August 2024 - Kernel - mailweb.openeuler.org

[PATCH OLK-5.10] s390/pkey: Wipe copies of clear-key structures on failure
by Li Huafei 12 Aug '24

12 Aug '24

From: Holger Dengler <dengler(a)linux.ibm.com> mainline inclusion from mainline-v6.10-rc1 commit d65d76a44ffe74c73298ada25b0f578680576073 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IAH6LY CVE: CVE-2024-42156 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- Wipe all sensitive data from stack for all IOCTLs, which convert a clear-key into a protected- or secure-key. Reviewed-by: Harald Freudenberger <freude(a)linux.ibm.com> Reviewed-by: Ingo Franzki <ifranzki(a)linux.ibm.com> Acked-by: Heiko Carstens <hca(a)linux.ibm.com> Signed-off-by: Holger Dengler <dengler(a)linux.ibm.com> Signed-off-by: Alexander Gordeev <agordeev(a)linux.ibm.com> Conflicts: drivers/s390/crypto/pkey_api.c [ Resolved context conflict due to commit 6d749b4e0208 ("s390/pkey: introduce dynamic debugging for pkey") not backport. ] Signed-off-by: Li Huafei <lihuafei1(a)huawei.com> --- drivers/s390/crypto/pkey_api.c | 20 +++++++++++--------- 1 file changed, 11 insertions(+), 9 deletions(-) diff --git a/drivers/s390/crypto/pkey_api.c b/drivers/s390/crypto/pkey_api.c index 4f73b39a1455..1025a7bcd4db 100644 --- a/drivers/s390/crypto/pkey_api.c +++ b/drivers/s390/crypto/pkey_api.c @@ -1151,10 +1151,8 @@ static long pkey_unlocked_ioctl(struct file *filp, unsigned int cmd, rc = cca_clr2seckey(kcs.cardnr, kcs.domain, kcs.keytype, kcs.clrkey.clrkey, kcs.seckey.seckey); DEBUG_DBG("%s cca_clr2seckey()=%d\n", __func__, rc); - if (rc) - break; - if (copy_to_user(ucs, &kcs, sizeof(kcs))) - rc = -EFAULT; + if (!rc && copy_to_user(ucs, &kcs, sizeof(kcs))) + return -EFAULT; memzero_explicit(&kcs, sizeof(kcs)); break; } @@ -1182,10 +1180,8 @@ static long pkey_unlocked_ioctl(struct file *filp, unsigned int cmd, rc = pkey_clr2protkey(kcp.keytype, &kcp.clrkey, &kcp.protkey); DEBUG_DBG("%s pkey_clr2protkey()=%d\n", __func__, rc); - if (rc) - break; - if (copy_to_user(ucp, &kcp, sizeof(kcp))) - rc = -EFAULT; + if (!rc && copy_to_user(ucp, &kcp, sizeof(kcp))) + return -EFAULT; memzero_explicit(&kcp, sizeof(kcp)); break; } @@ -1325,11 +1321,14 @@ static long pkey_unlocked_ioctl(struct file *filp, unsigned int cmd, if (copy_from_user(&kcs, ucs, sizeof(kcs))) return -EFAULT; apqns = _copy_apqns_from_user(kcs.apqns, kcs.apqn_entries); - if (IS_ERR(apqns)) + if (IS_ERR(apqns)) { + memzero_explicit(&kcs, sizeof(kcs)); return PTR_ERR(apqns); + } kkey = kmalloc(klen, GFP_KERNEL); if (!kkey) { kfree(apqns); + memzero_explicit(&kcs, sizeof(kcs)); return -ENOMEM; } rc = pkey_clr2seckey2(apqns, kcs.apqn_entries, @@ -1339,15 +1338,18 @@ static long pkey_unlocked_ioctl(struct file *filp, unsigned int cmd, kfree(apqns); if (rc) { kfree_sensitive(kkey); + memzero_explicit(&kcs, sizeof(kcs)); break; } if (kcs.key) { if (kcs.keylen < klen) { kfree_sensitive(kkey); + memzero_explicit(&kcs, sizeof(kcs)); return -EINVAL; } if (copy_to_user(kcs.key, kkey, klen)) { kfree_sensitive(kkey); + memzero_explicit(&kcs, sizeof(kcs)); return -EFAULT; } } -- 2.25.1

2 1

[PATCH OLK-5.10 V3] sched/fair: Prefer physical cores when migrating tasks
by Cheng Yu 12 Aug '24

