August 2019 - Linux-stable-mirror

[to-be-updated] mm-memcontrol-flush-percpu-vmstats-before-releasing-memcg.patch removed from -mm tree

by akpm＠linux-foundation.org

The patch titled Subject: mm: memcontrol: flush percpu vmstats before releasing memcg has been removed from the -mm tree. Its filename was mm-memcontrol-flush-percpu-vmstats-before-releasing-memcg.patch This patch was dropped because an updated version will be merged ------------------------------------------------------ From: Roman Gushchin <guro(a)fb.com> Subject: mm: memcontrol: flush percpu vmstats before releasing memcg Percpu caching of local vmstats with the conditional propagation by the cgroup tree leads to an accumulation of errors on non-leaf levels. Let's imagine two nested memory cgroups A and A/B. Say, a process belonging to A/B allocates 100 pagecache pages on the CPU 0. The percpu cache will spill 3 times, so that 32*3=96 pages will be accounted to A/B and A atomic vmstat counters, 4 pages will remain in the percpu cache. Imagine A/B is nearby memory.max, so that every following allocation triggers a direct reclaim on the local CPU. Say, each such attempt will free 16 pages on a new cpu. That means every percpu cache will have -16 pages, except the first one, which will have 4 - 16 = -12. A/B and A atomic counters will not be touched at all. Now a user removes A/B. All percpu caches are freed and corresponding vmstat numbers are forgotten. A has 96 pages more than expected. As memory cgroups are created and destroyed, errors do accumulate. Even 1-2 pages differences can accumulate into large numbers. To fix this issue let's accumulate and propagate percpu vmstat values before releasing the memory cgroup. At this point these numbers are stable and cannot be changed. Since on cpu hotplug we do flush percpu vmstats anyway, we can iterate only over online cpus. Link: http://lkml.kernel.org/r/20190812222911.2364802-2-guro@fb.com Fixes: 42a300353577 ("mm: memcontrol: fix recursive statistics correctness & scalabilty") Signed-off-by: Roman Gushchin <guro(a)fb.com> Acked-by: Michal Hocko <mhocko(a)suse.com> Cc: Johannes Weiner <hannes(a)cmpxchg.org> Cc: Vladimir Davydov <vdavydov.dev(a)gmail.com> Cc: <stable(a)vger.kernel.org> Signed-off-by: Andrew Morton <akpm(a)linux-foundation.org> --- mm/memcontrol.c | 40 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 40 insertions(+) --- a/mm/memcontrol.c~mm-memcontrol-flush-percpu-vmstats-before-releasing-memcg +++ a/mm/memcontrol.c @@ -3289,6 +3289,41 @@ static int memcg_online_kmem(struct mem_ return 0; } +static void memcg_flush_percpu_vmstats(struct mem_cgroup *memcg) +{ + unsigned long stat[MEMCG_NR_STAT]; + struct mem_cgroup *mi; + int node, cpu, i; + + for (i = 0; i < MEMCG_NR_STAT; i++) + stat[i] = 0; + + for_each_online_cpu(cpu) + for (i = 0; i < MEMCG_NR_STAT; i++) + stat[i] += raw_cpu_read(memcg->vmstats_percpu->stat[i]); + + for (mi = memcg; mi; mi = parent_mem_cgroup(mi)) + for (i = 0; i < MEMCG_NR_STAT; i++) + atomic_long_add(stat[i], &mi->vmstats[i]); + + for_each_node(node) { + struct mem_cgroup_per_node *pn = memcg->nodeinfo[node]; + struct mem_cgroup_per_node *pi; + + for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++) + stat[i] = 0; + + for_each_online_cpu(cpu) + for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++) + stat[i] += raw_cpu_read( + pn->lruvec_stat_cpu->count[i]); + + for (pi = pn; pi; pi = parent_nodeinfo(pi, node)) + for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++) + atomic_long_add(stat[i], &pi->lruvec_stat[i]); + } +} + static void memcg_offline_kmem(struct mem_cgroup *memcg) { struct cgroup_subsys_state *css; @@ -4682,6 +4717,11 @@ static void __mem_cgroup_free(struct mem { int node; + /* + * Flush percpu vmstats to guarantee the value correctness + * on parent's and all ancestor levels. + */ + memcg_flush_percpu_vmstats(memcg); for_each_node(node) free_mem_cgroup_per_node_info(memcg, node); free_percpu(memcg->vmstats_percpu); _ Patches currently in -mm which might be from guro(a)fb.com are mm-memcontrol-flush-percpu-slab-vmstats-on-kmem-offlining.patch mm-memcontrol-flush-percpu-slab-vmstats-on-kmem-offlining-fix.patch mm-memcontrol-flush-percpu-vmevents-before-releasing-memcg.patch mm-memcontrol-switch-to-rcu-protection-in-drain_all_stock.patch

