linux-kernel - Re: [PATCH] mm: Inline vma_needs

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20250619033037.1073740-1-jiangyunshui@kylinos.cn>
Date: Thu, 19 Jun 2025 11:30:07 +0800
From: Yunshui Jiang <jiangyunshui@...inos.cn>
To: liam.howlett@...cle.com
Cc: akpm@...ux-foundation.org,
	david@...hat.com,
	jiangyunshui@...inos.cn,
	linux-kernel@...r.kernel.org,
	linux-mm@...k.org,
	lorenzo.stoakes@...cle.com,
	vbabka@...e.cz
Subject: Re: [PATCH] mm: Inline vma_needs_copy

Initially, I added explicit inline hints in the code precisely because some
newer compilers might ignore such suggestions.
 

Based on everyone's feedback, I rechecked my compiler configurations

—ARM64 uses GCC 12.3, while x86 uses GCC 12.4. By analyzing the

disassembly of vmlinux via objdump, I confirmed that vma_needs_copy

is indeed inlined in both cases. In this case the score difference in spawn

might just be due to fluctuations.

 

Leave it unchanged might be better. If we later encounter issues where

older compilers fail to inline single-caller cases or multiple calls are

incurred due to code changes, we could re-discuss  whether manual

intervention is needed. 

 

Best regards,
Yunshui Jiang