* Re: [PATCH 2/2] powerpc/64s/hash: add torture_hpt kernel boot option to increase hash faults
From: Aneesh Kumar K.V @ 2020-05-04 7:36 UTC (permalink / raw)
To: Nicholas Piggin, linuxppc-dev; +Cc: Nicholas Piggin
In-Reply-To: <20200503082236.17991-2-npiggin@gmail.com>
Nicholas Piggin <npiggin@gmail.com> writes:
> This option increases the number of hash misses by limiting the number of
> kernel HPT entries. This helps stress test difficult to hit paths in the
> kernel.
>
It would nice if we can explain in commit message how we are limiting
the number of HPT entries.
> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
> ---
> .../admin-guide/kernel-parameters.txt | 9 +++
> arch/powerpc/include/asm/book3s/64/mmu-hash.h | 10 +++
> arch/powerpc/mm/book3s64/hash_4k.c | 3 +
> arch/powerpc/mm/book3s64/hash_64k.c | 8 +++
> arch/powerpc/mm/book3s64/hash_utils.c | 66 ++++++++++++++++++-
> 5 files changed, 95 insertions(+), 1 deletion(-)
....
> +void hpt_do_torture(unsigned long ea, unsigned long access,
> + unsigned long rflags, unsigned long hpte_group)
> +{
> + unsigned long last_group;
> + int cpu = raw_smp_processor_id();
> +
> + last_group = torture_hpt_last_group[cpu];
> + if (last_group != -1UL) {
> + while (mmu_hash_ops.hpte_remove(last_group) != -1)
> + ;
> + torture_hpt_last_group[cpu] = -1UL;
> + }
> +
> +#define QEMU_WORKAROUND 0
> +
> + if (ea >= PAGE_OFFSET) {
> + if (!QEMU_WORKAROUND && (access & (_PAGE_READ|_PAGE_WRITE)) &&
> + !(rflags & (HPTE_R_I|HPTE_R_G))) {
> + /* prefetch / prefetchw does not seem to set up a TLB
> + * entry with the powerpc systemsim (mambo) emulator,
> + * though it works with real hardware. An alternative
> + * approach that would work more reliably on quirky
> + * emulators like QEMU may be to remember the last
> + * insertion and remove that, rather than removing the
> + * current insertion. Then no prefetch is required.
> + */
> + if ((access & _PAGE_WRITE) && (access & _PAGE_READ))
> + atomic_add(0, (atomic_t *)(ea & ~0x3));
> + else if (access & _PAGE_READ)
> + *(volatile char *)ea;
> +
> + mb();
> +
> + while (mmu_hash_ops.hpte_remove(hpte_group) != -1)
> + ;
Do we get similar hpte faults rate, if we remove everything except the
current inserted entry?. If so that would largely simplify the code.
> + } else {
> + /* Can't prefetch cache-inhibited so clear next time. */
> + torture_hpt_last_group[cpu] = hpte_group;
> + }
> + }
> +}
> +
> +
> #ifdef CONFIG_DEBUG_PAGEALLOC
> static void kernel_map_linear_page(unsigned long vaddr, unsigned long lmi)
> {
> --
> 2.23.0
-aneesh
^ permalink raw reply
* Re: [PATCH v7 08/28] powerpc: Use a function for getting the instruction op code
From: Alistair Popple @ 2020-05-04 8:01 UTC (permalink / raw)
To: Jordan Niethe; +Cc: npiggin, bala24, naveen.n.rao, linuxppc-dev, dja
In-Reply-To: <20200501034220.8982-9-jniethe5@gmail.com>
Looks good to me in that it doesn't look to change the behaviour of any
existing code.
Reviewed-by: Alistair Popple <alistair@popple.id.au>
On Friday, 1 May 2020 1:42:00 PM AEST Jordan Niethe wrote:
> In preparation for using a data type for instructions that can not be
> directly used with the '>>' operator use a function for getting the op
> code of an instruction.
>
> Signed-off-by: Jordan Niethe <jniethe5@gmail.com>
> ---
> v4: New to series
> v6: - Rename ppc_inst_primary() to ppc_inst_primary_opcode()
> - Use in vecemu.c, fault.c, sstep.c
> - Move this patch after the ppc_inst_val() patch
> ---
> arch/powerpc/include/asm/inst.h | 5 +++++
> arch/powerpc/kernel/align.c | 2 +-
> arch/powerpc/kernel/vecemu.c | 3 ++-
> arch/powerpc/lib/code-patching.c | 4 ++--
> arch/powerpc/lib/sstep.c | 2 +-
> arch/powerpc/mm/fault.c | 3 ++-
> 6 files changed, 13 insertions(+), 6 deletions(-)
>
> diff --git a/arch/powerpc/include/asm/inst.h
> b/arch/powerpc/include/asm/inst.h index 8a9e73bfbd27..442a95f20de7 100644
> --- a/arch/powerpc/include/asm/inst.h
> +++ b/arch/powerpc/include/asm/inst.h
> @@ -13,4 +13,9 @@ static inline u32 ppc_inst_val(u32 x)
> return x;
> }
>
> +static inline int ppc_inst_primary_opcode(u32 x)
> +{
> + return ppc_inst_val(x) >> 26;
> +}
> +
> #endif /* _ASM_INST_H */
> diff --git a/arch/powerpc/kernel/align.c b/arch/powerpc/kernel/align.c
> index 44921001f84a..47dbba81a227 100644
> --- a/arch/powerpc/kernel/align.c
> +++ b/arch/powerpc/kernel/align.c
> @@ -314,7 +314,7 @@ int fix_alignment(struct pt_regs *regs)
> }
>
> #ifdef CONFIG_SPE
> - if ((ppc_inst_val(instr) >> 26) == 0x4) {
> + if (ppc_inst_primary_opcode(instr) == 0x4) {
> int reg = (ppc_inst_val(instr) >> 21) & 0x1f;
> PPC_WARN_ALIGNMENT(spe, regs);
> return emulate_spe(regs, reg, instr);
> diff --git a/arch/powerpc/kernel/vecemu.c b/arch/powerpc/kernel/vecemu.c
> index 1f5e3b4c8ae4..a544590b90e5 100644
> --- a/arch/powerpc/kernel/vecemu.c
> +++ b/arch/powerpc/kernel/vecemu.c
> @@ -10,6 +10,7 @@
> #include <asm/processor.h>
> #include <asm/switch_to.h>
> #include <linux/uaccess.h>
> +#include <asm/inst.h>
>
> /* Functions in vector.S */
> extern void vaddfp(vector128 *dst, vector128 *a, vector128 *b);
> @@ -268,7 +269,7 @@ int emulate_altivec(struct pt_regs *regs)
> return -EFAULT;
>
> word = ppc_inst_val(instr);
> - if ((word >> 26) != 4)
> + if (ppc_inst_primary_opcode(instr) != 4)
> return -EINVAL; /* not an altivec instruction */
> vd = (word >> 21) & 0x1f;
> va = (word >> 16) & 0x1f;
> diff --git a/arch/powerpc/lib/code-patching.c
> b/arch/powerpc/lib/code-patching.c index baa849b1a1f9..f5c6dcbac44b 100644
> --- a/arch/powerpc/lib/code-patching.c
> +++ b/arch/powerpc/lib/code-patching.c
> @@ -231,7 +231,7 @@ bool is_offset_in_branch_range(long offset)
> */
> bool is_conditional_branch(unsigned int instr)
> {
> - unsigned int opcode = instr >> 26;
> + unsigned int opcode = ppc_inst_primary_opcode(instr);
>
> if (opcode == 16) /* bc, bca, bcl, bcla */
> return true;
> @@ -289,7 +289,7 @@ int create_cond_branch(unsigned int *instr, const
> unsigned int *addr,
>
> static unsigned int branch_opcode(unsigned int instr)
> {
> - return (instr >> 26) & 0x3F;
> + return ppc_inst_primary_opcode(instr) & 0x3F;
> }
>
> static int instr_is_branch_iform(unsigned int instr)
> diff --git a/arch/powerpc/lib/sstep.c b/arch/powerpc/lib/sstep.c
> index 14c93ee4ffc8..7f7be154da7e 100644
> --- a/arch/powerpc/lib/sstep.c
> +++ b/arch/powerpc/lib/sstep.c
> @@ -1175,7 +1175,7 @@ int analyse_instr(struct instruction_op *op, const
> struct pt_regs *regs, word = ppc_inst_val(instr);
> op->type = COMPUTE;
>
> - opcode = instr >> 26;
> + opcode = ppc_inst_primary_opcode(instr);
> switch (opcode) {
> case 16: /* bc */
> op->type = BRANCH;
> diff --git a/arch/powerpc/mm/fault.c b/arch/powerpc/mm/fault.c
> index 9364921870df..0e7e145d5cad 100644
> --- a/arch/powerpc/mm/fault.c
> +++ b/arch/powerpc/mm/fault.c
> @@ -41,6 +41,7 @@
> #include <asm/siginfo.h>
> #include <asm/debug.h>
> #include <asm/kup.h>
> +#include <asm/inst.h>
>
> /*
> * Check whether the instruction inst is a store using
> @@ -52,7 +53,7 @@ static bool store_updates_sp(unsigned int inst)
> if (((ppc_inst_val(inst) >> 16) & 0x1f) != 1)
> return false;
> /* check major opcode */
> - switch (inst >> 26) {
> + switch (ppc_inst_primary_opcode(inst)) {
> case OP_STWU:
> case OP_STBU:
> case OP_STHU:
^ permalink raw reply
* Re: [PATCH 2/2] powerpc/64s/hash: add torture_hpt kernel boot option to increase hash faults
From: Nicholas Piggin @ 2020-05-04 8:06 UTC (permalink / raw)
To: Aneesh Kumar K.V, linuxppc-dev
In-Reply-To: <874kswm9s6.fsf@linux.ibm.com>
Excerpts from Aneesh Kumar K.V's message of May 4, 2020 5:36 pm:
> Nicholas Piggin <npiggin@gmail.com> writes:
>
>> This option increases the number of hash misses by limiting the number of
>> kernel HPT entries. This helps stress test difficult to hit paths in the
>> kernel.
>>
>
> It would nice if we can explain in commit message how we are limiting
> the number of HPT entries.
"limiting the number of kernel HPT entries by removing them as soon as
possible after they are installed"?
>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>> ---
>> .../admin-guide/kernel-parameters.txt | 9 +++
>> arch/powerpc/include/asm/book3s/64/mmu-hash.h | 10 +++
>> arch/powerpc/mm/book3s64/hash_4k.c | 3 +
>> arch/powerpc/mm/book3s64/hash_64k.c | 8 +++
>> arch/powerpc/mm/book3s64/hash_utils.c | 66 ++++++++++++++++++-
>> 5 files changed, 95 insertions(+), 1 deletion(-)
>
> ....
>
>
>> +void hpt_do_torture(unsigned long ea, unsigned long access,
>> + unsigned long rflags, unsigned long hpte_group)
>> +{
>> + unsigned long last_group;
>> + int cpu = raw_smp_processor_id();
>> +
>> + last_group = torture_hpt_last_group[cpu];
>> + if (last_group != -1UL) {
>> + while (mmu_hash_ops.hpte_remove(last_group) != -1)
>> + ;
>> + torture_hpt_last_group[cpu] = -1UL;
>> + }
>> +
>> +#define QEMU_WORKAROUND 0
>> +
>> + if (ea >= PAGE_OFFSET) {
>> + if (!QEMU_WORKAROUND && (access & (_PAGE_READ|_PAGE_WRITE)) &&
>> + !(rflags & (HPTE_R_I|HPTE_R_G))) {
>> + /* prefetch / prefetchw does not seem to set up a TLB
>> + * entry with the powerpc systemsim (mambo) emulator,
>> + * though it works with real hardware. An alternative
>> + * approach that would work more reliably on quirky
>> + * emulators like QEMU may be to remember the last
>> + * insertion and remove that, rather than removing the
>> + * current insertion. Then no prefetch is required.
>> + */
>> + if ((access & _PAGE_WRITE) && (access & _PAGE_READ))
>> + atomic_add(0, (atomic_t *)(ea & ~0x3));
>> + else if (access & _PAGE_READ)
>> + *(volatile char *)ea;
>> +
>> + mb();
>> +
>> + while (mmu_hash_ops.hpte_remove(hpte_group) != -1)
>> + ;
>
> Do we get similar hpte faults rate, if we remove everything except the
> current inserted entry?. If so that would largely simplify the code.
Well it would remove this one branch at least. It does actually help
cause more faults and helps (in theory) irritate cases where you have
two accesses to a vmalloc page in the kernel, where the first is okay
to take a fault but the second is buggy.
I actually like the interesting behaviour it exposes in emulators too.
We should really fix mambo to have prefetches bring in TLBs, and fix
qemu to bring in TLBs more like hardware, and it could have TLB vs PTE
consistency checks to catch bugs like mambo does.
So I prefer leaving this in.