5 years, 4 months

1
0
0 0

RE: [tip:irq/urgent] genirq: Properly pair kobject_del() with kobject_add()

by Michael Kelley

From: Sasha Levin <sashal(a)kernel.org> Sent: Monday, August 19, 2019 2:28 PM > > This commit has been processed because it contains a "Fixes:" tag, > fixing commit: ecb3f394c5db genirq: Expose interrupt information through sysfs. > > The bot has tested the following trees: v5.2.9, v4.19.67, v4.14.139, v4.9.189. > > v5.2.9: Build failed! Errors: > kernel/irq/irqdesc.c:446:6: error: ‘irq_kobj_base’ undeclared (first use in this function); > did you mean ‘irq_kobj_type’? > > v4.19.67: Build failed! Errors: > kernel/irq/irqdesc.c:445:6: error: ‘irq_kobj_base’ undeclared (first use in this function); > did you mean ‘irq_kobj_type’? > > v4.14.139: Build failed! Errors: > kernel/irq/irqdesc.c:428:6: error: ‘irq_kobj_base’ undeclared (first use in this function); > did you mean ‘irq_kobj_type’? > > v4.9.189: Build failed! Errors: > kernel/irq/irqdesc.c:414:6: error: ‘irq_kobj_base’ undeclared (first use in this function); > did you mean ‘irq_kobj_type’? > > > NOTE: The patch will not be queued to stable trees until it is upstream. > > How should we proceed with this patch? Compile error occurs when CONFIG_SYSFS is not selected. It's probably cleanest to revert the current patch. I'll send out a new version that fixes the problem. Michael

5 years, 4 months

1
0
0 0

[PATCH v2 3/3] mm: memcontrol: flush percpu vmevents before releasing memcg

by Roman Gushchin

Similar to vmstats, percpu caching of local vmevents leads to an accumulation of errors on non-leaf levels. This happens because some leftovers may remain in percpu caches, so that they are never propagated up by the cgroup tree and just disappear into nonexistence with on releasing of the memory cgroup. To fix this issue let's accumulate and propagate percpu vmevents values before releasing the memory cgroup similar to what we're doing with vmstats. Since on cpu hotplug we do flush percpu vmstats anyway, we can iterate only over online cpus. Fixes: 42a300353577 ("mm: memcontrol: fix recursive statistics correctness & scalabilty") Signed-off-by: Roman Gushchin <guro(a)fb.com> Acked-by: Michal Hocko <mhocko(a)suse.com> Cc: Johannes Weiner <hannes(a)cmpxchg.org> Cc: Vladimir Davydov <vdavydov.dev(a)gmail.com> Cc: <stable(a)vger.kernel.org> --- mm/memcontrol.c | 22 +++++++++++++++++++++- 1 file changed, 21 insertions(+), 1 deletion(-) diff --git a/mm/memcontrol.c b/mm/memcontrol.c index ebd72b80c90b..3137de6a46f0 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -3430,6 +3430,25 @@ static void memcg_flush_percpu_vmstats(struct mem_cgroup *memcg, bool slab_only) } } +static void memcg_flush_percpu_vmevents(struct mem_cgroup *memcg) +{ + unsigned long events[NR_VM_EVENT_ITEMS]; + struct mem_cgroup *mi; + int cpu, i; + + for (i = 0; i < NR_VM_EVENT_ITEMS; i++) + events[i] = 0; + + for_each_online_cpu(cpu) + for (i = 0; i < NR_VM_EVENT_ITEMS; i++) + events[i] += raw_cpu_read( + memcg->vmstats_percpu->events[i]); + + for (mi = memcg; mi; mi = parent_mem_cgroup(mi)) + for (i = 0; i < NR_VM_EVENT_ITEMS; i++) + atomic_long_add(events[i], &mi->vmevents[i]); +} + #ifdef CONFIG_MEMCG_KMEM static int memcg_online_kmem(struct mem_cgroup *memcg) { @@ -4860,10 +4879,11 @@ static void __mem_cgroup_free(struct mem_cgroup *memcg) int node; /* - * Flush percpu vmstats to guarantee the value correctness + * Flush percpu vmstats and vmevents to guarantee the value correctness * on parent's and all ancestor levels. */ memcg_flush_percpu_vmstats(memcg, false); + memcg_flush_percpu_vmevents(memcg); for_each_node(node) free_mem_cgroup_per_node_info(memcg, node); free_percpu(memcg->vmstats_percpu); -- 2.21.0