12 Aug '24

hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/IAJEHU CVE: NA -------------------------------- When cpu hyperthreading is enabled, one physical core can virtualize multiple logical cpus. Assume that physical core0 virtualizes two logical cpus, cpu0 and cpu1. Only when the load of cpu0 exceeds the set ratio to the capacity of cpu0, the task will be migrated to the cpu1, otherwise the task will not be migrated and the cpu0 will still be used. External impacts: 1) default config in arm64,x86: CONFIG_SCHED_KEEP_ON_CORE=y 2) sysctl: /proc/sys/kernel/sched_util_ratio 3) sched features: KEEP_ON_CORE (default NO_KEEP_ON_CORE) Signed-off-by: Cheng Yu <serein.chengyu(a)huawei.com> --- arch/arm64/configs/openeuler_defconfig | 1 + arch/x86/configs/openeuler_defconfig | 1 + include/linux/sched/sysctl.h | 4 ++++ init/Kconfig | 12 ++++++++++ kernel/sched/fair.c | 31 ++++++++++++++++++++++++++ kernel/sched/features.h | 4 ++++ kernel/sysctl.c | 10 +++++++++ 7 files changed, 63 insertions(+) diff --git a/arch/arm64/configs/openeuler_defconfig b/arch/arm64/configs/openeuler_defconfig index 69ff0b64ba59..8c95a5332b40 100644 --- a/arch/arm64/configs/openeuler_defconfig +++ b/arch/arm64/configs/openeuler_defconfig @@ -189,6 +189,7 @@ CONFIG_USER_NS=y CONFIG_PID_NS=y CONFIG_NET_NS=y CONFIG_SCHED_STEAL=y +CONFIG_SCHED_KEEP_ON_CORE=y CONFIG_CHECKPOINT_RESTORE=y CONFIG_SCHED_AUTOGROUP=y # CONFIG_SYSFS_DEPRECATED is not set diff --git a/arch/x86/configs/openeuler_defconfig b/arch/x86/configs/openeuler_defconfig index f3b810d0cf47..c2ebdb75bbe3 100644 --- a/arch/x86/configs/openeuler_defconfig +++ b/arch/x86/configs/openeuler_defconfig @@ -194,6 +194,7 @@ CONFIG_USER_NS=y CONFIG_PID_NS=y CONFIG_NET_NS=y CONFIG_SCHED_STEAL=y +# CONFIG_SCHED_KEEP_ON_CORE is not set CONFIG_CHECKPOINT_RESTORE=y CONFIG_SCHED_AUTOGROUP=y # CONFIG_SYSFS_DEPRECATED is not set diff --git a/include/linux/sched/sysctl.h b/include/linux/sched/sysctl.h index 5cd5b3c579d3..09214349bddf 100644 --- a/include/linux/sched/sysctl.h +++ b/include/linux/sched/sysctl.h @@ -35,6 +35,10 @@ extern unsigned int sysctl_sched_child_runs_first; extern int sysctl_sched_util_low_pct; #endif +#ifdef CONFIG_SCHED_KEEP_ON_CORE +extern int sysctl_sched_util_ratio; +#endif + #ifdef CONFIG_QOS_SCHED_SMART_GRID extern unsigned int sysctl_smart_grid_strategy_ctrl; extern int sysctl_affinity_adjust_delay_ms; diff --git a/init/Kconfig b/init/Kconfig index e552194efbea..9dcc12704729 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -1370,6 +1370,18 @@ config SCHED_STEAL If unsure, say N here. +config SCHED_KEEP_ON_CORE + bool "Prefer physical cores when migrating tasks" + depends on SCHED_SMT + default n + help + When cpu hyperthreading is enabled, one physical core can virtualize + multiple logical cpus. Assume that physical core0 virtualizes two + logical cpus, cpu0 and cpu1. Only when the load of cpu0 exceeds the + ratio to the capacity of cpu0, the task will be migrated to the cpu1, + otherwise the task will not be migrated and the cpu0 will still be + used. + config CHECKPOINT_RESTORE bool "Checkpoint/restore support" select PROC_CHILDREN diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 95d1841f8a20..8dc1d2afe1b7 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -7242,6 +7242,20 @@ static int select_idle_core(struct task_struct *p, int core, struct cpumask *cpu return -1; } +#ifdef CONFIG_SCHED_KEEP_ON_CORE +int sysctl_sched_util_ratio = 100; + +static int core_has_spare(int cpu) +{ + int core_id = cpumask_first(cpu_smt_mask(cpu)); + struct rq *rq = cpu_rq(core_id); + unsigned long util = rq->cfs.avg.util_avg; + unsigned long capacity = rq->cpu_capacity; + + return util * 100 < capacity * sysctl_sched_util_ratio; +} +#endif + #else /* CONFIG_SCHED_SMT */ static inline void set_idle_cores(int cpu, int val) @@ -8210,6 +8224,14 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_f } #endif +#ifdef CONFIG_SCHED_KEEP_ON_CORE + if (sched_feat(KEEP_ON_CORE) && + static_branch_likely(&sched_smt_present)) { + if (core_has_spare(new_cpu)) + new_cpu = cpumask_first(cpu_smt_mask((new_cpu))); + } +#endif + rcu_read_unlock(); #ifdef CONFIG_QOS_SCHED_DYNAMIC_AFFINITY @@ -9701,6 +9723,15 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env) } #endif +#ifdef CONFIG_SCHED_KEEP_ON_CORE + if (sched_feat(KEEP_ON_CORE) && + static_branch_likely(&sched_smt_present)) { + if (core_has_spare(env->dst_cpu) && + cpumask_first(cpu_smt_mask((env->dst_cpu))) != env->dst_cpu) + return 0; + } +#endif + /* * We do not migrate tasks that are: * 1) throttled_lb_pair, or diff --git a/kernel/sched/features.h b/kernel/sched/features.h index 76fade025c4b..fb885b20ba34 100644 --- a/kernel/sched/features.h +++ b/kernel/sched/features.h @@ -70,6 +70,10 @@ SCHED_FEAT(SIS_UTIL, false) SCHED_FEAT(STEAL, false) #endif +#ifdef CONFIG_SCHED_KEEP_ON_CORE +SCHED_FEAT(KEEP_ON_CORE, false) +#endif + /* * Issue a WARN when we do multiple update_rq_clock() calls * in a single rq->lock section. Default disabled because the diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 3941856c19d1..bd7b17be9ba4 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -2817,6 +2817,16 @@ static struct ctl_table kern_table[] = { .extra2 = &one_hundred, }, #endif +#ifdef CONFIG_SCHED_KEEP_ON_CORE + { + .procname = "sched_util_ratio", + .data = &sysctl_sched_util_ratio, + .maxlen = sizeof(sysctl_sched_util_ratio), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = SYSCTL_ZERO, + }, +#endif #ifdef CONFIG_QOS_SCHED_SMART_GRID { .procname = "smart_grid_strategy_ctrl", -- 2.25.1

