From: Junxiao Bi ocfs2-devel@oss.oracle.com

commit c80af0c250c8f8a3c978aa5aafbe9c39b336b813 upstream.

This reverts commit 912f655d78c5d4ad05eac287f23a435924df7144.

This commit introduced a regression that can cause mount hung. The changes in __ocfs2_find_empty_slot causes that any node with none-zero node number can grab the slot that was already taken by node 0, so node 1 will access the same journal with node 0, when it try to grab journal cluster lock, it will hung because it was already acquired by node 0. It's very easy to reproduce this, in one cluster, mount node 0 first, then node 1, you will see the following call trace from node 1.

[13148.735424] INFO: task mount.ocfs2:53045 blocked for more than 122 seconds. [13148.739691] Not tainted 5.15.0-2148.0.4.el8uek.mountracev2.x86_64 #2 [13148.742560] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [13148.745846] task:mount.ocfs2 state:D stack: 0 pid:53045 ppid: 53044 flags:0x00004000 [13148.749354] Call Trace: [13148.750718] <TASK> [13148.752019] ? usleep_range+0x90/0x89 [13148.753882] __schedule+0x210/0x567 [13148.755684] schedule+0x44/0xa8 [13148.757270] schedule_timeout+0x106/0x13c [13148.759273] ? __prepare_to_swait+0x53/0x78 [13148.761218] __wait_for_common+0xae/0x163 [13148.763144] __ocfs2_cluster_lock.constprop.0+0x1d6/0x870 [ocfs2] [13148.765780] ? ocfs2_inode_lock_full_nested+0x18d/0x398 [ocfs2] [13148.768312] ocfs2_inode_lock_full_nested+0x18d/0x398 [ocfs2] [13148.770968] ocfs2_journal_init+0x91/0x340 [ocfs2] [13148.773202] ocfs2_check_volume+0x39/0x461 [ocfs2] [13148.775401] ? iput+0x69/0xba [13148.777047] ocfs2_mount_volume.isra.0.cold+0x40/0x1f5 [ocfs2] [13148.779646] ocfs2_fill_super+0x54b/0x853 [ocfs2] [13148.781756] mount_bdev+0x190/0x1b7 [13148.783443] ? ocfs2_remount+0x440/0x440 [ocfs2] [13148.785634] legacy_get_tree+0x27/0x48 [13148.787466] vfs_get_tree+0x25/0xd0 [13148.789270] do_new_mount+0x18c/0x2d9 [13148.791046] __x64_sys_mount+0x10e/0x142 [13148.792911] do_syscall_64+0x3b/0x89 [13148.794667] entry_SYSCALL_64_after_hwframe+0x170/0x0 [13148.797051] RIP: 0033:0x7f2309f6e26e [13148.798784] RSP: 002b:00007ffdcee7d408 EFLAGS: 00000246 ORIG_RAX: 00000000000000a5 [13148.801974] RAX: ffffffffffffffda RBX: 00007ffdcee7d4a0 RCX: 00007f2309f6e26e [13148.804815] RDX: 0000559aa762a8ae RSI: 0000559aa939d340 RDI: 0000559aa93a22b0 [13148.807719] RBP: 00007ffdcee7d5b0 R08: 0000559aa93a2290 R09: 00007f230a0b4820 [13148.810659] R10: 0000000000000000 R11: 0000000000000246 R12: 00007ffdcee7d420 [13148.813609] R13: 0000000000000000 R14: 0000559aa939f000 R15: 0000000000000000 [13148.816564] </TASK>

To fix it, we can just fix __ocfs2_find_empty_slot. But original commit introduced the feature to mount ocfs2 locally even it is cluster based, that is a very dangerous, it can easily cause serious data corruption, there is no way to stop other nodes mounting the fs and corrupting it. Setup ha or other cluster-aware stack is just the cost that we have to take for avoiding corruption, otherwise we have to do it in kernel.

Link: https://lkml.kernel.org/r/20220603222801.42488-1-junxiao.bi@oracle.com Fixes: 912f655d78c5("ocfs2: mount shared volume without ha stack") Signed-off-by: Junxiao Bi junxiao.bi@oracle.com Acked-by: Joseph Qi joseph.qi@linux.alibaba.com Cc: Mark Fasheh mark@fasheh.com Cc: Joel Becker jlbec@evilplan.org Cc: Changwei Ge gechangwei@live.cn Cc: Gang He ghe@suse.com Cc: Jun Piao piaojun@huawei.com Cc: heming.zhao@suse.com Cc: stable@vger.kernel.org Signed-off-by: Andrew Morton akpm@linux-foundation.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- fs/ocfs2/ocfs2.h | 4 +--- fs/ocfs2/slot_map.c | 46 +++++++++++++++++++--------------------------- fs/ocfs2/super.c | 21 --------------------- 3 files changed, 20 insertions(+), 51 deletions(-)

--- a/fs/ocfs2/ocfs2.h +++ b/fs/ocfs2/ocfs2.h @@ -277,7 +277,6 @@ enum ocfs2_mount_options OCFS2_MOUNT_JOURNAL_ASYNC_COMMIT = 1 << 15, /* Journal Async Commit */ OCFS2_MOUNT_ERRORS_CONT = 1 << 16, /* Return EIO to the calling process on error */ OCFS2_MOUNT_ERRORS_ROFS = 1 << 17, /* Change filesystem to read-only on error */ - OCFS2_MOUNT_NOCLUSTER = 1 << 18, /* No cluster aware filesystem mount */ };

#define OCFS2_OSB_SOFT_RO 0x0001 @@ -673,8 +672,7 @@ static inline int ocfs2_cluster_o2cb_glo

static inline int ocfs2_mount_local(struct ocfs2_super *osb) { - return ((osb->s_feature_incompat & OCFS2_FEATURE_INCOMPAT_LOCAL_MOUNT) - || (osb->s_mount_opt & OCFS2_MOUNT_NOCLUSTER)); + return (osb->s_feature_incompat & OCFS2_FEATURE_INCOMPAT_LOCAL_MOUNT); }

static inline int ocfs2_uses_extended_slot_map(struct ocfs2_super *osb) --- a/fs/ocfs2/slot_map.c +++ b/fs/ocfs2/slot_map.c @@ -252,16 +252,14 @@ static int __ocfs2_find_empty_slot(struc int i, ret = -ENOSPC;

if ((preferred >= 0) && (preferred < si->si_num_slots)) { - if (!si->si_slots[preferred].sl_valid || - !si->si_slots[preferred].sl_node_num) { + if (!si->si_slots[preferred].sl_valid) { ret = preferred; goto out; } }

for(i = 0; i < si->si_num_slots; i++) { - if (!si->si_slots[i].sl_valid || - !si->si_slots[i].sl_node_num) { + if (!si->si_slots[i].sl_valid) { ret = i; break; } @@ -456,30 +454,24 @@ int ocfs2_find_slot(struct ocfs2_super * spin_lock(&osb->osb_lock); ocfs2_update_slot_info(si);

- if (ocfs2_mount_local(osb)) - /* use slot 0 directly in local mode */ - slot = 0; - else { - /* search for ourselves first and take the slot if it already - * exists. Perhaps we need to mark this in a variable for our - * own journal recovery? Possibly not, though we certainly - * need to warn to the user */ - slot = __ocfs2_node_num_to_slot(si, osb->node_num); + /* search for ourselves first and take the slot if it already + * exists. Perhaps we need to mark this in a variable for our + * own journal recovery? Possibly not, though we certainly + * need to warn to the user */ + slot = __ocfs2_node_num_to_slot(si, osb->node_num); + if (slot < 0) { + /* if no slot yet, then just take 1st available + * one. */ + slot = __ocfs2_find_empty_slot(si, osb->preferred_slot); if (slot < 0) { - /* if no slot yet, then just take 1st available - * one. */ - slot = __ocfs2_find_empty_slot(si, osb->preferred_slot); - if (slot < 0) { - spin_unlock(&osb->osb_lock); - mlog(ML_ERROR, "no free slots available!\n"); - status = -EINVAL; - goto bail; - } - } else - printk(KERN_INFO "ocfs2: Slot %d on device (%s) was " - "already allocated to this node!\n", - slot, osb->dev_str); - } + spin_unlock(&osb->osb_lock); + mlog(ML_ERROR, "no free slots available!\n"); + status = -EINVAL; + goto bail; + } + } else + printk(KERN_INFO "ocfs2: Slot %d on device (%s) was already " + "allocated to this node!\n", slot, osb->dev_str);

ocfs2_set_slot(si, slot, osb->node_num); osb->slot_num = slot; --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -173,7 +173,6 @@ enum { Opt_dir_resv_level, Opt_journal_async_commit, Opt_err_cont, - Opt_nocluster, Opt_err, };

@@ -207,7 +206,6 @@ static const match_table_t tokens = { {Opt_dir_resv_level, "dir_resv_level=%u"}, {Opt_journal_async_commit, "journal_async_commit"}, {Opt_err_cont, "errors=continue"}, - {Opt_nocluster, "nocluster"}, {Opt_err, NULL} };

@@ -619,13 +617,6 @@ static int ocfs2_remount(struct super_bl goto out; }

- tmp = OCFS2_MOUNT_NOCLUSTER; - if ((osb->s_mount_opt & tmp) != (parsed_options.mount_opt & tmp)) { - ret = -EINVAL; - mlog(ML_ERROR, "Cannot change nocluster option on remount\n"); - goto out; - } - tmp = OCFS2_MOUNT_HB_LOCAL | OCFS2_MOUNT_HB_GLOBAL | OCFS2_MOUNT_HB_NONE; if ((osb->s_mount_opt & tmp) != (parsed_options.mount_opt & tmp)) { @@ -866,7 +857,6 @@ static int ocfs2_verify_userspace_stack( }

if (ocfs2_userspace_stack(osb) && - !(osb->s_mount_opt & OCFS2_MOUNT_NOCLUSTER) && strncmp(osb->osb_cluster_stack, mopt->cluster_stack, OCFS2_STACK_LABEL_LEN)) { mlog(ML_ERROR, @@ -1145,11 +1135,6 @@ static int ocfs2_fill_super(struct super osb->s_mount_opt & OCFS2_MOUNT_DATA_WRITEBACK ? "writeback" : "ordered");

- if ((osb->s_mount_opt & OCFS2_MOUNT_NOCLUSTER) && - !(osb->s_feature_incompat & OCFS2_FEATURE_INCOMPAT_LOCAL_MOUNT)) - printk(KERN_NOTICE "ocfs2: The shared device (%s) is mounted " - "without cluster aware mode.\n", osb->dev_str); - atomic_set(&osb->vol_state, VOLUME_MOUNTED); wake_up(&osb->osb_mount_event);

@@ -1456,9 +1441,6 @@ static int ocfs2_parse_options(struct su case Opt_journal_async_commit: mopt->mount_opt |= OCFS2_MOUNT_JOURNAL_ASYNC_COMMIT; break; - case Opt_nocluster: - mopt->mount_opt |= OCFS2_MOUNT_NOCLUSTER; - break; default: mlog(ML_ERROR, "Unrecognized mount option "%s" " @@ -1570,9 +1552,6 @@ static int ocfs2_show_options(struct seq if (opts & OCFS2_MOUNT_JOURNAL_ASYNC_COMMIT) seq_printf(s, ",journal_async_commit");

- if (opts & OCFS2_MOUNT_NOCLUSTER) - seq_printf(s, ",nocluster"); - return 0; }

From: Andrei Vagin avagin@gmail.com

commit bdeb77bc2c405fa9f954c20269db175a0bd2793f upstream.

sendfile has to return EAGAIN if out_fd is nonblocking and the write into it would block.

Here is a small reproducer for the problem:

#define _GNU_SOURCE /* See feature_test_macros(7) */ #include <fcntl.h> #include <stdio.h> #include <unistd.h> #include <errno.h> #include <sys/stat.h> #include <sys/types.h> #include <sys/sendfile.h>

#define FILE_SIZE (1UL << 30) int main(int argc, char **argv) { int p[2], fd;

if (pipe2(p, O_NONBLOCK)) return 1;

fd = open(argv[1], O_RDWR | O_TMPFILE, 0666); if (fd < 0) return 1; ftruncate(fd, FILE_SIZE);

if (sendfile(p[1], fd, 0, FILE_SIZE) == -1) { fprintf(stderr, "FAIL\n"); } if (sendfile(p[1], fd, 0, FILE_SIZE) != -1 || errno != EAGAIN) { fprintf(stderr, "FAIL\n"); } return 0; }

It worked before b964bf53e540, it is stuck after b964bf53e540, and it works again with this fix.

