linux-kernel - Re: [PATCH] soc: qcom: icc-bwmon: Add tracepoints in bwmon_intr

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: <xigukwbvqld5snjykp6h7qwob3r55w4evacbrp7svnzkqrmau3@enk5xlu5bbqn>
Date: Thu, 20 Jun 2024 23:36:39 +0300
From: Dmitry Baryshkov <dmitry.baryshkov@...aro.org>
To: Konrad Dybcio <konrad.dybcio@...aro.org>
Cc: Shivnandan Kumar <quic_kshivnan@...cinc.com>, 
	Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>, Bjorn Andersson <andersson@...nel.org>, 
	linux-kernel@...r.kernel.org, linux-arm-msm@...r.kernel.org, quic_rgottimu@...cinc.com
Subject: Re: [PATCH] soc: qcom: icc-bwmon: Add tracepoints in
 bwmon_intr_thread

On Thu, Jun 20, 2024 at 12:48:04PM GMT, Konrad Dybcio wrote:
> 
> 
> On 6/20/24 10:55, Shivnandan Kumar wrote:
> > 
> > 
> > On 6/20/2024 12:46 AM, Konrad Dybcio wrote:
> > > 
> > > 
> > > On 6/19/24 15:51, Shivnandan Kumar wrote:
> > > > Add tracepoint for tracing the measured traffic in kbps,
> > > > up_kbps and down_kbps in bwmon. This information is valuable
> > > > for understanding what bwmon hw measures at the system cache
> > > > level and at the DDR level which is helpful in debugging
> > > > bwmon behavior.
> > > > 
> > > > Signed-off-by: Shivnandan Kumar <quic_kshivnan@...cinc.com>
> > > > ---
> > > >   MAINTAINERS                        |  1 +
> > > >   drivers/soc/qcom/icc-bwmon.c       |  7 +++--
> > > >   drivers/soc/qcom/trace_icc-bwmon.h | 49 ++++++++++++++++++++++++++++++
> > > >   3 files changed, 55 insertions(+), 2 deletions(-)
> > > >   create mode 100644 drivers/soc/qcom/trace_icc-bwmon.h
> > > > 
> > > > diff --git a/MAINTAINERS b/MAINTAINERS
> > > > index 242fc612fbc5..1b410c0183bb 100644
> > > > --- a/MAINTAINERS
> > > > +++ b/MAINTAINERS
> > > > @@ -18573,6 +18573,7 @@ M:    Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>
> > > >   L:    linux-arm-msm@...r.kernel.org
> > > >   S:    Maintained
> > > >   F: Documentation/devicetree/bindings/interconnect/qcom,msm8998-bwmon.yaml
> > > > +F:    drivers/soc/qcom/trace_icc-bwmon.h
> > > >   F:    drivers/soc/qcom/icc-bwmon.c
> > > > 
> > > >   QUALCOMM IOMMU
> > > > diff --git a/drivers/soc/qcom/icc-bwmon.c b/drivers/soc/qcom/icc-bwmon.c
> > > > index fb323b3364db..9b5ac1e62673 100644
> > > > --- a/drivers/soc/qcom/icc-bwmon.c
> > > > +++ b/drivers/soc/qcom/icc-bwmon.c
> > > > @@ -17,6 +17,8 @@
> > > >   #include <linux/pm_opp.h>
> > > >   #include <linux/regmap.h>
> > > >   #include <linux/sizes.h>
> > > > +#define CREATE_TRACE_POINTS
> > > > +#include "trace_icc-bwmon.h"
> > > > 
> > > >   /*
> > > >    * The BWMON samples data throughput within 'sample_ms' time. With three
> > > > @@ -645,9 +647,9 @@ static irqreturn_t bwmon_intr_thread(int irq, void *dev_id)
> > > >       struct icc_bwmon *bwmon = dev_id;
> > > >       unsigned int irq_enable = 0;
> > > >       struct dev_pm_opp *opp, *target_opp;
> > > > -    unsigned int bw_kbps, up_kbps, down_kbps;
> > > > +    unsigned int bw_kbps, up_kbps, down_kbps, meas_kbps;
> > > > 
> > > > -    bw_kbps = bwmon->target_kbps;
> > > > +    meas_kbps = bw_kbps = bwmon->target_kbps;
> > > > 
> > > >       target_opp = dev_pm_opp_find_bw_ceil(bwmon->dev, &bw_kbps, 0);
> > > 
> > > This breaks bwmon, as dev_pm_opp_find_bw_ceil is now fed a random
> > > (uninitialized variable) value
> > > 
> > 
> > Thank you for reviewing the patch.
> > I didn't get it, still the variable "bw_kbps" is being initialized along with "meas_kbps". Which variable are you referring to as being fed to dev_pm_opp_find_bw_ceil  with an uninitialized value?
> 
> Oh this one's on me, I skipped over the middle assignment.. Sorry!

Still:

CHECK: multiple assignments should be avoided
#57: FILE: drivers/soc/qcom/icc-bwmon.c:652:
+	meas_kbps = bw_kbps = bwmon->target_kbps;

While we are at it:

CHECK: Alignment should match open parenthesis
#88: FILE: drivers/soc/qcom/trace_icc-bwmon.h:14:
+TRACE_EVENT(qcom_bwmon_update,
+

CHECK: Alignment should match open parenthesis
#109: FILE: drivers/soc/qcom/trace_icc-bwmon.h:35:
+	TP_printk("name=%s meas_kbps=%u up_kbps=%u down_kbps=%u",
+		__get_str(name),


-- 
With best wishes
Dmitry