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-next>] [day] [month] [year] [list]
Message-ID: <20160720151211.BE19C3FACC@swsrvapps-01.diasemi.com>
Date:	Wed, 20 Jul 2016 15:53:58 +0100
From:	Steve Twiss <stwiss.opensource@...semi.com>
To:	LINUXKERNEL <linux-kernel@...r.kernel.org>,
	Liam Girdwood <lgirdwood@...il.com>,
	Mark Brown <broonie@...nel.org>
CC:	Support Opensource <support.opensource@...semi.com>
Subject: [PATCH V1] regulator: da9053/52: Fix incorrectly stated minimum and
 maximum voltage limits

From: Steve Twiss <stwiss.opensource@...semi.com>

This fix alters the minimum and maximum BUCK voltage limits for DA9052 and
DA9053. It does so for the following cases:

DA9052
- BUCK3 (MEM)
   min: 0.925V -> 0.950V
   max: 2.500V -> 2.525V

DA9053
- BUCK3 (MEM)
   min: 0.925V -> 0.950V
   max: 2.500V -> 2.525V
- BUCK4 (PERI)
   min: 0.925V -> 0.950V
   max: 2.500V -> 2.525V

The voltage range remains the same, but the limits are shifted by +0.025V.

This change is provided on DA9052:MEM, DA9053:MEM and DA9053:PERI
and is a voltage difference of 0.025V, compared to those measured before
this fix is applied. The patch has the effect of decreasing *all* measured
voltages on those BUCKs when compared against the previously measured
values for the same software voltage request.

For example, with this fix applied for DA9052:MEM, DA9053:MEM and
DA9053:PERI, the following is true.

Because the previous software defined slot 0 as being 0.925V, if a request
for 0.950V was previously sent, the slot 1 voltage would have been used.
This would have corresponded to an actual measured voltage of 0.975V. But,
with this patch fix, and with slot 0 properly aligned to 0.950V, if a
voltage of 0.950V is requested by software, a measured value of 0.950V will
be provided.

Tested-by: Steve Twiss <stwiss.opensource@...semi.com>
Signed-off-by: Steve Twiss <stwiss.opensource@...semi.com>

---
This patch applies against linux-next and v4.7-rc7

Hi Mark,

The advice for this patch is NOT to push these changes backwards into older
Linux stable kernels.

Although I have tested the patches for the current kernel linux-next/v4.7-rc7
and for the stable releases: v4.6.4, v4.5.7, v4.4.15 and v4.1.28; I have
NOT requested any Cc: <stable@...r.kernel.org> in this case.

The patch change will have the effect of decreasing *all* measured voltages
provided on DA9052:MEM, DA9053:MEM and DA9053:PERI by 0.025V compared to
those measured before the fix is applied.

Even though the new voltage measurement value will be correct, a small
decrease in the expected voltage could theoretically cause problems for
any system closely tuned to expect the previous voltage measure. This is
despite the voltage difference being only 0.025V (which at 1V will be below
the 3% tolerance limit of the BUCKs and LDOs). However, there could still be
a risk for existing systems that have been tuned exceptionally tightly,
even if this is a highly unlikely scenario.

That is the unknown here, and the basis for the recommendation. Even though
the risk is estimated to be very low in this case, because it is a
perceived risk to existing systems the patch should not be back-ported into
the linux-stable kernel trees.

Regards,
Steve


 drivers/regulator/da9052-regulator.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/regulator/da9052-regulator.c b/drivers/regulator/da9052-regulator.c
index 1050cb7..9ececfe 100644
--- a/drivers/regulator/da9052-regulator.c
+++ b/drivers/regulator/da9052-regulator.c
@@ -333,7 +333,7 @@ static const struct regulator_ops da9052_ldo_ops = {
 static struct da9052_regulator_info da9052_regulator_info[] = {
 	DA9052_DCDC(BUCK1, 25, 500, 2075, 6, 6, DA9052_SUPPLY_VBCOREGO),
 	DA9052_DCDC(BUCK2, 25, 500, 2075, 6, 6, DA9052_SUPPLY_VBPROGO),
-	DA9052_DCDC(BUCK3, 25, 925, 2500, 6, 6, DA9052_SUPPLY_VBMEMGO),
+	DA9052_DCDC(BUCK3, 25, 950, 2525, 6, 6, DA9052_SUPPLY_VBMEMGO),
 	DA9052_DCDC(BUCK4, 50, 1800, 3600, 5, 6, 0),
 	DA9052_LDO(LDO1, 50, 600, 1800, 5, 6, 0),
 	DA9052_LDO(LDO2, 25, 600, 1800, 6, 6, DA9052_SUPPLY_VLDO2GO),
@@ -350,8 +350,8 @@ static struct da9052_regulator_info da9052_regulator_info[] = {
 static struct da9052_regulator_info da9053_regulator_info[] = {
 	DA9052_DCDC(BUCK1, 25, 500, 2075, 6, 6, DA9052_SUPPLY_VBCOREGO),
 	DA9052_DCDC(BUCK2, 25, 500, 2075, 6, 6, DA9052_SUPPLY_VBPROGO),
-	DA9052_DCDC(BUCK3, 25, 925, 2500, 6, 6, DA9052_SUPPLY_VBMEMGO),
-	DA9052_DCDC(BUCK4, 25, 925, 2500, 6, 6, 0),
+	DA9052_DCDC(BUCK3, 25, 950, 2525, 6, 6, DA9052_SUPPLY_VBMEMGO),
+	DA9052_DCDC(BUCK4, 25, 950, 2525, 6, 6, 0),
 	DA9052_LDO(LDO1, 50, 600, 1800, 5, 6, 0),
 	DA9052_LDO(LDO2, 25, 600, 1800, 6, 6, DA9052_SUPPLY_VLDO2GO),
 	DA9052_LDO(LDO3, 25, 1725, 3300, 6, 6, DA9052_SUPPLY_VLDO3GO),
-- 
end-of-patch for PATCH V1

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