linux-kernel - Re: [PATCH v3 1/2] mm/memblock: Add reserved memory release function

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

Message-ID: <Z7RXGX/TyAvETRgc@kernel.org>
Date: Tue, 18 Feb 2025 11:47:05 +0200
From: Mike Rapoport <rppt@...nel.org>
To: Masami Hiramatsu <mhiramat@...nel.org>
Cc: Steven Rostedt <rostedt@...dmis.org>, linux-kernel@...r.kernel.org,
	linux-trace-kernel@...r.kernel.org,
	Mark Rutland <mark.rutland@....com>,
	Mathieu Desnoyers <mathieu.desnoyers@...icios.com>,
	Andrew Morton <akpm@...ux-foundation.org>
Subject: Re: [PATCH v3 1/2] mm/memblock: Add reserved memory release function

On Tue, Feb 18, 2025 at 05:42:57PM +0900, Masami Hiramatsu wrote:
> Hi Mike,
> 
> On Tue, 18 Feb 2025 09:24:53 +0200
> Mike Rapoport <rppt@...nel.org> wrote:
> 
> > Hi Masami,
> > 
> > On Tue, Feb 11, 2025 at 11:47:03PM +0900, Masami Hiramatsu (Google) wrote:
> > > From: Masami Hiramatsu (Google) <mhiramat@...nel.org>
> > > 
> > > Add reserve_mem_release_by_name() to release a reserved memory region
> > > with a given name. This allows us to release reserved memory which is
> > > defined by kernel cmdline, after boot.
> > > 
> > > Signed-off-by: Masami Hiramatsu (Google) <mhiramat@...nel.org>
> > > Cc: Andrew Morton <akpm@...ux-foundation.org>
> > > Cc: Mike Rapoport <rppt@...nel.org>
> > > Cc: linux-mm@...ck.org
> > 
> > ...
> > 
> > > +/**
> > > + * reserve_mem_release_by_name - Release reserved memory region with a given name
> > > + * @name: The name that is attatched to a reserved memory region
> > > + *
> > > + * Forcibly release the pages in the reserved memory region so that those memory
> > > + * can be used as free memory. After released the reserved region size becomes 0.
> > > + *
> > > + * Returns: 1 if released or 0 if not found.
> > > + */
> > > +int reserve_mem_release_by_name(const char *name)
> > > +{
> > > +	struct reserve_mem_table *map;
> > > +	unsigned int page_count;
> > > +	phys_addr_t start;
> > > +
> > > +	guard(mutex)(&reserve_mem_lock);
> > > +	map = reserve_mem_find_by_name_nolock(name);
> > > +	if (!map)
> > > +		return 0;
> > > +
> > > +	start = map->start;
> > > +	page_count = DIV_ROUND_UP(map->size, PAGE_SIZE);
> > > +
> > > +	for (int i = 0; i < page_count; i++) {
> > > +		phys_addr_t addr = start + i * PAGE_SIZE;
> > > +		struct page *page = pfn_to_page(addr >> PAGE_SHIFT);
> > > +
> > > +		page->flags &= ~BIT(PG_reserved);
> > > +		__free_page(page);
> > > +	}
> > 
> > We have free_reserved_area(), please use it here.
> > Otherwise
> > 
> > Acked-by: Mike Rapoport (Microsoft) <rppt@...nel.org>
> 
> Thanks! but I can not use free_reserved_area() here because it uses
> virt_to_page(). Here we only know the physical address in the map.
> I think we can use free_reserved_page() instead. Is that OK?

For reserve_mem ranges virt_to_phys() will work, they are allocated from the
memory that is covered by the direct map.
 
> Thank you,
> 
> -- 
> Masami Hiramatsu (Google) <mhiramat@...nel.org>

-- 
Sincerely yours,
Mike.