linux-kernel - Re: [PATCH] riscv: fix the global name pfn

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: <YQFTKtNmqorzsv6a@kernel.org>
Date:   Wed, 28 Jul 2021 15:52:58 +0300
From:   Mike Rapoport <rppt@...nel.org>
To:     Alex Ghiti <alex@...ti.fr>
Cc:     Kenneth Lee <nek.in.cn@...il.com>,
        Paul Walmsley <paul.walmsley@...ive.com>,
        Palmer Dabbelt <palmer@...belt.com>,
        Albert Ou <aou@...s.berkeley.edu>,
        Anup Patel <anup@...infault.org>,
        Atish Patra <atish.patra@....com>,
        Kenneth Lee <liguozhu@...ilicon.com>,
        Kefeng Wang <wangkefeng.wang@...wei.com>,
        Vitaly Wool <vitaly.wool@...sulko.com>,
        Guo Ren <guoren@...ux.alibaba.com>,
        Jisheng Zhang <jszhang@...nel.org>,
        Nick Kossifidis <mick@....forth.gr>,
        linux-riscv@...ts.infradead.org, linux-kernel@...r.kernel.org,
        guohanjun@...wei.com, wangzhou1@...ilicon.com
Subject: Re: [PATCH] riscv: fix the global name pfn_base confliction error

On Wed, Jul 28, 2021 at 09:19:16AM +0200, Alex Ghiti wrote:
> Hi Kenneth,
> 
> Le 28/07/2021 à 08:43, Kenneth Lee a écrit :
> > From: Kenneth Lee <liguozhu@...ilicon.com>
> > 
> > RISCV use a global variable pfn_base for page/pfn translation. But this
> > is a common name and will be used elsewhere. In those case,
> > the page-pfn macro which refer this name will refer to the local/input
> > variable of those function (such as in vfio_pin_pages_remote). This make
> > everything wrong.
> > 
> > This patch change the name from pfn_base to riscv_global_pfn_base to fix
> > this problem
> 
> What about removing this variable entirely and using
> PFN_DOWN(kernel_map.phys_addr) directly in ARCH_PFN_OFFSET definition? That
> would remove code from mm/init.c, which is nice :)

That would be nice, but such change would also change the generated code.
Probably nothing important, but someone would have to at least check what
size and bloat-o-meter say.
 
> Thanks,
> 
> Alex
> 
> > 
> > Signed-off-by: Kenneth Lee <liguozhu@...ilicon.com>
> > ---
> >   arch/riscv/include/asm/page.h | 4 ++--
> >   arch/riscv/mm/init.c          | 6 +++---
> >   2 files changed, 5 insertions(+), 5 deletions(-)
> > 
> > diff --git a/arch/riscv/include/asm/page.h b/arch/riscv/include/asm/page.h
> > index cca8764aed83..8711e415f37c 100644
> > --- a/arch/riscv/include/asm/page.h
> > +++ b/arch/riscv/include/asm/page.h
> > @@ -79,8 +79,8 @@ typedef struct page *pgtable_t;
> >   #endif
> >   #ifdef CONFIG_MMU
> > -extern unsigned long pfn_base;
> > -#define ARCH_PFN_OFFSET		(pfn_base)
> > +extern unsigned long riscv_global_pfn_base;
> > +#define ARCH_PFN_OFFSET		(riscv_global_pfn_base)
> >   #else
> >   #define ARCH_PFN_OFFSET		(PAGE_OFFSET >> PAGE_SHIFT)
> >   #endif /* CONFIG_MMU */
> > diff --git a/arch/riscv/mm/init.c b/arch/riscv/mm/init.c
> > index a14bf3910eec..2ce4e9a46ca0 100644
> > --- a/arch/riscv/mm/init.c
> > +++ b/arch/riscv/mm/init.c
> > @@ -228,8 +228,8 @@ static struct pt_alloc_ops _pt_ops __initdata;
> >   #define pt_ops _pt_ops
> >   #endif
> > -unsigned long pfn_base __ro_after_init;
> > -EXPORT_SYMBOL(pfn_base);
> > +unsigned long riscv_global_pfn_base __ro_after_init;
> > +EXPORT_SYMBOL(riscv_global_pfn_base);
> >   pgd_t swapper_pg_dir[PTRS_PER_PGD] __page_aligned_bss;
> >   pgd_t trampoline_pg_dir[PTRS_PER_PGD] __page_aligned_bss;
> > @@ -572,7 +572,7 @@ asmlinkage void __init setup_vm(uintptr_t dtb_pa)
> >   	kernel_map.va_kernel_pa_offset = kernel_map.virt_addr - kernel_map.phys_addr;
> >   #endif
> > -	pfn_base = PFN_DOWN(kernel_map.phys_addr);
> > +	riscv_global_pfn_base = PFN_DOWN(kernel_map.phys_addr);
> >   	/*
> >   	 * Enforce boot alignment requirements of RV32 and
> > 

-- 
Sincerely yours,
Mike.