[-next,V8,00/14] riscv: Add GENERIC_ENTRY support and related features

  From: Guo Ren <guoren@linux.alibaba.com>

The patches convert riscv to use the generic entry infrastructure from
kernel/entry/*. Additionally, add independent irq stacks (IRQ_STACKS)
for percpu to prevent kernel stack overflows. Add generic_entry based
STACKLEAK support. Some optimization for entry.S with new .macro and
merge ret_from_kernel_thread into ret_from_fork.

We have tested it with rv64, rv32, rv64 + 32rootfs, all are passed.

You can directly try it with:
[1] https://github.com/guoren83/linux/tree/generic_entry_v8

Any reviews and tests are helpful.

v8:
 - Rebase on palmer/for-next branch (20221102)
 - Add save/restore_from_x5_to_x31 .macro (JishengZhang)
 - Consolidate ret_from_kernel_thread into ret_from_fork (JishengZhang)
 - Optimize __noinstr_section comment (JiangshanLai)

v7:
https://lore.kernel.org/linux-riscv/20221015114702.3489989-1-guoren@kernel.org/
 - Fixup regs_irqs_disabled with SR_PIE
 - Optimize stackleak_erase -> stackleak_erase_on_task_stack (Thx Mark
   Rutland)
 - Add BUG_ON(!irqs_disabled()) in trap handlers
 - Using regs_irqs_disabled in __do_page_fault
 - Remove unnecessary irq disable in ret_from_exception and add comment 

v6:
https://lore.kernel.org/linux-riscv/20221002012451.2351127-1-guoren@kernel.org/
 - Use THEAD_SIZE_ORDER for thread size adjustment in kconfig (Thx Arnd)
 - Move call_on_stack to inline style (Thx Peter Zijlstra)
 - Fixup fp chain broken (Thx Chen Zhongjin)
 - Remove common entry modification, and fixup page_fault entry (Thx
   Peter Zijlstra)
 - Treat some traps as nmi entry (Thx Peter Zijlstra)

v5:
https://lore.kernel.org/linux-riscv/20220918155246.1203293-1-guoren@kernel.org/
 - Add riscv own stackleak patch instead of generic entry modification
   (by Mark Rutland)
 - Add EXPERT dependency for THREAD_SIZE (by Arnd)
 - Add EXPERT dependency for IRQ_STACK (by Sebastian, David Laight)
 - Corrected __trap_section (by Peter Zijlstra)
 - Add Tested-by (Yipeng Zou)
 - Use CONFIG_SOFTIRQ_ON_OWN_STACK replace "#ifndef CONFIG_PREEMPT_RT"
 - Fixup systrace_enter compile error
 - Fixup exit_to_user_mode_prepare preempt_disable warning

V4:
https://lore.kernel.org/linux-riscv/20220908022506.1275799-1-guoren@kernel.org/
 - Fixup entry.S with "la" bug (by Conor.Dooley)
 - Fixup missing noinstr bug (by Peter Zijlstra)

V3:
https://lore.kernel.org/linux-riscv/20220906035423.634617-1-guoren@kernel.org/
 - Fixup CONFIG_COMPAT=n compile error
 - Add THREAD_SIZE_ORDER config
 - Optimize elf_kexec.c warning fixup
 - Add static to irq_stack_ptr definition

V2:
https://lore.kernel.org/linux-riscv/20220904072637.8619-1-guoren@kernel.org/
 - Fixup compile error by include "riscv: ptrace: Remove duplicate
   operation"
 - Fixup compile warning
   Reported-by: kernel test robot <lkp@intel.com>
 - Add test repo link in cover letter

V1:
https://lore.kernel.org/linux-riscv/20220903163808.1954131-1-guoren@kernel.org/


Dao Lu (1):
  riscv: Add support for STACKLEAK gcc plugin

Guo Ren (9):
  riscv: elf_kexec: Fixup compile warning
  riscv: compat_syscall_table: Fixup compile warning
  riscv: ptrace: Remove duplicate operation
  riscv: traps: Add noinstr to prevent instrumentation inserted
  riscv: convert to generic entry
  riscv: Support HAVE_IRQ_EXIT_ON_IRQ_STACK
  riscv: Support HAVE_SOFTIRQ_ON_OWN_STACK
  riscv: Add config of thread stack size
  riscv: Typo fixup for addi -> andi in comment

Jisheng Zhang (3):
  riscv: remove extra level wrappers of trace_hardirqs_{on,off}
  riscv: consolidate ret_from_kernel_thread into ret_from_fork
  riscv: entry: consolidate general regs saving/restoring

