Re: [PATCH v11 09/14] entry: Rework syscall_exit_to_user_mode_work() for arch reuse

[Jinjie Ruan <ruanjinjie@huawei.com>]

On 2026/1/29 20:06, Kevin Brodsky wrote:
> On 28/01/2026 04:19, Jinjie Ruan wrote:
>> In the generic entry code, the beginning of
>> syscall_exit_to_user_mode_work() can be reused on arm64 so it makes
>> sense to rework it.
>>
>> In preparation for moving arm64 over to the generic entry
>> code, as nothing calls syscall_exit_to_user_mode_work() except for
>> syscall_exit_to_user_mode(), move local_irq_disable_exit_to_user() and
>> syscall_exit_to_user_mode_prepare() out from
>> syscall_exit_to_user_mode_work() to the only one caller.
>>
>> Also update the comment and no functional changes.
>>
>> Reviewed-by: Kevin Brodsky <kevin.brodsky@arm.com>
>> Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
>> Signed-off-by: Jinjie Ruan <ruanjinjie@huawei.com>
>> ---
>>  include/linux/entry-common.h | 16 ++++++++--------
>>  1 file changed, 8 insertions(+), 8 deletions(-)
>>
>> diff --git a/include/linux/entry-common.h b/include/linux/entry-common.h
>> index e4a8287af822..c4fea642d931 100644
>> --- a/include/linux/entry-common.h
>> +++ b/include/linux/entry-common.h
>> @@ -125,14 +125,14 @@ void syscall_exit_work(struct pt_regs *regs, unsigned long work);
>>   * syscall_exit_to_user_mode_work - Handle work before returning to user mode
>>   * @regs:	Pointer to currents pt_regs
>>   *
>> - * Same as step 1 and 2 of syscall_exit_to_user_mode() but without calling
>> + * Same as step 1 of syscall_exit_to_user_mode() but without calling
>> + * local_irq_disable(), syscall_exit_to_user_mode_prepare() and
>>   * exit_to_user_mode() to perform the final transition to user mode.
>>   *
>> - * Calling convention is the same as for syscall_exit_to_user_mode() and it
>> - * returns with all work handled and interrupts disabled. The caller must
>> - * invoke exit_to_user_mode() before actually switching to user mode to
>> - * make the final state transitions. Interrupts must stay disabled between
>> - * return from this function and the invocation of exit_to_user_mode().
>> + * Calling convention is the same as for syscall_exit_to_user_mode(). The
>> + * caller must invoke local_irq_disable(), __exit_to_user_mode_prepare() and
> 
> Shouldn't it be syscall_exit_to_user_mode_prepare() rather than
> __exit_to_user_mode_prepare()? The former has extra calls (e.g. rseq).

Perhaps we can just delete these comments — at present only generic
entry and arm64 use it, and nowhere else needs it; after the refactoring
the comments now seem rather unclear.

> 
> - Kevin
> 
>> + * exit_to_user_mode() before actually switching to user mode to
>> + * make the final state transitions.
>>   */
>>  static __always_inline void syscall_exit_to_user_mode_work(struct pt_regs *regs)
>>  {
>> @@ -155,8 +155,6 @@ static __always_inline void syscall_exit_to_user_mode_work(struct pt_regs *regs)
>>  	 */
>>  	if (unlikely(work & SYSCALL_WORK_EXIT))
>>  		syscall_exit_work(regs, work);
>> -	local_irq_disable_exit_to_user();
>> -	syscall_exit_to_user_mode_prepare(regs);
>>  }
>>  
>>  /**
>> @@ -192,6 +190,8 @@ static __always_inline void syscall_exit_to_user_mode(struct pt_regs *regs)
>>  {
>>  	instrumentation_begin();
>>  	syscall_exit_to_user_mode_work(regs);
>> +	local_irq_disable_exit_to_user();
>> +	syscall_exit_to_user_mode_prepare(regs);
>>  	instrumentation_end();
>>  	exit_to_user_mode();
>>  }
>