This regression occurred because do_splice_direct() calls pipe_write that handles O_NONBLOCK. Here is a trace log from the reproducer:

1) | __x64_sys_sendfile64() { 1) | do_sendfile() { 1) | __fdget() 1) | rw_verify_area() 1) | __fdget() 1) | rw_verify_area() 1) | do_splice_direct() { 1) | rw_verify_area() 1) | splice_direct_to_actor() { 1) | do_splice_to() { 1) | rw_verify_area() 1) | generic_file_splice_read() 1) + 74.153 us | } 1) | direct_splice_actor() { 1) | iter_file_splice_write() { 1) | __kmalloc() 1) 0.148 us | pipe_lock(); 1) 0.153 us | splice_from_pipe_next.part.0(); 1) 0.162 us | page_cache_pipe_buf_confirm(); ... 16 times 1) 0.159 us | page_cache_pipe_buf_confirm(); 1) | vfs_iter_write() { 1) | do_iter_write() { 1) | rw_verify_area() 1) | do_iter_readv_writev() { 1) | pipe_write() { 1) | mutex_lock() 1) 0.153 us | mutex_unlock(); 1) 1.368 us | } 1) 1.686 us | } 1) 5.798 us | } 1) 6.084 us | } 1) 0.174 us | kfree(); 1) 0.152 us | pipe_unlock(); 1) + 14.461 us | } 1) + 14.783 us | } 1) 0.164 us | page_cache_pipe_buf_release(); ... 16 times 1) 0.161 us | page_cache_pipe_buf_release(); 1) | touch_atime() 1) + 95.854 us | } 1) + 99.784 us | } 1) ! 107.393 us | } 1) ! 107.699 us | }

Link: https://lkml.kernel.org/r/20220415005015.525191-1-avagin@gmail.com Fixes: b964bf53e540 ("teach sendfile(2) to handle send-to-pipe directly") Signed-off-by: Andrei Vagin avagin@gmail.com Cc: Al Viro viro@zeniv.linux.org.uk Cc: stable@vger.kernel.org Signed-off-by: Andrew Morton akpm@linux-foundation.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- fs/read_write.c | 3 +++ 1 file changed, 3 insertions(+)

--- a/fs/read_write.c +++ b/fs/read_write.c @@ -1250,6 +1250,9 @@ static ssize_t do_sendfile(int out_fd, i count, fl); file_end_write(out.file); } else { + if (out.file->f_flags & O_NONBLOCK) + fl |= SPLICE_F_NONBLOCK; + retval = splice_file_to_pipe(in.file, opipe, &pos, count, fl); }

From: Josef Bacik josef@toxicpanda.com

commit 3fe2895cfecd03ac74977f32102b966b6589f481 upstream.

We have an application with a lot of threads that use a shared mmap backed by tmpfs mounted with -o huge=within_size. This application started leaking loads of huge pages when we upgraded to a recent kernel.

Using the page ref tracepoints and a BPF program written by Tejun Heo we were able to determine that these pages would have multiple refcounts from the page fault path, but when it came to unmap time we wouldn't drop the number of refs we had added from the faults.

I wrote a reproducer that mmap'ed a file backed by tmpfs with -o huge=always, and then spawned 20 threads all looping faulting random offsets in this map, while using madvise(MADV_DONTNEED) randomly for huge page aligned ranges. This very quickly reproduced the problem.

The problem here is that we check for the case that we have multiple threads faulting in a range that was previously unmapped. One thread maps the PMD, the other thread loses the race and then returns 0. However at this point we already have the page, and we are no longer putting this page into the processes address space, and so we leak the page. We actually did the correct thing prior to f9ce0be71d1f, however it looks like Kirill copied what we do in the anonymous page case. In the anonymous page case we don't yet have a page, so we don't have to drop a reference on anything. Previously we did the correct thing for file based faults by returning VM_FAULT_NOPAGE so we correctly drop the reference on the page we faulted in.

Fix this by returning VM_FAULT_NOPAGE in the pmd_devmap_trans_unstable() case, this makes us drop the ref on the page properly, and now my reproducer no longer leaks the huge pages.

[josef@toxicpanda.com: v2] Link: https://lkml.kernel.org/r/e90c8f0dbae836632b669c2afc434006a00d4a67.165772147... Link: https://lkml.kernel.org/r/2b798acfd95c9ab9395fe85e8d5a835e2e10a920.165705113... Fixes: f9ce0be71d1f ("mm: Cleanup faultaround and finish_fault() codepaths") Signed-off-by: Josef Bacik josef@toxicpanda.com Signed-off-by: Rik van Riel riel@surriel.com Signed-off-by: Chris Mason clm@fb.com Acked-by: Kirill A. Shutemov kirill.shutemov@linux.intel.com Cc: Matthew Wilcox (Oracle) willy@infradead.org Cc: stable@vger.kernel.org Signed-off-by: Andrew Morton akpm@linux-foundation.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- mm/memory.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-)

--- a/mm/memory.c +++ b/mm/memory.c @@ -4071,9 +4071,12 @@ vm_fault_t finish_fault(struct vm_fault } }

- /* See comment in handle_pte_fault() */ + /* + * See comment in handle_pte_fault() for how this scenario happens, we + * need to return NOPAGE so that we drop this page. + */ if (pmd_devmap_trans_unstable(vmf->pmd)) - return 0; + return VM_FAULT_NOPAGE;

vmf->pte = pte_offset_map_lock(vma->vm_mm, vmf->pmd, vmf->address, &vmf->ptl);

From: Lukas Bulwahn lukas.bulwahn@gmail.com

commit e2a619ca0b38f2114347b7078b8a67d72d457a3d upstream.

Commit 527701eda5f1 ("lib: Add a generic version of devmem_is_allowed()") introduces the config symbol GENERIC_LIB_DEVMEM_IS_ALLOWED, but then falsely refers to CONFIG_GENERIC_DEVMEM_IS_ALLOWED (note the missing LIB in the reference) in ./include/asm-generic/io.h.

Luckily, ./scripts/checkkconfigsymbols.py warns on non-existing configs:

GENERIC_DEVMEM_IS_ALLOWED Referencing files: include/asm-generic/io.h

The actual fix, though, is simply to not to make this function declaration dependent on any kernel config. For architectures that intend to use the generic version, the arch's 'select GENERIC_LIB_DEVMEM_IS_ALLOWED' will lead to picking the function definition, and for other architectures, this function is simply defined elsewhere.

The wrong '#ifndef' on a non-existing config symbol also always had the same effect (although more by mistake than by intent). So, there is no functional change.

Remove this broken and needless ifdef conditional.

Fixes: 527701eda5f1 ("lib: Add a generic version of devmem_is_allowed()") Signed-off-by: Lukas Bulwahn lukas.bulwahn@gmail.com Signed-off-by: Arnd Bergmann arnd@arndb.de Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- include/asm-generic/io.h | 2 -- 1 file changed, 2 deletions(-)

--- a/include/asm-generic/io.h +++ b/include/asm-generic/io.h @@ -1125,9 +1125,7 @@ static inline void memcpy_toio(volatile } #endif

-#ifndef CONFIG_GENERIC_DEVMEM_IS_ALLOWED extern int devmem_is_allowed(unsigned long pfn); -#endif

#endif /* __KERNEL__ */

From: Alistair Popple apopple@nvidia.com

commit 66cee9097e2b74ff3c8cc040ce5717c521a0c3fa upstream.

Users may request that pages from an OpenCL SVM allocation be migrated to the GPU with clEnqueueSVMMigrateMem(). In Nouveau this will call into nouveau_dmem_migrate_vma() to do the migration. If the total range to be migrated exceeds SG_MAX_SINGLE_ALLOC the pages will be migrated in chunks of size SG_MAX_SINGLE_ALLOC. However a typo in updating the starting address means that only the first chunk will get migrated.

Fix the calculation so that the entire range will get migrated if possible.

Signed-off-by: Alistair Popple apopple@nvidia.com Fixes: e3d8b0890469 ("drm/nouveau/svm: map pages after migration") Reviewed-by: Ralph Campbell rcampbell@nvidia.com Reviewed-by: Lyude Paul lyude@redhat.com Signed-off-by: Lyude Paul lyude@redhat.com Link: https://patchwork.freedesktop.org/patch/msgid/20220720062745.960701-1-apoppl... Cc: stable@vger.kernel.org # v5.8+ Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/gpu/drm/nouveau/nouveau_dmem.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)

--- a/drivers/gpu/drm/nouveau/nouveau_dmem.c +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c @@ -679,7 +679,11 @@ nouveau_dmem_migrate_vma(struct nouveau_ goto out_free_dma;

for (i = 0; i < npages; i += max) { - args.end = start + (max << PAGE_SHIFT); + if (args.start + (max << PAGE_SHIFT) > end) + args.end = end; + else + args.end = args.start + (max << PAGE_SHIFT); + ret = migrate_vma_setup(&args); if (ret) goto out_free_pfns;

From: David Howells dhowells@redhat.com

commit e0339f036ef4beb9b20f0b6532a1e0ece7f594c6 upstream.

Since __post_watch_notification() walks wlist->watchers with only the RCU read lock held, we need to use RCU methods to add to the list (we already use RCU methods to remove from the list).

Fix add_watch_to_object() to use hlist_add_head_rcu() instead of hlist_add_head() for that list.

Fixes: c73be61cede5 ("pipe: Add general notification queue support") Signed-off-by: David Howells dhowells@redhat.com Signed-off-by: Linus Torvalds torvalds@linux-foundation.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- kernel/watch_queue.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/kernel/watch_queue.c +++ b/kernel/watch_queue.c @@ -497,7 +497,7 @@ int add_watch_to_object(struct watch *wa unlock_wqueue(wqueue); }

- hlist_add_head(&watch->list_node, &wlist->watchers); + hlist_add_head_rcu(&watch->list_node, &wlist->watchers); return 0; } EXPORT_SYMBOL(add_watch_to_object);

From: Kuniyuki Iwashima kuniyu@amazon.com

commit 58ebb1c8b35a8ef38cd6927431e0fa7b173a632d upstream.

While reading sysctl_tcp_dsack, it can be changed concurrently. Thus, we need to add READ_ONCE() to its readers.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Kuniyuki Iwashima kuniyu@amazon.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- net/ipv4/tcp_input.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)

--- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -4419,7 +4419,7 @@ static void tcp_dsack_set(struct sock *s { struct tcp_sock *tp = tcp_sk(sk);

- if (tcp_is_sack(tp) && sock_net(sk)->ipv4.sysctl_tcp_dsack) { + if (tcp_is_sack(tp) && READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_dsack)) { int mib_idx;

if (before(seq, tp->rcv_nxt)) @@ -4466,7 +4466,7 @@ static void tcp_send_dupack(struct sock NET_INC_STATS(sock_net(sk), LINUX_MIB_DELAYEDACKLOST); tcp_enter_quickack_mode(sk, TCP_MAX_QUICKACKS);

- if (tcp_is_sack(tp) && sock_net(sk)->ipv4.sysctl_tcp_dsack) { + if (tcp_is_sack(tp) && READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_dsack)) { u32 end_seq = TCP_SKB_CB(skb)->end_seq;

tcp_rcv_spurious_retrans(sk, skb);

From: Kuniyuki Iwashima kuniyu@amazon.com

commit 36eeee75ef0157e42fb6593dcc65daab289b559e upstream.

While reading sysctl_tcp_adv_win_scale, it can be changed concurrently. Thus, we need to add READ_ONCE() to its reader.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Kuniyuki Iwashima kuniyu@amazon.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- include/net/tcp.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -1406,7 +1406,7 @@ void tcp_select_initial_window(const str

static inline int tcp_win_from_space(const struct sock *sk, int space) { - int tcp_adv_win_scale = sock_net(sk)->ipv4.sysctl_tcp_adv_win_scale; + int tcp_adv_win_scale = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_adv_win_scale);

return tcp_adv_win_scale <= 0 ? (space>>(-tcp_adv_win_scale)) :

From: Kuniyuki Iwashima kuniyu@amazon.com

commit 8499a2454d9e8a55ce616ede9f9580f36fd5b0f3 upstream.

