- Kernel - mailweb.openeuler.org

openEuler Kernel SIG双周例会
by openEuler conference 09 Nov '23

09 Nov '23

您好！ Kernel SIG 邀请您参加 2023-11-10 14:00 召开的Zoom会议(自动录制) 会议主题：openEuler Kernel SIG双周例会会议内容： 1. 进展update 2. 内容征集中（议题申报可回复邮件，或直接填入会议看板）会议链接：https://us06web.zoom.us/j/87016190729?pwd=QXivFoXObEngRNR23faQCqS1XGbrs3.1 会议纪要：https://etherpad.openeuler.org/p/Kernel-meetings 温馨提醒：建议接入会议后修改参会人的姓名，也可以使用您在gitee.com的ID 更多资讯尽在：https://openeuler.org/zh/ Hello! openEuler Kernel SIG invites you to attend the Zoom conference(auto recording) will be held at 2023-11-10 14:00, The subject of the conference is openEuler Kernel SIG双周例会, Summary: 1. 进展update 2. 内容征集中（议题申报可回复邮件，或直接填入会议看板） You can join the meeting at https://us06web.zoom.us/j/87016190729?pwd=QXivFoXObEngRNR23faQCqS1XGbrs3.1. Add topics at https://etherpad.openeuler.org/p/Kernel-meetings. Note: You are advised to change the participant name after joining the conference or use your ID at gitee.com. More information: https://openeuler.org/en/

1 0

[PATCH openEuler-1.0-LTS] sched/fair: Refill bandwidth before scaling
by Xia Fukun 08 Nov '23

08 Nov '23

From: Huaixin Chang <changhuaixin(a)linux.alibaba.com> mainline inclusion from mainline-v6.5-rc7 commit 5a6d6a6ccb5f48ca8cf7c6d64ff83fd9c7999390 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8EUCE CVE: NA -------------------------------- In order to prevent possible hardlockup of sched_cfs_period_timer() loop, loop count is introduced to denote whether to scale quota and period or not. However, scale is done between forwarding period timer and refilling cfs bandwidth runtime, which means that period timer is forwarded with old "period" while runtime is refilled with scaled "quota". Move do_sched_cfs_period_timer() before scaling to solve this. Fixes: 2e8e19226398 ("sched/fair: Limit sched_cfs_period_timer() loop to avoid hard lockup") Signed-off-by: Huaixin Chang <changhuaixin(a)linux.alibaba.com> Signed-off-by: Peter Zijlstra (Intel) <peterz(a)infradead.org> Reviewed-by: Ben Segall <bsegall(a)google.com> Reviewed-by: Phil Auld <pauld(a)redhat.com> Link: https://lkml.kernel.org/r/20200420024421.22442-3-changhuaixin@linux.alibaba… Signed-off-by: Xia Fukun <xiafukun(a)huawei.com> --- kernel/sched/fair.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 1c78e2f29901..ccd2a060c2df 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -5099,6 +5099,8 @@ static enum hrtimer_restart sched_cfs_period_timer(struct hrtimer *timer) if (!overrun) break; + idle = do_sched_cfs_period_timer(cfs_b, overrun, flags); + if (++count > 3) { u64 new, old = ktime_to_ns(cfs_b->period); @@ -5128,8 +5130,6 @@ static enum hrtimer_restart sched_cfs_period_timer(struct hrtimer *timer) /* reset count so we don't come right back in here */ count = 0; } - - idle = do_sched_cfs_period_timer(cfs_b, overrun, flags); } if (idle) cfs_b->period_active = 0; -- 2.34.1

2 1

[PATCH openEuler-1.0-LTS] signal: Properly set TRACE_SIGNAL_LOSE_INFO in __send_signal
by Xia Fukun 08 Nov '23

08 Nov '23

From: "Eric W. Biederman" <ebiederm(a)xmission.com> mainline inclusion from mainline-v6.5-rc7 commit 8917bef336f5301edd616cfa97b97d0319fd0496 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8ETTA CVE: NA -------------------------------- Any time siginfo is not stored in the signal queue information is lost. Therefore set TRACE_SIGNAL_LOSE_INFO every time the code does not allocate a signal queue entry, and a queue overflow abort is not triggered. Fixes: ba005e1f4172 ("tracepoint: Add signal loss events") Signed-off-by: "Eric W. Biederman" <ebiederm(a)xmission.com> Signed-off-by: Xia Fukun <xiafukun(a)huawei.com> --- kernel/signal.c | 33 ++++++++++++++++----------------- 1 file changed, 16 insertions(+), 17 deletions(-) diff --git a/kernel/signal.c b/kernel/signal.c index 69b9d8bff5a8..4166d22645f8 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -1163,23 +1163,22 @@ static int __send_signal(int sig, struct siginfo *info, struct task_struct *t, userns_fixup_signal_uid(&q->info, t); - } else if (!is_si_special(info)) { - if (sig >= SIGRTMIN && info->si_code != SI_USER) { - /* - * Queue overflow, abort. We may abort if the - * signal was rt and sent by user using something - * other than kill(). - */ - result = TRACE_SIGNAL_OVERFLOW_FAIL; - ret = -EAGAIN; - goto ret; - } else { - /* - * This is a silent loss of information. We still - * send the signal, but the *info bits are lost. - */ - result = TRACE_SIGNAL_LOSE_INFO; - } + } else if (!is_si_special(info) && + sig >= SIGRTMIN && info->si_code != SI_USER) { + /* + * Queue overflow, abort. We may abort if the + * signal was rt and sent by user using something + * other than kill(). + */ + result = TRACE_SIGNAL_OVERFLOW_FAIL; + ret = -EAGAIN; + goto ret; + } else { + /* + * This is a silent loss of information. We still + * send the signal, but the *info bits are lost. + */ + result = TRACE_SIGNAL_LOSE_INFO; } out_set: -- 2.34.1

2 1

[PATCH OLK-5.10 0/9] handle uninitialized numa nodes gracefully.
by Wupeng Ma 08 Nov '23

08 Nov '23

From: Ma Wupeng <mawupeng1(a)huawei.com> handle uninitialized numa nodes gracefully. Haifeng Xu (1): mm/memcontrol: do not tweak node in mem_cgroup_init() Michal Hocko (5): mm, memory_hotplug: make arch_alloc_nodedata independent on CONFIG_MEMORY_HOTPLUG mm: handle uninitialized numa nodes gracefully mm, memory_hotplug: drop arch_free_nodedata mm, memory_hotplug: reorganize new pgdat initialization mm: make free_area_init_node aware of memory less nodes Oscar Salvador (1): arch/x86/mm/numa: Do not initialize nodes twice Srikar Dronamraju (1): powerpc/numa: Handle partially initialized numa nodes Wei Yang (1): memcg: do not tweak node in alloc_mem_cgroup_per_node_info arch/ia64/mm/discontig.c | 11 +-- arch/powerpc/mm/numa.c | 2 +- arch/x86/mm/numa.c | 33 +++++---- include/linux/memory_hotplug.h | 118 ++++++++++++++++----------------- include/linux/mm.h | 1 - mm/internal.h | 2 + mm/memcontrol.c | 17 +---- mm/memory_hotplug.c | 55 +++------------ mm/page_alloc.c | 78 +++++++++++++++++++--- 9 files changed, 163 insertions(+), 154 deletions(-) -- 2.25.1

2 10

[PATCH OLK-5.10 v6 00/10] AGDI: Add driver for Arm Generic Diagnostic Dump and Reset device
by Qian Zou 08 Nov '23

08 Nov '23

Some use-cases, such as system management, require the ability to generate a non-maskable event to the OS to request the OS kernel to perform a diagnostic dump and reset the system. Arm Generic Diagnostic Dump and Reset device enables a maintainer to request OS to perform a diagnostic dump and reset a system via SDEI event or an interrupt. This patch implements SDEI path and discards the interrupted context before proceeding to the crash kernel. D Scott Phillips (1): arm64: sdei: abort running SDEI handlers during crash Ilkka Koskinen (4): ACPI: tables: Add AGDI to the list of known table signatures ACPI: AGDI: Add driver for Arm Generic Diagnostic Dump and Reset device ACPICA: iASL: Add suppport for AGDI table ACPI: AGDI: Fix missing prototype warning for acpi_agdi_init() Jia He (3): EDAC/ghes: Add a notifier for reporting memory errors EDAC/ghes: Prepare to make ghes_edac a proper module EDAC/ghes: Make ghes_edac a proper module Li Yang (1): APEI: GHES: correctly return NULL for ghes_get_devices() Shuai Xue (1): ACPI: APEI: explicit init of HEST and GHES in apci_init() arch/arm64/include/asm/sdei.h | 6 ++ arch/arm64/kernel/entry.S | 27 +++++++- arch/arm64/kernel/sdei.c | 3 + arch/arm64/kernel/smp.c | 8 +-- drivers/acpi/apei/ghes.c | 85 ++++++++++++++++++++---- drivers/acpi/arm64/Kconfig | 10 +++ drivers/acpi/arm64/Makefile | 1 + drivers/acpi/arm64/agdi.c | 117 ++++++++++++++++++++++++++++++++++ drivers/acpi/bus.c | 4 ++ drivers/acpi/pci_root.c | 3 - drivers/acpi/tables.c | 2 +- drivers/edac/Kconfig | 4 +- drivers/edac/ghes_edac.c | 90 ++++++++++++++++---------- drivers/firmware/Kconfig | 1 + drivers/firmware/arm_sdei.c | 32 ++++++---- include/acpi/actbl2.h | 20 ++++++ include/acpi/apei.h | 4 +- include/acpi/ghes.h | 34 +++------- include/linux/acpi_agdi.h | 13 ++++ include/linux/arm_sdei.h | 4 ++ 20 files changed, 371 insertions(+), 97 deletions(-) create mode 100644 drivers/acpi/arm64/agdi.c create mode 100644 include/linux/acpi_agdi.h -- 2.33.0

2 11

[PATCH v3 OLK-5.10 00/24] arm64/perf: Enable branch stack sampling
by Junhao He 08 Nov '23

08 Nov '23

This series enables perf branch stack sampling support on arm64 platform via a new arch feature called Branch Record Buffer Extension (BRBE). All relevant register definitions could be accessed here. v2-v3: - Remove indentation before macro #ifdef v1-v2: - Fix kabi breakage in struct perf_branch_entry Anshuman Khandual (16): perf tools: Add missing branch_sample_type to perf_event_attr__fprintf() perf: Add irq and exception return branch types perf: Add system error and not in transaction branch types perf: Extend branch type classification perf: Capture branch privilege information perf: Add PERF_BR_NEW_ARCH_[N] map for BRBE on arm64 platform perf branch: Add system error and not in transaction branch types perf branch: Extend branch type classification perf branch: Add branch privilege information request flag perf branch: Add PERF_BR_NEW_ARCH_[N] map for BRBE on arm64 platform perf: Consolidate branch sample filter helpers perf record: Add remaining branch filters: "no_cycles", "no_flags" & "hw_index" arm64/sysreg: Add BRBE registers and fields drivers: perf: arm_pmu: Add new sched_task() callback drivers: perf: arm_pmuv3: Enable branch stack sampling framework drivers: perf: arm_pmuv3: Enable branch stack sampling via FEAT_BRBE James Clark (5): perf evsel: Add error message for unsupported branch stack cases perf session: Print branch stack entry type in --dump-raw-trace perf script: Refactor branch stack printing perf script: Output branch sample type perf branch: Fix interpretation of branch records Junhao He (1): perf: Fix kabi breakage in struct perf_branch_entry Mark Brown (1): arm64/sysreg: Introduce helpers for access to sysreg fields Sandipan Das (1): perf/core: Add speculation info to branch entries arch/arm64/include/asm/perf_event.h | 56 ++ arch/arm64/include/asm/sysreg.h | 511 ++++++++++++++ arch/arm64/kernel/head.S | 48 ++ arch/arm64/kernel/perf_event.c | 90 ++- arch/x86/events/intel/lbr.c | 4 +- drivers/perf/Kconfig | 11 + drivers/perf/Makefile | 1 + drivers/perf/arm_brbe.c | 779 ++++++++++++++++++++++ drivers/perf/arm_brbe.h | 258 +++++++ drivers/perf/arm_pmu.c | 34 +- include/linux/perf/arm_pmu.h | 28 +- include/linux/perf_event.h | 27 + include/uapi/linux/perf_event.h | 53 ++ kernel/events/core.c | 9 +- tools/include/uapi/linux/perf_event.h | 38 +- tools/perf/builtin-script.c | 28 +- tools/perf/util/branch.c | 72 +- tools/perf/util/branch.h | 8 +- tools/perf/util/evsel.c | 4 + tools/perf/util/parse-branch-options.c | 4 + tools/perf/util/perf_event_attr_fprintf.c | 2 +- tools/perf/util/session.c | 5 +- 22 files changed, 2035 insertions(+), 35 deletions(-) create mode 100644 drivers/perf/arm_brbe.c create mode 100644 drivers/perf/arm_brbe.h -- 2.33.0

2 25

[PATCH v4 openEuler-1.0-LTS 0/3] handle uninitialized numa nodes gracefully.
by Wupeng Ma 08 Nov '23

08 Nov '23

From: Ma Wupeng <mawupeng1(a)huawei.com> handle uninitialized numa nodes gracefully. Michal Hocko (2): mm, memory_hotplug: make arch_alloc_nodedata independent on CONFIG_MEMORY_HOTPLUG mm: handle uninitialized numa nodes gracefully Oscar Salvador (1): arch/x86/mm/numa: Do not initialize nodes twice arch/ia64/mm/discontig.c | 6 +- arch/x86/mm/numa.c | 26 +++---- include/linux/memory_hotplug.h | 120 ++++++++++++++++----------------- mm/internal.h | 2 + mm/memory_hotplug.c | 24 +++---- mm/page_alloc.c | 45 +++++++++++-- 6 files changed, 123 insertions(+), 100 deletions(-) -- 2.25.1

2 4

[PATCH OLK-5.10 v6 00/10] AGDI: Add driver for Arm Generic Diagnostic Dump and Reset device
by Qian Zou 08 Nov '23

08 Nov '23

Some use-cases, such as system management, require the ability to generate a non-maskable event to the OS to request the OS kernel to perform a diagnostic dump and reset the system. Arm Generic Diagnostic Dump and Reset device enables a maintainer to request OS to perform a diagnostic dump and reset a system via SDEI event or an interrupt. This patch implements SDEI path and discards the interrupted context before proceeding to the crash kernel. D Scott Phillips (1): arm64: sdei: abort running SDEI handlers during crash Ilkka Koskinen (4): ACPI: tables: Add AGDI to the list of known table signatures ACPI: AGDI: Add driver for Arm Generic Diagnostic Dump and Reset device ACPICA: iASL: Add suppport for AGDI table ACPI: AGDI: Fix missing prototype warning for acpi_agdi_init() Jia He (3): EDAC/ghes: Add a notifier for reporting memory errors EDAC/ghes: Prepare to make ghes_edac a proper module EDAC/ghes: Make ghes_edac a proper module Li Yang (1): APEI: GHES: correctly return NULL for ghes_get_devices() Shuai Xue (1): ACPI: APEI: explicit init of HEST and GHES in apci_init() arch/arm64/include/asm/sdei.h | 6 ++ arch/arm64/kernel/entry.S | 27 +++++++- arch/arm64/kernel/sdei.c | 3 + arch/arm64/kernel/smp.c | 8 +-- drivers/acpi/apei/ghes.c | 85 ++++++++++++++++++++---- drivers/acpi/arm64/Kconfig | 10 +++ drivers/acpi/arm64/Makefile | 1 + drivers/acpi/arm64/agdi.c | 117 ++++++++++++++++++++++++++++++++++ drivers/acpi/bus.c | 4 ++ drivers/acpi/pci_root.c | 3 - drivers/acpi/tables.c | 2 +- drivers/edac/Kconfig | 4 +- drivers/edac/ghes_edac.c | 90 ++++++++++++++++---------- drivers/firmware/Kconfig | 1 + drivers/firmware/arm_sdei.c | 32 ++++++---- include/acpi/actbl2.h | 20 ++++++ include/acpi/apei.h | 4 +- include/acpi/ghes.h | 34 +++------- include/linux/acpi_agdi.h | 13 ++++ include/linux/arm_sdei.h | 4 ++ 20 files changed, 371 insertions(+), 97 deletions(-) create mode 100644 drivers/acpi/arm64/agdi.c create mode 100644 include/linux/acpi_agdi.h -- 2.33.0

1 9

[PATCH OLK-5.10 v5 0/9] AGDI: Add driver for Arm Generic Diagnostic Dump and Reset device
by Qian Zou 08 Nov '23

08 Nov '23

From: “z00447436” <zouqian4(a)huawei.com> Some use-cases, such as system management, require the ability to generate a non-maskable event to the OS to request the OS kernel to perform a diagnostic dump and reset the system. Arm Generic Diagnostic Dump and Reset device enables a maintainer to request OS to perform a diagnostic dump and reset a system via SDEI event or an interrupt. This patch implements SDEI path and discards the interrupted context before proceeding to the crash kernel. D Scott Phillips (1): arm64: sdei: abort running SDEI handlers during crash Ilkka Koskinen (4): ACPI: tables: Add AGDI to the list of known table signatures ACPI: AGDI: Add driver for Arm Generic Diagnostic Dump and Reset device ACPICA: iASL: Add suppport for AGDI table ACPI: AGDI: Fix missing prototype warning for acpi_agdi_init() Jia He (3): EDAC/ghes: Add a notifier for reporting memory errors EDAC/ghes: Prepare to make ghes_edac a proper module EDAC/ghes: Make ghes_edac a proper module Shuai Xue (1): ACPI: APEI: explicit init of HEST and GHES in apci_init() arch/arm64/include/asm/sdei.h | 6 ++ arch/arm64/kernel/entry.S | 27 +++++++- arch/arm64/kernel/sdei.c | 3 + arch/arm64/kernel/smp.c | 8 +-- drivers/acpi/apei/ghes.c | 83 ++++++++++++++++++++---- drivers/acpi/arm64/Kconfig | 10 +++ drivers/acpi/arm64/Makefile | 1 + drivers/acpi/arm64/agdi.c | 117 ++++++++++++++++++++++++++++++++++ drivers/acpi/bus.c | 4 ++ drivers/acpi/pci_root.c | 3 - drivers/acpi/tables.c | 2 +- drivers/edac/Kconfig | 4 +- drivers/edac/ghes_edac.c | 90 ++++++++++++++++---------- drivers/firmware/Kconfig | 1 + drivers/firmware/arm_sdei.c | 32 ++++++---- include/acpi/actbl2.h | 20 ++++++ include/acpi/apei.h | 4 +- include/acpi/ghes.h | 34 +++------- include/linux/acpi_agdi.h | 13 ++++ include/linux/arm_sdei.h | 4 ++ 20 files changed, 369 insertions(+), 97 deletions(-) create mode 100644 drivers/acpi/arm64/agdi.c create mode 100644 include/linux/acpi_agdi.h -- 2.33.0

2 10

[PATCH v2 OLK-5.10 00/24] arm64/perf: Enable branch stack sampling
by Junhao He 07 Nov '23

07 Nov '23

This series enables perf branch stack sampling support on arm64 platform via a new arch feature called Branch Record Buffer Extension (BRBE). All relevant register definitions could be accessed here. Anshuman Khandual (16): perf tools: Add missing branch_sample_type to perf_event_attr__fprintf() perf: Add irq and exception return branch types perf: Add system error and not in transaction branch types perf: Extend branch type classification perf: Capture branch privilege information perf: Add PERF_BR_NEW_ARCH_[N] map for BRBE on arm64 platform perf branch: Add system error and not in transaction branch types perf branch: Extend branch type classification perf branch: Add branch privilege information request flag perf branch: Add PERF_BR_NEW_ARCH_[N] map for BRBE on arm64 platform perf: Consolidate branch sample filter helpers perf record: Add remaining branch filters: "no_cycles", "no_flags" & "hw_index" arm64/sysreg: Add BRBE registers and fields drivers: perf: arm_pmu: Add new sched_task() callback drivers: perf: arm_pmuv3: Enable branch stack sampling framework drivers: perf: arm_pmuv3: Enable branch stack sampling via FEAT_BRBE James Clark (5): perf evsel: Add error message for unsupported branch stack cases perf session: Print branch stack entry type in --dump-raw-trace perf script: Refactor branch stack printing perf script: Output branch sample type perf branch: Fix interpretation of branch records Junhao He (1): perf: Fix kabi breakage in struct perf_branch_entry Mark Brown (1): arm64/sysreg: Introduce helpers for access to sysreg fields Sandipan Das (1): perf/core: Add speculation info to branch entries arch/arm64/include/asm/perf_event.h | 56 ++ arch/arm64/include/asm/sysreg.h | 511 ++++++++++++++ arch/arm64/kernel/head.S | 48 ++ arch/arm64/kernel/perf_event.c | 90 ++- arch/x86/events/intel/lbr.c | 4 +- drivers/perf/Kconfig | 11 + drivers/perf/Makefile | 1 + drivers/perf/arm_brbe.c | 779 ++++++++++++++++++++++ drivers/perf/arm_brbe.h | 258 +++++++ drivers/perf/arm_pmu.c | 34 +- include/linux/perf/arm_pmu.h | 28 +- include/linux/perf_event.h | 27 + include/uapi/linux/perf_event.h | 53 ++ kernel/events/core.c | 9 +- tools/include/uapi/linux/perf_event.h | 38 +- tools/perf/builtin-script.c | 28 +- tools/perf/util/branch.c | 72 +- tools/perf/util/branch.h | 8 +- tools/perf/util/evsel.c | 4 + tools/perf/util/parse-branch-options.c | 4 + tools/perf/util/perf_event_attr_fprintf.c | 2 +- tools/perf/util/session.c | 5 +- 22 files changed, 2035 insertions(+), 35 deletions(-) create mode 100644 drivers/perf/arm_brbe.c create mode 100644 drivers/perf/arm_brbe.h -- 2.33.0

2 25

2023年度openEuler社区优秀贡献者奖评选启动，欢迎大家参加
by openEuler 07 Nov '23

07 Nov '23

1 0

[PATCH v3 openEuler-1.0-LTS 0/3] handle uninitialized numa nodes gracefully
by Wupeng Ma 07 Nov '23

07 Nov '23

From: Ma Wupeng <mawupeng1(a)huawei.com> handle uninitialized numa nodes gracefully. Changelog since v2: - fix compile error. Changelog since v1: - add bugfix patch #3. Michal Hocko (2): mm, memory_hotplug: make arch_alloc_nodedata independent on CONFIG_MEMORY_HOTPLUG mm: handle uninitialized numa nodes gracefully Oscar Salvador (1): arch/x86/mm/numa: Do not initialize nodes twice arch/ia64/mm/discontig.c | 6 +- arch/x86/mm/numa.c | 26 +++---- include/linux/memory_hotplug.h | 120 ++++++++++++++++----------------- mm/internal.h | 2 + mm/memory_hotplug.c | 24 +++---- mm/page_alloc.c | 45 +++++++++++-- 6 files changed, 123 insertions(+), 100 deletions(-) -- 2.25.1

2 4

[PATCH OLK-5.10 v2 00/22] Add error handle for add_disk
by Li Nan 07 Nov '23

07 Nov '23

To make applying the mainline patch easier, reorder the nbd commit about the first_minor check here. Christoph Hellwig (5): block: fold register_disk into device_add_disk block: call blk_integrity_add earlier in device_add_disk block: add the events* attributes to disk_attrs block: fix error unwinding in device_add_disk block: clear ->slave_dir when dropping the main slave_dir reference Luis Chamberlain (4): block: return errors from blk_integrity_add block: return errors from disk_alloc_events block: add error handling for device_add_disk / add_disk block: fix device_add_disk() kobject_create_and_add() error handling Tetsuo Handa (1): block: check minor range in device_add_disk() Wen Yang (1): Revert "Revert "block: nbd: add sanity check for first_minor"" Yu Kuai (3): nbd: fix max value for 'first_minor' nbd: fix possible overflow for 'first_minor' in nbd_dev_add() block: fix memory leak for elevator on add_disk failure Zhang Wensheng (1): nbd: fix possible overflow on 'first_minor' in nbd_dev_add() Zhong Jinghua (7): nbd: Reorganize the messy commit log about the first_minor check block: return errors from blk_register_region block: Fix the kabi change in device_add_disk block: Fix the kabi change on blk_register_region block: call blk_get_queue earlier in __device_add_disk block: Fix minor range check in device_add_disk() block: Set memalloc_noio to false in the error path block/blk.h | 5 +- include/linux/genhd.h | 13 +++ block/blk-integrity.c | 12 ++- block/genhd.c | 246 +++++++++++++++++++++++++----------------- drivers/block/nbd.c | 15 +-- 5 files changed, 175 insertions(+), 116 deletions(-) -- 2.39.2

2 23

[PATCH OLK-5.10 v2 0/2] cpufreq: Abort show()/store() for half-initialized policies
by Jinjie Ruan 07 Nov '23

07 Nov '23

Fix cpu offline but cat scaling_cur_freq is the same as online bug. Schspa Shi (2): cpufreq: Abort show()/store() for half-initialized policies cpufreq: make interface functions and lock holding state clear drivers/cpufreq/cpufreq.c | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) -- 2.34.1

2 3

[PATCH openEuler-1.0-LTS v2 0/2] cpufreq: Abort show()/store() for half-initialized policies
by Jinjie Ruan 07 Nov '23

07 Nov '23

Fix cpu offline but cat scaling_cur_freq is the same as online bug. Changes in v2: - Fix the CHECKPATCH warnings. Schspa Shi (2): cpufreq: Abort show()/store() for half-initialized policies cpufreq: make interface functions and lock holding state clear drivers/cpufreq/cpufreq.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-) -- 2.34.1

2 3

[PATCH OLK-5.10 0/2] cpufreq: Abort show()/store() for half-initialized policies
by Jinjie Ruan 07 Nov '23

07 Nov '23

Fix cpu offline but cat scaling_cur_freq is the same as online bug. Schspa Shi (2): cpufreq: Abort show()/store() for half-initialized policies cpufreq: make interface functions and lock holding state clear drivers/cpufreq/cpufreq.c | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-) -- 2.34.1

2 3

[PATCH openEuler-1.0-LTS 0/2] cpufreq: Abort show()/store() for half-initialized policies
by Jinjie Ruan 07 Nov '23

07 Nov '23

Fix cpu offline but cat scaling_cur_freq is the same as online bug. Schspa Shi (2): cpufreq: Abort show()/store() for half-initialized policies cpufreq: make interface functions and lock holding state clear drivers/cpufreq/cpufreq.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-) -- 2.34.1

2 3

[PATCH v2 openEuler-1.0-LTS 0/3] handle uninitialized numa nodes gracefully
by Wupeng Ma 07 Nov '23

07 Nov '23

From: Ma Wupeng <mawupeng1(a)huawei.com> handle uninitialized numa nodes gracefully. Changelog since v1: - add bugfix patch #3. Michal Hocko (2): mm, memory_hotplug: make arch_alloc_nodedata independent on CONFIG_MEMORY_HOTPLUG mm: handle uninitialized numa nodes gracefully Oscar Salvador (1): arch/x86/mm/numa: Do not initialize nodes twice arch/ia64/mm/discontig.c | 6 +- arch/x86/mm/numa.c | 26 +++---- include/linux/memory_hotplug.h | 120 ++++++++++++++++----------------- mm/internal.h | 2 + mm/memory_hotplug.c | 24 +++---- mm/page_alloc.c | 45 +++++++++++-- 6 files changed, 123 insertions(+), 100 deletions(-) -- 2.25.1

2 4

[PATCH OLK-5.10 v3] blk-mq: avoid housekeeping CPUs scheduling a worker on a non-housekeeping CPU
by Xiongfeng Wang 07 Nov '23

07 Nov '23

hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I7ZBTV CVE: N/A ---------------------------------------------------- When NOHZ_FULL is enabled, such as in HPC situation, CPUs are divided into housekeeping CPUs and non-housekeeping CPUs. Non-housekeeping CPUs are NOHZ_FULL CPUs and are often monopolized by the userspace process, such HPC application process. Any sort of interruption is not expected. blk_mq_hctx_next_cpu() selects each cpu in 'hctx->cpumask' alternately to schedule the work thread blk_mq_run_work_fn(). When 'hctx->cpumask' contains housekeeping CPU and non-housekeeping CPU at the same time, a housekeeping CPU, which want to request a IO, may schedule a worker on a non-housekeeping CPU. This may affect the performance of the userspace application running on non-housekeeping CPUs. So let's just schedule the worker thread on the current CPU when the current CPU is housekeeping CPU. Signed-off-by: Xiongfeng Wang <wangxiongfeng2(a)huawei.com> --- block/blk-mq.c | 12 +++++++++++- include/linux/sched/isolation.h | 2 ++ kernel/sched/isolation.c | 8 ++++++++ 3 files changed, 21 insertions(+), 1 deletion(-) diff --git a/block/blk-mq.c b/block/blk-mq.c index 1d1200afb771..3245809453b8 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -23,6 +23,9 @@ #include <linux/sched/sysctl.h> #include <linux/sched/topology.h> #include <linux/sched/signal.h> +#ifndef __GENKSYMS__ +#include <linux/sched/isolation.h> +#endif #include <linux/delay.h> #include <linux/crash_dump.h> #include <linux/prefetch.h> @@ -1676,6 +1679,8 @@ static int blk_mq_hctx_next_cpu(struct blk_mq_hw_ctx *hctx) static void __blk_mq_delay_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async, unsigned long msecs) { + int work_cpu; + if (unlikely(blk_mq_hctx_stopped(hctx))) return; @@ -1697,7 +1702,12 @@ static void __blk_mq_delay_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async, if (!percpu_ref_tryget(&hctx->queue->q_usage_counter)) return; - kblockd_mod_delayed_work_on(blk_mq_hctx_next_cpu(hctx), &hctx->run_work, + if (enhanced_isolcpus && tick_nohz_full_enabled() && + housekeeping_cpu(raw_smp_processor_id(), HK_FLAG_WQ)) + work_cpu = raw_smp_processor_id(); + else + work_cpu = blk_mq_hctx_next_cpu(hctx); + kblockd_mod_delayed_work_on(work_cpu, &hctx->run_work, msecs_to_jiffies(msecs)); percpu_ref_put(&hctx->queue->q_usage_counter); } diff --git a/include/linux/sched/isolation.h b/include/linux/sched/isolation.h index cc9f393e2a70..2f93081ad7a0 100644 --- a/include/linux/sched/isolation.h +++ b/include/linux/sched/isolation.h @@ -18,6 +18,7 @@ enum hk_flags { }; #ifdef CONFIG_CPU_ISOLATION +extern bool enhanced_isolcpus; DECLARE_STATIC_KEY_FALSE(housekeeping_overridden); extern int housekeeping_any_cpu(enum hk_flags flags); extern const struct cpumask *housekeeping_cpumask(enum hk_flags flags); @@ -28,6 +29,7 @@ extern void __init housekeeping_init(void); #else +#define enhanced_isolcpus 0 static inline int housekeeping_any_cpu(enum hk_flags flags) { return smp_processor_id(); diff --git a/kernel/sched/isolation.c b/kernel/sched/isolation.c index 5a6ea03f9882..785ef5201116 100644 --- a/kernel/sched/isolation.c +++ b/kernel/sched/isolation.c @@ -198,3 +198,11 @@ static int __init housekeeping_isolcpus_setup(char *str) return housekeeping_setup(str, flags); } __setup("isolcpus=", housekeeping_isolcpus_setup); + +bool enhanced_isolcpus; +static int __init enhanced_isolcpus_setup(char *str) +{ + enhanced_isolcpus = true; + return 0; +} +__setup("enhanced_isolcpus", enhanced_isolcpus_setup); -- 2.20.1

2 1

[PATCH OLK-5.10 00/67] backport SME feature from mainline
by Wang ShaoBo 07 Nov '23

07 Nov '23

From: w00502114 <bobo.shaobowang(a)huawei.com> backport SME feature from mainline Alexandru Elisei (1): arm64: Do not trap PMSNEVFR_EL1 Marc Zyngier (10): KVM: arm64: Let vcpu_sve_pffr() handle HYP VAs KVM: arm64: Provide KVM's own save/restore SVE primitives KVM: arm64: Use {read,write}_sysreg_el1 to access ZCR_EL1 KVM: arm64: Introduce vcpu_sve_vq() helper arm64: sve: Provide a conditional update accessor for ZCR_ELx KVM: arm64: Map SVE context at EL2 when available KVM: arm64: Rework SVE host-save/guest-restore KVM: arm64: Save guest's ZCR_EL1 before saving the FPSIMD state KVM: arm64: Save/restore SVE state for nVHE KVM: arm64: Always start with clearing SME flag on load Mark Brown (54): arm64/sve: Remove redundant system_supports_sve() tests arm64/sve: Add compile time checks for SVE hooks in generic functions arm64/sve: Rework SVE access trap to convert state in registers arm64/sve: Split _sve_flush macro into separate Z and predicate flushes arm64/sve: Skip flushing Z registers with 128 bit vectors arm64/sve: Use the sve_flush macros in sve_load_from_fpsimd_state() arm64/sve: Remove sve_load_from_fpsimd_state() arm64/sve: Better handle failure to allocate SVE register storage arm64/fp: Reindent fpsimd_save() arm64/sve: Make access to FFR optional arm64/sve: Rename find_supported_vector_length() arm64/sve: Use accessor functions for vector lengths in thread_struct arm64/sve: Put system wide vector length information into structs arm64/sve: Explicitly load vector length when restoring SVE state arm64/sve: Track vector lengths for tasks in an array arm64/sve: Make sysctl interface for SVE reusable by SME arm64/sve: Generalise vector length configuration prctl() for SME arm64/sve: Minor clarification of ABI documentation arm64: cpufeature: Always specify and use a field width for capabilities arm64/sme: Provide ABI documentation for SME arm64/sme: System register and exception syndrome definitions arm64/sme: Manually encode SME instructions arm64: Disable fine grained traps on boot arm64/sme: Early CPU setup for SME arm64/sme: Basic enumeration support arm64/sme: Identify supported SME vector lengths at boot arm64/sme: Implement sysctl to set the default vector length arm64/sme: Implement vector length configuration prctl()s arm64/sme: Implement support for TPIDR2 arm64/sme: Implement SVCR context switching arm64/sme: Implement streaming SVE context switching arm64/sme: Implement ZA context switching arm64/sme: Implement traps and syscall handling for SME arm64/sme: Disable ZA and streaming mode when handling signals arm64/sme: Implement streaming SVE signal handling arm64/sme: Implement ZA signal handling arm64/sme: Implement ptrace support for streaming mode SVE registers arm64/sme: Add ptrace support for ZA arm64/sme: Disable streaming mode and ZA when flushing CPU state arm64/sme: Save and restore streaming mode over EFI runtime calls arm64/sme: Provide Kconfig for SME arm64/sme: Add ID_AA64SMFR0_EL1 to __read_sysreg_by_encoding() arm64/sme: More sensibly define the size for the ZA register set KVM: arm64: Hide SME system registers from guests KVM: arm64: Trap SME usage in guest KVM: arm64: Handle SME host state when running guests arm64/fp: Make SVE and SME length register definition match architecture arm64/fp: Rename SVE and SME LEN field name to _WIDTH arm64/sme: Drop SYS_ from SMIDR_EL1 defines arm64/sme: Standardise bitfield names for SVCR arm64/sme: Remove _EL0 from name of SVCR - FIXME sysreg.h arm64/sme: Fix tests for 0b1111 value ID registers arm64/sme: Fix SVE/SME typo in ABI documentation arm64/sme: Fix EFI save/restore Wan Jiabing (1): arm64/sme: Fix NULL check after kzalloc Xiaofei Tan (1): arm64: sve: Provide sve_cond_update_zcr_vq fallback when !ARM64_SVE Documentation/arm64/elf_hwcaps.rst | 34 + Documentation/arm64/index.rst | 1 + Documentation/arm64/sme.rst | 428 ++++++++++ Documentation/arm64/sve.rst | 72 +- arch/arm64/Kconfig | 11 + arch/arm64/include/asm/cpu.h | 4 + arch/arm64/include/asm/cpucaps.h | 2 + arch/arm64/include/asm/cpufeature.h | 25 + arch/arm64/include/asm/esr.h | 13 +- arch/arm64/include/asm/exception.h | 1 + arch/arm64/include/asm/fpsimd.h | 262 ++++++- arch/arm64/include/asm/fpsimdmacros.h | 108 ++- arch/arm64/include/asm/hwcap.h | 8 + arch/arm64/include/asm/kvm_arm.h | 1 + arch/arm64/include/asm/kvm_host.h | 14 +- arch/arm64/include/asm/kvm_hyp.h | 2 + arch/arm64/include/asm/processor.h | 78 +- arch/arm64/include/asm/signal_common.h | 19 +- arch/arm64/include/asm/sysreg.h | 79 +- arch/arm64/include/asm/thread_info.h | 4 +- arch/arm64/include/uapi/asm/hwcap.h | 8 + arch/arm64/include/uapi/asm/ptrace.h | 69 +- arch/arm64/include/uapi/asm/sigcontext.h | 55 +- arch/arm64/kernel/cpufeature.c | 282 +++++-- arch/arm64/kernel/cpuinfo.c | 13 + arch/arm64/kernel/entry-common.c | 11 + arch/arm64/kernel/entry-fpsimd.S | 72 +- arch/arm64/kernel/fpsimd.c | 944 ++++++++++++++++++----- arch/arm64/kernel/head.S | 87 +++ arch/arm64/kernel/process.c | 45 +- arch/arm64/kernel/ptrace.c | 373 +++++++-- arch/arm64/kernel/signal.c | 175 ++++- arch/arm64/kernel/syscall.c | 29 +- arch/arm64/kernel/traps.c | 1 + arch/arm64/kvm/fpsimd.c | 70 +- arch/arm64/kvm/guest.c | 6 +- arch/arm64/kvm/hyp/fpsimd.S | 12 + arch/arm64/kvm/hyp/include/hyp/switch.h | 75 +- arch/arm64/kvm/hyp/nvhe/switch.c | 38 +- arch/arm64/kvm/hyp/vhe/switch.c | 10 +- arch/arm64/kvm/reset.c | 14 +- arch/arm64/kvm/sys_regs.c | 8 +- include/uapi/linux/elf.h | 2 + include/uapi/linux/prctl.h | 9 + kernel/sys.c | 12 + 45 files changed, 3126 insertions(+), 460 deletions(-) create mode 100644 Documentation/arm64/sme.rst -- 2.25.1

1 67

[PATCH OLK-5.10 00/23] arm64/perf: Enable branch stack sampling
by Junhao He 07 Nov '23

07 Nov '23

This series enables perf branch stack sampling support on arm64 platform via a new arch feature called Branch Record Buffer Extension (BRBE). All relevant register definitions could be accessed here. Anshuman Khandual (16): perf tools: Add missing branch_sample_type to perf_event_attr__fprintf() perf: Add irq and exception return branch types perf: Add system error and not in transaction branch types perf: Extend branch type classification perf: Capture branch privilege information perf: Add PERF_BR_NEW_ARCH_[N] map for BRBE on arm64 platform perf branch: Add system error and not in transaction branch types perf branch: Extend branch type classification perf branch: Add branch privilege information request flag perf branch: Add PERF_BR_NEW_ARCH_[N] map for BRBE on arm64 platform perf: Consolidate branch sample filter helpers perf record: Add remaining branch filters: "no_cycles", "no_flags" & "hw_index" arm64/sysreg: Add BRBE registers and fields drivers: perf: arm_pmu: Add new sched_task() callback drivers: perf: arm_pmuv3: Enable branch stack sampling framework drivers: perf: arm_pmuv3: Enable branch stack sampling via FEAT_BRBE James Clark (5): perf evsel: Add error message for unsupported branch stack cases perf session: Print branch stack entry type in --dump-raw-trace perf script: Refactor branch stack printing perf script: Output branch sample type perf branch: Fix interpretation of branch records Mark Brown (1): arm64/sysreg: Introduce helpers for access to sysreg fields Sandipan Das (1): perf/core: Add speculation info to branch entries Documentation/arch/arm64/brbe.rst | 145 ++++ arch/arm64/include/asm/perf_event.h | 60 ++ arch/arm64/include/asm/sysreg.h | 499 ++++++++++++++ arch/arm64/kernel/head.S | 48 ++ arch/arm64/kernel/perf_event.c | 90 ++- arch/x86/events/intel/lbr.c | 4 +- drivers/perf/Kconfig | 11 + drivers/perf/Makefile | 1 + drivers/perf/arm_brbe.c | 779 ++++++++++++++++++++++ drivers/perf/arm_brbe.h | 261 ++++++++ drivers/perf/arm_pmu.c | 34 +- include/linux/perf/arm_pmu.h | 28 +- include/linux/perf_event.h | 27 + include/uapi/linux/perf_event.h | 51 +- kernel/events/core.c | 9 +- tools/include/uapi/linux/perf_event.h | 38 +- tools/perf/builtin-script.c | 28 +- tools/perf/util/branch.c | 72 +- tools/perf/util/branch.h | 8 +- tools/perf/util/evsel.c | 4 + tools/perf/util/parse-branch-options.c | 4 + tools/perf/util/perf_event_attr_fprintf.c | 2 +- tools/perf/util/session.c | 5 +- 23 files changed, 2172 insertions(+), 36 deletions(-) create mode 100644 Documentation/arch/arm64/brbe.rst create mode 100644 drivers/perf/arm_brbe.c create mode 100644 drivers/perf/arm_brbe.h -- 2.33.0

2 24

[PATCH OLK-5.10 00/67] backport SME feature from mainline
by w00502114 07 Nov '23