5 years, 4 months

1
0
0 0

[PATCH v2 1/3] mm: memcontrol: flush percpu vmstats before releasing memcg

by Roman Gushchin

Percpu caching of local vmstats with the conditional propagation by the cgroup tree leads to an accumulation of errors on non-leaf levels. Let's imagine two nested memory cgroups A and A/B. Say, a process belonging to A/B allocates 100 pagecache pages on the CPU 0. The percpu cache will spill 3 times, so that 32*3=96 pages will be accounted to A/B and A atomic vmstat counters, 4 pages will remain in the percpu cache. Imagine A/B is nearby memory.max, so that every following allocation triggers a direct reclaim on the local CPU. Say, each such attempt will free 16 pages on a new cpu. That means every percpu cache will have -16 pages, except the first one, which will have 4 - 16 = -12. A/B and A atomic counters will not be touched at all. Now a user removes A/B. All percpu caches are freed and corresponding vmstat numbers are forgotten. A has 96 pages more than expected. As memory cgroups are created and destroyed, errors do accumulate. Even 1-2 pages differences can accumulate into large numbers. To fix this issue let's accumulate and propagate percpu vmstat values before releasing the memory cgroup. At this point these numbers are stable and cannot be changed. Since on cpu hotplug we do flush percpu vmstats anyway, we can iterate only over online cpus. Fixes: 42a300353577 ("mm: memcontrol: fix recursive statistics correctness & scalabilty") Signed-off-by: Roman Gushchin <guro(a)fb.com> Acked-by: Michal Hocko <mhocko(a)suse.com> Cc: Johannes Weiner <hannes(a)cmpxchg.org> Cc: Vladimir Davydov <vdavydov.dev(a)gmail.com> Cc: <stable(a)vger.kernel.org> --- mm/memcontrol.c | 40 ++++++++++++++++++++++++++++++++++++++++ 1 file changed, 40 insertions(+) diff --git a/mm/memcontrol.c b/mm/memcontrol.c index 3e821f34399f..818165d8de3f 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -3383,6 +3383,41 @@ static u64 mem_cgroup_read_u64(struct cgroup_subsys_state *css, } } +static void memcg_flush_percpu_vmstats(struct mem_cgroup *memcg) +{ + unsigned long stat[MEMCG_NR_STAT]; + struct mem_cgroup *mi; + int node, cpu, i; + + for (i = 0; i < MEMCG_NR_STAT; i++) + stat[i] = 0; + + for_each_online_cpu(cpu) + for (i = 0; i < MEMCG_NR_STAT; i++) + stat[i] += raw_cpu_read(memcg->vmstats_percpu->stat[i]); + + for (mi = memcg; mi; mi = parent_mem_cgroup(mi)) + for (i = 0; i < MEMCG_NR_STAT; i++) + atomic_long_add(stat[i], &mi->vmstats[i]); + + for_each_node(node) { + struct mem_cgroup_per_node *pn = memcg->nodeinfo[node]; + struct mem_cgroup_per_node *pi; + + for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++) + stat[i] = 0; + + for_each_online_cpu(cpu) + for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++) + stat[i] += raw_cpu_read( + pn->lruvec_stat_cpu->count[i]); + + for (pi = pn; pi; pi = parent_nodeinfo(pi, node)) + for (i = 0; i < NR_VM_NODE_STAT_ITEMS; i++) + atomic_long_add(stat[i], &pi->lruvec_stat[i]); + } +} + #ifdef CONFIG_MEMCG_KMEM static int memcg_online_kmem(struct mem_cgroup *memcg) { @@ -4805,6 +4840,11 @@ static void __mem_cgroup_free(struct mem_cgroup *memcg) { int node; + /* + * Flush percpu vmstats to guarantee the value correctness + * on parent's and all ancestor levels. + */ + memcg_flush_percpu_vmstats(memcg); for_each_node(node) free_mem_cgroup_per_node_info(memcg, node); free_percpu(memcg->vmstats_percpu); -- 2.21.0