While reading sysctl_tcp_nometrics_save, it can be changed concurrently. Thus, we need to add READ_ONCE() to its reader.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Kuniyuki Iwashima kuniyu@amazon.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- net/ipv4/tcp_metrics.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/net/ipv4/tcp_metrics.c +++ b/net/ipv4/tcp_metrics.c @@ -329,7 +329,7 @@ void tcp_update_metrics(struct sock *sk) int m;

sk_dst_confirm(sk); - if (net->ipv4.sysctl_tcp_nometrics_save || !dst) + if (READ_ONCE(net->ipv4.sysctl_tcp_nometrics_save) || !dst) return;

rcu_read_lock();

From: Maciej Fijalkowski maciej.fijalkowski@intel.com

commit 283d736ff7c7e96ac5b32c6c0de40372f8eb171e upstream.

Tx side sets EOP and RS bits on descriptors to indicate that a particular descriptor is the last one and needs to generate an irq when it was sent. These bits should not be checked on completion path regardless whether it's the Tx or the Rx. DD bit serves this purpose and it indicates that a particular descriptor is either for Rx or was successfully Txed. EOF is also set as loopback test does not xmit fragmented frames.

Look at (DD | EOF) bits setting in ice_lbtest_receive_frames() instead of EOP and RS pair.

Fixes: 0e674aeb0b77 ("ice: Add handler for ethtool selftest") Signed-off-by: Maciej Fijalkowski maciej.fijalkowski@intel.com Tested-by: George Kuruvinakunnel george.kuruvinakunnel@intel.com Signed-off-by: Tony Nguyen anthony.l.nguyen@intel.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/net/ethernet/intel/ice/ice_ethtool.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)

--- a/drivers/net/ethernet/intel/ice/ice_ethtool.c +++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c @@ -651,7 +651,8 @@ static int ice_lbtest_receive_frames(str rx_desc = ICE_RX_DESC(rx_ring, i);

if (!(rx_desc->wb.status_error0 & - cpu_to_le16(ICE_TX_DESC_CMD_EOP | ICE_TX_DESC_CMD_RS))) + (cpu_to_le16(BIT(ICE_RX_FLEX_DESC_STATUS0_DD_S)) | + cpu_to_le16(BIT(ICE_RX_FLEX_DESC_STATUS0_EOF_S))))) continue;

rx_buf = &rx_ring->rx_buf[i];

From: Liang He windhl@126.com

commit a3435afba87dc6cd83f5595e7607f3c40f93ef01 upstream.

In ufshcd_populate_vreg(), we should hold the reference returned by of_parse_phandle() and then use it to call of_node_put() for refcount balance.

Link: https://lore.kernel.org/r/20220719071529.1081166-1-windhl@126.com Fixes: aa4976130934 ("ufs: Add regulator enable support") Reviewed-by: Bart Van Assche bvanassche@acm.org Signed-off-by: Liang He windhl@126.com Signed-off-by: Martin K. Petersen martin.petersen@oracle.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/scsi/ufs/ufshcd-pltfrm.c | 15 +++++++++++++-- 1 file changed, 13 insertions(+), 2 deletions(-)

--- a/drivers/scsi/ufs/ufshcd-pltfrm.c +++ b/drivers/scsi/ufs/ufshcd-pltfrm.c @@ -107,9 +107,20 @@ out: return ret; }

+static bool phandle_exists(const struct device_node *np, + const char *phandle_name, int index) +{ + struct device_node *parse_np = of_parse_phandle(np, phandle_name, index); + + if (parse_np) + of_node_put(parse_np); + + return parse_np != NULL; +} + #define MAX_PROP_SIZE 32 static int ufshcd_populate_vreg(struct device *dev, const char *name, - struct ufs_vreg **out_vreg) + struct ufs_vreg **out_vreg) { char prop_name[MAX_PROP_SIZE]; struct ufs_vreg *vreg = NULL; @@ -121,7 +132,7 @@ static int ufshcd_populate_vreg(struct d }

snprintf(prop_name, MAX_PROP_SIZE, "%s-supply", name); - if (!of_parse_phandle(np, prop_name, 0)) { + if (!phandle_exists(np, prop_name, 0)) { dev_info(dev, "%s: Unable to find %s regulator, assuming enabled\n", __func__, prop_name); goto out;

From: Subbaraya Sundeep sbhatta@marvell.com

commit 59e1be6f83b928a04189bbf3ab683a1fc6248db3 upstream.

Check the mask for non-zero value before installing tc filters for L4 source and destination ports. Otherwise installing a filter for source port installs destination port too and vice-versa.

Fixes: 1d4d9e42c240 ("octeontx2-pf: Add tc flower hardware offload on ingress traffic") Signed-off-by: Subbaraya Sundeep sbhatta@marvell.com Signed-off-by: Sunil Goutham sgoutham@marvell.com Signed-off-by: Paolo Abeni pabeni@redhat.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/net/ethernet/marvell/octeontx2/nic/otx2_tc.c | 30 +++++++++++-------- 1 file changed, 18 insertions(+), 12 deletions(-)

--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_tc.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_tc.c @@ -571,21 +571,27 @@ static int otx2_tc_prepare_flow(struct o

flow_spec->dport = match.key->dst; flow_mask->dport = match.mask->dst; - if (ip_proto == IPPROTO_UDP) - req->features |= BIT_ULL(NPC_DPORT_UDP); - else if (ip_proto == IPPROTO_TCP) - req->features |= BIT_ULL(NPC_DPORT_TCP); - else if (ip_proto == IPPROTO_SCTP) - req->features |= BIT_ULL(NPC_DPORT_SCTP); + + if (flow_mask->dport) { + if (ip_proto == IPPROTO_UDP) + req->features |= BIT_ULL(NPC_DPORT_UDP); + else if (ip_proto == IPPROTO_TCP) + req->features |= BIT_ULL(NPC_DPORT_TCP); + else if (ip_proto == IPPROTO_SCTP) + req->features |= BIT_ULL(NPC_DPORT_SCTP); + }

flow_spec->sport = match.key->src; flow_mask->sport = match.mask->src; - if (ip_proto == IPPROTO_UDP) - req->features |= BIT_ULL(NPC_SPORT_UDP); - else if (ip_proto == IPPROTO_TCP) - req->features |= BIT_ULL(NPC_SPORT_TCP); - else if (ip_proto == IPPROTO_SCTP) - req->features |= BIT_ULL(NPC_SPORT_SCTP); + + if (flow_mask->sport) { + if (ip_proto == IPPROTO_UDP) + req->features |= BIT_ULL(NPC_SPORT_UDP); + else if (ip_proto == IPPROTO_TCP) + req->features |= BIT_ULL(NPC_SPORT_TCP); + else if (ip_proto == IPPROTO_SCTP) + req->features |= BIT_ULL(NPC_SPORT_SCTP); + } }

return otx2_tc_parse_actions(nic, &rule->action, req, f, node);

From: Kuniyuki Iwashima kuniyu@amazon.com

commit 9fb90193fbd66b4c5409ef729fd081861f8b6351 upstream.

While reading sysctl_tcp_limit_output_bytes, it can be changed concurrently. Thus, we need to add READ_ONCE() to its reader.

Fixes: 46d3ceabd8d9 ("tcp: TCP Small Queues") Signed-off-by: Kuniyuki Iwashima kuniyu@amazon.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- net/ipv4/tcp_output.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -2503,7 +2503,7 @@ static bool tcp_small_queue_check(struct sk->sk_pacing_rate >> READ_ONCE(sk->sk_pacing_shift)); if (sk->sk_pacing_status == SK_PACING_NONE) limit = min_t(unsigned long, limit, - sock_net(sk)->ipv4.sysctl_tcp_limit_output_bytes); + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_limit_output_bytes)); limit <<= factor;

if (static_branch_unlikely(&tcp_tx_delay_enabled) &&

From: Jason Yan yanaijie@huawei.com

commit d9a434fa0c12ed5f7afe1e9dd30003ab5d059b85 upstream.

As explained in SG_IO howto[1]:

"If iovec_count is non-zero then 'dxfer_len' should be equal to the sum of iov_len lengths. If not, the minimum of the two is the transfer length."

When iovec_count is non-zero and dxfer_len is zero, the sg_io() just genarated a null bio, and finally caused a warning below. To fix it, skip generating a bio for this request if dxfer_len is zero.

[1] https://tldp.org/HOWTO/SCSI-Generic-HOWTO/x198.html

WARNING: CPU: 2 PID: 3643 at drivers/scsi/scsi_lib.c:1032 scsi_alloc_sgtables+0xc7d/0xf70 drivers/scsi/scsi_lib.c:1032 Modules linked in:

CPU: 2 PID: 3643 Comm: syz-executor397 Not tainted 5.17.0-rc3-syzkaller-00316-gb81b1829e7e3 #0 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.14.0-204/01/2014 RIP: 0010:scsi_alloc_sgtables+0xc7d/0xf70 drivers/scsi/scsi_lib.c:1032 Code: e7 fc 31 ff 44 89 f6 e8 c1 4e e7 fc 45 85 f6 0f 84 1a f5 ff ff e8 93 4c e7 fc 83 c5 01 0f b7 ed e9 0f f5 ff ff e8 83 4c e7 fc <0f> 0b 41 bc 0a 00 00 00 e9 2b fb ff ff 41 bc 09 00 00 00 e9 20 fb RSP: 0018:ffffc90000d07558 EFLAGS: 00010293 RAX: 0000000000000000 RBX: ffff88801bfc96a0 RCX: 0000000000000000 RDX: ffff88801c876000 RSI: ffffffff849060bd RDI: 0000000000000003 RBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000000 R10: ffffffff849055b9 R11: 0000000000000000 R12: ffff888012b8c000 R13: ffff88801bfc9580 R14: 0000000000000000 R15: ffff88801432c000 FS: 00007effdec8e700(0000) GS:ffff88802cc00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007effdec6d718 CR3: 00000000206d6000 CR4: 0000000000150ee0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: <TASK> scsi_setup_scsi_cmnd drivers/scsi/scsi_lib.c:1219 [inline] scsi_prepare_cmd drivers/scsi/scsi_lib.c:1614 [inline] scsi_queue_rq+0x283e/0x3630 drivers/scsi/scsi_lib.c:1730 blk_mq_dispatch_rq_list+0x6ea/0x22e0 block/blk-mq.c:1851 __blk_mq_sched_dispatch_requests+0x20b/0x410 block/blk-mq-sched.c:299 blk_mq_sched_dispatch_requests+0xfb/0x180 block/blk-mq-sched.c:332 __blk_mq_run_hw_queue+0xf9/0x350 block/blk-mq.c:1968 __blk_mq_delay_run_hw_queue+0x5b6/0x6c0 block/blk-mq.c:2045 blk_mq_run_hw_queue+0x30f/0x480 block/blk-mq.c:2096 blk_mq_sched_insert_request+0x340/0x440 block/blk-mq-sched.c:451 blk_execute_rq+0xcc/0x340 block/blk-mq.c:1231 sg_io+0x67c/0x1210 drivers/scsi/scsi_ioctl.c:485 scsi_ioctl_sg_io drivers/scsi/scsi_ioctl.c:866 [inline] scsi_ioctl+0xa66/0x1560 drivers/scsi/scsi_ioctl.c:921 sd_ioctl+0x199/0x2a0 drivers/scsi/sd.c:1576 blkdev_ioctl+0x37a/0x800 block/ioctl.c:588 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:874 [inline] __se_sys_ioctl fs/ioctl.c:860 [inline] __x64_sys_ioctl+0x193/0x200 fs/ioctl.c:860 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x44/0xae RIP: 0033:0x7effdecdc5d9 Code: 28 00 00 00 75 05 48 83 c4 28 c3 e8 81 14 00 00 90 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 b8 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007effdec8e2f8 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 RAX: ffffffffffffffda RBX: 00007effded664c0 RCX: 00007effdecdc5d9 RDX: 0000000020002300 RSI: 0000000000002285 RDI: 0000000000000004 RBP: 00007effded34034 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000003 R13: 00007effded34054 R14: 2f30656c69662f2e R15: 00007effded664c8

Link: https://lore.kernel.org/r/20220720025120.3226770-1-yanaijie@huawei.com Fixes: 25636e282fe9 ("block: fix SG_IO vector request data length handling") Reported-by: syzbot+d44b35ecfb807e5af0b5@syzkaller.appspotmail.com Reviewed-by: Christoph Hellwig hch@lst.de Reviewed-by: Bart Van Assche bvanassche@acm.org Signed-off-by: Jason Yan yanaijie@huawei.com Signed-off-by: Martin K. Petersen martin.petersen@oracle.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/scsi/scsi_ioctl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/drivers/scsi/scsi_ioctl.c +++ b/drivers/scsi/scsi_ioctl.c @@ -457,7 +457,7 @@ static int sg_io(struct scsi_device *sde goto out_free_cdb;

ret = 0; - if (hdr->iovec_count) { + if (hdr->iovec_count && hdr->dxfer_len) { struct iov_iter i; struct iovec *iov = NULL;

From: Kuniyuki Iwashima kuniyu@amazon.com

commit e27326009a3d247b831eda38878c777f6f4eb3d1 upstream.

