- Kernel - mailweb.openeuler.org

[PATCH OLK-6.6] netconsole: avoid OOB reads, msg is not nul-terminated
by Jinjiang Tu 10 Jun '26

10 Jun '26

From: Jakub Kicinski <kuba(a)kernel.org> mainline inclusion from mainline-v7.0-rc2 commit 82aec772fca2223bc5774bd9af486fd95766e578 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/14743 CVE: CVE-2026-43197 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… ------------------------------------------- msg passed to netconsole from the console subsystem is not guaranteed to be nul-terminated. Before recent commit 7eab73b18630 ("netconsole: convert to NBCON console infrastructure") the message would be placed in printk_shared_pbufs, a static global buffer, so KASAN had harder time catching OOB accesses. Now we see: printk: console [netcon_ext0] enabled BUG: KASAN: slab-out-of-bounds in string+0x1f7/0x240 Read of size 1 at addr ffff88813b6d4c00 by task pr/netcon_ext0/594 CPU: 65 UID: 0 PID: 594 Comm: pr/netcon_ext0 Not tainted 6.19.0-11754-g4246fd6547c9 Call Trace: kasan_report+0xe4/0x120 string+0x1f7/0x240 vsnprintf+0x655/0xba0 scnprintf+0xba/0x120 netconsole_write+0x3fe/0xa10 nbcon_emit_next_record+0x46e/0x860 nbcon_kthread_func+0x623/0x750 Allocated by task 1: nbcon_alloc+0x1ea/0x450 register_console+0x26b/0xe10 init_netconsole+0xbb0/0xda0 The buggy address belongs to the object at ffff88813b6d4000 which belongs to the cache kmalloc-4k of size 4096 The buggy address is located 0 bytes to the right of allocated 3072-byte region [ffff88813b6d4000, ffff88813b6d4c00) Fixes: c62c0a17f9b7 ("netconsole: Append kernel version to message") Signed-off-by: Jakub Kicinski <kuba(a)kernel.org> Reviewed-by: Simon Horman <horms(a)kernel.org> Link: https://patch.msgid.link/20260219195021.2099699-1-kuba@kernel.org Signed-off-by: Paolo Abeni <pabeni(a)redhat.com> Conflicts: drivers/net/netconsole.c [Context conflicts due to commit e7650d8d475c ("net: netconsole: split send_ext_msg_udp() function") isn't merged.] Signed-off-by: Jinjiang Tu <tujinjiang(a)huawei.com> --- drivers/net/netconsole.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c index fffffa3658d2..69d54f58aa97 100644 --- a/drivers/net/netconsole.c +++ b/drivers/net/netconsole.c @@ -854,7 +854,8 @@ static void send_ext_msg_udp(struct netconsole_target *nt, const char *msg, if (msg_len + release_len <= MAX_PRINT_CHUNK) { /* No fragmentation needed */ if (nt->release) { - scnprintf(buf, MAX_PRINT_CHUNK, "%s,%s", release, msg); + scnprintf(buf, MAX_PRINT_CHUNK, "%s,%.*s", release, + msg_len, msg); msg_len += release_len; msg_ready = buf; } -- 2.43.0

2 1

[PATCH OLK-6.6] bpf: fix UAF by restoring RCU-delayed inode freeing in bpffs
by Luo Gengkun 10 Jun '26

10 Jun '26

From: Deepanshu Kartikey <kartikey406(a)gmail.com> maillist inclusion category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/15537 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- commit 4f375ade6aa9 ("bpf: Avoid RCU context warning when unpinning htab with internal structs") moved inode cleanup from ->free_inode() into ->destroy_inode() to avoid sleeping in RCU context when calling bpf_any_put(). However this removed the RCU delay on freeing the inode itself and the cached symlink body (i_link), both of which can be accessed by RCU pathwalk (pick_link, may_lookup etc.). This causes a use-after-free when a concurrent unlinkat() drops the last inode reference and destroy_inode() frees the inode immediately, while another task is still walking the path in RCU mode and reads inode->i_opflags (offset +2) inside current_time() -> is_mgtime(). KASAN reports: BUG: KASAN: slab-use-after-free in is_mgtime include/linux/fs.h:2313 Read of size 2 at addr ffff8880407e4282 (offset +2 = i_opflags) The rules (per Al Viro): ->destroy_inode() called immediately, can sleep, use for blocking cleanup e.g. bpf_any_put() ->free_inode() called after RCU grace period, use for freeing inode and anything RCU-accessible e.g. i_link Fix: split the two concerns properly: - keep bpf_any_put() in bpf_destroy_inode() since it is blocking and needs to run promptly - introduce bpf_free_inode() to handle kfree(i_link) and free_inode_nonrcu() with proper RCU delay, preventing the UAF Fixes: 4f375ade6aa9 ("bpf: Avoid RCU context warning when unpinning htab with internal structs") Reported-by: syzbot+36e50496c8ac4bcde3f9(a)syzkaller.appspotmail.com Closes: https://syzkaller.appspot.com/bug?extid=36e50496c8ac4bcde3f9 Suggested-by: Al Viro <viro(a)zeniv.linux.org.uk> Link: https://lore.kernel.org/all/20260423043906.GN3518998@ZenIV/ Link: https://lore.kernel.org/all/20260602002607.110866-1-kartikey406@gmail.com/T/ [v1] Signed-off-by: Deepanshu Kartikey <kartikey406(a)gmail.com> Acked-by: Al Viro <viro(a)zeniv.linux.org.uk> Link: https://lore.kernel.org/r/20260602025249.113828-1-kartikey406@gmail.com Signed-off-by: Alexei Starovoitov <ast(a)kernel.org> Signed-off-by: Luo Gengkun <luogengkun2(a)huawei.com> --- kernel/bpf/inode.c | 13 +++++++++++-- 1 file changed, 11 insertions(+), 2 deletions(-) diff --git a/kernel/bpf/inode.c b/kernel/bpf/inode.c index 9a9630adcba4..99ec74325175 100644 --- a/kernel/bpf/inode.c +++ b/kernel/bpf/inode.c @@ -611,10 +611,18 @@ static void bpf_destroy_inode(struct inode *inode) { enum bpf_type type; - if (S_ISLNK(inode->i_mode)) - kfree(inode->i_link); if (!bpf_inode_type(inode, &type)) bpf_any_put(inode->i_private, type); +} + +/* + * Called after RCU grace period - safe to free inode and anything + * that might be accessed by RCU pathwalk (inode fields, i_link). + */ +static void bpf_free_inode(struct inode *inode) +{ + if (S_ISLNK(inode->i_mode)) + kfree(inode->i_link); free_inode_nonrcu(inode); } @@ -623,6 +631,7 @@ static const struct super_operations bpf_super_ops = { .drop_inode = generic_delete_inode, .show_options = bpf_show_options, .destroy_inode = bpf_destroy_inode, + .free_inode = bpf_free_inode, }; enum { -- 2.34.1

2 1

[PATCH OLK-5.10] netfilter: x_tables: ensure names are nul-terminated
by superdcc97＠163.com 10 Jun '26

10 Jun '26

From: Florian Westphal <fw(a)strlen.de> stable inclusion from stable-v5.10.253 commit bcac50ea0a29d430eedc5ac87b215393b567baa9 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/14560 CVE: CVE-2026-43028 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit a958a4f90ddd7de0800b33ca9d7b886b7d40f74e ] Reject names that lack a \0 character before feeding them to functions that expect c-strings. Fixes tag is the most recent commit that needs this change. Fixes: c38c4597e4bf ("netfilter: implement xt_cgroup cgroup2 path match") Signed-off-by: Florian Westphal <fw(a)strlen.de> Signed-off-by: Pablo Neira Ayuso <pablo(a)netfilter.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Dong Chenchen <dongchenchen2(a)huawei.com> --- net/netfilter/xt_cgroup.c | 6 ++++++ net/netfilter/xt_rateest.c | 5 +++++ 2 files changed, 11 insertions(+) diff --git a/net/netfilter/xt_cgroup.c b/net/netfilter/xt_cgroup.c index c0f5e9a4f3c6..bfc98719684e 100644 --- a/net/netfilter/xt_cgroup.c +++ b/net/netfilter/xt_cgroup.c @@ -53,6 +53,9 @@ static int cgroup_mt_check_v1(const struct xt_mtchk_param *par) info->priv = NULL; if (info->has_path) { + if (strnlen(info->path, sizeof(info->path)) >= sizeof(info->path)) + return -ENAMETOOLONG; + cgrp = cgroup_get_from_path(info->path); if (IS_ERR(cgrp)) { pr_info_ratelimited("invalid path, errno=%ld\n", @@ -85,6 +88,9 @@ static int cgroup_mt_check_v2(const struct xt_mtchk_param *par) info->priv = NULL; if (info->has_path) { + if (strnlen(info->path, sizeof(info->path)) >= sizeof(info->path)) + return -ENAMETOOLONG; + cgrp = cgroup_get_from_path(info->path); if (IS_ERR(cgrp)) { pr_info_ratelimited("invalid path, errno=%ld\n", diff --git a/net/netfilter/xt_rateest.c b/net/netfilter/xt_rateest.c index 72324bd976af..b1d736c15fcb 100644 --- a/net/netfilter/xt_rateest.c +++ b/net/netfilter/xt_rateest.c @@ -91,6 +91,11 @@ static int xt_rateest_mt_checkentry(const struct xt_mtchk_param *par) goto err1; } + if (strnlen(info->name1, sizeof(info->name1)) >= sizeof(info->name1)) + return -ENAMETOOLONG; + if (strnlen(info->name2, sizeof(info->name2)) >= sizeof(info->name2)) + return -ENAMETOOLONG; + ret = -ENOENT; est1 = xt_rateest_lookup(par->net, info->name1); if (!est1) -- 2.43.0

2 1

[PATCH OLK-5.10] mm/ksm: fix incorrect KSM counter handling in mm_struct during fork
by Wupeng Ma 09 Jun '26

09 Jun '26

From: Donet Tom <donettom(a)linux.ibm.com> stable inclusion from stable-v6.6.113 commit ad25061d1d73e9067fdddf9133b8f4cb6c89dc0d category: bugfix bugzilla: https://atomgit.com/openeuler/kernel/issues/9368 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 4d6fc29f36341d7795db1d1819b4c15fe9be7b23 ] Patch series "mm/ksm: Fix incorrect accounting of KSM counters during fork", v3. The first patch in this series fixes the incorrect accounting of KSM counters such as ksm_merging_pages, ksm_rmap_items, and the global ksm_zero_pages during fork. The following patch add a selftest to verify the ksm_merging_pages counter was updated correctly during fork. Test Results ============ Without the first patch ----------------------- # [RUN] test_fork_ksm_merging_page_count not ok 10 ksm_merging_page in child: 32 With the first patch -------------------- # [RUN] test_fork_ksm_merging_page_count ok 10 ksm_merging_pages is not inherited after fork This patch (of 2): Currently, the KSM-related counters in `mm_struct`, such as `ksm_merging_pages`, `ksm_rmap_items`, and `ksm_zero_pages`, are inherited by the child process during fork. This results in inconsistent accounting. When a process uses KSM, identical pages are merged and an rmap item is created for each merged page. The `ksm_merging_pages` and `ksm_rmap_items` counters are updated accordingly. However, after a fork, these counters are copied to the child while the corresponding rmap items are not. As a result, when the child later triggers an unmerge, there are no rmap items present in the child, so the counters remain stale, leading to incorrect accounting. A similar issue exists with `ksm_zero_pages`, which maintains both a global counter and a per-process counter. During fork, the per-process counter is inherited by the child, but the global counter is not incremented. Since the child also references zero pages, the global counter should be updated as well. Otherwise, during zero-page unmerge, both the global and per-process counters are decremented, causing the global counter to become inconsistent. To fix this, ksm_merging_pages and ksm_rmap_items are reset to 0 during fork, and the global ksm_zero_pages counter is updated with the per-process ksm_zero_pages value inherited by the child. This ensures that KSM statistics remain accurate and reflect the activity of each process correctly. Link: https://lkml.kernel.org/r/cover.1758648700.git.donettom@linux.ibm.com Link: https://lkml.kernel.org/r/7b9870eb67ccc0d79593940d9dbd4a0b39b5d396.17586487… Fixes: 7609385337a4 ("ksm: count ksm merging pages for each process") Fixes: cb4df4cae4f2 ("ksm: count allocated ksm rmap_items for each process") Fixes: e2942062e01d ("ksm: count all zero pages placed by KSM") Signed-off-by: Donet Tom <donettom(a)linux.ibm.com> Reviewed-by: Chengming Zhou <chengming.zhou(a)linux.dev> Acked-by: David Hildenbrand <david(a)redhat.com> Cc: Aboorva Devarajan <aboorvad(a)linux.ibm.com> Cc: David Hildenbrand <david(a)redhat.com> Cc: Donet Tom <donettom(a)linux.ibm.com> Cc: "Ritesh Harjani (IBM)" <ritesh.list(a)gmail.com> Cc: Wei Yang <richard.weiyang(a)gmail.com> Cc: xu xin <xu.xin16(a)zte.com.cn> Cc: <stable(a)vger.kernel.org> [6.6+] Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> [ changed mm_flags_test() to test_bit() ] Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Conflicts: include/linux/ksm.h [Conflicts due to commit e2942062e01d ("ksm: count all zero pages placed by KSM") no merged] Signed-off-by: Wupeng Ma <mawupeng1(a)huawei.com> --- include/linux/ksm.h | 3 +++ 1 file changed, 3 insertions(+) diff --git a/include/linux/ksm.h b/include/linux/ksm.h index debef5446114f..5d1430f7389cc 100644 --- a/include/linux/ksm.h +++ b/include/linux/ksm.h @@ -34,6 +34,9 @@ static inline int ksm_fork(struct mm_struct *mm, struct mm_struct *oldmm) int ret; if (test_bit(MMF_VM_MERGEABLE, &oldmm->flags)) { + mm->ksm_merging_pages = 0; + mm->ksm_rmap_items = 0; + ret = __ksm_enter(mm); if (ret) return ret; -- 2.43.0

2 1

[PATCH OLK-6.6] mm/ksm: fix incorrect KSM counter handling in mm_struct during fork
by Wupeng Ma 09 Jun '26

09 Jun '26

From: Donet Tom <donettom(a)linux.ibm.com> stable inclusion from stable-v6.6.113 commit ad25061d1d73e9067fdddf9133b8f4cb6c89dc0d category: bugfix bugzilla: https://atomgit.com/openeuler/kernel/issues/9368 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 4d6fc29f36341d7795db1d1819b4c15fe9be7b23 ] Patch series "mm/ksm: Fix incorrect accounting of KSM counters during fork", v3. The first patch in this series fixes the incorrect accounting of KSM counters such as ksm_merging_pages, ksm_rmap_items, and the global ksm_zero_pages during fork. The following patch add a selftest to verify the ksm_merging_pages counter was updated correctly during fork. Test Results ============ Without the first patch ----------------------- # [RUN] test_fork_ksm_merging_page_count not ok 10 ksm_merging_page in child: 32 With the first patch -------------------- # [RUN] test_fork_ksm_merging_page_count ok 10 ksm_merging_pages is not inherited after fork This patch (of 2): Currently, the KSM-related counters in `mm_struct`, such as `ksm_merging_pages`, `ksm_rmap_items`, and `ksm_zero_pages`, are inherited by the child process during fork. This results in inconsistent accounting. When a process uses KSM, identical pages are merged and an rmap item is created for each merged page. The `ksm_merging_pages` and `ksm_rmap_items` counters are updated accordingly. However, after a fork, these counters are copied to the child while the corresponding rmap items are not. As a result, when the child later triggers an unmerge, there are no rmap items present in the child, so the counters remain stale, leading to incorrect accounting. A similar issue exists with `ksm_zero_pages`, which maintains both a global counter and a per-process counter. During fork, the per-process counter is inherited by the child, but the global counter is not incremented. Since the child also references zero pages, the global counter should be updated as well. Otherwise, during zero-page unmerge, both the global and per-process counters are decremented, causing the global counter to become inconsistent. To fix this, ksm_merging_pages and ksm_rmap_items are reset to 0 during fork, and the global ksm_zero_pages counter is updated with the per-process ksm_zero_pages value inherited by the child. This ensures that KSM statistics remain accurate and reflect the activity of each process correctly. Link: https://lkml.kernel.org/r/cover.1758648700.git.donettom@linux.ibm.com Link: https://lkml.kernel.org/r/7b9870eb67ccc0d79593940d9dbd4a0b39b5d396.17586487… Fixes: 7609385337a4 ("ksm: count ksm merging pages for each process") Fixes: cb4df4cae4f2 ("ksm: count allocated ksm rmap_items for each process") Fixes: e2942062e01d ("ksm: count all zero pages placed by KSM") Signed-off-by: Donet Tom <donettom(a)linux.ibm.com> Reviewed-by: Chengming Zhou <chengming.zhou(a)linux.dev> Acked-by: David Hildenbrand <david(a)redhat.com> Cc: Aboorva Devarajan <aboorvad(a)linux.ibm.com> Cc: David Hildenbrand <david(a)redhat.com> Cc: Donet Tom <donettom(a)linux.ibm.com> Cc: "Ritesh Harjani (IBM)" <ritesh.list(a)gmail.com> Cc: Wei Yang <richard.weiyang(a)gmail.com> Cc: xu xin <xu.xin16(a)zte.com.cn> Cc: <stable(a)vger.kernel.org> [6.6+] Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> [ changed mm_flags_test() to test_bit() ] Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Conflicts: include/linux/ksm.h [Conflicts due to cleanup commit 283dfdd20e57 is merged.] Signed-off-by: Jinjiang Tu <tujinjiang(a)huawei.com> Signed-off-by: Wupeng Ma <mawupeng1(a)huawei.com> --- include/linux/ksm.h | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/include/linux/ksm.h b/include/linux/ksm.h index 691c1f54254e6..5c96c3c524c12 100644 --- a/include/linux/ksm.h +++ b/include/linux/ksm.h @@ -56,8 +56,15 @@ static inline long mm_ksm_zero_pages(struct mm_struct *mm) static inline void ksm_fork(struct mm_struct *mm, struct mm_struct *oldmm) { - if (test_bit(MMF_VM_MERGEABLE, &oldmm->flags)) + if (test_bit(MMF_VM_MERGEABLE, &oldmm->flags)) { + long nr_ksm_zero_pages = atomic_long_read(&mm->ksm_zero_pages); + + mm->ksm_merging_pages = 0; + mm->ksm_rmap_items = 0; + atomic_long_add(nr_ksm_zero_pages, &ksm_zero_pages); + __ksm_enter(mm); + } } static inline void ksm_exit(struct mm_struct *mm) -- 2.43.0

2 1

[PATCH OLK-5.10] netfilter: nf_conntrack_helper: pass helper to expect cleanup
by superdcc97＠163.com 09 Jun '26

09 Jun '26

From: Qi Tang <tpluszz77(a)gmail.com> stable inclusion from stable-v5.10.253 commit 5cf28d5c8dcbbe8af6d3b145babe491906d7bad1 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/14559 CVE: CVE-2026-43027 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit a242a9ae58aa46ff7dae51ce64150a93957abe65 ] nf_conntrack_helper_unregister() calls nf_ct_expect_iterate_destroy() to remove expectations belonging to the helper being unregistered. However, it passes NULL instead of the helper pointer as the data argument, so expect_iter_me() never matches any expectation and all of them survive the cleanup. After unregister returns, nfnl_cthelper_del() frees the helper object immediately. Subsequent expectation dumps or packet-driven init_conntrack() calls then dereference the freed exp->helper, causing a use-after-free. Pass the actual helper pointer so expectations referencing it are properly destroyed before the helper object is freed. BUG: KASAN: slab-use-after-free in string+0x38f/0x430 Read of size 1 at addr ffff888003b14d20 by task poc/103 Call Trace: string+0x38f/0x430 vsnprintf+0x3cc/0x1170 seq_printf+0x17a/0x240 exp_seq_show+0x2e5/0x560 seq_read_iter+0x419/0x1280 proc_reg_read+0x1ac/0x270 vfs_read+0x179/0x930 ksys_read+0xef/0x1c0 Freed by task 103: The buggy address is located 32 bytes inside of freed 192-byte region [ffff888003b14d00, ffff888003b14dc0) Fixes: ac7b84839003 ("netfilter: expect: add and use nf_ct_expect_iterate helpers") Signed-off-by: Qi Tang <tpluszz77(a)gmail.com> Reviewed-by: Phil Sutter <phil(a)nwl.cc> Signed-off-by: Pablo Neira Ayuso <pablo(a)netfilter.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Dong Chenchen <dongchenchen2(a)huawei.com> --- net/netfilter/nf_conntrack_helper.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/netfilter/nf_conntrack_helper.c b/net/netfilter/nf_conntrack_helper.c index 32cc91f5ba99..31f87eff5a36 100644 --- a/net/netfilter/nf_conntrack_helper.c +++ b/net/netfilter/nf_conntrack_helper.c @@ -468,7 +468,7 @@ void nf_conntrack_helper_unregister(struct nf_conntrack_helper *me) */ synchronize_rcu(); - nf_ct_expect_iterate_destroy(expect_iter_me, NULL); + nf_ct_expect_iterate_destroy(expect_iter_me, me); nf_ct_iterate_destroy(unhelp, me); /* Maybe someone has gotten the helper already when unhelp above. -- 2.43.0

2 1

[PATCH OLK-5.10] mm/ksm: fix incorrect KSM counter handling in mm_struct during fork
by Wupeng Ma 09 Jun '26

09 Jun '26

From: Donet Tom <donettom(a)linux.ibm.com> stable inclusion from stable-v6.6.113 commit ad25061d1d73e9067fdddf9133b8f4cb6c89dc0d category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/ID7OBU Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 4d6fc29f36341d7795db1d1819b4c15fe9be7b23 ] Patch series "mm/ksm: Fix incorrect accounting of KSM counters during fork", v3. The first patch in this series fixes the incorrect accounting of KSM counters such as ksm_merging_pages, ksm_rmap_items, and the global ksm_zero_pages during fork. The following patch add a selftest to verify the ksm_merging_pages counter was updated correctly during fork. Test Results ============ Without the first patch ----------------------- # [RUN] test_fork_ksm_merging_page_count not ok 10 ksm_merging_page in child: 32 With the first patch -------------------- # [RUN] test_fork_ksm_merging_page_count ok 10 ksm_merging_pages is not inherited after fork This patch (of 2): Currently, the KSM-related counters in `mm_struct`, such as `ksm_merging_pages`, `ksm_rmap_items`, and `ksm_zero_pages`, are inherited by the child process during fork. This results in inconsistent accounting. When a process uses KSM, identical pages are merged and an rmap item is created for each merged page. The `ksm_merging_pages` and `ksm_rmap_items` counters are updated accordingly. However, after a fork, these counters are copied to the child while the corresponding rmap items are not. As a result, when the child later triggers an unmerge, there are no rmap items present in the child, so the counters remain stale, leading to incorrect accounting. A similar issue exists with `ksm_zero_pages`, which maintains both a global counter and a per-process counter. During fork, the per-process counter is inherited by the child, but the global counter is not incremented. Since the child also references zero pages, the global counter should be updated as well. Otherwise, during zero-page unmerge, both the global and per-process counters are decremented, causing the global counter to become inconsistent. To fix this, ksm_merging_pages and ksm_rmap_items are reset to 0 during fork, and the global ksm_zero_pages counter is updated with the per-process ksm_zero_pages value inherited by the child. This ensures that KSM statistics remain accurate and reflect the activity of each process correctly. Link: https://lkml.kernel.org/r/cover.1758648700.git.donettom@linux.ibm.com Link: https://lkml.kernel.org/r/7b9870eb67ccc0d79593940d9dbd4a0b39b5d396.17586487… Fixes: 7609385337a4 ("ksm: count ksm merging pages for each process") Fixes: cb4df4cae4f2 ("ksm: count allocated ksm rmap_items for each process") Fixes: e2942062e01d ("ksm: count all zero pages placed by KSM") Signed-off-by: Donet Tom <donettom(a)linux.ibm.com> Reviewed-by: Chengming Zhou <chengming.zhou(a)linux.dev> Acked-by: David Hildenbrand <david(a)redhat.com> Cc: Aboorva Devarajan <aboorvad(a)linux.ibm.com> Cc: David Hildenbrand <david(a)redhat.com> Cc: Donet Tom <donettom(a)linux.ibm.com> Cc: "Ritesh Harjani (IBM)" <ritesh.list(a)gmail.com> Cc: Wei Yang <richard.weiyang(a)gmail.com> Cc: xu xin <xu.xin16(a)zte.com.cn> Cc: <stable(a)vger.kernel.org> [6.6+] Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> [ changed mm_flags_test() to test_bit() ] Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Conflicts: include/linux/ksm.h [Conflicts due to commit e2942062e01d ("ksm: count all zero pages placed by KSM") no merged] Signed-off-by: Wupeng Ma <mawupeng1(a)huawei.com> --- include/linux/ksm.h | 3 +++ 1 file changed, 3 insertions(+) diff --git a/include/linux/ksm.h b/include/linux/ksm.h index debef5446114f..5d1430f7389cc 100644 --- a/include/linux/ksm.h +++ b/include/linux/ksm.h @@ -34,6 +34,9 @@ static inline int ksm_fork(struct mm_struct *mm, struct mm_struct *oldmm) int ret; if (test_bit(MMF_VM_MERGEABLE, &oldmm->flags)) { + mm->ksm_merging_pages = 0; + mm->ksm_rmap_items = 0; + ret = __ksm_enter(mm); if (ret) return ret; -- 2.43.0

2 1

[PATCH OLK-6.6] mm/ksm: fix incorrect KSM counter handling in mm_struct during fork
by Wupeng Ma 09 Jun '26

09 Jun '26

From: Donet Tom <donettom(a)linux.ibm.com> stable inclusion from stable-v6.6.113 commit ad25061d1d73e9067fdddf9133b8f4cb6c89dc0d category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/ID7OBU Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 4d6fc29f36341d7795db1d1819b4c15fe9be7b23 ] Patch series "mm/ksm: Fix incorrect accounting of KSM counters during fork", v3. The first patch in this series fixes the incorrect accounting of KSM counters such as ksm_merging_pages, ksm_rmap_items, and the global ksm_zero_pages during fork. The following patch add a selftest to verify the ksm_merging_pages counter was updated correctly during fork. Test Results ============ Without the first patch ----------------------- # [RUN] test_fork_ksm_merging_page_count not ok 10 ksm_merging_page in child: 32 With the first patch -------------------- # [RUN] test_fork_ksm_merging_page_count ok 10 ksm_merging_pages is not inherited after fork This patch (of 2): Currently, the KSM-related counters in `mm_struct`, such as `ksm_merging_pages`, `ksm_rmap_items`, and `ksm_zero_pages`, are inherited by the child process during fork. This results in inconsistent accounting. When a process uses KSM, identical pages are merged and an rmap item is created for each merged page. The `ksm_merging_pages` and `ksm_rmap_items` counters are updated accordingly. However, after a fork, these counters are copied to the child while the corresponding rmap items are not. As a result, when the child later triggers an unmerge, there are no rmap items present in the child, so the counters remain stale, leading to incorrect accounting. A similar issue exists with `ksm_zero_pages`, which maintains both a global counter and a per-process counter. During fork, the per-process counter is inherited by the child, but the global counter is not incremented. Since the child also references zero pages, the global counter should be updated as well. Otherwise, during zero-page unmerge, both the global and per-process counters are decremented, causing the global counter to become inconsistent. To fix this, ksm_merging_pages and ksm_rmap_items are reset to 0 during fork, and the global ksm_zero_pages counter is updated with the per-process ksm_zero_pages value inherited by the child. This ensures that KSM statistics remain accurate and reflect the activity of each process correctly. Link: https://lkml.kernel.org/r/cover.1758648700.git.donettom@linux.ibm.com Link: https://lkml.kernel.org/r/7b9870eb67ccc0d79593940d9dbd4a0b39b5d396.17586487… Fixes: 7609385337a4 ("ksm: count ksm merging pages for each process") Fixes: cb4df4cae4f2 ("ksm: count allocated ksm rmap_items for each process") Fixes: e2942062e01d ("ksm: count all zero pages placed by KSM") Signed-off-by: Donet Tom <donettom(a)linux.ibm.com> Reviewed-by: Chengming Zhou <chengming.zhou(a)linux.dev> Acked-by: David Hildenbrand <david(a)redhat.com> Cc: Aboorva Devarajan <aboorvad(a)linux.ibm.com> Cc: David Hildenbrand <david(a)redhat.com> Cc: Donet Tom <donettom(a)linux.ibm.com> Cc: "Ritesh Harjani (IBM)" <ritesh.list(a)gmail.com> Cc: Wei Yang <richard.weiyang(a)gmail.com> Cc: xu xin <xu.xin16(a)zte.com.cn> Cc: <stable(a)vger.kernel.org> [6.6+] Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> [ changed mm_flags_test() to test_bit() ] Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Conflicts: include/linux/ksm.h [Conflicts due to cleanup commit 283dfdd20e57 is merged.] Signed-off-by: Jinjiang Tu <tujinjiang(a)huawei.com> Signed-off-by: Wupeng Ma <mawupeng1(a)huawei.com> --- include/linux/ksm.h | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/include/linux/ksm.h b/include/linux/ksm.h index 691c1f54254e6..5c96c3c524c12 100644 --- a/include/linux/ksm.h +++ b/include/linux/ksm.h @@ -56,8 +56,15 @@ static inline long mm_ksm_zero_pages(struct mm_struct *mm) static inline void ksm_fork(struct mm_struct *mm, struct mm_struct *oldmm) { - if (test_bit(MMF_VM_MERGEABLE, &oldmm->flags)) + if (test_bit(MMF_VM_MERGEABLE, &oldmm->flags)) { + long nr_ksm_zero_pages = atomic_long_read(&mm->ksm_zero_pages); + + mm->ksm_merging_pages = 0; + mm->ksm_rmap_items = 0; + atomic_long_add(nr_ksm_zero_pages, &ksm_zero_pages); + __ksm_enter(mm); + } } static inline void ksm_exit(struct mm_struct *mm) -- 2.43.0

2 1

[PATCH openEuler-1.0-LTS] scsi/hiraid: Support new RAID features and bug fixes
by LinKun 09 Jun '26