07 Nov '23

backport SME feature from mainline Alexandru Elisei (1): arm64: Do not trap PMSNEVFR_EL1 Marc Zyngier (10): KVM: arm64: Let vcpu_sve_pffr() handle HYP VAs KVM: arm64: Provide KVM's own save/restore SVE primitives KVM: arm64: Use {read,write}_sysreg_el1 to access ZCR_EL1 KVM: arm64: Introduce vcpu_sve_vq() helper arm64: sve: Provide a conditional update accessor for ZCR_ELx KVM: arm64: Map SVE context at EL2 when available KVM: arm64: Rework SVE host-save/guest-restore KVM: arm64: Save guest's ZCR_EL1 before saving the FPSIMD state KVM: arm64: Save/restore SVE state for nVHE KVM: arm64: Always start with clearing SME flag on load Mark Brown (54): arm64/sve: Remove redundant system_supports_sve() tests arm64/sve: Add compile time checks for SVE hooks in generic functions arm64/sve: Rework SVE access trap to convert state in registers arm64/sve: Split _sve_flush macro into separate Z and predicate flushes arm64/sve: Skip flushing Z registers with 128 bit vectors arm64/sve: Use the sve_flush macros in sve_load_from_fpsimd_state() arm64/sve: Remove sve_load_from_fpsimd_state() arm64/sve: Better handle failure to allocate SVE register storage arm64/fp: Reindent fpsimd_save() arm64/sve: Make access to FFR optional arm64/sve: Rename find_supported_vector_length() arm64/sve: Use accessor functions for vector lengths in thread_struct arm64/sve: Put system wide vector length information into structs arm64/sve: Explicitly load vector length when restoring SVE state arm64/sve: Track vector lengths for tasks in an array arm64/sve: Make sysctl interface for SVE reusable by SME arm64/sve: Generalise vector length configuration prctl() for SME arm64/sve: Minor clarification of ABI documentation arm64: cpufeature: Always specify and use a field width for capabilities arm64/sme: Provide ABI documentation for SME arm64/sme: System register and exception syndrome definitions arm64/sme: Manually encode SME instructions arm64: Disable fine grained traps on boot arm64/sme: Early CPU setup for SME arm64/sme: Basic enumeration support arm64/sme: Identify supported SME vector lengths at boot arm64/sme: Implement sysctl to set the default vector length arm64/sme: Implement vector length configuration prctl()s arm64/sme: Implement support for TPIDR2 arm64/sme: Implement SVCR context switching arm64/sme: Implement streaming SVE context switching arm64/sme: Implement ZA context switching arm64/sme: Implement traps and syscall handling for SME arm64/sme: Disable ZA and streaming mode when handling signals arm64/sme: Implement streaming SVE signal handling arm64/sme: Implement ZA signal handling arm64/sme: Implement ptrace support for streaming mode SVE registers arm64/sme: Add ptrace support for ZA arm64/sme: Disable streaming mode and ZA when flushing CPU state arm64/sme: Save and restore streaming mode over EFI runtime calls arm64/sme: Provide Kconfig for SME arm64/sme: Add ID_AA64SMFR0_EL1 to __read_sysreg_by_encoding() arm64/sme: More sensibly define the size for the ZA register set KVM: arm64: Hide SME system registers from guests KVM: arm64: Trap SME usage in guest KVM: arm64: Handle SME host state when running guests arm64/fp: Make SVE and SME length register definition match architecture arm64/fp: Rename SVE and SME LEN field name to _WIDTH arm64/sme: Drop SYS_ from SMIDR_EL1 defines arm64/sme: Standardise bitfield names for SVCR arm64/sme: Remove _EL0 from name of SVCR - FIXME sysreg.h arm64/sme: Fix tests for 0b1111 value ID registers arm64/sme: Fix SVE/SME typo in ABI documentation arm64/sme: Fix EFI save/restore Wan Jiabing (1): arm64/sme: Fix NULL check after kzalloc Xiaofei Tan (1): arm64: sve: Provide sve_cond_update_zcr_vq fallback when !ARM64_SVE Documentation/arm64/elf_hwcaps.rst | 34 + Documentation/arm64/index.rst | 1 + Documentation/arm64/sme.rst | 428 ++++++++++ Documentation/arm64/sve.rst | 72 +- arch/arm64/Kconfig | 11 + arch/arm64/include/asm/cpu.h | 4 + arch/arm64/include/asm/cpucaps.h | 2 + arch/arm64/include/asm/cpufeature.h | 25 + arch/arm64/include/asm/esr.h | 13 +- arch/arm64/include/asm/exception.h | 1 + arch/arm64/include/asm/fpsimd.h | 262 ++++++- arch/arm64/include/asm/fpsimdmacros.h | 108 ++- arch/arm64/include/asm/hwcap.h | 8 + arch/arm64/include/asm/kvm_arm.h | 1 + arch/arm64/include/asm/kvm_host.h | 14 +- arch/arm64/include/asm/kvm_hyp.h | 2 + arch/arm64/include/asm/processor.h | 78 +- arch/arm64/include/asm/signal_common.h | 19 +- arch/arm64/include/asm/sysreg.h | 79 +- arch/arm64/include/asm/thread_info.h | 4 +- arch/arm64/include/uapi/asm/hwcap.h | 8 + arch/arm64/include/uapi/asm/ptrace.h | 69 +- arch/arm64/include/uapi/asm/sigcontext.h | 55 +- arch/arm64/kernel/cpufeature.c | 282 +++++-- arch/arm64/kernel/cpuinfo.c | 13 + arch/arm64/kernel/entry-common.c | 11 + arch/arm64/kernel/entry-fpsimd.S | 72 +- arch/arm64/kernel/fpsimd.c | 944 ++++++++++++++++++----- arch/arm64/kernel/head.S | 87 +++ arch/arm64/kernel/process.c | 45 +- arch/arm64/kernel/ptrace.c | 373 +++++++-- arch/arm64/kernel/signal.c | 175 ++++- arch/arm64/kernel/syscall.c | 29 +- arch/arm64/kernel/traps.c | 1 + arch/arm64/kvm/fpsimd.c | 70 +- arch/arm64/kvm/guest.c | 6 +- arch/arm64/kvm/hyp/fpsimd.S | 12 + arch/arm64/kvm/hyp/include/hyp/switch.h | 75 +- arch/arm64/kvm/hyp/nvhe/switch.c | 38 +- arch/arm64/kvm/hyp/vhe/switch.c | 10 +- arch/arm64/kvm/reset.c | 14 +- arch/arm64/kvm/sys_regs.c | 8 +- include/uapi/linux/elf.h | 2 + include/uapi/linux/prctl.h | 9 + kernel/sys.c | 12 + 45 files changed, 3126 insertions(+), 460 deletions(-) create mode 100644 Documentation/arm64/sme.rst -- 2.25.1

1 67

[PATCH OLK-5.10 v2] blk-mq: avoid housekeeping CPUs scheduling a worker on a non-housekeeping CPU
by Xiongfeng Wang 07 Nov '23

07 Nov '23

hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I7ZBTV CVE: N/A ---------------------------------------------------- When NOHZ_FULL is enabled, such as in HPC situation, CPUs are divided into housekeeping CPUs and non-housekeeping CPUs. Non-housekeeping CPUs are NOHZ_FULL CPUs and are often monopolized by the userspace process, such HPC application process. Any sort of interruption is not expected. blk_mq_hctx_next_cpu() selects each cpu in 'hctx->cpumask' alternately to schedule the work thread blk_mq_run_work_fn(). When 'hctx->cpumask' contains housekeeping CPU and non-housekeeping CPU at the same time, a housekeeping CPU, which want to request a IO, may schedule a worker on a non-housekeeping CPU. This may affect the performance of the userspace application running on non-housekeeping CPUs. So let's just schedule the worker thread on the current CPU when the current CPU is housekeeping CPU. Signed-off-by: Xiongfeng Wang <wangxiongfeng2(a)huawei.com> --- block/blk-mq.c | 13 ++++++++++++- include/linux/sched/isolation.h | 2 ++ kernel/sched/isolation.c | 8 ++++++++ 3 files changed, 22 insertions(+), 1 deletion(-) diff --git a/block/blk-mq.c b/block/blk-mq.c index 1d1200afb771..b81d4a69056c 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -23,6 +23,9 @@ #include <linux/sched/sysctl.h> #include <linux/sched/topology.h> #include <linux/sched/signal.h> +#ifndef __GENKSYMS__ +#include <linux/sched/isolation.h> +#endif #include <linux/delay.h> #include <linux/crash_dump.h> #include <linux/prefetch.h> @@ -1676,6 +1679,8 @@ static int blk_mq_hctx_next_cpu(struct blk_mq_hw_ctx *hctx) static void __blk_mq_delay_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async, unsigned long msecs) { + int work_cpu; + if (unlikely(blk_mq_hctx_stopped(hctx))) return; @@ -1697,7 +1702,13 @@ static void __blk_mq_delay_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async, if (!percpu_ref_tryget(&hctx->queue->q_usage_counter)) return; - kblockd_mod_delayed_work_on(blk_mq_hctx_next_cpu(hctx), &hctx->run_work, + if (enhanced_isolcpus && tick_nohz_full_enabled() && + housekeeping_cpu(raw_smp_processor_id(), HK_FLAG_WQ)) + work_cpu = smp_processor_id(); + else + work_cpu = blk_mq_hctx_next_cpu(hctx); + + kblockd_mod_delayed_work_on(work_cpu, &hctx->run_work, msecs_to_jiffies(msecs)); percpu_ref_put(&hctx->queue->q_usage_counter); } diff --git a/include/linux/sched/isolation.h b/include/linux/sched/isolation.h index cc9f393e2a70..2f93081ad7a0 100644 --- a/include/linux/sched/isolation.h +++ b/include/linux/sched/isolation.h @@ -18,6 +18,7 @@ enum hk_flags { }; #ifdef CONFIG_CPU_ISOLATION +extern bool enhanced_isolcpus; DECLARE_STATIC_KEY_FALSE(housekeeping_overridden); extern int housekeeping_any_cpu(enum hk_flags flags); extern const struct cpumask *housekeeping_cpumask(enum hk_flags flags); @@ -28,6 +29,7 @@ extern void __init housekeeping_init(void); #else +#define enhanced_isolcpus 0 static inline int housekeeping_any_cpu(enum hk_flags flags) { return smp_processor_id(); diff --git a/kernel/sched/isolation.c b/kernel/sched/isolation.c index 5a6ea03f9882..785ef5201116 100644 --- a/kernel/sched/isolation.c +++ b/kernel/sched/isolation.c @@ -198,3 +198,11 @@ static int __init housekeeping_isolcpus_setup(char *str) return housekeeping_setup(str, flags); } __setup("isolcpus=", housekeeping_isolcpus_setup); + +bool enhanced_isolcpus; +static int __init enhanced_isolcpus_setup(char *str) +{ + enhanced_isolcpus = true; + return 0; +} +__setup("enhanced_isolcpus", enhanced_isolcpus_setup); -- 2.20.1

2 1

[PATCH OLK-5.10 00/22] Add error handle for add_disk
by Li Nan 07 Nov '23

07 Nov '23

To make applying the mainline patch easier, reorder the nbd commit about the first_minor check here. Christoph Hellwig (5): block: fold register_disk into device_add_disk block: call blk_integrity_add earlier in device_add_disk block: add the events* attributes to disk_attrs block: fix error unwinding in device_add_disk block: clear ->slave_dir when dropping the main slave_dir reference Luis Chamberlain (4): block: return errors from blk_integrity_add block: return errors from disk_alloc_events block: add error handling for device_add_disk / add_disk block: fix device_add_disk() kobject_create_and_add() error handling Tetsuo Handa (1): block: check minor range in device_add_disk() Wen Yang (1): Revert "Revert "block: nbd: add sanity check for first_minor"" Yu Kuai (3): nbd: fix max value for 'first_minor' nbd: fix possible overflow for 'first_minor' in nbd_dev_add() block: fix memory leak for elevator on add_disk failure Zhang Wensheng (1): nbd: fix possible overflow on 'first_minor' in nbd_dev_add() Zhong Jinghua (7): nbd: Reorganize the messy commit log about the first_minor check block: return errors from blk_register_region block: Fix the kabi change in device_add_disk block: Fix the kabi change on blk_register_region block: call blk_get_queue earlier in __device_add_disk block: Fix minor range check in device_add_disk() block: Set memalloc_noio to false in the error path block/blk.h | 5 +- include/linux/genhd.h | 13 +++ block/blk-integrity.c | 12 ++- block/genhd.c | 246 +++++++++++++++++++++++++----------------- drivers/block/nbd.c | 15 +-- 5 files changed, 175 insertions(+), 116 deletions(-) -- 2.39.2

2 23

[PATCH OLK-5.10 v2 00/42] Add error handle for add_disk
by Li Nan 07 Nov '23

07 Nov '23

Christoph Hellwig (14): block: fold register_disk into device_add_disk block: call blk_integrity_add earlier in device_add_disk block: add the events* attributes to disk_attrs block: fix error unwinding in device_add_disk block: clear ->slave_dir when dropping the main slave_dir reference block: add blk_alloc_disk and blk_cleanup_disk APIs blk-mq: add the blk_mq_alloc_disk APIs loop: use blk_mq_alloc_disk and blk_cleanup_disk loop: fix order of cleaning up the queue and freeing the tagset nbd: use blk_mq_alloc_disk and blk_cleanup_disk block: add a flag to make put_disk on partially initalized disks safer nvme: use blk_mq_alloc_disk brd: convert to blk_alloc_disk/blk_cleanup_disk ubi: use blk_mq_alloc_disk and blk_cleanup_disk Dan Carpenter (1): blk-mq: fix an IS_ERR() vs NULL bug Li Lingfeng (1): nbd: fix uaf in nbd_open Luis Chamberlain (10): block: return errors from blk_integrity_add block: return errors from disk_alloc_events block: add error handling for device_add_disk / add_disk block: fix device_add_disk() kobject_create_and_add() error handling loop: add error handling support for add_disk() nbd: add error handling support for add_disk() nvme: add error handling support for add_disk() block/brd: add error handling support for add_disk() scsi: sr: Add error handling support for add_disk() mtd/ubi/block: add error handling support for add_disk() Tetsuo Handa (1): block: check minor range in device_add_disk() Wang Qing (1): nbd: fix order of cleaning up the queue and freeing the tagset Wen Yang (1): Revert "Revert "block: nbd: add sanity check for first_minor"" Yu Kuai (3): nbd: fix max value for 'first_minor' nbd: fix possible overflow for 'first_minor' in nbd_dev_add() block: fix memory leak for elevator on add_disk failure Zhang Wensheng (1): nbd: fix possible overflow on 'first_minor' in nbd_dev_add() Zhong Jinghua (9): nbd: Reorganize the messy commit log about the first_minor check block: return errors from blk_register_region block: Fix the kabi change in device_add_disk block: Fix the kabi change on blk_register_region block: call blk_get_queue earlier in __device_add_disk block: Fix minor range check in device_add_disk() block: Set memalloc_noio to false in the error path mtd/ubi/block: Fix null pointer dereference issue in error path mtd/ubi/block: Fix uaf problem in ubiblock_cleanup block/blk.h | 5 +- include/linux/blk-mq.h | 12 ++ include/linux/genhd.h | 36 +++++ block/blk-integrity.c | 12 +- block/blk-mq.c | 19 +++ block/genhd.c | 286 ++++++++++++++++++++++++++------------- drivers/block/brd.c | 103 ++++++-------- drivers/block/loop.c | 26 ++-- drivers/block/nbd.c | 75 ++++------ drivers/mtd/ubi/block.c | 76 +++++------ drivers/nvme/host/core.c | 41 +++--- drivers/scsi/sr.c | 7 +- 12 files changed, 413 insertions(+), 285 deletions(-) -- 2.39.2

1 42

[PATCH openEuler-1.0-LTS] crypto: hisilicon/qm - alloc reserve buffer to set and get xqc
by wangyuan 06 Nov '23

06 Nov '23

From: Yu'an Wang <wangyuan46(a)huawei.com> driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8E6ED CVE: NA -------------------------------- If the temporarily applied memory is used to set or get the xqc information, the driver will release the memory after the mailbox timeout. However, the hardware still performs the operation. Therefore, the released memory may be written by hardware. So when load driver, reserve memory is applied for xqc configuration. Signed-off-by: Yu'an Wang <wangyuan46(a)huawei.com> --- drivers/crypto/hisilicon/qm.c | 544 +++++++++++++++------------------- drivers/crypto/hisilicon/qm.h | 13 + 2 files changed, 250 insertions(+), 307 deletions(-) diff --git a/drivers/crypto/hisilicon/qm.c b/drivers/crypto/hisilicon/qm.c index 64e8ef1bf8e7..25bb24b4c131 100644 --- a/drivers/crypto/hisilicon/qm.c +++ b/drivers/crypto/hisilicon/qm.c @@ -50,7 +50,7 @@ #define QM_SQ_TYPE_MASK 0xf -#define QM_SQ_TAIL_IDX(sqc) ((le16_to_cpu((sqc)->w11) >> 6) & 0x1) +#define QM_SQ_TAIL_IDX(sqc) ((le16_to_cpu((sqc).w11) >> 6) & 0x1) /* cqc shift */ #define QM_CQ_HOP_NUM_SHIFT 0 @@ -64,7 +64,7 @@ #define QM_QC_CQE_SIZE 4 -#define QM_CQ_TAIL_IDX(cqc) ((le16_to_cpu((cqc)->w11) >> 6) & 0x1) +#define QM_CQ_TAIL_IDX(cqc) ((le16_to_cpu((cqc).w11) >> 6) & 0x1) /* eqc shift */ #define QM_EQE_AEQE_SIZE (2UL << 12) @@ -217,19 +217,6 @@ #define QM_MK_SQC_DW3_V2(sqe_sz) \ ((QM_Q_DEPTH - 1) | ((u32)ilog2(sqe_sz) << QM_SQ_SQE_SIZE_SHIFT)) -#define INIT_QC_COMMON(qc, base, pasid) do { \ - (qc)->head = 0; \ - (qc)->tail = 0; \ - (qc)->base_l = cpu_to_le32(lower_32_bits(base)); \ - (qc)->base_h = cpu_to_le32(upper_32_bits(base)); \ - (qc)->dw3 = 0; \ - (qc)->w8 = 0; \ - (qc)->rsvd0 = 0; \ - (qc)->pasid = cpu_to_le16(pasid); \ - (qc)->w11 = 0; \ - (qc)->rsvd1 = 0; \ -} while (0) - #define QMC_ALIGN(sz) ALIGN(sz, 32) static int __hisi_qm_start(struct hisi_qm *qm); @@ -537,6 +524,61 @@ static int hisi_qm_mb_read(struct hisi_qm *qm, u64 *base, u8 cmd, u16 queue) return 0; } +/* op 0: set xqc info to hardware, 1: get xqc info from hardware. */ +static int qm_set_and_get_xqc(struct hisi_qm *qm, u8 cmd, void *xqc, u32 qp_id, + bool op) +{ + struct hisi_qm *pf_qm = pci_get_drvdata(pci_physfn(qm->pdev)); + struct qm_mailbox mailbox; + dma_addr_t xqc_dma; + void *tmp_xqc; + size_t size; + int ret; + + switch (cmd) { + case QM_MB_CMD_SQC: + size = sizeof(struct qm_sqc); + tmp_xqc = qm->xqc_buf.sqc; + xqc_dma = qm->xqc_buf.sqc_dma; + break; + case QM_MB_CMD_CQC: + size = sizeof(struct qm_cqc); + tmp_xqc = qm->xqc_buf.cqc; + xqc_dma = qm->xqc_buf.cqc_dma; + break; + case QM_MB_CMD_EQC: + size = sizeof(struct qm_eqc); + tmp_xqc = qm->xqc_buf.eqc; + xqc_dma = qm->xqc_buf.eqc_dma; + break; + case QM_MB_CMD_AEQC: + size = sizeof(struct qm_aeqc); + tmp_xqc = qm->xqc_buf.aeqc; + xqc_dma = qm->xqc_buf.aeqc_dma; + break; + } + + /* No need to judge if master OOO is blocked. */ + if (qm_check_dev_error(pf_qm)) { + dev_err(&qm->pdev->dev, + "QM mailbox operation failed since qm is stop!\n"); + return -EIO; + } + + mutex_lock(&qm->mailbox_lock); + if (!op) + memcpy(tmp_xqc, xqc, size); + + qm_mb_pre_init(&mailbox, cmd, xqc_dma, qp_id, op); + ret = qm_mb_nolock(qm, &mailbox); + if (!ret && op) + memcpy(xqc, tmp_xqc, size); + + mutex_unlock(&qm->mailbox_lock); + + return ret; +} + static void qm_db_v1(struct hisi_qm *qm, u16 qn, u8 cmd, u16 index, u8 priority) { u64 doorbell; @@ -1166,35 +1208,6 @@ static ssize_t qm_cmd_read(struct file *filp, char __user *buffer, return (*pos = len); } -static void *qm_ctx_alloc(struct hisi_qm *qm, size_t ctx_size, - dma_addr_t *dma_addr) -{ - struct device *dev = &qm->pdev->dev; - void *ctx_addr; - - ctx_addr = kzalloc(ctx_size, GFP_KERNEL); - if (!ctx_addr) - return ERR_PTR(-ENOMEM); - - *dma_addr = dma_map_single(dev, ctx_addr, ctx_size, DMA_FROM_DEVICE); - if (dma_mapping_error(dev, *dma_addr)) { - dev_err(dev, "DMA mapping error!\n"); - kfree(ctx_addr); - return ERR_PTR(-ENOMEM); - } - - return ctx_addr; -} - -static void qm_ctx_free(struct hisi_qm *qm, size_t ctx_size, - const void *ctx_addr, dma_addr_t *dma_addr) -{ - struct device *dev = &qm->pdev->dev; - - dma_unmap_single(dev, *dma_addr, ctx_size, DMA_FROM_DEVICE); - kfree(ctx_addr); -} - static int dump_show(struct hisi_qm *qm, void *info, unsigned int info_size, char *info_name) { @@ -1230,21 +1243,11 @@ static int dump_show(struct hisi_qm *qm, void *info, return 0; } -static int qm_dump_sqc_raw(struct hisi_qm *qm, dma_addr_t dma_addr, u16 qp_id) -{ - return hisi_qm_mb_write(qm, QM_MB_CMD_SQC, dma_addr, qp_id, 1); -} - -static int qm_dump_cqc_raw(struct hisi_qm *qm, dma_addr_t dma_addr, u16 qp_id) -{ - return hisi_qm_mb_write(qm, QM_MB_CMD_CQC, dma_addr, qp_id, 1); -} - static int qm_sqc_dump(struct hisi_qm *qm, const char *s) { struct device *dev = &qm->pdev->dev; - struct qm_sqc *sqc, *sqc_curr; - dma_addr_t sqc_dma; + struct qm_sqc *sqc_curr; + struct qm_sqc sqc; u32 qp_id; int ret; @@ -1257,40 +1260,28 @@ static int qm_sqc_dump(struct hisi_qm *qm, const char *s) return -EINVAL; } - sqc = qm_ctx_alloc(qm, sizeof(struct qm_sqc), &sqc_dma); - if (IS_ERR(sqc)) - return PTR_ERR(sqc); - - ret = qm_dump_sqc_raw(qm, sqc_dma, qp_id); - if (ret) { - down_read(&qm->qps_lock); - if (qm->sqc) { - sqc_curr = qm->sqc + qp_id; - - ret = dump_show(qm, sqc_curr, sizeof(struct qm_sqc), - "SOFT SQC"); - if (ret) - dev_info(dev, "Show soft sqc failed!\n"); - } - up_read(&qm->qps_lock); + ret = qm_set_and_get_xqc(qm, QM_MB_CMD_SQC, &sqc, qp_id, 1); + if (!ret) + return dump_show(qm, &sqc, sizeof(struct qm_sqc), "SQC"); - goto mailbox_fail; + down_read(&qm->qps_lock); + if (qm->sqc) { + sqc_curr = qm->sqc + qp_id; + ret = dump_show(qm, sqc_curr, sizeof(struct qm_sqc), + "SOFT SQC"); + if (ret) + dev_info(dev, "Show soft sqc failed!\n"); } + up_read(&qm->qps_lock); - ret = dump_show(qm, sqc, sizeof(struct qm_sqc), "SQC"); - if (ret) - dev_info(dev, "Show hw sqc failed!\n"); - -mailbox_fail: - qm_ctx_free(qm, sizeof(struct qm_sqc), sqc, &sqc_dma); return ret; } static int qm_cqc_dump(struct hisi_qm *qm, const char *s) { struct device *dev = &qm->pdev->dev; - struct qm_cqc *cqc, *cqc_curr; - dma_addr_t cqc_dma; + struct qm_cqc *cqc_curr; + struct qm_cqc cqc; u32 qp_id; int ret; @@ -1303,40 +1294,28 @@ static int qm_cqc_dump(struct hisi_qm *qm, const char *s) return -EINVAL; } - cqc = qm_ctx_alloc(qm, sizeof(struct qm_cqc), &cqc_dma); - if (IS_ERR(cqc)) - return PTR_ERR(cqc); - - ret = qm_dump_cqc_raw(qm, cqc_dma, qp_id); - if (ret) { - down_read(&qm->qps_lock); - if (qm->cqc) { - cqc_curr = qm->cqc + qp_id; + ret = qm_set_and_get_xqc(qm, QM_MB_CMD_CQC, &cqc, qp_id, 1); + if (!ret) + return dump_show(qm, &cqc, sizeof(struct qm_cqc), "CQC"); - ret = dump_show(qm, cqc_curr, sizeof(struct qm_cqc), - "SOFT CQC"); - if (ret) - dev_info(dev, "Show soft cqc failed!\n"); - } - up_read(&qm->qps_lock); + down_read(&qm->qps_lock); + if (qm->cqc) { + cqc_curr = qm->cqc + qp_id; - goto mailbox_fail; + ret = dump_show(qm, cqc_curr, sizeof(struct qm_cqc), + "SOFT CQC"); + if (ret) + dev_info(dev, "Show soft cqc failed!\n"); } + up_read(&qm->qps_lock); - ret = dump_show(qm, cqc, sizeof(struct qm_cqc), "CQC"); - if (ret) - dev_info(dev, "Show hw cqc failed!\n"); - -mailbox_fail: - qm_ctx_free(qm, sizeof(struct qm_cqc), cqc, &cqc_dma); return ret; } static int qm_eqc_dump(struct hisi_qm *qm, char *s) { struct device *dev = &qm->pdev->dev; - struct qm_eqc *eqc; - dma_addr_t eqc_dma; + struct qm_eqc eqc; int ret; if (strsep(&s, " ")) { @@ -1344,28 +1323,17 @@ static int qm_eqc_dump(struct hisi_qm *qm, char *s) return -EINVAL; } - eqc = qm_ctx_alloc(qm, sizeof(struct qm_eqc), &eqc_dma); - if (IS_ERR(eqc)) - return PTR_ERR(eqc); - - ret = hisi_qm_mb_write(qm, QM_MB_CMD_EQC, eqc_dma, 0, 1); - if (ret) - goto mailbox_fail; - - ret = dump_show(qm, eqc, sizeof(struct qm_eqc), "EQC"); + ret = qm_set_and_get_xqc(qm, QM_MB_CMD_EQC, &eqc, 0, 1); if (ret) - dev_info(dev, "Show eqc failed!\n"); + return ret; -mailbox_fail: - qm_ctx_free(qm, sizeof(struct qm_eqc), eqc, &eqc_dma); - return ret; + return dump_show(qm, &eqc, sizeof(struct qm_eqc), "EQC"); } static int qm_aeqc_dump(struct hisi_qm *qm, char *s) { struct device *dev = &qm->pdev->dev; - struct qm_aeqc *aeqc; - dma_addr_t aeqc_dma; + struct qm_aeqc aeqc; int ret; if (strsep(&s, " ")) { @@ -1373,21 +1341,11 @@ static int qm_aeqc_dump(struct hisi_qm *qm, char *s) return -EINVAL; } - aeqc = qm_ctx_alloc(qm, sizeof(struct qm_aeqc), &aeqc_dma); - if (IS_ERR(aeqc)) - return PTR_ERR(aeqc); - - ret = hisi_qm_mb_write(qm, QM_MB_CMD_AEQC, aeqc_dma, 0, 1); + ret = qm_set_and_get_xqc(qm, QM_MB_CMD_AEQC, &aeqc, 0, 1); if (ret) - goto mailbox_fail; - - ret = dump_show(qm, aeqc, sizeof(struct qm_aeqc), "AEQC"); - if (ret) - dev_info(dev, "Show hw aeqc failed!\n"); + return ret; -mailbox_fail: - qm_ctx_free(qm, sizeof(struct qm_aeqc), aeqc, &aeqc_dma); - return ret; + return dump_show(qm, &aeqc, sizeof(struct qm_aeqc), "AEQC"); } static int q_dump_param_parse(struct hisi_qm *qm, char *s, @@ -1432,7 +1390,6 @@ static int q_dump_param_parse(struct hisi_qm *qm, char *s, static int qm_sq_dump(struct hisi_qm *qm, char *s) { - struct device *dev = &qm->pdev->dev; struct hisi_qp *qp; u32 qp_id, sqe_id; void *sqe_curr; @@ -1447,16 +1404,11 @@ static int qm_sq_dump(struct hisi_qm *qm, char *s) memset(sqe_curr + qm->debug.sqe_mask_offset, SQE_ADDR_MASK, qm->debug.sqe_mask_len); - ret = dump_show(qm, sqe_curr, qm->sqe_size, "SQE"); - if (ret) - dev_info(dev, "Show sqe failed!\n"); - - return ret; + return dump_show(qm, sqe_curr, qm->sqe_size, "SQE"); } static int qm_cq_dump(struct hisi_qm *qm, char *s) { - struct device *dev = &qm->pdev->dev; struct qm_cqe *cqe_curr; struct hisi_qp *qp; u32 qp_id, cqe_id; @@ -1468,11 +1420,8 @@ static int qm_cq_dump(struct hisi_qm *qm, char *s) qp = &qm->qp_array[qp_id]; cqe_curr = qp->cqe + cqe_id; - ret = dump_show(qm, cqe_curr, sizeof(struct qm_cqe), "CQE"); - if (ret) - dev_info(dev, "Show cqe failed!\n"); - return ret; + return dump_show(qm, cqe_curr, sizeof(struct qm_cqe), "CQE"); } static int qm_eq_dump(struct hisi_qm *qm, const char *s) @@ -1895,79 +1844,46 @@ EXPORT_SYMBOL_GPL(hisi_qm_release_qp); static int qm_sq_ctx_cfg(struct hisi_qp *qp, int qp_id, int pasid) { struct hisi_qm *qm = qp->qm; - struct device *dev = &qm->pdev->dev; enum qm_hw_ver ver = qm->ver; - struct qm_sqc *sqc; - dma_addr_t sqc_dma; - int ret; + struct qm_sqc sqc = {0}; - sqc = kzalloc(sizeof(struct qm_sqc), GFP_KERNEL); - if (!sqc) - return -ENOMEM; - sqc_dma = dma_map_single(dev, sqc, sizeof(struct qm_sqc), - DMA_TO_DEVICE); - if (dma_mapping_error(dev, sqc_dma)) { - kfree(sqc); - return -ENOMEM; - } - - INIT_QC_COMMON(sqc, qp->sqe_dma, pasid); if (ver == QM_HW_V1) { - sqc->dw3 = cpu_to_le32(QM_MK_SQC_DW3_V1(0, 0, 0, qm->sqe_size)); - sqc->w8 = cpu_to_le16(QM_Q_DEPTH - 1); + sqc.dw3 = cpu_to_le32(QM_MK_SQC_DW3_V1(0, 0, 0, qm->sqe_size)); + sqc.w8 = cpu_to_le16(QM_Q_DEPTH - 1); } else if (ver == QM_HW_V2) { - sqc->dw3 = cpu_to_le32(QM_MK_SQC_DW3_V2(qm->sqe_size)); - sqc->w8 = 0; /* rand_qc */ + sqc.dw3 = cpu_to_le32(QM_MK_SQC_DW3_V2(qm->sqe_size)); + sqc.w8 = 0; /* rand_qc */ } - sqc->cq_num = cpu_to_le16(qp_id); - sqc->w13 = cpu_to_le16(QM_MK_SQC_W13(0, 1, qp->alg_type)); - - ret = hisi_qm_mb_write(qm, QM_MB_CMD_SQC, sqc_dma, qp_id, 0); - - dma_unmap_single(dev, sqc_dma, sizeof(struct qm_sqc), DMA_TO_DEVICE); - kfree(sqc); + sqc.cq_num = cpu_to_le16(qp_id); + sqc.w13 = cpu_to_le16(QM_MK_SQC_W13(0, 1, qp->alg_type)); + sqc.base_l = cpu_to_le32(lower_32_bits(qp->sqe_dma)); + sqc.base_h = cpu_to_le32(upper_32_bits(qp->sqe_dma)); + sqc.pasid = cpu_to_le16(pasid); - return ret; + return qm_set_and_get_xqc(qm, QM_MB_CMD_SQC, &sqc, qp_id, 0); } static int qm_cq_ctx_cfg(struct hisi_qp *qp, int qp_id, int pasid) { struct hisi_qm *qm = qp->qm; - struct device *dev = &qm->pdev->dev; enum qm_hw_ver ver = qm->ver; - struct qm_cqc *cqc; - dma_addr_t cqc_dma; - int ret; + struct qm_cqc cqc = {0}; - cqc = kzalloc(sizeof(struct qm_cqc), GFP_KERNEL); - if (!cqc) - return -ENOMEM; - - cqc_dma = dma_map_single(dev, cqc, sizeof(struct qm_cqc), - DMA_TO_DEVICE); - if (dma_mapping_error(dev, cqc_dma)) { - kfree(cqc); - return -ENOMEM; - } - - INIT_QC_COMMON(cqc, qp->cqe_dma, pasid); if (ver == QM_HW_V1) { - cqc->dw3 = cpu_to_le32(QM_MK_CQC_DW3_V1(0, 0, 0, + cqc.dw3 = cpu_to_le32(QM_MK_CQC_DW3_V1(0, 0, 0, QM_QC_CQE_SIZE)); - cqc->w8 = cpu_to_le16(QM_Q_DEPTH - 1); + cqc.w8 = cpu_to_le16(QM_Q_DEPTH - 1); } else if (ver == QM_HW_V2) { - cqc->dw3 = cpu_to_le32(QM_MK_CQC_DW3_V2(QM_QC_CQE_SIZE)); - cqc->w8 = 0; /* rand_qc */ + cqc.dw3 = cpu_to_le32(QM_MK_CQC_DW3_V2(QM_QC_CQE_SIZE)); + cqc.w8 = 0; /* rand_qc */ } - cqc->dw6 = cpu_to_le32(1 << QM_CQ_PHASE_SHIFT | - qp->c_flag << QM_CQ_FLAG_SHIFT); - - ret = hisi_qm_mb_write(qm, QM_MB_CMD_CQC, cqc_dma, qp_id, 0); + cqc.dw6 = cpu_to_le32(1 << QM_CQ_PHASE_SHIFT | + qp->c_flag << QM_CQ_FLAG_SHIFT); + cqc.base_l = cpu_to_le32(lower_32_bits(qp->cqe_dma)); + cqc.base_h = cpu_to_le32(upper_32_bits(qp->cqe_dma)); + cqc.pasid = cpu_to_le16(pasid); - dma_unmap_single(dev, cqc_dma, sizeof(struct qm_cqc), DMA_TO_DEVICE); - kfree(cqc); - - return ret; + return qm_set_and_get_xqc(qm, QM_MB_CMD_CQC, &cqc, qp_id, 0); } static int qm_qp_ctx_cfg(struct hisi_qp *qp, int qp_id, int pasid) @@ -2043,54 +1959,40 @@ static void qp_stop_fail_cb(struct hisi_qp *qp) static void qm_qp_has_no_task(struct hisi_qp *qp) { - size_t size = sizeof(struct qm_sqc) + sizeof(struct qm_cqc); - struct device *dev = &qp->qm->pdev->dev; - struct qm_sqc *sqc; - struct qm_cqc *cqc; - dma_addr_t dma_addr; - void *addr; - int i = 0; - int ret; + struct hisi_qm *qm = qp->qm; + struct device *dev = &qm->pdev->dev; + struct qm_sqc sqc; + struct qm_cqc cqc; + int ret, i = 0; if (qp->qm->err_ini.err_info.is_qm_ecc_mbit || qp->qm->err_ini.err_info.is_dev_ecc_mbit) return; - addr = qm_ctx_alloc(qp->qm, size, &dma_addr); - if (IS_ERR(addr)) { - dev_err(dev, "alloc ctx for sqc and cqc failed!\n"); - return; - } - while (++i) { - ret = qm_dump_sqc_raw(qp->qm, dma_addr, qp->qp_id); + ret = qm_set_and_get_xqc(qm, QM_MB_CMD_SQC, &sqc, qp->qp_id, 1); if (ret) { - dev_err(dev, "Failed to dump sqc!\n"); - break; + dev_err_ratelimited(dev, "Fail to dump sqc!\n"); + return; } - sqc = addr; - ret = qm_dump_cqc_raw(qp->qm, - (dma_addr + sizeof(struct qm_sqc)), qp->qp_id); + ret = qm_set_and_get_xqc(qm, QM_MB_CMD_CQC, &cqc, qp->qp_id, 1); if (ret) { - dev_err(dev, "Failed to dump cqc!\n"); - break; + dev_err_ratelimited(dev, "Fail to dump cqc!\n"); + return; } - cqc = addr + sizeof(struct qm_sqc); - if ((sqc->tail == cqc->tail) && - (QM_SQ_TAIL_IDX(sqc) == QM_CQ_TAIL_IDX(cqc))) - break; + if ((QM_SQ_TAIL_IDX(sqc) == QM_CQ_TAIL_IDX(cqc)) && + (sqc.tail == cqc.tail)) + return; if (i == MAX_WAIT_COUNTS) { - dev_err(dev, "Fail to wait for device stop!\n"); - break; + dev_err(dev, "Fail to empty queue %u!\n", qp->qp_id); + return; } usleep_range(WAIT_PERIOD_US_MIN, WAIT_PERIOD_US_MAX); } - - qm_ctx_free(qp->qm, size, addr, &dma_addr); } static int hisi_qm_stop_qp_nolock(struct hisi_qp *qp) @@ -2597,10 +2499,12 @@ static int hisi_qp_memory_init(struct hisi_qm *qm, size_t dma_size, int id) struct hisi_qp *qp; qp = &qm->qp_array[id]; - qp->qdma.va = dma_alloc_coherent(dev, dma_size, - &qp->qdma.dma, GFP_KERNEL); - if (!qp->qdma.va) + qp->qdma.va = dma_alloc_coherent(dev, dma_size, &qp->qdma.dma, + GFP_KERNEL); + if (!qp->qdma.va) { + dev_err(dev, "Fail to alloc qp dma buf size=%zx\n", dma_size); return -ENOMEM; + } qp->sqe = qp->qdma.va; qp->sqe_dma = qp->qdma.dma; @@ -2613,13 +2517,79 @@ static int hisi_qp_memory_init(struct hisi_qm *qm, size_t dma_size, int id) return 0; } +static int hisi_qp_alloc_memory(struct hisi_qm *qm) +{ + size_t qp_dma_size; + int i, ret; + + qm->qp_array = kcalloc(qm->qp_num, sizeof(struct hisi_qp), GFP_KERNEL); + if (!qm->qp_array) + return -ENOMEM; + + /* one more page for device or qp statuses */ + qp_dma_size = qm->sqe_size * QM_Q_DEPTH + + sizeof(struct cqe) * QM_Q_DEPTH; + qp_dma_size = PAGE_ALIGN(qp_dma_size) + PAGE_SIZE; + for (i = 0; i < qm->qp_num; i++) { + ret = hisi_qp_memory_init(qm, qp_dma_size, i); + if (ret) + goto err_init_qp_mem; + } + + return 0; + +err_init_qp_mem: + hisi_qp_memory_uninit(qm, i); + + return ret; +} + +static void hisi_qm_free_rsv_buf(struct hisi_qm *qm) +{ + struct qm_dma *xqc_dma = &qm->xqc_buf.qcdma; + struct device *dev = &qm->pdev->dev; + + dma_free_coherent(dev, xqc_dma->size, xqc_dma->va, xqc_dma->dma); +} + +static int hisi_qm_alloc_rsv_buf(struct hisi_qm *qm) +{ + struct qm_rsv_buf *xqc_buf = &qm->xqc_buf; + struct qm_dma *xqc_dma = &xqc_buf->qcdma; + struct device *dev = &qm->pdev->dev; + size_t off = 0; + +#define QM_XQC_BUF_INIT(xqc_buf, type) do { \ + (xqc_buf)->type = ((xqc_buf)->qcdma.va + (off)); \ + (xqc_buf)->type##_dma = (xqc_buf)->qcdma.dma + (off); \ + off += QMC_ALIGN(sizeof(struct qm_##type)); \ +} while (0) + + xqc_dma->size = QMC_ALIGN(sizeof(struct qm_eqc)) + + QMC_ALIGN(sizeof(struct qm_aeqc)) + + QMC_ALIGN(sizeof(struct qm_sqc)) + + QMC_ALIGN(sizeof(struct qm_cqc)); + + xqc_dma->va = dma_alloc_coherent(dev, xqc_dma->size, &xqc_dma->dma, + GFP_ATOMIC); + if (!xqc_dma->va) { + dev_err(dev, "Fail to alloc qcdma size=%zx\n", xqc_dma->size); + return -ENOMEM; + } + + QM_XQC_BUF_INIT(xqc_buf, eqc); + QM_XQC_BUF_INIT(xqc_buf, aeqc); + QM_XQC_BUF_INIT(xqc_buf, sqc); + QM_XQC_BUF_INIT(xqc_buf, cqc); + + return 0; +} + static int hisi_qm_memory_init(struct hisi_qm *qm) { struct device *dev = &qm->pdev->dev; - size_t qp_dma_size; + int ret = -ENOMEM; size_t off = 0; - int ret = 0; - int i; #define QM_INIT_BUF(qm, type, num) do { \ (qm)->type = ((qm)->qdma.va + (off)); \ @@ -2635,41 +2605,35 @@ static int hisi_qm_memory_init(struct hisi_qm *qm) QMC_ALIGN(sizeof(struct qm_aeqe) * QM_Q_DEPTH) + QMC_ALIGN(sizeof(struct qm_sqc) * qm->qp_num) + QMC_ALIGN(sizeof(struct qm_cqc) * qm->qp_num); - qm->qdma.va = dma_alloc_coherent(dev, qm->qdma.size, - &qm->qdma.dma, GFP_ATOMIC | __GFP_ZERO); - dev_dbg(dev, "allocate qm dma buf size=%zx)\n", qm->qdma.size); - if (!qm->qdma.va) - return -ENOMEM; + qm->qdma.va = dma_alloc_coherent(dev, qm->qdma.size, &qm->qdma.dma, + GFP_ATOMIC | __GFP_ZERO); + if (!qm->qdma.va) { + dev_err(dev, "Fail to alloc qdma size=%zx\n", qm->qdma.size); + goto err_destroy_idr; + } QM_INIT_BUF(qm, eqe, QM_EQ_DEPTH); QM_INIT_BUF(qm, aeqe, QM_Q_DEPTH); QM_INIT_BUF(qm, sqc, qm->qp_num); QM_INIT_BUF(qm, cqc, qm->qp_num); - qm->qp_array = kcalloc(qm->qp_num, sizeof(struct hisi_qp), GFP_KERNEL); - if (!qm->qp_array) { - ret = -ENOMEM; - goto err_alloc_qp_array; - } + ret = hisi_qm_alloc_rsv_buf(qm); + if (ret) + goto err_free_qdma; - /* one more page for device or qp statuses */ - qp_dma_size = qm->sqe_size * QM_Q_DEPTH + - sizeof(struct cqe) * QM_Q_DEPTH; - qp_dma_size = PAGE_ALIGN(qp_dma_size) + PAGE_SIZE; - for (i = 0; i < qm->qp_num; i++) { - ret = hisi_qp_memory_init(qm, qp_dma_size, i); - if (ret) - goto err_init_qp_mem; + ret = hisi_qp_alloc_memory(qm); + if (ret) + goto err_free_reserve_buf; - dev_dbg(dev, "allocate qp dma buf size=%zx)\n", qp_dma_size); - } + return 0; + +err_free_reserve_buf: + hisi_qm_free_rsv_buf(qm); +err_free_qdma: + dma_free_coherent(dev, qm->qdma.size, qm->qdma.va, qm->qdma.dma); +err_destroy_idr: + idr_destroy(&qm->qp_idr); - return ret; -err_init_qp_mem: - hisi_qp_memory_uninit(qm, i); -err_alloc_qp_array: - dma_free_coherent(dev, qm->qdma.size, - qm->qdma.va, qm->qdma.dma); return ret; } @@ -2861,8 +2825,7 @@ void hisi_qm_uninit(struct hisi_qm *qm) hisi_qp_memory_uninit(qm, qm->qp_num); idr_destroy(&qm->qp_idr); - - /* qm hardware buffer free on put_queue if no dma api */ + hisi_qm_free_rsv_buf(qm); if (qm->qdma.va) { hisi_qm_cache_wb(qm); dma_free_coherent(dev, qm->qdma.size, @@ -2962,59 +2925,26 @@ static void qm_init_eq_aeq_status(struct hisi_qm *qm) static int qm_eq_ctx_cfg(struct hisi_qm *qm) { - struct device *dev = &qm->pdev->dev; - struct qm_eqc *eqc; - dma_addr_t eqc_dma; - int ret; + struct qm_eqc eqc = {0}; - eqc = kzalloc(sizeof(struct qm_eqc), GFP_KERNEL); - if (!eqc) - return -ENOMEM; - eqc_dma = dma_map_single(dev, eqc, sizeof(struct qm_eqc), - DMA_TO_DEVICE); - if (dma_mapping_error(dev, eqc_dma)) { - kfree(eqc); - return -ENOMEM; - } - - eqc->base_l = cpu_to_le32(lower_32_bits(qm->eqe_dma)); - eqc->base_h = cpu_to_le32(upper_32_bits(qm->eqe_dma)); + eqc.base_l = cpu_to_le32(lower_32_bits(qm->eqe_dma)); + eqc.base_h = cpu_to_le32(upper_32_bits(qm->eqe_dma)); if (qm->ver == QM_HW_V1) - eqc->dw3 = cpu_to_le32(QM_EQE_AEQE_SIZE); - eqc->dw6 = cpu_to_le32((QM_EQ_DEPTH - 1) | (1 << QM_EQC_PHASE_SHIFT)); - ret = hisi_qm_mb_write(qm, QM_MB_CMD_EQC, eqc_dma, 0, 0); - dma_unmap_single(dev, eqc_dma, sizeof(struct qm_eqc), DMA_TO_DEVICE); - kfree(eqc); + eqc.dw3 = cpu_to_le32(QM_EQE_AEQE_SIZE); + eqc.dw6 = cpu_to_le32((QM_EQ_DEPTH - 1) | (1 << QM_EQC_PHASE_SHIFT)); - return ret; + return qm_set_and_get_xqc(qm, QM_MB_CMD_EQC, &eqc, 0, 0); } static int qm_aeq_ctx_cfg(struct hisi_qm *qm) { - struct device *dev = &qm->pdev->dev; - struct qm_aeqc *aeqc; - dma_addr_t aeqc_dma; - int ret; + struct qm_aeqc aeqc = {0}; - aeqc = kzalloc(sizeof(struct qm_aeqc), GFP_KERNEL); - if (!aeqc) - return -ENOMEM; - aeqc_dma = dma_map_single(dev, aeqc, sizeof(struct qm_aeqc), - DMA_TO_DEVICE); - if (dma_mapping_error(dev, aeqc_dma)) { - kfree(aeqc); - return -ENOMEM; - } + aeqc.base_l = cpu_to_le32(lower_32_bits(qm->aeqe_dma)); + aeqc.base_h = cpu_to_le32(upper_32_bits(qm->aeqe_dma)); + aeqc.dw6 = cpu_to_le32((QM_Q_DEPTH - 1) | (1 << QM_EQC_PHASE_SHIFT)); - aeqc->base_l = cpu_to_le32(lower_32_bits(qm->aeqe_dma)); - aeqc->base_h = cpu_to_le32(upper_32_bits(qm->aeqe_dma)); - aeqc->dw6 = cpu_to_le32((QM_Q_DEPTH - 1) | (1 << QM_EQC_PHASE_SHIFT)); - ret = hisi_qm_mb_write(qm, QM_MB_CMD_AEQC, aeqc_dma, 0, 0); - - dma_unmap_single(dev, aeqc_dma, sizeof(struct qm_aeqc), DMA_TO_DEVICE); - kfree(aeqc); - - return ret; + return qm_set_and_get_xqc(qm, QM_MB_CMD_AEQC, &aeqc, 0, 0); } static int qm_eq_aeq_ctx_cfg(struct hisi_qm *qm) diff --git a/drivers/crypto/hisilicon/qm.h b/drivers/crypto/hisilicon/qm.h index d7d23d1ec34c..9f5e440d7396 100644 --- a/drivers/crypto/hisilicon/qm.h +++ b/drivers/crypto/hisilicon/qm.h @@ -286,6 +286,18 @@ struct hisi_qm_list { bool (*check)(struct hisi_qm *qm); }; +struct qm_rsv_buf { + struct qm_sqc *sqc; + struct qm_cqc *cqc; + struct qm_eqc *eqc; + struct qm_aeqc *aeqc; + dma_addr_t sqc_dma; + dma_addr_t cqc_dma; + dma_addr_t eqc_dma; + dma_addr_t aeqc_dma; + struct qm_dma qcdma; +}; + struct hisi_qm { enum qm_hw_ver ver; enum qm_fun_type fun_type; @@ -309,6 +321,7 @@ struct hisi_qm { dma_addr_t cqc_dma; dma_addr_t eqe_dma; dma_addr_t aeqe_dma; + struct qm_rsv_buf xqc_buf; struct hisi_qm_status status; struct hisi_qm_err_ini err_ini; -- 2.30.0