2 1

[PATCH openEuler-1.0-LTS] crypto: aead,cipher - zeroize key buffer after use
by Yi Yang 12 Aug '24

12 Aug '24

From: Hailey Mothershead <hailmo(a)amazon.com> mainline inclusion from mainline-v6.10-rc1 commit 23e4099bdc3c8381992f9eb975c79196d6755210 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IAHJG9 CVE: CVE-2024-42229 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?… -------------------------------- I.G 9.7.B for FIPS 140-3 specifies that variables temporarily holding cryptographic information should be zeroized once they are no longer needed. Accomplish this by using kfree_sensitive for buffers that previously held the private key. Signed-off-by: Hailey Mothershead <hailmo(a)amazon.com> Signed-off-by: Herbert Xu <herbert(a)gondor.apana.org.au> Conflicts: crypto/cipher.c [context conflict] Signed-off-by: Yi Yang <yiyang13(a)huawei.com> --- crypto/aead.c | 3 +-- crypto/cipher.c | 3 +-- 2 files changed, 2 insertions(+), 4 deletions(-) diff --git a/crypto/aead.c b/crypto/aead.c index 9688ada13981..e6d7816fbf2f 100644 --- a/crypto/aead.c +++ b/crypto/aead.c @@ -45,8 +45,7 @@ static int setkey_unaligned(struct crypto_aead *tfm, const u8 *key, alignbuffer = (u8 *)ALIGN((unsigned long)buffer, alignmask + 1); memcpy(alignbuffer, key, keylen); ret = crypto_aead_alg(tfm)->setkey(tfm, alignbuffer, keylen); - memset(alignbuffer, 0, keylen); - kfree(buffer); + kfree_sensitive(buffer); return ret; } diff --git a/crypto/cipher.c b/crypto/cipher.c index 57836c30a49a..ba4193ba237b 100644 --- a/crypto/cipher.c +++ b/crypto/cipher.c @@ -38,8 +38,7 @@ static int setkey_unaligned(struct crypto_tfm *tfm, const u8 *key, alignbuffer = (u8 *)ALIGN((unsigned long)buffer, alignmask + 1); memcpy(alignbuffer, key, keylen); ret = cia->cia_setkey(tfm, alignbuffer, keylen); - memset(alignbuffer, 0, keylen); - kfree(buffer); + kfree_sensitive(buffer); return ret; } -- 2.25.1

