All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH v2 0/4] platform/x86: dell: dcdbas: Fix SMI completion wait and buffer handling
@ 2026-11-16 20:40 Rosen Penev
  2026-11-16 20:40 ` [PATCH v2 1/4] platform/x86: dell: dcdbas: Actually wait for the host control SMI Rosen Penev
                   ` (3 more replies)
  0 siblings, 4 replies; 6+ messages in thread
From: Rosen Penev @ 2026-11-16 20:40 UTC (permalink / raw)
  To: Stuart Hayes, Hans de Goede, Ilpo Järvinen, Andrew Morton,
	Doug Warzecha, Andy Shevchenko
  Cc: platform-driver-x86, linux-kernel

Fixes for dcdbas.

Patch 1 fixes the host control SMI wait for types 2 and 3, which the
compiler removed entirely because the status was a plain memory read,
and converts the type 1 wait along with it. Patches 2 and 3 fix the SMI
buffer lifetime on unbind and module removal. Patch 4 hardens the
validation of the firmware-provided WSMT buffer description.

These patches were developed with the help of an AI coding assistant
(Claude Opus 5.5): it reviewed the drivers, found the issues and drafted
the fixes and changelogs. I reviewed every change and I am responsible
for them.

All patches build cleanly with W=1.

Tested on a Dell Precision M4800 running v7.2.15, using the equivalent
backports built as out-of-tree modules:
the driver probes, and growing smi_data_buf_size and then unloading and
reloading dcdbas with dell-smbios on top is clean, with the buffer freed
on remove. The host control SMI (only used at power-off) and the WSMT
paths were not exercised.

Changes in v2:
- Squash the type 1 read_poll_timeout_atomic() conversion (v1 patch 5)
  into patch 1, so both waits in host_control_smi() change at once.
- Patch 1: include iopoll.h instead of io.h, and return the error from
  read_poll_timeout_atomic() (-ETIMEDOUT) instead of -ETIME.
- Patch 2: add a Fixes tag.
- Patch 3: make the dcdbas_release_buffer() comment a plain comment
  instead of malformed kernel-doc.
- Patch 4: the problems were found by code reading, so drop the Fixes
  tags and describe the patch as hardening; reject remap_size < 8
  instead of <= 8, as an 8 byte mapping just gives an empty buffer.
- v1: https://lore.kernel.org/platform-driver-x86/cover.1794633178.git.rosenp@gmail.com/

Rosen Penev (4):
  platform/x86: dell: dcdbas: Actually wait for the host control SMI
  platform/x86: dell: dcdbas: Suppress bind attributes of the platform
    driver
  platform/x86: dell: dcdbas: Release the SMI buffer in remove
  platform/x86: dell: dcdbas: Validate the WSMT buffer description

 drivers/platform/x86/dell/dcdbas.c | 82 ++++++++++++++++++------------
 drivers/platform/x86/dell/dcdbas.h |  1 -
 2 files changed, 50 insertions(+), 33 deletions(-)


base-commit: 2800868f63227c8237d70c41e70d441c919620d6
-- 
2.56.1


^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2026-11-17  7:57 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-11-16 20:40 [PATCH v2 0/4] platform/x86: dell: dcdbas: Fix SMI completion wait and buffer handling Rosen Penev
2026-11-16 20:40 ` [PATCH v2 1/4] platform/x86: dell: dcdbas: Actually wait for the host control SMI Rosen Penev
2026-11-17  7:54   ` Andy Shevchenko
2026-11-16 20:40 ` [PATCH v2 2/4] platform/x86: dell: dcdbas: Suppress bind attributes of the platform driver Rosen Penev
2026-11-16 20:40 ` [PATCH v2 3/4] platform/x86: dell: dcdbas: Release the SMI buffer in remove Rosen Penev
2026-11-16 20:40 ` [PATCH v2 4/4] platform/x86: dell: dcdbas: Validate the WSMT buffer description Rosen Penev

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.