- 30 5月, 2023 6 次提交
-
-
由 Peter Zijlstra 提交于
mainline inclusion from mainline-v5.13-rc1 commit e40f74c5 category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I6VS35 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=e40f74c535b8a0ecf3ef0388b51a34cdadb34fb5 -------------------------------- Introduce a cpumask that indicates (for each CPU) what direction the CPU hotplug is currently going. Notably, it tracks rollbacks. Eg. when an up fails and we do a roll-back down, it will accurately reflect the direction. Signed-off-by: NPeter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: NValentin Schneider <valentin.schneider@arm.com> Link: https://lkml.kernel.org/r/20210310150109.151441252@infradead.orgSigned-off-by: NZeng Heng <zengheng4@huawei.com>
-
由 openeuler-ci-bot 提交于
Merge Pull Request from: @hejunhao3 Some HiSilicon SMMU PMCG suffers the erratum that the global PMU disable control sometimes fail to disable each used the counters. This will lead to error or inaccurate data since before we enable the counters the counter's still counting for the event used in last perf session. This patch tries to fix this by hardening the global disable process. Before disable the PMU, writing an invalid event type (0xff) to focibly stop the counters. Link:https://gitee.com/openeuler/kernel/pulls/851 Reviewed-by: Yang Shen <shenyang39@huawei.com> Signed-off-by: Zheng Zengkai <zhengzengkai@huawei.com>
-
由 openeuler-ci-bot 提交于
Merge Pull Request from: @Hongchen_Zhang PV IPI with less VM exit can improve performance than iocsr emulation when sending ipi interrupt to multi cpus. This patch supports: 1. sending PV IPI by hypercall 2. recording and updating steal time for VM Link:https://gitee.com/openeuler/kernel/pulls/793 Reviewed-by: Guo Dongtai <guodongtai@kylinos.cn> Reviewed-by: Kevin Zhu <zhukeqian1@huawei.com> Signed-off-by: Zheng Zengkai <zhengzengkai@huawei.com>
-
由 openeuler-ci-bot 提交于
Merge Pull Request from: @jiayingbao there is new sysfs interface current_freq_khz added in upstream. Backport related patches for uncore-freq driver. below 3 patches is backported: ae7b2ce5 platform/x86/intel/uncore-freq: Use sysfs API to create. 414eef27 platform/x86/intel/uncore-freq: Display uncore current frequency. 8d75f7b4 platform/x86: intel-uncore-freq: Prevent driver loading in guests. test: build pass new sysfs current_freq_khz functional as expected Link:https://gitee.com/openeuler/kernel/pulls/840 Reviewed-by: Jason Zeng <jason.zeng@intel.com> Reviewed-by: Aichun Shi <aichun.shi@intel.com> Signed-off-by: Zheng Zengkai <zhengzengkai@huawei.com>
-
由 openeuler-ci-bot 提交于
Merge Pull Request from: @jiayingbao backport some patches for Intel_pstate no-HWP and OOB mode, and also one HWP update for all server platforms including below commit: cd23f02f cpufreq: intel_pstate: Add Ice Lake server to out-of-band IDs addca285 cpufreq: intel_pstate: Handle no_turbo in frequency invariance. bbd67f1b cpufreq: intel_pstate: Support Sapphire Rapids OOB mode. df51f287 cpufreq: intel_pstate: Add Sapphire Rapids support in no-HWP mode. 1f5e62f5 cpufreq: intel_pstate: Enable HWP IO boost for all servers. test: build pass intel_pstate initial and functional as expected Link:https://gitee.com/openeuler/kernel/pulls/839 Reviewed-by: Jason Zeng <jason.zeng@intel.com> Reviewed-by: Aichun Shi <aichun.shi@intel.com> Signed-off-by: Zheng Zengkai <zhengzengkai@huawei.com>
-
由 openeuler-ci-bot 提交于
