Re: [PATCH v2 1/6] powerpc64/bpf: Move tail_call_cnt to bottom of stack frame

[Hari Bathini <hbathini@linux.ibm.com>]

On 17/01/26 4:11 pm, adubey wrote:
> On 2026-01-17 15:41, Hari Bathini wrote:
>> On 14/01/26 5:14 pm, adubey@linux.ibm.com wrote:
>>> From: Abhishek Dubey <adubey@linux.ibm.com>
>>>
>>> In the conventional stack frame, the position of tail_call_cnt
>>> is after the NVR save area (BPF_PPC_STACK_SAVE). Whereas, the
>>> offset of tail_call_cnt in the trampoline frame is after the
>>> stack alignment padding. BPF JIT logic could become complex
>>> when dealing with frame-sensitive offset calculation of
>>> tail_call_cnt. Having the same offset in both frames is the
>>> desired objective.
>>>
>>> The trampoline frame does not have a BPF_PPC_STACK_SAVE area.
>>> Introducing it leads to under-utilization of extra memory meant
>>> only for the offset alignment of tail_call_cnt.
>>> Another challenge is the variable alignment padding sitting at
>>> the bottom of the trampoline frame, which requires additional
>>> handling to compute tail_call_cnt offset.
>>>
>>> This patch addresses the above issues by moving tail_call_cnt
>>> to the bottom of the stack frame at offset 0 for both types
>>> of frames. This saves additional bytes required by BPF_PPC_STACK_SAVE
>>> in trampoline frame, and a common offset computation for
>>> tail_call_cnt serves both frames.
>>>
>>> The changes in this patch are required by the third patch in the
>>> series, where the 'reference to tail_call_info' of the main frame
>>> is copied into the trampoline frame from the previous frame.
>>>
>>> Signed-off-by: Abhishek Dubey <adubey@linux.ibm.com>
>>> ---
>>>   arch/powerpc/net/bpf_jit.h        |  4 ++++
>>>   arch/powerpc/net/bpf_jit_comp64.c | 31 ++++++++++++++++++++-----------
>>>   2 files changed, 24 insertions(+), 11 deletions(-)
>>>
>>> diff --git a/arch/powerpc/net/bpf_jit.h b/arch/powerpc/net/bpf_jit.h
>>> index 8334cd667bba..45d419c0ee73 100644
>>> --- a/arch/powerpc/net/bpf_jit.h
>>> +++ b/arch/powerpc/net/bpf_jit.h
>>> @@ -72,6 +72,10 @@
>>>       } } while (0)
>>>     #ifdef CONFIG_PPC64
>>> +
>>> +/* for tailcall counter */
>>> +#define BPF_PPC_TAILCALL        8
>>> +
>>>   /* If dummy pass (!image), account for maximum possible 
>>> instructions */
>>>   #define PPC_LI64(d, i)        do {                          \
>>>       if (!image)                                  \
>>> diff --git a/arch/powerpc/net/bpf_jit_comp64.c b/arch/powerpc/net/ 
>>> bpf_jit_comp64.c
>>> index 1fe37128c876..39061cd742c1 100644
>>> --- a/arch/powerpc/net/bpf_jit_comp64.c
>>> +++ b/arch/powerpc/net/bpf_jit_comp64.c
>>> @@ -20,13 +20,15 @@
>>>   #include "bpf_jit.h"
>>>     /*
>>> - * Stack layout:
>>> + * Stack layout 1:
>>> + * Layout when setting up our own stack frame.
>>> + * Note: r1 at bottom, component offsets positive wrt r1.
>>>    * Ensure the top half (upto local_tmp_var) stays consistent
>>>    * with our redzone usage.
>>>    *
>>>    *        [    prev sp        ] <-------------
>>> - *        [   nv gpr save area    ] 6*8        |
>>>    *        [    tail_call_cnt    ] 8        |
>>> + *        [   nv gpr save area    ] 6*8        |
>>>    *        [    local_tmp_var    ] 24        |
>>>    * fp (r31) -->    [   ebpf stack space    ] upto 512    |
>>>    *        [     frame header    ] 32/112    |
>>> @@ -36,10 +38,12 @@
>>>   /* for gpr non volatile registers BPG_REG_6 to 10 */
>>>   #define BPF_PPC_STACK_SAVE    (6*8)
>>>   /* for bpf JIT code internal usage */
>>> -#define BPF_PPC_STACK_LOCALS    32
>>> +#define BPF_PPC_STACK_LOCALS    24
>>>   /* stack frame excluding BPF stack, ensure this is quadword aligned */
>>>   #define BPF_PPC_STACKFRAME    (STACK_FRAME_MIN_SIZE + \
>>> -                 BPF_PPC_STACK_LOCALS + BPF_PPC_STACK_SAVE)
>>> +                 BPF_PPC_STACK_LOCALS + \
>>> +                 BPF_PPC_STACK_SAVE   + \
>>> +                 BPF_PPC_TAILCALL)
>>>     /* BPF register usage */
>>>   #define TMP_REG_1    (MAX_BPF_JIT_REG + 0)
>>> @@ -87,27 +91,32 @@ static inline bool bpf_has_stack_frame(struct 
>>> codegen_context *ctx)
>>>   }
>>>
>>
>>>   /*
>>> + * Stack layout 2:
>>>    * When not setting up our own stackframe, the redzone (288 bytes) 
>>> usage is:
>>> + * Note: r1 from prev frame. Component offset negative wrt r1.
>>>    *
>>>    *        [    prev sp        ] <-------------
>>>    *        [      ...           ]         |
>>>    * sp (r1) --->    [    stack pointer    ] --------------
>>> - *        [   nv gpr save area    ] 6*8
>>>    *        [    tail_call_cnt    ] 8
>>> + *        [   nv gpr save area    ] 6*8
>>>    *        [    local_tmp_var    ] 24
>>>    *        [   unused red zone    ] 224
>>>    */
>>
>> Calling it stack layout 1 & 2 is inappropriate. The stack layout
>> is essentially the same. It just goes to show things with reference
>> to r1 when stack is setup explicitly vs when redzone is being used...
> Agree. I am using it as labels to refer in comment. Any better suggestions?
I think the comments could refer to has stack frame vs Redzone case..

- Hari