2 1

[PATCH OLK-5.10] crypto: aead,cipher - zeroize key buffer after use
by Yi Yang 12 Aug '24

12 Aug '24

From: Hailey Mothershead <hailmo(a)amazon.com> stable inclusion from stable-v5.10.222 commit 9db8c299a521813630fcb4154298cb60c37f3133 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IAHJG9 CVE: CVE-2024-42229 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 23e4099bdc3c8381992f9eb975c79196d6755210 ] I.G 9.7.B for FIPS 140-3 specifies that variables temporarily holding cryptographic information should be zeroized once they are no longer needed. Accomplish this by using kfree_sensitive for buffers that previously held the private key. Signed-off-by: Hailey Mothershead <hailmo(a)amazon.com> Signed-off-by: Herbert Xu <herbert(a)gondor.apana.org.au> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Yi Yang <yiyang13(a)huawei.com> --- crypto/aead.c | 3 +-- crypto/cipher.c | 3 +-- 2 files changed, 2 insertions(+), 4 deletions(-) diff --git a/crypto/aead.c b/crypto/aead.c index 16991095270d..c4ece86c45bc 100644 --- a/crypto/aead.c +++ b/crypto/aead.c @@ -35,8 +35,7 @@ static int setkey_unaligned(struct crypto_aead *tfm, const u8 *key, alignbuffer = (u8 *)ALIGN((unsigned long)buffer, alignmask + 1); memcpy(alignbuffer, key, keylen); ret = crypto_aead_alg(tfm)->setkey(tfm, alignbuffer, keylen); - memset(alignbuffer, 0, keylen); - kfree(buffer); + kfree_sensitive(buffer); return ret; } diff --git a/crypto/cipher.c b/crypto/cipher.c index fd78150deb1c..72c5606cc7f8 100644 --- a/crypto/cipher.c +++ b/crypto/cipher.c @@ -33,8 +33,7 @@ static int setkey_unaligned(struct crypto_cipher *tfm, const u8 *key, alignbuffer = (u8 *)ALIGN((unsigned long)buffer, alignmask + 1); memcpy(alignbuffer, key, keylen); ret = cia->cia_setkey(crypto_cipher_tfm(tfm), alignbuffer, keylen); - memset(alignbuffer, 0, keylen); - kfree(buffer); + kfree_sensitive(buffer); return ret; } -- 2.25.1

2 1

[PATCH openEuler-22.03-LTS-SP1] crypto: aead,cipher - zeroize key buffer after use
by Yi Yang 12 Aug '24

12 Aug '24

From: Hailey Mothershead <hailmo(a)amazon.com> stable inclusion from stable-v5.10.222 commit 9db8c299a521813630fcb4154298cb60c37f3133 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IAHJG9 CVE: CVE-2024-42229 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 23e4099bdc3c8381992f9eb975c79196d6755210 ] I.G 9.7.B for FIPS 140-3 specifies that variables temporarily holding cryptographic information should be zeroized once they are no longer needed. Accomplish this by using kfree_sensitive for buffers that previously held the private key. Signed-off-by: Hailey Mothershead <hailmo(a)amazon.com> Signed-off-by: Herbert Xu <herbert(a)gondor.apana.org.au> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Yi Yang <yiyang13(a)huawei.com> --- crypto/aead.c | 3 +-- crypto/cipher.c | 3 +-- 2 files changed, 2 insertions(+), 4 deletions(-) diff --git a/crypto/aead.c b/crypto/aead.c index 16991095270d..c4ece86c45bc 100644 --- a/crypto/aead.c +++ b/crypto/aead.c @@ -35,8 +35,7 @@ static int setkey_unaligned(struct crypto_aead *tfm, const u8 *key, alignbuffer = (u8 *)ALIGN((unsigned long)buffer, alignmask + 1); memcpy(alignbuffer, key, keylen); ret = crypto_aead_alg(tfm)->setkey(tfm, alignbuffer, keylen); - memset(alignbuffer, 0, keylen); - kfree(buffer); + kfree_sensitive(buffer); return ret; } diff --git a/crypto/cipher.c b/crypto/cipher.c index fd78150deb1c..72c5606cc7f8 100644 --- a/crypto/cipher.c +++ b/crypto/cipher.c @@ -33,8 +33,7 @@ static int setkey_unaligned(struct crypto_cipher *tfm, const u8 *key, alignbuffer = (u8 *)ALIGN((unsigned long)buffer, alignmask + 1); memcpy(alignbuffer, key, keylen); ret = cia->cia_setkey(crypto_cipher_tfm(tfm), alignbuffer, keylen); - memset(alignbuffer, 0, keylen); - kfree(buffer); + kfree_sensitive(buffer); return ret; } -- 2.25.1

