* Re: [Bug 214661] New: THREAD_SIZE on x86_64 is 4*PAGE_SIZE, not 2*PAGE_SIZE
[not found] <bug-214661-216477@https.bugzilla.kernel.org/>
@ 2021-10-09 18:51 ` Alejandro Colomar (man-pages)
0 siblings, 0 replies; only message in thread
From: Alejandro Colomar (man-pages) @ 2021-10-09 18:51 UTC (permalink / raw)
To: bugzilla-daemon, linux-kernel
CC += LKML
Hello,
On 10/9/21 7:19 AM, bugzilla-daemon@bugzilla.kernel.org wrote:
> https://bugzilla.kernel.org/show_bug.cgi?id=214661
>
> Bug ID: 214661
> Summary: THREAD_SIZE on x86_64 is 4*PAGE_SIZE, not 2*PAGE_SIZE
> Product: Documentation
> Version: unspecified
> Hardware: All
> OS: Linux
> Status: NEW
> Severity: normal
> Priority: P1
> Component: man-pages
This is incorrect.
The internal kernel documentation is not part of the man-pages.
I checked to see if the manual pages need an update, but I couldn't find
anything about THREAD_SIZE:
alx@sqli:~/src/linux/man-pages$ grep -rn THREAD_SIZE
alx@sqli:~/src/linux/man-pages$
I can't edit the component, so maybe you need to close the bug and open
a new one in a different component.
Anyway, thanks for the report!
Alex
> Assignee: documentation_man-pages@kernel-bugs.osdl.org
> Reporter: me@ieevee.com
> Regression: No
>
> According to '/root/linux/arch/x86/include/asm/page_64_types.h', THREAD_SIZE on
> x86_64 now is 4*PAGE_SIZE:
>
> #ifdef CONFIG_KASAN
> #define KASAN_STACK_ORDER 1
> #else
> #define KASAN_STACK_ORDER 0
> #endif
>
> #define THREAD_SIZE_ORDER (2 + KASAN_STACK_ORDER)
> #define THREAD_SIZE (PAGE_SIZE << THREAD_SIZE_ORDER)
>
> However, the Documentation/x86/kernel-stacks.rst shows that THREAD_SIZE on
> x86_64 is 2*PAGE_SIZE:
>
> Like all other architectures, x86_64 has a kernel stack for every
> active thread. These thread stacks are THREAD_SIZE (2*PAGE_SIZE) big.
>
> I am not quite sure which is correct, Maybe the document needs an upgrade?
> Thanks a lot for any reply.
>
--
Alejandro Colomar
Linux man-pages comaintainer; https://www.kernel.org/doc/man-pages/
http://www.alejandro-colomar.es/
^ permalink raw reply [flat|nested] only message in thread