linux-kernel - Re: [PATCH 1/6] seq_file: add seq_read

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: <20201116032942.GV3576660@ZenIV.linux.org.uk>
Date:   Mon, 16 Nov 2020 03:29:42 +0000
From:   Al Viro <viro@...iv.linux.org.uk>
To:     Nathan Chancellor <natechancellor@...il.com>
Cc:     Linus Torvalds <torvalds@...ux-foundation.org>,
        Christoph Hellwig <hch@....de>,
        Greg KH <gregkh@...uxfoundation.org>,
        Alexey Dobriyan <adobriyan@...il.com>,
        linux-fsdevel <linux-fsdevel@...r.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        kys@...rosoft.com, haiyangz@...rosoft.com, sthemmin@...rosoft.com,
        wei.liu@...nel.org, linux-hyperv@...r.kernel.org
Subject: Re: [PATCH 1/6] seq_file: add seq_read_iter

On Sun, Nov 15, 2020 at 05:34:16PM -0700, Nathan Chancellor wrote:
 
> Still good.
> 
> Tested-by: Nathan Chancellor <natechancellor@...il.com>

Pushed into #fixes

> > BTW, is that call of readv() really coming from init?  And if it
> > is, what version of init are you using?
> 
> I believe that it is but since this is WSL2, I believe that /init is a
> proprietary Microsoft implementation, rather than systemd or another
> init system:
> 
> https://wiki.ubuntu.com/WSL#Keeping_Ubuntu_Updated_in_WSL
> 
> So I am not sure how possible it is to see exactly what is going on or
> getting it improved.

Oh, well...  Anyway, as a regression test it's interesting:

#include <sys/uio.h>
#include <unistd.h>
#include <stdio.h>
#include <errno.h>
main()
{
	static char s[1024];
	static struct iovec v[2] = {{NULL, 0}, {s, 1024}};

	for(;;) {
		ssize_t n = readv(0, v, 2), m, w;

		if (n < 0) {
			perror("readv");
			return -1;
		}
		if (!n)
			return 0;
		for (m = 0; m < n; m += w) {
			w = write(1, s + m, n - m);
			if (w < 0)
				perror("write");
		}
	}
}

which ought to copy stdin to stdout; with this bug it would (on sufficiently
large seq_file-based files) fail with "readv: Bad address" (-EFAULT, that is).