2 1

[PATCH OLK-5.10 00/42] add error handling for add disk
by Li Nan 06 Nov '23

06 Nov '23

Christoph Hellwig (14): block: fold register_disk into device_add_disk block: call blk_integrity_add earlier in device_add_disk block: add the events* attributes to disk_attrs block: fix error unwinding in device_add_disk block: clear ->slave_dir when dropping the main slave_dir reference block: add blk_alloc_disk and blk_cleanup_disk APIs blk-mq: add the blk_mq_alloc_disk APIs loop: use blk_mq_alloc_disk and blk_cleanup_disk loop: fix order of cleaning up the queue and freeing the tagset nbd: use blk_mq_alloc_disk and blk_cleanup_disk block: add a flag to make put_disk on partially initalized disks safer nvme: use blk_mq_alloc_disk brd: convert to blk_alloc_disk/blk_cleanup_disk ubi: use blk_mq_alloc_disk and blk_cleanup_disk Dan Carpenter (1): blk-mq: fix an IS_ERR() vs NULL bug Li Lingfeng (1): nbd: fix uaf in nbd_open Luis Chamberlain (10): block: return errors from blk_integrity_add block: return errors from disk_alloc_events block: add error handling for device_add_disk / add_disk block: fix device_add_disk() kobject_create_and_add() error handling loop: add error handling support for add_disk() nbd: add error handling support for add_disk() nvme: add error handling support for add_disk() block/brd: add error handling support for add_disk() scsi: sr: Add error handling support for add_disk() mtd/ubi/block: add error handling support for add_disk() Tetsuo Handa (1): block: check minor range in device_add_disk() Wang Qing (1): nbd: fix order of cleaning up the queue and freeing the tagset Wen Yang (1): Revert "Revert "block: nbd: add sanity check for first_minor"" Yu Kuai (3): nbd: fix max value for 'first_minor' nbd: fix possible overflow for 'first_minor' in nbd_dev_add() block: fix memory leak for elevator on add_disk failure Zhang Wensheng (1): nbd: fix possible overflow on 'first_minor' in nbd_dev_add() Zhong Jinghua (9): nbd: Reorganize the messy commit log about the first_minor check block: return errors from blk_register_region block: Fix the kabi change in device_add_disk block: Fix the kabi change on blk_register_region block: call blk_get_queue earlier in __device_add_disk block: Fix minor range check in device_add_disk() block: Set memalloc_noio to false in the error path mtd/ubi/block: Fix null pointer dereference issue in error path mtd/ubi/block: Fix uaf problem in ubiblock_cleanup block/blk.h | 5 +- include/linux/blk-mq.h | 12 ++ include/linux/genhd.h | 36 +++++ block/blk-integrity.c | 12 +- block/blk-mq.c | 19 +++ block/genhd.c | 286 ++++++++++++++++++++++++++------------- drivers/block/brd.c | 103 ++++++-------- drivers/block/loop.c | 26 ++-- drivers/block/nbd.c | 75 ++++------ drivers/mtd/ubi/block.c | 76 +++++------ drivers/nvme/host/core.c | 41 +++--- drivers/scsi/sr.c | 7 +- 12 files changed, 413 insertions(+), 285 deletions(-) -- 2.39.2

1 42

[PATCH openEuler-1.0-LTS] crypto: hisilicon/qm - alloc reserve buffer to set and get xqc
by wangyuan 06 Nov '23

06 Nov '23

From: Yu'an Wang <wangyuan46(a)huawei.com> driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8E6ED CVE: NA -------------------------------- If the temporarily applied memory is used to set or get the xqc information, the driver will release the memory after the mailbox timeout. However, the hardware still performs the operation. Therefore, the released memory may be written by hardware. So when load driver, reserve memory is applied for xqc configuration. Signed-off-by: Yu'an Wang <wangyuan46(a)huawei.com> --- drivers/crypto/hisilicon/qm.c | 554 +++++++++++++++------------------- drivers/crypto/hisilicon/qm.h | 13 + 2 files changed, 255 insertions(+), 312 deletions(-) diff --git a/drivers/crypto/hisilicon/qm.c b/drivers/crypto/hisilicon/qm.c index 64e8ef1bf8e7..b4fd97e1e03e 100644 --- a/drivers/crypto/hisilicon/qm.c +++ b/drivers/crypto/hisilicon/qm.c @@ -50,7 +50,7 @@ #define QM_SQ_TYPE_MASK 0xf -#define QM_SQ_TAIL_IDX(sqc) ((le16_to_cpu((sqc)->w11) >> 6) & 0x1) +#define QM_SQ_TAIL_IDX(sqc) ((le16_to_cpu((sqc).w11) >> 6) & 0x1) /* cqc shift */ #define QM_CQ_HOP_NUM_SHIFT 0 @@ -64,7 +64,7 @@ #define QM_QC_CQE_SIZE 4 -#define QM_CQ_TAIL_IDX(cqc) ((le16_to_cpu((cqc)->w11) >> 6) & 0x1) +#define QM_CQ_TAIL_IDX(cqc) ((le16_to_cpu((cqc).w11) >> 6) & 0x1) /* eqc shift */ #define QM_EQE_AEQE_SIZE (2UL << 12) @@ -217,19 +217,6 @@ #define QM_MK_SQC_DW3_V2(sqe_sz) \ ((QM_Q_DEPTH - 1) | ((u32)ilog2(sqe_sz) << QM_SQ_SQE_SIZE_SHIFT)) -#define INIT_QC_COMMON(qc, base, pasid) do { \ - (qc)->head = 0; \ - (qc)->tail = 0; \ - (qc)->base_l = cpu_to_le32(lower_32_bits(base)); \ - (qc)->base_h = cpu_to_le32(upper_32_bits(base)); \ - (qc)->dw3 = 0; \ - (qc)->w8 = 0; \ - (qc)->rsvd0 = 0; \ - (qc)->pasid = cpu_to_le16(pasid); \ - (qc)->w11 = 0; \ - (qc)->rsvd1 = 0; \ -} while (0) - #define QMC_ALIGN(sz) ALIGN(sz, 32) static int __hisi_qm_start(struct hisi_qm *qm); @@ -537,6 +524,61 @@ static int hisi_qm_mb_read(struct hisi_qm *qm, u64 *base, u8 cmd, u16 queue) return 0; } +/* op 0: set xqc info to hardware, 1: get xqc info from hardware. */ +static int qm_set_and_get_xqc(struct hisi_qm *qm, u8 cmd, void *xqc, u32 qp_id, + bool op) +{ + struct hisi_qm *pf_qm = pci_get_drvdata(pci_physfn(qm->pdev)); + struct qm_mailbox mailbox; + dma_addr_t xqc_dma; + void *tmp_xqc; + size_t size; + int ret; + + switch (cmd) { + case QM_MB_CMD_SQC: + size = sizeof(struct qm_sqc); + tmp_xqc = qm->xqc_buf.sqc; + xqc_dma = qm->xqc_buf.sqc_dma; + break; + case QM_MB_CMD_CQC: + size = sizeof(struct qm_cqc); + tmp_xqc = qm->xqc_buf.cqc; + xqc_dma = qm->xqc_buf.cqc_dma; + break; + case QM_MB_CMD_EQC: + size = sizeof(struct qm_eqc); + tmp_xqc = qm->xqc_buf.eqc; + xqc_dma = qm->xqc_buf.eqc_dma; + break; + case QM_MB_CMD_AEQC: + size = sizeof(struct qm_aeqc); + tmp_xqc = qm->xqc_buf.aeqc; + xqc_dma = qm->xqc_buf.aeqc_dma; + break; + } + + /* No need to judge if master OOO is blocked. */ + if (qm_check_dev_error(pf_qm)) { + dev_err(&qm->pdev->dev, + "QM mailbox operation failed since qm is stop!\n"); + return -EIO; + } + + mutex_lock(&qm->mailbox_lock); + if (!op) + memcpy(tmp_xqc, xqc, size); + + qm_mb_pre_init(&mailbox, cmd, xqc_dma, qp_id, op); + ret = qm_mb_nolock(qm, &mailbox); + if (!ret && op) + memcpy(xqc, tmp_xqc, size); + + mutex_unlock(&qm->mailbox_lock); + + return ret; +} + static void qm_db_v1(struct hisi_qm *qm, u16 qn, u8 cmd, u16 index, u8 priority) { u64 doorbell; @@ -1166,35 +1208,6 @@ static ssize_t qm_cmd_read(struct file *filp, char __user *buffer, return (*pos = len); } -static void *qm_ctx_alloc(struct hisi_qm *qm, size_t ctx_size, - dma_addr_t *dma_addr) -{ - struct device *dev = &qm->pdev->dev; - void *ctx_addr; - - ctx_addr = kzalloc(ctx_size, GFP_KERNEL); - if (!ctx_addr) - return ERR_PTR(-ENOMEM); - - *dma_addr = dma_map_single(dev, ctx_addr, ctx_size, DMA_FROM_DEVICE); - if (dma_mapping_error(dev, *dma_addr)) { - dev_err(dev, "DMA mapping error!\n"); - kfree(ctx_addr); - return ERR_PTR(-ENOMEM); - } - - return ctx_addr; -} - -static void qm_ctx_free(struct hisi_qm *qm, size_t ctx_size, - const void *ctx_addr, dma_addr_t *dma_addr) -{ - struct device *dev = &qm->pdev->dev; - - dma_unmap_single(dev, *dma_addr, ctx_size, DMA_FROM_DEVICE); - kfree(ctx_addr); -} - static int dump_show(struct hisi_qm *qm, void *info, unsigned int info_size, char *info_name) { @@ -1205,8 +1218,10 @@ static int dump_show(struct hisi_qm *qm, void *info, #define BYTE_PER_DW 4 info_buf = kzalloc(info_size, GFP_KERNEL); - if (!info_buf) + if (!info_buf) { + dev_err(dev, "Fail to alloc dump info buf\n"); return -ENOMEM; + } for (i = 0; i < info_size; i++, info_curr++) { if (i % BYTE_PER_DW == 0) @@ -1230,21 +1245,11 @@ static int dump_show(struct hisi_qm *qm, void *info, return 0; } -static int qm_dump_sqc_raw(struct hisi_qm *qm, dma_addr_t dma_addr, u16 qp_id) -{ - return hisi_qm_mb_write(qm, QM_MB_CMD_SQC, dma_addr, qp_id, 1); -} - -static int qm_dump_cqc_raw(struct hisi_qm *qm, dma_addr_t dma_addr, u16 qp_id) -{ - return hisi_qm_mb_write(qm, QM_MB_CMD_CQC, dma_addr, qp_id, 1); -} - static int qm_sqc_dump(struct hisi_qm *qm, const char *s) { struct device *dev = &qm->pdev->dev; - struct qm_sqc *sqc, *sqc_curr; - dma_addr_t sqc_dma; + struct qm_sqc *sqc_curr; + struct qm_sqc sqc; u32 qp_id; int ret; @@ -1257,40 +1262,28 @@ static int qm_sqc_dump(struct hisi_qm *qm, const char *s) return -EINVAL; } - sqc = qm_ctx_alloc(qm, sizeof(struct qm_sqc), &sqc_dma); - if (IS_ERR(sqc)) - return PTR_ERR(sqc); - - ret = qm_dump_sqc_raw(qm, sqc_dma, qp_id); - if (ret) { - down_read(&qm->qps_lock); - if (qm->sqc) { - sqc_curr = qm->sqc + qp_id; - - ret = dump_show(qm, sqc_curr, sizeof(struct qm_sqc), - "SOFT SQC"); - if (ret) - dev_info(dev, "Show soft sqc failed!\n"); - } - up_read(&qm->qps_lock); + ret = qm_set_and_get_xqc(qm, QM_MB_CMD_SQC, &sqc, qp_id, 1); + if (!ret) + return dump_show(qm, &sqc, sizeof(struct qm_sqc), "SQC"); - goto mailbox_fail; + down_read(&qm->qps_lock); + if (qm->sqc) { + sqc_curr = qm->sqc + qp_id; + ret = dump_show(qm, sqc_curr, sizeof(struct qm_sqc), + "SOFT SQC"); + if (ret) + dev_info(dev, "Show soft sqc failed!\n"); } + up_read(&qm->qps_lock); - ret = dump_show(qm, sqc, sizeof(struct qm_sqc), "SQC"); - if (ret) - dev_info(dev, "Show hw sqc failed!\n"); - -mailbox_fail: - qm_ctx_free(qm, sizeof(struct qm_sqc), sqc, &sqc_dma); return ret; } static int qm_cqc_dump(struct hisi_qm *qm, const char *s) { struct device *dev = &qm->pdev->dev; - struct qm_cqc *cqc, *cqc_curr; - dma_addr_t cqc_dma; + struct qm_cqc *cqc_curr; + struct qm_cqc cqc; u32 qp_id; int ret; @@ -1303,40 +1296,28 @@ static int qm_cqc_dump(struct hisi_qm *qm, const char *s) return -EINVAL; } - cqc = qm_ctx_alloc(qm, sizeof(struct qm_cqc), &cqc_dma); - if (IS_ERR(cqc)) - return PTR_ERR(cqc); - - ret = qm_dump_cqc_raw(qm, cqc_dma, qp_id); - if (ret) { - down_read(&qm->qps_lock); - if (qm->cqc) { - cqc_curr = qm->cqc + qp_id; + ret = qm_set_and_get_xqc(qm, QM_MB_CMD_CQC, &cqc, qp_id, 1); + if (!ret) + return dump_show(qm, &cqc, sizeof(struct qm_cqc), "CQC"); - ret = dump_show(qm, cqc_curr, sizeof(struct qm_cqc), - "SOFT CQC"); - if (ret) - dev_info(dev, "Show soft cqc failed!\n"); - } - up_read(&qm->qps_lock); + down_read(&qm->qps_lock); + if (qm->cqc) { + cqc_curr = qm->cqc + qp_id; - goto mailbox_fail; + ret = dump_show(qm, cqc_curr, sizeof(struct qm_cqc), + "SOFT CQC"); + if (ret) + dev_info(dev, "Show soft cqc failed!\n"); } + up_read(&qm->qps_lock); - ret = dump_show(qm, cqc, sizeof(struct qm_cqc), "CQC"); - if (ret) - dev_info(dev, "Show hw cqc failed!\n"); - -mailbox_fail: - qm_ctx_free(qm, sizeof(struct qm_cqc), cqc, &cqc_dma); return ret; } static int qm_eqc_dump(struct hisi_qm *qm, char *s) { struct device *dev = &qm->pdev->dev; - struct qm_eqc *eqc; - dma_addr_t eqc_dma; + struct qm_eqc eqc; int ret; if (strsep(&s, " ")) { @@ -1344,28 +1325,17 @@ static int qm_eqc_dump(struct hisi_qm *qm, char *s) return -EINVAL; } - eqc = qm_ctx_alloc(qm, sizeof(struct qm_eqc), &eqc_dma); - if (IS_ERR(eqc)) - return PTR_ERR(eqc); - - ret = hisi_qm_mb_write(qm, QM_MB_CMD_EQC, eqc_dma, 0, 1); + ret = qm_set_and_get_xqc(qm, QM_MB_CMD_EQC, &eqc, 0, 1); if (ret) - goto mailbox_fail; - - ret = dump_show(qm, eqc, sizeof(struct qm_eqc), "EQC"); - if (ret) - dev_info(dev, "Show eqc failed!\n"); + return ret; -mailbox_fail: - qm_ctx_free(qm, sizeof(struct qm_eqc), eqc, &eqc_dma); - return ret; + return dump_show(qm, &eqc, sizeof(struct qm_eqc), "EQC"); } static int qm_aeqc_dump(struct hisi_qm *qm, char *s) { struct device *dev = &qm->pdev->dev; - struct qm_aeqc *aeqc; - dma_addr_t aeqc_dma; + struct qm_aeqc aeqc; int ret; if (strsep(&s, " ")) { @@ -1373,21 +1343,11 @@ static int qm_aeqc_dump(struct hisi_qm *qm, char *s) return -EINVAL; } - aeqc = qm_ctx_alloc(qm, sizeof(struct qm_aeqc), &aeqc_dma); - if (IS_ERR(aeqc)) - return PTR_ERR(aeqc); - - ret = hisi_qm_mb_write(qm, QM_MB_CMD_AEQC, aeqc_dma, 0, 1); - if (ret) - goto mailbox_fail; - - ret = dump_show(qm, aeqc, sizeof(struct qm_aeqc), "AEQC"); + ret = qm_set_and_get_xqc(qm, QM_MB_CMD_AEQC, &aeqc, 0, 1); if (ret) - dev_info(dev, "Show hw aeqc failed!\n"); + return ret; -mailbox_fail: - qm_ctx_free(qm, sizeof(struct qm_aeqc), aeqc, &aeqc_dma); - return ret; + return dump_show(qm, &aeqc, sizeof(struct qm_aeqc), "AEQC"); } static int q_dump_param_parse(struct hisi_qm *qm, char *s, @@ -1432,7 +1392,6 @@ static int q_dump_param_parse(struct hisi_qm *qm, char *s, static int qm_sq_dump(struct hisi_qm *qm, char *s) { - struct device *dev = &qm->pdev->dev; struct hisi_qp *qp; u32 qp_id, sqe_id; void *sqe_curr; @@ -1447,16 +1406,11 @@ static int qm_sq_dump(struct hisi_qm *qm, char *s) memset(sqe_curr + qm->debug.sqe_mask_offset, SQE_ADDR_MASK, qm->debug.sqe_mask_len); - ret = dump_show(qm, sqe_curr, qm->sqe_size, "SQE"); - if (ret) - dev_info(dev, "Show sqe failed!\n"); - - return ret; + return dump_show(qm, sqe_curr, qm->sqe_size, "SQE"); } static int qm_cq_dump(struct hisi_qm *qm, char *s) { - struct device *dev = &qm->pdev->dev; struct qm_cqe *cqe_curr; struct hisi_qp *qp; u32 qp_id, cqe_id; @@ -1468,11 +1422,8 @@ static int qm_cq_dump(struct hisi_qm *qm, char *s) qp = &qm->qp_array[qp_id]; cqe_curr = qp->cqe + cqe_id; - ret = dump_show(qm, cqe_curr, sizeof(struct qm_cqe), "CQE"); - if (ret) - dev_info(dev, "Show cqe failed!\n"); - return ret; + return dump_show(qm, cqe_curr, sizeof(struct qm_cqe), "CQE"); } static int qm_eq_dump(struct hisi_qm *qm, const char *s) @@ -1895,79 +1846,46 @@ EXPORT_SYMBOL_GPL(hisi_qm_release_qp); static int qm_sq_ctx_cfg(struct hisi_qp *qp, int qp_id, int pasid) { struct hisi_qm *qm = qp->qm; - struct device *dev = &qm->pdev->dev; enum qm_hw_ver ver = qm->ver; - struct qm_sqc *sqc; - dma_addr_t sqc_dma; - int ret; + struct qm_sqc sqc = {0}; - sqc = kzalloc(sizeof(struct qm_sqc), GFP_KERNEL); - if (!sqc) - return -ENOMEM; - sqc_dma = dma_map_single(dev, sqc, sizeof(struct qm_sqc), - DMA_TO_DEVICE); - if (dma_mapping_error(dev, sqc_dma)) { - kfree(sqc); - return -ENOMEM; - } - - INIT_QC_COMMON(sqc, qp->sqe_dma, pasid); if (ver == QM_HW_V1) { - sqc->dw3 = cpu_to_le32(QM_MK_SQC_DW3_V1(0, 0, 0, qm->sqe_size)); - sqc->w8 = cpu_to_le16(QM_Q_DEPTH - 1); + sqc.dw3 = cpu_to_le32(QM_MK_SQC_DW3_V1(0, 0, 0, qm->sqe_size)); + sqc.w8 = cpu_to_le16(QM_Q_DEPTH - 1); } else if (ver == QM_HW_V2) { - sqc->dw3 = cpu_to_le32(QM_MK_SQC_DW3_V2(qm->sqe_size)); - sqc->w8 = 0; /* rand_qc */ + sqc.dw3 = cpu_to_le32(QM_MK_SQC_DW3_V2(qm->sqe_size)); + sqc.w8 = 0; /* rand_qc */ } - sqc->cq_num = cpu_to_le16(qp_id); - sqc->w13 = cpu_to_le16(QM_MK_SQC_W13(0, 1, qp->alg_type)); - - ret = hisi_qm_mb_write(qm, QM_MB_CMD_SQC, sqc_dma, qp_id, 0); + sqc.cq_num = cpu_to_le16(qp_id); + sqc.w13 = cpu_to_le16(QM_MK_SQC_W13(0, 1, qp->alg_type)); + sqc.base_l = cpu_to_le32(lower_32_bits(qp->sqe_dma)); + sqc.base_h = cpu_to_le32(upper_32_bits(qp->sqe_dma)); + sqc.pasid = cpu_to_le16(pasid); - dma_unmap_single(dev, sqc_dma, sizeof(struct qm_sqc), DMA_TO_DEVICE); - kfree(sqc); - - return ret; + return qm_set_and_get_xqc(qm, QM_MB_CMD_SQC, &sqc, qp_id, 0); } static int qm_cq_ctx_cfg(struct hisi_qp *qp, int qp_id, int pasid) { struct hisi_qm *qm = qp->qm; - struct device *dev = &qm->pdev->dev; enum qm_hw_ver ver = qm->ver; - struct qm_cqc *cqc; - dma_addr_t cqc_dma; - int ret; - - cqc = kzalloc(sizeof(struct qm_cqc), GFP_KERNEL); - if (!cqc) - return -ENOMEM; - - cqc_dma = dma_map_single(dev, cqc, sizeof(struct qm_cqc), - DMA_TO_DEVICE); - if (dma_mapping_error(dev, cqc_dma)) { - kfree(cqc); - return -ENOMEM; - } + struct qm_cqc cqc = {0}; - INIT_QC_COMMON(cqc, qp->cqe_dma, pasid); if (ver == QM_HW_V1) { - cqc->dw3 = cpu_to_le32(QM_MK_CQC_DW3_V1(0, 0, 0, + cqc.dw3 = cpu_to_le32(QM_MK_CQC_DW3_V1(0, 0, 0, QM_QC_CQE_SIZE)); - cqc->w8 = cpu_to_le16(QM_Q_DEPTH - 1); + cqc.w8 = cpu_to_le16(QM_Q_DEPTH - 1); } else if (ver == QM_HW_V2) { - cqc->dw3 = cpu_to_le32(QM_MK_CQC_DW3_V2(QM_QC_CQE_SIZE)); - cqc->w8 = 0; /* rand_qc */ + cqc.dw3 = cpu_to_le32(QM_MK_CQC_DW3_V2(QM_QC_CQE_SIZE)); + cqc.w8 = 0; /* rand_qc */ } - cqc->dw6 = cpu_to_le32(1 << QM_CQ_PHASE_SHIFT | - qp->c_flag << QM_CQ_FLAG_SHIFT); + cqc.dw6 = cpu_to_le32(1 << QM_CQ_PHASE_SHIFT | + qp->c_flag << QM_CQ_FLAG_SHIFT); + cqc.base_l = cpu_to_le32(lower_32_bits(qp->cqe_dma)); + cqc.base_h = cpu_to_le32(upper_32_bits(qp->cqe_dma)); + cqc.pasid = cpu_to_le16(pasid); - ret = hisi_qm_mb_write(qm, QM_MB_CMD_CQC, cqc_dma, qp_id, 0); - - dma_unmap_single(dev, cqc_dma, sizeof(struct qm_cqc), DMA_TO_DEVICE); - kfree(cqc); - - return ret; + return qm_set_and_get_xqc(qm, QM_MB_CMD_CQC, &cqc, qp_id, 0); } static int qm_qp_ctx_cfg(struct hisi_qp *qp, int qp_id, int pasid) @@ -2043,54 +1961,40 @@ static void qp_stop_fail_cb(struct hisi_qp *qp) static void qm_qp_has_no_task(struct hisi_qp *qp) { - size_t size = sizeof(struct qm_sqc) + sizeof(struct qm_cqc); - struct device *dev = &qp->qm->pdev->dev; - struct qm_sqc *sqc; - struct qm_cqc *cqc; - dma_addr_t dma_addr; - void *addr; - int i = 0; - int ret; + struct hisi_qm *qm = qp->qm; + struct device *dev = &qm->pdev->dev; + struct qm_sqc sqc; + struct qm_cqc cqc; + int ret, i = 0; if (qp->qm->err_ini.err_info.is_qm_ecc_mbit || qp->qm->err_ini.err_info.is_dev_ecc_mbit) return; - addr = qm_ctx_alloc(qp->qm, size, &dma_addr); - if (IS_ERR(addr)) { - dev_err(dev, "alloc ctx for sqc and cqc failed!\n"); - return; - } - while (++i) { - ret = qm_dump_sqc_raw(qp->qm, dma_addr, qp->qp_id); + ret = qm_set_and_get_xqc(qm, QM_MB_CMD_SQC, &sqc, qp->qp_id, 1); if (ret) { - dev_err(dev, "Failed to dump sqc!\n"); - break; + dev_err_ratelimited(dev, "Fail to dump sqc!\n"); + return; } - sqc = addr; - ret = qm_dump_cqc_raw(qp->qm, - (dma_addr + sizeof(struct qm_sqc)), qp->qp_id); + ret = qm_set_and_get_xqc(qm, QM_MB_CMD_CQC, &cqc, qp->qp_id, 1); if (ret) { - dev_err(dev, "Failed to dump cqc!\n"); - break; + dev_err_ratelimited(dev, "Fail to dump cqc!\n"); + return; } - cqc = addr + sizeof(struct qm_sqc); - if ((sqc->tail == cqc->tail) && - (QM_SQ_TAIL_IDX(sqc) == QM_CQ_TAIL_IDX(cqc))) - break; + if ((QM_SQ_TAIL_IDX(sqc) == QM_CQ_TAIL_IDX(cqc)) && + (sqc.tail == cqc.tail)) + return; if (i == MAX_WAIT_COUNTS) { - dev_err(dev, "Fail to wait for device stop!\n"); - break; + dev_err(dev, "Fail to empty queue %u!\n", qp->qp_id); + return; } usleep_range(WAIT_PERIOD_US_MIN, WAIT_PERIOD_US_MAX); } - - qm_ctx_free(qp->qm, size, addr, &dma_addr); } static int hisi_qm_stop_qp_nolock(struct hisi_qp *qp) @@ -2597,10 +2501,12 @@ static int hisi_qp_memory_init(struct hisi_qm *qm, size_t dma_size, int id) struct hisi_qp *qp; qp = &qm->qp_array[id]; - qp->qdma.va = dma_alloc_coherent(dev, dma_size, - &qp->qdma.dma, GFP_KERNEL); - if (!qp->qdma.va) + qp->qdma.va = dma_alloc_coherent(dev, dma_size, &qp->qdma.dma, + GFP_KERNEL); + if (!qp->qdma.va) { + dev_err(dev, "Fail to alloc qp dma buf size=%zx\n", dma_size); return -ENOMEM; + } qp->sqe = qp->qdma.va; qp->sqe_dma = qp->qdma.dma; @@ -2613,13 +2519,79 @@ static int hisi_qp_memory_init(struct hisi_qm *qm, size_t dma_size, int id) return 0; } +static int hisi_qp_alloc_memory(struct hisi_qm *qm) +{ + size_t qp_dma_size; + int i, ret; + + qm->qp_array = kcalloc(qm->qp_num, sizeof(struct hisi_qp), GFP_KERNEL); + if (!qm->qp_array) + return -ENOMEM; + + /* one more page for device or qp statuses */ + qp_dma_size = qm->sqe_size * QM_Q_DEPTH + + sizeof(struct cqe) * QM_Q_DEPTH; + qp_dma_size = PAGE_ALIGN(qp_dma_size) + PAGE_SIZE; + for (i = 0; i < qm->qp_num; i++) { + ret = hisi_qp_memory_init(qm, qp_dma_size, i); + if (ret) + goto err_init_qp_mem; + } + + return 0; + +err_init_qp_mem: + hisi_qp_memory_uninit(qm, i); + + return ret; +} + +static void hisi_qm_free_rsv_buf(struct hisi_qm *qm) +{ + struct qm_dma *xqc_dma = &qm->xqc_buf.qcdma; + struct device *dev = &qm->pdev->dev; + + dma_free_coherent(dev, xqc_dma->size, xqc_dma->va, xqc_dma->dma); +} + +static int hisi_qm_alloc_rsv_buf(struct hisi_qm *qm) +{ + struct qm_rsv_buf *xqc_buf = &qm->xqc_buf; + struct qm_dma *xqc_dma = &xqc_buf->qcdma; + struct device *dev = &qm->pdev->dev; + size_t off = 0; + +#define QM_XQC_BUF_INIT(xqc_buf, type) do { \ + (xqc_buf)->type = ((xqc_buf)->qcdma.va + (off)); \ + (xqc_buf)->type##_dma = (xqc_buf)->qcdma.dma + (off); \ + off += QMC_ALIGN(sizeof(struct qm_##type)); \ +} while (0) + + xqc_dma->size = QMC_ALIGN(sizeof(struct qm_eqc)) + + QMC_ALIGN(sizeof(struct qm_aeqc)) + + QMC_ALIGN(sizeof(struct qm_sqc)) + + QMC_ALIGN(sizeof(struct qm_cqc)); + + xqc_dma->va = dma_alloc_coherent(dev, xqc_dma->size, &xqc_dma->dma, + GFP_ATOMIC); + if (!xqc_dma->va) { + dev_err(dev, "Fail to alloc qcdma size=%zx\n", xqc_dma->size); + return -ENOMEM; + } + + QM_XQC_BUF_INIT(xqc_buf, eqc); + QM_XQC_BUF_INIT(xqc_buf, aeqc); + QM_XQC_BUF_INIT(xqc_buf, sqc); + QM_XQC_BUF_INIT(xqc_buf, cqc); + + return 0; +} + static int hisi_qm_memory_init(struct hisi_qm *qm) { struct device *dev = &qm->pdev->dev; - size_t qp_dma_size; + int ret = -ENOMEM; size_t off = 0; - int ret = 0; - int i; #define QM_INIT_BUF(qm, type, num) do { \ (qm)->type = ((qm)->qdma.va + (off)); \ @@ -2635,41 +2607,35 @@ static int hisi_qm_memory_init(struct hisi_qm *qm) QMC_ALIGN(sizeof(struct qm_aeqe) * QM_Q_DEPTH) + QMC_ALIGN(sizeof(struct qm_sqc) * qm->qp_num) + QMC_ALIGN(sizeof(struct qm_cqc) * qm->qp_num); - qm->qdma.va = dma_alloc_coherent(dev, qm->qdma.size, - &qm->qdma.dma, GFP_ATOMIC | __GFP_ZERO); - dev_dbg(dev, "allocate qm dma buf size=%zx)\n", qm->qdma.size); - if (!qm->qdma.va) - return -ENOMEM; + qm->qdma.va = dma_alloc_coherent(dev, qm->qdma.size, &qm->qdma.dma, + GFP_ATOMIC | __GFP_ZERO); + if (!qm->qdma.va) { + dev_err(dev, "Fail to alloc qdma size=%zx\n", qm->qdma.size); + goto err_destroy_idr; + } QM_INIT_BUF(qm, eqe, QM_EQ_DEPTH); QM_INIT_BUF(qm, aeqe, QM_Q_DEPTH); QM_INIT_BUF(qm, sqc, qm->qp_num); QM_INIT_BUF(qm, cqc, qm->qp_num); - qm->qp_array = kcalloc(qm->qp_num, sizeof(struct hisi_qp), GFP_KERNEL); - if (!qm->qp_array) { - ret = -ENOMEM; - goto err_alloc_qp_array; - } + ret = hisi_qm_alloc_rsv_buf(qm); + if (ret) + goto err_free_qdma; - /* one more page for device or qp statuses */ - qp_dma_size = qm->sqe_size * QM_Q_DEPTH + - sizeof(struct cqe) * QM_Q_DEPTH; - qp_dma_size = PAGE_ALIGN(qp_dma_size) + PAGE_SIZE; - for (i = 0; i < qm->qp_num; i++) { - ret = hisi_qp_memory_init(qm, qp_dma_size, i); - if (ret) - goto err_init_qp_mem; + ret = hisi_qp_alloc_memory(qm); + if (ret) + goto err_free_reserve_buf; - dev_dbg(dev, "allocate qp dma buf size=%zx)\n", qp_dma_size); - } + return 0; + +err_free_reserve_buf: + hisi_qm_free_rsv_buf(qm); +err_free_qdma: + dma_free_coherent(dev, qm->qdma.size, qm->qdma.va, qm->qdma.dma); +err_destroy_idr: + idr_destroy(&qm->qp_idr); - return ret; -err_init_qp_mem: - hisi_qp_memory_uninit(qm, i); -err_alloc_qp_array: - dma_free_coherent(dev, qm->qdma.size, - qm->qdma.va, qm->qdma.dma); return ret; } @@ -2861,8 +2827,7 @@ void hisi_qm_uninit(struct hisi_qm *qm) hisi_qp_memory_uninit(qm, qm->qp_num); idr_destroy(&qm->qp_idr); - - /* qm hardware buffer free on put_queue if no dma api */ + hisi_qm_free_rsv_buf(qm); if (qm->qdma.va) { hisi_qm_cache_wb(qm); dma_free_coherent(dev, qm->qdma.size, @@ -2962,59 +2927,26 @@ static void qm_init_eq_aeq_status(struct hisi_qm *qm) static int qm_eq_ctx_cfg(struct hisi_qm *qm) { - struct device *dev = &qm->pdev->dev; - struct qm_eqc *eqc; - dma_addr_t eqc_dma; - int ret; + struct qm_eqc eqc = {0}; - eqc = kzalloc(sizeof(struct qm_eqc), GFP_KERNEL); - if (!eqc) - return -ENOMEM; - eqc_dma = dma_map_single(dev, eqc, sizeof(struct qm_eqc), - DMA_TO_DEVICE); - if (dma_mapping_error(dev, eqc_dma)) { - kfree(eqc); - return -ENOMEM; - } - - eqc->base_l = cpu_to_le32(lower_32_bits(qm->eqe_dma)); - eqc->base_h = cpu_to_le32(upper_32_bits(qm->eqe_dma)); + eqc.base_l = cpu_to_le32(lower_32_bits(qm->eqe_dma)); + eqc.base_h = cpu_to_le32(upper_32_bits(qm->eqe_dma)); if (qm->ver == QM_HW_V1) - eqc->dw3 = cpu_to_le32(QM_EQE_AEQE_SIZE); - eqc->dw6 = cpu_to_le32((QM_EQ_DEPTH - 1) | (1 << QM_EQC_PHASE_SHIFT)); - ret = hisi_qm_mb_write(qm, QM_MB_CMD_EQC, eqc_dma, 0, 0); - dma_unmap_single(dev, eqc_dma, sizeof(struct qm_eqc), DMA_TO_DEVICE); - kfree(eqc); + eqc.dw3 = cpu_to_le32(QM_EQE_AEQE_SIZE); + eqc.dw6 = cpu_to_le32((QM_EQ_DEPTH - 1) | (1 << QM_EQC_PHASE_SHIFT)); - return ret; + return qm_set_and_get_xqc(qm, QM_MB_CMD_EQC, &eqc, 0, 0); } static int qm_aeq_ctx_cfg(struct hisi_qm *qm) { - struct device *dev = &qm->pdev->dev; - struct qm_aeqc *aeqc; - dma_addr_t aeqc_dma; - int ret; + struct qm_aeqc aeqc = {0}; - aeqc = kzalloc(sizeof(struct qm_aeqc), GFP_KERNEL); - if (!aeqc) - return -ENOMEM; - aeqc_dma = dma_map_single(dev, aeqc, sizeof(struct qm_aeqc), - DMA_TO_DEVICE); - if (dma_mapping_error(dev, aeqc_dma)) { - kfree(aeqc); - return -ENOMEM; - } + aeqc.base_l = cpu_to_le32(lower_32_bits(qm->aeqe_dma)); + aeqc.base_h = cpu_to_le32(upper_32_bits(qm->aeqe_dma)); + aeqc.dw6 = cpu_to_le32((QM_Q_DEPTH - 1) | (1 << QM_EQC_PHASE_SHIFT)); - aeqc->base_l = cpu_to_le32(lower_32_bits(qm->aeqe_dma)); - aeqc->base_h = cpu_to_le32(upper_32_bits(qm->aeqe_dma)); - aeqc->dw6 = cpu_to_le32((QM_Q_DEPTH - 1) | (1 << QM_EQC_PHASE_SHIFT)); - ret = hisi_qm_mb_write(qm, QM_MB_CMD_AEQC, aeqc_dma, 0, 0); - - dma_unmap_single(dev, aeqc_dma, sizeof(struct qm_aeqc), DMA_TO_DEVICE); - kfree(aeqc); - - return ret; + return qm_set_and_get_xqc(qm, QM_MB_CMD_AEQC, &aeqc, 0, 0); } static int qm_eq_aeq_ctx_cfg(struct hisi_qm *qm) @@ -3841,10 +3773,8 @@ static int qm_vf_reset_prepare(struct pci_dev *pdev, ret = hisi_qm_stop(qm, stop_reason); if (ret) { - hisi_qm_set_hw_reset(qm, - QM_RESET_STOP_TX_OFFSET); - hisi_qm_set_hw_reset(qm, - QM_RESET_STOP_RX_OFFSET); + hisi_qm_set_hw_reset(qm, QM_RESET_STOP_TX_OFFSET); + hisi_qm_set_hw_reset(qm, QM_RESET_STOP_RX_OFFSET); atomic_set(&qm->status.flags, QM_STOP); } } diff --git a/drivers/crypto/hisilicon/qm.h b/drivers/crypto/hisilicon/qm.h index d7d23d1ec34c..9f5e440d7396 100644 --- a/drivers/crypto/hisilicon/qm.h +++ b/drivers/crypto/hisilicon/qm.h @@ -286,6 +286,18 @@ struct hisi_qm_list { bool (*check)(struct hisi_qm *qm); }; +struct qm_rsv_buf { + struct qm_sqc *sqc; + struct qm_cqc *cqc; + struct qm_eqc *eqc; + struct qm_aeqc *aeqc; + dma_addr_t sqc_dma; + dma_addr_t cqc_dma; + dma_addr_t eqc_dma; + dma_addr_t aeqc_dma; + struct qm_dma qcdma; +}; + struct hisi_qm { enum qm_hw_ver ver; enum qm_fun_type fun_type; @@ -309,6 +321,7 @@ struct hisi_qm { dma_addr_t cqc_dma; dma_addr_t eqe_dma; dma_addr_t aeqe_dma; + struct qm_rsv_buf xqc_buf; struct hisi_qm_status status; struct hisi_qm_err_ini err_ini; -- 2.30.0

