Re: [PATCH] MM: discard __GFP_ATOMIC
From: Matthew Wilcox <willy@infradead.org>
Date: 2021-11-17 13:18:33
Also in:
lkml
On Wed, Nov 17, 2021 at 03:39:30PM +1100, NeilBrown wrote:
quoted hunk ↗ jump to hunk
+++ b/drivers/iommu/tegra-smmu.c@@ -676,12 +676,12 @@ static struct page *as_get_pde_page(struct tegra_smmu_as *as, * allocate page in a sleeping context if GFP flags permit. Hence * spinlock needs to be unlocked and re-locked after allocation. */ - if (!(gfp & __GFP_ATOMIC)) + if (gfp & __GFP_DIRECT_RECLAIM) spin_unlock_irqrestore(&as->lock, *flags); page = alloc_page(gfp | __GFP_DMA | __GFP_ZERO); - if (!(gfp & __GFP_ATOMIC)) + if (gfp & __GFP_DIRECT_RECLAIM) spin_lock_irqsave(&as->lock, *flags); /*
Surely this should be gfpflags_allow_blocking() instead of poking about in the innards of gfp flags? This patch seems like a good simplification to me. Reviewed-by: Matthew Wilcox (Oracle) <willy@infradead.org>