ppma04dal.us.ibm.com (8.17.1.19/8.17.1.19) with ESMTP id 313K4G31008520; Fri, 3 Feb 2023 21:21:16 GMT Received: from smtprelay02.wdc07v.mail.ibm.com ([9.208.129.120]) by ppma04dal.us.ibm.com (PPS) with ESMTPS id 3ncvw3bfp4-1 (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NOT); Fri, 03 Feb 2023 21:21:16 +0000 Received: from smtpav05.dal12v.mail.ibm.com (smtpav05.dal12v.mail.ibm.com [10.241.53.104]) by smtprelay02.wdc07v.mail.ibm.com (8.14.9/8.14.9/NCO v10.0) with ESMTP id 313LLDnn64422286 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-GCM-SHA384 bits=256 verify=OK); Fri, 3 Feb 2023 21:21:14 GMT Received: from smtpav05.dal12v.mail.ibm.com (unknown [127.0.0.1]) by IMSVA (Postfix) with ESMTP id AED6058065; Fri, 3 Feb 2023 21:21:13 +0000 (GMT) Received: from smtpav05.dal12v.mail.ibm.com (unknown [127.0.0.1]) by IMSVA (Postfix) with ESMTP id 120E058056; Fri, 3 Feb 2023 21:21:13 +0000 (GMT) Received: from toto.the-meissners.org (unknown [9.65.233.34]) by smtpav05.dal12v.mail.ibm.com (Postfix) with ESMTPS; Fri, 3 Feb 2023 21:21:12 +0000 (GMT) Date: Fri, 3 Feb 2023 16:21:11 -0500 To: Michael Meissner , gcc-patches@gcc.gnu.org, Segher Boessenkool , "Kewen.Lin" , David Edelsohn , Peter Bergner , Will Schmidt Subject: [PATCH 1/8] PowerPC: Add -mcpu=future. Message-ID: Mail-Followup-To: Michael Meissner , gcc-patches@gcc.gnu.org, Segher Boessenkool , "Kewen.Lin" , David Edelsohn , Peter Bergner , Will Schmidt References: Content-Disposition: inline In-Reply-To: X-TM-AS-GCONF: 00 X-Proofpoint-ORIG-GUID: 6NMMyIf7DfUO-kZLtMHrsWvNoVTYgqlc X-Proofpoint-GUID: icI8Wl7AHetCBkQ0pCYrYpWt58So007m X-Proofpoint-UnRewURL: 0 URL was un-rewritten MIME-Version: 1.0 X-Proofpoint-Virus-Version: vendor=baseguard engine=ICAP:2.0.219,Aquarius:18.0.930,Hydra:6.0.562,FMLib:17.11.122.1 definitions=2023-02-03_19,2023-02-03_01,2022-06-22_01 X-Proofpoint-Spam-Details: rule=outbound_notspam policy=outbound score=0 impostorscore=0 mlxscore=0 malwarescore=0 spamscore=0 bulkscore=0 lowpriorityscore=0 suspectscore=0 priorityscore=1501 adultscore=0 phishscore=0 clxscore=1015 mlxlogscore=999 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.12.0-2212070000 definitions=main-2302030189 X-Spam-Status: No, score=-10.2 required=5.0 tests=BAYES_00, DKIM_SIGNED, DKIM_VALID, DKIM_VALID_EF, GIT_PATCH_0, KAM_MANYTO, KAM_SHORT, 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 List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-Patchwork-Original-From: Michael Meissner via Gcc-patches From: Michael Meissner Reply-To: Michael Meissner Errors-To: gcc-patches-bounces+patchwork=sourceware.org@gcc.gnu.org Sender: "Gcc-patches" These patches implement support for potential future PowerPC cpus. At this time, features enabled with -mcpu=future may or may not be in actual PowerPCs that will be delivered in the future. This patch adds support for the -mcpu=future and -mtune=future options. If you use -mcpu=future, the macro __ARCH_PWR_FUTURE__ is defined, and the assembler .machine directive "future" is used. Future patches in this series will add support for new instructions that may be present in future PowerPC processors. At the moment, we do not have any differences in tuning between power10 and future. It is anticipated that we may change the tuning characteristics for -mtune=future at a later time. The patches have been tested on the following platforms. I added the patches for PR target/107299 that I submitted on November 2nd before doing the builds so that GCC would build on systems using IEEE 128-bit long double. * https://gcc.gnu.org/pipermail/gcc-patches/2022-November/604834.html There were no regressions with doing bootstrap builds and running the regression tests: 1) Power10 LE using --with-cpu=power10 --with-long-double-format=ieee; 2) Power10 LE using --with-cpu=power10 --with-long-double-format=ibm; 