When we close ping6 sockets, some resources are left unfreed because pingv6_prot is missing sk->sk_prot->destroy(). As reported by syzbot [0], just three syscalls leak 96 bytes and easily cause OOM.

struct ipv6_sr_hdr *hdr; char data[24] = {0}; int fd;

hdr = (struct ipv6_sr_hdr *)data; hdr->hdrlen = 2; hdr->type = IPV6_SRCRT_TYPE_4;

fd = socket(AF_INET6, SOCK_DGRAM, NEXTHDR_ICMP); setsockopt(fd, IPPROTO_IPV6, IPV6_RTHDR, data, 24); close(fd);

To fix memory leaks, let's add a destroy function.

Note the socket() syscall checks if the GID is within the range of net.ipv4.ping_group_range. The default value is [1, 0] so that no GID meets the condition (1 <= GID <= 0). Thus, the local DoS does not succeed until we change the default value. However, at least Ubuntu/Fedora/RHEL loosen it.

$ cat /usr/lib/sysctl.d/50-default.conf ... -net.ipv4.ping_group_range = 0 2147483647

Also, there could be another path reported with these options, and some of them require CAP_NET_RAW.

setsockopt IPV6_ADDRFORM (inet6_sk(sk)->pktoptions) IPV6_RECVPATHMTU (inet6_sk(sk)->rxpmtu) IPV6_HOPOPTS (inet6_sk(sk)->opt) IPV6_RTHDRDSTOPTS (inet6_sk(sk)->opt) IPV6_RTHDR (inet6_sk(sk)->opt) IPV6_DSTOPTS (inet6_sk(sk)->opt) IPV6_2292PKTOPTIONS (inet6_sk(sk)->opt)

getsockopt IPV6_FLOWLABEL_MGR (inet6_sk(sk)->ipv6_fl_list)

For the record, I left a different splat with syzbot's one.

unreferenced object 0xffff888006270c60 (size 96): comm "repro2", pid 231, jiffies 4294696626 (age 13.118s) hex dump (first 32 bytes): 01 00 00 00 44 00 00 00 00 00 00 00 00 00 00 00 ....D........... 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace: [<00000000f6bc7ea9>] sock_kmalloc (net/core/sock.c:2564 net/core/sock.c:2554) [<000000006d699550>] do_ipv6_setsockopt.constprop.0 (net/ipv6/ipv6_sockglue.c:715) [<00000000c3c3b1f5>] ipv6_setsockopt (net/ipv6/ipv6_sockglue.c:1024) [<000000007096a025>] __sys_setsockopt (net/socket.c:2254) [<000000003a8ff47b>] __x64_sys_setsockopt (net/socket.c:2265 net/socket.c:2262 net/socket.c:2262) [<000000007c409dcb>] do_syscall_64 (arch/x86/entry/common.c:50 arch/x86/entry/common.c:80) [<00000000e939c4a9>] entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:120)

[0]: https://syzkaller.appspot.com/bug?extid=a8430774139ec3ab7176

Fixes: 6d0bfe226116 ("net: ipv6: Add IPv6 support to the ping socket.") Reported-by: syzbot+a8430774139ec3ab7176@syzkaller.appspotmail.com Reported-by: Ayushman Dutta ayudutta@amazon.com Signed-off-by: Kuniyuki Iwashima kuniyu@amazon.com Reviewed-by: David Ahern dsahern@kernel.org Reviewed-by: Eric Dumazet edumazet@google.com Link: https://lore.kernel.org/r/20220728012220.46918-1-kuniyu@amazon.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- net/ipv6/ping.c | 6 ++++++ 1 file changed, 6 insertions(+)

--- a/net/ipv6/ping.c +++ b/net/ipv6/ping.c @@ -22,6 +22,11 @@ #include <linux/proc_fs.h> #include <net/ping.h>

+static void ping_v6_destroy(struct sock *sk) +{ + inet6_destroy_sock(sk); +} + /* Compatibility glue so we can support IPv6 when it's compiled as a module */ static int dummy_ipv6_recv_error(struct sock *sk, struct msghdr *msg, int len, int *addr_len) @@ -166,6 +171,7 @@ struct proto pingv6_prot = { .owner = THIS_MODULE, .init = ping_init_sock, .close = ping_close, + .destroy = ping_v6_destroy, .connect = ip6_datagram_connect_v6_only, .disconnect = __udp_disconnect, .setsockopt = ipv6_setsockopt,

From: Maxim Mikityanskiy maximmi@nvidia.com

commit f6336724a4d4220c89a4ec38bca84b03b178b1a3 upstream.

tls_device_down takes a reference on all contexts it's going to move to the degraded state (software fallback). If sk_destruct runs afterwards, it can reduce the reference counter back to 1 and return early without destroying the context. Then tls_device_down will release the reference it took and call tls_device_free_ctx. However, the context will still stay in tls_device_down_list forever. The list will contain an item, memory for which is released, making a memory corruption possible.

Fix the above bug by properly removing the context from all lists before any call to tls_device_free_ctx.

Fixes: 3740651bf7e2 ("tls: Fix context leak on tls_device_down") Signed-off-by: Maxim Mikityanskiy maximmi@nvidia.com Reviewed-by: Tariq Toukan tariqt@nvidia.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- net/tls/tls_device.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)

--- a/net/tls/tls_device.c +++ b/net/tls/tls_device.c @@ -1351,8 +1351,13 @@ static int tls_device_down(struct net_de * by tls_device_free_ctx. rx_conf and tx_conf stay in TLS_HW. * Now release the ref taken above. */ - if (refcount_dec_and_test(&ctx->refcount)) + if (refcount_dec_and_test(&ctx->refcount)) { + /* sk_destruct ran after tls_device_down took a ref, and + * it returned early. Complete the destruction here. + */ + list_del(&ctx->list); tls_device_free_ctx(ctx); + } }

up_write(&device_offload_lock);

From: Vladimir Oltean vladimir.oltean@nxp.com

[ Upstream commit 27161db0904ee48e59140aa8d0835939a666c1f1 ]

While phylink_pcs_ops :: pcs_get_state does return void, xpcs_get_state() does check for a non-zero return code from xpcs_get_state_c37_sgmii() and prints that as a message to the kernel log.

However, a non-zero return code from xpcs_read() is translated into "return false" (i.e. zero as int) and the I/O error is therefore not printed. Fix that.

Fixes: b97b5331b8ab ("net: pcs: add C37 SGMII AN support for intel mGbE controller") Signed-off-by: Vladimir Oltean vladimir.oltean@nxp.com Link: https://lore.kernel.org/r/20220720112057.3504398-1-vladimir.oltean@nxp.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/pcs/pcs-xpcs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/pcs/pcs-xpcs.c b/drivers/net/pcs/pcs-xpcs.c index 7de631f5356f..fd4cbf8a55ad 100644 --- a/drivers/net/pcs/pcs-xpcs.c +++ b/drivers/net/pcs/pcs-xpcs.c @@ -890,7 +890,7 @@ static int xpcs_get_state_c37_sgmii(struct dw_xpcs *xpcs, */ ret = xpcs_read(xpcs, MDIO_MMD_VEND2, DW_VR_MII_AN_INTR_STS); if (ret < 0) - return false; + return ret;

if (ret & DW_VR_MII_C37_ANSGM_SP_LNKSTS) { int speed_value;

From: Kuniyuki Iwashima kuniyu@amazon.com

[ Upstream commit e0bb4ab9dfddd872622239f49fb2bd403b70853b ]

While reading sysctl_tcp_min_tso_segs, it can be changed concurrently. Thus, we need to add READ_ONCE() to its reader.

Fixes: 95bd09eb2750 ("tcp: TSO packets automatic sizing") Signed-off-by: Kuniyuki Iwashima kuniyu@amazon.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- net/ipv4/tcp_output.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

--- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -1986,7 +1986,7 @@ static u32 tcp_tso_segs(struct sock *sk,

min_tso = ca_ops->min_tso_segs ? ca_ops->min_tso_segs(sk) : - sock_net(sk)->ipv4.sysctl_tcp_min_tso_segs; + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_min_tso_segs);

tso_segs = tcp_tso_autosize(sk, mss_now, min_tso); return min_t(u32, tso_segs, sk->sk_gso_max_segs);

From: Kuniyuki Iwashima kuniyu@amazon.com

[ Upstream commit 85225e6f0a76e6745bc841c9f25169c509b573d8 ]

While reading sysctl_tcp_autocorking, it can be changed concurrently. Thus, we need to add READ_ONCE() to its reader.

Fixes: f54b311142a9 ("tcp: auto corking") Signed-off-by: Kuniyuki Iwashima kuniyu@amazon.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- net/ipv4/tcp.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 1abdb8712655..7ba9059c263a 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -694,7 +694,7 @@ static bool tcp_should_autocork(struct sock *sk, struct sk_buff *skb, int size_goal) { return skb->len < size_goal && - sock_net(sk)->ipv4.sysctl_tcp_autocorking && + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_autocorking) && !tcp_rtx_queue_empty(sk) && refcount_read(&sk->sk_wmem_alloc) > skb->truesize; }

From: Xin Long lucien.xin@gmail.com

[ Upstream commit aa709da0e032cee7c202047ecd75f437bb0126ed ]

Since commit 1033990ac5b2 ("sctp: implement memory accounting on tx path"), SCTP has supported memory accounting on tx path where 'sctp_wmem' is used by sk_wmem_schedule(). So we should fix the description for this option in ip-sysctl.rst accordingly.

v1->v2: - Improve the description as Marcelo suggested.

Fixes: 1033990ac5b2 ("sctp: implement memory accounting on tx path") Signed-off-by: Xin Long lucien.xin@gmail.com Acked-by: Marcelo Ricardo Leitner marcelo.leitner@gmail.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- Documentation/networking/ip-sysctl.rst | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/Documentation/networking/ip-sysctl.rst b/Documentation/networking/ip-sysctl.rst index b8b67041f955..ba0e8e6337c0 100644 --- a/Documentation/networking/ip-sysctl.rst +++ b/Documentation/networking/ip-sysctl.rst @@ -2808,7 +2808,14 @@ sctp_rmem - vector of 3 INTEGERs: min, default, max Default: 4K

sctp_wmem - vector of 3 INTEGERs: min, default, max - Currently this tunable has no effect. + Only the first value ("min") is used, "default" and "max" are + ignored. + + min: Minimum size of send buffer that can be used by SCTP sockets. + It is guaranteed to each SCTP socket (but not association) even + under moderate memory pressure. + + Default: 4K

addr_scope_policy - INTEGER Control IPv4 address scoping - draft-stewart-tsvwg-sctp-ipv4-00

From: Sabrina Dubroca sd@queasysnail.net

[ Upstream commit 3240eac4ff20e51b87600dbd586ed814daf313db ]

The expected length is MACSEC_SALT_LEN, not MACSEC_SA_ATTR_SALT.

