From: Aaron Tomlin <atomlin@redhat.com>
To: Christophe Leroy <christophe.leroy@csgroup.eu>
Cc: "mcgrof@kernel.org" <mcgrof@kernel.org>,
"cl@linux.com" <cl@linux.com>,
"pmladek@suse.com" <pmladek@suse.com>,
"mbenes@suse.cz" <mbenes@suse.cz>,
"akpm@linux-foundation.org" <akpm@linux-foundation.org>,
"jeyu@kernel.org" <jeyu@kernel.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"linux-modules@vger.kernel.org" <linux-modules@vger.kernel.org>,
"live-patching@vger.kernel.org" <live-patching@vger.kernel.org>,
"atomlin@atomlin.com" <atomlin@atomlin.com>,
"ghalat@redhat.com" <ghalat@redhat.com>,
"allen.lkml@gmail.com" <allen.lkml@gmail.com>,
"void@manifault.com" <void@manifault.com>,
"joe@perches.com" <joe@perches.com>,
"msuchanek@suse.de" <msuchanek@suse.de>,
"oleksandr@natalenko.name" <oleksandr@natalenko.name>
Subject: Re: [PATCH v5 09/13] module: Move kallsyms support into a separate file
Date: Fri, 11 Feb 2022 19:55:49 +0000 [thread overview]
Message-ID: <CANfR36go+eyhkX5rHDps4Re6Z2gt7MS3iGMM0vu6wEsYF1B07w@mail.gmail.com> (raw)
In-Reply-To: <1d6dde1d-e819-b659-0239-5d42ab9bd087@csgroup.eu>
On Thu 2022-02-10 13:43 +0000, Christophe Leroy wrote:
> Checkpatch reports:
>
> total: 3 errors, 1 warnings, 26 checks, 1103 lines checked
Christophe,
> Sparse reports the following:
>
> CHECK kernel/module/kallsyms.c
> kernel/module/kallsyms.c:174:23: warning: incorrect type in assignment
> (different address spaces)
> kernel/module/kallsyms.c:174:23: expected struct mod_kallsyms
> [noderef] __rcu *kallsyms
> kernel/module/kallsyms.c:174:23: got void *
Thanks once again for your review and feedback!
Indeed I can see the same via 'make C=2 kernel/module/'. Looking at struct
'module' declaration we see that field namely "kallsyms" has the __rcu
marker. So, If I understand correctly, perhaps this can be resolved as
follows, to be more explicit:
@@ -171,7 +171,7 @@ void add_kallsyms(struct module *mod, const struct
load_info *info)
Elf_Shdr *symsec = &info->sechdrs[info->index.sym];
/* Set up to point into init section. */
- mod->kallsyms = mod->init_layout.base + info->mod_kallsyms_init_off;
+ mod->kallsyms = (struct mod_kallsyms __rcu
*)mod->init_layout.base + info->mod_kallsyms_init_off;
> kernel/module/kallsyms.c:176:12: warning: dereference of noderef expression
> kernel/module/kallsyms.c:177:12: warning: dereference of noderef expression
> kernel/module/kallsyms.c:179:12: warning: dereference of noderef expression
> kernel/module/kallsyms.c:180:12: warning: dereference of noderef expression
> kernel/module/kallsyms.c:189:18: warning: dereference of noderef expression
> kernel/module/kallsyms.c:190:35: warning: dereference of noderef expression
> kernel/module/kallsyms.c:191:20: warning: dereference of noderef expression
> kernel/module/kallsyms.c:196:32: warning: dereference of noderef expression
> kernel/module/kallsyms.c:199:45: warning: dereference of noderef expression
I will use rcu_dereference*() for the above since the pointer should not be
accessed directly.
> >
> > diff --git a/kernel/module/Makefile b/kernel/module/Makefile
> > index 62c9fc91d411..868b13c06920 100644
> > --- a/kernel/module/Makefile
> > +++ b/kernel/module/Makefile
> > @@ -12,4 +12,5 @@ obj-$(CONFIG_LIVEPATCH) += livepatch.o
> > obj-$(CONFIG_MODULES_TREE_LOOKUP) += tree_lookup.o
> > obj-$(CONFIG_STRICT_MODULE_RWX) += strict_rwx.o
> > obj-$(CONFIG_DEBUG_KMEMLEAK) += debug_kmemleak.o
> > +obj-$(CONFIG_KALLSYMS) += kallsyms.o
> > endif
> > diff --git a/kernel/module/internal.h b/kernel/module/internal.h
> > index 33d7befd0602..7973666452c3 100644
> > --- a/kernel/module/internal.h
> > +++ b/kernel/module/internal.h
> > @@ -69,6 +69,11 @@ struct load_info {
> > };
> >
> > int mod_verify_sig(const void *mod, struct load_info *info);
> > +struct module *find_module_all(const char *name, size_t len, bool even_unformed);
> > +unsigned long kernel_symbol_value(const struct kernel_symbol *sym);
>
> This function is small enought to be a 'static inline' in internal.h
Fair enough.