2 1

[PATCH OLK-5.10 v4 0/7] AGDI: Add driver for Arm Generic Diagnostic Dump and Reset device
by Qian Zou 06 Nov '23

06 Nov '23

From: “z00447436” <zouqian4(a)huawei.com> Some use-cases, such as system management, require the ability to generate a non-maskable event to the OS to request the OS kernel to perform a diagnostic dump and reset the system. Arm Generic Diagnostic Dump and Reset device enables a maintainer to request OS to perform a diagnostic dump and reset a system via SDEI event or an interrupt. This patch implements SDEI path and discards the interrupted context before proceeding to the crash kernel. D Scott Phillips (1): arm64: sdei: abort running SDEI handlers during crash Ilkka Koskinen (4): ACPI: tables: Add AGDI to the list of known table signatures ACPI: AGDI: Add driver for Arm Generic Diagnostic Dump and Reset device ACPICA: iASL: Add suppport for AGDI table ACPI: AGDI: Fix missing prototype warning for acpi_agdi_init() Jia He (1): EDAC/ghes: Make ghes_edac a proper module Shuai Xue (1): ACPI: APEI: explicit init of HEST and GHES in apci_init() arch/arm64/include/asm/sdei.h | 6 ++ arch/arm64/kernel/entry.S | 27 +++++++- arch/arm64/kernel/sdei.c | 3 + arch/arm64/kernel/smp.c | 8 +-- drivers/acpi/apei/ghes.c | 23 +++---- drivers/acpi/arm64/Kconfig | 10 +++ drivers/acpi/arm64/Makefile | 1 + drivers/acpi/arm64/agdi.c | 117 ++++++++++++++++++++++++++++++++++ drivers/acpi/bus.c | 4 ++ drivers/acpi/pci_root.c | 3 - drivers/acpi/tables.c | 2 +- drivers/edac/Kconfig | 4 +- drivers/edac/ghes_edac.c | 40 +++++++++++- drivers/firmware/Kconfig | 1 + drivers/firmware/arm_sdei.c | 32 ++++++---- include/acpi/actbl2.h | 20 ++++++ include/acpi/apei.h | 4 +- include/acpi/ghes.h | 20 +----- include/linux/acpi_agdi.h | 13 ++++ include/linux/arm_sdei.h | 4 ++ 20 files changed, 283 insertions(+), 59 deletions(-) create mode 100644 drivers/acpi/arm64/agdi.c create mode 100644 include/linux/acpi_agdi.h -- 2.33.0

2 8

[PATCH OLK-5.10] blk-mq: avoid housekeeping CPUs scheduling a worker on a non-housekeeping CPU
by Xiongfeng Wang 06 Nov '23

06 Nov '23

hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I7ZBTV CVE: N/A ---------------------------------------------------- When NOHZ_FULL is enabled, such as in HPC situation, CPUs are divided into housekeeping CPUs and non-housekeeping CPUs. Non-housekeeping CPUs are NOHZ_FULL CPUs and are often monopolized by the userspace process, such HPC application process. Any sort of interruption is not expected. blk_mq_hctx_next_cpu() selects each cpu in 'hctx->cpumask' alternately to schedule the work thread blk_mq_run_work_fn(). When 'hctx->cpumask' contains housekeeping CPU and non-housekeeping CPU at the same time, a housekeeping CPU, which want to request a IO, may schedule a worker on a non-housekeeping CPU. This may affect the performance of the userspace application running on non-housekeeping CPUs. So let's just schedule the worker thread on the current CPU when the current CPU is housekeeping CPU. Signed-off-by: Xiongfeng Wang <wangxiongfeng2(a)huawei.com> --- block/blk-mq.c | 11 ++++++++++- include/linux/sched/isolation.h | 2 ++ kernel/sched/isolation.c | 8 ++++++++ 3 files changed, 20 insertions(+), 1 deletion(-) diff --git a/block/blk-mq.c b/block/blk-mq.c index 1d1200afb771..d0dce9ef9499 100644 --- a/block/blk-mq.c +++ b/block/blk-mq.c @@ -23,6 +23,7 @@ #include <linux/sched/sysctl.h> #include <linux/sched/topology.h> #include <linux/sched/signal.h> +#include <linux/sched/isolation.h> #include <linux/delay.h> #include <linux/crash_dump.h> #include <linux/prefetch.h> @@ -1676,6 +1677,8 @@ static int blk_mq_hctx_next_cpu(struct blk_mq_hw_ctx *hctx) static void __blk_mq_delay_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async, unsigned long msecs) { + int work_cpu; + if (unlikely(blk_mq_hctx_stopped(hctx))) return; @@ -1697,7 +1700,13 @@ static void __blk_mq_delay_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async, if (!percpu_ref_tryget(&hctx->queue->q_usage_counter)) return; - kblockd_mod_delayed_work_on(blk_mq_hctx_next_cpu(hctx), &hctx->run_work, + if (enhanced_isolcpus && tick_nohz_full_enabled() && + housekeeping_cpu(smp_processor_id(), HK_FLAG_WQ)) + work_cpu = smp_processor_id(); + else + work_cpu = blk_mq_hctx_next_cpu(hctx); + + kblockd_mod_delayed_work_on(work_cpu, &hctx->run_work, msecs_to_jiffies(msecs)); percpu_ref_put(&hctx->queue->q_usage_counter); } diff --git a/include/linux/sched/isolation.h b/include/linux/sched/isolation.h index cc9f393e2a70..2f93081ad7a0 100644 --- a/include/linux/sched/isolation.h +++ b/include/linux/sched/isolation.h @@ -18,6 +18,7 @@ enum hk_flags { }; #ifdef CONFIG_CPU_ISOLATION +extern bool enhanced_isolcpus; DECLARE_STATIC_KEY_FALSE(housekeeping_overridden); extern int housekeeping_any_cpu(enum hk_flags flags); extern const struct cpumask *housekeeping_cpumask(enum hk_flags flags); @@ -28,6 +29,7 @@ extern void __init housekeeping_init(void); #else +#define enhanced_isolcpus 0 static inline int housekeeping_any_cpu(enum hk_flags flags) { return smp_processor_id(); diff --git a/kernel/sched/isolation.c b/kernel/sched/isolation.c index 5a6ea03f9882..785ef5201116 100644 --- a/kernel/sched/isolation.c +++ b/kernel/sched/isolation.c @@ -198,3 +198,11 @@ static int __init housekeeping_isolcpus_setup(char *str) return housekeeping_setup(str, flags); } __setup("isolcpus=", housekeeping_isolcpus_setup); + +bool enhanced_isolcpus; +static int __init enhanced_isolcpus_setup(char *str) +{ + enhanced_isolcpus = true; + return 0; +} +__setup("enhanced_isolcpus", enhanced_isolcpus_setup); -- 2.20.1

3 3

[PATCH OLK-5.10 v3 0/7] AGDI: Add driver for Arm Generic Diagnostic Dump and Reset device
by Qian Zou 06 Nov '23

06 Nov '23

From: “z00447436” <zouqian4(a)huawei.com> Some use-cases, such as system management, require the ability to generate a non-maskable event to the OS to request the OS kernel to perform a diagnostic dump and reset the system. Arm Generic Diagnostic Dump and Reset device enables a maintainer to request OS to perform a diagnostic dump and reset a system via SDEI event or an interrupt. This patch implements SDEI path and discards the interrupted context before proceeding to the crash kernel. D Scott Phillips (1): arm64: sdei: abort running SDEI handlers during crash Ilkka Koskinen (4): ACPI: tables: Add AGDI to the list of known table signatures ACPI: AGDI: Add driver for Arm Generic Diagnostic Dump and Reset device ACPICA: iASL: Add suppport for AGDI table ACPI: AGDI: Fix missing prototype warning for acpi_agdi_init() Jia He (1): EDAC/ghes: Make ghes_edac a proper module Shuai Xue (1): ACPI: APEI: explicit init of HEST and GHES in apci_init() arch/arm64/include/asm/sdei.h | 6 ++ arch/arm64/kernel/entry.S | 27 +++++++- arch/arm64/kernel/sdei.c | 3 + arch/arm64/kernel/smp.c | 8 +-- drivers/acpi/apei/ghes.c | 23 +++---- drivers/acpi/arm64/Kconfig | 10 +++ drivers/acpi/arm64/Makefile | 1 + drivers/acpi/arm64/agdi.c | 117 ++++++++++++++++++++++++++++++++++ drivers/acpi/bus.c | 4 ++ drivers/acpi/pci_root.c | 3 - drivers/acpi/tables.c | 2 +- drivers/edac/Kconfig | 4 +- drivers/edac/ghes_edac.c | 40 +++++++++++- drivers/firmware/Kconfig | 1 + drivers/firmware/arm_sdei.c | 32 ++++++---- include/acpi/actbl2.h | 20 ++++++ include/acpi/apei.h | 4 +- include/acpi/ghes.h | 25 -------- include/linux/acpi_agdi.h | 13 ++++ include/linux/arm_sdei.h | 4 ++ 20 files changed, 281 insertions(+), 66 deletions(-) create mode 100644 drivers/acpi/arm64/agdi.c create mode 100644 include/linux/acpi_agdi.h -- 2.33.0

2 8

[PATCH OLK-5.10 v2 0/7] AGDI: Add driver for Arm Generic Diagnostic Dump and Reset device
by Qian Zou 06 Nov '23

06 Nov '23

From: “z00447436” <zouqian4(a)huawei.com> Some use-cases, such as system management, require the ability to generate a non-maskable event to the OS to request the OS kernel to perform a diagnostic dump and reset the system. Arm Generic Diagnostic Dump and Reset device enables a maintainer to request OS to perform a diagnostic dump and reset a system via SDEI event or an interrupt. This patch implements SDEI path and discards the interrupted context before proceeding to the crash kernel. D Scott Phillips (1): arm64: sdei: abort running SDEI handlers during crash Ilkka Koskinen (4): ACPI: tables: Add AGDI to the list of known table signatures ACPI: AGDI: Add driver for Arm Generic Diagnostic Dump and Reset device ACPICA: iASL: Add suppport for AGDI table ACPI: AGDI: Fix missing prototype warning for acpi_agdi_init() Jia He (1): EDAC/ghes: Make ghes_edac a proper module Shuai Xue (1): ACPI: APEI: explicit init of HEST and GHES in apci_init() arch/arm64/include/asm/sdei.h | 6 ++ arch/arm64/kernel/entry.S | 27 +++++++- arch/arm64/kernel/sdei.c | 3 + arch/arm64/kernel/smp.c | 8 +-- drivers/acpi/apei/ghes.c | 23 +++---- drivers/acpi/arm64/Kconfig | 10 +++ drivers/acpi/arm64/Makefile | 1 + drivers/acpi/arm64/agdi.c | 117 ++++++++++++++++++++++++++++++++++ drivers/acpi/bus.c | 4 ++ drivers/acpi/pci_root.c | 3 - drivers/acpi/tables.c | 2 +- drivers/edac/Kconfig | 4 +- drivers/edac/ghes_edac.c | 40 +++++++++++- drivers/firmware/Kconfig | 1 + drivers/firmware/arm_sdei.c | 32 ++++++---- include/acpi/actbl2.h | 20 ++++++ include/acpi/apei.h | 4 +- include/acpi/ghes.h | 25 -------- include/linux/acpi_agdi.h | 13 ++++ include/linux/arm_sdei.h | 4 ++ 20 files changed, 281 insertions(+), 66 deletions(-) create mode 100644 drivers/acpi/arm64/agdi.c create mode 100644 include/linux/acpi_agdi.h -- 2.33.0

2 8

[PATCH openEuler-1.0-LTS] fs: lockd: avoid possible wrong NULL parameter
by ZhaoLong Wang 06 Nov '23

06 Nov '23

From: Su Hui <suhui(a)nfschina.com> stable inclusion from stable-v4.19.295 commit 35f0749756b848ad4f4a165ad6b1dfa8d0e45a96 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8E5Q5 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit de8d38cf44bac43e83bad28357ba84784c412752 ] clang's static analysis warning: fs/lockd/mon.c: line 293, column 2: Null pointer passed as 2nd argument to memory copy function. Assuming 'hostname' is NULL and calling 'nsm_create_handle()', this will pass NULL as 2nd argument to memory copy function 'memcpy()'. So return NULL if 'hostname' is invalid. Fixes: 77a3ef33e2de ("NSM: More clean up of nsm_get_handle()") Signed-off-by: Su Hui <suhui(a)nfschina.com> Reviewed-by: Nick Desaulniers <ndesaulniers(a)google.com> Reviewed-by: Jeff Layton <jlayton(a)kernel.org> Signed-off-by: Chuck Lever <chuck.lever(a)oracle.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: ZhaoLong Wang <wangzhaolong1(a)huawei.com> --- fs/lockd/mon.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c index 654594ef4f94..68a2eac548c3 100644 --- a/fs/lockd/mon.c +++ b/fs/lockd/mon.c @@ -275,6 +275,9 @@ static struct nsm_handle *nsm_create_handle(const struct sockaddr *sap, { struct nsm_handle *new; + if (!hostname) + return NULL; + new = kzalloc(sizeof(*new) + hostname_len + 1, GFP_KERNEL); if (unlikely(new == NULL)) return NULL; -- 2.34.3

2 1

[PATCH openEuler-1.0-LTS] crypto: hisilicon - qm obtain the mailbox config at one time
by wangyuan 06 Nov '23

06 Nov '23

From: Yu'an Wang <wangyuan46(a)huawei.com> driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8E0N4 CVE: NA -------------------------------- The mailbox configuration needs to be obtained from the hardware at one time. If the mailbox configuration is obtained for multiple times, the read value may be incorrect. Use the instruction to read mailbox data instead of readl. Signed-off-by: Yu'an Wang <wangyuan46(a)huawei.com> --- drivers/crypto/hisilicon/qm.c | 187 +++++++++++++++++++++++++--------- 1 file changed, 137 insertions(+), 50 deletions(-) diff --git a/drivers/crypto/hisilicon/qm.c b/drivers/crypto/hisilicon/qm.c index f2706dc0d55e..64e8ef1bf8e7 100644 --- a/drivers/crypto/hisilicon/qm.c +++ b/drivers/crypto/hisilicon/qm.c @@ -37,6 +37,7 @@ #define QM_MB_OP_SHIFT 14 #define QM_MB_CMD_DATA_ADDR_L 0x304 #define QM_MB_CMD_DATA_ADDR_H 0x308 +#define QM_MB_STATUS_MASK GENMASK(12, 9) /* sqc shift */ #define QM_SQ_HOP_NUM_SHIFT 0 @@ -113,7 +114,7 @@ #define QM_SQC_VFT_NUM_MASK_v2 0x3ff #define QM_DFX_CNT_CLR_CE 0x100118 -#define QM_IN_IDLE_ST_REG 0x1040e4 +#define QM_IN_IDLE_ST_REG 0x1040e4 #define QM_ABNORMAL_INT_SOURCE 0x100000 #define QM_ABNORMAL_INT_MASK 0x100004 @@ -169,6 +170,8 @@ #define TASK_TIMEOUT 10000 #define WAIT_PERIOD 20 +#define QM_MB_WAIT_READY_CNT 10 +#define QM_MB_MAX_WAIT_CNT 21000 #define WAIT_PERIOD_US_MAX 200 #define WAIT_PERIOD_US_MIN 100 #define REMOVE_WAIT_DELAY 10 @@ -382,16 +385,6 @@ static bool qm_qp_avail_state(struct hisi_qm *qm, struct hisi_qp *qp, return avail; } -/* return 0 mailbox ready, -ETIMEDOUT hardware timeout */ -static int qm_wait_mb_ready(struct hisi_qm *qm) -{ - u32 val; - - return readl_relaxed_poll_timeout(qm->io_base + QM_MB_CMD_SEND_BASE, - val, !((val >> QM_MB_BUSY_SHIFT) & - 0x1), POLL_PERIOD, POLL_TIMEOUT); -} - /* 128 bit should be wrote to hardware at one time to trigger a mailbox */ static void qm_mb_write(struct hisi_qm *qm, const void *src) { @@ -400,7 +393,7 @@ static void qm_mb_write(struct hisi_qm *qm, const void *src) asm volatile("ldp %0, %1, %3\n" "stp %0, %1, %2\n" - "dsb sy\n" + "dmb oshst\n" : "=&r" (tmp0), "=&r" (tmp1), "+Q" (*((char __iomem *)fun_base)) @@ -408,45 +401,142 @@ static void qm_mb_write(struct hisi_qm *qm, const void *src) : "memory"); } -static int qm_mb(struct hisi_qm *qm, u8 cmd, dma_addr_t dma_addr, u16 queue, - bool op) +/* 128 bit should be read from hardware at one time*/ +static void qm_mb_read(struct hisi_qm *qm, void *dst) +{ + const void __iomem *fun_base = qm->io_base + QM_MB_CMD_SEND_BASE; + unsigned long tmp0 = 0, tmp1 = 0; + + asm volatile("ldp %0, %1, %3\n" + "stp %0, %1, %2\n" + "dmb oshst\n" + : "=&r" (tmp0), + "=&r" (tmp1), + "+Q" (*((char *)dst)) + : "Q" (*((char __iomem *)fun_base)) + : "memory"); +} + +static int qm_wait_mb_ready(struct hisi_qm *qm) { struct qm_mailbox mailbox; - int ret = 0; + int i = 0; + + while (i++ < QM_MB_WAIT_READY_CNT) { + qm_mb_read(qm, &mailbox); + if (!((le16_to_cpu(mailbox.w0) >> QM_MB_BUSY_SHIFT) & 0x1)) + return 0; - dev_dbg(&qm->pdev->dev, "QM mailbox request to q%u: %u\n", queue, cmd); + usleep_range(WAIT_PERIOD_US_MIN, WAIT_PERIOD_US_MAX); + } - mailbox.w0 = cpu_to_le16(cmd | - (op ? 0x1 << QM_MB_OP_SHIFT : 0) | - (0x1 << QM_MB_BUSY_SHIFT)); - mailbox.queue_num = cpu_to_le16(queue); - mailbox.base_l = cpu_to_le32(lower_32_bits(dma_addr)); - mailbox.base_h = cpu_to_le32(upper_32_bits(dma_addr)); - mailbox.rsvd = 0; + dev_err(&qm->pdev->dev, "QM mailbox is busy to start!\n"); - mutex_lock(&qm->mailbox_lock); + return -EBUSY; +} + +static int qm_wait_mb_finish(struct hisi_qm *qm, struct qm_mailbox *mailbox) +{ + int i = 0; + + while (++i) { + qm_mb_read(qm, mailbox); + if (!((le16_to_cpu(mailbox->w0) >> QM_MB_BUSY_SHIFT) & 0x1)) + break; + + if (i > QM_MB_MAX_WAIT_CNT) { + dev_err(&qm->pdev->dev, + "QM mailbox operation timeout!\n"); + return -ETIMEDOUT; + } - if (unlikely(qm_wait_mb_ready(qm))) { - ret = -EBUSY; - dev_err(&qm->pdev->dev, "QM mailbox is busy to start!\n"); - goto busy_unlock; + usleep_range(WAIT_PERIOD_US_MIN, WAIT_PERIOD_US_MAX); + } + + if (le16_to_cpu(mailbox->w0) & QM_MB_STATUS_MASK) { + dev_err(&qm->pdev->dev, "QM mailbox operation failed!\n"); + return -EIO; } - qm_mb_write(qm, &mailbox); + return 0; +} - if (unlikely(qm_wait_mb_ready(qm))) { - ret = -EBUSY; - dev_err(&qm->pdev->dev, "QM mailbox operation timeout!\n"); - goto busy_unlock; +static int qm_mb_nolock(struct hisi_qm *qm, struct qm_mailbox *mailbox) +{ + int ret; + + ret = qm_wait_mb_ready(qm); + if (ret) + goto mb_busy; + + qm_mb_write(qm, mailbox); + + ret = qm_wait_mb_finish(qm, mailbox); + if (ret) + goto mb_busy; + + return 0; + +mb_busy: + atomic64_inc(&qm->debug.dfx.mb_err_cnt); + return ret; +} + +static void qm_mb_pre_init(struct qm_mailbox *mailbox, u8 cmd, + u64 base, u16 queue, bool op) +{ + mailbox->w0 = cpu_to_le16((cmd) | + ((op) ? 0x1 << QM_MB_OP_SHIFT : 0) | + (0x1 << QM_MB_BUSY_SHIFT)); + mailbox->queue_num = cpu_to_le16(queue); + mailbox->base_l = cpu_to_le32(lower_32_bits(base)); + mailbox->base_h = cpu_to_le32(upper_32_bits(base)); + mailbox->rsvd = 0; +} + +static int qm_check_dev_error(struct hisi_qm *qm); +static int hisi_qm_mb_write(struct hisi_qm *qm, u8 cmd, dma_addr_t dma_addr, + u16 queue, bool op) +{ + struct hisi_qm *pf_qm = pci_get_drvdata(pci_physfn(qm->pdev)); + struct qm_mailbox mailbox; + int ret; + + dev_dbg(&qm->pdev->dev, "QM mailbox request to q%u: %u.\n", queue, cmd); + + /* No need to judge if master OOO is blocked. */ + if (qm_check_dev_error(pf_qm)) { + dev_err(&qm->pdev->dev, + "QM mailbox operation failed since qm is stop!\n"); + return -EIO; } -busy_unlock: + qm_mb_pre_init(&mailbox, cmd, dma_addr, queue, op); + mutex_lock(&qm->mailbox_lock); + ret = qm_mb_nolock(qm, &mailbox); mutex_unlock(&qm->mailbox_lock); - if (ret) - atomic64_inc(&qm->debug.dfx.mb_err_cnt); + return ret; } +static int hisi_qm_mb_read(struct hisi_qm *qm, u64 *base, u8 cmd, u16 queue) +{ + struct qm_mailbox mailbox; + int ret; + + qm_mb_pre_init(&mailbox, cmd, 0, queue, 1); + mutex_lock(&qm->mailbox_lock); + ret = qm_mb_nolock(qm, &mailbox); + mutex_unlock(&qm->mailbox_lock); + if (ret) + return ret; + + *base = le32_to_cpu(mailbox.base_l) | + ((u64)le32_to_cpu(mailbox.base_h) << 32); + + return 0; +} + static void qm_db_v1(struct hisi_qm *qm, u16 qn, u8 cmd, u16 index, u8 priority) { u64 doorbell; @@ -826,12 +916,10 @@ static int qm_get_vft_v2(struct hisi_qm *qm, u32 *base, u32 *number) u64 sqc_vft; int ret; - ret = qm_mb(qm, QM_MB_CMD_SQC_VFT_V2, 0, 0, 1); + ret = hisi_qm_mb_read(qm, &sqc_vft, QM_MB_CMD_SQC_VFT_V2, 0); if (ret) return ret; - sqc_vft = readl(qm->io_base + QM_MB_CMD_DATA_ADDR_L) | - ((u64)readl(qm->io_base + QM_MB_CMD_DATA_ADDR_H) << 32); *base = QM_SQC_VFT_BASE_MASK_V2 & (sqc_vft >> QM_SQC_VFT_BASE_SHIFT_V2); *number = (QM_SQC_VFT_NUM_MASK_v2 & (sqc_vft >> QM_SQC_VFT_NUM_SHIFT_V2)) + 1; @@ -1144,12 +1232,12 @@ static int dump_show(struct hisi_qm *qm, void *info, static int qm_dump_sqc_raw(struct hisi_qm *qm, dma_addr_t dma_addr, u16 qp_id) { - return qm_mb(qm, QM_MB_CMD_SQC, dma_addr, qp_id, 1); + return hisi_qm_mb_write(qm, QM_MB_CMD_SQC, dma_addr, qp_id, 1); } static int qm_dump_cqc_raw(struct hisi_qm *qm, dma_addr_t dma_addr, u16 qp_id) { - return qm_mb(qm, QM_MB_CMD_CQC, dma_addr, qp_id, 1); + return hisi_qm_mb_write(qm, QM_MB_CMD_CQC, dma_addr, qp_id, 1); } static int qm_sqc_dump(struct hisi_qm *qm, const char *s) @@ -1260,7 +1348,7 @@ static int qm_eqc_dump(struct hisi_qm *qm, char *s) if (IS_ERR(eqc)) return PTR_ERR(eqc); - ret = qm_mb(qm, QM_MB_CMD_EQC, eqc_dma, 0, 1); + ret = hisi_qm_mb_write(qm, QM_MB_CMD_EQC, eqc_dma, 0, 1); if (ret) goto mailbox_fail; @@ -1289,7 +1377,7 @@ static int qm_aeqc_dump(struct hisi_qm *qm, char *s) if (IS_ERR(aeqc)) return PTR_ERR(aeqc); - ret = qm_mb(qm, QM_MB_CMD_AEQC, aeqc_dma, 0, 1); + ret = hisi_qm_mb_write(qm, QM_MB_CMD_AEQC, aeqc_dma, 0, 1); if (ret) goto mailbox_fail; @@ -1834,7 +1922,7 @@ static int qm_sq_ctx_cfg(struct hisi_qp *qp, int qp_id, int pasid) sqc->cq_num = cpu_to_le16(qp_id); sqc->w13 = cpu_to_le16(QM_MK_SQC_W13(0, 1, qp->alg_type)); - ret = qm_mb(qm, QM_MB_CMD_SQC, sqc_dma, qp_id, 0); + ret = hisi_qm_mb_write(qm, QM_MB_CMD_SQC, sqc_dma, qp_id, 0); dma_unmap_single(dev, sqc_dma, sizeof(struct qm_sqc), DMA_TO_DEVICE); kfree(sqc); @@ -1874,7 +1962,7 @@ static int qm_cq_ctx_cfg(struct hisi_qp *qp, int qp_id, int pasid) cqc->dw6 = cpu_to_le32(1 << QM_CQ_PHASE_SHIFT | qp->c_flag << QM_CQ_FLAG_SHIFT); - ret = qm_mb(qm, QM_MB_CMD_CQC, cqc_dma, qp_id, 0); + ret = hisi_qm_mb_write(qm, QM_MB_CMD_CQC, cqc_dma, qp_id, 0); dma_unmap_single(dev, cqc_dma, sizeof(struct qm_cqc), DMA_TO_DEVICE); kfree(cqc); @@ -2894,7 +2982,7 @@ static int qm_eq_ctx_cfg(struct hisi_qm *qm) if (qm->ver == QM_HW_V1) eqc->dw3 = cpu_to_le32(QM_EQE_AEQE_SIZE); eqc->dw6 = cpu_to_le32((QM_EQ_DEPTH - 1) | (1 << QM_EQC_PHASE_SHIFT)); - ret = qm_mb(qm, QM_MB_CMD_EQC, eqc_dma, 0, 0); + ret = hisi_qm_mb_write(qm, QM_MB_CMD_EQC, eqc_dma, 0, 0); dma_unmap_single(dev, eqc_dma, sizeof(struct qm_eqc), DMA_TO_DEVICE); kfree(eqc); @@ -2921,7 +3009,7 @@ static int qm_aeq_ctx_cfg(struct hisi_qm *qm) aeqc->base_l = cpu_to_le32(lower_32_bits(qm->aeqe_dma)); aeqc->base_h = cpu_to_le32(upper_32_bits(qm->aeqe_dma)); aeqc->dw6 = cpu_to_le32((QM_Q_DEPTH - 1) | (1 << QM_EQC_PHASE_SHIFT)); - ret = qm_mb(qm, QM_MB_CMD_AEQC, aeqc_dma, 0, 0); + ret = hisi_qm_mb_write(qm, QM_MB_CMD_AEQC, aeqc_dma, 0, 0); dma_unmap_single(dev, aeqc_dma, sizeof(struct qm_aeqc), DMA_TO_DEVICE); kfree(aeqc); @@ -2969,11 +3057,11 @@ static int __hisi_qm_start(struct hisi_qm *qm) if (ret) return ret; - ret = qm_mb(qm, QM_MB_CMD_SQC_BT, qm->sqc_dma, 0, 0); + ret = hisi_qm_mb_write(qm, QM_MB_CMD_SQC_BT, qm->sqc_dma, 0, 0); if (ret) return ret; - ret = qm_mb(qm, QM_MB_CMD_CQC_BT, qm->cqc_dma, 0, 0); + ret = hisi_qm_mb_write(qm, QM_MB_CMD_CQC_BT, qm->cqc_dma, 0, 0); if (ret) return ret; @@ -3194,7 +3282,6 @@ static ssize_t qm_status_read(struct file *filp, char __user *buffer, return (*pos = len); } - static const struct file_operations qm_status_fops = { .owner = THIS_MODULE, .open = simple_open, -- 2.30.0

2 1

[PATCH V5 OLK-5.10] soc: hisilicon: hisi_hbmdev: Add hbm acls repair and query methods
by Zhang Zekun 06 Nov '23

06 Nov '23

hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8CCP5 ---------------------------------- Hbm memory device add support for acls hot repair. The patch add two methods for userpace: - query a paddr if it support acls repair - repair a paddr in hbm memory device The feature of ACLS hot repair can help to fix a memory error from userspace by passing through the error physical address to HBM hardware. Signed-off-by: Zhang Zekun <zhangzekun11(a)huawei.com> --- drivers/soc/hisilicon/Kconfig | 10 +++ drivers/soc/hisilicon/hisi_hbmdev.c | 134 ++++++++++++++++++++++++++++ 2 files changed, 144 insertions(+) diff --git a/drivers/soc/hisilicon/Kconfig b/drivers/soc/hisilicon/Kconfig index 383375f20cac..45754528873e 100644 --- a/drivers/soc/hisilicon/Kconfig +++ b/drivers/soc/hisilicon/Kconfig @@ -44,4 +44,14 @@ config HISI_HBMCACHE To compile the driver as a module, choose M here: the module will be called hisi_hbmcache. +config HISI_HBMDEV_ACLS + bool "Add support for HISI ACLS repair" + depends on HISI_HBMDEV + help + Add ACLS support for hbm device, which can be used to query and + repair hardware error in HBM devices. This feature need to work with + hardware firmwares. + + If not sure say no. + endmenu diff --git a/drivers/soc/hisilicon/hisi_hbmdev.c b/drivers/soc/hisilicon/hisi_hbmdev.c index 5b6b1618148c..a9cc78bde81b 100644 --- a/drivers/soc/hisilicon/hisi_hbmdev.c +++ b/drivers/soc/hisilicon/hisi_hbmdev.c @@ -11,6 +11,7 @@ #include <linux/node.h> #include <linux/arch_topology.h> #include <linux/memory_hotplug.h> +#include <linux/mm.h> #include "hisi_internal.h" @@ -25,6 +26,9 @@ struct cdev_node { struct memory_dev { struct kobject *memdev_kobj; struct kobject *topo_kobj; +#ifdef CONFIG_HISI_HBMDEV_ACLS + struct kobject *acls_kobj; +#endif struct cdev_node cdev_list; nodemask_t cluster_cpumask[MAX_NUMNODES]; }; @@ -85,6 +89,134 @@ static void memory_topo_init(void) kobject_put(mdev->topo_kobj); } +#ifdef CONFIG_HISI_HBMDEV_ACLS +static struct acpi_device *paddr_to_acpi_device(u64 paddr) +{ + unsigned long pfn; + int nid; + + pfn = __phys_to_pfn(paddr); + if (!pfn_valid(pfn)) + return NULL; + + nid = pfn_to_nid(pfn); + if (nid < 0 && nid >= MAX_NUMNODES) + return NULL; + + return hotplug_mdev[nid]; +} + +static ssize_t acls_query_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t count) +{ + struct acpi_object_list arg_list; + struct acpi_device *adev; + union acpi_object obj; + acpi_status status; + u64 paddr, res; + + if (kstrtoull(buf, 16, &paddr)) + return -EINVAL; + + adev = paddr_to_acpi_device(paddr); + if (!adev) + return -EINVAL; + + obj.type = ACPI_TYPE_INTEGER; + obj.integer.value = paddr; + arg_list.count = 1; + arg_list.pointer = &obj; + + status = acpi_evaluate_integer(adev->handle, "AQRY", &arg_list, &res); + if (ACPI_FAILURE(status)) + return -ENODEV; + + /* AQRY will return a positive error code to represent error status */ + if (IS_ERR_VALUE(-res)) + return -res; + else if (res) + return -ENODEV; + + return count; +} + +static struct kobj_attribute acls_query_store_attribute = + __ATTR(acls_query, 0200, NULL, acls_query_store); + +static ssize_t acls_repair_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t count) +{ + struct acpi_object_list arg_list; + struct acpi_device *adev; + union acpi_object obj; + acpi_status status; + u64 paddr, res; + + if (kstrtoull(buf, 16, &paddr)) + return -EINVAL; + + adev = paddr_to_acpi_device(paddr); + if (!adev) + return -EINVAL; + + obj.type = ACPI_TYPE_INTEGER; + obj.integer.value = paddr; + arg_list.count = 1; + arg_list.pointer = &obj; + + status = acpi_evaluate_integer(adev->handle, "AREP", &arg_list, &res); + if (ACPI_FAILURE(status)) + return -ENODEV; + + /* AREP will return a positive error code to represent error status */ + if (IS_ERR_VALUE(-res)) + return -res; + else if (res) + return -ENODEV; + + return count; +} +static struct kobj_attribute acls_repair_store_attribute = + __ATTR(acls_repair, 0200, NULL, acls_repair_store); + +static struct attribute *acls_attrs[] = { + &acls_query_store_attribute.attr, + &acls_repair_store_attribute.attr, + NULL, +}; + +static struct attribute_group acls_attr_group = { + .attrs = acls_attrs, +}; + +static void acls_init(void) +{ + int ret = -ENOMEM; + + mdev->acls_kobj = kobject_create_and_add("acls", mdev->memdev_kobj); + if (!mdev->acls_kobj) + goto out; + + ret = sysfs_create_group(mdev->acls_kobj, &acls_attr_group); + if (ret) + kobject_put(mdev->acls_kobj); + +out: + if (ret) + pr_err("ACLS hot repair is not enabled\n"); +} + +static void acls_remove(void) +{ + kobject_put(mdev->acls_kobj); +} +#else +static void acls_init(void) {} +static void acls_remove(void) {} +#endif + static int get_pxm(struct acpi_device *acpi_device, void *arg) { acpi_handle handle = acpi_device->handle; @@ -284,6 +416,7 @@ static int __init mdev_init(void) } memory_topo_init(); + acls_init(); return ret; } module_init(mdev_init); @@ -293,6 +426,7 @@ static void __exit mdev_exit(void) container_remove(); kobject_put(mdev->memdev_kobj); kobject_put(mdev->topo_kobj); + acls_remove(); kfree(mdev); } module_exit(mdev_exit); -- 2.17.1

3 2

[PATCH v4 OLK-5.10 00/29] Introduce some vdpa ops to support vdpa device live migrate
by Jiang Dongxu 06 Nov '23

06 Nov '23

From: jiangdongxu <jiangdongxu1(a)huawei.com> Patch 1-17: some bugfix and ops intruduced by upstream Patch 18-19: introduce vdpa device logging ops Patch 20-21: introduce vdpa device state ops Patch 22-23: introduce vdpa device migrate state ops Patch 24: introduce new vhost feature BYTEPMAPLOG Patch 25: export iommu_get_resv_regions/iommu_set_resv_regions Patch 26-27: some optimization about vhost-vdpa Patch 28: add vdpa/vhost-vdpa build config Patch 29: fix vhost-vdpa compile warnings Arnaldo Carvalho de Melo (1): tools include UAPI: Sync linux/vhost.h with the kernel sources Cindy Lu (2): vhost_vdpa: fix the crash in unmap a large memory vhost_vdpa: fix unmap process in no-batch mode Eugenio Pérez (1): vdpa: add get_backend_features vdpa operation Gautam Dawar (1): vhost-vdpa: free iommu domain after last use during cleanup Greg Kroah-Hartman (1): vhost-vdpa: vhost_vdpa_alloc_domain() should be using a const struct bus_type * Jason Gunthorpe (1): PCI/IOV: Add pci_iov_vf_id() to get VF index Sebastien Boeuf (3): vdpa: Add resume operation vhost-vdpa: Introduce RESUME backend feature bit vhost-vdpa: uAPI to resume the device Shannon Nelson (2): vhost_vdpa: tell vqs about the negotiated vhost_vdpa: support PACKED when setting-getting vring_base Shunsuke Mie (1): virtio: fix virtio transitional ids Stefano Garzarella (3): vhost-vdpa: fix an iotlb memory leak vdpa: add bind_mm/unbind_mm callbacks vhost-vdpa: use bind_mm/unbind_mm device callbacks Zhu Lingshan (1): virtio: update virtio id table, add transitional ids jiangdongxu (12): vdpa: add log operations vhost-vdpa: add uAPI for logging vdpa: add device state operations vhost-vdpa: add uAPI for device buffer vdpa: add vdpa device migration status ops vhost-vdpa: add uAPI for device migration status vhost: add VHOST feature VHOST_BACKEND_F_BYTEMAPLOG export iommu_get_resv_regions and iommu_set_resv_regions vhost-vdpa: Allow transparent MSI IOV vhost-vdpa: fix msi irq request err arm64: openeuler_defconfig: add VDPA config vhost-vdpa: fix compile warnings arch/arm64/configs/openeuler_defconfig | 6 +- drivers/iommu/iommu.c | 2 + drivers/pci/iov.c | 14 + drivers/vhost/vdpa.c | 383 ++++++++++++++++++++++--- include/linux/pci.h | 8 +- include/linux/vdpa.h | 60 +++- include/uapi/linux/vhost.h | 20 ++ include/uapi/linux/vhost_types.h | 21 ++ include/uapi/linux/virtio_ids.h | 12 + tools/include/uapi/linux/vhost.h | 8 + 10 files changed, 489 insertions(+), 45 deletions(-) -- 2.27.0