Fixes: 48ef50fa866a ("macsec: Netlink support of XPN cipher suites (IEEE 802.1AEbw)") Signed-off-by: Sabrina Dubroca sd@queasysnail.net Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/macsec.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index f72d4380374d..9ede0d7cd0b5 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -1768,7 +1768,7 @@ static int macsec_add_rxsa(struct sk_buff *skb, struct genl_info *info) if (nla_len(tb_sa[MACSEC_SA_ATTR_SALT]) != MACSEC_SALT_LEN) { pr_notice("macsec: nl: add_rxsa: bad salt length: %d != %d\n", nla_len(tb_sa[MACSEC_SA_ATTR_SALT]), - MACSEC_SA_ATTR_SALT); + MACSEC_SALT_LEN); rtnl_unlock(); return -EINVAL; } @@ -2010,7 +2010,7 @@ static int macsec_add_txsa(struct sk_buff *skb, struct genl_info *info) if (nla_len(tb_sa[MACSEC_SA_ATTR_SALT]) != MACSEC_SALT_LEN) { pr_notice("macsec: nl: add_txsa: bad salt length: %d != %d\n", nla_len(tb_sa[MACSEC_SA_ATTR_SALT]), - MACSEC_SA_ATTR_SALT); + MACSEC_SALT_LEN); rtnl_unlock(); return -EINVAL; }

From: Sabrina Dubroca sd@queasysnail.net

[ Upstream commit c630d1fe6219769049c87d1a6a0e9a6de55328a1 ]

Currently, MACSEC_SA_ATTR_PN is handled inconsistently, sometimes as a u32, sometimes forced into a u64 without checking the actual length of the attribute. Instead, we can use nla_get_u64 everywhere, which will read up to 64 bits into a u64, capped by the actual length of the attribute coming from userspace.

This fixes several issues: - the check in validate_add_rxsa doesn't work with 32-bit attributes - the checks in validate_add_txsa and validate_upd_sa incorrectly reject X << 32 (with X != 0)

Fixes: 48ef50fa866a ("macsec: Netlink support of XPN cipher suites (IEEE 802.1AEbw)") Signed-off-by: Sabrina Dubroca sd@queasysnail.net Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/macsec.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/net/macsec.c b/drivers/net/macsec.c index 1f2eb576533c..3e74dcc1f875 100644 --- a/drivers/net/macsec.c +++ b/drivers/net/macsec.c @@ -1696,7 +1696,7 @@ static bool validate_add_rxsa(struct nlattr **attrs) return false;

if (attrs[MACSEC_SA_ATTR_PN] && - *(u64 *)nla_data(attrs[MACSEC_SA_ATTR_PN]) == 0) + nla_get_u64(attrs[MACSEC_SA_ATTR_PN]) == 0) return false;

if (attrs[MACSEC_SA_ATTR_ACTIVE]) { @@ -1939,7 +1939,7 @@ static bool validate_add_txsa(struct nlattr **attrs) if (nla_get_u8(attrs[MACSEC_SA_ATTR_AN]) >= MACSEC_NUM_AN) return false;

- if (nla_get_u32(attrs[MACSEC_SA_ATTR_PN]) == 0) + if (nla_get_u64(attrs[MACSEC_SA_ATTR_PN]) == 0) return false;

if (attrs[MACSEC_SA_ATTR_ACTIVE]) { @@ -2293,7 +2293,7 @@ static bool validate_upd_sa(struct nlattr **attrs) if (nla_get_u8(attrs[MACSEC_SA_ATTR_AN]) >= MACSEC_NUM_AN) return false;

- if (attrs[MACSEC_SA_ATTR_PN] && nla_get_u32(attrs[MACSEC_SA_ATTR_PN]) == 0) + if (attrs[MACSEC_SA_ATTR_PN] && nla_get_u64(attrs[MACSEC_SA_ATTR_PN]) == 0) return false;

if (attrs[MACSEC_SA_ATTR_ACTIVE]) {

From: Taehee Yoo ap420073@gmail.com

[ Upstream commit 3e7d18b9dca388940a19cae30bfc1f76dccd8c28 ]

mld_{query | report}_work() processes queued events. If there are too many events in the queue, it re-queue a work. And then, it returns without in6_dev_put(). But if queuing is failed, it should call in6_dev_put(), but it doesn't. So, a reference count leak would occur.

THREAD0 THREAD1 mld_report_work() spin_lock_bh() if (!mod_delayed_work()) in6_dev_hold(); spin_unlock_bh() spin_lock_bh() schedule_delayed_work() spin_unlock_bh()

Script to reproduce(by Hangbin Liu): ip netns add ns1 ip netns add ns2 ip netns exec ns1 sysctl -w net.ipv6.conf.all.force_mld_version=1 ip netns exec ns2 sysctl -w net.ipv6.conf.all.force_mld_version=1

ip -n ns1 link add veth0 type veth peer name veth0 netns ns2 ip -n ns1 link set veth0 up ip -n ns2 link set veth0 up

for i in `seq 50`; do for j in `seq 100`; do ip -n ns1 addr add 2021:${i}::${j}/64 dev veth0 ip -n ns2 addr add 2022:${i}::${j}/64 dev veth0 done done modprobe -r veth ip -a netns del

splat looks like: unregister_netdevice: waiting for veth0 to become free. Usage count = 2 leaked reference. ipv6_add_dev+0x324/0xec0 addrconf_notify+0x481/0xd10 raw_notifier_call_chain+0xe3/0x120 call_netdevice_notifiers+0x106/0x160 register_netdevice+0x114c/0x16b0 veth_newlink+0x48b/0xa50 [veth] rtnl_newlink+0x11a2/0x1a40 rtnetlink_rcv_msg+0x63f/0xc00 netlink_rcv_skb+0x1df/0x3e0 netlink_unicast+0x5de/0x850 netlink_sendmsg+0x6c9/0xa90 ____sys_sendmsg+0x76a/0x780 __sys_sendmsg+0x27c/0x340 do_syscall_64+0x43/0x90 entry_SYSCALL_64_after_hwframe+0x63/0xcd

Tested-by: Hangbin Liu liuhangbin@gmail.com Fixes: f185de28d9ae ("mld: add new workqueues for process mld events") Signed-off-by: Taehee Yoo ap420073@gmail.com Reviewed-by: Eric Dumazet edumazet@google.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- net/ipv6/mcast.c | 14 ++++++++------ 1 file changed, 8 insertions(+), 6 deletions(-)

diff --git a/net/ipv6/mcast.c b/net/ipv6/mcast.c index 7f695c39d9a8..87c699d57b36 100644 --- a/net/ipv6/mcast.c +++ b/net/ipv6/mcast.c @@ -1522,7 +1522,6 @@ static void mld_query_work(struct work_struct *work)

if (++cnt >= MLD_MAX_QUEUE) { rework = true; - schedule_delayed_work(&idev->mc_query_work, 0); break; } } @@ -1533,8 +1532,10 @@ static void mld_query_work(struct work_struct *work) __mld_query_work(skb); mutex_unlock(&idev->mc_lock);

- if (!rework) - in6_dev_put(idev); + if (rework && queue_delayed_work(mld_wq, &idev->mc_query_work, 0)) + return; + + in6_dev_put(idev); }

/* called with rcu_read_lock() */ @@ -1624,7 +1625,6 @@ static void mld_report_work(struct work_struct *work)

if (++cnt >= MLD_MAX_QUEUE) { rework = true; - schedule_delayed_work(&idev->mc_report_work, 0); break; } } @@ -1635,8 +1635,10 @@ static void mld_report_work(struct work_struct *work) __mld_report_work(skb); mutex_unlock(&idev->mc_lock);

- if (!rework) - in6_dev_put(idev); + if (rework && queue_delayed_work(mld_wq, &idev->mc_report_work, 0)) + return; + + in6_dev_put(idev); }

static bool is_in(struct ifmcaddr6 *pmc, struct ip6_sf_list *psf, int type,

From: Kuniyuki Iwashima kuniyu@amazon.com

[ Upstream commit 02739545951ad4c1215160db7fbf9b7a918d3c0b ]

While reading these sysctl variables, they can be changed concurrently. Thus, we need to add READ_ONCE() to their readers.

- .sysctl_rmem - .sysctl_rwmem - .sysctl_rmem_offset - .sysctl_wmem_offset - sysctl_tcp_rmem[1, 2] - sysctl_tcp_wmem[1, 2] - sysctl_decnet_rmem[1] - sysctl_decnet_wmem[1] - sysctl_tipc_rmem[1]

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Kuniyuki Iwashima kuniyu@amazon.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- include/net/sock.h | 8 ++++---- net/decnet/af_decnet.c | 4 ++-- net/ipv4/tcp.c | 6 +++--- net/ipv4/tcp_input.c | 13 +++++++------ net/ipv4/tcp_output.c | 2 +- net/mptcp/protocol.c | 6 +++--- net/tipc/socket.c | 2 +- 7 files changed, 21 insertions(+), 20 deletions(-)

diff --git a/include/net/sock.h b/include/net/sock.h index 96f51d4b1649..819c53965ef3 100644 --- a/include/net/sock.h +++ b/include/net/sock.h @@ -2765,18 +2765,18 @@ static inline int sk_get_wmem0(const struct sock *sk, const struct proto *proto) { /* Does this proto have per netns sysctl_wmem ? */ if (proto->sysctl_wmem_offset) - return *(int *)((void *)sock_net(sk) + proto->sysctl_wmem_offset); + return READ_ONCE(*(int *)((void *)sock_net(sk) + proto->sysctl_wmem_offset));

- return *proto->sysctl_wmem; + return READ_ONCE(*proto->sysctl_wmem); }

static inline int sk_get_rmem0(const struct sock *sk, const struct proto *proto) { /* Does this proto have per netns sysctl_rmem ? */ if (proto->sysctl_rmem_offset) - return *(int *)((void *)sock_net(sk) + proto->sysctl_rmem_offset); + return READ_ONCE(*(int *)((void *)sock_net(sk) + proto->sysctl_rmem_offset));

- return *proto->sysctl_rmem; + return READ_ONCE(*proto->sysctl_rmem); }

/* Default TCP Small queue budget is ~1 ms of data (1sec >> 10) diff --git a/net/decnet/af_decnet.c b/net/decnet/af_decnet.c index dc92a67baea3..7d542eb46172 100644 --- a/net/decnet/af_decnet.c +++ b/net/decnet/af_decnet.c @@ -480,8 +480,8 @@ static struct sock *dn_alloc_sock(struct net *net, struct socket *sock, gfp_t gf sk->sk_family = PF_DECnet; sk->sk_protocol = 0; sk->sk_allocation = gfp; - sk->sk_sndbuf = sysctl_decnet_wmem[1]; - sk->sk_rcvbuf = sysctl_decnet_rmem[1]; + sk->sk_sndbuf = READ_ONCE(sysctl_decnet_wmem[1]); + sk->sk_rcvbuf = READ_ONCE(sysctl_decnet_rmem[1]);

/* Initialization of DECnet Session Control Port */ scp = DN_SK(sk); diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c index 7ba9059c263a..2097eeaf30a6 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -458,8 +458,8 @@ void tcp_init_sock(struct sock *sk)

icsk->icsk_sync_mss = tcp_sync_mss;

- WRITE_ONCE(sk->sk_sndbuf, sock_net(sk)->ipv4.sysctl_tcp_wmem[1]); - WRITE_ONCE(sk->sk_rcvbuf, sock_net(sk)->ipv4.sysctl_tcp_rmem[1]); + WRITE_ONCE(sk->sk_sndbuf, READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_wmem[1])); + WRITE_ONCE(sk->sk_rcvbuf, READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[1]));

sk_sockets_allocated_inc(sk); sk->sk_route_forced_caps = NETIF_F_GSO; @@ -1722,7 +1722,7 @@ int tcp_set_rcvlowat(struct sock *sk, int val) if (sk->sk_userlocks & SOCK_RCVBUF_LOCK) cap = sk->sk_rcvbuf >> 1; else - cap = sock_net(sk)->ipv4.sysctl_tcp_rmem[2] >> 1; + cap = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[2]) >> 1; val = min(val, cap); WRITE_ONCE(sk->sk_rcvlowat, val ? : 1);

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 566745f527fe..e007bdc20e82 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -426,7 +426,7 @@ static void tcp_sndbuf_expand(struct sock *sk)

if (sk->sk_sndbuf < sndmem) WRITE_ONCE(sk->sk_sndbuf, - min(sndmem, sock_net(sk)->ipv4.sysctl_tcp_wmem[2])); + min(sndmem, READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_wmem[2]))); }

/* 2. Tuning advertised window (window_clamp, rcv_ssthresh) @@ -461,7 +461,7 @@ static int __tcp_grow_window(const struct sock *sk, const struct sk_buff *skb, struct tcp_sock *tp = tcp_sk(sk); /* Optimize this! */ int truesize = tcp_win_from_space(sk, skbtruesize) >> 1; - int window = tcp_win_from_space(sk, sock_net(sk)->ipv4.sysctl_tcp_rmem[2]) >> 1; + int window = tcp_win_from_space(sk, READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[2])) >> 1;

while (tp->rcv_ssthresh <= window) { if (truesize <= skb->len) @@ -566,16 +566,17 @@ static void tcp_clamp_window(struct sock *sk) struct tcp_sock *tp = tcp_sk(sk); struct inet_connection_sock *icsk = inet_csk(sk); struct net *net = sock_net(sk); + int rmem2;

icsk->icsk_ack.quick = 0; + rmem2 = READ_ONCE(net->ipv4.sysctl_tcp_rmem[2]);

