From patchwork Mon May 15 14:33:12 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Richard Henderson X-Patchwork-Id: 681968 Delivered-To: patch@linaro.org Received: by 2002:adf:fd8f:0:0:0:0:0 with SMTP id d15csp1152677wrr; Mon, 15 May 2023 07:44:23 -0700 (PDT) X-Google-Smtp-Source: ACHHUZ5oGBfPExF0nrCYWBbRR7bfzSAVODILWTdDe4xXfzjS+W1ty5Xk1Lpa1VCspHdWy+ilyDSq X-Received: by 2002:adf:fdd1:0:b0:306:2ff1:5227 with SMTP id i17-20020adffdd1000000b003062ff15227mr24111241wrs.23.1684161862792; Mon, 15 May 2023 07:44:22 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1684161862; cv=none; d=google.com; s=arc-20160816; b=nxHxTpKaqd4rFXJk3fDLj5FoUtcTdSKYeck7lSnwHPl6dusQ60jh3c/z1PEP/Cuw/p 5VjTLHB0L1yNphaK7MA8qXTKtxkjMPkwdiutRDYnULitMdM4NlaGqG4c/rZy7yzRmVfu RxlKn/XJsU1YvEG+BydOkce37a11lxwJlAoRvS2RqOmzTI/8TKy0yAAM6hDS0SqO20Av 7cdXDcqy/E/xxBulD0MlbSq/RDCEsfpQK4XjrAosrlX4pRiIzhEVv8N8GDWuiSgdEBe2 SYBK76bbRj0aBLIdG8AoqFLrwtTFinsTu6Co/w13KGxS/qJbBb3xY8JpwVEPYukTy5GZ 5LRg== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=sender:errors-to:list-subscribe:list-help:list-post:list-archive :list-unsubscribe:list-id:precedence:content-transfer-encoding :mime-version:references:in-reply-to:message-id:date:subject:cc:to :from:dkim-signature; bh=a3C62E7I5KSH6hopm1N3+kJjnyiF8W6DIOatywqt9qE=; b=ILaMzlVLWPVtzG2IbbjJxsY0D1LmJyTzKBHRmZV0XpxBpQHuy+JXip+yOLK+hJ7Sbi hjDgn6mbbBfkR0ZwvpupKdvFIw+sU/x2bN41Bm4GUadMoc6OBpQNJg+2CjRo5N4eSZlG OKOpzg4jf9Dy8uJeS/fu/AYGpEABavWUqr9oftquYKzN1kyvupMsz2+nPhXzRcg5QkdX 7djGGmVkEC+nWaYAzTDk1Rr4We37voLRwi3GFXX1wO6lk8iLL7ydckVKbJ+G0Wy+Uuc+ KL77NInyFht3k1T+MJR8Zt9j+XqwKRttEV8VOWFPfFQXOSq4Xg4DaJGkt7BxsWJ/AtUh OegA== ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@linaro.org header.s=google header.b=kcvh4loy; spf=pass (google.com: domain of qemu-devel-bounces+patch=linaro.org@nongnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom="qemu-devel-bounces+patch=linaro.org@nongnu.org"; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=linaro.org Return-Path: Received: from lists.gnu.org (lists.gnu.org. [209.51.188.17]) by mx.google.com with ESMTPS id z18-20020adff752000000b002ff15b4136fsi14652790wrp.418.2023.05.15.07.44.22 for (version=TLS1_2 cipher=ECDHE-ECDSA-CHACHA20-POLY1305 bits=256/256); Mon, 15 May 2023 07:44:22 -0700 (PDT) Received-SPF: pass (google.com: domain of qemu-devel-bounces+patch=linaro.org@nongnu.org designates 209.51.188.17 as permitted sender) client-ip=209.51.188.17; Authentication-Results: mx.google.com; dkim=pass header.i=@linaro.org header.s=google header.b=kcvh4loy; spf=pass (google.com: domain of qemu-devel-bounces+patch=linaro.org@nongnu.org designates 209.51.188.17 as permitted sender) smtp.mailfrom="qemu-devel-bounces+patch=linaro.org@nongnu.org"; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=linaro.org Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1pyZLN-0008Qr-Ph; Mon, 15 May 2023 10:38:33 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1pyZK7-0005ti-0V for qemu-devel@nongnu.org; Mon, 15 May 2023 10:37:20 -0400 Received: from mail-pl1-x634.google.com ([2607:f8b0:4864:20::634]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.90_1) (envelope-from ) id 1pyZJs-00062d-Rj