tjh.dev / kernel
Linux kernel mirror (for testing) git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
kernel os linux
1
fork

Configure Feed

Select the types of activity you want to include in your feed.

vmalloc: align nr_vmalloc_pages and vmap_lazy_nr

Currently both atomics share one cache-line:

<snip>
...
ffffffff83eab400 b vmap_lazy_nr
ffffffff83eab408 b nr_vmalloc_pages
...
<snip>

those are global variables and they are only 8 bytes apart. Since they
are modified by different threads this causes a false sharing. This can
lead to a performance drop due to unnecessary cache invalidations.

After this patch it is aligned to a cache line boundary:

<snip>
...
ffffffff8260a600 d vmap_lazy_nr
ffffffff8260a640 d nr_vmalloc_pages
...
<snip>

Link: https://lkml.kernel.org/r/20250417161216.88318-4-urezki@gmail.com
Signed-off-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
Reviewed-by: Baoquan He <bhe@redhat.com>
Reviewed-by: Adrian Huang <ahuang12@lenovo.com>
Tested-by: Adrian Huang <ahuang12@lenovo.com>
Cc: Mateusz Guzik <mjguzik@gmail.com>
Cc: Christop Hellwig <hch@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

authored by

Uladzislau Rezki (Sony) and committed by
Andrew Morton d0966120 7a6fe587

+2 -3
+2 -3
mm/vmalloc.c
··· 1008 1008 static void drain_vmap_area_work(struct work_struct *work); 1009 1009 static DECLARE_WORK(drain_vmap_work, drain_vmap_area_work); 1010 1010 1011 - static atomic_long_t nr_vmalloc_pages; 1011 + static __cacheline_aligned_in_smp atomic_long_t nr_vmalloc_pages; 1012 + static __cacheline_aligned_in_smp atomic_long_t vmap_lazy_nr; 1012 1013 1013 1014 unsigned long vmalloc_nr_pages(void) 1014 1015 { ··· 2117 2116 2118 2117 return log * (32UL * 1024 * 1024 / PAGE_SIZE); 2119 2118 } 2120 - 2121 - static atomic_long_t vmap_lazy_nr = ATOMIC_LONG_INIT(0); 2122 2119 2123 2120 /* 2124 2121 * Serialize vmap purging. There is no actual critical section protected