3) Power9 LE using --with-cpu=power9 --with-long-double-format=ibm; and 4) Power8 BE using --with-cpu=power8 (both 32-bit & 64-bit tested). Can I check this patch into the GCC 13 master branch? Note, I will be on vacation from Tuesday February 7th through Tuesday February 14th. 2023-02-03 Michael Meissner gcc/ * config/rs6000/power10.md (power10-load): Temporarily treat -mcpu=future the same as -mcpu=power10. (power10-fused-load): Likewise. (power10-prefixed-load): Likewise. (power10-prefixed-load): Likewise. (power10-load-update): Likewise. (power10-fpload-double): Likewise. (power10-fpload-double): Likewise. (power10-prefixed-fpload-double): Likewise. (power10-prefixed-fpload-double): Likewise. (power10-fpload-update-double): Likewise. (power10-fpload-single): Likewise. (power10-fpload-update-single): Likewise. (power10-vecload): Likewise. (power10-vecload-pair): Likewise. (power10-store): Likewise. (power10-fused-store): Likewise. (power10-prefixed-store): Likewise. (power10-prefixed-store): Likewise. (power10-store-update): Likewise. (power10-vecstore-pair): Likewise. (power10-larx): Likewise. (power10-lq): Likewise. (power10-stcx): Likewise. (power10-stq): Likewise. (power10-sync): Likewise. (power10-sync): Likewise. (power10-alu): Likewise. (power10-fused_alu): Likewise. (power10-paddi): Likewise. (power10-rot): Likewise. (power10-rot-compare): Likewise. (power10-alu2): Likewise. (power10-cmp): Likewise. (power10-two): Likewise. (power10-three): Likewise. (power10-mul): Likewise. (power10-mul-compare): Likewise. (power10-div): Likewise. (power10-div-compare): Likewise. (power10-crlogical): Likewise. (power10-mfcrf): Likewise. (power10-mfcr): Likewise. (power10-mtcr): Likewise. (power10-mtjmpr): Likewise. (power10-mfjmpr): Likewise. (power10-mfjmpr): Likewise. (power10-fpsimple): Likewise. (power10-fp): Likewise. (power10-fpcompare): Likewise. (power10-sdiv): Likewise. (power10-ddiv): Likewise. (power10-sqrt): Likewise. (power10-dsqrt): Likewise. (power10-vec-2cyc): Likewise. (power10-fused-vec): Likewise. (power10-veccmp): Likewise. (power10-vecsimple): Likewise. (power10-vecnormal): Likewise. (power10-qp): Likewise. (power10-vecperm): Likewise. (power10-vecperm-compare): Likewise. (power10-prefixed-vecperm): Likewise. (power10-veccomplex): Likewise. (power10-vecfdiv): Likewise. (power10-vecdiv): Likewise. (power10-qpdiv): Likewise. (power10-qpmul): Likewise. (power10-mtvsr): Likewise. (power10-mfvsr): Likewise. (power10-mfvsr): Likewise. (power10-branch): Likewise. (power10-fused-branch): Likewise. (power10-crypto): Likewise. (power10-htm): Likewise. (power10-htm): Likewise. (power10-dfp): Likewise. (power10-dfpq): Likewise. (power10-mma): Likewise. (power10-prefixed-mma): Likewise. * config/rs6000/rs6000-c.cc (rs6000_target_modify_macros): Define __ARCH_PWR_FUTURE__ if -mcpu=future. * config/rs6000/rs6000-cpus.def (ISA_FUTURE_MASKS): New macro. (POWERPC_MASKS): Add -mcpu=future. * config/rs6000/rs6000-opts.h (enum processor_type): Add PROCESSOR_FUTURE. * config/rs6000/rs6000-tables.opt: Regenerate. * config/rs6000/rs6000.cc (future_costs): Add -mcpu=future support. Make -mtune=future act like -mtune=power10 for now. (rs6000_option_override_internal): (rs6000_machine_from_flags): Likewise. (rs6000_reassociation_width): Likewise. (rs6000_adjust_cost): Likewise. (rs6000_issue_rate): Likewise. (rs6000_sched_reorder): Likewise. (rs6000_sched_reorder2): Likewise. (rs6000_register_move_cost): Likewise. (rs6000_opt_masks): Add -mfuture. * config/rs6000/rs6000.h (ASM_CPU_SUPPORT): Likewise. * config/rs6000/rs6000.md (cpu attribute): Add -mcpu=future support. * config/rs6000/rs6000.opt (-mfuture): New undocumented debug switch. * doc/invoke.texi (IBM RS/6000 and PowerPC Options): Document -mcpu=future. --- gcc/config/rs6000/power10.md | 142 ++++++++++++++-------------- gcc/config/rs6000/rs6000-c.cc | 2 + gcc/config/rs6000/rs6000-cpus.def | 6 ++ gcc/config/rs6000/rs6000-opts.h | 4 +- gcc/config/rs6000/rs6000-tables.opt | 3 + gcc/config/rs6000/rs6000.cc | 51 ++++++++-- gcc/config/rs6000/rs6000.h | 1 + gcc/config/rs6000/rs6000.md | 2 +- gcc/config/rs6000/rs6000.opt | 4 + gcc/doc/invoke.texi | 2 +- 10 files changed, 137 insertions(+), 80 deletions(-) diff --git a/gcc/config/rs6000/power10.md b/gcc/config/rs6000/power10.md index 8e1d4e1afc6..caed2d53668 100644 --- a/gcc/config/rs6000/power10.md +++ b/gcc/config/rs6000/power10.md @@ -97,12 +97,12 @@ (define_insn_reservation "power10-load" 4 (eq_attr "update" "no") (eq_attr "size" "!128") (eq_attr "prefixed" "no") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,LU_power10") (define_insn_reservation "power10-fused-load" 4 (and (eq_attr "type" "fused_load_cmpi,fused_addis_load,fused_load_load") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,LU_power10") (define_insn_reservation "power10-prefixed-load" 4 @@ -110,13 +110,13 @@ (define_insn_reservation "power10-prefixed-load" 4 (eq_attr "update" "no") (eq_attr "size" "!128") (eq_attr "prefixed" "yes") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,LU_power10") (define_insn_reservation "power10-load-update" 4 (and (eq_attr "type" "load") (eq_attr "update" "yes") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,LU_power10+SXU_power10") (define_insn_reservation "power10-fpload-double" 4 @@ -124,7 +124,7 @@ (define_insn_reservation "power10-fpload-double" 4 (eq_attr "update" "no") (eq_attr "size" "64") (eq_attr "prefixed" "no") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,LU_power10") (define_insn_reservation "power10-prefixed-fpload-double" 4 @@ -132,14 +132,14 @@ (define_insn_reservation "power10-prefixed-fpload-double" 4 (eq_attr "update" "no") (eq_attr "size" "64") (eq_attr "prefixed" "yes") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,LU_power10") (define_insn_reservation "power10-fpload-update-double" 4 (and (eq_attr "type" "fpload") (eq_attr "update" "yes") (eq_attr "size" "64") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,LU_power10+SXU_power10") ; SFmode loads are cracked and have additional 3 cycles over DFmode @@ -148,27 +148,27 @@ (define_insn_reservation "power10-fpload-single" 7 (and (eq_attr "type" "fpload") (eq_attr "update" "no") (eq_attr "size" "32") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,LU_power10") (define_insn_reservation "power10-fpload-update-single" 7 (and (eq_attr "type" "fpload") (eq_attr "update" "yes") (eq_attr "size" "32") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,LU_power10+SXU_power10") (define_insn_reservation "power10-vecload" 4 (and (eq_attr "type" "vecload") (eq_attr "size" "!256") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,LU_power10") ; lxvp (define_insn_reservation "power10-vecload-pair" 4 (and (eq_attr "type" "vecload") (eq_attr "size" "256") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,LU_power10+SXU_power10") ; Store Unit @@ -178,12 +178,12 @@ (define_insn_reservation "power10-store" 0 (eq_attr "prefixed" "no") (eq_attr "size" "!128") (eq_attr "size" "!256") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,STU_power10") (define_insn_reservation "power10-fused-store" 0 (and (eq_attr "type" "fused_store_store") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,STU_power10") (define_insn_reservation "power10-prefixed-store" 0 @@ -191,52 +191,52 @@ (define_insn_reservation "power10-prefixed-store" 0 (eq_attr "prefixed" "yes") (eq_attr "size" "!128") (eq_attr "size" "!256") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,STU_power10") ; Update forms have 2 cycle latency for updated addr reg (define_insn_reservation "power10-store-update" 2 (and (eq_attr "type" "store,fpstore") (eq_attr "update" "yes") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,STU_power10") ; stxvp (define_insn_reservation "power10-vecstore-pair" 0 (and (eq_attr "type" "vecstore") (eq_attr "size" "256") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,stu0_power10+stu1_power10") (define_insn_reservation "power10-larx" 4 (and (eq_attr "type" "load_l") (eq_attr "size" "!128") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,LU_power10") ; All load quad forms (define_insn_reservation "power10-lq" 4 (and (eq_attr "type" "load,load_l") (eq_attr "size" "128") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,LU_power10+SXU_power10") (define_insn_reservation "power10-stcx" 0 (and (eq_attr "type" "store_c") (eq_attr "size" "!128") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,STU_power10") ; All store quad forms (define_insn_reservation "power10-stq" 0 (and (eq_attr "type" "store,store_c") (eq_attr "size" "128") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,stu0_power10+stu1_power10") (define_insn_reservation "power10-sync" 1 (and (eq_attr "type" "sync,isync") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,STU_power10") @@ -248,7 +248,7 @@ (define_insn_reservation "power10-sync" 1 (define_insn_reservation "power10-alu" 2 (and (eq_attr "type" "add,exts,integer,logical,isel") (eq_attr "prefixed" "no") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") ; 4 cycle CR latency (define_bypass 4 "power10-alu" @@ -256,28 +256,28 @@ (define_bypass 4 "power10-alu" (define_insn_reservation "power10-fused_alu" 2 (and (eq_attr "type" "fused_arith_logical,fused_cmp_isel,fused_carry") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,EXU_power10") ; paddi (define_insn_reservation "power10-paddi" 2 (and (eq_attr "type" "add") (eq_attr "prefixed" "yes") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,EXU_power10") ; Rotate/shift (non-record form) (define_insn_reservation "power10-rot" 2 (and (eq_attr "type" "insert,shift") (eq_attr "dot" "no") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") ; Record form rotate/shift (define_insn_reservation "power10-rot-compare" 3 (and (eq_attr "type" "insert,shift") (eq_attr "dot" "yes") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") ; 5 cycle CR latency (define_bypass 5 "power10-rot-compare" @@ -285,7 +285,7 @@ (define_bypass 5 "power10-rot-compare" (define_insn_reservation "power10-alu2" 3 (and (eq_attr "type" "cntlz,popcnt,trap") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") ; 5 cycle CR latency (define_bypass 5 "power10-alu2" @@ -293,24 +293,24 @@ (define_bypass 5 "power10-alu2" (define_insn_reservation "power10-cmp" 2 (and (eq_attr "type" "cmp") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") ; Treat 'two' and 'three' types as 2 or 3 way cracked (define_insn_reservation "power10-two" 4 (and (eq_attr "type" "two") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,EXU_power10") (define_insn_reservation "power10-three" 6 (and (eq_attr "type" "three") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_all_power10,EXU_power10") (define_insn_reservation "power10-mul" 5 (and (eq_attr "type" "mul") (eq_attr "dot" "no") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") ; 4 cycle MUL->MUL latency (define_bypass 4 "power10-mul" @@ -319,7 +319,7 @@ (define_bypass 4 "power10-mul" (define_insn_reservation "power10-mul-compare" 5 (and (eq_attr "type" "mul") (eq_attr "dot" "yes") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,EXU_power10") ; 4 cycle MUL->MUL latency (define_bypass 4 "power10-mul-compare" @@ -331,13 +331,13 @@ (define_bypass 7 "power10-mul-compare" (define_insn_reservation "power10-div" 12 (and (eq_attr "type" "div") (eq_attr "dot" "no") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-div-compare" 12 (and (eq_attr "type" "div") (eq_attr "dot" "yes") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,EXU_power10") ; 