09 Jun '26

From: 岳智超 <yuezhichao1(a)h-partners.com> driver inclusion category: feature bugzilla: https://atomgit.com/openeuler/kernel/issues/9101 CVE: NA -------------------------------- - Fix issue where throughput may drop to 0 under large I/O conditions - Fix drive letter misalignment when creating/ deleting drive letters too quickly - Fix edge case in timeout handling where equal to timeout value is not handled - Fix OS failure to start in single queue mode - Adapt to configurable management command timeout - Adapt to configurable I/O queue depth - Simplify PRP and SGL workflow - Simplify stream identification feature - Add driver fast recovery after linkdown/linkup - Add boot drive priority reporting during drive letter enumeration - Add FLR support with firmware adaptation for fast driver recovery after FLR - Add NCQ feature for SATA drives - Remove legacy driver template class definitions Signed-off-by: Zhichao Yue <yuezhichao1(a)h-partners.com> --- drivers/scsi/hisi_raid/hiraid.h | 87 ++- drivers/scsi/hisi_raid/hiraid_main.c | 919 ++++++++++++++------------- 2 files changed, 536 insertions(+), 470 deletions(-) diff --git a/drivers/scsi/hisi_raid/hiraid.h b/drivers/scsi/hisi_raid/hiraid.h index cb90d3d..25df43b 100644 --- a/drivers/scsi/hisi_raid/hiraid.h +++ b/drivers/scsi/hisi_raid/hiraid.h @@ -50,6 +50,10 @@ #define HIRAID_DEV_INFO_FLAG_VALID(flag) ((flag) & 0x01) #define HIRAID_DEV_INFO_FLAG_CHANGE(flag) ((flag) & 0x02) +#define HIRAID_DEV_INFO_FLAG_DEV_MAGIC_DEV(flag) ((flag) >> 2) +#define HIRAID_DEV_INFO_FLAG_DELETE_OR_ADD(flag, org_flag) \ + (HIRAID_DEV_INFO_FLAG_DEV_MAGIC_DEV(flag) != \ + HIRAID_DEV_INFO_FLAG_DEV_MAGIC_DEV(org_flag) ? 1 : 0) #define HIRAID_CAP_MQES(cap) ((cap) & 0xffff) #define HIRAID_CAP_STRIDE(cap) (((cap) >> 32) & 0xf) @@ -74,7 +78,19 @@ #define PCI_VENDOR_ID_HUAWEI_LOGIC 0x19E5 #define HIRAID_SERVER_DEVICE_HBA_DID 0x3858 +#define HIRAID_SERVER_DEVICE_HBAS_DID 0x3918 #define HIRAID_SERVER_DEVICE_RAID_DID 0x3758 +#define HIRAID_SERVER_DEVICE_RAIDS_DID 0x38D8 + +#define HIRAID_SCSI_VPD_LEN SCSI_VPD_PG_LEN +#define SCSI_RSVD_INIT_SHIFT 0XFC +#define NCQ_PRIO_SUPPORT_BYTE 213 +#define VPD_NCQ_PRIO_SUPPORT_BIT 4 + +#define HIRAID_CAP_CAC_STRIDE 8 + +#define HIRAID_MAX_PD_NUM (40 + 1) +#define HIRAID_MAX_STREAM_NUM 8 enum { HIRAID_SC_SUCCESS = 0x0, @@ -90,6 +106,7 @@ enum { enum { HIRAID_REG_CAP = 0x0000, + HIRAID_REG_VS = 0x0008, HIRAID_REG_CC = 0x0014, HIRAID_REG_CSTS = 0x001c, HIRAID_REG_AQA = 0x0024, @@ -236,6 +253,20 @@ enum { DISPATCH_BY_DISK, }; +enum hiraid_stream_type { + HIRAID_STREAM_TYPE_TOTAL, + HIRAID_STREAM_TYPE_WRITE, + HIRAID_STREAM_TYPE_READ, + HIRAID_STREAM_TYPE_CLEAN, + HIRAID_STREAM_TYPE_BOTTOM +}; + +enum hiraid_stream_io_operation_type { + HIRAID_STREAM_TYPE_DELETE_SINGLE_IO = 1, + HIRAID_STREAM_TYPE_DELETE_SINGLE_IO_LIST, + HIRAID_STREAM_TYPE_DELETE_ALL_IO_LIST +}; + struct hiraid_completion { __le32 result; union { @@ -271,6 +302,19 @@ struct hiraid_ctrl_info { __u8 rsvd1[4020]; }; +struct hiraid_stream { + /* recog-window */ + u64 stream_lba; + u32 stream_len; + u16 did; + u16 type; + /* aging ctrl */ + int aging_credit; + int aging_grade; + u16 stream_id; + u16 stream_is_using; +}; + struct hiraid_dev { struct pci_dev *pdev; struct device *dev; @@ -318,6 +362,18 @@ struct hiraid_dev { u8 hdd_dispatch; struct request_queue *bsg_queue; + + u16 hiraid_stream_io_count; + u64 hiraid_stream_aging_time; + u64 hiraid_stream_sent_io_size[HIRAID_MAX_PD_NUM] + [HIRAID_MAX_STREAM_NUM]; + u16 hiraid_stream_io_num_per_pd[HIRAID_MAX_PD_NUM] + [HIRAID_STREAM_TYPE_BOTTOM + 1]; + spinlock_t hiraid_stream_array_lock; + struct hiraid_stream hiraid_stream_array[HIRAID_MAX_PD_NUM] + [HIRAID_MAX_STREAM_NUM]; + struct task_struct *hiraid_stream_submit_task; + u64 hiraid_stream_io_last_pull_time[HIRAID_MAX_PD_NUM]; }; struct hiraid_sgl_desc { @@ -696,6 +752,10 @@ struct hiraid_mapmange { u32 sge_cnt; u32 len; bool use_sgl; + /* This field is used in the I/O read/write process. + * It indicates the TRANSFER LENGTH field in the CDB. + */ + u32 cdb_data_len; dma_addr_t first_dma; void *sense_buffer_virt; dma_addr_t sense_buffer_phy; @@ -755,38 +815,19 @@ struct hiraid_sdev_hostdata { u8 flag; u8 rg_id; u8 hwq; + u8 sata_ncq_prio_enable; + u8 rsvd[3]; u16 pend_count; }; -enum stream_type { - TYPE_TOTAL, - TYPE_WRITE, - TYPE_READ, - TYPE_CLEAN, - TYPE_BOTTOM -}; - -struct HIRAID_STREAM_S { - /* recog-window */ - u64 stream_lba; - u32 stream_len; - u16 did; - u16 type; - /* aging ctrl */ - int aging_credit; - int aging_grade; - u16 stream_id; - u16 using; -}; - -struct IO_LIST_S { +struct hiraid_stream_io_list { struct list_head list; struct hiraid_scsi_io_cmd io_cmd; struct hiraid_queue *submit_queue; unsigned int sector_size; }; -struct mutex_list_head_s { +struct mutex_list_head { struct list_head list; struct mutex lock; }; diff --git a/drivers/scsi/hisi_raid/hiraid_main.c b/drivers/scsi/hisi_raid/hiraid_main.c index cdba466..9251fb8 100644 --- a/drivers/scsi/hisi_raid/hiraid_main.c +++ b/drivers/scsi/hisi_raid/hiraid_main.c @@ -26,6 +26,10 @@ #include <linux/bsg-lib.h> #include <asm/unaligned.h> #include <linux/sort.h> +#include <linux/kthread.h> +#include <linux/mutex.h> +#include <linux/sched.h> +#include <linux/sched/prio.h> #include <target/target_core_backend.h> #include <scsi/scsi.h> @@ -35,14 +39,10 @@ #include <scsi/scsi_transport.h> #include <scsi/scsi_dbg.h> #include <scsi/sg.h> -#include <linux/kthread.h> -#include <linux/mutex.h> -#include <linux/sched.h> -#include <linux/sched/prio.h> #include "hiraid.h" -static u32 admin_tmout = 60; +static u32 admin_tmout = 180; module_param(admin_tmout, uint, 0644); MODULE_PARM_DESC(admin_tmout, "admin commands timeout (seconds)"); @@ -101,7 +101,7 @@ static const struct kernel_param_ops io_queue_depth_ops = { .get = param_get_uint, }; -static u32 io_queue_depth = 1024; +static u32 io_queue_depth = 4096; module_param_cb(io_queue_depth, &io_queue_depth_ops, &io_queue_depth, 0644); MODULE_PARM_DESC(io_queue_depth, "set io queue depth, should >= 2"); @@ -143,8 +143,6 @@ static void hiraid_handle_async_notice(struct hiraid_dev *hdev, u32 result); static void hiraid_handle_async_vs(struct hiraid_dev *hdev, u32 result, u32 result1); -static struct class *hiraid_class; - #define HIRAID_CAP_TIMEOUT_UNIT_MS (HZ / 2) static struct workqueue_struct *work_queue; @@ -155,7 +153,7 @@ static struct workqueue_struct *work_queue; __func__, ##__VA_ARGS__); \ } while (0) -#define HIRAID_DRV_VERSION "1.1.0.1" +#define HIRAID_DRV_VERSION "2.1.0.1" #define ADMIN_TIMEOUT (admin_tmout * HZ) #define USRCMD_TIMEOUT (180 * HZ) @@ -178,8 +176,12 @@ static struct workqueue_struct *work_queue; #define MIN_CREDIT 0 #define MAX_CREDIT 64 #define CREDIT_THRES 32 +#ifndef MIN #define MIN(a, b) (((a) < (b)) ? (a) : (b)) +#endif +#ifndef MAX #define MAX(a, b) (((a) > (b)) ? (a) : (b)) +#endif enum SENSE_STATE_CODE { SENSE_STATE_OK = 0, @@ -362,95 +364,8 @@ static u32 hiraid_get_max_cmd_size(struct hiraid_dev *hdev) return sizeof(struct hiraid_mapmange) + alloc_size; } -static int hiraid_build_passthru_prp(struct hiraid_dev *hdev, - struct hiraid_mapmange *mapbuf) -{ - struct scatterlist *sg = mapbuf->sgl; - __le64 *phy_regpage, *prior_list; - u64 buf_addr = sg_dma_address(sg); - int buf_length = sg_dma_len(sg); - u32 page_size = hdev->page_size; - int offset = buf_addr & (page_size - 1); - void **list = hiraid_mapbuf_list(mapbuf); - int maplen = mapbuf->len; - struct dma_pool *pool; - dma_addr_t buffer_phy; - int i; - - maplen -= (page_size - offset); - if (maplen <= 0) { - mapbuf->first_dma = 0; - return 0; - } - - buf_length -= (page_size - offset); - if (buf_length) { - buf_addr += (page_size - offset); - } else { - sg = sg_next(sg); - buf_addr = sg_dma_address(sg); - buf_length = sg_dma_len(sg); - } - - if (maplen <= page_size) { - mapbuf->first_dma = buf_addr; - return 0; - } - - pool = hdev->prp_page_pool; - mapbuf->page_cnt = 1; - - phy_regpage = dma_pool_alloc(pool, GFP_ATOMIC, &buffer_phy); - if (!phy_regpage) { - dev_err_ratelimited(hdev->dev, "allocate first admin prp_list memory failed\n"); - mapbuf->first_dma = buf_addr; - mapbuf->page_cnt = -1; - return -ENOMEM; - } - list[0] = phy_regpage; - mapbuf->first_dma = buffer_phy; - i = 0; - for (;;) { - if (i == page_size / PRP_ENTRY_SIZE) { - prior_list = phy_regpage; - - phy_regpage = dma_pool_alloc(pool, GFP_ATOMIC, - &buffer_phy); - if (!phy_regpage) { - dev_err_ratelimited(hdev->dev, "allocate [%d]th admin prp list memory failed\n", - mapbuf->page_cnt + 1); - return -ENOMEM; - } - list[mapbuf->page_cnt++] = phy_regpage; - phy_regpage[0] = prior_list[i - 1]; - prior_list[i - 1] = cpu_to_le64(buffer_phy); - i = 1; - } - phy_regpage[i++] = cpu_to_le64(buf_addr); - buf_addr += page_size; - buf_length -= page_size; - maplen -= page_size; - if (maplen <= 0) - break; - if (buf_length > 0) - continue; - if (unlikely(buf_length < 0)) - goto bad_admin_sgl; - sg = sg_next(sg); - buf_addr = sg_dma_address(sg); - buf_length = sg_dma_len(sg); - } - - return 0; - -bad_admin_sgl: - dev_err(hdev->dev, "setup prps, invalid admin SGL for payload[%d] nents[%d]\n", - mapbuf->len, mapbuf->sge_cnt); - return -EIO; -} - static int hiraid_build_prp(struct hiraid_dev *hdev, - struct hiraid_mapmange *mapbuf) + struct hiraid_mapmange *mapbuf) { struct scatterlist *sg = mapbuf->sgl; __le64 *phy_regpage, *prior_list; @@ -656,67 +571,9 @@ static void hiraid_sgl_set_seg(struct hiraid_sgl_desc *sge, } } -static int hiraid_build_passthru_sgl(struct hiraid_dev *hdev, - struct hiraid_admin_command *admin_cmd, - struct hiraid_mapmange *mapbuf) -{ - struct hiraid_sgl_desc *sg_list, *link, *old_sg_list; - struct scatterlist *sg = mapbuf->sgl; - void **list = hiraid_mapbuf_list(mapbuf); - struct dma_pool *pool; - int nsge = mapbuf->sge_cnt; - dma_addr_t buffer_phy; - int i = 0; - - admin_cmd->common.flags |= SQE_FLAG_SGL_METABUF; - - if (nsge == 1) { - hiraid_sgl_set_data(&admin_cmd->common.dptr.sgl, sg); - return 0; - } - - pool = hdev->prp_page_pool; - mapbuf->page_cnt = 1; - - sg_list = dma_pool_alloc(pool, GFP_ATOMIC, &buffer_phy); - if (!sg_list) { - dev_err_ratelimited(hdev->dev, "allocate first admin sgl_list failed\n"); - mapbuf->page_cnt = -1; - return -ENOMEM; - } - - list[0] = sg_list; - mapbuf->first_dma = buffer_phy; - hiraid_sgl_set_seg(&admin_cmd->common.dptr.sgl, buffer_phy, nsge); - do { - if (i == SGES_PER_PAGE) { - old_sg_list = sg_list; - link = &old_sg_list[SGES_PER_PAGE - 1]; - - sg_list = dma_pool_alloc(pool, GFP_ATOMIC, &buffer_phy); - if (!sg_list) { - dev_err_ratelimited(hdev->dev, "allocate [%d]th admin sgl_list failed\n", - mapbuf->page_cnt + 1); - return -ENOMEM; - } - list[mapbuf->page_cnt++] = sg_list; - - i = 0; - memcpy(&sg_list[i++], link, sizeof(*link)); - hiraid_sgl_set_seg(link, buffer_phy, nsge); - } - - hiraid_sgl_set_data(&sg_list[i++], sg); - sg = sg_next(sg); - } while (--nsge > 0); - - return 0; -} - - static int hiraid_build_sgl(struct hiraid_dev *hdev, - struct hiraid_scsi_io_cmd *io_cmd, - struct hiraid_mapmange *mapbuf) + struct hiraid_scsi_io_cmd *io_cmd, + struct hiraid_mapmange *mapbuf) { struct hiraid_sgl_desc *sg_list, *link, *old_sg_list; struct scatterlist *sg = mapbuf->sgl; @@ -776,102 +633,52 @@ static int hiraid_build_sgl(struct hiraid_dev *hdev, return 0; } -#define MAX_PD_NUM (40 + 1) -#define MAX_STREAM_NUM 8 -#define PER_MB (1024 * 1024) -#define MAX_IO_NUM (200 * PER_MB) -#define STREAM_LEN (4 * PER_MB) +#define MAX_IO_NUM (200 * 1024 * 1024) +#define STREAM_LEN (4 * 1024 * 1024) #define MAX_IO_NUM_ONCE 100 #define IO_SUBMIT_TIME_OUT 100 -#define MAX_AGING_NUM 100 - +#define MAX_AGING_NUM 5000 +#define MAX_AGING_TIME 16 +#define AGING_DEGRADE (-2) #define MIN_IO_SEND_TIME 10 #define MAX_IO_SEND_TIME 50 -enum io_operation_type { - TYPE_DELETE_SINGLE_IO = 1, - TYPE_DELETE_SINGLE_IO_LIST, - TYPE_DELETE_ALL_IO_LIST -}; - -struct HIRAID_STREAM_S stream_array[MAX_PD_NUM][MAX_STREAM_NUM] = {0}; -struct mutex_list_head_s io_heads_per_stream[MAX_PD_NUM * MAX_STREAM_NUM]; -spinlock_t stream_array_lock; +struct mutex_list_head hiraid_io_heads_per_stream[ + HIRAID_MAX_PD_NUM * HIRAID_MAX_STREAM_NUM]; DEFINE_MUTEX(g_stream_operation_mutex); -u64 g_io_transport_num[MAX_PD_NUM][MAX_STREAM_NUM] = {0}; -u16 g_io_stream_num[MAX_PD_NUM][TYPE_BOTTOM] = {0}; -u16 g_io_count = 1; - -void hiraid_inc_io_transport_num(u16 disk_id, u16 streamd_id, u16 nlb) -{ - g_io_transport_num[disk_id][streamd_id] += nlb; -} - -void hiraid_refresh_io_transport_num(u16 disk_id, u16 streamd_id) -{ - g_io_transport_num[disk_id][streamd_id] = 0; -} - -void hiraid_inc_stream_num(u16 disk_id) -{ - spin_lock(&stream_array_lock); - g_io_stream_num[disk_id][TYPE_TOTAL]++; - spin_unlock(&stream_array_lock); -} - -void hiraid_dec_stream_num(u16 disk_id) -{ - spin_lock(&stream_array_lock); - if (g_io_stream_num[disk_id][TYPE_TOTAL] > 0) - g_io_stream_num[disk_id][TYPE_TOTAL]--; - spin_unlock(&stream_array_lock); -} - -static bool hiraid_io_recog_check_stream_exceed(u16 disk_id) +static inline bool hiraid_io_recog_check_stream_exceed(struct hiraid_dev *hdev, + u16 disk_id) { bool exceed_flag; - spin_lock(&stream_array_lock); - exceed_flag = (g_io_stream_num[disk_id][TYPE_TOTAL] >= MAX_STREAM_NUM); - spin_unlock(&stream_array_lock); + spin_lock(&hdev->hiraid_stream_array_lock); + exceed_flag = + (hdev->hiraid_stream_io_num_per_pd[disk_id][HIRAID_STREAM_TYPE_TOTAL] >= + HIRAID_MAX_STREAM_NUM); + spin_unlock(&hdev->hiraid_stream_array_lock); return exceed_flag; } -static u16 hiraid_get_stream_num(u16 disk_id) -{ - return g_io_stream_num[disk_id][TYPE_TOTAL]; -} - -static inline struct HIRAID_STREAM_S *hiraid_get_stream(u16 disk_id, - u16 stream_id) -{ - return &stream_array[disk_id][stream_id]; -} - -static inline struct mutex_list_head_s *hiraid_get_io_head(u16 disk_id) -{ - return &(io_heads_per_stream[disk_id]); -} - -static bool hiraid_recognition_acknowledge(const struct HIRAID_STREAM_S *stream) +static inline bool hiraid_recognition_acknowledge( + const struct hiraid_stream *stream) { return (stream->aging_credit >= CREDIT_THRES) ? true : false; } -void hiraid_io_recognition_init(void) +void hiraid_io_recognition_init(struct hiraid_dev *hdev) { u16 i; - spin_lock_init(&stream_array_lock); - for (i = 0; i < (MAX_PD_NUM * MAX_STREAM_NUM); i++) { - INIT_LIST_HEAD(&hiraid_get_io_head(i)->list); - mutex_init(&hiraid_get_io_head(i)->lock); + spin_lock_init(&hdev->hiraid_stream_array_lock); + for (i = 0; i < (HIRAID_MAX_PD_NUM * HIRAID_MAX_STREAM_NUM); i++) { + INIT_LIST_HEAD(&hiraid_io_heads_per_stream[i].list); + mutex_init(&hiraid_io_heads_per_stream[i].lock); } } -static void hiraid_io_recognition_iterator(struct HIRAID_STREAM_S *stream, - int direction) +static void hiraid_io_recognition_iterator(struct hiraid_stream *stream, + int direction) { stream->aging_grade = stream->aging_grade + direction * INC_GRADE; stream->aging_grade = MAX(stream->aging_grade, MAX_DECREASE_GRADE); @@ -881,46 +688,52 @@ static void hiraid_io_recognition_iterator(struct HIRAID_STREAM_S *stream, stream->aging_credit = MIN(stream->aging_credit, MAX_CREDIT); } -struct HIRAID_STREAM_S *hiraid_io_pick_stream( +struct hiraid_stream *hiraid_io_pick_stream(struct hiraid_dev *hdev, struct hiraid_scsi_rw_cmd *req, u16 type, u16 actual_id) { - struct HIRAID_STREAM_S *first_hit_stream = NULL; - struct HIRAID_STREAM_S *temp_stream = NULL; - u16 pick_flag = 0; + struct hiraid_stream *first_hit_stream = NULL; + struct hiraid_stream *temp_stream = NULL; + bool pick_flag = false; u8 i; - for (i = 0; i < MAX_STREAM_NUM; i++) { - temp_stream = &stream_array[actual_id][i]; + for (i = 0; i < HIRAID_MAX_STREAM_NUM; i++) { + temp_stream = &hdev->hiraid_stream_array[actual_id][i]; temp_stream->stream_id = i; if (req->slba < temp_stream->stream_lba || req->slba >= temp_stream->stream_lba + temp_stream->stream_len || - temp_stream->type != type) { + temp_stream->type != type || + !temp_stream->stream_is_using) { continue; } - if (!pick_flag) { + if (pick_flag == false) { temp_stream->stream_lba = req->slba; first_hit_stream = temp_stream; - pick_flag = 1; + pick_flag = true; continue; } - hiraid_dec_stream_num(actual_id); - memset(temp_stream, 0, - sizeof(struct HIRAID_STREAM_S)); // 去重影 + spin_lock(&hdev->hiraid_stream_array_lock); + if (hdev->hiraid_stream_io_num_per_pd[actual_id] + [HIRAID_STREAM_TYPE_TOTAL] > 0) + hdev->hiraid_stream_io_num_per_pd[actual_id] + [HIRAID_STREAM_TYPE_TOTAL]--; + + memset(temp_stream, 0, sizeof(struct hiraid_stream)); + spin_unlock(&hdev->hiraid_stream_array_lock); } return first_hit_stream; } -static struct HIRAID_STREAM_S *hiraid_init_flow_stream( +static struct hiraid_stream *hiraid_init_flow_stream(struct hiraid_dev *hdev, struct hiraid_scsi_rw_cmd *req, u16 type, u16 actual_id) { int i; - struct HIRAID_STREAM_S *stream = NULL; + struct hiraid_stream *stream = NULL; - for (i = 0; i < MAX_STREAM_NUM; i++) { - stream = hiraid_get_stream(actual_id, i); - if (!stream->using) { - stream->using = 1; + for (i = 0; i < HIRAID_MAX_STREAM_NUM; i++) { + stream = &hdev->hiraid_stream_array[actual_id][i]; + if (!stream->stream_is_using) { + stream->stream_is_using = 1; stream->stream_id = i; break; } @@ -934,82 +747,96 @@ static struct HIRAID_STREAM_S *hiraid_init_flow_stream( return stream; } -static struct HIRAID_STREAM_S *hiraid_stream_detect(struct hiraid_dev *hdev, - struct hiraid_scsi_rw_cmd *io_cmd, u16 actual_id) +static struct hiraid_stream *hiraid_stream_detect(struct hiraid_dev *hdev, + struct hiraid_scsi_rw_cmd *io_cmd, u16 actual_id) { - u16 type = io_cmd->opcode == HIRAID_CMD_WRITE ? TYPE_WRITE : TYPE_READ; - struct HIRAID_STREAM_S *stream = hiraid_io_pick_stream(io_cmd, type, - actual_id); + u16 type = HIRAID_STREAM_TYPE_READ; + struct hiraid_stream *stream = NULL; + + if (io_cmd->opcode == HIRAID_CMD_WRITE) + type = HIRAID_STREAM_TYPE_WRITE; + + stream = hiraid_io_pick_stream(hdev, + io_cmd, type, actual_id); - if (stream != NULL) { /* 可以命中一个stream */ + if (stream != NULL) return stream; - } - if (hiraid_io_recog_check_stream_exceed(actual_id)) + if (hiraid_io_recog_check_stream_exceed(hdev, actual_id)) return NULL; - stream = hiraid_init_flow_stream(io_cmd, type, actual_id); - hiraid_inc_stream_num(actual_id); + + stream = hiraid_init_flow_stream(hdev, io_cmd, type, actual_id); + spin_lock(&hdev->hiraid_stream_array_lock); + hdev->hiraid_stream_io_num_per_pd[actual_id] + [HIRAID_STREAM_TYPE_TOTAL]++; + spin_unlock(&hdev->hiraid_stream_array_lock); return stream; } -u64 g_io_last_pull_time[MAX_PD_NUM] = {0}; - static u16 hiraid_get_submit_io_stream(u16 did, struct hiraid_dev *hdev) { u64 temp_num, i; - static u16 stream_num[MAX_PD_NUM] = {0}; + static u16 stream_num[HIRAID_MAX_PD_NUM] = {0}; - if (g_io_last_pull_time[did] == 0) - g_io_last_pull_time[did] = jiffies_to_msecs(jiffies); + if (hdev->hiraid_stream_io_last_pull_time[did] == 0) + hdev->hiraid_stream_io_last_pull_time[did] = + jiffies_to_msecs(jiffies); - for (i = 0; i < MAX_STREAM_NUM; i++) { - temp_num = g_io_transport_num[did][i]; + for (i = 0; i < HIRAID_MAX_STREAM_NUM; i++) { + temp_num = hdev->hiraid_stream_sent_io_size[did][i]; if (temp_num != 0) { if ((temp_num < MAX_IO_NUM) && - ((jiffies_to_msecs(jiffies) - g_io_last_pull_time[did]) - < IO_SUBMIT_TIME_OUT)) { + ((jiffies_to_msecs(jiffies) - + hdev->hiraid_stream_io_last_pull_time[did]) < + IO_SUBMIT_TIME_OUT)) { stream_num[did] = i; return i; } - g_io_last_pull_time[did] = jiffies_to_msecs(jiffies); - hiraid_refresh_io_transport_num(did, i); - stream_num[did] = ((i+1) % MAX_STREAM_NUM); - return ((i+1) % MAX_STREAM_NUM); + hdev->hiraid_stream_io_last_pull_time[did] = + jiffies_to_msecs(jiffies); + hdev->hiraid_stream_sent_io_size[did][i] = 0; + stream_num[did] = ((i+1) % + HIRAID_MAX_STREAM_NUM); + return ((i+1) % HIRAID_MAX_STREAM_NUM); + } } - g_io_last_pull_time[did] = jiffies_to_msecs(jiffies); - return ((stream_num[did]++) % MAX_STREAM_NUM); + hdev->hiraid_stream_io_last_pull_time[did] = + jiffies_to_msecs(jiffies); + return ((stream_num[did]++) % HIRAID_MAX_STREAM_NUM); } static void hiraid_submit_io_stream(u16 hdid, struct hiraid_dev *hdev) { - struct mutex_list_head_s *io_slist = NULL; + struct mutex_list_head *io_slist = NULL; struct list_head *node = NULL; struct list_head *next_node = NULL; struct list_head temp_header; - struct hiraid_scsi_io_cmd io_cmd = {0}; + struct hiraid_scsi_io_cmd io_cmd; u16 submit_stream_id; - struct IO_LIST_S *temp_io_stream = NULL; + struct hiraid_stream_io_list *temp_io_stream = NULL; u16 count = 0; INIT_LIST_HEAD(&temp_header); submit_stream_id = hiraid_get_submit_io_stream(hdid, hdev); - io_slist = hiraid_get_io_head(hdid * MAX_STREAM_NUM + submit_stream_id); + io_slist = &hiraid_io_heads_per_stream[hdid * HIRAID_MAX_STREAM_NUM + + submit_stream_id]; mutex_lock(&io_slist->lock); list_for_each_safe(node, next_node, &io_slist->list) { list_del(node); list_add_tail(node, &temp_header); - if (++count >= MAX_IO_NUM_ONCE) + if (count++ >= MAX_IO_NUM_ONCE) break; } mutex_unlock(&io_slist->lock); list_for_each_safe(node, next_node, &temp_header) { - temp_io_stream = list_entry(node, struct IO_LIST_S, list); + temp_io_stream = list_entry(node, + struct hiraid_stream_io_list, list); io_cmd = temp_io_stream->io_cmd; hiraid_submit_cmd(temp_io_stream->submit_queue, &io_cmd); - hiraid_inc_io_transport_num(hdid, submit_stream_id, - io_cmd.rw.nlb * temp_io_stream->sector_size); + hdev->hiraid_stream_sent_io_size[hdid][submit_stream_id] += + io_cmd.rw.nlb * temp_io_stream->sector_size; list_del(node); kfree(temp_io_stream); } @@ -1017,59 +844,90 @@ static void hiraid_submit_io_stream(u16 hdid, struct hiraid_dev *hdev) list_del_init(&temp_header); } -static u8 hiraid_detect_if_aging(void) +static u8 hiraid_aging_detect(struct hiraid_dev *hdev) { - if (++g_io_count == MAX_AGING_NUM) { - g_io_count = 0; + if (hdev->hiraid_stream_aging_time == 0) + hdev->hiraid_stream_aging_time = jiffies_to_msecs(jiffies); + if (++hdev->hiraid_stream_io_count > MAX_AGING_NUM) { + hdev->hiraid_stream_io_count = 0; + if ((jiffies_to_msecs(jiffies) - + hdev->hiraid_stream_aging_time) < + MAX_AGING_TIME) + return 0; + hdev->hiraid_stream_aging_time = jiffies_to_msecs(jiffies); return 1; } return 0; } +static void hiraid_sync_using_stream(struct hiraid_dev *hdev, u16 hdid) +{ + u8 i; + struct hiraid_stream *temp_stream = NULL; + + hdev->hiraid_stream_io_num_per_pd[hdid][HIRAID_STREAM_TYPE_TOTAL] = 0; + for (i = 0; i < HIRAID_MAX_STREAM_NUM; i++) { + temp_stream = &hdev->hiraid_stream_array[hdid][i]; + if (temp_stream->stream_is_using) + hdev->hiraid_stream_io_num_per_pd[hdid] + [HIRAID_STREAM_TYPE_TOTAL]++; + } +} + static void hiraid_aging(struct hiraid_dev *hdev) { - struct HIRAID_STREAM_S *temp_stream = NULL; + struct hiraid_stream *temp_stream = NULL; int i = 0; int j = 0; - - for (i = 0; i < MAX_PD_NUM; i++) { - for (j = 0; j < MAX_STREAM_NUM; j++) { - temp_stream = hiraid_get_stream(i, j); - if (temp_stream->using) { - hiraid_io_recognition_iterator(temp_stream, -1); + u16 *tmp_io_num = NULL; + + for (i = 0; i < HIRAID_MAX_PD_NUM; i++) { + for (j = 0; j < HIRAID_MAX_STREAM_NUM; j++) { + temp_stream = &hdev->hiraid_stream_array[i][j]; + if (temp_stream->stream_is_using) { + hiraid_io_recognition_iterator(temp_stream, + AGING_DEGRADE); + spin_lock(&hdev->hiraid_stream_array_lock); + tmp_io_num = &hdev->hiraid_stream_io_num_per_pd + [i][HIRAID_STREAM_TYPE_TOTAL]; if (temp_stream->aging_credit <= 0) { - hiraid_dec_stream_num(i); + if (*tmp_io_num > 0) + (*tmp_io_num)--; memset(temp_stream, - 0, sizeof(struct HIRAID_STREAM_S)); // 老化 + 0, sizeof(struct hiraid_stream)); } + spin_unlock(&hdev->hiraid_stream_array_lock); } } + hiraid_sync_using_stream(hdev, i); } } -static u8 hiraid_io_list_operation(u32 hdid, u16 cid, u16 hwq, u8 operation) +static u8 hiraid_stream_delete_io_list(struct hiraid_dev *hdev, u32 hdid, + u16 cid, u16 hwq, u8 delete_operation) { int i, j; - - struct mutex_list_head_s *io_slist = NULL; + struct mutex_list_head *io_slist = NULL; struct list_head *node = NULL; struct list_head *next_node = NULL; struct hiraid_scsi_io_cmd *io_cmd = NULL; struct hiraid_queue *hiraidq = NULL; - struct IO_LIST_S *temp_io_stream = NULL; + struct hiraid_stream_io_list *temp_io_stream = NULL; + u8 max_hd_num = delete_operation == + HIRAID_STREAM_TYPE_DELETE_ALL_IO_LIST ? HIRAID_MAX_PD_NUM : hdid + 1; - u8 max_hd_num = operation == TYPE_DELETE_ALL_IO_LIST ? - MAX_PD_NUM : hdid + 1; for (i = hdid; i < max_hd_num; i++) { - for (j = 0; j < MAX_STREAM_NUM; j++) { - io_slist = hiraid_get_io_head(i * MAX_STREAM_NUM + j); + for (j = 0; j < HIRAID_MAX_STREAM_NUM; j++) { + io_slist = + &hiraid_io_heads_per_stream[i * HIRAID_MAX_STREAM_NUM + j]; mutex_lock(&io_slist->lock); list_for_each_safe(node, next_node, &io_slist->list) { - temp_io_stream = list_entry(node, - struct IO_LIST_S, list); + temp_io_stream = + list_entry(node, struct hiraid_stream_io_list, list); io_cmd = &(temp_io_stream->io_cmd); hiraidq = temp_io_stream->submit_queue; - if (operation >= TYPE_DELETE_SINGLE_IO_LIST) { + if (delete_operation >= + HIRAID_STREAM_TYPE_DELETE_SINGLE_IO_LIST) { list_del_init(node); kfree(temp_io_stream); temp_io_stream = NULL; @@ -1089,54 +947,27 @@ static u8 hiraid_io_list_operation(u32 hdid, u16 cid, u16 hwq, u8 operation) return 0; } -static u8 hiraid_check_io_list(u32 hdid, u16 cid, u16 hwq) -{ - u8 ret; - - mutex_lock(&g_stream_operation_mutex); - ret = hiraid_io_list_operation(hdid, cid, hwq, TYPE_DELETE_SINGLE_IO); - mutex_unlock(&g_stream_operation_mutex); - return ret; -} - -static u8 hiraid_delete_single_pd_io_list(u32 hdid) +static u8 hiraid_add_io_to_list(struct hiraid_dev *hdev, + struct hiraid_queue *submit_queue, struct hiraid_stream *tmp_stream, + struct hiraid_scsi_io_cmd io_cmd, unsigned int sector_size, + u16 actual_id) { - u8 ret; + struct mutex_list_head *io_slist = NULL; + struct hiraid_stream_io_list *new_io_node = NULL; + u16 temp_io_id = actual_id * HIRAID_MAX_STREAM_NUM + + tmp_stream->stream_id; - mutex_lock(&g_stream_operation_mutex); - ret = hiraid_io_list_operation(hdid, 0, 0, TYPE_DELETE_SINGLE_IO_LIST); - mutex_unlock(&g_stream_operation_mutex); - return ret; -} - -static u8 hiraid_delete_all_io_list(void) -{ - u8 ret; - - mutex_lock(&g_stream_operation_mutex); - ret = hiraid_io_list_operation(0, 0, 0, TYPE_DELETE_ALL_IO_LIST); - mutex_unlock(&g_stream_operation_mutex); - return ret; -} - -static u8 hiraid_add_io_to_list(struct hiraid_queue *submit_queue, - struct HIRAID_STREAM_S *tmp_stream, struct hiraid_scsi_io_cmd io_cmd, - unsigned int sector_size, u16 actual_id) -{ - struct mutex_list_head_s *io_slist = NULL; - struct IO_LIST_S *new_io_node = NULL; - - new_io_node = kmalloc(sizeof(struct IO_LIST_S), GFP_KERNEL); + new_io_node = kmalloc(sizeof(struct hiraid_stream_io_list), GFP_KERNEL); if (!new_io_node) return 0; new_io_node->io_cmd = io_cmd; new_io_node->submit_queue = submit_queue; new_io_node->sector_size = sector_size; - io_slist = hiraid_get_io_head(actual_id * - MAX_STREAM_NUM + tmp_stream->stream_id); + io_slist = &hiraid_io_heads_per_stream[temp_io_id]; mutex_lock(&io_slist->lock); INIT_LIST_HEAD(&(new_io_node->list)); - list_add_tail(&(new_io_node->list), &io_slist->list); + list_add_tail(&(new_io_node->list), + &hiraid_io_heads_per_stream[temp_io_id].list); mutex_unlock(&io_slist->lock); return 1; } @@ -1146,10 +977,8 @@ static void hiraid_submit_io_threading(struct hiraid_dev *hdev) int i = 0; while (!kthread_should_stop()) { - mutex_lock(&g_stream_operation_mutex); - for (i = 0; i < MAX_PD_NUM; i++) + for (i = 0; i < HIRAID_MAX_PD_NUM; i++) hiraid_submit_io_stream(i, hdev); - mutex_unlock(&g_stream_operation_mutex); usleep_range(MIN_IO_SEND_TIME, MAX_IO_SEND_TIME); } } @@ -1158,27 +987,34 @@ static void hiraid_destroy_io_stream_resource(struct hiraid_dev *hdev) { u16 i; - for (i = 0; i < (MAX_PD_NUM * MAX_STREAM_NUM); i++) - list_del_init(&hiraid_get_io_head(i)->list); + for (i = 0; i < (HIRAID_MAX_PD_NUM * HIRAID_MAX_STREAM_NUM); i++) { + list_del_init(&hiraid_io_heads_per_stream[i].list); + mutex_destroy(&hiraid_io_heads_per_stream[i].lock); + } } -struct task_struct *g_hiraid_submit_task; static void hiraid_init_io_stream(struct hiraid_dev *hdev) { - hiraid_io_recognition_init(); - g_hiraid_submit_task = kthread_run((void *)hiraid_submit_io_threading, - hdev, "hiraid_submit_thread"); + hiraid_io_recognition_init(hdev); + hdev->hiraid_stream_submit_task = + kthread_run((void *)hiraid_submit_io_threading, + hdev, "hiraid_stream_submit_task"); } #define HIRAID_RW_FUA BIT(14) +#define RW_LENGTH_ZERO (67) static int hiraid_setup_rw_cmd(struct hiraid_dev *hdev, struct hiraid_scsi_rw_cmd *io_cmd, - struct scsi_cmnd *scmd) + struct scsi_cmnd *scmd, + struct hiraid_mapmange *mapbuf) { + u32 ret = 0; u32 start_lba_lo, start_lba_hi; u32 datalength = 0; u16 control = 0; + struct scsi_device *sdev = scmd->device; + u32 buf_len = cpu_to_le32(scsi_bufflen(scmd)); start_lba_lo = 0; start_lba_hi = 0; @@ -1187,6 +1023,8 @@ static int hiraid_setup_rw_cmd(struct hiraid_dev *hdev, io_cmd->opcode = HIRAID_CMD_WRITE; } else if (scmd->sc_data_direction == DMA_FROM_DEVICE) { io_cmd->opcode = HIRAID_CMD_READ; + } else if (scmd->sc_data_direction == DMA_NONE) { + ret = RW_LENGTH_ZERO; } else { dev_err(hdev->dev, "invalid RW_IO for unsupported data direction[%d]\n", scmd->sc_data_direction); @@ -1194,6 +1032,9 @@ static int hiraid_setup_rw_cmd(struct hiraid_dev *hdev, return -EINVAL; } + if (ret == RW_LENGTH_ZERO) + return ret; + /* 6-byte READ(0x08) or WRITE(0x0A) cdb */ if (scmd->cmd_len == 6) { datalength = (u32)(scmd->cmnd[4] == 0 ? @@ -1230,18 +1071,32 @@ static int hiraid_setup_rw_cmd(struct hiraid_dev *hdev, control |= HIRAID_RW_FUA; } - if (unlikely(datalength > U16_MAX || datalength == 0)) { - dev_err(hdev->dev, "invalid IO for err trans data length[%u]\n", + if (unlikely(datalength > U16_MAX)) { + dev_err(hdev->dev, "invalid IO for illegal transfer data length[%u]\n", datalength); WARN_ON(1); return -EINVAL; } + if (unlikely(datalength == 0)) + return RW_LENGTH_ZERO; + io_cmd->slba = cpu_to_le64(((u64)start_lba_hi << 32) | start_lba_lo); /* 0base for nlb */ io_cmd->nlb = cpu_to_le16((u16)(datalength - 1)); io_cmd->control = cpu_to_le16(control); + mapbuf->cdb_data_len = (u32)((io_cmd->nlb + 1) * sdev->sector_size); + if (mapbuf->cdb_data_len > buf_len) { + /* return DID_ERROR */ + dev_err(hdev->dev, "error: buf len[0x%x] is smaller than actual length[0x%x] sectorsize[0x%x]\n", + buf_len, mapbuf->cdb_data_len, sdev->sector_size); + return -EINVAL; + } else if (mapbuf->cdb_data_len < buf_len) { + dev_warn(hdev->dev, "warn: buf_len[0x%x] cdb_data_len[0x%x] nlb[0x%x] sectorsize[0x%x]\n", + buf_len, mapbuf->cdb_data_len, + io_cmd->nlb, sdev->sector_size); + } return 0; } @@ -1296,14 +1151,17 @@ static bool hiraid_disk_is_hdd_rawdrive(u8 attr) } static int hiraid_setup_io_cmd(struct hiraid_dev *hdev, - struct hiraid_scsi_io_cmd *io_cmd, - struct scsi_cmnd *scmd) + struct hiraid_scsi_io_cmd *io_cmd, struct scsi_cmnd *scmd, + struct hiraid_mapmange *mapbuf) { memcpy(io_cmd->common.cdb, scmd->cmnd, scmd->cmd_len); io_cmd->common.cdb_len = scmd->cmd_len; + /* init cdb_data_len */ + mapbuf->cdb_data_len = cpu_to_le32(scsi_bufflen(scmd)); + if (hiraid_is_rw_scmd(scmd)) - return hiraid_setup_rw_cmd(hdev, &io_cmd->rw, scmd); + return hiraid_setup_rw_cmd(hdev, &io_cmd->rw, scmd, mapbuf); else return hiraid_setup_nonrw_cmd(hdev, &io_cmd->nonrw, scmd); } @@ -1360,7 +1218,7 @@ static int hiraid_io_map_data(struct hiraid_dev *hdev, ret = scsi_dma_map(scmd); if (unlikely(ret < 0)) - return ret; + return SCSI_MLQUEUE_HOST_BUSY; mapbuf->sge_cnt = ret; /* No data to DMA, it may be scsi no-rw command */ @@ -1390,7 +1248,12 @@ static void hiraid_check_status(struct hiraid_mapmange *mapbuf, struct scsi_cmnd *scmd, struct hiraid_completion *cqe) { - scsi_set_resid(scmd, 0); + u32 datalength = cpu_to_le32(scsi_bufflen(scmd)); + + if (datalength > mapbuf->cdb_data_len) + scsi_set_resid(scmd, datalength - mapbuf->cdb_data_len); + else + scsi_set_resid(scmd, 0); switch ((le16_to_cpu(cqe->status) >> 1) & 0x7f) { case SENSE_STATE_OK: @@ -1443,6 +1306,35 @@ static inline void hiraid_query_scmd_tag(struct scsi_cmnd *scmd, u16 *qid, *cid = blk_mq_unique_tag_to_tag(tag); } +static void hiraid_set_ncq_prio(struct scsi_cmnd *scmd, + struct hiraid_scsi_io_cmd *io_cmd, + struct hiraid_sdev_hostdata *hostdata, + struct hiraid_dev *hdev) +{ + struct request *rq = scmd->request; + int class = 0; + + if (hostdata->sata_ncq_prio_enable) { + class = IOPRIO_PRIO_CLASS(req_get_ioprio(rq)); + switch (class) { + case IOPRIO_CLASS_NONE: + break; + case IOPRIO_CLASS_RT: + io_cmd->rw.rsvd2 &= SCSI_RSVD_INIT_SHIFT; + io_cmd->rw.rsvd2 |= (1 << IOPRIO_CLASS_RT); + break; + case IOPRIO_CLASS_BE: + break; + case IOPRIO_CLASS_IDLE: + break; + default: + break; + } + } + dev_log_dbg(hdev->dev, "blk_ncq_prio = %d, scmd_ncq_prio = 0x%02x\n", + class, io_cmd->rw.rsvd2); +} + static int hiraid_queue_command(struct Scsi_Host *shost, struct scsi_cmnd *scmd) { @@ -1452,7 +1344,7 @@ static int hiraid_queue_command(struct Scsi_Host *shost, struct hiraid_sdev_hostdata *hostdata; struct hiraid_scsi_io_cmd io_cmd; struct hiraid_queue *ioq; - struct HIRAID_STREAM_S *tmp_stm = NULL; + struct hiraid_stream *tmp_stream; u16 hwq, cid; int ret; @@ -1482,9 +1374,16 @@ static int hiraid_queue_command(struct Scsi_Host *shost, io_cmd.rw.hdid = cpu_to_le32(hostdata->hdid); io_cmd.rw.cmd_id = cpu_to_le16(cid); - ret = hiraid_setup_io_cmd(hdev, &io_cmd, scmd); + hiraid_set_ncq_prio(scmd, &io_cmd, hostdata, hdev); + + ret = hiraid_setup_io_cmd(hdev, &io_cmd, scmd, mapbuf); if (unlikely(ret)) { - set_host_byte(scmd, DID_ERROR); + if (ret == RW_LENGTH_ZERO) { + scsi_set_resid(scmd, scsi_bufflen(scmd)); + set_host_byte(scmd, DID_OK); + } else { + set_host_byte(scmd, DID_ERROR); + } scmd->scsi_done(scmd); atomic_dec(&ioq->inflight); return 0; @@ -1507,6 +1406,10 @@ static int hiraid_queue_command(struct Scsi_Host *shost, mapbuf->cid = cid; ret = hiraid_io_map_data(hdev, mapbuf, scmd, &io_cmd); if (unlikely(ret)) { + if (ret == SCSI_MLQUEUE_HOST_BUSY) { + atomic_dec(&ioq->inflight); + return ret; + } dev_err(hdev->dev, "io map data err\n"); set_host_byte(scmd, DID_ERROR); scmd->scsi_done(scmd); @@ -1515,20 +1418,20 @@ static int hiraid_queue_command(struct Scsi_Host *shost, } WRITE_ONCE(mapbuf->state, CMD_FLIGHT); - if (hiraid_is_rw_scmd(scmd) && hiraid_disk_is_hdd_rawdrive(hostdata->attr)) { - if (hiraid_detect_if_aging()) + if (hiraid_aging_detect(hdev)) hiraid_aging(hdev); - tmp_stm = hiraid_stream_detect(hdev, &(io_cmd.rw), sdev->id); - if (tmp_stm != NULL) { - hiraid_io_recognition_iterator(tmp_stm, 1); - if (hiraid_recognition_acknowledge(tmp_stm) && - (hiraid_get_stream_num(sdev->id) > 1)) { - if (hiraid_add_io_to_list(ioq, - tmp_stm, io_cmd, sdev->sector_size, sdev->id)) { + tmp_stream = hiraid_stream_detect(hdev, &(io_cmd.rw), sdev->id); + if (tmp_stream != NULL) { + hiraid_io_recognition_iterator(tmp_stream, 1); + if (hiraid_recognition_acknowledge(tmp_stream) && + (sdev->id < HIRAID_MAX_PD_NUM) && + (hdev->hiraid_stream_io_num_per_pd[sdev->id][HIRAID_STREAM_TYPE_TOTAL] > + 1)) { + if (hiraid_add_io_to_list(hdev, ioq, tmp_stream, + io_cmd, sdev->sector_size, sdev->id)) return 0; - } } } } @@ -1666,6 +1569,7 @@ static int hiraid_slave_configure(struct scsi_device *sdev) static void hiraid_shost_init(struct hiraid_dev *hdev) { struct pci_dev *pdev = hdev->pdev; + struct scsi_host_template *hostt; u8 domain, bus; u32 dev_func; @@ -1674,8 +1578,10 @@ static void hiraid_shost_init(struct hiraid_dev *hdev) dev_func = pdev->devfn; hdev->shost->nr_hw_queues = work_mode ? 1 : hdev->online_queues - 1; - hdev->shost->can_queue = hdev->scsi_qd; - + hdev->shost->can_queue = min(hdev->scsi_qd, + le16_to_cpu(hdev->ctrl_info->max_cmds) - + HIRAID_AQ_DEPTH - + HIRAID_TOTAL_PTCMDS(hdev->online_queues - 1)); hdev->shost->sg_tablesize = le16_to_cpu(hdev->ctrl_info->max_num_sge); /* 512B per sector */ hdev->shost->max_sectors = @@ -1689,7 +1595,13 @@ static void hiraid_shost_init(struct hiraid_dev *hdev) hdev->shost->this_id = -1; hdev->shost->unique_id = (domain << 16) | (bus << 8) | dev_func; hdev->shost->max_cmd_len = MAX_CDB_LEN; - hdev->shost->hostt->cmd_size = hiraid_get_max_cmd_size(hdev); + + hostt = (struct scsi_host_template *)hdev->shost->hostt; + hostt->cmd_size = hiraid_get_max_cmd_size(hdev); + + dev_info(hdev->dev, "nr_hw_queues[%u] can_queue[%u] unique_id[%u] cmd_size[%u]\n", + hdev->shost->nr_hw_queues, hdev->shost->can_queue, + hdev->shost->unique_id, hdev->shost->hostt->cmd_size); } static int hiraid_alloc_queue(struct hiraid_dev *hdev, u16 qid, u16 depth) @@ -2505,6 +2417,14 @@ static int hiraid_get_queue_cnt(struct hiraid_dev *hdev, u32 *cnt) return 0; } +static bool hiraid_pci_is_present(struct hiraid_dev *hdev) +{ + if (!pci_device_is_present(hdev->pdev) || + readl(hdev->bar + HIRAID_REG_VS) == 0Xffffffff) + return false; + return true; +} + static int hiraid_setup_io_queues(struct hiraid_dev *hdev) { struct hiraid_queue *adminq = &hdev->queues[0]; @@ -2579,7 +2499,7 @@ static void hiraid_delete_io_queues(struct hiraid_dev *hdev) u8 opcode = HIRAID_ADMIN_DELETE_SQ; u16 i, pass; - if (!pci_device_is_present(hdev->pdev)) { + if (!hiraid_pci_is_present(hdev)) { dev_err(hdev->dev, "pci_device is not present, skip disable io queues\n"); return; } @@ -2619,7 +2539,7 @@ static void hiraid_disable_admin_queue(struct hiraid_dev *hdev, struct hiraid_queue *adminq = &hdev->queues[0]; u16 start, end; - if (pci_device_is_present(hdev->pdev)) { + if (hiraid_pci_is_present(hdev)) { if (shutdown) hiraid_shutdown_control(hdev); else @@ -2850,11 +2770,16 @@ static int hiraid_luntarget_sort(const void *l, const void *r) const struct hiraid_dev_info *rn = r; int l_attr = HIRAID_DEV_INFO_ATTR_BOOT(ln->attr); int r_attr = HIRAID_DEV_INFO_ATTR_BOOT(rn->attr); - /* boot first */ if (l_attr != r_attr) return (r_attr - l_attr); + l_attr = HIRAID_DEV_INFO_ATTR_VD(ln->attr); + r_attr = HIRAID_DEV_INFO_ATTR_VD(rn->attr); + /* vd second */ + if (l_attr != r_attr) + return (r_attr - l_attr); + if (ln->channel == rn->channel) return le16_to_cpu(ln->target) - le16_to_cpu(rn->target); @@ -2892,6 +2817,19 @@ static void hiraid_scan_work(struct work_struct *work) if (HIRAID_DEV_INFO_FLAG_VALID(flag)) { if (!HIRAID_DEV_INFO_FLAG_VALID(org_flag)) { + down_write(&hdev->dev_rwsem); + memcpy(&old_dev[i], &dev[i], + sizeof(struct hiraid_dev_info)); + memcpy(&new_dev[count++], &dev[i], + sizeof(struct hiraid_dev_info)); + up_write(&hdev->dev_rwsem); + } else if (HIRAID_DEV_INFO_FLAG_DELETE_OR_ADD(flag, + org_flag)) { + down_write(&hdev->dev_rwsem); + old_dev[i].flag &= 0xfe; + up_write(&hdev->dev_rwsem); + hiraid_delete_device(hdev, &old_dev[i]); + down_write(&hdev->dev_rwsem); memcpy(&old_dev[i], &dev[i], sizeof(struct hiraid_dev_info)); @@ -3044,7 +2982,7 @@ static void hiraid_bsg_buf_unmap(struct hiraid_dev *hdev, struct bsg_job *job) } static int hiraid_bsg_buf_map(struct hiraid_dev *hdev, struct bsg_job *job, - struct hiraid_admin_command *cmd) + struct hiraid_admin_command *cmd, u16 qid) { struct hiraid_bsg_request *bsg_req = job->request; struct request *rq = blk_mq_rq_from_pdu(job); @@ -3058,24 +2996,26 @@ static int hiraid_bsg_buf_map(struct hiraid_dev *hdev, struct bsg_job *job, mapbuf->sgl = job->request_payload.sg_list; mapbuf->len = job->request_payload.payload_len; mapbuf->page_cnt = -1; + mapbuf->hiraidq = &hdev->queues[qid]; if (unlikely(mapbuf->sge_cnt == 0)) goto out; - mapbuf->use_sgl = !hiraid_is_prp(hdev, mapbuf->sgl, mapbuf->sge_cnt); - ret = dma_map_sg_attrs(hdev->dev, mapbuf->sgl, mapbuf->sge_cnt, - dma_dir, DMA_ATTR_NO_WARN); + dma_dir, DMA_ATTR_NO_WARN); if (!ret) goto out; - if ((mapbuf->use_sgl == (bool)true) && + mapbuf->use_sgl = !hiraid_is_prp(hdev, mapbuf->sgl, mapbuf->sge_cnt); + + if ((mapbuf->use_sgl == true) && (bsg_req->msgcode == HIRAID_BSG_IOPTHRU) && - (hdev->ctrl_info->pt_use_sgl != (bool)false)) { - ret = hiraid_build_passthru_sgl(hdev, cmd, mapbuf); + (hdev->ctrl_info->pt_use_sgl != false)) { + ret = hiraid_build_sgl(hdev, (struct hiraid_scsi_io_cmd *)cmd, + mapbuf); } else { mapbuf->use_sgl = false; - ret = hiraid_build_passthru_prp(hdev, mapbuf); + ret = hiraid_build_prp(hdev, mapbuf); cmd->common.dptr.prp1 = cpu_to_le64(sg_dma_address(mapbuf->sgl)); cmd->common.dptr.prp2 = cpu_to_le64(mapbuf->first_dma); @@ -3156,9 +3096,7 @@ static int hiraid_init_control_info(struct hiraid_dev *hdev) if (hdev->ctrl_info->asynevent > HIRAID_ASYN_COMMANDS) hdev->ctrl_info->asynevent = HIRAID_ASYN_COMMANDS; - hdev->scsi_qd = work_mode ? - le16_to_cpu(hdev->ctrl_info->max_cmds) : - (hdev->ioq_depth - HIRAID_PTHRU_CMDS_PERQ); + hdev->scsi_qd = hdev->ioq_depth - HIRAID_PTHRU_CMDS_PERQ; return 0; } @@ -3191,7 +3129,7 @@ static int hiraid_user_send_admcmd(struct hiraid_dev *hdev, struct bsg_job *job) admin_cmd.common.cdw14 = cpu_to_le32(ptcmd->cdw14); admin_cmd.common.cdw15 = cpu_to_le32(ptcmd->cdw15); - status = hiraid_bsg_buf_map(hdev, job, &admin_cmd); + status = hiraid_bsg_buf_map(hdev, job, &admin_cmd, 0); if (status) { dev_err(hdev->dev, "err, map data failed\n"); return status; @@ -3251,19 +3189,13 @@ static void hiraid_free_io_ptcmds(struct hiraid_dev *hdev) } static int hiraid_put_io_sync_request(struct hiraid_dev *hdev, - struct hiraid_scsi_io_cmd *io_cmd, - u32 *result, u32 *reslen, u32 timeout) + struct hiraid_scsi_io_cmd *io_cmd, struct hiraid_cmd *pt_cmd, + u32 *result, u32 *reslen, u32 timeout) { int ret; dma_addr_t buffer_phy; struct hiraid_queue *ioq; void *sense_addr = NULL; - struct hiraid_cmd *pt_cmd = hiraid_get_cmd(hdev, HIRAID_CMD_PTHRU); - - if (!pt_cmd) { - dev_err(hdev->dev, "err, get ioq cmd failed\n"); - return -EFAULT; - } timeout = timeout ? timeout : ADMIN_TIMEOUT; @@ -3271,8 +3203,9 @@ static int hiraid_put_io_sync_request(struct hiraid_dev *hdev, ioq = &hdev->queues[pt_cmd->qid]; if (work_mode) { - ret = ((pt_cmd->qid - 1) * HIRAID_PTHRU_CMDS_PERQ + - pt_cmd->cid) * SCSI_SENSE_BUFFERSIZE; + ret = ((pt_cmd->qid - 1) * + HIRAID_PTHRU_CMDS_PERQ + pt_cmd->cid) * + SCSI_SENSE_BUFFERSIZE; sense_addr = hdev->sense_buffer_virt + ret; buffer_phy = hdev->sense_buffer_phy + ret; } else { @@ -3293,8 +3226,6 @@ static int hiraid_put_io_sync_request(struct hiraid_dev *hdev, (le32_to_cpu(io_cmd->common.cdw3[0]) & 0xffff)); hiraid_admin_timeout(hdev, pt_cmd); - - hiraid_put_cmd(hdev, pt_cmd, HIRAID_CMD_PTHRU); return -ETIME; } @@ -3305,8 +3236,6 @@ static int hiraid_put_io_sync_request(struct hiraid_dev *hdev, } } - hiraid_put_cmd(hdev, pt_cmd, HIRAID_CMD_PTHRU); - return pt_cmd->status; } @@ -3316,6 +3245,7 @@ static int hiraid_user_send_ptcmd(struct hiraid_dev *hdev, struct bsg_job *job) (struct hiraid_bsg_request *)(job->request); struct hiraid_passthru_io_cmd *cmd = &(bsg_req->pthrucmd); struct hiraid_scsi_io_cmd pthru_cmd; + struct hiraid_cmd *pt_cmd; int status = 0; u32 timeout = msecs_to_jiffies(cmd->timeout_ms); // data len is 4k before use sgl, now len is 1M @@ -3358,16 +3288,21 @@ static int hiraid_user_send_ptcmd(struct hiraid_dev *hdev, struct bsg_job *job) pthru_cmd.common.cdw26[2] = cpu_to_le32(cmd->cdw26[2]); pthru_cmd.common.cdw26[3] = cpu_to_le32(cmd->cdw26[3]); + pt_cmd = hiraid_get_cmd(hdev, HIRAID_CMD_PTHRU); + if (pt_cmd == NULL) { + dev_err(hdev->dev, "err, get ioq cmd failed\n"); + return -EFAULT; + } + status = hiraid_bsg_buf_map(hdev, job, - (struct hiraid_admin_command *)&pthru_cmd); + (struct hiraid_admin_command *)&pthru_cmd, pt_cmd->qid); if (status) { dev_err(hdev->dev, "err, map data failed\n"); - return status; + goto ret; } - status = hiraid_put_io_sync_request(hdev, &pthru_cmd, job->reply, - &job->reply_len, timeout); - + status = hiraid_put_io_sync_request(hdev, &pthru_cmd, pt_cmd, + job->reply, &job->reply_len, timeout); if (status) dev_info(hdev->dev, "opcode[0x%x] subopcode[0x%x] status[0x%x] replylen[%d]\n", cmd->opcode, cmd->info_1.subopcode, @@ -3375,6 +3310,9 @@ static int hiraid_user_send_ptcmd(struct hiraid_dev *hdev, struct bsg_job *job) hiraid_bsg_buf_unmap(hdev, job); +ret: + hiraid_put_cmd(hdev, pt_cmd, HIRAID_CMD_PTHRU); + return status; } @@ -3396,23 +3334,28 @@ static bool hiraid_check_scmd_finished(struct scsi_cmnd *scmd) return false; } -static enum blk_eh_timer_return hiraid_timed_out(struct scsi_cmnd *scmd) +#define EH_TIMED_RET_TP enum blk_eh_timer_return +#define EH_TIMED_RET_DONE BLK_EH_DONE +#define EH_TIMED_RET_NOT_HANDLED BLK_EH_DONE +#define EH_TIMED_RET_RESET_TIMER BLK_EH_RESET_TIMER +static EH_TIMED_RET_TP hiraid_timed_out(struct scsi_cmnd *scmd) { struct hiraid_mapmange *mapbuf = scsi_cmd_priv(scmd); unsigned int timeout = scmd->device->request_queue->rq_timeout; if (hiraid_check_scmd_finished(scmd)) - goto out; + return EH_TIMED_RET_DONE; - if (time_after(jiffies, scmd->jiffies_at_alloc + timeout)) { + if (time_after_eq(jiffies, scmd->jiffies_at_alloc + timeout)) { + scmd_printk(KERN_WARNING, scmd, "timeout is work[%d], need abort\n", + mapbuf->state); if (cmpxchg(&mapbuf->state, CMD_FLIGHT, CMD_TIMEOUT) == CMD_FLIGHT) - return BLK_EH_DONE; + return EH_TIMED_RET_NOT_HANDLED; } -out: - return BLK_EH_RESET_TIMER; -} + return EH_TIMED_RET_RESET_TIMER; +} /* send abort command by admin queue temporary */ static int hiraid_send_abort_cmd(struct hiraid_dev *hdev, u32 hdid, u16 qid, u16 cid) @@ -3535,7 +3478,7 @@ static int hiraid_dev_disable(struct hiraid_dev *hdev, bool shutdown) struct hiraid_queue *adminq = &hdev->queues[0]; u16 start, end; - if (pci_device_is_present(hdev->pdev)) { + if (hiraid_pci_is_present(hdev)) { if (shutdown) hiraid_shutdown_control(hdev); else @@ -3733,11 +3676,19 @@ static int hiraid_abort(struct scsi_cmnd *scmd) hiraid_check_scmd_finished(scmd)) return SUCCESS; + if (hdev->state != DEV_LIVE) + return SUCCESS; + hostdata = scmd->device->hostdata; cid = mapbuf->cid; hwq = mapbuf->hiraidq->qid; - if (hiraid_check_io_list(hostdata->hdid, cid, hwq)) { + mutex_lock(&g_stream_operation_mutex); + ret = hiraid_stream_delete_io_list(hdev, hostdata->hdid, + cid, hwq, HIRAID_STREAM_TYPE_DELETE_SINGLE_IO); + mutex_unlock(&g_stream_operation_mutex); + + if (ret == 1) { dev_warn(hdev->dev, "find cid[%d] qid[%d] in host, abort succ\n", cid, hwq); return SUCCESS; @@ -3749,14 +3700,14 @@ static int hiraid_abort(struct scsi_cmnd *scmd) ret = hiraid_wait_io_completion(mapbuf); if (ret) { dev_warn(hdev->dev, "cid[%d] qid[%d] abort failed, not found\n", - cid, hwq); + cid, hwq); return FAILED; } dev_warn(hdev->dev, "cid[%d] qid[%d] abort succ\n", cid, hwq); return SUCCESS; } dev_warn(hdev->dev, "cid[%d] qid[%d] abort failed, timeout\n", - cid, hwq); + cid, hwq); return FAILED; } @@ -3778,7 +3729,10 @@ static int hiraid_scsi_reset(struct scsi_cmnd *scmd, enum hiraid_rst_type rst) if ((ret == 0) || (ret == FW_EH_DEV_NONE && rst == HIRAID_RESET_TARGET)) { if (rst == HIRAID_RESET_TARGET) { - hiraid_delete_single_pd_io_list(hostdata->hdid); + mutex_lock(&g_stream_operation_mutex); + hiraid_stream_delete_io_list(hdev, hostdata->hdid, + 0, 0, HIRAID_STREAM_TYPE_DELETE_SINGLE_IO_LIST); + mutex_unlock(&g_stream_operation_mutex); ret = wait_tgt_reset_io_done(scmd); if (ret) { dev_warn(hdev->dev, "sdev[%d:%d] target has %d peding cmd, target reset failed\n", @@ -3818,9 +3772,13 @@ static int hiraid_host_reset(struct scsi_cmnd *scmd) dev_warn(hdev->dev, "sdev[%d:%d] send host reset\n", scmd->device->channel, scmd->device->id); - hiraid_delete_all_io_list(); - if (hiraid_reset_work_sync(hdev) == -EBUSY) + mutex_lock(&g_stream_operation_mutex); + hiraid_stream_delete_io_list(hdev, 0, 0, 0, + HIRAID_STREAM_TYPE_DELETE_ALL_IO_LIST); + mutex_unlock(&g_stream_operation_mutex); + if (hiraid_reset_work_sync(hdev) == -EBUSY) { flush_work(&hdev->reset_work); + } if (hdev->state != DEV_LIVE) { dev_warn(hdev->dev, "sdev[%d:%d] host reset failed\n", @@ -3852,7 +3810,10 @@ static pci_ers_result_t hiraid_pci_error_detected(struct pci_dev *pdev, scsi_block_requests(hdev->shost); hiraid_dev_state_trans(hdev, DEV_RESETTING); - hiraid_delete_all_io_list(); + mutex_lock(&g_stream_operation_mutex); + hiraid_stream_delete_io_list(hdev, 0, 0, 0, + HIRAID_STREAM_TYPE_DELETE_ALL_IO_LIST); + mutex_unlock(&g_stream_operation_mutex); return PCI_ERS_RESULT_NEED_RESET; case pci_channel_io_perm_failure: @@ -3888,9 +3849,24 @@ static void hiraid_reset_pci_finish(struct pci_dev *pdev) { struct hiraid_dev *hdev = pci_get_drvdata(pdev); + if (hiraid_reset_work_sync(hdev) == -EBUSY) + flush_work(&hdev->reset_work); + dev_info(hdev->dev, "enter hiraid reset finish\n"); } +static void hiraid_reset_pci_prepare(struct pci_dev *pdev) +{ + struct hiraid_dev *hdev = pci_get_drvdata(pdev); + + mutex_lock(&g_stream_operation_mutex); + hiraid_stream_delete_io_list(hdev, 0, 0, 0, + HIRAID_STREAM_TYPE_DELETE_ALL_IO_LIST); + mutex_unlock(&g_stream_operation_mutex); + msleep(100); + dev_info(hdev->dev, "exit hiraid reset prepare\n"); +} + static ssize_t csts_pp_show(struct device *cdev, struct device_attribute *attr, char *buf) { @@ -3898,7 +3874,7 @@ static ssize_t csts_pp_show(struct device *cdev, struct hiraid_dev *hdev = shost_priv(shost); int ret = -1; - if (pci_device_is_present(hdev->pdev)) { + if (hiraid_pci_is_present(hdev)) { ret = (readl(hdev->bar + HIRAID_REG_CSTS) & HIRAID_CSTS_PP_MASK); ret >>= HIRAID_CSTS_PP_SHIFT; @@ -3914,7 +3890,7 @@ static ssize_t csts_shst_show(struct device *cdev, struct hiraid_dev *hdev = shost_priv(shost); int ret = -1; - if (pci_device_is_present(hdev->pdev)) { + if (hiraid_pci_is_present(hdev)) { ret = (readl(hdev->bar + HIRAID_REG_CSTS) & HIRAID_CSTS_SHST_MASK); ret >>= HIRAID_CSTS_SHST_SHIFT; @@ -3930,7 +3906,7 @@ static ssize_t csts_cfs_show(struct device *cdev, struct hiraid_dev *hdev = shost_priv(shost); int ret = -1; - if (pci_device_is_present(hdev->pdev)) { + if (hiraid_pci_is_present(hdev)) { ret = (readl(hdev->bar + HIRAID_REG_CSTS) & HIRAID_CSTS_CFS_MASK); ret >>= HIRAID_CSTS_CFS_SHIFT; @@ -3946,7 +3922,7 @@ static ssize_t csts_rdy_show(struct device *cdev, struct hiraid_dev *hdev = shost_priv(shost); int ret = -1; - if (pci_device_is_present(hdev->pdev)) + if (hiraid_pci_is_present(hdev)) ret = (readl(hdev->bar + HIRAID_REG_CSTS) & HIRAID_CSTS_RDY); return snprintf(buf, PAGE_SIZE, "%d\n", ret); @@ -4239,22 +4215,78 @@ static ssize_t dispatch_hwq_store(struct device *dev, return strlen(buf); } +static ssize_t sata_ncq_prio_enable_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct hiraid_sdev_hostdata *hostdata; + + hostdata = to_scsi_device(dev)->hostdata; + return snprintf(buf, PAGE_SIZE, "%d\n", hostdata->sata_ncq_prio_enable); +} + +bool scsi_ncq_prio_support(struct scsi_device *sdev, struct hiraid_dev *hdev) +{ + unsigned char *buf; + bool ncq_prio_supp = false; + + if (!scsi_device_supports_vpd(sdev)) { + dev_err(hdev->dev, "scsi device not support vpd\n"); + return ncq_prio_supp; + } + + buf = kzalloc(HIRAID_SCSI_VPD_LEN, GFP_KERNEL); + if (!buf) + return ncq_prio_supp; + + if (!scsi_get_vpd_page(sdev, 0x89, buf, HIRAID_SCSI_VPD_LEN)) + ncq_prio_supp = (buf[NCQ_PRIO_SUPPORT_BYTE] >> + VPD_NCQ_PRIO_SUPPORT_BIT) & 1; + + kfree(buf); + return ncq_prio_supp; +} + +static ssize_t sata_ncq_prio_enable_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + struct scsi_device *sdev = to_scsi_device(dev); + struct hiraid_sdev_hostdata *hostdata = sdev->hostdata; + struct hiraid_dev *hdev = shost_priv(sdev->host); + bool ncq_prio_enable = 0; + + if (kstrtobool(buf, &ncq_prio_enable)) + return -EINVAL; + + if (!scsi_ncq_prio_support(sdev, hdev)) { + dev_err(hdev->dev, "sdev[%d:%d] not support ncq prio\n", + sdev->channel, sdev->id); + return -EINVAL; + } + + hostdata->sata_ncq_prio_enable = ncq_prio_enable; + + return strlen(buf); +} + static DEVICE_ATTR_RO(raid_level); static DEVICE_ATTR_RO(raid_state); static DEVICE_ATTR_RO(raid_resync); static DEVICE_ATTR_RW(dispatch_hwq); +static DEVICE_ATTR_RW(sata_ncq_prio_enable); static struct device_attribute *hiraid_dev_attrs[] = { &dev_attr_raid_state, &dev_attr_raid_level, &dev_attr_raid_resync, &dev_attr_dispatch_hwq, + &dev_attr_sata_ncq_prio_enable, NULL, }; static struct pci_error_handlers hiraid_err_handler = { .error_detected = hiraid_pci_error_detected, .slot_reset = hiraid_pci_slot_reset, + .reset_prepare = hiraid_reset_pci_prepare, .reset_done = hiraid_reset_pci_finish, }; @@ -4518,14 +4550,17 @@ static void hiraid_remove(struct pci_dev *pdev) dev_info(hdev->dev, "enter hiraid remove\n"); - kthread_stop(g_hiraid_submit_task); - hiraid_delete_all_io_list(); + kthread_stop(hdev->hiraid_stream_submit_task); + mutex_lock(&g_stream_operation_mutex); + hiraid_stream_delete_io_list(hdev, 0, 0, 0, + HIRAID_STREAM_TYPE_DELETE_ALL_IO_LIST); + mutex_unlock(&g_stream_operation_mutex); hiraid_destroy_io_stream_resource(hdev); hiraid_dev_state_trans(hdev, DEV_DELETING); flush_work(&hdev->reset_work); - if (!pci_device_is_present(pdev)) + if (!hiraid_pci_is_present(hdev)) hiraid_flush_running_cmds(hdev); hiraid_unregist_bsg(hdev); @@ -4548,8 +4583,12 @@ static void hiraid_remove(struct pci_dev *pdev) static const struct pci_device_id hiraid_hw_card_ids[] = { { PCI_DEVICE(PCI_VENDOR_ID_HUAWEI_LOGIC, HIRAID_SERVER_DEVICE_HBA_DID) }, + { PCI_DEVICE(PCI_VENDOR_ID_HUAWEI_LOGIC, + HIRAID_SERVER_DEVICE_HBAS_DID) }, { PCI_DEVICE(PCI_VENDOR_ID_HUAWEI_LOGIC, HIRAID_SERVER_DEVICE_RAID_DID) }, + { PCI_DEVICE(PCI_VENDOR_ID_HUAWEI_LOGIC, + HIRAID_SERVER_DEVICE_RAIDS_DID) }, { 0, } }; MODULE_DEVICE_TABLE(pci, hiraid_hw_card_ids); @@ -4572,22 +4611,9 @@ static int __init hiraid_init(void) if (!work_queue) return -ENOMEM; - hiraid_class = class_create(THIS_MODULE, "hiraid"); - if (IS_ERR(hiraid_class)) { - ret = PTR_ERR(hiraid_class); - goto destroy_wq; - } - ret = pci_register_driver(&hiraid_driver); if (ret < 0) - goto destroy_class; - - return 0; - -destroy_class: - class_destroy(hiraid_class); -destroy_wq: - destroy_workqueue(work_queue); + destroy_workqueue(work_queue); return ret; } @@ -4595,7 +4621,6 @@ destroy_wq: static void __exit hiraid_exit(void) { pci_unregister_driver(&hiraid_driver); - class_destroy(hiraid_class); destroy_workqueue(work_queue); } -- 2.45.1.windows.1

