- Kernel - mailweb.openeuler.org

[PATCH OLK-5.10] RDMA/hns: Fix DCA mmap area PUAF
by Chengchang Tang 06 Nov '24

06 Nov '24

driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IB2DWH ---------------------------------------------------------------------- DCA mmap area would be freed after driver reseted or hot unplugged. At this time, the vma is still valid, which may lead to a PUAF. Using rdma_user_mmap_io() to remap this area to ensure the vma is zapped after dirver removal. Fixes: d3caaebdbfe9 ("RDMA/hns: Optimize user DCA perfermance by sharing DCA status") Signed-off-by: Chengchang Tang <tangchengchang(a)huawei.com> Signed-off-by: Xinghai Cen <cenxinghai(a)h-partners.com> --- drivers/infiniband/hw/hns/hns_roce_dca.c | 3 +- drivers/infiniband/hw/hns/hns_roce_main.c | 39 ++++------------------- 2 files changed, 9 insertions(+), 33 deletions(-) diff --git a/drivers/infiniband/hw/hns/hns_roce_dca.c b/drivers/infiniband/hw/hns/hns_roce_dca.c index 91b6b65c83df..65ed7666a826 100644 --- a/drivers/infiniband/hw/hns/hns_roce_dca.c +++ b/drivers/infiniband/hw/hns/hns_roce_dca.c @@ -1181,7 +1181,8 @@ static void init_udca_status(struct hns_roce_ucontext *uctx, int udca_max_qps, return; ctx->dca_mmap_entry = hns_roce_user_mmap_entry_insert(ib_uctx, - (u64)kaddr, size, HNS_ROCE_MMAP_TYPE_DCA); + (u64)virt_to_phys(kaddr), size, + HNS_ROCE_MMAP_TYPE_DCA); if (!ctx->dca_mmap_entry) { free_pages_exact(kaddr, size); return; diff --git a/drivers/infiniband/hw/hns/hns_roce_main.c b/drivers/infiniband/hw/hns/hns_roce_main.c index 27b723a40abe..3872e777c74b 100644 --- a/drivers/infiniband/hw/hns/hns_roce_main.c +++ b/drivers/infiniband/hw/hns/hns_roce_main.c @@ -674,36 +674,6 @@ static void hns_roce_dealloc_ucontext(struct ib_ucontext *ibcontext) ida_free(&hr_dev->uar_ida.ida, (int)context->uar.logic_idx); } -static int mmap_dca(struct ib_ucontext *context, struct vm_area_struct *vma) -{ - struct hns_roce_ucontext *uctx = to_hr_ucontext(context); - struct hns_roce_dca_ctx *ctx = &uctx->dca_ctx; - struct page **pages; - unsigned long num; - int ret; - - if ((vma->vm_end - vma->vm_start != (ctx->status_npage * PAGE_SIZE) || - return -EINVAL; - - if (!(vma->vm_flags & VM_WRITE) || (vma->vm_flags & VM_EXEC)) - return -EPERM; - - if (!ctx->buf_status) - return -EOPNOTSUPP; - - pages = kcalloc(ctx->status_npage, sizeof(struct page *), GFP_KERNEL); - if (!pages) - return -ENOMEM; - - for (num = 0; num < ctx->status_npage; num++) - pages[num] = virt_to_page(ctx->buf_status + num * PAGE_SIZE); - - ret = vm_insert_pages(vma, vma->vm_start, pages, &num); - kfree(pages); - return ret; -} - static int hns_roce_mmap(struct ib_ucontext *uctx, struct vm_area_struct *vma) { struct hns_roce_dev *hr_dev = to_hr_dev(uctx->device); @@ -728,8 +698,13 @@ static int hns_roce_mmap(struct ib_ucontext *uctx, struct vm_area_struct *vma) prot = pgprot_device(vma->vm_page_prot); break; case HNS_ROCE_MMAP_TYPE_DCA: - ret = mmap_dca(uctx, vma); - goto out; + if (!(vma->vm_flags & VM_WRITE) || (vma->vm_flags & VM_EXEC)) { + ret = -EPERM; + goto out; + } + vma->vm_flags |= VM_DONTEXPAND; + prot = vma->vm_page_prot; + break; case HNS_ROCE_MMAP_TYPE_RESET: if (vma->vm_flags & (VM_WRITE | VM_EXEC)) { ret = -EINVAL; -- 2.33.0

