LKML Archive mirror
 help / color / mirror / Atom feed
From: "Michał Cłapiński" <mclapinski@google.com>
To: Ard Biesheuvel <ardb@kernel.org>
Cc: Thomas Gleixner <tglx@linutronix.de>,
	Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
	 Dave Hansen <dave.hansen@linux.intel.com>,
	x86@kernel.org,  "H. Peter Anvin" <hpa@zytor.com>,
	Dan Williams <dan.j.williams@intel.com>,
	 Pasha Tatashin <pasha.tatashin@soleen.com>,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH 1/1] x86/boot/compressed: Fix avoiding memmap in physical KASLR
Date: Sat, 21 Jun 2025 11:51:10 +0200	[thread overview]
Message-ID: <CAAi7L5dmAYJoB0BfKvoT0iKOeWuc5hLqFPBPQr99TkdEB1dtHg@mail.gmail.com> (raw)
In-Reply-To: <CAMj1kXErKwmiM5AiGOCE5D39U+3o_MXz5vqJmE5pkjQV308d9w@mail.gmail.com>

Hi Ard,

On Sat, Jun 21, 2025 at 10:19 AM Ard Biesheuvel <ardb@kernel.org> wrote:
>
> Hi Michal,
>
> On Tue, 10 Jun 2025 at 23:42, Michal Clapinski <mclapinski@google.com> wrote:
> >
> > The intent of the code was to cancel KASLR if there are more than 4
> > memmap args. Unfortunately, it was only doing that if the memmap args
> > were comma delimited, not if they were entirely separate.
> > This change fixes it.
> >
> > Signed-off-by: Michal Clapinski <mclapinski@google.com>
> > ---
> > I would like KASLR to support more than 4 memmap args. Do you think
> > I can just increase the MAX_MEMMAP_REGIONS or should I implement
> > support for the general case?
> >
> >  arch/x86/boot/compressed/kaslr.c | 3 ---
> >  1 file changed, 3 deletions(-)
> >
> > diff --git a/arch/x86/boot/compressed/kaslr.c b/arch/x86/boot/compressed/kaslr.c
> > index f03d59ea6e40..4aa9c9781ca7 100644
> > --- a/arch/x86/boot/compressed/kaslr.c
> > +++ b/arch/x86/boot/compressed/kaslr.c
> > @@ -162,9 +162,6 @@ static void mem_avoid_memmap(char *str)
> >  {
> >         static int i;
> >
> > -       if (i >= MAX_MEMMAP_REGIONS)
> > -               return;
> > -
>
> It isn't obvious at all why simply dropping this condition is fine.
> Could you elaborate?

Of course. Let's look at the whole function without my change:

static void mem_avoid_memmap(char *str)
{
        static int i;

        if (i >= MAX_MEMMAP_REGIONS)
                return;

        while (str && (i < MAX_MEMMAP_REGIONS)) {
                int rc;
                u64 start, size;
                char *k = strchr(str, ',');

                if (k)
                        *k++ = 0;

                rc = parse_memmap(str, &start, &size);
                if (rc < 0)
                        break;
                str = k;

                if (start == 0) {
                        /* Store the specified memory limit if size > 0 */
                        if (size > 0 && size < mem_limit)
                                mem_limit = size;

                        continue;
                }

                mem_avoid[MEM_AVOID_MEMMAP_BEGIN + i].start = start;
                mem_avoid[MEM_AVOID_MEMMAP_BEGIN + i].size = size;
                i++;
        }

        /* More than 4 memmaps, fail kaslr */
        if ((i >= MAX_MEMMAP_REGIONS) && str)
                memmap_too_large = true;
}

This function is called for every memmap= param. Let's say we supply
only separate memmap= params (instead of comma delimited). Then on the
4th param, `i` will be equal to MAX_MEMMAP_REGIONS but the last `if`
won't execute since `str` will be null. Then on the 5th param the
first `if` (the one I want to remove) will execute and
`memmap_too_large` will never be set.

With my change, while parsing the 5th param, the `while` loop will be
skipped since `i` is not smaller than MAX_MEMMAP_REGIONS and the last
`if` will execute setting `memmap_too_large`. Basically, my change is
safe because the `if` I want to remove is already baked into the
`while` loop condition.


>
> >         while (str && (i < MAX_MEMMAP_REGIONS)) {
> >                 int rc;
> >                 u64 start, size;
> > --
> > 2.50.0.rc0.642.g800a2b2222-goog
> >

  reply	other threads:[~2025-06-21  9:51 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-06-10 21:41 [PATCH 1/1] x86/boot/compressed: Fix avoiding memmap in physical KASLR Michal Clapinski
2025-06-10 22:44 ` Pasha Tatashin
2025-06-21  8:18 ` Ard Biesheuvel
2025-06-21  9:51   ` Michał Cłapiński [this message]
2025-07-02 15:57     ` Chris Li

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CAAi7L5dmAYJoB0BfKvoT0iKOeWuc5hLqFPBPQr99TkdEB1dtHg@mail.gmail.com \
    --to=mclapinski@google.com \
    --cc=ardb@kernel.org \
    --cc=bp@alien8.de \
    --cc=dan.j.williams@intel.com \
    --cc=dave.hansen@linux.intel.com \
    --cc=hpa@zytor.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=pasha.tatashin@soleen.com \
    --cc=tglx@linutronix.de \
    --cc=x86@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
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).