Merge Pull Request from: @Hongchen_Zhang enable memory and pci hotplug related configs for LoongArch, used by LoongArch virtual machine now. Link:https://gitee.com/openeuler/kernel/pulls/809 Reviewed-by: Guo Dongtai <guodongtai@kylinos.cn> Reviewed-by: Liu Chao <liuchao173@huawei.com> Signed-off-by: Zheng Zengkai <zhengzengkai@huawei.com>
-
- 29 5月, 2023 9 次提交
-
-
由 openeuler-ci-bot 提交于
Merge Pull Request from: @leoliu-oc Recent Zhaoxin/Centaur CPUs support X86_FEATURE_IDA and the turbo boost can be dynamically enabled or disabled through MSR 0x1a0[38] in the same way as Intel. So add turbo boost control support for these CPUs too. Issue https://gitee.com/openeuler/kernel/issues/I6SKVN Link:https://gitee.com/openeuler/kernel/pulls/547 Reviewed-by: Zheng Zengkai <zhengzengkai@huawei.com> Signed-off-by: Zheng Zengkai <zhengzengkai@huawei.com>
-
由 openeuler-ci-bot 提交于
Merge Pull Request from: @leoliu-oc Zhaoxin CPUs which use CENTAUR as vendor id, also have NONSTOP TSC feature. So enable the ACPI driver support for it too. Issue https://gitee.com/openeuler/kernel/issues/I6SJQ8 Link:https://gitee.com/openeuler/kernel/pulls/544 Reviewed-by: Zheng Zengkai <zhengzengkai@huawei.com> Signed-off-by: Zheng Zengkai <zhengzengkai@huawei.com>
-
由 openeuler-ci-bot 提交于
Merge Pull Request from: @hejunhao3 1. Added support for HiSilicon T6 ETM 2. Fix CPU hold issue caused by hip09 ETM overflow Link:https://gitee.com/openeuler/kernel/pulls/848 Signed-off-by: Zheng Zengkai <zhengzengkai@huawei.com>
-
由 Yicong Yang 提交于
driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I798Y2 CVE: NA ---------------------------------------------------------------------- Some HiSilicon SMMU PMCG suffers the erratum that the global PMU disable control sometimes fail to disable each used the counters. This will lead to error or inaccurate data since before we enable the counters the counter's still counting for the event used in last perf session. This patch tries to fix this by hardening the global disable process. Before disable the PMU, writing an invalid event type (0xff) to focibly stop the counters. Signed-off-by: NYicong Yang <yangyicong@hisilicon.com> Signed-off-by: NJunhao He <hejunhao3@huawei.com>
-
由 Junhao He 提交于
driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I79882 CVE: NA ---------------------------------------------------------------------- When setting up SMT, not every process has an ETM, so the path ".../cs_etm/cpux/trcidr/trcidr0" does not exist, and the function perf_pmu__scan_file() will return an error. Log a error when read fails. Signed-off-by: NJunhao He <hejunhao3@huawei.com>
-
由 Junhao He 提交于
driver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I79882 CVE: NA ---------------------------------------------------------------------- Workaround this issue on HiSilicon ETM by setting bit 13 of TRCAUXCTLR which is used to indicate that the ETM is in the idle state. Signed-off-by: NJunhao He <hejunhao3@huawei.com>
-
由 Junhao He 提交于
driver inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I79882 CVE: NA ---------------------------------------------------------------------- Add ETMv4 periperhal ID for HiSilicon T6 platform. Signed-off-by: NJunhao He <hejunhao3@huawei.com>
-
由 leoliuoc 提交于
mainline inclusion commit d6f89596 category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I6SKVN CVE: NA -------------------------------------------- Recent Zhaoxin/Centaur CPUs support X86_FEATURE_IDA and the turbo boost can be dynamically enabled or disabled through MSR 0x1a0[38] in the same way as Intel. So add turbo boost control support for these CPUs too. Signed-off-by: NTony W Wang-oc <TonyWWang-oc@zhaoxin.com> Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: NLeoLiu-oc <leoliu-oc@zhaoxin.com>
