[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <Zn7q3oL1AE8jdM-g@arm.com>
Date: Fri, 28 Jun 2024 17:54:54 +0100
From: Catalin Marinas <catalin.marinas@....com>
To: Yang Shi <yang@...amperecomputing.com>
Cc: will@...nel.org, anshuman.khandual@....com, david@...hat.com,
scott@...amperecomputing.com, cl@...two.org,
linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [v5 PATCH] arm64: mm: force write fault for atomic RMW
instructions
On Wed, Jun 26, 2024 at 12:18:30PM -0700, Yang Shi wrote:
> @@ -568,6 +596,12 @@ static int __kprobes do_page_fault(unsigned long far, unsigned long esr,
> if (!vma)
> goto lock_mmap;
>
> + if ((vm_flags & VM_READ) && (vma->vm_flags & VM_WRITE) &&
> + is_el0_atomic_instr(regs)) {
> + vm_flags = VM_WRITE;
> + mm_flags |= FAULT_FLAG_WRITE;
> + }
The patch looks fine now and AFAICT there's no ABI change.
However, before deciding whether to merge this patch, I'd like to
understand why OpenJDK cannot use madvise(MADV_POPULATE_WRITE). This
would be the portable (Linux) solution that works better on
architectures without such atomic instructions (e.g. arm64 without LSE
atomics). So fixing user-space would be my preferred solution.
(I poked some people in Arm working in the area, hopefully I'll get some
more information)
--
Catalin
Powered by blists - more mailing lists