2 5

[PATCH OLK-6.6] mm/sharepool: Support numa_maps
by Yin Tirui 09 Jun '26

09 Jun '26

Support numa_maps Signed-off-by: Yin Tirui <yintirui(a)huawei.com> --- mm/share_pool.c | 130 ++++++++++++++++++++++++++++++++++++++++++++++-- 1 file changed, 126 insertions(+), 4 deletions(-) diff --git a/mm/share_pool.c b/mm/share_pool.c index b6e543f1cefc..4e11c60db2fc 100644 --- a/mm/share_pool.c +++ b/mm/share_pool.c @@ -3984,11 +3984,125 @@ static void get_process_non_sp_res(unsigned long total_rss, unsigned long shmem, static void print_process_prot(struct seq_file *seq, unsigned long prot) { if (prot == PROT_READ) - seq_puts(seq, "R"); + seq_printf(seq, "%-4s ", "R"); else if (prot == (PROT_READ | PROT_WRITE)) - seq_puts(seq, "RW"); + seq_printf(seq, "%-4s ", "RW"); else - seq_puts(seq, "-"); + seq_printf(seq, "%-4s ", "-"); +} + +static int sp_numa_pte_entry(pte_t *pte, unsigned long addr, + unsigned long next, struct mm_walk *walk) +{ + unsigned long *size_array = walk->private; + pte_t entry = ptep_get(pte); + struct page *page; + int nid; + + if (pte_present(entry)) { + if (is_zero_pfn(pte_pfn(entry))) + return 0; + + page = pte_page(entry); + if (page && !PageReserved(page)) { + nid = page_to_nid(page); + if (nid >= 0 && nid < MAX_NUMNODES) + size_array[nid] += PAGE_SIZE; + } + } + return 0; +} + +static int sp_numa_pmd_entry(pmd_t *pmd, unsigned long addr, + unsigned long next, struct mm_walk *walk) +{ + unsigned long *size_array = walk->private; + pmd_t entry = pmdp_get(pmd); + struct page *page; + int nid; + + if (pmd_present(entry) && pmd_huge(entry)) { + if (is_zero_pfn(pmd_pfn(entry))) + goto skip; + + page = pmd_page(entry); + if (page && !PageReserved(page)) { + nid = page_to_nid(page); + if (nid >= 0 && nid < MAX_NUMNODES) + size_array[nid] += PMD_SIZE; + } +skip: + walk->action = ACTION_CONTINUE; + } + return 0; +} + +static int sp_numa_hugetlb_entry(pte_t *ptep, unsigned long hmask, + unsigned long addr, unsigned long next, + struct mm_walk *walk) +{ + unsigned long *size_array = walk->private; + pte_t pte = huge_ptep_get(ptep); + struct page *page; + int nid; + + if (pte_present(pte)) { + if (is_zero_pfn(pte_pfn(pte))) + return 0; + page = pte_page(pte); + if (page && !PageReserved(page)) { + nid = page_to_nid(page); + if (nid >= 0 && nid < MAX_NUMNODES) + size_array[nid] += (~hmask + 1); + } + } + return 0; +} + +static const struct mm_walk_ops sp_numa_walk_ops = { + .pte_entry = sp_numa_pte_entry, + .pmd_entry = sp_numa_pmd_entry, + .hugetlb_entry = sp_numa_hugetlb_entry, +}; + +static void print_process_numa_maps(struct seq_file *seq, struct sp_group_node *spg_node) +{ + struct mm_struct *mm = spg_node->master->mm; + struct sp_group *spg = spg_node->spg; + unsigned long size[MAX_NUMNODES] = {0}; + struct rb_node *p, *n; + struct sp_area *spa; + int nid, ret; + + if (!mmget_not_zero(mm)) + goto out; + + for (p = rb_first(&spg->spa_root); p; p = n) { + n = rb_next(p); + spa = container_of(p, struct sp_area, spg_link); + + if (spa->type != SPA_TYPE_ALLOC || + spa->applier != spg_node->master->tgid) + continue; + + mmap_read_lock(mm); + ret = walk_page_range(mm, spa->va_start, spa->va_start + spa_size(spa), + &sp_numa_walk_ops, size); + mmap_read_unlock(mm); + if (ret < 0) { + pr_err_ratelimited("walk_page_range failed %d for 0x%lx.\n", + ret, spa->va_start); + continue; + } + + cond_resched(); + } + + mmput_async(mm); + +out: + for_each_node_state(nid, N_MEMORY) + seq_printf(seq, "%-9ld ", byte2kb(size[nid])); } static void spa_stat_of_mapping_show(struct seq_file *seq, struct sp_mapping *spm) @@ -4180,6 +4294,7 @@ static int proc_usage_by_group(int id, void *p, void *data) page2kb(mm->total_vm), page2kb(total_rss), page2kb(shmem)); print_process_prot(seq, spg_node->prot); + print_process_numa_maps(seq, spg_node); seq_putc(seq, '\n'); } up_read(&spg->rw_lock); @@ -4190,6 +4305,8 @@ static int proc_usage_by_group(int id, void *p, void *data) static int proc_group_usage_show(struct seq_file *seq, void *offset) { + int nid; + if (!should_show_statistics()) return -EPERM; @@ -4197,10 +4314,15 @@ static int proc_group_usage_show(struct seq_file *seq, void *offset) spa_overview_show(seq); /* print the file header */ - seq_printf(seq, "%-8s %-8s %-9s %-9s %-9s %-8s %-7s %-7s %-4s\n", + seq_printf(seq, "%-8s %-8s %-9s %-9s %-9s %-8s %-7s %-7s %-4s ", "PID", "Group_ID", "SP_ALLOC", "SP_K2U", "SP_RES", "VIRT", "RES", "Shm", "PROT"); + for_each_node_state(nid, N_MEMORY) + seq_printf(seq, "N%-8d ", nid); + + seq_putc(seq, '\n'); + down_read(&sp_global_sem); idr_for_each(&sp_group_idr, proc_usage_by_group, seq); up_read(&sp_global_sem); -- 2.43.0

2 1

[PATCH OLK-6.6 0/4] Share Pool Sysctl Support
by Yin Tirui 09 Jun '26

09 Jun '26

-- 2.43.0

2 5

[PATCH OLK-6.6 0/11] Share Pool Remote Support
by Yin Tirui 09 Jun '26

09 Jun '26

-- 2.43.0

2 12

[PATCH] tipc: restrict socket queue dumps in enqueue tracepoints
by Li Xiasong 09 Jun '26

09 Jun '26

tipc_sk_enqueue() runs with sk->sk_lock.slock held while the socket is owned by user context. The spinlock protects the backlog queue in this path, but it does not serialize against the socket owner consuming or purging sk_receive_queue. KASAN reported: CPU: 14 UID: 0 PID: 1050 Comm: tipc3 Not tainted 7.1.0-rc6+ #126 PREEMPT(lazy) Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014 Call Trace: <TASK> dump_stack_lvl+0x76/0xa0 lib/dump_stack.c:123 print_report+0xce/0x5b0 mm/kasan/report.c:482 kasan_report+0xc6/0x100 mm/kasan/report.c:597 __asan_report_load4_noabort+0x14/0x30 mm/kasan/report_generic.c:380 tipc_skb_dump+0x1327/0x16f0 net/tipc/trace.c:73 tipc_list_dump+0x208/0x2e0 net/tipc/trace.c:187 tipc_sk_dump+0xaf6/0xd60 net/tipc/socket.c:3996 trace_event_raw_event_tipc_sk_class+0x312/0x5a0 net/tipc/trace.h:188 tipc_sk_rcv+0xb1d/0x1d50 net/tipc/socket.c:2497 tipc_node_xmit+0x1c3/0x1440 net/tipc/node.c:1689 __tipc_sendmsg+0x97a/0x1440 net/tipc/socket.c:1512 tipc_sendmsg+0x52/0x80 net/tipc/socket.c:1400 sock_sendmsg+0x2f6/0x3e0 net/socket.c:825 splice_to_socket+0x7f9/0x1010 fs/splice.c:884 do_splice+0xe21/0x2330 fs/splice.c:936 __do_splice+0x153/0x260 fs/splice.c:1431 __x64_sys_splice+0x150/0x230 fs/splice.c:1616 x64_sys_call+0xeb5/0x2790 arch/x86/entry/syscall_64.c:41 do_syscall_64+0xf3/0x620 arch/x86/entry/syscall_64.c:63 entry_SYSCALL_64_after_hwframe+0x76/0x7e arch/x86/entry/entry_64.S:130 RIP: 0033:0x71624e8aafe2 Code: 08 0f 85 71 3a ff ff 49 89 fb 48 89 f0 48 89 d7 48 89 ce 4c 89 c2 4d 89 ca 4c 8b 44 24 08 4c 8b 4c 24 10 4c 89 5c 24 08 0f 05 <c3> 66 2e 0f 1f 84 00 00 00 00 00 66 2e 0f 1f 84 00 00 00 00 00 66 RSP: 002b:0000716157ffed68 EFLAGS: 00000246 ORIG_RAX: 0000000000000113 RAX: ffffffffffffffda RBX: 0000716157fff6c0 RCX: 000071624e8aafe2 RDX: 000000000000005f RSI: 0000000000000000 RDI: 0000000000000066 RBP: 0000716157ffed90 R08: 0000000000008000 R09: 0000000000000001 R10: 0000000000000000 R11: 0000000000000246 R12: ffffffffffffff00 R13: 0000000000000021 R14: 0000000000000000 R15: 00007fff89799c40 </TASK> The TIPC_DUMP_ALL tracepoints in tipc_sk_enqueue() also dump sk_receive_queue and can therefore dereference skbs that the socket owner has already dequeued or freed. Restrict these dumps to TIPC_DUMP_SK_BKLGQ, which matches the queue protected by the held spinlock. Keep the change limited to the enqueue path, where the unsafe queue dump is reachable while the socket is owned by user context. Fixes: 01e661ebfbad ("tipc: add trace_events for tipc socket") Cc: stable(a)vger.kernel.org Signed-off-by: Li Xiasong <lixiasong1(a)huawei.com> --- net/tipc/socket.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/net/tipc/socket.c b/net/tipc/socket.c index 9329919fb07f..6b761003bcd1 100644 --- a/net/tipc/socket.c +++ b/net/tipc/socket.c @@ -2452,17 +2452,17 @@ static void tipc_sk_enqueue(struct sk_buff_head *inputq, struct sock *sk, atomic_set(dcnt, 0); lim = rcvbuf_limit(sk, skb) + atomic_read(dcnt); if (likely(!sk_add_backlog(sk, skb, lim))) { - trace_tipc_sk_overlimit1(sk, skb, TIPC_DUMP_ALL, + trace_tipc_sk_overlimit1(sk, skb, TIPC_DUMP_SK_BKLGQ, "bklg & rcvq >90% allocated!"); continue; } - trace_tipc_sk_dump(sk, skb, TIPC_DUMP_ALL, "err_overload!"); + trace_tipc_sk_dump(sk, skb, TIPC_DUMP_SK_BKLGQ, "err_overload!"); /* Overload => reject message back to sender */ onode = tipc_own_addr(sock_net(sk)); sk_drops_inc(sk); if (tipc_msg_reverse(onode, &skb, TIPC_ERR_OVERLOAD)) { - trace_tipc_sk_rej_msg(sk, skb, TIPC_DUMP_ALL, + trace_tipc_sk_rej_msg(sk, skb, TIPC_DUMP_SK_BKLGQ, "@sk_enqueue!"); __skb_queue_tail(xmitq, skb); } -- 2.34.1

1 0

[PATCH openEuler-1.0-LTS 0/4] mm: introduce memcg-based ZRAM-Reclaim policy
by Ze Zuo 09 Jun '26

09 Jun '26

This series introduces ZRAM-Reclaim, a memcg-aware reclaim policy for ZRAM + HDD environments. The policy uses recent anon swapin cost and file refault cost to adjust reclaim scan balance under memory pressure. The feature is optional and controlled by CONFIG_ZRAM_RECLAIM plus the per-memcg memory.zram_reclaim interface. The interface accepts "<enable> <ratio>". Enabling without an explicit ratio uses the default ratio, and disabling is represented as "0 0". The current implementation limits ratio to 60. Patch 1 adds the policy core, memcg control file, Kconfig wiring, sysctl documentation and scan-balance hook. Patch 2 connects the anon swapin and file refault paths to the policy input signals. Patch 3 adds an optional node reclaim mode for waking kswapd during node reclaim. Patch 4 enables CONFIG_ZRAM_RECLAIM in the arm64 openEuler defconfig. Ze Zuo (4): mm: add ZRAM-Reclaim infrastructure and policy core mm: hook ZRAM-Reclaim into swap and refault paths mm: node_reclaim: add optional kswapd wakeup mode config: enable ZRAM_RECLAIM in arm64 defconfig Documentation/sysctl/vm.txt | 1 + arch/arm64/configs/openeuler_defconfig | 1 + include/linux/memcontrol.h | 7 + include/linux/zram_reclaim.h | 29 ++ mm/Kconfig | 23 ++ mm/Makefile | 1 + mm/internal.h | 5 +- mm/memcontrol.c | 6 + mm/memory.c | 17 + mm/page_alloc.c | 3 +- mm/vmscan.c | 21 +- mm/workingset.c | 5 + mm/zram_reclaim.c | 411 +++++++++++++++++++++++++ 13 files changed, 526 insertions(+), 4 deletions(-) create mode 100644 include/linux/zram_reclaim.h create mode 100644 mm/zram_reclaim.c -- 2.25.1

2 5

[PATCH 0/4] rmfork code
by Qi Xi 09 Jun '26

09 Jun '26

Qi Xi (4): skip-pages: add --skip-pages option to omit pages.img generation rmfork: add --enable-rmfork for kernel-side checkpoint/restore rmfork: add restorer blob support for kernel-side memory restore rmfork: integrate into dump/restore infrastructure criu/Makefile.crtools | 1 + criu/config.c | 26 +++++ criu/cr-dump.c | 18 +++- criu/cr-restore.c | 14 +++ criu/include/cr_options.h | 2 + criu/include/linux/rseq.h | 2 +- criu/include/restorer.h | 3 + criu/include/rmfork.h | 28 ++++++ criu/mem.c | 21 +++- criu/page-xfer.c | 44 ++++++-- criu/pie/restorer.c | 204 +++++++++++++++++++++++++++++--------- criu/rmfork.c | 156 +++++++++++++++++++++++++++++ criu/shmem.c | 19 ++-- criu/tty.c | 2 +- 14 files changed, 471 insertions(+), 69 deletions(-) create mode 100644 criu/include/rmfork.h create mode 100644 criu/rmfork.c -- 2.53.0

1 0

[PATCH 0/4] rmfork code
by Qi Xi 09 Jun '26

09 Jun '26

Qi Xi (4): rmfork: add kernel-side memory checkpoint/restore for CRIU rmfork arm64: add pmemmem kernel parameter for persistent memory rmfork: refactor ubmem allocator, add sysfs reset, expand limits rmfork: switch page restore from vm_insert_page to copy_to_user Kconfig | 1 + arch/arm64/include/uapi/asm/unistd.h | 6 + arch/arm64/kernel/setup.c | 5 + arch/arm64/mm/init.c | 75 +++ include/linux/ioport.h | 1 + include/linux/mm.h | 4 + include/uapi/linux/rmfork.h | 26 + kernel/Kconfig.rmfork | 8 + kernel/Makefile | 1 + kernel/rmfork.c | 830 +++++++++++++++++++++++++++ kernel/sys_ni.c | 3 + lib/Kconfig | 6 + 12 files changed, 966 insertions(+) create mode 100644 include/uapi/linux/rmfork.h create mode 100644 kernel/Kconfig.rmfork create mode 100644 kernel/rmfork.c -- 2.53.0

1 4

[PATCH] first code
by Qi Xi 09 Jun '26

09 Jun '26

--- .../admin-guide/kernel-parameters.txt | 3 + arch/arm64/Kconfig | 11 ++ arch/arm64/include/asm/asm-uaccess.h | 13 ++ arch/arm64/include/asm/cputype.h | 2 + arch/arm64/include/asm/uaccess.h | 21 ++- arch/arm64/kernel/cpufeature.c | 51 +++++ arch/arm64/lib/copy_from_user.S | 33 ++++ arch/arm64/lib/copy_template_opt.S | 175 ++++++++++++++++++ arch/arm64/tools/cpucaps | 4 +- 9 files changed, 309 insertions(+), 4 deletions(-) create mode 100644 arch/arm64/lib/copy_template_opt.S diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index b443a9665e03..e2adafc903c1 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -423,6 +423,9 @@ arm64.nomops [ARM64] Unconditionally disable Memory Copy and Memory Set instructions support + copy_opt_disable [ARM64] Disable optimized copy_from_user + implementation + arm64.nomte [ARM64] Unconditionally disable Memory Tagging Extension support diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index 81d2baafdcd6..757d4bbe4251 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -1317,6 +1317,14 @@ config HISILICON_ERRATUM_165010801 system watchdog hardlockup detection might be triggered. The arch_timer driver addresses this by proactively increasing affected interrupt priorities. +config ARM64_COPY_FROM_USER_OPT + bool "Optimized copy_from_user for Hisilicon CPUs" + depends on ARCH_HISI + default y + help + Enable an optimized copy_from_user implementation for Hisilicon + CPUs that benefit from LDP instruction based copy routines. + config QCOM_FALKOR_ERRATUM_1003 bool "Falkor E1003: Incorrect translation due to ASID change" default y @@ -2061,6 +2069,9 @@ config ARM64_PAN config AS_HAS_LSE_ATOMICS def_bool $(as-instr,.arch_extension lse) +config AS_HAS_LSUI + def_bool $(as-instr,.arch_extension lsui) + config ARM64_LSE_ATOMICS bool default ARM64_USE_LSE_ATOMICS diff --git a/arch/arm64/include/asm/asm-uaccess.h b/arch/arm64/include/asm/asm-uaccess.h index 7bbebfa5b710..342e77b0d7f2 100644 --- a/arch/arm64/include/asm/asm-uaccess.h +++ b/arch/arm64/include/asm/asm-uaccess.h @@ -94,4 +94,17 @@ alternative_else_nop_endif _asm_extable_uaccess 8888b, \l; .endm + + .macro user_ldpair l, reg1, reg2, addr, val +8888: ldp \reg1, \reg2, [\addr, \val]; + + _asm_extable_uaccess 8888b, \l; + .endm + + .macro user_ldtpair l, reg1, reg2, addr, val +8888: .arch_extension lsui + ldtp \reg1, \reg2, [\addr, \val]; + + _asm_extable_uaccess 8888b, \l; + .endm #endif diff --git a/arch/arm64/include/asm/cputype.h b/arch/arm64/include/asm/cputype.h index 490c2ac36ac0..16c4d867f199 100644 --- a/arch/arm64/include/asm/cputype.h +++ b/arch/arm64/include/asm/cputype.h @@ -142,6 +142,7 @@ #define HISI_CPU_PART_TSV110 0xD01 #define HISI_CPU_PART_LINXICORE9100 0xD02 +#define HISI_CPU_PART_HIP11 0xD22 #define HISI_CPU_PART_HIP12 0xD06 #define APPLE_CPU_PART_M1_ICESTORM 0x022 @@ -230,6 +231,7 @@ #define MIDR_FUJITSU_A64FX MIDR_CPU_MODEL(ARM_CPU_IMP_FUJITSU, FUJITSU_CPU_PART_A64FX) #define MIDR_HISI_TSV110 MIDR_CPU_MODEL(ARM_CPU_IMP_HISI, HISI_CPU_PART_TSV110) #define MIDR_HISI_LINXICORE9100 MIDR_CPU_MODEL(ARM_CPU_IMP_HISI, HISI_CPU_PART_LINXICORE9100) +#define MIDR_HISI_HIP11 MIDR_CPU_MODEL(ARM_CPU_IMP_HISI, HISI_CPU_PART_HIP11) #define MIDR_HISI_HIP12 MIDR_CPU_MODEL(ARM_CPU_IMP_HISI, HISI_CPU_PART_HIP12) #define MIDR_APPLE_M1_ICESTORM MIDR_CPU_MODEL(ARM_CPU_IMP_APPLE, APPLE_CPU_PART_M1_ICESTORM) #define MIDR_APPLE_M1_FIRESTORM MIDR_CPU_MODEL(ARM_CPU_IMP_APPLE, APPLE_CPU_PART_M1_FIRESTORM) diff --git a/arch/arm64/include/asm/uaccess.h b/arch/arm64/include/asm/uaccess.h index dd0877a75922..ed307e1425b4 100644 --- a/arch/arm64/include/asm/uaccess.h +++ b/arch/arm64/include/asm/uaccess.h @@ -392,12 +392,29 @@ do { \ } while(0) extern unsigned long __must_check __arch_copy_from_user(void *to, const void __user *from, unsigned long n); +#ifdef CONFIG_ARM64_COPY_FROM_USER_OPT +#define COPY_OPT_THRESHOLD 4096 +static __always_inline bool use_copy_opt(unsigned long n) +{ + if (alternative_has_cap_unlikely(ARM64_HAS_LSUI)) + return true; + return alternative_has_cap_unlikely(ARM64_HAS_COPY_OPT) && n >= COPY_OPT_THRESHOLD; +} +extern unsigned long __must_check __arch_copy_from_user_opt(void *to, + const void __user *from, unsigned long n); +#else +static __always_inline bool use_copy_opt(unsigned long n) { return false; } +#endif #define raw_copy_from_user(to, from, n) \ ({ \ unsigned long __acfu_ret; \ uaccess_ttbr0_enable(); \ - __acfu_ret = __arch_copy_from_user((to), \ - __uaccess_mask_ptr(from), (n)); \ + if (use_copy_opt(n)) \ + __acfu_ret = __arch_copy_from_user_opt((to), \ + __uaccess_mask_ptr(from), (n)); \ + else \ + __acfu_ret = __arch_copy_from_user((to), \ + __uaccess_mask_ptr(from), (n)); \ uaccess_ttbr0_disable(); \ __acfu_ret; \ }) diff --git a/arch/arm64/kernel/cpufeature.c b/arch/arm64/kernel/cpufeature.c index c8e3f16387cb..1ed2850ba72d 100644 --- a/arch/arm64/kernel/cpufeature.c +++ b/arch/arm64/kernel/cpufeature.c @@ -2338,6 +2338,37 @@ static void cpu_enable_dit(const struct arm64_cpu_capabilities *__unused) set_pstate_dit(1); } +#ifdef CONFIG_ARM64_COPY_FROM_USER_OPT +static bool copy_opt_disable __ro_after_init; + +static int __init copy_opt_disable_param(char *str) +{ + copy_opt_disable = true; + return 0; +} +early_param("copy_opt_disable", copy_opt_disable_param); + +static bool has_copy_opt(const struct arm64_cpu_capabilities *entry, int scope) +{ + static const struct midr_range copy_opt_cpus[] = { + MIDR_ALL_VERSIONS(MIDR_HISI_LINXICORE9100), + MIDR_ALL_VERSIONS(MIDR_HISI_HIP11), + MIDR_ALL_VERSIONS(MIDR_HISI_HIP12), + { /* sentinel */ } + }; + + if (copy_opt_disable) + return false; + + return is_midr_in_range_list(copy_opt_cpus); // todo +} + +static void cpu_enable_copy_opt(const struct arm64_cpu_capabilities *__unused) +{ + pr_info("copy_from_user: optimized implementation enabled\n"); +} +#endif + static void cpu_enable_mops(const struct arm64_cpu_capabilities *__unused) { sysreg_clear_set(sctlr_el1, 0, SCTLR_EL1_MSCEn); @@ -3154,6 +3185,26 @@ static const struct arm64_cpu_capabilities arm64_features[] = { .matches = has_arch_xcall_xint_support, .cpu_enable = cpu_enable_arch_xcall_xint, }, +#endif + { + .desc = "Unprivileged Load Store Instructions", + .capability = ARM64_HAS_LSUI, + .type = ARM64_CPUCAP_SYSTEM_FEATURE, + .sys_reg = SYS_ID_AA64ISAR2_EL1, + .sign = FTR_UNSIGNED, + .field_pos = 4, + .field_width = 4, + .min_field_value = 1, + .matches = has_cpuid_feature, + }, +#ifdef CONFIG_ARM64_COPY_FROM_USER_OPT + { + .desc = "Optimized copy_from_user", + .capability = ARM64_HAS_COPY_OPT, + .type = ARM64_CPUCAP_BOOT_CPU_FEATURE, + .matches = has_copy_opt, + .cpu_enable = cpu_enable_copy_opt, + }, #endif {}, }; diff --git a/arch/arm64/lib/copy_from_user.S b/arch/arm64/lib/copy_from_user.S index 34e317907524..d090836aa9c0 100644 --- a/arch/arm64/lib/copy_from_user.S +++ b/arch/arm64/lib/copy_from_user.S @@ -52,6 +52,18 @@ stp \reg1, \reg2, [\ptr], \val .endm + .macro ldp2 reg1, reg2, ptr, val + alternative_if_not ARM64_HAS_LSUI + user_ldpair 9997f, \reg1, \reg2, \ptr, \val + alternative_else + user_ldtpair 9997f, \reg1, \reg2, \ptr, \val + alternative_endif + .endm + + .macro stp2 reg1, reg2, ptr, val + stp \reg1, \reg2, [\ptr, \val] + .endm + end .req x5 srcin .req x15 SYM_FUNC_START(__arch_copy_from_user) @@ -71,3 +83,24 @@ USER(9998f, ldtrb tmp1w, [srcin]) ret SYM_FUNC_END(__arch_copy_from_user) EXPORT_SYMBOL(__arch_copy_from_user) + +#ifdef CONFIG_ARM64_COPY_FROM_USER_OPT +SYM_FUNC_START(__arch_copy_from_user_opt) + add end, x0, x2 + mov srcin, x1 + +#include "copy_template_opt.S" + mov x0, #0 // Nothing to copy + ret + + // Exception fixups +9997: cmp dst, dstin + b.ne 9998f + // Before being absolutely sure we couldn't copy anything, try harder +USER(9998f, ldtrb tmp1w, [srcin]) + strb tmp1w, [dst], #1 +9998: sub x0, end, dst // bytes not copied + ret +SYM_FUNC_END(__arch_copy_from_user_opt) +EXPORT_SYMBOL(__arch_copy_from_user_opt) +#endif diff --git a/arch/arm64/lib/copy_template_opt.S b/arch/arm64/lib/copy_template_opt.S new file mode 100644 index 000000000000..df6c2be11dc0 --- /dev/null +++ b/arch/arm64/lib/copy_template_opt.S @@ -0,0 +1,175 @@ +/* SPDX-License-Identifier: GPL-2.0-only */ +/* + * Copyright (C) 2026 - Huawei Ltd. + */ + +/* + * Copy a buffer from src to dest (alignment handled by the hardware) + * + * Parameters: + * x0 - dest + * x1 - src + * x2 - n + * Returns: + * x0 - dest + * + * Note: register aliases (dstin, src, count, tmp1, etc.) are defined by + * copy_template.S which is included earlier in copy_from_user.S. + */ + + mov dst, dstin + cmp count, #16 + /*When memory length is less than 16, the accesses are not aligned.*/ + b.lo .Ltiny15_opt + + neg tmp2, src + ands tmp2, tmp2, #15/* Bytes to reach alignment. */ + b.eq .LSrcAligned_opt + sub count, count, tmp2 + /* + * Copy the leading memory data from src to dst in an increasing + * address order.By this way,the risk of overwriting the source + * memory data is eliminated when the distance between src and + * dst is less than 16. The memory accesses here are alignment. + */ + tbz tmp2, #0, 1f + ldrb1 tmp1w, src, #1 + strb1 tmp1w, dst, #1 +1: + tbz tmp2, #1, 2f + ldrh1 tmp1w, src, #2 + strh1 tmp1w, dst, #2 +2: + tbz tmp2, #2, 3f + ldr1 tmp1w, src, #4 + str1 tmp1w, dst, #4 +3: + tbz tmp2, #3, .LSrcAligned_opt + ldr1 tmp1, src, #8 + str1 tmp1, dst, #8 + +.LSrcAligned_opt: + cmp count, #64 + b.ge .Lcpy_over64_opt + /* + * Deal with small copies quickly by dropping straight into the + * exit block. + */ +.Ltail63_opt: + /* + * Copy up to 48 bytes of data. At this point we only need the + * bottom 6 bits of count to be accurate. + */ + ands tmp1, count, #0x30 + b.eq .Ltiny15_opt + cmp tmp1w, #0x20 + b.eq 1f + b.lt 2f + ldp2 A_l, A_h, src, #0 + stp2 A_l, A_h, dst, #0 + add src, src, #16 + add dst, dst, #16 +1: + ldp2 A_l, A_h, src, #0 + stp2 A_l, A_h, dst, #0 + add src, src, #16 + add dst, dst, #16 +2: + ldp2 A_l, A_h, src, #0 + stp2 A_l, A_h, dst, #0 + add src, src, #16 + add dst, dst, #16 +.Ltiny15_opt: + /* + * Prefer to break one ldp/stp into several load/store to access + * memory in an increasing address order,rather than to load/store 16 + * bytes from (src-16) to (dst-16) and to backward the src to aligned + * address,which way is used in original cortex memcpy. If keeping + * the original memcpy process here, memmove need to satisfy the + * precondition that src address is at least 16 bytes bigger than dst + * address,otherwise some source data will be overwritten when memove + * call memcpy directly. To make memmove simpler and decouple the + * memcpy's dependency on memmove, withdrew the original process. + */ + tbz count, #3, 1f + ldr1 tmp1, src, #8 + str1 tmp1, dst, #8 +1: + tbz count, #2, 2f + ldr1 tmp1w, src, #4 + str1 tmp1w, dst, #4 +2: + tbz count, #1, 3f + ldrh1 tmp1w, src, #2 + strh1 tmp1w, dst, #2 +3: + tbz count, #0, .Lexitfunc_opt + ldrb1 tmp1w, src, #1 + strb1 tmp1w, dst, #1 + + b .Lexitfunc_opt + +.Lcpy_over64_opt: + subs count, count, #128 + b.ge .Lcpy_body_large_opt + /* + * Less than 128 bytes to copy, so handle 64 here and then jump + * to the tail. + */ + ldp2 A_l, A_h, src, #0 + stp2 A_l, A_h, dst, #0 + ldp2 B_l, B_h, src, #16 + ldp2 C_l, C_h, src, #32 + stp2 B_l, B_h, dst, #16 + stp2 C_l, C_h, dst, #32 + ldp2 D_l, D_h, src, #48 + stp2 D_l, D_h, dst, #48 + add src, src, #64 + add dst, dst, #64 + + tst count, #0x3f + b.ne .Ltail63_opt + b .Lexitfunc_opt + + /* + * Critical loop. Start at a new cache line boundary. Assuming + * 64 bytes per line this ensures the entire loop is in one line. + */ + .p2align L1_CACHE_SHIFT +.Lcpy_body_large_opt: + + /* pre-get 64 bytes data. */ + ldp2 A_l, A_h, src, #0 + ldp2 B_l, B_h, src, #16 + ldp2 C_l, C_h, src, #32 + ldp2 D_l, D_h, src, #48 + add src, src, #64 +1: + /* + * interlace the load of next 64 bytes data block with store of the last + * loaded 64 bytes data. + */ + stp2 A_l, A_h, dst, #0 + ldp2 A_l, A_h, src, #0 + stp2 B_l, B_h, dst, #16 + ldp2 B_l, B_h, src, #16 + stp2 C_l, C_h, dst, #32 + ldp2 C_l, C_h, src, #32 + stp2 D_l, D_h, dst, #48 + ldp2 D_l, D_h, src, #48 + add dst, dst, #64 + add src, src, #64 + subs count, count, #64 + b.ge 1b + + /* Post-loop: store the last block of data using stp2 */ + /* (without post-increment) */ + stp2 A_l, A_h, dst, #0 + stp2 B_l, B_h, dst, #16 + stp2 C_l, C_h, dst, #32 + stp2 D_l, D_h, dst, #48 + add dst, dst, #64 + + tst count, #0x3f + b.ne .Ltail63_opt +.Lexitfunc_opt: diff --git a/arch/arm64/tools/cpucaps b/arch/arm64/tools/cpucaps index d8f2db273def..f6445266e886 100644 --- a/arch/arm64/tools/cpucaps +++ b/arch/arm64/tools/cpucaps @@ -114,8 +114,8 @@ HAS_LS64 HAS_LS64_V HAS_HW_XCALL_XINT WORKAROUND_PHYTIUM_FT3386 -KABI_RESERVE_7 -KABI_RESERVE_8 +HAS_COPY_OPT +HAS_LSUI KABI_RESERVE_9 KABI_RESERVE_10 KABI_RESERVE_11 -- 2.53.0

1 0

[PATCH openEuler-1.0-LTS] scsi/hiraid: Support new RAID features and bug fixes
by LinKun 09 Jun '26