Thanks,
Nick
^ permalink raw reply
* Re: [PATCH 1/2] powerpc/64s/hash: add torture_slb kernel boot option to increase SLB faults
From: Nicholas Piggin @ 2020-05-04 8:19 UTC (permalink / raw)
To: Aneesh Kumar K.V, linuxppc-dev
In-Reply-To: <877dxsma7p.fsf@linux.ibm.com>
Excerpts from Aneesh Kumar K.V's message of May 4, 2020 5:27 pm:
> Nicholas Piggin <npiggin@gmail.com> writes:
>
>> This option increases the number of SLB misses by limiting the number of
>> kernel SLB entries, and increased flushing of cached lookaside information.
>> This helps stress test difficult to hit paths in the kernel.
>>
>> Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
>
> ....
>
>> +{
>> + unsigned long slbie_data = get_paca()->slb_cache[index];
>> + unsigned long ksp = get_paca()->kstack;
>> +
>> + slbie_data <<= SID_SHIFT;
>> + slbie_data |= 0xc000000000000000ULL;
>> + if ((ksp & slb_esid_mask(mmu_kernel_ssize)) == slbie_data)
>> + return;
>> + slbie_data |= mmu_kernel_ssize << SLBIE_SSIZE_SHIFT;
>> +
>> + asm volatile("slbie %0" : : "r" (slbie_data));
>> +}
>> +
>> +static void slb_cache_slbie(unsigned int index)
>
> May be slb_cache_slbie_user()? Similar to _kernel above?
Yeah that'd help.
>> +{
>> + unsigned long slbie_data = get_paca()->slb_cache[index];
>> +
>> + slbie_data <<= SID_SHIFT;
>> + slbie_data |= user_segment_size(slbie_data) << SLBIE_SSIZE_SHIFT;
>> + slbie_data |= SLBIE_C; /* user slbs have C=1 */
>> +
>> + asm volatile("slbie %0" : : "r" (slbie_data));
>> +}
>>
>> /* Flush all user entries from the segment table of the current processor. */
>> void switch_slb(struct task_struct *tsk, struct mm_struct *mm)
>> @@ -414,8 +449,14 @@ void switch_slb(struct task_struct *tsk, struct mm_struct *mm)
>> * which would update the slb_cache/slb_cache_ptr fields in the PACA.
>> */
>> hard_irq_disable();
>> - asm volatile("isync" : : : "memory");
>> - if (cpu_has_feature(CPU_FTR_ARCH_300)) {
>> + isync();
>> + if (torture_slb()) {
>> + __slb_flush_and_restore_bolted(0);
>
> s/0/SLIBA_IH_ALL or something like that?
IH isn't so simple. 0 isn't all, it's clear all SLBE except index zero,
and flush all EA lookaside data.
Maybe callers should use a bool parameter though to flush kernel
lookaside data.
Thanks,
Nick
^ permalink raw reply
* Re: [PATCH v7 12/28] powerpc: Use a function for reading instructions
From: Alistair Popple @ 2020-05-04 8:26 UTC (permalink / raw)
To: Jordan Niethe; +Cc: npiggin, bala24, naveen.n.rao, linuxppc-dev, dja
In-Reply-To: <20200501034220.8982-13-jniethe5@gmail.com>
> diff --git a/arch/powerpc/kernel/uprobes.c b/arch/powerpc/kernel/uprobes.c
> index 31c870287f2b..6893d40a48c5 100644
> --- a/arch/powerpc/kernel/uprobes.c
> +++ b/arch/powerpc/kernel/uprobes.c
> @@ -174,7 +174,7 @@ bool arch_uprobe_skip_sstep(struct arch_uprobe *auprobe,
> struct pt_regs *regs) * emulate_step() returns 1 if the insn was
> successfully emulated. * For all other cases, we need to single-step in
> hardware.
> */
> - ret = emulate_step(regs, auprobe->insn);
> + ret = emulate_step(regs, ppc_inst_read(&auprobe->insn));
I'm not a uprobe expert so I don't follow why we need this read here but the
rest of the patch looked ok in that it shouldn't change behaviour (and in
practice neither should the above) so:
Reviewed-by: Alistair Popple <alistair@popple.id.au>
> if (ret > 0)
> return true;
>
^ permalink raw reply
* Re: [PATCH v7 13/28] powerpc: Add a probe_user_read_inst() function
From: Alistair Popple @ 2020-05-04 8:30 UTC (permalink / raw)
To: Jordan Niethe; +Cc: npiggin, bala24, naveen.n.rao, linuxppc-dev, dja
In-Reply-To: <20200501034220.8982-14-jniethe5@gmail.com>
Looks good.
Reviewed-by: Alistair Popple <alistair@popple.id.au>
On Friday, 1 May 2020 1:42:05 PM AEST Jordan Niethe wrote:
> Introduce a probe_user_read_inst() function to use in cases where
> probe_user_read() is used for getting an instruction. This will be more
> useful for prefixed instructions.
>
> Signed-off-by: Jordan Niethe <jniethe5@gmail.com>
> ---
> v6: - New to series
> ---
> arch/powerpc/include/asm/inst.h | 3 +++
> arch/powerpc/lib/Makefile | 2 +-
> arch/powerpc/lib/inst.c | 18 ++++++++++++++++++
> arch/powerpc/mm/fault.c | 2 +-
> 4 files changed, 23 insertions(+), 2 deletions(-)
> create mode 100644 arch/powerpc/lib/inst.c
>
> diff --git a/arch/powerpc/include/asm/inst.h
> b/arch/powerpc/include/asm/inst.h index 552e953bf04f..3e9a58420151 100644
> --- a/arch/powerpc/include/asm/inst.h
> +++ b/arch/powerpc/include/asm/inst.h
> @@ -37,4 +37,7 @@ static inline bool ppc_inst_equal(struct ppc_inst x,
> struct ppc_inst y) return ppc_inst_val(x) == ppc_inst_val(y);
> }
>
> +int probe_user_read_inst(struct ppc_inst *inst,
> + struct ppc_inst *nip);
> +
> #endif /* _ASM_INST_H */
> diff --git a/arch/powerpc/lib/Makefile b/arch/powerpc/lib/Makefile
> index b8de3be10eb4..546591848219 100644
> --- a/arch/powerpc/lib/Makefile
> +++ b/arch/powerpc/lib/Makefile
> @@ -16,7 +16,7 @@ CFLAGS_code-patching.o += -DDISABLE_BRANCH_PROFILING
> CFLAGS_feature-fixups.o += -DDISABLE_BRANCH_PROFILING
> endif
>
> -obj-y += alloc.o code-patching.o feature-fixups.o pmem.o
> +obj-y += alloc.o code-patching.o feature-fixups.o pmem.o inst.o
>
> ifndef CONFIG_KASAN
> obj-y += string.o memcmp_$(BITS).o
> diff --git a/arch/powerpc/lib/inst.c b/arch/powerpc/lib/inst.c
> new file mode 100644
> index 000000000000..eaf786afad2b
> --- /dev/null
> +++ b/arch/powerpc/lib/inst.c
> @@ -0,0 +1,18 @@
> +// SPDX-License-Identifier: GPL-2.0-or-later
> +/*
> + * Copyright 2020, IBM Corporation.
