stable.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg KH <greg@kroah.com>
To: "Zidenberg, Tsahi" <tsahee@amazon.com>
Cc: stable@vger.kernel.org
Subject: Re: [PATCH 2/2] tracing/kprobes: handle userspace access on unified probes
Date: Sat, 10 Apr 2021 13:29:27 +0200	[thread overview]
Message-ID: <YHGMFzQlHomDtZYG@kroah.com> (raw)
In-Reply-To: <ea2d7cd2-9891-573e-ebcb-bfeebd79661a@amazon.com>

On Mon, Mar 29, 2021 at 01:59:48PM +0300, Zidenberg, Tsahi wrote:
> 
> commit 9de1fec50b23117f0a19f7609cc837ca72e764a6 upstream.
> 
> This is an adaptation of parts from the above commit to kernel 5.4.
> 
> Allow Kprobes to access userspace data correctly in architectures with no
> overlap between kernel and userspace addresses.
> 
> Cc: <stable@vger.kernel.org> # 5.4.x
> Signed-off-by: Tsahi Zidenberg <tsahee@amazon.com>
> ---
>  kernel/trace/trace_kprobe.c | 15 +++++++++++++++
>  1 file changed, 15 insertions(+)
> 
> diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c
> index 233322c77b76..cbd72a1c9530 100644
> --- a/kernel/trace/trace_kprobe.c
> +++ b/kernel/trace/trace_kprobe.c
> @@ -1043,6 +1043,11 @@ fetch_store_strlen(unsigned long addr)
>      int ret, len = 0;
>      u8 c;
>  
> +#ifdef CONFIG_ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE
> +    if (addr < TASK_SIZE)
> +        return fetch_store_strlen_user(addr);
> +#endif
> +
>      do {
>          ret = probe_kernel_read(&c, (u8 *)addr + len, 1);
>          len++;
> @@ -1071,6 +1076,11 @@ fetch_store_string(unsigned long addr, void *dest, void *base)
>      void *__dest;
>      long ret;
>  
> +#ifdef CONFIG_ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE
> +    if (addr < TASK_SIZE)
> +        return fetch_store_string_user(addr, dest, base);
> +#endif
> +
>      if (unlikely(!maxlen))
>          return -ENOMEM;
>  
> @@ -1114,6 +1124,11 @@ fetch_store_string_user(unsigned long addr, void *dest, void *base)
>  static nokprobe_inline int
>  probe_mem_read(void *dest, void *src, size_t size)
>  {
> +#ifdef CONFIG_ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE
> +    if ((unsigned long)src < TASK_SIZE)
> +        return probe_mem_read_user(dest, src, size);
> +#endif
> +
>      return probe_kernel_read(dest, src, size);
>  }
>  
> -- 
> 2.25.1

What problem is this fixing?

thanks,

greg k-h

  reply	other threads:[~2021-04-10 11:29 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-03-29 10:56 [PATCH 0/2] fix userspace access on arm64 for linux 5.4 Zidenberg, Tsahi
2021-03-29 10:58 ` [PATCH 1/2] bpf: fix userspace access for bpf_probe_read{, str}() Zidenberg, Tsahi
2021-03-30 17:21   ` Sasha Levin
2021-03-31 18:37     ` Zidenberg, Tsahi
2021-04-03  9:56       ` Greg KH
2021-04-04  9:13         ` Zidenberg, Tsahi
2021-04-10 11:29           ` Greg KH
2021-04-12 20:01             ` Zidenberg, Tsahi
2021-04-13  7:28               ` Greg KH
2021-03-29 10:59 ` [PATCH 2/2] tracing/kprobes: handle userspace access on unified probes Zidenberg, Tsahi
2021-04-10 11:29   ` Greg KH [this message]
2021-04-10 11:30 ` [PATCH 0/2] fix userspace access on arm64 for linux 5.4 Greg KH
2021-04-12 19:46   ` Zidenberg, Tsahi
2021-04-13  7:27     ` Greg KH
2021-04-21 13:04       ` Zidenberg, Tsahi
2021-04-21 13:26         ` Greg KH

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=YHGMFzQlHomDtZYG@kroah.com \
    --to=greg@kroah.com \
    --cc=stable@vger.kernel.org \
    --cc=tsahee@amazon.com \
    /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).