2 1

[PATCH OLK-6.6] RDMA/hns: Fix DCA mmap area PUAF
by Chengchang Tang 06 Nov '24

06 Nov '24

driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IB2H4P ---------------------------------------------------------------------- DCA mmap area would be freed after driver reseted or hot unplugged. At this time, the vma is still valid, which may lead to a PUAF. Using rdma_user_mmap_io() to remap this area to ensure the vma is zapped after dirver removal. Fixes: d3caaebdbfe9 ("RDMA/hns: Optimize user DCA perfermance by sharing DCA status") Signed-off-by: Chengchang Tang <tangchengchang(a)huawei.com> Signed-off-by: Xinghai Cen <cenxinghai(a)h-partners.com> --- drivers/infiniband/hw/hns/hns_roce_dca.c | 3 +- drivers/infiniband/hw/hns/hns_roce_main.c | 39 ++++------------------- 2 files changed, 9 insertions(+), 33 deletions(-) diff --git a/drivers/infiniband/hw/hns/hns_roce_dca.c b/drivers/infiniband/hw/hns/hns_roce_dca.c index 49adf43261d6..4cef41591795 100644 --- a/drivers/infiniband/hw/hns/hns_roce_dca.c +++ b/drivers/infiniband/hw/hns/hns_roce_dca.c @@ -1132,7 +1132,8 @@ static void init_udca_status(struct hns_roce_ucontext *uctx, int udca_max_qps, return; ctx->dca_mmap_entry = hns_roce_user_mmap_entry_insert(ib_uctx, - (u64)kaddr, size, HNS_ROCE_MMAP_TYPE_DCA); + (u64)virt_to_phys(kaddr), size, + HNS_ROCE_MMAP_TYPE_DCA); if (!ctx->dca_mmap_entry) { free_pages_exact(kaddr, size); return; diff --git a/drivers/infiniband/hw/hns/hns_roce_main.c b/drivers/infiniband/hw/hns/hns_roce_main.c index 1302c58c3937..85223d25995d 100644 --- a/drivers/infiniband/hw/hns/hns_roce_main.c +++ b/drivers/infiniband/hw/hns/hns_roce_main.c @@ -644,36 +644,6 @@ static void hns_roce_dealloc_ucontext(struct ib_ucontext *ibcontext) ida_free(&hr_dev->uar_ida.ida, (int)context->uar.logic_idx); } -static int mmap_dca(struct ib_ucontext *context, struct vm_area_struct *vma) -{ - struct hns_roce_ucontext *uctx = to_hr_ucontext(context); - struct hns_roce_dca_ctx *ctx = &uctx->dca_ctx; - struct page **pages; - unsigned long num; - int ret; - - if ((vma->vm_end - vma->vm_start != (ctx->status_npage * PAGE_SIZE) || - return -EINVAL; - - if (!(vma->vm_flags & VM_WRITE) || (vma->vm_flags & VM_EXEC)) - return -EPERM; - - if (!ctx->buf_status) - return -EOPNOTSUPP; - - pages = kcalloc(ctx->status_npage, sizeof(struct page *), GFP_KERNEL); - if (!pages) - return -ENOMEM; - - for (num = 0; num < ctx->status_npage; num++) - pages[num] = virt_to_page(ctx->buf_status + num * PAGE_SIZE); - - ret = vm_insert_pages(vma, vma->vm_start, pages, &num); - kfree(pages); - return ret; -} - static int hns_roce_mmap(struct ib_ucontext *uctx, struct vm_area_struct *vma) { struct hns_roce_dev *hr_dev = to_hr_dev(uctx->device); @@ -703,8 +673,13 @@ static int hns_roce_mmap(struct ib_ucontext *uctx, struct vm_area_struct *vma) prot = pgprot_device(vma->vm_page_prot); break; case HNS_ROCE_MMAP_TYPE_DCA: - ret = mmap_dca(uctx, vma); - goto out; + if (!(vma->vm_flags & VM_WRITE) || (vma->vm_flags & VM_EXEC)) { + ret = -EPERM; + goto out; + } + vma->vm_flags |= VM_DONTEXPAND; + prot = vma->vm_page_prot; + break; case HNS_ROCE_MMAP_TYPE_RESET: if (vma->vm_flags & (VM_WRITE | VM_EXEC)) { ret = -EINVAL; -- 2.33.0