2 1

[PATCH OLK-5.10] crypto: aead,cipher - zeroize key buffer after use
by Yi Yang 12 Aug '24

12 Aug '24

From: Hailey Mothershead <hailmo(a)amazon.com> stable inclusion from stable-v5.10.222 commit 9db8c299a521813630fcb4154298cb60c37f3133 category: bugfix bugzilla: 190216 CVE: CVE-2024-42229 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 23e4099bdc3c8381992f9eb975c79196d6755210 ] I.G 9.7.B for FIPS 140-3 specifies that variables temporarily holding cryptographic information should be zeroized once they are no longer needed. Accomplish this by using kfree_sensitive for buffers that previously held the private key. Signed-off-by: Hailey Mothershead <hailmo(a)amazon.com> Signed-off-by: Herbert Xu <herbert(a)gondor.apana.org.au> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Yi Yang <yiyang13(a)huawei.com> --- crypto/aead.c | 3 +-- crypto/cipher.c | 3 +-- 2 files changed, 2 insertions(+), 4 deletions(-) diff --git a/crypto/aead.c b/crypto/aead.c index 16991095270d..c4ece86c45bc 100644 --- a/crypto/aead.c +++ b/crypto/aead.c @@ -35,8 +35,7 @@ static int setkey_unaligned(struct crypto_aead *tfm, const u8 *key, alignbuffer = (u8 *)ALIGN((unsigned long)buffer, alignmask + 1); memcpy(alignbuffer, key, keylen); ret = crypto_aead_alg(tfm)->setkey(tfm, alignbuffer, keylen); - memset(alignbuffer, 0, keylen); - kfree(buffer); + kfree_sensitive(buffer); return ret; } diff --git a/crypto/cipher.c b/crypto/cipher.c index fd78150deb1c..72c5606cc7f8 100644 --- a/crypto/cipher.c +++ b/crypto/cipher.c @@ -33,8 +33,7 @@ static int setkey_unaligned(struct crypto_cipher *tfm, const u8 *key, alignbuffer = (u8 *)ALIGN((unsigned long)buffer, alignmask + 1); memcpy(alignbuffer, key, keylen); ret = cia->cia_setkey(crypto_cipher_tfm(tfm), alignbuffer, keylen); - memset(alignbuffer, 0, keylen); - kfree(buffer); + kfree_sensitive(buffer); return ret; } -- 2.25.1

2 1

[PATCH OLK-5.10 V2] sched/fair: Prefer physical cores when migrating tasks
by Cheng Yu 12 Aug '24

