mailing list of musl libc
 help / color / mirror / code / Atom feed
From: Szabolcs Nagy <nsz@port70.net>
To: musl@lists.openwall.com
Subject: Re: [PATCH] Add support for mkostemp, mkstemps and mkostemps
Date: Thu, 31 Jan 2013 01:28:09 +0100	[thread overview]
Message-ID: <20130131002808.GI6181@port70.net> (raw)
In-Reply-To: <20130130192233.GP20323@brightrain.aerifal.cx>

* Rich Felker <dalias@aerifal.cx> [2013-01-30 14:22:33 -0500]:
> On Wed, Jan 30, 2013 at 08:12:58PM +0100, Szabolcs Nagy wrote:
> > void __randname(char *p)
> > {
> > 	struct timespec ts;
> > 	unsigned long r;
> > 	int i;
> > 
> > 	clock_gettime(CLOCK_REALTIME, &ts);
> > 	r = ts.tv_nsec*65537 ^ (uintptr_t)&ts / 16 + (uintptr_t)p;
> > 	for (i=0; i<6; i++, r>>=5)
> > 		p[i] = 'A'+(r&15)+(r&16)*2;
> > }
> > 
> > this uses 30bits of r and mixes the random low bits of nsec
> > into the high bits
> 
> Keep in mind it might be bits 8-15 that are most valuable with ASLR
> (assuming the randomization only adjusts by small amounts and not so
> much to waste lots of address space). I think this needs a little bit
> more consideration.

hm then use *69069 (a favourite lcg multiplier of the
above mentioned george marsaglia according to knuth,
[taocp vol2 p108] so it probably mixes the low bits
of nsec well)

> > > Other implementations probably use 36 bits or slightly less (base64
> > > perhaps modified base64).
> > > 
> > > I could see it being feasible to increase this slightly and maybe even
> > 
> > <= 36bits is probably ok
> 
> You mean >=36? Or..?

i wanted to say that using <=64 chars in the filename is reasonable
(i thought there were only 62 robust chars: lower,upper,digits
so going above 36 does not make sense and we have 32bit input anyway)

but maybe #%+-._~ are ok as well (these are the ascii chars that
are not escaped by bash for whatever reason), except #+-.~ should
not be first char

in urls only -._ are the always safe extra chars

in regex #%-_~ are safe from the above set

but if we go above 62chars then more than 32bit input is needed


another proposal:

uint64_t r=1;
uint64_t a[]={stackptr, inputptr, sec, ns, pid, retrycount}; // entropy sources
for (i = 0; i < 6; i++) {
	r += a[i];
	r *= 6364136223846793005ull;
	r ^= r>>24;
}
for (i = 0; i < 6; i++, r /= 63)
	p[i] = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789_"[r%63];


  reply	other threads:[~2013-01-31  0:28 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-01-28  5:06 Anthony G. Basile
2013-01-28  8:47 ` John Spencer
2013-01-28  9:37 ` Szabolcs Nagy
2013-01-28 17:33   ` Szabolcs Nagy
2013-01-29 23:16   ` Anthony G. Basile
2013-01-30  7:21     ` Rich Felker
2013-01-30  7:59       ` Hardy Falk
2013-01-30 13:45         ` Szabolcs Nagy
2013-01-30 16:51           ` Rich Felker
2013-01-30 19:12             ` Szabolcs Nagy
2013-01-30 19:22               ` Rich Felker
2013-01-31  0:28                 ` Szabolcs Nagy [this message]
2013-01-28 16:33 ` Rich Felker
2013-02-02  3:48 ` Rich Felker
2013-02-02 18:46   ` Anthony G. Basile
  -- strict thread matches above, loose matches on Subject: below --
2013-02-02 21:15 Anthony G. Basile
2013-02-02 21:17 ` Anthony G. Basile
2013-02-03  3:30 ` Rich Felker
2013-02-02 18:45 Anthony G. Basile
2013-02-02 19:51 ` John Spencer
2013-02-02 19:59   ` John Spencer
2013-02-02 22:11   ` Szabolcs Nagy
2013-02-02 20:14 ` Szabolcs Nagy
2013-02-02 20:38   ` Anthony G. Basile
2013-02-02 22:22     ` Szabolcs Nagy
2013-01-28  2:36 Anthony G. Basile
2013-01-28  5:01 ` Anthony G. Basile

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=20130131002808.GI6181@port70.net \
    --to=nsz@port70.net \
    --cc=musl@lists.openwall.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.
Code repositories for project(s) associated with this public inbox

	https://git.vuxu.org/mirror/musl/

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).