-
由 leoliuoc 提交于
zhaoxin inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I6SJQ8 CVE: NA ---------------------------------------- Zhaoxin CPUs which use CENTAUR as vendor id, also have NONSTOP TSC feature. So enable the ACPI driver support for it too. Signed-off-by: Nleoliuoc <leoliu-oc@zhaoxin.com>
-
- 26 5月, 2023 12 次提交
-
-
由 openeuler-ci-bot 提交于
Merge Pull Request from: @aspiresky01 The NIC driver supports the following features: Supports IPv4/IPv6 TCP/UDP checksum offload, TSO (TCP Segmentation Offload), LRO (Large Receive Offload) offload and RSS (Receive Side Scaling) functions Supports interrupt aggregation parameter configuration and interrupt adaptation. Supports 802.1Q VLAN (Virtual Local Area Network) offloading and filtering. Supports NIC SR-IOV (Single Root I/O Virtualization). Support PF promiscuous mode, unicast list filtering, multicast list filtering, and full multicast mode. Support VF unicast list filtering, multicast list filtering, and full multicast mode. Supports VF QinQ mode. Supports VF link state configuration and QoS configuration. Support VF MAC address management. Support VF spoofchk check. Loopback testing is supported. Support port lighting. Support Ethernet mouth self-negotiation, support pause frame. Link:https://gitee.com/openeuler/kernel/pulls/835 Reviewed-by: Zheng Zengkai <zhengzengkai@huawei.com> Reviewed-by: Chiqijun <chiqijun@huawei.com> Signed-off-by: Zheng Zengkai <zhengzengkai@huawei.com>
-
由 yangyinglu 提交于
LoongArch inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I77Q98 -------------------------------- Signed-off-by: Nlixianglai <lixianglai@loongson.cn> Signed-off-by: Nyangyinglu <yangyinglu@loongson.cn> Change-Id: I6b4ca700f830c05c2cf62695c22bb110dd6f017b
-
由 yangyinglu 提交于
LoongArch inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I77Q98 -------------------------------- Signed-off-by: Nlixianglai <lixianglai@loongson.cn> Signed-off-by: Nyangyinglu <yangyinglu@loongson.cn> Change-Id: I640f550277271f97317e084aecb03c9a5da3cd01
-
由 Srinivas Pandruvada 提交于
mainline inclusion from mainline-v5.18-rc5 commit 8d75f7b4 category: feature bugzilla: https://gitee.com/openeuler/intel-kernel/issues/I78KDD CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=8d75f7b4a3dfd5714a5dc87cfdaa27bd2d14aa48 Intel-SIG: commit 8d75f7b4 platform/x86: intel-uncore-freq: Prevent driver loading in guests. Backport for intel-uncore-freqe driver ------------------------------------------------- Loading this driver in guests results in unchecked MSR access error for MSR 0x620. There is no use of reading and modifying package/die scope uncore MSRs in guests. So check for CPU feature X86_FEATURE_HYPERVISOR to prevent loading of this driver in guests. Fixes: dbce412a ("platform/x86/intel-uncore-freq: Split common and enumeration part") Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=215870Suggested-by: NBorislav Petkov <bp@suse.de> Signed-off-by: NSrinivas Pandruvada <srinivas.pandruvada@linux.intel.com> Link: https://lore.kernel.org/r/20220427100304.2562990-1-srinivas.pandruvada@linux.intel.comSigned-off-by: NHans de Goede <hdegoede@redhat.com> Signed-off-by: Nyingbao jia <yingbao.jia@intel.com>
-
由 Srinivas Pandruvada 提交于
