Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Mime-Version: 1.0
Subject: RE: [PATCH modules-next 1/1] kallsyms: enhance %pS/s/b printing
 when KALLSYSMS is disabled
Reply-To: maninder1.s@samsung.com
Sender: Maninder Singh <maninder1.s@samsung.com>
From:   Maninder Singh <maninder1.s@samsung.com>
To:     Kees Cook <keescook@chromium.org>
CC:     "mcgrof@kernel.org" <mcgrof@kernel.org>,
        "pmladek@suse.com" <pmladek@suse.com>,
        "rostedt@goodmis.org" <rostedt@goodmis.org>,
        "senozhatsky@chromium.org" <senozhatsky@chromium.org>,
        "andriy.shevchenko@linux.intel.com" 
        <andriy.shevchenko@linux.intel.com>,
        "linux@rasmusvillemoes.dk" <linux@rasmusvillemoes.dk>,
        "akpm@linux-foundation.org" <akpm@linux-foundation.org>,
        "wangkefeng.wang@huawei.com" <wangkefeng.wang@huawei.com>,
        Vaneet Narang <v.narang@samsung.com>,
        "swboyd@chromium.org" <swboyd@chromium.org>,
        "ojeda@kernel.or" <ojeda@kernel.or>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "linux-modules@vger.kernel.org" <linux-modules@vger.kernel.org>,
        "avimalin@gmail.com" <avimalin@gmail.com>,
        "atomlin@redhat.com" <atomlin@redhat.com>,
        Onkarnath <onkarnath.1@samsung.com>
In-Reply-To: <202205111513.3AD646936@keescook>
Message-ID: <20220512034437epcms5p1b384e18e75f0f56ee70f67e62f7ab0ce@epcms5p1>
Date:   Thu, 12 May 2022 09:14:37 +0530
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset="utf-8"
CMS-TYPE: 105P
References: <202205111513.3AD646936@keescook>
        <20220421041542.9195-1-maninder1.s@samsung.com>
        <CGME20220421041604epcas5p414cf2e851d1bd828a151dc6b2b004680@epcms5p1>
Precedence: bulk

Hi,

> On Wed, May 11, 2022 at 01:36:56PM +0530, Maninder Singh wrote:
> > kallsyms_show_value return false if KALLSYMS is disabled,
> > but its usage is done by module.c also.
> > Thus when KALLSYMS is disabled, system will not print module
> > load address:
> 
> Eek, I hadn't see the other changes this depends on. I think those
> changes need to be reworked first. Notably in the other patch, this is
> no good:
> 
>         /* address belongs to module */
>         if (add_offset)
>                 len = sprintf(buf, "0x%p+0x%lx", base, offset);
>         else
>                 len = sprintf(buf, "0x%lx", value);
> 
> This is printing raw kernel addresses with no hashing, as far as I can
> tell. That's not okay at all.
>

yes same was suggested by Petr also, because earlier we were printing base address also as raw address.

https://lkml.org/lkml/2022/2/28/847

but then modified approach to print base address as hash when we are going to show offset of module address,
but when we print complete address then we thought of keeping it same as it was:

original:
 [12.487424] ps 0xffff800000eb008c
with patch:
 [9.624152] ps 0xffff800001bd008c [crash]

But if its has to be hashed, will fix that also.

> Once that other patch gets fixed, this one then can be revisited.
> 

I will check detailed comments on that also

> And just on naming: "kallsyms_tiny" is a weird name: it's just "ksyms"
> -- there's no "all".  :)

Ok :)

Will name it as knosyms.c (if it seems ok).


Thanks,
Maninder Singh