Lai Jiangshan (1):
  compiler_types.h: Add __noinstr_section() for noinstr

 arch/riscv/Kconfig                    |  21 ++
 arch/riscv/include/asm/asm.h          |  63 +++++
 arch/riscv/include/asm/csr.h          |   1 -
 arch/riscv/include/asm/entry-common.h |   8 +
 arch/riscv/include/asm/ptrace.h       |  10 +-
 arch/riscv/include/asm/stacktrace.h   |   5 +
 arch/riscv/include/asm/syscall.h      |   6 +
 arch/riscv/include/asm/thread_info.h  |  27 +--
 arch/riscv/include/asm/vmap_stack.h   |  28 +++
 arch/riscv/kernel/Makefile            |   3 +-
 arch/riscv/kernel/elf_kexec.c         |   2 +-
 arch/riscv/kernel/entry.S             | 327 +++-----------------------
 arch/riscv/kernel/irq.c               | 110 +++++++++
 arch/riscv/kernel/mcount-dyn.S        |  56 +----
 arch/riscv/kernel/process.c           |   5 +-
 arch/riscv/kernel/ptrace.c            |  44 ----
 arch/riscv/kernel/signal.c            |  21 +-
 arch/riscv/kernel/sys_riscv.c         |  27 +++
 arch/riscv/kernel/trace_irq.c         |  27 ---
 arch/riscv/kernel/trace_irq.h         |  11 -
 arch/riscv/kernel/traps.c             |  74 ++++--
 arch/riscv/mm/fault.c                 |  16 +-
 drivers/firmware/efi/libstub/Makefile |   2 +-
 include/linux/compiler_types.h        |  15 +-
 24 files changed, 407 insertions(+), 502 deletions(-)
 create mode 100644 arch/riscv/include/asm/entry-common.h
 create mode 100644 arch/riscv/include/asm/vmap_stack.h
 delete mode 100644 arch/riscv/kernel/trace_irq.c
 delete mode 100644 arch/riscv/kernel/trace_irq.h
  

On Thu, Nov 03, 2022 at 03:50:33AM -0400, guoren@kernel.org wrote:
> From: Guo Ren <guoren@linux.alibaba.com>
> 
> The patches convert riscv to use the generic entry infrastructure from
> kernel/entry/*. Additionally, add independent irq stacks (IRQ_STACKS)
> for percpu to prevent kernel stack overflows. Add generic_entry based
> STACKLEAK support. Some optimization for entry.S with new .macro and
> merge ret_from_kernel_thread into ret_from_fork.
> 
> We have tested it with rv64, rv32, rv64 + 32rootfs, all are passed.
> 
> You can directly try it with:
> [1] https://github.com/guoren83/linux/tree/generic_entry_v8
> 
> Any reviews and tests are helpful.