09 Jun '26

From: 岳智超 <yuezhichao1(a)h-partners.com> driver inclusion category: feature bugzilla: https://atomgit.com/openeuler/kernel/issues/9101 CVE: NA -------------------------------- Fix high traffic IO disconnection issues Fix drive letter misalignment issue Fix timeout work mode Fix boot issue in single work mode Change adm timeout Change io queue depth Change prp、sgl Change stream detect Add linkdown fast recovery Add boot disk first scan Add flr Add ncq feature Delete driver class Signed-off-by: Zhichao Yue <yuezhichao1(a)h-partners.com> --- drivers/scsi/hisi_raid/hiraid.h | 87 ++- drivers/scsi/hisi_raid/hiraid_main.c | 919 ++++++++++++++------------- 2 files changed, 536 insertions(+), 470 deletions(-) diff --git a/drivers/scsi/hisi_raid/hiraid.h b/drivers/scsi/hisi_raid/hiraid.h index cb90d3d..25df43b 100644 --- a/drivers/scsi/hisi_raid/hiraid.h +++ b/drivers/scsi/hisi_raid/hiraid.h @@ -50,6 +50,10 @@ #define HIRAID_DEV_INFO_FLAG_VALID(flag) ((flag) & 0x01) #define HIRAID_DEV_INFO_FLAG_CHANGE(flag) ((flag) & 0x02) +#define HIRAID_DEV_INFO_FLAG_DEV_MAGIC_DEV(flag) ((flag) >> 2) +#define HIRAID_DEV_INFO_FLAG_DELETE_OR_ADD(flag, org_flag) \ + (HIRAID_DEV_INFO_FLAG_DEV_MAGIC_DEV(flag) != \ + HIRAID_DEV_INFO_FLAG_DEV_MAGIC_DEV(org_flag) ? 1 : 0) #define HIRAID_CAP_MQES(cap) ((cap) & 0xffff) #define HIRAID_CAP_STRIDE(cap) (((cap) >> 32) & 0xf) @@ -74,7 +78,19 @@ #define PCI_VENDOR_ID_HUAWEI_LOGIC 0x19E5 #define HIRAID_SERVER_DEVICE_HBA_DID 0x3858 +#define HIRAID_SERVER_DEVICE_HBAS_DID 0x3918 #define HIRAID_SERVER_DEVICE_RAID_DID 0x3758 +#define HIRAID_SERVER_DEVICE_RAIDS_DID 0x38D8 + +#define HIRAID_SCSI_VPD_LEN SCSI_VPD_PG_LEN +#define SCSI_RSVD_INIT_SHIFT 0XFC +#define NCQ_PRIO_SUPPORT_BYTE 213 +#define VPD_NCQ_PRIO_SUPPORT_BIT 4 + +#define HIRAID_CAP_CAC_STRIDE 8 + +#define HIRAID_MAX_PD_NUM (40 + 1) +#define HIRAID_MAX_STREAM_NUM 8 enum { HIRAID_SC_SUCCESS = 0x0, @@ -90,6 +106,7 @@ enum { enum { HIRAID_REG_CAP = 0x0000, + HIRAID_REG_VS = 0x0008, HIRAID_REG_CC = 0x0014, HIRAID_REG_CSTS = 0x001c, HIRAID_REG_AQA = 0x0024, @@ -236,6 +253,20 @@ enum { DISPATCH_BY_DISK, }; +enum hiraid_stream_type { + HIRAID_STREAM_TYPE_TOTAL, + HIRAID_STREAM_TYPE_WRITE, + HIRAID_STREAM_TYPE_READ, + HIRAID_STREAM_TYPE_CLEAN, + HIRAID_STREAM_TYPE_BOTTOM +}; + +enum hiraid_stream_io_operation_type { + HIRAID_STREAM_TYPE_DELETE_SINGLE_IO = 1, + HIRAID_STREAM_TYPE_DELETE_SINGLE_IO_LIST, + HIRAID_STREAM_TYPE_DELETE_ALL_IO_LIST +}; + struct hiraid_completion { __le32 result; union { @@ -271,6 +302,19 @@ struct hiraid_ctrl_info { __u8 rsvd1[4020]; }; +struct hiraid_stream { + /* recog-window */ + u64 stream_lba; + u32 stream_len; + u16 did; + u16 type; + /* aging ctrl */ + int aging_credit; + int aging_grade; + u16 stream_id; + u16 stream_is_using; +}; + struct hiraid_dev { struct pci_dev *pdev; struct device *dev; @@ -318,6 +362,18 @@ struct hiraid_dev { u8 hdd_dispatch; struct request_queue *bsg_queue; + + u16 hiraid_stream_io_count; + u64 hiraid_stream_aging_time; + u64 hiraid_stream_sent_io_size[HIRAID_MAX_PD_NUM] + [HIRAID_MAX_STREAM_NUM]; + u16 hiraid_stream_io_num_per_pd[HIRAID_MAX_PD_NUM] + [HIRAID_STREAM_TYPE_BOTTOM + 1]; + spinlock_t hiraid_stream_array_lock; + struct hiraid_stream hiraid_stream_array[HIRAID_MAX_PD_NUM] + [HIRAID_MAX_STREAM_NUM]; + struct task_struct *hiraid_stream_submit_task; + u64 hiraid_stream_io_last_pull_time[HIRAID_MAX_PD_NUM]; }; struct hiraid_sgl_desc { @@ -696,6 +752,10 @@ struct hiraid_mapmange { u32 sge_cnt; u32 len; bool use_sgl; + /* This field is used in the I/O read/write process. + * It indicates the TRANSFER LENGTH field in the CDB. + */ + u32 cdb_data_len; dma_addr_t first_dma; void *sense_buffer_virt; dma_addr_t sense_buffer_phy; @@ -755,38 +815,19 @@ struct hiraid_sdev_hostdata { u8 flag; u8 rg_id; u8 hwq; + u8 sata_ncq_prio_enable; + u8 rsvd[3]; u16 pend_count; }; -enum stream_type { - TYPE_TOTAL, - TYPE_WRITE, - TYPE_READ, - TYPE_CLEAN, - TYPE_BOTTOM -}; - -struct HIRAID_STREAM_S { - /* recog-window */ - u64 stream_lba; - u32 stream_len; - u16 did; - u16 type; - /* aging ctrl */ - int aging_credit; - int aging_grade; - u16 stream_id; - u16 using; -}; - -struct IO_LIST_S { +struct hiraid_stream_io_list { struct list_head list; struct hiraid_scsi_io_cmd io_cmd; struct hiraid_queue *submit_queue; unsigned int sector_size; }; -struct mutex_list_head_s { +struct mutex_list_head { struct list_head list; struct mutex lock; }; diff --git a/drivers/scsi/hisi_raid/hiraid_main.c b/drivers/scsi/hisi_raid/hiraid_main.c index cdba466..9251fb8 100644 --- a/drivers/scsi/hisi_raid/hiraid_main.c +++ b/drivers/scsi/hisi_raid/hiraid_main.c @@ -26,6 +26,10 @@ #include <linux/bsg-lib.h> #include <asm/unaligned.h> #include <linux/sort.h> +#include <linux/kthread.h> +#include <linux/mutex.h> +#include <linux/sched.h> +#include <linux/sched/prio.h> #include <target/target_core_backend.h> #include <scsi/scsi.h> @@ -35,14 +39,10 @@ #include <scsi/scsi_transport.h> #include <scsi/scsi_dbg.h> #include <scsi/sg.h> -#include <linux/kthread.h> -#include <linux/mutex.h> -#include <linux/sched.h> -#include <linux/sched/prio.h> #include "hiraid.h" -static u32 admin_tmout = 60; +static u32 admin_tmout = 180; module_param(admin_tmout, uint, 0644); MODULE_PARM_DESC(admin_tmout, "admin commands timeout (seconds)"); @@ -101,7 +101,7 @@ static const struct kernel_param_ops io_queue_depth_ops = { .get = param_get_uint, }; -static u32 io_queue_depth = 1024; +static u32 io_queue_depth = 4096; module_param_cb(io_queue_depth, &io_queue_depth_ops, &io_queue_depth, 0644); MODULE_PARM_DESC(io_queue_depth, "set io queue depth, should >= 2"); @@ -143,8 +143,6 @@ static void hiraid_handle_async_notice(struct hiraid_dev *hdev, u32 result); static void hiraid_handle_async_vs(struct hiraid_dev *hdev, u32 result, u32 result1); -static struct class *hiraid_class; - #define HIRAID_CAP_TIMEOUT_UNIT_MS (HZ / 2) static struct workqueue_struct *work_queue; @@ -155,7 +153,7 @@ static struct workqueue_struct *work_queue; __func__, ##__VA_ARGS__); \ } while (0) -#define HIRAID_DRV_VERSION "1.1.0.1" +#define HIRAID_DRV_VERSION "2.1.0.1" #define ADMIN_TIMEOUT (admin_tmout * HZ) #define USRCMD_TIMEOUT (180 * HZ) @@ -178,8 +176,12 @@ static struct workqueue_struct *work_queue; #define MIN_CREDIT 0 #define MAX_CREDIT 64 #define CREDIT_THRES 32 +#ifndef MIN #define MIN(a, b) (((a) < (b)) ? (a) : (b)) +#endif +#ifndef MAX #define MAX(a, b) (((a) > (b)) ? (a) : (b)) +#endif enum SENSE_STATE_CODE { SENSE_STATE_OK = 0, @@ -362,95 +364,8 @@ static u32 hiraid_get_max_cmd_size(struct hiraid_dev *hdev) return sizeof(struct hiraid_mapmange) + alloc_size; } -static int hiraid_build_passthru_prp(struct hiraid_dev *hdev, - struct hiraid_mapmange *mapbuf) -{ - struct scatterlist *sg = mapbuf->sgl; - __le64 *phy_regpage, *prior_list; - u64 buf_addr = sg_dma_address(sg); - int buf_length = sg_dma_len(sg); - u32 page_size = hdev->page_size; - int offset = buf_addr & (page_size - 1); - void **list = hiraid_mapbuf_list(mapbuf); - int maplen = mapbuf->len; - struct dma_pool *pool; - dma_addr_t buffer_phy; - int i; - - maplen -= (page_size - offset); - if (maplen <= 0) { - mapbuf->first_dma = 0; - return 0; - } - - buf_length -= (page_size - offset); - if (buf_length) { - buf_addr += (page_size - offset); - } else { - sg = sg_next(sg); - buf_addr = sg_dma_address(sg); - buf_length = sg_dma_len(sg); - } - - if (maplen <= page_size) { - mapbuf->first_dma = buf_addr; - return 0; - } - - pool = hdev->prp_page_pool; - mapbuf->page_cnt = 1; - - phy_regpage = dma_pool_alloc(pool, GFP_ATOMIC, &buffer_phy); - if (!phy_regpage) { - dev_err_ratelimited(hdev->dev, "allocate first admin prp_list memory failed\n"); - mapbuf->first_dma = buf_addr; - mapbuf->page_cnt = -1; - return -ENOMEM; - } - list[0] = phy_regpage; - mapbuf->first_dma = buffer_phy; - i = 0; - for (;;) { - if (i == page_size / PRP_ENTRY_SIZE) { - prior_list = phy_regpage; - - phy_regpage = dma_pool_alloc(pool, GFP_ATOMIC, - &buffer_phy); - if (!phy_regpage) { - dev_err_ratelimited(hdev->dev, "allocate [%d]th admin prp list memory failed\n", - mapbuf->page_cnt + 1); - return -ENOMEM; - } - list[mapbuf->page_cnt++] = phy_regpage; - phy_regpage[0] = prior_list[i - 1]; - prior_list[i - 1] = cpu_to_le64(buffer_phy); - i = 1; - } - phy_regpage[i++] = cpu_to_le64(buf_addr); - buf_addr += page_size; - buf_length -= page_size; - maplen -= page_size; - if (maplen <= 0) - break; - if (buf_length > 0) - continue; - if (unlikely(buf_length < 0)) - goto bad_admin_sgl; - sg = sg_next(sg); - buf_addr = sg_dma_address(sg); - buf_length = sg_dma_len(sg); - } - - return 0; - -bad_admin_sgl: - dev_err(hdev->dev, "setup prps, invalid admin SGL for payload[%d] nents[%d]\n", - mapbuf->len, mapbuf->sge_cnt); - return -EIO; -} - static int hiraid_build_prp(struct hiraid_dev *hdev, - struct hiraid_mapmange *mapbuf) + struct hiraid_mapmange *mapbuf) { struct scatterlist *sg = mapbuf->sgl; __le64 *phy_regpage, *prior_list; @@ -656,67 +571,9 @@ static void hiraid_sgl_set_seg(struct hiraid_sgl_desc *sge, } } -static int hiraid_build_passthru_sgl(struct hiraid_dev *hdev, - struct hiraid_admin_command *admin_cmd, - struct hiraid_mapmange *mapbuf) -{ - struct hiraid_sgl_desc *sg_list, *link, *old_sg_list; - struct scatterlist *sg = mapbuf->sgl; - void **list = hiraid_mapbuf_list(mapbuf); - struct dma_pool *pool; - int nsge = mapbuf->sge_cnt; - dma_addr_t buffer_phy; - int i = 0; - - admin_cmd->common.flags |= SQE_FLAG_SGL_METABUF; - - if (nsge == 1) { - hiraid_sgl_set_data(&admin_cmd->common.dptr.sgl, sg); - return 0; - } - - pool = hdev->prp_page_pool; - mapbuf->page_cnt = 1; - - sg_list = dma_pool_alloc(pool, GFP_ATOMIC, &buffer_phy); - if (!sg_list) { - dev_err_ratelimited(hdev->dev, "allocate first admin sgl_list failed\n"); - mapbuf->page_cnt = -1; - return -ENOMEM; - } - - list[0] = sg_list; - mapbuf->first_dma = buffer_phy; - hiraid_sgl_set_seg(&admin_cmd->common.dptr.sgl, buffer_phy, nsge); - do { - if (i == SGES_PER_PAGE) { - old_sg_list = sg_list; - link = &old_sg_list[SGES_PER_PAGE - 1]; - - sg_list = dma_pool_alloc(pool, GFP_ATOMIC, &buffer_phy); - if (!sg_list) { - dev_err_ratelimited(hdev->dev, "allocate [%d]th admin sgl_list failed\n", - mapbuf->page_cnt + 1); - return -ENOMEM; - } - list[mapbuf->page_cnt++] = sg_list; - - i = 0; - memcpy(&sg_list[i++], link, sizeof(*link)); - hiraid_sgl_set_seg(link, buffer_phy, nsge); - } - - hiraid_sgl_set_data(&sg_list[i++], sg); - sg = sg_next(sg); - } while (--nsge > 0); - - return 0; -} - - static int hiraid_build_sgl(struct hiraid_dev *hdev, - struct hiraid_scsi_io_cmd *io_cmd, - struct hiraid_mapmange *mapbuf) + struct hiraid_scsi_io_cmd *io_cmd, + struct hiraid_mapmange *mapbuf) { struct hiraid_sgl_desc *sg_list, *link, *old_sg_list; struct scatterlist *sg = mapbuf->sgl; @@ -776,102 +633,52 @@ static int hiraid_build_sgl(struct hiraid_dev *hdev, return 0; } -#define MAX_PD_NUM (40 + 1) -#define MAX_STREAM_NUM 8 -#define PER_MB (1024 * 1024) -#define MAX_IO_NUM (200 * PER_MB) -#define STREAM_LEN (4 * PER_MB) +#define MAX_IO_NUM (200 * 1024 * 1024) +#define STREAM_LEN (4 * 1024 * 1024) #define MAX_IO_NUM_ONCE 100 #define IO_SUBMIT_TIME_OUT 100 -#define MAX_AGING_NUM 100 - +#define MAX_AGING_NUM 5000 +#define MAX_AGING_TIME 16 +#define AGING_DEGRADE (-2) #define MIN_IO_SEND_TIME 10 #define MAX_IO_SEND_TIME 50 -enum io_operation_type { - TYPE_DELETE_SINGLE_IO = 1, - TYPE_DELETE_SINGLE_IO_LIST, - TYPE_DELETE_ALL_IO_LIST -}; - -struct HIRAID_STREAM_S stream_array[MAX_PD_NUM][MAX_STREAM_NUM] = {0}; -struct mutex_list_head_s io_heads_per_stream[MAX_PD_NUM * MAX_STREAM_NUM]; -spinlock_t stream_array_lock; +struct mutex_list_head hiraid_io_heads_per_stream[ + HIRAID_MAX_PD_NUM * HIRAID_MAX_STREAM_NUM]; DEFINE_MUTEX(g_stream_operation_mutex); -u64 g_io_transport_num[MAX_PD_NUM][MAX_STREAM_NUM] = {0}; -u16 g_io_stream_num[MAX_PD_NUM][TYPE_BOTTOM] = {0}; -u16 g_io_count = 1; - -void hiraid_inc_io_transport_num(u16 disk_id, u16 streamd_id, u16 nlb) -{ - g_io_transport_num[disk_id][streamd_id] += nlb; -} - -void hiraid_refresh_io_transport_num(u16 disk_id, u16 streamd_id) -{ - g_io_transport_num[disk_id][streamd_id] = 0; -} - -void hiraid_inc_stream_num(u16 disk_id) -{ - spin_lock(&stream_array_lock); - g_io_stream_num[disk_id][TYPE_TOTAL]++; - spin_unlock(&stream_array_lock); -} - -void hiraid_dec_stream_num(u16 disk_id) -{ - spin_lock(&stream_array_lock); - if (g_io_stream_num[disk_id][TYPE_TOTAL] > 0) - g_io_stream_num[disk_id][TYPE_TOTAL]--; - spin_unlock(&stream_array_lock); -} - -static bool hiraid_io_recog_check_stream_exceed(u16 disk_id) +static inline bool hiraid_io_recog_check_stream_exceed(struct hiraid_dev *hdev, + u16 disk_id) { bool exceed_flag; - spin_lock(&stream_array_lock); - exceed_flag = (g_io_stream_num[disk_id][TYPE_TOTAL] >= MAX_STREAM_NUM); - spin_unlock(&stream_array_lock); + spin_lock(&hdev->hiraid_stream_array_lock); + exceed_flag = + (hdev->hiraid_stream_io_num_per_pd[disk_id][HIRAID_STREAM_TYPE_TOTAL] >= + HIRAID_MAX_STREAM_NUM); + spin_unlock(&hdev->hiraid_stream_array_lock); return exceed_flag; } -static u16 hiraid_get_stream_num(u16 disk_id) -{ - return g_io_stream_num[disk_id][TYPE_TOTAL]; -} - -static inline struct HIRAID_STREAM_S *hiraid_get_stream(u16 disk_id, - u16 stream_id) -{ - return &stream_array[disk_id][stream_id]; -} - -static inline struct mutex_list_head_s *hiraid_get_io_head(u16 disk_id) -{ - return &(io_heads_per_stream[disk_id]); -} - -static bool hiraid_recognition_acknowledge(const struct HIRAID_STREAM_S *stream) +static inline bool hiraid_recognition_acknowledge( + const struct hiraid_stream *stream) { return (stream->aging_credit >= CREDIT_THRES) ? true : false; } -void hiraid_io_recognition_init(void) +void hiraid_io_recognition_init(struct hiraid_dev *hdev) { u16 i; - spin_lock_init(&stream_array_lock); - for (i = 0; i < (MAX_PD_NUM * MAX_STREAM_NUM); i++) { - INIT_LIST_HEAD(&hiraid_get_io_head(i)->list); - mutex_init(&hiraid_get_io_head(i)->lock); + spin_lock_init(&hdev->hiraid_stream_array_lock); + for (i = 0; i < (HIRAID_MAX_PD_NUM * HIRAID_MAX_STREAM_NUM); i++) { + INIT_LIST_HEAD(&hiraid_io_heads_per_stream[i].list); + mutex_init(&hiraid_io_heads_per_stream[i].lock); } } -static void hiraid_io_recognition_iterator(struct HIRAID_STREAM_S *stream, - int direction) +static void hiraid_io_recognition_iterator(struct hiraid_stream *stream, + int direction) { stream->aging_grade = stream->aging_grade + direction * INC_GRADE; stream->aging_grade = MAX(stream->aging_grade, MAX_DECREASE_GRADE); @@ -881,46 +688,52 @@ static void hiraid_io_recognition_iterator(struct HIRAID_STREAM_S *stream, stream->aging_credit = MIN(stream->aging_credit, MAX_CREDIT); } -struct HIRAID_STREAM_S *hiraid_io_pick_stream( +struct hiraid_stream *hiraid_io_pick_stream(struct hiraid_dev *hdev, struct hiraid_scsi_rw_cmd *req, u16 type, u16 actual_id) { - struct HIRAID_STREAM_S *first_hit_stream = NULL; - struct HIRAID_STREAM_S *temp_stream = NULL; - u16 pick_flag = 0; + struct hiraid_stream *first_hit_stream = NULL; + struct hiraid_stream *temp_stream = NULL; + bool pick_flag = false; u8 i; - for (i = 0; i < MAX_STREAM_NUM; i++) { - temp_stream = &stream_array[actual_id][i]; + for (i = 0; i < HIRAID_MAX_STREAM_NUM; i++) { + temp_stream = &hdev->hiraid_stream_array[actual_id][i]; temp_stream->stream_id = i; if (req->slba < temp_stream->stream_lba || req->slba >= temp_stream->stream_lba + temp_stream->stream_len || - temp_stream->type != type) { + temp_stream->type != type || + !temp_stream->stream_is_using) { continue; } - if (!pick_flag) { + if (pick_flag == false) { temp_stream->stream_lba = req->slba; first_hit_stream = temp_stream; - pick_flag = 1; + pick_flag = true; continue; } - hiraid_dec_stream_num(actual_id); - memset(temp_stream, 0, - sizeof(struct HIRAID_STREAM_S)); // 去重影 + spin_lock(&hdev->hiraid_stream_array_lock); + if (hdev->hiraid_stream_io_num_per_pd[actual_id] + [HIRAID_STREAM_TYPE_TOTAL] > 0) + hdev->hiraid_stream_io_num_per_pd[actual_id] + [HIRAID_STREAM_TYPE_TOTAL]--; + + memset(temp_stream, 0, sizeof(struct hiraid_stream)); + spin_unlock(&hdev->hiraid_stream_array_lock); } return first_hit_stream; } -static struct HIRAID_STREAM_S *hiraid_init_flow_stream( +static struct hiraid_stream *hiraid_init_flow_stream(struct hiraid_dev *hdev, struct hiraid_scsi_rw_cmd *req, u16 type, u16 actual_id) { int i; - struct HIRAID_STREAM_S *stream = NULL; + struct hiraid_stream *stream = NULL; - for (i = 0; i < MAX_STREAM_NUM; i++) { - stream = hiraid_get_stream(actual_id, i); - if (!stream->using) { - stream->using = 1; + for (i = 0; i < HIRAID_MAX_STREAM_NUM; i++) { + stream = &hdev->hiraid_stream_array[actual_id][i]; + if (!stream->stream_is_using) { + stream->stream_is_using = 1; stream->stream_id = i; break; } @@ -934,82 +747,96 @@ static struct HIRAID_STREAM_S *hiraid_init_flow_stream( return stream; } -static struct HIRAID_STREAM_S *hiraid_stream_detect(struct hiraid_dev *hdev, - struct hiraid_scsi_rw_cmd *io_cmd, u16 actual_id) +static struct hiraid_stream *hiraid_stream_detect(struct hiraid_dev *hdev, + struct hiraid_scsi_rw_cmd *io_cmd, u16 actual_id) { - u16 type = io_cmd->opcode == HIRAID_CMD_WRITE ? TYPE_WRITE : TYPE_READ; - struct HIRAID_STREAM_S *stream = hiraid_io_pick_stream(io_cmd, type, - actual_id); + u16 type = HIRAID_STREAM_TYPE_READ; + struct hiraid_stream *stream = NULL; + + if (io_cmd->opcode == HIRAID_CMD_WRITE) + type = HIRAID_STREAM_TYPE_WRITE; + + stream = hiraid_io_pick_stream(hdev, + io_cmd, type, actual_id); - if (stream != NULL) { /* 可以命中一个stream */ + if (stream != NULL) return stream; - } - if (hiraid_io_recog_check_stream_exceed(actual_id)) + if (hiraid_io_recog_check_stream_exceed(hdev, actual_id)) return NULL; - stream = hiraid_init_flow_stream(io_cmd, type, actual_id); - hiraid_inc_stream_num(actual_id); + + stream = hiraid_init_flow_stream(hdev, io_cmd, type, actual_id); + spin_lock(&hdev->hiraid_stream_array_lock); + hdev->hiraid_stream_io_num_per_pd[actual_id] + [HIRAID_STREAM_TYPE_TOTAL]++; + spin_unlock(&hdev->hiraid_stream_array_lock); return stream; } -u64 g_io_last_pull_time[MAX_PD_NUM] = {0}; - static u16 hiraid_get_submit_io_stream(u16 did, struct hiraid_dev *hdev) { u64 temp_num, i; - static u16 stream_num[MAX_PD_NUM] = {0}; + static u16 stream_num[HIRAID_MAX_PD_NUM] = {0}; - if (g_io_last_pull_time[did] == 0) - g_io_last_pull_time[did] = jiffies_to_msecs(jiffies); + if (hdev->hiraid_stream_io_last_pull_time[did] == 0) + hdev->hiraid_stream_io_last_pull_time[did] = + jiffies_to_msecs(jiffies); - for (i = 0; i < MAX_STREAM_NUM; i++) { - temp_num = g_io_transport_num[did][i]; + for (i = 0; i < HIRAID_MAX_STREAM_NUM; i++) { + temp_num = hdev->hiraid_stream_sent_io_size[did][i]; if (temp_num != 0) { if ((temp_num < MAX_IO_NUM) && - ((jiffies_to_msecs(jiffies) - g_io_last_pull_time[did]) - < IO_SUBMIT_TIME_OUT)) { + ((jiffies_to_msecs(jiffies) - + hdev->hiraid_stream_io_last_pull_time[did]) < + IO_SUBMIT_TIME_OUT)) { stream_num[did] = i; return i; } - g_io_last_pull_time[did] = jiffies_to_msecs(jiffies); - hiraid_refresh_io_transport_num(did, i); - stream_num[did] = ((i+1) % MAX_STREAM_NUM); - return ((i+1) % MAX_STREAM_NUM); + hdev->hiraid_stream_io_last_pull_time[did] = + jiffies_to_msecs(jiffies); + hdev->hiraid_stream_sent_io_size[did][i] = 0; + stream_num[did] = ((i+1) % + HIRAID_MAX_STREAM_NUM); + return ((i+1) % HIRAID_MAX_STREAM_NUM); + } } - g_io_last_pull_time[did] = jiffies_to_msecs(jiffies); - return ((stream_num[did]++) % MAX_STREAM_NUM); + hdev->hiraid_stream_io_last_pull_time[did] = + jiffies_to_msecs(jiffies); + return ((stream_num[did]++) % HIRAID_MAX_STREAM_NUM); } static void hiraid_submit_io_stream(u16 hdid, struct hiraid_dev *hdev) { - struct mutex_list_head_s *io_slist = NULL; + struct mutex_list_head *io_slist = NULL; struct list_head *node = NULL; struct list_head *next_node = NULL; struct list_head temp_header; - struct hiraid_scsi_io_cmd io_cmd = {0}; + struct hiraid_scsi_io_cmd io_cmd; u16 submit_stream_id; - struct IO_LIST_S *temp_io_stream = NULL; + struct hiraid_stream_io_list *temp_io_stream = NULL; u16 count = 0; INIT_LIST_HEAD(&temp_header); submit_stream_id = hiraid_get_submit_io_stream(hdid, hdev); - io_slist = hiraid_get_io_head(hdid * MAX_STREAM_NUM + submit_stream_id); + io_slist = &hiraid_io_heads_per_stream[hdid * HIRAID_MAX_STREAM_NUM + + submit_stream_id]; mutex_lock(&io_slist->lock); list_for_each_safe(node, next_node, &io_slist->list) { list_del(node); list_add_tail(node, &temp_header); - if (++count >= MAX_IO_NUM_ONCE) + if (count++ >= MAX_IO_NUM_ONCE) break; } mutex_unlock(&io_slist->lock); list_for_each_safe(node, next_node, &temp_header) { - temp_io_stream = list_entry(node, struct IO_LIST_S, list); + temp_io_stream = list_entry(node, + struct hiraid_stream_io_list, list); io_cmd = temp_io_stream->io_cmd; hiraid_submit_cmd(temp_io_stream->submit_queue, &io_cmd); - hiraid_inc_io_transport_num(hdid, submit_stream_id, - io_cmd.rw.nlb * temp_io_stream->sector_size); + hdev->hiraid_stream_sent_io_size[hdid][submit_stream_id] += + io_cmd.rw.nlb * temp_io_stream->sector_size; list_del(node); kfree(temp_io_stream); } @@ -1017,59 +844,90 @@ static void hiraid_submit_io_stream(u16 hdid, struct hiraid_dev *hdev) list_del_init(&temp_header); } -static u8 hiraid_detect_if_aging(void) +static u8 hiraid_aging_detect(struct hiraid_dev *hdev) { - if (++g_io_count == MAX_AGING_NUM) { - g_io_count = 0; + if (hdev->hiraid_stream_aging_time == 0) + hdev->hiraid_stream_aging_time = jiffies_to_msecs(jiffies); + if (++hdev->hiraid_stream_io_count > MAX_AGING_NUM) { + hdev->hiraid_stream_io_count = 0; + if ((jiffies_to_msecs(jiffies) - + hdev->hiraid_stream_aging_time) < + MAX_AGING_TIME) + return 0; + hdev->hiraid_stream_aging_time = jiffies_to_msecs(jiffies); return 1; } return 0; } +static void hiraid_sync_using_stream(struct hiraid_dev *hdev, u16 hdid) +{ + u8 i; + struct hiraid_stream *temp_stream = NULL; + + hdev->hiraid_stream_io_num_per_pd[hdid][HIRAID_STREAM_TYPE_TOTAL] = 0; + for (i = 0; i < HIRAID_MAX_STREAM_NUM; i++) { + temp_stream = &hdev->hiraid_stream_array[hdid][i]; + if (temp_stream->stream_is_using) + hdev->hiraid_stream_io_num_per_pd[hdid] + [HIRAID_STREAM_TYPE_TOTAL]++; + } +} + static void hiraid_aging(struct hiraid_dev *hdev) { - struct HIRAID_STREAM_S *temp_stream = NULL; + struct hiraid_stream *temp_stream = NULL; int i = 0; int j = 0; - - for (i = 0; i < MAX_PD_NUM; i++) { - for (j = 0; j < MAX_STREAM_NUM; j++) { - temp_stream = hiraid_get_stream(i, j); - if (temp_stream->using) { - hiraid_io_recognition_iterator(temp_stream, -1); + u16 *tmp_io_num = NULL; + + for (i = 0; i < HIRAID_MAX_PD_NUM; i++) { + for (j = 0; j < HIRAID_MAX_STREAM_NUM; j++) { + temp_stream = &hdev->hiraid_stream_array[i][j]; + if (temp_stream->stream_is_using) { + hiraid_io_recognition_iterator(temp_stream, + AGING_DEGRADE); + spin_lock(&hdev->hiraid_stream_array_lock); + tmp_io_num = &hdev->hiraid_stream_io_num_per_pd + [i][HIRAID_STREAM_TYPE_TOTAL]; if (temp_stream->aging_credit <= 0) { - hiraid_dec_stream_num(i); + if (*tmp_io_num > 0) + (*tmp_io_num)--; memset(temp_stream, - 0, sizeof(struct HIRAID_STREAM_S)); // 老化 + 0, sizeof(struct hiraid_stream)); } + spin_unlock(&hdev->hiraid_stream_array_lock); } } + hiraid_sync_using_stream(hdev, i); } } -static u8 hiraid_io_list_operation(u32 hdid, u16 cid, u16 hwq, u8 operation) +static u8 hiraid_stream_delete_io_list(struct hiraid_dev *hdev, u32 hdid, + u16 cid, u16 hwq, u8 delete_operation) { int i, j; - - struct mutex_list_head_s *io_slist = NULL; + struct mutex_list_head *io_slist = NULL; struct list_head *node = NULL; struct list_head *next_node = NULL; struct hiraid_scsi_io_cmd *io_cmd = NULL; struct hiraid_queue *hiraidq = NULL; - struct IO_LIST_S *temp_io_stream = NULL; + struct hiraid_stream_io_list *temp_io_stream = NULL; + u8 max_hd_num = delete_operation == + HIRAID_STREAM_TYPE_DELETE_ALL_IO_LIST ? HIRAID_MAX_PD_NUM : hdid + 1; - u8 max_hd_num = operation == TYPE_DELETE_ALL_IO_LIST ? - MAX_PD_NUM : hdid + 1; for (i = hdid; i < max_hd_num; i++) { - for (j = 0; j < MAX_STREAM_NUM; j++) { - io_slist = hiraid_get_io_head(i * MAX_STREAM_NUM + j); + for (j = 0; j < HIRAID_MAX_STREAM_NUM; j++) { + io_slist = + &hiraid_io_heads_per_stream[i * HIRAID_MAX_STREAM_NUM + j]; mutex_lock(&io_slist->lock); list_for_each_safe(node, next_node, &io_slist->list) { - temp_io_stream = list_entry(node, - struct IO_LIST_S, list); + temp_io_stream = + list_entry(node, struct hiraid_stream_io_list, list); io_cmd = &(temp_io_stream->io_cmd); hiraidq = temp_io_stream->submit_queue; - if (operation >= TYPE_DELETE_SINGLE_IO_LIST) { + if (delete_operation >= + HIRAID_STREAM_TYPE_DELETE_SINGLE_IO_LIST) { list_del_init(node); kfree(temp_io_stream); temp_io_stream = NULL; @@ -1089,54 +947,27 @@ static u8 hiraid_io_list_operation(u32 hdid, u16 cid, u16 hwq, u8 operation) return 0; } -static u8 hiraid_check_io_list(u32 hdid, u16 cid, u16 hwq) -{ - u8 ret; - - mutex_lock(&g_stream_operation_mutex); - ret = hiraid_io_list_operation(hdid, cid, hwq, TYPE_DELETE_SINGLE_IO); - mutex_unlock(&g_stream_operation_mutex); - return ret; -} - -static u8 hiraid_delete_single_pd_io_list(u32 hdid) +static u8 hiraid_add_io_to_list(struct hiraid_dev *hdev, + struct hiraid_queue *submit_queue, struct hiraid_stream *tmp_stream, + struct hiraid_scsi_io_cmd io_cmd, unsigned int sector_size, + u16 actual_id) { - u8 ret; + struct mutex_list_head *io_slist = NULL; + struct hiraid_stream_io_list *new_io_node = NULL; + u16 temp_io_id = actual_id * HIRAID_MAX_STREAM_NUM + + tmp_stream->stream_id; - mutex_lock(&g_stream_operation_mutex); - ret = hiraid_io_list_operation(hdid, 0, 0, TYPE_DELETE_SINGLE_IO_LIST); - mutex_unlock(&g_stream_operation_mutex); - return ret; -} - -static u8 hiraid_delete_all_io_list(void) -{ - u8 ret; - - mutex_lock(&g_stream_operation_mutex); - ret = hiraid_io_list_operation(0, 0, 0, TYPE_DELETE_ALL_IO_LIST); - mutex_unlock(&g_stream_operation_mutex); - return ret; -} - -static u8 hiraid_add_io_to_list(struct hiraid_queue *submit_queue, - struct HIRAID_STREAM_S *tmp_stream, struct hiraid_scsi_io_cmd io_cmd, - unsigned int sector_size, u16 actual_id) -{ - struct mutex_list_head_s *io_slist = NULL; - struct IO_LIST_S *new_io_node = NULL; - - new_io_node = kmalloc(sizeof(struct IO_LIST_S), GFP_KERNEL); + new_io_node = kmalloc(sizeof(struct hiraid_stream_io_list), GFP_KERNEL); if (!new_io_node) return 0; new_io_node->io_cmd = io_cmd; new_io_node->submit_queue = submit_queue; new_io_node->sector_size = sector_size; - io_slist = hiraid_get_io_head(actual_id * - MAX_STREAM_NUM + tmp_stream->stream_id); + io_slist = &hiraid_io_heads_per_stream[temp_io_id]; mutex_lock(&io_slist->lock); INIT_LIST_HEAD(&(new_io_node->list)); - list_add_tail(&(new_io_node->list), &io_slist->list); + list_add_tail(&(new_io_node->list), + &hiraid_io_heads_per_stream[temp_io_id].list); mutex_unlock(&io_slist->lock); return 1; } @@ -1146,10 +977,8 @@ static void hiraid_submit_io_threading(struct hiraid_dev *hdev) int i = 0; while (!kthread_should_stop()) { - mutex_lock(&g_stream_operation_mutex); - for (i = 0; i < MAX_PD_NUM; i++) + for (i = 0; i < HIRAID_MAX_PD_NUM; i++) hiraid_submit_io_stream(i, hdev); - mutex_unlock(&g_stream_operation_mutex); usleep_range(MIN_IO_SEND_TIME, MAX_IO_SEND_TIME); } } @@ -1158,27 +987,34 @@ static void hiraid_destroy_io_stream_resource(struct hiraid_dev *hdev) { u16 i; - for (i = 0; i < (MAX_PD_NUM * MAX_STREAM_NUM); i++) - list_del_init(&hiraid_get_io_head(i)->list); + for (i = 0; i < (HIRAID_MAX_PD_NUM * HIRAID_MAX_STREAM_NUM); i++) { + list_del_init(&hiraid_io_heads_per_stream[i].list); + mutex_destroy(&hiraid_io_heads_per_stream[i].lock); + } } -struct task_struct *g_hiraid_submit_task; static void hiraid_init_io_stream(struct hiraid_dev *hdev) { - hiraid_io_recognition_init(); - g_hiraid_submit_task = kthread_run((void *)hiraid_submit_io_threading, - hdev, "hiraid_submit_thread"); + hiraid_io_recognition_init(hdev); + hdev->hiraid_stream_submit_task = + kthread_run((void *)hiraid_submit_io_threading, + hdev, "hiraid_stream_submit_task"); } #define HIRAID_RW_FUA BIT(14) +#define RW_LENGTH_ZERO (67) static int hiraid_setup_rw_cmd(struct hiraid_dev *hdev, struct hiraid_scsi_rw_cmd *io_cmd, - struct scsi_cmnd *scmd) + struct scsi_cmnd *scmd, + struct hiraid_mapmange *mapbuf) { + u32 ret = 0; u32 start_lba_lo, start_lba_hi; u32 datalength = 0; u16 control = 0; + struct scsi_device *sdev = scmd->device; + u32 buf_len = cpu_to_le32(scsi_bufflen(scmd)); start_lba_lo = 0; start_lba_hi = 0; @@ -1187,6 +1023,8 @@ static int hiraid_setup_rw_cmd(struct hiraid_dev *hdev, io_cmd->opcode = HIRAID_CMD_WRITE; } else if (scmd->sc_data_direction == DMA_FROM_DEVICE) { io_cmd->opcode = HIRAID_CMD_READ; + } else if (scmd->sc_data_direction == DMA_NONE) { + ret = RW_LENGTH_ZERO; } else { dev_err(hdev->dev, "invalid RW_IO for unsupported data direction[%d]\n", scmd->sc_data_direction); @@ -1194,6 +1032,9 @@ static int hiraid_setup_rw_cmd(struct hiraid_dev *hdev, return -EINVAL; } + if (ret == RW_LENGTH_ZERO) + return ret; + /* 6-byte READ(0x08) or WRITE(0x0A) cdb */ if (scmd->cmd_len == 6) { datalength = (u32)(scmd->cmnd[4] == 0 ? @@ -1230,18 +1071,32 @@ static int hiraid_setup_rw_cmd(struct hiraid_dev *hdev, control |= HIRAID_RW_FUA; } - if (unlikely(datalength > U16_MAX || datalength == 0)) { - dev_err(hdev->dev, "invalid IO for err trans data length[%u]\n", + if (unlikely(datalength > U16_MAX)) { + dev_err(hdev->dev, "invalid IO for illegal transfer data length[%u]\n", datalength); WARN_ON(1); return -EINVAL; } + if (unlikely(datalength == 0)) + return RW_LENGTH_ZERO; + io_cmd->slba = cpu_to_le64(((u64)start_lba_hi << 32) | start_lba_lo); /* 0base for nlb */ io_cmd->nlb = cpu_to_le16((u16)(datalength - 1)); io_cmd->control = cpu_to_le16(control); + mapbuf->cdb_data_len = (u32)((io_cmd->nlb + 1) * sdev->sector_size); + if (mapbuf->cdb_data_len > buf_len) { + /* return DID_ERROR */ + dev_err(hdev->dev, "error: buf len[0x%x] is smaller than actual length[0x%x] sectorsize[0x%x]\n", + buf_len, mapbuf->cdb_data_len, sdev->sector_size); + return -EINVAL; + } else if (mapbuf->cdb_data_len < buf_len) { + dev_warn(hdev->dev, "warn: buf_len[0x%x] cdb_data_len[0x%x] nlb[0x%x] sectorsize[0x%x]\n", + buf_len, mapbuf->cdb_data_len, + io_cmd->nlb, sdev->sector_size); + } return 0; } @@ -1296,14 +1151,17 @@ static bool hiraid_disk_is_hdd_rawdrive(u8 attr) } static int hiraid_setup_io_cmd(struct hiraid_dev *hdev, - struct hiraid_scsi_io_cmd *io_cmd, - struct scsi_cmnd *scmd) + struct hiraid_scsi_io_cmd *io_cmd, struct scsi_cmnd *scmd, + struct hiraid_mapmange *mapbuf) { memcpy(io_cmd->common.cdb, scmd->cmnd, scmd->cmd_len); io_cmd->common.cdb_len = scmd->cmd_len; + /* init cdb_data_len */ + mapbuf->cdb_data_len = cpu_to_le32(scsi_bufflen(scmd)); + if (hiraid_is_rw_scmd(scmd)) - return hiraid_setup_rw_cmd(hdev, &io_cmd->rw, scmd); + return hiraid_setup_rw_cmd(hdev, &io_cmd->rw, scmd, mapbuf); else return hiraid_setup_nonrw_cmd(hdev, &io_cmd->nonrw, scmd); } @@ -1360,7 +1218,7 @@ static int hiraid_io_map_data(struct hiraid_dev *hdev, ret = scsi_dma_map(scmd); if (unlikely(ret < 0)) - return ret; + return SCSI_MLQUEUE_HOST_BUSY; mapbuf->sge_cnt = ret; /* No data to DMA, it may be scsi no-rw command */ @@ -1390,7 +1248,12 @@ static void hiraid_check_status(struct hiraid_mapmange *mapbuf, struct scsi_cmnd *scmd, struct hiraid_completion *cqe) { - scsi_set_resid(scmd, 0); + u32 datalength = cpu_to_le32(scsi_bufflen(scmd)); + + if (datalength > mapbuf->cdb_data_len) + scsi_set_resid(scmd, datalength - mapbuf->cdb_data_len); + else + scsi_set_resid(scmd, 0); switch ((le16_to_cpu(cqe->status) >> 1) & 0x7f) { case SENSE_STATE_OK: @@ -1443,6 +1306,35 @@ static inline void hiraid_query_scmd_tag(struct scsi_cmnd *scmd, u16 *qid, *cid = blk_mq_unique_tag_to_tag(tag); } +static void hiraid_set_ncq_prio(struct scsi_cmnd *scmd, + struct hiraid_scsi_io_cmd *io_cmd, + struct hiraid_sdev_hostdata *hostdata, + struct hiraid_dev *hdev) +{ + struct request *rq = scmd->request; + int class = 0; + + if (hostdata->sata_ncq_prio_enable) { + class = IOPRIO_PRIO_CLASS(req_get_ioprio(rq)); + switch (class) { + case IOPRIO_CLASS_NONE: + break; + case IOPRIO_CLASS_RT: + io_cmd->rw.rsvd2 &= SCSI_RSVD_INIT_SHIFT; + io_cmd->rw.rsvd2 |= (1 << IOPRIO_CLASS_RT); + break; + case IOPRIO_CLASS_BE: + break; + case IOPRIO_CLASS_IDLE: + break; + default: + break; + } + } + dev_log_dbg(hdev->dev, "blk_ncq_prio = %d, scmd_ncq_prio = 0x%02x\n", + class, io_cmd->rw.rsvd2); +} + static int hiraid_queue_command(struct Scsi_Host *shost, struct scsi_cmnd *scmd) { @@ -1452,7 +1344,7 @@ static int hiraid_queue_command(struct Scsi_Host *shost, struct hiraid_sdev_hostdata *hostdata; struct hiraid_scsi_io_cmd io_cmd; struct hiraid_queue *ioq; - struct HIRAID_STREAM_S *tmp_stm = NULL; + struct hiraid_stream *tmp_stream; u16 hwq, cid; int ret; @@ -1482,9 +1374,16 @@ static int hiraid_queue_command(struct Scsi_Host *shost, io_cmd.rw.hdid = cpu_to_le32(hostdata->hdid); io_cmd.rw.cmd_id = cpu_to_le16(cid); - ret = hiraid_setup_io_cmd(hdev, &io_cmd, scmd); + hiraid_set_ncq_prio(scmd, &io_cmd, hostdata, hdev); + + ret = hiraid_setup_io_cmd(hdev, &io_cmd, scmd, mapbuf); if (unlikely(ret)) { - set_host_byte(scmd, DID_ERROR); + if (ret == RW_LENGTH_ZERO) { + scsi_set_resid(scmd, scsi_bufflen(scmd)); + set_host_byte(scmd, DID_OK); + } else { + set_host_byte(scmd, DID_ERROR); + } scmd->scsi_done(scmd); atomic_dec(&ioq->inflight); return 0; @@ -1507,6 +1406,10 @@ static int hiraid_queue_command(struct Scsi_Host *shost, mapbuf->cid = cid; ret = hiraid_io_map_data(hdev, mapbuf, scmd, &io_cmd); if (unlikely(ret)) { + if (ret == SCSI_MLQUEUE_HOST_BUSY) { + atomic_dec(&ioq->inflight); + return ret; + } dev_err(hdev->dev, "io map data err\n"); set_host_byte(scmd, DID_ERROR); scmd->scsi_done(scmd); @@ -1515,20 +1418,20 @@ static int hiraid_queue_command(struct Scsi_Host *shost, } WRITE_ONCE(mapbuf->state, CMD_FLIGHT); - if (hiraid_is_rw_scmd(scmd) && hiraid_disk_is_hdd_rawdrive(hostdata->attr)) { - if (hiraid_detect_if_aging()) + if (hiraid_aging_detect(hdev)) hiraid_aging(hdev); - tmp_stm = hiraid_stream_detect(hdev, &(io_cmd.rw), sdev->id); - if (tmp_stm != NULL) { - hiraid_io_recognition_iterator(tmp_stm, 1); - if (hiraid_recognition_acknowledge(tmp_stm) && - (hiraid_get_stream_num(sdev->id) > 1)) { - if (hiraid_add_io_to_list(ioq, - tmp_stm, io_cmd, sdev->sector_size, sdev->id)) { + tmp_stream = hiraid_stream_detect(hdev, &(io_cmd.rw), sdev->id); + if (tmp_stream != NULL) { + hiraid_io_recognition_iterator(tmp_stream, 1); + if (hiraid_recognition_acknowledge(tmp_stream) && + (sdev->id < HIRAID_MAX_PD_NUM) && + (hdev->hiraid_stream_io_num_per_pd[sdev->id][HIRAID_STREAM_TYPE_TOTAL] > + 1)) { + if (hiraid_add_io_to_list(hdev, ioq, tmp_stream, + io_cmd, sdev->sector_size, sdev->id)) return 0; - } } } } @@ -1666,6 +1569,7 @@ static int hiraid_slave_configure(struct scsi_device *sdev) static void hiraid_shost_init(struct hiraid_dev *hdev) { struct pci_dev *pdev = hdev->pdev; + struct scsi_host_template *hostt; u8 domain, bus; u32 dev_func; @@ -1674,8 +1578,10 @@ static void hiraid_shost_init(struct hiraid_dev *hdev) dev_func = pdev->devfn; hdev->shost->nr_hw_queues = work_mode ? 1 : hdev->online_queues - 1; - hdev->shost->can_queue = hdev->scsi_qd; - + hdev->shost->can_queue = min(hdev->scsi_qd, + le16_to_cpu(hdev->ctrl_info->max_cmds) - + HIRAID_AQ_DEPTH - + HIRAID_TOTAL_PTCMDS(hdev->online_queues - 1)); hdev->shost->sg_tablesize = le16_to_cpu(hdev->ctrl_info->max_num_sge); /* 512B per sector */ hdev->shost->max_sectors = @@ -1689,7 +1595,13 @@ static void hiraid_shost_init(struct hiraid_dev *hdev) hdev->shost->this_id = -1; hdev->shost->unique_id = (domain << 16) | (bus << 8) | dev_func; hdev->shost->max_cmd_len = MAX_CDB_LEN; - hdev->shost->hostt->cmd_size = hiraid_get_max_cmd_size(hdev); + + hostt = (struct scsi_host_template *)hdev->shost->hostt; + hostt->cmd_size = hiraid_get_max_cmd_size(hdev); + + dev_info(hdev->dev, "nr_hw_queues[%u] can_queue[%u] unique_id[%u] cmd_size[%u]\n", + hdev->shost->nr_hw_queues, hdev->shost->can_queue, + hdev->shost->unique_id, hdev->shost->hostt->cmd_size); } static int hiraid_alloc_queue(struct hiraid_dev *hdev, u16 qid, u16 depth) @@ -2505,6 +2417,14 @@ static int hiraid_get_queue_cnt(struct hiraid_dev *hdev, u32 *cnt) return 0; } +static bool hiraid_pci_is_present(struct hiraid_dev *hdev) +{ + if (!pci_device_is_present(hdev->pdev) || + readl(hdev->bar + HIRAID_REG_VS) == 0Xffffffff) + return false; + return true; +} + static int hiraid_setup_io_queues(struct hiraid_dev *hdev) { struct hiraid_queue *adminq = &hdev->queues[0]; @@ -2579,7 +2499,7 @@ static void hiraid_delete_io_queues(struct hiraid_dev *hdev) u8 opcode = HIRAID_ADMIN_DELETE_SQ; u16 i, pass; - if (!pci_device_is_present(hdev->pdev)) { + if (!hiraid_pci_is_present(hdev)) { dev_err(hdev->dev, "pci_device is not present, skip disable io queues\n"); return; } @@ -2619,7 +2539,7 @@ static void hiraid_disable_admin_queue(struct hiraid_dev *hdev, struct hiraid_queue *adminq = &hdev->queues[0]; u16 start, end; - if (pci_device_is_present(hdev->pdev)) { + if (hiraid_pci_is_present(hdev)) { if (shutdown) hiraid_shutdown_control(hdev); else @@ -2850,11 +2770,16 @@ static int hiraid_luntarget_sort(const void *l, const void *r) const struct hiraid_dev_info *rn = r; int l_attr = HIRAID_DEV_INFO_ATTR_BOOT(ln->attr); int r_attr = HIRAID_DEV_INFO_ATTR_BOOT(rn->attr); - /* boot first */ if (l_attr != r_attr) return (r_attr - l_attr); + l_attr = HIRAID_DEV_INFO_ATTR_VD(ln->attr); + r_attr = HIRAID_DEV_INFO_ATTR_VD(rn->attr); + /* vd second */ + if (l_attr != r_attr) + return (r_attr - l_attr); + if (ln->channel == rn->channel) return le16_to_cpu(ln->target) - le16_to_cpu(rn->target); @@ -2892,6 +2817,19 @@ static void hiraid_scan_work(struct work_struct *work) if (HIRAID_DEV_INFO_FLAG_VALID(flag)) { if (!HIRAID_DEV_INFO_FLAG_VALID(org_flag)) { + down_write(&hdev->dev_rwsem); + memcpy(&old_dev[i], &dev[i], + sizeof(struct hiraid_dev_info)); + memcpy(&new_dev[count++], &dev[i], + sizeof(struct hiraid_dev_info)); + up_write(&hdev->dev_rwsem); + } else if (HIRAID_DEV_INFO_FLAG_DELETE_OR_ADD(flag, + org_flag)) { + down_write(&hdev->dev_rwsem); + old_dev[i].flag &= 0xfe; + up_write(&hdev->dev_rwsem); + hiraid_delete_device(hdev, &old_dev[i]); + down_write(&hdev->dev_rwsem); memcpy(&old_dev[i], &dev[i], sizeof(struct hiraid_dev_info)); @@ -3044,7 +2982,7 @@ static void hiraid_bsg_buf_unmap(struct hiraid_dev *hdev, struct bsg_job *job) } static int hiraid_bsg_buf_map(struct hiraid_dev *hdev, struct bsg_job *job, - struct hiraid_admin_command *cmd) + struct hiraid_admin_command *cmd, u16 qid) { struct hiraid_bsg_request *bsg_req = job->request; struct request *rq = blk_mq_rq_from_pdu(job); @@ -3058,24 +2996,26 @@ static int hiraid_bsg_buf_map(struct hiraid_dev *hdev, struct bsg_job *job, mapbuf->sgl = job->request_payload.sg_list; mapbuf->len = job->request_payload.payload_len; mapbuf->page_cnt = -1; + mapbuf->hiraidq = &hdev->queues[qid]; if (unlikely(mapbuf->sge_cnt == 0)) goto out; - mapbuf->use_sgl = !hiraid_is_prp(hdev, mapbuf->sgl, mapbuf->sge_cnt); - ret = dma_map_sg_attrs(hdev->dev, mapbuf->sgl, mapbuf->sge_cnt, - dma_dir, DMA_ATTR_NO_WARN); + dma_dir, DMA_ATTR_NO_WARN); if (!ret) goto out; - if ((mapbuf->use_sgl == (bool)true) && + mapbuf->use_sgl = !hiraid_is_prp(hdev, mapbuf->sgl, mapbuf->sge_cnt); + + if ((mapbuf->use_sgl == true) && (bsg_req->msgcode == HIRAID_BSG_IOPTHRU) && - (hdev->ctrl_info->pt_use_sgl != (bool)false)) { - ret = hiraid_build_passthru_sgl(hdev, cmd, mapbuf); + (hdev->ctrl_info->pt_use_sgl != false)) { + ret = hiraid_build_sgl(hdev, (struct hiraid_scsi_io_cmd *)cmd, + mapbuf); } else { mapbuf->use_sgl = false; - ret = hiraid_build_passthru_prp(hdev, mapbuf); + ret = hiraid_build_prp(hdev, mapbuf); cmd->common.dptr.prp1 = cpu_to_le64(sg_dma_address(mapbuf->sgl)); cmd->common.dptr.prp2 = cpu_to_le64(mapbuf->first_dma); @@ -3156,9 +3096,7 @@ static int hiraid_init_control_info(struct hiraid_dev *hdev) if (hdev->ctrl_info->asynevent > HIRAID_ASYN_COMMANDS) hdev->ctrl_info->asynevent = HIRAID_ASYN_COMMANDS; - hdev->scsi_qd = work_mode ? - le16_to_cpu(hdev->ctrl_info->max_cmds) : - (hdev->ioq_depth - HIRAID_PTHRU_CMDS_PERQ); + hdev->scsi_qd = hdev->ioq_depth - HIRAID_PTHRU_CMDS_PERQ; return 0; } @@ -3191,7 +3129,7 @@ static int hiraid_user_send_admcmd(struct hiraid_dev *hdev, struct bsg_job *job) admin_cmd.common.cdw14 = cpu_to_le32(ptcmd->cdw14); admin_cmd.common.cdw15 = cpu_to_le32(ptcmd->cdw15); - status = hiraid_bsg_buf_map(hdev, job, &admin_cmd); + status = hiraid_bsg_buf_map(hdev, job, &admin_cmd, 0); if (status) { dev_err(hdev->dev, "err, map data failed\n"); return status; @@ -3251,19 +3189,13 @@ static void hiraid_free_io_ptcmds(struct hiraid_dev *hdev) } static int hiraid_put_io_sync_request(struct hiraid_dev *hdev, - struct hiraid_scsi_io_cmd *io_cmd, - u32 *result, u32 *reslen, u32 timeout) + struct hiraid_scsi_io_cmd *io_cmd, struct hiraid_cmd *pt_cmd, + u32 *result, u32 *reslen, u32 timeout) { int ret; dma_addr_t buffer_phy; struct hiraid_queue *ioq; void *sense_addr = NULL; - struct hiraid_cmd *pt_cmd = hiraid_get_cmd(hdev, HIRAID_CMD_PTHRU); - - if (!pt_cmd) { - dev_err(hdev->dev, "err, get ioq cmd failed\n"); - return -EFAULT; - } timeout = timeout ? timeout : ADMIN_TIMEOUT; @@ -3271,8 +3203,9 @@ static int hiraid_put_io_sync_request(struct hiraid_dev *hdev, ioq = &hdev->queues[pt_cmd->qid]; if (work_mode) { - ret = ((pt_cmd->qid - 1) * HIRAID_PTHRU_CMDS_PERQ + - pt_cmd->cid) * SCSI_SENSE_BUFFERSIZE; + ret = ((pt_cmd->qid - 1) * + HIRAID_PTHRU_CMDS_PERQ + pt_cmd->cid) * + SCSI_SENSE_BUFFERSIZE; sense_addr = hdev->sense_buffer_virt + ret; buffer_phy = hdev->sense_buffer_phy + ret; } else { @@ -3293,8 +3226,6 @@ static int hiraid_put_io_sync_request(struct hiraid_dev *hdev, (le32_to_cpu(io_cmd->common.cdw3[0]) & 0xffff)); hiraid_admin_timeout(hdev, pt_cmd); - - hiraid_put_cmd(hdev, pt_cmd, HIRAID_CMD_PTHRU); return -ETIME; } @@ -3305,8 +3236,6 @@ static int hiraid_put_io_sync_request(struct hiraid_dev *hdev, } } - hiraid_put_cmd(hdev, pt_cmd, HIRAID_CMD_PTHRU); - return pt_cmd->status; } @@ -3316,6 +3245,7 @@ static int hiraid_user_send_ptcmd(struct hiraid_dev *hdev, struct bsg_job *job) (struct hiraid_bsg_request *)(job->request); struct hiraid_passthru_io_cmd *cmd = &(bsg_req->pthrucmd); struct hiraid_scsi_io_cmd pthru_cmd; + struct hiraid_cmd *pt_cmd; int status = 0; u32 timeout = msecs_to_jiffies(cmd->timeout_ms); // data len is 4k before use sgl, now len is 1M @@ -3358,16 +3288,21 @@ static int hiraid_user_send_ptcmd(struct hiraid_dev *hdev, struct bsg_job *job) pthru_cmd.common.cdw26[2] = cpu_to_le32(cmd->cdw26[2]); pthru_cmd.common.cdw26[3] = cpu_to_le32(cmd->cdw26[3]); + pt_cmd = hiraid_get_cmd(hdev, HIRAID_CMD_PTHRU); + if (pt_cmd == NULL) { + dev_err(hdev->dev, "err, get ioq cmd failed\n"); + return -EFAULT; + } + status = hiraid_bsg_buf_map(hdev, job, - (struct hiraid_admin_command *)&pthru_cmd); + (struct hiraid_admin_command *)&pthru_cmd, pt_cmd->qid); if (status) { dev_err(hdev->dev, "err, map data failed\n"); - return status; + goto ret; } - status = hiraid_put_io_sync_request(hdev, &pthru_cmd, job->reply, - &job->reply_len, timeout); - + status = hiraid_put_io_sync_request(hdev, &pthru_cmd, pt_cmd, + job->reply, &job->reply_len, timeout); if (status) dev_info(hdev->dev, "opcode[0x%x] subopcode[0x%x] status[0x%x] replylen[%d]\n", cmd->opcode, cmd->info_1.subopcode, @@ -3375,6 +3310,9 @@ static int hiraid_user_send_ptcmd(struct hiraid_dev *hdev, struct bsg_job *job) hiraid_bsg_buf_unmap(hdev, job); +ret: + hiraid_put_cmd(hdev, pt_cmd, HIRAID_CMD_PTHRU); + return status; } @@ -3396,23 +3334,28 @@ static bool hiraid_check_scmd_finished(struct scsi_cmnd *scmd) return false; } -static enum blk_eh_timer_return hiraid_timed_out(struct scsi_cmnd *scmd) +#define EH_TIMED_RET_TP enum blk_eh_timer_return +#define EH_TIMED_RET_DONE BLK_EH_DONE +#define EH_TIMED_RET_NOT_HANDLED BLK_EH_DONE +#define EH_TIMED_RET_RESET_TIMER BLK_EH_RESET_TIMER +static EH_TIMED_RET_TP hiraid_timed_out(struct scsi_cmnd *scmd) { struct hiraid_mapmange *mapbuf = scsi_cmd_priv(scmd); unsigned int timeout = scmd->device->request_queue->rq_timeout; if (hiraid_check_scmd_finished(scmd)) - goto out; + return EH_TIMED_RET_DONE; - if (time_after(jiffies, scmd->jiffies_at_alloc + timeout)) { + if (time_after_eq(jiffies, scmd->jiffies_at_alloc + timeout)) { + scmd_printk(KERN_WARNING, scmd, "timeout is work[%d], need abort\n", + mapbuf->state); if (cmpxchg(&mapbuf->state, CMD_FLIGHT, CMD_TIMEOUT) == CMD_FLIGHT) - return BLK_EH_DONE; + return EH_TIMED_RET_NOT_HANDLED; } -out: - return BLK_EH_RESET_TIMER; -} + return EH_TIMED_RET_RESET_TIMER; +} /* send abort command by admin queue temporary */ static int hiraid_send_abort_cmd(struct hiraid_dev *hdev, u32 hdid, u16 qid, u16 cid) @@ -3535,7 +3478,7 @@ static int hiraid_dev_disable(struct hiraid_dev *hdev, bool shutdown) struct hiraid_queue *adminq = &hdev->queues[0]; u16 start, end; - if (pci_device_is_present(hdev->pdev)) { + if (hiraid_pci_is_present(hdev)) { if (shutdown) hiraid_shutdown_control(hdev); else @@ -3733,11 +3676,19 @@ static int hiraid_abort(struct scsi_cmnd *scmd) hiraid_check_scmd_finished(scmd)) return SUCCESS; + if (hdev->state != DEV_LIVE) + return SUCCESS; + hostdata = scmd->device->hostdata; cid = mapbuf->cid; hwq = mapbuf->hiraidq->qid; - if (hiraid_check_io_list(hostdata->hdid, cid, hwq)) { + mutex_lock(&g_stream_operation_mutex); + ret = hiraid_stream_delete_io_list(hdev, hostdata->hdid, + cid, hwq, HIRAID_STREAM_TYPE_DELETE_SINGLE_IO); + mutex_unlock(&g_stream_operation_mutex); + + if (ret == 1) { dev_warn(hdev->dev, "find cid[%d] qid[%d] in host, abort succ\n", cid, hwq); return SUCCESS; @@ -3749,14 +3700,14 @@ static int hiraid_abort(struct scsi_cmnd *scmd) ret = hiraid_wait_io_completion(mapbuf); if (ret) { dev_warn(hdev->dev, "cid[%d] qid[%d] abort failed, not found\n", - cid, hwq); + cid, hwq); return FAILED; } dev_warn(hdev->dev, "cid[%d] qid[%d] abort succ\n", cid, hwq); return SUCCESS; } dev_warn(hdev->dev, "cid[%d] qid[%d] abort failed, timeout\n", - cid, hwq); + cid, hwq); return FAILED; } @@ -3778,7 +3729,10 @@ static int hiraid_scsi_reset(struct scsi_cmnd *scmd, enum hiraid_rst_type rst) if ((ret == 0) || (ret == FW_EH_DEV_NONE && rst == HIRAID_RESET_TARGET)) { if (rst == HIRAID_RESET_TARGET) { - hiraid_delete_single_pd_io_list(hostdata->hdid); + mutex_lock(&g_stream_operation_mutex); + hiraid_stream_delete_io_list(hdev, hostdata->hdid, + 0, 0, HIRAID_STREAM_TYPE_DELETE_SINGLE_IO_LIST); + mutex_unlock(&g_stream_operation_mutex); ret = wait_tgt_reset_io_done(scmd); if (ret) { dev_warn(hdev->dev, "sdev[%d:%d] target has %d peding cmd, target reset failed\n", @@ -3818,9 +3772,13 @@ static int hiraid_host_reset(struct scsi_cmnd *scmd) dev_warn(hdev->dev, "sdev[%d:%d] send host reset\n", scmd->device->channel, scmd->device->id); - hiraid_delete_all_io_list(); - if (hiraid_reset_work_sync(hdev) == -EBUSY) + mutex_lock(&g_stream_operation_mutex); + hiraid_stream_delete_io_list(hdev, 0, 0, 0, + HIRAID_STREAM_TYPE_DELETE_ALL_IO_LIST); + mutex_unlock(&g_stream_operation_mutex); + if (hiraid_reset_work_sync(hdev) == -EBUSY) { flush_work(&hdev->reset_work); + } if (hdev->state != DEV_LIVE) { dev_warn(hdev->dev, "sdev[%d:%d] host reset failed\n", @@ -3852,7 +3810,10 @@ static pci_ers_result_t hiraid_pci_error_detected(struct pci_dev *pdev, scsi_block_requests(hdev->shost); hiraid_dev_state_trans(hdev, DEV_RESETTING); - hiraid_delete_all_io_list(); + mutex_lock(&g_stream_operation_mutex); + hiraid_stream_delete_io_list(hdev, 0, 0, 0, + HIRAID_STREAM_TYPE_DELETE_ALL_IO_LIST); + mutex_unlock(&g_stream_operation_mutex); return PCI_ERS_RESULT_NEED_RESET; case pci_channel_io_perm_failure: @@ -3888,9 +3849,24 @@ static void hiraid_reset_pci_finish(struct pci_dev *pdev) { struct hiraid_dev *hdev = pci_get_drvdata(pdev); + if (hiraid_reset_work_sync(hdev) == -EBUSY) + flush_work(&hdev->reset_work); + dev_info(hdev->dev, "enter hiraid reset finish\n"); } +static void hiraid_reset_pci_prepare(struct pci_dev *pdev) +{ + struct hiraid_dev *hdev = pci_get_drvdata(pdev); + + mutex_lock(&g_stream_operation_mutex); + hiraid_stream_delete_io_list(hdev, 0, 0, 0, + HIRAID_STREAM_TYPE_DELETE_ALL_IO_LIST); + mutex_unlock(&g_stream_operation_mutex); + msleep(100); + dev_info(hdev->dev, "exit hiraid reset prepare\n"); +} + static ssize_t csts_pp_show(struct device *cdev, struct device_attribute *attr, char *buf) { @@ -3898,7 +3874,7 @@ static ssize_t csts_pp_show(struct device *cdev, struct hiraid_dev *hdev = shost_priv(shost); int ret = -1; - if (pci_device_is_present(hdev->pdev)) { + if (hiraid_pci_is_present(hdev)) { ret = (readl(hdev->bar + HIRAID_REG_CSTS) & HIRAID_CSTS_PP_MASK); ret >>= HIRAID_CSTS_PP_SHIFT; @@ -3914,7 +3890,7 @@ static ssize_t csts_shst_show(struct device *cdev, struct hiraid_dev *hdev = shost_priv(shost); int ret = -1; - if (pci_device_is_present(hdev->pdev)) { + if (hiraid_pci_is_present(hdev)) { ret = (readl(hdev->bar + HIRAID_REG_CSTS) & HIRAID_CSTS_SHST_MASK); ret >>= HIRAID_CSTS_SHST_SHIFT; @@ -3930,7 +3906,7 @@ static ssize_t csts_cfs_show(struct device *cdev, struct hiraid_dev *hdev = shost_priv(shost); int ret = -1; - if (pci_device_is_present(hdev->pdev)) { + if (hiraid_pci_is_present(hdev)) { ret = (readl(hdev->bar + HIRAID_REG_CSTS) & HIRAID_CSTS_CFS_MASK); ret >>= HIRAID_CSTS_CFS_SHIFT; @@ -3946,7 +3922,7 @@ static ssize_t csts_rdy_show(struct device *cdev, struct hiraid_dev *hdev = shost_priv(shost); int ret = -1; - if (pci_device_is_present(hdev->pdev)) + if (hiraid_pci_is_present(hdev)) ret = (readl(hdev->bar + HIRAID_REG_CSTS) & HIRAID_CSTS_RDY); return snprintf(buf, PAGE_SIZE, "%d\n", ret); @@ -4239,22 +4215,78 @@ static ssize_t dispatch_hwq_store(struct device *dev, return strlen(buf); } +static ssize_t sata_ncq_prio_enable_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct hiraid_sdev_hostdata *hostdata; + + hostdata = to_scsi_device(dev)->hostdata; + return snprintf(buf, PAGE_SIZE, "%d\n", hostdata->sata_ncq_prio_enable); +} + +bool scsi_ncq_prio_support(struct scsi_device *sdev, struct hiraid_dev *hdev) +{ + unsigned char *buf; + bool ncq_prio_supp = false; + + if (!scsi_device_supports_vpd(sdev)) { + dev_err(hdev->dev, "scsi device not support vpd\n"); + return ncq_prio_supp; + } + + buf = kzalloc(HIRAID_SCSI_VPD_LEN, GFP_KERNEL); + if (!buf) + return ncq_prio_supp; + + if (!scsi_get_vpd_page(sdev, 0x89, buf, HIRAID_SCSI_VPD_LEN)) + ncq_prio_supp = (buf[NCQ_PRIO_SUPPORT_BYTE] >> + VPD_NCQ_PRIO_SUPPORT_BIT) & 1; + + kfree(buf); + return ncq_prio_supp; +} + +static ssize_t sata_ncq_prio_enable_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + struct scsi_device *sdev = to_scsi_device(dev); + struct hiraid_sdev_hostdata *hostdata = sdev->hostdata; + struct hiraid_dev *hdev = shost_priv(sdev->host); + bool ncq_prio_enable = 0; + + if (kstrtobool(buf, &ncq_prio_enable)) + return -EINVAL; + + if (!scsi_ncq_prio_support(sdev, hdev)) { + dev_err(hdev->dev, "sdev[%d:%d] not support ncq prio\n", + sdev->channel, sdev->id); + return -EINVAL; + } + + hostdata->sata_ncq_prio_enable = ncq_prio_enable; + + return strlen(buf); +} + static DEVICE_ATTR_RO(raid_level); static DEVICE_ATTR_RO(raid_state); static DEVICE_ATTR_RO(raid_resync); static DEVICE_ATTR_RW(dispatch_hwq); +static DEVICE_ATTR_RW(sata_ncq_prio_enable); static struct device_attribute *hiraid_dev_attrs[] = { &dev_attr_raid_state, &dev_attr_raid_level, &dev_attr_raid_resync, &dev_attr_dispatch_hwq, + &dev_attr_sata_ncq_prio_enable, NULL, }; static struct pci_error_handlers hiraid_err_handler = { .error_detected = hiraid_pci_error_detected, .slot_reset = hiraid_pci_slot_reset, + .reset_prepare = hiraid_reset_pci_prepare, .reset_done = hiraid_reset_pci_finish, }; @@ -4518,14 +4550,17 @@ static void hiraid_remove(struct pci_dev *pdev) dev_info(hdev->dev, "enter hiraid remove\n"); - kthread_stop(g_hiraid_submit_task); - hiraid_delete_all_io_list(); + kthread_stop(hdev->hiraid_stream_submit_task); + mutex_lock(&g_stream_operation_mutex); + hiraid_stream_delete_io_list(hdev, 0, 0, 0, + HIRAID_STREAM_TYPE_DELETE_ALL_IO_LIST); + mutex_unlock(&g_stream_operation_mutex); hiraid_destroy_io_stream_resource(hdev); hiraid_dev_state_trans(hdev, DEV_DELETING); flush_work(&hdev->reset_work); - if (!pci_device_is_present(pdev)) + if (!hiraid_pci_is_present(hdev)) hiraid_flush_running_cmds(hdev); hiraid_unregist_bsg(hdev); @@ -4548,8 +4583,12 @@ static void hiraid_remove(struct pci_dev *pdev) static const struct pci_device_id hiraid_hw_card_ids[] = { { PCI_DEVICE(PCI_VENDOR_ID_HUAWEI_LOGIC, HIRAID_SERVER_DEVICE_HBA_DID) }, + { PCI_DEVICE(PCI_VENDOR_ID_HUAWEI_LOGIC, + HIRAID_SERVER_DEVICE_HBAS_DID) }, { PCI_DEVICE(PCI_VENDOR_ID_HUAWEI_LOGIC, HIRAID_SERVER_DEVICE_RAID_DID) }, + { PCI_DEVICE(PCI_VENDOR_ID_HUAWEI_LOGIC, + HIRAID_SERVER_DEVICE_RAIDS_DID) }, { 0, } }; MODULE_DEVICE_TABLE(pci, hiraid_hw_card_ids); @@ -4572,22 +4611,9 @@ static int __init hiraid_init(void) if (!work_queue) return -ENOMEM; - hiraid_class = class_create(THIS_MODULE, "hiraid"); - if (IS_ERR(hiraid_class)) { - ret = PTR_ERR(hiraid_class); - goto destroy_wq; - } - ret = pci_register_driver(&hiraid_driver); if (ret < 0) - goto destroy_class; - - return 0; - -destroy_class: - class_destroy(hiraid_class); -destroy_wq: - destroy_workqueue(work_queue); + destroy_workqueue(work_queue); return ret; } @@ -4595,7 +4621,6 @@ destroy_wq: static void __exit hiraid_exit(void) { pci_unregister_driver(&hiraid_driver); - class_destroy(hiraid_class); destroy_workqueue(work_queue); } -- 2.45.1.windows.1