> + */
> +
> +#include <linux/uaccess.h>
> +#include <asm/inst.h>
> +
> +int probe_user_read_inst(struct ppc_inst *inst,
> + struct ppc_inst *nip)
> +{
> + unsigned int val;
> + int err;
> +
> + err = probe_user_read(&val, nip, sizeof(val));
> + *inst = ppc_inst(val);
> + return err;
> +}
> diff --git a/arch/powerpc/mm/fault.c b/arch/powerpc/mm/fault.c
> index 4a50f125ec18..f3a943eae305 100644
> --- a/arch/powerpc/mm/fault.c
> +++ b/arch/powerpc/mm/fault.c
> @@ -281,7 +281,7 @@ static bool bad_stack_expansion(struct pt_regs *regs,
> unsigned long address, access_ok(nip, sizeof(*nip))) {
> struct ppc_inst inst;
>
> - if (!probe_user_read(&inst, nip, sizeof(inst)))
> + if (!probe_user_read_inst(&inst, (struct ppc_inst __user *)nip))
> return !store_updates_sp(inst);
> *must_retry = true;
> }
^ permalink raw reply
* Re: [PATCH v6 1/4] powerpc: Document details on H_SCM_HEALTH hcall
From: Vaibhav Jain @ 2020-05-04 8:52 UTC (permalink / raw)
To: Michael Ellerman, linuxppc-dev, linux-nvdimm; +Cc: Aneesh Kumar K . V
In-Reply-To: <87r1w5echa.fsf@mpe.ellerman.id.au>
Thanks for looking into this patch Mpe,
Michael Ellerman <mpe@ellerman.id.au> writes:
> Vaibhav Jain <vaibhav@linux.ibm.com> writes:
>
>> Add documentation to 'papr_hcalls.rst' describing the bitmap flags
>> that are returned from H_SCM_HEALTH hcall as per the PAPR-SCM
>> specification.
>>
>> Cc: Dan Williams <dan.j.williams@intel.com>
>> Cc: Michael Ellerman <mpe@ellerman.id.au>
>> Cc: "Aneesh Kumar K . V" <aneesh.kumar@linux.ibm.com>
>> Signed-off-by: Vaibhav Jain <vaibhav@linux.ibm.com>
>> ---
>> Changelog:
>>
>> v5..v6
>> * New patch in the series
>> ---
>> Documentation/powerpc/papr_hcalls.rst | 43 ++++++++++++++++++++++++---
>> 1 file changed, 39 insertions(+), 4 deletions(-)
>>
>> diff --git a/Documentation/powerpc/papr_hcalls.rst b/Documentation/powerpc/papr_hcalls.rst
>> index 3493631a60f8..9a5ba5eaf323 100644
>> --- a/Documentation/powerpc/papr_hcalls.rst
>> +++ b/Documentation/powerpc/papr_hcalls.rst
>> @@ -220,13 +220,48 @@ from the LPAR memory.
>> **H_SCM_HEALTH**
>>
>> | Input: drcIndex
>> -| Out: *health-bitmap, health-bit-valid-bitmap*
>> +| Out: *health-bitmap (r4), health-bit-valid-bitmap (r5)*
>> | Return Value: *H_Success, H_Parameter, H_Hardware*
>>
>> Given a DRC Index return the info on predictive failure and overall health of
>> -the NVDIMM. The asserted bits in the health-bitmap indicate a single predictive
>> -failure and health-bit-valid-bitmap indicate which bits in health-bitmap are
>> -valid.
>> +the NVDIMM. The asserted bits in the health-bitmap indicate one or more states
>> +(described in table below) of the NVDIMM and health-bit-valid-bitmap indicate
>> +which bits in health-bitmap are valid.
>> +
>> +Health Bitmap Flags:
>> +
>> ++------+-----------------------------------------------------------------------+
>> +| Bit | Definition |
>> ++======+=======================================================================+
>> +| 00 | SCM device is unable to persist memory contents. |
>> +| | If the system is powered down, nothing will be saved. |
>> ++------+-----------------------------------------------------------------------+
>
> Are these correct bit numbers or backward IBM big endian bit numbers?
>
> ie. which bit is LSB?
These bit numbers index to a 64-bit dword laid in IBM big endian
format. So LSB would be at the located at a higher address. For example
0xC400000000000000 indicates bits 0, 1, and 5 are valid.
>
> cheers
> _______________________________________________
> Linux-nvdimm mailing list -- linux-nvdimm@lists.01.org
> To unsubscribe send an email to linux-nvdimm-leave@lists.01.org
^ permalink raw reply
* [PATCH v2 1/2] powerpc/64s/hash: add torture_slb kernel boot option to increase SLB faults
From: Nicholas Piggin @ 2020-05-04 9:06 UTC (permalink / raw)
To: linuxppc-dev; +Cc: Aneesh Kumar K . V, Nicholas Piggin
This option increases the number of SLB misses by limiting the number of
kernel SLB entries, and increased flushing of cached lookaside information.
This helps stress test difficult to hit paths in the kernel.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
v2:
- Address some comments from Aneesh about function names and types
.../admin-guide/kernel-parameters.txt | 5 +
arch/powerpc/include/asm/book3s/64/mmu-hash.h | 7 +
arch/powerpc/mm/book3s64/hash_utils.c | 13 ++
arch/powerpc/mm/book3s64/slb.c | 152 ++++++++++++------
4 files changed, 132 insertions(+), 45 deletions(-)
diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index f2a93c8679e8..5a34b7dd9ebe 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -871,6 +871,11 @@
can be useful when debugging issues that require an SLB
miss to occur.
+ torture_slb [PPC]
+ Limits the number of kernel SLB entries, and flushes
+ them frequently to increase the rate of SLB faults
+ on kernel addresses.