2 30

[PATCH V4 OLK-5.10] soc: hisilicon: hisi_hbmdev: Add hbm acls repair and query methods
by Zhang Zekun 06 Nov '23

06 Nov '23

hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8CCP5 ---------------------------------- Hbm memory device add support for acls hot repair. The patch add two methods for userpace: - query a paddr if it support acls repair - repair a paddr in hbm memory device The feature of ACLS hot repair can help to fix a memory error from userspace by passing through the error physical address to HBM hardware. Signed-off-by: Zhang Zekun <zhangzekun11(a)huawei.com> --- v4 - remove "Offering HULK" to pass CI patch format check v3 - Add #include <linux/mm.h> to avoid compile failure in arch x86_64 v2: - Change "u64 pfn" ot "unsigned long pfn" to avoid compile error - Add two stub to pretiffy the code drivers/soc/hisilicon/Kconfig | 10 +++ drivers/soc/hisilicon/hisi_hbmdev.c | 132 ++++++++++++++++++++++++++++ 2 files changed, 142 insertions(+) diff --git a/drivers/soc/hisilicon/Kconfig b/drivers/soc/hisilicon/Kconfig index 383375f20cac..45754528873e 100644 --- a/drivers/soc/hisilicon/Kconfig +++ b/drivers/soc/hisilicon/Kconfig @@ -44,4 +44,14 @@ config HISI_HBMCACHE To compile the driver as a module, choose M here: the module will be called hisi_hbmcache. +config HISI_HBMDEV_ACLS + bool "Add support for HISI ACLS repair" + depends on HISI_HBMDEV + help + Add ACLS support for hbm device, which can be used to query and + repair hardware error in HBM devices. This feature need to work with + hardware firmwares. + + If not sure say no. + endmenu diff --git a/drivers/soc/hisilicon/hisi_hbmdev.c b/drivers/soc/hisilicon/hisi_hbmdev.c index 5b6b1618148c..8348abc6144c 100644 --- a/drivers/soc/hisilicon/hisi_hbmdev.c +++ b/drivers/soc/hisilicon/hisi_hbmdev.c @@ -11,6 +11,7 @@ #include <linux/node.h> #include <linux/arch_topology.h> #include <linux/memory_hotplug.h> +#include <linux/mm.h> #include "hisi_internal.h" @@ -25,6 +26,9 @@ struct cdev_node { struct memory_dev { struct kobject *memdev_kobj; struct kobject *topo_kobj; +#ifdef CONFIG_HISI_HBMDEV_ACLS + struct kobject *acls_kobj; +#endif struct cdev_node cdev_list; nodemask_t cluster_cpumask[MAX_NUMNODES]; }; @@ -85,6 +89,132 @@ static void memory_topo_init(void) kobject_put(mdev->topo_kobj); } +#ifdef CONFIG_HISI_HBMDEV_ACLS +static struct acpi_device *paddr_to_acpi_device(u64 paddr) +{ + unsigned long pfn; + int nid; + + pfn = __phys_to_pfn(paddr); + if (!pfn_valid(pfn)) + return NULL; + + nid = pfn_to_nid(pfn); + if (nid < 0 && nid >= MAX_NUMNODES) + return NULL; + + return hotplug_mdev[nid]; +} + +static ssize_t acls_query_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t count) +{ + struct acpi_object_list arg_list; + struct acpi_device *adev; + union acpi_object obj; + acpi_status status; + u64 paddr, res; + + if (kstrtoull(buf, 16, &paddr)) + return -EINVAL; + + adev = paddr_to_acpi_device(paddr); + if (!adev) + return -EINVAL; + + obj.type = ACPI_TYPE_INTEGER; + obj.integer.value = paddr; + arg_list.count = 1; + arg_list.pointer = &obj; + + status = acpi_evaluate_integer(adev->handle, "AQRY", &arg_list, &res); + if (ACPI_FAILURE(status)) + return -ENODEV; + + if (IS_ERR_VALUE(-res)) + return -res; + else if (res) + return -ENODEV; + + return count; +} + +static struct kobj_attribute acls_query_store_attribute = + __ATTR(acls_query, 0200, NULL, acls_query_store); + +static ssize_t acls_repair_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t count) +{ + struct acpi_object_list arg_list; + struct acpi_device *adev; + union acpi_object obj; + acpi_status status; + u64 paddr, res; + + if (kstrtoull(buf, 16, &paddr)) + return -EINVAL; + + adev = paddr_to_acpi_device(paddr); + if (!adev) + return -EINVAL; + + obj.type = ACPI_TYPE_INTEGER; + obj.integer.value = paddr; + arg_list.count = 1; + arg_list.pointer = &obj; + + status = acpi_evaluate_integer(adev->handle, "AREP", &arg_list, &res); + if (ACPI_FAILURE(status)) + return -ENODEV; + + if (IS_ERR_VALUE(-res)) + return -res; + else if (res) + return -ENODEV; + + return count; +} +static struct kobj_attribute acls_repair_store_attribute = + __ATTR(acls_repair, 0200, NULL, acls_repair_store); + +static struct attribute *acls_attrs[] = { + &acls_query_store_attribute.attr, + &acls_repair_store_attribute.attr, + NULL, +}; + +static struct attribute_group acls_attr_group = { + .attrs = acls_attrs, +}; + +static void acls_init(void) +{ + int ret = -ENOMEM; + + mdev->acls_kobj = kobject_create_and_add("acls", mdev->memdev_kobj); + if (!mdev->acls_kobj) + goto out; + + ret = sysfs_create_group(mdev->acls_kobj, &acls_attr_group); + if (ret) + kobject_put(mdev->acls_kobj); + +out: + if (ret) + pr_err("ACLS hot repair is not enabled\n"); +} + +static void acls_remove(void) +{ + kobject_put(mdev->acls_kobj); +} +#else +static void acls_init(void) {} +static void acls_remove(void) {} +#endif + static int get_pxm(struct acpi_device *acpi_device, void *arg) { acpi_handle handle = acpi_device->handle; @@ -284,6 +414,7 @@ static int __init mdev_init(void) } memory_topo_init(); + acls_init(); return ret; } module_init(mdev_init); @@ -293,6 +424,7 @@ static void __exit mdev_exit(void) container_remove(); kobject_put(mdev->memdev_kobj); kobject_put(mdev->topo_kobj); + acls_remove(); kfree(mdev); } module_exit(mdev_exit); -- 2.17.1

2 1

[PATCH V3 OLK-5.10] soc: hisilicon: hisi_hbmdev: Add hbm acls repair and query methods
by Zhang Zekun 06 Nov '23

06 Nov '23

Offering: HULK hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8CCP5 ---------------------------------- Hbm memory device add support for acls hot repair. The patch add two methods for userpace: - query a paddr if it support acls repair - repair a paddr in hbm memory device The feature of ACLS hot repair can help to fix a memory error from userspace by passing through the error physical address to HBM hardware. Signed-off-by: Zhang Zekun <zhangzekun11(a)huawei.com> --- v3 - Add #include <linux/mm.h> to avoid compile failure in arch x86_64 v2: - Change "u64 pfn" ot "unsigned long pfn" to avoid compile error - Add two stub to pretiffy the code drivers/soc/hisilicon/Kconfig | 10 +++ drivers/soc/hisilicon/hisi_hbmdev.c | 132 ++++++++++++++++++++++++++++ 2 files changed, 142 insertions(+) diff --git a/drivers/soc/hisilicon/Kconfig b/drivers/soc/hisilicon/Kconfig index 383375f20cac..45754528873e 100644 --- a/drivers/soc/hisilicon/Kconfig +++ b/drivers/soc/hisilicon/Kconfig @@ -44,4 +44,14 @@ config HISI_HBMCACHE To compile the driver as a module, choose M here: the module will be called hisi_hbmcache. +config HISI_HBMDEV_ACLS + bool "Add support for HISI ACLS repair" + depends on HISI_HBMDEV + help + Add ACLS support for hbm device, which can be used to query and + repair hardware error in HBM devices. This feature need to work with + hardware firmwares. + + If not sure say no. + endmenu diff --git a/drivers/soc/hisilicon/hisi_hbmdev.c b/drivers/soc/hisilicon/hisi_hbmdev.c index 5b6b1618148c..8348abc6144c 100644 --- a/drivers/soc/hisilicon/hisi_hbmdev.c +++ b/drivers/soc/hisilicon/hisi_hbmdev.c @@ -11,6 +11,7 @@ #include <linux/node.h> #include <linux/arch_topology.h> #include <linux/memory_hotplug.h> +#include <linux/mm.h> #include "hisi_internal.h" @@ -25,6 +26,9 @@ struct cdev_node { struct memory_dev { struct kobject *memdev_kobj; struct kobject *topo_kobj; +#ifdef CONFIG_HISI_HBMDEV_ACLS + struct kobject *acls_kobj; +#endif struct cdev_node cdev_list; nodemask_t cluster_cpumask[MAX_NUMNODES]; }; @@ -85,6 +89,132 @@ static void memory_topo_init(void) kobject_put(mdev->topo_kobj); } +#ifdef CONFIG_HISI_HBMDEV_ACLS +static struct acpi_device *paddr_to_acpi_device(u64 paddr) +{ + unsigned long pfn; + int nid; + + pfn = __phys_to_pfn(paddr); + if (!pfn_valid(pfn)) + return NULL; + + nid = pfn_to_nid(pfn); + if (nid < 0 && nid >= MAX_NUMNODES) + return NULL; + + return hotplug_mdev[nid]; +} + +static ssize_t acls_query_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t count) +{ + struct acpi_object_list arg_list; + struct acpi_device *adev; + union acpi_object obj; + acpi_status status; + u64 paddr, res; + + if (kstrtoull(buf, 16, &paddr)) + return -EINVAL; + + adev = paddr_to_acpi_device(paddr); + if (!adev) + return -EINVAL; + + obj.type = ACPI_TYPE_INTEGER; + obj.integer.value = paddr; + arg_list.count = 1; + arg_list.pointer = &obj; + + status = acpi_evaluate_integer(adev->handle, "AQRY", &arg_list, &res); + if (ACPI_FAILURE(status)) + return -ENODEV; + + if (IS_ERR_VALUE(-res)) + return -res; + else if (res) + return -ENODEV; + + return count; +} + +static struct kobj_attribute acls_query_store_attribute = + __ATTR(acls_query, 0200, NULL, acls_query_store); + +static ssize_t acls_repair_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t count) +{ + struct acpi_object_list arg_list; + struct acpi_device *adev; + union acpi_object obj; + acpi_status status; + u64 paddr, res; + + if (kstrtoull(buf, 16, &paddr)) + return -EINVAL; + + adev = paddr_to_acpi_device(paddr); + if (!adev) + return -EINVAL; + + obj.type = ACPI_TYPE_INTEGER; + obj.integer.value = paddr; + arg_list.count = 1; + arg_list.pointer = &obj; + + status = acpi_evaluate_integer(adev->handle, "AREP", &arg_list, &res); + if (ACPI_FAILURE(status)) + return -ENODEV; + + if (IS_ERR_VALUE(-res)) + return -res; + else if (res) + return -ENODEV; + + return count; +} +static struct kobj_attribute acls_repair_store_attribute = + __ATTR(acls_repair, 0200, NULL, acls_repair_store); + +static struct attribute *acls_attrs[] = { + &acls_query_store_attribute.attr, + &acls_repair_store_attribute.attr, + NULL, +}; + +static struct attribute_group acls_attr_group = { + .attrs = acls_attrs, +}; + +static void acls_init(void) +{ + int ret = -ENOMEM; + + mdev->acls_kobj = kobject_create_and_add("acls", mdev->memdev_kobj); + if (!mdev->acls_kobj) + goto out; + + ret = sysfs_create_group(mdev->acls_kobj, &acls_attr_group); + if (ret) + kobject_put(mdev->acls_kobj); + +out: + if (ret) + pr_err("ACLS hot repair is not enabled\n"); +} + +static void acls_remove(void) +{ + kobject_put(mdev->acls_kobj); +} +#else +static void acls_init(void) {} +static void acls_remove(void) {} +#endif + static int get_pxm(struct acpi_device *acpi_device, void *arg) { acpi_handle handle = acpi_device->handle; @@ -284,6 +414,7 @@ static int __init mdev_init(void) } memory_topo_init(); + acls_init(); return ret; } module_init(mdev_init); @@ -293,6 +424,7 @@ static void __exit mdev_exit(void) container_remove(); kobject_put(mdev->memdev_kobj); kobject_put(mdev->topo_kobj); + acls_remove(); kfree(mdev); } module_exit(mdev_exit); -- 2.17.1

3 2

[PATCH openEuler-1.0-LTS 0/2] Sync LTS patches for openEuler-1.0-LTS
by Zhang Zekun 06 Nov '23

06 Nov '23

Sync 4.19 LTS patches for openEuler-1.0-LTS 1. Fix the wrong alloc flags which may cause sleep in atomic. 2. Fix the wrong bit number. Add a cover letter for these patches, which can help to generate PR. Dan Carpenter (1): regmap: rbtree: Use alloc_flags for memory allocations Richard Fitzgerald (1): regmap: rbtree: Fix wrong register marked as in-cache when creating new node drivers/base/regmap/regcache-rbtree.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) -- 2.17.1

2 3

[PATCH openEuler-22.03-LTS 0/4] fix memcgv1 oom meminfo bug
by Lu Jialin 06 Nov '23

06 Nov '23

Sergey Senozhatsky (1): seq_buf: Add seq_buf_do_printk() helper Steven Rostedt (VMware) (1): seq_buf: Add seq_buf_terminate() API Yosry Ahmed (2): memcg: use seq_buf_do_printk() with mem_cgroup_print_oom_meminfo() memcg: dump memory.stat during cgroup OOM for v1 include/linux/seq_buf.h | 27 +++++++++++++ lib/seq_buf.c | 32 +++++++++++++++ mm/memcontrol.c | 87 +++++++++++++++++++++++------------------ 3 files changed, 108 insertions(+), 38 deletions(-) -- 2.34.1

2 5

[PATCH OLK-5.10 0/5] AGDI: Add driver for Arm Generic Diagnostic Dump and Reset device
by Qian Zou 04 Nov '23

04 Nov '23

From: “z00447436” <zouqian4(a)huawei.com> Some use-cases, such as system management, require the ability to generate a non-maskable event to the OS to request the OS kernel to perform a diagnostic dump and reset the system. Arm Generic Diagnostic Dump and Reset device enables a maintainer to request OS to perform a diagnostic dump and reset a system via SDEI event or an interrupt. This patch implements SDEI path and discards the interrupted context before proceeding to the crash kernel. D Scott Phillips (1): arm64: sdei: abort running SDEI handlers during crash Ilkka Koskinen (3): ACPI: tables: Add AGDI to the list of known table signatures ACPI: AGDI: Add driver for Arm Generic Diagnostic Dump and Reset device ACPICA: iASL: Add suppport for AGDI table Shuai Xue (1): ACPI: APEI: explicit init of HEST and GHES in apci_init() arch/arm64/include/asm/sdei.h | 6 ++ arch/arm64/kernel/entry.S | 27 +++++++- arch/arm64/kernel/sdei.c | 3 + arch/arm64/kernel/smp.c | 8 +-- drivers/acpi/apei/ghes.c | 19 +++--- drivers/acpi/arm64/Kconfig | 10 +++ drivers/acpi/arm64/Makefile | 1 + drivers/acpi/arm64/agdi.c | 116 ++++++++++++++++++++++++++++++++++ drivers/acpi/bus.c | 4 ++ drivers/acpi/pci_root.c | 3 - drivers/acpi/tables.c | 2 +- drivers/firmware/Kconfig | 1 + drivers/firmware/arm_sdei.c | 32 ++++++---- include/acpi/actbl2.h | 20 ++++++ include/acpi/apei.h | 4 +- include/linux/acpi_agdi.h | 13 ++++ include/linux/arm_sdei.h | 4 ++ 17 files changed, 240 insertions(+), 33 deletions(-) create mode 100644 drivers/acpi/arm64/agdi.c create mode 100644 include/linux/acpi_agdi.h -- 2.33.0

2 6

[PATCH OLK-5.10 0/4] AGDI: Add driver for Arm Generic Diagnostic Dump and Reset device
by Qian Zou 04 Nov '23

04 Nov '23

From: “z00447436” <zouqian4(a)huawei.com> Some use-cases, such as system management, require the ability to generate a non-maskable event to the OS to request the OS kernel to perform a diagnostic dump and reset the system. Arm Generic Diagnostic Dump and Reset device enables a maintainer to request OS to perform a diagnostic dump and reset a system via SDEI event or an interrupt. This patch implements SDEI path and discards the interrupted context before proceeding to the crash kernel. D Scott Phillips (1): arm64: sdei: abort running SDEI handlers during crash Ilkka Koskinen (3): ACPI: tables: Add AGDI to the list of known table signatures ACPI: AGDI: Add driver for Arm Generic Diagnostic Dump and Reset device ACPICA: iASL: Add suppport for AGDI table arch/arm64/include/asm/sdei.h | 6 ++++++ arch/arm64/kernel/entry.S | 27 +++++++++++++++++++++++++-- arch/arm64/kernel/sdei.c | 3 +++ arch/arm64/kernel/smp.c | 8 ++++---- drivers/acpi/arm64/Kconfig | 10 ++++++++++ drivers/acpi/arm64/Makefile | 1 + drivers/acpi/bus.c | 2 ++ drivers/acpi/tables.c | 2 +- drivers/firmware/arm_sdei.c | 19 +++++++++++++++++++ include/acpi/actbl2.h | 20 ++++++++++++++++++++ include/linux/acpi_agdi.h | 13 +++++++++++++ include/linux/arm_sdei.h | 2 ++ 12 files changed, 106 insertions(+), 7 deletions(-) create mode 100644 include/linux/acpi_agdi.h -- 2.33.0

2 5

[PATCH OLK-5.10 00/42] add error handling for add disk
by Li Nan 03 Nov '23

03 Nov '23

Christoph Hellwig (14): block: fold register_disk into device_add_disk block: call blk_integrity_add earlier in device_add_disk block: add the events* attributes to disk_attrs block: fix error unwinding in device_add_disk block: clear ->slave_dir when dropping the main slave_dir reference block: add blk_alloc_disk and blk_cleanup_disk APIs blk-mq: add the blk_mq_alloc_disk APIs loop: use blk_mq_alloc_disk and blk_cleanup_disk loop: fix order of cleaning up the queue and freeing the tagset nbd: use blk_mq_alloc_disk and blk_cleanup_disk block: add a flag to make put_disk on partially initalized disks safer nvme: use blk_mq_alloc_disk brd: convert to blk_alloc_disk/blk_cleanup_disk ubi: use blk_mq_alloc_disk and blk_cleanup_disk Dan Carpenter (1): blk-mq: fix an IS_ERR() vs NULL bug Li Lingfeng (1): nbd: fix uaf in nbd_open Luis Chamberlain (10): block: return errors from blk_integrity_add block: return errors from disk_alloc_events block: add error handling for device_add_disk / add_disk block: fix device_add_disk() kobject_create_and_add() error handling loop: add error handling support for add_disk() nbd: add error handling support for add_disk() nvme: add error handling support for add_disk() block/brd: add error handling support for add_disk() scsi: sr: Add error handling support for add_disk() mtd/ubi/block: add error handling support for add_disk() Tetsuo Handa (1): block: check minor range in device_add_disk() Wang Qing (1): nbd: fix order of cleaning up the queue and freeing the tagset Wen Yang (1): Revert "Revert "block: nbd: add sanity check for first_minor"" Yu Kuai (3): nbd: fix max value for 'first_minor' nbd: fix possible overflow for 'first_minor' in nbd_dev_add() block: fix memory leak for elevator on add_disk failure Zhang Wensheng (1): nbd: fix possible overflow on 'first_minor' in nbd_dev_add() Zhong Jinghua (9): nbd: Reorganize the messy commit log about the first_minor check block: return errors from blk_register_region block: Fix the kabi change in device_add_disk block: Fix the kabi change on blk_register_region block: call blk_get_queue earlier in __device_add_disk block: Fix minor range check in device_add_disk() block: Set memalloc_noio to false in the error path mtd/ubi/block: Fix null pointer dereference issue in error path mtd/ubi/block: Fix uaf problem in ubiblock_cleanup block/blk.h | 5 +- include/linux/blk-mq.h | 12 ++ include/linux/genhd.h | 36 +++++ block/blk-integrity.c | 12 +- block/blk-mq.c | 19 +++ block/genhd.c | 286 ++++++++++++++++++++++++++------------- drivers/block/brd.c | 103 ++++++-------- drivers/block/loop.c | 26 ++-- drivers/block/nbd.c | 75 ++++------ drivers/mtd/ubi/block.c | 76 +++++------ drivers/nvme/host/core.c | 41 +++--- drivers/scsi/sr.c | 7 +- 12 files changed, 413 insertions(+), 285 deletions(-) -- 2.39.2

1 42

[PATCH OLK-5.10 01/42] nbd: Reorganize the messy commit log about the first_minor check
by Li Nan 03 Nov '23

03 Nov '23

From: Zhong Jinghua <zhongjinghua(a)huawei.com> hulk inclusion category: bugfix bugzilla: 188217 ---------------------------------------- Commits on our branch with check issues on first_minor had serious confusion, resulting in duplicate checks in nbd_dev_add and nbd_genl_connect. So I revert the messy commit, backport lts patch. Revert 881885f30261 [Backport] nbd: fix max value for 'first_minor' Revert 4d759cee3f18 [Backport] Revert "Revert "block: nbd: add sanity check for first_minor"" Revert b95487500a4f [Huawei] nbd: fix assignment error for first_minor in nbd_dev_add Revert 93c4218b2f4d [Backport] nbd: fix possible overflow on 'first_minor' in nbd_dev_add() Revert 60141517d289 nbd: Fix use-after-free in blk_mq_free_rqs Revert 98d3ad1d2589 nbd: add sanity check for first_minor Signed-off-by: Zhong Jinghua <zhongjinghua(a)huawei.com> Signed-off-by: Li Nan <linan122(a)huawei.com> --- drivers/block/nbd.c | 24 +----------------------- 1 file changed, 1 insertion(+), 23 deletions(-) diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c index fa08f380f5be..29884a0f1fca 100644 --- a/drivers/block/nbd.c +++ b/drivers/block/nbd.c @@ -1875,17 +1875,7 @@ static int nbd_dev_add(int index) refcount_set(&nbd->refs, 1); INIT_LIST_HEAD(&nbd->list); disk->major = NBD_MAJOR; - - /* Too big first_minor can cause duplicate creation of - * sysfs files/links, since index << part_shift might overflow, or - * MKDEV() expect that the max bits of first_minor is 20. - */ disk->first_minor = index << part_shift; - if (disk->first_minor < index || disk->first_minor > MINORMASK) { - err = -EINVAL; - goto out_free_idr; - } - disk->fops = &nbd_fops; disk->private_data = nbd; sprintf(disk->disk_name, "nbd%d", index); @@ -1972,20 +1962,8 @@ static int nbd_genl_connect(struct sk_buff *skb, struct genl_info *info) if (!netlink_capable(skb, CAP_SYS_ADMIN)) return -EPERM; - if (info->attrs[NBD_ATTR_INDEX]) { + if (info->attrs[NBD_ATTR_INDEX]) index = nla_get_u32(info->attrs[NBD_ATTR_INDEX]); - - /* - * Too big first_minor can cause duplicate creation of - * sysfs files/links, since index << part_shift might - * overflow, or MKDEV() expect that the max bits of - * first_minor is 20. - */ - if (index < 0 || index > MINORMASK >> part_shift) { - printk(KERN_ERR "nbd: illegal input index %d\n", index); - return -EINVAL; - } - } if (!info->attrs[NBD_ATTR_SOCKETS]) { printk(KERN_ERR "nbd: must specify at least one socket\n"); return -EINVAL; -- 2.39.2

1 3

[PATCH RESEND openEuler-1.0-LTS 0/2] Sync LTS patches for openEuler-1.0-LTS
by Zhang Zekun 03 Nov '23

03 Nov '23

Sync 4.19 LTS patches for openEuler-1.0-LTS 1. Fix the wrong alloc flags which may cause sleep in atomic. 2. Fix the wrong bit number. Add a cover letter for these patches, which can help to generate PR. Dan Carpenter (1): regmap: rbtree: Use alloc_flags for memory allocations Richard Fitzgerald (1): regmap: rbtree: Fix wrong register marked as in-cache when creating new node drivers/base/regmap/regcache-rbtree.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) -- 2.17.1

1 2

[PATCH V2 OLK-5.10] soc: hisilicon: hisi_hbmdev: Add hbm acls repair and query methods
by Zhang Zekun 03 Nov '23

03 Nov '23

Offering: HULK hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8CCP5 ---------------------------------- Hbm memory device add support for acls hot repair. The patch add two methods for userpace: - query a paddr if it support acls repair - repair a paddr in hbm memory device The feature of ACLS hot repair can help to fix a memory error from userspace by passing through the error physical address to HBM hardware. Signed-off-by: Zhang Zekun <zhangzekun11(a)huawei.com> --- v2: - Change "u64 pfn" ot "unsigned long pfn" to avoid compile error - Add two stub to pretiffy the code drivers/soc/hisilicon/Kconfig | 10 +++ drivers/soc/hisilicon/hisi_hbmdev.c | 131 ++++++++++++++++++++++++++++ 2 files changed, 141 insertions(+) diff --git a/drivers/soc/hisilicon/Kconfig b/drivers/soc/hisilicon/Kconfig index 383375f20cac..45754528873e 100644 --- a/drivers/soc/hisilicon/Kconfig +++ b/drivers/soc/hisilicon/Kconfig @@ -44,4 +44,14 @@ config HISI_HBMCACHE To compile the driver as a module, choose M here: the module will be called hisi_hbmcache. +config HISI_HBMDEV_ACLS + bool "Add support for HISI ACLS repair" + depends on HISI_HBMDEV + help + Add ACLS support for hbm device, which can be used to query and + repair hardware error in HBM devices. This feature need to work with + hardware firmwares. + + If not sure say no. + endmenu diff --git a/drivers/soc/hisilicon/hisi_hbmdev.c b/drivers/soc/hisilicon/hisi_hbmdev.c index 5b6b1618148c..caa2711b2de7 100644 --- a/drivers/soc/hisilicon/hisi_hbmdev.c +++ b/drivers/soc/hisilicon/hisi_hbmdev.c @@ -25,6 +25,9 @@ struct cdev_node { struct memory_dev { struct kobject *memdev_kobj; struct kobject *topo_kobj; +#ifdef CONFIG_HISI_HBMDEV_ACLS + struct kobject *acls_kobj; +#endif struct cdev_node cdev_list; nodemask_t cluster_cpumask[MAX_NUMNODES]; }; @@ -85,6 +88,132 @@ static void memory_topo_init(void) kobject_put(mdev->topo_kobj); } +#ifdef CONFIG_HISI_HBMDEV_ACLS +static struct acpi_device* paddr_to_acpi_device(u64 paddr) +{ + unsigned long pfn; + int nid; + + pfn = __phys_to_pfn(paddr); + if (!pfn_valid(pfn)) + return NULL; + + nid = pfn_to_nid(pfn); + if (nid < 0 && nid >= MAX_NUMNODES) + return NULL; + + return hotplug_mdev[nid]; +} + +static ssize_t acls_query_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t count) +{ + struct acpi_object_list arg_list; + struct acpi_device *adev; + union acpi_object obj; + acpi_status status; + u64 paddr, res; + + if (kstrtoull(buf, 16, &paddr)) + return -EINVAL; + + adev = paddr_to_acpi_device(paddr); + if (!adev) + return -EINVAL; + + obj.type = ACPI_TYPE_INTEGER; + obj.integer.value = paddr; + arg_list.count = 1; + arg_list.pointer = &obj; + + status = acpi_evaluate_integer(adev->handle, "AQRY", &arg_list, &res); + if (ACPI_FAILURE(status)) + return -ENODEV; + + if (IS_ERR_VALUE(-res)) + return -res; + else if (res) + return -ENODEV; + + return count; +} + +static struct kobj_attribute acls_query_store_attribute = + __ATTR(acls_query, 0200, NULL, acls_query_store); + +static ssize_t acls_repair_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t count) +{ + struct acpi_object_list arg_list; + struct acpi_device *adev; + union acpi_object obj; + acpi_status status; + u64 paddr, res; + + if (kstrtoull(buf, 16, &paddr)) + return -EINVAL; + + adev = paddr_to_acpi_device(paddr); + if (!adev) + return -EINVAL; + + obj.type = ACPI_TYPE_INTEGER; + obj.integer.value = paddr; + arg_list.count = 1; + arg_list.pointer = &obj; + + status = acpi_evaluate_integer(adev->handle, "AREP", &arg_list, &res); + if (ACPI_FAILURE(status)) + return -ENODEV; + + if (IS_ERR_VALUE(-res)) + return -res; + else if (res) + return -ENODEV; + + return count; +} +static struct kobj_attribute acls_repair_store_attribute = + __ATTR(acls_repair, 0200, NULL, acls_repair_store); + +static struct attribute *acls_attrs[] = { + &acls_query_store_attribute.attr, + &acls_repair_store_attribute.attr, + NULL, +}; + +static struct attribute_group acls_attr_group = { + .attrs = acls_attrs, +}; + +static void acls_init(void) +{ + int ret = -ENOMEM; + + mdev->acls_kobj = kobject_create_and_add("acls", mdev->memdev_kobj); + if (!mdev->acls_kobj) + goto out; + + ret = sysfs_create_group(mdev->acls_kobj, &acls_attr_group); + if (ret) + kobject_put(mdev->acls_kobj); + +out: + if (ret) + pr_err("ACLS hot repair is not enabled\n"); +} + +static void acls_remove(void) +{ + kobject_put(mdev->acls_kobj); +} +#else +static void acls_init(void) {} +static void acls_remove(void) {} +#endif + static int get_pxm(struct acpi_device *acpi_device, void *arg) { acpi_handle handle = acpi_device->handle; @@ -284,6 +413,7 @@ static int __init mdev_init(void) } memory_topo_init(); + acls_init(); return ret; } module_init(mdev_init); @@ -293,6 +423,7 @@ static void __exit mdev_exit(void) container_remove(); kobject_put(mdev->memdev_kobj); kobject_put(mdev->topo_kobj); + acls_remove(); kfree(mdev); } module_exit(mdev_exit); -- 2.17.1

2 1

[PATCH openEuler-1.0-LTS 1/2] regmap: rbtree: Use alloc_flags for memory allocations
by Zhang Zekun 03 Nov '23

03 Nov '23

From: Dan Carpenter <dan.carpenter(a)linaro.org> stable inclusion from stable-v4.19.295 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8DDE2 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… ---------------------------------------------------------- [ Upstream commit 0c8b0bf42c8cef56f7cd9cd876fbb7ece9217064 ] The kunit tests discovered a sleeping in atomic bug. The allocations in the regcache-rbtree code should use the map->alloc_flags instead of GFP_KERNEL. [ 5.005510] BUG: sleeping function called from invalid context at include/linux/sched/mm.h:306 [ 5.005960] in_atomic(): 1, irqs_disabled(): 128, non_block: 0, pid: 117, name: kunit_try_catch [ 5.006219] preempt_count: 1, expected: 0 [ 5.006414] 1 lock held by kunit_try_catch/117: [ 5.006590] #0: 833b9010 (regmap_kunit:86:(config)->lock){....}-{2:2}, at: regmap_lock_spinlock+0x14/0x1c [ 5.007493] irq event stamp: 162 [ 5.007627] hardirqs last enabled at (161): [<80786738>] crng_make_state+0x1a0/0x294 [ 5.007871] hardirqs last disabled at (162): [<80c531ec>] _raw_spin_lock_irqsave+0x7c/0x80 [ 5.008119] softirqs last enabled at (0): [<801110ac>] copy_process+0x810/0x2138 [ 5.008356] softirqs last disabled at (0): [<00000000>] 0x0 [ 5.008688] CPU: 0 PID: 117 Comm: kunit_try_catch Tainted: G N 6.4.4-rc3-g0e8d2fdfb188 #1 [ 5.009011] Hardware name: Generic DT based system [ 5.009277] unwind_backtrace from show_stack+0x18/0x1c [ 5.009497] show_stack from dump_stack_lvl+0x38/0x5c [ 5.009676] dump_stack_lvl from __might_resched+0x188/0x2d0 [ 5.009860] __might_resched from __kmem_cache_alloc_node+0x1dc/0x25c [ 5.010061] __kmem_cache_alloc_node from kmalloc_trace+0x30/0xc8 [ 5.010254] kmalloc_trace from regcache_rbtree_write+0x26c/0x468 [ 5.010446] regcache_rbtree_write from _regmap_write+0x88/0x140 [ 5.010634] _regmap_write from regmap_write+0x44/0x68 [ 5.010803] regmap_write from basic_read_write+0x8c/0x270 [ 5.010980] basic_read_write from kunit_try_run_case+0x48/0xa0 Fixes: 28644c809f44 ("regmap: Add the rbtree cache support") Reported-by: Guenter Roeck <linux(a)roeck-us.net> Closes: https://lore.kernel.org/all/ee59d128-413c-48ad-a3aa-d9d350c80042@roeck-us.n… Signed-off-by: Dan Carpenter <dan.carpenter(a)linaro.org> Tested-by: Guenter Roeck <linux(a)roeck-us.net> Link: https://lore.kernel.org/r/58f12a07-5f4b-4a8f-ab84-0a42d1908cb9@moroto.mount… Signed-off-by: Mark Brown <broonie(a)kernel.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Zhang Zekun <zhangzekun11(a)huawei.com> --- drivers/base/regmap/regcache-rbtree.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) diff --git a/drivers/base/regmap/regcache-rbtree.c b/drivers/base/regmap/regcache-rbtree.c index b1e9aae9a5d0..f04007b388ea 100644 --- a/drivers/base/regmap/regcache-rbtree.c +++ b/drivers/base/regmap/regcache-rbtree.c @@ -291,14 +291,14 @@ static int regcache_rbtree_insert_to_block(struct regmap *map, blk = krealloc(rbnode->block, blklen * map->cache_word_size, - GFP_KERNEL); + map->alloc_flags); if (!blk) return -ENOMEM; if (BITS_TO_LONGS(blklen) > BITS_TO_LONGS(rbnode->blklen)) { present = krealloc(rbnode->cache_present, BITS_TO_LONGS(blklen) * sizeof(*present), - GFP_KERNEL); + map->alloc_flags); if (!present) { kfree(blk); return -ENOMEM; @@ -335,7 +335,7 @@ regcache_rbtree_node_alloc(struct regmap *map, unsigned int reg) const struct regmap_range *range; int i; - rbnode = kzalloc(sizeof(*rbnode), GFP_KERNEL); + rbnode = kzalloc(sizeof(*rbnode), map->alloc_flags); if (!rbnode) return NULL; @@ -361,13 +361,13 @@ regcache_rbtree_node_alloc(struct regmap *map, unsigned int reg) } rbnode->block = kmalloc_array(rbnode->blklen, map->cache_word_size, - GFP_KERNEL); + map->alloc_flags); if (!rbnode->block) goto err_free; rbnode->cache_present = kcalloc(BITS_TO_LONGS(rbnode->blklen), sizeof(*rbnode->cache_present), - GFP_KERNEL); + map->alloc_flags); if (!rbnode->cache_present) goto err_free_block; -- 2.17.1

2 2

[PATCH OLK-5.10] soc: hisilicon: hisi_hbmdev: Add hbm acls repair and query methods
by Zhang Zekun 03 Nov '23

03 Nov '23

Offering: HULK hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8CCP5 ---------------------------------- Hbm memory device add support for acls hot repair. The patch add two methods for userpace: - query a paddr if it support acls repair - repair a paddr in hbm memory device The feature of ACLS hot repair can help to fix a memory error from userspace by passing through the error physical address to HBM hardware. Signed-off-by: Zhang Zekun <zhangzekun11(a)huawei.com> --- drivers/soc/hisilicon/Kconfig | 10 +++ drivers/soc/hisilicon/hisi_hbmdev.c | 134 ++++++++++++++++++++++++++++ 2 files changed, 144 insertions(+) diff --git a/drivers/soc/hisilicon/Kconfig b/drivers/soc/hisilicon/Kconfig index 383375f20cac..45754528873e 100644 --- a/drivers/soc/hisilicon/Kconfig +++ b/drivers/soc/hisilicon/Kconfig @@ -44,4 +44,14 @@ config HISI_HBMCACHE To compile the driver as a module, choose M here: the module will be called hisi_hbmcache. +config HISI_HBMDEV_ACLS + bool "Add support for HISI ACLS repair" + depends on HISI_HBMDEV + help + Add ACLS support for hbm device, which can be used to query and + repair hardware error in HBM devices. This feature need to work with + hardware firmwares. + + If not sure say no. + endmenu diff --git a/drivers/soc/hisilicon/hisi_hbmdev.c b/drivers/soc/hisilicon/hisi_hbmdev.c index 5b6b1618148c..5cde7a123033 100644 --- a/drivers/soc/hisilicon/hisi_hbmdev.c +++ b/drivers/soc/hisilicon/hisi_hbmdev.c @@ -26,6 +26,9 @@ struct memory_dev { struct kobject *memdev_kobj; struct kobject *topo_kobj; struct cdev_node cdev_list; +#ifdef CONFIG_HISI_HBMDEV_ACLS + struct kobject *acls_kobj; +#endif nodemask_t cluster_cpumask[MAX_NUMNODES]; }; @@ -85,6 +88,131 @@ static void memory_topo_init(void) kobject_put(mdev->topo_kobj); } +#ifdef CONFIG_HISI_HBMDEV_ACLS +static struct acpi_device* paddr_to_acpi_device(u64 paddr) +{ + int nid; + u64 pfn; + + pfn = __phys_to_pfn(paddr); + if (!pfn_valid(pfn)) + return NULL; + + nid = pfn_to_nid(pfn); + if (nid < 0 && nid >= MAX_NUMNODES) + return NULL; + + return hotplug_mdev[nid]; +} + +static ssize_t acls_query_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t count) +{ + struct acpi_object_list arg_list; + struct acpi_device *adev; + union acpi_object obj; + acpi_status status; + u64 paddr, res; + + if (kstrtoull(buf, 16, &paddr)) + return -EINVAL; + + adev = paddr_to_acpi_device(paddr); + if (!adev) + return -EINVAL; + + obj.type = ACPI_TYPE_INTEGER; + obj.integer.value = paddr; + arg_list.count = 1; + arg_list.pointer = &obj; + + status = acpi_evaluate_integer(adev->handle, "AQRY", &arg_list, &res); + if (ACPI_FAILURE(status)) + return -ENODEV; + + if (IS_ERR_VALUE(-res)) + return -res; + else if (res) + return -ENODEV; + + return count; +} + +static struct kobj_attribute acls_query_store_attribute = + __ATTR(acls_query, 0200, NULL, acls_query_store); + +static ssize_t acls_repair_store(struct kobject *kobj, + struct kobj_attribute *attr, + const char *buf, size_t count) +{ + struct acpi_object_list arg_list; + struct acpi_device *adev; + union acpi_object obj; + acpi_status status; + u64 paddr, res; + + if (kstrtoull(buf, 16, &paddr)) + return -EINVAL; + + adev = paddr_to_acpi_device(paddr); + if (!adev) + return -EINVAL; + + obj.type = ACPI_TYPE_INTEGER; + obj.integer.value = paddr; + arg_list.count = 1; + arg_list.pointer = &obj; + + status = acpi_evaluate_integer(adev->handle, "AREP", &arg_list, &res); + if (ACPI_FAILURE(status)) + return -ENODEV; + + if (IS_ERR_VALUE(-res)) + return -res; + else if (res) + return -ENODEV; + + return count; +} +static struct kobj_attribute acls_repair_store_attribute = + __ATTR(acls_repair, 0200, NULL, acls_repair_store); + +static struct attribute *acls_attrs[] = { + &acls_query_store_attribute.attr, + &acls_repair_store_attribute.attr, + NULL, +}; + +static struct attribute_group acls_attr_group = { + .attrs = acls_attrs, +}; + +static void acls_init(void) +{ + int ret = -ENOMEM; + + mdev->acls_kobj = kobject_create_and_add("acls", mdev->memdev_kobj); + if (!mdev->acls_kobj) + goto out; + + ret = sysfs_create_group(mdev->acls_kobj, &acls_attr_group); + if (ret) + kobject_put(mdev->acls_kobj); + +out: + if (ret) + pr_err("ACLS hot repair is not enabled\n"); + + return; +} + +static void acls_remove(void) +{ + kobject_put(mdev->acls_kobj); +} +#endif + static int get_pxm(struct acpi_device *acpi_device, void *arg) { acpi_handle handle = acpi_device->handle; @@ -284,6 +412,9 @@ static int __init mdev_init(void) } memory_topo_init(); +#ifdef CONFIG_HISI_HBMDEV_ACLS + acls_init(); +#endif return ret; } module_init(mdev_init); @@ -293,6 +424,9 @@ static void __exit mdev_exit(void) container_remove(); kobject_put(mdev->memdev_kobj); kobject_put(mdev->topo_kobj); +#ifdef CONFIG_HISI_HBMDEV_ACLS + acls_remove(); +#endif kfree(mdev); } module_exit(mdev_exit); -- 2.17.1

2 1

[PATCH OLK-5.10 v2] perf auxtrace ptt: Record whether an auxtrace mmap is needed
by Junhao He 03 Nov '23