2 1

[PATCH OLK-6.6] drm/amd/display: Fix index out of bounds in degamma hardware format translation
by Liu Chuang 06 Nov '24

06 Nov '24

From: Srinivasan Shanmugam <srinivasan.shanmugam(a)amd.com> stable inclusion from stable-v6.6.55 commit 122e3a7a8c7bcbe3aacddd6103f67f9f36bed473 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IAYR9A CVE: CVE-2024-49894 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit b7e99058eb2e86aabd7a10761e76cae33d22b49f ] Fixes index out of bounds issue in `cm_helper_translate_curve_to_degamma_hw_format` function. The issue could occur when the index 'i' exceeds the number of transfer function points (TRANSFER_FUNC_POINTS). The fix adds a check to ensure 'i' is within bounds before accessing the transfer function points. If 'i' is out of bounds the function returns false to indicate an error. Reported by smatch: drivers/gpu/drm/amd/amdgpu/../display/dc/dcn10/dcn10_cm_common.c:594 cm_helper_translate_curve_to_degamma_hw_format() error: buffer overflow 'output_tf->tf_pts.red' 1025 <= s32max drivers/gpu/drm/amd/amdgpu/../display/dc/dcn10/dcn10_cm_common.c:595 cm_helper_translate_curve_to_degamma_hw_format() error: buffer overflow 'output_tf->tf_pts.green' 1025 <= s32max drivers/gpu/drm/amd/amdgpu/../display/dc/dcn10/dcn10_cm_common.c:596 cm_helper_translate_curve_to_degamma_hw_format() error: buffer overflow 'output_tf->tf_pts.blue' 1025 <= s32max Cc: Tom Chung <chiahsuan.chung(a)amd.com> Cc: Rodrigo Siqueira <Rodrigo.Siqueira(a)amd.com> Cc: Roman Li <roman.li(a)amd.com> Cc: Alex Hung <alex.hung(a)amd.com> Cc: Aurabindo Pillai <aurabindo.pillai(a)amd.com> Cc: Harry Wentland <harry.wentland(a)amd.com> Cc: Hamza Mahfooz <hamza.mahfooz(a)amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam(a)amd.com> Reviewed-by: Tom Chung <chiahsuan.chung(a)amd.com> Signed-off-by: Alex Deucher <alexander.deucher(a)amd.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Liu Chuang <liuchuang40(a)huawei.com> --- drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c index c0372aa4ec83..684e30f9cf89 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_cm_common.c @@ -571,6 +571,8 @@ bool cm_helper_translate_curve_to_degamma_hw_format( i += increment) { if (j == hw_points - 1) break; + if (i >= TRANSFER_FUNC_POINTS) + return false; rgb_resulted[j].red = output_tf->tf_pts.red[i]; rgb_resulted[j].green = output_tf->tf_pts.green[i]; rgb_resulted[j].blue = output_tf->tf_pts.blue[i]; -- 2.34.1

2 1

[Cancel] openEuler TC 临时会议
by openEuler conference 06 Nov '24

06 Nov '24

Sorry! The WeLink meeting will be held at 2024-11-09 10:00 scheduled by Kernel SIG has been cancelled.

