Re: [PATCH ath-next] wifi: ath10k: spread the PCI copy engine interrupts over the CPUs

From: Julius Bairaktaris

Date: Sat Nov 21 2026 - 12:54:42 EST


On 11/18/2026 11:57 AM, Jeff Johnson wrote:
> I'm concerned that you are taking an approach to benefit routers which can
> actually harm mobile clients (which want to minimize the number of CPU cores
> in use to reduce power consumption). I'm very hesitant to consider a series
> for upstream that could be detrimental to mobile devices.

Agreed. In v2 only QCA988X, QCA9887, QCA99X0, QCA9984 and QCA9888 take
the vector block, set by ce_msi_spread in their ath10k_hw_values.
QCA6174, QCA6164 and QCA9377 keep one MSI vector and no affinity hint.

> This looks like something that should stay within OpenWrt, which is focused on
> the AP, and not come upstream where we have to treat mobile devices nicely.

Without ce_msi_spread, ath10k_pci_init_irq() calls
pci_alloc_irq_vectors(pdev, 1, 1, PCI_IRQ_MSI), the call
pci_enable_msi() makes, and request, free, sync and deinit make the
same calls as ath-next, so the client chips see no change.

v2: https://lore.kernel.org/all/20261121175402.644803-1-julius@xxxxxxxxxxxxxx/

Julius