mainline inclusion from mainline-v5.18-rc1 commit 414eef27 category: feature bugzilla: https://gitee.com/openeuler/intel-kernel/issues/I78KDD CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=414eef27283a2133a0998beb8e3e35f84eaac961 Intel-SIG: commit 414eef27 platform/x86/intel/uncore-freq: Display uncore current frequency. Backport for intel-uncore-freqe driver ------------------------------------------------- Add a new sysfs attribute "current_freq_khz" to display current uncore frequency. This value is read from MSR 0x621. Root user permission is required to read uncore current frequency. Signed-off-by: NSrinivas Pandruvada <srinivas.pandruvada@intel.com> Acked-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com> Link: https://lore.kernel.org/r/20220204000306.2517447-4-srinivas.pandruvada@linux.intel.comSigned-off-by: NHans de Goede <hdegoede@redhat.com> Signed-off-by: Nyingbao jia <yingbao.jia@intel.com>
-
由 Srinivas Pandruvada 提交于
mainline inclusion from mainline-v5.18-rc1 commit ae7b2ce5 category: feature bugzilla: https://gitee.com/openeuler/intel-kernel/issues/I78KDD CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=ae7b2ce578513adf0a77f0127e27ee4447d50443 Intel-SIG: commit ae7b2ce5 platform/x86/intel/uncore-freq: Use sysfs API to create attributes. Backport for intel-uncore-freqe driver ------------------------------------------------- Use of sysfs API is always preferable over using kobject calls to create attributes. Remove usage of kobject_init_and_add() and use sysfs_create_group(). To create relationship between sysfs attribute and uncore instance use device_attribute*, which is defined per uncore instance. To create uniform locking for both read and write attributes take lock in the sysfs callbacks, not in the actual functions where the MSRs are read or updated. No functional changes are expected. Signed-off-by: NSrinivas Pandruvada <srinivas.pandruvada@intel.com> Acked-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com> Link: https://lore.kernel.org/r/20220204000306.2517447-3-srinivas.pandruvada@linux.intel.comSigned-off-by: NHans de Goede <hdegoede@redhat.com> Signed-off-by: Nyingbao jia <yingbao.jia@intel.com>
-
由 Srinivas Pandruvada 提交于
mainline inclusion from mainline-v6.4-rc1 commit 1f5e62f5 category: feature bugzilla: https://gitee.com/openeuler/intel-kernel/issues/I78I87 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=1f5e62f5fb217f2c1e003236be7d03cf606c26c4 Intel-SIG: commit 1f5e62f5 cpufreq: intel_pstate: Enable HWP IO boost for all servers. Backport for intel-pstate driver ------------------------------------------------- The HWP IO boost results in slight improvements for IO performance on both Ice Lake and Sapphire Rapid servers. Currently there is a CPU model check for Skylake desktop and server along with the ACPI PM profile for performance and enterprise servers to enable IO boost. Remove the CPU model check, so that all current server models enable HWP IO boost by default. Signed-off-by: NSrinivas Pandruvada <srinivas.pandruvada@linux.intel.com> Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Nyingbao jia <yingbao.jia@intel.com>
-
由 Giovanni Gherdovich 提交于
mainline inclusion from mainline-v6.2-rc1 commit df51f287 category: feature bugzilla: https://gitee.com/openeuler/intel-kernel/issues/I78I87 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=df51f287b5de3b9d4fd39593eafd1f8298d711c7 Intel-SIG: commit df51f287 cpufreq: intel_pstate: Add Sapphire Rapids support in no-HWP mode. Backport for intel-pstate driver ------------------------------------------------- Users may disable HWP in firmware, in which case intel_pstate wouldn't load unless the CPU model is explicitly supported. See also the following past commits: commit d8de7a44 ("cpufreq: intel_pstate: Add Skylake servers support") commit 706c5328 ("cpufreq: intel_pstate: Add Cometlake support in no-HWP mode") commit fbdc21e9 ("cpufreq: intel_pstate: Add Icelake servers support in no-HWP mode") commit 71bb5c82 ("cpufreq: intel_pstate: Add Tigerlake support in no-HWP mode") Signed-off-by: NGiovanni Gherdovich <ggherdovich@suse.cz> Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Nyingbao jia <yingbao.jia@intel.com>
