Re: [PATCH v14 1/3] /proc/pid/status: Add support for architecture specific output

On Tue, Apr 9, 2019 at 7:20 PM Li, Aubrey [off-list ref] wrote:

On 2019/4/10 9:58, Andy Lutomirski wrote:

quoted

On Tue, Apr 9, 2019 at 6:55 PM Aubrey Li [off-list ref] wrote:

quoted

The architecture specific information of the running processes could
be useful to the userland. Add support to examine process architecture
specific information externally.

Signed-off-by: Aubrey Li <redacted>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Andi Kleen <redacted>
Cc: Tim Chen <redacted>
Cc: Dave Hansen <redacted>
Cc: Arjan van de Ven <redacted>
Cc: Linux API <redacted>
Cc: Alexey Dobriyan <redacted>
Cc: Andrew Morton <akpm@linux-foundation.org>
---
 fs/proc/array.c         | 5 +++++
 include/linux/proc_fs.h | 2 ++
 2 files changed, 7 insertions(+)

diff --git a/fs/proc/array.c b/fs/proc/array.c
index 2edbb657f859..331592a61718 100644
--- a/fs/proc/array.c
+++ b/fs/proc/array.c

@@ -401,6 +401,10 @@ static inline void task_thp_status(struct seq_file *m, struct mm_struct *mm)
        seq_printf(m, "THP_enabled:\t%d\n", thp_enabled);
 }

+void __weak arch_proc_pid_status(struct seq_file *m, struct task_struct *task)
+{
+}

This pointlessly bloats other architectures.  Do this instead in an
appropriate header:

#ifndef arch_proc_pid_status
static inline void arch_proc_pid_status(...)
{
}
#endif

I saw a bunch of similar weak functions, is it not acceptable?

fs/proc$ grep weak *.c
cpuinfo.c:__weak void arch_freq_prepare_all(void)
meminfo.c:void __attribute__((weak)) arch_report_meminfo(struct seq_file *m)
vmcore.c:int __weak elfcorehdr_alloc(unsigned long long *addr, unsigned long long *size)
vmcore.c:void __weak elfcorehdr_free(unsigned long long addr)
vmcore.c:ssize_t __weak elfcorehdr_read(char *buf, size_t count, u64 *ppos)
vmcore.c:ssize_t __weak elfcorehdr_read_notes(char *buf, size_t count, u64 *ppos)
vmcore.c:int __weak remap_oldmem_pfn_range(struct vm_area_struct *vma,
vmcore.c:ssize_t __weak

I think they're acceptable, but I don't personally like them.

quoted

Or add /proc/PID/x86_status, which sounds better in most respects to me.

I didn't figure out how to make /proc/PID/x86_status invisible to other
architectures in an appropriate way, do you have any suggestions?

#ifdef CONFIG_X86?