+
disable= [IPV6]
See Documentation/networking/ipv6.txt.
diff --git a/arch/powerpc/include/asm/book3s/64/mmu-hash.h b/arch/powerpc/include/asm/book3s/64/mmu-hash.h
index 3fa1b962dc27..758de1e0f676 100644
--- a/arch/powerpc/include/asm/book3s/64/mmu-hash.h
+++ b/arch/powerpc/include/asm/book3s/64/mmu-hash.h
@@ -317,6 +317,13 @@ extern unsigned long tce_alloc_start, tce_alloc_end;
*/
extern int mmu_ci_restrictions;
+extern bool torture_slb_enabled;
+DECLARE_STATIC_KEY_FALSE(torture_slb_key);
+static inline bool torture_slb(void)
+{
+ return static_branch_unlikely(&torture_slb_key);
+}
+
/*
* This computes the AVPN and B fields of the first dword of a HPTE,
* for use when we want to match an existing PTE. The bottom 7 bits
diff --git a/arch/powerpc/mm/book3s64/hash_utils.c b/arch/powerpc/mm/book3s64/hash_utils.c
index 8ed2411c3f39..9c487b5782ef 100644
--- a/arch/powerpc/mm/book3s64/hash_utils.c
+++ b/arch/powerpc/mm/book3s64/hash_utils.c
@@ -354,6 +354,7 @@ int htab_remove_mapping(unsigned long vstart, unsigned long vend,
}
static bool disable_1tb_segments = false;
+bool torture_slb_enabled __read_mostly = false;
static int __init parse_disable_1tb_segments(char *p)
{
@@ -362,6 +363,13 @@ static int __init parse_disable_1tb_segments(char *p)
}
early_param("disable_1tb_segments", parse_disable_1tb_segments);
+static int __init parse_torture_slb(char *p)
+{
+ torture_slb_enabled = true;
+ return 0;
+}
+early_param("torture_slb", parse_torture_slb);
+
static int __init htab_dt_scan_seg_sizes(unsigned long node,
const char *uname, int depth,
void *data)
@@ -854,6 +862,8 @@ static void __init hash_init_partition_table(phys_addr_t hash_table,
pr_info("Partition table %p\n", partition_tb);
}
+DEFINE_STATIC_KEY_FALSE(torture_slb_key);
+
static void __init htab_initialize(void)
{
unsigned long table;
@@ -870,6 +880,9 @@ static void __init htab_initialize(void)
printk(KERN_INFO "Using 1TB segments\n");
}
+ if (torture_slb_enabled)
+ static_branch_enable(&torture_slb_key);
+
/*
* Calculate the required size of the htab. We want the number of
* PTEGs to equal one half the number of real pages.
diff --git a/arch/powerpc/mm/book3s64/slb.c b/arch/powerpc/mm/book3s64/slb.c
index 716204aee3da..cf41c4b63d95 100644
--- a/arch/powerpc/mm/book3s64/slb.c
+++ b/arch/powerpc/mm/book3s64/slb.c
@@ -68,7 +68,7 @@ static void assert_slb_presence(bool present, unsigned long ea)
* slbfee. requires bit 24 (PPC bit 39) be clear in RB. Hardware
* ignores all other bits from 0-27, so just clear them all.
*/
- ea &= ~((1UL << 28) - 1);
+ ea &= ~((1UL << SID_SHIFT) - 1);
asm volatile(__PPC_SLBFEE_DOT(%0, %1) : "=r"(tmp) : "r"(ea) : "cr0");
WARN_ON(present == (tmp == 0));
@@ -153,14 +153,42 @@ void slb_flush_all_realmode(void)
asm volatile("slbmte %0,%0; slbia" : : "r" (0));
}
+static __always_inline void __slb_flush_and_restore_bolted(bool preserve_kernel_lookaside)
+{
+ struct slb_shadow *p = get_slb_shadow();
+ unsigned long ksp_esid_data, ksp_vsid_data;
+ u32 ih;
+
+ /*
+ * SLBIA IH=1 on ISA v2.05 and newer processors may preserve lookaside
+ * information created with Class=0 entries, which we use for kernel
+ * SLB entries (the SLB entries themselves are still invalidated).
+ *
+ * Older processors will ignore this optimisation. Over-invalidation
+ * is fine because we never rely on lookaside information existing.
+ */
+ if (preserve_kernel_lookaside)
+ ih = 1;
+ else
+ ih = 0;
+
+ ksp_esid_data = be64_to_cpu(p->save_area[KSTACK_INDEX].esid);
+ ksp_vsid_data = be64_to_cpu(p->save_area[KSTACK_INDEX].vsid);
+
+ asm volatile(PPC_SLBIA(%0)" \n"
+ "slbmte %1, %2 \n"
+ :: "i" (ih),
+ "r" (ksp_vsid_data),
+ "r" (ksp_esid_data)
+ : "memory");
+}
+
/*
* This flushes non-bolted entries, it can be run in virtual mode. Must
* be called with interrupts disabled.
*/
void slb_flush_and_restore_bolted(void)
{
- struct slb_shadow *p = get_slb_shadow();
-
BUILD_BUG_ON(SLB_NUM_BOLTED != 2);
WARN_ON(!irqs_disabled());
@@ -171,13 +199,10 @@ void slb_flush_and_restore_bolted(void)
*/
hard_irq_disable();
- asm volatile("isync\n"
- "slbia\n"
- "slbmte %0, %1\n"
- "isync\n"
- :: "r" (be64_to_cpu(p->save_area[KSTACK_INDEX].vsid)),
- "r" (be64_to_cpu(p->save_area[KSTACK_INDEX].esid))
- : "memory");
+ isync();
+ __slb_flush_and_restore_bolted(false);
+ isync();
+
assert_slb_presence(true, get_paca()->kstack);
get_paca()->slb_cache_ptr = 0;
@@ -400,6 +425,30 @@ void preload_new_slb_context(unsigned long start, unsigned long sp)
local_irq_enable();
}
+static void slb_cache_slbie_kernel(unsigned int index)
+{
+ unsigned long slbie_data = get_paca()->slb_cache[index];
+ unsigned long ksp = get_paca()->kstack;
+
+ slbie_data <<= SID_SHIFT;
+ slbie_data |= 0xc000000000000000ULL;
+ if ((ksp & slb_esid_mask(mmu_kernel_ssize)) == slbie_data)
+ return;
+ slbie_data |= mmu_kernel_ssize << SLBIE_SSIZE_SHIFT;
+
+ asm volatile("slbie %0" : : "r" (slbie_data));
+}
+
+static void slb_cache_slbie_user(unsigned int index)
+{
+ unsigned long slbie_data = get_paca()->slb_cache[index];
+
+ slbie_data <<= SID_SHIFT;
+ slbie_data |= user_segment_size(slbie_data) << SLBIE_SSIZE_SHIFT;
+ slbie_data |= SLBIE_C; /* user slbs have C=1 */
+
+ asm volatile("slbie %0" : : "r" (slbie_data));
+}
/* Flush all user entries from the segment table of the current processor. */
void switch_slb(struct task_struct *tsk, struct mm_struct *mm)
@@ -414,8 +463,14 @@ void switch_slb(struct task_struct *tsk, struct mm_struct *mm)
* which would update the slb_cache/slb_cache_ptr fields in the PACA.
*/
hard_irq_disable();
- asm volatile("isync" : : : "memory");
- if (cpu_has_feature(CPU_FTR_ARCH_300)) {
+ isync();
+ if (torture_slb()) {
+ __slb_flush_and_restore_bolted(false);
+ isync();
+ get_paca()->slb_cache_ptr = 0;
+ get_paca()->slb_kern_bitmap = (1U << SLB_NUM_BOLTED) - 1;
+
+ } else if (cpu_has_feature(CPU_FTR_ARCH_300)) {
/*
* SLBIA IH=3 invalidates all Class=1 SLBEs and their
* associated lookaside structures, which matches what
@@ -423,47 +478,29 @@ void switch_slb(struct task_struct *tsk, struct mm_struct *mm)
* cache.
*/
asm volatile(PPC_SLBIA(3));
+
} else {
unsigned long offset = get_paca()->slb_cache_ptr;
if (!mmu_has_feature(MMU_FTR_NO_SLBIE_B) &&
offset <= SLB_CACHE_ENTRIES) {
- unsigned long slbie_data = 0;
-
- for (i = 0; i < offset; i++) {
- unsigned long ea;
-
- ea = (unsigned long)
- get_paca()->slb_cache[i] << SID_SHIFT;
- /*
- * Could assert_slb_presence(true) here, but
- * hypervisor or machine check could have come
- * in and removed the entry at this point.
- */
-
- slbie_data = ea;
- slbie_data |= user_segment_size(slbie_data)
- << SLBIE_SSIZE_SHIFT;
- slbie_data |= SLBIE_C; /* user slbs have C=1 */
- asm volatile("slbie %0" : : "r" (slbie_data));
- }
+ /*
+ * Could assert_slb_presence(true) here, but
+ * hypervisor or machine check could have come
+ * in and removed the entry at this point.
+ */
+
+ for (i = 0; i < offset; i++)
+ slb_cache_slbie_user(i);
/* Workaround POWER5 < DD2.1 issue */
if (!cpu_has_feature(CPU_FTR_ARCH_207S) && offset == 1)
- asm volatile("slbie %0" : : "r" (slbie_data));
+ slb_cache_slbie_user(0);
} else {
- struct slb_shadow *p = get_slb_shadow();
- unsigned long ksp_esid_data =
- be64_to_cpu(p->save_area[KSTACK_INDEX].esid);
- unsigned long ksp_vsid_data =
- be64_to_cpu(p->save_area[KSTACK_INDEX].vsid);
-
- asm volatile(PPC_SLBIA(1) "\n"
- "slbmte %0,%1\n"
- "isync"
- :: "r"(ksp_vsid_data),
- "r"(ksp_esid_data));
+ /* Flush but retain kernel lookaside information */
+ __slb_flush_and_restore_bolted(true);
+ isync();
get_paca()->slb_kern_bitmap = (1U << SLB_NUM_BOLTED) - 1;
}
@@ -503,7 +540,7 @@ void switch_slb(struct task_struct *tsk, struct mm_struct *mm)
* address accesses by the kernel (user mode won't happen until
* rfid, which is safe).
*/
- asm volatile("isync" : : : "memory");
+ isync();
}
void slb_set_size(u16 size)
@@ -571,6 +608,9 @@ static void slb_cache_update(unsigned long esid_data)
if (cpu_has_feature(CPU_FTR_ARCH_300))
return; /* ISAv3.0B and later does not use slb_cache */
+ if (torture_slb())
+ return;
+
/*
* Now update slb cache entries
*/
@@ -580,7 +620,7 @@ static void slb_cache_update(unsigned long esid_data)
* We have space in slb cache for optimized switch_slb().
* Top 36 bits from esid_data as per ISA
*/
- local_paca->slb_cache[slb_cache_index++] = esid_data >> 28;
+ local_paca->slb_cache[slb_cache_index++] = esid_data >> SID_SHIFT;
local_paca->slb_cache_ptr++;
} else {
/*
@@ -671,6 +711,28 @@ static long slb_insert_entry(unsigned long ea, unsigned long context,
* accesses user memory before it returns to userspace with rfid.
*/
assert_slb_presence(false, ea);
+ if (torture_slb()) {
+ int slb_cache_index = local_paca->slb_cache_ptr;
+
+ /*
+ * torture_slb() does not use slb cache, repurpose as a
+ * cache of inserted (non-bolted) kernel SLB entries. All
+ * non-bolted kernel entries are flushed on any user fault,
+ * or if there are already 3 non-boled kernel entries.
+ */
+ BUILD_BUG_ON(SLB_CACHE_ENTRIES < 3);
+ if (!kernel || slb_cache_index == 3) {
+ int i;
+
+ for (i = 0; i < slb_cache_index; i++)
+ slb_cache_slbie_kernel(i);
+ slb_cache_index = 0;
+ }
+
+ if (kernel)
+ local_paca->slb_cache[slb_cache_index++] = esid_data >> SID_SHIFT;
+ local_paca->slb_cache_ptr = slb_cache_index;
+ }
asm volatile("slbmte %0, %1" : : "r" (vsid_data), "r" (esid_data));
barrier();
--
2.23.0
^ permalink raw reply related
* [PATCH v2 2/2] powerpc/64s/hash: add torture_hpt kernel boot option to increase hash faults
From: Nicholas Piggin @ 2020-05-04 9:06 UTC (permalink / raw)
To: linuxppc-dev; +Cc: Aneesh Kumar K . V, Nicholas Piggin
In-Reply-To: <20200504090658.44996-1-npiggin@gmail.com>
This option increases the number of hash misses by limiting the number of
kernel HPT entries, by accessing the address immediately after installing
the PTE, then removing it again (except in the case of CI entries that
must not be accessed, these are removed on the next hash fault).
This helps stress test difficult to hit paths in the kernel.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
v2:
- Add a bit more to the changelog.
- Add some kuap stuff for when Aneesh's kaup for hash patches are
merged.
.../admin-guide/kernel-parameters.txt | 9 +++
arch/powerpc/include/asm/book3s/64/mmu-hash.h | 10 +++
arch/powerpc/mm/book3s64/hash_4k.c | 3 +
arch/powerpc/mm/book3s64/hash_64k.c | 8 ++
arch/powerpc/mm/book3s64/hash_utils.c | 76 ++++++++++++++++++-
5 files changed, 105 insertions(+), 1 deletion(-)
diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index 5a34b7dd9ebe..1ec6a32a717a 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -876,6 +876,15 @@
them frequently to increase the rate of SLB faults
on kernel addresses.