1 0

[PATCH OLK-6.6] ksmbd: add refcnt to ksmbd_conn struct
by Long Li 06 Nov '24

06 Nov '24

From: Namjae Jeon <linkinjeon(a)kernel.org> mainline inclusion from mainline-v6.10-rc2 commit ee426bfb9d09b29987369b897fe9b6485ac2be27 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IAYRCR CVE: CVE-2024-49988 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- When sending an oplock break request, opinfo->conn is used, But freed ->conn can be used on multichannel. This patch add a reference count to the ksmbd_conn struct so that it can be freed when it is no longer used. Signed-off-by: Namjae Jeon <linkinjeon(a)kernel.org> Signed-off-by: Steve French <stfrench(a)microsoft.com> Signed-off-by: Long Li <leo.lilong(a)huawei.com> --- fs/smb/server/connection.c | 4 ++- fs/smb/server/connection.h | 1 + fs/smb/server/oplock.c | 55 +++++++++++--------------------------- fs/smb/server/vfs_cache.c | 3 +++ 4 files changed, 23 insertions(+), 40 deletions(-) diff --git a/fs/smb/server/connection.c b/fs/smb/server/connection.c index 7889df8112b4..cac80e7bfefc 100644 --- a/fs/smb/server/connection.c +++ b/fs/smb/server/connection.c @@ -39,7 +39,8 @@ void ksmbd_conn_free(struct ksmbd_conn *conn) xa_destroy(&conn->sessions); kvfree(conn->request_buf); kfree(conn->preauth_info); - kfree(conn); + if (atomic_dec_and_test(&conn->refcnt)) + kfree(conn); } /** @@ -68,6 +69,7 @@ struct ksmbd_conn *ksmbd_conn_alloc(void) conn->um = NULL; atomic_set(&conn->req_running, 0); atomic_set(&conn->r_count, 0); + atomic_set(&conn->refcnt, 1); conn->total_credits = 1; conn->outstanding_credits = 0; diff --git a/fs/smb/server/connection.h b/fs/smb/server/connection.h index b93e5437793e..82343afc8d04 100644 --- a/fs/smb/server/connection.h +++ b/fs/smb/server/connection.h @@ -106,6 +106,7 @@ struct ksmbd_conn { bool signing_negotiated; __le16 signing_algorithm; bool binding; + atomic_t refcnt; }; struct ksmbd_conn_ops { diff --git a/fs/smb/server/oplock.c b/fs/smb/server/oplock.c index e546ffa57b55..8ee86478287f 100644 --- a/fs/smb/server/oplock.c +++ b/fs/smb/server/oplock.c @@ -51,6 +51,7 @@ static struct oplock_info *alloc_opinfo(struct ksmbd_work *work, init_waitqueue_head(&opinfo->oplock_brk); atomic_set(&opinfo->refcount, 1); atomic_set(&opinfo->breaking_cnt, 0); + atomic_inc(&opinfo->conn->refcnt); return opinfo; } @@ -124,6 +125,8 @@ static void free_opinfo(struct oplock_info *opinfo) { if (opinfo->is_lease) free_lease(opinfo); + if (opinfo->conn && atomic_dec_and_test(&opinfo->conn->refcnt)) + kfree(opinfo->conn); kfree(opinfo); } @@ -163,9 +166,7 @@ static struct oplock_info *opinfo_get_list(struct ksmbd_inode *ci) !atomic_inc_not_zero(&opinfo->refcount)) opinfo = NULL; else { - atomic_inc(&opinfo->conn->r_count); if (ksmbd_conn_releasing(opinfo->conn)) { - atomic_dec(&opinfo->conn->r_count); atomic_dec(&opinfo->refcount); opinfo = NULL; } @@ -177,26 +178,11 @@ static struct oplock_info *opinfo_get_list(struct ksmbd_inode *ci) return opinfo; } -static void opinfo_conn_put(struct oplock_info *opinfo) +void opinfo_put(struct oplock_info *opinfo) { - struct ksmbd_conn *conn; - if (!opinfo) return; - conn = opinfo->conn; - /* - * Checking waitqueue to dropping pending requests on - * disconnection. waitqueue_active is safe because it - * uses atomic operation for condition. - */ - if (!atomic_dec_return(&conn->r_count) && waitqueue_active(&conn->r_count_q)) - wake_up(&conn->r_count_q); - opinfo_put(opinfo); -} - -void opinfo_put(struct oplock_info *opinfo) -{ if (!atomic_dec_and_test(&opinfo->refcount)) return; @@ -1127,14 +1113,11 @@ void smb_send_parent_lease_break_noti(struct ksmbd_file *fp, if (!atomic_inc_not_zero(&opinfo->refcount)) continue; - atomic_inc(&opinfo->conn->r_count); - if (ksmbd_conn_releasing(opinfo->conn)) { - atomic_dec(&opinfo->conn->r_count); + if (ksmbd_conn_releasing(opinfo->conn)) continue; - } oplock_break(opinfo, SMB2_OPLOCK_LEVEL_NONE); - opinfo_conn_put(opinfo); + opinfo_put(opinfo); } } up_read(&p_ci->m_lock); @@ -1167,13 +1150,10 @@ void smb_lazy_parent_lease_break_close(struct ksmbd_file *fp) if (!atomic_inc_not_zero(&opinfo->refcount)) continue; - atomic_inc(&opinfo->conn->r_count); - if (ksmbd_conn_releasing(opinfo->conn)) { - atomic_dec(&opinfo->conn->r_count); + if (ksmbd_conn_releasing(opinfo->conn)) continue; - } oplock_break(opinfo, SMB2_OPLOCK_LEVEL_NONE); - opinfo_conn_put(opinfo); + opinfo_put(opinfo); } } up_read(&p_ci->m_lock); @@ -1252,7 +1232,7 @@ int smb_grant_oplock(struct ksmbd_work *work, int req_op_level, u64 pid, prev_opinfo = opinfo_get_list(ci); if (!prev_opinfo || (prev_opinfo->level == SMB2_OPLOCK_LEVEL_NONE && lctx)) { - opinfo_conn_put(prev_opinfo); + opinfo_put(prev_opinfo); goto set_lev; } prev_op_has_lease = prev_opinfo->is_lease; @@ -1262,19 +1242,19 @@ int smb_grant_oplock(struct ksmbd_work *work, int req_op_level, u64 pid, if (share_ret < 0 && prev_opinfo->level == SMB2_OPLOCK_LEVEL_EXCLUSIVE) { err = share_ret; - opinfo_conn_put(prev_opinfo); + opinfo_put(prev_opinfo); goto err_out; } if (prev_opinfo->level != SMB2_OPLOCK_LEVEL_BATCH && prev_opinfo->level != SMB2_OPLOCK_LEVEL_EXCLUSIVE) { - opinfo_conn_put(prev_opinfo); + opinfo_put(prev_opinfo); goto op_break_not_needed; } list_add(&work->interim_entry, &prev_opinfo->interim_list); err = oplock_break(prev_opinfo, SMB2_OPLOCK_LEVEL_II); - opinfo_conn_put(prev_opinfo); + opinfo_put(prev_opinfo); if (err == -ENOENT) goto set_lev; /* Check all oplock was freed by close */ @@ -1337,14 +1317,14 @@ static void smb_break_all_write_oplock(struct ksmbd_work *work, return; if (brk_opinfo->level != SMB2_OPLOCK_LEVEL_BATCH && brk_opinfo->level != SMB2_OPLOCK_LEVEL_EXCLUSIVE) { - opinfo_conn_put(brk_opinfo); + opinfo_put(brk_opinfo); return; } brk_opinfo->open_trunc = is_trunc; list_add(&work->interim_entry, &brk_opinfo->interim_list); oplock_break(brk_opinfo, SMB2_OPLOCK_LEVEL_II); - opinfo_conn_put(brk_opinfo); + opinfo_put(brk_opinfo); } /** @@ -1376,11 +1356,8 @@ void smb_break_all_levII_oplock(struct ksmbd_work *work, struct ksmbd_file *fp, if (!atomic_inc_not_zero(&brk_op->refcount)) continue; - atomic_inc(&brk_op->conn->r_count); - if (ksmbd_conn_releasing(brk_op->conn)) { - atomic_dec(&brk_op->conn->r_count); + if (ksmbd_conn_releasing(brk_op->conn)) continue; - } rcu_read_unlock(); if (brk_op->is_lease && (brk_op->o_lease->state & @@ -1411,7 +1388,7 @@ void smb_break_all_levII_oplock(struct ksmbd_work *work, struct ksmbd_file *fp, brk_op->open_trunc = is_trunc; oplock_break(brk_op, SMB2_OPLOCK_LEVEL_NONE); next: - opinfo_conn_put(brk_op); + opinfo_put(brk_op); rcu_read_lock(); } rcu_read_unlock(); diff --git a/fs/smb/server/vfs_cache.c b/fs/smb/server/vfs_cache.c index 8b2e37c8716e..271a23abc82f 100644 --- a/fs/smb/server/vfs_cache.c +++ b/fs/smb/server/vfs_cache.c @@ -710,6 +710,8 @@ static bool session_fd_check(struct ksmbd_tree_connect *tcon, list_for_each_entry_rcu(op, &ci->m_op_list, op_entry) { if (op->conn != conn) continue; + if (op->conn && atomic_dec_and_test(&op->conn->refcnt)) + kfree(op->conn); op->conn = NULL; } up_write(&ci->m_lock); @@ -807,6 +809,7 @@ int ksmbd_reopen_durable_fd(struct ksmbd_work *work, struct ksmbd_file *fp) if (op->conn) continue; op->conn = fp->conn; + atomic_inc(&op->conn->refcnt); } up_write(&ci->m_lock); -- 2.39.2

