linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Stanimir Varbanov <stanimir.varbanov@linaro.org>
To: Dikshita Agarwal <dikshita@codeaurora.org>, linux-media@vger.kernel.org
Cc: linux-kernel@vger.kernel.org, vgarodia@codeaurora.org
Subject: Re: [PATCH 1/7] venus: firmware: enable no tz fw loading for sc7280
Date: Mon, 31 May 2021 12:01:05 +0300	[thread overview]
Message-ID: <73299cba-3ea4-5cb9-45ee-228ef698a48d@linaro.org> (raw)
In-Reply-To: <1621417008-6117-2-git-send-email-dikshita@codeaurora.org>



On 5/19/21 12:36 PM, Dikshita Agarwal wrote:
> - Enable no tz FW loading.
> - add routine to reset XTSS.
> 
> Signed-off-by: Dikshita Agarwal <dikshita@codeaurora.org>
> ---
>  drivers/media/platform/qcom/venus/firmware.c     | 42 ++++++++++++++++++++----
>  drivers/media/platform/qcom/venus/hfi_venus_io.h |  2 ++
>  2 files changed, 38 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/media/platform/qcom/venus/firmware.c b/drivers/media/platform/qcom/venus/firmware.c
> index 227bd3b..e05e01a 100644
> --- a/drivers/media/platform/qcom/venus/firmware.c
> +++ b/drivers/media/platform/qcom/venus/firmware.c
> @@ -42,6 +42,22 @@ static void venus_reset_cpu(struct venus_core *core)
>  	writel(0, wrapper_base + WRAPPER_A9SS_SW_RESET);
>  }
>  
> +static void venus_reset_cpu_V6(struct venus_core *core)
> +{
> +	u32 fw_size = core->fw.mapped_mem_size;
> +	void __iomem *wrapper_tz_base = core->wrapper_tz_base;
> +
> +	writel(0, wrapper_tz_base + WRAPPER_FW_START_ADDR);
> +	writel(fw_size, wrapper_tz_base + WRAPPER_FW_END_ADDR);
> +	writel(0, wrapper_tz_base + WRAPPER_CPA_START_ADDR);
> +	writel(fw_size, wrapper_tz_base + WRAPPER_CPA_END_ADDR);
> +	writel(fw_size, wrapper_tz_base + WRAPPER_NONPIX_START_ADDR);
> +	writel(fw_size, wrapper_tz_base + WRAPPER_NONPIX_END_ADDR);
> +
> +	/* Bring XTSS out of reset */
> +	writel(0, wrapper_tz_base + WRAPPER_TZ_XTSS_SW_RESET);
> +}

Could you squash this v6 cpu reset into venus_reset_cpu() and add
IS_V6() checks there ...

> +
>  int venus_set_hw_state(struct venus_core *core, bool resume)
>  {
>  	int ret;
> @@ -54,10 +70,15 @@ int venus_set_hw_state(struct venus_core *core, bool resume)
>  	}
>  
>  	if (resume) {
> -		venus_reset_cpu(core);
> +		if (IS_V6(core))
> +			venus_reset_cpu_V6(core);
> +		else
> +			venus_reset_cpu(core);

... then this IS_V6() is not needed.

>  	} else {
>  		if (!IS_V6(core))
>  			writel(1, core->wrapper_base + WRAPPER_A9SS_SW_RESET);
> +		else
> +			writel(1, core->wrapper_tz_base + WRAPPER_TZ_XTSS_SW_RESET);

Could you invert the logic here:

if (IS_V6(core)
	writel(1, core->wrapper_tz_base + WRAPPER_TZ_XTSS_SW_RESET);
else
	writel(1, core->wrapper_base + WRAPPER_A9SS_SW_RESET);

