linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [RESEND 1/3] hwrng: st: dt: Fix trivial typo in node address
@ 2015-10-06 16:12 Lee Jones
  2015-10-06 16:12 ` [RESEND 2/3] hwrng: st: Use real-world device timings for timeout Lee Jones
  2015-10-06 16:12 ` [RESEND 3/3] hwrng: st: Report correct FIFO size Lee Jones
  0 siblings, 2 replies; 5+ messages in thread
From: Lee Jones @ 2015-10-06 16:12 UTC (permalink / raw)
  To: linux-arm-kernel, linux-kernel
  Cc: kernel, herbert, linux-crypto, peter, festevam, pankaj.dev,
	daniel.thompson, Lee Jones

DT nodes should not append their addresses with '0x'.

Suggested-by: Stephen Boyd <sboyd@codeaurora.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
---
 Documentation/devicetree/bindings/rng/st,rng.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/rng/st,rng.txt b/Documentation/devicetree/bindings/rng/st,rng.txt
index dbc64e6..35734bc 100644
--- a/Documentation/devicetree/bindings/rng/st,rng.txt
+++ b/Documentation/devicetree/bindings/rng/st,rng.txt
@@ -8,7 +8,7 @@ clocks		: Phandle to device's clock (See: ../clocks/clock-bindings.txt)
 
 Example:
 
-rng@0xfee80000 {
+rng@fee80000 {
 	compatible      = "st,rng";
 	reg		= <0xfee80000 0x1000>;
 	clocks          = <&clk_sysin>;
-- 
1.9.1


^ permalink raw reply related	[flat|nested] 5+ messages in thread

* [RESEND 2/3] hwrng: st: Use real-world device timings for timeout
  2015-10-06 16:12 [RESEND 1/3] hwrng: st: dt: Fix trivial typo in node address Lee Jones
@ 2015-10-06 16:12 ` Lee Jones
  2015-10-06 16:12 ` [RESEND 3/3] hwrng: st: Report correct FIFO size Lee Jones
  1 sibling, 0 replies; 5+ messages in thread
From: Lee Jones @ 2015-10-06 16:12 UTC (permalink / raw)
  To: linux-arm-kernel, linux-kernel
  Cc: kernel, herbert, linux-crypto, peter, festevam, pankaj.dev,
	daniel.thompson, Lee Jones

Samples are documented to be available every 0.667us, so in theory
the 8 sample deep FIFO should take 5.336us to fill.  However, during
thorough testing, it became apparent that filling the FIFO actually
takes closer to 12us.

Signed-off-by: Lee Jones <lee.jones@linaro.org>
---
 drivers/char/hw_random/st-rng.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)

diff --git a/drivers/char/hw_random/st-rng.c b/drivers/char/hw_random/st-rng.c
index 8c8a435..a1ff080 100644
--- a/drivers/char/hw_random/st-rng.c
+++ b/drivers/char/hw_random/st-rng.c
@@ -32,8 +32,13 @@
 #define ST_RNG_FIFO_SIZE		8
 #define ST_RNG_SAMPLE_SIZE		2 /* 2 Byte (16bit) samples */
 
-/* Samples are available every 0.667us, which we round to 1us */
-#define ST_RNG_FILL_FIFO_TIMEOUT   (1 * (ST_RNG_FIFO_SIZE / ST_RNG_SAMPLE_SIZE))
+/*
+ * Samples are documented to be available every 0.667us, so in theory
+ * the 8 sample deep FIFO should take 5.336us to fill.  However, during
+ * thorough testing, it became apparent that filling the FIFO actually
+ * takes closer to 12us.
+ */
+#define ST_RNG_FILL_FIFO_TIMEOUT	12
 
 struct st_rng_data {
 	void __iomem	*base;
-- 
1.9.1


^ permalink raw reply related	[flat|nested] 5+ messages in thread

* [RESEND 3/3] hwrng: st: Report correct FIFO size
  2015-10-06 16:12 [RESEND 1/3] hwrng: st: dt: Fix trivial typo in node address Lee Jones
  2015-10-06 16:12 ` [RESEND 2/3] hwrng: st: Use real-world device timings for timeout Lee Jones
@ 2015-10-06 16:12 ` Lee Jones
  1 sibling, 0 replies; 5+ messages in thread
From: Lee Jones @ 2015-10-06 16:12 UTC (permalink / raw)
  To: linux-arm-kernel, linux-kernel
  Cc: kernel, herbert, linux-crypto, peter, festevam, pankaj.dev,
	daniel.thompson, Lee Jones

The values supplied to the 'read random data from FIFO' arithmetic
are not correct.  The value fed in to initialise the iterator
describes the FIFO depth, but then the iterator is treated in
Bytes and subsequently increased by 2 in value for every read
word.  This means only 4 of the 8 available values are being read
during each invocation of .read().