2 1

openEuler TC 临时会议
by openEuler conference 06 Nov '24

06 Nov '24

您好！ Kernel 邀请您参加 2024-11-09 10:00 召开的WeLink会议(自动录制) 会议主题：openEuler TC 临时会议会议链接：https://meeting.huaweicloud.com:36443/#/j/983087267 会议纪要：https://etherpad.openeuler.org/p/Kernel-meetings 更多资讯尽在：https://www.openeuler.org/zh/ Hello! Kernel invites you to attend the WeLink conference(auto recording) will be held at 2024-11-09 10:00, The subject of the conference is openEuler TC 临时会议, You can join the meeting at https://meeting.huaweicloud.com:36443/#/j/983087267. Add topics at https://etherpad.openeuler.org/p/Kernel-meetings. More information: https://www.openeuler.org/en/

1 0

[PATCH v4 OLK-5.10 0/7] mm: support poison recover for more
by Wupeng Ma 06 Nov '24

06 Nov '24

From: Ma Wupeng <mawupeng1(a)huawei.com> Add machine check safe support for cow & migrate_pages. Avoid kernel panic during arm64_do_kernel_sea if no valid info is reported. Changelog since v3: - backport bugfix commit 88e4f525002b ("mm/hugetlb: add missing VM_FAULT_SET_HINDEX in hugetlb_wp") Changelog since v1: - fix incorrect patch format Kefeng Wang (2): mm: support poison recovery from copy_present_page() mm: support poison recovery from do_cow_fault() Liu Shixin (1): mm: hwpoison: support recovery from HugePage copy-on-write faults Ma Wupeng (2): arm64: mm: Add copy mc support for all migrate_page arm64: send sig fault for user task when apei_claim_sea fails Oscar Salvador (1): mm/hugetlb: add missing VM_FAULT_SET_HINDEX in hugetlb_wp Tong Tiangen (1): make copy_[user]_highpage_mc have return value arch/arm64/include/asm/page.h | 6 +-- arch/arm64/lib/copy_page_mc.S | 6 ++- arch/arm64/mm/copypage.c | 21 +++++--- arch/arm64/mm/fault.c | 17 +++++-- include/linux/highmem.h | 13 ++++- include/linux/mm.h | 8 ++-- mm/hugetlb.c | 7 ++- mm/memory.c | 90 +++++++++++++++++++++++------------ mm/migrate.c | 5 +- 9 files changed, 118 insertions(+), 55 deletions(-) -- 2.25.1

2 8

[PATCH OLK-6.6] sched/fair: Take the scheduling domain into account in select_idle_smt()
by Wenyu Huang 06 Nov '24