- if (sk->sk_rcvbuf < net->ipv4.sysctl_tcp_rmem[2] && + if (sk->sk_rcvbuf < rmem2 && !(sk->sk_userlocks & SOCK_RCVBUF_LOCK) && !tcp_under_memory_pressure(sk) && sk_memory_allocated(sk) < sk_prot_mem_limits(sk, 0)) { WRITE_ONCE(sk->sk_rcvbuf, - min(atomic_read(&sk->sk_rmem_alloc), - net->ipv4.sysctl_tcp_rmem[2])); + min(atomic_read(&sk->sk_rmem_alloc), rmem2)); } if (atomic_read(&sk->sk_rmem_alloc) > sk->sk_rcvbuf) tp->rcv_ssthresh = min(tp->window_clamp, 2U * tp->advmss); @@ -737,7 +738,7 @@ void tcp_rcv_space_adjust(struct sock *sk)

do_div(rcvwin, tp->advmss); rcvbuf = min_t(u64, rcvwin * rcvmem, - sock_net(sk)->ipv4.sysctl_tcp_rmem[2]); + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[2])); if (rcvbuf > sk->sk_rcvbuf) { WRITE_ONCE(sk->sk_rcvbuf, rcvbuf);

diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 51f31311fdb6..9c9a0f7a3dee 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -238,7 +238,7 @@ void tcp_select_initial_window(const struct sock *sk, int __space, __u32 mss, *rcv_wscale = 0; if (wscale_ok) { /* Set window scaling on max possible window */ - space = max_t(u32, space, sock_net(sk)->ipv4.sysctl_tcp_rmem[2]); + space = max_t(u32, space, READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[2])); space = max_t(u32, space, sysctl_rmem_max); space = min_t(u32, space, *window_clamp); *rcv_wscale = clamp_t(int, ilog2(space) - 15, diff --git a/net/mptcp/protocol.c b/net/mptcp/protocol.c index 01ede89e3c46..7f96e0c42a09 100644 --- a/net/mptcp/protocol.c +++ b/net/mptcp/protocol.c @@ -1899,7 +1899,7 @@ static void mptcp_rcv_space_adjust(struct mptcp_sock *msk, int copied)

do_div(rcvwin, advmss); rcvbuf = min_t(u64, rcvwin * rcvmem, - sock_net(sk)->ipv4.sysctl_tcp_rmem[2]); + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[2]));

if (rcvbuf > sk->sk_rcvbuf) { u32 window_clamp; @@ -2532,8 +2532,8 @@ static int mptcp_init_sock(struct sock *sk) icsk->icsk_ca_ops = NULL;

sk_sockets_allocated_inc(sk); - sk->sk_rcvbuf = sock_net(sk)->ipv4.sysctl_tcp_rmem[1]; - sk->sk_sndbuf = sock_net(sk)->ipv4.sysctl_tcp_wmem[1]; + sk->sk_rcvbuf = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_rmem[1]); + sk->sk_sndbuf = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_wmem[1]);

return 0; } diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 43509c7e90fc..f1c3b8eb4b3d 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -517,7 +517,7 @@ static int tipc_sk_create(struct net *net, struct socket *sock, timer_setup(&sk->sk_timer, tipc_sk_timeout, 0); sk->sk_shutdown = 0; sk->sk_backlog_rcv = tipc_sk_backlog_rcv; - sk->sk_rcvbuf = sysctl_tipc_rmem[1]; + sk->sk_rcvbuf = READ_ONCE(sysctl_tipc_rmem[1]); sk->sk_data_ready = tipc_data_ready; sk->sk_write_space = tipc_write_space; sk->sk_destruct = tipc_sock_destruct;

From: Kuniyuki Iwashima kuniyu@amazon.com

[ Upstream commit 22396941a7f343d704738360f9ef0e6576489d43 ]

While reading sysctl_tcp_comp_sack_slack_ns, it can be changed concurrently. Thus, we need to add READ_ONCE() to its reader.

Fixes: a70437cc09a1 ("tcp: add hrtimer slack to sack compression") Signed-off-by: Kuniyuki Iwashima kuniyu@amazon.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- net/ipv4/tcp_input.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index 486ca1d5b436..7b593865b4ae 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -5517,7 +5517,7 @@ static void __tcp_ack_snd_check(struct sock *sk, int ofo_possible) rtt * (NSEC_PER_USEC >> 3)/20); sock_hold(sk); hrtimer_start_range_ns(&tp->compressed_ack_timer, ns_to_ktime(delay), - sock_net(sk)->ipv4.sysctl_tcp_comp_sack_slack_ns, + READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_comp_sack_slack_ns), HRTIMER_MODE_REL_PINNED_SOFT); }

From: Kuniyuki Iwashima kuniyu@amazon.com

[ Upstream commit 870e3a634b6a6cb1543b359007aca73fe6a03ac5 ]

While reading sysctl_tcp_reflect_tos, it can be changed concurrently. Thus, we need to add READ_ONCE() to its readers.

Fixes: ac8f1710c12b ("tcp: reflect tos value received in SYN to the socket") Signed-off-by: Kuniyuki Iwashima kuniyu@amazon.com Acked-by: Wei Wang weiwan@google.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- net/ipv4/tcp_ipv4.c | 4 ++-- net/ipv6/tcp_ipv6.c | 4 ++-- 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index fba02cf6b468..dae0776c4948 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1004,7 +1004,7 @@ static int tcp_v4_send_synack(const struct sock *sk, struct dst_entry *dst, if (skb) { __tcp_v4_send_check(skb, ireq->ir_loc_addr, ireq->ir_rmt_addr);

- tos = sock_net(sk)->ipv4.sysctl_tcp_reflect_tos ? + tos = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_reflect_tos) ? (tcp_rsk(req)->syn_tos & ~INET_ECN_MASK) | (inet_sk(sk)->tos & INET_ECN_MASK) : inet_sk(sk)->tos; @@ -1590,7 +1590,7 @@ struct sock *tcp_v4_syn_recv_sock(const struct sock *sk, struct sk_buff *skb, /* Set ToS of the new socket based upon the value of incoming SYN. * ECT bits are set later in tcp_init_transfer(). */ - if (sock_net(sk)->ipv4.sysctl_tcp_reflect_tos) + if (READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_reflect_tos)) newinet->tos = tcp_rsk(req)->syn_tos & ~INET_ECN_MASK;

if (!dst) { diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index beaa0c2ada23..8ab39cf57d43 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -542,7 +542,7 @@ static int tcp_v6_send_synack(const struct sock *sk, struct dst_entry *dst, if (np->repflow && ireq->pktopts) fl6->flowlabel = ip6_flowlabel(ipv6_hdr(ireq->pktopts));

- tclass = sock_net(sk)->ipv4.sysctl_tcp_reflect_tos ? + tclass = READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_reflect_tos) ? (tcp_rsk(req)->syn_tos & ~INET_ECN_MASK) | (np->tclass & INET_ECN_MASK) : np->tclass; @@ -1364,7 +1364,7 @@ static struct sock *tcp_v6_syn_recv_sock(const struct sock *sk, struct sk_buff * /* Set ToS of the new socket based upon the value of incoming SYN. * ECT bits are set later in tcp_init_transfer(). */ - if (sock_net(sk)->ipv4.sysctl_tcp_reflect_tos) + if (READ_ONCE(sock_net(sk)->ipv4.sysctl_tcp_reflect_tos)) newnp->tclass = tcp_rsk(req)->syn_tos & ~INET_ECN_MASK;

/* Clone native IPv6 options from listening socket (if any)

From: Michal Maloszewski michal.maloszewski@intel.com

[ Upstream commit 5fcbb711024aac6d4db385623e6f2fdf019f7782 ]

Fix the inability to bring an interface up on a setup with only MSI interrupts enabled (no MSI-X). Solution is to add a default number of QPs = 1. This is enough, since without MSI-X support driver enables only a basic feature set.

Fixes: bc6d33c8d93f ("i40e: Fix the number of queues available to be mapped for use") Signed-off-by: Dawid Lukwinski dawid.lukwinski@intel.com Signed-off-by: Michal Maloszewski michal.maloszewski@intel.com Tested-by: Dave Switzer david.switzer@intel.com Signed-off-by: Tony Nguyen anthony.l.nguyen@intel.com Link: https://lore.kernel.org/r/20220722175401.112572-1-anthony.l.nguyen@intel.com Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/intel/i40e/i40e_main.c | 4 ++++ 1 file changed, 4 insertions(+)

diff --git a/drivers/net/ethernet/intel/i40e/i40e_main.c b/drivers/net/ethernet/intel/i40e/i40e_main.c index c801b128e5b2..b07d55c99317 100644 --- a/drivers/net/ethernet/intel/i40e/i40e_main.c +++ b/drivers/net/ethernet/intel/i40e/i40e_main.c @@ -1908,11 +1908,15 @@ static void i40e_vsi_setup_queue_map(struct i40e_vsi *vsi, * non-zero req_queue_pairs says that user requested a new * queue count via ethtool's set_channels, so use this * value for queues distribution across traffic classes + * We need at least one queue pair for the interface + * to be usable as we see in else statement. */ if (vsi->req_queue_pairs > 0) vsi->num_queue_pairs = vsi->req_queue_pairs; else if (pf->flags & I40E_FLAG_MSIX_ENABLED) vsi->num_queue_pairs = pf->num_lan_msix; + else + vsi->num_queue_pairs = 1; }

/* Number of queues per enabled TC */

From: Sunil Goutham sgoutham@marvell.com

[ Upstream commit b354eaeec8637d87003945439209251d76a2bb95 ]

NIX_AF_TLXX_PIR/CIR register format has changed from OcteonTx2 to CN10K. CN10K supports larger burst size. Fix burst exponent and burst mantissa configuration for CN10K.

Also fixed 'maxrate' from u32 to u64 since 'police.rate_bytes_ps' passed by stack is also u64.

Fixes: e638a83f167e ("octeontx2-pf: TC_MATCHALL egress ratelimiting offload") Signed-off-by: Sunil Goutham sgoutham@marvell.com Signed-off-by: Subbaraya Sundeep sbhatta@marvell.com Signed-off-by: Paolo Abeni pabeni@redhat.com Signed-off-by: Sasha Levin sashal@kernel.org --- .../ethernet/marvell/octeontx2/nic/otx2_tc.c | 76 ++++++++++++++----- 1 file changed, 55 insertions(+), 21 deletions(-)

diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_tc.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_tc.c index 626961a41089..ff569e261be4 100644 --- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_tc.c +++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_tc.c @@ -28,6 +28,9 @@ #define MAX_RATE_EXPONENT 0x0FULL #define MAX_RATE_MANTISSA 0xFFULL

+#define CN10K_MAX_BURST_MANTISSA 0x7FFFULL +#define CN10K_MAX_BURST_SIZE 8453888ULL + /* Bitfields in NIX_TLX_PIR register */ #define TLX_RATE_MANTISSA GENMASK_ULL(8, 1) #define TLX_RATE_EXPONENT GENMASK_ULL(12, 9) @@ -35,6 +38,9 @@ #define TLX_BURST_MANTISSA GENMASK_ULL(36, 29) #define TLX_BURST_EXPONENT GENMASK_ULL(40, 37)

+#define CN10K_TLX_BURST_MANTISSA GENMASK_ULL(43, 29) +#define CN10K_TLX_BURST_EXPONENT GENMASK_ULL(47, 44) + struct otx2_tc_flow_stats { u64 bytes; u64 pkts; @@ -77,33 +83,42 @@ int otx2_tc_alloc_ent_bitmap(struct otx2_nic *nic) } EXPORT_SYMBOL(otx2_tc_alloc_ent_bitmap);

-static void otx2_get_egress_burst_cfg(u32 burst, u32 *burst_exp, - u32 *burst_mantissa) +static void otx2_get_egress_burst_cfg(struct otx2_nic *nic, u32 burst, + u32 *burst_exp, u32 *burst_mantissa) { + int max_burst, max_mantissa; unsigned int tmp;