2 5

[PATCH OLK-6.6 0/4] hungtask fix patch and mainline backport
by Chen Jinghuang 09 Jun '26

09 Jun '26

*** BLURB HERE *** Chen Jinghuang (1): sched: Fix kabi broken of struct sched_entity Peter Zijlstra (2): sched/fair: Fix lag clamp sched/fair: Revert 6d71a9c61604 ("sched/fair: Fix EEVDF entity placement bug causing scheduling lag") Zhan Xusheng (1): sched/fair: Fix math notation errors in avg_vruntime comment include/linux/sched.h | 2 +- kernel/sched/fair.c | 188 +++++++++++++++++++++++++++++++++++------- 2 files changed, 161 insertions(+), 29 deletions(-) -- 2.34.1

2 5

[PATCH OLK-6.6 0/5]
by ZongYu Wu 09 Jun '26

09 Jun '26

From: parm64 <parm64(a)huawei.com> Chenghai Huang (1): crypto: hisilicon/zip - add backlog support for zip Wenkai Lin (3): crypto: hisilicon/sec2 - fix UAF in sec_alg_send_backlog crypto: hisilicon/sec2 - fix resource leakage issues in non-backlog mode crypto: hisilicon/sec2 - restore iv for ctr mode lizhi (1): crypto: hisilicon/hpre: implement full backlog support for hpre driver drivers/crypto/hisilicon/hpre/hpre_crypto.c | 287 ++++++++++++------ drivers/crypto/hisilicon/sec2/sec_crypto.c | 48 ++- drivers/crypto/hisilicon/zip/zip_crypto.c | 313 +++++++++++++------- 3 files changed, 426 insertions(+), 222 deletions(-) -- 2.43.0

2 6

[PATCH OLK-6.6 0/2] Fix livepatch and kretprobe conflict warning
by Tengda Wu 08 Jun '26

08 Jun '26

This series is going to fix livepatch and kretprobe conflict warning. The livepatch has stop_machine() to avoid race condition of tsk->on_cpu state check. It's ok to replace task_is_running to tsk->on_cpu. Kui-Feng Lee (1): rethook: Remove warning messages printed for finding return address of a frame. Tengda Wu (1): rethook: Use tsk->on_cpu to check task execution state kernel/trace/rethook.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) -- 2.34.1

2 3

[PATCH OLK-6.6 00/29] backport minios patches from linux mainline
by Qinxin Xia 08 Jun '26

08 Jun '26

From: Hongye Lin <linhongye(a)h-partners.com> driver inclusion category: feature bugzilla: https://atomgit.com/openeuler/kernel/issues/8904 ---------------------------------------------------------------------- backport minios patches from linux mainline Mark Rutland (8): arm64/fpsimd: Add fpsimd_save_and_flush_current_state() arm64/fpsimd: signal32: Always save+flush state early arm64/fpsimd: signal: Always save+flush state early arm64: ptrace: fix partial SETREGSET for NT_ARM_FPMR arm64/fpsimd: Reset FPMR upon exec() arm64/fpsimd: Don't corrupt FPMR when streaming mode changes arm64/fpsimd: Do not discard modified SVE state arm64/fpsimd: Avoid warning when sve_to_fpsimd() is unused Qinxin Xia (18): arm64/sysreg: Add definition for ID_AA64PFR2_EL1 arm64/sysreg: Add definition for ID_AA64ISAR3_EL1 arm64/sysreg: Add definition for ID_AA64FPFR0_EL1 arm64/sysreg: Add definition for FPMR arm64/sysreg: Add EnFPM field for SCTLR_EL1 and HCRX_EL2 arm64/sysreg: Add LUT field for ID_AA64ISAR2_EL1 arm64/cpufeature: Hook new identification registers up to cpufeature arm64/hwcap: Define hwcaps for 2023 DPISA features kselftest/arm64: Add basic FPMR test kselftest/arm64: Handle FPMR context in generic signal frame parser kselftest/arm64: Add 2023 DPISA hwcap test coverage arm64: Kconfig: Detect toolchain support for LSUI arm64: cpufeature: add FEAT_LSUI arm64/signal: Add FPMR signal handling arm64/ptrace: Expose FPMR via ptrace Add support "arm64.nocnp" for start option arm64/fpsimd: Support FEAT_FPMR arm64: syscall: switch to fpsimd_syscall_enter() Yifan Wu (3): arm64/hwcap: Add support for FEAT_CMPBR kselftest/arm64: Add FEAT_CMPBR to the hwcap selftest arm64/sysreg: Update ID_AA64ISAR2_EL1 for FEAT_CMPBR Documentation/arch/arm64/elf_hwcaps.rst | 30 +++++ arch/arm64/Kconfig | 5 + arch/arm64/include/asm/cpu.h | 3 + arch/arm64/include/asm/cpufeature.h | 6 + arch/arm64/include/asm/fpsimd.h | 7 +- arch/arm64/include/asm/hwcap.h | 10 ++ arch/arm64/include/asm/kvm_host.h | 1 + arch/arm64/include/asm/processor.h | 4 + arch/arm64/include/asm/signal_common.h | 20 ++- arch/arm64/include/uapi/asm/hwcap.h | 10 ++ arch/arm64/include/uapi/asm/sigcontext.h | 8 ++ arch/arm64/kernel/cpufeature.c | 70 +++++++++- arch/arm64/kernel/cpuinfo.c | 3 + arch/arm64/kernel/entry-common.c | 50 +++++-- arch/arm64/kernel/fpsimd.c | 74 +++++------ arch/arm64/kernel/hwcap_str.h | 10 ++ arch/arm64/kernel/idreg-override.c | 11 ++ arch/arm64/kernel/ptrace.c | 44 +++++++ arch/arm64/kernel/signal.c | 102 ++++++++------- arch/arm64/kernel/signal32.c | 11 +- arch/arm64/kvm/fpsimd.c | 1 + arch/arm64/tools/cpucaps | 2 + arch/arm64/tools/sysreg | 122 +++++++++++++++++- include/uapi/linux/elf.h | 1 + tools/testing/selftests/arm64/abi/hwcap.c | 116 +++++++++++++++++ .../testing/selftests/arm64/signal/.gitignore | 1 + .../arm64/signal/testcases/fpmr_siginfo.c | 82 ++++++++++++ .../arm64/signal/testcases/testcases.c | 8 ++ .../arm64/signal/testcases/testcases.h | 1 + 29 files changed, 704 insertions(+), 109 deletions(-) create mode 100644 tools/testing/selftests/arm64/signal/testcases/fpmr_siginfo.c -- 2.33.0

2 30

[PATCH OLK-5.10] ALSA: ctxfi: Add fallback to default RSR for S/PDIF
by Zhang Yuwei 08 Jun '26

08 Jun '26

From: Harin Lee <me(a)harin.net> mainline inclusion from mainline-v7.1-rc1 commit 7d61662197ecdc458e33e475b6ada7f6da61d364 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/15293 CVE: CVE-2026-46049 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- spdif_passthru_playback_get_resources() uses atc->pll_rate as the RSR for the MSR calculation loop. However, pll_rate is only updated in atc_pll_init() and not in hw_pll_init(), so it remains 0 after the card init. When spdif_passthru_playback_setup() skips atc_pll_init() for 32000 Hz, (rsr * desc.msr) always becomes 0, causing the loop to spin indefinitely. Add fallback to use atc->rsr when atc->pll_rate is 0. This reflects the hardware state, since hw_card_init() already configures the PLL to the default RSR. Fixes: 8cc72361481f ("ALSA: SB X-Fi driver merge") Cc: stable(a)vger.kernel.org Signed-off-by: Harin Lee <me(a)harin.net> Link: https://patch.msgid.link/20260406074913.217374-1-me@harin.net Signed-off-by: Takashi Iwai <tiwai(a)suse.de> Signed-off-by: Zhang Yuwei <zhangyuwei20(a)huawei.com> --- sound/pci/ctxfi/ctatc.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/sound/pci/ctxfi/ctatc.c b/sound/pci/ctxfi/ctatc.c index 06775519dab0..fd821303999a 100644 --- a/sound/pci/ctxfi/ctatc.c +++ b/sound/pci/ctxfi/ctatc.c @@ -791,7 +791,8 @@ static int spdif_passthru_playback_get_resources(struct ct_atc *atc, struct src *src; int err; int n_amixer = apcm->substream->runtime->channels, i; - unsigned int pitch, rsr = atc->pll_rate; + unsigned int pitch; + unsigned int rsr = atc->pll_rate ? atc->pll_rate : atc->rsr; /* first release old resources */ atc_pcm_release_resources(atc, apcm); -- 2.22.0

2 1

[PATCH OLK-6.6] iommu/vt-d: Clear Present bit before tearing down context entry
by Zhang Yuwei 08 Jun '26

08 Jun '26

From: Lu Baolu <baolu.lu(a)linux.intel.com> mainline inclusion from mainline-v7.0-rc1 commit c1e4f1dccbe9d7656d1c6872ebeadb5992d0aaa2 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/15190 CVE: CVE-2026-45944 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- When tearing down a context entry, the current implementation zeros the entire 128-bit entry using multiple 64-bit writes. This creates a window where the hardware can fetch a "torn" entry — where some fields are already zeroed while the 'Present' bit is still set — leading to unpredictable behavior or spurious faults. While x86 provides strong write ordering, the compiler may reorder writes to the two 64-bit halves of the context entry. Even without compiler reordering, the hardware fetch is not guaranteed to be atomic with respect to multiple CPU writes. Align with the "Guidance to Software for Invalidations" in the VT-d spec (Section 6.5.3.3) by implementing the recommended ownership handshake: 1. Clear only the 'Present' (P) bit of the context entry first to signal the transition of ownership from hardware to software. 2. Use dma_wmb() to ensure the cleared bit is visible to the IOMMU. 3. Perform the required cache and context-cache invalidation to ensure hardware no longer has cached references to the entry. 4. Fully zero out the entry only after the invalidation is complete. Also, add a dma_wmb() to context_set_present() to ensure the entry is fully initialized before the 'Present' bit becomes visible. Fixes: ba39592764ed ("Intel IOMMU: Intel IOMMU driver") Reported-by: Dmytro Maluka <dmaluka(a)chromium.org> Closes: https://lore.kernel.org/all/aTG7gc7I5wExai3S@google.com/ Signed-off-by: Lu Baolu <baolu.lu(a)linux.intel.com> Reviewed-by: Dmytro Maluka <dmaluka(a)chromium.org> Reviewed-by: Samiullah Khawaja <skhawaja(a)google.com> Reviewed-by: Kevin Tian <kevin.tian(a)intel.com> Link: https://lore.kernel.org/r/20260120061816.2132558-3-baolu.lu@linux.intel.com Signed-off-by: Joerg Roedel <joerg.roedel(a)amd.com> Conflicts: drivers/iommu/intel/pasid.c drivers/iommu/intel/iommu.c drivers/iommu/intel/iommu.h [contxt conflict] Signed-off-by: Zhang Yuwei <zhangyuwei20(a)huawei.com> --- drivers/iommu/intel/iommu.c | 4 +++- drivers/iommu/intel/iommu.h | 21 ++++++++++++++++++++- 2 files changed, 23 insertions(+), 2 deletions(-) diff --git a/drivers/iommu/intel/iommu.c b/drivers/iommu/intel/iommu.c index 86d713aefe91..ba4b298d456b 100644 --- a/drivers/iommu/intel/iommu.c +++ b/drivers/iommu/intel/iommu.c @@ -2180,7 +2180,7 @@ static void domain_context_clear_one(struct device_domain_info *info, u8 bus, u8 did_old = context_domain_id(context); } - context_clear_entry(context); + context_clear_present(context); __iommu_flush_cache(iommu, context, sizeof(*context)); spin_unlock(&iommu->lock); iommu->flush.flush_context(iommu, @@ -2199,6 +2199,8 @@ static void domain_context_clear_one(struct device_domain_info *info, u8 bus, u8 DMA_TLB_DSI_FLUSH); __iommu_flush_dev_iotlb(info, 0, MAX_AGAW_PFN_WIDTH); + context_clear_entry(context); + __iommu_flush_cache(iommu, context, sizeof(*context)); } static int domain_setup_first_level(struct intel_iommu *iommu, diff --git a/drivers/iommu/intel/iommu.h b/drivers/iommu/intel/iommu.h index 9e566a2f22df..791326fa95ba 100644 --- a/drivers/iommu/intel/iommu.h +++ b/drivers/iommu/intel/iommu.h @@ -929,7 +929,26 @@ static inline unsigned long virt_to_dma_pfn(void *p) static inline void context_set_present(struct context_entry *context) { - context->lo |= 1; + u64 val; + + dma_wmb(); + val = READ_ONCE(context->lo) | 1; + WRITE_ONCE(context->lo, val); +} + +/* + * Clear the Present (P) bit (bit 0) of a context table entry. This initiates + * the transition of the entry's ownership from hardware to software. The + * caller is responsible for fulfilling the invalidation handshake recommended + * by the VT-d spec, Section 6.5.3.3 (Guidance to Software for Invalidations). + */ +static inline void context_clear_present(struct context_entry *context) +{ + u64 val; + + val = READ_ONCE(context->lo) & GENMASK_ULL(63, 1); + WRITE_ONCE(context->lo, val); + dma_wmb(); } static inline void context_set_fault_enable(struct context_entry *context) -- 2.22.0

2 1

[PATCH OLK-5.10] iommu/amd: move wait_on_sem() out of spinlock
by Zhang Yuwei 08 Jun '26

08 Jun '26

From: Ankit Soni <Ankit.Soni(a)amd.com> mainline inclusion from mainline-v7.0-rc1 commit d2a0cac10597068567d336e85fa3cbdbe8ca62bf category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/14797 CVE: CVE-2026-43253 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- With iommu.strict=1, the existing completion wait path can cause soft lockups under stressed environment, as wait_on_sem() busy-waits under the spinlock with interrupts disabled. Move the completion wait in iommu_completion_wait() out of the spinlock. wait_on_sem() only polls the hardware-updated cmd_sem and does not require iommu->lock, so holding the lock during the busy wait unnecessarily increases contention and extends the time with interrupts disabled. Signed-off-by: Ankit Soni <Ankit.Soni(a)amd.com> Reviewed-by: Vasant Hegde <vasant.hegde(a)amd.com> Signed-off-by: Joerg Roedel <joerg.roedel(a)amd.com> Conflicts: drivers/iommu/amd/iommu.c [commit 1ce018df87640 not merged] Signed-off-by: Zhang Yuwei <zhangyuwei20(a)huawei.com> --- drivers/iommu/amd/iommu.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-) diff --git a/drivers/iommu/amd/iommu.c b/drivers/iommu/amd/iommu.c index ae63eb3fe3d0..c141f98465b7 100644 --- a/drivers/iommu/amd/iommu.c +++ b/drivers/iommu/amd/iommu.c @@ -882,7 +882,12 @@ static int wait_on_sem(struct amd_iommu *iommu, u64 data) { int i = 0; - while (*iommu->cmd_sem != data && i < LOOP_TIMEOUT) { + /* + * cmd_sem holds a monotonically non-decreasing completion sequence + * number. + */ + while ((__s64)(READ_ONCE(*iommu->cmd_sem) - data) < 0 && + i < LOOP_TIMEOUT) { udelay(1); i += 1; } @@ -1143,14 +1148,13 @@ static int iommu_completion_wait(struct amd_iommu *iommu) build_completion_wait(&cmd, iommu, data); ret = __iommu_queue_command_sync(iommu, &cmd, false); + raw_spin_unlock_irqrestore(&iommu->lock, flags); + if (ret) - goto out_unlock; + return ret; ret = wait_on_sem(iommu, data); -out_unlock: - raw_spin_unlock_irqrestore(&iommu->lock, flags); - return ret; } -- 2.22.0

2 1

[PATCH OLK-6.6 0/2] irqchip/gic-v3: Fix ALLINT masking logic by decoupling from GIC NMI support
by Jinjie Ruan 08 Jun '26

08 Jun '26

Fix ALLINT masking logic by decoupling from GIC NMI support. Jinjie Ruan (2): Revert "irqchip/gic-v3: Fix ALLINT masking logic by decoupling from GIC NMI support" irqchip/gic-v3: Fix ALLINT masking logic by decoupling from GIC NMI support arch/arm64/include/asm/nmi.h | 5 +++++ drivers/irqchip/irq-gic-v3.c | 14 ++++++++------ 2 files changed, 13 insertions(+), 6 deletions(-) -- 2.34.1

2 3

[PATCH OLK-6.6 0/2] irqchip/gic-v3: Fix ALLINT masking logic by decoupling from GIC NMI support
by Jinjie Ruan 08 Jun '26

08 Jun '26

Fix ALLINT masking logic by decoupling from GIC NMI support. Jinjie Ruan (2): Revert "irqchip/gic-v3: Fix ALLINT masking logic by decoupling from GIC NMI support" irqchip/gic-v3: Fix ALLINT masking logic by decoupling from GIC NMI support arch/arm64/include/asm/nmi.h | 5 +++++ drivers/irqchip/irq-gic-v3.c | 14 ++++++++------ 2 files changed, 13 insertions(+), 6 deletions(-) -- 2.34.1

2 3

[PATCH OLK-5.10] irqchip/gic-v3: Fix ALLINT masking logic by decoupling from GIC NMI support
by Jinjie Ruan 08 Jun '26

08 Jun '26

hulk inclusion category: bugfix bugzilla: https://atomgit.com/openeuler/kernel/issues/9343 -------------------------------- The ARM64 FEAT_NMI implementation relies on two independent hardware validation paths: 1. CPU Feature Check (`system_uses_nmi()`): Validates if the PE supports FEAT_NMI via ID_AA64PFR1_EL1.NMI. If pseudo-NMI is disabled and this bit is set, enables SCTLR_EL1.NMI (map to SCTLR_EL2.NMI in VHE mode) and clears SCTLR_EL1.SPINTMASK. Once active, the hardware automatically sets PSTATE.ALLINT upon taking an exception to EL1/EL2, effectively masking interrupts. `sysreg_clear_set(sctlr_el1, SCTLR_EL1_SPINTMASK, SCTLR_EL1_NMI)` 2. GICv3 Feature Check (gic_data.has_nmi): Validates if the interrupt controller supports Non-maskable Interrupts via GICD_TYPER.NMI. `gic_data.has_nmi = !!(typer & GICD_TYPER_NMI)` And PSTATE.ALLINT is strictly a PE-level state register. Its existence and behavior are entirely independent of whether the underlying GICv3 hardware implementation has its own FEAT_NMI capabilities enabled. Currently, __gic_handle_irq_from_irqson() uses has_v3_3_nmi() to guard the clearing of PSTATE.ALLINT. This creates a destructive dependency where the GIC's NMI status dictates PE register management. static inline bool has_v3_3_nmi(void) { return gic_data.has_nmi && system_uses_nmi(); } If a platform features a PE that supports FEAT_NMI but has a GICv3 that does not support (or has disabled) the NMI feature, the current implementation fails to clear ALLINT during IRQ handling. Because ALLINT is automatically set by the hardware exception entry, it remains permanently set. As a consequence, local_irq_enable() only clears PSTATE.I and F bits, leaving ALLINT untouched. Since ALLINT masks both standard IRQs and NMIs, even if PSTATE.I cleared, the system becomes completely incapable of responding to subsequent NMIs and standard IRQs during downstream processing (such as softirq execution). handle_softirqs() -> local_irq_enable() -> asm volatile("msr daifclr, #3") -> do pending softirq action -> local_irq_disable() -> asm volatile("msr daifset, #3") Fix this by replacing the has_v3_3_nmi() check with system_uses_nmi() when clearing ALLINT. This ensures that PE state management correctly reflects PE capabilities. Additionally, explicitly log whether the GIC supports hardware NMI based on the GICD_TYPER.NMI bit during initialization. Since FEAT_NMI deployment relies on both subsystems, printing this status provides critical diagnostic information for system verification. Fixes: eefea6156921 ("irqchip/gic-v3: Fix hard LOCKUP caused by NMI being masked") Signed-off-by: Jinjie Ruan <ruanjinjie(a)huawei.com> --- arch/arm64/include/asm/daifflags.h | 5 +++++ drivers/irqchip/irq-gic-v3.c | 8 +++++--- 2 files changed, 10 insertions(+), 3 deletions(-) diff --git a/arch/arm64/include/asm/daifflags.h b/arch/arm64/include/asm/daifflags.h index e4aec9469dcf..f04c5ccbd176 100644 --- a/arch/arm64/include/asm/daifflags.h +++ b/arch/arm64/include/asm/daifflags.h @@ -17,6 +17,7 @@ #define DAIF_ERRCTX (PSR_I_BIT | PSR_A_BIT) #define DAIF_MASK (PSR_D_BIT | PSR_A_BIT | PSR_I_BIT | PSR_F_BIT) +#ifdef CONFIG_ARM64_NMI static __always_inline void _allint_clear(void) { asm volatile(__msr_s(SYS_ALLINT_CLR, "xzr")); @@ -26,6 +27,10 @@ static __always_inline void _allint_set(void) { asm volatile(__msr_s(SYS_ALLINT_SET, "xzr")); } +#else +static __always_inline void _allint_clear(void) { } +static __always_inline void _allint_set(void) { } +#endif /* mask/save/unmask/restore all exceptions, including interrupts. */ static inline void local_daif_mask(void) diff --git a/drivers/irqchip/irq-gic-v3.c b/drivers/irqchip/irq-gic-v3.c index 9870ad4cc766..38a429c29fdf 100644 --- a/drivers/irqchip/irq-gic-v3.c +++ b/drivers/irqchip/irq-gic-v3.c @@ -821,11 +821,12 @@ static void __gic_handle_irq_from_irqson(struct pt_regs *regs) if (gic_prio_masking_enabled()) { gic_pmr_mask_irqs(); gic_arch_enable_irqs(); - } else if (has_v3_3_nmi()) { -#ifdef CONFIG_ARM64_NMI + } + +#ifdef CONFIG_ARM64 + if (system_uses_nmi()) _allint_clear(); #endif - } if (!is_nmi) __gic_handle_irq(irqnr, regs); @@ -2273,6 +2274,7 @@ static int __init gic_init_bases(void __iomem *dist_base, gic_data.has_nmi = !!(typer & GICD_TYPER_NMI); pr_info("Distributor has %sRange Selector support\n", gic_data.has_rss ? "" : "no "); + pr_info("GICD_TYPER NMI is%s supported.\n", gic_data.has_nmi ? "" : " not"); if (typer & GICD_TYPER_MBIS) { err = mbi_init(handle, gic_data.domain); -- 2.34.1

2 1

[PATCH OLK-6.6 0/2] arm64: Fix ALLINT masking logic by decoupling from GIC NMI support
by Jinjie Ruan 08 Jun '26

08 Jun '26

Fix ALLINT masking logic by decoupling from GIC NMI support Jinjie Ruan (2): Revert "irqchip/gic-v3: Fix ALLINT masking logic by decoupling from GIC NMI support" irqchip/gic-v3: Fix ALLINT masking logic by decoupling from GIC NMI support drivers/irqchip/irq-gic-v3.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) -- 2.34.1

2 3

[PATCH OLK-5.10] iommu/vt-d: Flush dev-IOTLB only when PCIe device is accessible in scalable mode
by Zhang Yuwei 08 Jun '26

08 Jun '26

From: Jinhui Guo <guojinhui.liam(a)bytedance.com> stable inclusion from stable-v5.10.252 commit 581ce094d9eafb78ec4f9de77bd24b780c151236 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/14677 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 10e60d87813989e20eac1f3eda30b3bae461e7f9 ] Commit 4fc82cd907ac ("iommu/vt-d: Don't issue ATS Invalidation request when device is disconnected") relies on pci_dev_is_disconnected() to skip ATS invalidation for safely-removed devices, but it does not cover link-down caused by faults, which can still hard-lock the system. For example, if a VM fails to connect to the PCIe device, "virsh destroy" is executed to release resources and isolate the fault, but a hard-lockup occurs while releasing the group fd. Call Trace: qi_submit_sync qi_flush_dev_iotlb intel_pasid_tear_down_entry device_block_translation blocking_domain_attach_dev __iommu_attach_device __iommu_device_set_domain __iommu_group_set_domain_internal iommu_detach_group vfio_iommu_type1_detach_group vfio_group_detach_container vfio_group_fops_release __fput Although pci_device_is_present() is slower than pci_dev_is_disconnected(), it still takes only ~70 µs on a ConnectX-5 (8 GT/s, x2) and becomes even faster as PCIe speed and width increase. Besides, devtlb_invalidation_with_pasid() is called only in the paths below, which are far less frequent than memory map/unmap. 1. mm-struct release 2. {attach,release}_dev 3. set/remove PASID 4. dirty-tracking setup The gain in system stability far outweighs the negligible cost of using pci_device_is_present() instead of pci_dev_is_disconnected() to decide when to skip ATS invalidation, especially under GDR high-load conditions. Fixes: 4fc82cd907ac ("iommu/vt-d: Don't issue ATS Invalidation request when device is disconnected") Cc: stable(a)vger.kernel.org Signed-off-by: Jinhui Guo <guojinhui.liam(a)bytedance.com> Link: https://lore.kernel.org/r/20251211035946.2071-3-guojinhui.liam@bytedance.com Signed-off-by: Lu Baolu <baolu.lu(a)linux.intel.com> Signed-off-by: Joerg Roedel <joerg.roedel(a)amd.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Lin Yujun <linyujun809(a)h-partners.com> --- drivers/iommu/intel/pasid.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iommu/intel/pasid.c b/drivers/iommu/intel/pasid.c index cc913f00f296..33c85d9e3a45 100644 --- a/drivers/iommu/intel/pasid.c +++ b/drivers/iommu/intel/pasid.c @@ -513,7 +513,7 @@ devtlb_invalidation_with_pasid(struct intel_iommu *iommu, if (!info || !info->ats_enabled) return; - if (pci_dev_is_disconnected(to_pci_dev(dev))) + if (!pci_device_is_present(to_pci_dev(dev))) return; sid = info->bus << 8 | info->devfn; -- 2.22.0

2 1

[PATCH OLK-5.10] netfilter: flowtable: strictly check for maximum number of actions
by Zhang Changzhong 06 Jun '26

