platform/kernel/linux-rpi.git
22 months agoMAINTAINERS: update FPU EMULATOR web page
Randy Dunlap [Tue, 14 Feb 2023 17:02:08 +0000 (09:02 -0800)]
MAINTAINERS: update FPU EMULATOR web page

The web page entry for the FPU EMULATOR no longer works.  I notified Bill
of this and he asked me to update it to this new entry.

Link: https://lkml.kernel.org/r/20230214170208.17287-1-rdunlap@infradead.org
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Acked-by: Bill Metzenthen <billm@melbpc.org.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agomm/MADV_COLLAPSE: set EAGAIN on unexpected page refcount
Zach O'Keefe [Wed, 25 Jan 2023 01:57:37 +0000 (17:57 -0800)]
mm/MADV_COLLAPSE: set EAGAIN on unexpected page refcount

During collapse, in a few places we check to see if a given small page has
any unaccounted references.  If the refcount on the page doesn't match our
expectations, it must be there is an unknown user concurrently interested
in the page, and so it's not safe to move the contents elsewhere.
However, the unaccounted pins are likely an ephemeral state.

In this situation, MADV_COLLAPSE returns -EINVAL when it should return
-EAGAIN.  This could cause userspace to conclude that the syscall
failed, when it in fact could succeed by retrying.

Link: https://lkml.kernel.org/r/20230125015738.912924-1-zokeefe@google.com
Fixes: 7d8faaf15545 ("mm/madvise: introduce MADV_COLLAPSE sync hugepage collapse")
Signed-off-by: Zach O'Keefe <zokeefe@google.com>
Reported-by: Hugh Dickins <hughd@google.com>
Acked-by: Hugh Dickins <hughd@google.com>
Reviewed-by: Yang Shi <shy828301@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agomm/filemap: fix page end in filemap_get_read_batch
Qian Yingjin [Wed, 8 Feb 2023 02:24:00 +0000 (10:24 +0800)]
mm/filemap: fix page end in filemap_get_read_batch

I was running traces of the read code against an RAID storage system to
understand why read requests were being misaligned against the underlying
RAID strips.  I found that the page end offset calculation in
filemap_get_read_batch() was off by one.

When a read is submitted with end offset 1048575, then it calculates the
end page for read of 256 when it should be 255.  "last_index" is the index
of the page beyond the end of the read and it should be skipped when get a
batch of pages for read in @filemap_get_read_batch().

The below simple patch fixes the problem.  This code was introduced in
kernel 5.12.

Link: https://lkml.kernel.org/r/20230208022400.28962-1-coolqyj@163.com
Fixes: cbd59c48ae2b ("mm/filemap: use head pages in generic_file_buffered_read")
Signed-off-by: Qian Yingjin <qian@ddn.com>
Reviewed-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agoof: reserved_mem: Have kmemleak ignore dynamically allocated reserved mem
Isaac J. Manjarres [Wed, 8 Feb 2023 23:20:00 +0000 (15:20 -0800)]
of: reserved_mem: Have kmemleak ignore dynamically allocated reserved mem

Patch series "Fix kmemleak crashes when scanning CMA regions", v2.

When trying to boot a device with an ARM64 kernel with the following
config options enabled:

CONFIG_DEBUG_PAGEALLOC=y
CONFIG_DEBUG_PAGEALLOC_ENABLE_DEFAULT=y
CONFIG_DEBUG_KMEMLEAK=y

a crash is encountered when kmemleak starts to scan the list of gray
or allocated objects that it maintains. Upon closer inspection, it was
observed that these page-faults always occurred when kmemleak attempted
to scan a CMA region.

At the moment, kmemleak is made aware of CMA regions that are specified
through the devicetree to be dynamically allocated within a range of
addresses. However, kmemleak should not need to scan CMA regions or any
reserved memory region, as those regions can be used for DMA transfers
between drivers and peripherals, and thus wouldn't contain anything
useful for kmemleak.

Additionally, since CMA regions are unmapped from the kernel's address
space when they are freed to the buddy allocator at boot when
CONFIG_DEBUG_PAGEALLOC is enabled, kmemleak shouldn't attempt to access
those memory regions, as that will trigger a crash. Thus, kmemleak
should ignore all dynamically allocated reserved memory regions.

This patch (of 1):

Currently, kmemleak ignores dynamically allocated reserved memory regions
that don't have a kernel mapping.  However, regions that do retain a
kernel mapping (e.g.  CMA regions) do get scanned by kmemleak.

This is not ideal for two reasons:

1  kmemleak works by scanning memory regions for pointers to allocated
   objects to determine if those objects have been leaked or not.
   However, reserved memory regions can be used between drivers and
   peripherals for DMA transfers, and thus, would not contain pointers to
   allocated objects, making it unnecessary for kmemleak to scan these
   reserved memory regions.

2  When CONFIG_DEBUG_PAGEALLOC is enabled, along with kmemleak, the
   CMA reserved memory regions are unmapped from the kernel's address
   space when they are freed to buddy at boot.  These CMA reserved regions
   are still tracked by kmemleak, however, and when kmemleak attempts to
   scan them, a crash will happen, as accessing the CMA region will result
   in a page-fault, since the regions are unmapped.

Thus, use kmemleak_ignore_phys() for all dynamically allocated reserved
memory regions, instead of those that do not have a kernel mapping
associated with them.

Link: https://lkml.kernel.org/r/20230208232001.2052777-1-isaacmanjarres@google.com
Link: https://lkml.kernel.org/r/20230208232001.2052777-2-isaacmanjarres@google.com
Fixes: a7259df76702 ("memblock: make memblock_find_in_range method private")
Signed-off-by: Isaac J. Manjarres <isaacmanjarres@google.com>
Acked-by: Mike Rapoport (IBM) <rppt@kernel.org>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Cc: Frank Rowand <frowand.list@gmail.com>
Cc: Kirill A. Shutemov <kirill.shtuemov@linux.intel.com>
Cc: Nick Kossifidis <mick@ics.forth.gr>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Cc: Rob Herring <robh@kernel.org>
Cc: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Cc: Saravana Kannan <saravanak@google.com>
Cc: <stable@vger.kernel.org> [5.15+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agoscripts/gdb: fix 'lx-current' for x86
Jeff Xie [Sat, 4 Feb 2023 09:01:39 +0000 (17:01 +0800)]
scripts/gdb: fix 'lx-current' for x86

When printing the name of the current process, it will report an error:
(gdb) p $lx_current().comm Python Exception <class 'gdb.error'> No symbol
"current_task" in current context.: Error occurred in Python: No symbol
"current_task" in current context.

Because e57ef2ed97c1 ("x86: Put hot per CPU variables into a struct")
changed it.

Link: https://lkml.kernel.org/r/20230204090139.1789264-1-xiehuan09@gmail.com
Fixes: e57ef2ed97c1 ("x86: Put hot per CPU variables into a struct")
Signed-off-by: Jeff Xie <xiehuan09@gmail.com>
Cc: Jan Kiszka <jan.kiszka@siemens.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agolib: parser: optimize match_NUMBER apis to use local array
Li Lingfeng [Fri, 20 Jan 2023 03:23:52 +0000 (11:23 +0800)]
lib: parser: optimize match_NUMBER apis to use local array

Memory will be allocated to store substring_t in match_strdup(), which
means the caller of match_strdup() may need to be scheduled out to wait
for reclaiming memory.  smatch complains that this can cuase sleeping in
an atoic context.

Using local array to store substring_t to remove the restriction.

Link: https://lkml.kernel.org/r/20230120032352.242767-1-lilingfeng3@huawei.com
Link: https://lore.kernel.org/all/20221104023938.2346986-5-yukuai1@huaweicloud.com/
Link: https://lkml.kernel.org/r/20230120032352.242767-1-lilingfeng3@huawei.com
Fixes: 2c0647988433 ("blk-iocost: don't release 'ioc->lock' while updating params")
Signed-off-by: Li Lingfeng <lilingfeng3@huawei.com>
Reported-by: Yu Kuai <yukuai1@huaweicloud.com>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: BingJing Chang <bingjingc@synology.com>
Cc: Eric Biggers <ebiggers@google.com>
Cc: Hou Tao <houtao1@huawei.com>
Cc: James Smart <james.smart@broadcom.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: yangerkun <yangerkun@huawei.com>
Cc: Zhang Yi <yi.zhang@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agomm: shrinkers: fix deadlock in shrinker debugfs
Qi Zheng [Thu, 2 Feb 2023 10:56:12 +0000 (18:56 +0800)]
mm: shrinkers: fix deadlock in shrinker debugfs

The debugfs_remove_recursive() is invoked by unregister_shrinker(), which
is holding the write lock of shrinker_rwsem.  It will waits for the
handler of debugfs file complete.  The handler also needs to hold the read
lock of shrinker_rwsem to do something.  So it may cause the following
deadlock:

  CPU0 CPU1

debugfs_file_get()
shrinker_debugfs_count_show()/shrinker_debugfs_scan_write()

      unregister_shrinker()
--> down_write(&shrinker_rwsem);
    debugfs_remove_recursive()
// wait for (A)
    --> wait_for_completion();

    // wait for (B)
--> down_read_killable(&shrinker_rwsem)
debugfs_file_put() -- (A)

    up_write() -- (B)

The down_read_killable() can be killed, so that the above deadlock can be
recovered.  But it still requires an extra kill action, otherwise it will
block all subsequent shrinker-related operations, so it's better to fix
it.

[akpm@linux-foundation.org: fix CONFIG_SHRINKER_DEBUG=n stub]
Link: https://lkml.kernel.org/r/20230202105612.64641-1-zhengqi.arch@bytedance.com
Fixes: 5035ebc644ae ("mm: shrinkers: introduce debugfs interface for memory shrinkers")
Signed-off-by: Qi Zheng <zhengqi.arch@bytedance.com>
Reviewed-by: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Kent Overstreet <kent.overstreet@gmail.com>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agomm: hwpoison: support recovery from ksm_might_need_to_copy()
Kefeng Wang [Fri, 9 Dec 2022 07:28:01 +0000 (15:28 +0800)]
mm: hwpoison: support recovery from ksm_might_need_to_copy()

When the kernel copies a page from ksm_might_need_to_copy(), but runs into
an uncorrectable error, it will crash since poisoned page is consumed by
kernel, this is similar to the issue recently fixed by Copy-on-write
poison recovery.

When an error is detected during the page copy, return VM_FAULT_HWPOISON
in do_swap_page(), and install a hwpoison entry in unuse_pte() when
swapoff, which help us to avoid system crash.  Note, memory failure on a
KSM page will be skipped, but still call memory_failure_queue() to be
consistent with general memory failure process, and we could support KSM
page recovery in the feature.

[wangkefeng.wang@huawei.com: enhance unuse_pte(), fix issue found by lkp]
Link: https://lkml.kernel.org/r/20221213120523.141588-1-wangkefeng.wang@huawei.com
[wangkefeng.wang@huawei.com: update changelog, alter ksm_might_need_to_copy(), restore unlikely() in unuse_pte()]
Link: https://lkml.kernel.org/r/20230201074433.96641-1-wangkefeng.wang@huawei.com
Link: https://lkml.kernel.org/r/20221209072801.193221-1-wangkefeng.wang@huawei.com
Signed-off-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Reviewed-by: Naoya Horiguchi <naoya.horiguchi@nec.com>
Cc: Miaohe Lin <linmiaohe@huawei.com>
Cc: Tony Luck <tony.luck@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agokasan: fix Oops due to missing calls to kasan_arch_is_ready()
Christophe Leroy [Thu, 26 Jan 2023 07:04:47 +0000 (08:04 +0100)]
kasan: fix Oops due to missing calls to kasan_arch_is_ready()

On powerpc64, you can build a kernel with KASAN as soon as you build it
with RADIX MMU support.  However if the CPU doesn't have RADIX MMU, KASAN
isn't enabled at init and the following Oops is encountered.

  [    0.000000][    T0] KASAN not enabled as it requires radix!

  [    4.484295][   T26] BUG: Unable to handle kernel data access at 0xc00e000000804a04
  [    4.485270][   T26] Faulting instruction address: 0xc00000000062ec6c
  [    4.485748][   T26] Oops: Kernel access of bad area, sig: 11 [#1]
  [    4.485920][   T26] BE PAGE_SIZE=64K MMU=Hash SMP NR_CPUS=2048 NUMA pSeries
  [    4.486259][   T26] Modules linked in:
  [    4.486637][   T26] CPU: 0 PID: 26 Comm: kworker/u2:2 Not tainted 6.2.0-rc3-02590-gf8a023b0a805 #249
  [    4.486907][   T26] Hardware name: IBM pSeries (emulated by qemu) POWER9 (raw) 0x4e1200 0xf000005 of:SLOF,HEAD pSeries
  [    4.487445][   T26] Workqueue: eval_map_wq .tracer_init_tracefs_work_func
  [    4.488744][   T26] NIP:  c00000000062ec6c LR: c00000000062bb84 CTR: c0000000002ebcd0
  [    4.488867][   T26] REGS: c0000000049175c0 TRAP: 0380   Not tainted  (6.2.0-rc3-02590-gf8a023b0a805)
  [    4.489028][   T26] MSR:  8000000002009032 <SF,VEC,EE,ME,IR,DR,RI>  CR: 44002808  XER: 00000000
  [    4.489584][   T26] CFAR: c00000000062bb80 IRQMASK: 0
  [    4.489584][   T26] GPR00: c0000000005624d4 c000000004917860 c000000001cfc000 1800000000804a04
  [    4.489584][   T26] GPR04: c0000000003a2650 0000000000000cc0 c00000000000d3d8 c00000000000d3d8
  [    4.489584][   T26] GPR08: c0000000049175b0 a80e000000000000 0000000000000000 0000000017d78400
  [    4.489584][   T26] GPR12: 0000000044002204 c000000003790000 c00000000435003c c0000000043f1c40
  [    4.489584][   T26] GPR16: c0000000043f1c68 c0000000043501a0 c000000002106138 c0000000043f1c08
  [    4.489584][   T26] GPR20: c0000000043f1c10 c0000000043f1c20 c000000004146c40 c000000002fdb7f8
  [    4.489584][   T26] GPR24: c000000002fdb834 c000000003685e00 c000000004025030 c000000003522e90
  [    4.489584][   T26] GPR28: 0000000000000cc0 c0000000003a2650 c000000004025020 c000000004025020
  [    4.491201][   T26] NIP [c00000000062ec6c] .kasan_byte_accessible+0xc/0x20
  [    4.491430][   T26] LR [c00000000062bb84] .__kasan_check_byte+0x24/0x90
  [    4.491767][   T26] Call Trace:
  [    4.491941][   T26] [c000000004917860] [c00000000062ae70] .__kasan_kmalloc+0xc0/0x110 (unreliable)
  [    4.492270][   T26] [c0000000049178f0] [c0000000005624d4] .krealloc+0x54/0x1c0
  [    4.492453][   T26] [c000000004917990] [c0000000003a2650] .create_trace_option_files+0x280/0x530
  [    4.492613][   T26] [c000000004917a90] [c000000002050d90] .tracer_init_tracefs_work_func+0x274/0x2c0
  [    4.492771][   T26] [c000000004917b40] [c0000000001f9948] .process_one_work+0x578/0x9f0
  [    4.492927][   T26] [c000000004917c30] [c0000000001f9ebc] .worker_thread+0xfc/0x950
  [    4.493084][   T26] [c000000004917d60] [c00000000020be84] .kthread+0x1a4/0x1b0
  [    4.493232][   T26] [c000000004917e10] [c00000000000d3d8] .ret_from_kernel_thread+0x58/0x60
  [    4.495642][   T26] Code: 60000000 7cc802a6 38a00000 4bfffc78 60000000 7cc802a6 38a00001 4bfffc68 60000000 3d20a80e 7863e8c2 792907c6 <7c6348ae20630007 78630fe0 68630001
  [    4.496704][   T26] ---[ end trace 0000000000000000 ]---

The Oops is due to kasan_byte_accessible() not checking the readiness of
KASAN.  Add missing call to kasan_arch_is_ready() and bail out when not
ready.  The same problem is observed with ____kasan_kfree_large() so fix
it the same.

Also, as KASAN is not available and no shadow area is allocated for linear
memory mapping, there is no point in allocating shadow mem for vmalloc
memory as shown below in /sys/kernel/debug/kernel_page_tables

  ---[ kasan shadow mem start ]---
  0xc00f000000000000-0xc00f00000006ffff  0x00000000040f0000       448K         r  w       pte  valid  present        dirty  accessed
  0xc00f000000860000-0xc00f00000086ffff  0x000000000ac10000        64K         r  w       pte  valid  present        dirty  accessed
  0xc00f3ffffffe0000-0xc00f3fffffffffff  0x0000000004d10000       128K         r  w       pte  valid  present        dirty  accessed
  ---[ kasan shadow mem end ]---

So, also verify KASAN readiness before allocating and poisoning
shadow mem for VMAs.

Link: https://lkml.kernel.org/r/150768c55722311699fdcf8f5379e8256749f47d.1674716617.git.christophe.leroy@csgroup.eu
Fixes: 41b7a347bf14 ("powerpc: Book3S 64-bit outline-only KASAN support")
Signed-off-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Reported-by: Nathan Lynch <nathanl@linux.ibm.com>
Suggested-by: Michael Ellerman <mpe@ellerman.id.au>
Cc: Alexander Potapenko <glider@google.com>
Cc: Andrey Konovalov <andreyknvl@gmail.com>
Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Vincenzo Frascino <vincenzo.frascino@arm.com>
Cc: <stable@vger.kernel.org> [5.19+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agorevert "squashfs: harden sanity check in squashfs_read_xattr_id_table"
Andrew Morton [Fri, 3 Feb 2023 02:07:35 +0000 (18:07 -0800)]
revert "squashfs: harden sanity check in squashfs_read_xattr_id_table"

This fix was nacked by Philip, for reasons identified in the email linked
below.

Link: https://lkml.kernel.org/r/68f15d67-8945-2728-1f17-5b53a80ec52d@squashfs.org.uk
Fixes: 72e544b1b28325 ("squashfs: harden sanity check in squashfs_read_xattr_id_table")
Cc: Alexey Khoroshilov <khoroshilov@ispras.ru>
Cc: Fedor Pchelkin <pchelkin@ispras.ru>
Cc: Phillip Lougher <phillip@squashfs.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agofsdax: dax_unshare_iter() should return a valid length
Shiyang Ruan [Thu, 2 Feb 2023 12:33:47 +0000 (12:33 +0000)]
fsdax: dax_unshare_iter() should return a valid length

The copy_mc_to_kernel() will return 0 if it executed successfully.  Then
the return value should be set to the length it copied.