5 years, 4 months

1
0
0 0

Re: [PATCH] HID: wacom: correct misreported EKR ring values

by Sasha Levin

On Mon, Aug 19, 2019 at 10:13:31AM -0700, Aaron Armstrong Skomra wrote: >On Mon, Aug 19, 2019 at 10:02 AM Sasha Levin <sashal(a)kernel.org> wrote: > >> On Mon, Aug 19, 2019 at 09:41:54AM -0700, Aaron Armstrong Skomra wrote: >> >The EKR ring claims a range of 0 to 71 but actually reports >> >values 1 to 72. The ring is used in relative mode so this >> >change should not affect users. >> > >> >Signed-off-by: Aaron Armstrong Skomra <aaron.skomra(a)wacom.com> >> >Fixes: 72b236d60218f ("HID: wacom: Add support for Express Key Remote.") >> >Cc: <stable(a)vger.kernel.org> # v4.3+ >> >Reviewed-by: Ping Cheng <ping.cheng(a)wacom.com> >> >Reviewed-by: Jason Gerecke <jason.gerecke(a)wacom.com> >> >--- >> >Patch specifically targeted to v4.9.189 >> >> Is this not a problem upstream as well? Why not? >> >> If it is, this patch will need to go upstream first, and then it'll get >> to stable branches from there. >> >> Hi Sasha, >I neglected my "--in-reply-to" in git send-email, I will resend. My >apologies. Ah, I see what happened. Looks good now, thanks! -- Thanks, Sasha

5 years, 4 months

1
0
0 0

[PATCH] HID: wacom: correct misreported EKR ring values

by Aaron Armstrong Skomra

The EKR ring claims a range of 0 to 71 but actually reports values 1 to 72. The ring is used in relative mode so this change should not affect users. Signed-off-by: Aaron Armstrong Skomra <aaron.skomra(a)wacom.com> Fixes: 72b236d60218f ("HID: wacom: Add support for Express Key Remote.") Cc: <stable(a)vger.kernel.org> # v4.3+ Reviewed-by: Ping Cheng <ping.cheng(a)wacom.com> Reviewed-by: Jason Gerecke <jason.gerecke(a)wacom.com> --- My first attempt at sending this patch specifically targeted to v4.9.189 neglected the "in-reply-to" in git send-email. My apologies. drivers/hid/wacom_wac.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/hid/wacom_wac.c b/drivers/hid/wacom_wac.c index 6c3bf8846b52..949761dd29ca 100644 --- a/drivers/hid/wacom_wac.c +++ b/drivers/hid/wacom_wac.c @@ -819,7 +819,7 @@ static int wacom_remote_irq(struct wacom_wac *wacom_wac, size_t len) input_report_key(input, BTN_BASE2, (data[11] & 0x02)); if (data[12] & 0x80) - input_report_abs(input, ABS_WHEEL, (data[12] & 0x7f)); + input_report_abs(input, ABS_WHEEL, (data[12] & 0x7f) - 1); else input_report_abs(input, ABS_WHEEL, 0); -- 2.17.1

