qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: "Daniel P. Berrangé" <berrange@redhat.com>
To: Helge Deller <deller@gmx.de>
Cc: Laurent Vivier <laurent@vivier.eu>,
	Richard Henderson <richard.henderson@linaro.org>,
	qemu-devel@nongnu.org
Subject: Re: [PATCH] linux-user: Fix openat() emulation to correctly detect accesses to /proc
Date: Tue, 1 Aug 2023 17:09:45 +0100	[thread overview]
Message-ID: <ZMkuSZL/fZ6CO934@redhat.com> (raw)
In-Reply-To: <ZMkdsWHTeHT2+lF2@p100>

On Tue, Aug 01, 2023 at 04:58:57PM +0200, Helge Deller wrote:
> In qemu we catch accesses to files like /proc/cpuinfo or /proc/net/route
> and return to the guest contents which would be visible on a real system
> (instead what the host would show).
> 
> This patch fixes a bug, where for example the accesses
>     cat /proc////cpuinfo
> or
>     cd /proc && cat cpuinfo
> will not be recognized by qemu and where qemu will wrongly show
> the contents of the host's /proc/cpuinfo file.
> 
> Signed-off-by: Helge Deller <deller@gmx.de>
> 
> diff --git a/linux-user/syscall.c b/linux-user/syscall.c
> index 917c388073..bb864c2bb3 100644
> --- a/linux-user/syscall.c
> +++ b/linux-user/syscall.c
> @@ -8531,9 +8531,11 @@ static int open_cpuinfo(CPUArchState *cpu_env, int fd)
>  }
>  #endif
> 
> -int do_guest_openat(CPUArchState *cpu_env, int dirfd, const char *pathname,
> +int do_guest_openat(CPUArchState *cpu_env, int dirfd, const char *fname,
>                      int flags, mode_t mode, bool safe)
>  {
> +    char proc_name[PATH_MAX];

No PATH_MAX buffers declared on the stack please.

> +    const char *pathname;
>      struct fake_open {
>          const char *filename;
>          int (*fill)(CPUArchState *cpu_env, int fd);
> @@ -8560,6 +8562,13 @@ int do_guest_openat(CPUArchState *cpu_env, int dirfd, const char *pathname,
>          { NULL, NULL, NULL }
>      };
> 
> +    /* if this is a file from /proc/ filesystem, expand full name */
> +    if (realpath(fname, proc_name) && strncmp(proc_name, "/proc/", 6) == 0) {

QEMU relies on the extended semantics of realpath() where passing NULL
for 'proc_name' causes it to allocate a buffer of the correct size and
return the new buffer. Using that avoids PATH_MAX variables.

> +        pathname = proc_name;
> +    } else {
> +        pathname = fname;
> +    }

With regards,
Daniel
-- 
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|



      reply	other threads:[~2023-08-01 16:10 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-08-01 14:58 [PATCH] linux-user: Fix openat() emulation to correctly detect accesses to /proc Helge Deller
2023-08-01 16:09 ` Daniel P. Berrangé [this message]

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=ZMkuSZL/fZ6CO934@redhat.com \
    --to=berrange@redhat.com \
    --cc=deller@gmx.de \
    --cc=laurent@vivier.eu \
    --cc=qemu-devel@nongnu.org \
    --cc=richard.henderson@linaro.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 NNTP newsgroup(s).