This change increased the device bandwidth by a factor of 2.

Reported-by: Daniel Thompson <daniel.thompson@linaro.org>
Reviewed-by: Daniel Thompson <daniel.thompson@linaro.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
---
 drivers/char/hw_random/st-rng.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/char/hw_random/st-rng.c b/drivers/char/hw_random/st-rng.c
index a1ff080..3b1432c 100644
--- a/drivers/char/hw_random/st-rng.c
+++ b/drivers/char/hw_random/st-rng.c
@@ -29,8 +29,9 @@
 #define ST_RNG_STATUS_BAD_ALTERNANCE	BIT(1)
 #define ST_RNG_STATUS_FIFO_FULL		BIT(5)
 
-#define ST_RNG_FIFO_SIZE		8
 #define ST_RNG_SAMPLE_SIZE		2 /* 2 Byte (16bit) samples */
+#define ST_RNG_FIFO_DEPTH		8
+#define ST_RNG_FIFO_SIZE		(ST_RNG_FIFO_DEPTH * ST_RNG_SAMPLE_SIZE)
 
 /*
  * Samples are documented to be available every 0.667us, so in theory
-- 
1.9.1


^ permalink raw reply related	[flat|nested] 5+ messages in thread

* Re: [RESEND 1/3] hwrng: st: dt: Fix trivial typo in node address
  2015-10-06 14:43 [RESEND 1/3] hwrng: st: dt: Fix trivial typo in node address Lee Jones
@ 2015-10-06 15:24 ` Herbert Xu
  0 siblings, 0 replies; 5+ messages in thread
From: Herbert Xu @ 2015-10-06 15:24 UTC (permalink / raw)
  To: Lee Jones
  Cc: linux-arm-kernel, linux-kernel, kernel, peter, festevam,
	pankaj.dev, daniel.thompson

On Tue, Oct 06, 2015 at 03:43:58PM +0100, Lee Jones wrote:
> DT nodes should not append their addresses with '0x'.
> 
> Suggested-by: Stephen Boyd <sboyd@codeaurora.org>
> Signed-off-by: Lee Jones <lee.jones@linaro.org>

If this is supposed to go in via the crypto tree it needs to be
posted to linux-crypto@vger.kernel.org.

Thanks,
-- 
Email: Herbert Xu <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [RESEND 1/3] hwrng: st: dt: Fix trivial typo in node address
@ 2015-10-06 14:43 Lee Jones
  2015-10-06 15:24 ` Herbert Xu
  0 siblings, 1 reply; 5+ messages in thread
From: Lee Jones @ 2015-10-06 14:43 UTC (permalink / raw)
  To: linux-arm-kernel, linux-kernel
  Cc: kernel, herbert, peter, festevam, pankaj.dev, daniel.thompson, Lee Jones

DT nodes should not append their addresses with '0x'.

Suggested-by: Stephen Boyd <sboyd@codeaurora.org>
Signed-off-by: Lee Jones <lee.jones@linaro.org>
---
 Documentation/devicetree/bindings/rng/st,rng.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/rng/st,rng.txt b/Documentation/devicetree/bindings/rng/st,rng.txt
index dbc64e6..35734bc 100644
--- a/Documentation/devicetree/bindings/rng/st,rng.txt
+++ b/Documentation/devicetree/bindings/rng/st,rng.txt
@@ -8,7 +8,7 @@ clocks		: Phandle to device's clock (See: ../clocks/clock-bindings.txt)
 
 Example:
 
-rng@0xfee80000 {
+rng@fee80000 {
 	compatible      = "st,rng";
 	reg		= <0xfee80000 0x1000>;
 	clocks          = <&clk_sysin>;
-- 
1.9.1


^ permalink raw reply related	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2015-10-06 16:15 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2015-10-06 16:12 [RESEND 1/3] hwrng: st: dt: Fix trivial typo in node address Lee Jones
2015-10-06 16:12 ` [RESEND 2/3] hwrng: st: Use real-world device timings for timeout Lee Jones
2015-10-06 16:12 ` [RESEND 3/3] hwrng: st: Report correct FIFO size Lee Jones
  -- strict thread matches above, loose matches on Subject: below --
2015-10-06 14:43 [RESEND 1/3] hwrng: st: dt: Fix trivial typo in node address Lee Jones
2015-10-06 15:24 ` Herbert Xu

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).