Devicetree
 help / color / mirror / Atom feed
From: Conor Dooley <conor@kernel.org>
To: Mina Hui-Min Chou <minachou@andestech.com>
Cc: pjw@kernel.org, palmer@dabbelt.com, aou@eecs.berkeley.edu,
	alex@ghiti.fr, jic23@kernel.org, robh@kernel.org,
	krzk+dt@kernel.org, conor+dt@kernel.org, geert+renesas@glider.be,
	magnus.damm@gmail.com, prabhakar.mahadev-lad.rj@bp.renesas.com,
	linux-riscv@lists.infradead.org, devicetree@vger.kernel.org,
	linux-renesas-soc@vger.kernel.org, linux-kernel@vger.kernel.org,
	randolph@andestech.com, cyyang772@andestech.com,
	tim609@andestech.com, az70021@gmail.com
Subject: Re: [PATCH v2 17/19] riscv: errata: andes: generalize cache errata for Andes cores
Date: Fri, 13 Nov 2026 13:59:11 +0000	[thread overview]
Message-ID: <20261113-voice-pledge-d0b4e1420676@spud> (raw)
In-Reply-To: <20261113071143.1446167-18-minachou@andestech.com>

[-- Attachment #1: Type: text/plain, Size: 4292 bytes --]

On Fri, Nov 13, 2026 at 03:11:41PM +0800, Mina Hui-Min Chou wrote:
> The Andes cache errata probe only runs on AX45MP: it matches
> ANDES_AX45MP_MARCHID/MIMPID and then asks the Andes SBI extension
> whether IOCP is missing. Other Andes cores that need CCTL-based cache
> maintenance, such as the A27L2, which has neither IOCP nor Zicbom, are
> silently skipped.
> 
> The IOCP_SW_WORKAROUND SBI call cannot simply be used on every Andes
> core: OpenSBI checks the IOCP bit in mmsc_cfg[47], which only exists on
> RV64 cores. RV32 cores such as the A45MP report IOCP in mmsc_cfg2
> instead, and cores such as the A25/A27 do not support IOCP at all.
> 
> Keep the existing SBI-based check for AX45MP. On other Andes cores, use
> the Andes CCTL operations for cache maintenance when the core does not
> implement Zicbom.
> 
> As the probe is no longer only about IOCP, rename errata_probe_iocp()
> to errata_probe_andes_cache(), ERRATA_ANDES_NO_IOCP to
> ERRATA_ANDES_CACHE, and the ERRATA_ANDES prompt from "Andes AX45MP
> errata" to "Andes errata".
> 
> Signed-off-by: Mina Hui-Min Chou <minachou@andestech.com>
> ---
>  arch/riscv/Kconfig.errata                    |  2 +-
>  arch/riscv/errata/andes/errata.c             | 13 +++++++------
>  arch/riscv/include/asm/errata_list_vendors.h |  2 +-
>  3 files changed, 9 insertions(+), 8 deletions(-)
> 
> diff --git a/arch/riscv/Kconfig.errata b/arch/riscv/Kconfig.errata
> index 3c43e065aa01..ca8e52bee497 100644
> --- a/arch/riscv/Kconfig.errata
> +++ b/arch/riscv/Kconfig.errata
> @@ -1,7 +1,7 @@
>  menu "CPU errata selection"
>  
>  config ERRATA_ANDES
> -	bool "Andes AX45MP errata"
> +	bool "Andes errata"
>  	depends on RISCV_SBI
>  	help
>  	  All Andes errata Kconfig depend on this Kconfig. Disabling
> diff --git a/arch/riscv/errata/andes/errata.c b/arch/riscv/errata/andes/errata.c
> index dcc9d1ee5ffd..530d2df216e0 100644
> --- a/arch/riscv/errata/andes/errata.c
> +++ b/arch/riscv/errata/andes/errata.c
> @@ -39,7 +39,7 @@ static long ax45mp_iocp_sw_workaround(void)
>  	return ret.error ? 0 : ret.value;
>  }
>  
> -static void errata_probe_iocp(unsigned int stage, unsigned long arch_id, unsigned long impid)
> +static void errata_probe_andes_cache(unsigned int stage, unsigned long arch_id, unsigned long impid)

Just call it probe so that you dont rename it again when the next
erratum shows up?

>  {
>  	static bool done;
>  
> @@ -51,11 +51,12 @@ static void errata_probe_iocp(unsigned int stage, unsigned long arch_id, unsigne
>  
>  	done = true;
>  
> -	if (arch_id != ANDES_AX45MP_MARCHID || impid != ANDES_AX45MP_MIMPID)
> -		return;
> -
> -	if (!ax45mp_iocp_sw_workaround())
> +	if (arch_id == ANDES_AX45MP_MARCHID && impid == ANDES_AX45MP_MIMPID) {
> +		if (!ax45mp_iocp_sw_workaround())
> +			return;
> +	} else if (riscv_isa_extension_available(NULL, ZICBOM)) {
>  		return;
> +	}

Is this correct? The code goes on to always mark noncoherent stuff
supported, so this means that every andes device must then support
noncoherent. The commit message's wording isn't that strong.

>  
>  	/* Set this just to make core cbo code happy */

This comment is no longer accurate because you gave this value special
meaning elsewhere in the series.
That patch needs to change this comment.

Cheers,
Conor.