06 Jun '26

From: Pablo Neira Ayuso <pablo(a)netfilter.org> mainline inclusion from mainline-v7.0-rc7 commit 76522fcdbc3a02b568f5d957f7e66fc194abb893 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/14887 CVE: CVE-2026-43329 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- The maximum number of flowtable hardware offload actions in IPv6 is: * ethernet mangling (4 payload actions, 2 for each ethernet address) * SNAT (4 payload actions) * DNAT (4 payload actions) * Double VLAN (4 vlan actions, 2 for popping vlan, and 2 for pushing) for QinQ. * Redirect (1 action) Which makes 17, while the maximum is 16. But act_ct supports for tunnels actions too. Note that payload action operates at 32-bit word level, so mangling an IPv6 address takes 4 payload actions. Update flow_action_entry_next() calls to check for the maximum number of supported actions. While at it, rise the maximum number of actions per flow from 16 to 24 so this works fine with IPv6 setups. Fixes: c29f74e0df7a ("netfilter: nf_flow_table: hardware offload support") Reported-by: Hyunwoo Kim <imv4bel(a)gmail.com> Signed-off-by: Pablo Neira Ayuso <pablo(a)netfilter.org> Conflicts: net/netfilter/nf_flow_table_offload.c [context conflict because commits 2ed37183abb7 and eeff3000f240 were not merged] Signed-off-by: Zhang Changzhong <zhangchangzhong(a)huawei.com> --- net/netfilter/nf_flow_table_offload.c | 173 +++++++++++++++++++++++----------- 1 file changed, 116 insertions(+), 57 deletions(-) diff --git a/net/netfilter/nf_flow_table_offload.c b/net/netfilter/nf_flow_table_offload.c index f6275d9..dc53bf4 100644 --- a/net/netfilter/nf_flow_table_offload.c +++ b/net/netfilter/nf_flow_table_offload.c @@ -13,6 +13,8 @@ #include <net/netfilter/nf_conntrack_core.h> #include <net/netfilter/nf_conntrack_tuple.h> +#define NF_FLOW_RULE_ACTION_MAX 24 + static struct workqueue_struct *nf_flow_offload_wq; struct flow_offload_work { @@ -165,7 +167,12 @@ static void flow_offload_mangle(struct flow_action_entry *entry, static inline struct flow_action_entry * flow_action_entry_next(struct nf_flow_rule *flow_rule) { - int i = flow_rule->rule->action.num_entries++; + int i; + + if (unlikely(flow_rule->rule->action.num_entries >= NF_FLOW_RULE_ACTION_MAX)) + return NULL; + + i = flow_rule->rule->action.num_entries++; return &flow_rule->rule->action.entries[i]; } @@ -182,6 +189,9 @@ static int flow_offload_eth_src(struct net *net, u32 mask, val; u16 val16; + if (!entry0 || !entry1) + return -E2BIG; + dev = dev_get_by_index(net, tuple->iifidx); if (!dev) return -ENOENT; @@ -216,6 +226,9 @@ static int flow_offload_eth_dst(struct net *net, u8 nud_state; u16 val16; + if (!entry0 || !entry1) + return -E2BIG; + dst_cache = flow->tuplehash[dir].tuple.dst_cache; n = dst_neigh_lookup(dst_cache, daddr); if (!n) @@ -246,16 +259,19 @@ static int flow_offload_eth_dst(struct net *net, return 0; } -static void flow_offload_ipv4_snat(struct net *net, - const struct flow_offload *flow, - enum flow_offload_tuple_dir dir, - struct nf_flow_rule *flow_rule) +static int flow_offload_ipv4_snat(struct net *net, + const struct flow_offload *flow, + enum flow_offload_tuple_dir dir, + struct nf_flow_rule *flow_rule) { struct flow_action_entry *entry = flow_action_entry_next(flow_rule); u32 mask = ~htonl(0xffffffff); __be32 addr; u32 offset; + if (!entry) + return -E2BIG; + switch (dir) { case FLOW_OFFLOAD_DIR_ORIGINAL: addr = flow->tuplehash[FLOW_OFFLOAD_DIR_REPLY].tuple.dst_v4.s_addr; @@ -266,23 +282,27 @@ static void flow_offload_ipv4_snat(struct net *net, offset = offsetof(struct iphdr, daddr); break; default: - return; + return -EOPNOTSUPP; } flow_offload_mangle(entry, FLOW_ACT_MANGLE_HDR_TYPE_IP4, offset, &addr, &mask); + return 0; } -static void flow_offload_ipv4_dnat(struct net *net, - const struct flow_offload *flow, - enum flow_offload_tuple_dir dir, - struct nf_flow_rule *flow_rule) +static int flow_offload_ipv4_dnat(struct net *net, + const struct flow_offload *flow, + enum flow_offload_tuple_dir dir, + struct nf_flow_rule *flow_rule) { struct flow_action_entry *entry = flow_action_entry_next(flow_rule); u32 mask = ~htonl(0xffffffff); __be32 addr; u32 offset; + if (!entry) + return -E2BIG; + switch (dir) { case FLOW_OFFLOAD_DIR_ORIGINAL: addr = flow->tuplehash[FLOW_OFFLOAD_DIR_REPLY].tuple.src_v4.s_addr; @@ -293,14 +313,15 @@ static void flow_offload_ipv4_dnat(struct net *net, offset = offsetof(struct iphdr, saddr); break; default: - return; + return -EOPNOTSUPP; } flow_offload_mangle(entry, FLOW_ACT_MANGLE_HDR_TYPE_IP4, offset, &addr, &mask); + return 0; } -static void flow_offload_ipv6_mangle(struct nf_flow_rule *flow_rule, +static int flow_offload_ipv6_mangle(struct nf_flow_rule *flow_rule, unsigned int offset, const __be32 *addr, const __be32 *mask) { @@ -309,15 +330,20 @@ static void flow_offload_ipv6_mangle(struct nf_flow_rule *flow_rule, for (i = 0; i < sizeof(struct in6_addr) / sizeof(u32); i++) { entry = flow_action_entry_next(flow_rule); + if (!entry) + return -E2BIG; + flow_offload_mangle(entry, FLOW_ACT_MANGLE_HDR_TYPE_IP6, offset + i * sizeof(u32), &addr[i], mask); } + + return 0; } -static void flow_offload_ipv6_snat(struct net *net, - const struct flow_offload *flow, - enum flow_offload_tuple_dir dir, - struct nf_flow_rule *flow_rule) +static int flow_offload_ipv6_snat(struct net *net, + const struct flow_offload *flow, + enum flow_offload_tuple_dir dir, + struct nf_flow_rule *flow_rule) { u32 mask = ~htonl(0xffffffff); const __be32 *addr; @@ -333,16 +359,16 @@ static void flow_offload_ipv6_snat(struct net *net, offset = offsetof(struct ipv6hdr, daddr); break; default: - return; + return -EOPNOTSUPP; } - flow_offload_ipv6_mangle(flow_rule, offset, addr, &mask); + return flow_offload_ipv6_mangle(flow_rule, offset, addr, &mask); } -static void flow_offload_ipv6_dnat(struct net *net, - const struct flow_offload *flow, - enum flow_offload_tuple_dir dir, - struct nf_flow_rule *flow_rule) +static int flow_offload_ipv6_dnat(struct net *net, + const struct flow_offload *flow, + enum flow_offload_tuple_dir dir, + struct nf_flow_rule *flow_rule) { u32 mask = ~htonl(0xffffffff); const __be32 *addr; @@ -358,10 +384,10 @@ static void flow_offload_ipv6_dnat(struct net *net, offset = offsetof(struct ipv6hdr, saddr); break; default: - return; + return -EOPNOTSUPP; } - flow_offload_ipv6_mangle(flow_rule, offset, addr, &mask); + return flow_offload_ipv6_mangle(flow_rule, offset, addr, &mask); } static int flow_offload_l4proto(const struct flow_offload *flow) @@ -383,15 +409,18 @@ static int flow_offload_l4proto(const struct flow_offload *flow) return type; } -static void flow_offload_port_snat(struct net *net, - const struct flow_offload *flow, - enum flow_offload_tuple_dir dir, - struct nf_flow_rule *flow_rule) +static int flow_offload_port_snat(struct net *net, + const struct flow_offload *flow, + enum flow_offload_tuple_dir dir, + struct nf_flow_rule *flow_rule) { struct flow_action_entry *entry = flow_action_entry_next(flow_rule); u32 mask, port; u32 offset; + if (!entry) + return -E2BIG; + switch (dir) { case FLOW_OFFLOAD_DIR_ORIGINAL: port = ntohs(flow->tuplehash[FLOW_OFFLOAD_DIR_REPLY].tuple.dst_port); @@ -406,22 +435,26 @@ static void flow_offload_port_snat(struct net *net, mask = ~htonl(0xffff); break; default: - return; + return -EOPNOTSUPP; } flow_offload_mangle(entry, flow_offload_l4proto(flow), offset, &port, &mask); + return 0; } -static void flow_offload_port_dnat(struct net *net, - const struct flow_offload *flow, - enum flow_offload_tuple_dir dir, - struct nf_flow_rule *flow_rule) +static int flow_offload_port_dnat(struct net *net, + const struct flow_offload *flow, + enum flow_offload_tuple_dir dir, + struct nf_flow_rule *flow_rule) { struct flow_action_entry *entry = flow_action_entry_next(flow_rule); u32 mask, port; u32 offset; + if (!entry) + return -E2BIG; + switch (dir) { case FLOW_OFFLOAD_DIR_ORIGINAL: port = ntohs(flow->tuplehash[FLOW_OFFLOAD_DIR_REPLY].tuple.src_port); @@ -436,20 +469,24 @@ static void flow_offload_port_dnat(struct net *net, mask = ~htonl(0xffff0000); break; default: - return; + return -EOPNOTSUPP; } flow_offload_mangle(entry, flow_offload_l4proto(flow), offset, &port, &mask); + return 0; } -static void flow_offload_ipv4_checksum(struct net *net, - const struct flow_offload *flow, - struct nf_flow_rule *flow_rule) +static int flow_offload_ipv4_checksum(struct net *net, + const struct flow_offload *flow, + struct nf_flow_rule *flow_rule) { u8 protonum = flow->tuplehash[FLOW_OFFLOAD_DIR_ORIGINAL].tuple.l4proto; struct flow_action_entry *entry = flow_action_entry_next(flow_rule); + if (!entry) + return -E2BIG; + entry->id = FLOW_ACTION_CSUM; entry->csum_flags = TCA_CSUM_UPDATE_FLAG_IPV4HDR; @@ -461,22 +498,29 @@ static void flow_offload_ipv4_checksum(struct net *net, entry->csum_flags |= TCA_CSUM_UPDATE_FLAG_UDP; break; } + + return 0; } -static void flow_offload_redirect(const struct flow_offload *flow, +static int flow_offload_redirect(const struct flow_offload *flow, enum flow_offload_tuple_dir dir, struct nf_flow_rule *flow_rule) { struct flow_action_entry *entry = flow_action_entry_next(flow_rule); struct rtable *rt; + if (!entry) + return -E2BIG; + rt = (struct rtable *)flow->tuplehash[dir].tuple.dst_cache; entry->id = FLOW_ACTION_REDIRECT; entry->dev = rt->dst.dev; dev_hold(rt->dst.dev); + + return 0; } -static void flow_offload_encap_tunnel(const struct flow_offload *flow, +static int flow_offload_encap_tunnel(const struct flow_offload *flow, enum flow_offload_tuple_dir dir, struct nf_flow_rule *flow_rule) { @@ -490,13 +534,17 @@ static void flow_offload_encap_tunnel(const struct flow_offload *flow, tun_info = lwt_tun_info(dst->lwtstate); if (tun_info && (tun_info->mode & IP_TUNNEL_INFO_TX)) { entry = flow_action_entry_next(flow_rule); + if (!entry) + return -E2BIG; entry->id = FLOW_ACTION_TUNNEL_ENCAP; entry->tunnel = tun_info; } } + + return 0; } -static void flow_offload_decap_tunnel(const struct flow_offload *flow, +static int flow_offload_decap_tunnel(const struct flow_offload *flow, enum flow_offload_tuple_dir dir, struct nf_flow_rule *flow_rule) { @@ -510,35 +558,44 @@ static void flow_offload_decap_tunnel(const struct flow_offload *flow, tun_info = lwt_tun_info(dst->lwtstate); if (tun_info && (tun_info->mode & IP_TUNNEL_INFO_TX)) { entry = flow_action_entry_next(flow_rule); + if (!entry) + return -E2BIG; entry->id = FLOW_ACTION_TUNNEL_DECAP; } } + + return 0; } int nf_flow_rule_route_ipv4(struct net *net, const struct flow_offload *flow, enum flow_offload_tuple_dir dir, struct nf_flow_rule *flow_rule) { - flow_offload_decap_tunnel(flow, dir, flow_rule); - flow_offload_encap_tunnel(flow, dir, flow_rule); + if (flow_offload_decap_tunnel(flow, dir, flow_rule) < 0 || + flow_offload_encap_tunnel(flow, dir, flow_rule) < 0) + return -1; if (flow_offload_eth_src(net, flow, dir, flow_rule) < 0 || flow_offload_eth_dst(net, flow, dir, flow_rule) < 0) return -1; if (test_bit(NF_FLOW_SNAT, &flow->flags)) { - flow_offload_ipv4_snat(net, flow, dir, flow_rule); - flow_offload_port_snat(net, flow, dir, flow_rule); + if (flow_offload_ipv4_snat(net, flow, dir, flow_rule) < 0 || + flow_offload_port_snat(net, flow, dir, flow_rule) < 0) + return -1; } if (test_bit(NF_FLOW_DNAT, &flow->flags)) { - flow_offload_ipv4_dnat(net, flow, dir, flow_rule); - flow_offload_port_dnat(net, flow, dir, flow_rule); + if (flow_offload_ipv4_dnat(net, flow, dir, flow_rule) < 0 || + flow_offload_port_dnat(net, flow, dir, flow_rule) < 0) + return -1; } if (test_bit(NF_FLOW_SNAT, &flow->flags) || test_bit(NF_FLOW_DNAT, &flow->flags)) - flow_offload_ipv4_checksum(net, flow, flow_rule); + if (flow_offload_ipv4_checksum(net, flow, flow_rule) < 0) + return -1; - flow_offload_redirect(flow, dir, flow_rule); + if (flow_offload_redirect(flow, dir, flow_rule) < 0) + return -1; return 0; } @@ -548,30 +605,32 @@ int nf_flow_rule_route_ipv6(struct net *net, const struct flow_offload *flow, enum flow_offload_tuple_dir dir, struct nf_flow_rule *flow_rule) { - flow_offload_decap_tunnel(flow, dir, flow_rule); - flow_offload_encap_tunnel(flow, dir, flow_rule); + if (flow_offload_decap_tunnel(flow, dir, flow_rule) < 0 || + flow_offload_encap_tunnel(flow, dir, flow_rule) < 0) + return -1; if (flow_offload_eth_src(net, flow, dir, flow_rule) < 0 || flow_offload_eth_dst(net, flow, dir, flow_rule) < 0) return -1; if (test_bit(NF_FLOW_SNAT, &flow->flags)) { - flow_offload_ipv6_snat(net, flow, dir, flow_rule); - flow_offload_port_snat(net, flow, dir, flow_rule); + if (flow_offload_ipv6_snat(net, flow, dir, flow_rule) < 0 || + flow_offload_port_snat(net, flow, dir, flow_rule) < 0) + return -1; } if (test_bit(NF_FLOW_DNAT, &flow->flags)) { - flow_offload_ipv6_dnat(net, flow, dir, flow_rule); - flow_offload_port_dnat(net, flow, dir, flow_rule); + if (flow_offload_ipv6_dnat(net, flow, dir, flow_rule) < 0 || + flow_offload_port_dnat(net, flow, dir, flow_rule) < 0) + return -1; } - flow_offload_redirect(flow, dir, flow_rule); + if (flow_offload_redirect(flow, dir, flow_rule) < 0) + return -1; return 0; } EXPORT_SYMBOL_GPL(nf_flow_rule_route_ipv6); -#define NF_FLOW_RULE_ACTION_MAX 16 - static struct nf_flow_rule * nf_flow_offload_rule_alloc(struct net *net, const struct flow_offload_work *offload, -- 2.9.5

2 1

[PATCH OLK-5.10] iommu/vt-d: Flush dev-IOTLB only when PCIe device is accessible in scalable mode
by Zhang Yuwei 06 Jun '26

06 Jun '26

From: Jinhui Guo <guojinhui.liam(a)bytedance.com> stable inclusion from stable-v5.10.252 commit 581ce094d9eafb78ec4f9de77bd24b780c151236 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/15190 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 10e60d87813989e20eac1f3eda30b3bae461e7f9 ] Commit 4fc82cd907ac ("iommu/vt-d: Don't issue ATS Invalidation request when device is disconnected") relies on pci_dev_is_disconnected() to skip ATS invalidation for safely-removed devices, but it does not cover link-down caused by faults, which can still hard-lock the system. For example, if a VM fails to connect to the PCIe device, "virsh destroy" is executed to release resources and isolate the fault, but a hard-lockup occurs while releasing the group fd. Call Trace: qi_submit_sync qi_flush_dev_iotlb intel_pasid_tear_down_entry device_block_translation blocking_domain_attach_dev __iommu_attach_device __iommu_device_set_domain __iommu_group_set_domain_internal iommu_detach_group vfio_iommu_type1_detach_group vfio_group_detach_container vfio_group_fops_release __fput Although pci_device_is_present() is slower than pci_dev_is_disconnected(), it still takes only ~70 µs on a ConnectX-5 (8 GT/s, x2) and becomes even faster as PCIe speed and width increase. Besides, devtlb_invalidation_with_pasid() is called only in the paths below, which are far less frequent than memory map/unmap. 1. mm-struct release 2. {attach,release}_dev 3. set/remove PASID 4. dirty-tracking setup The gain in system stability far outweighs the negligible cost of using pci_device_is_present() instead of pci_dev_is_disconnected() to decide when to skip ATS invalidation, especially under GDR high-load conditions. Fixes: 4fc82cd907ac ("iommu/vt-d: Don't issue ATS Invalidation request when device is disconnected") Cc: stable(a)vger.kernel.org Signed-off-by: Jinhui Guo <guojinhui.liam(a)bytedance.com> Link: https://lore.kernel.org/r/20251211035946.2071-3-guojinhui.liam@bytedance.com Signed-off-by: Lu Baolu <baolu.lu(a)linux.intel.com> Signed-off-by: Joerg Roedel <joerg.roedel(a)amd.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Lin Yujun <linyujun809(a)h-partners.com> --- drivers/iommu/intel/pasid.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/iommu/intel/pasid.c b/drivers/iommu/intel/pasid.c index cc913f00f296..33c85d9e3a45 100644 --- a/drivers/iommu/intel/pasid.c +++ b/drivers/iommu/intel/pasid.c @@ -513,7 +513,7 @@ devtlb_invalidation_with_pasid(struct intel_iommu *iommu, if (!info || !info->ats_enabled) return; - if (pci_dev_is_disconnected(to_pci_dev(dev))) + if (!pci_device_is_present(to_pci_dev(dev))) return; sid = info->bus << 8 | info->devfn; -- 2.22.0

2 1

[PATCH OLK-5.10] iommu/vt-d: Clear Present bit before tearing down context entry
by Zhang Yuwei 06 Jun '26

06 Jun '26

From: Lu Baolu <baolu.lu(a)linux.intel.com> mainline inclusion from mainline-v7.0-rc1 commit c1e4f1dccbe9d7656d1c6872ebeadb5992d0aaa2 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/15190 CVE: CVE-2026-45944 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- When tearing down a context entry, the current implementation zeros the entire 128-bit entry using multiple 64-bit writes. This creates a window where the hardware can fetch a "torn" entry — where some fields are already zeroed while the 'Present' bit is still set — leading to unpredictable behavior or spurious faults. While x86 provides strong write ordering, the compiler may reorder writes to the two 64-bit halves of the context entry. Even without compiler reordering, the hardware fetch is not guaranteed to be atomic with respect to multiple CPU writes. Align with the "Guidance to Software for Invalidations" in the VT-d spec (Section 6.5.3.3) by implementing the recommended ownership handshake: 1. Clear only the 'Present' (P) bit of the context entry first to signal the transition of ownership from hardware to software. 2. Use dma_wmb() to ensure the cleared bit is visible to the IOMMU. 3. Perform the required cache and context-cache invalidation to ensure hardware no longer has cached references to the entry. 4. Fully zero out the entry only after the invalidation is complete. Also, add a dma_wmb() to context_set_present() to ensure the entry is fully initialized before the 'Present' bit becomes visible. Fixes: ba39592764ed ("Intel IOMMU: Intel IOMMU driver") Reported-by: Dmytro Maluka <dmaluka(a)chromium.org> Closes: https://lore.kernel.org/all/aTG7gc7I5wExai3S@google.com/ Signed-off-by: Lu Baolu <baolu.lu(a)linux.intel.com> Reviewed-by: Dmytro Maluka <dmaluka(a)chromium.org> Reviewed-by: Samiullah Khawaja <skhawaja(a)google.com> Reviewed-by: Kevin Tian <kevin.tian(a)intel.com> Link: https://lore.kernel.org/r/20260120061816.2132558-3-baolu.lu@linux.intel.com Signed-off-by: Joerg Roedel <joerg.roedel(a)amd.com> Conflicts: drivers/iommu/intel/pasid.c drivers/iommu/intel/iommu.c drivers/iommu/intel/iommu.h [contxt conflict] Signed-off-by: Zhang Yuwei <zhangyuwei20(a)huawei.com> --- drivers/iommu/intel/iommu.c | 25 +++++++++++++++++++++++-- 1 file changed, 23 insertions(+), 2 deletions(-) diff --git a/drivers/iommu/intel/iommu.c b/drivers/iommu/intel/iommu.c index a68d1bfc7444..245fef1026c3 100644 --- a/drivers/iommu/intel/iommu.c +++ b/drivers/iommu/intel/iommu.c @@ -210,7 +210,26 @@ static phys_addr_t root_entry_uctp(struct root_entry *re) static inline void context_set_present(struct context_entry *context) { - context->lo |= 1; + u64 val; + + dma_wmb(); + val = READ_ONCE(context->lo) | 1; + WRITE_ONCE(context->lo, val); +} + +/* + * Clear the Present (P) bit (bit 0) of a context table entry. This initiates + * the transition of the entry's ownership from hardware to software. The + * caller is responsible for fulfilling the invalidation handshake recommended + * by the VT-d spec, Section 6.5.3.3 (Guidance to Software for Invalidations). + */ +static inline void context_clear_present(struct context_entry *context) +{ + u64 val; + + val = READ_ONCE(context->lo) & GENMASK_ULL(63, 1); + WRITE_ONCE(context->lo, val); + dma_wmb(); } static inline void context_set_fault_enable(struct context_entry *context) @@ -2617,7 +2636,7 @@ static void domain_context_clear_one(struct device_domain_info *info, u8 bus, u8 did_old = context_domain_id(context); } - context_clear_entry(context); + context_clear_present(context); __iommu_flush_cache(iommu, context, sizeof(*context)); spin_unlock_irqrestore(&iommu->lock, flags); iommu->flush.flush_context(iommu, @@ -2636,6 +2655,8 @@ static void domain_context_clear_one(struct device_domain_info *info, u8 bus, u8 DMA_TLB_DSI_FLUSH); __iommu_flush_dev_iotlb(info, 0, MAX_AGAW_PFN_WIDTH); + context_clear_entry(context); + __iommu_flush_cache(iommu, context, sizeof(*context)); } static inline void unlink_domain_info(struct device_domain_info *info) -- 2.22.0

2 1

[PATCH OLK-5.10] iommu/vt-d: Clear Present bit before tearing down PASID entry
by Zhang Yuwei 06 Jun '26

06 Jun '26

From: Lu Baolu <baolu.lu(a)linux.intel.com> mainline inclusion from mainline-v7.0-rc1 commit 75ed00055c059dedc47b5daaaa2f8a7a019138ff category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/15137 CVE: CVE-2026-45894 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- The Intel VT-d Scalable Mode PASID table entry consists of 512 bits (64 bytes). When tearing down an entry, the current implementation zeros the entire 64-byte structure immediately using multiple 64-bit writes. Since the IOMMU hardware may fetch these 64 bytes using multiple internal transactions (e.g., four 128-bit bursts), updating or zeroing the entire entry while it is active (P=1) risks a "torn" read. If a hardware fetch occurs simultaneously with the CPU zeroing the entry, the hardware could observe an inconsistent state, leading to unpredictable behavior or spurious faults. Follow the "Guidance to Software for Invalidations" in the VT-d spec (Section 6.5.3.3) by implementing the recommended ownership handshake: 1. Clear only the 'Present' (P) bit of the PASID entry. 2. Use a dma_wmb() to ensure the cleared bit is visible to hardware before proceeding. 3. Execute the required invalidation sequence (PASID cache, IOTLB, and Device-TLB flush) to ensure the hardware has released all cached references. 4. Only after the flushes are complete, zero out the remaining fields of the PASID entry. Also, add a dma_wmb() in pasid_set_present() to ensure that all other fields of the PASID entry are visible to the hardware before the Present bit is set. Fixes: 0bbeb01a4faf ("iommu/vt-d: Manage scalalble mode PASID tables") Signed-off-by: Lu Baolu <baolu.lu(a)linux.intel.com> Reviewed-by: Dmytro Maluka <dmaluka(a)chromium.org> Reviewed-by: Samiullah Khawaja <skhawaja(a)google.com> Reviewed-by: Kevin Tian <kevin.tian(a)intel.com> Link: https://lore.kernel.org/r/20260120061816.2132558-2-baolu.lu@linux.intel.com Signed-off-by: Joerg Roedel <joerg.roedel(a)amd.com> Conflicts: drivers/iommu/intel/pasid.c drivers/iommu/intel/pasid.h [commit 1f5e307ca16c0 not merged] Signed-off-by: Zhang Yuwei <zhangyuwei20(a)huawei.com> --- drivers/iommu/intel/pasid.c | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-) diff --git a/drivers/iommu/intel/pasid.c b/drivers/iommu/intel/pasid.c index 9d909e99b7e8..cc913f00f296 100644 --- a/drivers/iommu/intel/pasid.c +++ b/drivers/iommu/intel/pasid.c @@ -411,9 +411,23 @@ static inline void pasid_set_sre(struct pasid_entry *pe) */ static inline void pasid_set_present(struct pasid_entry *pe) { + dma_wmb(); pasid_set_bits(&pe->val[0], 1 << 0, 1); } +/* + * Clear the Present (P) bit (bit 0) of a scalable-mode PASID table entry. + * This initiates the transition of the entry's ownership from hardware + * to software. The caller is responsible for fulfilling the invalidation + * handshake recommended by the VT-d spec, Section 6.5.3.3 (Guidance to + * Software for Invalidations). + */ +static inline void pasid_clear_present(struct pasid_entry *pe) +{ + pasid_set_bits(&pe->val[0], 1 << 0, 0); + dma_wmb(); +} + /* * Setup Page Walk Snoop bit (Bit 87) of a scalable mode PASID * entry. @@ -531,7 +545,7 @@ void intel_pasid_tear_down_entry(struct intel_iommu *iommu, struct device *dev, did = pasid_get_domain_id(pte); pgtt = pasid_pte_get_pgtt(pte); - intel_pasid_clear_entry(dev, pasid, fault_ignore); + pasid_clear_present(pte); if (!ecap_coherent(iommu->ecap)) clflush_cache_range(pte, sizeof(*pte)); @@ -546,6 +560,10 @@ void intel_pasid_tear_down_entry(struct intel_iommu *iommu, struct device *dev, /* Device IOTLB doesn't need to be flushed in caching mode. */ if (!cap_caching_mode(iommu->cap)) devtlb_invalidation_with_pasid(iommu, dev, pasid); + + intel_pasid_clear_entry(dev, pasid, fault_ignore); + if (!ecap_coherent(iommu->ecap)) + clflush_cache_range(pte, sizeof(*pte)); } static void pasid_flush_caches(struct intel_iommu *iommu, -- 2.22.0

2 1

[PATCH OLK-6.6] iommu/vt-d: Clear Present bit before tearing down PASID entry
by Zhang Yuwei 06 Jun '26

06 Jun '26

From: Lu Baolu <baolu.lu(a)linux.intel.com> mainline inclusion from mainline-v7.0-rc1 commit 75ed00055c059dedc47b5daaaa2f8a7a019138ff category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/15137 CVE: CVE-2026-45894 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- The Intel VT-d Scalable Mode PASID table entry consists of 512 bits (64 bytes). When tearing down an entry, the current implementation zeros the entire 64-byte structure immediately using multiple 64-bit writes. Since the IOMMU hardware may fetch these 64 bytes using multiple internal transactions (e.g., four 128-bit bursts), updating or zeroing the entire entry while it is active (P=1) risks a "torn" read. If a hardware fetch occurs simultaneously with the CPU zeroing the entry, the hardware could observe an inconsistent state, leading to unpredictable behavior or spurious faults. Follow the "Guidance to Software for Invalidations" in the VT-d spec (Section 6.5.3.3) by implementing the recommended ownership handshake: 1. Clear only the 'Present' (P) bit of the PASID entry. 2. Use a dma_wmb() to ensure the cleared bit is visible to hardware before proceeding. 3. Execute the required invalidation sequence (PASID cache, IOTLB, and Device-TLB flush) to ensure the hardware has released all cached references. 4. Only after the flushes are complete, zero out the remaining fields of the PASID entry. Also, add a dma_wmb() in pasid_set_present() to ensure that all other fields of the PASID entry are visible to the hardware before the Present bit is set. Fixes: 0bbeb01a4faf ("iommu/vt-d: Manage scalalble mode PASID tables") Signed-off-by: Lu Baolu <baolu.lu(a)linux.intel.com> Reviewed-by: Dmytro Maluka <dmaluka(a)chromium.org> Reviewed-by: Samiullah Khawaja <skhawaja(a)google.com> Reviewed-by: Kevin Tian <kevin.tian(a)intel.com> Link: https://lore.kernel.org/r/20260120061816.2132558-2-baolu.lu@linux.intel.com Signed-off-by: Joerg Roedel <joerg.roedel(a)amd.com> Conflicts: drivers/iommu/intel/pasid.c [commit 1f5e307ca16c0 not merged] Signed-off-by: Zhang Yuwei <zhangyuwei20(a)huawei.com> --- drivers/iommu/intel/pasid.c | 6 +++++- drivers/iommu/intel/pasid.h | 14 ++++++++++++++ 2 files changed, 19 insertions(+), 1 deletion(-) diff --git a/drivers/iommu/intel/pasid.c b/drivers/iommu/intel/pasid.c index 6ef582bfaea5..89a2c7a17489 100644 --- a/drivers/iommu/intel/pasid.c +++ b/drivers/iommu/intel/pasid.c @@ -248,7 +248,7 @@ void intel_pasid_tear_down_entry(struct intel_iommu *iommu, struct device *dev, did = pasid_get_domain_id(pte); pgtt = pasid_pte_get_pgtt(pte); - intel_pasid_clear_entry(dev, pasid, fault_ignore); + pasid_clear_present(pte); spin_unlock(&iommu->lock); if (!ecap_coherent(iommu->ecap)) @@ -264,6 +264,10 @@ void intel_pasid_tear_down_entry(struct intel_iommu *iommu, struct device *dev, /* Device IOTLB doesn't need to be flushed in caching mode. */ if (!cap_caching_mode(iommu->cap)) devtlb_invalidation_with_pasid(iommu, dev, pasid); + + intel_pasid_clear_entry(dev, pasid, fault_ignore); + if (!ecap_coherent(iommu->ecap)) + clflush_cache_range(pte, sizeof(*pte)); } /* diff --git a/drivers/iommu/intel/pasid.h b/drivers/iommu/intel/pasid.h index 42fda97fd851..979fcfea6b2c 100644 --- a/drivers/iommu/intel/pasid.h +++ b/drivers/iommu/intel/pasid.h @@ -235,9 +235,23 @@ static inline void pasid_set_wpe(struct pasid_entry *pe) */ static inline void pasid_set_present(struct pasid_entry *pe) { + dma_wmb(); pasid_set_bits(&pe->val[0], 1 << 0, 1); } +/* + * Clear the Present (P) bit (bit 0) of a scalable-mode PASID table entry. + * This initiates the transition of the entry's ownership from hardware + * to software. The caller is responsible for fulfilling the invalidation + * handshake recommended by the VT-d spec, Section 6.5.3.3 (Guidance to + * Software for Invalidations). + */ +static inline void pasid_clear_present(struct pasid_entry *pe) +{ + pasid_set_bits(&pe->val[0], 1 << 0, 0); + dma_wmb(); +} + /* * Setup Page Walk Snoop bit (Bit 87) of a scalable mode PASID * entry. -- 2.22.0

2 1

[PATCH OLK-5.10] sound: ua101: fix division by zero at probe
by Yin Tirui 06 Jun '26

06 Jun '26

From: SeungJu Cheon <suunj1331(a)gmail.com> mainline inclusion from mainline-v7.1-rc2 commit d1f73f169c1014463b5060e3f60813e13ddc7b87 category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/15431 CVE: CVE-2026-46184 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- Add a missing sanity check for bNrChannels in detect_usb_format() to prevent a division by zero in playback_urb_complete() and capture_urb_complete(). USB core does not validate class-specific descriptor fields such as bNrChannels, so drivers must verify them before use. If a device provides bNrChannels = 0, frame_bytes becomes zero and is later used as a divisor in the URB completion handlers, leading to a kernel crash. Fixes: 63978ab3e3e9 ("sound: add Edirol UA-101 support") Cc: stable(a)vger.kernel.org Signed-off-by: SeungJu Cheon <suunj1331(a)gmail.com> Link: https://patch.msgid.link/20260426111239.103296-1-suunj1331@gmail.com Signed-off-by: Takashi Iwai <tiwai(a)suse.de> Signed-off-by: Yin Tirui <yintirui(a)huawei.com> --- sound/usb/misc/ua101.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/sound/usb/misc/ua101.c b/sound/usb/misc/ua101.c index 5dd1cbe29d12..e9ce9f903cc6 100644 --- a/sound/usb/misc/ua101.c +++ b/sound/usb/misc/ua101.c @@ -995,6 +995,13 @@ static int detect_usb_format(struct ua101 *ua) ua->capture.channels = fmt_capture->bNrChannels; ua->playback.channels = fmt_playback->bNrChannels; + if (!ua->capture.channels || !ua->playback.channels) { + dev_err(&ua->dev->dev, + "invalid channel count: capture %u, playback %u\n", + ua->capture.channels, ua->playback.channels); + return -EINVAL; + } + ua->capture.frame_bytes = fmt_capture->bSubframeSize * ua->capture.channels; ua->playback.frame_bytes = -- 2.43.0

2 1

[PATCH OLK-5.10] iommu/vt-d: Flush cache for PASID table before using it
by Yin Tirui 06 Jun '26

06 Jun '26

From: Dmytro Maluka <dmaluka(a)chromium.org> stable inclusion from stable-v5.10.252 commit cd75e77125c8a51754ca4cd60b4ca083ed735d1d category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/15130 CVE: CVE-2026-45862 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 22d169bdd2849fe6bd18c2643742e1c02be6451c ] When writing the address of a freshly allocated zero-initialized PASID table to a PASID directory entry, do that after the CPU cache flush for this PASID table, not before it, to avoid the time window when this PASID table may be already used by non-coherent IOMMU hardware while its contents in RAM is still some random old data, not zero-initialized. Fixes: 194b3348bdbb ("iommu/vt-d: Fix PASID directory pointer coherency") Signed-off-by: Dmytro Maluka <dmaluka(a)chromium.org> Reviewed-by: Kevin Tian <kevin.tian(a)intel.com> Link: https://lore.kernel.org/r/20251221123508.37495-1-dmaluka@chromium.org Signed-off-by: Lu Baolu <baolu.lu(a)linux.intel.com> Signed-off-by: Joerg Roedel <joerg.roedel(a)amd.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Yin Tirui <yintirui(a)huawei.com> --- drivers/iommu/intel/pasid.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/drivers/iommu/intel/pasid.c b/drivers/iommu/intel/pasid.c index 9d909e99b7e8..69f19ac300a0 100644 --- a/drivers/iommu/intel/pasid.c +++ b/drivers/iommu/intel/pasid.c @@ -268,6 +268,9 @@ struct pasid_entry *intel_pasid_get_entry(struct device *dev, u32 pasid) if (!entries) return NULL; + if (!ecap_coherent(info->iommu->ecap)) + clflush_cache_range(entries, VTD_PAGE_SIZE); + /* * The pasid directory table entry won't be freed after * allocation. No worry about the race with free and @@ -279,10 +282,8 @@ struct pasid_entry *intel_pasid_get_entry(struct device *dev, u32 pasid) free_pgtable_page(entries); goto retry; } - if (!ecap_coherent(info->iommu->ecap)) { - clflush_cache_range(entries, VTD_PAGE_SIZE); + if (!ecap_coherent(info->iommu->ecap)) clflush_cache_range(&dir[dir_index].val, sizeof(*dir)); - } } return &entries[index]; -- 2.43.0

2 1

[PATCH OLK-5.10] Revert "hwmon: (ibmpex) fix use-after-free in high/low store"
by Yin Tirui 06 Jun '26

06 Jun '26

From: Guenter Roeck <linux(a)roeck-us.net> stable inclusion from stable-v5.10.252 commit 05112ba67c824ab416cd54307c0b50aba9f0047a category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/15222 CVE: CVE-2026-45914 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 8bde3e395a85017f12af2b0ba5c3684f5af9c006 ] This reverts commit 6946c726c3f4c36f0f049e6f97e88c510b15f65d. Jean Delvare points out that the patch does not completely fix the reported problem, that it in fact introduces a (new) race condition, and that it may actually not be needed in the first place. Various AI reviews agree. Specific and relevant AI feedback: " This reordering sets the driver data to NULL before removing the sensor attributes in the loop below. ibmpex_show_sensor() retrieves this driver data via dev_get_drvdata() but does not check if it is NULL before dereferencing it to access data->sensors[]. If a userspace process reads a sensor file (like temp1_input) while this delete function is running, could it race with the dev_set_drvdata(..., NULL) call here and crash in ibmpex_show_sensor()? Would it be safer to keep the original order where device_remove_file() is called before clearing the driver data? device_remove_file() should wait for any active sysfs callbacks to complete, which might already prevent the use-after-free this patch intends to fix. " Revert the offending patch. If it can be shown that the originally reported alleged race condition does indeed exist, it can always be re-introduced with a complete fix. Reported-by: Jean Delvare <jdelvare(a)suse.de> Closes: https://lore.kernel.org/linux-hwmon/20260121095342.73e723cb@endymion/ Cc: Jean Delvare <jdelvare(a)suse.de> Cc: Junrui Luo <moonafterrain(a)outlook.com> Fixes: 6946c726c3f4 ("hwmon: (ibmpex) fix use-after-free in high/low store") Reviewed-by: Jean Delvare <jdelvare(a)suse.de> Signed-off-by: Guenter Roeck <linux(a)roeck-us.net> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Yin Tirui <yintirui(a)huawei.com> --- drivers/hwmon/ibmpex.c | 9 ++------- 1 file changed, 2 insertions(+), 7 deletions(-) diff --git a/drivers/hwmon/ibmpex.c b/drivers/hwmon/ibmpex.c index 235d56e96879..fe90f0536d76 100644 --- a/drivers/hwmon/ibmpex.c +++ b/drivers/hwmon/ibmpex.c @@ -282,9 +282,6 @@ static ssize_t ibmpex_high_low_store(struct device *dev, { struct ibmpex_bmc_data *data = dev_get_drvdata(dev); - if (!data) - return -ENODEV; - ibmpex_reset_high_low_data(data); return count; @@ -517,9 +514,6 @@ static void ibmpex_bmc_delete(struct ibmpex_bmc_data *data) { int i, j; - hwmon_device_unregister(data->hwmon_dev); - dev_set_drvdata(data->bmc_device, NULL); - device_remove_file(data->bmc_device, &sensor_dev_attr_reset_high_low.dev_attr); device_remove_file(data->bmc_device, &sensor_dev_attr_name.dev_attr); @@ -533,7 +527,8 @@ static void ibmpex_bmc_delete(struct ibmpex_bmc_data *data) } list_del(&data->list); - + dev_set_drvdata(data->bmc_device, NULL); + hwmon_device_unregister(data->hwmon_dev); ipmi_destroy_user(data->user); kfree(data->sensors); kfree(data); -- 2.43.0

2 1

[PATCH OLK-5.10] spi: fix resource leaks on device setup failure
by Yin Tirui 06 Jun '26

06 Jun '26

From: Johan Hovold <johan(a)kernel.org> mainline inclusion from mainline-v7.1-rc1 commit db357034f7e0cf23f233f414a8508312dfe8fbbe category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/15327 CVE: CVE-2026-46083 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- Make sure to call controller cleanup() if spi_setup() fails while registering a device to avoid leaking any resources allocated by setup(). Fixes: c7299fea6769 ("spi: Fix spi device unregister flow") Cc: stable(a)vger.kernel.org # 5.13 Cc: Saravana Kannan <saravanak(a)kernel.org> Signed-off-by: Johan Hovold <johan(a)kernel.org> Link: https://patch.msgid.link/20260410154907.129248-2-johan@kernel.org Signed-off-by: Mark Brown <broonie(a)kernel.org> Conflicts: drivers/spi/spi.c [Old SPI core lacks upstream __spi_setup()/CS-timing refactor.] Signed-off-by: Yin Tirui <yintirui(a)huawei.com> --- drivers/spi/spi.c | 59 +++++++++++++++++++++++++++++------------------ 1 file changed, 36 insertions(+), 23 deletions(-) diff --git a/drivers/spi/spi.c b/drivers/spi/spi.c index a05877fff8343..309488589b810 100644 --- a/drivers/spi/spi.c +++ b/drivers/spi/spi.c @@ -41,6 +41,8 @@ EXPORT_TRACEPOINT_SYMBOL(spi_transfer_stop); #include "internals.h" +static int __spi_setup(struct spi_device *spi, bool initial_setup); + static DEFINE_IDR(spi_master_idr); static void spidev_release(struct device *dev) @@ -588,7 +590,7 @@ int spi_add_device(struct spi_device *spi) * normally rely on the device being setup. Devices * using SPI_CS_HIGH can't coexist well otherwise... */ - status = spi_setup(spi); + status = __spi_setup(spi, true); if (status < 0) { dev_err(dev, "can't setup %s, status %d\n", dev_name(&spi->dev), status); @@ -3348,27 +3350,7 @@ static int __spi_validate_bits_per_word(struct spi_controller *ctlr, return 0; } -/** - * spi_setup - setup SPI mode and clock rate - * @spi: the device whose settings are being modified - * Context: can sleep, and no requests are queued to the device - * - * SPI protocol drivers may need to update the transfer mode if the - * device doesn't work with its default. They may likewise need - * to update clock rates or word sizes from initial values. This function - * changes those settings, and must be called from a context that can sleep. - * Except for SPI_CS_HIGH, which takes effect immediately, the changes take - * effect the next time the device is selected and data is transferred to - * or from it. When this function returns, the spi device is deselected. - * - * Note that this call will fail if the protocol driver specifies an option - * that the underlying controller or its driver does not support. For - * example, not all hardware supports wire transfers using nine bit words, - * LSB-first wire encoding, or active-high chipselects. - * - * Return: zero on success, else a negative error code. - */ -int spi_setup(struct spi_device *spi) +static int __spi_setup(struct spi_device *spi, bool initial_setup) { unsigned bad_bits, ugly_bits; int status; @@ -3444,7 +3426,7 @@ int spi_setup(struct spi_device *spi) pm_runtime_put_noidle(spi->controller->dev.parent); dev_err(&spi->controller->dev, "Failed to power device: %d\n", status); - return status; + goto err_cleanup; } /* @@ -3479,6 +3461,37 @@ int spi_setup(struct spi_device *spi) status); return status; + +err_cleanup: + if (initial_setup) + spi_cleanup(spi); + + return status; +} + +/** + * spi_setup - setup SPI mode and clock rate + * @spi: the device whose settings are being modified + * Context: can sleep, and no requests are queued to the device + * + * SPI protocol drivers may need to update the transfer mode if the + * device doesn't work with its default. They may likewise need + * to update clock rates or word sizes from initial values. This function + * changes those settings, and must be called from a context that can sleep. + * Except for SPI_CS_HIGH, which takes effect immediately, the changes take + * effect the next time the device is selected and data is transferred to + * or from it. When this function returns, the spi device is deselected. + * + * Note that this call will fail if the protocol driver specifies an option + * that the underlying controller or its driver does not support. For + * example, not all hardware supports wire transfers using nine bit words, + * LSB-first wire encoding, or active-high chipselects. + * + * Return: zero on success, else a negative error code. + */ +int spi_setup(struct spi_device *spi) +{ + return __spi_setup(spi, false); } EXPORT_SYMBOL_GPL(spi_setup); -- 2.43.0

2 1

[PATCH OLK-6.6] vhost: move vdpa group bound check to vhost_vdpa
by Yin Tirui 06 Jun '26

06 Jun '26

From: Eugenio Pérez <eperezma(a)redhat.com> mainline inclusion from mainline-v7.0-rc1 commit cd025c1e876b4e262e71398236a1550486a73ede category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/14792 CVE: CVE-2026-43248 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- Remove duplication by consolidating these here. This reduces the posibility of a parent driver missing them. While we're at it, fix a bug in vdpa_sim where a valid ASID can be assigned to a group equal to ngroups, causing an out of bound write. Cc: stable(a)vger.kernel.org Fixes: bda324fd037a ("vdpasim: control virtqueue support") Acked-by: Jason Wang <jasowang(a)redhat.com> Signed-off-by: Eugenio Pérez <eperezma(a)redhat.com> Signed-off-by: Michael S. Tsirkin <mst(a)redhat.com> Message-Id: <20260119143306.1818855-2-eperezma(a)redhat.com> Conflicts: drivers/vhost/vdpa.c drivers/vdpa/mlx5/net/mlx5_vnet.c [context conflicts.] Signed-off-by: Yin Tirui <yintirui(a)huawei.com> --- drivers/vdpa/mlx5/net/mlx5_vnet.c | 3 --- drivers/vdpa/vdpa_sim/vdpa_sim.c | 6 ------ drivers/vhost/vdpa.c | 2 +- 3 files changed, 1 insertion(+), 10 deletions(-) diff --git a/drivers/vdpa/mlx5/net/mlx5_vnet.c b/drivers/vdpa/mlx5/net/mlx5_vnet.c index a8b76b3e2620e..3095d9da45a0e 100644 --- a/drivers/vdpa/mlx5/net/mlx5_vnet.c +++ b/drivers/vdpa/mlx5/net/mlx5_vnet.c @@ -3178,9 +3178,6 @@ static int mlx5_set_group_asid(struct vdpa_device *vdev, u32 group, { struct mlx5_vdpa_dev *mvdev = to_mvdev(vdev); - if (group >= MLX5_VDPA_NUMVQ_GROUPS) - return -EINVAL; - mvdev->group2asid[group] = asid; return 0; } diff --git a/drivers/vdpa/vdpa_sim/vdpa_sim.c b/drivers/vdpa/vdpa_sim/vdpa_sim.c index 8ac56a895f029..af59d90ba9a82 100644 --- a/drivers/vdpa/vdpa_sim/vdpa_sim.c +++ b/drivers/vdpa/vdpa_sim/vdpa_sim.c @@ -584,12 +584,6 @@ static int vdpasim_set_group_asid(struct vdpa_device *vdpa, unsigned int group, struct vhost_iotlb *iommu; int i; - if (group > vdpasim->dev_attr.ngroups) - return -EINVAL; - - if (asid >= vdpasim->dev_attr.nas) - return -EINVAL; - iommu = &vdpasim->iommu[asid]; mutex_lock(&vdpasim->mutex); diff --git a/drivers/vhost/vdpa.c b/drivers/vhost/vdpa.c index e64400c4c76c8..a4c8bbee1df78 100644 --- a/drivers/vhost/vdpa.c +++ b/drivers/vhost/vdpa.c @@ -730,7 +730,7 @@ static long vhost_vdpa_vring_ioctl(struct vhost_vdpa *v, unsigned int cmd, case VHOST_VDPA_SET_GROUP_ASID: if (copy_from_user(&s, argp, sizeof(s))) return -EFAULT; - if (s.num >= vdpa->nas) + if (idx >= vdpa->ngroups || s.num >= vdpa->nas) return -EINVAL; if (!ops->set_group_asid) return -EOPNOTSUPP; -- 2.43.0

2 1

[PATCH OLK-5.10 0/3] CVE-2026-46065
by Liu Kai 05 Jun '26

05 Jun '26

CVE-2026-46065 Liu Kai (1): fbdev: defio: fix KABI breakage in struct fb_deferred_io and struct fb_info Thomas Zimmermann (2): fbdev: Refactor implementation of page_mkwrite fbdev: defio: Disconnect deferred I/O from the lifetime of struct fb_info drivers/video/fbdev/core/fb_defio.c | 184 ++++++++++++++++++++++++---- include/linux/fb.h | 4 +- include/linux/workqueue.h | 4 + 3 files changed, 167 insertions(+), 25 deletions(-) -- 2.34.1

2 4

[PATCH OLK-5.10] irqchip/gic-v3: Fix ALLINT masking logic by decoupling from GIC NMI support
by Jinjie Ruan 05 Jun '26

05 Jun '26

hulk inclusion category: bugfix bugzilla: https://atomgit.com/openeuler/kernel/issues/9343 -------------------------------- The ARM64 FEAT_NMI implementation relies on two independent hardware validation paths: 1. CPU Feature Check (`system_uses_nmi()`): Validates if the PE supports FEAT_NMI via ID_AA64PFR1_EL1.NMI. If pseudo-NMI is disabled and this bit is set, enables SCTLR_EL1.NMI (map to SCTLR_EL2.NMI in VHE mode) and clears SCTLR_EL1.SPINTMASK. Once active, the hardware automatically sets PSTATE.ALLINT upon taking an exception to EL1/EL2, effectively masking interrupts. `sysreg_clear_set(sctlr_el1, SCTLR_EL1_SPINTMASK, SCTLR_EL1_NMI)` 2. GICv3 Feature Check (gic_data.has_nmi): Validates if the interrupt controller supports Non-maskable Interrupts via GICD_TYPER.NMI. `gic_data.has_nmi = !!(typer & GICD_TYPER_NMI)` And PSTATE.ALLINT is strictly a PE-level state register. Its existence and behavior are entirely independent of whether the underlying GICv3 hardware implementation has its own FEAT_NMI capabilities enabled. Currently, __gic_handle_irq_from_irqson() uses has_v3_3_nmi() to guard the clearing of PSTATE.ALLINT. This creates a destructive dependency where the GIC's NMI status dictates PE register management. static inline bool has_v3_3_nmi(void) { return gic_data.has_nmi && system_uses_nmi(); } If a platform features a PE that supports FEAT_NMI but has a GICv3 that does not support (or has disabled) the NMI feature, the current implementation fails to clear ALLINT during IRQ handling. Because ALLINT is automatically set by the hardware exception entry, it remains permanently set. As a consequence, local_irq_enable() only clears PSTATE.I and F bits, leaving ALLINT untouched. Since ALLINT masks both standard IRQs and NMIs, even if PSTATE.I cleared, the system becomes completely incapable of responding to subsequent NMIs and standard IRQs during downstream processing (such as softirq execution). handle_softirqs() -> local_irq_enable() -> asm volatile("msr daifclr, #3") -> do pending softirq action -> local_irq_disable() -> asm volatile("msr daifset, #3") Fix this by replacing the has_v3_3_nmi() check with system_uses_nmi() when clearing ALLINT. This ensures that PE state management correctly reflects PE capabilities. Additionally, explicitly log whether the GIC supports hardware NMI based on the GICD_TYPER.NMI bit during initialization. Since FEAT_NMI deployment relies on both subsystems, printing this status provides critical diagnostic information for system verification. Fixes: ee945094e78a ("irqchip/gic-v3: Fix hard LOCKUP caused by NMI being masked") Signed-off-by: Jinjie Ruan <ruanjinjie(a)huawei.com> --- drivers/irqchip/irq-gic-v3.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/drivers/irqchip/irq-gic-v3.c b/drivers/irqchip/irq-gic-v3.c index 9870ad4cc766..6e4b3f64bb2c 100644 --- a/drivers/irqchip/irq-gic-v3.c +++ b/drivers/irqchip/irq-gic-v3.c @@ -821,7 +821,7 @@ static void __gic_handle_irq_from_irqson(struct pt_regs *regs) if (gic_prio_masking_enabled()) { gic_pmr_mask_irqs(); gic_arch_enable_irqs(); - } else if (has_v3_3_nmi()) { + } else if (system_uses_nmi()) { #ifdef CONFIG_ARM64_NMI _allint_clear(); #endif @@ -2273,6 +2273,7 @@ static int __init gic_init_bases(void __iomem *dist_base, gic_data.has_nmi = !!(typer & GICD_TYPER_NMI); pr_info("Distributor has %sRange Selector support\n", gic_data.has_rss ? "" : "no "); + pr_info("GICD_TYPER NMI is%s supported.\n", gic_data.has_nmi ? "" : " not"); if (typer & GICD_TYPER_MBIS) { err = mbi_init(handle, gic_data.domain); -- 2.34.1

2 1

[PATCH OLK-5.10] spi: spidev: fix lock inversion between spi_lock and buf_lock
by Lin Ruifeng 05 Jun '26

