Re: [PATCH] arm: dma-mapping: fix potential endless loop in __dma_page_dev_to_cpu()

From: Matthew Wilcox
Date: Mon Aug 07 2023 - 12:24:07 EST


On Mon, Aug 07, 2023 at 05:26:57PM +0200, Marek Szyprowski wrote:
> It is possible that the folio_size() of the next folio returns zero, so

What? How can folio_size() return zero?

return PAGE_SIZE << folio_order(folio);

It is a minimum of PAGE_SIZE.

> avoid looping with 'left' equals to zero in D-cache cleaning loop.
>
> This fixes the following endless loop observed by RCU stall:
> --->8---
> rcu: INFO: rcu_sched self-detected stall on CPU
> rcu: 0-....: (27320 ticks this GP) idle=e414/1/0x40000002 softirq=36/36 fqs=13044
> rcu: (t=27385 jiffies g=-1067 q=34 ncpus=8)
> CPU: 0 PID: 93 Comm: kworker/0:1H Not tainted 6.5.0-rc5-next-20230807 #6981
> Hardware name: Samsung Exynos (Flattened Device Tree)
> Workqueue: mmc_complete mmc_blk_mq_complete_work
> PC is at _set_bit+0x28/0x44
> LR is at __dma_page_dev_to_cpu+0xdc/0x170
> ..
> _set_bit from __dma_page_dev_to_cpu+0xdc/0x170
> __dma_page_dev_to_cpu from dma_direct_unmap_sg+0x100/0x130
> dma_direct_unmap_sg from dw_mci_post_req+0x68/0x6c
> dw_mci_post_req from mmc_blk_mq_post_req+0x34/0x100

I don't know what you've actually hit here, but the explanation is wrong.