5 years, 4 months

1
0
0 0

perf build failures in {v4.4,v4.9,v4.14}.y.queue

by Guenter Roeck

util/header.c: In function ‘perf_session__read_header’: util/header.c:2860:10: error: ‘data’ undeclared Culprit is "perf header: Fix divide by zero error if f_header.attr_size==0". Fix might be to replace "data->file.path" with "file->path" in the affected branches. Guenter

5 years, 4 months

3
2
0 0

[PATCH AUTOSEL 4.4 01/14] xtensa: fix build for cores with coprocessors

by Sasha Levin

From: Max Filippov <jcmvbkbc(a)gmail.com> [ Upstream commit e3cacb73e626d885b8cf24103fed0ae26518e3c4 ] Assembly entry/return abstraction change didn't add asmmacro.h include statement to coprocessor.S, resulting in references to undefined macros abi_entry and abi_ret on cores that define XTENSA_HAVE_COPROCESSORS. Fix that by including asm/asmmacro.h from the coprocessor.S. Signed-off-by: Max Filippov <jcmvbkbc(a)gmail.com> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- arch/xtensa/kernel/coprocessor.S | 1 + 1 file changed, 1 insertion(+) diff --git a/arch/xtensa/kernel/coprocessor.S b/arch/xtensa/kernel/coprocessor.S index a482df5df2b2c..c5d6285201252 100644 --- a/arch/xtensa/kernel/coprocessor.S +++ b/arch/xtensa/kernel/coprocessor.S @@ -14,6 +14,7 @@ #include <linux/linkage.h> #include <asm/asm-offsets.h> +#include <asm/asmmacro.h> #include <asm/processor.h> #include <asm/coprocessor.h> #include <asm/thread_info.h> -- 2.20.1

5 years, 4 months

5
17
0 0

[PATCH] HID: wacom: correct misreported EKR ring values

by Aaron Armstrong Skomra

The EKR ring claims a range of 0 to 71 but actually reports values 1 to 72. The ring is used in relative mode so this change should not affect users. Signed-off-by: Aaron Armstrong Skomra <aaron.skomra(a)wacom.com> Fixes: 72b236d60218f ("HID: wacom: Add support for Express Key Remote.") Cc: <stable(a)vger.kernel.org> # v4.3+ Reviewed-by: Ping Cheng <ping.cheng(a)wacom.com> Reviewed-by: Jason Gerecke <jason.gerecke(a)wacom.com> --- Patch specifically targeted to v4.9.189 drivers/hid/wacom_wac.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/hid/wacom_wac.c b/drivers/hid/wacom_wac.c index 6c3bf8846b52..949761dd29ca 100644 --- a/drivers/hid/wacom_wac.c +++ b/drivers/hid/wacom_wac.c @@ -819,7 +819,7 @@ static int wacom_remote_irq(struct wacom_wac *wacom_wac, size_t len) input_report_key(input, BTN_BASE2, (data[11] & 0x02)); if (data[12] & 0x80) - input_report_abs(input, ABS_WHEEL, (data[12] & 0x7f)); + input_report_abs(input, ABS_WHEEL, (data[12] & 0x7f) - 1); else input_report_abs(input, ABS_WHEEL, 0); -- 2.17.1

