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 phobos.denx.de (phobos.denx.de [85.214.62.61]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id E40CCC4332F for ; Sat, 15 Oct 2022 13:29:28 +0000 (UTC) Received: from h2850616.stratoserver.net (localhost [IPv6:::1]) by phobos.denx.de (Postfix) with ESMTP id 41AE284F06; Sat, 15 Oct 2022 15:29:26 +0200 (CEST) Authentication-Results: phobos.denx.de; dmarc=pass (p=none dis=none) header.from=kernel.org Authentication-Results: phobos.denx.de; spf=pass smtp.mailfrom=u-boot-bounces@lists.denx.de Authentication-Results: phobos.denx.de; dkim=pass (2048-bit key; unprotected) header.d=kernel.org header.i=@kernel.org header.b="d4ceG46f"; dkim-atps=neutral Received: by phobos.denx.de (Postfix, from userid 109) id 8462084F1E; Sat, 15 Oct 2022 15:29:24 +0200 (CEST) Received: from dfw.source.kernel.org (dfw.source.kernel.org [139.178.84.217]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by phobos.denx.de (Postfix) with ESMTPS id 48EE884EE0 for ; Sat, 15 Oct 2022 15:29:21 +0200 (CEST) Authentication-Results: phobos.denx.de; dmarc=pass (p=none dis=none) header.from=kernel.org Authentication-Results: phobos.denx.de; spf=pass smtp.mailfrom=rogerq@kernel.org Received: from smtp.kernel.org (relay.kernel.org [52.25.139.140]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by dfw.source.kernel.org (Postfix) with ESMTPS id D31E960C27; Sat, 15 Oct 2022 13:29:19 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id 498C8C433D6; Sat, 15 Oct 2022 13:29:18 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1665840559; bh=DfpQpOgCOUsAoCGSX3wYwhHlDL7OtCJ2JykOPlCl/G8=; h=Date:Subject:To:Cc:References:From:In-Reply-To:From; b=d4ceG46ftnlBVEL6ZESKzpiQaLnbIRYCPDn87eoLl0Rx0NOMMvij0CdVZnypcsfWz TcQdCgAIirBpizPfrviFGCiu30oRpC79ZCKXtoC9FrD3CdGhEwI5jL3sA0TLGTswMy xecfW6adUCSp3XIF0AK3cLCTiRYAx+ncqroYRVo5DQt+tt+hqwxJ2mzQmTMe7+WdQx FAICS389znx2TMc/udSLlUY0hCE2/wXbjARxabNL/ihDVE67SetDvXQnMioh/eUrNA wHPH3jCS7EeczZL4WoY4SSuChU2tj0dx57UCrRLsN8QqPJiBv2Pr9/TuWkXZwFtPoF tfMofl5L1qJug== Message-ID: <32f321a3-35ad-fed4-12d8-a98877286cd9@kernel.org> Date: Sat, 15 Oct 2022 16:29:15 +0300 MIME-Version: 1.0 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101 Thunderbird/91.9.1 Subject: Re: [u-boot][PATCH 04/14] mtd: rawnand: omap_gpmc: Optimize NAND reads Content-Language: en-US To: Michael Nazzareno Trimarchi Cc: dario.binacchi@amarulasolutions.com, trini@konsulko.com, u-boot@lists.denx.de References: <20221011115012.6181-1-rogerq@kernel.org> <20221011115012.6181-5-rogerq@kernel.org> From: Roger Quadros In-Reply-To: Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit X-BeenThere: u-boot@lists.denx.de X-Mailman-Version: 2.1.39 Precedence: list List-Id: U-Boot discussion List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: u-boot-bounces@lists.denx.de Sender: "U-Boot" X-Virus-Scanned: clamav-milter 0.103.6 at phobos.denx.de X-Virus-Status: Clean Hi Michael, On 15/10/2022 10:24, Michael Nazzareno Trimarchi wrote: > Hi > > On Tue, Oct 11, 2022 at 1:50 PM Roger Quadros wrote: >> >> Rename omap_nand_read() to omap_nand_read_buf() to reflect >> actual behaviour. >> >> Use FIFO read address instead of raw read address for reads. >> >> The GPMC automatically converts 32-bit/16-bit reads to NAND >> device specific reads (8/16 bit). Use the largest possible >> read granularity size for more efficient reads. >> >> Signed-off-by: Roger Quadros >> --- >> drivers/mtd/nand/raw/omap_gpmc.c | 49 ++++++++++++++++++-------------- >> 1 file changed, 28 insertions(+), 21 deletions(-) >> >> diff --git a/drivers/mtd/nand/raw/omap_gpmc.c b/drivers/mtd/nand/raw/omap_gpmc.c >> index d62c3e6fce..b36fe762b3 100644 >> --- a/drivers/mtd/nand/raw/omap_gpmc.c >> +++ b/drivers/mtd/nand/raw/omap_gpmc.c >> @@ -55,6 +55,7 @@ struct omap_nand_info { >> enum omap_ecc ecc_scheme; >> uint8_t cs; >> uint8_t ws; /* wait status pin (0,1) */ >> + void __iomem *fifo; >> }; >> >> /* We are wasting a bit of memory but al least we are safe */ >> @@ -350,6 +351,20 @@ static int omap_calculate_ecc(struct mtd_info *mtd, const uint8_t *dat, >> return 0; >> } >> >> +static inline void omap_nand_read_buf(struct mtd_info *mtd, uint8_t *buf, int len) >> +{ >> + struct nand_chip *chip = mtd_to_nand(mtd); >> + struct omap_nand_info *info = nand_get_controller_data(chip); >> + u32 alignment = ((uintptr_t)buf | len) & 3; >> + >> + if (alignment & 1) >> + readsb(info->fifo, buf, len); >> + else if (alignment & 3) >> + readsw(info->fifo, buf, len >> 1); >> + else >> + readsl(info->fifo, buf, len >> 2); >> +} >> + >> #ifdef CONFIG_NAND_OMAP_GPMC_PREFETCH >> >> #define PREFETCH_CONFIG1_CS_SHIFT 24 >> @@ -415,7 +430,7 @@ static int __read_prefetch_aligned(struct nand_chip *chip, uint32_t *buf, int le >> cnt = PREFETCH_STATUS_FIFO_CNT(cnt); >> >> for (i = 0; i < cnt / 4; i++) { >> - *buf++ = readl(CONFIG_SYS_NAND_BASE); >> + *buf++ = readl(info->fifo); >> len -= 4; >> } >> } while (len); >> @@ -425,16 +440,6 @@ static int __read_prefetch_aligned(struct nand_chip *chip, uint32_t *buf, int le >> return 0; >> } >> >> -static inline void omap_nand_read(struct mtd_info *mtd, uint8_t *buf, int len) >> -{ >> - struct nand_chip *chip = mtd_to_nand(mtd); >> - >> - if (chip->options & NAND_BUSWIDTH_16) >> - nand_read_buf16(mtd, buf, len); >> - else >> - nand_read_buf(mtd, buf, len); >> -} >> - >> static void omap_nand_read_prefetch(struct mtd_info *mtd, uint8_t *buf, int len) >> { >> int ret; >> @@ -447,7 +452,7 @@ static void omap_nand_read_prefetch(struct mtd_info *mtd, uint8_t *buf, int len) >> */ >> head = ((uintptr_t)buf) % 4; >> if (head) { >> - omap_nand_read(mtd, buf, head); >> + omap_nand_read_buf(mtd, buf, head); >> buf += head; >> len -= head; >> } >> @@ -461,10 +466,10 @@ static void omap_nand_read_prefetch(struct mtd_info *mtd, uint8_t *buf, int len) >> ret = __read_prefetch_aligned(chip, (uint32_t *)buf, len - tail); >> if (ret < 0) { >> /* fallback in case the prefetch engine is busy */ >> - omap_nand_read(mtd, buf, len); >> + omap_nand_read_buf(mtd, buf, len); >> } else if (tail) { >> buf += len - tail; >> - omap_nand_read(mtd, buf, tail); >> + omap_nand_read_buf(mtd, buf, tail); >> } >> } >> #endif /* CONFIG_NAND_OMAP_GPMC_PREFETCH */ >> @@ -1001,6 +1006,8 @@ int board_nand_init(struct nand_chip *nand) >> int32_t gpmc_config = 0; >> int cs = cs_next++; >> int err = 0; >> + struct omap_nand_info *info; >> + >> /* >> * xloader/Uboot's gpmc configuration would have configured GPMC for >> * nand type of memory. The following logic scans and latches on to the >> @@ -1029,9 +1036,12 @@ int board_nand_init(struct nand_chip *nand) >> >> nand->IO_ADDR_R = (void __iomem *)&gpmc_cfg->cs[cs].nand_dat; >> nand->IO_ADDR_W = (void __iomem *)&gpmc_cfg->cs[cs].nand_cmd; >> - omap_nand_info[cs].control = NULL; >> - omap_nand_info[cs].cs = cs; >> - omap_nand_info[cs].ws = wscfg[cs]; >> + >> + info = &omap_nand_info[cs]; >> + info->control = NULL; >> + info->cs = cs; >> + info->ws = wscfg[cs]; >> + info->fifo = (void __iomem *)CONFIG_SYS_NAND_BASE; >> nand_set_controller_data(nand, &omap_nand_info[cs]); >> nand->cmd_ctrl = omap_nand_hwcontrol; >> nand->options |= NAND_NO_PADDING | NAND_CACHEPRG; >> @@ -1062,10 +1072,7 @@ int board_nand_init(struct nand_chip *nand) >> #ifdef CONFIG_NAND_OMAP_GPMC_PREFETCH >> nand->read_buf = omap_nand_read_prefetch; >> #else >> - if (nand->options & NAND_BUSWIDTH_16) >> - nand->read_buf = nand_read_buf16; >> - else >> - nand->read_buf = nand_read_buf; >> + nand->read_buf = omap_nand_read_buf; >> #endif >> >> nand->dev_ready = omap_dev_ready; >> -- >> 2.17.1 >> > > Is possible for you to split in use of fifo and optimize read/alignment Sure, I'll remember to do that when I re-spin this series. > > Otherwise > > Reviewed-by: Michael Trimarchi > > Thanks! -- cheers, -roger