12 Aug '24

hulk inclusion category: feature bugzilla: https://gitee.com/openeuler/kernel/issues/IAJEHU CVE: NA -------------------------------- When cpu hyperthreading is enabled, one physical core can virtualize multiple logical cpus. Assume that physical core0 virtualizes two logical cpus, cpu0 and cpu1. Only when the load of cpu0 exceeds the set ratio to the capacity of cpu0, the task will be migrated to the cpu1, otherwise the task will not be migrated and the cpu0 will still be used. External impacts: 1) default config in arm64,x86: CONFIG_SCHED_KEEP_ON_CORE=y 2) sysctl: /proc/sys/kernel/sched_util_ratio 3) sched features: KEEP_ON_CORE (default NO_KEEP_ON_CORE) Signed-off-by: Cheng Yu <serein.chengyu(a)huawei.com> --- arch/arm64/configs/openeuler_defconfig | 1 + include/linux/sched/sysctl.h | 4 ++++ init/Kconfig | 12 ++++++++++ kernel/sched/fair.c | 31 ++++++++++++++++++++++++++ kernel/sched/features.h | 4 ++++ kernel/sysctl.c | 10 +++++++++ 6 files changed, 62 insertions(+) diff --git a/arch/arm64/configs/openeuler_defconfig b/arch/arm64/configs/openeuler_defconfig index 69ff0b64ba59..8c95a5332b40 100644 --- a/arch/arm64/configs/openeuler_defconfig +++ b/arch/arm64/configs/openeuler_defconfig @@ -189,6 +189,7 @@ CONFIG_USER_NS=y CONFIG_PID_NS=y CONFIG_NET_NS=y CONFIG_SCHED_STEAL=y +CONFIG_SCHED_KEEP_ON_CORE=y CONFIG_CHECKPOINT_RESTORE=y CONFIG_SCHED_AUTOGROUP=y # CONFIG_SYSFS_DEPRECATED is not set diff --git a/include/linux/sched/sysctl.h b/include/linux/sched/sysctl.h index 5cd5b3c579d3..09214349bddf 100644 --- a/include/linux/sched/sysctl.h +++ b/include/linux/sched/sysctl.h @@ -35,6 +35,10 @@ extern unsigned int sysctl_sched_child_runs_first; extern int sysctl_sched_util_low_pct; #endif +#ifdef CONFIG_SCHED_KEEP_ON_CORE +extern int sysctl_sched_util_ratio; +#endif + #ifdef CONFIG_QOS_SCHED_SMART_GRID extern unsigned int sysctl_smart_grid_strategy_ctrl; extern int sysctl_affinity_adjust_delay_ms; diff --git a/init/Kconfig b/init/Kconfig index e552194efbea..9dcc12704729 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -1370,6 +1370,18 @@ config SCHED_STEAL If unsure, say N here. +config SCHED_KEEP_ON_CORE + bool "Prefer physical cores when migrating tasks" + depends on SCHED_SMT + default n + help + When cpu hyperthreading is enabled, one physical core can virtualize + multiple logical cpus. Assume that physical core0 virtualizes two + logical cpus, cpu0 and cpu1. Only when the load of cpu0 exceeds the + ratio to the capacity of cpu0, the task will be migrated to the cpu1, + otherwise the task will not be migrated and the cpu0 will still be + used. + config CHECKPOINT_RESTORE bool "Checkpoint/restore support" select PROC_CHILDREN diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 95d1841f8a20..8dc1d2afe1b7 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -7242,6 +7242,20 @@ static int select_idle_core(struct task_struct *p, int core, struct cpumask *cpu return -1; } +#ifdef CONFIG_SCHED_KEEP_ON_CORE +int sysctl_sched_util_ratio = 100; + +static int core_has_spare(int cpu) +{ + int core_id = cpumask_first(cpu_smt_mask(cpu)); + struct rq *rq = cpu_rq(core_id); + unsigned long util = rq->cfs.avg.util_avg; + unsigned long capacity = rq->cpu_capacity; + + return util * 100 < capacity * sysctl_sched_util_ratio; +} +#endif + #else /* CONFIG_SCHED_SMT */ static inline void set_idle_cores(int cpu, int val) @@ -8210,6 +8224,14 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_f } #endif +#ifdef CONFIG_SCHED_KEEP_ON_CORE + if (sched_feat(KEEP_ON_CORE) && + static_branch_likely(&sched_smt_present)) { + if (core_has_spare(new_cpu)) + new_cpu = cpumask_first(cpu_smt_mask((new_cpu))); + } +#endif + rcu_read_unlock(); #ifdef CONFIG_QOS_SCHED_DYNAMIC_AFFINITY @@ -9701,6 +9723,15 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env) } #endif +#ifdef CONFIG_SCHED_KEEP_ON_CORE + if (sched_feat(KEEP_ON_CORE) && + static_branch_likely(&sched_smt_present)) { + if (core_has_spare(env->dst_cpu) && + cpumask_first(cpu_smt_mask((env->dst_cpu))) != env->dst_cpu) + return 0; + } +#endif + /* * We do not migrate tasks that are: * 1) throttled_lb_pair, or diff --git a/kernel/sched/features.h b/kernel/sched/features.h index 76fade025c4b..fb885b20ba34 100644 --- a/kernel/sched/features.h +++ b/kernel/sched/features.h @@ -70,6 +70,10 @@ SCHED_FEAT(SIS_UTIL, false) SCHED_FEAT(STEAL, false) #endif +#ifdef CONFIG_SCHED_KEEP_ON_CORE +SCHED_FEAT(KEEP_ON_CORE, false) +#endif + /* * Issue a WARN when we do multiple update_rq_clock() calls * in a single rq->lock section. Default disabled because the diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 3941856c19d1..bd7b17be9ba4 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -2817,6 +2817,16 @@ static struct ctl_table kern_table[] = { .extra2 = &one_hundred, }, #endif +#ifdef CONFIG_SCHED_KEEP_ON_CORE + { + .procname = "sched_util_ratio", + .data = &sysctl_sched_util_ratio, + .maxlen = sizeof(sysctl_sched_util_ratio), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = SYSCTL_ZERO, + }, +#endif #ifdef CONFIG_QOS_SCHED_SMART_GRID { .procname = "smart_grid_strategy_ctrl", -- 2.25.1

2 1

[PATCH openEuler-22.03-LTS-SP1] orangefs: fix out-of-bounds fsid access
by Liu Shixin 12 Aug '24