-
由 Srinivas Pandruvada 提交于
mainline inclusion from mainline-v5.19-rc1 commit bbd67f1b category: feature bugzilla: https://gitee.com/openeuler/intel-kernel/issues/I78I87 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=bbd67f1b5a9449b090560ca5288aef0cfe623275 Intel-SIG: commit bbd67f1b cpufreq: intel_pstate: Support Sapphire Rapids OOB mode. Backport for intel-pstate driver ------------------------------------------------- Prevent intel_pstate to load when OOB (Out Of Band) P-states mode is enabled in Sapphire Rapids. The OOB identifying bits are same as the prior generation CPUs like Ice Lake servers. So, also add Sapphire Rapids to intel_pstate_cpu_oob_ids list. Signed-off-by: NSrinivas Pandruvada <srinivas.pandruvada@linux.intel.com> Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Nyingbao jia <yingbao.jia@intel.com>
-
由 Chen Yu 提交于
mainline inclusion from mainline-v5.19-rc1 commit addca285 category: bugfix bugzilla: https://gitee.com/openeuler/intel-kernel/issues/I78I87 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=addca285120b0edf2fef795f7809c83774cf74b7 Intel-SIG: commit addca285 cpufreq: intel_pstate: Handle no_turbo in frequency invariance. Backport for intel-pstate driver ------------------------------------------------- Problem statement: Once the user has disabled turbo frequency by the cfs_rq's util_avg becomes quite small when compared with CPU capacity. Step to reproduce: would launch 1 thread and bind it to CPU3, lasting for 100 seconds, with a CPU utilization of 99%. [1] top result: %Cpu3 : 98.4 us, 0.0 sy, 0.0 ni, 1.6 id, 0.0 wa, 0.0 hi, 0.0 si, 0.0 st check util_avg: cat /sys/kernel/debug/sched/debug | grep "cfs_rq\[3\]" -A 20 | grep util_avg .util_avg : 611 So the util_avg/cpu capacity is 611/1024, which is much smaller than 98.4% shown in the top result. This might impact some logic in the scheduler. For example, group_is_overloaded() would compare the group_capacity and group_util in the sched group, to check if this sched group is overloaded or not. With this gap, even when there is a nearly 100% workload, the sched group will not be regarded as overloaded. Besides group_is_overloaded(), there are also other victims. There is a ongoing work that aims to optimize the task wakeup in a LLC domain. The main idea is to stop searching idle CPUs if the sched domain is overloaded[2]. This proposal also relies on the util_avg/CPU capacity to decide whether the LLC domain is overloaded. Analysis: CPU frequency invariance has caused this difference. In summary, the util_sum of cfs rq would decay quite fast when the CPU is in idle, when the CPU frequency invariance is enabled. The detail is as followed: As depicted in update_rq_clock_pelt(), when the frequency invariance is enabled, there would be two clock variables on each rq, clock_task and clock_pelt: The clock_pelt scales the time to reflect the effective amount of computation done during the running delta time but then syncs back to clock_task when rq is idle. absolute time | 1| 2| 3| 4| 