linux-kernel - Re: [PATCH 05/11] riscv: enable landing pad enforcement

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: <d6643810-094b-46d8-b223-b987783d23db@canonical.com>
Date: Fri, 25 Jul 2025 16:43:20 +0200
From: Heinrich Schuchardt <heinrich.schuchardt@...onical.com>
To: Deepak Gupta <debug@...osinc.com>
Cc: linux-riscv@...ts.infradead.org, linux-kernel@...r.kernel.org,
 linux-kbuild@...r.kernel.org, linux-mm@...ck.org, llvm@...ts.linux.dev,
 rick.p.edgecombe@...el.com, broonie@...nel.org, cleger@...osinc.com,
 samitolvanen@...gle.com, apatel@...tanamicro.com, ajones@...tanamicro.com,
 conor.dooley@...rochip.com, charlie@...osinc.com, samuel.holland@...ive.com,
 bjorn@...osinc.com, fweimer@...hat.com, jeffreyalaw@...il.com,
 andrew@...ive.com, ved@...osinc.com, Paul Walmsley
 <paul.walmsley@...ive.com>, Palmer Dabbelt <palmer@...belt.com>,
 Albert Ou <aou@...s.berkeley.edu>, Alexandre Ghiti <alex@...ti.fr>,
 Masahiro Yamada <masahiroy@...nel.org>, Nathan Chancellor
 <nathan@...nel.org>, Nicolas Schier <nicolas.schier@...ux.dev>,
 Andrew Morton <akpm@...ux-foundation.org>,
 David Hildenbrand <david@...hat.com>,
 Lorenzo Stoakes <lorenzo.stoakes@...cle.com>,
 "Liam R. Howlett" <Liam.Howlett@...cle.com>, Vlastimil Babka
 <vbabka@...e.cz>, Mike Rapoport <rppt@...nel.org>,
 Suren Baghdasaryan <surenb@...gle.com>, Michal Hocko <mhocko@...e.com>,
 Nick Desaulniers <nick.desaulniers+lkml@...il.com>,
 Bill Wendling <morbo@...gle.com>, Monk Chiang <monk.chiang@...ive.com>,
 Kito Cheng <kito.cheng@...ive.com>, Justin Stitt <justinstitt@...gle.com>
Subject: Re: [PATCH 05/11] riscv: enable landing pad enforcement

On 25.07.25 16:20, Deepak Gupta wrote:
> On Fri, Jul 25, 2025 at 08:33:46AM +0200, Heinrich Schuchardt wrote:
>> On 25.07.25 01:36, Deepak Gupta wrote:
>>> Enables landing pad enforcement by invoking a SBI FWFT call.
>>>
>>> Signed-off-by: Deepak Gupta <debug@...osinc.com>
>>> ---
>>>  arch/riscv/kernel/asm-offsets.c |  1 +
>>>  arch/riscv/kernel/head.S        | 19 +++++++++++++++++++
>>>  2 files changed, 20 insertions(+)
>>>
>>> diff --git a/arch/riscv/kernel/asm-offsets.c b/arch/riscv/kernel/asm- 
>>> offsets.c
>>> index e4d55126dc3e..e6a9fad86fae 100644
>>> --- a/arch/riscv/kernel/asm-offsets.c
>>> +++ b/arch/riscv/kernel/asm-offsets.c
>>> @@ -536,6 +536,7 @@ void asm_offsets(void)
>>>      DEFINE(SBI_EXT_FWFT, SBI_EXT_FWFT);
>>>      DEFINE(SBI_EXT_FWFT_SET, SBI_EXT_FWFT_SET);
>>>      DEFINE(SBI_FWFT_SHADOW_STACK, SBI_FWFT_SHADOW_STACK);
>>> +    DEFINE(SBI_FWFT_LANDING_PAD, SBI_FWFT_LANDING_PAD);
>>>      DEFINE(SBI_FWFT_SET_FLAG_LOCK, SBI_FWFT_SET_FLAG_LOCK);
>>>  #endif
>>>  }
>>> diff --git a/arch/riscv/kernel/head.S b/arch/riscv/kernel/head.S
>>> index 9c99c5ad6fe8..59af044bf85c 100644
>>> --- a/arch/riscv/kernel/head.S
>>> +++ b/arch/riscv/kernel/head.S
>>> @@ -185,6 +185,16 @@ secondary_start_sbi:
>>>  1:
>>>  #endif
>>>      scs_load_current
>>> +
>>> +#if defined(CONFIG_RISCV_SBI) && defined(CONFIG_RISCV_KERNEL_CFI)
>>> +    li a7, SBI_EXT_FWFT
>>> +    li a6, SBI_EXT_FWFT_SET
>>> +    li a0, SBI_FWFT_LANDING_PAD
>>> +    li a1, 1 /* enable landing pad for supervisor */
>>> +    li a2, SBI_FWFT_SET_FLAG_LOCK
>>> +    ecall    /* check for error condition and take appropriate 
>>> action */
>>> +#endif
>>> +
>>>      call smp_callin
>>>  #endif /* CONFIG_SMP */
>>> @@ -359,6 +369,15 @@ SYM_CODE_START(_start_kernel)
>>>  #endif
>>>      scs_load_current
>>> +#if defined(CONFIG_RISCV_SBI) && defined(CONFIG_RISCV_KERNEL_CFI)
>>> +    li a7, SBI_EXT_FWFT
>>> +    li a6, SBI_EXT_FWFT_SET
>>> +    li a0, SBI_FWFT_LANDING_PAD
>>> +    li a1, 1 /* enable landing pad for supervisor */
>>
>> The SBI specification calls BIT(0) "LOCK".
>> Shouldn't we define a constant for the lock bit instead of using a 
>> magic value?
> 
> See below `li a2, SBI_FWFT_SET_FLAG_LOCK`.
> 
> "li a1, 1 /* enable landing pad for supervisor */>" --> this is enabling.
> Had we done "li a1, 0 /* enable landing pad for supervisor */" --> this is
> asking firmware to disable the feature (turn off the bit in menvcfg CSR)

So we lack constants for

"Enable landing pad for supervisor-mode"
"Disable landing pad for supervisor-mode"

Best regards

Heinrich

>>
>>> +    li a2, SBI_FWFT_SET_FLAG_LOCK
>>> +    ecall    /* check for error condition and take appropriate 
>>> action */
>>> +#endif
>>> +
>>>  #ifdef CONFIG_KASAN
>>>      call kasan_early_init
>>>  #endif
>>>
>>