12 Aug '24

From: Mike Marshall <hubcap(a)omnibond.com> stable inclusion from stable-v5.10.222 commit 1617249e24bd04c8047956afb43feec4876d1715 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IAGPSE CVE: CVE-2024-42143 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit 53e4efa470d5fc6a96662d2d3322cfc925818517 ] Arnd Bergmann sent a patch to fsdevel, he says: "orangefs_statfs() copies two consecutive fields of the superblock into the statfs structure, which triggers a warning from the string fortification helpers" Jan Kara suggested an alternate way to do the patch to make it more readable. I ran both ideas through xfstests and both seem fine. This patch is based on Jan Kara's suggestion. Signed-off-by: Mike Marshall <hubcap(a)omnibond.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Liu Shixin <liushixin2(a)huawei.com> --- fs/orangefs/super.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/fs/orangefs/super.c b/fs/orangefs/super.c index 2f2e430461b2..b48aef43b51d 100644 --- a/fs/orangefs/super.c +++ b/fs/orangefs/super.c @@ -200,7 +200,8 @@ static int orangefs_statfs(struct dentry *dentry, struct kstatfs *buf) (long)new_op->downcall.resp.statfs.files_avail); buf->f_type = sb->s_magic; - memcpy(&buf->f_fsid, &ORANGEFS_SB(sb)->fs_id, sizeof(buf->f_fsid)); + buf->f_fsid.val[0] = ORANGEFS_SB(sb)->fs_id; + buf->f_fsid.val[1] = ORANGEFS_SB(sb)->id; buf->f_bsize = new_op->downcall.resp.statfs.block_size; buf->f_namelen = ORANGEFS_NAME_MAX; -- 2.25.1

2 1

[PATCH openEuler-22.03-LTS-SP1] drm/panel: ilitek-ili9881c: Fix warning with GPIO controllers that sleep
by Liu Shixin 12 Aug '24

12 Aug '24

From: Laurent Pinchart <laurent.pinchart(a)ideasonboard.com> stable inclusion from stable-v5.10.221 commit cae52f61fda0f5d2949dc177f984c9e187d4c6a0 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IAGEP0 CVE: CVE-2024-42087 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- [ Upstream commit ee7860cd8b5763017f8dc785c2851fecb7a0c565 ] The ilitek-ili9881c controls the reset GPIO using the non-sleeping gpiod_set_value() function. This complains loudly when the GPIO controller needs to sleep. As the caller can sleep, use gpiod_set_value_cansleep() to fix the issue. Signed-off-by: Laurent Pinchart <laurent.pinchart(a)ideasonboard.com> Reviewed-by: Neil Armstrong <neil.armstrong(a)linaro.org> Link: https://lore.kernel.org/r/20240317154839.21260-1-laurent.pinchart@ideasonbo… Signed-off-by: Neil Armstrong <neil.armstrong(a)linaro.org> Link: https://patchwork.freedesktop.org/patch/msgid/20240317154839.21260-1-lauren… Signed-off-by: Sasha Levin <sashal(a)kernel.org> Signed-off-by: Liu Shixin <liushixin2(a)huawei.com> --- drivers/gpu/drm/panel/panel-ilitek-ili9881c.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c b/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c index 534dd7414d42..917cb322bab1 100644 --- a/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c +++ b/drivers/gpu/drm/panel/panel-ilitek-ili9881c.c @@ -506,10 +506,10 @@ static int ili9881c_prepare(struct drm_panel *panel) msleep(5); /* And reset it */ - gpiod_set_value(ctx->reset, 1); + gpiod_set_value_cansleep(ctx->reset, 1); msleep(20); - gpiod_set_value(ctx->reset, 0); + gpiod_set_value_cansleep(ctx->reset, 0); msleep(20); for (i = 0; i < ctx->desc->init_length; i++) { @@ -564,7 +564,7 @@ static int ili9881c_unprepare(struct drm_panel *panel) mipi_dsi_dcs_enter_sleep_mode(ctx->dsi); regulator_disable(ctx->power); - gpiod_set_value(ctx->reset, 1); + gpiod_set_value_cansleep(ctx->reset, 1); return 0; } -- 2.25.1

2 1

[PATCH openEuler-22.03-LTS-SP1] drm/i915/gt: Fix potential UAF by revoke of fence registers
by Liu Shixin 12 Aug '24