[akpm@linux-foundation.org: don't mess up `ret', per Matthew]
Link: https://lkml.kernel.org/r/1675341227-14-1-git-send-email-ruansy.fnst@fujitsu.com
Fixes: d984648e428b ("fsdax,xfs: port unshare to fsdax")
Signed-off-by: Shiyang Ruan <ruansy.fnst@fujitsu.com>
Cc: Darrick J. Wong <djwong@kernel.org>
Cc: Alistair Popple <apopple@nvidia.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Jason Gunthorpe <jgg@nvidia.com>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agomm/gup: add folio to list when folio_isolate_lru() succeed
Kuan-Ying Lee [Tue, 31 Jan 2023 06:32:06 +0000 (14:32 +0800)]
mm/gup: add folio to list when folio_isolate_lru() succeed

If we call folio_isolate_lru() successfully, we will get return value 0.
We need to add this folio to the movable_pages_list.

Link: https://lkml.kernel.org/r/20230131063206.28820-1-Kuan-Ying.Lee@mediatek.com
Fixes: 67e139b02d99 ("mm/gup.c: refactor check_and_migrate_movable_pages()")
Signed-off-by: Kuan-Ying Lee <Kuan-Ying.Lee@mediatek.com>
Reviewed-by: Alistair Popple <apopple@nvidia.com>
Acked-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: Andrew Yang <andrew.yang@mediatek.com>
Cc: Chinwen Chang <chinwen.chang@mediatek.com>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: Matthias Brugger <matthias.bgg@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agoaio: fix mremap after fork null-deref
Seth Jenkins [Tue, 31 Jan 2023 17:25:55 +0000 (12:25 -0500)]
aio: fix mremap after fork null-deref

Commit e4a0d3e720e7 ("aio: Make it possible to remap aio ring") introduced
a null-deref if mremap is called on an old aio mapping after fork as
mm->ioctx_table will be set to NULL.

[jmoyer@redhat.com: fix 80 column issue]
Link: https://lkml.kernel.org/r/x49sffq4nvg.fsf@segfault.boston.devel.redhat.com
Fixes: e4a0d3e720e7 ("aio: Make it possible to remap aio ring")
Signed-off-by: Seth Jenkins <sethjenkins@google.com>
Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Benjamin LaHaise <bcrl@kvack.org>
Cc: Jann Horn <jannh@google.com>
Cc: Pavel Emelyanov <xemul@parallels.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agomailmap: add entry for Alexander Mikhalitsyn
Alexander Mikhalitsyn [Tue, 31 Jan 2023 12:34:56 +0000 (13:34 +0100)]
mailmap: add entry for Alexander Mikhalitsyn

My old email <alexander.mikhalitsyn@virtuozzo.com> isn't working
anymore.

Link: https://lkml.kernel.org/r/20230131123456.192657-1-aleksandr.mikhalitsyn@canonical.com
Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agomm: extend max struct page size for kmsan
Arnd Bergmann [Mon, 30 Jan 2023 13:07:26 +0000 (14:07 +0100)]
mm: extend max struct page size for kmsan

After x86 enabled support for KMSAN, it has become possible to have larger
'struct page' than was expected when commit 5470dea49f53 ("mm: use
mm_zero_struct_page from SPARC on all 64b architectures") was merged:

include/linux/mm.h:156:10: warning: no case matching constant switch condition '96'
        switch (sizeof(struct page)) {

Extend the maximum accordingly.

Link: https://lkml.kernel.org/r/20230130130739.563628-1-arnd@kernel.org
Fixes: 5470dea49f53 ("mm: use mm_zero_struct_page from SPARC on all 64b architectures")
Fixes: 4ca8cc8d1bbe ("x86: kmsan: enable KMSAN builds for x86")
Fixes: f80be4571b19 ("kmsan: add KMSAN runtime core")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Michal Hocko <mhocko@suse.com>
Reviewed-by: Pasha Tatashin <pasha.tatashin@soleen.com>
Cc: Alexander Duyck <alexander.h.duyck@linux.intel.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Alex Sierra <alex.sierra@amd.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: John Hubbard <jhubbard@nvidia.com>
Cc: Liam R. Howlett <Liam.Howlett@Oracle.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Naoya Horiguchi <naoya.horiguchi@nec.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agomm: memcg: fix NULL pointer in mem_cgroup_track_foreign_dirty_slowpath()
Kefeng Wang [Sun, 29 Jan 2023 04:09:45 +0000 (12:09 +0800)]
mm: memcg: fix NULL pointer in mem_cgroup_track_foreign_dirty_slowpath()

As commit 18365225f044 ("hwpoison, memcg: forcibly uncharge LRU pages"),
hwpoison will forcibly uncharg a LRU hwpoisoned page, the folio_memcg
could be NULl, then, mem_cgroup_track_foreign_dirty_slowpath() could
occurs a NULL pointer dereference, let's do not record the foreign
writebacks for folio memcg is null in mem_cgroup_track_foreign_dirty() to
fix it.

Link: https://lkml.kernel.org/r/20230129040945.180629-1-wangkefeng.wang@huawei.com
Fixes: 97b27821b485 ("writeback, memcg: Implement foreign dirty flushing")
Signed-off-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Reported-by: Ma Wupeng <mawupeng1@huawei.com>
Tested-by: Miko Larsson <mikoxyzzz@gmail.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: Ma Wupeng <mawupeng1@huawei.com>
Cc: Naoya Horiguchi <naoya.horiguchi@nec.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agoKconfig.debug: fix the help description in SCHED_DEBUG
ye xingchen [Sun, 29 Jan 2023 02:13:57 +0000 (10:13 +0800)]
Kconfig.debug: fix the help description in SCHED_DEBUG

The correct file path for SCHED_DEBUG is /sys/kernel/debug/sched.

Link: https://lkml.kernel.org/r/202301291013573466558@zte.com.cn
Signed-off-by: ye xingchen <ye.xingchen@zte.com.cn>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Geert Uytterhoeven <geert+renesas@glider.be>
Cc: Josh Poimboeuf <jpoimboe@kernel.org>
Cc: Kees Cook <keescook@chromium.org>
Cc: Miguel Ojeda <ojeda@kernel.org>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Nick Desaulniers <ndesaulniers@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Randy Dunlap <rdunlap@infradead.org>
Cc: Rasmus Villemoes <linux@rasmusvillemoes.dk>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Zhaoyang Huang <zhaoyang.huang@unisoc.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agomm/swapfile: add cond_resched() in get_swap_pages()
Longlong Xia [Sat, 28 Jan 2023 09:47:57 +0000 (09:47 +0000)]
mm/swapfile: add cond_resched() in get_swap_pages()

The softlockup still occurs in get_swap_pages() under memory pressure.  64
CPU cores, 64GB memory, and 28 zram devices, the disksize of each zram
device is 50MB with same priority as si.  Use the stress-ng tool to
increase memory pressure, causing the system to oom frequently.

The plist_for_each_entry_safe() loops in get_swap_pages() could reach tens
of thousands of times to find available space (extreme case:
cond_resched() is not called in scan_swap_map_slots()).  Let's add
cond_resched() into get_swap_pages() when failed to find available space
to avoid softlockup.

Link: https://lkml.kernel.org/r/20230128094757.1060525-1-xialonglong1@huawei.com
Signed-off-by: Longlong Xia <xialonglong1@huawei.com>
Reviewed-by: "Huang, Ying" <ying.huang@intel.com>
Cc: Chen Wandun <chenwandun@huawei.com>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Kefeng Wang <wangkefeng.wang@huawei.com>
Cc: Nanyong Sun <sunnanyong@huawei.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agomm: use stack_depot_early_init for kmemleak
Zhaoyang Huang [Thu, 19 Jan 2023 01:22:25 +0000 (09:22 +0800)]
mm: use stack_depot_early_init for kmemleak

Mirsad report the below error which is caused by stack_depot_init()
failure in kvcalloc.  Solve this by having stackdepot use
stack_depot_early_init().

On 1/4/23 17:08, Mirsad Goran Todorovac wrote:
I hate to bring bad news again, but there seems to be a problem with the output of /sys/kernel/debug/kmemleak:

[root@pc-mtodorov ~]# cat /sys/kernel/debug/kmemleak
unreferenced object 0xffff951c118568b0 (size 16):
comm "kworker/u12:2", pid 56, jiffies 4294893952 (age 4356.548s)
hex dump (first 16 bytes):
    6d 65 6d 73 74 69 63 6b 30 00 00 00 00 00 00 00 memstick0.......
    backtrace:
[root@pc-mtodorov ~]#

Apparently, backtrace of called functions on the stack is no longer
printed with the list of memory leaks.  This appeared on Lenovo desktop
10TX000VCR, with AlmaLinux 8.7 and BIOS version M22KT49A (11/10/2022) and
6.2-rc1 and 6.2-rc2 builds.  This worked on 6.1 with the same
CONFIG_KMEMLEAK=y and MGLRU enabled on a vanilla mainstream kernel from
Mr.  Torvalds' tree.  I don't know if this is deliberate feature for some
reason or a bug.  Please find attached the config, lshw and kmemleak
output.

[vbabka@suse.cz: remove stack_depot_init() call]
Link: https://lore.kernel.org/all/5272a819-ef74-65ff-be61-4d2d567337de@alu.unizg.hr/
Link: https://lkml.kernel.org/r/1674091345-14799-2-git-send-email-zhaoyang.huang@unisoc.com
Fixes: 56a61617dd22 ("mm: use stack_depot for recording kmemleak's backtrace")
Reported-by: Mirsad Todorovac <mirsad.todorovac@alu.unizg.hr>
Suggested-by: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Zhaoyang Huang <zhaoyang.huang@unisoc.com>
Acked-by: Mike Rapoport (IBM) <rppt@kernel.org>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Tested-by: Borislav Petkov (AMD) <bp@alien8.de>
Cc: ke.wang <ke.wang@unisoc.com>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agoSquashfs: fix handling and sanity checking of xattr_ids count
Phillip Lougher [Fri, 27 Jan 2023 06:18:42 +0000 (06:18 +0000)]
Squashfs: fix handling and sanity checking of xattr_ids count

A Sysbot [1] corrupted filesystem exposes two flaws in the handling and
sanity checking of the xattr_ids count in the filesystem.  Both of these
flaws cause computation overflow due to incorrect typing.

In the corrupted filesystem the xattr_ids value is 4294967071, which
stored in a signed variable becomes the negative number -225.

Flaw 1 (64-bit systems only):

The signed integer xattr_ids variable causes sign extension.

This causes variable overflow in the SQUASHFS_XATTR_*(A) macros.  The
variable is first multiplied by sizeof(struct squashfs_xattr_id) where the
type of the sizeof operator is "unsigned long".

On a 64-bit system this is 64-bits in size, and causes the negative number
to be sign extended and widened to 64-bits and then become unsigned.  This
produces the very large number 18446744073709548016 or 2^64 - 3600.  This
number when rounded up by SQUASHFS_METADATA_SIZE - 1 (8191 bytes) and
divided by SQUASHFS_METADATA_SIZE overflows and produces a length of 0
(stored in len).

Flaw 2 (32-bit systems only):

On a 32-bit system the integer variable is not widened by the unsigned
long type of the sizeof operator (32-bits), and the signedness of the
variable has no effect due it always being treated as unsigned.

The above corrupted xattr_ids value of 4294967071, when multiplied
overflows and produces the number 4294963696 or 2^32 - 3400.  This number
when rounded up by SQUASHFS_METADATA_SIZE - 1 (8191 bytes) and divided by
SQUASHFS_METADATA_SIZE overflows again and produces a length of 0.

The effect of the 0 length computation:

In conjunction with the corrupted xattr_ids field, the filesystem also has
a corrupted xattr_table_start value, where it matches the end of
filesystem value of 850.

This causes the following sanity check code to fail because the
incorrectly computed len of 0 matches the incorrect size of the table
reported by the superblock (0 bytes).

    len = SQUASHFS_XATTR_BLOCK_BYTES(*xattr_ids);
    indexes = SQUASHFS_XATTR_BLOCKS(*xattr_ids);

    /*
     * The computed size of the index table (len bytes) should exactly
     * match the table start and end points
    */
    start = table_start + sizeof(*id_table);
    end = msblk->bytes_used;

    if (len != (end - start))
            return ERR_PTR(-EINVAL);

Changing the xattr_ids variable to be "usigned int" fixes the flaw on a
64-bit system.  This relies on the fact the computation is widened by the
unsigned long type of the sizeof operator.

Casting the variable to u64 in the above macro fixes this flaw on a 32-bit
system.

It also means 64-bit systems do not implicitly rely on the type of the
sizeof operator to widen the computation.

[1] https://lore.kernel.org/lkml/000000000000cd44f005f1a0f17f@google.com/

Link: https://lkml.kernel.org/r/20230127061842.10965-1-phillip@squashfs.org.uk
Fixes: 506220d2ba21 ("squashfs: add more sanity checks in xattr id lookup")
Signed-off-by: Phillip Lougher <phillip@squashfs.org.uk>
Reported-by: <syzbot+082fa4af80a5bb1a9843@syzkaller.appspotmail.com>
Cc: Alexey Khoroshilov <khoroshilov@ispras.ru>
Cc: Fedor Pchelkin <pchelkin@ispras.ru>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agosh: define RUNTIME_DISCARD_EXIT
Tom Saeger [Tue, 24 Jan 2023 00:09:35 +0000 (17:09 -0700)]
sh: define RUNTIME_DISCARD_EXIT

sh vmlinux fails to link with GNU ld < 2.40 (likely < 2.36) since
commit 99cb0d917ffa ("arch: fix broken BuildID for arm64 and riscv").

This is similar to fixes for powerpc and s390:
commit 4b9880dbf3bd ("powerpc/vmlinux.lds: Define RUNTIME_DISCARD_EXIT").
commit a494398bde27 ("s390: define RUNTIME_DISCARD_EXIT to fix link error
with GNU ld < 2.36").

  $ sh4-linux-gnu-ld --version | head -n1
  GNU ld (GNU Binutils for Debian) 2.35.2

  $ make ARCH=sh CROSS_COMPILE=sh4-linux-gnu- microdev_defconfig
  $ make ARCH=sh CROSS_COMPILE=sh4-linux-gnu-

  `.exit.text' referenced in section `__bug_table' of crypto/algboss.o:
  defined in discarded section `.exit.text' of crypto/algboss.o
  `.exit.text' referenced in section `__bug_table' of
  drivers/char/hw_random/core.o: defined in discarded section
  `.exit.text' of drivers/char/hw_random/core.o
  make[2]: *** [scripts/Makefile.vmlinux:34: vmlinux] Error 1
  make[1]: *** [Makefile:1252: vmlinux] Error 2

arch/sh/kernel/vmlinux.lds.S keeps EXIT_TEXT:

/*
 * .exit.text is discarded at runtime, not link time, to deal with
 * references from __bug_table
 */
.exit.text : AT(ADDR(.exit.text)) { EXIT_TEXT }

However, EXIT_TEXT is thrown away by
DISCARD(include/asm-generic/vmlinux.lds.h) because
sh does not define RUNTIME_DISCARD_EXIT.

GNU ld 2.40 does not have this issue and builds fine.
This corresponds with Masahiro's comments in a494398bde27:
"Nathan [Chancellor] also found that binutils
commit 21401fc7bf67 ("Duplicate output sections in scripts") cured this
issue, so we cannot reproduce it with binutils 2.36+, but it is better
to not rely on it."

Link: https://lkml.kernel.org/r/9166a8abdc0f979e50377e61780a4bba1dfa2f52.1674518464.git.tom.saeger@oracle.com
Fixes: 99cb0d917ffa ("arch: fix broken BuildID for arm64 and riscv")
Link: https://lore.kernel.org/all/Y7Jal56f6UBh1abE@dev-arch.thelio-3990X/
Link: https://lore.kernel.org/all/20230123194218.47ssfzhrpnv3xfez@oracle.com/
Signed-off-by: Tom Saeger <tom.saeger@oracle.com>
Tested-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Cc: Ard Biesheuvel <ardb@kernel.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Dennis Gilmore <dennis@ausil.us>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Masahiro Yamada <masahiroy@kernel.org>
Cc: Naresh Kamboju <naresh.kamboju@linaro.org>
Cc: Nathan Chancellor <nathan@kernel.org>
Cc: Palmer Dabbelt <palmer@rivosinc.com>
Cc: Rich Felker <dalias@libc.org>
Cc: Yoshinori Sato <ysato@users.sourceforge.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agohighmem: round down the address passed to kunmap_flush_on_unmap()
Matthew Wilcox (Oracle) [Thu, 26 Jan 2023 20:07:27 +0000 (20:07 +0000)]
highmem: round down the address passed to kunmap_flush_on_unmap()

We already round down the address in kunmap_local_indexed() which is the
other implementation of __kunmap_local().  The only implementation of
kunmap_flush_on_unmap() is PA-RISC which is expecting a page-aligned
address.  This may be causing PA-RISC to be flushing the wrong addresses
currently.

Link: https://lkml.kernel.org/r/20230126200727.1680362-1-willy@infradead.org
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Fixes: 298fa1ad5571 ("highmem: Provide generic variant of kmap_atomic*")
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Cc: "Fabio M. De Francesco" <fmdefrancesco@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Helge Deller <deller@gmx.de>
Cc: Alexander Potapenko <glider@google.com>
Cc: Andrey Konovalov <andreyknvl@gmail.com>
Cc: Bagas Sanjaya <bagasdotme@gmail.com>
Cc: David Sterba <dsterba@suse.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Tony Luck <tony.luck@intel.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agomigrate: hugetlb: check for hugetlb shared PMD in node migration
Mike Kravetz [Thu, 26 Jan 2023 22:27:21 +0000 (14:27 -0800)]
migrate: hugetlb: check for hugetlb shared PMD in node migration

migrate_pages/mempolicy semantics state that CAP_SYS_NICE is required to
move pages shared with another process to a different node.  page_mapcount
> 1 is being used to determine if a hugetlb page is shared.  However, a
hugetlb page will have a mapcount of 1 if mapped by multiple processes via
a shared PMD.  As a result, hugetlb pages shared by multiple processes and
mapped with a shared PMD can be moved by a process without CAP_SYS_NICE.