> > +int cmp_name(const void *name, const void *sym);
> > +long get_offset(struct module *mod, unsigned int *size, Elf_Shdr *sechdr,
> > + unsigned int section);
>
> Having a non static function called get_offset() seems dangerous.
>
> There are already several get_offset() functions in the kernel allthough
> they are all static.
>
> It takes a struct module as an argument so it could be called
> module_get_offset()
The rename is a good idea.
> > +bool sect_empty(const Elf_Shdr *sect);
>
> sect_empty() is small enough to remain a static inline.
Yes and moved to kernel/module/internal.h.
>
> > +const char *find_kallsyms_symbol(struct module *mod, unsigned long addr,
> > + unsigned long *size, unsigned long *offset);
>
> This is not used outside kallsyms.c, no need to have it in internal.h
Agreed.
>
> > +#else /* !CONFIG_KALLSYMS */
> > +static inline void layout_symtab(struct module *mod, struct load_info *info) { }
> > +static inline void add_kallsyms(struct module *mod, const struct load_info *info) { }
> > +static inline char *find_kallsyms_symbol(struct module *mod, unsigned long addr,
> > + unsigned long *size, unsigned long *offset)
>
> This is not used outside kallsyms.c, no need to have it when
> !CONFIG_KALLSYMS
Agreed.
>
> > +{
> > + return NULL;
> > +}
> > +#endif /* CONFIG_KALLSYMS */
> > diff --git a/kernel/module/kallsyms.c b/kernel/module/kallsyms.c
> > new file mode 100644
> > index 000000000000..ed28f6310701
> > --- /dev/null
> > +++ b/kernel/module/kallsyms.c
> > @@ -0,0 +1,502 @@
> ...
> > +
> > +/* Given a module and name of symbol, find and return the symbol's value */
> > +static unsigned long find_kallsyms_symbol_value(struct module *mod, const char *name)
>
> This function is called from main.c, it can't be static and must be
> defined in internal.h
Agreed. This was an unfortunate oversight.
> > -static unsigned long kernel_symbol_value(const struct kernel_symbol *sym)
> > +unsigned long kernel_symbol_value(const struct kernel_symbol *sym)
>
> This function is small enought to become a 'static inline' in internal.h
Agreed.
> > +int cmp_name(const void *name, const void *sym)
>
> This function is small enought to become a 'static inline' in internal.h
Agreed.
Kind regards,
--
Aaron Tomlin
next prev parent reply other threads:[~2022-02-11 19:56 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-02-09 17:08 [PATCH v5 07/13] module: Move extra signature support out of core code Aaron Tomlin
2022-02-09 17:08 ` [PATCH v5 08/13] module: Move kmemleak support to a separate file Aaron Tomlin
2022-02-10 13:07 ` Christophe Leroy
2022-02-11 15:42 ` Aaron Tomlin
2022-02-09 17:08 ` [PATCH v5 09/13] module: Move kallsyms support into " Aaron Tomlin
2022-02-10 13:43 ` Christophe Leroy
2022-02-11 19:55 ` Aaron Tomlin [this message]
2022-02-09 17:08 ` [PATCH v5 10/13] module: Move procfs " Aaron Tomlin
2022-02-10 13:46 ` Christophe Leroy
2022-02-09 20:48 ` [PATCH v5 07/13] module: Move extra signature support out of core code Michal Suchánek
2022-02-10 9:59 ` Aaron Tomlin
2022-02-10 13:01 ` Christophe Leroy
2022-02-11 13:51 ` Aaron Tomlin
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=CANfR36go+eyhkX5rHDps4Re6Z2gt7MS3iGMM0vu6wEsYF1B07w@mail.gmail.com \
--to=atomlin@redhat.com \
--cc=akpm@linux-foundation.org \
--cc=allen.lkml@gmail.com \
--cc=atomlin@atomlin.com \
--cc=christophe.leroy@csgroup.eu \
--cc=cl@linux.com \
--cc=ghalat@redhat.com \
--cc=jeyu@kernel.org \
--cc=joe@perches.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-modules@vger.kernel.org \
--cc=live-patching@vger.kernel.org \
--cc=mbenes@suse.cz \
--cc=mcgrof@kernel.org \
--cc=msuchanek@suse.de \
--cc=oleksandr@natalenko.name \
--cc=pmladek@suse.com \
--cc=void@manifault.com \
/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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).