5 years, 4 months

2
1
0 0

[PATCH 4.14 1/4] bpf: get rid of pure_initcall dependency to enable jits

by Ben Hutchings

From: Daniel Borkmann <daniel(a)iogearbox.net> commit fa9dd599b4dae841924b022768354cfde9affecb upstream. Having a pure_initcall() callback just to permanently enable BPF JITs under CONFIG_BPF_JIT_ALWAYS_ON is unnecessary and could leave a small race window in future where JIT is still disabled on boot. Since we know about the setting at compilation time anyway, just initialize it properly there. Also consolidate all the individual bpf_jit_enable variables into a single one and move them under one location. Moreover, don't allow for setting unspecified garbage values on them. Signed-off-by: Daniel Borkmann <daniel(a)iogearbox.net> Acked-by: Alexei Starovoitov <ast(a)kernel.org> Signed-off-by: Alexei Starovoitov <ast(a)kernel.org> [bwh: Backported to 4.14 as dependency of commit 2e4a30983b0f "bpf: restrict access to core bpf sysctls": - Adjust context] Signed-off-by: Ben Hutchings <ben.hutchings(a)codethink.co.uk> --- arch/arm/net/bpf_jit_32.c | 2 -- arch/arm64/net/bpf_jit_comp.c | 2 -- arch/mips/net/bpf_jit.c | 2 -- arch/mips/net/ebpf_jit.c | 2 -- arch/powerpc/net/bpf_jit_comp.c | 2 -- arch/powerpc/net/bpf_jit_comp64.c | 2 -- arch/s390/net/bpf_jit_comp.c | 2 -- arch/sparc/net/bpf_jit_comp_32.c | 2 -- arch/sparc/net/bpf_jit_comp_64.c | 2 -- arch/x86/net/bpf_jit_comp.c | 2 -- kernel/bpf/core.c | 19 ++++++++++++------- net/core/sysctl_net_core.c | 18 ++++++++++++------ net/socket.c | 9 --------- 13 files changed, 24 insertions(+), 42 deletions(-) diff --git a/arch/arm/net/bpf_jit_32.c b/arch/arm/net/bpf_jit_32.c index dafeb5f81353..b18fb70c5dcf 100644 --- a/arch/arm/net/bpf_jit_32.c +++ b/arch/arm/net/bpf_jit_32.c @@ -25,8 +25,6 @@ #include "bpf_jit_32.h" -int bpf_jit_enable __read_mostly; - /* * eBPF prog stack layout: * diff --git a/arch/arm64/net/bpf_jit_comp.c b/arch/arm64/net/bpf_jit_comp.c index b742171bfef7..1bbb457c293f 100644 --- a/arch/arm64/net/bpf_jit_comp.c +++ b/arch/arm64/net/bpf_jit_comp.c @@ -31,8 +31,6 @@ #include "bpf_jit.h" -int bpf_jit_enable __read_mostly; - #define TMP_REG_1 (MAX_BPF_JIT_REG + 0) #define TMP_REG_2 (MAX_BPF_JIT_REG + 1) #define TCALL_CNT (MAX_BPF_JIT_REG + 2) diff --git a/arch/mips/net/bpf_jit.c b/arch/mips/net/bpf_jit.c index 44b925005dd3..4d8cb9bb8365 100644 --- a/arch/mips/net/bpf_jit.c +++ b/arch/mips/net/bpf_jit.c @@ -1207,8 +1207,6 @@ static int build_body(struct jit_ctx *ctx) return 0; } -int bpf_jit_enable __read_mostly; - void bpf_jit_compile(struct bpf_prog *fp) { struct jit_ctx ctx; diff --git a/arch/mips/net/ebpf_jit.c b/arch/mips/net/ebpf_jit.c index 8004bfcfb033..42faa95ce664 100644 --- a/arch/mips/net/ebpf_jit.c +++ b/arch/mips/net/ebpf_jit.c @@ -177,8 +177,6 @@ static u32 b_imm(unsigned int tgt, struct jit_ctx *ctx) (ctx->idx * 4) - 4; } -int bpf_jit_enable __read_mostly; - enum which_ebpf_reg { src_reg, src_reg_no_fp, diff --git