05 Jun '26

From: Fabian Godehardt <fg(a)emlix.com> stable inclusion from stable-v6.12.75 commit f8431b8672231d378b03176fe74c95adfd3522cf category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/14877 CVE: CVE-2026-43319 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 40534d19ed2afb880ecf202dab26a8e7a5808d16 ] The spidev driver previously used two mutexes, spi_lock and buf_lock, but acquired them in different orders depending on the code path: write()/read(): buf_lock -> spi_lock ioctl(): spi_lock -> buf_lock This AB-BA locking pattern triggers lockdep warnings and can cause real deadlocks: WARNING: possible circular locking dependency detected spidev_ioctl() -> mutex_lock(&spidev->buf_lock) spidev_sync_write() -> mutex_lock(&spidev->spi_lock) *** DEADLOCK *** The issue is reproducible with a simple userspace program that performs write() and SPI_IOC_WR_MAX_SPEED_HZ ioctl() calls from separate threads on the same spidev file descriptor. Fix this by simplifying the locking model and removing the lock inversion entirely. spidev_sync() no longer performs any locking, and all callers serialize access using spi_lock. buf_lock is removed since its functionality is fully covered by spi_lock, eliminating the possibility of lock ordering issues. This removes the lock inversion and prevents deadlocks without changing userspace ABI or behaviour. Signed-off-by: Fabian Godehardt <fg(a)emlix.com> Link: https://patch.msgid.link/20260211072616.489522-1-fg@emlix.com Signed-off-by: Mark Brown <broonie(a)kernel.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Conflicts: drivers/spi/spidev.c [Context Conflicts] Signed-off-by: Lin Ruifeng <linruifeng4(a)huawei.com> --- drivers/spi/spidev.c | 64 +++++++++++++++----------------------------- 1 file changed, 22 insertions(+), 42 deletions(-) diff --git a/drivers/spi/spidev.c b/drivers/spi/spidev.c index 1c072a6e947e..f7b658ec0e7f 100644 --- a/drivers/spi/spidev.c +++ b/drivers/spi/spidev.c @@ -72,7 +72,6 @@ struct spidev_data { struct list_head device_entry; /* TX/RX buffers are NULL unless this device is open (users > 0) */ - struct mutex buf_lock; unsigned users; u8 *tx_buffer; u8 *rx_buffer; @@ -100,25 +99,6 @@ spidev_sync_unlocked(struct spi_device *spi, struct spi_message *message) return status; } -static ssize_t -spidev_sync(struct spidev_data *spidev, struct spi_message *message) -{ - ssize_t status; - struct spi_device *spi; - - mutex_lock(&spidev->spi_lock); - spi = spidev->spi; - - if (spi == NULL) - status = -ESHUTDOWN; - else - status = spidev_sync_unlocked(spi, message); - - mutex_unlock(&spidev->spi_lock); - - return status; -} - static inline ssize_t spidev_sync_write(struct spidev_data *spidev, size_t len) { @@ -131,7 +111,8 @@ spidev_sync_write(struct spidev_data *spidev, size_t len) spi_message_init(&m); spi_message_add_tail(&t, &m); - return spidev_sync(spidev, &m); + + return spidev_sync_unlocked(spidev->spi, &m); } static inline ssize_t @@ -146,7 +127,8 @@ spidev_sync_read(struct spidev_data *spidev, size_t len) spi_message_init(&m); spi_message_add_tail(&t, &m); - return spidev_sync(spidev, &m); + + return spidev_sync_unlocked(spidev->spi, &m); } /*-------------------------------------------------------------------------*/ @@ -156,7 +138,7 @@ static ssize_t spidev_read(struct file *filp, char __user *buf, size_t count, loff_t *f_pos) { struct spidev_data *spidev; - ssize_t status; + ssize_t status = -ESHUTDOWN; /* chipselect only toggles at start or end of operation */ if (count > bufsiz) @@ -164,7 +146,11 @@ spidev_read(struct file *filp, char __user *buf, size_t count, loff_t *f_pos) spidev = filp->private_data; - mutex_lock(&spidev->buf_lock); + mutex_lock(&spidev->spi_lock); + + if (spidev->spi == NULL) + goto err_spi_removed; + status = spidev_sync_read(spidev, count); if (status > 0) { unsigned long missing; @@ -175,7 +161,9 @@ spidev_read(struct file *filp, char __user *buf, size_t count, loff_t *f_pos) else status = status - missing; } - mutex_unlock(&spidev->buf_lock); + +err_spi_removed: + mutex_unlock(&spidev->spi_lock); return status; } @@ -186,7 +174,7 @@ spidev_write(struct file *filp, const char __user *buf, size_t count, loff_t *f_pos) { struct spidev_data *spidev; - ssize_t status; + ssize_t status = -ESHUTDOWN; unsigned long missing; /* chipselect only toggles at start or end of operation */ @@ -195,13 +183,19 @@ spidev_write(struct file *filp, const char __user *buf, spidev = filp->private_data; - mutex_lock(&spidev->buf_lock); + mutex_lock(&spidev->spi_lock); + + if (spidev->spi == NULL) + goto err_spi_removed; + missing = copy_from_user(spidev->tx_buffer, buf, count); if (missing == 0) status = spidev_sync_write(spidev, count); else status = -EFAULT; - mutex_unlock(&spidev->buf_lock); + +err_spi_removed: + mutex_unlock(&spidev->spi_lock); return status; } @@ -375,14 +369,6 @@ spidev_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) return -ESHUTDOWN; } - /* use the buffer lock here for triple duty: - * - prevent I/O (from us) so calling spi_setup() is safe; - * - prevent concurrent SPI_IOC_WR_* from morphing - * data fields while SPI_IOC_RD_* reads them; - * - SPI_IOC_MESSAGE needs the buffer locked "normally". - */ - mutex_lock(&spidev->buf_lock); - switch (cmd) { /* read requests */ case SPI_IOC_RD_MODE: @@ -508,7 +494,6 @@ spidev_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) break; } - mutex_unlock(&spidev->buf_lock); spi_dev_put(spi); mutex_unlock(&spidev->spi_lock); return retval; @@ -539,9 +524,6 @@ spidev_compat_ioc_message(struct file *filp, unsigned int cmd, return -ESHUTDOWN; } - /* SPI_IOC_MESSAGE needs the buffer locked "normally" */ - mutex_lock(&spidev->buf_lock); - /* Check message and copy into scratch area */ ioc = spidev_get_ioc_message(cmd, u_ioc, &n_ioc); if (IS_ERR(ioc)) { @@ -562,7 +544,6 @@ spidev_compat_ioc_message(struct file *filp, unsigned int cmd, kfree(ioc); done: - mutex_unlock(&spidev->buf_lock); spi_dev_put(spi); mutex_unlock(&spidev->spi_lock); return retval; @@ -772,7 +753,6 @@ static int spidev_probe(struct spi_device *spi) /* Initialize the driver data */ spidev->spi = spi; mutex_init(&spidev->spi_lock); - mutex_init(&spidev->buf_lock); INIT_LIST_HEAD(&spidev->device_entry); -- 2.43.0

2 1

[PATCH OLK-6.6] spi: spidev: fix lock inversion between spi_lock and buf_lock
by Lin Ruifeng 05 Jun '26

05 Jun '26

From: Fabian Godehardt <fg(a)emlix.com> stable inclusion from stable-v6.12.75 commit f8431b8672231d378b03176fe74c95adfd3522cf category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/14877 CVE: CVE-2026-43319 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 40534d19ed2afb880ecf202dab26a8e7a5808d16 ] The spidev driver previously used two mutexes, spi_lock and buf_lock, but acquired them in different orders depending on the code path: write()/read(): buf_lock -> spi_lock ioctl(): spi_lock -> buf_lock This AB-BA locking pattern triggers lockdep warnings and can cause real deadlocks: WARNING: possible circular locking dependency detected spidev_ioctl() -> mutex_lock(&spidev->buf_lock) spidev_sync_write() -> mutex_lock(&spidev->spi_lock) *** DEADLOCK *** The issue is reproducible with a simple userspace program that performs write() and SPI_IOC_WR_MAX_SPEED_HZ ioctl() calls from separate threads on the same spidev file descriptor. Fix this by simplifying the locking model and removing the lock inversion entirely. spidev_sync() no longer performs any locking, and all callers serialize access using spi_lock. buf_lock is removed since its functionality is fully covered by spi_lock, eliminating the possibility of lock ordering issues. This removes the lock inversion and prevents deadlocks without changing userspace ABI or behaviour. Signed-off-by: Fabian Godehardt <fg(a)emlix.com> Link: https://patch.msgid.link/20260211072616.489522-1-fg@emlix.com Signed-off-by: Mark Brown <broonie(a)kernel.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Conflicts: drivers/spi/spidev.c [Context Conflicts] Signed-off-by: Lin Ruifeng <linruifeng4(a)huawei.com> --- drivers/spi/spidev.c | 63 ++++++++++++++++---------------------------- 1 file changed, 22 insertions(+), 41 deletions(-) diff --git a/drivers/spi/spidev.c b/drivers/spi/spidev.c index 16bb4fc3a4ba..2024532dfc44 100644 --- a/drivers/spi/spidev.c +++ b/drivers/spi/spidev.c @@ -74,7 +74,6 @@ struct spidev_data { struct list_head device_entry; /* TX/RX buffers are NULL unless this device is open (users > 0) */ - struct mutex buf_lock; unsigned users; u8 *tx_buffer; u8 *rx_buffer; @@ -102,24 +101,6 @@ spidev_sync_unlocked(struct spi_device *spi, struct spi_message *message) return status; } -static ssize_t -spidev_sync(struct spidev_data *spidev, struct spi_message *message) -{ - ssize_t status; - struct spi_device *spi; - - mutex_lock(&spidev->spi_lock); - spi = spidev->spi; - - if (spi == NULL) - status = -ESHUTDOWN; - else - status = spidev_sync_unlocked(spi, message); - - mutex_unlock(&spidev->spi_lock); - return status; -} - static inline ssize_t spidev_sync_write(struct spidev_data *spidev, size_t len) { @@ -132,7 +113,8 @@ spidev_sync_write(struct spidev_data *spidev, size_t len) spi_message_init(&m); spi_message_add_tail(&t, &m); - return spidev_sync(spidev, &m); + + return spidev_sync_unlocked(spidev->spi, &m); } static inline ssize_t @@ -147,7 +129,8 @@ spidev_sync_read(struct spidev_data *spidev, size_t len) spi_message_init(&m); spi_message_add_tail(&t, &m); - return spidev_sync(spidev, &m); + + return spidev_sync_unlocked(spidev->spi, &m); } /*-------------------------------------------------------------------------*/ @@ -157,7 +140,7 @@ static ssize_t spidev_read(struct file *filp, char __user *buf, size_t count, loff_t *f_pos) { struct spidev_data *spidev; - ssize_t status; + ssize_t status = -ESHUTDOWN; /* chipselect only toggles at start or end of operation */ if (count > bufsiz) @@ -165,7 +148,11 @@ spidev_read(struct file *filp, char __user *buf, size_t count, loff_t *f_pos) spidev = filp->private_data; - mutex_lock(&spidev->buf_lock); + mutex_lock(&spidev->spi_lock); + + if (spidev->spi == NULL) + goto err_spi_removed; + status = spidev_sync_read(spidev, count); if (status > 0) { unsigned long missing; @@ -176,7 +163,9 @@ spidev_read(struct file *filp, char __user *buf, size_t count, loff_t *f_pos) else status = status - missing; } - mutex_unlock(&spidev->buf_lock); + +err_spi_removed: + mutex_unlock(&spidev->spi_lock); return status; } @@ -187,7 +176,7 @@ spidev_write(struct file *filp, const char __user *buf, size_t count, loff_t *f_pos) { struct spidev_data *spidev; - ssize_t status; + ssize_t status = -ESHUTDOWN; unsigned long missing; /* chipselect only toggles at start or end of operation */ @@ -196,13 +185,19 @@ spidev_write(struct file *filp, const char __user *buf, spidev = filp->private_data; - mutex_lock(&spidev->buf_lock); + mutex_lock(&spidev->spi_lock); + + if (spidev->spi == NULL) + goto err_spi_removed; + missing = copy_from_user(spidev->tx_buffer, buf, count); if (missing == 0) status = spidev_sync_write(spidev, count); else status = -EFAULT; - mutex_unlock(&spidev->buf_lock); + +err_spi_removed: + mutex_unlock(&spidev->spi_lock); return status; } @@ -376,14 +371,6 @@ spidev_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) return -ESHUTDOWN; } - /* use the buffer lock here for triple duty: - * - prevent I/O (from us) so calling spi_setup() is safe; - * - prevent concurrent SPI_IOC_WR_* from morphing - * data fields while SPI_IOC_RD_* reads them; - * - SPI_IOC_MESSAGE needs the buffer locked "normally". - */ - mutex_lock(&spidev->buf_lock); - switch (cmd) { /* read requests */ case SPI_IOC_RD_MODE: @@ -516,7 +503,6 @@ spidev_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) break; } - mutex_unlock(&spidev->buf_lock); spi_dev_put(spi); mutex_unlock(&spidev->spi_lock); return retval; @@ -547,9 +533,6 @@ spidev_compat_ioc_message(struct file *filp, unsigned int cmd, return -ESHUTDOWN; } - /* SPI_IOC_MESSAGE needs the buffer locked "normally" */ - mutex_lock(&spidev->buf_lock); - /* Check message and copy into scratch area */ ioc = spidev_get_ioc_message(cmd, u_ioc, &n_ioc); if (IS_ERR(ioc)) { @@ -570,7 +553,6 @@ spidev_compat_ioc_message(struct file *filp, unsigned int cmd, kfree(ioc); done: - mutex_unlock(&spidev->buf_lock); spi_dev_put(spi); mutex_unlock(&spidev->spi_lock); return retval; @@ -795,7 +777,6 @@ static int spidev_probe(struct spi_device *spi) /* Initialize the driver data */ spidev->spi = spi; mutex_init(&spidev->spi_lock); - mutex_init(&spidev->buf_lock); INIT_LIST_HEAD(&spidev->device_entry); -- 2.43.0

2 1

[PATCH OLK-6.6 0/3] Fix PMD sharing disabled by missing SPLIT_PMD_PTLOCKS
by Qi Xi 05 Jun '26

05 Jun '26

Patch 1: Define CONFIG_SPLIT_PTE_PTLOCKS and CONFIG_SPLIT_PMD_PTLOCKS by turning the old USE_SPLIT_* macros into proper Kconfig options. Patch 2: in the same series with patch 1. Patch 3: new bugfix for patch 1. David Hildenbrand (2): mm: turn USE_SPLIT_PTE_PTLOCKS / USE_SPLIT_PTE_PTLOCKS into Kconfig options powerpc/8xx: document and enforce that split PT locks are not used Guenter Roeck (1): mm: make SPLIT_PTE_PTLOCKS depend on SMP arch/arm/mm/fault-armv.c | 6 +++--- arch/powerpc/mm/pgtable.c | 4 ++++ arch/x86/xen/mmu_pv.c | 7 ++++--- drivers/misc/zcopy/zcopy.c | 4 ++-- include/linux/mm.h | 8 ++++---- include/linux/mm_types.h | 2 +- include/linux/mm_types_task.h | 3 --- kernel/fork.c | 4 ++-- mm/Kconfig | 19 ++++++++++++------- mm/memory.c | 2 +- 10 files changed, 33 insertions(+), 26 deletions(-) -- 2.53.0

2 4

[PATCH OLK-6.6 0/5] crypto: hisilicon - add backlog support
by ZongYu Wu 05 Jun '26

05 Jun '26

From: parm64 <parm64(a)huawei.com> Chenghai Huang (1): crypto: hisilicon/zip - add backlog support for zip Wenkai Lin (3): crypto: hisilicon/sec2 - fix UAF in sec_alg_send_backlog crypto: hisilicon/sec2 - fix resource leakage issues in non-backlog mode crypto: hisilicon/sec2 - restore iv for ctr mode lizhi (1): crypto: hisilicon/hpre: implement full backlog support for hpre driver drivers/crypto/hisilicon/hpre/hpre_crypto.c | 287 ++++++++++++------ drivers/crypto/hisilicon/sec2/sec_crypto.c | 47 ++- drivers/crypto/hisilicon/zip/zip_crypto.c | 313 +++++++++++++------- 3 files changed, 426 insertions(+), 221 deletions(-) -- 2.43.0

2 6

[PATCH openEuler-1.0-LTS] scsi/hiraid: Support New RAID features
by LinKun 05 Jun '26