>  	riscv_cbom_block_size = 1;
> @@ -69,7 +70,7 @@ void __init_or_module andes_errata_patch_func(struct alt_entry *begin, struct al
>  	BUILD_BUG_ON(ERRATA_ANDES_NUMBER >= RISCV_VENDOR_EXT_ALTERNATIVES_BASE);
>  
>  	if (stage == RISCV_ALTERNATIVES_BOOT)
> -		errata_probe_iocp(stage, archid, impid);
> +		errata_probe_andes_cache(stage, archid, impid);
>  
>  	/* we have nothing to patch here ATM so just return back */
>  }
> diff --git a/arch/riscv/include/asm/errata_list_vendors.h b/arch/riscv/include/asm/errata_list_vendors.h
> index 02b5ab0d23a8..d2fbef69f421 100644
> --- a/arch/riscv/include/asm/errata_list_vendors.h
> +++ b/arch/riscv/include/asm/errata_list_vendors.h
> @@ -4,7 +4,7 @@
>  #define ASM_ERRATA_LIST_VENDORS_H
>  
>  #ifdef CONFIG_ERRATA_ANDES
> -#define	ERRATA_ANDES_NO_IOCP 0
> +#define	ERRATA_ANDES_CACHE 0
>  #define	ERRATA_ANDES_NUMBER 1
>  #endif
>  
> -- 
> 2.34.1
> 

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]

  reply	other threads:[~2026-11-13 13:59 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-13  7:11 [PATCH v2 00/19] riscv: andes: make the LLC driver and cache errata generic Mina Hui-Min Chou
2026-11-13  7:11 ` [PATCH v2 01/19] cache: ax45mp_cache: use __initconst for ax45mp_cmo_ops Mina Hui-Min Chou
2026-11-13  7:11 ` [PATCH v2 02/19] cache: ax45mp_cache: rename AX45MP-specific identifiers to generic ANDES Mina Hui-Min Chou
2026-11-13  7:11 ` [PATCH v2 03/19] cache: ax45mp_cache: rename L2 identifiers to LLC Mina Hui-Min Chou
2026-11-13  7:11 ` [PATCH v2 04/19] cache: ax45mp_cache: add shared header for Andes CSRs Mina Hui-Min Chou
2026-11-13  7:11 ` [PATCH v2 05/19] cache: ax45mp_cache: standardize macro naming for LLC and L1D registers Mina Hui-Min Chou
2026-11-13  7:11 ` [PATCH v2 06/19] cache: ax45mp_cache: rename ax45mp_cache.c to andes_llcache.c Mina Hui-Min Chou
2026-11-13  7:11 ` [PATCH v2 07/19] cache: andes_llcache: clean up cache inv/wback functions Mina Hui-Min Chou
2026-11-13  7:11 ` [PATCH v2 08/19] cache: andes_llcache: use u32 and fix the cache-line-size error message Mina Hui-Min Chou
2026-11-13  7:11 ` [PATCH v2 09/19] cache: andes_llcache: write the full physical address for CCTL Mina Hui-Min Chou
2026-11-13  7:11 ` [PATCH v2 10/19] cache: andes_llcache: register cache ops only when the errata needs them Mina Hui-Min Chou
2026-11-13  7:11 ` [PATCH v2 11/19] cache: andes_llcache: use the hart ID to index the CCTL registers Mina Hui-Min Chou
2026-11-13  7:11 ` [PATCH v2 12/19] dt-bindings: cache: andestech: update schema for generic LLC support Mina Hui-Min Chou
2026-11-13 14:05   ` Conor Dooley
2026-11-13  7:11 ` [PATCH v2 13/19] dt-bindings: cache: rename andestech,ax45mp-cache to andestech,llcache Mina Hui-Min Chou
2026-11-13 14:01   ` Conor Dooley
2026-11-13  7:11 ` [PATCH v2 14/19] cache: andes_llcache: add generic andestech,llcache compatible Mina Hui-Min Chou
2026-11-13  7:11 ` [PATCH v2 15/19] MAINTAINERS: Add maintainer for Andes cache driver Mina Hui-Min Chou
2026-11-13  7:11 ` [PATCH v2 16/19] riscv: errata: andes: allow ERRATA_ANDES_CMO on all Andes SoCs Mina Hui-Min Chou
2026-11-13 14:00   ` Conor Dooley
2026-11-13  7:11 ` [PATCH v2 17/19] riscv: errata: andes: generalize cache errata for Andes cores Mina Hui-Min Chou
2026-11-13 13:59   ` Conor Dooley [this message]
2026-11-13  7:11 ` [PATCH v2 18/19] riscv: errata: andes: limit the IOCP workaround to RZ/Five Mina Hui-Min Chou
2026-11-13 13:53   ` Conor Dooley

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=20261113-voice-pledge-d0b4e1420676@spud \
    --to=conor@kernel.org \
    --cc=alex@ghiti.fr \
    --cc=aou@eecs.berkeley.edu \
    --cc=az70021@gmail.com \
    --cc=conor+dt@kernel.org \
    --cc=cyyang772@andestech.com \
    --cc=devicetree@vger.kernel.org \
    --cc=geert+renesas@glider.be \
    --cc=jic23@kernel.org \
    --cc=krzk+dt@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-renesas-soc@vger.kernel.org \
    --cc=linux-riscv@lists.infradead.org \
    --cc=magnus.damm@gmail.com \
    --cc=minachou@andestech.com \
    --cc=palmer@dabbelt.com \
    --cc=pjw@kernel.org \
    --cc=prabhakar.mahadev-lad.rj@bp.renesas.com \
    --cc=randolph@andestech.com \
    --cc=robh@kernel.org \
    --cc=tim609@andestech.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox