Re: [PATCH 7/9] lib/cpumask: add num_{possible,present,active}_cpus_{eq,gt,le}
From: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Date: 2021-11-28 05:12:03
Also in:
kvm, linux-arm-kernel, linux-crypto, linux-mips, linux-mm, linux-perf-users, linux-riscv, linux-s390, linuxppc-dev, lkml
Possibly related (same subject, not in this thread)
- 2021-11-28 · Re: [PATCH 7/9] lib/cpumask: add num_{possible,present,active}_cpus_{eq,gt,le} · Yury Norov <yury.norov@gmail.com>
- 2021-11-28 · Re: [PATCH 7/9] lib/cpumask: add num_{possible,present,active}_cpus_{eq,gt,le} · Emil Renner Berthing <hidden>
- 2021-11-28 · Re: [PATCH 7/9] lib/cpumask: add num_{possible,present,active}_cpus_{eq,gt,le} · Joe Perches <joe@perches.com>
- 2021-11-28 · Re: [PATCH 7/9] lib/cpumask: add num_{possible,present,active}_cpus_{eq,gt,le} · Dennis Zhou <dennis@kernel.org>
- 2021-11-28 · Re: [PATCH 7/9] lib/cpumask: add num_{possible,present,active}_cpus_{eq,gt,le} · Yury Norov <yury.norov@gmail.com>
On Sun, Nov 28, 2021 at 05:56:51AM +0100, Micha³ Miros³aw wrote:
On Sat, Nov 27, 2021 at 07:57:02PM -0800, Yury Norov wrote:quoted
Add num_{possible,present,active}_cpus_{eq,gt,le} and replace num_*_cpus() with one of new functions where appropriate. This allows num_*_cpus_*() to return earlier depending on the condition.[...]quoted
@@ -3193,7 +3193,7 @@ int __init pcpu_page_first_chunk(size_t reserved_size, /* allocate pages */ j = 0; - for (unit = 0; unit < num_possible_cpus(); unit++) { + for (unit = 0; num_possible_cpus_gt(unit); unit++) {This looks dubious. The old version I could hope the compiler would call num_possible_cpus() only once if it's marked const or pure, but the alternative is going to count the bits every time making this a guaranteed O(n^2) even though the bitmap doesn't change.
Hmm. This code already unnecessarily calls num_possible_cpus() multiple times. Since it doesn't change after early init I would suggest just calling it once here. Best Regards Micha³ Miros³aw