03 Nov '23

driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I7HI80 CVE: NA ---------------------------------------------------------------------- Add a flag needs_auxtrace_mmap to record whether an auxtrace mmap is needed, in preparation for correctly determining whether or not an auxtrace mmap is needed. The change was added in the mainline [1], but it was deleted when upstream openEuler for fixes build error. openEuler OLK-5.10 now has merge evsel->needs_auxtrace_mmap. Therefore, backport this modification to OpenEuler OLK-5.10. [1] commit 057381a7ece1 ("perf auxtrace arm64: Add support for HiSilicon PCIe Tune and Trace device driver") Signed-off-by: Junhao He <hejunhao3(a)huawei.com> --- tools/perf/arch/arm64/util/hisi-ptt.c | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/perf/arch/arm64/util/hisi-ptt.c b/tools/perf/arch/arm64/util/hisi-ptt.c index 110b2edf3e6b..97696b6e1dd5 100644 --- a/tools/perf/arch/arm64/util/hisi-ptt.c +++ b/tools/perf/arch/arm64/util/hisi-ptt.c @@ -113,6 +113,7 @@ static int hisi_ptt_recording_options(struct auxtrace_record *itr, } evsel->core.attr.freq = 0; evsel->core.attr.sample_period = 1; + evsel->needs_auxtrace_mmap = true; hisi_ptt_evsel = evsel; opts->full_auxtrace = true; } -- 2.33.0

2 1

[PATCH OLK-5.10 00/13] arm64: Add sysreg header generation scripting
by Junhao He 03 Nov '23

03 Nov '23

The arm64 kernel requires some metadata for each system register it may need to access. Currently we have: * A SYS_<regname> definition which sorresponds to a sys_reg() macro. This is used both to look up a sysreg by encoding (e.g. in KVM), and also to generate code to access a sysreg where the assembler is unaware of the specific sysreg encoding. Where assemblers support the S3_<op1>_C<crn>_C<crm>_<op2> syntax for system registers, we could use this rather than manually assembling the instructions. However, we don't have consistent definitions for these and we currently still need to handle toolchains that lack this feature. * A set of <regname>_<fieldname>_SHIFT and <regname>_<fieldname>_MASK definitions, which can be used to extract fields from the register, or to construct a register from a set of fields. These do not follow the convention used by <linux/bitfield.h>, and the masks are not shifted into place, preventing their use in FIELD_PREP() and FIELD_GET(). We require the SHIFT definitions for inline assembly (and WIDTH definitions would be helpful for UBFX/SBFX), so we cannot only define a shifted MASK. Defining a SHIFT, WIDTH, shifted MASK and unshifted MASK is tedious and error-prone and life is much easier when they can be relied up to exist when writing code. * A set of <regname>_<fieldname>_<valname> definitions for each enumerated value a field may hold. These are used when identifying the presence of features. Atop of this, other code has to build up metadata at runtime (e.g. the sets of RES0/RES1 bits in a register). Alejandro Tafalla (1): arm64/sysreg: Fix typo in Enum element regex James Morse (1): arm64/sysreg: Extend the maximum width of a register and symbol name Marc Zyngier (1): arm64: Allow the definition of UNKNOWN system register fields Mark Brown (6): arm64/sysreg: Enable automatic generation of system register definitions arm64/sysreg: Introduce helpers for access to sysreg fields arm64/sysreg: Add SYS_FIELD_GET() helper arm64/sysreg: Support generation of RAZ fields arm64/sysreg: Allow leading blanks on comments in sysreg file arm64/sysreg: Allow enumerations to be declared as signed or unsigned Mark Rutland (4): arm64: Add sysreg header generation scripting arm64/sysreg: improve comment for regs without fields arm64/sysreg: fix odd line spacing arm64/sysreg: allow *Enum blocks in SysregFields blocks arch/arm64/Makefile | 3 + arch/arm64/include/asm/Kbuild | 2 + arch/arm64/include/asm/sysreg.h | 17 ++ arch/arm64/tools/Makefile | 18 ++ arch/arm64/tools/gen-sysreg.awk | 336 ++++++++++++++++++++++++++++++++ arch/arm64/tools/sysreg | 50 +++++ 6 files changed, 426 insertions(+) create mode 100644 arch/arm64/tools/Makefile create mode 100755 arch/arm64/tools/gen-sysreg.awk create mode 100644 arch/arm64/tools/sysreg -- 2.33.0

2 14

[PATCH OLK-5.10] PCI/IOV: Add pci_sriov_numvfs_lock to support enable pci sriov concurrently
by Jialin Zhang 02 Nov '23

02 Nov '23

hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8BLQX CVE: NA -------------------------------- "echo sriov_totalvfs > /sys/bus/pci/devices/$PF_BDF/sriov_numvfs" concurrently may cause the following warnings: Warning 1: sysfs: cannot create duplicate filename '/devices/pci0000:95/0000:95:00.0/pci_bus/0000:97' Call trace: dump_backtrace+0x0/0x200 show_stack+0x20/0x30 dump_stack+0xf0/0x138 sysfs_warn_dup+0x6c/0x90 sysfs_create_dir_ns+0xf8/0x11c create_dir+0x30/0x18c kobject_add_internal+0x5c/0x190 kobject_add+0x98/0x110 device_add+0x100/0x4a0 device_register+0x28/0x40 pci_alloc_child_bus+0x184/0x24c pci_add_new_bus+0x20/0xa4 pci_iov_add_virtfn+0x2a8/0x31c sriov_enable+0x20c/0x4d0 pci_enable_sriov+0x38/0x50 virtio_pci_sriov_configure+0x3c/0xf4 [virtio_pci] sriov_numvfs_store+0xb0/0x1b0 dev_attr_store+0x20/0x34 sysfs_kf_write+0x4c/0x5c kernfs_fop_write_iter+0x130/0x1c0 new_sync_write+0xf0/0x194 vfs_write+0x224/0x2c0 ksys_write+0x74/0x104 __arm64_sys_write+0x24/0x30 el0_svc_common.constprop.0+0x7c/0x1bc do_el0_svc+0x2c/0xa4 el0_svc+0x20/0x30 el0_sync_handler+0xb0/0xb4 el0_sync+0x160/0x180 The reason is that different VFs may create the same pci bus number and try to add new bus concurrently in virtfn_add_bus. Warning 2: proc_dir_entry 'pci/0000:97' already registered Call trace: proc_register+0x100/0x190 proc_mkdir+0x6c/0xa0 pci_proc_attach_device+0xfc/0x120 pci_bus_add_device+0x40/0xd0 pci_iov_add_virtfn+0x2c8/0x31c sriov_enable+0x20c/0x4d0 pci_enable_sriov+0x38/0x50 virtio_pci_sriov_configure+0x3c/0xf4 [virtio_pci] sriov_numvfs_store+0xb0/0x1b0 dev_attr_store+0x20/0x34 sysfs_kf_write+0x4c/0x5c kernfs_fop_write_iter+0x130/0x1c0 new_sync_write+0xf0/0x194 vfs_write+0x224/0x2c0 ksys_write+0x74/0x104 __arm64_sys_write+0x24/0x30 el0_svc_common.constprop.0+0x7c/0x1bc do_el0_svc+0x2c/0xa4 el0_svc+0x20/0x30 el0_sync_handler+0xb0/0xb4 el0_sync+0x160/0x180 The reason is that different VFs may create '/proc/bus/pci/bus_number' directory using the same bus number in pci_proc_attach_device concurrently. Mutex lock can avoid potential conflict. With the patch below the warnings above are no longer appear. Signed-off-by: Jialin Zhang <zhangjialin11(a)huawei.com> --- drivers/pci/iov.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/pci/iov.c b/drivers/pci/iov.c index 4afd4ee4f7f0..958f349345ca 100644 --- a/drivers/pci/iov.c +++ b/drivers/pci/iov.c @@ -16,6 +16,8 @@ #define VIRTFN_ID_LEN 16 +static DEFINE_MUTEX(pci_sriov_numvfs_lock); + int pci_iov_virtfn_bus(struct pci_dev *dev, int vf_id) { if (!dev->is_physfn) @@ -299,6 +301,7 @@ static ssize_t sriov_numvfs_store(struct device *dev, if (num_vfs > pci_sriov_get_totalvfs(pdev)) return -ERANGE; + mutex_lock(&pci_sriov_numvfs_lock); device_lock(&pdev->dev); if (num_vfs == pdev->sriov->num_VFs) @@ -335,6 +338,7 @@ static ssize_t sriov_numvfs_store(struct device *dev, exit: device_unlock(&pdev->dev); + mutex_unlock(&pci_sriov_numvfs_lock); if (ret < 0) return ret; -- 2.25.1

2 1

[PATCH openEuler-1.0-LTS] PCI/IOV: Add pci_sriov_numvfs_lock to support enable pci sriov concurrently
by Jialin Zhang 02 Nov '23

02 Nov '23

hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8BLQX CVE: NA -------------------------------- "echo sriov_totalvfs > /sys/bus/pci/devices/$PF_BDF/sriov_numvfs" concurrently may cause the following warnings: Warning 1: sysfs: cannot create duplicate filename '/devices/pci0000:95/0000:95:00.0/pci_bus/0000:97' Call trace: dump_backtrace+0x0/0x200 show_stack+0x20/0x30 dump_stack+0xf0/0x138 sysfs_warn_dup+0x6c/0x90 sysfs_create_dir_ns+0xf8/0x11c create_dir+0x30/0x18c kobject_add_internal+0x5c/0x190 kobject_add+0x98/0x110 device_add+0x100/0x4a0 device_register+0x28/0x40 pci_alloc_child_bus+0x184/0x24c pci_add_new_bus+0x20/0xa4 pci_iov_add_virtfn+0x2a8/0x31c sriov_enable+0x20c/0x4d0 pci_enable_sriov+0x38/0x50 virtio_pci_sriov_configure+0x3c/0xf4 [virtio_pci] sriov_numvfs_store+0xb0/0x1b0 dev_attr_store+0x20/0x34 sysfs_kf_write+0x4c/0x5c kernfs_fop_write_iter+0x130/0x1c0 new_sync_write+0xf0/0x194 vfs_write+0x224/0x2c0 ksys_write+0x74/0x104 __arm64_sys_write+0x24/0x30 el0_svc_common.constprop.0+0x7c/0x1bc do_el0_svc+0x2c/0xa4 el0_svc+0x20/0x30 el0_sync_handler+0xb0/0xb4 el0_sync+0x160/0x180 The reason is that different VFs may create the same pci bus number and try to add new bus concurrently in virtfn_add_bus. Warning 2: proc_dir_entry 'pci/0000:97' already registered Call trace: proc_register+0x100/0x190 proc_mkdir+0x6c/0xa0 pci_proc_attach_device+0xfc/0x120 pci_bus_add_device+0x40/0xd0 pci_iov_add_virtfn+0x2c8/0x31c sriov_enable+0x20c/0x4d0 pci_enable_sriov+0x38/0x50 virtio_pci_sriov_configure+0x3c/0xf4 [virtio_pci] sriov_numvfs_store+0xb0/0x1b0 dev_attr_store+0x20/0x34 sysfs_kf_write+0x4c/0x5c kernfs_fop_write_iter+0x130/0x1c0 new_sync_write+0xf0/0x194 vfs_write+0x224/0x2c0 ksys_write+0x74/0x104 __arm64_sys_write+0x24/0x30 el0_svc_common.constprop.0+0x7c/0x1bc do_el0_svc+0x2c/0xa4 el0_svc+0x20/0x30 el0_sync_handler+0xb0/0xb4 el0_sync+0x160/0x180 The reason is that different VFs may create '/proc/bus/pci/bus_number' directory using the same bus number in pci_proc_attach_device concurrently. Mutex lock can avoid potential conflict. With the patch below the warnings above are no longer appear. Signed-off-by: Jialin Zhang <zhangjialin11(a)huawei.com> --- drivers/pci/pci-sysfs.c | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/drivers/pci/pci-sysfs.c b/drivers/pci/pci-sysfs.c index 48c56cb08652..e35c2f0ee28c 100644 --- a/drivers/pci/pci-sysfs.c +++ b/drivers/pci/pci-sysfs.c @@ -31,6 +31,8 @@ static int sysfs_initialized; /* = 0 */ +static DEFINE_MUTEX(pci_sriov_numvfs_lock); + /* show configuration fields */ #define pci_config_attr(field, format_string) \ static ssize_t \ @@ -604,6 +606,7 @@ static ssize_t sriov_numvfs_store(struct device *dev, if (num_vfs > pci_sriov_get_totalvfs(pdev)) return -ERANGE; + mutex_lock(&pci_sriov_numvfs_lock); device_lock(&pdev->dev); if (num_vfs == pdev->sriov->num_VFs) @@ -640,6 +643,7 @@ static ssize_t sriov_numvfs_store(struct device *dev, exit: device_unlock(&pdev->dev); + mutex_unlock(&pci_sriov_numvfs_lock); if (ret < 0) return ret; -- 2.25.1

2 1

[PATCH OLK-5.10] coresight: core: fix memory leak in dict->fwnode_list
by Junhao He 02 Nov '23

02 Nov '23

driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8DBW8 CVE: NA ---------------------------------------------------------------------- There are memory leaks reported by kmemleak: ... unreferenced object 0xffff2020103c3200 (size 256): comm "insmod", pid 4476, jiffies 4294978252 (age 50072.536s) hex dump (first 32 bytes): 10 60 40 06 28 20 ff ff 10 c0 59 06 20 20 ff ff .`@.( ....Y. .. 10 e0 47 06 28 20 ff ff 10 00 49 06 28 20 ff ff ..G.( ....I.( .. backtrace: [<0000000034ec4724>] __kmem_cache_alloc_node+0x2f8/0x348 [<0000000057fbc15d>] __kmalloc_node_track_caller+0x5c/0x110 [<00000055d5e34b>] krealloc+0x8c/0x178 [<00000000a4635beb>] coresight_alloc_device_name+0x128/0x188 [coresight] [<00000000a92ddfee>] funnel_cs_ops+0x10/0xfffffffffffedaa0 [coresight_funnel] [<00000000449e20f8>] dynamic_funnel_ids+0x80/0xfffffffffffed840 [coresight_funnel] ... when remove driver, the golab variables defined by the macro DEFINE_CORESIGHT_DEVLIST will be released, dict->nr_idx and dict->fwnode_list are cleared to 0. The lifetime of the golab variable has ended. So the buffer pointer is lost. Use the callback of devm_add_action_or_reset() to free memory. Fixes: 0f5f9b6ba9e1 ("coresight: Use platform agnostic names") Signed-off-by: Junhao He <hejunhao3(a)huawei.com> --- drivers/hwtracing/coresight/coresight-core.c | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-) diff --git a/drivers/hwtracing/coresight/coresight-core.c b/drivers/hwtracing/coresight/coresight-core.c index 9696f402a328..55ec00876989 100644 --- a/drivers/hwtracing/coresight/coresight-core.c +++ b/drivers/hwtracing/coresight/coresight-core.c @@ -1739,6 +1739,20 @@ bool coresight_loses_context_with_cpu(struct device *dev) } EXPORT_SYMBOL_GPL(coresight_loses_context_with_cpu); +void coresight_release_dev_list(void *data) +{ + struct coresight_dev_list *dict = data; + + mutex_lock(&coresight_mutex); + + if (dict->nr_idx) { + kfree(dict->fwnode_list); + dict->nr_idx = 0; + } + + mutex_unlock(&coresight_mutex); +} + /* * coresight_alloc_device_name - Get an index for a given device in the * device index list specific to a driver. An index is allocated for a @@ -1749,12 +1763,16 @@ EXPORT_SYMBOL_GPL(coresight_loses_context_with_cpu); char *coresight_alloc_device_name(struct coresight_dev_list *dict, struct device *dev) { - int idx; + int idx, ret; char *name = NULL; struct fwnode_handle **list; mutex_lock(&coresight_mutex); + ret = devm_add_action_or_reset(dev, coresight_release_dev_list, dict); + if (ret) + goto done; + idx = coresight_search_device_idx(dict, dev_fwnode(dev)); if (idx < 0) { /* Make space for the new entry */ -- 2.33.0

2 1

[PATCH OLK-5.10 v2] perf auxtrace ptt: Record whether an auxtrace mmap is needed
by Junhao He 02 Nov '23

02 Nov '23

driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I7HI80 CVE: NA ---------------------------------------------------------------------- Add a flag needs_auxtrace_mmap to record whether an auxtrace mmap is needed, in preparation for correctly determining whether or not an auxtrace mmap is needed. The change was added in the mainline patch [1], but it was deleted when upstream openEuler for fixes build error. openEuler OLK-5.10 now has merge evsel->needs_auxtrace_mmap. Therefore, backport this modification to OpenEuler OLK-5.10. [1] 057381a7ece1 "perf auxtrace arm64: Add support for HiSilicon PCIe Tune and Trace device driver" Signed-off-by: Junhao He <hejunhao3(a)huawei.com> --- tools/perf/arch/arm64/util/hisi-ptt.c | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/perf/arch/arm64/util/hisi-ptt.c b/tools/perf/arch/arm64/util/hisi-ptt.c index 110b2edf3e6b..97696b6e1dd5 100644 --- a/tools/perf/arch/arm64/util/hisi-ptt.c +++ b/tools/perf/arch/arm64/util/hisi-ptt.c @@ -113,6 +113,7 @@ static int hisi_ptt_recording_options(struct auxtrace_record *itr, } evsel->core.attr.freq = 0; evsel->core.attr.sample_period = 1; + evsel->needs_auxtrace_mmap = true; hisi_ptt_evsel = evsel; opts->full_auxtrace = true; } -- 2.33.0

2 1

[PATCH] perf auxtrace ptt: Record whether an auxtrace mmap is needed
by Junhao He 02 Nov '23

02 Nov '23

driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I7HI80 CVE: NA ---------------------------------------------------------------------- Add a flag needs_auxtrace_mmap to record whether an auxtrace mmap is needed, in preparation for correctly determining whether or not an auxtrace mmap is needed. The change was added in the mainline patch [1], but it was deleted when upstream openEuler for fixes build error. openEuler OLK-5.10 now has merge evsel->needs_auxtrace_mmap. Therefore, backport this modification to OpenEuler OLK-5.10. [1] 057381a7ece1 "perf auxtrace arm64: Add support for HiSilicon PCIe Tune and Trace device driver" Signed-off-by: Junhao He <hejunhao3(a)huawei.com> --- tools/perf/arch/arm64/util/hisi-ptt.c | 1 + 1 file changed, 1 insertion(+) diff --git a/tools/perf/arch/arm64/util/hisi-ptt.c b/tools/perf/arch/arm64/util/hisi-ptt.c index 110b2edf3e6b..97696b6e1dd5 100644 --- a/tools/perf/arch/arm64/util/hisi-ptt.c +++ b/tools/perf/arch/arm64/util/hisi-ptt.c @@ -113,6 +113,7 @@ static int hisi_ptt_recording_options(struct auxtrace_record *itr, } evsel->core.attr.freq = 0; evsel->core.attr.sample_period = 1; + evsel->needs_auxtrace_mmap = true; hisi_ptt_evsel = evsel; opts->full_auxtrace = true; } -- 2.33.0

1 0

[PATCH v3 openEuler-1.0-LTS] kernel/trace: Fix do not unregister tracepoints when register sched_migrate_task fail
by Wenyu Huang 02 Nov '23

02 Nov '23

From: Kaitao Cheng <pilgrimtao(a)gmail.com> stable inclusion from stable-v4.19.297 commit 50f9ad607ea891a9308e67b81f774c71736d1098 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8BAW8 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- In the function, if register_trace_sched_migrate_task() returns error, sched_switch/sched_wakeup_new/sched_wakeup won't unregister. That is why fail_deprobe_sched_switch was added. Link: http://lkml.kernel.org/r/20191231133530.2794-1-pilgrimtao@gmail.com Cc: stable(a)vger.kernel.org Fixes: 478142c39c8c2 ("tracing: do not grab lock in wakeup latency function tracing") Signed-off-by: Kaitao Cheng <pilgrimtao(a)gmail.com> Signed-off-by: Steven Rostedt (VMware) <rostedt(a)goodmis.org> Signed-off-by: Wenyu Huang <huangwenyu4(a)huawei.com> --- kernel/trace/trace_sched_wakeup.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index 0db1b024e786..8041bd5e4262 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -642,7 +642,7 @@ static void start_wakeup_tracer(struct trace_array *tr) if (ret) { pr_info("wakeup trace: Couldn't activate tracepoint" " probe to kernel_sched_migrate_task\n"); - return; + goto fail_deprobe_sched_switch; } wakeup_reset(tr); @@ -660,6 +660,8 @@ static void start_wakeup_tracer(struct trace_array *tr) printk(KERN_ERR "failed to start wakeup tracer\n"); return; +fail_deprobe_sched_switch: + unregister_trace_sched_switch(probe_wakeup_sched_switch, NULL); fail_deprobe_wake_new: unregister_trace_sched_wakeup_new(probe_wakeup, NULL); fail_deprobe: -- 2.34.1

2 1

[PATCH openEuler-1.0-LTS] net: sched: sch_qfq: Use non-work-conserving warning handler
by Liu Jian 02 Nov '23

02 Nov '23

mainline inclusion from mainline-v6.7-rc1 commit 6d25d1dc76bf5943a5c1f4bb74d66d5eac58eb77 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8D8S8 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… --------------------------- A helper function for printing non-work-conserving alarms is added in commit b00355db3f88 ("pkt_sched: sch_hfsc: sch_htb: Add non-work-conserving warning handler."). In this commit, use qdisc_warn_nonwc() instead of WARN_ONCE() to handle the non-work-conserving warning in qfq Qdisc. Signed-off-by: Liu Jian <liujian56(a)huawei.com> Link: https://lore.kernel.org/r/20231023064729.370649-1-liujian56@huawei.com Signed-off-by: Paolo Abeni <pabeni(a)redhat.com> Signed-off-by: Liu Jian <liujian56(a)huawei.com> --- net/sched/sch_qfq.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c index 5b7b149a2b9f..14d009151a5b 100644 --- a/net/sched/sch_qfq.c +++ b/net/sched/sch_qfq.c @@ -1018,7 +1018,7 @@ static inline struct sk_buff *qfq_peek_skb(struct qfq_aggregate *agg, *cl = list_first_entry(&agg->active, struct qfq_class, alist); skb = (*cl)->qdisc->ops->peek((*cl)->qdisc); if (skb == NULL) - WARN_ONCE(1, "qfq_dequeue: non-workconserving leaf\n"); + qdisc_warn_nonwc("qfq_dequeue", (*cl)->qdisc); else *len = qdisc_pkt_len(skb); -- 2.34.1

2 1

[PATCH OLK-5.10] net: sched: sch_qfq: Use non-work-conserving warning handler
by Liu Jian 02 Nov '23

02 Nov '23

mainline inclusion from mainline-v6.7-rc1 commit 6d25d1dc76bf5943a5c1f4bb74d66d5eac58eb77 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8D8S8 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… --------------------------- A helper function for printing non-work-conserving alarms is added in commit b00355db3f88 ("pkt_sched: sch_hfsc: sch_htb: Add non-work-conserving warning handler."). In this commit, use qdisc_warn_nonwc() instead of WARN_ONCE() to handle the non-work-conserving warning in qfq Qdisc. Signed-off-by: Liu Jian <liujian56(a)huawei.com> Link: https://lore.kernel.org/r/20231023064729.370649-1-liujian56@huawei.com Signed-off-by: Paolo Abeni <pabeni(a)redhat.com> Signed-off-by: Liu Jian <liujian56(a)huawei.com> --- net/sched/sch_qfq.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/sched/sch_qfq.c b/net/sched/sch_qfq.c index 9447f486141d..e48c3baa059e 100644 --- a/net/sched/sch_qfq.c +++ b/net/sched/sch_qfq.c @@ -1005,7 +1005,7 @@ static inline struct sk_buff *qfq_peek_skb(struct qfq_aggregate *agg, *cl = list_first_entry(&agg->active, struct qfq_class, alist); skb = (*cl)->qdisc->ops->peek((*cl)->qdisc); if (skb == NULL) - WARN_ONCE(1, "qfq_dequeue: non-workconserving leaf\n"); + qdisc_warn_nonwc("qfq_dequeue", (*cl)->qdisc); else *len = qdisc_pkt_len(skb); -- 2.34.1

2 1

[PATCH OLK-5.10] ext4: recheck buffer valid after page unlock
by yangerkun 02 Nov '23

02 Nov '23

hulk inclusion category: bugfix bugzilla: 189306, https://gitee.com/openeuler/kernel/issues/I8BBWH CVE: NA ---------------------------------------- We need destination address when we do dio read, and this addr can come from mmap results for a journal data mode inode. Then dio_bio_complete will dirty the page which the mmap addr point to(since we have fill dirty data for this page). ext4_journalled_set_page_dirty will first set PageChecked and then dirty page(do not dirty buffer), which leave __ext4_journalled_writepage in ext4_writepage to do the rest thing needed for journal data mode. We need first start handle and then lock the page, so in __ext4_journalled_writepage we first unlock the page and latter call ext4_journal_start; after we relock the page, we do some check to prevent the concurrence truncate and then walk through buffer to help join journal. Actually, once we unlock the page, since we has not add extra buffer refcount, so the buffer can also gone(concurrent happened for jbd2_journal_commit_transaction and jbd2_log_do_checkpoint can remove the extra buffer head ref and clear buffer dirty, so drop cache can release buffer), and upper walk through buffer will trigger the BUG_ON in page_buffers. The problem does not exist in mainline since 3f079114bf52 ("ext4: Convert data=journal writeback to use ext4_writepages()") delete all this code, and this patchset seems too complex to do the backport. So we just fix it with a simpler way, check buffer valid before walk through buffer. Signed-off-by: yangerkun <yangerkun(a)huawei.com> --- fs/ext4/inode.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 9379a062dba4..0b5521cec637 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1911,6 +1911,13 @@ static int __ext4_journalled_writepage(struct page *page, goto out; } + if (!page_has_buffers(page)) { + /* Check buffer valid since we ever unlock this page */ + ext4_journal_stop(handle); + ClearPageDirty(page); + goto out; + } + if (inline_data) { ret = ext4_mark_inode_dirty(handle, inode); } else { -- 2.39.2

2 1

[PATCH openEuler-1.0-LTS v3 0/3] bugfixes for CVE-2022-45884
by liwei 02 Nov '23

02 Nov '23

bugfixes for CVE-2022-45884 Dinghao Liu (1): media: dvbdev: Fix memleak in dvb_register_device Fuqian Huang (1): media: media/dvb: Use kmemdup rather than duplicating its implementation Hyunwoo Kim (1): media: dvb-core: Fix use-after-free due to race at dvb_register_device() drivers/media/dvb-core/dvbdev.c | 86 ++++++++++++++++----- drivers/media/dvb-frontends/drx39xyj/drxj.c | 5 +- include/media/dvbdev.h | 15 ++++ 3 files changed, 82 insertions(+), 24 deletions(-) -- 2.25.1

2 4

[PATCH openEuler-1.0-LTS v4 0/7] Fix netfilter conntrack
by Lu Wei 02 Nov '23

02 Nov '23

Florian Westphal (7): netfilter: conntrack: tell compiler to not inline nf_ct_resolve_clash netfilter: conntrack: remove two args from resolve_clash netfilter: conntrack: place confirm-bit setting in a helper netfilter: conntrack: split resolve_clash function netfilter: conntrack: allow insertion of clashing entries netfilter: conntrack: do not auto-delete clash entries on reply netfilter: conntrack: fix infinite loop on rmmod include/linux/rculist_nulls.h | 7 + .../linux/netfilter/nf_conntrack_common.h | 12 +- net/netfilter/nf_conntrack_core.c | 215 +++++++++++++++--- net/netfilter/nf_conntrack_proto_udp.c | 8 +- net/netfilter/nft_flow_offload.c | 2 +- 5 files changed, 206 insertions(+), 38 deletions(-) -- 2.34.1

2 8

[PATCH openEuler-1.0-LTS v2 0/4] ACPI for MPAM 2.0
by Yu Liao 01 Nov '23

01 Nov '23

v1 -> v2: modified commit log Erik Kaneda (1): ACPICA: ACPI 6.4: PPTT: add new version of subtable type 1 Hesham Almatary (1): ACPICA: Add support for Arm's MPAM ACPI table version 2 Yu Liao (2): ACPI / PPTT: Find PPTT processor node by cache id ACPI/MPAM: Adapt to Arm's MPAM ACPI table version 2 arch/arm64/kernel/mpam/mpam_device.c | 2 +- drivers/acpi/arm64/Makefile | 2 +- drivers/acpi/arm64/mpam.c | 21 +++- drivers/acpi/arm64/mpam_v2.c | 175 +++++++++++++++++++++++++++ drivers/acpi/pptt.c | 55 +++++++++ include/acpi/actbl2.h | 118 ++++++++++++++++++ include/linux/acpi.h | 5 + include/linux/arm_mpam.h | 2 +- 8 files changed, 373 insertions(+), 7 deletions(-) create mode 100644 drivers/acpi/arm64/mpam_v2.c -- 2.25.1

2 5

[PATCH openEuler-1.0-LTS v2 0/3] bugfixes for CVE-2022-45884
by liwei 01 Nov '23

01 Nov '23

bugfixes for CVE-2022-45884 Dinghao Liu (1): media: dvbdev: Fix memleak in dvb_register_device Fuqian Huang (1): media: media/dvb: Use kmemdup rather than duplicating its implementation Hyunwoo Kim (1): media: dvb-core: Fix use-after-free due to race at dvb_register_device() drivers/media/dvb-core/dvbdev.c | 86 ++++++++++++++++----- drivers/media/dvb-frontends/drx39xyj/drxj.c | 5 +- include/media/dvbdev.h | 15 ++++ 3 files changed, 82 insertions(+), 24 deletions(-) -- 2.25.1

2 4

[PATCH OLK-5.10 v2 0/7] add sample sockmap code for redis
by Liu Jian 31 Oct '23

31 Oct '23

v1->v2: Move local_skb to a reserved field instead of inserting it to a structure hole. Fix a cgroup ref leak bug. Hengqi Chen (1): libbpf: Support uniform BTF-defined key/value specification across all BPF maps Liu Jian (4): cgroup: make cgroup_bpf_prog_attach work when cgroup2 is not mounted net: let sockops can use bpf_get_current_comm() net: add local_skb parameter to identify local tcp connection tools: add sample sockmap code for redis Xu Kuohai (1): bpf, sockmap: Fix map type error in sock_map_del_link Yosry Ahmed (1): cgroup: add cgroup_v1v2_get_from_[fd/file]() include/linux/cgroup.h | 2 + include/linux/filter.h | 1 + include/linux/skbuff.h | 2 +- include/uapi/linux/bpf.h | 1 + kernel/bpf/cgroup.c | 8 +- kernel/cgroup/cgroup.c | 72 +++++++- net/core/filter.c | 9 + net/core/sock_map.c | 10 +- net/ipv4/tcp_input.c | 4 +- net/ipv4/tcp_output.c | 4 + tools/include/uapi/linux/bpf.h | 1 + tools/lib/bpf/libbpf.c | 24 +++ tools/netacc/Makefile | 24 +++ tools/netacc/bpf_sockmap.h | 167 +++++++++++++++++++ tools/netacc/net-acc | 35 ++++ tools/netacc/netacc.c | 296 +++++++++++++++++++++++++++++++++ tools/netacc/netaccsockmap.c | 176 ++++++++++++++++++++ 17 files changed, 820 insertions(+), 16 deletions(-) create mode 100644 tools/netacc/Makefile create mode 100644 tools/netacc/bpf_sockmap.h create mode 100755 tools/netacc/net-acc create mode 100644 tools/netacc/netacc.c create mode 100644 tools/netacc/netaccsockmap.c -- 2.34.1

2 8

[PATCH openEuler-1.0-LTS v3 0/7] Fix netfilter conntrack
by Lu Wei 31 Oct '23

31 Oct '23

Florian Westphal (7): netfilter: conntrack: tell compiler to not inline nf_ct_resolve_clash netfilter: conntrack: remove two args from resolve_clash netfilter: conntrack: place confirm-bit setting in a helper netfilter: conntrack: split resolve_clash function netfilter: conntrack: allow insertion of clashing entries netfilter: conntrack: do not auto-delete clash entries on reply netfilter: conntrack: fix infinite loop on rmmod include/linux/rculist_nulls.h | 7 + .../linux/netfilter/nf_conntrack_common.h | 12 +- net/netfilter/nf_conntrack_core.c | 215 +++++++++++++++--- net/netfilter/nf_conntrack_proto_udp.c | 7 +- net/netfilter/nft_flow_offload.c | 2 +- 5 files changed, 205 insertions(+), 38 deletions(-) -- V3: modify commit message 2.34.1

2 8

[PATCH OLK-5.10 0/3] LoongArch: add 32/64 pc relative relocation type support
by Hongchen Zhang 31 Oct '23

31 Oct '23

Hongchen Zhang (3): LoongArch: Define relocation types for ABI v2.10 LoongArch: Add support for 32_PCREL relocation type LoongArch: Add support for 64_PCREL relocation type arch/loongarch/include/asm/elf.h | 9 +++++++++ arch/loongarch/kernel/module.c | 22 +++++++++++++++++++++- 2 files changed, 30 insertions(+), 1 deletion(-) -- 2.33.0

2 4

[PATCH OLK-5.10 1/3] LoongArch: Define relocation types for ABI v2.10
by Hongchen Zhang 31 Oct '23

31 Oct '23

LoongArch inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8C3BV ------------------------------------------ The relocation types from 101 to 109 are used by GNU binutils >= 2.41, add their definitions to use them in later patches. Link: https://sourceware.org/git/?p=binutils-gdb.git;a=blob;f=include/elf/loongar… Cc: <stable(a)vger.kernel.org> Signed-off-by: Tiezhu Yang <yangtiezhu(a)loongson.cn> Signed-off-by: Huacai Chen <chenhuacai(a)loongson.cn> Signed-off-by: Hongchen Zhang <zhanghongchen(a)loongson.cn> --- arch/loongarch/include/asm/elf.h | 9 +++++++++ arch/loongarch/kernel/module.c | 2 +- 2 files changed, 10 insertions(+), 1 deletion(-) diff --git a/arch/loongarch/include/asm/elf.h b/arch/loongarch/include/asm/elf.h index 7af0cebf28d7..b9a4ab54285c 100644 --- a/arch/loongarch/include/asm/elf.h +++ b/arch/loongarch/include/asm/elf.h @@ -111,6 +111,15 @@ #define R_LARCH_TLS_GD_HI20 98 #define R_LARCH_32_PCREL 99 #define R_LARCH_RELAX 100 +#define R_LARCH_DELETE 101 +#define R_LARCH_ALIGN 102 +#define R_LARCH_PCREL20_S2 103 +#define R_LARCH_CFA 104 +#define R_LARCH_ADD6 105 +#define R_LARCH_SUB6 106 +#define R_LARCH_ADD_ULEB128 107 +#define R_LARCH_SUB_ULEB128 108 +#define R_LARCH_64_PCREL 109 #ifndef ELF_ARCH diff --git a/arch/loongarch/kernel/module.c b/arch/loongarch/kernel/module.c index 8127df5dd0ad..0b8ce1e0f83b 100644 --- a/arch/loongarch/kernel/module.c +++ b/arch/loongarch/kernel/module.c @@ -387,7 +387,7 @@ typedef int (*reloc_rela_handler)(struct module *mod, u32 *location, Elf_Addr v, /* The handlers for known reloc types */ static reloc_rela_handler reloc_rela_handlers[] = { - [R_LARCH_NONE ... R_LARCH_RELAX] = apply_r_larch_error, + [R_LARCH_NONE ... R_LARCH_64_PCREL] = apply_r_larch_error, [R_LARCH_NONE] = apply_r_larch_none, [R_LARCH_32] = apply_r_larch_32, -- 2.33.0

2 2

[PATCH OLK-5.10 3/3] LoongArch: Add support for 64_PCREL relocation type
by Hongchen Zhang 31 Oct '23

31 Oct '23

LoongArch inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8C3BV ------------------------------------------ When build and update kernel with the latest upstream binutils and loongson3_defconfig, module loader fails with: kmod: zsmalloc: Unknown relocation type 109 kmod: fuse: Unknown relocation type 109 kmod: fuse: Unknown relocation type 109 kmod: radeon: Unknown relocation type 109 kmod: nf_tables: Unknown relocation type 109 kmod: nf_tables: Unknown relocation type 109 This is because the latest upstream binutils replaces a pair of ADD64 and SUB64 with 64_PCREL, so add support for 64_PCREL relocation type. Link: https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=ecb802d02eeb Cc: <stable(a)vger.kernel.org> Signed-off-by: Tiezhu Yang <yangtiezhu(a)loongson.cn> Signed-off-by: Huacai Chen <chenhuacai(a)loongson.cn> Signed-off-by: Hongchen Zhang <zhanghongchen(a)loongson.cn> --- arch/loongarch/kernel/module.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/arch/loongarch/kernel/module.c b/arch/loongarch/kernel/module.c index a67f9def7927..b4cf3383a8bc 100644 --- a/arch/loongarch/kernel/module.c +++ b/arch/loongarch/kernel/module.c @@ -381,6 +381,15 @@ static int apply_r_larch_32_pcrel(struct module *mod, u32 *location, Elf_Addr v, return 0; } +static int apply_r_larch_64_pcrel(struct module *mod, u32 *location, Elf_Addr v, + s64 *rela_stack, size_t *rela_stack_top, unsigned int type) +{ + ptrdiff_t offset = (void *)v - (void *)location; + + *(u64 *)location = offset; + return 0; +} + /* * reloc_handlers_rela() - Apply a particular relocation to a module * @mod: the module to apply the reloc to @@ -414,6 +423,7 @@ static reloc_rela_handler reloc_rela_handlers[] = { [R_LARCH_PCALA_HI20...R_LARCH_PCALA64_HI12] = apply_r_larch_pcala, [R_LARCH_GOT_PC_HI20...R_LARCH_GOT_PC_LO12] = apply_r_larch_got_pc, [R_LARCH_32_PCREL] = apply_r_larch_32_pcrel, + [R_LARCH_64_PCREL] = apply_r_larch_64_pcrel, }; int apply_relocate_add(Elf_Shdr *sechdrs, const char *strtab, -- 2.33.0

1 0

[PATCH OLK-5.10 2/3] LoongArch: Add support for 32_PCREL relocation type
by Hongchen Zhang 31 Oct '23

31 Oct '23

LoongArch inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I8C3BV ------------------------------------------ When build and update kernel with the latest upstream binutils and loongson3_defconfig, module loader fails with: kmod: zsmalloc: Unsupport relocation type 99, please add its support. kmod: fuse: Unsupport relocation type 99, please add its support. kmod: ipmi_msghandler: Unsupport relocation type 99, please add its support. kmod: ipmi_msghandler: Unsupport relocation type 99, please add its support. kmod: pstore: Unsupport relocation type 99, please add its support. kmod: drm_display_helper: Unsupport relocation type 99, please add its support. kmod: drm_display_helper: Unsupport relocation type 99, please add its support. kmod: drm_display_helper: Unsupport relocation type 99, please add its support. kmod: fuse: Unsupport relocation type 99, please add its support. kmod: fat: Unsupport relocation type 99, please add its support. This is because the latest upstream binutils replaces a pair of ADD32 and SUB32 with 32_PCREL, so add support for 32_PCREL relocation type. Link: https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=ecb802d02eeb Cc: <stable(a)vger.kernel.org> Co-developed-by: Youling Tang <tangyouling(a)loongson.cn> Signed-off-by: Youling Tang <tangyouling(a)loongson.cn> Signed-off-by: Tiezhu Yang <yangtiezhu(a)loongson.cn> Signed-off-by: Huacai Chen <chenhuacai(a)loongson.cn> Signed-off-by: Hongchen Zhang <zhanghongchen(a)loongson.cn> --- arch/loongarch/kernel/module.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/arch/loongarch/kernel/module.c b/arch/loongarch/kernel/module.c index 0b8ce1e0f83b..a67f9def7927 100644 --- a/arch/loongarch/kernel/module.c +++ b/arch/loongarch/kernel/module.c @@ -372,6 +372,15 @@ static int apply_r_larch_got_pc(struct module *mod, u32 *location, Elf_Addr v, return apply_r_larch_pcala(mod, location, got, rela_stack, rela_stack_top, type); } +static int apply_r_larch_32_pcrel(struct module *mod, u32 *location, Elf_Addr v, + s64 *rela_stack, size_t *rela_stack_top, unsigned int type) +{ + ptrdiff_t offset = (void *)v - (void *)location; + + *(u32 *)location = offset; + return 0; +} + /* * reloc_handlers_rela() - Apply a particular relocation to a module * @mod: the module to apply the reloc to @@ -404,6 +413,7 @@ static reloc_rela_handler reloc_rela_handlers[] = { [R_LARCH_B26] = apply_r_larch_b26, [R_LARCH_PCALA_HI20...R_LARCH_PCALA64_HI12] = apply_r_larch_pcala, [R_LARCH_GOT_PC_HI20...R_LARCH_GOT_PC_LO12] = apply_r_larch_got_pc, + [R_LARCH_32_PCREL] = apply_r_larch_32_pcrel, }; int apply_relocate_add(Elf_Shdr *sechdrs, const char *strtab, -- 2.33.0

1 0

[PATCH OLK-5.10] nvmet-tcp: Fix a possible UAF in queue intialization setup
by Li Lingfeng 31 Oct '23

31 Oct '23

From: Sagi Grimberg <sagi(a)grimberg.me> mainline inclusion from mainline-v6.6-rc7 commit d920abd1e7c4884f9ecd0749d1921b7ab19ddfbd category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I88DLS CVE: CVE-2023-5178 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… ------------------------------------------------- From Alon: "Due to a logical bug in the NVMe-oF/TCP subsystem in the Linux kernel, a malicious user can cause a UAF and a double free, which may lead to RCE (may also lead to an LPE in case the attacker already has local privileges)." Hence, when a queue initialization fails after the ahash requests are allocated, it is guaranteed that the queue removal async work will be called, hence leave the deallocation to the queue removal. Also, be extra careful not to continue processing the socket, so set queue rcv_state to NVMET_TCP_RECV_ERR upon a socket error. Cc: stable(a)vger.kernel.org Reported-by: Alon Zahavi <zahavi.alon(a)gmail.com> Tested-by: Alon Zahavi <zahavi.alon(a)gmail.com> Signed-off-by: Sagi Grimberg <sagi(a)grimberg.me> Reviewed-by: Christoph Hellwig <hch(a)lst.de> Reviewed-by: Chaitanya Kulkarni <kch(a)nvidia.com> Signed-off-by: Keith Busch <kbusch(a)kernel.org> Signed-off-by: Li Lingfeng <lilingfeng3(a)huawei.com> --- drivers/nvme/target/tcp.c | 7 ++----- 1 file changed, 2 insertions(+), 5 deletions(-) diff --git a/drivers/nvme/target/tcp.c b/drivers/nvme/target/tcp.c index 2ddbd4f4f628..7ce22d173fc7 100644 --- a/drivers/nvme/target/tcp.c +++ b/drivers/nvme/target/tcp.c @@ -336,6 +336,7 @@ static void nvmet_tcp_fatal_error(struct nvmet_tcp_queue *queue) static void nvmet_tcp_socket_error(struct nvmet_tcp_queue *queue, int status) { + queue->rcv_state = NVMET_TCP_RECV_ERR; if (status == -EPIPE || status == -ECONNRESET) kernel_sock_shutdown(queue->sock, SHUT_RDWR); else @@ -882,15 +883,11 @@ static int nvmet_tcp_handle_icreq(struct nvmet_tcp_queue *queue) iov.iov_len = sizeof(*icresp); ret = kernel_sendmsg(queue->sock, &msg, &iov, 1, iov.iov_len); if (ret < 0) - goto free_crypto; + return ret; /* queue removal will cleanup */ queue->state = NVMET_TCP_Q_LIVE; nvmet_prepare_receive_pdu(queue); return 0; -free_crypto: - if (queue->hdr_digest || queue->data_digest) - nvmet_tcp_free_crypto(queue); - return ret; } static void nvmet_tcp_handle_req_failure(struct nvmet_tcp_queue *queue, -- 2.31.1

2 1

[PATCH OLK-5.10] integrity: Fix possible multiple allocation in integrity_inode_get()
by felix 31 Oct '23

31 Oct '23

From: Tianjia Zhang <tianjia.zhang(a)linux.alibaba.com> mainline inclusion from mainline-v6.5-rc1 commit 9df6a4870dc371136e90330cfbbc51464ee66993 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8C2DB CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- When integrity_inode_get() is querying and inserting the cache, there is a conditional race in the concurrent environment. The race condition is the result of not properly implementing "double-checked locking". In this case, it first checks to see if the iint cache record exists before taking the lock, but doesn't check again after taking the integrity_iint_lock. Fixes: bf2276d10ce5 ("ima: allocating iint improvements") Signed-off-by: Tianjia Zhang <tianjia.zhang(a)linux.alibaba.com> Cc: Dmitry Kasatkin <dmitry.kasatkin(a)gmail.com> Cc: <stable(a)vger.kernel.org> # v3.10+ Signed-off-by: Mimi Zohar <zohar(a)linux.ibm.com> Signed-off-by: Felix Fu <fuzhen5(a)huawei.com> --- security/integrity/iint.c | 15 +++++++++------ 1 file changed, 9 insertions(+), 6 deletions(-) diff --git a/security/integrity/iint.c b/security/integrity/iint.c index 1668cfbeb671..2428037d9e14 100644 --- a/security/integrity/iint.c +++ b/security/integrity/iint.c @@ -43,12 +43,10 @@ static struct integrity_iint_cache *__integrity_iint_find(struct inode *inode) else if (inode > iint->inode) n = n->rb_right; else - break; + return iint; } - if (!n) - return NULL; - return iint; + return NULL; } /* @@ -111,10 +109,15 @@ struct integrity_iint_cache *integrity_inode_get(struct inode *inode) parent = *p; test_iint = rb_entry(parent, struct integrity_iint_cache, rb_node); - if (inode < test_iint->inode) + if (inode < test_iint->inode) { p = &(*p)->rb_left; - else + } else if (inode > test_iint->inode) { p = &(*p)->rb_right; + } else { + write_unlock(&integrity_iint_lock); + kmem_cache_free(iint_cache, iint); + return test_iint; + } } iint->inode = inode; -- 2.34.1

2 1

[PATCH OLK-5.10 0/3] LoongArch: add 32/64 pc relative relocation type support
by Hongchen Zhang 31 Oct '23

31 Oct '23

Hongchen Zhang (3): LoongArch: Define relocation types for ABI v2.10 LoongArch: Add support for 32_PCREL relocation type LoongArch: Add support for 64_PCREL relocation type arch/loongarch/include/asm/elf.h | 9 +++++++++ arch/loongarch/kernel/module.c | 22 +++++++++++++++++++++- 2 files changed, 30 insertions(+), 1 deletion(-) -- 2.33.0

2 4

[PATCH openEuler-1.0-LTS v2 0/7] Fix netfilter nf_conntrack
by Lu Wei 31 Oct '23

31 Oct '23

Florian Westphal (6): netfilter: conntrack: tell compiler to not inline nf_ct_resolve_clash netfilter: conntrack: remove two args from resolve_clash netfilter: conntrack: place confirm-bit setting in a helper netfilter: conntrack: split resolve_clash function netfilter: conntrack: allow insertion of clashing entries netfilter: conntrack: do not auto-delete clash entries on reply Zhihao Cheng (1): ubi: Refuse attaching if mtd's erasesize is 0 -- V2: add a new bugfix patch 2.34.1

2 8

[PATCH OLK-5.10 0/4] fix memcgv1 oom meminfo bug
by Lu Jialin 31 Oct '23

31 Oct '23

Sergey Senozhatsky (1): seq_buf: Add seq_buf_do_printk() helper Steven Rostedt (VMware) (1): seq_buf: Add seq_buf_terminate() API Yosry Ahmed (2): memcg: use seq_buf_do_printk() with mem_cgroup_print_oom_meminfo() memcg: dump memory.stat during cgroup OOM for v1 include/linux/seq_buf.h | 27 +++++++++++++ lib/seq_buf.c | 32 +++++++++++++++ mm/memcontrol.c | 87 +++++++++++++++++++++++------------------ 3 files changed, 108 insertions(+), 38 deletions(-) -- 2.34.1

2 5

openEuler 测试讨论
by openEuler conference 31 Oct '23

31 Oct '23

您好！ Kernel SIG 邀请您参加 2023-10-31 15:30 召开的WeLink会议会议主题：openEuler 测试讨论会议链接：https://bmeeting.huaweicloud.com:36443/#/j/980828033 会议纪要：https://etherpad.openeuler.org/p/Kernel-meetings 温馨提醒：建议接入会议后修改参会人的姓名，也可以使用您在gitee.com的ID 更多资讯尽在：https://openeuler.org/zh/ Hello! openEuler Kernel SIG invites you to attend the WeLink conference will be held at 2023-10-31 15:30, The subject of the conference is openEuler 测试讨论, You can join the meeting at https://bmeeting.huaweicloud.com:36443/#/j/980828033. Add topics at https://etherpad.openeuler.org/p/Kernel-meetings. Note: You are advised to change the participant name after joining the conference or use your ID at gitee.com. More information: https://openeuler.org/en/

1 0

[PATCH OLK-5.10 v2 00/38] Add error handling for device_add_disk
by Zhong Jinghua 31 Oct '23

31 Oct '23

Befor these paths, block is not know when drvier is add fail. Now, add error handling for drivers. Christoph Hellwig (14): block: fold register_disk into device_add_disk block: call blk_integrity_add earlier in device_add_disk block: add the events* attributes to disk_attrs block: fix error unwinding in device_add_disk block: clear ->slave_dir when dropping the main slave_dir reference block: add blk_alloc_disk and blk_cleanup_disk APIs blk-mq: add the blk_mq_alloc_disk APIs loop: use blk_mq_alloc_disk and blk_cleanup_disk loop: fix order of cleaning up the queue and freeing the tagset nbd: use blk_mq_alloc_disk and blk_cleanup_disk block: add a flag to make put_disk on partially initalized disks safer nvme: use blk_mq_alloc_disk brd: convert to blk_alloc_disk/blk_cleanup_disk ubi: use blk_mq_alloc_disk and blk_cleanup_disk Dan Carpenter (1): blk-mq: fix an IS_ERR() vs NULL bug Luis Chamberlain (11): block: return errors from blk_integrity_add block: return errors from disk_alloc_events block: add error handling for device_add_disk / add_disk block: fix device_add_disk() kobject_create_and_add() error handling loop: add error handling support for add_disk() nbd: add error handling support for add_disk() nvme: add error handling support for add_disk() block/brd: add error handling support for add_disk() scsi: sd: Add error handling support for add_disk() scsi: sr: Add error handling support for add_disk() mtd/ubi/block: add error handling support for add_disk() Tetsuo Handa (1): block: check minor range in device_add_disk() Wang Qing (1): nbd: fix order of cleaning up the queue and freeing the tagset Yu Kuai (1): block: fix memory leak for elevator on add_disk failure Zhong Jinghua (9): block: return errors from blk_register_region block: Fix the kabi change in device_add_disk block: Fix the kabi change on blk_register_region block: call blk_get_queue earlier in __device_add_disk block: Fix minor range check in device_add_disk() block: Set memalloc_noio to false in the error path scsi: sd: Clean up sdkp if device_add_disk() failed mtd/ubi/block: Fix null pointer dereference issue in error path mtd/ubi/block: Fix uaf problem in ubiblock_cleanup block/blk-integrity.c | 12 +- block/blk-mq.c | 19 +++ block/blk.h | 5 +- block/genhd.c | 286 ++++++++++++++++++++++++++------------- drivers/block/brd.c | 103 ++++++-------- drivers/block/loop.c | 26 ++-- drivers/block/nbd.c | 59 ++++---- drivers/mtd/ubi/block.c | 76 +++++------ drivers/nvme/host/core.c | 41 +++--- drivers/scsi/sd.c | 8 +- drivers/scsi/sr.c | 7 +- include/linux/blk-mq.h | 12 ++ include/linux/genhd.h | 36 +++++ 13 files changed, 417 insertions(+), 273 deletions(-) -- 2.31.1

2 39

[PATCH OLK-5.10 1/3] LoongArch: Define relocation types for ABI v2.10
by Hongchen Zhang 31 Oct '23

31 Oct '23

The relocation types from 101 to 109 are used by GNU binutils >= 2.41, add their definitions to use them in later patches. Link: https://sourceware.org/git/?p=binutils-gdb.git;a=blob;f=include/elf/loongar… Cc: <stable(a)vger.kernel.org> Signed-off-by: Tiezhu Yang <yangtiezhu(a)loongson.cn> Signed-off-by: Huacai Chen <chenhuacai(a)loongson.cn> Signed-off-by: Hongchen Zhang <zhanghongchen(a)loongson.cn> --- arch/loongarch/include/asm/elf.h | 9 +++++++++ arch/loongarch/kernel/module.c | 2 +- 2 files changed, 10 insertions(+), 1 deletion(-) diff --git a/arch/loongarch/include/asm/elf.h b/arch/loongarch/include/asm/elf.h index 7af0cebf28d7..b9a4ab54285c 100644 --- a/arch/loongarch/include/asm/elf.h +++ b/arch/loongarch/include/asm/elf.h @@ -111,6 +111,15 @@ #define R_LARCH_TLS_GD_HI20 98 #define R_LARCH_32_PCREL 99 #define R_LARCH_RELAX 100 +#define R_LARCH_DELETE 101 +#define R_LARCH_ALIGN 102 +#define R_LARCH_PCREL20_S2 103 +#define R_LARCH_CFA 104 +#define R_LARCH_ADD6 105 +#define R_LARCH_SUB6 106 +#define R_LARCH_ADD_ULEB128 107 +#define R_LARCH_SUB_ULEB128 108 +#define R_LARCH_64_PCREL 109 #ifndef ELF_ARCH diff --git a/arch/loongarch/kernel/module.c b/arch/loongarch/kernel/module.c index 8127df5dd0ad..0b8ce1e0f83b 100644 --- a/arch/loongarch/kernel/module.c +++ b/arch/loongarch/kernel/module.c @@ -387,7 +387,7 @@ typedef int (*reloc_rela_handler)(struct module *mod, u32 *location, Elf_Addr v, /* The handlers for known reloc types */ static reloc_rela_handler reloc_rela_handlers[] = { - [R_LARCH_NONE ... R_LARCH_RELAX] = apply_r_larch_error, + [R_LARCH_NONE ... R_LARCH_64_PCREL] = apply_r_larch_error, [R_LARCH_NONE] = apply_r_larch_none, [R_LARCH_32] = apply_r_larch_32, -- 2.33.0

2 1

[PATCH OLK-5.10 3/3] LoongArch: Add support for 64_PCREL relocation type
by Hongchen Zhang 31 Oct '23

31 Oct '23

When build and update kernel with the latest upstream binutils and loongson3_defconfig, module loader fails with: kmod: zsmalloc: Unknown relocation type 109 kmod: fuse: Unknown relocation type 109 kmod: fuse: Unknown relocation type 109 kmod: radeon: Unknown relocation type 109 kmod: nf_tables: Unknown relocation type 109 kmod: nf_tables: Unknown relocation type 109 This is because the latest upstream binutils replaces a pair of ADD64 and SUB64 with 64_PCREL, so add support for 64_PCREL relocation type. Link: https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=ecb802d02eeb Cc: <stable(a)vger.kernel.org> Signed-off-by: Tiezhu Yang <yangtiezhu(a)loongson.cn> Signed-off-by: Huacai Chen <chenhuacai(a)loongson.cn> Signed-off-by: Hongchen Zhang <zhanghongchen(a)loongson.cn> --- arch/loongarch/kernel/module.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/arch/loongarch/kernel/module.c b/arch/loongarch/kernel/module.c index a67f9def7927..b4cf3383a8bc 100644 --- a/arch/loongarch/kernel/module.c +++ b/arch/loongarch/kernel/module.c @@ -381,6 +381,15 @@ static int apply_r_larch_32_pcrel(struct module *mod, u32 *location, Elf_Addr v, return 0; } +static int apply_r_larch_64_pcrel(struct module *mod, u32 *location, Elf_Addr v, + s64 *rela_stack, size_t *rela_stack_top, unsigned int type) +{ + ptrdiff_t offset = (void *)v - (void *)location; + + *(u64 *)location = offset; + return 0; +} + /* * reloc_handlers_rela() - Apply a particular relocation to a module * @mod: the module to apply the reloc to @@ -414,6 +423,7 @@ static reloc_rela_handler reloc_rela_handlers[] = { [R_LARCH_PCALA_HI20...R_LARCH_PCALA64_HI12] = apply_r_larch_pcala, [R_LARCH_GOT_PC_HI20...R_LARCH_GOT_PC_LO12] = apply_r_larch_got_pc, [R_LARCH_32_PCREL] = apply_r_larch_32_pcrel, + [R_LARCH_64_PCREL] = apply_r_larch_64_pcrel, }; int apply_relocate_add(Elf_Shdr *sechdrs, const char *strtab, -- 2.33.0

1 0

[PATCH OLK-5.10 2/3] LoongArch: Add support for 32_PCREL relocation type
by Hongchen Zhang 31 Oct '23

31 Oct '23

When build and update kernel with the latest upstream binutils and loongson3_defconfig, module loader fails with: kmod: zsmalloc: Unsupport relocation type 99, please add its support. kmod: fuse: Unsupport relocation type 99, please add its support. kmod: ipmi_msghandler: Unsupport relocation type 99, please add its support. kmod: ipmi_msghandler: Unsupport relocation type 99, please add its support. kmod: pstore: Unsupport relocation type 99, please add its support. kmod: drm_display_helper: Unsupport relocation type 99, please add its support. kmod: drm_display_helper: Unsupport relocation type 99, please add its support. kmod: drm_display_helper: Unsupport relocation type 99, please add its support. kmod: fuse: Unsupport relocation type 99, please add its support. kmod: fat: Unsupport relocation type 99, please add its support. This is because the latest upstream binutils replaces a pair of ADD32 and SUB32 with 32_PCREL, so add support for 32_PCREL relocation type. Link: https://sourceware.org/git/?p=binutils-gdb.git;a=commit;h=ecb802d02eeb Cc: <stable(a)vger.kernel.org> Co-developed-by: Youling Tang <tangyouling(a)loongson.cn> Signed-off-by: Youling Tang <tangyouling(a)loongson.cn> Signed-off-by: Tiezhu Yang <yangtiezhu(a)loongson.cn> Signed-off-by: Huacai Chen <chenhuacai(a)loongson.cn> Signed-off-by: Hongchen Zhang <zhanghongchen(a)loongson.cn> --- arch/loongarch/kernel/module.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/arch/loongarch/kernel/module.c b/arch/loongarch/kernel/module.c index 0b8ce1e0f83b..a67f9def7927 100644 --- a/arch/loongarch/kernel/module.c +++ b/arch/loongarch/kernel/module.c @@ -372,6 +372,15 @@ static int apply_r_larch_got_pc(struct module *mod, u32 *location, Elf_Addr v, return apply_r_larch_pcala(mod, location, got, rela_stack, rela_stack_top, type); } +static int apply_r_larch_32_pcrel(struct module *mod, u32 *location, Elf_Addr v, + s64 *rela_stack, size_t *rela_stack_top, unsigned int type) +{ + ptrdiff_t offset = (void *)v - (void *)location; + + *(u32 *)location = offset; + return 0; +} + /* * reloc_handlers_rela() - Apply a particular relocation to a module * @mod: the module to apply the reloc to @@ -404,6 +413,7 @@ static reloc_rela_handler reloc_rela_handlers[] = { [R_LARCH_B26] = apply_r_larch_b26, [R_LARCH_PCALA_HI20...R_LARCH_PCALA64_HI12] = apply_r_larch_pcala, [R_LARCH_GOT_PC_HI20...R_LARCH_GOT_PC_LO12] = apply_r_larch_got_pc, + [R_LARCH_32_PCREL] = apply_r_larch_32_pcrel, }; int apply_relocate_add(Elf_Shdr *sechdrs, const char *strtab, -- 2.33.0

1 0

[PATCH OLK-5.10] io_uring/fdinfo: lock SQ thread while retrieving thread cpu/pid
by Baokun Li 30 Oct '23

30 Oct '23

From: Jens Axboe <axboe(a)kernel.dk> mainline inclusion from mainline-v6.7 commit 7644b1a1c9a7ae8ab99175989bfc8676055edb46 category: bugfix bugzilla: 189322, https://gitee.com/src-openeuler/kernel/issues/I8BQSX CVE: CVE-2023-46862 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- We could race with SQ thread exit, and if we do, we'll hit a NULL pointer dereference when the thread is cleared. Grab the SQPOLL data lock before attempting to get the task cpu and pid for fdinfo, this ensures we have a stable view of it. Cc: stable(a)vger.kernel.org Link: https://bugzilla.kernel.org/show_bug.cgi?id=218032 Reviewed-by: Gabriel Krisman Bertazi <krisman(a)suse.de> Signed-off-by: Jens Axboe <axboe(a)kernel.dk> Conflicts: io_uring/io_uring.c Signed-off-by: Baokun Li <libaokun1(a)huawei.com> --- io_uring/io_uring.c | 18 ++++++++++++------ 1 file changed, 12 insertions(+), 6 deletions(-) diff --git a/io_uring/io_uring.c b/io_uring/io_uring.c index 2e61a67938a7..3fa584b0ee0a 100644 --- a/io_uring/io_uring.c +++ b/io_uring/io_uring.c @@ -10180,7 +10180,7 @@ static int io_uring_show_cred(struct seq_file *m, unsigned int id, static void __io_uring_show_fdinfo(struct io_ring_ctx *ctx, struct seq_file *m) { - struct io_sq_data *sq = NULL; + int sq_pid = -1, sq_cpu = -1; bool has_lock; int i; @@ -10193,13 +10193,19 @@ static void __io_uring_show_fdinfo(struct io_ring_ctx *ctx, struct seq_file *m) has_lock = mutex_trylock(&ctx->uring_lock); if (has_lock && (ctx->flags & IORING_SETUP_SQPOLL)) { - sq = ctx->sq_data; - if (!sq->thread) - sq = NULL; + struct io_sq_data *sq = ctx->sq_data; + + if (mutex_trylock(&sq->lock)) { + if (sq->thread) { + sq_pid = task_pid_nr(sq->thread); + sq_cpu = task_cpu(sq->thread); + } + mutex_unlock(&sq->lock); + } } - seq_printf(m, "SqThread:\t%d\n", sq ? task_pid_nr(sq->thread) : -1); - seq_printf(m, "SqThreadCpu:\t%d\n", sq ? task_cpu(sq->thread) : -1); + seq_printf(m, "SqThread:\t%d\n", sq_pid); + seq_printf(m, "SqThreadCpu:\t%d\n", sq_cpu); seq_printf(m, "UserFiles:\t%u\n", ctx->nr_user_files); for (i = 0; has_lock && i < ctx->nr_user_files; i++) { struct file *f = io_file_from_index(ctx, i); -- 2.31.1

2 1

[PATCH openEuler-1.0-LTS 0/4] bugfixes for CVE-2022-45884
by liwei 30 Oct '23

30 Oct '23

bugfixes for CVE-2022-45884 Hyunwoo Kim (1): media: dvb-core: Fix use-after-free due to race at dvb_register_device() Keita Suzuki (1): media: dvb-core: Fix double free in dvb_register_device() Lin Ma (2): media: dvbdev: adopts refcnt to avoid UAF media: dvbdev: fix refcnt bug drivers/media/dvb-core/dvb_ca_en50221.c | 2 +- drivers/media/dvb-core/dvb_frontend.c | 2 +- drivers/media/dvb-core/dvbdev.c | 122 ++++++++++++++++++------ include/media/dvbdev.h | 46 ++++++--- 4 files changed, 128 insertions(+), 44 deletions(-) -- 2.25.1

2 5

[openeuler 0/8] libhns: Support STARS over RDMA
by Chengchang Tang 30 Oct '23

30 Oct '23

STARS is a HW scheduler. These patches support hns RoCE working in STARS mode which means RoCE will be scheduled by STARS. Chengchang Tang (8): Update kernel headers for libhns query_device() libhns: Support query HW ID by hnsdv_query_device() Update kernel headers for supporting POE CQs libhns: Add support for POE CQs Update kernel headers for supporting STARS QP in HNS libhns: Support STARS mode QP Update kernel headers for supporting write with notify libhns: Support write with notify kernel-headers/rdma/hns-abi.h | 38 ++++++- providers/hns/hns_roce_u.c | 50 ++++++--- providers/hns/hns_roce_u.h | 1 + providers/hns/hns_roce_u_abi.h | 3 + providers/hns/hns_roce_u_hw_v2.c | 65 +++++++++-- providers/hns/hns_roce_u_hw_v2.h | 1 + providers/hns/hns_roce_u_verbs.c | 229 +++++++++++++++++++++++++++++++++++---- providers/hns/hnsdv.h | 50 +++++++++ providers/hns/libhns.map | 3 + 9 files changed, 393 insertions(+), 47 deletions(-) -- 2.9.5

1 8

[openeuler 0/6] RDMA/hns: Support STARS over RDMA
by Chengchang Tang 30 Oct '23

30 Oct '23

STARS is a HW scheduler. These patches support STARS taking over HNS RoCE. Chengchang Tang (6): RDMA/hns: Support query HW ID from user space. RDMA/hns: Fix print after query hw id failed. RDMA/hns: Support configuring POE channels and creating POE CQs RDMA/hns: Support STARS mode QP RDMA/hns: Support kernel ULP querying HW ID RDMA/hns: Support write with notify drivers/infiniband/hw/hns/Makefile | 3 +- drivers/infiniband/hw/hns/hns_roce_common.h | 3 + drivers/infiniband/hw/hns/hns_roce_cq.c | 114 +++++++++++++++++- drivers/infiniband/hw/hns/hns_roce_debugfs.c | 111 ++++++++++++++++++ drivers/infiniband/hw/hns/hns_roce_device.h | 63 ++++++++++ drivers/infiniband/hw/hns/hns_roce_ext.c | 143 +++++++++++++++++++++++ drivers/infiniband/hw/hns/hns_roce_ext.h | 66 +++++++++++ drivers/infiniband/hw/hns/hns_roce_hw_v2.c | 169 +++++++++++++++++++++++++-- drivers/infiniband/hw/hns/hns_roce_hw_v2.h | 52 +++++++++ drivers/infiniband/hw/hns/hns_roce_main.c | 62 +++++++++- drivers/infiniband/hw/hns/hns_roce_poe.c | 97 +++++++++++++++ drivers/infiniband/hw/hns/hns_roce_qp.c | 62 +++++++++- include/uapi/rdma/hns-abi.h | 37 +++++- 13 files changed, 966 insertions(+), 16 deletions(-) create mode 100644 drivers/infiniband/hw/hns/hns_roce_ext.c create mode 100644 drivers/infiniband/hw/hns/hns_roce_ext.h create mode 100644 drivers/infiniband/hw/hns/hns_roce_poe.c -- 2.9.5

1 6

[PATCH OLK-5.10 v2 00/38] Add error handling for device_add_disk
by Zhong Jinghua 30 Oct '23

30 Oct '23

From: Zhong Jinghua <zhongjinghua(a)huaweicloud.com> Befor these paths, block is not know when drvier is add fail. Now, add error handling for drivers. Christoph Hellwig (14): block: fold register_disk into device_add_disk block: call blk_integrity_add earlier in device_add_disk block: add the events* attributes to disk_attrs block: fix error unwinding in device_add_disk block: clear ->slave_dir when dropping the main slave_dir reference block: add blk_alloc_disk and blk_cleanup_disk APIs blk-mq: add the blk_mq_alloc_disk APIs loop: use blk_mq_alloc_disk and blk_cleanup_disk loop: fix order of cleaning up the queue and freeing the tagset nbd: use blk_mq_alloc_disk and blk_cleanup_disk block: add a flag to make put_disk on partially initalized disks safer nvme: use blk_mq_alloc_disk brd: convert to blk_alloc_disk/blk_cleanup_disk ubi: use blk_mq_alloc_disk and blk_cleanup_disk Dan Carpenter (1): blk-mq: fix an IS_ERR() vs NULL bug Luis Chamberlain (11): block: return errors from blk_integrity_add block: return errors from disk_alloc_events block: add error handling for device_add_disk / add_disk block: fix device_add_disk() kobject_create_and_add() error handling loop: add error handling support for add_disk() nbd: add error handling support for add_disk() nvme: add error handling support for add_disk() block/brd: add error handling support for add_disk() scsi: sd: Add error handling support for add_disk() scsi: sr: Add error handling support for add_disk() mtd/ubi/block: add error handling support for add_disk() Tetsuo Handa (1): block: check minor range in device_add_disk() Wang Qing (1): nbd: fix order of cleaning up the queue and freeing the tagset Yu Kuai (1): block: fix memory leak for elevator on add_disk failure Zhong Jinghua (9): block: return errors from blk_register_region block: Fix the kabi change in device_add_disk block: Fix the kabi change on blk_register_region block: call blk_get_queue earlier in __device_add_disk block: Fix minor range check in device_add_disk() block: Set memalloc_noio to false in the error path scsi: sd: Clean up sdkp if device_add_disk() failed mtd/ubi/block: Fix null pointer dereference issue in error path mtd/ubi/block: Fix uaf problem in ubiblock_cleanup block/blk-integrity.c | 12 +- block/blk-mq.c | 19 +++ block/blk.h | 5 +- block/genhd.c | 286 ++++++++++++++++++++++++++------------- drivers/block/brd.c | 103 ++++++-------- drivers/block/loop.c | 26 ++-- drivers/block/nbd.c | 59 ++++---- drivers/mtd/ubi/block.c | 76 +++++------ drivers/nvme/host/core.c | 41 +++--- drivers/scsi/sd.c | 8 +- drivers/scsi/sr.c | 7 +- include/linux/blk-mq.h | 12 ++ include/linux/genhd.h | 36 +++++ 13 files changed, 417 insertions(+), 273 deletions(-) -- 2.31.1

1 38

[PATCH OLK-5.10 0/3] fix CVE-2023-46813
by Yu Liao 30 Oct '23

30 Oct '23

fix CVE-2023-46813 An issue was discovered in the Linux kernel before 6.5.9, exploitable by local users with userspace access to MMIO registers. Incorrect access checking in the #VC handler and instruction emulation of the SEV-ES emulation of MMIO accesses could lead to arbitrary write access to kernel memory (and thus privilege escalation). This depends on a race condition through which userspace can replace an instruction before the #VC handler reads it. Borislav Petkov (AMD) (1): [Backport] x86/sev: Disable MMIO emulation from user mode Joerg Roedel (2): [Backport] x86/sev: Check IOBM for IOIO exceptions from user-space [Backport] x86/sev: Check for user-space IOIO pointing to kernel space arch/x86/boot/compressed/sev-es.c | 10 ++++++ arch/x86/kernel/sev-es-shared.c | 53 +++++++++++++++++++++++++------ arch/x86/kernel/sev-es.c | 30 +++++++++++++++++ 3 files changed, 84 insertions(+), 9 deletions(-) -- 2.25.1

2 4

[PATCH OLK-5.10 v2 00/38] Add error handling for device_add_disk
by Zhong Jinghua 30 Oct '23

30 Oct '23

From: Zhong Jinghua <zhongjinghua(a)huaweicloud.com> Befor these paths, block is not know when drvier is add fail. Now, add error handling for drivers. Christoph Hellwig (14): block: fold register_disk into device_add_disk block: call blk_integrity_add earlier in device_add_disk block: add the events* attributes to disk_attrs block: fix error unwinding in device_add_disk block: clear ->slave_dir when dropping the main slave_dir reference block: add blk_alloc_disk and blk_cleanup_disk APIs blk-mq: add the blk_mq_alloc_disk APIs loop: use blk_mq_alloc_disk and blk_cleanup_disk loop: fix order of cleaning up the queue and freeing the tagset nbd: use blk_mq_alloc_disk and blk_cleanup_disk block: add a flag to make put_disk on partially initalized disks safer nvme: use blk_mq_alloc_disk brd: convert to blk_alloc_disk/blk_cleanup_disk ubi: use blk_mq_alloc_disk and blk_cleanup_disk Dan Carpenter (1): blk-mq: fix an IS_ERR() vs NULL bug Luis Chamberlain (11): block: return errors from blk_integrity_add block: return errors from disk_alloc_events block: add error handling for device_add_disk / add_disk block: fix device_add_disk() kobject_create_and_add() error handling loop: add error handling support for add_disk() nbd: add error handling support for add_disk() nvme: add error handling support for add_disk() block/brd: add error handling support for add_disk() scsi: sd: Add error handling support for add_disk() scsi: sr: Add error handling support for add_disk() mtd/ubi/block: add error handling support for add_disk() Tetsuo Handa (1): block: check minor range in device_add_disk() Wang Qing (1): nbd: fix order of cleaning up the queue and freeing the tagset Yu Kuai (1): block: fix memory leak for elevator on add_disk failure Zhong Jinghua (9): block: return errors from blk_register_region block: Fix the kabi change in device_add_disk block: Fix the kabi change on blk_register_region block: call blk_get_queue earlier in __device_add_disk block: Fix minor range check in device_add_disk() block: Set memalloc_noio to false in the error path scsi: sd: Clean up sdkp if device_add_disk() failed mtd/ubi/block: Fix null pointer dereference issue in error path mtd/ubi/block: Fix uaf problem in ubiblock_cleanup block/blk-integrity.c | 12 +- block/blk-mq.c | 19 +++ block/blk.h | 5 +- block/genhd.c | 286 ++++++++++++++++++++++++++------------- drivers/block/brd.c | 103 ++++++-------- drivers/block/loop.c | 26 ++-- drivers/block/nbd.c | 59 ++++---- drivers/mtd/ubi/block.c | 76 +++++------ drivers/nvme/host/core.c | 41 +++--- drivers/scsi/sd.c | 8 +- drivers/scsi/sr.c | 7 +- include/linux/blk-mq.h | 12 ++ include/linux/genhd.h | 36 +++++ 13 files changed, 417 insertions(+), 273 deletions(-) -- 2.31.1

2 39

[openEuler-1.0-LTS 00/15] Fix the UAF when task fails.
by Zhong Jinghua 30 Oct '23

30 Oct '23

The NetLlink mechanism is used to send unreliable notifications when a task fails. Bharath Ravi (1): scsi: iscsi: Perform connection failure entirely in kernel space Gabriel Krisman Bertazi (2): scsi: iscsi: Report connection state in sysfs scsi: iscsi: Fix deadlock on recovery path during GFP_IO reclaim Gulam Mohamed (1): scsi: iscsi: Fix race condition between login and sync thread Mike Christie (10): scsi: iscsi: Fix iSCSI cls conn state scsi: iscsi: Force immediate failure during shutdown scsi: iscsi: Rel ref after iscsi_lookup_endpoint() scsi: iscsi: Fix in-kernel conn failure handling scsi: iscsi: Fix set_param() handling scsi: iscsi: Move iscsi_ep_disconnect() scsi: iscsi: Fix offload conn cleanup when iscsid restarts scsi: iscsi: Fix conn cleanup and stop race during iscsid restart scsi: iscsi: Fix endpoint reuse regression scsi: iscsi: Fix unbound endpoint error handling Zhong Jinghua (1): scsi: iscsi: Fix kabi change in iscsi_cls_conn drivers/infiniband/ulp/iser/iscsi_iser.c | 1 + drivers/scsi/be2iscsi/be_iscsi.c | 19 +- drivers/scsi/bnx2i/bnx2i_iscsi.c | 23 +- drivers/scsi/cxgbi/libcxgbi.c | 12 +- drivers/scsi/libiscsi.c | 21 +- drivers/scsi/qedi/qedi_iscsi.c | 25 +- drivers/scsi/qla4xxx/ql4_os.c | 1 + drivers/scsi/scsi_transport_iscsi.c | 487 ++++++++++++++++++----- include/scsi/scsi_transport_iscsi.h | 28 ++ 9 files changed, 485 insertions(+), 132 deletions(-) -- 2.31.1

1 15

[PATCH OLK-5.10] net: xfrm: Fix xfrm_address_filter OOB read
by Dong Chenchen 30 Oct '23

30 Oct '23

From: Lin Ma <linma(a)zju.edu.cn> stable inclusion from stable-v5.10.192 commit 7e50815d29037e08d3d26f3ebc41bcec729847b7 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/I85DZ6 CVE: CVE-2023-39194 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit dfa73c17d55b921e1d4e154976de35317e43a93a ] We found below OOB crash: [ 44.211730] ================================================================== [ 44.212045] BUG: KASAN: slab-out-of-bounds in memcmp+0x8b/0xb0 [ 44.212045] Read of size 8 at addr ffff88800870f320 by task poc.xfrm/97 [ 44.212045] [ 44.212045] CPU: 0 PID: 97 Comm: poc.xfrm Not tainted 6.4.0-rc7-00072-gdad9774deaf1-dirty #4 [ 44.212045] Call Trace: [ 44.212045] <TASK> [ 44.212045] dump_stack_lvl+0x37/0x50 [ 44.212045] print_report+0xcc/0x620 [ 44.212045] ? __virt_addr_valid+0xf3/0x170 [ 44.212045] ? memcmp+0x8b/0xb0 [ 44.212045] kasan_report+0xb2/0xe0 [ 44.212045] ? memcmp+0x8b/0xb0 [ 44.212045] kasan_check_range+0x39/0x1c0 [ 44.212045] memcmp+0x8b/0xb0 [ 44.212045] xfrm_state_walk+0x21c/0x420 [ 44.212045] ? __pfx_dump_one_state+0x10/0x10 [ 44.212045] xfrm_dump_sa+0x1e2/0x290 [ 44.212045] ? __pfx_xfrm_dump_sa+0x10/0x10 [ 44.212045] ? __kernel_text_address+0xd/0x40 [ 44.212045] ? kasan_unpoison+0x27/0x60 [ 44.212045] ? mutex_lock+0x60/0xe0 [ 44.212045] ? __pfx_mutex_lock+0x10/0x10 [ 44.212045] ? kasan_save_stack+0x22/0x50 [ 44.212045] netlink_dump+0x322/0x6c0 [ 44.212045] ? __pfx_netlink_dump+0x10/0x10 [ 44.212045] ? mutex_unlock+0x7f/0xd0 [ 44.212045] ? __pfx_mutex_unlock+0x10/0x10 [ 44.212045] __netlink_dump_start+0x353/0x430 [ 44.212045] xfrm_user_rcv_msg+0x3a4/0x410 [ 44.212045] ? __pfx__raw_spin_lock_irqsave+0x10/0x10 [ 44.212045] ? __pfx_xfrm_user_rcv_msg+0x10/0x10 [ 44.212045] ? __pfx_xfrm_dump_sa+0x10/0x10 [ 44.212045] ? __pfx_xfrm_dump_sa_done+0x10/0x10 [ 44.212045] ? __stack_depot_save+0x382/0x4e0 [ 44.212045] ? filter_irq_stacks+0x1c/0x70 [ 44.212045] ? kasan_save_stack+0x32/0x50 [ 44.212045] ? kasan_save_stack+0x22/0x50 [ 44.212045] ? kasan_set_track+0x25/0x30 [ 44.212045] ? __kasan_slab_alloc+0x59/0x70 [ 44.212045] ? kmem_cache_alloc_node+0xf7/0x260 [ 44.212045] ? kmalloc_reserve+0xab/0x120 [ 44.212045] ? __alloc_skb+0xcf/0x210 [ 44.212045] ? netlink_sendmsg+0x509/0x700 [ 44.212045] ? sock_sendmsg+0xde/0xe0 [ 44.212045] ? __sys_sendto+0x18d/0x230 [ 44.212045] ? __x64_sys_sendto+0x71/0x90 [ 44.212045] ? do_syscall_64+0x3f/0x90 [ 44.212045] ? entry_SYSCALL_64_after_hwframe+0x72/0xdc [ 44.212045] ? netlink_sendmsg+0x509/0x700 [ 44.212045] ? sock_sendmsg+0xde/0xe0 [ 44.212045] ? __sys_sendto+0x18d/0x230 [ 44.212045] ? __x64_sys_sendto+0x71/0x90 [ 44.212045] ? do_syscall_64+0x3f/0x90 [ 44.212045] ? entry_SYSCALL_64_after_hwframe+0x72/0xdc [ 44.212045] ? kasan_save_stack+0x22/0x50 [ 44.212045] ? kasan_set_track+0x25/0x30 [ 44.212045] ? kasan_save_free_info+0x2e/0x50 [ 44.212045] ? __kasan_slab_free+0x10a/0x190 [ 44.212045] ? kmem_cache_free+0x9c/0x340 [ 44.212045] ? netlink_recvmsg+0x23c/0x660 [ 44.212045] ? sock_recvmsg+0xeb/0xf0 [ 44.212045] ? __sys_recvfrom+0x13c/0x1f0 [ 44.212045] ? __x64_sys_recvfrom+0x71/0x90 [ 44.212045] ? do_syscall_64+0x3f/0x90 [ 44.212045] ? entry_SYSCALL_64_after_hwframe+0x72/0xdc [ 44.212045] ? copyout+0x3e/0x50 [ 44.212045] netlink_rcv_skb+0xd6/0x210 [ 44.212045] ? __pfx_xfrm_user_rcv_msg+0x10/0x10 [ 44.212045] ? __pfx_netlink_rcv_skb+0x10/0x10 [ 44.212045] ? __pfx_sock_has_perm+0x10/0x10 [ 44.212045] ? mutex_lock+0x8d/0xe0 [ 44.212045] ? __pfx_mutex_lock+0x10/0x10 [ 44.212045] xfrm_netlink_rcv+0x44/0x50 [ 44.212045] netlink_unicast+0x36f/0x4c0 [ 44.212045] ? __pfx_netlink_unicast+0x10/0x10 [ 44.212045] ? netlink_recvmsg+0x500/0x660 [ 44.212045] netlink_sendmsg+0x3b7/0x700 [ 44.212045] ? __pfx_netlink_sendmsg+0x10/0x10 [ 44.212045] ? __pfx_netlink_sendmsg+0x10/0x10 [ 44.212045] sock_sendmsg+0xde/0xe0 [ 44.212045] __sys_sendto+0x18d/0x230 [ 44.212045] ? __pfx___sys_sendto+0x10/0x10 [ 44.212045] ? rcu_core+0x44a/0xe10 [ 44.212045] ? __rseq_handle_notify_resume+0x45b/0x740 [ 44.212045] ? _raw_spin_lock_irq+0x81/0xe0 [ 44.212045] ? __pfx___rseq_handle_notify_resume+0x10/0x10 [ 44.212045] ? __pfx_restore_fpregs_from_fpstate+0x10/0x10 [ 44.212045] ? __pfx_blkcg_maybe_throttle_current+0x10/0x10 [ 44.212045] ? __pfx_task_work_run+0x10/0x10 [ 44.212045] __x64_sys_sendto+0x71/0x90 [ 44.212045] do_syscall_64+0x3f/0x90 [ 44.212045] entry_SYSCALL_64_after_hwframe+0x72/0xdc [ 44.212045] RIP: 0033:0x44b7da [ 44.212045] RSP: 002b:00007ffdc8838548 EFLAGS: 00000246 ORIG_RAX: 000000000000002c [ 44.212045] RAX: ffffffffffffffda RBX: 00007ffdc8839978 RCX: 000000000044b7da [ 44.212045] RDX: 0000000000000038 RSI: 00007ffdc8838770 RDI: 0000000000000003 [ 44.212045] RBP: 00007ffdc88385b0 R08: 00007ffdc883858c R09: 000000000000000c [ 44.212045] R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000001 [ 44.212045] R13: 00007ffdc8839968 R14: 00000000004c37d0 R15: 0000000000000001 [ 44.212045] </TASK> [ 44.212045] [ 44.212045] Allocated by task 97: [ 44.212045] kasan_save_stack+0x22/0x50 [ 44.212045] kasan_set_track+0x25/0x30 [ 44.212045] __kasan_kmalloc+0x7f/0x90 [ 44.212045] __kmalloc_node_track_caller+0x5b/0x140 [ 44.212045] kmemdup+0x21/0x50 [ 44.212045] xfrm_dump_sa+0x17d/0x290 [ 44.212045] netlink_dump+0x322/0x6c0 [ 44.212045] __netlink_dump_start+0x353/0x430 [ 44.212045] xfrm_user_rcv_msg+0x3a4/0x410 [ 44.212045] netlink_rcv_skb+0xd6/0x210 [ 44.212045] xfrm_netlink_rcv+0x44/0x50 [ 44.212045] netlink_unicast+0x36f/0x4c0 [ 44.212045] netlink_sendmsg+0x3b7/0x700 [ 44.212045] sock_sendmsg+0xde/0xe0 [ 44.212045] __sys_sendto+0x18d/0x230 [ 44.212045] __x64_sys_sendto+0x71/0x90 [ 44.212045] do_syscall_64+0x3f/0x90 [ 44.212045] entry_SYSCALL_64_after_hwframe+0x72/0xdc [ 44.212045] [ 44.212045] The buggy address belongs to the object at ffff88800870f300 [ 44.212045] which belongs to the cache kmalloc-64 of size 64 [ 44.212045] The buggy address is located 32 bytes inside of [ 44.212045] allocated 36-byte region [ffff88800870f300, ffff88800870f324) [ 44.212045] [ 44.212045] The buggy address belongs to the physical page: [ 44.212045] page:00000000e4de16ee refcount:1 mapcount:0 mapping:000000000 ... [ 44.212045] flags: 0x100000000000200(slab|node=0|zone=1) [ 44.212045] page_type: 0xffffffff() [ 44.212045] raw: 0100000000000200 ffff888004c41640 dead000000000122 0000000000000000 [ 44.212045] raw: 0000000000000000 0000000080200020 00000001ffffffff 0000000000000000 [ 44.212045] page dumped because: kasan: bad access detected [ 44.212045] [ 44.212045] Memory state around the buggy address: [ 44.212045] ffff88800870f200: fa fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc [ 44.212045] ffff88800870f280: 00 00 00 00 00 fc fc fc fc fc fc fc fc fc fc fc [ 44.212045] >ffff88800870f300: 00 00 00 00 04 fc fc fc fc fc fc fc fc fc fc fc [ 44.212045] ^ [ 44.212045] ffff88800870f380: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc [ 44.212045] ffff88800870f400: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc [ 44.212045] ================================================================== By investigating the code, we find the root cause of this OOB is the lack of checks in xfrm_dump_sa(). The buggy code allows a malicious user to pass arbitrary value of filter->splen/dplen. Hence, with crafted xfrm states, the attacker can achieve 8 bytes heap OOB read, which causes info leak. if (attrs[XFRMA_ADDRESS_FILTER]) { filter = kmemdup(nla_data(attrs[XFRMA_ADDRESS_FILTER]), sizeof(*filter), GFP_KERNEL); if (filter == NULL) return -ENOMEM; // NO MORE CHECKS HERE !!! } This patch fixes the OOB by adding necessary boundary checks, just like the code in pfkey_dump() function. Fixes: d3623099d350 ("ipsec: add support of limited SA dump") Signed-off-by: Lin Ma <linma(a)zju.edu.cn> Signed-off-by: Steffen Klassert <steffen.klassert(a)secunet.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Dong Chenchen <dongchenchen2(a)huawei.com> --- net/xfrm/xfrm_user.c | 9 +++++++++ 1 file changed, 9 insertions(+) diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c index 138f71c67969..9956acbaffcc 100644 --- a/net/xfrm/xfrm_user.c +++ b/net/xfrm/xfrm_user.c @@ -1049,6 +1049,15 @@ static int xfrm_dump_sa(struct sk_buff *skb, struct netlink_callback *cb) sizeof(*filter), GFP_KERNEL); if (filter == NULL) return -ENOMEM; + + /* see addr_match(), (prefix length >> 5) << 2 + * will be used to compare xfrm_address_t + */ + if (filter->splen > (sizeof(xfrm_address_t) << 3) || + filter->dplen > (sizeof(xfrm_address_t) << 3)) { + kfree(filter); + return -EINVAL; + } } if (attrs[XFRMA_PROTO]) -- 2.25.1

