* [PATCH] arm64/fpsimd: simplify sme_setup()
@ 2025-09-13 0:09 Yury Norov (NVIDIA)
2025-09-16 21:13 ` Will Deacon
0 siblings, 1 reply; 2+ messages in thread
From: Yury Norov (NVIDIA) @ 2025-09-13 0:09 UTC (permalink / raw)
To: Catalin Marinas, Will Deacon, Mark Rutland, Mark Brown,
Ard Biesheuvel, linux-arm-kernel, linux-kernel
Cc: Yury Norov (NVIDIA)
The function checks info->vq_map for emptiness right before calling
find_last_bit().
We can use the find_last_bit() output and save on bitmap_empty() call,
which is O(N).
Signed-off-by: Yury Norov (NVIDIA) <yury.norov@gmail.com>
---
arch/arm64/kernel/fpsimd.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/arch/arm64/kernel/fpsimd.c b/arch/arm64/kernel/fpsimd.c
index c37f02d7194e..e3f8f51748bc 100644
--- a/arch/arm64/kernel/fpsimd.c
+++ b/arch/arm64/kernel/fpsimd.c
@@ -1265,6 +1265,8 @@ void __init sme_setup(void)
if (!system_supports_sme())
return;
+ min_bit = find_last_bit(info->vq_map, SVE_VQ_MAX);
+
/*
* SME doesn't require any particular vector length be
* supported but it does require at least one. We should have
@@ -1272,9 +1274,8 @@ void __init sme_setup(void)
* let's double check here. The bitmap is SVE_VQ_MAP sized for
* sharing with SVE.
*/
- WARN_ON(bitmap_empty(info->vq_map, SVE_VQ_MAX));
+ WARN_ON(min_bit >= SVE_VQ_MAX);
- min_bit = find_last_bit(info->vq_map, SVE_VQ_MAX);
info->min_vl = sve_vl_from_vq(__bit_to_vq(min_bit));
max_bit = find_first_bit(info->vq_map, SVE_VQ_MAX);
--
2.43.0
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH] arm64/fpsimd: simplify sme_setup()
2025-09-13 0:09 [PATCH] arm64/fpsimd: simplify sme_setup() Yury Norov (NVIDIA)
@ 2025-09-16 21:13 ` Will Deacon
0 siblings, 0 replies; 2+ messages in thread
From: Will Deacon @ 2025-09-16 21:13 UTC (permalink / raw)
To: Catalin Marinas, Mark Rutland, Mark Brown, Ard Biesheuvel,
linux-arm-kernel, linux-kernel, Yury Norov (NVIDIA)
Cc: kernel-team, Will Deacon
On Fri, 12 Sep 2025 20:09:05 -0400, Yury Norov (NVIDIA) wrote:
> The function checks info->vq_map for emptiness right before calling
> find_last_bit().
>
> We can use the find_last_bit() output and save on bitmap_empty() call,
> which is O(N).
>
>
> [...]
Applied to arm64 (for-next/misc), thanks!
[1/1] arm64/fpsimd: simplify sme_setup()
https://git.kernel.org/arm64/c/19dd484cd19c
Cheers,
--
Will
https://fixes.arm64.dev
https://next.arm64.dev
https://will.arm64.dev
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2025-09-16 21:14 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-09-13 0:09 [PATCH] arm64/fpsimd: simplify sme_setup() Yury Norov (NVIDIA)
2025-09-16 21:13 ` Will Deacon
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).