From: E Shattow <e@freeshell.de>
To: Anirudh Srinivasan <asrinivasan@oss.tenstorrent.com>,
Yao Zi <me@ziyao.cc>,
u-boot@lists.u-boot-project.org
Cc: Tim Ouyang <tim609@andestech.com>,
Leo Liang <leo.liang@sifive.com>, Tom Rini <trini@konsulko.com>,
Troy Mitchell <troy.mitchell@linux.spacemit.com>
Subject: Re: [PATCH v3] riscv: Disable M-IMSIC and M-APLIC nodes in DT passed to next stage in S mode
Date: Thu, 19 Nov 2026 13:31:27 -0800 [thread overview]
Message-ID: <ae7bc060-5d53-4d55-b006-89ac5fcbada5@freeshell.de> (raw)
In-Reply-To: <20261119-mimsic_maplic_fixup-v3-1-e028e28fb958@oss.tenstorrent.com>
On 11/19/26 07:47, Anirudh Srinivasan wrote:
> When U-Boot is running in S mode, the DT passed to the next S mode stage
> can either be the bundled in DT U-Boot and the previous M mode stage
> use, or be loaded by U-Boot from some boot media. In the former case,
> the M mode stage can fix up the DT to disable the M-IMSIC and M-APLIC
> nodes on an AIA system (for example, OpenSBI does this fixup). These
> controllers are to be managed by M-mode software, so this is done to
> prevent S mode software from trying to probe them.
>
> This fixup doesn't happen in the latter case because U-Boot loads a
> fresh DT from boot media. Add a fdt fixup to do this similar to OpenSBI.
> This gets rid of prints like this during linux boots.
>
> [ 0.000000] riscv-imsic: interrupt-controller@a0000000: no parent irqs available
> [ 0.000000] riscv-imsic: interrupt-controller@a0000000: failed to setup state (error -22)
> [ 0.000000] OF: of_irq_init: Failed to init /soc/interrupt-controller@a0000000 ((____ptrval____)), parent (____ptrval____)
>
> [ 10.727605] platform cc000000.interrupt-controller: deferred probe pending: platform: supplier a0000000.interrupt-controller not ready
>
> Reviewed-by: E Shattow <e@freeshell.de>
> Signed-off-by: Anirudh Srinivasan <asrinivasan@oss.tenstorrent.com>
> ---
> Changes in v3:
> - Reword commit title to say M-APLIC/M-IMSIC and drop references to M Mode to avoid any confusion
> - Link to v2: https://patch.msgid.link/20261119-mimsic_maplic_fixup-v2-1-e29e04bf1b76@oss.tenstorrent.com
>
> Changes in v2:
> - Rename riscv_aia_mmode_aplic_imsic_fixup to riscv_aia_mimsic_maplic_fixup
> - Drop most references to OpenSBI/Linux in commit message and just use M-mode/s-mode software
> - Link to v1: https://patch.msgid.link/20261118-mimsic_maplic_fixup-v1-1-52dea8f39779@oss.tenstorrent.com
> ---
> arch/riscv/lib/fdt_fixup.c | 86 ++++++++++++++++++++++++++++++++++++++++++++++
> 1 file changed, 86 insertions(+)
>
> diff --git a/arch/riscv/lib/fdt_fixup.c b/arch/riscv/lib/fdt_fixup.c
> index 6dfd1a2d5b8..49f5b3b5bb6 100644
> --- a/arch/riscv/lib/fdt_fixup.c
> +++ b/arch/riscv/lib/fdt_fixup.c
> @@ -117,6 +117,85 @@ int riscv_board_reserved_mem_fixup(void *fdt)
> return 0;
> }
>
> +/**
> + * riscv_node_handles_mmode_external_interrupts() - Check if interrupt controller handles M-mode
> + * external interrupts
> + * @fdt: Pointer to the device tree
> + * @offset: Offset of node in DT to check
> + *
> + * Return: true if controller handles M-mode external interrupts IRQ_M_EXT (11), else false
> + */
> +static bool riscv_node_handles_mmode_external_interrupts(void *fdt, int offset)
> +{
> + int i, len;
> + const fdt32_t *interrupts_extended = fdt_getprop(fdt, offset, "interrupts-extended", &len);
> +
> + /*
> + * Each entry is <phandle irq>. The phandle points to a riscv,cpu-intc
> + * node, whose DT binding sets #interrupt-cells = <1>, so each entry is
> + * 2 cells.
> + */
> + if (!interrupts_extended || !len || len % (2 * sizeof(u32)))
> + return false;
> + for (i = 1; i < len / sizeof(u32); i += 2) {
> + if (fdt32_to_cpu(interrupts_extended[i]) == IRQ_M_EXT)
> + return true;
> + }
> + return false;
> +}
> +
> +/**
> + * riscv_aia_mimsic_maplic_fixup() - Disable M-APLIC and M-IMSIC nodes in DT
> + * @fdt: Pointer to the device tree in which nodes are to be disabled
> + *
> + * On RISC-V AIA systems, M-mode software manages the M-APLIC and M-IMSIC.
> + * If S-mode U-Boot loads a device tree from boot media, the DT nodes for these
> + * could still be enabled, so fixup the DT to disable these nodes before passing
> + * it to the next S-mode stage.
> + *
> + * Return: 0 on success or error otherwise.
> + */
> +static int riscv_aia_mimsic_maplic_fixup(void *fdt)
> +{
> + int offset, err;
> +
> + /* Disable M-IMSIC in APLIC-IMSIC configuration */
> + fdt_for_each_node_by_compatible(offset, fdt, -1, "riscv,imsics") {
> + if (riscv_node_handles_mmode_external_interrupts(fdt, offset)) {
> + err = fdt_status_disabled(fdt, offset);
> + if (err < 0)
> + return err;
> + }
> + }
> +
> + /* Disable M-APLIC in APLIC only configuration */
> + fdt_for_each_node_by_compatible(offset, fdt, -1, "riscv,aplic") {
> + if (riscv_node_handles_mmode_external_interrupts(fdt, offset)) {
> + err = fdt_status_disabled(fdt, offset);
> + if (err < 0)
> + return err;
> + }
> + }
> +
> + /* Disable M-APLIC in APLIC-IMSIC configuration */
> + fdt_for_each_node_by_compatible(offset, fdt, -1, "riscv,aplic") {
> + int len, imsic_offset;
> + const fdt32_t *msi_parent = fdt_getprop(fdt, offset, "msi-parent", &len);
> +
> + if (msi_parent && len >= sizeof(fdt32_t)) {
> + imsic_offset = fdt_node_offset_by_phandle(fdt, fdt32_to_cpu(*msi_parent));
> + if (imsic_offset >= 0 &&
> + riscv_node_handles_mmode_external_interrupts(fdt, imsic_offset)) {
> + err = fdt_status_disabled(fdt, offset);
> + if (err < 0)
> + return err;
> + }
> + }
> + }
> +
> + return 0;
> +}
> +
> #ifdef CONFIG_OF_BOARD_FIXUP
> int board_fix_fdt(void *fdt)
> {
> @@ -165,5 +244,12 @@ int arch_fixup_fdt(void *blob)
> if (err < 0)
> return err;
>
> + /* Disable M-IMSIC and M-APLIC nodes when we're not in M-mode */
> + if (!CONFIG_IS_ENABLED(RISCV_MMODE)) {
> + err = riscv_aia_mimsic_maplic_fixup(blob);
> + if (err < 0)
> + return err;
> + }
> +
> return 0;
> }
>
> ---
> base-commit: 83b1ec64ede9998566e6a82b22848c20a33f4aff
> change-id: 20261118-mimsic_maplic_fixup-051bd38d0985
>
> Best regards,
> --
> Anirudh Srinivasan <asrinivasan@oss.tenstorrent.com>
>
The code and function call-out comments LGTM, thanks for updating those.
nit: If I am understanding this correctly, then I suggest to drop
references to "boot media" in the commit description of where DT
originates from; we simply have the U-Boot control Fdt (in-memory dtb
with whatever modifications happened before it gets to our present
state) as it was passed from M-mode or SBI S-mode handoff regardless of
where it originates from. It exists, and we need to validate in U-Boot
runtime for uniformity with the current operating mode since U-Boot
source tree is a likely source of overrides [1] and non-uniformity in
handoff to S-mode kernels.
From what I understand of reading the code then that commit message
could be clearer as e.g.:
"Introduce fixup of initial control Fdt in S-mode for the current
operating mode to disable non-applicable M-mode specific M-IMSIC and
M-APLIC nodes.
This fixup is evaluated once for the control Fdt passed from the
previous stage and does not affect any replacement Fdt loaded from boot
media."
...unless this does filter again on subsequent actions loading DT from
boot media, as that's what the present commit message has me confused
about now. I don't think that it does, from what I understand reading
the code.
Keep my R-by tag, and up to you to refine the commit message or not as
you like. I appreciate your help with this as a follow-up to [1]. Thanks!
-E
1: https://lore.kernel.org/lkml/20261029162343.1896134-1-e@freeshell.de/
next prev parent reply other threads:[~2026-11-19 21:32 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-19 15:47 [PATCH v3] riscv: Disable M-IMSIC and M-APLIC nodes in DT passed to next stage in S mode Anirudh Srinivasan
2026-11-19 21:31 ` E Shattow [this message]
2026-11-20 1:04 ` Anirudh Srinivasan
2026-11-20 10:46 ` Yao Zi
2026-11-20 16:09 ` Anirudh Srinivasan
2026-11-20 17:07 ` Yao Zi
2026-11-20 17:24 ` Anirudh Srinivasan
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=ae7bc060-5d53-4d55-b006-89ac5fcbada5@freeshell.de \
--to=e@freeshell.de \
--cc=asrinivasan@oss.tenstorrent.com \
--cc=leo.liang@sifive.com \
--cc=me@ziyao.cc \
--cc=tim609@andestech.com \
--cc=trini@konsulko.com \
--cc=troy.mitchell@linux.spacemit.com \
--cc=u-boot@lists.u-boot-project.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 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.