>  	}
>  
>  	return 0;
> @@ -149,7 +170,10 @@ static int venus_boot_no_tz(struct venus_core *core, phys_addr_t mem_phys,
>  		return ret;
>  	}
>  
> -	venus_reset_cpu(core);
> +	if (IS_V6(core))
> +		venus_reset_cpu_V6(core);
> +	else
> +		venus_reset_cpu(core);
>  
>  	return 0;
>  }
> @@ -162,12 +186,18 @@ static int venus_shutdown_no_tz(struct venus_core *core)
>  	u32 reg;
>  	struct device *dev = core->fw.dev;
>  	void __iomem *wrapper_base = core->wrapper_base;
> +	void __iomem *wrapper_tz_base = core->wrapper_tz_base;
>  
> +	if (IS_V6(core)) {
> +		reg = readl_relaxed(wrapper_tz_base + WRAPPER_TZ_XTSS_SW_RESET);
> +		reg |= WRAPPER_XTSS_SW_RESET_BIT;
> +		writel_relaxed(reg, wrapper_tz_base + WRAPPER_TZ_XTSS_SW_RESET);
> +	} else {
>  	/* Assert the reset to ARM9 */

This comment should be moved above.

> -	reg = readl_relaxed(wrapper_base + WRAPPER_A9SS_SW_RESET);
> -	reg |= WRAPPER_A9SS_SW_RESET_BIT;
> -	writel_relaxed(reg, wrapper_base + WRAPPER_A9SS_SW_RESET);
> -
> +		reg = readl_relaxed(wrapper_base + WRAPPER_A9SS_SW_RESET);
> +		reg |= WRAPPER_A9SS_SW_RESET_BIT;
> +		writel_relaxed(reg, wrapper_base + WRAPPER_A9SS_SW_RESET);
> +	}
>  	/* Make sure reset is asserted before the mapping is removed */
>  	mb();
>  
> diff --git a/drivers/media/platform/qcom/venus/hfi_venus_io.h b/drivers/media/platform/qcom/venus/hfi_venus_io.h
> index 300c6e47..9735a24 100644
> --- a/drivers/media/platform/qcom/venus/hfi_venus_io.h
> +++ b/drivers/media/platform/qcom/venus/hfi_venus_io.h
> @@ -149,6 +149,8 @@
>  /* Wrapper TZ 6xx */
>  #define WRAPPER_TZ_BASE_V6			0x000c0000
>  #define WRAPPER_TZ_CPU_STATUS_V6		0x10
> +#define WRAPPER_TZ_XTSS_SW_RESET		0x1000
> +#define WRAPPER_XTSS_SW_RESET_BIT		BIT(0)
>  
>  /* Venus AON */
>  #define AON_BASE_V6				0x000e0000
> 

-- 
regards,
Stan

  reply	other threads:[~2021-05-31  9:01 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-19  9:36 [PATCH 0/7] media: venus: Enable venus support on sc7280 Dikshita Agarwal
2021-05-19  9:36 ` [PATCH 1/7] venus: firmware: enable no tz fw loading for sc7280 Dikshita Agarwal
2021-05-31  9:01   ` Stanimir Varbanov [this message]
2021-05-19  9:36 ` [PATCH 2/7] media: venus: core: Add sc7280 DT compatible and resource data Dikshita Agarwal
2021-05-31 11:18   ` Stanimir Varbanov
2021-05-19  9:36 ` [PATCH 3/7] media: venus: Add num_vpp_pipes to resource structure Dikshita Agarwal
2021-05-31 11:19   ` Stanimir Varbanov
2021-05-19  9:36 ` [PATCH 4/7] media: venus: hfi: Skip AON register programming for V6 1pipe Dikshita Agarwal
2021-05-31 11:19   ` Stanimir Varbanov
2021-05-19  9:36 ` [PATCH 5/7] venus: vdec: set work route to fw Dikshita Agarwal
2021-05-31 10:32   ` Stanimir Varbanov
2021-05-19  9:36 ` [PATCH 6/7] media: venus: helpers: update NUM_MBS macro calculation Dikshita Agarwal
2021-05-31 11:21   ` Stanimir Varbanov
2021-05-19  9:36 ` [PATCH 7/7] media: venus: Set buffer to FW based on FW min count requirement Dikshita Agarwal
2021-05-31 11:28   ` Stanimir Varbanov

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=73299cba-3ea4-5cb9-45ee-228ef698a48d@linaro.org \
    --to=stanimir.varbanov@linaro.org \
    --cc=dikshita@codeaurora.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-media@vger.kernel.org \
    --cc=vgarodia@codeaurora.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;
as well as URLs for NNTP newsgroup(s).