2 1

[PATCH openEuler-22.03-LTS v2 00/16] arm64: fix a concurrency issue in emulation_proc_handler()
by Jinjie Ruan 30 Oct '23

30 Oct '23

In linux-6.1, the related code is refactored in commit 124c49b1b5d9 ("arm64: armv8_deprecated: rework deprected instruction handling") and this issue was incidentally fixed. This patch set try to adapt the refactoring stable 5.10 patches to solve the problem of repeated addition of linked lists described below. How to reproduce: CONFIG_ARMV8_DEPRECATED=y, CONFIG_SWP_EMULATION=y, and CONFIG_DEBUG_LIST=y, then launch two shell executions: #!/bin/bash while [ 1 ]; do echo 1 > /proc/sys/abi/swp done or "echo 1 > /proc/sys/abi/swp" and then launch two shell executions: #!/bin/bash while [ 1 ]; do echo 0 > /proc/sys/abi/swp done In emulation_proc_handler(), read and write operations are performed on insn->current_mode. In the concurrency scenario, mutex only protects writing insn->current_mode, and not protects the read. Suppose there are two concurrent tasks, task1 updates insn->current_mode to INSN_EMULATE in the critical section, the prev_mode of task2 is still the old data INSN_UNDEF of insn->current_mode. As a result, two tasks call update_insn_emulation_mode twice with prev_mode = INSN_UNDEF and current_mode = INSN_EMULATE, then call register_emulation_hooks twice, resulting in a list_add double problem. commit 124c49b1b5d9 ("arm64: armv8_deprecated: rework deprected instruction handling") remove the dynamic registration and unregistration so remove the register_undef_hook() function, so the below problem was incidentally fixed. Call trace: __list_add_valid+0xd8/0xe4 register_undef_hook+0x94/0x13c update_insn_emulation_mode+0xd0/0x12c emulation_proc_handler+0xd8/0xf4 proc_sys_call_handler+0x140/0x250 proc_sys_write+0x1c/0x2c new_sync_write+0xec/0x18c vfs_write+0x214/0x2ac ksys_write+0x70/0xfc __arm64_sys_write+0x24/0x30 el0_svc_common.constprop.0+0x7c/0x1bc do_el0_svc+0x2c/0x94 el0_svc+0x20/0x30 el0_sync_handler+0xb0/0xb4 el0_sync+0x160/0x180 Call trace: __list_del_entry_valid+0xac/0x110 unregister_undef_hook+0x34/0x80 update_insn_emulation_mode+0xf0/0x180 emulation_proc_handler+0x8c/0xd8 proc_sys_call_handler+0x1d8/0x208 proc_sys_write+0x14/0x20 new_sync_write+0xf0/0x190 vfs_write+0x304/0x388 ksys_write+0x6c/0x100 __arm64_sys_write+0x1c/0x28 el0_svc_common.constprop.4+0x68/0x188 do_el0_svc+0x24/0xa0 el0_svc+0x14/0x20 el0_sync_handler+0x90/0xb8 el0_sync+0x160/0x180 The first patch is the self-developed patch, so revert it. The next 5 patches is a patch set which provides context for subsequent refactoring 9 patches, especially commit 0f2cb928a154 ("arm64: consistently pass ESR_ELx to die()") which modify do_undefinstr() to add a ESR_ELx value arg, and then commit 61d64a376ea8 ("arm64: split EL0/EL1 UNDEF handlers") splits do_undefinstr() handler into separate do_el0_undef() and do_el1_undef() handlers. The 9 patches after that is another refactoring patch set, which is in preparation for the main rework commit 124c49b1b5d9 ("arm64: armv8_deprecated: rework deprected instruction handling"). To remove struct undef_hook, commit bff8f413c71f ("arm64: factor out EL1 SSBS emulation hook") factor out EL1 SSBS emulation hook, which also avoid call call_undef_hook() in do_el1_undef(), commit f5962add74b6 ("arm64: rework EL0 MRS emulation") factor out EL0 MRS emulation hook, which also prepare for replacing call_undef_hook() in do_el0_undef(). To replace call_undef_hook() function, commit 61d64a376ea8 ("arm64: split EL0/EL1 UNDEF handlers") split the do_undefinstr() into do_el0_undef() and do_el1_undef() functions, and commit dbfbd87efa79 ("arm64: factor insn read out of call_undef_hook()") factor user_insn_read() from call_undef_hook() so the main rework patch can replace the call_undef_hook() in do_el0_undef(). The last patch is a bugfix for the main rework patch. Fixes: af483947d472 ("arm64: fix oops in concurrently setting insn_emulation sysctls") Signed-off-by: Jinjie Ruan <ruanjinjie(a)huawei.com> Jinjie Ruan (1): Revert "arm64: fix a concurrency issue in emulation_proc_handler()" Mark Rutland (14): arm64: report EL1 UNDEFs better arm64: die(): pass 'err' as long arm64: consistently pass ESR_ELx to die() arm64: rework FPAC exception handling arm64: rework BTI exception handling arm64: allow kprobes on EL0 handlers arm64: split EL0/EL1 UNDEF handlers arm64: factor out EL1 SSBS emulation hook arm64: factor insn read out of call_undef_hook() arm64: rework EL0 MRS emulation arm64: armv8_deprecated: fold ops into insn_emulation arm64: armv8_deprecated move emulation functions arm64: armv8_deprecated: move aarch32 helper earlier arm64: armv8_deprecated: rework deprected instruction handling Ren Zhijie (1): arm64: armv8_deprecated: fix unused-function error arch/arm64/include/asm/cpufeature.h | 3 +- arch/arm64/include/asm/exception.h | 13 +- arch/arm64/include/asm/spectre.h | 2 + arch/arm64/include/asm/system_misc.h | 2 +- arch/arm64/include/asm/traps.h | 19 +- arch/arm64/kernel/armv8_deprecated.c | 574 +++++++++++++-------------- arch/arm64/kernel/cpufeature.c | 23 +- arch/arm64/kernel/entry-common.c | 36 +- arch/arm64/kernel/proton-pack.c | 26 +- arch/arm64/kernel/traps.c | 125 +++--- 10 files changed, 396 insertions(+), 427 deletions(-) -- 2.34.1

2 17

[PATCH openEuler-22.03-LTS-SP1 00/15] arm64: fix a concurrency issue in emulation_proc_handler()
by Jinjie Ruan 30 Oct '23

30 Oct '23

In linux-6.1, the related code is refactored in commit 124c49b1b5d9 ("arm64: armv8_deprecated: rework deprected instruction handling") and this issue was incidentally fixed. This patch set try to adapt the refactoring stable 5.10 patches to solve the problem of repeated addition of linked lists described below. How to reproduce: CONFIG_ARMV8_DEPRECATED=y, CONFIG_SWP_EMULATION=y, and CONFIG_DEBUG_LIST=y, then launch two shell executions: #!/bin/bash while [ 1 ]; do echo 1 > /proc/sys/abi/swp done or "echo 1 > /proc/sys/abi/swp" and then launch two shell executions: #!/bin/bash while [ 1 ]; do echo 0 > /proc/sys/abi/swp done In emulation_proc_handler(), read and write operations are performed on insn->current_mode. In the concurrency scenario, mutex only protects writing insn->current_mode, and not protects the read. Suppose there are two concurrent tasks, task1 updates insn->current_mode to INSN_EMULATE in the critical section, the prev_mode of task2 is still the old data INSN_UNDEF of insn->current_mode. As a result, two tasks call update_insn_emulation_mode twice with prev_mode = INSN_UNDEF and current_mode = INSN_EMULATE, then call register_emulation_hooks twice, resulting in a list_add double problem. commit 124c49b1b5d9 ("arm64: armv8_deprecated: rework deprected instruction handling") remove the dynamic registration and unregistration so remove the register_undef_hook() function, so the below problem was incidentally fixed. Call trace: __list_add_valid+0xd8/0xe4 register_undef_hook+0x94/0x13c update_insn_emulation_mode+0xd0/0x12c emulation_proc_handler+0xd8/0xf4 proc_sys_call_handler+0x140/0x250 proc_sys_write+0x1c/0x2c new_sync_write+0xec/0x18c vfs_write+0x214/0x2ac ksys_write+0x70/0xfc __arm64_sys_write+0x24/0x30 el0_svc_common.constprop.0+0x7c/0x1bc do_el0_svc+0x2c/0x94 el0_svc+0x20/0x30 el0_sync_handler+0xb0/0xb4 el0_sync+0x160/0x180 Call trace: __list_del_entry_valid+0xac/0x110 unregister_undef_hook+0x34/0x80 update_insn_emulation_mode+0xf0/0x180 emulation_proc_handler+0x8c/0xd8 proc_sys_call_handler+0x1d8/0x208 proc_sys_write+0x14/0x20 new_sync_write+0xf0/0x190 vfs_write+0x304/0x388 ksys_write+0x6c/0x100 __arm64_sys_write+0x1c/0x28 el0_svc_common.constprop.4+0x68/0x188 do_el0_svc+0x24/0xa0 el0_svc+0x14/0x20 el0_sync_handler+0x90/0xb8 el0_sync+0x160/0x180 The first patch is the self-developed patch, so revert it. The next 4 patches is a patch set which provides context for subsequent refactoring 9 patches, especially commit 0f2cb928a154 ("arm64: consistently pass ESR_ELx to die()") which modify do_undefinstr() to add a ESR_ELx value arg, and then commit 61d64a376ea8 ("arm64: split EL0/EL1 UNDEF handlers") splits do_undefinstr() handler into separate do_el0_undef() and do_el1_undef() handlers. The 9 patches after that is another refactoring patch set, which is in preparation for the main rework commit 124c49b1b5d9 ("arm64: armv8_deprecated: rework deprected instruction handling"). To remove struct undef_hook, commit bff8f413c71f ("arm64: factor out EL1 SSBS emulation hook") factor out EL1 SSBS emulation hook, which also avoid call call_undef_hook() in do_el1_undef(), commit f5962add74b6 ("arm64: rework EL0 MRS emulation") factor out EL0 MRS emulation hook, which also prepare for replacing call_undef_hook() in do_el0_undef(). To replace call_undef_hook() function, commit 61d64a376ea8 ("arm64: split EL0/EL1 UNDEF handlers") split the do_undefinstr() into do_el0_undef() and do_el1_undef() functions, and commit dbfbd87efa79 ("arm64: factor insn read out of call_undef_hook()") factor user_insn_read() from call_undef_hook() so the main rework patch can replace the call_undef_hook() in do_el0_undef(). The last patch is a bugfix for the main rework patch. Fixes: af483947d472 ("arm64: fix oops in concurrently setting insn_emulation sysctls") Signed-off-by: Jinjie Ruan <ruanjinjie(a)huawei.com> Jinjie Ruan (1): Revert "arm64: fix a concurrency issue in emulation_proc_handler()" Mark Rutland (13): arm64: die(): pass 'err' as long arm64: consistently pass ESR_ELx to die() arm64: rework FPAC exception handling arm64: rework BTI exception handling arm64: allow kprobes on EL0 handlers arm64: split EL0/EL1 UNDEF handlers arm64: factor out EL1 SSBS emulation hook arm64: factor insn read out of call_undef_hook() arm64: rework EL0 MRS emulation arm64: armv8_deprecated: fold ops into insn_emulation arm64: armv8_deprecated move emulation functions arm64: armv8_deprecated: move aarch32 helper earlier arm64: armv8_deprecated: rework deprected instruction handling Ren Zhijie (1): arm64: armv8_deprecated: fix unused-function error arch/arm64/include/asm/cpufeature.h | 3 +- arch/arm64/include/asm/exception.h | 13 +- arch/arm64/include/asm/spectre.h | 2 + arch/arm64/include/asm/system_misc.h | 2 +- arch/arm64/include/asm/traps.h | 19 +- arch/arm64/kernel/armv8_deprecated.c | 574 +++++++++++++-------------- arch/arm64/kernel/cpufeature.c | 23 +- arch/arm64/kernel/entry-common.c | 36 +- arch/arm64/kernel/proton-pack.c | 26 +- arch/arm64/kernel/traps.c | 125 +++--- 10 files changed, 395 insertions(+), 428 deletions(-) -- 2.34.1

2 16

[PATCH openEuler-22.03-LTS 00/16] arm64: fix a concurrency issue in emulation_proc_handler()
by Jinjie Ruan 30 Oct '23

30 Oct '23

In linux-6.1, the related code is refactored in commit 124c49b1b5d9 ("arm64: armv8_deprecated: rework deprected instruction handling") and this issue was incidentally fixed. This patch set try to adapt the refactoring stable 5.10 patches to solve the problem of repeated addition of linked lists described below. How to reproduce: CONFIG_ARMV8_DEPRECATED=y, CONFIG_SWP_EMULATION=y, and CONFIG_DEBUG_LIST=y, then launch two shell executions: #!/bin/bash while [ 1 ]; do echo 1 > /proc/sys/abi/swp done or "echo 1 > /proc/sys/abi/swp" and then launch two shell executions: #!/bin/bash while [ 1 ]; do echo 0 > /proc/sys/abi/swp done In emulation_proc_handler(), read and write operations are performed on insn->current_mode. In the concurrency scenario, mutex only protects writing insn->current_mode, and not protects the read. Suppose there are two concurrent tasks, task1 updates insn->current_mode to INSN_EMULATE in the critical section, the prev_mode of task2 is still the old data INSN_UNDEF of insn->current_mode. As a result, two tasks call update_insn_emulation_mode twice with prev_mode = INSN_UNDEF and current_mode = INSN_EMULATE, then call register_emulation_hooks twice, resulting in a list_add double problem. commit 124c49b1b5d9 ("arm64: armv8_deprecated: rework deprected instruction handling") remove the dynamic registration and unregistration so remove the register_undef_hook() function, so the below problem was incidentally fixed. Call trace: __list_add_valid+0xd8/0xe4 register_undef_hook+0x94/0x13c update_insn_emulation_mode+0xd0/0x12c emulation_proc_handler+0xd8/0xf4 proc_sys_call_handler+0x140/0x250 proc_sys_write+0x1c/0x2c new_sync_write+0xec/0x18c vfs_write+0x214/0x2ac ksys_write+0x70/0xfc __arm64_sys_write+0x24/0x30 el0_svc_common.constprop.0+0x7c/0x1bc do_el0_svc+0x2c/0x94 el0_svc+0x20/0x30 el0_sync_handler+0xb0/0xb4 el0_sync+0x160/0x180 Call trace: __list_del_entry_valid+0xac/0x110 unregister_undef_hook+0x34/0x80 update_insn_emulation_mode+0xf0/0x180 emulation_proc_handler+0x8c/0xd8 proc_sys_call_handler+0x1d8/0x208 proc_sys_write+0x14/0x20 new_sync_write+0xf0/0x190 vfs_write+0x304/0x388 ksys_write+0x6c/0x100 __arm64_sys_write+0x1c/0x28 el0_svc_common.constprop.4+0x68/0x188 do_el0_svc+0x24/0xa0 el0_svc+0x14/0x20 el0_sync_handler+0x90/0xb8 el0_sync+0x160/0x180 The first patch is the self-developed patch, so revert it. The next 5 patches is a patch set which provides context for subsequent refactoring 9 patches, especially commit 0f2cb928a154 ("arm64: consistently pass ESR_ELx to die()") which modify do_undefinstr() to add a ESR_ELx value arg, and then commit 61d64a376ea8 ("arm64: split EL0/EL1 UNDEF handlers") splits do_undefinstr() handler into separate do_el0_undef() and do_el1_undef() handlers. The 9 patches after that is another refactoring patch set, which is in preparation for the main rework commit 124c49b1b5d9 ("arm64: armv8_deprecated: rework deprected instruction handling"). To remove struct undef_hook, commit bff8f413c71f ("arm64: factor out EL1 SSBS emulation hook") factor out EL1 SSBS emulation hook, which also avoid call call_undef_hook() in do_el1_undef(), commit f5962add74b6 ("arm64: rework EL0 MRS emulation") factor out EL0 MRS emulation hook, which also prepare for replacing call_undef_hook() in do_el0_undef(). To replace call_undef_hook() function, commit 61d64a376ea8 ("arm64: split EL0/EL1 UNDEF handlers") split the do_undefinstr() into do_el0_undef() and do_el1_undef() functions, and commit dbfbd87efa79 ("arm64: factor insn read out of call_undef_hook()") factor user_insn_read() from call_undef_hook() so the main rework patch can replace the call_undef_hook() in do_el0_undef(). The last patch is a bugfix for the main rework patch. Fixes: af483947d472 ("arm64: fix oops in concurrently setting insn_emulation sysctls") Signed-off-by: Jinjie Ruan <ruanjinjie(a)huawei.com> Jinjie Ruan (1): Revert "arm64: fix a concurrency issue in emulation_proc_handler()" Mark Rutland (14): arm64: report EL1 UNDEFs better arm64: die(): pass 'err' as long arm64: consistently pass ESR_ELx to die() arm64: rework FPAC exception handling arm64: rework BTI exception handling arm64: allow kprobes on EL0 handlers arm64: split EL0/EL1 UNDEF handlers arm64: factor out EL1 SSBS emulation hook arm64: factor insn read out of call_undef_hook() arm64: rework EL0 MRS emulation arm64: armv8_deprecated: fold ops into insn_emulation arm64: armv8_deprecated move emulation functions arm64: armv8_deprecated: move aarch32 helper earlier arm64: armv8_deprecated: rework deprected instruction handling Ren Zhijie (1): arm64: armv8_deprecated: fix unused-function error arch/arm64/include/asm/cpufeature.h | 3 +- arch/arm64/include/asm/exception.h | 13 +- arch/arm64/include/asm/spectre.h | 2 + arch/arm64/include/asm/system_misc.h | 2 +- arch/arm64/include/asm/traps.h | 19 +- arch/arm64/kernel/armv8_deprecated.c | 574 +++++++++++++-------------- arch/arm64/kernel/cpufeature.c | 23 +- arch/arm64/kernel/entry-common.c | 36 +- arch/arm64/kernel/proton-pack.c | 26 +- arch/arm64/kernel/traps.c | 125 +++--- 10 files changed, 396 insertions(+), 427 deletions(-) -- 2.34.1

2 17

[PATCH OLK-5.10 0/3] Synchronize mainline hisilicon uncore pmu driver bugfix to openEuler-OLK-5.10
by Junhao He 28 Oct '23

28 Oct '23

Synchronize mainline hisilicon uncore pmu driver bugfix to openEuler-OLK-5.10 This patchset includes 2 minor updates for the hisi_pcie_pmu: - fix issue that we may touch others events in some case - modify the event->cpu only on the success pmu::event_init() - dix use-after-free when register hisilicon ddrc pmu fails Junhao He (1): perf: hisi: Fix use-after-free when register pmu fails Yicong Yang (2): drivers/perf: hisi_pcie: Check the type first in pmu::event_init() drivers/perf: hisi_pcie: Initialize event->cpu only on success drivers/perf/hisilicon/hisi_pcie_pmu.c | 9 +++++---- drivers/perf/hisilicon/hisi_uncore_pa_pmu.c | 4 ++-- drivers/perf/hisilicon/hisi_uncore_sllc_pmu.c | 4 ++-- 3 files changed, 9 insertions(+), 8 deletions(-) -- 2.33.0

2 4

[PATCH OLK-5.10 0/7] coresight: trbe: Enable ACPI based devices
by Junhao He 28 Oct '23

28 Oct '23

This series enables detection of ACPI based TRBE devices via a stand alone purpose built representative platform device. But as a pre-requisite this changes coresight_platform_data structure assignment for the TRBE device. | commit from OLK-5.10 | patch | Conflict or Not | description | | --------------- | ------------ | --------------- | ------------ | | 003e440c95a0 | coresight: trbe: Enable ACPI based TRBE devices | No | | | c64a976a27cd | coresight: trbe: Add a representative coresight_platform_data for TRBE | No | | | eb630e5b2a47 | arm_pmu: acpi: Add a representative platform device for TRBE | Yes | Resolve code conflicts without changing functions | | a25b79f824ec | arm_pmu: acpi: Refactor arm_spe_acpi_register_device() | No | | | 54c422019d57 | Revert "coresight: Return the pointer of @pdata when not "fwnode"" | No | mainline`s patch "c64a976a27cd" fixes the issue | | 7e4acd18f0d0 | Revert "arm64/trbe: Add initial MADT/SPE probing" | No | mainline`s patch "eb630e5b2a47" adds ARM TRBE device support | | fbdfedf671b9 | Revert "coresight: trbe: Enable ACPI/Platform automatic module loading" | No | mainline`s patch "003e440c95a0" adds ARM TRBE ACPI support | Anshuman Khandual (4): arm_pmu: acpi: Refactor arm_spe_acpi_register_device() arm_pmu: acpi: Add a representative platform device for TRBE coresight: trbe: Add a representative coresight_platform_data for TRBE coresight: trbe: Enable ACPI based TRBE devices Junhao He (3): Revert "coresight: trbe: Enable ACPI/Platform automatic module loading" Revert "arm64/trbe: Add initial MADT/SPE probing" Revert "coresight: Return the pointer of @pdata when not "fwnode"" arch/arm64/configs/openeuler_defconfig | 1 - arch/arm64/kernel/Makefile | 1 - arch/arm64/kernel/acpi_trbe.c | 81 ----------- drivers/hwtracing/coresight/Kconfig | 4 - .../hwtracing/coresight/coresight-platform.c | 6 +- drivers/hwtracing/coresight/coresight-trbe.c | 26 +++- drivers/hwtracing/coresight/coresight-trbe.h | 2 + drivers/perf/arm_pmu_acpi.c | 137 +++++++++++++----- include/linux/coresight.h | 2 - include/linux/perf/arm_pmu.h | 1 + 10 files changed, 123 insertions(+), 138 deletions(-) delete mode 100644 arch/arm64/kernel/acpi_trbe.c -- 2.33.0

2 8

[PATCH OLK-5.10 v3 00/15] arm64: fix a concurrency issue in emulation_proc_handler()
by Jinjie Ruan 28 Oct '23

28 Oct '23

In linux-6.1, the related code is refactored in commit 124c49b1b5d9 ("arm64: armv8_deprecated: rework deprected instruction handling") and this issue was incidentally fixed. This patch set try to adapt the refactoring stable 5.10 patches to solve the problem of repeated addition of linked lists described below. How to reproduce: CONFIG_ARMV8_DEPRECATED=y, CONFIG_SWP_EMULATION=y, and CONFIG_DEBUG_LIST=y, then launch two shell executions: #!/bin/bash while [ 1 ]; do echo 1 > /proc/sys/abi/swp done or "echo 1 > /proc/sys/abi/swp" and then launch two shell executions: #!/bin/bash while [ 1 ]; do echo 0 > /proc/sys/abi/swp done In emulation_proc_handler(), read and write operations are performed on insn->current_mode. In the concurrency scenario, mutex only protects writing insn->current_mode, and not protects the read. Suppose there are two concurrent tasks, task1 updates insn->current_mode to INSN_EMULATE in the critical section, the prev_mode of task2 is still the old data INSN_UNDEF of insn->current_mode. As a result, two tasks call update_insn_emulation_mode twice with prev_mode = INSN_UNDEF and current_mode = INSN_EMULATE, then call register_emulation_hooks twice, resulting in a list_add double problem. commit 124c49b1b5d9 ("arm64: armv8_deprecated: rework deprected instruction handling") remove the dynamic registration and unregistration so remove the register_undef_hook() function, so the below problem was incidentally fixed. Call trace: __list_add_valid+0xd8/0xe4 register_undef_hook+0x94/0x13c update_insn_emulation_mode+0xd0/0x12c emulation_proc_handler+0xd8/0xf4 proc_sys_call_handler+0x140/0x250 proc_sys_write+0x1c/0x2c new_sync_write+0xec/0x18c vfs_write+0x214/0x2ac ksys_write+0x70/0xfc __arm64_sys_write+0x24/0x30 el0_svc_common.constprop.0+0x7c/0x1bc do_el0_svc+0x2c/0x94 el0_svc+0x20/0x30 el0_sync_handler+0xb0/0xb4 el0_sync+0x160/0x180 Call trace: __list_del_entry_valid+0xac/0x110 unregister_undef_hook+0x34/0x80 update_insn_emulation_mode+0xf0/0x180 emulation_proc_handler+0x8c/0xd8 proc_sys_call_handler+0x1d8/0x208 proc_sys_write+0x14/0x20 new_sync_write+0xf0/0x190 vfs_write+0x304/0x388 ksys_write+0x6c/0x100 __arm64_sys_write+0x1c/0x28 el0_svc_common.constprop.4+0x68/0x188 do_el0_svc+0x24/0xa0 el0_svc+0x14/0x20 el0_sync_handler+0x90/0xb8 el0_sync+0x160/0x180 The first patch is the self-developed patch, so revert it. The next 4 patches is a patch set which provides context for subsequent refactoring 9 patches, especially commit 0f2cb928a154 ("arm64: consistently pass ESR_ELx to die()") which modify do_undefinstr() to add a ESR_ELx value arg, and then commit 61d64a376ea8 ("arm64: split EL0/EL1 UNDEF handlers") splits do_undefinstr() handler into separate do_el0_undef() and do_el1_undef() handlers. The 9 patches after that is another refactoring patch set, which is in preparation for the main rework commit 124c49b1b5d9 ("arm64: armv8_deprecated: rework deprected instruction handling"). To remove struct undef_hook, commit bff8f413c71f ("arm64: factor out EL1 SSBS emulation hook") factor out EL1 SSBS emulation hook, which also avoid call call_undef_hook() in do_el1_undef(), commit f5962add74b6 ("arm64: rework EL0 MRS emulation") factor out EL0 MRS emulation hook, which also prepare for replacing call_undef_hook() in do_el0_undef(). To replace call_undef_hook() function, commit 61d64a376ea8 ("arm64: split EL0/EL1 UNDEF handlers") split the do_undefinstr() into do_el0_undef() and do_el1_undef() functions, and commit dbfbd87efa79 ("arm64: factor insn read out of call_undef_hook()") factor user_insn_read() from call_undef_hook() so the main rework patch can replace the call_undef_hook() in do_el0_undef(). The last patch is a bugfix for the main rework patch. I've tested this with userspace programs which use each of the deprecated instructions on Raspberry Pi 4B KVM/Qemu, and I've concurrently modified the support level for each of the features back-and-forth between HW and emulated to check that there are no oops or above repeated addition or deletion call trace. Changes in v3: - Add conflicts message. Fixes: af483947d472 ("arm64: fix oops in concurrently setting insn_emulation sysctls") Signed-off-by: Jinjie Ruan <ruanjinjie(a)huawei.com> Jinjie Ruan (1): Revert "arm64: fix a concurrency issue in emulation_proc_handler()" Mark Rutland (13): arm64: die(): pass 'err' as long arm64: consistently pass ESR_ELx to die() arm64: rework FPAC exception handling arm64: rework BTI exception handling arm64: allow kprobes on EL0 handlers arm64: split EL0/EL1 UNDEF handlers arm64: factor out EL1 SSBS emulation hook arm64: factor insn read out of call_undef_hook() arm64: rework EL0 MRS emulation arm64: armv8_deprecated: fold ops into insn_emulation arm64: armv8_deprecated move emulation functions arm64: armv8_deprecated: move aarch32 helper earlier arm64: armv8_deprecated: rework deprected instruction handling Ren Zhijie (1): arm64: armv8_deprecated: fix unused-function error arch/arm64/include/asm/cpufeature.h | 3 +- arch/arm64/include/asm/exception.h | 13 +- arch/arm64/include/asm/spectre.h | 2 + arch/arm64/include/asm/system_misc.h | 2 +- arch/arm64/include/asm/traps.h | 19 +- arch/arm64/kernel/armv8_deprecated.c | 574 +++++++++++++-------------- arch/arm64/kernel/cpufeature.c | 23 +- arch/arm64/kernel/entry-common.c | 36 +- arch/arm64/kernel/proton-pack.c | 26 +- arch/arm64/kernel/traps.c | 125 +++--- 10 files changed, 395 insertions(+), 428 deletions(-) -- 2.34.1

2 16

[PATCH openEuler-1.0-LTS] sched/membarrier: fix missing local execution of ipi_sync_rq_state()
by Xia Fukun 28 Oct '23

28 Oct '23

From: Mathieu Desnoyers <mathieu.desnoyers(a)efficios.com> mainline inclusion from mainline-v6.5-rc7 commit ce29ddc47b91f97e7f69a0fb7cbb5845f52a9825 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8BN94 CVE: NA -------------------------------- The function sync_runqueues_membarrier_state() should copy the membarrier state from the @mm received as parameter to each runqueue currently running tasks using that mm. However, the use of smp_call_function_many() skips the current runqueue, which is unintended. Replace by a call to on_each_cpu_mask(). Fixes: 227a4aadc75b ("sched/membarrier: Fix p->mm->membarrier_state racy load") Reported-by: Nadav Amit <nadav.amit(a)gmail.com> Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers(a)efficios.com> Signed-off-by: Peter Zijlstra (Intel) <peterz(a)infradead.org> Signed-off-by: Ingo Molnar <mingo(a)kernel.org> Cc: stable(a)vger.kernel.org # 5.4.x+ Link: https://lore.kernel.org/r/74F1E842-4A84-47BF-B6C2-5407DFDD4A4A@gmail.com Signed-off-by: Xia Fukun <xiafukun(a)huawei.com> --- kernel/sched/membarrier.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) diff --git a/kernel/sched/membarrier.c b/kernel/sched/membarrier.c index 8c4e14e6544a..1757074be994 100644 --- a/kernel/sched/membarrier.c +++ b/kernel/sched/membarrier.c @@ -255,9 +255,7 @@ static int sync_runqueues_membarrier_state(struct mm_struct *mm) } rcu_read_unlock(); - preempt_disable(); - smp_call_function_many(tmpmask, ipi_sync_rq_state, mm, 1); - preempt_enable(); + on_each_cpu_mask(tmpmask, ipi_sync_rq_state, mm, true); free_cpumask_var(tmpmask); cpus_read_unlock(); -- 2.34.1

2 1

[PATCH openEuler-1.0-LTS] sched/cpuacct: Fix charge cpuacct.usage_sys
by Xia Fukun 28 Oct '23

28 Oct '23

From: Muchun Song <songmuchun(a)bytedance.com> mainline inclusion from mainline-v6.5-rc7 commit dbe9337109c2705f08e6a00392f991eb2d2570a5 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8BN4B CVE: NA -------------------------------- The user_mode(task_pt_regs(tsk)) always return true for user thread, and false for kernel thread. So it means that the cpuacct.usage_sys is the time that kernel thread uses not the time that thread uses in the kernel mode. We can try get_irq_regs() first, if it is NULL, then we can fall back to task_pt_regs(). Signed-off-by: Muchun Song <songmuchun(a)bytedance.com> Signed-off-by: Peter Zijlstra (Intel) <peterz(a)infradead.org> Link: https://lkml.kernel.org/r/20200420070453.76815-1-songmuchun@bytedance.com Signed-off-by: Xia Fukun <xiafukun(a)huawei.com> --- kernel/sched/cpuacct.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/kernel/sched/cpuacct.c b/kernel/sched/cpuacct.c index 9fbb10383434..254a60c99435 100644 --- a/kernel/sched/cpuacct.c +++ b/kernel/sched/cpuacct.c @@ -5,6 +5,7 @@ * Based on the work by Paul Menage (menage(a)google.com) and Balbir Singh * (balbir(a)in.ibm.com) */ +#include <asm/irq_regs.h> #include "sched.h" /* Time spent by the tasks of the CPU accounting group executing in ... */ @@ -339,7 +340,7 @@ void cpuacct_charge(struct task_struct *tsk, u64 cputime) { struct cpuacct *ca; int index = CPUACCT_STAT_SYSTEM; - struct pt_regs *regs = task_pt_regs(tsk); + struct pt_regs *regs = get_irq_regs() ? : task_pt_regs(tsk); if (regs && user_mode(regs)) index = CPUACCT_STAT_USER; -- 2.34.1

2 1

[PATCH openEuler-1.0-LTS] kernel/trace: Fix do not unregister tracepoints when register sched_migrate_task fail
by Wenyu Huang 28 Oct '23

28 Oct '23

From: Kaitao Cheng <pilgrimtao(a)gmail.com> stable inclusion from stable-4.19 commit 50f9ad607ea891a9308e67b81f774c71736d1098 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8BAW8 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- In the function, if register_trace_sched_migrate_task() returns error, sched_switch/sched_wakeup_new/sched_wakeup won't unregister. That is why fail_deprobe_sched_switch was added. Link: http://lkml.kernel.org/r/20191231133530.2794-1-pilgrimtao@gmail.com Cc: stable(a)vger.kernel.org Fixes: 478142c39c8c2 ("tracing: do not grab lock in wakeup latency function tracing") Signed-off-by: Kaitao Cheng <pilgrimtao(a)gmail.com> Signed-off-by: Steven Rostedt (VMware) <rostedt(a)goodmis.org> Signed-off-by: Wenyu Huang <huangwenyu4(a)huawei.com> --- kernel/trace/trace_sched_wakeup.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index 0db1b024e786..8041bd5e4262 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -642,7 +642,7 @@ static void start_wakeup_tracer(struct trace_array *tr) if (ret) { pr_info("wakeup trace: Couldn't activate tracepoint" " probe to kernel_sched_migrate_task\n"); - return; + goto fail_deprobe_sched_switch; } wakeup_reset(tr); @@ -660,6 +660,8 @@ static void start_wakeup_tracer(struct trace_array *tr) printk(KERN_ERR "failed to start wakeup tracer\n"); return; +fail_deprobe_sched_switch: + unregister_trace_sched_switch(probe_wakeup_sched_switch, NULL); fail_deprobe_wake_new: unregister_trace_sched_wakeup_new(probe_wakeup, NULL); fail_deprobe: -- 2.34.1

2 1

[PATCH OLK-5.10 v5 0/6] Synchronize coresight driver bugfix patches to openEuler
by Junhao He 28 Oct '23

28 Oct '23

1. Fix memory leak coresight: Fix memory leak in acpi_buffer->pointer 2. Fix TRBE bug coresight: trbe: Fix TRBE potential sleep in atomic context coresight: trbe: Allocate platform data per device coresight: trbe: Fix return value check in arm_trbe_register_coresight_cpu() 3. Fix ETR bug coresight: Fix run time warnings while reusing ETR buffer coresight: tmc-etr: Disable warnings for allocation failures | mainline commit | patch | Conflict or Not | description | | --------------- | ------------ | --------------- | ------------ | | 34149bd82c35 | coresight: Fix memory leak in acpi_buffer->pointer | No | | | 2e0fdff6f374 | coresight: Fix run time warnings while reusing ETR buffer | Yes | Resolve code conflicts without changing functions | | d3ce83b13fbd | coresight: trbe: Fix TRBE potential sleep in atomic context | Yes | Resolve code conflicts without changing functions | | 91ee6ec9039e | coresight: Fix memory leak in acpi_buffer->pointer | Yes | Dependency cleanup patch-set does not applied to OLK-5.10 | | e911e9760c77 | coresight: trbe: Allocate platform data per device | No | depend on ��c7e95621feea�� | | c7e95621feea | coresight: trbe: Fix return value check in arm_trbe_register_coresight_cpu() | No | | Junhao He (2) coresight: Fix memory leak in acpi_buffer->pointer coresight: trbe: Fix TRBE potential sleep in atomic context Linu Cherian (1): coresight: Fix run time warnings while reusing ETR buffer Suzuki K Poulose (2): coresight: trbe: Allocate platform data per device coresight: tmc-etr: Disable warnings for allocation failures Wei Yongjun (1): coresight: trbe: Fix return value check in arm_trbe_register_coresight_cpu() .../hwtracing/coresight/coresight-platform.c | 46 ++++++++++++------- .../hwtracing/coresight/coresight-tmc-etr.c | 23 +++++----- drivers/hwtracing/coresight/coresight-trbe.c | 41 ++++++++--------- 3 files changed, 62 insertions(+), 48 deletions(-) -- 2.33.0

2 7

[PATCH OLK-5.10 0/2] Fix CVE-2023-5717
by Yang Jihong 28 Oct '23

28 Oct '23

Peter Zijlstra (1): perf: Disallow mis-matched inherited group reads Yang Jihong (1): perf: Fix kabi breakage in struct perf_event include/linux/perf_event.h | 2 ++ kernel/events/core.c | 40 ++++++++++++++++++++++++++++++++------ 2 files changed, 36 insertions(+), 6 deletions(-) -- 2.34.1

2 3

[PATCH v2 openEuler-1.0-LTS] kernel/trace: Fix do not unregister tracepoints when register sched_migrate_task fail
by Wenyu Huang 27 Oct '23

27 Oct '23

From: Kaitao Cheng <pilgrimtao(a)gmail.com> stable inclusion from stable-v4.19.297 commit 50f9ad607ea891a9308e67b81f774c71736d1098 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I8BAW8 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- In the function, if register_trace_sched_migrate_task() returns error, sched_switch/sched_wakeup_new/sched_wakeup won't unregister. That is why fail_deprobe_sched_switch was added. Link: http://lkml.kernel.org/r/20191231133530.2794-1-pilgrimtao@gmail.com Cc: stable(a)vger.kernel.org Fixes: 478142c39c8c2 ("tracing: do not grab lock in wakeup latency function tracing") Signed-off-by: Kaitao Cheng <pilgrimtao(a)gmail.com> Signed-off-by: Steven Rostedt (VMware) <rostedt(a)goodmis.org> Signed-off-by: Wenyu Huang <huangwenyu4(a)huawei.com> --- kernel/trace/trace_sched_wakeup.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index 0db1b024e786..8041bd5e4262 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -642,7 +642,7 @@ static void start_wakeup_tracer(struct trace_array *tr) if (ret) { pr_info("wakeup trace: Couldn't activate tracepoint" " probe to kernel_sched_migrate_task\n"); - return; + goto fail_deprobe_sched_switch; } wakeup_reset(tr); @@ -660,6 +660,8 @@ static void start_wakeup_tracer(struct trace_array *tr) printk(KERN_ERR "failed to start wakeup tracer\n"); return; +fail_deprobe_sched_switch: + unregister_trace_sched_switch(probe_wakeup_sched_switch, NULL); fail_deprobe_wake_new: unregister_trace_sched_wakeup_new(probe_wakeup, NULL); fail_deprobe: -- 2.34.1

2 1