From: Michael Kelley Date: Thu, 3 Oct 2024 03:53:30 +0000 (-0700) Subject: Drivers: hv: Don't assume cpu_possible_mask is dense X-Git-Tag: v6.14-rc1~95^2~9 X-Git-Url: http://git.ipfire.org/?a=commitdiff_plain;h=16b18fdf6bc7292ae0edbf33d2d693af3240e49d;p=thirdparty%2Flinux.git Drivers: hv: Don't assume cpu_possible_mask is dense Current code allocates the hv_vp_index array with size num_possible_cpus(). This code assumes cpu_possible_mask is dense, which is not true in the general case per [1]. If cpu_possible_mask is sparse, the array might be indexed by a value beyond the size of the array. However, the configurations that Hyper-V provides to guest VMs on x86 and ARM64 hardware, in combination with how architecture specific code assigns Linux CPU numbers, *does* always produce a dense cpu_possible_mask. So the dense assumption is not currently causing failures. But for robustness against future changes in how cpu_possible_mask is populated, update the code to no longer assume dense. The correct approach is to allocate and initialize the array using size "nr_cpu_ids". While this leaves unused array entries corresponding to holes in cpu_possible_mask, the holes are assumed to be minimal and hence the amount of memory wasted by unused entries is minimal. Using nr_cpu_ids also reduces initialization time, in that the loop to initialize the array currently rescans cpu_possible_mask on each iteration. This is n-squared in the number of CPUs, which could be significant for large CPU counts. [1] https://lore.kernel.org/lkml/SN6PR02MB4157210CC36B2593F8572E5ED4692@SN6PR02MB4157.namprd02.prod.outlook.com/ Signed-off-by: Michael Kelley Acked-by: Peter Zijlstra (Intel) Link: https://lore.kernel.org/r/20241003035333.49261-3-mhklinux@outlook.com Signed-off-by: Wei Liu Message-ID: <20241003035333.49261-3-mhklinux@outlook.com> --- diff --git a/drivers/hv/hv_common.c b/drivers/hv/hv_common.c index c4fd07d9bf1ae..c6ed3ba4bf61e 100644 --- a/drivers/hv/hv_common.c +++ b/drivers/hv/hv_common.c @@ -345,14 +345,14 @@ int __init hv_common_init(void) BUG_ON(!hyperv_pcpu_output_arg); } - hv_vp_index = kmalloc_array(num_possible_cpus(), sizeof(*hv_vp_index), + hv_vp_index = kmalloc_array(nr_cpu_ids, sizeof(*hv_vp_index), GFP_KERNEL); if (!hv_vp_index) { hv_common_free(); return -ENOMEM; } - for (i = 0; i < num_possible_cpus(); i++) + for (i = 0; i < nr_cpu_ids; i++) hv_vp_index[i] = VP_INVAL; return 0;