X86 platform drivers
 help / color / mirror / Atom feed
From: Hans de Goede <hdegoede@redhat.com>
To: Shyam Sundar S K <Shyam-sundar.S-k@amd.com>, markgross@kernel.org
Cc: Sanket.Goswami@amd.com, mario.limonciello@amd.com,
	platform-driver-x86@vger.kernel.org,
	Harsh Jain <Harsh.Jain@amd.com>
Subject: Re: [PATCH v4 2/2] platform/x86/amd/pmc: Add dump_custom_stb module parameter
Date: Mon, 18 Sep 2023 14:27:29 +0200	[thread overview]
Message-ID: <b3e4acdc-c031-cf00-b972-bc11e3933799@redhat.com> (raw)
In-Reply-To: <20230910142034.2967110-2-Shyam-sundar.S-k@amd.com>

Hi Shyam,

On 9/10/23 16:20, Shyam Sundar S K wrote:
> There have been instances when the default size (1M) of the STB is not
> sufficient to get the complete traces of the failure. In such scenarios
> we can use a module_param to enable full trace that shall contain more
> debugging data. This is not a regular case and hence not enabling this
> capability by default.
> 
> With this change, there will be two cases on how the driver fetches the
> stb data:
> 1) A special case (proposed now) - which is required only for certain
> platforms. Here, a new module param will be supplied to the driver that
> will have a special PMFW supporting enhanced dram sizes for getting
> the stb data. Without the special PMFW support, just setting the module
> param will not help to get the enhanced stb data.
> 
> 2) Current code branch which fetches the stb data based on the parameters
> like the num_samples, fsize and the r/w pointer.
> 
> Co-developed-by: Harsh Jain <Harsh.Jain@amd.com>
> Signed-off-by: Harsh Jain <Harsh.Jain@amd.com>
> Signed-off-by: Sanket Goswami <Sanket.Goswami@amd.com>
> Signed-off-by: Shyam Sundar S K <Shyam-sundar.S-k@amd.com>
> ---
> v3->v4:
>  - Update code branches and commit-msg as per Ilpo's remark.
> 
> v2->v3:
>  - no change
> 
> v1->v2:
>  - rebase to 'review-hans' branch
>  - drop 2/4 of v1
>    (https://patchwork.kernel.org/project/platform-driver-x86/list/?series=775324&state=%2A&archive=both)
> 
>  drivers/platform/x86/amd/pmc/pmc.c | 43 +++++++++++++++++++++---------
>  drivers/platform/x86/amd/pmc/pmc.h |  1 +
>  2 files changed, 32 insertions(+), 12 deletions(-)
> 
> diff --git a/drivers/platform/x86/amd/pmc/pmc.c b/drivers/platform/x86/amd/pmc/pmc.c
> index 443bb78ea5f4..7e907cb50787 100644
> --- a/drivers/platform/x86/amd/pmc/pmc.c
> +++ b/drivers/platform/x86/amd/pmc/pmc.c
> @@ -53,6 +53,7 @@
>  
>  /* STB Spill to DRAM Parameters */
>  #define S2D_TELEMETRY_BYTES_MAX		0x100000
> +#define S2D_TELEMETRY_FSIZE_MAX		0x200000
>  #define S2D_TELEMETRY_DRAMBYTES_MAX	0x1000000
>  
>  /* STB Spill to DRAM Message Definition */
> @@ -160,6 +161,10 @@ static bool disable_workarounds;
>  module_param(disable_workarounds, bool, 0644);
>  MODULE_PARM_DESC(disable_workarounds, "Disable workarounds for platform bugs");
>  
> +static bool dump_custom_stb;
> +module_param(dump_custom_stb, bool, 0644);
> +MODULE_PARM_DESC(dump_custom_stb, "Enable to dump full STB buffer");
> +
>  static struct amd_pmc_dev pmc;
>  static int amd_pmc_send_cmd(struct amd_pmc_dev *dev, u32 arg, u32 *data, u8 msg, bool ret);
>  static int amd_pmc_read_stb(struct amd_pmc_dev *dev, u32 *buf);
> @@ -239,7 +244,7 @@ static const struct file_operations amd_pmc_stb_debugfs_fops = {
>  static int amd_pmc_stb_debugfs_open_v2(struct inode *inode, struct file *filp)
>  {
>  	struct amd_pmc_dev *dev = filp->f_inode->i_private;
> -	u32 *buf, fsize, num_samples, val, stb_rdptr_offset = 0;
> +	u32 *buf, num_samples, val, stb_rdptr_offset = 0;
>  	int ret;
>  
>  	/* Write dummy postcode while reading the STB buffer */
> @@ -247,10 +252,6 @@ static int amd_pmc_stb_debugfs_open_v2(struct inode *inode, struct file *filp)
>  	if (ret)
>  		dev_err(dev->dev, "error writing to STB: %d\n", ret);
>  
> -	buf = kzalloc(S2D_TELEMETRY_BYTES_MAX, GFP_KERNEL);
> -	if (!buf)
> -		return -ENOMEM;
> -
>  	/* Spill to DRAM num_samples uses separate SMU message port */
>  	dev->msg_port = 1;
>  
> @@ -264,20 +265,36 @@ static int amd_pmc_stb_debugfs_open_v2(struct inode *inode, struct file *filp)
>  	dev->msg_port = 0;
>  	if (ret) {
>  		dev_err(dev->dev, "error: S2D_NUM_SAMPLES not supported : %d\n", ret);
> -		kfree(buf);
>  		return ret;
>  	}
>  
>  	/* Start capturing data from the last push location */
> -	if (num_samples > S2D_TELEMETRY_BYTES_MAX) {
> -		fsize  = S2D_TELEMETRY_BYTES_MAX;
> -		stb_rdptr_offset = num_samples - fsize;
> +	if (dump_custom_stb &&
> +	    (dev->dram_size - S2D_TELEMETRY_BYTES_MAX <= S2D_TELEMETRY_FSIZE_MAX)) {
> +		/*
> +		 * we have a custom stb size and the PMFW is supposed to give
> +		 * the enhanced dram size. Note that we land here only for the
> +		 * platforms that support enhanced dram size reporting.
> +		 */
> +		dev->fsize = dev->dram_size - S2D_TELEMETRY_BYTES_MAX;
> +		stb_rdptr_offset = 0;

I don't understand this part. Why is num_samples not taken into account
anymore ?  and why substract S2D_TELEMETRY_BYTES_MAX from dram_size ?



> +	} else if (num_samples > S2D_TELEMETRY_BYTES_MAX) {
> +		/*
> +		 * This is for general cases, where the stb limits are meant for
> +		 * standard usage
> +		 */
> +		dev->fsize  = S2D_TELEMETRY_BYTES_MAX;
> +		stb_rdptr_offset = num_samples - dev->fsize;

This assumes that num_samples is in the S2D_TELEMETRY_BYTES_MAX+1 .. 2*S2D_TELEMETRY_BYTES_MAX
rang, what if it is more ?

I think that what you want here is:

		stb_rdptr_offset = num_samples % S2D_TELEMETRY_BYTES_MAX;

Note this is a pre-existing problem but I just noticed this now.

Regards,

Hans

>  	} else {
> -		fsize = num_samples;
> +		dev->fsize = num_samples;
>  		stb_rdptr_offset = 0;
>  	}
>  
> -	memcpy_fromio(buf, dev->stb_virt_addr + stb_rdptr_offset, fsize);
> +	buf = kzalloc(dev->fsize, GFP_KERNEL);
> +	if (!buf)
> +		return -ENOMEM;
> +
> +	memcpy_fromio(buf, dev->stb_virt_addr + stb_rdptr_offset, dev->fsize);
>  	filp->private_data = buf;
>  
>  	return 0;
> @@ -286,11 +303,13 @@ static int amd_pmc_stb_debugfs_open_v2(struct inode *inode, struct file *filp)
>  static ssize_t amd_pmc_stb_debugfs_read_v2(struct file *filp, char __user *buf, size_t size,
>  					   loff_t *pos)
>  {
> +	struct amd_pmc_dev *dev = filp->f_inode->i_private;
> +
>  	if (!filp->private_data)
>  		return -EINVAL;
>  
>  	return simple_read_from_buffer(buf, size, pos, filp->private_data,
> -					S2D_TELEMETRY_BYTES_MAX);
> +					dev->fsize);
>  }
>  
>  static int amd_pmc_stb_debugfs_release_v2(struct inode *inode, struct file *filp)
> diff --git a/drivers/platform/x86/amd/pmc/pmc.h b/drivers/platform/x86/amd/pmc/pmc.h
> index c27bd6a5642f..f73d265430b8 100644
> --- a/drivers/platform/x86/amd/pmc/pmc.h
> +++ b/drivers/platform/x86/amd/pmc/pmc.h
> @@ -26,6 +26,7 @@ struct amd_pmc_dev {
>  	u32 dram_size;
>  	u32 num_ips;
>  	u32 s2d_msg_id;
> +	u32 fsize;
>  /* SMU version information */
>  	u8 smu_program;
>  	u8 major;


  reply	other threads:[~2023-09-18 12:29 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-09-10 14:20 [PATCH v4 1/2] platform/x86/amd/pmc: Add PMFW command id to support S2D force flush Shyam Sundar S K
2023-09-10 14:20 ` [PATCH v4 2/2] platform/x86/amd/pmc: Add dump_custom_stb module parameter Shyam Sundar S K
2023-09-18 12:27   ` Hans de Goede [this message]
2023-09-25 10:48     ` Shyam Sundar S K
2023-10-04  9:14       ` Hans de Goede
2023-10-04 12:11         ` Shyam Sundar S K
2023-10-04 15:10           ` Hans de Goede
2023-10-05 10:15             ` Ilpo Järvinen
2023-10-05 10:51               ` Hans de Goede
2023-10-06 10:27                 ` Ilpo Järvinen
2023-10-09  5:57                   ` Shyam Sundar S K
2023-10-09  9:25         ` Shyam Sundar S K

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=b3e4acdc-c031-cf00-b972-bc11e3933799@redhat.com \
    --to=hdegoede@redhat.com \
    --cc=Harsh.Jain@amd.com \
    --cc=Sanket.Goswami@amd.com \
    --cc=Shyam-sundar.S-k@amd.com \
    --cc=mario.limonciello@amd.com \
    --cc=markgross@kernel.org \
    --cc=platform-driver-x86@vger.kernel.org \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox