From: Mario Limonciello <superm1@kernel.org>
To: "Mario Limonciello" <superm1@kernel.org>,
"Ilpo Järvinen" <ilpo.jarvinen@linux.intel.com>
Cc: Hans de Goede <hansg@kernel.org>, Ma Jun <Jun.Ma2@amd.com>,
Evan Quan <quanliangl@hotmail.com>,
platform-driver-x86@vger.kernel.org,
linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2] platform/x86/amd: send WBRF DSM payload as buffer
Date: Wed, 18 Nov 2026 09:38:14 -0600 [thread overview]
Message-ID: <fa625e27-a1bd-4a22-8611-cca13fc9d11a@kernel.org> (raw)
In-Reply-To: <20261102-superm1-wbrf-v2-1-28b02bb4ce61@kernel.org>
On 11/2/26 14:58, Mario Limonciello wrote:
> Spec defines the WBRF record _DSM argument as a single buffer that packs
> the header and frequency ranges. The driver instead built a package of
> integers, so firmware on some systems abort \WBRF with AE_AML_OPERAND_TYPE
> when dereferencing Arg2[0].
>
> Encode the payload into a little-endian buffer and hand it as the lone
> package element.
>
> Fixes: 58e82a62669d ("platform/x86/amd: Add support for AMD ACPI based Wifi band RFI mitigation feature")
> Signed-off-by: Mario Limonciello <superm1@kernel.org>
> ---
> Changes in v2:
> - Declare payload with __free(kfree) at the kzalloc() site, after input
> validation, instead of initializing it to NULL at the top of
> wbrf_record() (Sashiko)
> - Compute payload length with struct_size_t() so it no longer depends on
> payload being declared earlier (Sashiko)
> - Link to v1: https://patch.msgid.link/20261102-superm1-wbrf-v1-1-b58d8d85277c@kernel.org
Ilpo,
FYI - this was found on some existing systems in the wild so I would
prefer if it could go to -fixes.
I sent right after the merge window so just wanted to make sure it's
still in your queue.
Thanks,
> ---
> drivers/platform/x86/amd/wbrf.c | 75 +++++++++++++++++++++++------------------
> 1 file changed, 43 insertions(+), 32 deletions(-)
>
> diff --git a/drivers/platform/x86/amd/wbrf.c b/drivers/platform/x86/amd/wbrf.c
> index 3281357185d1f..2afb574e4327d 100644
> --- a/drivers/platform/x86/amd/wbrf.c
> +++ b/drivers/platform/x86/amd/wbrf.c
> @@ -6,6 +6,8 @@
>
> #include <linux/acpi.h>
> #include <linux/acpi_amd_wbrf.h>
> +#include <asm/byteorder.h>
> +#include <linux/slab.h>
>
> /*
> * Functions bit vector for WBRF method
> @@ -29,6 +31,17 @@ struct amd_wbrf_ranges_out {
> struct freq_band_range band_list[MAX_NUM_OF_WBRF_RANGES];
> } __packed;
>
> +struct wbrf_record_entry {
> + __le64 start;
> + __le64 end;
> +} __packed;
> +
> +struct wbrf_record_payload {
> + __le32 num_of_ranges;
> + __le32 action;
> + struct wbrf_record_entry ranges[];
> +} __packed;
> +
> static const guid_t wifi_acpi_dsm_guid =
> GUID_INIT(0x7b7656cf, 0xdc3d, 0x4c1c,
> 0x83, 0xe9, 0x66, 0xe7, 0x21, 0xde, 0x30, 0x70);
> @@ -42,8 +55,9 @@ static BLOCKING_NOTIFIER_HEAD(wbrf_chain_head);
> static int wbrf_record(struct acpi_device *adev, uint8_t action, struct wbrf_ranges_in_out *in)
> {
> union acpi_object argv4;
> + union acpi_object args[1];
> u32 num_of_ranges = 0;
> - u32 num_of_elements;
> + size_t payload_len;
> u32 arg_idx = 0;
> int ret;
> u32 i;
> @@ -64,45 +78,39 @@ static int wbrf_record(struct acpi_device *adev, uint8_t action, struct wbrf_ran
> if (num_of_ranges != in->num_of_ranges)
> return -EINVAL;
>
> - /*
> - * Every input frequency band comes with two end points(start/end)
> - * and each is accounted as an element. Meanwhile the range count
> - * and action type are accounted as an element each.
> - * So, the total element count = 2 * num_of_ranges + 1 + 1.
> - */
> - num_of_elements = 2 * num_of_ranges + 2;
> -
> - union acpi_object *tmp __free(kfree) = kzalloc_objs(*tmp,
> - num_of_elements);
> - if (!tmp)
> + payload_len = struct_size_t(struct wbrf_record_payload, ranges, num_of_ranges);
> + struct wbrf_record_payload *payload __free(kfree) =
> + kzalloc(payload_len, GFP_KERNEL);
> + if (!payload)
> return -ENOMEM;
>
> - argv4.package.type = ACPI_TYPE_PACKAGE;
> - argv4.package.count = num_of_elements;
> - argv4.package.elements = tmp;
> -
> - /* save the number of ranges*/
> - tmp[0].integer.type = ACPI_TYPE_INTEGER;
> - tmp[0].integer.value = num_of_ranges;
> -
> - /* save the action(WBRF_RECORD_ADD/REMOVE/RETRIEVE) */
> - tmp[1].integer.type = ACPI_TYPE_INTEGER;
> - tmp[1].integer.value = action;
> + payload->num_of_ranges = cpu_to_le32(num_of_ranges);
> + payload->action = cpu_to_le32(action);
>
> - arg_idx = 2;
> + arg_idx = 0;
> for (i = 0; i < ARRAY_SIZE(in->band_list); i++) {
> if (!in->band_list[i].start || !in->band_list[i].end)
> continue;
>
> - tmp[arg_idx].integer.type = ACPI_TYPE_INTEGER;
> - tmp[arg_idx++].integer.value = in->band_list[i].start;
> - tmp[arg_idx].integer.type = ACPI_TYPE_INTEGER;
> - tmp[arg_idx++].integer.value = in->band_list[i].end;
> + payload->ranges[arg_idx].start = cpu_to_le64(in->band_list[i].start);
> + payload->ranges[arg_idx].end = cpu_to_le64(in->band_list[i].end);
> + arg_idx++;
> }
>
> + argv4.package.type = ACPI_TYPE_PACKAGE;
> + argv4.package.count = ARRAY_SIZE(args);
> + argv4.package.elements = args;
> +
> + args[0].type = ACPI_TYPE_BUFFER;
> + args[0].buffer.length = payload_len;
> + args[0].buffer.pointer = (u8 *)payload;
> +
> union acpi_object *obj __free(kfree) =
> - acpi_evaluate_dsm(adev->handle, &wifi_acpi_dsm_guid,
> - WBRF_REVISION, WBRF_RECORD, &argv4);
> + acpi_evaluate_dsm(adev->handle,
> + &wifi_acpi_dsm_guid,
> + WBRF_REVISION,
> + WBRF_RECORD,
> + &argv4);
>
> if (!obj)
> return -EINVAL;
> @@ -236,8 +244,11 @@ int amd_wbrf_retrieve_freq_band(struct device *dev, struct wbrf_ranges_in_out *o
> param.string.length = 0;
> param.string.pointer = NULL;
>
> - obj = acpi_evaluate_dsm(adev->handle, &wifi_acpi_dsm_guid,
> - WBRF_REVISION, WBRF_RETRIEVE, ¶m);
> + obj = acpi_evaluate_dsm(adev->handle,
> + &wifi_acpi_dsm_guid,
> + WBRF_REVISION,
> + WBRF_RETRIEVE,
> + ¶m);
> if (!obj)
> return -EINVAL;
>
>
> ---
> base-commit: 6628e7d4c6b25c8d18f1345e3a20d21829aae7e4
> change-id: 20261029-superm1-wbrf-d94050a28212
>
> Best regards,
> --
> Mario Limonciello <superm1@kernel.org>
next prev parent reply other threads:[~2026-11-18 15:38 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-02 20:58 [PATCH v2] platform/x86/amd: send WBRF DSM payload as buffer Mario Limonciello
2026-11-18 15:38 ` Mario Limonciello [this message]
2026-11-18 19:56 ` Ilpo Järvinen
2026-11-20 10:12 ` Ilpo Järvinen
2026-11-20 13:44 ` David Laight
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=fa625e27-a1bd-4a22-8611-cca13fc9d11a@kernel.org \
--to=superm1@kernel.org \
--cc=Jun.Ma2@amd.com \
--cc=hansg@kernel.org \
--cc=ilpo.jarvinen@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=platform-driver-x86@vger.kernel.org \
--cc=quanliangl@hotmail.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 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.