+ if (is_dev_otx2(nic->pdev)) { + max_burst = MAX_BURST_SIZE; + max_mantissa = MAX_BURST_MANTISSA; + } else { + max_burst = CN10K_MAX_BURST_SIZE; + max_mantissa = CN10K_MAX_BURST_MANTISSA; + } + /* Burst is calculated as * ((256 + BURST_MANTISSA) << (1 + BURST_EXPONENT)) / 256 * Max supported burst size is 130,816 bytes. */ - burst = min_t(u32, burst, MAX_BURST_SIZE); + burst = min_t(u32, burst, max_burst); if (burst) { *burst_exp = ilog2(burst) ? ilog2(burst) - 1 : 0; tmp = burst - rounddown_pow_of_two(burst); - if (burst < MAX_BURST_MANTISSA) + if (burst < max_mantissa) *burst_mantissa = tmp * 2; else *burst_mantissa = tmp / (1ULL << (*burst_exp - 7)); } else { *burst_exp = MAX_BURST_EXPONENT; - *burst_mantissa = MAX_BURST_MANTISSA; + *burst_mantissa = max_mantissa; } }

-static void otx2_get_egress_rate_cfg(u32 maxrate, u32 *exp, +static void otx2_get_egress_rate_cfg(u64 maxrate, u32 *exp, u32 *mantissa, u32 *div_exp) { - unsigned int tmp; + u64 tmp;

/* Rate calculation by hardware * @@ -132,21 +147,44 @@ static void otx2_get_egress_rate_cfg(u32 maxrate, u32 *exp, } }

-static int otx2_set_matchall_egress_rate(struct otx2_nic *nic, u32 burst, u32 maxrate) +static u64 otx2_get_txschq_rate_regval(struct otx2_nic *nic, + u64 maxrate, u32 burst) { - struct otx2_hw *hw = &nic->hw; - struct nix_txschq_config *req; u32 burst_exp, burst_mantissa; u32 exp, mantissa, div_exp; + u64 regval = 0; + + /* Get exponent and mantissa values from the desired rate */ + otx2_get_egress_burst_cfg(nic, burst, &burst_exp, &burst_mantissa); + otx2_get_egress_rate_cfg(maxrate, &exp, &mantissa, &div_exp); + + if (is_dev_otx2(nic->pdev)) { + regval = FIELD_PREP(TLX_BURST_EXPONENT, (u64)burst_exp) | + FIELD_PREP(TLX_BURST_MANTISSA, (u64)burst_mantissa) | + FIELD_PREP(TLX_RATE_DIVIDER_EXPONENT, div_exp) | + FIELD_PREP(TLX_RATE_EXPONENT, exp) | + FIELD_PREP(TLX_RATE_MANTISSA, mantissa) | BIT_ULL(0); + } else { + regval = FIELD_PREP(CN10K_TLX_BURST_EXPONENT, (u64)burst_exp) | + FIELD_PREP(CN10K_TLX_BURST_MANTISSA, (u64)burst_mantissa) | + FIELD_PREP(TLX_RATE_DIVIDER_EXPONENT, div_exp) | + FIELD_PREP(TLX_RATE_EXPONENT, exp) | + FIELD_PREP(TLX_RATE_MANTISSA, mantissa) | BIT_ULL(0); + } + + return regval; +} + +static int otx2_set_matchall_egress_rate(struct otx2_nic *nic, + u32 burst, u64 maxrate) +{ + struct otx2_hw *hw = &nic->hw; + struct nix_txschq_config *req; int txschq, err;

/* All SQs share the same TL4, so pick the first scheduler */ txschq = hw->txschq_list[NIX_TXSCH_LVL_TL4][0];

- /* Get exponent and mantissa values from the desired rate */ - otx2_get_egress_burst_cfg(burst, &burst_exp, &burst_mantissa); - otx2_get_egress_rate_cfg(maxrate, &exp, &mantissa, &div_exp); - mutex_lock(&nic->mbox.lock); req = otx2_mbox_alloc_msg_nix_txschq_cfg(&nic->mbox); if (!req) { @@ -157,11 +195,7 @@ static int otx2_set_matchall_egress_rate(struct otx2_nic *nic, u32 burst, u32 ma req->lvl = NIX_TXSCH_LVL_TL4; req->num_regs = 1; req->reg[0] = NIX_AF_TL4X_PIR(txschq); - req->regval[0] = FIELD_PREP(TLX_BURST_EXPONENT, burst_exp) | - FIELD_PREP(TLX_BURST_MANTISSA, burst_mantissa) | - FIELD_PREP(TLX_RATE_DIVIDER_EXPONENT, div_exp) | - FIELD_PREP(TLX_RATE_EXPONENT, exp) | - FIELD_PREP(TLX_RATE_MANTISSA, mantissa) | BIT_ULL(0); + req->regval[0] = otx2_get_txschq_rate_regval(nic, maxrate, burst);

err = otx2_sync_mbox_msg(&nic->mbox); mutex_unlock(&nic->mbox.lock); @@ -196,7 +230,7 @@ static int otx2_tc_egress_matchall_install(struct otx2_nic *nic, struct netlink_ext_ack *extack = cls->common.extack; struct flow_action *actions = &cls->rule->action; struct flow_action_entry *entry; - u32 rate; + u64 rate; int err;

err = otx2_tc_validate_flow(nic, actions, extack); @@ -218,7 +252,7 @@ static int otx2_tc_egress_matchall_install(struct otx2_nic *nic, } /* Convert bytes per second to Mbps */ rate = entry->police.rate_bytes_ps * 8; - rate = max_t(u32, rate / 1000000, 1); + rate = max_t(u64, rate / 1000000, 1); err = otx2_set_matchall_egress_rate(nic, entry->police.burst, rate); if (err) return err;

From: Jason Wang jasowang@redhat.com

[ Upstream commit 5a159128faff151b7fe5f4eb0f310b1e0a2d56bf ]

We try using cancel_delayed_work_sync() to prevent the work from enabling NAPI. This is insufficient since we don't disable the source of the refill work scheduling. This means an NAPI poll callback after cancel_delayed_work_sync() can schedule the refill work then can re-enable the NAPI that leads to use-after-free [1].

Since the work can enable NAPI, we can't simply disable NAPI before calling cancel_delayed_work_sync(). So fix this by introducing a dedicated boolean to control whether or not the work could be scheduled from NAPI.

[1] ================================================================== BUG: KASAN: use-after-free in refill_work+0x43/0xd4 Read of size 2 at addr ffff88810562c92e by task kworker/2:1/42

CPU: 2 PID: 42 Comm: kworker/2:1 Not tainted 5.19.0-rc1+ #480 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014 Workqueue: events refill_work Call Trace: <TASK> dump_stack_lvl+0x34/0x44 print_report.cold+0xbb/0x6ac ? _printk+0xad/0xde ? refill_work+0x43/0xd4 kasan_report+0xa8/0x130 ? refill_work+0x43/0xd4 refill_work+0x43/0xd4 process_one_work+0x43d/0x780 worker_thread+0x2a0/0x6f0 ? process_one_work+0x780/0x780 kthread+0x167/0x1a0 ? kthread_exit+0x50/0x50 ret_from_fork+0x22/0x30 </TASK> ...

Fixes: b2baed69e605c ("virtio_net: set/cancel work on ndo_open/ndo_stop") Signed-off-by: Jason Wang jasowang@redhat.com Acked-by: Michael S. Tsirkin mst@redhat.com Reviewed-by: Xuan Zhuo xuanzhuo@linux.alibaba.com Signed-off-by: David S. Miller davem@davemloft.net Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/virtio_net.c | 37 ++++++++++++++++++++++++++++++++++--- 1 file changed, 34 insertions(+), 3 deletions(-)

--- a/drivers/net/virtio_net.c +++ b/drivers/net/virtio_net.c @@ -213,9 +213,15 @@ struct virtnet_info { /* Packet virtio header size */ u8 hdr_len;

- /* Work struct for refilling if we run low on memory. */ + /* Work struct for delayed refilling if we run low on memory. */ struct delayed_work refill;

+ /* Is delayed refill enabled? */ + bool refill_enabled; + + /* The lock to synchronize the access to refill_enabled */ + spinlock_t refill_lock; + /* Work struct for config space updates */ struct work_struct config_work;

@@ -319,6 +325,20 @@ static struct page *get_a_page(struct re return p; }

+static void enable_delayed_refill(struct virtnet_info *vi) +{ + spin_lock_bh(&vi->refill_lock); + vi->refill_enabled = true; + spin_unlock_bh(&vi->refill_lock); +} + +static void disable_delayed_refill(struct virtnet_info *vi) +{ + spin_lock_bh(&vi->refill_lock); + vi->refill_enabled = false; + spin_unlock_bh(&vi->refill_lock); +} + static void virtqueue_napi_schedule(struct napi_struct *napi, struct virtqueue *vq) { @@ -1454,8 +1474,12 @@ static int virtnet_receive(struct receiv }

if (rq->vq->num_free > min((unsigned int)budget, virtqueue_get_vring_size(rq->vq)) / 2) { - if (!try_fill_recv(vi, rq, GFP_ATOMIC)) - schedule_delayed_work(&vi->refill, 0); + if (!try_fill_recv(vi, rq, GFP_ATOMIC)) { + spin_lock(&vi->refill_lock); + if (vi->refill_enabled) + schedule_delayed_work(&vi->refill, 0); + spin_unlock(&vi->refill_lock); + } }

u64_stats_update_begin(&rq->stats.syncp); @@ -1578,6 +1602,8 @@ static int virtnet_open(struct net_devic struct virtnet_info *vi = netdev_priv(dev); int i, err;

+ enable_delayed_refill(vi); + for (i = 0; i < vi->max_queue_pairs; i++) { if (i < vi->curr_queue_pairs) /* Make sure we have some buffers: if oom use wq. */ @@ -1958,6 +1984,8 @@ static int virtnet_close(struct net_devi struct virtnet_info *vi = netdev_priv(dev); int i;

+ /* Make sure NAPI doesn't schedule refill work */ + disable_delayed_refill(vi); /* Make sure refill_work doesn't re-enable napi! */ cancel_delayed_work_sync(&vi->refill);

@@ -2455,6 +2483,8 @@ static int virtnet_restore_up(struct vir

virtio_device_ready(vdev);

+ enable_delayed_refill(vi); + if (netif_running(vi->dev)) { err = virtnet_open(vi->dev); if (err) @@ -3162,6 +3192,7 @@ static int virtnet_probe(struct virtio_d vdev->priv = vi;

INIT_WORK(&vi->config_work, virtnet_config_changed_work); + spin_lock_init(&vi->refill_lock);

/* If we can receive ANY GSO packets, we must allocate large ones. */ if (virtio_has_feature(vdev, VIRTIO_NET_F_GUEST_TSO4) ||

From: Alejandro Lucero alejandro.lucero-palau@amd.com

[ Upstream commit 67c3b611d92fc238c43734878bc3e232ab570c79 ]

Sending a PTP packet can imply to use the normal TX driver datapath but invoked from the driver's ptp worker. The kernel generic TX code disables softirqs and preemption before calling specific driver TX code, but the ptp worker does not. Although current ptp driver functionality does not require it, there are several reasons for doing so:

1) The invoked code is always executed with softirqs disabled for non PTP packets. 2) Better if a ptp packet transmission is not interrupted by softirq handling which could lead to high latencies. 3) netdev_xmit_more used by the TX code requires preemption to be disabled.

Indeed a solution for dealing with kernel preemption state based on static kernel configuration is not possible since the introduction of dynamic preemption level configuration at boot time using the static calls functionality.