06 Nov '24

From: Keisuke Nishimura <keisuke.nishimura(a)inria.fr> stable inclusion from stable-v6.6.23 commit 495b7cb952003973ff1fe4f70639bf3df91fbfb1 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IB2B23 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=… -------------------------------- [ Upstream commit 8aeaffef8c6eceab0e1498486fdd4f3dc3b7066c ] When picking a CPU on task wakeup, select_idle_smt() has to take into account the scheduling domain of @target. This is because the "isolcpus" kernel command line option can remove CPUs from the domain to isolate them from other SMT siblings. This fix checks if the candidate CPU is in the target scheduling domain. Commit: df3cb4ea1fb6 ("sched/fair: Fix wrong cpu selecting from isolated domain") ... originally introduced this fix by adding the check of the scheduling domain in the loop. However, commit: 3e6efe87cd5cc ("sched/fair: Remove redundant check in select_idle_smt()") ... accidentally removed the check. Bring it back. Fixes: 3e6efe87cd5c ("sched/fair: Remove redundant check in select_idle_smt()") Signed-off-by: Keisuke Nishimura <keisuke.nishimura(a)inria.fr> Signed-off-by: Julia Lawall <julia.lawall(a)inria.fr> Signed-off-by: Ingo Molnar <mingo(a)kernel.org> Reviewed-by: Vincent Guittot <vincent.guittot(a)linaro.org> Link: https://lore.kernel.org/r/20240110131707.437301-1-keisuke.nishimura@inria.fr Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Wenyu Huang <huangwenyu5(a)huawei.com> --- kernel/sched/fair.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 4ef95b2210b5e..d5ae744deb39d 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -8070,7 +8070,7 @@ static int select_idle_core(struct task_struct *p, int core, struct cpumask *cpu /* * Scan the local SMT mask for idle CPUs. */ -static int select_idle_smt(struct task_struct *p, int target) +static int select_idle_smt(struct task_struct *p, struct sched_domain *sd, int target) { int cpu; @@ -8081,6 +8081,12 @@ static int select_idle_smt(struct task_struct *p, int target) #endif if (cpu == target) continue; + /* + * Check if the CPU is in the LLC scheduling domain of @target. + * Due to isolcpus, there is no guarantee that all the siblings are in the domain. + */ + if (!cpumask_test_cpu(cpu, sched_domain_span(sd))) + continue; if (available_idle_cpu(cpu) || sched_idle_cpu(cpu)) return cpu; } @@ -8104,7 +8110,7 @@ static inline int select_idle_core(struct task_struct *p, int core, struct cpuma return __select_idle_cpu(core, p); } -static inline int select_idle_smt(struct task_struct *p, int target) +static inline int select_idle_smt(struct task_struct *p, struct sched_domain *sd, int target) { return -1; } @@ -8441,7 +8447,7 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target) has_idle_core = test_idle_cores(target); if (!has_idle_core && cpus_share_cache(prev, target)) { - i = select_idle_smt(p, prev); + i = select_idle_smt(p, sd, prev); if ((unsigned int)i < nr_cpumask_bits) return i; } -- 2.34.1

2 1

[PATCH OLK-5.10] serial: core: Fix atomicity violation in uart_tiocmget
by Yi Yang 06 Nov '24

06 Nov '24

From: Gui-Dong Han <2045gemini(a)gmail.com> mainline inclusion from mainline-v6.8-rc3 commit 30926783a46841c2d1bbf3f74067ba85d304fd0d category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IB1MRY CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- In uart_tiocmget(): result = uport->mctrl; uart_port_lock_irq(uport); result |= uport->ops->get_mctrl(uport); uart_port_unlock_irq(uport); ... return result; In uart_update_mctrl(): uart_port_lock_irqsave(port, &flags); ... port->mctrl = (old & ~clear) | set; ... port->ops->set_mctrl(port, port->mctrl); ... uart_port_unlock_irqrestore(port, flags); An atomicity violation is identified due to the concurrent execution of uart_tiocmget() and uart_update_mctrl(). After assigning result = uport->mctrl, the mctrl value may change in uart_update_mctrl(), leading to a mismatch between the value returned by uport->ops->get_mctrl(uport) and the mctrl value previously read. This can result in uart_tiocmget() returning an incorrect value. This possible bug is found by an experimental static analysis tool developed by our team, BassCheck[1]. This tool analyzes the locking APIs to extract function pairs that can be concurrently executed, and then analyzes the instructions in the paired functions to identify possible concurrency bugs including data races and atomicity violations. The above possible bug is reported when our tool analyzes the source code of Linux 5.17. To address this issue, it is suggested to move the line result = uport->mctrl inside the uart_port_lock block to ensure atomicity and prevent the mctrl value from being altered during the execution of uart_tiocmget(). With this patch applied, our tool no longer reports the bug, with the kernel configuration allyesconfig for x86_64. Due to the absence of the requisite hardware, we are unable to conduct runtime testing of the patch. Therefore, our verification is solely based on code logic analysis. [1] https://sites.google.com/view/basscheck/ Fixes: c5f4644e6c8b ("[PATCH] Serial: Adjust serial locking") Cc: stable(a)vger.kernel.org Signed-off-by: Gui-Dong Han <2045gemini(a)gmail.com> Link: https://lore.kernel.org/r/20240112113624.17048-1-2045gemini@gmail.com Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Conflicts: drivers/tty/serial/serial_core.c [Commit 559c7ff4e324("serial: core: Use port lock wrappers") not merged, no functional change.] Signed-off-by: Yi Yang <yiyang13(a)huawei.com> --- drivers/tty/serial/serial_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/tty/serial/serial_core.c b/drivers/tty/serial/serial_core.c index c7adcf97e2a3..f86bef623953 100644 --- a/drivers/tty/serial/serial_core.c +++ b/drivers/tty/serial/serial_core.c @@ -1092,8 +1092,8 @@ static int uart_tiocmget(struct tty_struct *tty) goto out; if (!tty_io_error(tty)) { - result = uport->mctrl; spin_lock_irq(&uport->lock); + result = uport->mctrl; result |= uport->ops->get_mctrl(uport); spin_unlock_irq(&uport->lock); } -- 2.25.1

