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 for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAPGkw+wtE0HMQmYsMkFEt_BPqqB2j_TQ6zwATp6zyXLyxxwTOA@mail.gmail.com>
Date:   Mon, 30 Aug 2021 00:11:22 +0200
From:   Krish Jain <krishjain02939@...il.com>
To:     Bryan Brattlof <hello@...anbrattlof.com>
Cc:     Greg KH <gregkh@...uxfoundation.org>,
        linux-staging@...ts.linux.dev, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] Declare the file_operations struct as const

On Sun, Aug 29, 2021 at 11:00 PM Bryan Brattlof <hello@...anbrattlof.com> wrote:
>
> On this day, August 29, 2021, thus sayeth Krish Jain:
> > Keeping you updated. Small win. The "Symbol version dump
> > "Module.symvers" is missing. " error disappeared. Now I still don't
> > know why
> >
>
> Whoop! Any win, no matter their size, always feel great. I ran around
> the house yesterday after cross compiling DOOM! for an armel chip. It's
> that "win" feeling you get that keeps me involved.
>
> It is important that you find out why though. What is the importance to
> having Module.symvers? and why is it a WARNING and not an ERROR?

 When a module is loaded/used, the values contained in the kernel are
compared with similar values in the module; if they are not equal, the
kernel refuses to load the module. I don't need it in my case.

> What would happen if we didn't have the proper symbols when compiling or
> installing this driver?
> How and what generates the Module.symvers file when we *do* need it?

The kernel would refuse to load the module.





> How can we turn this warning off when we don't need it?
>
> This is covered in chapter "6. Module Versioning"
>
>   https://www.kernel.org/doc/html/latest/kbuild/modules.html
>
> >
> > ERROR: Kernel configuration is invalid."; \
> > echo >&2 "         include/generated/autoconf.h or
> > include/config/auto.conf are missing.";\
> > echo >&2 "         Run 'make oldconfig && make prepare' on kernel src
> > to fix it."; \
> >
> >
> > is still present.
> >
> > How can I fix this?
> >
>
> Are there any other 'make *config' options we could try?

Yes, like main menuconfig. I tried it but it still doesn't work.

> What does 'make prepare' even do?


Prepares for different architectures etc.


> Why do we even need a configuration file?
>
>   https://www.kernel.org/doc/html/latest/kbuild/kconfig.html
>
> >
> > Best Regards
> >
> > On Sun, Aug 29, 2021 at 8:28 PM Krish Jain <krishjain02939@...il.com> wrote:
> > >
> > > Basically it says "you must have a prebuilt kernel available that
> > > contains the configuration and header files used in the build." Since
> > > for the staging kernel  "make oldconfig" asked me for  more
> > > configurations apart from my old configuration file (as it reads the
> > > existing .config file that was used for an old kernel and prompts the
> > > user for options in the current kernel source that are not found in
> > > the file) . So I *don't* currently have a prebuilt kernel that
> > > contains all the configuration in my staging kernel's .config file. So
> > > do I have to build the kernel once before I can just build the module
> > > with "make CCFLAGS=-Werror W=1 M=drivers/staging/android" ?
> > >
>
> What do all these other configuration settings turn on and off anyway?
>
> Do we really need CONFIG_INFINIBAND turned on if we're working in the
> drivers/staging tree of the kernel?


No, we don't. I removed it.

> What would we gain from having a compiled kernel if we want to test a
> single staging driver?

No need to compile the entire kernel I guess for my use case. But
after all this reading :( I still don't get why " sudo make
CCFLAGS=-Werror W=1 M=drivers/staging/android/  V=1" worked for you
but not for me. I still get the following errors


test -e include/generated/autoconf.h -a -e include/config/auto.conf || ( \
echo >&2; \
echo >&2 "  ERROR: Kernel configuration is invalid."; \
echo >&2 "         include/generated/autoconf.h or
include/config/auto.conf are missing.";\
echo >&2 "         Run 'make oldconfig && make prepare' on kernel src
to fix it."; \
echo >&2 ; \
/bin/false)
.....


How can I fix this?




> If you found what Module.symvers does, you should know this.
>
> > > > >
> > > > > Again, do not allow others to rob you of learning how to solve these
> > > > > issues yourself. I *strongly* encourage you to familiarize yourself with
> > > > > the Kernel Build System in the Documentation.
> > > > >
> > > > >   https://www.kernel.org/doc/html/latest/kbuild/modules.html
> > > > >
> > > > > Specifically the first paragraph of "2. How to Build External Modules"
> > > > >
> > > > > It may seem like a lot for such a simple issue but it *is* worth it.
> > > > > ~Bryan
> > > > >
> > > >
> > > > That section says
> > > >
> > > >
> > > > "To build external modules, *you must have a prebuilt kernel
> > > > available* that contains the configuration and header files used in
> > > > the build. Also, the kernel must have been built with modules enabled.
> > > > If you are using a distribution kernel, there will be a package for
> > > > the kernel you are running provided by your distribution.
> > > >
> > > > An alternative is to use the “make” target “modules_prepare.” This
> > > > will make sure the kernel contains the information required. The
> > > > target exists solely as a simple way to prepare a kernel source tree
> > > > for building external modules.
> > > >
> > > > NOTE: “modules_prepare” will not build Module.symvers even if
> > > > CONFIG_MODVERSIONS is set; therefore, *a full kernel build needs to be
> > > > executed to make module versioning work.*"
> > > >
> > > > So I am just trying to confirm with you whether I have to first build
> > > > the kernel with like "make" or not? As you can imagine my hardware
> > > > takes *very* long to build a kernel as I did in my last attempt so I
> > > > am asking whether it is needed. Hope you understand.
> > > >
>
> I understand. Though I still don't wish to rob you of this opportunity.
>
> Your ability to come up with these questions and answer them yourself is
> what will make you a better programmer and developer.
>
> Don't get me wrong. Greg knows all too well the garbage I can shovel his
> way. It's not about knowing the answer. It about knowing how to find the
> answer yourself.
>
> ~Bryan
>

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