Re: [PATCH v11 10/16] iommu/arm-smmu-v3: Factor out arm_smmu_handle_gerror()
flat view
From: Pranjal Shrivastava <praan@google.com>
Date: 2026-09-30 20:00:40
Also in:
driver-core, linux-iommu, linux-pci, lkml
On Wed, Sep 30, 2026 at 11:34:40AM -0700, Nicolin Chen wrote:
On Tue, Sep 29, 2026 at 03:45:04AM +0000, Pranjal Shrivastava wrote:quoted
The GERROR register's state might be lost when the SMMU is powered down during runtime suspend, requiring the suspend sequence to handle any pending errors before the hardware state is lost. Refactor the gerror handling logic into a helper function. Subsequent patches will invoke it from the runtime suspend callback after disabling the SMMU, ensuring that any late-breaking gerrors are logged and ack'ed before the hardware state is lost.CMDQV has a separate IRQ, so I think it likely needs a counterpart. I have a similar patch in one of my series: https://lore.kernel.org/linux-iommu/16ae24cc2a002802b990afa1a69fe4ceab1e0da2.1790188510.git.nicolinc@nvidia.com/ (local) Mine only handles CMDQ error though..
I'll add a counterpart to arm_smmu_handle_gerror() in v12, I'll factor out the body of tegra241_cmdqv_isr() into something like tegra241_cmdqv_handle_errors() (bailing out early if VINTF_ERR_MAP is clear, so we don't warn on every suspend). The new impl->suspend op would call it after draining the lvcmdqs, with cmdqv->irq disabled around it, right before hw_deinit(vintf0). Since yours is for the CMD_SYNC timeout path, I'll keep this independent? WDYT? Praan