To fix, check for a shared PMD if mapcount is 1.  If a shared PMD is found
consider the page shared.

Link: https://lkml.kernel.org/r/20230126222721.222195-3-mike.kravetz@oracle.com
Fixes: e2d8cf405525 ("migrate: add hugepage migration code to migrate_pages()")
Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
Acked-by: Peter Xu <peterx@redhat.com>
Acked-by: David Hildenbrand <david@redhat.com>
Cc: James Houghton <jthoughton@google.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: Naoya Horiguchi <naoya.horiguchi@linux.dev>
Cc: Vishal Moola (Oracle) <vishal.moola@gmail.com>
Cc: Yang Shi <shy828301@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agomm: hugetlb: proc: check for hugetlb shared PMD in /proc/PID/smaps
Mike Kravetz [Thu, 26 Jan 2023 22:27:20 +0000 (14:27 -0800)]
mm: hugetlb: proc: check for hugetlb shared PMD in /proc/PID/smaps

Patch series "Fixes for hugetlb mapcount at most 1 for shared PMDs".

This issue of mapcount in hugetlb pages referenced by shared PMDs was
discussed in [1].  The following two patches address user visible behavior
caused by this issue.

[1] https://lore.kernel.org/linux-mm/Y9BF+OCdWnCSilEu@monkey/

This patch (of 2):

A hugetlb page will have a mapcount of 1 if mapped by multiple processes
via a shared PMD.  This is because only the first process increases the
map count, and subsequent processes just add the shared PMD page to their
page table.

page_mapcount is being used to decide if a hugetlb page is shared or
private in /proc/PID/smaps.  Pages referenced via a shared PMD were
incorrectly being counted as private.

To fix, check for a shared PMD if mapcount is 1.  If a shared PMD is found
count the hugetlb page as shared.  A new helper to check for a shared PMD
is added.

[akpm@linux-foundation.org: simplification, per David]
[akpm@linux-foundation.org: hugetlb.h: include page_ref.h for page_count()]
Link: https://lkml.kernel.org/r/20230126222721.222195-2-mike.kravetz@oracle.com
Fixes: 25ee01a2fca0 ("mm: hugetlb: proc: add hugetlb-related fields to /proc/PID/smaps")
Signed-off-by: Mike Kravetz <mike.kravetz@oracle.com>
Acked-by: Peter Xu <peterx@redhat.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: James Houghton <jthoughton@google.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: Naoya Horiguchi <naoya.horiguchi@linux.dev>
Cc: Vishal Moola (Oracle) <vishal.moola@gmail.com>
Cc: Yang Shi <shy828301@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agomm/MADV_COLLAPSE: catch !none !huge !bad pmd lookups
Zach O'Keefe [Wed, 25 Jan 2023 22:53:58 +0000 (14:53 -0800)]
mm/MADV_COLLAPSE: catch !none !huge !bad pmd lookups

In commit 34488399fa08 ("mm/madvise: add file and shmem support to
MADV_COLLAPSE") we make the following change to find_pmd_or_thp_or_none():

-       if (!pmd_present(pmde))
-               return SCAN_PMD_NULL;
+       if (pmd_none(pmde))
+               return SCAN_PMD_NONE;

This was for-use by MADV_COLLAPSE file/shmem codepaths, where
MADV_COLLAPSE might identify a pte-mapped hugepage, only to have
khugepaged race-in, free the pte table, and clear the pmd.  Such codepaths
include:

A) If we find a suitably-aligned compound page of order HPAGE_PMD_ORDER
   already in the pagecache.
B) In retract_page_tables(), if we fail to grab mmap_lock for the target
   mm/address.

In these cases, collapse_pte_mapped_thp() really does expect a none (not
just !present) pmd, and we want to suitably identify that case separate
from the case where no pmd is found, or it's a bad-pmd (of course, many
things could happen once we drop mmap_lock, and the pmd could plausibly
undergo multiple transitions due to intervening fault, split, etc).
Regardless, the code is prepared install a huge-pmd only when the existing
pmd entry is either a genuine pte-table-mapping-pmd, or the none-pmd.

However, the commit introduces a logical hole; namely, that we've allowed
!none- && !huge- && !bad-pmds to be classified as genuine
pte-table-mapping-pmds.  One such example that could leak through are swap
entries.  The pmd values aren't checked again before use in
pte_offset_map_lock(), which is expecting nothing less than a genuine
pte-table-mapping-pmd.

We want to put back the !pmd_present() check (below the pmd_none() check),
but need to be careful to deal with subtleties in pmd transitions and
treatments by various arch.

The issue is that __split_huge_pmd_locked() temporarily clears the present
bit (or otherwise marks the entry as invalid), but pmd_present() and
pmd_trans_huge() still need to return true while the pmd is in this
transitory state.  For example, x86's pmd_present() also checks the
_PAGE_PSE , riscv's version also checks the _PAGE_LEAF bit, and arm64 also
checks a PMD_PRESENT_INVALID bit.

Covering all 4 cases for x86 (all checks done on the same pmd value):

1) pmd_present() && pmd_trans_huge()
   All we actually know here is that the PSE bit is set. Either:
   a) We aren't racing with __split_huge_page(), and PRESENT or PROTNONE
      is set.
      => huge-pmd
   b) We are currently racing with __split_huge_page().  The danger here
      is that we proceed as-if we have a huge-pmd, but really we are
      looking at a pte-mapping-pmd.  So, what is the risk of this
      danger?

      The only relevant path is:

madvise_collapse() -> collapse_pte_mapped_thp()

      Where we might just incorrectly report back "success", when really
      the memory isn't pmd-backed.  This is fine, since split could
      happen immediately after (actually) successful madvise_collapse().
      So, it should be safe to just assume huge-pmd here.

2) pmd_present() && !pmd_trans_huge()
   Either:
   a) PSE not set and either PRESENT or PROTNONE is.
      => pte-table-mapping pmd (or PROT_NONE)
   b) devmap.  This routine can be called immediately after
      unlocking/locking mmap_lock -- or called with no locks held (see
      khugepaged_scan_mm_slot()), so previous VMA checks have since been
      invalidated.

3) !pmd_present() && pmd_trans_huge()
  Not possible.

4) !pmd_present() && !pmd_trans_huge()
  Neither PRESENT nor PROTNONE set
  => not present

I've checked all archs that implement pmd_trans_huge() (arm64, riscv,
powerpc, longarch, x86, mips, s390) and this logic roughly translates
(though devmap treatment is unique to x86 and powerpc, and (3) doesn't
necessarily hold in general -- but that doesn't matter since
!pmd_present() always takes failure path).

Also, add a comment above find_pmd_or_thp_or_none() to help future
travelers reason about the validity of the code; namely, the possible
mutations that might happen out from under us, depending on how mmap_lock
is held (if at all).

Link: https://lkml.kernel.org/r/20230125225358.2576151-1-zokeefe@google.com
Fixes: 34488399fa08 ("mm/madvise: add file and shmem support to MADV_COLLAPSE")
Signed-off-by: Zach O'Keefe <zokeefe@google.com>
Reported-by: Hugh Dickins <hughd@google.com>
Reviewed-by: Yang Shi <shy828301@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agoRevert "mm: kmemleak: alloc gray object for reserved region with direct map"
Isaac J. Manjarres [Tue, 24 Jan 2023 23:02:54 +0000 (15:02 -0800)]
Revert "mm: kmemleak: alloc gray object for reserved region with direct map"

This reverts commit 972fa3a7c17c9d60212e32ecc0205dc585b1e769.

Kmemleak operates by periodically scanning memory regions for pointers to
allocated memory blocks to determine if they are leaked or not.  However,
reserved memory regions can be used for DMA transactions between a device
and a CPU, and thus, wouldn't contain pointers to allocated memory blocks,
making them inappropriate for kmemleak to scan.  Thus, revert this commit.

Link: https://lkml.kernel.org/r/20230124230254.295589-1-isaacmanjarres@google.com
Fixes: 972fa3a7c17c9 ("mm: kmemleak: alloc gray object for reserved region with direct map")
Signed-off-by: Isaac J. Manjarres <isaacmanjarres@google.com>
Acked-by: Catalin Marinas <catalin.marinas@arm.com>
Cc: Calvin Zhang <calvinzhang.cool@gmail.com>
Cc: Frank Rowand <frowand.list@gmail.com>
Cc: Rob Herring <robh+dt@kernel.org>
Cc: Saravana Kannan <saravanak@google.com>
Cc: <stable@vger.kernel.org> [5.17+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agofreevxfs: Kconfig: fix spelling
Randy Dunlap [Tue, 24 Jan 2023 18:16:38 +0000 (10:16 -0800)]
freevxfs: Kconfig: fix spelling

Fix a spello in freevxfs Kconfig.
(reported by codespell)

Link: https://lkml.kernel.org/r/20230124181638.15604-1-rdunlap@infradead.org
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Cc: Christoph Hellwig <hch@infradead.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agomaple_tree: should get pivots boundary by type
Wei Yang [Sat, 12 Nov 2022 23:43:08 +0000 (23:43 +0000)]
maple_tree: should get pivots boundary by type

We should get pivots boundary by type.  Fixes a potential overindexing of
mt_pivots[].

Link: https://lkml.kernel.org/r/20221112234308.23823-1-richard.weiyang@gmail.com
Fixes: 54a611b60590 ("Maple Tree: add new data structure")
Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
Reviewed-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months ago.mailmap: update e-mail address for Eugen Hristev
Eugen Hristev [Thu, 19 Jan 2023 07:22:29 +0000 (09:22 +0200)]
.mailmap: update e-mail address for Eugen Hristev

Update e-mail address.

Link: https://lkml.kernel.org/r/20230119072229.99603-1-eugen.hristev@collabora.com
Signed-off-by: Eugen Hristev <eugen.hristev@collabora.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agomm, mremap: fix mremap() expanding for vma's with vm_ops->close()
Vlastimil Babka [Tue, 17 Jan 2023 10:19:39 +0000 (11:19 +0100)]
mm, mremap: fix mremap() expanding for vma's with vm_ops->close()

Fabian has reported another regression in 6.1 due to ca3d76b0aa80 ("mm:
add merging after mremap resize").  The problem is that vma_merge() can
fail when vma has a vm_ops->close() method, causing is_mergeable_vma()
test to be negative.  This was happening for vma mapping a file from
fuse-overlayfs, which does have the method.  But when we are simply
expanding the vma, we never remove it due to the "merge" with the added
area, so the test should not prevent the expansion.

As a quick fix, check for such vmas and expand them using vma_adjust()
directly as was done before commit ca3d76b0aa80.  For a more robust long
term solution we should try to limit the check for vma_ops->close only to
cases that actually result in vma removal, so that no merge would be
prevented unnecessarily.

[akpm@linux-foundation.org: fix indenting whitespace, reflow comment]
Link: https://lkml.kernel.org/r/20230117101939.9753-1-vbabka@suse.cz
Fixes: ca3d76b0aa80 ("mm: add merging after mremap resize")
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Reported-by: Fabian Vogt <fvogt@suse.com>
Link: https://bugzilla.suse.com/show_bug.cgi?id=1206359#c35
Tested-by: Fabian Vogt <fvogt@suse.com>
Cc: Jakub Matěna <matenajakub@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agosquashfs: harden sanity check in squashfs_read_xattr_id_table
Fedor Pchelkin [Tue, 17 Jan 2023 10:52:26 +0000 (13:52 +0300)]
squashfs: harden sanity check in squashfs_read_xattr_id_table

While mounting a corrupted filesystem, a signed integer '*xattr_ids' can
become less than zero.  This leads to the incorrect computation of 'len'
and 'indexes' values which can cause null-ptr-deref in copy_bio_to_actor()
or out-of-bounds accesses in the next sanity checks inside
squashfs_read_xattr_id_table().

Found by Linux Verification Center (linuxtesting.org) with Syzkaller.

Link: https://lkml.kernel.org/r/20230117105226.329303-2-pchelkin@ispras.ru
Fixes: 506220d2ba21 ("squashfs: add more sanity checks in xattr id lookup")
Reported-by: <syzbot+082fa4af80a5bb1a9843@syzkaller.appspotmail.com>
Signed-off-by: Fedor Pchelkin <pchelkin@ispras.ru>
Signed-off-by: Alexey Khoroshilov <khoroshilov@ispras.ru>
Cc: Phillip Lougher <phillip@squashfs.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agoia64: fix build error due to switch case label appearing next to declaration
James Morse [Tue, 17 Jan 2023 15:16:32 +0000 (15:16 +0000)]
ia64: fix build error due to switch case label appearing next to declaration

Since commit aa06a9bd8533 ("ia64: fix clock_getres(CLOCK_MONOTONIC) to
report ITC frequency"), gcc 10.1.0 fails to build ia64 with the gnomic:
| ../arch/ia64/kernel/sys_ia64.c: In function 'ia64_clock_getres':
| ../arch/ia64/kernel/sys_ia64.c:189:3: error: a label can only be part of a statement and a declaration is not a statement
|   189 |   s64 tick_ns = DIV_ROUND_UP(NSEC_PER_SEC, local_cpu_data->itc_freq);

This line appears immediately after a case label in a switch.

Move the declarations out of the case, to the top of the function.

Link: https://lkml.kernel.org/r/20230117151632.393836-1-james.morse@arm.com
Fixes: aa06a9bd8533 ("ia64: fix clock_getres(CLOCK_MONOTONIC) to report ITC frequency")
Signed-off-by: James Morse <james.morse@arm.com>
Reviewed-by: Sergei Trofimovich <slyich@gmail.com>
Cc: Émeric Maschino <emeric.maschino@gmail.com>
Cc: matoro <matoro_mailinglist_kernel@matoro.tk>
Cc: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agomm: multi-gen LRU: fix crash during cgroup migration
Yu Zhao [Mon, 16 Jan 2023 03:44:05 +0000 (20:44 -0700)]
mm: multi-gen LRU: fix crash during cgroup migration

lru_gen_migrate_mm() assumes lru_gen_add_mm() runs prior to itself.  This
isn't true for the following scenario:

    CPU 1                         CPU 2

  clone()
    cgroup_can_fork()
                                cgroup_procs_write()
    cgroup_post_fork()
                                  task_lock()
                                  lru_gen_migrate_mm()
                                  task_unlock()
    task_lock()
    lru_gen_add_mm()
    task_unlock()

And when the above happens, kernel crashes because of linked list
corruption (mm_struct->lru_gen.list).

Link: https://lore.kernel.org/r/20230115134651.30028-1-msizanoen@qtmlabs.xyz/
Link: https://lkml.kernel.org/r/20230116034405.2960276-1-yuzhao@google.com
Fixes: bd74fdaea146 ("mm: multi-gen LRU: support page table walks")
Signed-off-by: Yu Zhao <yuzhao@google.com>
Reported-by: msizanoen <msizanoen@qtmlabs.xyz>
Tested-by: msizanoen <msizanoen@qtmlabs.xyz>
Cc: <stable@vger.kernel.org> [6.1+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agoRevert "mm: add nodes= arg to memory.reclaim"
Michal Hocko [Fri, 16 Dec 2022 09:46:33 +0000 (10:46 +0100)]
Revert "mm: add nodes= arg to memory.reclaim"

This reverts commit 12a5d3955227b0d7e04fb793ccceeb2a1dd275c5.

Although it is recognized that a finer grained pro-active reclaim is
something we need and want the semantic of this implementation is really
ambiguous.

In a follow up discussion it became clear that there are two essential
usecases here.  One is to use memory.reclaim to pro-actively reclaim
memory and expectation is that the requested and reported amount of memory
is uncharged from the memcg.  Another usecase focuses on pro-active
demotion when the memory is merely shuffled around to demotion targets
while the overall charged memory stays unchanged.

The current implementation considers demoted pages as reclaimed and that
break both usecases.  [1] has tried to address the reporting part but
there are more issues with that summarized in [2] and follow up emails.

Let's revert the nodemask based extension of the memcg pro-active
reclaim for now until we settle with a more robust semantic.

[1] http://lkml.kernel.org/r/http://lkml.kernel.org/r/20221206023406.3182800-1-almasrymina@google.com
[2] http://lkml.kernel.org/r/Y5bsmpCyeryu3Zz1@dhcp22.suse.cz

Link: https://lkml.kernel.org/r/Y5xASNe1x8cusiTx@dhcp22.suse.cz
Fixes: 12a5d3955227b0d ("mm: add nodes= arg to memory.reclaim")
Signed-off-by: Michal Hocko <mhocko@suse.com>
Cc: Bagas Sanjaya <bagasdotme@gmail.com>
Cc: Huang Ying <ying.huang@intel.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Mina Almasry <almasrymina@google.com>
Cc: Muchun Song <songmuchun@bytedance.com>
Cc: Roman Gushchin <roman.gushchin@linux.dev>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Wei Xu <weixugc@google.com>
Cc: Yang Shi <yang.shi@linux.alibaba.com>
Cc: Yosry Ahmed <yosryahmed@google.com>
Cc: zefan li <lizefan.x@bytedance.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agozsmalloc: fix a race with deferred_handles storing
Nhat Pham [Tue, 10 Jan 2023 23:17:01 +0000 (15:17 -0800)]
zsmalloc: fix a race with deferred_handles storing

Currently, there is a race between zs_free() and zs_reclaim_page():
zs_reclaim_page() finds a handle to an allocated object, but before the
eviction happens, an independent zs_free() call to the same handle could
come in and overwrite the object value stored at the handle with the last
deferred handle.  When zs_reclaim_page() finally gets to call the eviction
handler, it will see an invalid object value (i.e the previous deferred
handle instead of the original object value).

This race happens quite infrequently.  We only managed to produce it with
out-of-tree developmental code that triggers zsmalloc writeback with a
much higher frequency than usual.

This patch fixes this race by storing the deferred handle in the object
header instead.  We differentiate the deferred handle from the other two
cases (handle for allocated object, and linkage for free object) with a
new tag.  If zspage reclamation succeeds, we will free these deferred
handles by walking through the zspage objects.  On the other hand, if
zspage reclamation fails, we reconstruct the zspage freelist (with the
deferred handle tag and allocated tag) before trying again with the
reclamation.

[arnd@arndb.de: avoid unused-function warning]
Link: https://lkml.kernel.org/r/20230117170507.2651972-1-arnd@kernel.org
Link: https://lkml.kernel.org/r/20230110231701.326724-1-nphamcs@gmail.com
Fixes: 9997bc017549 ("zsmalloc: implement writeback mechanism for zsmalloc")
Signed-off-by: Nhat Pham <nphamcs@gmail.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Suggested-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Dan Streetman <ddstreet@ieee.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Nitin Gupta <ngupta@vflare.org>
Cc: Sergey Senozhatsky <senozhatsky@chromium.org>
Cc: Seth Jennings <sjenning@redhat.com>
Cc: Vitaly Wool <vitaly.wool@konsulko.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agomm/khugepaged: fix ->anon_vma race
Jann Horn [Wed, 11 Jan 2023 13:33:51 +0000 (14:33 +0100)]
mm/khugepaged: fix ->anon_vma race

If an ->anon_vma is attached to the VMA, collapse_and_free_pmd() requires
it to be locked.

Page table traversal is allowed under any one of the mmap lock, the
anon_vma lock (if the VMA is associated with an anon_vma), and the
mapping lock (if the VMA is associated with a mapping); and so to be
able to remove page tables, we must hold all three of them.
retract_page_tables() bails out if an ->anon_vma is attached, but does
this check before holding the mmap lock (as the comment above the check
explains).

If we racily merged an existing ->anon_vma (shared with a child
process) from a neighboring VMA, subsequent rmap traversals on pages
belonging to the child will be able to see the page tables that we are
concurrently removing while assuming that nothing else can access them.

Repeat the ->anon_vma check once we hold the mmap lock to ensure that
there really is no concurrent page table access.

Hitting this bug causes a lockdep warning in collapse_and_free_pmd(),
in the line "lockdep_assert_held_write(&vma->anon_vma->root->rwsem)".
It can also lead to use-after-free access.

Link: https://lore.kernel.org/linux-mm/CAG48ez3434wZBKFFbdx4M9j6eUwSUVPd4dxhzW_k_POneSDF+A@mail.gmail.com/
Link: https://lkml.kernel.org/r/20230111133351.807024-1-jannh@google.com
Fixes: f3f0e1d2150b ("khugepaged: add support of collapse for tmpfs/shmem pages")
Signed-off-by: Jann Horn <jannh@google.com>
Reported-by: Zach O'Keefe <zokeefe@google.com>
Acked-by: Kirill A. Shutemov <kirill.shutemov@intel.linux.com>
Reviewed-by: Yang Shi <shy828301@gmail.com>
Cc: David Hildenbrand <david@redhat.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
22 months agomaple_tree: fix mas_empty_area_rev() lower bound validation
Liam Howlett [Wed, 11 Jan 2023 20:02:07 +0000 (20:02 +0000)]
maple_tree: fix mas_empty_area_rev() lower bound validation

mas_empty_area_rev() was not correctly validating the start of a gap
against the lower limit.  This could lead to the range starting lower than
the requested minimum.

Fix the issue by better validating a gap once one is found.

This commit also adds tests to the maple tree test suite for this issue
and tests the mas_empty_area() function for similar bound checking.

Link: https://lkml.kernel.org/r/20230111200136.1851322-1-Liam.Howlett@oracle.com
Link: https://bugzilla.kernel.org/show_bug.cgi?id=216911
Fixes: 54a611b60590 ("Maple Tree: add new data structure")
Signed-off-by: Liam R. Howlett <Liam.Howlett@oracle.com>
Reported-by: <amanieu@gmail.com>
Link: https://lore.kernel.org/linux-mm/0b9f5425-08d4-8013-aa4c-e620c3b10bb2@leemhuis.info/
Tested-by: Holger Hoffsttte <holger@applied-asynchrony.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
23 months agoselftests/filesystems: grant executable permission to run_fat_tests.sh
Pengfei Xu [Wed, 11 Jan 2023 08:15:31 +0000 (16:15 +0800)]
selftests/filesystems: grant executable permission to run_fat_tests.sh

When use tools/testing/selftests/kselftest_install.sh to make the
kselftest-list.txt under tools/testing/selftests/kselftest_install.

Then use tools/testing/selftests/kselftest_install/run_kselftest.sh to run
all the kselftests in kselftest-list.txt, it will be blocked by case
"filesystems/fat: run_fat_tests.sh" with "Warning: file run_fat_tests.sh
is not executable", so grant executable permission to run_fat_tests.sh to
fix this issue.

Link: https://lkml.kernel.org/r/dfdbba6df8a1ab34bb1e81cd8bd7ca3f9ed5c369.1673424747.git.pengfei.xu@intel.com
Fixes: dd7c9be330d8 ("selftests/filesystems: add a vfat RENAME_EXCHANGE test")
Signed-off-by: Pengfei Xu <pengfei.xu@intel.com>
Reviewed-by: Javier Martinez Canillas <javierm@redhat.com>
Cc: Shuah Khan <shuah@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
23 months agoselftests/vm: remove __USE_GNU in hugetlb-madvise.c
Peter Xu [Mon, 16 Jan 2023 16:33:07 +0000 (11:33 -0500)]
selftests/vm: remove __USE_GNU in hugetlb-madvise.c

__USE_GNU should be an internal macro only used inside glibc.  Either
memfd_create() or fallocate() requires _GNU_SOURCE per man page, where
__USE_GNU will further be defined by glibc headers include/features.h:

  #ifdef _GNU_SOURCE
  # define __USE_GNU 1
  #endif

This fixes:

   >> hugetlb-madvise.c:20: warning: "__USE_GNU" redefined
      20 | #define __USE_GNU
         |
   In file included from /usr/include/x86_64-linux-gnu/bits/libc-header-start.h:33,
                    from /usr/include/stdlib.h:26,
                    from hugetlb-madvise.c:16:
   /usr/include/features.h:407: note: this is the location of the previous definition
     407 | # define __USE_GNU      1
         |

Link: https://lkml.kernel.org/r/Y8V9z+z6Tk7NetI3@x1n
Signed-off-by: Peter Xu <peterx@redhat.com>
Reported-by: kernel test robot <lkp@intel.com>
Cc: Mike Kravetz <mike.kravetz@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
23 months agomm: fix a few rare cases of using swapin error pte marker
Peter Xu [Wed, 14 Dec 2022 20:04:53 +0000 (15:04 -0500)]
mm: fix a few rare cases of using swapin error pte marker

This patch should harden commit 15520a3f0469 ("mm: use pte markers for
swap errors") on using pte markers for swapin errors on a few corner
cases.

1. Propagate swapin errors across fork()s: if there're swapin errors in
   the parent mm, after fork()s the child should sigbus too when an error
   page is accessed.

2. Fix a rare condition race in pte_marker_clear() where a uffd-wp pte
   marker can be quickly switched to a swapin error.

3. Explicitly ignore swapin error pte markers in change_protection().

I mostly don't worry on (2) or (3) at all, but we should still have them.
Case (1) is special because it can potentially cause silent data corrupt
on child when parent has swapin error triggered with swapoff, but since
swapin error is rare itself already it's probably not easy to trigger
either.

Currently there is a priority difference between the uffd-wp bit and the
swapin error entry, in which the swapin error always has higher priority
(e.g.  we don't need to wr-protect a swapin error pte marker).

If there will be a 3rd bit introduced, we'll probably need to consider a
more involved approach so we may need to start operate on the bits.  Let's
leave that for later.

This patch is tested with case (1) explicitly where we'll get corrupted
data before in the child if there's existing swapin error pte markers, and
after patch applied the child can be rightfully killed.

We don't need to copy stable for this one since 15520a3f0469 just landed
as part of v6.2-rc1, only "Fixes" applied.

Link: https://lkml.kernel.org/r/20221214200453.1772655-3-peterx@redhat.com
Fixes: 15520a3f0469 ("mm: use pte markers for swap errors")
Signed-off-by: Peter Xu <peterx@redhat.com>
Acked-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Miaohe Lin <linmiaohe@huawei.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: "Huang, Ying" <ying.huang@intel.com>
Cc: Nadav Amit <nadav.amit@gmail.com>
Cc: Pengfei Xu <pengfei.xu@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
23 months agomm/uffd: fix pte marker when fork() without fork event
Peter Xu [Wed, 14 Dec 2022 20:04:52 +0000 (15:04 -0500)]
mm/uffd: fix pte marker when fork() without fork event

Patch series "mm: Fixes on pte markers".

Patch 1 resolves the syzkiller report from Pengfei.

Patch 2 further harden pte markers when used with the recent swapin error
markers.  The major case is we should persist a swapin error marker after
fork(), so child shouldn't read a corrupted page.

This patch (of 2):

When fork(), dst_vma is not guaranteed to have VM_UFFD_WP even if src may
have it and has pte marker installed.  The warning is improper along with
the comment.  The right thing is to inherit the pte marker when needed, or
keep the dst pte empty.

A vague guess is this happened by an accident when there's the prior patch
to introduce src/dst vma into this helper during the uffd-wp feature got
developed and I probably messed up in the rebase, since if we replace
dst_vma with src_vma the warning & comment it all makes sense too.

Hugetlb did exactly the right here (copy_hugetlb_page_range()).  Fix the
general path.

Reproducer:

https://github.com/xupengfe/syzkaller_logs/blob/main/221208_115556_copy_page_range/repro.c

Bugzilla report: https://bugzilla.kernel.org/show_bug.cgi?id=216808

Link: https://lkml.kernel.org/r/20221214200453.1772655-1-peterx@redhat.com
Link: https://lkml.kernel.org/r/20221214200453.1772655-2-peterx@redhat.com
Fixes: c56d1b62cce8 ("mm/shmem: handle uffd-wp during fork()")
Signed-off-by: Peter Xu <peterx@redhat.com>
Reported-by: Pengfei Xu <pengfei.xu@intel.com>
Acked-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Miaohe Lin <linmiaohe@huawei.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: "Huang, Ying" <ying.huang@intel.com>
Cc: Nadav Amit <nadav.amit@gmail.com>
Cc: <stable@vger.kernel.org> # 5.19+
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
23 months agoSync with v6.2-rc4
Andrew Morton [Thu, 19 Jan 2023 00:52:20 +0000 (16:52 -0800)]
Sync with v6.2-rc4

Merge branch 'master' into mm-hotfixes-stable

23 months agoLinux 6.2-rc4
Linus Torvalds [Sun, 15 Jan 2023 15:22:43 +0000 (09:22 -0600)]
Linux 6.2-rc4

23 months agoMerge tag 'x86_urgent_for_v6.2_rc4' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 15 Jan 2023 13:17:44 +0000 (07:17 -0600)]
Merge tag 'x86_urgent_for_v6.2_rc4' of git://git./linux/kernel/git/tip/tip

Pull x86 fixes from Borislav Petkov:

 - Make sure the poking PGD is pinned for Xen PV as it requires it this
   way

 - Fixes for two resctrl races when moving a task or creating a new
   monitoring group

 - Fix SEV-SNP guests running under HyperV where MTRRs are disabled to
   not return a UC- type mapping type on memremap() and thus cause a
   serious slowdown

 - Fix insn mnemonics in bioscall.S now that binutils is starting to fix
   confusing insn suffixes

* tag 'x86_urgent_for_v6.2_rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/mm: fix poking_init() for Xen PV guests
  x86/resctrl: Fix event counts regression in reused RMIDs
  x86/resctrl: Fix task CLOSID/RMID update race
  x86/pat: Fix pat_x_mtrr_type() for MTRR disabled case
  x86/boot: Avoid using Intel mnemonics in AT&T syntax asm

23 months agoMerge tag 'edac_urgent_for_v6.2_rc4' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Sun, 15 Jan 2023 13:12:58 +0000 (07:12 -0600)]
Merge tag 'edac_urgent_for_v6.2_rc4' of git://git./linux/kernel/git/ras/ras

Pull EDAC fixes from Borislav Petkov:

 - Fix the EDAC device's confusion in the polling setting units

 - Fix a memory leak in highbank's probing function

* tag 'edac_urgent_for_v6.2_rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/ras/ras:
  EDAC/highbank: Fix memory leak in highbank_mc_probe()
  EDAC/device: Fix period calculation in edac_device_reset_delay_period()

23 months agoMerge tag 'powerpc-6.2-3' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc...
Linus Torvalds [Sun, 15 Jan 2023 13:09:41 +0000 (07:09 -0600)]
Merge tag 'powerpc-6.2-3' of git://git./linux/kernel/git/powerpc/linux

Pull powerpc fixes from Michael Ellerman:

 - Fix a build failure with some versions of ld that have an odd version
   string

 - Fix incorrect use of mutex in the IMC PMU driver

Thanks to Kajol Jain, Michael Petlan, Ojaswin Mujoo, Peter Zijlstra, and
Yang Yingliang.

* tag 'powerpc-6.2-3' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux:
  powerpc/64s/hash: Make stress_hpt_timer_fn() static
  powerpc/imc-pmu: Fix use of mutex in IRQs disabled section
  powerpc/boot: Fix incorrect version calculation issue in ld_version

23 months agoMerge tag 'iommu-fixes-v6.2-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Sat, 14 Jan 2023 16:48:15 +0000 (10:48 -0600)]
Merge tag 'iommu-fixes-v6.2-rc3' of git://git./linux/kernel/git/joro/iommu

Pull iommu fixes from Joerg Roedel:

 - Core: Fix an iommu-group refcount leak

 - Fix overflow issue in IOVA alloc path

 - ARM-SMMU fixes from Will:
    - Fix VFIO regression on NXP SoCs by reporting IOMMU_CAP_CACHE_COHERENCY
    - Fix SMMU shutdown paths to avoid device unregistration race

 - Error handling fix for Mediatek IOMMU driver

* tag 'iommu-fixes-v6.2-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu:
  iommu/mediatek-v1: Fix an error handling path in mtk_iommu_v1_probe()
  iommu/iova: Fix alloc iova overflows issue
  iommu: Fix refcount leak in iommu_device_claim_dma_owner
  iommu/arm-smmu-v3: Don't unregister on shutdown
  iommu/arm-smmu: Don't unregister on shutdown
  iommu/arm-smmu: Report IOMMU_CAP_CACHE_COHERENCY even betterer

23 months agoMerge tag 'fixes-2023-01-14' of git://git.kernel.org/pub/scm/linux/kernel/git/rppt...
Linus Torvalds [Sat, 14 Jan 2023 16:08:08 +0000 (10:08 -0600)]
Merge tag 'fixes-2023-01-14' of git://git./linux/kernel/git/rppt/memblock

Pull memblock fix from Mike Rapoport:
 "memblock: always release pages to the buddy allocator in
  memblock_free_late()

  If CONFIG_DEFERRED_STRUCT_PAGE_INIT is enabled, memblock_free_pages()
  only releases pages to the buddy allocator if they are not in the
  deferred range. This is correct for free pages (as defined by
  for_each_free_mem_pfn_range_in_zone()) because free pages in the
  deferred range will be initialized and released as part of the
  deferred init process.

  memblock_free_pages() is called by memblock_free_late(), which is used
  to free reserved ranges after memblock_free_all() has run. All pages
  in reserved ranges have been initialized at that point, and
  accordingly, those pages are not touched by the deferred init process.

  This means that currently, if the pages that memblock_free_late()
  intends to release are in the deferred range, they will never be
  released to the buddy allocator. They will forever be reserved.

  In addition, memblock_free_pages() calls kmsan_memblock_free_pages(),
  which is also correct for free pages but is not correct for reserved
  pages. KMSAN metadata for reserved pages is initialized by
  kmsan_init_shadow(), which runs shortly before memblock_free_all().

  For both of these reasons, memblock_free_pages() should only be called
  for free pages, and memblock_free_late() should call
  __free_pages_core() directly instead.

  One case where this issue can occur in the wild is EFI boot on x86_64.
  The x86 EFI code reserves all EFI boot services memory ranges via
  memblock_reserve() and frees them later via memblock_free_late()
  (efi_reserve_boot_services() and efi_free_boot_services(),
  respectively).

  If any of those ranges happens to fall within the deferred init range,
  the pages will not be released and that memory will be unavailable.

  For example, on an Amazon EC2 t3.micro VM (1 GB) booting via EFI:

    v6.2-rc2:
    Node 0, zone      DMA
          spanned  4095
          present  3999
          managed  3840
    Node 0, zone    DMA32
          spanned  246652
          present  245868
          managed  178867

    v6.2-rc2 + patch:
    Node 0, zone      DMA
          spanned  4095
          present  3999
          managed  3840
    Node 0, zone    DMA32
          spanned  246652
          present  245868
          managed  222816   # +43,949 pages"

* tag 'fixes-2023-01-14' of git://git.kernel.org/pub/scm/linux/kernel/git/rppt/memblock:
  mm: Always release pages to the buddy allocator in memblock_free_late().

