在 2023/3/15 15:54, Greg Kroah-Hartman 写道:
On Thu, Mar 02, 2023 at 12:18:04AM +0800, wenyang.linux@xxxxxxxxxxx wrote:
From: Arnd Bergmann <arnd@xxxxxxxx>
commit 4b692e861619353ce069e547a67c8d0e32d9ef3d upstream.
Patch series "compat: remove compat_alloc_user_space", v5.
Going through compat_alloc_user_space() to convert indirect system call
arguments tends to add complexity compared to handling the native and
compat logic in the same code.
This patch (of 6):
What about the other 6?
Hi,
Both 4b692e861619 ("kexec: move locking into do_kexec_load") and
7bb5da0d490b
("kexec: turn all kexec_mutex acquisitions into trylocks") are prerequisites
for 05c6257433b7 ("panic, kexec: make __crash_kexec() NMI safe").
In addition, although 4b692e861619 ("kexec: move locking into
do_kexec_load")
is part of patch series "compat: remove compat_alloc_user_space", it is also
independent and a general optimization, and here we just need is it, as
follows:
Arnd Bergmann (6):
kexec: move locking into do_kexec_load
kexec: avoid compat_alloc_user_space
mm: simplify compat_sys_move_pages
mm: simplify compat numa syscalls
compat: remove some compat entry points
arch: remove compat_alloc_user_space
https://lore.kernel.org/all/20210727144859.4150043-7-arnd@xxxxxxxxxx/T/#u
And what kernel is this going to, just 5.10.y?
For 5.10.y, these three patches are needed:
4b692e861619 ("kexec: move locking into do_kexec_load")
7bb5da0d490b ("kexec: turn all kexec_mutex acquisitions into trylocks")
05c6257433b7 ("panic, kexec: make __crash_kexec() NMI safe").
For 5.15.y, only these two patches are needed:
7bb5da0d490b ("kexec: turn all kexec_mutex acquisitions into trylocks")
05c6257433b7 ("panic, kexec: make __crash_kexec() NMI safe").
Can you resend this as an actual patch series linked together? They do
not show up properly for some reason (same for your 5.15.y patches.)
Try using git send-email to send them out.
OK, we'll resend it later.
--
Best wishes,
Wen