From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id 393CDC433EF for ; Thu, 2 Jun 2022 22:37:10 +0000 (UTC) Received: from localhost ([::1]:43644 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1nwtRF-0005tL-BZ for qemu-devel@archiver.kernel.org; Thu, 02 Jun 2022 18:37:09 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:38426) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1nwskG-0001cr-5f for qemu-devel@nongnu.org; Thu, 02 Jun 2022 17:52:49 -0400 Received: from mail-pf1-x431.google.com ([2607:f8b0:4864:20::431]:33524) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.90_1) (envelope-from ) id 1nwskC-0000uE-WD for qemu-devel@nongnu.org; Thu, 02 Jun 2022 17:52:43 -0400 Received: by mail-pf1-x431.google.com with SMTP id w21so5850085pfc.0 for ; Thu, 02 Jun 2022 14:52:40 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=linaro.org; s=google; h=from:to:cc:subject:date:message-id:in-reply-to:references :mime-version:content-transfer-encoding; bh=l9/P4k41tvaBx3FynADzNq0fNHgArONzp5FUlS6UNac=; b=EFMg04W1++2hdCw6/xl11577eFiJHLXPJMFV4slglvOPf2P5trJcAlDnWoTx0YRXwv 61b/xeV0h0Qt6bwKZXNu2rl4oK95L+fONcQZrvjZDfsWYaJpJd7soBYsBX4m1plVoFsk UUEBTzaxSSGyZY/CsTsxuG9hMLgCyO0l3Vzk/ypRkE3+HQKSG5nqMw2yBOih+mOCRlN4 Udyi9TJx6KG9OR6t7F1Mi4pYDlf//DT5zRQhUQW4XuGEurPNjt+pQos6LMxPSjU9xNE6 MfLKXoaAD57Zattfx60TjI4uwMM1WMkyI7gKytpUq7fMwXP8rDQr9UUsHZP5ohQ2+wbN o5wA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references:mime-version:content-transfer-encoding; bh=l9/P4k41tvaBx3FynADzNq0fNHgArONzp5FUlS6UNac=; b=TXEmL1QRxGBj/NSpaHa4UMrxqZ5J7PO8SDhP3AaSjHzBoP4JTeUUCWP52Yj1dOoFyv p2pPZZA38/rFZajUAE8AvPh67Yo3YWK9ccPJtMVI9HE79PT7bnBNTF0/ps8jQgx0fj/0 mIvzE/HBO/pRRHIxolOMjogTk58zMSIK8DQ7vpWE8am53Iv5OJvLfwk0Xudtre4gv/20 IlZbTFZ25d7auqdiJ923X+1DNmmXzE6gt21GjceEwSWLs4gE8q8zoiENangJjv2QSDKv FQrRwvRjPNay8UTmobUx+7YOAQifZ/ZXOMeAeH4bZTYlQlfSLS7OTmKJt4/zBhSzsx1N tTVw== X-Gm-Message-State: AOAM5327g4PYURJMF9fKNQTduJHH1fdkpnCQI+fufjOV7Y0KJdtTNWEb l1LKc+st0XFv9Rz22TqOJUa3BW9lgzDKfA== X-Google-Smtp-Source: ABdhPJwFzH14KgPRMVsHwOu8SNwWr3I7BxOH6rFKTIknJMIDdPTR0tUgxNbrukTjBqFDpPpRyudC6w== X-Received: by 2002:a65:6b8a:0:b0:3db:7dc5:fec2 with SMTP id d10-20020a656b8a000000b003db7dc5fec2mr5765553pgw.223.1654206759577; Thu, 02 Jun 2022 14:52:39 -0700 (PDT) Received: from stoup.. (174-21-71-225.tukw.qwest.net. [174.21.71.225]) by smtp.gmail.com with ESMTPSA id e14-20020a170902ed8e00b0015edfccfdb5sm4039605plj.50.2022.06.02.14.52.38 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Thu, 02 Jun 2022 14:52:39 -0700 (PDT) From: Richard Henderson To: qemu-devel@nongnu.org Cc: qemu-arm@nongnu.org Subject: [PATCH 51/71] target/arm: Implement BFMOPA, BFMOPS Date: Thu, 2 Jun 2022 14:48:33 -0700 Message-Id: <20220602214853.496211-52-richard.henderson@linaro.org> X-Mailer: git-send-email 2.34.1 In-Reply-To: <20220602214853.496211-1-richard.henderson@linaro.org> References: <20220602214853.496211-1-richard.henderson@linaro.org> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Received-SPF: pass client-ip=2607:f8b0:4864:20::431; envelope-from=richard.henderson@linaro.org; helo=mail-pf1-x431.