23 months agoMerge tag 'hardening-v6.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/kees...
Linus Torvalds [Sat, 14 Jan 2023 16:04:00 +0000 (10:04 -0600)]
Merge tag 'hardening-v6.2-rc4' of git://git./linux/kernel/git/kees/linux

Pull kernel hardening fixes from Kees Cook:

 - Fix CFI hash randomization with KASAN (Sami Tolvanen)

 - Check size of coreboot table entry and use flex-array

* tag 'hardening-v6.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
  kbuild: Fix CFI hash randomization with KASAN
  firmware: coreboot: Check size of table entry and use flex-array

23 months agoMerge tag 'modules-6.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/mcgrof...
Linus Torvalds [Sat, 14 Jan 2023 14:17:27 +0000 (08:17 -0600)]
Merge tag 'modules-6.2-rc4' of git://git./linux/kernel/git/mcgrof/linux

Pull module fix from Luis Chamberlain:
 "Just one fix for modules by Nick"

* tag 'modules-6.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/mcgrof/linux:
  kallsyms: Fix scheduling with interrupts disabled in self-test

23 months agoMerge tag '6.2-rc3-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6
Linus Torvalds [Sat, 14 Jan 2023 14:08:25 +0000 (08:08 -0600)]
Merge tag '6.2-rc3-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6

Pull cifs fixes from Steve French:

 - memory leak and double free fix

 - two symlink fixes

 - minor cleanup fix

 - two smb1 fixes

* tag '6.2-rc3-smb3-client-fixes' of git://git.samba.org/sfrench/cifs-2.6:
  cifs: Fix uninitialized memory read for smb311 posix symlink create
  cifs: fix potential memory leaks in session setup
  cifs: do not query ifaces on smb1 mounts
  cifs: fix double free on failed kerberos auth
  cifs: remove redundant assignment to the variable match
  cifs: fix file info setting in cifs_open_file()
  cifs: fix file info setting in cifs_query_path_info()

23 months agoMerge tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Linus Torvalds [Sat, 14 Jan 2023 13:57:25 +0000 (07:57 -0600)]
Merge tag 'scsi-fixes' of git://git./linux/kernel/git/jejb/scsi

Pull SCSI fixes from James Bottomley:
 "Two minor fixes in the hisi_sas driver which only impact enterprise
  style multi-expander and shared disk situations and no core changes"

* tag 'scsi-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi:
  scsi: hisi_sas: Set a port invalid only if there are no devices attached when refreshing port id
  scsi: hisi_sas: Use abort task set to reset SAS disks when discovered

23 months agoMerge tag 'ata-6.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/dlemoal...
Linus Torvalds [Sat, 14 Jan 2023 13:52:11 +0000 (07:52 -0600)]
Merge tag 'ata-6.2-rc4' of git://git./linux/kernel/git/dlemoal/libata

Pull ATA fix from Damien Le Moal:
 "A single fix to prevent building the pata_cs5535 driver with user mode
  linux as it uses msr operations that are not defined with UML"

* tag 'ata-6.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/dlemoal/libata:
  ata: pata_cs5535: Don't build on UML

23 months agoMerge tag 'block-6.2-2023-01-13' of git://git.kernel.dk/linux
Linus Torvalds [Fri, 13 Jan 2023 23:41:19 +0000 (17:41 -0600)]
Merge tag 'block-6.2-2023-01-13' of git://git.kernel.dk/linux

Pull block fixes from Jens Axboe:
 "Nothing major in here, just a collection of NVMe fixes and dropping a
  wrong might_sleep() that static checkers tripped over but which isn't
  valid"

* tag 'block-6.2-2023-01-13' of git://git.kernel.dk/linux:
  MAINTAINERS: stop nvme matching for nvmem files
  nvme: don't allow unprivileged passthrough on partitions
  nvme: replace the "bool vec" arguments with flags in the ioctl path
  nvme: remove __nvme_ioctl
  nvme-pci: fix error handling in nvme_pci_enable()
  nvme-pci: add NVME_QUIRK_IDENTIFY_CNS quirk to Apple T2 controllers
  nvme-apple: add NVME_QUIRK_IDENTIFY_CNS quirk to fix regression
  block: Drop spurious might_sleep() from blk_put_queue()

23 months agoMerge tag 'io_uring-6.2-2023-01-13' of git://git.kernel.dk/linux
Linus Torvalds [Fri, 13 Jan 2023 23:37:09 +0000 (17:37 -0600)]
Merge tag 'io_uring-6.2-2023-01-13' of git://git.kernel.dk/linux

Pull io_uring fixes from Jens Axboe:
 "A fix for a regression that happened last week, rest is fixes that
  will be headed to stable as well. In detail:

   - Fix for a regression added with the leak fix from last week (me)

   - In writing a test case for that leak, inadvertently discovered a
     case where we a poll request can race. So fix that up and mark it
     for stable, and also ensure that fdinfo covers both the poll tables
     that we have. The latter was an oversight when the split poll table
     were added (me)

   - Fix for a lockdep reported issue with IOPOLL (Pavel)"

* tag 'io_uring-6.2-2023-01-13' of git://git.kernel.dk/linux:
  io_uring: lock overflowing for IOPOLL
  io_uring/poll: attempt request issue after racy poll wakeup
  io_uring/fdinfo: include locked hash table in fdinfo output
  io_uring/poll: add hash if ready poll request can't complete inline
  io_uring/io-wq: only free worker if it was allocated for creation

23 months agoMerge tag 'pci-v6.2-fixes-1' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci
Linus Torvalds [Fri, 13 Jan 2023 23:32:22 +0000 (17:32 -0600)]
Merge tag 'pci-v6.2-fixes-1' of git://git./linux/kernel/git/helgaas/pci

Pull pci fixes from Bjorn Helgaas:

 - Work around apparent firmware issue that made Linux reject MMCONFIG
   space, which broke PCI extended config space (Bjorn Helgaas)

 - Fix CONFIG_PCIE_BT1 dependency due to mid-air collision between a
   PCI_MSI_IRQ_DOMAIN -> PCI_MSI change and addition of PCIE_BT1 (Lukas
   Bulwahn)

* tag 'pci-v6.2-fixes-1' of git://git.kernel.org/pub/scm/linux/kernel/git/helgaas/pci:
  x86/pci: Treat EfiMemoryMappedIO as reservation of ECAM space
  x86/pci: Simplify is_mmconf_reserved() messages
  PCI: dwc: Adjust to recent removal of PCI_MSI_IRQ_DOMAIN

23 months agokbuild: Fix CFI hash randomization with KASAN
Sami Tolvanen [Thu, 12 Jan 2023 22:49:48 +0000 (22:49 +0000)]
kbuild: Fix CFI hash randomization with KASAN

Clang emits a asan.module_ctor constructor to each object file
when KASAN is enabled, and these functions are indirectly called
in do_ctors. With CONFIG_CFI_CLANG, the compiler also emits a CFI
type hash before each address-taken global function so they can
pass indirect call checks.

However, in commit 0c3e806ec0f9 ("x86/cfi: Add boot time hash
randomization"), x86 implemented boot time hash randomization,
which relies on the .cfi_sites section generated by objtool. As
objtool is run against vmlinux.o instead of individual object
files with X86_KERNEL_IBT (enabled by default), CFI types in
object files that are not part of vmlinux.o end up not being
included in .cfi_sites, and thus won't get randomized and trip
CFI when called.

Only .vmlinux.export.o and init/version-timestamp.o are linked
into vmlinux separately from vmlinux.o. As these files don't
contain any functions, disable KASAN for both of them to avoid
breaking hash randomization.

Link: https://github.com/ClangBuiltLinux/linux/issues/1742
Fixes: 0c3e806ec0f9 ("x86/cfi: Add boot time hash randomization")
Signed-off-by: Sami Tolvanen <samitolvanen@google.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Kees Cook <keescook@chromium.org>
Link: https://lore.kernel.org/r/20230112224948.1479453-2-samitolvanen@google.com
23 months agofirmware: coreboot: Check size of table entry and use flex-array
Kees Cook [Thu, 12 Jan 2023 23:03:16 +0000 (15:03 -0800)]
firmware: coreboot: Check size of table entry and use flex-array

The memcpy() of the data following a coreboot_table_entry couldn't
be evaluated by the compiler under CONFIG_FORTIFY_SOURCE. To make it
easier to reason about, add an explicit flexible array member to struct
coreboot_device so the entire entry can be copied at once. Additionally,
validate the sizes before copying. Avoids this run-time false positive
warning:

  memcpy: detected field-spanning write (size 168) of single field "&device->entry" at drivers/firmware/google/coreboot_table.c:103 (size 8)

Reported-by: Paul Menzel <pmenzel@molgen.mpg.de>
Link: https://lore.kernel.org/all/03ae2704-8c30-f9f0-215b-7cdf4ad35a9a@molgen.mpg.de/
Cc: Jack Rosenthal <jrosenth@chromium.org>
Cc: Guenter Roeck <groeck@chromium.org>
Cc: Julius Werner <jwerner@chromium.org>
Cc: Brian Norris <briannorris@chromium.org>
Cc: Stephen Boyd <swboyd@chromium.org>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Kees Cook <keescook@chromium.org>
Reviewed-by: Julius Werner <jwerner@chromium.org>
Reviewed-by: Guenter Roeck <groeck@chromium.org>
Link: https://lore.kernel.org/r/20230107031406.gonna.761-kees@kernel.org
Reviewed-by: Stephen Boyd <swboyd@chromium.org>
Reviewed-by: Jack Rosenthal <jrosenth@chromium.org>
Link: https://lore.kernel.org/r/20230112230312.give.446-kees@kernel.org
23 months agokallsyms: Fix scheduling with interrupts disabled in self-test
Nicholas Piggin [Thu, 12 Jan 2023 10:54:26 +0000 (20:54 +1000)]
kallsyms: Fix scheduling with interrupts disabled in self-test

kallsyms_on_each* may schedule so must not be called with interrupts
disabled. The iteration function could disable interrupts, but this
also changes lookup_symbol() to match the change to the other timing
code.

Reported-by: Erhard F. <erhard_f@mailbox.org>
Link: https://lore.kernel.org/all/bug-216902-206035@https.bugzilla.kernel.org%2F/
Reported-by: kernel test robot <oliver.sang@intel.com>
Link: https://lore.kernel.org/oe-lkp/202212251728.8d0872ff-oliver.sang@intel.com
Fixes: 30f3bb09778d ("kallsyms: Add self-test facility")
Tested-by: "Erhard F." <erhard_f@mailbox.org>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Luis Chamberlain <mcgrof@kernel.org>
23 months agoata: pata_cs5535: Don't build on UML
Peter Foley [Fri, 13 Jan 2023 04:37:06 +0000 (23:37 -0500)]
ata: pata_cs5535: Don't build on UML

This driver uses MSR functions that aren't implemented under UML.
Avoid building it to prevent tripping up allyesconfig.

e.g.
/usr/lib/gcc/x86_64-pc-linux-gnu/12/../../../../x86_64-pc-linux-gnu/bin/ld: pata_cs5535.c:(.text+0x3a3): undefined reference to `__tracepoint_read_msr'
/usr/lib/gcc/x86_64-pc-linux-gnu/12/../../../../x86_64-pc-linux-gnu/bin/ld: pata_cs5535.c:(.text+0x3d2): undefined reference to `__tracepoint_write_msr'
/usr/lib/gcc/x86_64-pc-linux-gnu/12/../../../../x86_64-pc-linux-gnu/bin/ld: pata_cs5535.c:(.text+0x457): undefined reference to `__tracepoint_write_msr'
/usr/lib/gcc/x86_64-pc-linux-gnu/12/../../../../x86_64-pc-linux-gnu/bin/ld: pata_cs5535.c:(.text+0x481): undefined reference to `do_trace_write_msr'
/usr/lib/gcc/x86_64-pc-linux-gnu/12/../../../../x86_64-pc-linux-gnu/bin/ld: pata_cs5535.c:(.text+0x4d5): undefined reference to `do_trace_write_msr'
/usr/lib/gcc/x86_64-pc-linux-gnu/12/../../../../x86_64-pc-linux-gnu/bin/ld: pata_cs5535.c:(.text+0x4f5): undefined reference to `do_trace_read_msr'
/usr/lib/gcc/x86_64-pc-linux-gnu/12/../../../../x86_64-pc-linux-gnu/bin/ld: pata_cs5535.c:(.text+0x51c): undefined reference to `do_trace_write_msr'

Signed-off-by: Peter Foley <pefoley2@pefoley.com>
Reviewed-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Damien Le Moal <damien.lemoal@opensource.wdc.com>
23 months agoMerge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Linus Torvalds [Fri, 13 Jan 2023 20:41:50 +0000 (14:41 -0600)]
Merge tag 'for-linus' of git://git./virt/kvm/kvm

Pull kvm fixes from Paolo Bonzini:
 "ARM:

   - Fix the PMCR_EL0 reset value after the PMU rework

   - Correctly handle S2 fault triggered by a S1 page table walk by not
     always classifying it as a write, as this breaks on R/O memslots

   - Document why we cannot exit with KVM_EXIT_MMIO when taking a write
     fault from a S1 PTW on a R/O memslot

   - Put the Apple M2 on the naughty list for not being able to
     correctly implement the vgic SEIS feature, just like the M1 before
     it

   - Reviewer updates: Alex is stepping down, replaced by Zenghui

  x86:

   - Fix various rare locking issues in Xen emulation and teach lockdep
     to detect them

   - Documentation improvements

   - Do not return host topology information from KVM_GET_SUPPORTED_CPUID"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  KVM: x86/xen: Avoid deadlock by adding kvm->arch.xen.xen_lock leaf node lock
  KVM: Ensure lockdep knows about kvm->lock vs. vcpu->mutex ordering rule
  KVM: x86/xen: Fix potential deadlock in kvm_xen_update_runstate_guest()
  KVM: x86/xen: Fix lockdep warning on "recursive" gpc locking
  Documentation: kvm: fix SRCU locking order docs
  KVM: x86: Do not return host topology information from KVM_GET_SUPPORTED_CPUID
  KVM: nSVM: clarify recalc_intercepts() wrt CR8
  MAINTAINERS: Remove myself as a KVM/arm64 reviewer
  MAINTAINERS: Add Zenghui Yu as a KVM/arm64 reviewer
  KVM: arm64: vgic: Add Apple M2 cpus to the list of broken SEIS implementations
  KVM: arm64: Convert FSC_* over to ESR_ELx_FSC_*
  KVM: arm64: Document the behaviour of S1PTW faults on RO memslots
  KVM: arm64: Fix S1PTW handling on RO memslots
  KVM: arm64: PMU: Fix PMCR_EL0 reset value

23 months agolockref: stop doing cpu_relax in the cmpxchg loop
Mateusz Guzik [Fri, 13 Jan 2023 18:44:47 +0000 (19:44 +0100)]
lockref: stop doing cpu_relax in the cmpxchg loop

On the x86-64 architecture even a failing cmpxchg grants exclusive
access to the cacheline, making it preferable to retry the failed op
immediately instead of stalling with the pause instruction.

To illustrate the impact, below are benchmark results obtained by
running various will-it-scale tests on top of the 6.2-rc3 kernel and
Cascade Lake (2 sockets * 24 cores * 2 threads) CPU.

All results in ops/s.  Note there is some variance in re-runs, but the
code is consistently faster when contention is present.

  open3 ("Same file open/close"):
  proc          stock       no-pause
     1         805603         814942       (+%1)
     2        1054980        1054781       (-0%)
     8        1544802        1822858      (+18%)
    24        1191064        2199665      (+84%)
    48         851582        1469860      (+72%)
    96         609481        1427170     (+134%)

  fstat2 ("Same file fstat"):
  proc          stock       no-pause
     1        3013872        3047636       (+1%)
     2        4284687        4400421       (+2%)
     8        3257721        5530156      (+69%)
    24        2239819        5466127     (+144%)
    48        1701072        5256609     (+209%)
    96        1269157        6649326     (+423%)

Additionally, a kernel with a private patch to help access() scalability:
access2 ("Same file access"):

  proc          stock        patched      patched
                                         +nopause
    24        2378041        2005501      5370335  (-15% / +125%)

That is, fixing the problems in access itself *reduces* scalability
after the cacheline ping-pong only happens in lockref with the pause
instruction.

Note that fstat and access benchmarks are not currently integrated into
will-it-scale, but interested parties can find them in pull requests to
said project.

Code at hand has a rather tortured history.  First modification showed
up in commit d472d9d98b46 ("lockref: Relax in cmpxchg loop"), written
with Itanium in mind.  Later it got patched up to use an arch-dependent
macro to stop doing it on s390 where it caused a significant regression.
Said macro had undergone revisions and was ultimately eliminated later,
going back to cpu_relax.

While I intended to only remove cpu_relax for x86-64, I got the
following comment from Linus:

    I would actually prefer just removing it entirely and see if
    somebody else hollers. You have the numbers to prove it hurts on
    real hardware, and I don't think we have any numbers to the
    contrary.

    So I think it's better to trust the numbers and remove it as a
    failure, than say "let's just remove it on x86-64 and leave
    everybody else with the potentially broken code"

Additionally, Will Deacon (maintainer of the arm64 port, one of the
architectures previously benchmarked):

    So, from the arm64 side of the fence, I'm perfectly happy just
    removing the cpu_relax() calls from lockref.

As such, come back full circle in history and whack it altogether.

Signed-off-by: Mateusz Guzik <mjguzik@gmail.com>
Link: https://lore.kernel.org/all/CAGudoHHx0Nqg6DE70zAVA75eV-HXfWyhVMWZ-aSeOofkA_=WdA@mail.gmail.com/
Acked-by: Tony Luck <tony.luck@intel.com> # ia64
Acked-by: Nicholas Piggin <npiggin@gmail.com> # powerpc
Acked-by: Will Deacon <will@kernel.org> # arm64
Acked-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
23 months agox86/pci: Treat EfiMemoryMappedIO as reservation of ECAM space
Bjorn Helgaas [Tue, 10 Jan 2023 18:02:43 +0000 (12:02 -0600)]
x86/pci: Treat EfiMemoryMappedIO as reservation of ECAM space

Normally we reject ECAM space unless it is reported as reserved in the E820
table or via a PNP0C02 _CRS method (PCI Firmware, r3.3, sec 4.1.2).

07eab0901ede ("efi/x86: Remove EfiMemoryMappedIO from E820 map"), removes
E820 entries that correspond to EfiMemoryMappedIO regions because some
other firmware uses EfiMemoryMappedIO for PCI host bridge windows, and the
E820 entries prevent Linux from allocating BAR space for hot-added devices.

Some firmware doesn't report ECAM space via PNP0C02 _CRS methods, but does
mention it as an EfiMemoryMappedIO region via EFI GetMemoryMap(), which is
normally converted to an E820 entry by a bootloader or EFI stub.  After
07eab0901ede, that E820 entry is removed, so we reject this ECAM space,
which makes PCI extended config space (offsets 0x100-0xfff) inaccessible.

The lack of extended config space breaks anything that relies on it,
including perf, VSEC telemetry, EDAC, QAT, SR-IOV, etc.

Allow use of ECAM for extended config space when the region is covered by
an EfiMemoryMappedIO region, even if it's not included in E820 or PNP0C02
_CRS.