Thanks for this series,

Tested-by: Jisheng Zhang <jszhang@kernel.org>

> 
> v8:
>  - Rebase on palmer/for-next branch (20221102)
>  - Add save/restore_from_x5_to_x31 .macro (JishengZhang)
>  - Consolidate ret_from_kernel_thread into ret_from_fork (JishengZhang)
>  - Optimize __noinstr_section comment (JiangshanLai)
> 
> v7:
> https://lore.kernel.org/linux-riscv/20221015114702.3489989-1-guoren@kernel.org/
>  - Fixup regs_irqs_disabled with SR_PIE
>  - Optimize stackleak_erase -> stackleak_erase_on_task_stack (Thx Mark
>    Rutland)
>  - Add BUG_ON(!irqs_disabled()) in trap handlers
>  - Using regs_irqs_disabled in __do_page_fault
>  - Remove unnecessary irq disable in ret_from_exception and add comment 
> 
> v6:
> https://lore.kernel.org/linux-riscv/20221002012451.2351127-1-guoren@kernel.org/
>  - Use THEAD_SIZE_ORDER for thread size adjustment in kconfig (Thx Arnd)
>  - Move call_on_stack to inline style (Thx Peter Zijlstra)
>  - Fixup fp chain broken (Thx Chen Zhongjin)
>  - Remove common entry modification, and fixup page_fault entry (Thx
>    Peter Zijlstra)
>  - Treat some traps as nmi entry (Thx Peter Zijlstra)
> 
> v5:
> https://lore.kernel.org/linux-riscv/20220918155246.1203293-1-guoren@kernel.org/
>  - Add riscv own stackleak patch instead of generic entry modification
>    (by Mark Rutland)
>  - Add EXPERT dependency for THREAD_SIZE (by Arnd)
>  - Add EXPERT dependency for IRQ_STACK (by Sebastian, David Laight)
>  - Corrected __trap_section (by Peter Zijlstra)
>  - Add Tested-by (Yipeng Zou)
>  - Use CONFIG_SOFTIRQ_ON_OWN_STACK replace "#ifndef CONFIG_PREEMPT_RT"
>  - Fixup systrace_enter compile error
>  - Fixup exit_to_user_mode_prepare preempt_disable warning
> 
> V4:
> https://lore.kernel.org/linux-riscv/20220908022506.1275799-1-guoren@kernel.org/
>  - Fixup entry.S with "la" bug (by Conor.Dooley)
>  - Fixup missing noinstr bug (by Peter Zijlstra)
> 
> V3:
> https://lore.kernel.org/linux-riscv/20220906035423.634617-1-guoren@kernel.org/
>  - Fixup CONFIG_COMPAT=n compile error
>  - Add THREAD_SIZE_ORDER config
>  - Optimize elf_kexec.c warning fixup
>  - Add static to irq_stack_ptr definition
> 
> V2:
> https://lore.kernel.org/linux-riscv/20220904072637.8619-1-guoren@kernel.org/
>  - Fixup compile error by include "riscv: ptrace: Remove duplicate
>    operation"
>  - Fixup compile warning
>    Reported-by: kernel test robot <lkp@intel.com>
>  - Add test repo link in cover letter
> 
> V1:
> https://lore.kernel.org/linux-riscv/20220903163808.1954131-1-guoren@kernel.org/
> 
> 
> Dao Lu (1):
>   riscv: Add support for STACKLEAK gcc plugin
> 
> Guo Ren (9):
>   riscv: elf_kexec: Fixup compile warning
>   riscv: compat_syscall_table: Fixup compile warning
>   riscv: ptrace: Remove duplicate operation
>   riscv: traps: Add noinstr to prevent instrumentation inserted
>   riscv: convert to generic entry
>   riscv: Support HAVE_IRQ_EXIT_ON_IRQ_STACK
>   riscv: Support HAVE_SOFTIRQ_ON_OWN_STACK
>   riscv: Add config of thread stack size
>   riscv: Typo fixup for addi -> andi in comment
> 
> Jisheng Zhang (3):
>   riscv: remove extra level wrappers of trace_hardirqs_{on,off}
>   riscv: consolidate ret_from_kernel_thread into ret_from_fork
>   riscv: entry: consolidate general regs saving/restoring
> 
> Lai Jiangshan (1):
>   compiler_types.h: Add __noinstr_section() for noinstr
> 
>  arch/riscv/Kconfig                    |  21 ++
>  arch/riscv/include/asm/asm.h          |  63 +++++
>  arch/riscv/include/asm/csr.h          |   1 -
>  arch/riscv/include/asm/entry-common.h |   8 +
>  arch/riscv/include/asm/ptrace.h       |  10 +-
>  arch/riscv/include/asm/stacktrace.h   |   5 +
>  arch/riscv/include/asm/syscall.h      |   6 +
>  arch/riscv/include/asm/thread_info.h  |  27 +--
>  arch/riscv/include/asm/vmap_stack.h   |  28 +++
>  arch/riscv/kernel/Makefile            |   3 +-
>  arch/riscv/kernel/elf_kexec.c         |   2 +-
>  arch/riscv/kernel/entry.S             | 327 +++-----------------------
>  arch/riscv/kernel/irq.c               | 110 +++++++++
>  arch/riscv/kernel/mcount-dyn.S        |  56 +----
>  arch/riscv/kernel/process.c           |   5 +-
>  arch/riscv/kernel/ptrace.c            |  44 ----
>  arch/riscv/kernel/signal.c            |  21 +-
>  arch/riscv/kernel/sys_riscv.c         |  27 +++
>  arch/riscv/kernel/trace_irq.c         |  27 ---
>  arch/riscv/kernel/trace_irq.h         |  11 -
>  arch/riscv/kernel/traps.c             |  74 ++++--
>  arch/riscv/mm/fault.c                 |  16 +-
>  drivers/firmware/efi/libstub/Makefile |   2 +-
>  include/linux/compiler_types.h        |  15 +-
>  24 files changed, 407 insertions(+), 502 deletions(-)
>  create mode 100644 arch/riscv/include/asm/entry-common.h
>  create mode 100644 arch/riscv/include/asm/vmap_stack.h
>  delete mode 100644 arch/riscv/kernel/trace_irq.c
>  delete mode 100644 arch/riscv/kernel/trace_irq.h
> 
> -- 
> 2.36.1
>
  

guoren@kernel.org writes:

> From: Guo Ren <guoren@linux.alibaba.com>
>
> The patches convert riscv to use the generic entry infrastructure from
> kernel/entry/*. Additionally, add independent irq stacks (IRQ_STACKS)
> for percpu to prevent kernel stack overflows. Add generic_entry based
> STACKLEAK support. Some optimization for entry.S with new .macro and
> merge ret_from_kernel_thread into ret_from_fork.
>
> We have tested it with rv64, rv32, rv64 + 32rootfs, all are passed.
>
> You can directly try it with:
> [1] https://github.com/guoren83/linux/tree/generic_entry_v8

Guo, this is a really nice work, and I'm looking forward having generic
entry support for RV. However, there are many patches in this series
that really shouldn't be part of the series.

Patch 2, 3, 4, and 10 should defintely be pulled out.

I'm not sure 7, 8, and 9 belong to series, as it's really a separate
feature.

Dito for patch 11, it just makes the series harder to review.

For GENERIC_ENTRY support only patch 1, 5, 6, 12, 13, and 14, really
required. The others are unrelated.


Thanks,
Björn
  

On Mon, Dec 5, 2022 at 5:46 PM Björn Töpel <bjorn@kernel.org> wrote:
>
> guoren@kernel.org writes:
>
> > From: Guo Ren <guoren@linux.alibaba.com>
> >
> > The patches convert riscv to use the generic entry infrastructure from
> > kernel/entry/*. Additionally, add independent irq stacks (IRQ_STACKS)
> > for percpu to prevent kernel stack overflows. Add generic_entry based
> > STACKLEAK support. Some optimization for entry.S with new .macro and
> > merge ret_from_kernel_thread into ret_from_fork.
> >
> > We have tested it with rv64, rv32, rv64 + 32rootfs, all are passed.
> >
> > You can directly try it with:
> > [1] https://github.com/guoren83/linux/tree/generic_entry_v8
>
> Guo, this is a really nice work, and I'm looking forward having generic
> entry support for RV. However, there are many patches in this series
> that really shouldn't be part of the series.
>
> Patch 2, 3, 4, and 10 should defintely be pulled out.
Okay.

>
> I'm not sure 7, 8, and 9 belong to series, as it's really a separate
> feature.
The separate irq/softirq stack patches dpend on generic_entry tightly,
so I recommand put them together.

>
> Dito for patch 11, it just makes the series harder to review.
The 11 is not so necessary as above, I would remove it from this
series. And send it again after generic_entry merged.

>
> For GENERIC_ENTRY support only patch 1, 5, 6, 12, 13, and 14, really
> required. The others are unrelated.
Thx for the reminder; I will re-organize them.

>
>
> Thanks,
> Björn
  

Guo Ren <guoren@kernel.org> writes:

> On Mon, Dec 5, 2022 at 5:46 PM Björn Töpel <bjorn@kernel.org> wrote:
>>
>> guoren@kernel.org writes:
>>
>> > From: Guo Ren <guoren@linux.alibaba.com>
>> >
>> > The patches convert riscv to use the generic entry infrastructure from
>> > kernel/entry/*. Additionally, add independent irq stacks (IRQ_STACKS)
>> > for percpu to prevent kernel stack overflows. Add generic_entry based
>> > STACKLEAK support. Some optimization for entry.S with new .macro and
>> > merge ret_from_kernel_thread into ret_from_fork.
>> >
>> > We have tested it with rv64, rv32, rv64 + 32rootfs, all are passed.
>> >
>> > You can directly try it with:
>> > [1] https://github.com/guoren83/linux/tree/generic_entry_v8
>>
>> Guo, this is a really nice work, and I'm looking forward having generic
>> entry support for RV. However, there are many patches in this series
>> that really shouldn't be part of the series.
>>
>> Patch 2, 3, 4, and 10 should defintely be pulled out.
> Okay.
>
>>
>> I'm not sure 7, 8, and 9 belong to series, as it's really a separate
>> feature.
> The separate irq/softirq stack patches dpend on generic_entry tightly,
> so I recommand put them together.

I'd say *depends* on the series, not required for generic entry. Again,
keeping the series orthogonal and small is a good thing.

>> For GENERIC_ENTRY support only patch 1, 5, 6, 12, 13, and 14, really
>> required. The others are unrelated.
> Thx for the reminder; I will re-organize them.

This is a really nice clean up, and the reason I'm pushing for removing
unrelated patches in the series, is because I want to see it land in the
RV tree sooner, rather than later. Send these as follow-ups to generic
entry, once it lands in the tree.


Thank you for the hard work!
Björn