Re: [PATCH v20 1/9] PCI/AER: Introduce AER-CXL protocol error kfifo

[sashiko-bot@kernel.org]

> CXL VH RAS handling requires the AER driver to hand off CXL protocol
> errors to cxl_core for logging and recovery before PCIe AER recovery
> tears down the device. Introduce pci/pcie/aer_cxl_vh.c to implement
> this handoff via a kfifo-backed work item.
> 
> The producer, cxl_forward_error(), is gated by is_cxl_error() and
> enqueues the error source PCI device and severity. cxl_core registers a
> consumer via cxl_register_proto_err_work(); the consumer drains the
> kfifo with for_each_cxl_proto_err(). For uncorrectable errors,
> cxl_proto_err_wait_for_empty() lets the AER path block until the CXL
> plane has finished so recovery does not race device teardown.
> 
> A rwsem serializes registration, deregistration, enqueue, and dequeue
> against concurrent AER IRQ threads; a spinlock serializes concurrent
> kfifo writers. is_aer_internal_error() moves into this file and now
> evaluates info->status & ~info->mask rather than the raw info->status,
> so a masked internal-error bit is treated as not-set. For the RCH RCEC
> path this is equivalent because cxl_rch_enable_rcec() first calls
> pci_aer_unmask_internal_errors(), which clears those mask bits in
> hardware before the AER status is read back.
> 
> A subsequent patch wires cxl_forward_error() into handle_error_source().
> 
> Add MAINTAINERS entries for aer_cxl_vh.c and aer_cxl_rch.c under the CXL
> entry.
> 
> Co-developed-by: Dan Williams <djbw@kernel.org>
> Signed-off-by: Dan Williams <djbw@kernel.org>
> Signed-off-by: Terry Bowman <terry.bowman@amd.com>

Sashiko has reviewed this patch and found no issues. It looks great!

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260902133933.2992457-1-terry.bowman@amd.com?part=1