5| 6| 7| 8| 9|10|11|12|13|14|15|16 @ max frequency ------******---------------******--------------- @ half frequency ------************---------************--------- clock pelt | 1| 2| 3| 4| 7| 8| 9| 10| 11|14|15|16 The fast decay of util_sum during idle is due to: 1. rq->clock_pelt is always behind rq->clock_task 2. rq->last_update is updated to rq->clock_pelt' after invoking ___update_load_sum() 3. Then the CPU becomes idle, the rq->clock_pelt' would be suddenly increased a lot to rq->clock_task 4. Enters ___update_load_sum() again, the idle period is calculated by rq->clock_task - rq->last_update, AKA, rq->clock_task - rq->clock_pelt'. The lower the CPU frequency is, the larger the delta = rq->clock_task - rq->clock_pelt' will be. Since the idle period will be used to decay the util_sum only, the util_sum drops significantly during idle period. Proposal: This symptom is not only caused by disabling turbo frequency, but it would also appear if the user limits the max frequency at runtime. Because, if the frequency is always lower than the max frequency, CPU frequency invariance would decay the util_sum quite fast during idle. As some end users would disable turbo after boot up, this patch aims to present this symptom and deals with turbo scenarios for now. It might be ideal if CPU frequency invariance is aware of the max CPU frequency (user specified) at runtime in the future. Link: https://github.com/yu-chen-surf/x86_cpuload.git #1 Link: https://lore.kernel.org/lkml/20220310005228.11737-1-yu.c.chen@intel.com/ #2 Signed-off-by: NChen Yu <yu.c.chen@intel.com> Acked-by: NPeter Zijlstra (Intel) <peterz@infradead.org> Reviewed-by: NGiovanni Gherdovich <ggherdovich@suse.cz> Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Nyingbao jia <yingbao.jia@intel.com>
-
由 zhoujiadong 提交于
driver inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I6V4RY CVE: NA --------------------------------- The NIC driver supports the following features: Supports IPv4/IPv6 TCP/UDP checksum, TSO, LRO offload and RSS functions. Supports interrupt aggregation parameter configuration and interrupt adaptation. Supports 802.1Q VLAN (Virtual Local Area Network) offloading and filtering. Supports NIC SR-IOV (Single Root I/O Virtualization). Support PF promiscuous mode Supports VF QinQ mode. Supports VF link state configuration and QoS configuration. Support VF MAC address management. Support VF spoofchk check. Support port lighting. Support Ethernet mouth self-negotiation, support pause frame. Signed-off-by: Nzhoujiadong <zhoujiadong5@huawei.com> Reviewed-by: NWulike (Collin) <wulike1@huawei.com>
-
由 Adamos Ttofari 提交于
mainline inclusion from mainline-v5.16 commit cd23f02f category: feature bugzilla: https://gitee.com/openeuler/intel-kernel/issues/I78I87 CVE: NA Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=d23f02f166892603eb9f2d488152b975872b682 Intel-SIG: commit cd23f02f cpufreq: intel_pstate: Add Ice Lake server to out-of-band IDs. Backport for intel-pstate driver ------------------------------------------------- Commit fbdc21e9 ("cpufreq: intel_pstate: Add Icelake servers support in no-HWP mode") enabled the use of Intel P-State driver for Ice Lake servers. But it doesn't cover the case when OS can't control P-States. Therefore, for Ice Lake server, if MSR_MISC_PWR_MGMT bits 8 or 18 are enabled, then the Intel P-State driver should exit as OS can't control P-States. Fixes: fbdc21e9 ("cpufreq: intel_pstate: Add Icelake servers support in no-HWP mode") Signed-off-by: NAdamos Ttofari <attofari@amazon.de> Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Nyingbao jia <yingbao.jia@intel.com>