12 Aug '24

From: Janusz Krzysztofik <janusz.krzysztofik(a)linux.intel.com> stable inclusion from stable-v5.10.221 commit f771b91f21c46ad1217328d05e72a2c7e3add535 category: bugfix bugzilla: https://gitee.com/src-openeuler/kernel/issues/IAGEOK CVE: CVE-2024-41092 Reference: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?id… -------------------------------- commit 996c3412a06578e9d779a16b9e79ace18125ab50 upstream. CI has been sporadically reporting the following issue triggered by igt@i915_selftest@live@hangcheck on ADL-P and similar machines: <6> [414.049203] i915: Running intel_hangcheck_live_selftests/igt_reset_evict_fence ... <6> [414.068804] i915 0000:00:02.0: [drm] GT0: GUC: submission enabled <6> [414.068812] i915 0000:00:02.0: [drm] GT0: GUC: SLPC enabled <3> [414.070354] Unable to pin Y-tiled fence; err:-4 <3> [414.071282] i915_vma_revoke_fence:301 GEM_BUG_ON(!i915_active_is_idle(&fence->active)) ... <4>[ 609.603992] ------------[ cut here ]------------ <2>[ 609.603995] kernel BUG at drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c:301! <4>[ 609.604003] invalid opcode: 0000 [#1] PREEMPT SMP NOPTI <4>[ 609.604006] CPU: 0 PID: 268 Comm: kworker/u64:3 Tainted: G U W 6.9.0-CI_DRM_14785-g1ba62f8cea9c+ #1 <4>[ 609.604008] Hardware name: Intel Corporation Alder Lake Client Platform/AlderLake-P DDR4 RVP, BIOS RPLPFWI1.R00.4035.A00.2301200723 01/20/2023 <4>[ 609.604010] Workqueue: i915 __i915_gem_free_work [i915] <4>[ 609.604149] RIP: 0010:i915_vma_revoke_fence+0x187/0x1f0 [i915] ... <4>[ 609.604271] Call Trace: <4>[ 609.604273] <TASK> ... <4>[ 609.604716] __i915_vma_evict+0x2e9/0x550 [i915] <4>[ 609.604852] __i915_vma_unbind+0x7c/0x160 [i915] <4>[ 609.604977] force_unbind+0x24/0xa0 [i915] <4>[ 609.605098] i915_vma_destroy+0x2f/0xa0 [i915] <4>[ 609.605210] __i915_gem_object_pages_fini+0x51/0x2f0 [i915] <4>[ 609.605330] __i915_gem_free_objects.isra.0+0x6a/0xc0 [i915] <4>[ 609.605440] process_scheduled_works+0x351/0x690 ... In the past, there were similar failures reported by CI from other IGT tests, observed on other platforms. Before commit 63baf4f3d587 ("drm/i915/gt: Only wait for GPU activity before unbinding a GGTT fence"), i915_vma_revoke_fence() was waiting for idleness of vma->active via fence_update(). That commit introduced vma->fence->active in order for the fence_update() to be able to wait selectively on that one instead of vma->active since only idleness of fence registers was needed. But then, another commit 0d86ee35097a ("drm/i915/gt: Make fence revocation unequivocal") replaced the call to fence_update() in i915_vma_revoke_fence() with only fence_write(), and also added that GEM_BUG_ON(!i915_active_is_idle(&fence->active)) in front. No justification was provided on why we might then expect idleness of vma->fence->active without first waiting on it. The issue can be potentially caused by a race among revocation of fence registers on one side and sequential execution of signal callbacks invoked on completion of a request that was using them on the other, still processed in parallel to revocation of those fence registers. Fix it by waiting for idleness of vma->fence->active in i915_vma_revoke_fence(). Fixes: 0d86ee35097a ("drm/i915/gt: Make fence revocation unequivocal") Closes: https://gitlab.freedesktop.org/drm/intel/issues/10021 Signed-off-by: Janusz Krzysztofik <janusz.krzysztofik(a)linux.intel.com> Cc: stable(a)vger.kernel.org # v5.8+ Reviewed-by: Andi Shyti <andi.shyti(a)linux.intel.com> Signed-off-by: Andi Shyti <andi.shyti(a)linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240603195446.297690-2-janus… (cherry picked from commit 24bb052d3dd499c5956abad5f7d8e4fd07da7fb1) Signed-off-by: Jani Nikula <jani.nikula(a)intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh(a)linuxfoundation.org> Signed-off-by: Liu Shixin <liushixin2(a)huawei.com> --- drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c b/drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c index cd71631bef0c..ddf76a4e4fa2 100644 --- a/drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c +++ b/drivers/gpu/drm/i915/gt/intel_ggtt_fencing.c @@ -309,6 +309,7 @@ void i915_vma_revoke_fence(struct i915_vma *vma) return; GEM_BUG_ON(fence->vma != vma); + i915_active_wait(&fence->active); GEM_BUG_ON(!i915_active_is_idle(&fence->active)); GEM_BUG_ON(atomic_read(&fence->pin_count)); -- 2.25.1

2 1

2024

2023

2022

2021

2020

2019

Kernel August 2024