+ torture_hpt [PPC]
+ Limits the number of kernel HPT entries in the hash
+ page table to increase the rate of hash page table
+ faults on kernel addresses.
+
+ This may hang when run on processors / emulators which
+ do not have a TLB, or flush it more often than
+ required, QEMU seems to have problems.
+
disable= [IPV6]
See Documentation/networking/ipv6.txt.
diff --git a/arch/powerpc/include/asm/book3s/64/mmu-hash.h b/arch/powerpc/include/asm/book3s/64/mmu-hash.h
index 758de1e0f676..539e3d91eac4 100644
--- a/arch/powerpc/include/asm/book3s/64/mmu-hash.h
+++ b/arch/powerpc/include/asm/book3s/64/mmu-hash.h
@@ -324,6 +324,16 @@ static inline bool torture_slb(void)
return static_branch_unlikely(&torture_slb_key);
}
+extern bool torture_hpt_enabled;
+DECLARE_STATIC_KEY_FALSE(torture_hpt_key);
+static inline bool torture_hpt(void)
+{
+ return static_branch_unlikely(&torture_hpt_key);
+}
+
+void hpt_do_torture(unsigned long ea, unsigned long access,
+ unsigned long rflags, unsigned long hpte_group);
+
/*
* This computes the AVPN and B fields of the first dword of a HPTE,
* for use when we want to match an existing PTE. The bottom 7 bits
diff --git a/arch/powerpc/mm/book3s64/hash_4k.c b/arch/powerpc/mm/book3s64/hash_4k.c
index 22e787123cdf..54e4ff8c558d 100644
--- a/arch/powerpc/mm/book3s64/hash_4k.c
+++ b/arch/powerpc/mm/book3s64/hash_4k.c
@@ -118,6 +118,9 @@ int __hash_page_4K(unsigned long ea, unsigned long access, unsigned long vsid,
}
new_pte = (new_pte & ~_PAGE_HPTEFLAGS) | H_PAGE_HASHPTE;
new_pte |= pte_set_hidx(ptep, rpte, 0, slot, PTRS_PER_PTE);
+
+ if (torture_hpt())
+ hpt_do_torture(ea, access, rflags, hpte_group);
}
*ptep = __pte(new_pte & ~H_PAGE_BUSY);
return 0;
diff --git a/arch/powerpc/mm/book3s64/hash_64k.c b/arch/powerpc/mm/book3s64/hash_64k.c
index 7084ce2951e6..19ea0fc145a9 100644
--- a/arch/powerpc/mm/book3s64/hash_64k.c
+++ b/arch/powerpc/mm/book3s64/hash_64k.c
@@ -216,6 +216,9 @@ int __hash_page_4K(unsigned long ea, unsigned long access, unsigned long vsid,
new_pte |= pte_set_hidx(ptep, rpte, subpg_index, slot, PTRS_PER_PTE);
new_pte |= H_PAGE_HASHPTE;
+ if (torture_hpt())
+ hpt_do_torture(ea, access, rflags, hpte_group);
+
*ptep = __pte(new_pte & ~H_PAGE_BUSY);
return 0;
}
@@ -327,7 +330,12 @@ int __hash_page_64K(unsigned long ea, unsigned long access,
new_pte = (new_pte & ~_PAGE_HPTEFLAGS) | H_PAGE_HASHPTE;
new_pte |= pte_set_hidx(ptep, rpte, 0, slot, PTRS_PER_PTE);
+
+ if (torture_hpt())
+ hpt_do_torture(ea, access, rflags, hpte_group);
}
+
*ptep = __pte(new_pte & ~H_PAGE_BUSY);
+
return 0;
}
diff --git a/arch/powerpc/mm/book3s64/hash_utils.c b/arch/powerpc/mm/book3s64/hash_utils.c
index 9c487b5782ef..845da1e8ca4f 100644
--- a/arch/powerpc/mm/book3s64/hash_utils.c
+++ b/arch/powerpc/mm/book3s64/hash_utils.c
@@ -353,8 +353,12 @@ int htab_remove_mapping(unsigned long vstart, unsigned long vend,
return ret;
}
-static bool disable_1tb_segments = false;
+static bool disable_1tb_segments __read_mostly = false;
bool torture_slb_enabled __read_mostly = false;
+bool torture_hpt_enabled __read_mostly = false;
+
+/* per-CPU array allocated if we enable torture_hpt. */
+static unsigned long *torture_hpt_last_group;
static int __init parse_disable_1tb_segments(char *p)
{
@@ -370,6 +374,13 @@ static int __init parse_torture_slb(char *p)
}
early_param("torture_slb", parse_torture_slb);
+static int __init parse_torture_hpt(char *p)
+{
+ torture_hpt_enabled = true;
+ return 0;
+}
+early_param("torture_hpt", parse_torture_hpt);
+
static int __init htab_dt_scan_seg_sizes(unsigned long node,
const char *uname, int depth,
void *data)
@@ -863,6 +874,7 @@ static void __init hash_init_partition_table(phys_addr_t hash_table,
}
DEFINE_STATIC_KEY_FALSE(torture_slb_key);
+DEFINE_STATIC_KEY_FALSE(torture_hpt_key);
static void __init htab_initialize(void)
{
@@ -882,6 +894,15 @@ static void __init htab_initialize(void)
if (torture_slb_enabled)
static_branch_enable(&torture_slb_key);
+ if (torture_hpt_enabled) {
+ unsigned long tmp;
+ static_branch_enable(&torture_hpt_key);
+ tmp = memblock_phys_alloc_range(sizeof(unsigned long) * NR_CPUS,
+ 0,
+ 0, MEMBLOCK_ALLOC_ANYWHERE);
+ memset((void *)tmp, 0xff, sizeof(unsigned long) * NR_CPUS);
+ torture_hpt_last_group = __va(tmp);
+ }
/*
* Calculate the required size of the htab. We want the number of
@@ -1901,6 +1922,59 @@ long hpte_insert_repeating(unsigned long hash, unsigned long vpn,
return slot;
}
+void hpt_do_torture(unsigned long ea, unsigned long access,
+ unsigned long rflags, unsigned long hpte_group)
+{
+ unsigned long last_group;
+ int cpu = raw_smp_processor_id();
+
+ last_group = torture_hpt_last_group[cpu];
+ if (last_group != -1UL) {
+ while (mmu_hash_ops.hpte_remove(last_group) != -1)
+ ;
+ torture_hpt_last_group[cpu] = -1UL;
+ }
+
+#define QEMU_WORKAROUND 0
+
+ if (ea >= PAGE_OFFSET) {
+ if (!QEMU_WORKAROUND && (access & (_PAGE_READ|_PAGE_WRITE)) &&
+ !(rflags & (HPTE_R_I|HPTE_R_G))) {
+ /* prefetch / prefetchw does not seem to set up a TLB
+ * entry with the powerpc systemsim (mambo) emulator,
+ * though it works with real hardware. An alternative
+ * approach that would work more reliably on quirky
+ * emulators like QEMU may be to remember the last
+ * insertion and remove that, rather than removing the
+ * current insertion. Then no prefetch is required.