-
- 25 5月, 2023 5 次提交
-
-
由 openeuler-ci-bot 提交于
Merge Pull Request from: @remimin https://gitee.com/openeuler/kernel/issues/I6XQ34 Link:https://gitee.com/openeuler/kernel/pulls/601 Reviewed-by: Jialin Zhang <zhangjialin11@huawei.com> Reviewed-by: Kevin Zhu <zhukeqian1@huawei.com> Signed-off-by: Jialin Zhang <zhangjialin11@huawei.com>
-
由 openeuler-ci-bot 提交于
Merge Pull Request from: @remimin https://gitee.com/openeuler/kernel/issues/I72190 Link:https://gitee.com/openeuler/kernel/pulls/739 Reviewed-by: Jialin Zhang <zhangjialin11@huawei.com> Reviewed-by: Kevin Zhu <zhukeqian1@huawei.com> Reviewed-by: sanglipeng <sanglipeng1@jd.com> Signed-off-by: Jialin Zhang <zhangjialin11@huawei.com>
-
由 openeuler-ci-bot 提交于
Merge Pull Request from: @lyyue233 issue: [#I72XT8](https://gitee.com/openeuler/kernel/issues/I72XT8) Uninitialized variables can cause various errors in a program, such as unexpected behavior, segmentation faults, or even security vulnerabilities. To avoid these errors, we should always initialize variables before using them.Initialize variable “ err ” to address potential issues. Link:https://gitee.com/openeuler/kernel/pulls/767 Reviewed-by: Kevin Zhu <zhukeqian1@huawei.com> Signed-off-by: Zheng Zengkai <zhengzengkai@huawei.com> Acked-by: Zheng Zengkai <zhengzengkai@huawei.com>
-
由 zhoumin 提交于
LoongArch inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/I76XVK -------------------------------- PV IPI with less VM exit can improve performance than iocsr emulation when sending ipi interrupt to multi cpus. This patch supports: 1. sending PV IPI by hypercall 2. recording and updating steal time for VM Change-Id: Ia3ce493b83c067cc9e7c66b3094e4d87ff630f1d Signed-off-by: Nzhoumin <zhoumin@loongson.cn>
-
由 openeuler-ci-bot 提交于
Merge Pull Request from: @guzitao These patches generally cover the following tasks: 1.optimize kernel cores, remove unused codes, fix compile errors 2.fixes for perf, iommu, kvm, cpufreq, efi, struct pt_regs 3.add support, add support for restartable sequences, trace user task unalignment, add libbfd support in perf 4.optimize DIV and MOD instructions for bpf Link:https://gitee.com/openeuler/kernel/pulls/777 Reviewed-by: Jialin Zhang <zhangjialin11@huawei.com> Signed-off-by: Jialin Zhang <zhangjialin11@huawei.com> Acked-by: Xie XiuQi <xiexiuqi@huawei.com>
-
- 24 5月, 2023 6 次提交
-
-
由 openeuler-ci-bot 提交于
Merge Pull Request from: @ci-robot PR sync from: Jialin Zhang <zhangjialin11@huawei.com> https://mailweb.openeuler.org/hyperkitty/list/kernel@openeuler.org/thread/MYPG5M6GX66MEDMYSKLAETDPV37U7LWQ/ Link:https://gitee.com/openeuler/kernel/pulls/824 Reviewed-by: Zheng Zengkai <zhengzengkai@huawei.com> Signed-off-by: Zheng Zengkai <zhengzengkai@huawei.com>
-
由 Jialin Zhang 提交于
3snic inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I77U9U CVE: NA --------------------------- Net: ethernet: 3snic 3s9xx network Kconfig add "select NET_DEVLINK" Signed-off-by: NSteven Song <steven.song@3snic.com> Signed-off-by: NJialin Zhang <zhangjialin11@huawei.com>
-
由 openeuler-ci-bot 提交于
Merge Pull Request from: @zhangjialin11 This reverts commit 932afc18. Fix build error in riscv64: build failed: riscv, allmodconfig arch/riscv/kernel/smpboot.o: In function `smp_callin': (.text+0x160): undefined reference to `store_cpu_topology' arch/riscv/kernel/smpboot.o: In function `smp_prepare_cpus': (.init.text+0x212): undefined reference to `store_cpu_topology' make: *** [vmlinux] Error 1 Link:https://gitee.com/openeuler/kernel/pulls/821 Reviewed-by: Zheng Zengkai <zhengzengkai@huawei.com> Signed-off-by: Zheng Zengkai <zhengzengkai@huawei.com>