14 cycle CR latency (define_bypass 14 "power10-div-compare" @@ -345,34 +345,34 @@ (define_bypass 14 "power10-div-compare" (define_insn_reservation "power10-crlogical" 2 (and (eq_attr "type" "cr_logical") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-mfcrf" 2 (and (eq_attr "type" "mfcrf") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-mfcr" 3 (and (eq_attr "type" "mfcr") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,EXU_power10") ; Should differentiate between 1 cr field and > 1 since target of > 1 cr ; is cracked (define_insn_reservation "power10-mtcr" 3 (and (eq_attr "type" "mtcr") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-mtjmpr" 3 (and (eq_attr "type" "mtjmpr") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-mfjmpr" 2 (and (eq_attr "type" "mfjmpr") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") @@ -380,126 +380,126 @@ (define_insn_reservation "power10-mfjmpr" 2 (define_insn_reservation "power10-fpsimple" 3 (and (eq_attr "type" "fpsimple") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-fp" 5 (and (eq_attr "type" "fp,dmul") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-fpcompare" 3 (and (eq_attr "type" "fpcompare") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-sdiv" 22 (and (eq_attr "type" "sdiv") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-ddiv" 27 (and (eq_attr "type" "ddiv") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-sqrt" 26 (and (eq_attr "type" "ssqrt") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-dsqrt" 36 (and (eq_attr "type" "dsqrt") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-vec-2cyc" 2 (and (eq_attr "type" "vecmove,veclogical,vecexts,veccmpfx") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-fused-vec" 2 (and (eq_attr "type" "fused_vector") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,EXU_power10") (define_insn_reservation "power10-veccmp" 3 (and (eq_attr "type" "veccmp") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-vecsimple" 2 (and (eq_attr "type" "vecsimple") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-vecnormal" 5 (and (eq_attr "type" "vecfloat,vecdouble") (eq_attr "size" "!128") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-qp" 12 (and (eq_attr "type" "vecfloat,vecdouble") (eq_attr "size" "128") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-vecperm" 3 (and (eq_attr "type" "vecperm") (eq_attr "prefixed" "no") (eq_attr "dot" "no") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-vecperm-compare" 3 (and (eq_attr "type" "vecperm") (eq_attr "dot" "yes") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,EXU_power10") (define_insn_reservation "power10-prefixed-vecperm" 3 (and (eq_attr "type" "vecperm") (eq_attr "prefixed" "yes") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,EXU_power10") (define_insn_reservation "power10-veccomplex" 6 (and (eq_attr "type" "veccomplex") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-vecfdiv" 24 (and (eq_attr "type" "vecfdiv") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-vecdiv" 27 (and (eq_attr "type" "vecdiv") (eq_attr "size" "!128") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-qpdiv" 56 (and (eq_attr "type" "vecdiv") (eq_attr "size" "128") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-qpmul" 24 (and (eq_attr "type" "qmul") (eq_attr "size" "128") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-mtvsr" 2 (and (eq_attr "type" "mtvsr") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-mfvsr" 2 (and (eq_attr "type" "mfvsr") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") @@ -507,26 +507,26 @@ (define_insn_reservation "power10-mfvsr" 2 ; Branch is 2 cycles, grouped with STU for issue (define_insn_reservation "power10-branch" 2 (and (eq_attr "type" "jmpreg,branch") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,STU_power10") (define_insn_reservation "power10-fused-branch" 3 (and (eq_attr "type" "fused_mtbc") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,STU_power10") ; Crypto (define_insn_reservation "power10-crypto" 4 (and (eq_attr "type" "crypto") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") ; HTM (define_insn_reservation "power10-htm" 2 (and (eq_attr "type" "htmsimple,htm") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") @@ -535,26 +535,26 @@ (define_insn_reservation "power10-htm" 2 (define_insn_reservation "power10-dfp" 12 (and (eq_attr "type" "dfp") (eq_attr "size" "!128") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_power10") (define_insn_reservation "power10-dfpq" 12 (and (eq_attr "type" "dfp") (eq_attr "size" "128") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,EXU_power10") ; MMA (define_insn_reservation "power10-mma" 9 (and (eq_attr "type" "mma") (eq_attr "prefixed" "no") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_any_power10,EXU_super_power10") (define_insn_reservation "power10-prefixed-mma" 9 (and (eq_attr "type" "mma") (eq_attr "prefixed" "yes") - (eq_attr "cpu" "power10")) + (eq_attr "cpu" "power10,future")) "DU_even_power10,EXU_super_power10") ; 4 cycle MMA->MMA latency (define_bypass 4 "power10-mma,power10-prefixed-mma" diff --git a/gcc/config/rs6000/rs6000-c.cc b/gcc/config/rs6000/rs6000-c.cc index 8555174d36e..2803014f2b6 100644 --- a/gcc/config/rs6000/rs6000-c.cc +++ b/gcc/config/rs6000/rs6000-c.cc @@ -447,6 +447,8 @@ rs6000_target_modify_macros (bool define_p, HOST_WIDE_INT flags) rs6000_define_or_undefine_macro (define_p, "_ARCH_PWR9"); if ((flags & OPTION_MASK_POWER10) != 0) rs6000_define_or_undefine_macro (define_p, "_ARCH_PWR10"); + if ((flags & OPTION_MASK_FUTURE) != 0) + rs6000_define_or_undefine_macro (define_p, "_ARCH_PWR_FUTURE"); if ((flags & OPTION_MASK_SOFT_FLOAT) != 0) rs6000_define_or_undefine_macro (define_p, "_SOFT_FLOAT"); if ((flags & OPTION_MASK_RECIP_PRECISION) != 0) diff --git a/gcc/config/rs6000/rs6000-cpus.def b/gcc/config/rs6000/rs6000-cpus.def index 4f350da378c..deb4ea1c980 100644 --- a/gcc/config/rs6000/rs6000-cpus.def +++ b/gcc/config/rs6000/rs6000-cpus.def @@ -86,6 +86,10 @@ | OPTION_MASK_POWER10 \ | OTHER_POWER10_MASKS) +/* Flags for a potential future processor that may or may not be delivered. */ +#define ISA_FUTURE_MASKS (ISA_3_1_MASKS_SERVER \ + | OPTION_MASK_FUTURE) + /* Flags that need to be turned off if -mno-power9-vector. */ #define OTHER_P9_VECTOR_MASKS (OPTION_MASK_FLOAT128_HW \ | OPTION_MASK_P9_MINMAX) @@ -132,6 +136,7 @@ | OPTION_MASK_FPRND \ | OPTION_MASK_POWER10 \ | OPTION_MASK_P10_FUSION \ + | OPTION_MASK_FUTURE \ | OPTION_MASK_HTM \ | OPTION_MASK_ISEL \ | OPTION_MASK_MFCRF \ @@ -263,3 +268,4 @@ RS6000_CPU ("powerpc64", PROCESSOR_POWERPC64, OPTION_MASK_PPC_GFXOPT RS6000_CPU ("powerpc64le", PROCESSOR_POWER8, MASK_POWERPC64 | ISA_2_7_MASKS_SERVER | OPTION_MASK_HTM) RS6000_CPU ("rs64", PROCESSOR_RS64A, OPTION_MASK_PPC_GFXOPT | MASK_POWERPC64) +RS6000_CPU ("future", PROCESSOR_FUTURE, MASK_POWERPC64 | ISA_FUTURE_MASKS) diff --git a/gcc/config/rs6000/rs6000-opts.h b/gcc/config/rs6000/rs6000-opts.h index 8040cfdc06e..f56f01d6fa5 100644 --- a/gcc/config/rs6000/rs6000-opts.h +++ b/gcc/config/rs6000/rs6000-opts.h @@ -67,7 +67,9 @@ enum processor_type PROCESSOR_MPCCORE, PROCESSOR_CELL, PROCESSOR_PPCA2, - PROCESSOR_TITAN + PROCESSOR_TITAN, + + PROCESSOR_FUTURE }; diff --git a/gcc/config/rs6000/rs6000-tables.opt b/gcc/config/rs6000/rs6000-tables.opt index b82f8205fa1..3ff28e39f6c 100644 --- a/gcc/config/rs6000/rs6000-tables.opt +++ b/gcc/config/rs6000/rs6000-tables.opt @@ -197,3 +197,6 @@ Enum(rs6000_cpu_opt_value) String(powerpc64le) Value(55) EnumValue Enum(rs6000_cpu_opt_value) String(rs64) Value(56) +EnumValue +Enum(rs6000_cpu_opt_value) String(future) Value(57) + diff --git a/gcc/config/rs6000/rs6000.cc b/gcc/config/rs6000/rs6000.cc index 7e76c37fdab..ed26e2755cc 100644 --- a/gcc/config/rs6000/rs6000.cc +++ b/gcc/config/rs6000/rs6000.cc @@ -1085,6 +1085,27 @@ struct processor_costs power10_cost = { COSTS_N_INSNS (2), /* SF->DF convert */ }; +/* Instruction costs on Future processors. At the moment, this is a copy of + the power10 costs, but it is expected to change over time.. */ +static const +struct processor_costs future_cost = { + COSTS_N_INSNS (2), /* mulsi */ + COSTS_N_INSNS (2), /* mulsi_const */ + COSTS_N_INSNS (2), /* mulsi_const9 */ + COSTS_N_INSNS (2), /* muldi */ + COSTS_N_INSNS (6), /* divsi */ + COSTS_N_INSNS (6), /* divdi */ + COSTS_N_INSNS (2), /* fp */ + COSTS_N_INSNS (2), /* dmul */ + COSTS_N_INSNS (11), /* sdiv */ + COSTS_N_INSNS (13), /* ddiv */ + 128, /* cache line size */ + 32, /* l1 cache */ + 512, /* l2 cache */ + 16, /* prefetch streams */ + COSTS_N_INSNS (2), /* SF->DF convert */ +}; + /* Instruction costs on POWER A2 processors. */ static const struct processor_costs ppca2_cost = { @@ -4430,6 +4451,7 @@ rs6000_option_override_internal (bool global_init_p) && rs6000_tune != PROCESSOR_POWER8 && rs6000_tune != PROCESSOR_POWER9 && rs6000_tune != PROCESSOR_POWER10 + && rs6000_tune != PROCESSOR_FUTURE && rs6000_tune != PROCESSOR_PPCA2 && rs6000_tune != PROCESSOR_CELL && rs6000_tune != PROCESSOR_PPC476); @@ -4444,6 +4466,7 @@ rs6000_option_override_internal (bool global_init_p) || rs6000_tune == PROCESSOR_POWER8 || rs6000_tune == PROCESSOR_POWER9 || rs6000_tune == PROCESSOR_POWER10 + || rs6000_tune == PROCESSOR_FUTURE || rs6000_tune == PROCESSOR_PPCE500MC || rs6000_tune == PROCESSOR_PPCE500MC64 || rs6000_tune == PROCESSOR_PPCE5500 @@ -4746,6 +4769,10 @@ rs6000_option_override_internal (bool global_init_p) rs6000_cost = &power10_cost; break; + case PROCESSOR_FUTURE: + rs6000_cost = &future_cost; + break; + case PROCESSOR_PPCA2: rs6000_cost = &ppca2_cost; break; @@ -5902,6 +5929,8 @@ rs6000_machine_from_flags (void) /* Disable the flags that should never influence the .machine selection. */ flags &= ~(OPTION_MASK_PPC_GFXOPT | OPTION_MASK_PPC_GPOPT | OPTION_MASK_ISEL); + if ((flags & (ISA_FUTURE_MASKS & ~ISA_3_1_MASKS_SERVER)) != 0) + return "future"; if ((flags & (ISA_3_1_MASKS_SERVER & ~ISA_3_0_MASKS_SERVER)) != 0) return "power10"; if ((flags & (ISA_3_0_MASKS_SERVER & ~ISA_2_7_MASKS_SERVER)) != 0) @@ -10113,6 +10142,7 @@ rs6000_reassociation_width (unsigned int opc ATTRIBUTE_UNUSED, case PROCESSOR_POWER8: case PROCESSOR_POWER9: case PROCESSOR_POWER10: + case PROCESSOR_FUTURE: if (DECIMAL_FLOAT_MODE_P (mode)) return 1; if (VECTOR_MODE_P (mode)) @@ -17912,7 +17942,8 @@ rs6000_adjust_cost (rtx_insn *insn, int dep_type, rtx_insn *dep_insn, int cost, /* Separate a load from a narrower, dependent store. */ if ((rs6000_sched_groups || rs6000_tune == PROCESSOR_POWER9 - || rs6000_tune == PROCESSOR_POWER10) + || rs6000_tune == PROCESSOR_POWER10 + || rs6000_tune == PROCESSOR_FUTURE) && GET_CODE (PATTERN (insn)) == SET && GET_CODE (PATTERN (dep_insn)) == SET && MEM_P (XEXP (PATTERN (insn), 1)) @@ -17951,6 +17982,7 @@ rs6000_adjust_cost (rtx_insn *insn, int dep_type, rtx_insn *dep_insn, int cost, || rs6000_tune == PROCESSOR_POWER8 || rs6000_tune == PROCESSOR_POWER9 || rs6000_tune == PROCESSOR_POWER10 + || rs6000_tune == PROCESSOR_FUTURE || rs6000_tune == PROCESSOR_CELL) && recog_memoized (dep_insn) && (INSN_CODE (dep_insn) >= 0)) @@ -18525,6 +18557,7 @@ rs6000_issue_rate (void) case PROCESSOR_POWER9: return 6; case PROCESSOR_POWER10: + case PROCESSOR_FUTURE: return 8; default: return 1; @@ -19240,8 +19273,10 @@ rs6000_sched_reorder (FILE *dump ATTRIBUTE_UNUSED, int sched_verbose, if (rs6000_tune == PROCESSOR_POWER6) load_store_pendulum = 0; - /* Do Power10 dependent reordering. */ - if (rs6000_tune == PROCESSOR_POWER10 && last_scheduled_insn) + /* Do Power10 dependent reordering. For now, assume "future" has the same + dependent reordering as power10. */ + if ((rs6000_tune == PROCESSOR_POWER10 + || rs6000_tune == PROCESSOR_FUTURE) && last_scheduled_insn) power10_sched_reorder (ready, n_ready - 1); return rs6000_issue_rate (); @@ -19265,8 +19300,10 @@ rs6000_sched_reorder2 (FILE *dump, int sched_verbose, rtx_insn **ready, && recog_memoized (last_scheduled_insn) >= 0) return power9_sched_reorder2 (ready, *pn_ready - 1); - /* Do Power10 dependent reordering. */ - if (rs6000_tune == PROCESSOR_POWER10 && last_scheduled_insn) + /* Do Power10 dependent reordering. For now, assume "future" has the same + dependent reordering as power10. */ + if ((rs6000_tune == PROCESSOR_POWER10 + || rs6000_tune == PROCESSOR_FUTURE) && last_scheduled_insn) return power10_sched_reorder (ready, *pn_ready - 1); return cached_can_issue_more; @@ -22481,7 +22518,8 @@ rs6000_register_move_cost (machine_mode mode, allocation a move within the same class might turn out to be a nop. */ if (rs6000_tune == PROCESSOR_POWER9 - || rs6000_tune == PROCESSOR_POWER10) + || rs6000_tune == PROCESSOR_POWER10 + || rs6000_tune == PROCESSOR_FUTURE) ret = 3 * hard_regno_nregs (FIRST_GPR_REGNO, mode); else ret = 4 * hard_regno_nregs (FIRST_GPR_REGNO, mode); @@ -24139,6 +24177,7 @@ static struct rs6000_opt_mask const rs6000_opt_masks[] = { "float128-hardware", OPTION_MASK_FLOAT128_HW, false, true }, { "fprnd", OPTION_MASK_FPRND, false, true }, { "power10", OPTION_MASK_POWER10, false, true }, + { "future", OPTION_MASK_FUTURE, false, true }, { "hard-dfp", OPTION_MASK_DFP, false, true }, { "htm", OPTION_MASK_HTM, false, true }, { "isel", OPTION_MASK_ISEL, false, true }, diff --git a/gcc/config/rs6000/rs6000.h b/gcc/config/rs6000/rs6000.h index 3503614efbd..44fa355a061 100644 --- a/gcc/config/rs6000/rs6000.h +++ b/gcc/config/rs6000/rs6000.h @@ -163,6 +163,7 @@ mcpu=e5500: -me5500; \ mcpu=e6500: -me6500; \ mcpu=titan: -mtitan; \ + mcpu=future: -mfuture; \ !mcpu*: %{mpower9-vector: -mpower9; \ mpower8-vector|mcrypto|mdirect-move|mhtm: -mpower8; \ mvsx: -mpower7; \ diff --git a/gcc/config/rs6000/rs6000.md b/gcc/config/rs6000/rs6000.md index 4a7812fa592..5f933bede93 100644 --- a/gcc/config/rs6000/rs6000.md +++ b/gcc/config/rs6000/rs6000.md @@ -350,7 +350,7 @@ (define_attr "cpu" ppc403,ppc405,ppc440,ppc476, ppc8540,ppc8548,ppce300c2,ppce300c3,ppce500mc,ppce500mc64,ppce5500,ppce6500, power4,power5,power6,power7,power8,power9,power10, - rs64a,mpccore,cell,ppca2,titan" + rs64a,mpccore,cell,ppca2,titan,future" (const (symbol_ref "(enum attr_cpu) rs6000_tune"))) ;; The ISA we implement. diff --git a/gcc/config/rs6000/rs6000.opt b/gcc/config/rs6000/rs6000.opt index bde6d3ff664..04532a774b9 100644 --- a/gcc/config/rs6000/rs6000.opt +++ b/gcc/config/rs6000/rs6000.opt @@ -620,6 +620,10 @@ mieee128-constant Target Var(TARGET_IEEE128_CONSTANT) Init(1) Save Generate (do not generate) code that uses the LXVKQ instruction. +mfuture +Target Undocumented Mask(FUTURE) Var(rs6000_isa_flags) +Generate (do not generate) future instructions. + ; Documented parameters -param=rs6000-vect-unroll-limit= diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi index 1eda0e0396b..696f941c1c5 100644 --- a/gcc/doc/invoke.texi +++ b/gcc/doc/invoke.texi @@ -29209,7 +29209,7 @@ Supported values for @var{cpu_type} are @samp{401}, @samp{403}, @samp{titan}, @samp{power3}, @samp{power4}, @samp{power5}, @samp{power5+}, @samp{power6}, @samp{power6x}, @samp{power7}, @samp{power8}, @samp{power9}, @samp{power10}, @samp{powerpc}, @samp{powerpc64}, -@samp{powerpc64le}, @samp{rs64}, and @samp{native}. +@samp{powerpc64le}, @samp{rs64}, @samp{future}, and @samp{native}. @option{-mcpu=powerpc}, @option{-mcpu=powerpc64}, and @option{-mcpu=powerpc64le} specify pure 32-bit PowerPC (either