| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <CAPDyKFpt-ZmSGUWBukLvYvY6DexOr6g23FMWdY14d3gBKxzAmA@mail.gmail.com>
Date: Tue, 3 Oct 2023 13:09:25 +0200
From: Ulf Hansson <ulf.hansson@...aro.org>
To: Victor Shih <victorshihgli@...il.com>
Cc: adrian.hunter@...el.com, linux-mmc@...r.kernel.org,
linux-kernel@...r.kernel.org, benchuanggli@...il.com,
HL.Liu@...esyslogic.com.tw, Greg.tu@...esyslogic.com.tw,
takahiro.akashi@...aro.org, dlunev@...omium.org,
Ben Chuang <ben.chuang@...esyslogic.com.tw>,
Victor Shih <victor.shih@...esyslogic.com.tw>
Subject: Re: [PATCH V12 15/23] mmc: sdhci-uhs2: add detect_init() to detect
the interface
On Fri, 15 Sept 2023 at 11:44, Victor Shih <victorshihgli@...il.com> wrote:
>
> From: Victor Shih <victor.shih@...esyslogic.com.tw>
>
> Sdhci_uhs2_do_detect_init() is a sdhci version of mmc's uhs2_detect_init
> operation. After detected, the host's UHS-II capabilities will be set up
> here and interrupts will also be enabled.
$subject patch is adding a bunch of static functions, which isn't
really being used until later. If you compile this patch it will
trigger warnings about unused function, we don't want that. Each patch
in the series should build nicely without warning and errors.
To deal with these problems, I suggest that you move the introduction
of the sdhci_uhs2_control() from patch17 to $subject patch - or
possibly make that as a standalone patch, preceeding $subject patch.
Step by step you can then add support for each of the "enum
sd_uhs2_operation" to sdhci_uhs2_control().
Moreover, please work at the commit message a bit, it's not entirely
easy to understand by reading what goes on here.
>
> Signed-off-by: Ben Chuang <ben.chuang@...esyslogic.com.tw>
> Signed-off-by: AKASHI Takahiro <takahiro.akashi@...aro.org>
> Signed-off-by: Victor Shih <victor.shih@...esyslogic.com.tw>
> Acked-by: Adrian Hunter <adrian.hunter@...el.com>
> ---
>
> Updates in V8:
> - usleep_range() to instead of udelay() in sdhci_uhs2_interface_detect().
> - read_poll_timeout() to instead of read_poll_timeout_atomic()
> in sdhci_uhs2_interface_detect().
> - Modify return value in sdhci_uhs2_do_detect_init().
>
> Updates in V7:
> - Drop using uhs2_reset ops and use sdhci_uhs2_reset()
> in sdhci_uhs2_do_detect_init().
>
> Updates in V6:
> - Remove unnecessary functions.
> - Wrap at 100 columns in some functions.
>
> ---
>
> drivers/mmc/host/sdhci-uhs2.c | 112 ++++++++++++++++++++++++++++++++++
> 1 file changed, 112 insertions(+)
>
> diff --git a/drivers/mmc/host/sdhci-uhs2.c b/drivers/mmc/host/sdhci-uhs2.c
> index ad791c48f681..4c2a56629ab3 100644
> --- a/drivers/mmc/host/sdhci-uhs2.c
> +++ b/drivers/mmc/host/sdhci-uhs2.c
> @@ -335,6 +335,118 @@ static int sdhci_uhs2_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
> * *
> \*****************************************************************************/
>
> +static int sdhci_uhs2_interface_detect(struct sdhci_host *host)
> +{
> + int timeout = 100000; /* 100ms */
Please use define instead.
> + u32 val;
> +
> + usleep_range(50, 200); /* wait for 50us - 200us before check */
Why? Comment?
And use defines.
> +
> + if (read_poll_timeout(sdhci_readl, val, (val & SDHCI_UHS2_IF_DETECT),
> + 100, timeout, true, host, SDHCI_PRESENT_STATE)) {
> + pr_warn("%s: not detect UHS2 interface in 100ms.\n", mmc_hostname(host->mmc));
> + sdhci_dumpregs(host);
> + return -EIO;
> + }
> +
> + /* Enable UHS2 error interrupts */
> + sdhci_uhs2_clear_set_irqs(host, SDHCI_INT_ALL_MASK, SDHCI_UHS2_INT_ERROR_MASK);
> +
> + /* 150ms */
> + timeout = 150000;
Ditto.