05 Jun '26

From: 岳智超 <yuezhichao1(a)h-partners.com> driver inclusion category: feature bugzilla: https://atomgit.com/openeuler/kernel/issues/9101 CVE: NA -------------------------------- Fix high traffic IO disconnection issues Fix drive letter misalignment issue Fix timeout work mode Fix boot issue in single work mode Change adm timeout Change io queue depth Change prp、sgl Change stream detect Add linkdown fast recovery Add boot disk first scan Add flr Add ncq feature Delete driver class Signed-off-by: 岳智超 <yuezhichao1(a)h-partners.com> --- drivers/scsi/hisi_raid/hiraid.h | 87 ++- drivers/scsi/hisi_raid/hiraid_main.c | 919 ++++++++++++++------------- 2 files changed, 536 insertions(+), 470 deletions(-) diff --git a/drivers/scsi/hisi_raid/hiraid.h b/drivers/scsi/hisi_raid/hiraid.h index cb90d3d..25df43b 100644 --- a/drivers/scsi/hisi_raid/hiraid.h +++ b/drivers/scsi/hisi_raid/hiraid.h @@ -50,6 +50,10 @@ #define HIRAID_DEV_INFO_FLAG_VALID(flag) ((flag) & 0x01) #define HIRAID_DEV_INFO_FLAG_CHANGE(flag) ((flag) & 0x02) +#define HIRAID_DEV_INFO_FLAG_DEV_MAGIC_DEV(flag) ((flag) >> 2) +#define HIRAID_DEV_INFO_FLAG_DELETE_OR_ADD(flag, org_flag) \ + (HIRAID_DEV_INFO_FLAG_DEV_MAGIC_DEV(flag) != \ + HIRAID_DEV_INFO_FLAG_DEV_MAGIC_DEV(org_flag) ? 1 : 0) #define HIRAID_CAP_MQES(cap) ((cap) & 0xffff) #define HIRAID_CAP_STRIDE(cap) (((cap) >> 32) & 0xf) @@ -74,7 +78,19 @@ #define PCI_VENDOR_ID_HUAWEI_LOGIC 0x19E5 #define HIRAID_SERVER_DEVICE_HBA_DID 0x3858 +#define HIRAID_SERVER_DEVICE_HBAS_DID 0x3918 #define HIRAID_SERVER_DEVICE_RAID_DID 0x3758 +#define HIRAID_SERVER_DEVICE_RAIDS_DID 0x38D8 + +#define HIRAID_SCSI_VPD_LEN SCSI_VPD_PG_LEN +#define SCSI_RSVD_INIT_SHIFT 0XFC +#define NCQ_PRIO_SUPPORT_BYTE 213 +#define VPD_NCQ_PRIO_SUPPORT_BIT 4 + +#define HIRAID_CAP_CAC_STRIDE 8 + +#define HIRAID_MAX_PD_NUM (40 + 1) +#define HIRAID_MAX_STREAM_NUM 8 enum { HIRAID_SC_SUCCESS = 0x0, @@ -90,6 +106,7 @@ enum { enum { HIRAID_REG_CAP = 0x0000, + HIRAID_REG_VS = 0x0008, HIRAID_REG_CC = 0x0014, HIRAID_REG_CSTS = 0x001c, HIRAID_REG_AQA = 0x0024, @@ -236,6 +253,20 @@ enum { DISPATCH_BY_DISK, }; +enum hiraid_stream_type { + HIRAID_STREAM_TYPE_TOTAL, + HIRAID_STREAM_TYPE_WRITE, + HIRAID_STREAM_TYPE_READ, + HIRAID_STREAM_TYPE_CLEAN, + HIRAID_STREAM_TYPE_BOTTOM +}; + +enum hiraid_stream_io_operation_type { + HIRAID_STREAM_TYPE_DELETE_SINGLE_IO = 1, + HIRAID_STREAM_TYPE_DELETE_SINGLE_IO_LIST, + HIRAID_STREAM_TYPE_DELETE_ALL_IO_LIST +}; + struct hiraid_completion { __le32 result; union { @@ -271,6 +302,19 @@ struct hiraid_ctrl_info { __u8 rsvd1[4020]; }; +struct hiraid_stream { + /* recog-window */ + u64 stream_lba; + u32 stream_len; + u16 did; + u16 type; + /* aging ctrl */ + int aging_credit; + int aging_grade; + u16 stream_id; + u16 stream_is_using; +}; + struct hiraid_dev { struct pci_dev *pdev; struct device *dev; @@ -318,6 +362,18 @@ struct hiraid_dev { u8 hdd_dispatch; struct request_queue *bsg_queue; + + u16 hiraid_stream_io_count; + u64 hiraid_stream_aging_time; + u64 hiraid_stream_sent_io_size[HIRAID_MAX_PD_NUM] + [HIRAID_MAX_STREAM_NUM]; + u16 hiraid_stream_io_num_per_pd[HIRAID_MAX_PD_NUM] + [HIRAID_STREAM_TYPE_BOTTOM + 1]; + spinlock_t hiraid_stream_array_lock; + struct hiraid_stream hiraid_stream_array[HIRAID_MAX_PD_NUM] + [HIRAID_MAX_STREAM_NUM]; + struct task_struct *hiraid_stream_submit_task; + u64 hiraid_stream_io_last_pull_time[HIRAID_MAX_PD_NUM]; }; struct hiraid_sgl_desc { @@ -696,6 +752,10 @@ struct hiraid_mapmange { u32 sge_cnt; u32 len; bool use_sgl; + /* This field is used in the I/O read/write process. + * It indicates the TRANSFER LENGTH field in the CDB. + */ + u32 cdb_data_len; dma_addr_t first_dma; void *sense_buffer_virt; dma_addr_t sense_buffer_phy; @@ -755,38 +815,19 @@ struct hiraid_sdev_hostdata { u8 flag; u8 rg_id; u8 hwq; + u8 sata_ncq_prio_enable; + u8 rsvd[3]; u16 pend_count; }; -enum stream_type { - TYPE_TOTAL, - TYPE_WRITE, - TYPE_READ, - TYPE_CLEAN, - TYPE_BOTTOM -}; - -struct HIRAID_STREAM_S { - /* recog-window */ - u64 stream_lba; - u32 stream_len; - u16 did; - u16 type; - /* aging ctrl */ - int aging_credit; - int aging_grade; - u16 stream_id; - u16 using; -}; - -struct IO_LIST_S { +struct hiraid_stream_io_list { struct list_head list; struct hiraid_scsi_io_cmd io_cmd; struct hiraid_queue *submit_queue; unsigned int sector_size; }; -struct mutex_list_head_s { +struct mutex_list_head { struct list_head list; struct mutex lock; }; diff --git a/drivers/scsi/hisi_raid/hiraid_main.c b/drivers/scsi/hisi_raid/hiraid_main.c index cdba466..9251fb8 100644 --- a/drivers/scsi/hisi_raid/hiraid_main.c +++ b/drivers/scsi/hisi_raid/hiraid_main.c @@ -26,6 +26,10 @@ #include <linux/bsg-lib.h> #include <asm/unaligned.h> #include <linux/sort.h> +#include <linux/kthread.h> +#include <linux/mutex.h> +#include <linux/sched.h> +#include <linux/sched/prio.h> #include <target/target_core_backend.h> #include <scsi/scsi.h> @@ -35,14 +39,10 @@ #include <scsi/scsi_transport.h> #include <scsi/scsi_dbg.h> #include <scsi/sg.h> -#include <linux/kthread.h> -#include <linux/mutex.h> -#include <linux/sched.h> -#include <linux/sched/prio.h> #include "hiraid.h" -static u32 admin_tmout = 60; +static u32 admin_tmout = 180; module_param(admin_tmout, uint, 0644); MODULE_PARM_DESC(admin_tmout, "admin commands timeout (seconds)"); @@ -101,7 +101,7 @@ static const struct kernel_param_ops io_queue_depth_ops = { .get = param_get_uint, }; -static u32 io_queue_depth = 1024; +static u32 io_queue_depth = 4096; module_param_cb(io_queue_depth, &io_queue_depth_ops, &io_queue_depth, 0644); MODULE_PARM_DESC(io_queue_depth, "set io queue depth, should >= 2"); @@ -143,8 +143,6 @@ static void hiraid_handle_async_notice(struct hiraid_dev *hdev, u32 result); static void hiraid_handle_async_vs(struct hiraid_dev *hdev, u32 result, u32 result1); -static struct class *hiraid_class; - #define HIRAID_CAP_TIMEOUT_UNIT_MS (HZ / 2) static struct workqueue_struct *work_queue; @@ -155,7 +153,7 @@ static struct workqueue_struct *work_queue; __func__, ##__VA_ARGS__); \ } while (0) -#define HIRAID_DRV_VERSION "1.1.0.1" +#define HIRAID_DRV_VERSION "2.1.0.1" #define ADMIN_TIMEOUT (admin_tmout * HZ) #define USRCMD_TIMEOUT (180 * HZ) @@ -178,8 +176,12 @@ static struct workqueue_struct *work_queue; #define MIN_CREDIT 0 #define MAX_CREDIT 64 #define CREDIT_THRES 32 +#ifndef MIN #define MIN(a, b) (((a) < (b)) ? (a) : (b)) +#endif +#ifndef MAX #define MAX(a, b) (((a) > (b)) ? (a) : (b)) +#endif enum SENSE_STATE_CODE { SENSE_STATE_OK = 0, @@ -362,95 +364,8 @@ static u32 hiraid_get_max_cmd_size(struct hiraid_dev *hdev) return sizeof(struct hiraid_mapmange) + alloc_size; } -static int hiraid_build_passthru_prp(struct hiraid_dev *hdev, - struct hiraid_mapmange *mapbuf) -{ - struct scatterlist *sg = mapbuf->sgl; - __le64 *phy_regpage, *prior_list; - u64 buf_addr = sg_dma_address(sg); - int buf_length = sg_dma_len(sg); - u32 page_size = hdev->page_size; - int offset = buf_addr & (page_size - 1); - void **list = hiraid_mapbuf_list(mapbuf); - int maplen = mapbuf->len; - struct dma_pool *pool; - dma_addr_t buffer_phy; - int i; - - maplen -= (page_size - offset); - if (maplen <= 0) { - mapbuf->first_dma = 0; - return 0; - } - - buf_length -= (page_size - offset); - if (buf_length) { - buf_addr += (page_size - offset); - } else { - sg = sg_next(sg); - buf_addr = sg_dma_address(sg); - buf_length = sg_dma_len(sg); - } - - if (maplen <= page_size) { - mapbuf->first_dma = buf_addr; - return 0; - } - - pool = hdev->prp_page_pool; - mapbuf->page_cnt = 1; - - phy_regpage = dma_pool_alloc(pool, GFP_ATOMIC, &buffer_phy); - if (!phy_regpage) { - dev_err_ratelimited(hdev->dev, "allocate first admin prp_list memory failed\n"); - mapbuf->first_dma = buf_addr; - mapbuf->page_cnt = -1; - return -ENOMEM; - } - list[0] = phy_regpage; - mapbuf->first_dma = buffer_phy; - i = 0; - for (;;) { - if (i == page_size / PRP_ENTRY_SIZE) { - prior_list = phy_regpage; - - phy_regpage = dma_pool_alloc(pool, GFP_ATOMIC, - &buffer_phy); - if (!phy_regpage) { - dev_err_ratelimited(hdev->dev, "allocate [%d]th admin prp list memory failed\n", - mapbuf->page_cnt + 1); - return -ENOMEM; - } - list[mapbuf->page_cnt++] = phy_regpage; - phy_regpage[0] = prior_list[i - 1]; - prior_list[i - 1] = cpu_to_le64(buffer_phy); - i = 1; - } - phy_regpage[i++] = cpu_to_le64(buf_addr); - buf_addr += page_size; - buf_length -= page_size; - maplen -= page_size; - if (maplen <= 0) - break; - if (buf_length > 0) - continue; - if (unlikely(buf_length < 0)) - goto bad_admin_sgl; - sg = sg_next(sg); - buf_addr = sg_dma_address(sg); - buf_length = sg_dma_len(sg); - } - - return 0; - -bad_admin_sgl: - dev_err(hdev->dev, "setup prps, invalid admin SGL for payload[%d] nents[%d]\n", - mapbuf->len, mapbuf->sge_cnt); - return -EIO; -} - static int hiraid_build_prp(struct hiraid_dev *hdev, - struct hiraid_mapmange *mapbuf) + struct hiraid_mapmange *mapbuf) { struct scatterlist *sg = mapbuf->sgl; __le64 *phy_regpage, *prior_list; @@ -656,67 +571,9 @@ static void hiraid_sgl_set_seg(struct hiraid_sgl_desc *sge, } } -static int hiraid_build_passthru_sgl(struct hiraid_dev *hdev, - struct hiraid_admin_command *admin_cmd, - struct hiraid_mapmange *mapbuf) -{ - struct hiraid_sgl_desc *sg_list, *link, *old_sg_list; - struct scatterlist *sg = mapbuf->sgl; - void **list = hiraid_mapbuf_list(mapbuf); - struct dma_pool *pool; - int nsge = mapbuf->sge_cnt; - dma_addr_t buffer_phy; - int i = 0; - - admin_cmd->common.flags |= SQE_FLAG_SGL_METABUF; - - if (nsge == 1) { - hiraid_sgl_set_data(&admin_cmd->common.dptr.sgl, sg); - return 0; - } - - pool = hdev->prp_page_pool; - mapbuf->page_cnt = 1; - - sg_list = dma_pool_alloc(pool, GFP_ATOMIC, &buffer_phy); - if (!sg_list) { - dev_err_ratelimited(hdev->dev, "allocate first admin sgl_list failed\n"); - mapbuf->page_cnt = -1; - return -ENOMEM; - } - - list[0] = sg_list; - mapbuf->first_dma = buffer_phy; - hiraid_sgl_set_seg(&admin_cmd->common.dptr.sgl, buffer_phy, nsge); - do { - if (i == SGES_PER_PAGE) { - old_sg_list = sg_list; - link = &old_sg_list[SGES_PER_PAGE - 1]; - - sg_list = dma_pool_alloc(pool, GFP_ATOMIC, &buffer_phy); - if (!sg_list) { - dev_err_ratelimited(hdev->dev, "allocate [%d]th admin sgl_list failed\n", - mapbuf->page_cnt + 1); - return -ENOMEM; - } - list[mapbuf->page_cnt++] = sg_list; - - i = 0; - memcpy(&sg_list[i++], link, sizeof(*link)); - hiraid_sgl_set_seg(link, buffer_phy, nsge); - } - - hiraid_sgl_set_data(&sg_list[i++], sg); - sg = sg_next(sg); - } while (--nsge > 0); - - return 0; -} - - static int hiraid_build_sgl(struct hiraid_dev *hdev, - struct hiraid_scsi_io_cmd *io_cmd, - struct hiraid_mapmange *mapbuf) + struct hiraid_scsi_io_cmd *io_cmd, + struct hiraid_mapmange *mapbuf) { struct hiraid_sgl_desc *sg_list, *link, *old_sg_list; struct scatterlist *sg = mapbuf->sgl; @@ -776,102 +633,52 @@ static int hiraid_build_sgl(struct hiraid_dev *hdev, return 0; } -#define MAX_PD_NUM (40 + 1) -#define MAX_STREAM_NUM 8 -#define PER_MB (1024 * 1024) -#define MAX_IO_NUM (200 * PER_MB) -#define STREAM_LEN (4 * PER_MB) +#define MAX_IO_NUM (200 * 1024 * 1024) +#define STREAM_LEN (4 * 1024 * 1024) #define MAX_IO_NUM_ONCE 100 #define IO_SUBMIT_TIME_OUT 100 -#define MAX_AGING_NUM 100 - +#define MAX_AGING_NUM 5000 +#define MAX_AGING_TIME 16 +#define AGING_DEGRADE (-2) #define MIN_IO_SEND_TIME 10 #define MAX_IO_SEND_TIME 50 -enum io_operation_type { - TYPE_DELETE_SINGLE_IO = 1, - TYPE_DELETE_SINGLE_IO_LIST, - TYPE_DELETE_ALL_IO_LIST -}; - -struct HIRAID_STREAM_S stream_array[MAX_PD_NUM][MAX_STREAM_NUM] = {0}; -struct mutex_list_head_s io_heads_per_stream[MAX_PD_NUM * MAX_STREAM_NUM]; -spinlock_t stream_array_lock; +struct mutex_list_head hiraid_io_heads_per_stream[ + HIRAID_MAX_PD_NUM * HIRAID_MAX_STREAM_NUM]; DEFINE_MUTEX(g_stream_operation_mutex); -u64 g_io_transport_num[MAX_PD_NUM][MAX_STREAM_NUM] = {0}; -u16 g_io_stream_num[MAX_PD_NUM][TYPE_BOTTOM] = {0}; -u16 g_io_count = 1; - -void hiraid_inc_io_transport_num(u16 disk_id, u16 streamd_id, u16 nlb) -{ - g_io_transport_num[disk_id][streamd_id] += nlb; -} - -void hiraid_refresh_io_transport_num(u16 disk_id, u16 streamd_id) -{ - g_io_transport_num[disk_id][streamd_id] = 0; -} - -void hiraid_inc_stream_num(u16 disk_id) -{ - spin_lock(&stream_array_lock); - g_io_stream_num[disk_id][TYPE_TOTAL]++; - spin_unlock(&stream_array_lock); -} - -void hiraid_dec_stream_num(u16 disk_id) -{ - spin_lock(&stream_array_lock); - if (g_io_stream_num[disk_id][TYPE_TOTAL] > 0) - g_io_stream_num[disk_id][TYPE_TOTAL]--; - spin_unlock(&stream_array_lock); -} - -static bool hiraid_io_recog_check_stream_exceed(u16 disk_id) +static inline bool hiraid_io_recog_check_stream_exceed(struct hiraid_dev *hdev, + u16 disk_id) { bool exceed_flag; - spin_lock(&stream_array_lock); - exceed_flag = (g_io_stream_num[disk_id][TYPE_TOTAL] >= MAX_STREAM_NUM); - spin_unlock(&stream_array_lock); + spin_lock(&hdev->hiraid_stream_array_lock); + exceed_flag = + (hdev->hiraid_stream_io_num_per_pd[disk_id][HIRAID_STREAM_TYPE_TOTAL] >= + HIRAID_MAX_STREAM_NUM); + spin_unlock(&hdev->hiraid_stream_array_lock); return exceed_flag; } -static u16 hiraid_get_stream_num(u16 disk_id) -{ - return g_io_stream_num[disk_id][TYPE_TOTAL]; -} - -static inline struct HIRAID_STREAM_S *hiraid_get_stream(u16 disk_id, - u16 stream_id) -{ - return &stream_array[disk_id][stream_id]; -} - -static inline struct mutex_list_head_s *hiraid_get_io_head(u16 disk_id) -{ - return &(io_heads_per_stream[disk_id]); -} - -static bool hiraid_recognition_acknowledge(const struct HIRAID_STREAM_S *stream) +static inline bool hiraid_recognition_acknowledge( + const struct hiraid_stream *stream) { return (stream->aging_credit >= CREDIT_THRES) ? true : false; } -void hiraid_io_recognition_init(void) +void hiraid_io_recognition_init(struct hiraid_dev *hdev) { u16 i; - spin_lock_init(&stream_array_lock); - for (i = 0; i < (MAX_PD_NUM * MAX_STREAM_NUM); i++) { - INIT_LIST_HEAD(&hiraid_get_io_head(i)->list); - mutex_init(&hiraid_get_io_head(i)->lock); + spin_lock_init(&hdev->hiraid_stream_array_lock); + for (i = 0; i < (HIRAID_MAX_PD_NUM * HIRAID_MAX_STREAM_NUM); i++) { + INIT_LIST_HEAD(&hiraid_io_heads_per_stream[i].list); + mutex_init(&hiraid_io_heads_per_stream[i].lock); } } -static void hiraid_io_recognition_iterator(struct HIRAID_STREAM_S *stream, - int direction) +static void hiraid_io_recognition_iterator(struct hiraid_stream *stream, + int direction) { stream->aging_grade = stream->aging_grade + direction * INC_GRADE; stream->aging_grade = MAX(stream->aging_grade, MAX_DECREASE_GRADE); @@ -881,46 +688,52 @@ static void hiraid_io_recognition_iterator(struct HIRAID_STREAM_S *stream, stream->aging_credit = MIN(stream->aging_credit, MAX_CREDIT); } -struct HIRAID_STREAM_S *hiraid_io_pick_stream( +struct hiraid_stream *hiraid_io_pick_stream(struct hiraid_dev *hdev, struct hiraid_scsi_rw_cmd *req, u16 type, u16 actual_id) { - struct HIRAID_STREAM_S *first_hit_stream = NULL; - struct HIRAID_STREAM_S *temp_stream = NULL; - u16 pick_flag = 0; + struct hiraid_stream *first_hit_stream = NULL; + struct hiraid_stream *temp_stream = NULL; + bool pick_flag = false; u8 i; - for (i = 0; i < MAX_STREAM_NUM; i++) { - temp_stream = &stream_array[actual_id][i]; + for (i = 0; i < HIRAID_MAX_STREAM_NUM; i++) { + temp_stream = &hdev->hiraid_stream_array[actual_id][i]; temp_stream->stream_id = i; if (req->slba < temp_stream->stream_lba || req->slba >= temp_stream->stream_lba + temp_stream->stream_len || - temp_stream->type != type) { + temp_stream->type != type || + !temp_stream->stream_is_using) { continue; } - if (!pick_flag) { + if (pick_flag == false) { temp_stream->stream_lba = req->slba; first_hit_stream = temp_stream; - pick_flag = 1; + pick_flag = true; continue; } - hiraid_dec_stream_num(actual_id); - memset(temp_stream, 0, - sizeof(struct HIRAID_STREAM_S)); // 去重影 + spin_lock(&hdev->hiraid_stream_array_lock); + if (hdev->hiraid_stream_io_num_per_pd[actual_id] + [HIRAID_STREAM_TYPE_TOTAL] > 0) + hdev->hiraid_stream_io_num_per_pd[actual_id] + [HIRAID_STREAM_TYPE_TOTAL]--; + + memset(temp_stream, 0, sizeof(struct hiraid_stream)); + spin_unlock(&hdev->hiraid_stream_array_lock); } return first_hit_stream; } -static struct HIRAID_STREAM_S *hiraid_init_flow_stream( +static struct hiraid_stream *hiraid_init_flow_stream(struct hiraid_dev *hdev, struct hiraid_scsi_rw_cmd *req, u16 type, u16 actual_id) { int i; - struct HIRAID_STREAM_S *stream = NULL; + struct hiraid_stream *stream = NULL; - for (i = 0; i < MAX_STREAM_NUM; i++) { - stream = hiraid_get_stream(actual_id, i); - if (!stream->using) { - stream->using = 1; + for (i = 0; i < HIRAID_MAX_STREAM_NUM; i++) { + stream = &hdev->hiraid_stream_array[actual_id][i]; + if (!stream->stream_is_using) { + stream->stream_is_using = 1; stream->stream_id = i; break; } @@ -934,82 +747,96 @@ static struct HIRAID_STREAM_S *hiraid_init_flow_stream( return stream; } -static struct HIRAID_STREAM_S *hiraid_stream_detect(struct hiraid_dev *hdev, - struct hiraid_scsi_rw_cmd *io_cmd, u16 actual_id) +static struct hiraid_stream *hiraid_stream_detect(struct hiraid_dev *hdev, + struct hiraid_scsi_rw_cmd *io_cmd, u16 actual_id) { - u16 type = io_cmd->opcode == HIRAID_CMD_WRITE ? TYPE_WRITE : TYPE_READ; - struct HIRAID_STREAM_S *stream = hiraid_io_pick_stream(io_cmd, type, - actual_id); + u16 type = HIRAID_STREAM_TYPE_READ; + struct hiraid_stream *stream = NULL; + + if (io_cmd->opcode == HIRAID_CMD_WRITE) + type = HIRAID_STREAM_TYPE_WRITE; + + stream = hiraid_io_pick_stream(hdev, + io_cmd, type, actual_id); - if (stream != NULL) { /* 可以命中一个stream */ + if (stream != NULL) return stream; - } - if (hiraid_io_recog_check_stream_exceed(actual_id)) + if (hiraid_io_recog_check_stream_exceed(hdev, actual_id)) return NULL; - stream = hiraid_init_flow_stream(io_cmd, type, actual_id); - hiraid_inc_stream_num(actual_id); + + stream = hiraid_init_flow_stream(hdev, io_cmd, type, actual_id); + spin_lock(&hdev->hiraid_stream_array_lock); + hdev->hiraid_stream_io_num_per_pd[actual_id] + [HIRAID_STREAM_TYPE_TOTAL]++; + spin_unlock(&hdev->hiraid_stream_array_lock); return stream; } -u64 g_io_last_pull_time[MAX_PD_NUM] = {0}; - static u16 hiraid_get_submit_io_stream(u16 did, struct hiraid_dev *hdev) { u64 temp_num, i; - static u16 stream_num[MAX_PD_NUM] = {0}; + static u16 stream_num[HIRAID_MAX_PD_NUM] = {0}; - if (g_io_last_pull_time[did] == 0) - g_io_last_pull_time[did] = jiffies_to_msecs(jiffies); + if (hdev->hiraid_stream_io_last_pull_time[did] == 0) + hdev->hiraid_stream_io_last_pull_time[did] = + jiffies_to_msecs(jiffies); - for (i = 0; i < MAX_STREAM_NUM; i++) { - temp_num = g_io_transport_num[did][i]; + for (i = 0; i < HIRAID_MAX_STREAM_NUM; i++) { + temp_num = hdev->hiraid_stream_sent_io_size[did][i]; if (temp_num != 0) { if ((temp_num < MAX_IO_NUM) && - ((jiffies_to_msecs(jiffies) - g_io_last_pull_time[did]) - < IO_SUBMIT_TIME_OUT)) { + ((jiffies_to_msecs(jiffies) - + hdev->hiraid_stream_io_last_pull_time[did]) < + IO_SUBMIT_TIME_OUT)) { stream_num[did] = i; return i; } - g_io_last_pull_time[did] = jiffies_to_msecs(jiffies); - hiraid_refresh_io_transport_num(did, i); - stream_num[did] = ((i+1) % MAX_STREAM_NUM); - return ((i+1) % MAX_STREAM_NUM); + hdev->hiraid_stream_io_last_pull_time[did] = + jiffies_to_msecs(jiffies); + hdev->hiraid_stream_sent_io_size[did][i] = 0; + stream_num[did] = ((i+1) % + HIRAID_MAX_STREAM_NUM); + return ((i+1) % HIRAID_MAX_STREAM_NUM); + } } - g_io_last_pull_time[did] = jiffies_to_msecs(jiffies); - return ((stream_num[did]++) % MAX_STREAM_NUM); + hdev->hiraid_stream_io_last_pull_time[did] = + jiffies_to_msecs(jiffies); + return ((stream_num[did]++) % HIRAID_MAX_STREAM_NUM); } static void hiraid_submit_io_stream(u16 hdid, struct hiraid_dev *hdev) { - struct mutex_list_head_s *io_slist = NULL; + struct mutex_list_head *io_slist = NULL; struct list_head *node = NULL; struct list_head *next_node = NULL; struct list_head temp_header; - struct hiraid_scsi_io_cmd io_cmd = {0}; + struct hiraid_scsi_io_cmd io_cmd; u16 submit_stream_id; - struct IO_LIST_S *temp_io_stream = NULL; + struct hiraid_stream_io_list *temp_io_stream = NULL; u16 count = 0; INIT_LIST_HEAD(&temp_header); submit_stream_id = hiraid_get_submit_io_stream(hdid, hdev); - io_slist = hiraid_get_io_head(hdid * MAX_STREAM_NUM + submit_stream_id); + io_slist = &hiraid_io_heads_per_stream[hdid * HIRAID_MAX_STREAM_NUM + + submit_stream_id]; mutex_lock(&io_slist->lock); list_for_each_safe(node, next_node, &io_slist->list) { list_del(node); list_add_tail(node, &temp_header); - if (++count >= MAX_IO_NUM_ONCE) + if (count++ >= MAX_IO_NUM_ONCE) break; } mutex_unlock(&io_slist->lock); list_for_each_safe(node, next_node, &temp_header) { - temp_io_stream = list_entry(node, struct IO_LIST_S, list); + temp_io_stream = list_entry(node, + struct hiraid_stream_io_list, list); io_cmd = temp_io_stream->io_cmd; hiraid_submit_cmd(temp_io_stream->submit_queue, &io_cmd); - hiraid_inc_io_transport_num(hdid, submit_stream_id, - io_cmd.rw.nlb * temp_io_stream->sector_size); + hdev->hiraid_stream_sent_io_size[hdid][submit_stream_id] += + io_cmd.rw.nlb * temp_io_stream->sector_size; list_del(node); kfree(temp_io_stream); } @@ -1017,59 +844,90 @@ static void hiraid_submit_io_stream(u16 hdid, struct hiraid_dev *hdev) list_del_init(&temp_header); } -static u8 hiraid_detect_if_aging(void) +static u8 hiraid_aging_detect(struct hiraid_dev *hdev) { - if (++g_io_count == MAX_AGING_NUM) { - g_io_count = 0; + if (hdev->hiraid_stream_aging_time == 0) + hdev->hiraid_stream_aging_time = jiffies_to_msecs(jiffies); + if (++hdev->hiraid_stream_io_count > MAX_AGING_NUM) { + hdev->hiraid_stream_io_count = 0; + if ((jiffies_to_msecs(jiffies) - + hdev->hiraid_stream_aging_time) < + MAX_AGING_TIME) + return 0; + hdev->hiraid_stream_aging_time = jiffies_to_msecs(jiffies); return 1; } return 0; } +static void hiraid_sync_using_stream(struct hiraid_dev *hdev, u16 hdid) +{ + u8 i; + struct hiraid_stream *temp_stream = NULL; + + hdev->hiraid_stream_io_num_per_pd[hdid][HIRAID_STREAM_TYPE_TOTAL] = 0; + for (i = 0; i < HIRAID_MAX_STREAM_NUM; i++) { + temp_stream = &hdev->hiraid_stream_array[hdid][i]; + if (temp_stream->stream_is_using) + hdev->hiraid_stream_io_num_per_pd[hdid] + [HIRAID_STREAM_TYPE_TOTAL]++; + } +} + static void hiraid_aging(struct hiraid_dev *hdev) { - struct HIRAID_STREAM_S *temp_stream = NULL; + struct hiraid_stream *temp_stream = NULL; int i = 0; int j = 0; - - for (i = 0; i < MAX_PD_NUM; i++) { - for (j = 0; j < MAX_STREAM_NUM; j++) { - temp_stream = hiraid_get_stream(i, j); - if (temp_stream->using) { - hiraid_io_recognition_iterator(temp_stream, -1); + u16 *tmp_io_num = NULL; + + for (i = 0; i < HIRAID_MAX_PD_NUM; i++) { + for (j = 0; j < HIRAID_MAX_STREAM_NUM; j++) { + temp_stream = &hdev->hiraid_stream_array[i][j]; + if (temp_stream->stream_is_using) { + hiraid_io_recognition_iterator(temp_stream, + AGING_DEGRADE); + spin_lock(&hdev->hiraid_stream_array_lock); + tmp_io_num = &hdev->hiraid_stream_io_num_per_pd + [i][HIRAID_STREAM_TYPE_TOTAL]; if (temp_stream->aging_credit <= 0) { - hiraid_dec_stream_num(i); + if (*tmp_io_num > 0) + (*tmp_io_num)--; memset(temp_stream, - 0, sizeof(struct HIRAID_STREAM_S)); // 老化 + 0, sizeof(struct hiraid_stream)); } + spin_unlock(&hdev->hiraid_stream_array_lock); } } + hiraid_sync_using_stream(hdev, i); } } -static u8 hiraid_io_list_operation(u32 hdid, u16 cid, u16 hwq, u8 operation) +static u8 hiraid_stream_delete_io_list(struct hiraid_dev *hdev, u32 hdid, + u16 cid, u16 hwq, u8 delete_operation) { int i, j; - - struct mutex_list_head_s *io_slist = NULL; + struct mutex_list_head *io_slist = NULL; struct list_head *node = NULL; struct list_head *next_node = NULL; struct hiraid_scsi_io_cmd *io_cmd = NULL; struct hiraid_queue *hiraidq = NULL; - struct IO_LIST_S *temp_io_stream = NULL; + struct hiraid_stream_io_list *temp_io_stream = NULL; + u8 max_hd_num = delete_operation == + HIRAID_STREAM_TYPE_DELETE_ALL_IO_LIST ? HIRAID_MAX_PD_NUM : hdid + 1; - u8 max_hd_num = operation == TYPE_DELETE_ALL_IO_LIST ? - MAX_PD_NUM : hdid + 1; for (i = hdid; i < max_hd_num; i++) { - for (j = 0; j < MAX_STREAM_NUM; j++) { - io_slist = hiraid_get_io_head(i * MAX_STREAM_NUM + j); + for (j = 0; j < HIRAID_MAX_STREAM_NUM; j++) { + io_slist = + &hiraid_io_heads_per_stream[i * HIRAID_MAX_STREAM_NUM + j]; mutex_lock(&io_slist->lock); list_for_each_safe(node, next_node, &io_slist->list) { - temp_io_stream = list_entry(node, - struct IO_LIST_S, list); + temp_io_stream = + list_entry(node, struct hiraid_stream_io_list, list); io_cmd = &(temp_io_stream->io_cmd); hiraidq = temp_io_stream->submit_queue; - if (operation >= TYPE_DELETE_SINGLE_IO_LIST) { + if (delete_operation >= + HIRAID_STREAM_TYPE_DELETE_SINGLE_IO_LIST) { list_del_init(node); kfree(temp_io_stream); temp_io_stream = NULL; @@ -1089,54 +947,27 @@ static u8 hiraid_io_list_operation(u32 hdid, u16 cid, u16 hwq, u8 operation) return 0; } -static u8 hiraid_check_io_list(u32 hdid, u16 cid, u16 hwq) -{ - u8 ret; - - mutex_lock(&g_stream_operation_mutex); - ret = hiraid_io_list_operation(hdid, cid, hwq, TYPE_DELETE_SINGLE_IO); - mutex_unlock(&g_stream_operation_mutex); - return ret; -} - -static u8 hiraid_delete_single_pd_io_list(u32 hdid) +static u8 hiraid_add_io_to_list(struct hiraid_dev *hdev, + struct hiraid_queue *submit_queue, struct hiraid_stream *tmp_stream, + struct hiraid_scsi_io_cmd io_cmd, unsigned int sector_size, + u16 actual_id) { - u8 ret; + struct mutex_list_head *io_slist = NULL; + struct hiraid_stream_io_list *new_io_node = NULL; + u16 temp_io_id = actual_id * HIRAID_MAX_STREAM_NUM + + tmp_stream->stream_id; - mutex_lock(&g_stream_operation_mutex); - ret = hiraid_io_list_operation(hdid, 0, 0, TYPE_DELETE_SINGLE_IO_LIST); - mutex_unlock(&g_stream_operation_mutex); - return ret; -} - -static u8 hiraid_delete_all_io_list(void) -{ - u8 ret; - - mutex_lock(&g_stream_operation_mutex); - ret = hiraid_io_list_operation(0, 0, 0, TYPE_DELETE_ALL_IO_LIST); - mutex_unlock(&g_stream_operation_mutex); - return ret; -} - -static u8 hiraid_add_io_to_list(struct hiraid_queue *submit_queue, - struct HIRAID_STREAM_S *tmp_stream, struct hiraid_scsi_io_cmd io_cmd, - unsigned int sector_size, u16 actual_id) -{ - struct mutex_list_head_s *io_slist = NULL; - struct IO_LIST_S *new_io_node = NULL; - - new_io_node = kmalloc(sizeof(struct IO_LIST_S), GFP_KERNEL); + new_io_node = kmalloc(sizeof(struct hiraid_stream_io_list), GFP_KERNEL); if (!new_io_node) return 0; new_io_node->io_cmd = io_cmd; new_io_node->submit_queue = submit_queue; new_io_node->sector_size = sector_size; - io_slist = hiraid_get_io_head(actual_id * - MAX_STREAM_NUM + tmp_stream->stream_id); + io_slist = &hiraid_io_heads_per_stream[temp_io_id]; mutex_lock(&io_slist->lock); INIT_LIST_HEAD(&(new_io_node->list)); - list_add_tail(&(new_io_node->list), &io_slist->list); + list_add_tail(&(new_io_node->list), + &hiraid_io_heads_per_stream[temp_io_id].list); mutex_unlock(&io_slist->lock); return 1; } @@ -1146,10 +977,8 @@ static void hiraid_submit_io_threading(struct hiraid_dev *hdev) int i = 0; while (!kthread_should_stop()) { - mutex_lock(&g_stream_operation_mutex); - for (i = 0; i < MAX_PD_NUM; i++) + for (i = 0; i < HIRAID_MAX_PD_NUM; i++) hiraid_submit_io_stream(i, hdev); - mutex_unlock(&g_stream_operation_mutex); usleep_range(MIN_IO_SEND_TIME, MAX_IO_SEND_TIME); } } @@ -1158,27 +987,34 @@ static void hiraid_destroy_io_stream_resource(struct hiraid_dev *hdev) { u16 i; - for (i = 0; i < (MAX_PD_NUM * MAX_STREAM_NUM); i++) - list_del_init(&hiraid_get_io_head(i)->list); + for (i = 0; i < (HIRAID_MAX_PD_NUM * HIRAID_MAX_STREAM_NUM); i++) { + list_del_init(&hiraid_io_heads_per_stream[i].list); + mutex_destroy(&hiraid_io_heads_per_stream[i].lock); + } } -struct task_struct *g_hiraid_submit_task; static void hiraid_init_io_stream(struct hiraid_dev *hdev) { - hiraid_io_recognition_init(); - g_hiraid_submit_task = kthread_run((void *)hiraid_submit_io_threading, - hdev, "hiraid_submit_thread"); + hiraid_io_recognition_init(hdev); + hdev->hiraid_stream_submit_task = + kthread_run((void *)hiraid_submit_io_threading, + hdev, "hiraid_stream_submit_task"); } #define HIRAID_RW_FUA BIT(14) +#define RW_LENGTH_ZERO (67) static int hiraid_setup_rw_cmd(struct hiraid_dev *hdev, struct hiraid_scsi_rw_cmd *io_cmd, - struct scsi_cmnd *scmd) + struct scsi_cmnd *scmd, + struct hiraid_mapmange *mapbuf) { + u32 ret = 0; u32 start_lba_lo, start_lba_hi; u32 datalength = 0; u16 control = 0; + struct scsi_device *sdev = scmd->device; + u32 buf_len = cpu_to_le32(scsi_bufflen(scmd)); start_lba_lo = 0; start_lba_hi = 0; @@ -1187,6 +1023,8 @@ static int hiraid_setup_rw_cmd(struct hiraid_dev *hdev, io_cmd->opcode = HIRAID_CMD_WRITE; } else if (scmd->sc_data_direction == DMA_FROM_DEVICE) { io_cmd->opcode = HIRAID_CMD_READ; + } else if (scmd->sc_data_direction == DMA_NONE) { + ret = RW_LENGTH_ZERO; } else { dev_err(hdev->dev, "invalid RW_IO for unsupported data direction[%d]\n", scmd->sc_data_direction); @@ -1194,6 +1032,9 @@ static int hiraid_setup_rw_cmd(struct hiraid_dev *hdev, return -EINVAL; } + if (ret == RW_LENGTH_ZERO) + return ret; + /* 6-byte READ(0x08) or WRITE(0x0A) cdb */ if (scmd->cmd_len == 6) { datalength = (u32)(scmd->cmnd[4] == 0 ? @@ -1230,18 +1071,32 @@ static int hiraid_setup_rw_cmd(struct hiraid_dev *hdev, control |= HIRAID_RW_FUA; } - if (unlikely(datalength > U16_MAX || datalength == 0)) { - dev_err(hdev->dev, "invalid IO for err trans data length[%u]\n", + if (unlikely(datalength > U16_MAX)) { + dev_err(hdev->dev, "invalid IO for illegal transfer data length[%u]\n", datalength); WARN_ON(1); return -EINVAL; } + if (unlikely(datalength == 0)) + return RW_LENGTH_ZERO; + io_cmd->slba = cpu_to_le64(((u64)start_lba_hi << 32) | start_lba_lo); /* 0base for nlb */ io_cmd->nlb = cpu_to_le16((u16)(datalength - 1)); io_cmd->control = cpu_to_le16(control); + mapbuf->cdb_data_len = (u32)((io_cmd->nlb + 1) * sdev->sector_size); + if (mapbuf->cdb_data_len > buf_len) { + /* return DID_ERROR */ + dev_err(hdev->dev, "error: buf len[0x%x] is smaller than actual length[0x%x] sectorsize[0x%x]\n", + buf_len, mapbuf->cdb_data_len, sdev->sector_size); + return -EINVAL; + } else if (mapbuf->cdb_data_len < buf_len) { + dev_warn(hdev->dev, "warn: buf_len[0x%x] cdb_data_len[0x%x] nlb[0x%x] sectorsize[0x%x]\n", + buf_len, mapbuf->cdb_data_len, + io_cmd->nlb, sdev->sector_size); + } return 0; } @@ -1296,14 +1151,17 @@ static bool hiraid_disk_is_hdd_rawdrive(u8 attr) } static int hiraid_setup_io_cmd(struct hiraid_dev *hdev, - struct hiraid_scsi_io_cmd *io_cmd, - struct scsi_cmnd *scmd) + struct hiraid_scsi_io_cmd *io_cmd, struct scsi_cmnd *scmd, + struct hiraid_mapmange *mapbuf) { memcpy(io_cmd->common.cdb, scmd->cmnd, scmd->cmd_len); io_cmd->common.cdb_len = scmd->cmd_len; + /* init cdb_data_len */ + mapbuf->cdb_data_len = cpu_to_le32(scsi_bufflen(scmd)); + if (hiraid_is_rw_scmd(scmd)) - return hiraid_setup_rw_cmd(hdev, &io_cmd->rw, scmd); + return hiraid_setup_rw_cmd(hdev, &io_cmd->rw, scmd, mapbuf); else return hiraid_setup_nonrw_cmd(hdev, &io_cmd->nonrw, scmd); } @@ -1360,7 +1218,7 @@ static int hiraid_io_map_data(struct hiraid_dev *hdev, ret = scsi_dma_map(scmd); if (unlikely(ret < 0)) - return ret; + return SCSI_MLQUEUE_HOST_BUSY; mapbuf->sge_cnt = ret; /* No data to DMA, it may be scsi no-rw command */ @@ -1390,7 +1248,12 @@ static void hiraid_check_status(struct hiraid_mapmange *mapbuf, struct scsi_cmnd *scmd, struct hiraid_completion *cqe) { - scsi_set_resid(scmd, 0); + u32 datalength = cpu_to_le32(scsi_bufflen(scmd)); + + if (datalength > mapbuf->cdb_data_len) + scsi_set_resid(scmd, datalength - mapbuf->cdb_data_len); + else + scsi_set_resid(scmd, 0); switch ((le16_to_cpu(cqe->status) >> 1) & 0x7f) { case SENSE_STATE_OK: @@ -1443,6 +1306,35 @@ static inline void hiraid_query_scmd_tag(struct scsi_cmnd *scmd, u16 *qid, *cid = blk_mq_unique_tag_to_tag(tag); } +static void hiraid_set_ncq_prio(struct scsi_cmnd *scmd, + struct hiraid_scsi_io_cmd *io_cmd, + struct hiraid_sdev_hostdata *hostdata, + struct hiraid_dev *hdev) +{ + struct request *rq = scmd->request; + int class = 0; + + if (hostdata->sata_ncq_prio_enable) { + class = IOPRIO_PRIO_CLASS(req_get_ioprio(rq)); + switch (class) { + case IOPRIO_CLASS_NONE: + break; + case IOPRIO_CLASS_RT: + io_cmd->rw.rsvd2 &= SCSI_RSVD_INIT_SHIFT; + io_cmd->rw.rsvd2 |= (1 << IOPRIO_CLASS_RT); + break; + case IOPRIO_CLASS_BE: + break; + case IOPRIO_CLASS_IDLE: + break; + default: + break; + } + } + dev_log_dbg(hdev->dev, "blk_ncq_prio = %d, scmd_ncq_prio = 0x%02x\n", + class, io_cmd->rw.rsvd2); +} + static int hiraid_queue_command(struct Scsi_Host *shost, struct scsi_cmnd *scmd) { @@ -1452,7 +1344,7 @@ static int hiraid_queue_command(struct Scsi_Host *shost, struct hiraid_sdev_hostdata *hostdata; struct hiraid_scsi_io_cmd io_cmd; struct hiraid_queue *ioq; - struct HIRAID_STREAM_S *tmp_stm = NULL; + struct hiraid_stream *tmp_stream; u16 hwq, cid; int ret; @@ -1482,9 +1374,16 @@ static int hiraid_queue_command(struct Scsi_Host *shost, io_cmd.rw.hdid = cpu_to_le32(hostdata->hdid); io_cmd.rw.cmd_id = cpu_to_le16(cid); - ret = hiraid_setup_io_cmd(hdev, &io_cmd, scmd); + hiraid_set_ncq_prio(scmd, &io_cmd, hostdata, hdev); + + ret = hiraid_setup_io_cmd(hdev, &io_cmd, scmd, mapbuf); if (unlikely(ret)) { - set_host_byte(scmd, DID_ERROR); + if (ret == RW_LENGTH_ZERO) { + scsi_set_resid(scmd, scsi_bufflen(scmd)); + set_host_byte(scmd, DID_OK); + } else { + set_host_byte(scmd, DID_ERROR); + } scmd->scsi_done(scmd); atomic_dec(&ioq->inflight); return 0; @@ -1507,6 +1406,10 @@ static int hiraid_queue_command(struct Scsi_Host *shost, mapbuf->cid = cid; ret = hiraid_io_map_data(hdev, mapbuf, scmd, &io_cmd); if (unlikely(ret)) { + if (ret == SCSI_MLQUEUE_HOST_BUSY) { + atomic_dec(&ioq->inflight); + return ret; + } dev_err(hdev->dev, "io map data err\n"); set_host_byte(scmd, DID_ERROR); scmd->scsi_done(scmd); @@ -1515,20 +1418,20 @@ static int hiraid_queue_command(struct Scsi_Host *shost, } WRITE_ONCE(mapbuf->state, CMD_FLIGHT); - if (hiraid_is_rw_scmd(scmd) && hiraid_disk_is_hdd_rawdrive(hostdata->attr)) { - if (hiraid_detect_if_aging()) + if (hiraid_aging_detect(hdev)) hiraid_aging(hdev); - tmp_stm = hiraid_stream_detect(hdev, &(io_cmd.rw), sdev->id); - if (tmp_stm != NULL) { - hiraid_io_recognition_iterator(tmp_stm, 1); - if (hiraid_recognition_acknowledge(tmp_stm) && - (hiraid_get_stream_num(sdev->id) > 1)) { - if (hiraid_add_io_to_list(ioq, - tmp_stm, io_cmd, sdev->sector_size, sdev->id)) { + tmp_stream = hiraid_stream_detect(hdev, &(io_cmd.rw), sdev->id); + if (tmp_stream != NULL) { + hiraid_io_recognition_iterator(tmp_stream, 1); + if (hiraid_recognition_acknowledge(tmp_stream) && + (sdev->id < HIRAID_MAX_PD_NUM) && + (hdev->hiraid_stream_io_num_per_pd[sdev->id][HIRAID_STREAM_TYPE_TOTAL] > + 1)) { + if (hiraid_add_io_to_list(hdev, ioq, tmp_stream, + io_cmd, sdev->sector_size, sdev->id)) return 0; - } } } } @@ -1666,6 +1569,7 @@ static int hiraid_slave_configure(struct scsi_device *sdev) static void hiraid_shost_init(struct hiraid_dev *hdev) { struct pci_dev *pdev = hdev->pdev; + struct scsi_host_template *hostt; u8 domain, bus; u32 dev_func; @@ -1674,8 +1578,10 @@ static void hiraid_shost_init(struct hiraid_dev *hdev) dev_func = pdev->devfn; hdev->shost->nr_hw_queues = work_mode ? 1 : hdev->online_queues - 1; - hdev->shost->can_queue = hdev->scsi_qd; - + hdev->shost->can_queue = min(hdev->scsi_qd, + le16_to_cpu(hdev->ctrl_info->max_cmds) - + HIRAID_AQ_DEPTH - + HIRAID_TOTAL_PTCMDS(hdev->online_queues - 1)); hdev->shost->sg_tablesize = le16_to_cpu(hdev->ctrl_info->max_num_sge); /* 512B per sector */ hdev->shost->max_sectors = @@ -1689,7 +1595,13 @@ static void hiraid_shost_init(struct hiraid_dev *hdev) hdev->shost->this_id = -1; hdev->shost->unique_id = (domain << 16) | (bus << 8) | dev_func; hdev->shost->max_cmd_len = MAX_CDB_LEN; - hdev->shost->hostt->cmd_size = hiraid_get_max_cmd_size(hdev); + + hostt = (struct scsi_host_template *)hdev->shost->hostt; + hostt->cmd_size = hiraid_get_max_cmd_size(hdev); + + dev_info(hdev->dev, "nr_hw_queues[%u] can_queue[%u] unique_id[%u] cmd_size[%u]\n", + hdev->shost->nr_hw_queues, hdev->shost->can_queue, + hdev->shost->unique_id, hdev->shost->hostt->cmd_size); } static int hiraid_alloc_queue(struct hiraid_dev *hdev, u16 qid, u16 depth) @@ -2505,6 +2417,14 @@ static int hiraid_get_queue_cnt(struct hiraid_dev *hdev, u32 *cnt) return 0; } +static bool hiraid_pci_is_present(struct hiraid_dev *hdev) +{ + if (!pci_device_is_present(hdev->pdev) || + readl(hdev->bar + HIRAID_REG_VS) == 0Xffffffff) + return false; + return true; +} + static int hiraid_setup_io_queues(struct hiraid_dev *hdev) { struct hiraid_queue *adminq = &hdev->queues[0]; @@ -2579,7 +2499,7 @@ static void hiraid_delete_io_queues(struct hiraid_dev *hdev) u8 opcode = HIRAID_ADMIN_DELETE_SQ; u16 i, pass; - if (!pci_device_is_present(hdev->pdev)) { + if (!hiraid_pci_is_present(hdev)) { dev_err(hdev->dev, "pci_device is not present, skip disable io queues\n"); return; } @@ -2619,7 +2539,7 @@ static void hiraid_disable_admin_queue(struct hiraid_dev *hdev, struct hiraid_queue *adminq = &hdev->queues[0]; u16 start, end; - if (pci_device_is_present(hdev->pdev)) { + if (hiraid_pci_is_present(hdev)) { if (shutdown) hiraid_shutdown_control(hdev); else @@ -2850,11 +2770,16 @@ static int hiraid_luntarget_sort(const void *l, const void *r) const struct hiraid_dev_info *rn = r; int l_attr = HIRAID_DEV_INFO_ATTR_BOOT(ln->attr); int r_attr = HIRAID_DEV_INFO_ATTR_BOOT(rn->attr); - /* boot first */ if (l_attr != r_attr) return (r_attr - l_attr); + l_attr = HIRAID_DEV_INFO_ATTR_VD(ln->attr); + r_attr = HIRAID_DEV_INFO_ATTR_VD(rn->attr); + /* vd second */ + if (l_attr != r_attr) + return (r_attr - l_attr); + if (ln->channel == rn->channel) return le16_to_cpu(ln->target) - le16_to_cpu(rn->target); @@ -2892,6 +2817,19 @@ static void hiraid_scan_work(struct work_struct *work) if (HIRAID_DEV_INFO_FLAG_VALID(flag)) { if (!HIRAID_DEV_INFO_FLAG_VALID(org_flag)) { + down_write(&hdev->dev_rwsem); + memcpy(&old_dev[i], &dev[i], + sizeof(struct hiraid_dev_info)); + memcpy(&new_dev[count++], &dev[i], + sizeof(struct hiraid_dev_info)); + up_write(&hdev->dev_rwsem); + } else if (HIRAID_DEV_INFO_FLAG_DELETE_OR_ADD(flag, + org_flag)) { + down_write(&hdev->dev_rwsem); + old_dev[i].flag &= 0xfe; + up_write(&hdev->dev_rwsem); + hiraid_delete_device(hdev, &old_dev[i]); + down_write(&hdev->dev_rwsem); memcpy(&old_dev[i], &dev[i], sizeof(struct hiraid_dev_info)); @@ -3044,7 +2982,7 @@ static void hiraid_bsg_buf_unmap(struct hiraid_dev *hdev, struct bsg_job *job) } static int hiraid_bsg_buf_map(struct hiraid_dev *hdev, struct bsg_job *job, - struct hiraid_admin_command *cmd) + struct hiraid_admin_command *cmd, u16 qid) { struct hiraid_bsg_request *bsg_req = job->request; struct request *rq = blk_mq_rq_from_pdu(job); @@ -3058,24 +2996,26 @@ static int hiraid_bsg_buf_map(struct hiraid_dev *hdev, struct bsg_job *job, mapbuf->sgl = job->request_payload.sg_list; mapbuf->len = job->request_payload.payload_len; mapbuf->page_cnt = -1; + mapbuf->hiraidq = &hdev->queues[qid]; if (unlikely(mapbuf->sge_cnt == 0)) goto out; - mapbuf->use_sgl = !hiraid_is_prp(hdev, mapbuf->sgl, mapbuf->sge_cnt); - ret = dma_map_sg_attrs(hdev->dev, mapbuf->sgl, mapbuf->sge_cnt, - dma_dir, DMA_ATTR_NO_WARN); + dma_dir, DMA_ATTR_NO_WARN); if (!ret) goto out; - if ((mapbuf->use_sgl == (bool)true) && + mapbuf->use_sgl = !hiraid_is_prp(hdev, mapbuf->sgl, mapbuf->sge_cnt); + + if ((mapbuf->use_sgl == true) && (bsg_req->msgcode == HIRAID_BSG_IOPTHRU) && - (hdev->ctrl_info->pt_use_sgl != (bool)false)) { - ret = hiraid_build_passthru_sgl(hdev, cmd, mapbuf); + (hdev->ctrl_info->pt_use_sgl != false)) { + ret = hiraid_build_sgl(hdev, (struct hiraid_scsi_io_cmd *)cmd, + mapbuf); } else { mapbuf->use_sgl = false; - ret = hiraid_build_passthru_prp(hdev, mapbuf); + ret = hiraid_build_prp(hdev, mapbuf); cmd->common.dptr.prp1 = cpu_to_le64(sg_dma_address(mapbuf->sgl)); cmd->common.dptr.prp2 = cpu_to_le64(mapbuf->first_dma); @@ -3156,9 +3096,7 @@ static int hiraid_init_control_info(struct hiraid_dev *hdev) if (hdev->ctrl_info->asynevent > HIRAID_ASYN_COMMANDS) hdev->ctrl_info->asynevent = HIRAID_ASYN_COMMANDS; - hdev->scsi_qd = work_mode ? - le16_to_cpu(hdev->ctrl_info->max_cmds) : - (hdev->ioq_depth - HIRAID_PTHRU_CMDS_PERQ); + hdev->scsi_qd = hdev->ioq_depth - HIRAID_PTHRU_CMDS_PERQ; return 0; } @@ -3191,7 +3129,7 @@ static int hiraid_user_send_admcmd(struct hiraid_dev *hdev, struct bsg_job *job) admin_cmd.common.cdw14 = cpu_to_le32(ptcmd->cdw14); admin_cmd.common.cdw15 = cpu_to_le32(ptcmd->cdw15); - status = hiraid_bsg_buf_map(hdev, job, &admin_cmd); + status = hiraid_bsg_buf_map(hdev, job, &admin_cmd, 0); if (status) { dev_err(hdev->dev, "err, map data failed\n"); return status; @@ -3251,19 +3189,13 @@ static void hiraid_free_io_ptcmds(struct hiraid_dev *hdev) } static int hiraid_put_io_sync_request(struct hiraid_dev *hdev, - struct hiraid_scsi_io_cmd *io_cmd, - u32 *result, u32 *reslen, u32 timeout) + struct hiraid_scsi_io_cmd *io_cmd, struct hiraid_cmd *pt_cmd, + u32 *result, u32 *reslen, u32 timeout) { int ret; dma_addr_t buffer_phy; struct hiraid_queue *ioq; void *sense_addr = NULL; - struct hiraid_cmd *pt_cmd = hiraid_get_cmd(hdev, HIRAID_CMD_PTHRU); - - if (!pt_cmd) { - dev_err(hdev->dev, "err, get ioq cmd failed\n"); - return -EFAULT; - } timeout = timeout ? timeout : ADMIN_TIMEOUT; @@ -3271,8 +3203,9 @@ static int hiraid_put_io_sync_request(struct hiraid_dev *hdev, ioq = &hdev->queues[pt_cmd->qid]; if (work_mode) { - ret = ((pt_cmd->qid - 1) * HIRAID_PTHRU_CMDS_PERQ + - pt_cmd->cid) * SCSI_SENSE_BUFFERSIZE; + ret = ((pt_cmd->qid - 1) * + HIRAID_PTHRU_CMDS_PERQ + pt_cmd->cid) * + SCSI_SENSE_BUFFERSIZE; sense_addr = hdev->sense_buffer_virt + ret; buffer_phy = hdev->sense_buffer_phy + ret; } else { @@ -3293,8 +3226,6 @@ static int hiraid_put_io_sync_request(struct hiraid_dev *hdev, (le32_to_cpu(io_cmd->common.cdw3[0]) & 0xffff)); hiraid_admin_timeout(hdev, pt_cmd); - - hiraid_put_cmd(hdev, pt_cmd, HIRAID_CMD_PTHRU); return -ETIME; } @@ -3305,8 +3236,6 @@ static int hiraid_put_io_sync_request(struct hiraid_dev *hdev, } } - hiraid_put_cmd(hdev, pt_cmd, HIRAID_CMD_PTHRU); - return pt_cmd->status; } @@ -3316,6 +3245,7 @@ static int hiraid_user_send_ptcmd(struct hiraid_dev *hdev, struct bsg_job *job) (struct hiraid_bsg_request *)(job->request); struct hiraid_passthru_io_cmd *cmd = &(bsg_req->pthrucmd); struct hiraid_scsi_io_cmd pthru_cmd; + struct hiraid_cmd *pt_cmd; int status = 0; u32 timeout = msecs_to_jiffies(cmd->timeout_ms); // data len is 4k before use sgl, now len is 1M @@ -3358,16 +3288,21 @@ static int hiraid_user_send_ptcmd(struct hiraid_dev *hdev, struct bsg_job *job) pthru_cmd.common.cdw26[2] = cpu_to_le32(cmd->cdw26[2]); pthru_cmd.common.cdw26[3] = cpu_to_le32(cmd->cdw26[3]); + pt_cmd = hiraid_get_cmd(hdev, HIRAID_CMD_PTHRU); + if (pt_cmd == NULL) { + dev_err(hdev->dev, "err, get ioq cmd failed\n"); + return -EFAULT; + } + status = hiraid_bsg_buf_map(hdev, job, - (struct hiraid_admin_command *)&pthru_cmd); + (struct hiraid_admin_command *)&pthru_cmd, pt_cmd->qid); if (status) { dev_err(hdev->dev, "err, map data failed\n"); - return status; + goto ret; } - status = hiraid_put_io_sync_request(hdev, &pthru_cmd, job->reply, - &job->reply_len, timeout); - + status = hiraid_put_io_sync_request(hdev, &pthru_cmd, pt_cmd, + job->reply, &job->reply_len, timeout); if (status) dev_info(hdev->dev, "opcode[0x%x] subopcode[0x%x] status[0x%x] replylen[%d]\n", cmd->opcode, cmd->info_1.subopcode, @@ -3375,6 +3310,9 @@ static int hiraid_user_send_ptcmd(struct hiraid_dev *hdev, struct bsg_job *job) hiraid_bsg_buf_unmap(hdev, job); +ret: + hiraid_put_cmd(hdev, pt_cmd, HIRAID_CMD_PTHRU); + return status; } @@ -3396,23 +3334,28 @@ static bool hiraid_check_scmd_finished(struct scsi_cmnd *scmd) return false; } -static enum blk_eh_timer_return hiraid_timed_out(struct scsi_cmnd *scmd) +#define EH_TIMED_RET_TP enum blk_eh_timer_return +#define EH_TIMED_RET_DONE BLK_EH_DONE +#define EH_TIMED_RET_NOT_HANDLED BLK_EH_DONE +#define EH_TIMED_RET_RESET_TIMER BLK_EH_RESET_TIMER +static EH_TIMED_RET_TP hiraid_timed_out(struct scsi_cmnd *scmd) { struct hiraid_mapmange *mapbuf = scsi_cmd_priv(scmd); unsigned int timeout = scmd->device->request_queue->rq_timeout; if (hiraid_check_scmd_finished(scmd)) - goto out; + return EH_TIMED_RET_DONE; - if (time_after(jiffies, scmd->jiffies_at_alloc + timeout)) { + if (time_after_eq(jiffies, scmd->jiffies_at_alloc + timeout)) { + scmd_printk(KERN_WARNING, scmd, "timeout is work[%d], need abort\n", + mapbuf->state); if (cmpxchg(&mapbuf->state, CMD_FLIGHT, CMD_TIMEOUT) == CMD_FLIGHT) - return BLK_EH_DONE; + return EH_TIMED_RET_NOT_HANDLED; } -out: - return BLK_EH_RESET_TIMER; -} + return EH_TIMED_RET_RESET_TIMER; +} /* send abort command by admin queue temporary */ static int hiraid_send_abort_cmd(struct hiraid_dev *hdev, u32 hdid, u16 qid, u16 cid) @@ -3535,7 +3478,7 @@ static int hiraid_dev_disable(struct hiraid_dev *hdev, bool shutdown) struct hiraid_queue *adminq = &hdev->queues[0]; u16 start, end; - if (pci_device_is_present(hdev->pdev)) { + if (hiraid_pci_is_present(hdev)) { if (shutdown) hiraid_shutdown_control(hdev); else @@ -3733,11 +3676,19 @@ static int hiraid_abort(struct scsi_cmnd *scmd) hiraid_check_scmd_finished(scmd)) return SUCCESS; + if (hdev->state != DEV_LIVE) + return SUCCESS; + hostdata = scmd->device->hostdata; cid = mapbuf->cid; hwq = mapbuf->hiraidq->qid; - if (hiraid_check_io_list(hostdata->hdid, cid, hwq)) { + mutex_lock(&g_stream_operation_mutex); + ret = hiraid_stream_delete_io_list(hdev, hostdata->hdid, + cid, hwq, HIRAID_STREAM_TYPE_DELETE_SINGLE_IO); + mutex_unlock(&g_stream_operation_mutex); + + if (ret == 1) { dev_warn(hdev->dev, "find cid[%d] qid[%d] in host, abort succ\n", cid, hwq); return SUCCESS; @@ -3749,14 +3700,14 @@ static int hiraid_abort(struct scsi_cmnd *scmd) ret = hiraid_wait_io_completion(mapbuf); if (ret) { dev_warn(hdev->dev, "cid[%d] qid[%d] abort failed, not found\n", - cid, hwq); + cid, hwq); return FAILED; } dev_warn(hdev->dev, "cid[%d] qid[%d] abort succ\n", cid, hwq); return SUCCESS; } dev_warn(hdev->dev, "cid[%d] qid[%d] abort failed, timeout\n", - cid, hwq); + cid, hwq); return FAILED; } @@ -3778,7 +3729,10 @@ static int hiraid_scsi_reset(struct scsi_cmnd *scmd, enum hiraid_rst_type rst) if ((ret == 0) || (ret == FW_EH_DEV_NONE && rst == HIRAID_RESET_TARGET)) { if (rst == HIRAID_RESET_TARGET) { - hiraid_delete_single_pd_io_list(hostdata->hdid); + mutex_lock(&g_stream_operation_mutex); + hiraid_stream_delete_io_list(hdev, hostdata->hdid, + 0, 0, HIRAID_STREAM_TYPE_DELETE_SINGLE_IO_LIST); + mutex_unlock(&g_stream_operation_mutex); ret = wait_tgt_reset_io_done(scmd); if (ret) { dev_warn(hdev->dev, "sdev[%d:%d] target has %d peding cmd, target reset failed\n", @@ -3818,9 +3772,13 @@ static int hiraid_host_reset(struct scsi_cmnd *scmd) dev_warn(hdev->dev, "sdev[%d:%d] send host reset\n", scmd->device->channel, scmd->device->id); - hiraid_delete_all_io_list(); - if (hiraid_reset_work_sync(hdev) == -EBUSY) + mutex_lock(&g_stream_operation_mutex); + hiraid_stream_delete_io_list(hdev, 0, 0, 0, + HIRAID_STREAM_TYPE_DELETE_ALL_IO_LIST); + mutex_unlock(&g_stream_operation_mutex); + if (hiraid_reset_work_sync(hdev) == -EBUSY) { flush_work(&hdev->reset_work); + } if (hdev->state != DEV_LIVE) { dev_warn(hdev->dev, "sdev[%d:%d] host reset failed\n", @@ -3852,7 +3810,10 @@ static pci_ers_result_t hiraid_pci_error_detected(struct pci_dev *pdev, scsi_block_requests(hdev->shost); hiraid_dev_state_trans(hdev, DEV_RESETTING); - hiraid_delete_all_io_list(); + mutex_lock(&g_stream_operation_mutex); + hiraid_stream_delete_io_list(hdev, 0, 0, 0, + HIRAID_STREAM_TYPE_DELETE_ALL_IO_LIST); + mutex_unlock(&g_stream_operation_mutex); return PCI_ERS_RESULT_NEED_RESET; case pci_channel_io_perm_failure: @@ -3888,9 +3849,24 @@ static void hiraid_reset_pci_finish(struct pci_dev *pdev) { struct hiraid_dev *hdev = pci_get_drvdata(pdev); + if (hiraid_reset_work_sync(hdev) == -EBUSY) + flush_work(&hdev->reset_work); + dev_info(hdev->dev, "enter hiraid reset finish\n"); } +static void hiraid_reset_pci_prepare(struct pci_dev *pdev) +{ + struct hiraid_dev *hdev = pci_get_drvdata(pdev); + + mutex_lock(&g_stream_operation_mutex); + hiraid_stream_delete_io_list(hdev, 0, 0, 0, + HIRAID_STREAM_TYPE_DELETE_ALL_IO_LIST); + mutex_unlock(&g_stream_operation_mutex); + msleep(100); + dev_info(hdev->dev, "exit hiraid reset prepare\n"); +} + static ssize_t csts_pp_show(struct device *cdev, struct device_attribute *attr, char *buf) { @@ -3898,7 +3874,7 @@ static ssize_t csts_pp_show(struct device *cdev, struct hiraid_dev *hdev = shost_priv(shost); int ret = -1; - if (pci_device_is_present(hdev->pdev)) { + if (hiraid_pci_is_present(hdev)) { ret = (readl(hdev->bar + HIRAID_REG_CSTS) & HIRAID_CSTS_PP_MASK); ret >>= HIRAID_CSTS_PP_SHIFT; @@ -3914,7 +3890,7 @@ static ssize_t csts_shst_show(struct device *cdev, struct hiraid_dev *hdev = shost_priv(shost); int ret = -1; - if (pci_device_is_present(hdev->pdev)) { + if (hiraid_pci_is_present(hdev)) { ret = (readl(hdev->bar + HIRAID_REG_CSTS) & HIRAID_CSTS_SHST_MASK); ret >>= HIRAID_CSTS_SHST_SHIFT; @@ -3930,7 +3906,7 @@ static ssize_t csts_cfs_show(struct device *cdev, struct hiraid_dev *hdev = shost_priv(shost); int ret = -1; - if (pci_device_is_present(hdev->pdev)) { + if (hiraid_pci_is_present(hdev)) { ret = (readl(hdev->bar + HIRAID_REG_CSTS) & HIRAID_CSTS_CFS_MASK); ret >>= HIRAID_CSTS_CFS_SHIFT; @@ -3946,7 +3922,7 @@ static ssize_t csts_rdy_show(struct device *cdev, struct hiraid_dev *hdev = shost_priv(shost); int ret = -1; - if (pci_device_is_present(hdev->pdev)) + if (hiraid_pci_is_present(hdev)) ret = (readl(hdev->bar + HIRAID_REG_CSTS) & HIRAID_CSTS_RDY); return snprintf(buf, PAGE_SIZE, "%d\n", ret); @@ -4239,22 +4215,78 @@ static ssize_t dispatch_hwq_store(struct device *dev, return strlen(buf); } +static ssize_t sata_ncq_prio_enable_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct hiraid_sdev_hostdata *hostdata; + + hostdata = to_scsi_device(dev)->hostdata; + return snprintf(buf, PAGE_SIZE, "%d\n", hostdata->sata_ncq_prio_enable); +} + +bool scsi_ncq_prio_support(struct scsi_device *sdev, struct hiraid_dev *hdev) +{ + unsigned char *buf; + bool ncq_prio_supp = false; + + if (!scsi_device_supports_vpd(sdev)) { + dev_err(hdev->dev, "scsi device not support vpd\n"); + return ncq_prio_supp; + } + + buf = kzalloc(HIRAID_SCSI_VPD_LEN, GFP_KERNEL); + if (!buf) + return ncq_prio_supp; + + if (!scsi_get_vpd_page(sdev, 0x89, buf, HIRAID_SCSI_VPD_LEN)) + ncq_prio_supp = (buf[NCQ_PRIO_SUPPORT_BYTE] >> + VPD_NCQ_PRIO_SUPPORT_BIT) & 1; + + kfree(buf); + return ncq_prio_supp; +} + +static ssize_t sata_ncq_prio_enable_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t count) +{ + struct scsi_device *sdev = to_scsi_device(dev); + struct hiraid_sdev_hostdata *hostdata = sdev->hostdata; + struct hiraid_dev *hdev = shost_priv(sdev->host); + bool ncq_prio_enable = 0; + + if (kstrtobool(buf, &ncq_prio_enable)) + return -EINVAL; + + if (!scsi_ncq_prio_support(sdev, hdev)) { + dev_err(hdev->dev, "sdev[%d:%d] not support ncq prio\n", + sdev->channel, sdev->id); + return -EINVAL; + } + + hostdata->sata_ncq_prio_enable = ncq_prio_enable; + + return strlen(buf); +} + static DEVICE_ATTR_RO(raid_level); static DEVICE_ATTR_RO(raid_state); static DEVICE_ATTR_RO(raid_resync); static DEVICE_ATTR_RW(dispatch_hwq); +static DEVICE_ATTR_RW(sata_ncq_prio_enable); static struct device_attribute *hiraid_dev_attrs[] = { &dev_attr_raid_state, &dev_attr_raid_level, &dev_attr_raid_resync, &dev_attr_dispatch_hwq, + &dev_attr_sata_ncq_prio_enable, NULL, }; static struct pci_error_handlers hiraid_err_handler = { .error_detected = hiraid_pci_error_detected, .slot_reset = hiraid_pci_slot_reset, + .reset_prepare = hiraid_reset_pci_prepare, .reset_done = hiraid_reset_pci_finish, }; @@ -4518,14 +4550,17 @@ static void hiraid_remove(struct pci_dev *pdev) dev_info(hdev->dev, "enter hiraid remove\n"); - kthread_stop(g_hiraid_submit_task); - hiraid_delete_all_io_list(); + kthread_stop(hdev->hiraid_stream_submit_task); + mutex_lock(&g_stream_operation_mutex); + hiraid_stream_delete_io_list(hdev, 0, 0, 0, + HIRAID_STREAM_TYPE_DELETE_ALL_IO_LIST); + mutex_unlock(&g_stream_operation_mutex); hiraid_destroy_io_stream_resource(hdev); hiraid_dev_state_trans(hdev, DEV_DELETING); flush_work(&hdev->reset_work); - if (!pci_device_is_present(pdev)) + if (!hiraid_pci_is_present(hdev)) hiraid_flush_running_cmds(hdev); hiraid_unregist_bsg(hdev); @@ -4548,8 +4583,12 @@ static void hiraid_remove(struct pci_dev *pdev) static const struct pci_device_id hiraid_hw_card_ids[] = { { PCI_DEVICE(PCI_VENDOR_ID_HUAWEI_LOGIC, HIRAID_SERVER_DEVICE_HBA_DID) }, + { PCI_DEVICE(PCI_VENDOR_ID_HUAWEI_LOGIC, + HIRAID_SERVER_DEVICE_HBAS_DID) }, { PCI_DEVICE(PCI_VENDOR_ID_HUAWEI_LOGIC, HIRAID_SERVER_DEVICE_RAID_DID) }, + { PCI_DEVICE(PCI_VENDOR_ID_HUAWEI_LOGIC, + HIRAID_SERVER_DEVICE_RAIDS_DID) }, { 0, } }; MODULE_DEVICE_TABLE(pci, hiraid_hw_card_ids); @@ -4572,22 +4611,9 @@ static int __init hiraid_init(void) if (!work_queue) return -ENOMEM; - hiraid_class = class_create(THIS_MODULE, "hiraid"); - if (IS_ERR(hiraid_class)) { - ret = PTR_ERR(hiraid_class); - goto destroy_wq; - } - ret = pci_register_driver(&hiraid_driver); if (ret < 0) - goto destroy_class; - - return 0; - -destroy_class: - class_destroy(hiraid_class); -destroy_wq: - destroy_workqueue(work_queue); + destroy_workqueue(work_queue); return ret; } @@ -4595,7 +4621,6 @@ destroy_wq: static void __exit hiraid_exit(void) { pci_unregister_driver(&hiraid_driver); - class_destroy(hiraid_class); destroy_workqueue(work_queue); } -- 2.45.1.windows.1

2 6

[PATCH OLK-6.6] KVM: SVM: Add missing save/restore handling of LBR MSRs
by Zhang Kunbo 05 Jun '26

05 Jun '26

From: Yosry Ahmed <yosry(a)kernel.org> stable inclusion from stable-v6.18.27 commit 2b922a42b531a82d7881add14a7698dcdc5e1f0a category: bugfix bugzilla: https://atomgit.com/src-openeuler/kernel/issues/15261 CVE: CVE-2026-46014 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit 3700f0788da6acf73b2df56690f4b201aa4aefd2 upstream. MSR_IA32_DEBUGCTLMSR and LBR MSRs are currently not enumerated by KVM_GET_MSR_INDEX_LIST, and LBR MSRs cannot be set with KVM_SET_MSRS. So save/restore is completely broken. Fix it by adding the MSRs to msrs_to_save_base, and allowing writes to LBR MSRs from userspace only (as they are read-only MSRs) if LBR virtualization is enabled. Additionally, to correctly restore L1's LBRs while L2 is running, make sure the LBRs are copied from the captured VMCB01 save area in svm_copy_vmrun_state(). Note, for VMX, this also fixes a flaw where MSR_IA32_DEBUGCTLMSR isn't reported as an MSR to save/restore. Note #2, over-reporting MSR_IA32_LASTxxx on Intel is ok, as KVM already handles unsupported reads and writes thanks to commit b5e2fec0ebc3 ("KVM: Ignore DEBUGCTL MSRs with no effect") (kvm_do_msr_access() will morph the unsupported userspace write into a nop). Fixes: 24e09cbf480a ("KVM: SVM: enable LBR virtualization") Cc: stable(a)vger.kernel.org Reported-by: Jim Mattson <jmattson(a)google.com> Signed-off-by: Yosry Ahmed <yosry(a)kernel.org> Link: https://patch.msgid.link/20260303003421.2185681-4-yosry@kernel.org [sean: guard with lbrv checks, massage changelog] Signed-off-by: Sean Christopherson <seanjc(a)google.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Conflicts: arch/x86/kvm/svm/nested.c arch/x86/kvm/svm/svm.c arch/x86/kvm/x86.c [context conflicts and a3f0981a5a0e ("KVM: nSVM: Avoid clearing VMCB_LBR in vmcb12") not merged, which delete the vmcb_mark_dirty from `svm_copy_lbrs`] Signed-off-by: Zhang Kunbo <zhangkunbo(a)huawei.com> --- arch/x86/kvm/svm/nested.c | 10 ++++++++++ arch/x86/kvm/svm/svm.c | 42 ++++++++++++++++++++++++++++++++++----- arch/x86/kvm/x86.c | 3 +++ 3 files changed, 50 insertions(+), 5 deletions(-) diff --git a/arch/x86/kvm/svm/nested.c b/arch/x86/kvm/svm/nested.c index 2cd9299b250e..146ab1ea390e 100644 --- a/arch/x86/kvm/svm/nested.c +++ b/arch/x86/kvm/svm/nested.c @@ -969,6 +969,16 @@ void svm_copy_vmrun_state(struct vmcb_save_area *to_save, to_save->rsp = from_save->rsp; to_save->rip = from_save->rip; to_save->cpl = 0; + + if (kvm_cpu_cap_has(X86_FEATURE_LBRV)) { + to_save->dbgctl = from_save->dbgctl; + to_save->br_from = from_save->br_from; + to_save->br_to = from_save->br_to; + to_save->last_excp_from = from_save->last_excp_from; + to_save->last_excp_to = from_save->last_excp_to; + + to_save->dbgctl &= ~DEBUGCTL_RESERVED_BITS; + } } void svm_copy_vmloadsave_state(struct vmcb *to_vmcb, struct vmcb *from_vmcb) diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c index 06d0d8706ba4..8a299fa2f47a 100644 --- a/arch/x86/kvm/svm/svm.c +++ b/arch/x86/kvm/svm/svm.c @@ -2941,19 +2941,19 @@ static int svm_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) msr_info->data = svm->tsc_aux; break; case MSR_IA32_DEBUGCTLMSR: - msr_info->data = svm_get_lbr_vmcb(svm)->save.dbgctl; + msr_info->data = lbrv ? svm_get_lbr_vmcb(svm)->save.dbgctl : 0; break; case MSR_IA32_LASTBRANCHFROMIP: - msr_info->data = svm_get_lbr_vmcb(svm)->save.br_from; + msr_info->data = lbrv ? svm_get_lbr_vmcb(svm)->save.br_from : 0; break; case MSR_IA32_LASTBRANCHTOIP: - msr_info->data = svm_get_lbr_vmcb(svm)->save.br_to; + msr_info->data = lbrv ? svm_get_lbr_vmcb(svm)->save.br_to : 0; break; case MSR_IA32_LASTINTFROMIP: - msr_info->data = svm_get_lbr_vmcb(svm)->save.last_excp_from; + msr_info->data = lbrv ? svm_get_lbr_vmcb(svm)->save.last_excp_from : 0; break; case MSR_IA32_LASTINTTOIP: - msr_info->data = svm_get_lbr_vmcb(svm)->save.last_excp_to; + msr_info->data = lbrv ? svm_get_lbr_vmcb(svm)->save.last_excp_to : 0; break; case MSR_VM_HSAVE_PA: msr_info->data = svm->nested.hsave_msr; @@ -3239,6 +3239,38 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr) vmcb_mark_dirty(svm->vmcb, VMCB_LBR); svm_update_lbrv(vcpu); break; + case MSR_IA32_LASTBRANCHFROMIP: + if (!lbrv) + return KVM_MSR_RET_UNSUPPORTED; + if (!msr->host_initiated) + return 1; + svm->vmcb->save.br_from = data; + vmcb_mark_dirty(svm->vmcb, VMCB_LBR); + break; + case MSR_IA32_LASTBRANCHTOIP: + if (!lbrv) + return KVM_MSR_RET_UNSUPPORTED; + if (!msr->host_initiated) + return 1; + svm->vmcb->save.br_to = data; + vmcb_mark_dirty(svm->vmcb, VMCB_LBR); + break; + case MSR_IA32_LASTINTFROMIP: + if (!lbrv) + return KVM_MSR_RET_UNSUPPORTED; + if (!msr->host_initiated) + return 1; + svm->vmcb->save.last_excp_from = data; + vmcb_mark_dirty(svm->vmcb, VMCB_LBR); + break; + case MSR_IA32_LASTINTTOIP: + if (!lbrv) + return KVM_MSR_RET_UNSUPPORTED; + if (!msr->host_initiated) + return 1; + svm->vmcb->save.last_excp_to = data; + vmcb_mark_dirty(svm->vmcb, VMCB_LBR); + break; case MSR_VM_HSAVE_PA: /* * Old kernels did not validate the value written to diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index b6c46697b023..8fbaa4216260 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -1524,6 +1524,9 @@ static const u32 msrs_to_save_base[] = { MSR_IA32_UMWAIT_CONTROL, MSR_IA32_XFD, MSR_IA32_XFD_ERR, + MSR_IA32_DEBUGCTLMSR, + MSR_IA32_LASTBRANCHFROMIP, MSR_IA32_LASTBRANCHTOIP, + MSR_IA32_LASTINTFROMIP, MSR_IA32_LASTINTTOIP, }; static const u32 msrs_to_save_pmu[] = { -- 2.34.1

2 1

[PATCH OLK-6.6 0/5] CVE-2026-46065
by Liu Kai 05 Jun '26

05 Jun '26

CVE-2026-46065 Liu Kai (2): fbdev: defio: fix KABI breakage in struct fb_deferred_io and struct fb_info fbdev: defio: Remove unused variable 'fbdefio' to fix build warning Thomas Zimmermann (3): fbdev/deferred-io: Clean up pageref on lastclose fbdev/deferred-io: Always call get_page() for framebuffer pages fbdev: defio: Disconnect deferred I/O from the lifetime of struct fb_info drivers/video/fbdev/core/fb_defio.c | 205 +++++++++++++++++++++------- include/linux/fb.h | 8 +- 2 files changed, 165 insertions(+), 48 deletions(-) -- 2.34.1

2 6

[PATCH OLK-5.10 0/3] CVE-2026-46065
by Liu Kai 05 Jun '26

05 Jun '26

CVE-2026-46065 Liu Kai (1): fbdev: defio: fix KABI breakage in struct fb_deferred_io and struct fb_info Thomas Zimmermann (2): fbdev: Refactor implementation of page_mkwrite fbdev: defio: Disconnect deferred I/O from the lifetime of struct fb_info drivers/video/fbdev/core/fb_defio.c | 184 ++++++++++++++++++++++++---- include/linux/fb.h | 3 +- include/linux/workqueue.h | 4 + 3 files changed, 166 insertions(+), 25 deletions(-) -- 2.34.1

2 4