public inbox for linux-arch@vger.kernel.org
 help / color / mirror / Atom feed
From: Andrew Morton <akpm@osdl.org>
To: Anton Blanchard <anton@samba.org>
Cc: arnd@arndb.de, richm@oldelvet.org.uk, tytso@mit.edu,
	linux-arch@vger.kernel.org
Subject: Re: Using get_cycles for add_timer_randomness
Date: Sun, 15 Aug 2004 15:48:21 -0700	[thread overview]
Message-ID: <20040815154821.2d6edb03.akpm@osdl.org> (raw)
In-Reply-To: <20040814183623.GB5637@krispykreme>


Things have changed a bit - sparc now has special-case code in there too. 
Hopefully we can remove that now.

Is everyone OK with this?


From: Anton Blanchard <anton@samba.org>

I tested how long it took to do a dd from /dev/random on ppc64 before and
after this patch, while doing a ping flood from another machine.

before:
# /usr/bin/time dd if=/dev/random of=/dev/zero count=1k
0+51 records in
Command terminated by signal 2
0.00user 0.00system 19:18.46elapsed 0%CPU (0avgtext+0avgdata 0maxresident)k

I gave up after 19 minutes.

after:
# /usr/bin/time dd if=/dev/random of=/dev/zero count=1k
0+1024 records in
0.00user 0.00system 0:33.38elapsed 0%CPU (0avgtext+0avgdata 0maxresident)k

Just over 33 seconds. Better.

From: Arnd Bergmann <arnd@arndb.de>

I noticed that only i386 and x86-64 are currently using a high resolution
timer source when adding randomness.  Since many architectures have a
working get_cycles() implementation, it seems rather straightforward to use
that.

Has this been discussed before, or can anyone comment on the implementation
below?

This patch attempts to take into account the size of cycles_t, which is
either 32 or 64 bits wide but independent of the architecture's word size.

The behavior should be nearly identical to the old one on i386, x86-64 and
all architectures without a time stamp counter, while finding more entropy
on the other architectures.

Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---

 25-akpm/drivers/char/random.c    |   31 ++++++++++++-------------------
 25-akpm/include/asm-i386/timex.h |   13 ++++++++-----
 2 files changed, 20 insertions(+), 24 deletions(-)

diff -puN drivers/char/random.c~using-get_cycles-for-add_timer_randomness drivers/char/random.c
--- 25/drivers/char/random.c~using-get_cycles-for-add_timer_randomness	2004-08-15 15:45:13.514016424 -0700
+++ 25-akpm/drivers/char/random.c	2004-08-15 15:46:14.773703536 -0700
@@ -781,8 +781,8 @@ static void batch_entropy_process(void *
 
 /* There is one of these per entropy source */
 struct timer_rand_state {
-	__u32		last_time;
-	__s32		last_delta,last_delta2;
+	cycles_t	last_time;
+	long		last_delta,last_delta2;
 	int		dont_count_entropy:1;
 };
 
@@ -799,14 +799,12 @@ static struct timer_rand_state *irq_time
  * The number "num" is also added to the pool - it should somehow describe
  * the type of event which just happened.  This is currently 0-255 for
  * keyboard scan codes, and 256 upwards for interrupts.
- * On the i386, this is assumed to be at most 16 bits, and the high bits
- * are used for a high-resolution timer.
  *
  */
 static void add_timer_randomness(struct timer_rand_state *state, unsigned num)
 {
-	__u32		time;
-	__s32		delta, delta2, delta3;
+	cycles_t	time;
+	long		delta, delta2, delta3;
 	int		entropy = 0;
 
 	/* if over the trickle threshold, use only 1 in 4096 samples */
@@ -814,22 +812,17 @@ static void add_timer_randomness(struct 
 	     (__get_cpu_var(trickle_count)++ & 0xfff))
 		return;
 
-#if defined (__i386__) || defined (__x86_64__)
-	if (cpu_has_tsc) {
-		__u32 high;
-		rdtsc(time, high);
-		num ^= high;
+	/*
+	 * Use get_cycles() if implemented, otherwise fall back to
+	 * jiffies.
+	 */
+	time = get_cycles();
+	if (time != 0) {
+		if (sizeof(time) > 4)
+			num ^= (u32)(time >> 32);
 	} else {
 		time = jiffies;
 	}
-#elif defined (__sparc_v9__)
-	unsigned long tick = tick_ops->get_tick();
-
-	time = (unsigned int) tick;
-	num ^= (tick >> 32UL);
-#else
-	time = jiffies;
-#endif
 
 	/*
 	 * Calculate number of bits of randomness we probably added.
diff -puN include/asm-i386/timex.h~using-get_cycles-for-add_timer_randomness include/asm-i386/timex.h
--- 25/include/asm-i386/timex.h~using-get_cycles-for-add_timer_randomness	2004-08-15 15:45:13.516016120 -0700
+++ 25-akpm/include/asm-i386/timex.h	2004-08-15 15:45:13.522015208 -0700
@@ -7,7 +7,7 @@
 #define _ASMi386_TIMEX_H
 
 #include <linux/config.h>
-#include <asm/msr.h>
+#include <asm/processor.h>
 
 #ifdef CONFIG_X86_ELAN
 #  define CLOCK_TICK_RATE 1189200 /* AMD Elan has different frequency! */
@@ -40,14 +40,17 @@ extern cycles_t cacheflush_time;
 
 static inline cycles_t get_cycles (void)
 {
+	unsigned long long ret=0;
+
 #ifndef CONFIG_X86_TSC
-	return 0;
-#else
-	unsigned long long ret;
+	if (!cpu_has_tsc)
+		return 0;
+#endif
 
+#if defined(CONFIG_X86_GENERIC) || defined(CONFIG_X86_TSC)
 	rdtscll(ret);
-	return ret;
 #endif
+	return ret;
 }
 
 extern unsigned long cpu_khz;
_

       reply	other threads:[~2004-08-15 22:50 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <200308160303.17612.arnd@arndb.de>
     [not found] ` <20040810162435.GK24690@krispykreme>
     [not found]   ` <20040814183623.GB5637@krispykreme>
2004-08-15 22:48     ` Andrew Morton [this message]
2004-08-15 23:46       ` Using get_cycles for add_timer_randomness Richard Mortimer
2004-08-15 23:59       ` David S. Miller
2004-08-16 16:17       ` Theodore Ts'o
2004-08-16 16:22         ` William Lee Irwin III
2004-08-16 16:44           ` Ralf Baechle
2004-08-16 16:49             ` William Lee Irwin III
2004-08-19 18:26               ` Ralf Baechle
2004-08-16 18:53 Luck, Tony
  -- strict thread matches above, loose matches on Subject: below --
2004-08-16 21:29 Luck, Tony

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=20040815154821.2d6edb03.akpm@osdl.org \
    --to=akpm@osdl.org \
    --cc=anton@samba.org \
    --cc=arnd@arndb.de \
    --cc=linux-arch@vger.kernel.org \
    --cc=richm@oldelvet.org.uk \
    --cc=tytso@mit.edu \
    /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