> + if (read_poll_timeout(sdhci_readl, val, (val & SDHCI_UHS2_LANE_SYNC),
> + 100, timeout, true, host, SDHCI_PRESENT_STATE)) {
> + pr_warn("%s: UHS2 Lane sync fail in 150ms.\n", mmc_hostname(host->mmc));
> + sdhci_dumpregs(host);
> + return -EIO;
> + }
> +
> + DBG("%s: UHS2 Lane synchronized in UHS2 mode, PHY is initialized.\n",
> + mmc_hostname(host->mmc));
> + return 0;
> +}
> +
> +static int sdhci_uhs2_init(struct sdhci_host *host)
> +{
> + u16 caps_ptr = 0;
> + u32 caps_gen = 0;
> + u32 caps_phy = 0;
> + u32 caps_tran[2] = {0, 0};
> + struct mmc_host *mmc = host->mmc;
> +
> + caps_ptr = sdhci_readw(host, SDHCI_UHS2_CAPS_PTR);
> + if (caps_ptr < 0x100 || caps_ptr > 0x1FF) {
> + pr_err("%s: SDHCI_UHS2_CAPS_PTR(%d) is wrong.\n",
> + mmc_hostname(mmc), caps_ptr);
> + return -ENODEV;
> + }
> + caps_gen = sdhci_readl(host, caps_ptr + SDHCI_UHS2_CAPS_OFFSET);
> + caps_phy = sdhci_readl(host, caps_ptr + SDHCI_UHS2_CAPS_PHY_OFFSET);
> + caps_tran[0] = sdhci_readl(host, caps_ptr + SDHCI_UHS2_CAPS_TRAN_OFFSET);
> + caps_tran[1] = sdhci_readl(host, caps_ptr + SDHCI_UHS2_CAPS_TRAN_1_OFFSET);
> +
> + /* General Caps */
> + mmc->uhs2_caps.dap = caps_gen & SDHCI_UHS2_CAPS_DAP_MASK;
> + mmc->uhs2_caps.gap = FIELD_GET(SDHCI_UHS2_CAPS_GAP_MASK, caps_gen);
> + mmc->uhs2_caps.n_lanes = FIELD_GET(SDHCI_UHS2_CAPS_LANE_MASK, caps_gen);
> + mmc->uhs2_caps.addr64 = (caps_gen & SDHCI_UHS2_CAPS_ADDR_64) ? 1 : 0;
> + mmc->uhs2_caps.card_type = FIELD_GET(SDHCI_UHS2_CAPS_DEV_TYPE_MASK, caps_gen);
> +
> + /* PHY Caps */
> + mmc->uhs2_caps.phy_rev = caps_phy & SDHCI_UHS2_CAPS_PHY_REV_MASK;
> + mmc->uhs2_caps.speed_range = FIELD_GET(SDHCI_UHS2_CAPS_PHY_RANGE_MASK, caps_phy);
> + mmc->uhs2_caps.n_lss_sync = FIELD_GET(SDHCI_UHS2_CAPS_PHY_N_LSS_SYN_MASK, caps_phy);
> + mmc->uhs2_caps.n_lss_dir = FIELD_GET(SDHCI_UHS2_CAPS_PHY_N_LSS_DIR_MASK, caps_phy);
> + if (mmc->uhs2_caps.n_lss_sync == 0)
> + mmc->uhs2_caps.n_lss_sync = 16 << 2;
> + else
> + mmc->uhs2_caps.n_lss_sync <<= 2;
> + if (mmc->uhs2_caps.n_lss_dir == 0)
> + mmc->uhs2_caps.n_lss_dir = 16 << 3;
> + else
> + mmc->uhs2_caps.n_lss_dir <<= 3;
> +
> + /* LINK/TRAN Caps */
> + mmc->uhs2_caps.link_rev = caps_tran[0] & SDHCI_UHS2_CAPS_TRAN_LINK_REV_MASK;
> + mmc->uhs2_caps.n_fcu = FIELD_GET(SDHCI_UHS2_CAPS_TRAN_N_FCU_MASK, caps_tran[0]);
> + if (mmc->uhs2_caps.n_fcu == 0)
> + mmc->uhs2_caps.n_fcu = 256;
> + mmc->uhs2_caps.host_type = FIELD_GET(SDHCI_UHS2_CAPS_TRAN_HOST_TYPE_MASK, caps_tran[0]);
> + mmc->uhs2_caps.maxblk_len = FIELD_GET(SDHCI_UHS2_CAPS_TRAN_BLK_LEN_MASK, caps_tran[0]);
> + mmc->uhs2_caps.n_data_gap = caps_tran[1] & SDHCI_UHS2_CAPS_TRAN_1_N_DATA_GAP_MASK;
> +
> + return 0;
> +}
> +
> +static int sdhci_uhs2_do_detect_init(struct mmc_host *mmc)
> +{
> + struct sdhci_host *host = mmc_priv(mmc);
> +
> + DBG("Begin do uhs2 detect init.\n");
> +
> + if (sdhci_uhs2_interface_detect(host)) {
> + pr_warn("%s: cannot detect UHS2 interface.\n", mmc_hostname(host->mmc));
Does this really deserve a warning to be printed to the log?
> + return -EIO;
> + }
> +
> + if (sdhci_uhs2_init(host)) {
> + pr_warn("%s: UHS2 init fail.\n", mmc_hostname(host->mmc));
> + return -EIO;
> + }
> +
> + /* Init complete, do soft reset and enable UHS2 error irqs. */
> + sdhci_uhs2_reset(host, SDHCI_UHS2_SW_RESET_SD);
> + sdhci_uhs2_clear_set_irqs(host, SDHCI_INT_ALL_MASK, SDHCI_UHS2_INT_ERROR_MASK);
> + /*
> + * N.B SDHCI_INT_ENABLE and SDHCI_SIGNAL_ENABLE was cleared
> + * by SDHCI_UHS2_SW_RESET_SD
> + */
> + sdhci_writel(host, host->ier, SDHCI_INT_ENABLE);
> + sdhci_writel(host, host->ier, SDHCI_SIGNAL_ENABLE);
> +
> + return 0;
> +}
> +
> static int sdhci_uhs2_host_ops_init(struct sdhci_host *host)
> {
> host->mmc_host_ops.start_signal_voltage_switch =
Kind regards
Uffe
Powered by blists - more mailing lists