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 65687C433F5 for ; Tue, 15 Feb 2022 16:22:06 +0000 (UTC) Received: from localhost ([::1]:53630 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1nK0aa-0001kT-3e for qemu-devel@archiver.kernel.org; Tue, 15 Feb 2022 11:22:05 -0500 Received: from eggs.gnu.org ([209.51.188.92]:50358) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1nK0Z0-0000n9-G3; Tue, 15 Feb 2022 11:20:26 -0500 Received: from gloria.sntech.de ([185.11.138.130]:54332) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1nK0Yx-0000cr-1S; Tue, 15 Feb 2022 11:20:26 -0500 Received: from ip5b412258.dynamic.kabel-deutschland.de ([91.65.34.88] helo=diego.localnet) by gloria.sntech.de with esmtpsa (TLS1.3:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1nK0Yr-0000cd-9w; Tue, 15 Feb 2022 17:20:17 +0100 From: Heiko =?ISO-8859-1?Q?St=FCbner?= To: qemu-devel@nongnu.org Cc: Alistair Francis , Bin Meng , Atish Patra , Palmer Dabbelt , qemu-riscv@nongnu.org, Atish Patra Subject: Re: [PATCH] target/riscv: Add isa extenstion strings to the device tree Date: Tue, 15 Feb 2022 17:20:16 +0100 Message-ID: <2828317.Et2jP947se@diego> In-Reply-To: <20220215090530.911828-1-atishp@rivosinc.com> References: <20220215090530.911828-1-atishp@rivosinc.com> MIME-Version: 1.0 Content-Transfer-Encoding: 7Bit Content-Type: text/plain; charset="us-ascii" Received-SPF: none client-ip=185.11.138.130; envelope-from=heiko@sntech.de; helo=gloria.sntech.de X-Spam_score_int: -18 X-Spam_score: -1.9 X-Spam_bar: - X-Spam_report: (-1.9 / 5.0 requ) BAYES_00=-1.9, SPF_NONE=0.001, T_SCC_BODY_TEXT_LINE=-0.01, T_SPF_HELO_TEMPERROR=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" Am Dienstag, 15. Februar 2022, 10:05:30 CET schrieb Atish Patra: > Append the available ISA extensions to the "riscv,isa" string if it > is enabled so that kernel can process it. > > Signed-off-by: Atish Patra > --- > target/riscv/cpu.c | 23 ++++++++++++++++++++++- > 1 file changed, 22 insertions(+), 1 deletion(-) > > diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c > index b0a40b83e7a8..c70260d0df15 100644 > --- a/target/riscv/cpu.c > +++ b/target/riscv/cpu.c > @@ -34,6 +34,9 @@ > > /* RISC-V CPU definitions */ > > +/* This includes the null terminated character '\0' */ > +#define MAX_ISA_EXT_LEN 256 > + > static const char riscv_exts[26] = "IEMAFDQCLBJTPVNSUHKORWXYZG"; > > const char * const riscv_int_regnames[] = { > @@ -881,10 +884,26 @@ static void riscv_cpu_class_init(ObjectClass *c, void *data) > device_class_set_props(dc, riscv_cpu_properties); > } > > +static void riscv_isa_string_ext(RISCVCPU *cpu, char *isa_str, int max_str_len) > +{ > + int offset = strnlen(isa_str, max_str_len); > + > + if (cpu->cfg.ext_svpbmt) { > + offset += snprintf(isa_str + offset, max_str_len, "_%s", "_svpbmt"); > + } > + if ((offset < max_str_len) && cpu->cfg.ext_svinval) { shouldn't offset + strlen("svinval") +1 be < max_str_len? snprintf will write partial strings but this would throw off a qemu client completely I guess. > + offset += snprintf(isa_str + offset, max_str_len, "_%s", "_svinval"); > + } > + if ((offset < max_str_len) && (cpu->cfg.ext_svnapot)) { > + offset += snprintf(isa_str + offset, max_str_len, "_%s", "_svnapot"); > + } wouldn't it make more sense to do something like: + struct { + const char *value; + bool enabled; + } extensions[] = { + { "svpbmt", cpu->cfg.ext_svpbmt }, + { "svinval", cpu->cfg.ext_svinval }, + { "svnapot", cpu->cfg.ext_svnapot }, + }; + int i; + + for (i = 0; i < ARRAY_SIZE(extensions); i++) { + if (!extensions[i].enabled) + continue; + + /* check available space */ + if (offset + strlen(extensions[i].value) + 1 > max_str_len) { + //do warn about exceeded length + return; + } + + offset += snprintf(isa_str + offset, max_str_len, "_%s", + extensions[i].value); + } instead? Because that list will get longer over time and repeating checks and snprintf calls will get harder to keep in sync over time? Heiko > +} > + > char *riscv_isa_string(RISCVCPU *cpu) > { > int i; > - const size_t maxlen = sizeof("rv128") + sizeof(riscv_exts) + 1; > + const size_t maxlen = sizeof("rv128") + sizeof(riscv_exts) + > + MAX_ISA_EXT_LEN; > char *isa_str = g_new(char, maxlen); > char *p = isa_str + snprintf(isa_str, maxlen, "rv%d", TARGET_LONG_BITS); > for (i = 0; i < sizeof(riscv_exts); i++) { > @@ -893,6 +912,8 @@ char *riscv_isa_string(RISCVCPU *cpu) > } > } > *p = '\0'; > + riscv_isa_string_ext(cpu, isa_str, maxlen); > + > return isa_str; > } > >