Link: https://lore.kernel.org/r/ac2693d8-8ba3-72e0-5b66-b3ae008d539d@linux.intel.com
Link: https://bugzilla.kernel.org/show_bug.cgi?id=216891
Fixes: 07eab0901ede ("efi/x86: Remove EfiMemoryMappedIO from E820 map")
Link: https://lore.kernel.org/r/20230110180243.1590045-3-helgaas@kernel.org
Reported-by: Kan Liang <kan.liang@linux.intel.com>
Reported-by: Tony Luck <tony.luck@intel.com>
Reported-by: Giovanni Cabiddu <giovanni.cabiddu@intel.com>
Reported-by: Yunying Sun <yunying.sun@intel.com>
Reported-by: Baowen Zheng <baowen.zheng@corigine.com>
Reported-by: Zhenzhong Duan <zhenzhong.duan@intel.com>
Reported-by: Yang Lixiao <lixiao.yang@intel.com>
Tested-by: Tony Luck <tony.luck@intel.com>
Tested-by: Giovanni Cabiddu <giovanni.cabiddu@intel.com>
Tested-by: Kan Liang <kan.liang@linux.intel.com>
Tested-by: Yunying Sun <yunying.sun@intel.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: Rafael J. Wysocki <rafael@kernel.org>
23 months agoMerge tag 'efi-fixes-for-v6.2-1' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Fri, 13 Jan 2023 16:37:10 +0000 (10:37 -0600)]
Merge tag 'efi-fixes-for-v6.2-1' of git://git./linux/kernel/git/efi/efi

Pull EFI fixes from Ard Biesheuvel:

 - avoid a potential crash on the efi_subsys_init() error path

 - use more appropriate error code for runtime services calls issued
   after a crash in the firmware occurred

 - avoid READ_ONCE() for accessing firmware tables that may appear
   misaligned in memory

* tag 'efi-fixes-for-v6.2-1' of git://git.kernel.org/pub/scm/linux/kernel/git/efi/efi:
  efi: tpm: Avoid READ_ONCE() for accessing the event log
  efi: rt-wrapper: Add missing include
  efi: fix userspace infinite retry read efivars after EFI runtime services page fault
  efi: fix NULL-deref in init error path

23 months agoMerge tag 'docs-6.2-fixes' of git://git.lwn.net/linux
Linus Torvalds [Fri, 13 Jan 2023 16:35:26 +0000 (10:35 -0600)]
Merge tag 'docs-6.2-fixes' of git://git.lwn.net/linux

Pull documentation fixes from Jonathan Corbet:
 "Three documentation fixes (or rather two and one warning):

   - Sphinx 6.0 broke our configuration mechanism, so fix it

   - I broke our configuration for non-Alabaster themes; Akira fixed it

   - Deprecate Sphinx < 2.4 with an eye toward future removal"

* tag 'docs-6.2-fixes' of git://git.lwn.net/linux:
  docs/conf.py: Use about.html only in sidebar of alabaster theme
  docs: Deprecate use of Sphinx < 2.4.x
  docs: Fix the docs build with Sphinx 6.0

23 months agoefi: tpm: Avoid READ_ONCE() for accessing the event log
Ard Biesheuvel [Mon, 9 Jan 2023 09:44:31 +0000 (10:44 +0100)]
efi: tpm: Avoid READ_ONCE() for accessing the event log

Nathan reports that recent kernels built with LTO will crash when doing
EFI boot using Fedora's GRUB and SHIM. The culprit turns out to be a
misaligned load from the TPM event log, which is annotated with
READ_ONCE(), and under LTO, this gets translated into a LDAR instruction
which does not tolerate misaligned accesses.

Interestingly, this does not happen when booting the same kernel
straight from the UEFI shell, and so the fact that the event log may
appear misaligned in memory may be caused by a bug in GRUB or SHIM.

However, using READ_ONCE() to access firmware tables is slightly unusual
in any case, and here, we only need to ensure that 'event' is not
dereferenced again after it gets unmapped, but this is already taken
care of by the implicit barrier() semantics of the early_memunmap()
call.

Cc: <stable@vger.kernel.org>
Cc: Peter Jones <pjones@redhat.com>
Cc: Jarkko Sakkinen <jarkko@kernel.org>
Cc: Matthew Garrett <mjg59@srcf.ucam.org>
Reported-by: Nathan Chancellor <nathan@kernel.org>
Tested-by: Nathan Chancellor <nathan@kernel.org>
Link: https://github.com/ClangBuiltLinux/linux/issues/1782
Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
23 months agoio_uring: lock overflowing for IOPOLL
Pavel Begunkov [Thu, 12 Jan 2023 13:08:56 +0000 (13:08 +0000)]
io_uring: lock overflowing for IOPOLL

syzbot reports an issue with overflow filling for IOPOLL:

WARNING: CPU: 0 PID: 28 at io_uring/io_uring.c:734 io_cqring_event_overflow+0x1c0/0x230 io_uring/io_uring.c:734
CPU: 0 PID: 28 Comm: kworker/u4:1 Not tainted 6.2.0-rc3-syzkaller-16369-g358a161a6a9e #0
Workqueue: events_unbound io_ring_exit_work
Call trace:
 io_cqring_event_overflow+0x1c0/0x230 io_uring/io_uring.c:734
 io_req_cqe_overflow+0x5c/0x70 io_uring/io_uring.c:773
 io_fill_cqe_req io_uring/io_uring.h:168 [inline]
 io_do_iopoll+0x474/0x62c io_uring/rw.c:1065
 io_iopoll_try_reap_events+0x6c/0x108 io_uring/io_uring.c:1513
 io_uring_try_cancel_requests+0x13c/0x258 io_uring/io_uring.c:3056
 io_ring_exit_work+0xec/0x390 io_uring/io_uring.c:2869
 process_one_work+0x2d8/0x504 kernel/workqueue.c:2289
 worker_thread+0x340/0x610 kernel/workqueue.c:2436
 kthread+0x12c/0x158 kernel/kthread.c:376
 ret_from_fork+0x10/0x20 arch/arm64/kernel/entry.S:863

There is no real problem for normal IOPOLL as flush is also called with
uring_lock taken, but it's getting more complicated for IOPOLL|SQPOLL,
for which __io_cqring_overflow_flush() happens from the CQ waiting path.

Reported-and-tested-by: syzbot+6805087452d72929404e@syzkaller.appspotmail.com
Cc: stable@vger.kernel.org # 5.10+
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
23 months agoMerge tag 'sound-6.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai...
Linus Torvalds [Fri, 13 Jan 2023 14:20:29 +0000 (08:20 -0600)]
Merge tag 'sound-6.2-rc4' of git://git./linux/kernel/git/tiwai/sound

Pull sound fixes from Takashi Iwai:
 "This became a slightly big update, but it's more or less expected, as
  the first batch after holidays.

  All changes (but for the last two last-minute fixes) have been stewed
  in linux-next long enough, so it's fairly safe to take:

   - PCM UAF fix in 32bit compat layer

   - ASoC board-specific fixes for Intel, AMD, Medathek, Qualcomm

   - SOF power management fixes

   - ASoC Intel link failure fixes

   - A series of fixes for USB-audio regressions

   - CS35L41 HD-audio codec regression fixes

   - HD-audio device-specific fixes / quirks

  Note that one SPI patch has been taken in ASoC subtree mistakenly, and
  the same fix is found in spi tree, but it should be OK to apply"

* tag 'sound-6.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound: (39 commits)
  ALSA: pcm: Move rwsem lock inside snd_ctl_elem_read to prevent UAF
  ALSA: usb-audio: Fix possible NULL pointer dereference in snd_usb_pcm_has_fixed_rate()
  ALSA: hda/realtek: Enable mute/micmute LEDs on HP Spectre x360 13-aw0xxx
  ASoC: fsl-asoc-card: Fix naming of AC'97 CODEC widgets
  ASoC: fsl_ssi: Rename AC'97 streams to avoid collisions with AC'97 CODEC
  ALSA: hda/hdmi: Add a HP device 0x8715 to force connect list
  ALSA: control-led: use strscpy in set_led_id()
  ALSA: usb-audio: Always initialize fixed_rate in snd_usb_find_implicit_fb_sync_format()
  ASoC: dt-bindings: qcom,lpass-tx-macro: correct clocks on SC7280
  ASoC: dt-bindings: qcom,lpass-wsa-macro: correct clocks on SM8250
  ASoC: qcom: Fix building APQ8016 machine driver without SOUNDWIRE
  ALSA: hda: cs35l41: Check runtime suspend capability at runtime_idle
  ALSA: hda: cs35l41: Don't return -EINVAL from system suspend/resume
  ASoC: fsl_micfil: Correct the number of steps on SX controls
  ALSA: hda/realtek: fix mute/micmute LEDs don't work for a HP platform
  Revert "ALSA: usb-audio: Drop superfluous interface setup at parsing"
  ALSA: usb-audio: More refactoring of hw constraint rules
  ALSA: usb-audio: Relax hw constraints for implicit fb sync
  ALSA: usb-audio: Make sure to stop endpoints before closing EPs
  ALSA: hda - Enable headset mic on another Dell laptop with ALC3254
  ...

23 months agoMerge tag 'pm-6.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Linus Torvalds [Fri, 13 Jan 2023 13:38:14 +0000 (07:38 -0600)]
Merge tag 'pm-6.2-rc4' of git://git./linux/kernel/git/rafael/linux-pm

Pull power management fixes from Rafael Wysocki:
 "These fix assorted issues in the ARM cpufreq drivers and in the AMD
  P-state driver.

  Specifics:

   - Fix cpufreq policy reference counting in amd-pstate to prevent it
     from crashing on removal (Perry Yuan)

   - Fix double initialization and set suspend-freq for Apple's cpufreq
     driver (Arnd Bergmann, Hector Martin)

   - Fix reading of "reg" property, update cpufreq-dt's blocklist and
     update DT documentation for Qualcomm's cpufreq driver (Konrad
     Dybcio, Krzysztof Kozlowski)

   - Replace 0 with NULL in the Armada cpufreq driver (Miles Chen)

   - Fix potential overflows in the CPPC cpufreq driver (Pierre Gondois)

   - Update blocklist for the Tegra234 Soc cpufreq driver (Sumit Gupta)"

* tag 'pm-6.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  cpufreq: amd-pstate: fix kernel hang issue while amd-pstate unregistering
  cpufreq: armada-37xx: stop using 0 as NULL pointer
  cpufreq: apple-soc: Switch to the lowest frequency on suspend
  dt-bindings: cpufreq: cpufreq-qcom-hw: document interrupts
  cpufreq: Add SM6375 to cpufreq-dt-platdev blocklist
  cpufreq: Add Tegra234 to cpufreq-dt-platdev blocklist
  cpufreq: qcom-hw: Fix reading "reg" with address/size-cells != 2
  cpufreq: CPPC: Add u64 casts to avoid overflowing
  cpufreq: apple: remove duplicate intializer

23 months agoMerge tag 'acpi-6.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael...
Linus Torvalds [Fri, 13 Jan 2023 13:32:55 +0000 (07:32 -0600)]
Merge tag 'acpi-6.2-rc4' of git://git./linux/kernel/git/rafael/linux-pm

Pull ACPI fixes from Rafael Wysocki:
 "These add one more ACPI IRQ override quirk, improve ACPI companion
  lookup for backlight devices and add missing kernel command line
  option values for backlight detection.

  Specifics:

   - Improve ACPI companion lookup for backlight devices in the cases
     when there is more than one candidate ACPI device object (Hans de
     Goede)

   - Add missing support for manual selection of NVidia-WMI-EC or Apple
     GMUX backlight in the kernel command line to the ACPI backlight
     driver (Hans de Goede)

   - Skip ACPI IRQ override on Asus Expertbook B2402CBA (Tamim Khan)"

* tag 'acpi-6.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  ACPI: Fix selecting wrong ACPI fwnode for the iGPU on some Dell laptops
  ACPI: video: Allow selecting NVidia-WMI-EC or Apple GMUX backlight from the cmdline
  ACPI: resource: Skip IRQ override on Asus Expertbook B2402CBA

23 months agoMerge tag 'platform-drivers-x86-v6.2-2' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 13 Jan 2023 13:26:40 +0000 (07:26 -0600)]
Merge tag 'platform-drivers-x86-v6.2-2' of git://git./linux/kernel/git/pdx86/platform-drivers-x86

Pull x86 platform driver fixes from Hans de Goede:
 "A set of assorted fixes and hardware-id additions"

* tag 'platform-drivers-x86-v6.2-2' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86:
  platform/x86: thinkpad_acpi: Fix profile mode display in AMT mode
  platform/x86: int3472/discrete: Ensure the clk/power enable pins are in output mode
  platform/x86/amd: Fix refcount leak in amd_pmc_probe
  platform/x86: intel/pmc/core: Add Meteor Lake mobile support
  platform/x86: simatic-ipc: add another model
  platform/x86: simatic-ipc: correct name of a model
  platform/x86: dell-privacy: Only register SW_CAMERA_LENS_COVER if present
  platform/x86: dell-privacy: Fix SW_CAMERA_LENS_COVER reporting
  platform/x86: asus-wmi: Don't load fan curves without fan
  platform/x86: asus-wmi: Ignore fan on E410MA
  platform/x86: asus-wmi: Add quirk wmi_ignore_fan
  platform/x86: asus-nb-wmi: Add alternate mapping for KEY_SCREENLOCK
  platform/x86: asus-nb-wmi: Add alternate mapping for KEY_CAMERA
  platform/surface: aggregator: Add missing call to ssam_request_sync_free()
  platform/surface: aggregator: Ignore command messages not intended for us
  platform/x86: touchscreen_dmi: Add info for the CSL Panther Tab HD
  platform/x86: ideapad-laptop: Add Legion 5 15ARH05 DMI id to set_fn_lock_led_list[]
  platform/x86: sony-laptop: Don't turn off 0x153 keyboard backlight during probe

23 months agoMerge tag 'drm-fixes-2023-01-13' of git://anongit.freedesktop.org/drm/drm
Linus Torvalds [Fri, 13 Jan 2023 13:18:59 +0000 (07:18 -0600)]
Merge tag 'drm-fixes-2023-01-13' of git://anongit.freedesktop.org/drm/drm

Pull drm fixes from Dave Airlie:
 "There is a bit of a post-holiday build up here I expect, small fixes
  across the board, amdgpu and msm being the main leaders, with others
  having a few. One code removal patch for nouveau:

  buddy:
   - benchmark regression fix for top-down buddy allocation

  panel:
   - add Lenovo panel orientation quirk

  ttm:
   - fix kernel oops regression

  amdgpu:
   - fix missing fence references
   - fix missing pipeline sync fencing
   - SMU13 fan speed fix
   - SMU13 fix power cap handling
   - SMU13 BACO fix
   - Fix a possible segfault in bo validation error case
   - Delay removal of firmware framebuffer
   - Fix error when unloading

  amdkfd:
   - SVM fix when clearing vram
   - GC11 fix for multi-GPU

  i915:
   - Reserve enough fence slot for i915_vma_unbind_vsync
   - Fix potential use after free
   - Reset engines twice in case of reset failure
   - Use multi-cast registers for SVG Unit registers

  msm:
   - display:
   - doc warning fixes
   - dt attribs cleanups
   - memory leak fix
   - error handing in hdmi probe fix
   - dp_aux_isr incorrect signalling fix
   - shutdown path fix
   - accel:
   - a5xx: fix quirks to be a bitmask
   - a6xx: fix gx halt to avoid 1s hang
   - kexec shutdown fix
   - fix potential double free

  vmwgfx:
   - drop rcu usage to make code more robust

  virtio:
   - fix use-after-free in gem handle code

  nouveau:
   - drop unused nouveau_fbcon.c"

* tag 'drm-fixes-2023-01-13' of git://anongit.freedesktop.org/drm/drm: (35 commits)
  drm: Optimize drm buddy top-down allocation method
  drm/ttm: Fix a regression causing kernel oops'es
  drm/i915/gt: Cover rest of SVG unit MCR registers
  drm/nouveau: Remove file nouveau_fbcon.c
  drm/amdkfd: Fix NULL pointer error for GC 11.0.1 on mGPU
  drm/amd/pm/smu13: BACO is supported when it's in BACO state
  drm/amdkfd: Add sync after creating vram bo
  drm/i915/gt: Reset twice
  drm/amdgpu: fix pipeline sync v2
  drm/vmwgfx: Remove rcu locks from user resources
  drm/virtio: Fix GEM handle creation UAF
  drm/amdgpu: Fixed bug on error when unloading amdgpu
  drm/amd: Delay removal of the firmware framebuffer
  drm/amdgpu: Fix potential NULL dereference
  drm/i915: Fix potential context UAFs
  drm/i915: Reserve enough fence slot for i915_vma_unbind_async
  drm: Add orientation quirk for Lenovo ideapad D330-10IGL
  drm/msm/a6xx: Avoid gx gbit halt during rpm suspend
  drm/msm/adreno: Make adreno quirks not overwrite each other
  drm/msm: another fix for the headless Adreno GPU
  ...

23 months agoALSA: pcm: Move rwsem lock inside snd_ctl_elem_read to prevent UAF
Clement Lecigne [Fri, 13 Jan 2023 12:07:45 +0000 (13:07 +0100)]
ALSA: pcm: Move rwsem lock inside snd_ctl_elem_read to prevent UAF

Takes rwsem lock inside snd_ctl_elem_read instead of snd_ctl_elem_read_user
like it was done for write in commit 1fa4445f9adf1 ("ALSA: control - introduce
snd_ctl_notify_one() helper"). Doing this way we are also fixing the following
locking issue happening in the compat path which can be easily triggered and
turned into an use-after-free.

64-bits:
snd_ctl_ioctl
  snd_ctl_elem_read_user
    [takes controls_rwsem]
    snd_ctl_elem_read [lock properly held, all good]
    [drops controls_rwsem]

32-bits:
snd_ctl_ioctl_compat
  snd_ctl_elem_write_read_compat
    ctl_elem_write_read
      snd_ctl_elem_read [missing lock, not good]

CVE-2023-0266 was assigned for this issue.

Cc: stable@kernel.org # 5.13+
Signed-off-by: Clement Lecigne <clecigne@google.com>
Reviewed-by: Jaroslav Kysela <perex@perex.cz>
Link: https://lore.kernel.org/r/20230113120745.25464-1-tiwai@suse.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
23 months agoMerge tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux
Linus Torvalds [Fri, 13 Jan 2023 13:11:45 +0000 (07:11 -0600)]
Merge tag 'arm64-fixes' of git://git./linux/kernel/git/arm64/linux

Pull arm64 fixes from Will Deacon:
 "Here's a sizeable batch of Friday the 13th arm64 fixes for -rc4. What
  could possibly go wrong?

  The obvious reason we have so much here is because of the holiday
  season right after the merge window, but we've also brought back an
  erratum workaround that was previously dropped at the last minute and
  there's an MTE coredumping fix that strays outside of the arch/arm64
  directory.

  Summary:

   - Fix PAGE_TABLE_CHECK failures on hugepage splitting path

   - Fix PSCI encoding of MEM_PROTECT_RANGE function in UAPI header

   - Fix NULL deref when accessing debugfs node if PSCI is not present

   - Fix MTE core dumping when VMA list is being updated concurrently

   - Fix SME signal frame handling when SVE is not implemented by the
     CPU

   - Fix asm constraints for cmpxchg_double() to hazard both words

   - Fix build failure with stack tracer and older versions of Clang

   - Bring back workaround for Cortex-A715 erratum 2645198"

* tag 'arm64-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm64/linux:
  arm64: Fix build with CC=clang, CONFIG_FTRACE=y and CONFIG_STACK_TRACER=y
  arm64/mm: Define dummy pud_user_exec() when using 2-level page-table
  arm64: errata: Workaround possible Cortex-A715 [ESR|FAR]_ELx corruption
  firmware/psci: Don't register with debugfs if PSCI isn't available
  firmware/psci: Fix MEM_PROTECT_RANGE function numbers
  arm64/signal: Always allocate SVE signal frames on SME only systems
  arm64/signal: Always accept SVE signal frames on SME only systems
  arm64/sme: Fix context switch for SME only systems
  arm64: cmpxchg_double*: hazard against entire exchange variable
  arm64/uprobes: change the uprobe_opcode_t typedef to fix the sparse warning
  arm64: mte: Avoid the racy walk of the vma list during core dump
  elfcore: Add a cprm parameter to elf_core_extra_{phdrs,data_size}
  arm64: mte: Fix double-freeing of the temporary tag storage during coredump
  arm64: ptrace: Use ARM64_SME to guard the SME register enumerations
  arm64/mm: add pud_user_exec() check in pud_user_accessible_page()
  arm64/mm: fix incorrect file_map_count for invalid pmd

23 months agoiommu/mediatek-v1: Fix an error handling path in mtk_iommu_v1_probe()
Christophe JAILLET [Mon, 19 Dec 2022 18:06:22 +0000 (19:06 +0100)]
iommu/mediatek-v1: Fix an error handling path in mtk_iommu_v1_probe()

A clk, prepared and enabled in mtk_iommu_v1_hw_init(), is not released in
the error handling path of mtk_iommu_v1_probe().

Add the corresponding clk_disable_unprepare(), as already done in the
remove function.

Fixes: b17336c55d89 ("iommu/mediatek: add support for mtk iommu generation one HW")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Yong Wu <yong.wu@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Reviewed-by: Matthias Brugger <matthias.bgg@gmail.com>
Link: https://lore.kernel.org/r/593e7b7d97c6e064b29716b091a9d4fd122241fb.1671473163.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Joerg Roedel <jroedel@suse.de>
23 months agoiommu/iova: Fix alloc iova overflows issue
Yunfei Wang [Wed, 11 Jan 2023 06:38:00 +0000 (14:38 +0800)]
iommu/iova: Fix alloc iova overflows issue

In __alloc_and_insert_iova_range, there is an issue that retry_pfn
overflows. The value of iovad->anchor.pfn_hi is ~0UL, then when
iovad->cached_node is iovad->anchor, curr_iova->pfn_hi + 1 will
overflow. As a result, if the retry logic is executed, low_pfn is
updated to 0, and then new_pfn < low_pfn returns false to make the
allocation successful.

This issue occurs in the following two situations:
1. The first iova size exceeds the domain size. When initializing
iova domain, iovad->cached_node is assigned as iovad->anchor. For
example, the iova domain size is 10M, start_pfn is 0x1_F000_0000,
and the iova size allocated for the first time is 11M. The
following is the log information, new->pfn_lo is smaller than
iovad->cached_node.

Example log as follows:
[  223.798112][T1705487] sh: [name:iova&]__alloc_and_insert_iova_range
start_pfn:0x1f0000,retry_pfn:0x0,size:0xb00,limit_pfn:0x1f0a00
[  223.799590][T1705487] sh: [name:iova&]__alloc_and_insert_iova_range
success start_pfn:0x1f0000,new->pfn_lo:0x1efe00,new->pfn_hi:0x1f08ff

2. The node with the largest iova->pfn_lo value in the iova domain
is deleted, iovad->cached_node will be updated to iovad->anchor,
and then the alloc iova size exceeds the maximum iova size that can
be allocated in the domain.

After judging that retry_pfn is less than limit_pfn, call retry_pfn+1
to fix the overflow issue.

Signed-off-by: jianjiao zeng <jianjiao.zeng@mediatek.com>
Signed-off-by: Yunfei Wang <yf.wang@mediatek.com>
Cc: <stable@vger.kernel.org> # 5.15.*
Fixes: 4e89dce72521 ("iommu/iova: Retry from last rb tree node if iova search fails")
Acked-by: Robin Murphy <robin.murphy@arm.com>
Link: https://lore.kernel.org/r/20230111063801.25107-1-yf.wang@mediatek.com
Signed-off-by: Joerg Roedel <jroedel@suse.de>
23 months agoiommu: Fix refcount leak in iommu_device_claim_dma_owner
Miaoqian Lin [Fri, 30 Dec 2022 08:31:00 +0000 (12:31 +0400)]
iommu: Fix refcount leak in iommu_device_claim_dma_owner

iommu_group_get() returns the group with the reference incremented.
Move iommu_group_get() after owner check to fix the refcount leak.

Fixes: 89395ccedbc1 ("iommu: Add device-centric DMA ownership interfaces")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Reviewed-by: Lu Baolu <baolu.lu@linux.intel.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/20221230083100.1489569-1-linmq006@gmail.com
[ joro: Remove *group = NULL initialization ]
Signed-off-by: Joerg Roedel <jroedel@suse.de>
23 months agoiommu/arm-smmu-v3: Don't unregister on shutdown
Vladimir Oltean [Thu, 15 Dec 2022 14:12:51 +0000 (16:12 +0200)]
iommu/arm-smmu-v3: Don't unregister on shutdown

Similar to SMMUv2, this driver calls iommu_device_unregister() from the
shutdown path, which removes the IOMMU groups with no coordination
whatsoever with their users - shutdown methods are optional in device
drivers. This can lead to NULL pointer dereferences in those drivers'
DMA API calls, or worse.

Instead of calling the full arm_smmu_device_remove() from
arm_smmu_device_shutdown(), let's pick only the relevant function call -
arm_smmu_device_disable() - more or less the reverse of
arm_smmu_device_reset() - and call just that from the shutdown path.

Fixes: 57365a04c921 ("iommu: Move bus setup to IOMMU device registration")
Suggested-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Link: https://lore.kernel.org/r/20221215141251.3688780-2-vladimir.oltean@nxp.com
Signed-off-by: Will Deacon <will@kernel.org>
23 months agoiommu/arm-smmu: Don't unregister on shutdown
Vladimir Oltean [Thu, 15 Dec 2022 14:12:50 +0000 (16:12 +0200)]
iommu/arm-smmu: Don't unregister on shutdown

Michael Walle says he noticed the following stack trace while performing
a shutdown with "reboot -f". He suggests he got "lucky" and just hit the
correct spot for the reboot while there was a packet transmission in
flight.

Unable to handle kernel NULL pointer dereference at virtual address 0000000000000098
CPU: 0 PID: 23 Comm: kworker/0:1 Not tainted 6.1.0-rc5-00088-gf3600ff8e322 #1930
Hardware name: Kontron KBox A-230-LS (DT)
pc : iommu_get_dma_domain+0x14/0x20
lr : iommu_dma_map_page+0x9c/0x254
Call trace:
 iommu_get_dma_domain+0x14/0x20
 dma_map_page_attrs+0x1ec/0x250
 enetc_start_xmit+0x14c/0x10b0
 enetc_xmit+0x60/0xdc
 dev_hard_start_xmit+0xb8/0x210
 sch_direct_xmit+0x11c/0x420
 __dev_queue_xmit+0x354/0xb20
 ip6_finish_output2+0x280/0x5b0
 __ip6_finish_output+0x15c/0x270
 ip6_output+0x78/0x15c
 NF_HOOK.constprop.0+0x50/0xd0
 mld_sendpack+0x1bc/0x320
 mld_ifc_work+0x1d8/0x4dc
 process_one_work+0x1e8/0x460
 worker_thread+0x178/0x534
 kthread+0xe0/0xe4
 ret_from_fork+0x10/0x20
Code: d503201f f9416800 d503233f d50323bf (f9404c00)
---[ end trace 0000000000000000 ]---
Kernel panic - not syncing: Oops: Fatal exception in interrupt

This appears to be reproducible when the board has a fixed IP address,
is ping flooded from another host, and "reboot -f" is used.

The following is one more manifestation of the issue:

$ reboot -f
kvm: exiting hardware virtualization
cfg80211: failed to load regulatory.db
arm-smmu 5000000.iommu: disabling translation
sdhci-esdhc 2140000.mmc: Removing from iommu group 11
sdhci-esdhc 2150000.mmc: Removing from iommu group 12
fsl-edma 22c0000.dma-controller: Removing from iommu group 17
dwc3 3100000.usb: Removing from iommu group 9
dwc3 3110000.usb: Removing from iommu group 10
ahci-qoriq 3200000.sata: Removing from iommu group 2
fsl-qdma 8380000.dma-controller: Removing from iommu group 20
platform f080000.display: Removing from iommu group 0
etnaviv-gpu f0c0000.gpu: Removing from iommu group 1
etnaviv etnaviv: Removing from iommu group 1
caam_jr 8010000.jr: Removing from iommu group 13
caam_jr 8020000.jr: Removing from iommu group 14
caam_jr 8030000.jr: Removing from iommu group 15
caam_jr 8040000.jr: Removing from iommu group 16
fsl_enetc 0000:00:00.0: Removing from iommu group 4
arm-smmu 5000000.iommu: Blocked unknown Stream ID 0x429; boot with "arm-smmu.disable_bypass=0" to allow, but this may have security implications
arm-smmu 5000000.iommu:         GFSR 0x80000002, GFSYNR0 0x00000002, GFSYNR1 0x00000429, GFSYNR2 0x00000000
fsl_enetc 0000:00:00.1: Removing from iommu group 5
arm-smmu 5000000.iommu: Blocked unknown Stream ID 0x429; boot with "arm-smmu.disable_bypass=0" to allow, but this may have security implications
arm-smmu 5000000.iommu:         GFSR 0x80000002, GFSYNR0 0x00000002, GFSYNR1 0x00000429, GFSYNR2 0x00000000
arm-smmu 5000000.iommu: Blocked unknown Stream ID 0x429; boot with "arm-smmu.disable_bypass=0" to allow, but this may have security implications
arm-smmu 5000000.iommu:         GFSR 0x80000002, GFSYNR0 0x00000000, GFSYNR1 0x00000429, GFSYNR2 0x00000000
fsl_enetc 0000:00:00.2: Removing from iommu group 6
fsl_enetc_mdio 0000:00:00.3: Removing from iommu group 8
mscc_felix 0000:00:00.5: Removing from iommu group 3
fsl_enetc 0000:00:00.6: Removing from iommu group 7
pcieport 0001:00:00.0: Removing from iommu group 18
arm-smmu 5000000.iommu: Blocked unknown Stream ID 0x429; boot with "arm-smmu.disable_bypass=0" to allow, but this may have security implications
arm-smmu 5000000.iommu:         GFSR 0x00000002, GFSYNR0 0x00000000, GFSYNR1 0x00000429, GFSYNR2 0x00000000
pcieport 0002:00:00.0: Removing from iommu group 19
Unable to handle kernel NULL pointer dereference at virtual address 00000000000000a8
pc : iommu_get_dma_domain+0x14/0x20
lr : iommu_dma_unmap_page+0x38/0xe0
Call trace:
 iommu_get_dma_domain+0x14/0x20
 dma_unmap_page_attrs+0x38/0x1d0
 enetc_unmap_tx_buff.isra.0+0x6c/0x80
 enetc_poll+0x170/0x910
 __napi_poll+0x40/0x1e0
 net_rx_action+0x164/0x37c
 __do_softirq+0x128/0x368
 run_ksoftirqd+0x68/0x90
 smpboot_thread_fn+0x14c/0x190
Code: d503201f f9416800 d503233f d50323bf (f9405400)
---[ end trace 0000000000000000 ]---
Kernel panic - not syncing: Oops: Fatal exception in interrupt
---[ end Kernel panic - not syncing: Oops: Fatal exception in interrupt ]---

The problem seems to be that iommu_group_remove_device() is allowed to
run with no coordination whatsoever with the shutdown procedure of the
enetc PCI device. In fact, it almost seems as if it implies that the
pci_driver :: shutdown() method is mandatory if DMA is used with an
IOMMU, otherwise this is inevitable. That was never the case; shutdown
methods are optional in device drivers.

This is the call stack that leads to iommu_group_remove_device() during
reboot:

kernel_restart
-> device_shutdown
   -> platform_shutdown
      -> arm_smmu_device_shutdown
         -> arm_smmu_device_remove
            -> iommu_device_unregister
               -> bus_for_each_dev
                  -> remove_iommu_group
                     -> iommu_release_device
                        -> iommu_group_remove_device

I don't know much about the arm_smmu driver, but
arm_smmu_device_shutdown() invoking arm_smmu_device_remove() looks
suspicious, since it causes the IOMMU device to unregister and that's
where everything starts to unravel. It forces all other devices which
depend on IOMMU groups to also point their ->shutdown() to ->remove(),
which will make reboot slower overall.

There are 2 moments relevant to this behavior. First was commit
b06c076ea962 ("Revert "iommu/arm-smmu: Make arm-smmu explicitly
non-modular"") when arm_smmu_device_shutdown() was made to run the exact
same thing as arm_smmu_device_remove(). Prior to that, there was no
iommu_device_unregister() call in arm_smmu_device_shutdown(). However,
that was benign until commit 57365a04c921 ("iommu: Move bus setup to
IOMMU device registration"), which made iommu_device_unregister() call
remove_iommu_group().

Restore the old shutdown behavior by making remove() call shutdown(),
but shutdown() does not call the remove() specific bits.

Fixes: 57365a04c921 ("iommu: Move bus setup to IOMMU device registration")
Reported-by: Michael Walle <michael@walle.cc>
Tested-by: Michael Walle <michael@walle.cc> # on kontron-sl28
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Link: https://lore.kernel.org/r/20221215141251.3688780-1-vladimir.oltean@nxp.com
Signed-off-by: Will Deacon <will@kernel.org>
23 months agoiommu/arm-smmu: Report IOMMU_CAP_CACHE_COHERENCY even betterer
Robin Murphy [Thu, 15 Dec 2022 16:51:55 +0000 (16:51 +0000)]
iommu/arm-smmu: Report IOMMU_CAP_CACHE_COHERENCY even betterer

Although it's vanishingly unlikely that anyone would integrate an SMMU
within a coherent interconnect without also making the pagetable walk
interface coherent, the same effect happens if a coherent SMMU fails to
advertise CTTW correctly. This turns out to be the case on some popular
NXP SoCs, where VFIO started failing the IOMMU_CAP_CACHE_COHERENCY test,
even though IOMMU_CACHE *was* previously achieving the desired effect
anyway thanks to the underlying integration.

While those SoCs stand to gain some more general benefits from a
firmware update to override CTTW correctly in DT/ACPI, it's also easy
to work around this in Linux as well, to avoid imposing too much on
affected users - since the upstream client devices *are* correctly
marked as coherent, we can trivially infer their coherent paths through
the SMMU as well.

Reported-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Fixes: df198b37e72c ("iommu/arm-smmu: Report IOMMU_CAP_CACHE_COHERENCY better")
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Tested-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Link: https://lore.kernel.org/r/d6dc41952961e5c7b21acac08a8bf1eb0f69e124.1671123115.git.robin.murphy@arm.com
Signed-off-by: Will Deacon <will@kernel.org>
23 months agoplatform/x86: thinkpad_acpi: Fix profile mode display in AMT mode
Mark Pearson [Thu, 12 Jan 2023 22:12:28 +0000 (17:12 -0500)]
platform/x86: thinkpad_acpi: Fix profile mode display in AMT mode

Recently AMT mode was enabled (somewhat unexpectedly) on the Lenovo
Z13 platform. The FW is advertising it is available and the driver tries
to use it - unfortunately it reports the profile mode incorrectly.

Note, there is also some extra work needed to enable the dynamic aspect
of AMT support that I will be following up with; but more testing is
needed first. This patch just fixes things so the profiles are reported
correctly.

Link: https://gitlab.freedesktop.org/hadess/power-profiles-daemon/-/issues/115
Fixes: 46dcbc61b739 ("platform/x86: thinkpad-acpi: Add support for automatic mode transitions")
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Mark Pearson <mpearson-lenovo@squebb.ca>
Link: https://lore.kernel.org/r/20230112221228.490946-1-mpearson-lenovo@squebb.ca
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
23 months agoMerge branches 'acpi-resource' and 'acpi-video'
Rafael J. Wysocki [Fri, 13 Jan 2023 10:11:05 +0000 (11:11 +0100)]
Merge branches 'acpi-resource' and 'acpi-video'

Merge an ACPI resource management quirk and an ACPI backlight driver fix
for 6.2-rc4:

 - Skip ACPI  IRQ override on Asus Expertbook B2402CBA (Tamim Khan).

 - Add missing support for manual selection of NVidia-WMI-EC or Apple
   GMUX backlight in the kernel command line to the ACPI backlight
   driver (Hans de Goede).

* acpi-resource:
  ACPI: resource: Skip IRQ override on Asus Expertbook B2402CBA

* acpi-video:
  ACPI: video: Allow selecting NVidia-WMI-EC or Apple GMUX backlight from the cmdline

23 months agoALSA: usb-audio: Fix possible NULL pointer dereference in snd_usb_pcm_has_fixed_rate()
Jaroslav Kysela [Fri, 13 Jan 2023 08:53:11 +0000 (09:53 +0100)]
ALSA: usb-audio: Fix possible NULL pointer dereference in snd_usb_pcm_has_fixed_rate()

The subs function argument may be NULL, so do not use it before the NULL check.

Fixes: 291e9da91403 ("ALSA: usb-audio: Always initialize fixed_rate in snd_usb_find_implicit_fb_sync_format()")
Reported-by: coverity-bot <keescook@chromium.org>
Link: https://lore.kernel.org/alsa-devel/202301121424.4A79A485@keescook/
Signed-off-by: Jaroslav Kysela <perex@perex.cz>
Link: https://lore.kernel.org/r/20230113085311.623325-1-perex@perex.cz
Signed-off-by: Takashi Iwai <tiwai@suse.de>
23 months agoMerge tag 'drm-msm-fixes-2023-01-12' of https://gitlab.freedesktop.org/drm/msm into...
Dave Airlie [Fri, 13 Jan 2023 03:01:22 +0000 (13:01 +1000)]
Merge tag 'drm-msm-fixes-2023-01-12' of https://gitlab.freedesktop.org/drm/msm into drm-fixes

msm-fixes for v6.3-rc4

Display Fixes:

- Fix the documentation for dpu_encoder_phys_wb_init() and
  dpu_encoder_phys_wb_setup_fb() APIs to address doc warnings
- Remove vcca-supply and vdds-supply as mandatory for 14nm PHY and
  10nm PHY DT schemas respectively as they are not present on some
  SOCs using these PHYs
- Add the dsi-phy-regulator-ldo-mode to dsi-phy-28nm.yaml as it was
  missed out during txt to yaml migration
- Remove operating-points-v2 and power-domain as a required property
  for the DSI controller as thats not the case for every SOC
- Fix the description from display escape clock to display core
  clock in the dsi controller yaml
- Fix the memory leak for mdp1-mem path for the cases when we return
  early after failing to get mdp0-mem ICC paths for msm
- Fix error handling path in msm_hdmi_dev_probe() to release the phy
  ref count when devm_pm_runtime_enable() fails
- Fix the dp_aux_isr() routine to make sure it doesnt incorrectly
  signal the aux transaction as complete if the ISR was not an AUX
  isr. This fixes a big hitter stability bug on chromebooks.
- Add protection against null pointer dereference when there is no
  kms object as in the case of headless adreno GPU in the shutdown
  path.

GPU Fixes:

- a5xx: fix quirks to actually be a bitmask and not overwrite each
  other
- a6xx: fix gx halt sequence to avoid 1000ms hang on some devices
- kexec shutdown fix
- fix potential double free

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Rob Clark <robdclark@gmail.com>
Link: https://patchwork.freedesktop.org/patch/msgid/CAF6AEGv7=in_MHW3kdkhqh7ZFoVCmnikmr29YYHCXR=7aOEneg@mail.gmail.com
23 months agoMerge tag 'drm-intel-fixes-2023-01-12' of git://anongit.freedesktop.org/drm/drm-intel...
Dave Airlie [Fri, 13 Jan 2023 02:44:21 +0000 (12:44 +1000)]
Merge tag 'drm-intel-fixes-2023-01-12' of git://anongit.freedesktop.org/drm/drm-intel into drm-fixes

- Reserve enough fence slot for i915_vma_unbind_vsync (Nirmoy)
- Fix potential use after free (Rob Clark)
- Reset engines twice in case of reset failure (Chris)
- Use multi-cast registers for SVG Unit registers (Gustavo)

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/Y8AbHelGeXc5eQ8U@intel.com
23 months agoMerge tag 'amd-drm-fixes-6.2-2023-01-11' of https://gitlab.freedesktop.org/agd5f...
Dave Airlie [Fri, 13 Jan 2023 02:37:24 +0000 (12:37 +1000)]
Merge tag 'amd-drm-fixes-6.2-2023-01-11' of https://gitlab.freedesktop.org/agd5f/linux into drm-fixes

amd-drm-fixes-6.2-2023-01-11:

amdgpu:
- SMU13 fan speed fix
- SMU13 fix power cap handling
- SMU13 BACO fix
- Fix a possible segfault in bo validation error case
- Delay removal of firmware framebuffer
- Fix error when unloading

amdkfd:
- SVM fix when clearing vram
- GC11 fix for multi-GPU

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20230112033004.8184-1-alexander.deucher@amd.com
23 months agoMerge tag 'drm-misc-fixes-2023-01-12' of git://anongit.freedesktop.org/drm/drm-misc...
Dave Airlie [Fri, 13 Jan 2023 01:44:54 +0000 (11:44 +1000)]
Merge tag 'drm-misc-fixes-2023-01-12' of git://anongit.freedesktop.org/drm/drm-misc into drm-fixes

Several fixes for amdgpu (all addressing issues with fences), yet
another orientation quirk for a Lenovo device, a use-after-free fix for
virtio, a regression fix in TTM and a performance regression in drm
buddy.

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Maxime Ripard <maxime@cerno.tech>
Link: https://patchwork.freedesktop.org/patch/msgid/20230112130954.pxt77g3a7rokha42@houat
23 months agoMerge tag 'net-6.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net
Linus Torvalds [Fri, 13 Jan 2023 00:20:44 +0000 (18:20 -0600)]
Merge tag 'net-6.2-rc4' of git://git./linux/kernel/git/netdev/net

Pull networking fixes from Paolo Abeni:
 "Including fixes from rxrpc.

  The rxrpc changes are noticeable large: to address a recent regression
  has been necessary completing the threaded refactor.

  Current release - regressions:

   - rxrpc:
       - only disconnect calls in the I/O thread
       - move client call connection to the I/O thread
       - fix incoming call setup race

   - eth: mlx5:
       - restore pkt rate policing support
       - fix memory leak on updating vport counters

  Previous releases - regressions:

   - gro: take care of DODGY packets

   - ipv6: deduct extension header length in rawv6_push_pending_frames

   - tipc: fix unexpected link reset due to discovery messages

  Previous releases - always broken:

   - sched: disallow noqueue for qdisc classes

   - eth: ice: fix potential memory leak in ice_gnss_tty_write()

   - eth: ixgbe: fix pci device refcount leak

   - eth: mlx5:
       - fix command stats access after free
       - fix macsec possible null dereference when updating MAC security
         entity (SecY)"

* tag 'net-6.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (64 commits)
  r8152: add vendor/device ID pair for Microsoft Devkit
  net: stmmac: add aux timestamps fifo clearance wait
  bnxt: make sure we return pages to the pool
  net: hns3: fix wrong use of rss size during VF rss config
  ipv6: raw: Deduct extension header length in rawv6_push_pending_frames
  net: lan966x: check for ptp to be enabled in lan966x_ptp_deinit()
  net: sched: disallow noqueue for qdisc classes
  iavf/iavf_main: actually log ->src mask when talking about it
  igc: Fix PPS delta between two synchronized end-points
  ixgbe: fix pci device refcount leak
  octeontx2-pf: Fix resource leakage in VF driver unbind
  selftests/net: l2_tos_ttl_inherit.sh: Ensure environment cleanup on failure.
  selftests/net: l2_tos_ttl_inherit.sh: Run tests in their own netns.
  selftests/net: l2_tos_ttl_inherit.sh: Set IPv6 addresses with "nodad".
  net/mlx5e: Fix macsec possible null dereference when updating MAC security entity (SecY)
  net/mlx5e: Fix macsec ssci attribute handling in offload path
  net/mlx5: E-switch, Coverity: overlapping copy
  net/mlx5e: Don't support encap rules with gbp option
  net/mlx5: Fix ptp max frequency adjustment range
  net/mlx5e: Fix memory leak on updating vport counters
  ...

23 months agoMerge tag 's390-6.2-2' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux
Linus Torvalds [Thu, 12 Jan 2023 23:09:20 +0000 (17:09 -0600)]
Merge tag 's390-6.2-2' of git://git./linux/kernel/git/s390/linux

Pull s390 fixes from Heiko Carstens:

 - Add various missing READ_ONCE() to cmpxchg() loops prevent the
   compiler from potentially generating incorrect code. This includes a
   rather large change to the s390 specific hardware sampling code and
   its current use of cmpxchg_double().

   Do the fix now to get it out of the way of Peter Zijlstra's
   cmpxchg128() work, and have something that can be backported. The
   added new code includes a private 128 bit cmpxchg variant which will
   be removed again after Peter's rework is available. Also note that
   this 128 bit cmpxchg variant is used to implement 128 bit
   READ_ONCE(), while strictly speaking it wouldn't be necessary, and
   _READ_ONCE() should also be sufficient; even though it isn't obvious
   for all converted locations that this is the case. Therefore use this
   implementation for for the sake of clarity and consistency for now.

 - Fix ipl report address handling to avoid kdump failures/hangs.

 - Fix misuse of #(el)if in kernel decompressor.

 - Define RUNTIME_DISCARD_EXIT to fix link error with GNU ld < 2.36,
   caused by the recently changed discard behaviour.

 - Make sure _edata and _end symbols are always page aligned.

 - The current header guard DEBUG_H in one of the s390 specific header
   files is too generic and conflicts with the ath9k wireless driver.
   Add an _ASM_S390_ prefix to the guard to make it unique.

 - Update defconfigs.

* tag 's390-6.2-2' of git://git.kernel.org/pub/scm/linux/kernel/git/s390/linux:
  s390: update defconfigs
  KVM: s390: interrupt: use READ_ONCE() before cmpxchg()
  s390/percpu: add READ_ONCE() to arch_this_cpu_to_op_simple()
  s390/cpum_sf: add READ_ONCE() semantics to compare and swap loops
  s390/kexec: fix ipl report address for kdump
  s390: fix -Wundef warning for CONFIG_KERNEL_ZSTD
  s390: define RUNTIME_DISCARD_EXIT to fix link error with GNU ld < 2.36
  s390: expicitly align _edata and _end symbols on page boundary
  s390/debug: add _ASM_S390_ prefix to header guard

23 months agoMerge tag 'for-linus-6.2-rc4-tag' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Thu, 12 Jan 2023 23:02:20 +0000 (17:02 -0600)]
Merge tag 'for-linus-6.2-rc4-tag' of git://git./linux/kernel/git/xen/tip

Pull xen fixes from Juergen Gross:

 - two cleanup patches

 - a fix of a memory leak in the Xen pvfront driver

 - a fix of a locking issue in the Xen hypervisor console driver

* tag 'for-linus-6.2-rc4-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip:
  xen/pvcalls: free active map buffer on pvcalls_front_free_map
  hvc/xen: lock console list traversal
  x86/xen: Remove the unused function p2m_index()
  xen: make remove callback of xen driver void returned

23 months agoMerge tag 'timers-urgent-2023-01-12' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Thu, 12 Jan 2023 22:53:39 +0000 (16:53 -0600)]
Merge tag 'timers-urgent-2023-01-12' of git://git./linux/kernel/git/tip/tip

Pull timer doc fixes from Ingo Molnar:

 - Fix various DocBook formatting errors in kernel/time/ that generated
   (justified) warnings during a kernel-doc build.

* tag 'timers-urgent-2023-01-12' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  time: Fix various kernel-doc problems

23 months agoMerge tag 'perf-urgent-2023-01-12' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Thu, 12 Jan 2023 22:47:32 +0000 (16:47 -0600)]
Merge tag 'perf-urgent-2023-01-12' of git://git./linux/kernel/git/tip/tip

Pull perf events hw enablement from Ingo Molnar:

 - More hardware-enablement for Intel Meteor Lake & Emerald Rapid
   systems: pure model ID enumeration additions that do not affect other
   systems.

* tag 'perf-urgent-2023-01-12' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  perf/x86/intel/uncore: Add Emerald Rapids
  perf/x86/msr: Add Emerald Rapids
  perf/x86/msr: Add Meteor Lake support
  perf/x86/cstate: Add Meteor Lake support

23 months agoMerge tag 'sched-urgent-2023-01-12' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Thu, 12 Jan 2023 22:39:43 +0000 (16:39 -0600)]
Merge tag 'sched-urgent-2023-01-12' of git://git./linux/kernel/git/tip/tip

Pull scheduler fixes from Ingo Molnar:

 - Fix scheduler frequency invariance bug related to overly long
   tickless periods triggering an integer overflow and disabling the
   feature.

 - Fix use-after-free bug in dup_user_cpus_ptr().

 - Fix do_set_cpus_allowed() deadlock scenarios related to calling
   kfree() with the pi_lock held. NOTE: the rcu_free() is the 'lazy'
   solution here - we looked at patches to free the structure after the
   pi_lock got dropped, but that looked quite a bit messier - and none
   of this is truly performance critical. We can revisit this if it's
   too lazy of a solution ...

* tag 'sched-urgent-2023-01-12' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  sched/core: Use kfree_rcu() in do_set_cpus_allowed()
  sched/core: Fix use-after-free bug in dup_user_cpus_ptr()
  sched/core: Fix arch_scale_freq_tick() on tickless systems

23 months agoMerge tag 'core-urgent-2023-01-12' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Thu, 12 Jan 2023 20:41:32 +0000 (14:41 -0600)]
Merge tag 'core-urgent-2023-01-12' of git://git./linux/kernel/git/tip/tip

Pull objtool fix from Ingo Molnar:

 - Fix objtool to be more permissive with hand-written assembly that
   uses non-function symbols in executable sections.

* tag 'core-urgent-2023-01-12' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  objtool: Tolerate STT_NOTYPE symbols at end of section

23 months agoMerge tag 'urgent-nolibc.2023.01.09a' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Thu, 12 Jan 2023 20:38:22 +0000 (14:38 -0600)]
Merge tag 'urgent-nolibc.2023.01.09a' of git://git./linux/kernel/git/paulmck/linux-rcu

Pull nolibc fixes from Paul McKenney:

 - The fd_set structure was incorrectly defined as arrays of u32 instead
   of long, which breaks BE64. Fix courtesy of Sven Schnelle.

 - S_ISxxx macros were incorrectly testing the bits after applying them
   instead of bitwise ANDing S_FMT with the value. Fix from Warner Losh.

 - The mips code was randomly broken due to an unprotected "noreorder"
   directive in the _start code that could prevent the assembler from
   filling delayed slots. This in turn resulted in random other
   instructions being placed into those slots. Fix courtesy of Willy
   Tarreau.

 - The current nolibc header layout refrains from including files that
   are not explicitly included by the code using nolibc. Unfortunately,
   this causes build failures when such files contain definitions that
   are used (for example) by libgcc. Example definitions include raise()
   and memset(), which are called by some architectures, but only at
   certain optimization levels. Fix courtesy of Willy Tarreau.

 - gcc 11.3 in ARM thumb2 mode at -O2 recognized a memset() construction
   inside the memset() definition. The compiler replaced this
   construction with a call to... memset(). Userland cannot be forced to
   build with -ffreestanding, so an empty asm() statement was introduced
   into the loop the loop in order to prevent the compiler from making
   this unproductive transformation. Fix courtesy of Willy Tarreau.

 - Most of the O_* macros were wrong on RISCV because their octal values
   were coded as hexadecimal. This resulted in the getdents64() selftest
   failing. Fix courtesy of Willy Tarreau.

This was tested on x86_64, i386, armv5, armv7, thumb1, thumb2, mips and
riscv, all at -O0, -Os and -O3.

* tag 'urgent-nolibc.2023.01.09a' of git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu:
  tools/nolibc: fix the O_* fcntl/open macro definitions for riscv
  tools/nolibc: prevent gcc from making memset() loop over itself
  tools/nolibc: fix missing includes causing build issues at -O0
  tools/nolibc: restore mips branch ordering in the _start block
  tools/nolibc: Fix S_ISxxx macros
  nolibc: fix fd_set type

23 months agoplatform/x86: int3472/discrete: Ensure the clk/power enable pins are in output mode
Hans de Goede [Wed, 11 Jan 2023 20:14:26 +0000 (21:14 +0100)]
platform/x86: int3472/discrete: Ensure the clk/power enable pins are in output mode

acpi_get_and_request_gpiod() does not take a gpio_lookup_flags argument
specifying that the pins direction should be initialized to a specific
value.

This means that in some cases the pins might be left in input mode, causing
the gpiod_set() calls made to enable the clk / regulator to not work.

One example of this problem is the clk-enable GPIO for the ov01a1s sensor
on a Dell Latitude 9420 being left in input mode causing the clk to
never get enabled.

Explicitly set the direction of the pins to output to fix this.

Fixes: 5de691bffe57 ("platform/x86: Add intel_skl_int3472 driver")
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Andy Shevchenko <andy@kernel.org>
Reviewed-by: Daniel Scally <djrscally@gmail.com>
Reviewed-by: Sakari Ailus <sakari.ailus@linux.intel.com>
Link: https://lore.kernel.org/r/20230111201426.947853-1-hdegoede@redhat.com
23 months agoplatform/x86/amd: Fix refcount leak in amd_pmc_probe
Miaoqian Lin [Thu, 29 Dec 2022 07:25:33 +0000 (11:25 +0400)]
platform/x86/amd: Fix refcount leak in amd_pmc_probe

pci_get_domain_bus_and_slot() takes reference, the caller should release
the reference by calling pci_dev_put() after use. Call pci_dev_put() in
the error path to fix this.

Fixes: 3d7d407dfb05 ("platform/x86: amd-pmc: Add support for AMD Spill to DRAM STB feature")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Link: https://lore.kernel.org/r/20221229072534.1381432-1-linmq006@gmail.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
23 months agoplatform/x86: intel/pmc/core: Add Meteor Lake mobile support
Gayatri Kammela [Wed, 28 Dec 2022 23:05:53 +0000 (15:05 -0800)]
platform/x86: intel/pmc/core: Add Meteor Lake mobile support

Add Meteor Lake mobile support to pmc core driver. Meteor Lake mobile
parts reuse all the Meteor Lake PCH IPs.

Cc: David E Box <david.e.box@linux.intel.com>
Signed-off-by: Gayatri Kammela <gayatri.kammela@linux.intel.com>
Link: https://lore.kernel.org/r/20221228230553.2497183-1-gayatri.kammela@linux.intel.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
23 months agoplatform/x86: simatic-ipc: add another model
Henning Schild [Thu, 22 Dec 2022 10:37:20 +0000 (11:37 +0100)]
platform/x86: simatic-ipc: add another model

Add IPC PX-39A support.

Signed-off-by: Henning Schild <henning.schild@siemens.com>
Link: https://lore.kernel.org/r/20221222103720.8546-3-henning.schild@siemens.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
23 months agoplatform/x86: simatic-ipc: correct name of a model
Henning Schild [Thu, 22 Dec 2022 10:37:19 +0000 (11:37 +0100)]
platform/x86: simatic-ipc: correct name of a model

What we called IPC427G should be renamed to BX-39A to be more in line
with the actual product name.

Signed-off-by: Henning Schild <henning.schild@siemens.com>
Link: https://lore.kernel.org/r/20221222103720.8546-2-henning.schild@siemens.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>