2 1

[PATCH OLK-6.6] serial: core: Fix atomicity violation in uart_tiocmget
by Yi Yang 06 Nov '24

06 Nov '24

From: Gui-Dong Han <2045gemini(a)gmail.com> mainline inclusion from mainline-v6.8-rc3 commit 30926783a46841c2d1bbf3f74067ba85d304fd0d category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/IB1MRY CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- In uart_tiocmget(): result = uport->mctrl; uart_port_lock_irq(uport); result |= uport->ops->get_mctrl(uport); uart_port_unlock_irq(uport); ... return result; In uart_update_mctrl(): uart_port_lock_irqsave(port, &flags); ... port->mctrl = (old & ~clear) | set; ... port->ops->set_mctrl(port, port->mctrl); ... uart_port_unlock_irqrestore(port, flags); An atomicity violation is identified due to the concurrent execution of uart_tiocmget() and uart_update_mctrl(). After assigning result = uport->mctrl, the mctrl value may change in uart_update_mctrl(), leading to a mismatch between the value returned by uport->ops->get_mctrl(uport) and the mctrl value previously read. This can result in uart_tiocmget() returning an incorrect value. This possible bug is found by an experimental static analysis tool developed by our team, BassCheck[1]. This tool analyzes the locking APIs to extract function pairs that can be concurrently executed, and then analyzes the instructions in the paired functions to identify possible concurrency bugs including data races and atomicity violations. The above possible bug is reported when our tool analyzes the source code of Linux 5.17. To address this issue, it is suggested to move the line result = uport->mctrl inside the uart_port_lock block to ensure atomicity and prevent the mctrl value from being altered during the execution of uart_tiocmget(). With this patch applied, our tool no longer reports the bug, with the kernel configuration allyesconfig for x86_64. Due to the absence of the requisite hardware, we are unable to conduct runtime testing of the patch. Therefore, our verification is solely based on code logic analysis. [1] https://sites.google.com/view/basscheck/ Fixes: c5f4644e6c8b ("[PATCH] Serial: Adjust serial locking") Cc: stable(a)vger.kernel.org Signed-off-by: Gui-Dong Han <2045gemini(a)gmail.com> Link: https://lore.kernel.org/r/20240112113624.17048-1-2045gemini@gmail.com Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Conflicts: drivers/tty/serial/serial_core.c [Commit 559c7ff4e324("serial: core: Use port lock wrappers") not merged, no functional change.] Signed-off-by: Yi Yang <yiyang13(a)huawei.com> --- drivers/tty/serial/serial_core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/tty/serial/serial_core.c b/drivers/tty/serial/serial_core.c index 5d5570bebfeb..938e36b6cd4f 100644 --- a/drivers/tty/serial/serial_core.c +++ b/drivers/tty/serial/serial_core.c @@ -1105,8 +1105,8 @@ static int uart_tiocmget(struct tty_struct *tty) goto out; if (!tty_io_error(tty)) { - result = uport->mctrl; spin_lock_irq(&uport->lock); + result = uport->mctrl; result |= uport->ops->get_mctrl(uport); spin_unlock_irq(&uport->lock); } -- 2.25.1

2 1