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=ham 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+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" Signed-off-by: Richard Henderson --- target/arm/helper-sme.h | 2 ++ target/arm/sme.decode | 2 ++ target/arm/sme_helper.c | 52 ++++++++++++++++++++++++++++++++++++++ target/arm/translate-sme.c | 29 +++++++++++++++++++++ 4 files changed, 85 insertions(+) diff --git a/target/arm/helper-sme.h b/target/arm/helper-sme.h index 727095a3eb..6b36542133 100644 --- a/target/arm/helper-sme.h +++ b/target/arm/helper-sme.h @@ -124,3 +124,5 @@ DEF_HELPER_FLAGS_7(sme_fmopa_s, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, ptr, ptr, i32) DEF_HELPER_FLAGS_7(sme_fmopa_d, TCG_CALL_NO_RWG, void, ptr, ptr, ptr, ptr, ptr, ptr, i32) +DEF_HELPER_FLAGS_6(sme_bfmopa, TCG_CALL_NO_RWG, + void, ptr, ptr, ptr, ptr, ptr, i32) diff --git a/target/arm/sme.decode b/target/arm/sme.decode index ba4774d174..afd9c0dffd 100644 --- a/target/arm/sme.decode +++ b/target/arm/sme.decode @@ -73,3 +73,5 @@ ADDVA_d 11000000 11 01000 1 ... ... ..... 00 ... @adda_64 FMOPA_s 10000000 100 ..... ... ... ..... . 00 .. @op_32 FMOPA_d 10000000 110 ..... ... ... ..... . 0 ... @op_64 + +BFMOPA 10000001 100 ..... ... ... ..... . 00 .. @op_32 diff --git a/target/arm/sme_helper.c b/target/arm/sme_helper.c index 16655c86a2..69e4252abc 100644 --- a/target/arm/sme_helper.c +++ b/target/arm/sme_helper.c @@ -963,3 +963,55 @@ void HELPER(sme_fmopa_d)(void *vza, void *vzn, void *vzm, void *vpn, set_default_nan_mode(save_dn, vst); } + +/* + * Alter PAIR as needed for controlling predicates being false, + * and for NEG on an enabled row element. + */ +static inline uint32_t f16mop_adj_pair(uint32_t pair, uint32_t pg, uint32_t neg) +{ + pair ^= neg; + if (!(pg & 1)) { + pair &= 0xffff0000u; + } + if (!(pg & 4)) { + pair &= 0x0000ffffu; + } + return pair; +} + +void HELPER(sme_bfmopa)(void *vza, void *vzn, void *vzm, void *vpn, + void *vpm, uint32_t desc) +{ + intptr_t row, col, oprsz = simd_maxsz(desc); + uint32_t neg = simd_data(desc) << 15; + uint16_t *pn = vpn, *pm = vpm; + + for (row = 0; row < oprsz; ) { + uint16_t pa = pn[H2(row >> 4)]; + do { + void *vza_row = vza + row * sizeof(ARMVectorReg); + uint32_t n = *(uint32_t *)(vzn + row); + + n = f16mop_adj_pair(n, pa, neg); + + for (col = 0; col < oprsz; ) { + uint16_t pb = pm[H2(col >> 4)]; + do { + if ((pa & 0b0101) == 0b0101 || (pb & 0b0101) == 0b0101) { + uint32_t *a = vza_row + col; + uint32_t m = *(uint32_t *)(vzm + col); + + m = f16mop_adj_pair(m, pb, neg); + *a = bfdotadd(*a, n, m); + + col += 4; + pb >>= 4; + } + } while (col & 15); + } + row += 4; + pa >>= 4; + } while (row & 15); + } +} diff --git a/target/arm/translate-sme.c b/target/arm/translate-sme.c index e6e4541e76..581bf9174f 100644 --- a/target/arm/translate-sme.c +++ b/target/arm/translate-sme.c @@ -274,6 +274,32 @@ TRANS_FEAT(ADDVA_s, aa64_sme, do_adda, a, MO_32, gen_helper_sme_addva_s) TRANS_FEAT(ADDHA_d, aa64_sme_i16i64, do_adda, a, MO_64, gen_helper_sme_addha_d) TRANS_FEAT(ADDVA_d, aa64_sme_i16i64, do_adda, a, MO_64, gen_helper_sme_addva_d) +static bool do_outprod(DisasContext *s, arg_op *a, MemOp esz, + gen_helper_gvec_5 *fn) +{ + uint32_t desc = simd_desc(s->svl, s->svl, a->sub); + TCGv_ptr za, zn, zm, pn, pm; + + if (!sme_smza_enabled_check(s)) { + return true; + } + + /* Sum XZR+zad to find ZAd. */ + za = get_tile_rowcol(s, esz, 31, a->zad, false); + zn = vec_full_reg_ptr(s, a->zn); + zm = vec_full_reg_ptr(s, a->zm); + pn = pred_full_reg_ptr(s, a->pn); + pm = pred_full_reg_ptr(s, a->pm); + + fn(za, zn, zm, pn, pm, tcg_constant_i32(desc)); + + tcg_temp_free_ptr(za); + tcg_temp_free_ptr(zn); + tcg_temp_free_ptr(pn); + tcg_temp_free_ptr(pm); + return true; +} + static bool do_outprod_fpst(DisasContext *s, arg_op *a, MemOp esz, gen_helper_gvec_5_ptr *fn) { @@ -306,3 +332,6 @@ TRANS_FEAT(FMOPA_s, aa64_sme, do_outprod_fpst, a, MO_32, gen_helper_sme_fmopa_s) TRANS_FEAT(FMOPA_d, aa64_sme_f64f64, do_outprod_fpst, a, MO_64, gen_helper_sme_fmopa_d) + +/* TODO: FEAT_EBF16 */ +TRANS_FEAT(BFMOPA, aa64_sme, do_outprod, a, MO_32, gen_helper_sme_bfmopa) -- 2.34.1