a/arch/powerpc/net/bpf_jit_comp.c b/arch/powerpc/net/bpf_jit_comp.c index f760494ecd66..a9636d8cba15 100644 --- a/arch/powerpc/net/bpf_jit_comp.c +++ b/arch/powerpc/net/bpf_jit_comp.c @@ -18,8 +18,6 @@ #include "bpf_jit32.h" -int bpf_jit_enable __read_mostly; - static inline void bpf_flush_icache(void *start, void *end) { smp_wmb(); diff --git a/arch/powerpc/net/bpf_jit_comp64.c b/arch/powerpc/net/bpf_jit_comp64.c index 70e8216a39f0..28434040cfb6 100644 --- a/arch/powerpc/net/bpf_jit_comp64.c +++ b/arch/powerpc/net/bpf_jit_comp64.c @@ -21,8 +21,6 @@ #include "bpf_jit64.h" -int bpf_jit_enable __read_mostly; - static void bpf_jit_fill_ill_insns(void *area, unsigned int size) { memset32(area, BREAKPOINT_INSTRUCTION, size/4); diff --git a/arch/s390/net/bpf_jit_comp.c b/arch/s390/net/bpf_jit_comp.c index 6b1474fa99ab..bc9431aace05 100644 --- a/arch/s390/net/bpf_jit_comp.c +++ b/arch/s390/net/bpf_jit_comp.c @@ -30,8 +30,6 @@ #include <asm/set_memory.h> #include "bpf_jit.h" -int bpf_jit_enable __read_mostly; - struct bpf_jit { u32 seen; /* Flags to remember seen eBPF instructions */ u32 seen_reg[16]; /* Array to remember which registers are used */ diff --git a/arch/sparc/net/bpf_jit_comp_32.c b/arch/sparc/net/bpf_jit_comp_32.c index 09e318eb34ee..3bd8ca95e521 100644 --- a/arch/sparc/net/bpf_jit_comp_32.c +++ b/arch/sparc/net/bpf_jit_comp_32.c @@ -11,8 +11,6 @@ #include "bpf_jit_32.h" -int bpf_jit_enable __read_mostly; - static inline bool is_simm13(unsigned int value) { return value + 0x1000 < 0x2000; diff --git a/arch/sparc/net/bpf_jit_comp_64.c b/arch/sparc/net/bpf_jit_comp_64.c index ff5f9cb3039a..adfb4581bd80 100644 --- a/arch/sparc/net/bpf_jit_comp_64.c +++ b/arch/sparc/net/bpf_jit_comp_64.c @@ -12,8 +12,6 @@ #include "bpf_jit_64.h" -int bpf_jit_enable __read_mostly; - static inline bool is_simm13(unsigned int value) { return value + 0x1000 < 0x2000; diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c index a9deb2b0397d..cdb386fa7101 100644 --- a/arch/x86/net/bpf_jit_comp.c +++ b/arch/x86/net/bpf_jit_comp.c @@ -16,8 +16,6 @@ #include <asm/nospec-branch.h> #include <linux/bpf.h> -int bpf_jit_enable __read_mostly; - /* * assembly code in arch/x86/net/bpf_jit.S */ diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c index e46106c6ac39..661fb837b168 100644 --- a/kernel/bpf/core.c +++ b/kernel/bpf/core.c @@ -290,6 +290,11 @@ struct bpf_prog *bpf_patch_insn_single(struct bpf_prog *prog, u32 off, } #ifdef CONFIG_BPF_JIT +/* All BPF JIT sysctl knobs here. */ +int bpf_jit_enable __read_mostly = IS_BUILTIN(CONFIG_BPF_JIT_ALWAYS_ON); +int bpf_jit_harden __read_mostly; +int bpf_jit_kallsyms __read_mostly; + static __always_inline void bpf_get_prog_addr_region(const struct bpf_prog *prog, unsigned long *symbol_start, @@ -358,8 +363,6 @@ static DEFINE_SPINLOCK(bpf_lock); static LIST_HEAD(bpf_kallsyms); static struct latch_tree_root bpf_tree __cacheline_aligned; -int bpf_jit_kallsyms __read_mostly; - static void bpf_prog_ksym_node_add(struct bpf_prog_aux *aux) { WARN_ON_ONCE(!list_empty(&aux->ksym_lnode)); @@ -540,8 +543,6 @@ void __weak bpf_jit_free(struct bpf_prog *fp) bpf_prog_unlock_free(fp); } -int bpf_jit_harden __read_mostly; - static int bpf_jit_blind_insn(const struct bpf_insn *from, const struct bpf_insn *aux, struct bpf_insn *to_buff) @@ -1327,9 +1328,13 @@ EVAL4(PROG_NAME_LIST, 416, 448, 480, 512) }; #else -static unsigned int __bpf_prog_ret0(const void *ctx, - const struct bpf_insn *insn) +static unsigned int __bpf_prog_ret0_warn(const void *ctx, + const struct bpf_insn *insn) { + /* If this handler ever gets executed, then BPF_JIT_ALWAYS_ON + * is not working properly, so warn about it! + */ + WARN_ON_ONCE(1); return 0; } #endif @@ -1386,7 +1391,7 @@ struct bpf_prog *bpf_prog_select_runtime(struct bpf_prog *fp, int *err) fp->bpf_func = interpreters[(round_up(stack_depth, 32) / 32) - 1]; #else - fp->bpf_func = __bpf_prog_ret0; + fp->bpf_func = __bpf_prog_ret0_warn; #endif /* eBPF JITs can rewrite the program in case constant diff --git a/net/core/sysctl_net_core.c b/net/core/sysctl_net_core.c index a47ad6cd41c0..6d39b4c01fc6 100644 --- a/net/core/sysctl_net_core.c +++ b/net/core/sysctl_net_core.c @@ -25,6 +25,7 @@ static int zero = 0; static int one = 1; +static int two __maybe_unused = 2; static int min_sndbuf = SOCK_MIN_SNDBUF; static int min_rcvbuf = SOCK_MIN_RCVBUF; static int max_skb_frags = MAX_SKB_FRAGS; @@ -325,13 +326,14 @@ static struct ctl_table net_core_table[] = { .data = &bpf_jit_enable, .maxlen = sizeof(int), .mode = 0644, -#ifndef CONFIG_BPF_JIT_ALWAYS_ON - .proc_handler = proc_dointvec -#else .proc_handler = proc_dointvec_minmax, +# ifdef CONFIG_BPF_JIT_ALWAYS_ON .extra1 = &one, .extra2 = &one, -#endif +# else + .extra1 = &zero, + .extra2 = &two, +# endif }, # ifdef CONFIG_HAVE_EBPF_JIT { @@ -339,14 +341,18 @@ static struct ctl_table net_core_table[] = { .data = &bpf_jit_harden, .maxlen = sizeof(int), .mode = 0600, - .proc_handler = proc_dointvec, + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &two, }, { .procname = "bpf_jit_kallsyms", .data = &bpf_jit_kallsyms, .maxlen = sizeof(int), .mode = 0600, - .proc_handler = proc_dointvec, + .proc_handler = proc_dointvec_minmax, + .extra1 = &zero, + .extra2 = &one, }, # endif #endif diff --git a/net/socket.c b/net/socket.c index 6d8f0c248c7e..aab65277314d 100644 --- a/net/socket.c +++ b/net/socket.c @@ -2656,15 +2656,6 @@ static int __init sock_init(void) core_initcall(sock_init); /* early initcall */ -static int __init jit_init(void) -{ -#ifdef CONFIG_BPF_JIT_ALWAYS_ON - bpf_jit_enable = 1; -#endif - return 0; -} -pure_initcall(jit_init); - #ifdef CONFIG_PROC_FS void socket_seq_show(struct seq_file *seq) { -- Ben Hutchings, Software Developer Codethink Ltd https://www.codethink.co.uk/ Dale House, 35 Dale Street Manchester, M1 2HF, United Kingdom

5 years, 4 months

3
19
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

Linux-stable-mirror August 2019