for qemu-devel@nongnu.org; Mon, 15 May 2023 10:37:14 -0400 Received: by mail-pl1-x634.google.com with SMTP id d9443c01a7336-1aaf2ede38fso124301315ad.2 for ; Mon, 15 May 2023 07:37:00 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linaro.org; s=google; t=1684161419; x=1686753419; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=a3C62E7I5KSH6hopm1N3+kJjnyiF8W6DIOatywqt9qE=; b=kcvh4loy5A+0KhiHPS28qq4x7z0Isll6eS2rML2ZU/gSpriYoEF9BahWdJ0w1nG7X/ imkfTW/oopVyrTDbxhDctOAs3HzXbJHeG1e8EBEDrbKsKlVNDix3dl1sVodXrTXlbyoO 3paDA2eqQL8ySpIWItxUhG/Phd8OXL/c+BDF2AjnJEmusPsLXqucO5gpbQHziLU/JODh 2KAY4cU73TGc0qS2jA75wOD/Gv14C9mUaILz8P489RftO6qWXAc1b0ToHSKKs0EupQHH J90xTO/3sI2aObEO60vryMXXB5n+9bT3CkXrSHPPfgj/HFL2oicl4HkdkrNaRY+SqRCx 2mfA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1684161419; x=1686753419; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=a3C62E7I5KSH6hopm1N3+kJjnyiF8W6DIOatywqt9qE=; b=cLHN5a3bY2yLcbR4dYJZo36WStS8hQYHGyujLMoU30H+o8ifkb+E+OrjsRwON4nvYu EeB8Mvkh803ox5KcsaGLfn0MKo/0YQFNmXz2PkGUySBSqPZ2ilEMl1o5G45XSxmiqR7m 94hY180TC379f0/BLxbC2tnF93paDl2owDm0qZBgq1JHjw79LkEADVos7dYoz101xM93 jBxVkr9GaRvB1D1flHSR9J8MUTJpPDk++SrP1aqyZAsDWIhflQAyFL6F6plNUEgVeSrw 6iZsY9plSoiMUe0GhwtR/pj5vOcOvILm22GlGzismnLTAbGIqEzqFblca07eElbw1u1M TH3w== X-Gm-Message-State: AC+VfDyyd46CZN3vFMG0iEqSHfj1BjPHZrE8YwLE4wYnIErExKC5dBv5 mAVpb9yogCo/IWan5XyHj4xBQNe0VLHRVwrUIFc= X-Received: by 2002:a17:902:f547:b0:1ac:4a41:d38d with SMTP id h7-20020a170902f54700b001ac4a41d38dmr40019890plf.51.1684161419504; Mon, 15 May 2023 07:36:59 -0700 (PDT) Received: from stoup.. ([2602:ae:1598:4c01:9902:96ac:8d8c:4366]) by smtp.gmail.com with ESMTPSA id i6-20020a1709026ac600b001aadfdfcd06sm8942735plt.299.2023.05.15.07.36.58 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Mon, 15 May 2023 07:36:59 -0700 (PDT) From: Richard Henderson To: qemu-devel@nongnu.org Cc: qemu-arm@nongnu.org, qemu-s390x@nongnu.org, Daniel Henrique Barboza Subject: [PATCH v5 53/54] tcg/ppc: Support 128-bit load/store Date: Mon, 15 May 2023 07:33:12 -0700 Message-Id: <20230515143313.734053-54-richard.henderson@linaro.org> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20230515143313.734053-1-richard.henderson@linaro.org> References: <20230515143313.734053-1-richard.henderson@linaro.org> MIME-Version: 1.0 Received-SPF: pass client-ip=2607:f8b0:4864:20::634; envelope-from=richard.henderson@linaro.org; helo=mail-pl1-x634.google.com X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, T_SCC_BODY_TEXT_LINE=-0.01 autolearn=unavailable autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+patch=linaro.org@nongnu.org Sender: qemu-devel-bounces+patch=linaro.org@nongnu.org Use LQ/STQ with ISA v2.07, and 16-byte atomicity is required. Note that these instructions do not require 16-byte alignment. Reviewed-by: Daniel Henrique Barboza Signed-off-by: Richard Henderson --- tcg/ppc/tcg-target-con-set.h | 2 + tcg/ppc/tcg-target-con-str.h | 1 + tcg/ppc/tcg-target.h | 3 +- tcg/ppc/tcg-target.c.inc | 115 +++++++++++++++++++++++++++++++---- 4 files changed, 108 insertions(+), 13 deletions(-) diff --git a/tcg/ppc/tcg-target-con-set.h b/tcg/ppc/tcg-target-con-set.h index f206b29205..bbd7b21247 100644 --- a/tcg/ppc/tcg-target-con-set.h +++ b/tcg/ppc/tcg-target-con-set.h @@ -14,6 +14,7 @@ C_O0_I2(r, r) C_O0_I2(r, ri) C_O0_I2(v, r) C_O0_I3(r, r, r) +C_O0_I3(o, m, r) C_O0_I4(r, r, ri, ri) C_O0_I4(r, r, r, r) C_O1_I1(r, r) @@ -34,6 +35,7 @@ C_O1_I3(v, v, v, v) C_O1_I4(r, r, ri, rZ, rZ) C_O1_I4(r, r, r, ri, ri) C_O2_I1(r, r, r) +C_O2_I1(o, m, r) C_O2_I2(r, r, r, r) C_O2_I4(r, r, rI, rZM, r, r) C_O2_I4(r, r, r, r, rI, rZM) diff --git a/tcg/ppc/tcg-target-con-str.h b/tcg/ppc/tcg-target-con-str.h index 094613cbcb..20846901de 100644 --- a/tcg/ppc/tcg-target-con-str.h +++ b/tcg/ppc/tcg-target-con-str.h @@ -9,6 +9,7 @@ * REGS(letter, register_mask) */ REGS('r', ALL_GENERAL_REGS) +REGS('o', ALL_GENERAL_REGS & 0xAAAAAAAAu) /* odd registers */ REGS('v', ALL_VECTOR_REGS) /* diff --git a/tcg/ppc/tcg-target.h b/tcg/ppc/tcg-target.h index 0914380bd7..204b70f86a 100644 --- a/tcg/ppc/tcg-target.h +++ b/tcg/ppc/tcg-target.h @@ -149,7 +149,8 @@ extern bool have_vsx; #define TCG_TARGET_HAS_mulsh_i64 1 #endif -#define TCG_TARGET_HAS_qemu_ldst_i128 0 +#define TCG_TARGET_HAS_qemu_ldst_i128 \ + (TCG_TARGET_REG_BITS == 64 && have_isa_2_07) /* * While technically Altivec could support V64, it has no 64-bit store diff --git a/tcg/ppc/tcg-target.c.inc b/tcg/ppc/tcg-target.c.inc index b5c49895f3..c3a1527856 100644 --- a/tcg/ppc/tcg-target.c.inc +++ b/tcg/ppc/tcg-target.c.inc @@ -295,25 +295,27 @@ static bool tcg_target_const_match(int64_t val, TCGType type, int ct) #define B OPCD( 18) #define BC OPCD( 16) + #define LBZ OPCD( 34) #define LHZ OPCD( 40) #define LHA OPCD( 42) #define LWZ OPCD( 32) #define LWZUX XO31( 55) -#define STB OPCD( 38) -#define STH OPCD( 44) -#define STW OPCD( 36) - -#define STD XO62( 0) -#define STDU XO62( 1) -#define STDX XO31(149) - #define LD XO58( 0) #define LDX XO31( 21) #define LDU XO58( 1) #define LDUX XO31( 53) #define LWA XO58( 2) #define LWAX XO31(341) +#define LQ OPCD( 56) + +#define STB OPCD( 38) +#define STH OPCD( 44) +#define STW OPCD( 36) +#define STD XO62( 0) +#define STDU XO62( 1) +#define STDX XO31(149) +#define STQ XO62( 2) #define ADDIC OPCD( 12) #define ADDI OPCD( 14) @@ -2020,7 +2022,18 @@ typedef struct { bool tcg_target_has_memory_bswap(MemOp memop) { - return true; + TCGAtomAlign aa; + + if ((memop & MO_SIZE) <= MO_64) { + return true; + } + + /* + * Reject 16-byte memop with 16-byte atomicity, + * but do allow a pair of 64-bit operations. + */ + aa = atom_and_align_for_opc(tcg_ctx, memop, MO_ATOM_IFALIGN, true); + return aa.atom <= MO_64; } /* @@ -2035,7 +2048,7 @@ static TCGLabelQemuLdst *prepare_host_addr(TCGContext *s, HostAddress *h, { TCGLabelQemuLdst *ldst = NULL; MemOp opc = get_memop(oi); - MemOp a_bits; + MemOp a_bits, s_bits; /* * Book II, Section 1.4, Single-Copy Atomicity, specifies: @@ -2047,10 +2060,11 @@ static TCGLabelQemuLdst *prepare_host_addr(TCGContext *s, HostAddress *h, * As of 3.0, "the non-atomic access is performed as described in * the corresponding list", which matches MO_ATOM_SUBALIGN. */ + s_bits = opc & MO_SIZE; h->aa = atom_and_align_for_opc(s, opc, have_isa_3_00 ? MO_ATOM_SUBALIGN : MO_ATOM_IFALIGN, - false); + s_bits == MO_128); a_bits = h->aa.align; #ifdef CONFIG_SOFTMMU @@ -2060,7 +2074,6 @@ static TCGLabelQemuLdst *prepare_host_addr(TCGContext *s, HostAddress *h, int fast_off = TLB_MASK_TABLE_OFS(mem_index); int mask_off = fast_off + offsetof(CPUTLBDescFast, mask); int table_off = fast_off + offsetof(CPUTLBDescFast, table); - unsigned s_bits = opc & MO_SIZE; ldst = new_ldst_label(s); ldst->is_ld = is_ld; @@ -2303,6 +2316,70 @@ static void tcg_out_qemu_st(TCGContext *s, TCGReg datalo, TCGReg datahi, } } +static TCGLabelQemuLdst * +prepare_host_addr_index_only(TCGContext *s, HostAddress *h, TCGReg addr_reg, + MemOpIdx oi, bool is_ld) +{ + TCGLabelQemuLdst *ldst; + + ldst = prepare_host_addr(s, h, addr_reg, -1, oi, true); + + /* Compose the final address, as LQ/STQ have no indexing. */ + if (h->base != 0) { + tcg_out32(s, ADD | TAB(TCG_REG_TMP1, h->base, h->index)); + h->index = TCG_REG_TMP1; + h->base = 0; + } + + return ldst; +} + +static void tcg_out_qemu_ldst_i128(TCGContext *s, TCGReg datalo, TCGReg datahi, + TCGReg addr_reg, MemOpIdx oi, bool is_ld) +{ + TCGLabelQemuLdst *ldst; + HostAddress h; + bool need_bswap; + uint32_t insn; + + ldst = prepare_host_addr_index_only(s, &h, addr_reg, oi, is_ld); + need_bswap = get_memop(oi) & MO_BSWAP; + + if (h.aa.atom == MO_128) { + tcg_debug_assert(!need_bswap); + tcg_debug_assert(datalo & 1); + tcg_debug_assert(datahi == datalo - 1); + insn = is_ld ? LQ : STQ; + tcg_out32(s, insn | TAI(datahi, h.index, 0)); + } else { + TCGReg d1, d2; + + if (HOST_BIG_ENDIAN ^ need_bswap) { + d1 = datahi, d2 = datalo; + } else { + d1 = datalo, d2 = datahi; + } + + if (need_bswap) { + tcg_out_movi(s, TCG_TYPE_PTR, TCG_REG_R0, 8); + insn = is_ld ? LDBRX : STDBRX; + tcg_out32(s, insn | TAB(d1, 0, h.index)); + tcg_out32(s, insn | TAB(d2, h.index, TCG_REG_R0)); + } else { + insn = is_ld ? LD : STD; + tcg_out32(s, insn | TAI(d1, h.index, 0)); + tcg_out32(s, insn | TAI(d2, h.index, 8)); + } + } + + if (ldst) { + ldst->type = TCG_TYPE_I128; + ldst->datalo_reg = datalo; + ldst->datahi_reg = datahi; + ldst->raddr = tcg_splitwx_to_rx(s->code_ptr); + } +} + static void tcg_out_nop_fill(tcg_insn_unit *p, int count) { int i; @@ -2853,6 +2930,11 @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, args[4], TCG_TYPE_I64); } break; + case INDEX_op_qemu_ld_i128: + tcg_debug_assert(TCG_TARGET_REG_BITS == 64); + tcg_out_qemu_ldst_i128(s, args[0], args[1], args[2], args[3], true); + break; + case INDEX_op_qemu_st_i32: if (TCG_TARGET_REG_BITS >= TARGET_LONG_BITS) { tcg_out_qemu_st(s, args[0], -1, args[1], -1, @@ -2874,6 +2956,10 @@ static void tcg_out_op(TCGContext *s, TCGOpcode opc, args[4], TCG_TYPE_I64); } break; + case INDEX_op_qemu_st_i128: + tcg_debug_assert(TCG_TARGET_REG_BITS == 64); + tcg_out_qemu_ldst_i128(s, args[0], args[1], args[2], args[3], false); + break; case INDEX_op_setcond_i32: tcg_out_setcond(s, TCG_TYPE_I32, args[3], args[0], args[1], args[2], @@ -3709,6 +3795,11 @@ static TCGConstraintSetIndex tcg_target_op_def(TCGOpcode op) : TARGET_LONG_BITS == 32 ? C_O0_I3(r, r, r) : C_O0_I4(r, r, r, r)); + case INDEX_op_qemu_ld_i128: + return C_O2_I1(o, m, r); + case INDEX_op_qemu_st_i128: + return C_O0_I3(o, m, r); + case INDEX_op_add_vec: case INDEX_op_sub_vec: case INDEX_op_mul_vec: