All of lore.kernel.org
 help / color / mirror / Atom feed
From: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
To: Bart Van Assche <bvanassche@acm.org>
Cc: "Martin K . Petersen" <martin.petersen@oracle.com>,
	linux-scsi@vger.kernel.org,
	"James E.J. Bottomley" <James.Bottomley@HansenPartnership.com>,
	Peter Wang <peter.wang@mediatek.com>,
	Minwoo Im <minwoo.im@samsung.com>,
	Stanley Jhu <chu.stanley@gmail.com>,
	ChanWoo Lee <cw9316.lee@samsung.com>,
	Yang Li <yang.lee@linux.alibaba.com>,
	"Bao D. Nguyen" <quic_nguyenb@quicinc.com>,
	Avri Altman <avri.altman@wdc.com>,
	Andrew Halaney <ahalaney@redhat.com>,
	Bean Huo <beanhuo@micron.com>,
	Maramaina Naresh <quic_mnaresh@quicinc.com>,
	Akinobu Mita <akinobu.mita@gmail.com>
Subject: Re: [PATCH v4 9/9] scsi: ufs: Make .get_hba_mac() optional
Date: Wed, 3 Jul 2024 18:52:02 +0530	[thread overview]
Message-ID: <20240703132202.GE3498@thinkpad> (raw)
In-Reply-To: <20240702204020.2489324-10-bvanassche@acm.org>

On Tue, Jul 02, 2024 at 01:39:17PM -0700, Bart Van Assche wrote:
> UFSHCI controllers that are compliant with the UFSHCI 4.0 standard report
> the maximum number of supported commands in the controller capabilities
> register. Use that value if .get_hba_mac == NULL.
> 
> Signed-off-by: Bart Van Assche <bvanassche@acm.org>
> ---
>  drivers/ufs/core/ufs-mcq.c     | 20 ++++++++++++++------
>  drivers/ufs/core/ufshcd-priv.h |  1 +
>  drivers/ufs/core/ufshcd.c      |  6 ++++--
>  include/ufs/ufshcd.h           |  4 +++-
>  include/ufs/ufshci.h           |  1 +
>  5 files changed, 23 insertions(+), 9 deletions(-)
> 
> diff --git a/drivers/ufs/core/ufs-mcq.c b/drivers/ufs/core/ufs-mcq.c
> index 0482c7a1e419..b2cf34a1fe48 100644
> --- a/drivers/ufs/core/ufs-mcq.c
> +++ b/drivers/ufs/core/ufs-mcq.c
> @@ -138,18 +138,21 @@ EXPORT_SYMBOL_GPL(ufshcd_mcq_queue_cfg_addr);
>   *
>   * MAC - Max. Active Command of the Host Controller (HC)
>   * HC wouldn't send more than this commands to the device.
> - * It is mandatory to implement get_hba_mac() to enable MCQ mode.
>   * Calculates and adjusts the queue depth based on the depth
>   * supported by the HC and ufs device.
>   */
>  int ufshcd_mcq_decide_queue_depth(struct ufs_hba *hba)
>  {
> -	int mac = -EOPNOTSUPP;
> +	int mac;
>  
> -	if (!hba->vops || !hba->vops->get_hba_mac)
> -		goto err;
> -
> -	mac = hba->vops->get_hba_mac(hba);
> +	if (!hba->vops || !hba->vops->get_hba_mac) {
> +		hba->capabilities =
> +			ufshcd_readl(hba, REG_CONTROLLER_CAPABILITIES);
> +		mac = hba->capabilities & MASK_TRANSFER_REQUESTS_SLOTS_MCQ;
> +		mac++;

Can you add a comment to state that the MAC value is 0 based?

> +	} else {
> +		mac = hba->vops->get_hba_mac(hba);
> +	}
>  	if (mac < 0)
>  		goto err;
>  
> @@ -423,6 +426,11 @@ void ufshcd_mcq_enable(struct ufs_hba *hba)
>  }
>  EXPORT_SYMBOL_GPL(ufshcd_mcq_enable);
>  
> +void ufshcd_mcq_disable(struct ufs_hba *hba)
> +{
> +	ufshcd_rmwl(hba, MCQ_MODE_SELECT, 0, REG_UFS_MEM_CFG);
> +}
> +
>  void ufshcd_mcq_config_esi(struct ufs_hba *hba, struct msi_msg *msg)
>  {
>  	ufshcd_writel(hba, msg->address_lo, REG_UFS_ESILBA);
> diff --git a/drivers/ufs/core/ufshcd-priv.h b/drivers/ufs/core/ufshcd-priv.h
> index 88ce93748305..ce36154ce963 100644
> --- a/drivers/ufs/core/ufshcd-priv.h
> +++ b/drivers/ufs/core/ufshcd-priv.h
> @@ -64,6 +64,7 @@ void ufshcd_auto_hibern8_update(struct ufs_hba *hba, u32 ahit);
>  void ufshcd_compl_one_cqe(struct ufs_hba *hba, int task_tag,
>  			  struct cq_entry *cqe);
>  int ufshcd_mcq_init(struct ufs_hba *hba);
> +void ufshcd_mcq_disable(struct ufs_hba *hba);
>  int ufshcd_mcq_decide_queue_depth(struct ufs_hba *hba);
>  int ufshcd_mcq_memory_alloc(struct ufs_hba *hba);
>  struct ufs_hw_queue *ufshcd_mcq_req_to_hwq(struct ufs_hba *hba,
> diff --git a/drivers/ufs/core/ufshcd.c b/drivers/ufs/core/ufshcd.c
> index b3444f9ce130..9e0290c6c2d3 100644
> --- a/drivers/ufs/core/ufshcd.c
> +++ b/drivers/ufs/core/ufshcd.c
> @@ -8753,13 +8753,15 @@ static int ufshcd_device_init(struct ufs_hba *hba, bool init_dev_params)
>  		if (ret)
>  			return ret;
>  		if (is_mcq_supported(hba) && !hba->scsi_host_added) {
> +			ufshcd_mcq_enable(hba);
> +			hba->mcq_enabled = true;

If the 'mcq_enabled' assignment goes hand in hand with
ufshcd_mcq_{enable/disable}, why shouldn't it be moved inside?

- Mani

>  			ret = ufshcd_alloc_mcq(hba);
>  			if (!ret) {
>  				ufshcd_config_mcq(hba);
> -				ufshcd_mcq_enable(hba);
> -				hba->mcq_enabled = true;
>  			} else {
>  				/* Continue with SDB mode */
> +				ufshcd_mcq_disable(hba);
> +				hba->mcq_enabled = false;
>  				use_mcq_mode = false;
>  				dev_err(hba->dev, "MCQ mode is disabled, err=%d\n",
>  					 ret);
> diff --git a/include/ufs/ufshcd.h b/include/ufs/ufshcd.h
> index c0e28a512b3c..6518997930f3 100644
> --- a/include/ufs/ufshcd.h
> +++ b/include/ufs/ufshcd.h
> @@ -325,7 +325,9 @@ struct ufs_pwr_mode_info {
>   * @event_notify: called to notify important events
>   * @reinit_notify: called to notify reinit of UFSHCD during max gear switch
>   * @mcq_config_resource: called to configure MCQ platform resources
> - * @get_hba_mac: called to get vendor specific mac value, mandatory for mcq mode
> + * @get_hba_mac: reports maximum number of outstanding commands supported by
> + *	the controller. Should be implemented for UFSHCI 4.0 or later
> + *	controllers that are not compliant with the UFSHCI 4.0 specification.
>   * @op_runtime_config: called to config Operation and runtime regs Pointers
>   * @get_outstanding_cqs: called to get outstanding completion queues
>   * @config_esi: called to config Event Specific Interrupt
> diff --git a/include/ufs/ufshci.h b/include/ufs/ufshci.h
> index 8d0cc73537c6..38fe97971a65 100644
> --- a/include/ufs/ufshci.h
> +++ b/include/ufs/ufshci.h
> @@ -68,6 +68,7 @@ enum {
>  /* Controller capability masks */
>  enum {
>  	MASK_TRANSFER_REQUESTS_SLOTS_SDB	= 0x0000001F,
> +	MASK_TRANSFER_REQUESTS_SLOTS_MCQ	= 0x000000FF,
>  	MASK_NUMBER_OUTSTANDING_RTT		= 0x0000FF00,
>  	MASK_TASK_MANAGEMENT_REQUEST_SLOTS	= 0x00070000,
>  	MASK_EHSLUTRD_SUPPORTED			= 0x00400000,

-- 
மணிவண்ணன் சதாசிவம்

  parent reply	other threads:[~2024-07-03 13:22 UTC|newest]

Thread overview: 25+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-07-02 20:39 [PATCH v4 0/9] UFS patches for kernel 6.11 Bart Van Assche
2024-07-02 20:39 ` [PATCH v4 1/9] scsi: ufs: Declare functions once Bart Van Assche
2024-07-02 20:39 ` [PATCH v4 2/9] scsi: ufs: Initialize struct uic_command once Bart Van Assche
2024-07-02 20:39 ` [PATCH v4 3/9] scsi: ufs: Remove two constants Bart Van Assche
2024-07-02 20:39 ` [PATCH v4 4/9] scsi: ufs: Rename the MASK_TRANSFER_REQUESTS_SLOTS constant Bart Van Assche
2024-07-03 12:59   ` Manivannan Sadhasivam
2024-07-02 20:39 ` [PATCH v4 5/9] scsi: ufs: Initialize hba->reserved_slot earlier Bart Van Assche
2024-07-03 13:01   ` Manivannan Sadhasivam
2024-07-02 20:39 ` [PATCH v4 6/9] scsi: ufs: Inline is_mcq_enabled() Bart Van Assche
2024-07-03  8:58   ` Peter Wang (王信友)
2024-07-03 13:03   ` Manivannan Sadhasivam
2024-07-02 20:39 ` [PATCH v4 7/9] scsi: ufs: Move the ufshcd_mcq_enable() call Bart Van Assche
2024-07-03  8:59   ` Peter Wang (王信友)
2024-07-03 13:10   ` Manivannan Sadhasivam
2024-07-03 20:32     ` Bart Van Assche
2024-07-02 20:39 ` [PATCH v4 8/9] scsi: ufs: Inline ufshcd_mcq_vops_get_hba_mac() Bart Van Assche
2024-07-02 20:39 ` [PATCH v4 9/9] scsi: ufs: Make .get_hba_mac() optional Bart Van Assche
2024-07-03  9:00   ` Peter Wang (王信友)
2024-07-03 13:22   ` Manivannan Sadhasivam [this message]
2024-07-03 20:36     ` Bart Van Assche
2024-07-06 16:33       ` Manivannan Sadhasivam
2024-07-07  3:24         ` Bart Van Assche
2024-07-08 10:44           ` Manivannan Sadhasivam
2024-07-08 17:10             ` Bart Van Assche
2024-07-08 18:06               ` Manivannan Sadhasivam

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20240703132202.GE3498@thinkpad \
    --to=manivannan.sadhasivam@linaro.org \
    --cc=James.Bottomley@HansenPartnership.com \
    --cc=ahalaney@redhat.com \
    --cc=akinobu.mita@gmail.com \
    --cc=avri.altman@wdc.com \
    --cc=beanhuo@micron.com \
    --cc=bvanassche@acm.org \
    --cc=chu.stanley@gmail.com \
    --cc=cw9316.lee@samsung.com \
    --cc=linux-scsi@vger.kernel.org \
    --cc=martin.petersen@oracle.com \
    --cc=minwoo.im@samsung.com \
    --cc=peter.wang@mediatek.com \
    --cc=quic_mnaresh@quicinc.com \
    --cc=quic_nguyenb@quicinc.com \
    --cc=yang.lee@linux.alibaba.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.