Fixes: f79c957a0b537 ("drivers: net: sfc: use netdev_xmit_more helper") Signed-off-by: Alejandro Lucero alejandro.lucero-palau@amd.com Link: https://lore.kernel.org/r/20220726064504.49613-1-alejandro.lucero-palau@amd.... Signed-off-by: Jakub Kicinski kuba@kernel.org Signed-off-by: Sasha Levin sashal@kernel.org --- drivers/net/ethernet/sfc/ptp.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+)

diff --git a/drivers/net/ethernet/sfc/ptp.c b/drivers/net/ethernet/sfc/ptp.c index 725b0f38813a..a2b4e3befa59 100644 --- a/drivers/net/ethernet/sfc/ptp.c +++ b/drivers/net/ethernet/sfc/ptp.c @@ -1100,7 +1100,29 @@ static void efx_ptp_xmit_skb_queue(struct efx_nic *efx, struct sk_buff *skb)

tx_queue = efx_channel_get_tx_queue(ptp_data->channel, type); if (tx_queue && tx_queue->timestamping) { + /* This code invokes normal driver TX code which is always + * protected from softirqs when called from generic TX code, + * which in turn disables preemption. Look at __dev_queue_xmit + * which uses rcu_read_lock_bh disabling preemption for RCU + * plus disabling softirqs. We do not need RCU reader + * protection here. + * + * Although it is theoretically safe for current PTP TX/RX code + * running without disabling softirqs, there are three good + * reasond for doing so: + * + * 1) The code invoked is mainly implemented for non-PTP + * packets and it is always executed with softirqs + * disabled. + * 2) This being a single PTP packet, better to not + * interrupt its processing by softirqs which can lead + * to high latencies. + * 3) netdev_xmit_more checks preemption is disabled and + * triggers a BUG_ON if not. + */ + local_bh_disable(); efx_enqueue_skb(tx_queue, skb); + local_bh_enable(); } else { WARN_ONCE(1, "PTP channel has no timestamped tx queue\n"); dev_kfree_skb_any(skb);

From: Greg Kroah-Hartman gregkh@linuxfoundation.org

The gcc build warning prevents all clang-built kernels from working properly, so comment it out to fix the build.

This is a -stable kernel only patch for now, it will be resolved differently in mainline releases in the future.

Cc: "Jason A. Donenfeld" Jason@zx2c4.com Cc: "Justin M. Forbes" jforbes@fedoraproject.org Cc: Ard Biesheuvel ardb@kernel.org Acked-by: Arnd Bergmann arnd@arndb.de Cc: Nicolas Pitre nico@linaro.org Cc: Nathan Chancellor nathan@kernel.org Cc: Nick Desaulniers ndesaulniers@google.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- arch/arm/lib/xor-neon.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)

--- a/arch/arm/lib/xor-neon.c +++ b/arch/arm/lib/xor-neon.c @@ -26,8 +26,9 @@ MODULE_LICENSE("GPL"); * While older versions of GCC do not generate incorrect code, they fail to * recognize the parallel nature of these functions, and emit plain ARM code, * which is known to be slower than the optimized ARM code in asm-arm/xor.h. + * + * #warning This code requires at least version 4.6 of GCC */ -#warning This code requires at least version 4.6 of GCC #endif

#pragma GCC diagnostic ignored "-Wunused-variable"

From: Jaewon Kim jaewon31.kim@samsung.com

commit 9282012fc0aa248b77a69f5eb802b67c5a16bb13 upstream.

There was a report that a task is waiting at the throttle_direct_reclaim. The pgscan_direct_throttle in vmstat was increasing.

This is a bug where zone_watermark_fast returns true even when the free is very low. The commit f27ce0e14088 ("page_alloc: consider highatomic reserve in watermark fast") changed the watermark fast to consider highatomic reserve. But it did not handle a negative value case which can be happened when reserved_highatomic pageblock is bigger than the actual free.

If watermark is considered as ok for the negative value, allocating contexts for order-0 will consume all free pages without direct reclaim, and finally free page may become depleted except highatomic free.

Then allocating contexts may fall into throttle_direct_reclaim. This symptom may easily happen in a system where wmark min is low and other reclaimers like kswapd does not make free pages quickly.

Handle the negative case by using MIN.

Link: https://lkml.kernel.org/r/20220725095212.25388-1-jaewon31.kim@samsung.com Fixes: f27ce0e14088 ("page_alloc: consider highatomic reserve in watermark fast") Signed-off-by: Jaewon Kim jaewon31.kim@samsung.com Reported-by: GyeongHwan Hong gh21.hong@samsung.com Acked-by: Mel Gorman mgorman@techsingularity.net Cc: Minchan Kim minchan@kernel.org Cc: Baoquan He bhe@redhat.com Cc: Vlastimil Babka vbabka@suse.cz Cc: Johannes Weiner hannes@cmpxchg.org Cc: Michal Hocko mhocko@kernel.org Cc: Yong-Taek Lee ytk.lee@samsung.com Cc: stable@vger.kerenl.org Signed-off-by: Andrew Morton akpm@linux-foundation.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- mm/page_alloc.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-)

--- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -3928,11 +3928,15 @@ static inline bool zone_watermark_fast(s * need to be calculated. */ if (!order) { - long fast_free; + long usable_free; + long reserved;

- fast_free = free_pages; - fast_free -= __zone_watermark_unusable_free(z, 0, alloc_flags); - if (fast_free > mark + z->lowmem_reserve[highest_zoneidx]) + usable_free = free_pages; + reserved = __zone_watermark_unusable_free(z, 0, alloc_flags); + + /* reserved may over estimate high-atomic reserves. */ + usable_free -= min(usable_free, reserved); + if (usable_free > mark + z->lowmem_reserve[highest_zoneidx]) return true; }

From: Toshi Kani toshi.kani@hpe.com

commit 5e2805d5379619c4a2e3ae4994e73b36439f4bad upstream.

The commit

cb51a371d08e ("EDAC/ghes: Setup DIMM label from DMI and use it in error reports")

enforced that both the bank and device strings passed to dimm_setup_label() are not NULL.

However, there are BIOSes, for example on a

HPE ProLiant DL360 Gen10/ProLiant DL360 Gen10, BIOS U32 03/15/2019

which don't populate both strings:

Handle 0x0020, DMI type 17, 84 bytes Memory Device Array Handle: 0x0013 Error Information Handle: Not Provided Total Width: 72 bits Data Width: 64 bits Size: 32 GB Form Factor: DIMM Set: None Locator: PROC 1 DIMM 1 <===== device Bank Locator: Not Specified <===== bank

This results in a buffer overflow because ghes_edac_register() calls strlen() on an uninitialized label, which had non-zero values left over from krealloc_array():

detected buffer overflow in __fortify_strlen ------------[ cut here ]------------ kernel BUG at lib/string_helpers.c:983! invalid opcode: 0000 [#1] PREEMPT SMP NOPTI CPU: 1 PID: 1 Comm: swapper/0 Tainted: G I 5.18.6-200.fc36.x86_64 #1 Hardware name: HPE ProLiant DL360 Gen10/ProLiant DL360 Gen10, BIOS U32 03/15/2019 RIP: 0010:fortify_panic ... Call Trace: <TASK> ghes_edac_register.cold ghes_probe platform_probe really_probe __driver_probe_device driver_probe_device __driver_attach ? __device_attach_driver bus_for_each_dev bus_add_driver driver_register acpi_ghes_init acpi_init ? acpi_sleep_proc_init do_one_initcall

The label contains garbage because the commit in Fixes reallocs the DIMMs array while scanning the system but doesn't clear the newly allocated memory.

Change dimm_setup_label() to always initialize the label to fix the issue. Set it to the empty string in case BIOS does not provide both bank and device so that ghes_edac_register() can keep the default label given by edac_mc_alloc_dimms().

[ bp: Rewrite commit message. ]

Fixes: b9cae27728d1f ("EDAC/ghes: Scan the system once on driver init") Co-developed-by: Robert Richter rric@kernel.org Signed-off-by: Robert Richter rric@kernel.org Signed-off-by: Toshi Kani toshi.kani@hpe.com Signed-off-by: Borislav Petkov bp@suse.de Tested-by: Robert Elliott elliott@hpe.com Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20220719220124.760359-1-toshi.kani@hpe.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- drivers/edac/ghes_edac.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-)

--- a/drivers/edac/ghes_edac.c +++ b/drivers/edac/ghes_edac.c @@ -101,9 +101,14 @@ static void dimm_setup_label(struct dimm

dmi_memdev_name(handle, &bank, &device);

- /* both strings must be non-zero */ - if (bank && *bank && device && *device) - snprintf(dimm->label, sizeof(dimm->label), "%s %s", bank, device); + /* + * Set to a NULL string when both bank and device are zero. In this case, + * the label assigned by default will be preserved. + */ + snprintf(dimm->label, sizeof(dimm->label), "%s%s%s", + (bank && *bank) ? bank : "", + (bank && *bank && device && *device) ? " " : "", + (device && *device) ? device : ""); }

static void assign_dmi_dimm_info(struct dimm_info *dimm, struct memdev_dmi_entry *entry)

From: Waiman Long longman@redhat.com

commit 6eebd5fb20838f5971ba17df9f55cc4f84a31053 upstream.

With commit d257cc8cb8d5 ("locking/rwsem: Make handoff bit handling more consistent"), the writer that sets the handoff bit can be interrupted out without clearing the bit if the wait queue isn't empty. This disables reader and writer optimistic lock spinning and stealing.

Now if a non-first writer in the queue is somehow woken up or a new waiter enters the slowpath, it can't acquire the lock. This is not the case before commit d257cc8cb8d5 as the writer that set the handoff bit will clear it when exiting out via the out_nolock path. This is less efficient as the busy rwsem stays in an unlock state for a longer time.

In some cases, this new behavior may cause lockups as shown in [1] and [2].

This patch allows a non-first writer to ignore the handoff bit if it is not originally set or initiated by the first waiter. This patch is shown to be effective in fixing the lockup problem reported in [1].

[1] https://lore.kernel.org/lkml/20220617134325.GC30825@techsingularity.net/ [2] https://lore.kernel.org/lkml/3f02975c-1a9d-be20-32cf-f1d8e3dfafcc@oracle.com...

Fixes: d257cc8cb8d5 ("locking/rwsem: Make handoff bit handling more consistent") Signed-off-by: Waiman Long longman@redhat.com Signed-off-by: Peter Zijlstra (Intel) peterz@infradead.org Acked-by: John Donnelly john.p.donnelly@oracle.com Tested-by: Mel Gorman mgorman@techsingularity.net Link: https://lore.kernel.org/r/20220622200419.778799-1-longman@redhat.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- kernel/locking/rwsem.c | 30 ++++++++++++++++++++---------- 1 file changed, 20 insertions(+), 10 deletions(-)

--- a/kernel/locking/rwsem.c +++ b/kernel/locking/rwsem.c @@ -335,8 +335,6 @@ struct rwsem_waiter { struct task_struct *task; enum rwsem_waiter_type type; unsigned long timeout; - - /* Writer only, not initialized in reader */ bool handoff_set; }; #define rwsem_first_waiter(sem) \ @@ -456,10 +454,12 @@ static void rwsem_mark_wake(struct rw_se * to give up the lock), request a HANDOFF to * force the issue. */ - if (!(oldcount & RWSEM_FLAG_HANDOFF) && - time_after(jiffies, waiter->timeout)) { - adjustment -= RWSEM_FLAG_HANDOFF; - lockevent_inc(rwsem_rlock_handoff); + if (time_after(jiffies, waiter->timeout)) { + if (!(oldcount & RWSEM_FLAG_HANDOFF)) { + adjustment -= RWSEM_FLAG_HANDOFF; + lockevent_inc(rwsem_rlock_handoff); + } + waiter->handoff_set = true; }

atomic_long_add(-adjustment, &sem->count); @@ -569,7 +569,7 @@ static void rwsem_mark_wake(struct rw_se static inline bool rwsem_try_write_lock(struct rw_semaphore *sem, struct rwsem_waiter *waiter) { - bool first = rwsem_first_waiter(sem) == waiter; + struct rwsem_waiter *first = rwsem_first_waiter(sem); long count, new;

lockdep_assert_held(&sem->wait_lock); @@ -579,11 +579,20 @@ static inline bool rwsem_try_write_lock( bool has_handoff = !!(count & RWSEM_FLAG_HANDOFF);

if (has_handoff) { - if (!first) + /* + * Honor handoff bit and yield only when the first + * waiter is the one that set it. Otherwisee, we + * still try to acquire the rwsem. + */ + if (first->handoff_set && (waiter != first)) return false;

- /* First waiter inherits a previously set handoff bit */ - waiter->handoff_set = true; + /* + * First waiter can inherit a previously set handoff + * bit and spin on rwsem if lock acquisition fails. + */ + if (waiter == first) + waiter->handoff_set = true; }

new = count; @@ -978,6 +987,7 @@ queue: waiter.task = current; waiter.type = RWSEM_WAITING_FOR_READ; waiter.timeout = jiffies + RWSEM_WAIT_TIMEOUT; + waiter.handoff_set = false;

raw_spin_lock_irq(&sem->wait_lock); if (list_empty(&sem->wait_list)) {