-
由 openeuler-ci-bot 提交于
Merge Pull Request from: @nebula_matrix M1600 dirver depend on CONFIG_HWMON, let's add the dependency. Bugfix:https://gitee.com/openeuler/kernel/issues/I77UIM?from=project-issue Link:https://gitee.com/openeuler/kernel/pulls/820 Reviewed-by: Jialin Zhang <zhangjialin11@huawei.com> Signed-off-by: Jialin Zhang <zhangjialin11@huawei.com>
-
由 Jialin Zhang 提交于
hulk inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I73HJ0 -------------------------------- This reverts commit 932afc18. Fix build error in riscv64: build failed: riscv, allmodconfig arch/riscv/kernel/smpboot.o: In function `smp_callin': (.text+0x160): undefined reference to `store_cpu_topology' arch/riscv/kernel/smpboot.o: In function `smp_prepare_cpus': (.init.text+0x212): undefined reference to `store_cpu_topology' make: *** [vmlinux] Error 1 Signed-off-by: NJialin Zhang <zhangjialin11@huawei.com>
-
由 nebula_matrix 提交于
m1600-dirver inclusion category: bugfix bugzilla: https://gitee.com/openeuler/kernel/issues/I77UIM CVE: NA -------------------------------- M1600 dirver depend on CONFIG_HWMON, let's add the dependency. Signed-off-by: NYi Chen <open@nebula-matrix.com>
-
- 23 5月, 2023 2 次提交
-
-
由 openeuler-ci-bot 提交于
Merge Pull Request from: @Hongchen_Zhang a double free problem in stmmac driver was triggered when doing mugen test, After apply this patch the double free problem disappear. Link:https://gitee.com/openeuler/kernel/pulls/761 Reviewed-by: Jialin Zhang <zhangjialin11@huawei.com> Reviewed-by: Guo Dongtai <guodongtai@kylinos.cn> Reviewed-by: Yue Haibing <yuehaibing@huawei.com> Signed-off-by: Jialin Zhang <zhangjialin11@huawei.com>
-
由 openeuler-ci-bot 提交于
Merge Pull Request from: @sanglipeng Backport 5.10.153 LTS patches from upstream. Conflicts: Already merged(3): 935a8b620210 mm/memory: add non-anonymous page check in the copy_present_page() 568e3812b177 mm,hugetlb: take hugetlb_lock before decrementing h->resv_huge_pages (Already merged mainline commit 12df140f mm,hugetlb: take hugetlb_lock before decrementing h->resv_huge_pages) d523384766fd scsi: sd: Revert "scsi: sd: Remove a local variable" (Already merged commit 7d7fe3e4 scsi: sd: Revert "scsi: sd: Remove a local variable") Context confilict(1): 3b250824b6d3 xhci: Add quirk to reset host back to default state at shutdown Rejected(4): ce605b68db53 net: hinic: fix incorrect assignment issue in hinic_set_interrupt_cfg() (The involved file drivers/net/ethernet/huawei/hinic/hinic_hw_dev.c is already deleted by the commit f1168d29 net/hinic: Update Hardware Abstract Layer) bb01910763f9 net: hinic: fix memory leak when reading function table (The involved file drivers/net/ethernet/huawei/hinic/hinic_debugfs.c is already deleted by the commit f1168d29 net/hinic: Update Hardware Abstract Layer) 6603843c80b1 net: hinic: fix the issue of CMDQ memory leaks (The involved file drivers/net/ethernet/huawei/hinic/hinic_hw_cmdq.c is already deleted by the commit f1168d29 net/hinic: Update Hardware Abstract Layer)) 0ce1ef335300 net: hinic: fix the issue of double release MBOX callback of VF (The involved function hinic_vf_func_init() is moved to other file.) Total patches: 91 - 3 - 4 = 84 Link:https://gitee.com/openeuler/kernel/pulls/807 Reviewed-by: Jialin Zhang <zhangjialin11@huawei.com> Signed-off-by: Jialin Zhang <zhangjialin11@huawei.com>
-