linux-kernel - Re: [PATCH v2 2/5] i3c: mipi-i3c-hci: Add a quirk to set PIO mode

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

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <a0f5ea6b-2be4-4fac-8c22-34fc21df84b2@linux.intel.com>
Date: Fri, 2 Aug 2024 16:58:32 +0300
From: Jarkko Nikula <jarkko.nikula@...ux.intel.com>
To: Shyam Sundar S K <Shyam-sundar.S-k@....com>,
 Alexandre Belloni <alexandre.belloni@...tlin.com>
Cc: Guruvendra Punugupati <Guruvendra.Punugupati@....com>,
 Krishnamoorthi M <krishnamoorthi.m@....com>, linux-i3c@...ts.infradead.org,
 linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2 2/5] i3c: mipi-i3c-hci: Add a quirk to set PIO mode

Hi

On 7/24/24 10:12 AM, Shyam Sundar S K wrote:
> The AMD HCI controller currently only supports PIO mode but exposes DMA
> rings to the OS, which leads to the controller being configured in DMA
> mode. To address this, add a quirk to avoid configuring the controller in
> DMA mode and default to PIO mode.
> 
> Additionally, introduce a generic quirk infrastructure to the mipi-i3c-hci
> driver to facilitate seamless future quirk additions.
> 
> Co-developed-by: Krishnamoorthi M <krishnamoorthi.m@....com>
> Signed-off-by: Krishnamoorthi M <krishnamoorthi.m@....com>
> Co-developed-by: Guruvendra Punugupati <Guruvendra.Punugupati@....com>
> Signed-off-by: Guruvendra Punugupati <Guruvendra.Punugupati@....com>
> Signed-off-by: Shyam Sundar S K <Shyam-sundar.S-k@....com>
> ---
>   drivers/i3c/master/mipi-i3c-hci/Makefile |  3 ++-
>   drivers/i3c/master/mipi-i3c-hci/core.c   | 15 ++++++++++++++-
>   drivers/i3c/master/mipi-i3c-hci/hci.h    |  3 +++
>   3 files changed, 19 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/i3c/master/mipi-i3c-hci/Makefile b/drivers/i3c/master/mipi-i3c-hci/Makefile
> index a658e7b8262c..1f8cd5c48fde 100644
> --- a/drivers/i3c/master/mipi-i3c-hci/Makefile
> +++ b/drivers/i3c/master/mipi-i3c-hci/Makefile
> @@ -3,4 +3,5 @@
>   obj-$(CONFIG_MIPI_I3C_HCI)		+= mipi-i3c-hci.o
>   mipi-i3c-hci-y				:= core.o ext_caps.o pio.o dma.o \
>   					   cmd_v1.o cmd_v2.o \
> -					   dat_v1.o dct_v1.o
> +					   dat_v1.o dct_v1.o \
> +					   hci_quirks.o

This doesn't build since hci_quirks.c is added by the patch 4/5. One 
idea below.

> diff --git a/drivers/i3c/master/mipi-i3c-hci/core.c b/drivers/i3c/master/mipi-i3c-hci/core.c
> index dbc8c38bd962..8bb422ab1d01 100644
> --- a/drivers/i3c/master/mipi-i3c-hci/core.c
> +++ b/drivers/i3c/master/mipi-i3c-hci/core.c
> @@ -33,6 +33,7 @@
>   #define reg_clear(r, v)		reg_write(r, reg_read(r) & ~(v))
>   
>   #define HCI_VERSION			0x00	/* HCI Version (in BCD) */
> +#define HCI_VERSION_V1			0x100	/* MIPI HCI Version number V1.0 */
>   
>   #define HC_CONTROL			0x04
>   #define HC_CONTROL_BUS_ENABLE		BIT(31)
> @@ -745,6 +746,14 @@ static int i3c_hci_init(struct i3c_hci *hci)
>   		return -EINVAL;
>   	}
>   
> +	/* Initialize quirks for AMD platforms */
> +	amd_i3c_hci_quirks_init(hci);
> +
> +	regval = reg_read(HCI_VERSION);
> +
> +	if (hci->quirks & HCI_QUIRK_AMD_PIO_MODE)
> +		hci->RHS_regs = NULL;
> +
>   	/* Try activating DMA operations first */
>   	if (hci->RHS_regs) {
>   		reg_clear(HC_CONTROL, HC_CONTROL_PIO_MODE);
> @@ -760,7 +769,11 @@ static int i3c_hci_init(struct i3c_hci *hci)
>   	/* If no DMA, try PIO */
>   	if (!hci->io && hci->PIO_regs) {
>   		reg_set(HC_CONTROL, HC_CONTROL_PIO_MODE);
> -		if (!(reg_read(HC_CONTROL) & HC_CONTROL_PIO_MODE)) {
> +		/*
> +		 * HC_CONTROL_PIO_MODE bit not present in HC_CONTROL register w.r.t V1.0
> +		 * specification. So skip checking PIO_MODE bit status
> +		 */
> +		if (regval != HCI_VERSION_V1 && !(reg_read(HC_CONTROL) & HC_CONTROL_PIO_MODE)) {
>   			dev_err(&hci->master.dev, "DMA mode is stuck\n");
>   			ret = -EIO;
>   		} else {

This is true, I see this now from pre-v1.0, v1.0. v1.1 and v1.2 specs 
too, HC_CONTROL_PIO_MODE bit is present only after v1.0. And therefore 
version != HCI_VERSION_V1 check is not fully correct since bit is not 
present in pre-v1.0 HW versions either.

I'd split this patch and do version check alone here (perhaps as a first 
patch) and do quirk stuff later where hci_quirks.c is added.