+ */
+ if ((access & _PAGE_WRITE) && (access & _PAGE_READ)) {
+ void __user *mem = (void *)(ea & ~0x3);
+
+ allow_read_write_user(mem, mem, sizeof(atomic_t));
+ atomic_add(0, (atomic_t *)mem);
+ prevent_read_write_user(mem, mem, sizeof(atomic_t));
+
+ } else if (access & _PAGE_READ) {
+ void __user *mem = (void *)ea;
+
+ allow_read_from_user(mem, 1);
+ (void)*(volatile char *)mem;
+ prevent_read_from_user(mem, 1);
+ }
+
+ mb();
+
+ while (mmu_hash_ops.hpte_remove(hpte_group) != -1)
+ ;
+ } else {
+ /* Can't prefetch cache-inhibited so clear next time. */
+ torture_hpt_last_group[cpu] = hpte_group;
+ }
+ }
+}
+
+
#ifdef CONFIG_DEBUG_PAGEALLOC
static void kernel_map_linear_page(unsigned long vaddr, unsigned long lmi)
{
--
2.23.0
^ permalink raw reply related
* Re: [PATCH v7 14/28] powerpc: Add a probe_kernel_read_inst() function
From: Alistair Popple @ 2020-05-04 9:24 UTC (permalink / raw)
To: Jordan Niethe; +Cc: npiggin, bala24, naveen.n.rao, linuxppc-dev, dja
In-Reply-To: <20200501034220.8982-15-jniethe5@gmail.com>
> @@ -524,7 +524,10 @@ __ftrace_make_call(struct dyn_ftrace *rec, unsigned
> long addr) struct module *mod = rec->arch.mod;
>
> /* read where this goes */
> - if (probe_kernel_read(op, ip, sizeof(op)))
> + if (probe_kernel_read_inst(op, ip))
> + return -EFAULT;
> +
> + if (probe_kernel_read_inst(op + 1, ip + 4))
> return -EFAULT;
I had to double check the above for what happens when we introduce prefix
instructions but it looks mostly correct. There does however look to be a
corner case that could alter the behaviour once prefix instructions are
introduced.
With prefix instructions probe_kernel_read_inst() will read 8 bytes if the first
4 bytes are a valid prefix. Therefore the above could end up trying to read 12
bytes in total if the ip is a normal instruction and ip+4 is prefixed.
Obviously this is never going to match the expected nop sequence, and prefixed
instructions shouldn't cross page boundaries so the extra 4 bytes should never
be the cause of a fault either. The only difference we might see is
ftrace_make_call() incorrectly returning -EFAULT instead of -EINVAL for an
invalid (ie. crossing a 64 byte boundary) prefix instruction sequence.
In practice this doesn't seem like it would cause any real issues and the rest
of the patch does not appear to change any existing behaviour.
Reviewed-by: Alistair Popple <alistair@popple.id.au>
>
> if (!expected_nop_sequence(ip, op[0], op[1])) {
> @@ -587,7 +590,7 @@ __ftrace_make_call(struct dyn_ftrace *rec, unsigned long
> addr) unsigned long ip = rec->ip;
>
> /* read where this goes */
> - if (probe_kernel_read(&op, (void *)ip, MCOUNT_INSN_SIZE))
> + if (probe_kernel_read_inst(&op, (void *)ip))
> return -EFAULT;
>
> /* It should be pointing to a nop */
> @@ -643,7 +646,7 @@ static int __ftrace_make_call_kernel(struct dyn_ftrace
> *rec, unsigned long addr) }
>
> /* Make sure we have a nop */
> - if (probe_kernel_read(&op, ip, sizeof(op))) {
> + if (probe_kernel_read_inst(&op, ip)) {
> pr_err("Unable to read ftrace location %p\n", ip);
> return -EFAULT;
> }
> @@ -721,7 +724,7 @@ __ftrace_modify_call(struct dyn_ftrace *rec, unsigned
> long old_addr, }
>
> /* read where this goes */
> - if (probe_kernel_read(&op, (void *)ip, sizeof(int))) {
> + if (probe_kernel_read_inst(&op, (void *)ip)) {
> pr_err("Fetching opcode failed.\n");
> return -EFAULT;
> }
> diff --git a/arch/powerpc/lib/inst.c b/arch/powerpc/lib/inst.c
> index eaf786afad2b..08dedd927268 100644
> --- a/arch/powerpc/lib/inst.c
> +++ b/arch/powerpc/lib/inst.c
> @@ -16,3 +16,14 @@ int probe_user_read_inst(struct ppc_inst *inst,
> *inst = ppc_inst(val);
> return err;
> }
> +
> +int probe_kernel_read_inst(struct ppc_inst *inst,
> + struct ppc_inst *src)
> +{
> + unsigned int val;
> + int err;
> +
> + err = probe_kernel_read(&val, src, sizeof(val));
> + *inst = ppc_inst(val);
> + return err;
> +}
^ permalink raw reply
* Re: [PATCH v2 3/3] mm/page_alloc: Keep memoryless cpuless node 0 offline
From: Michal Hocko @ 2020-05-04 9:37 UTC (permalink / raw)
To: Srikar Dronamraju
Cc: Linus Torvalds, linux-kernel, linux-mm, Mel Gorman,
Kirill A. Shutemov, Andrew Morton, linuxppc-dev,
Christopher Lameter, Vlastimil Babka
In-Reply-To: <20200430071820.GF19958@linux.vnet.ibm.com>
On Thu 30-04-20 12:48:20, Srikar Dronamraju wrote:
> * Michal Hocko <mhocko@kernel.org> [2020-04-29 14:22:11]:
>
> > On Wed 29-04-20 07:11:45, Srikar Dronamraju wrote:
> > > > >
> > > > > By marking, N_ONLINE as NODE_MASK_NONE, lets stop assuming that Node 0 is
> > > > > always online.
> > > > >
> > > > > ...
> > > > >
> > > > > --- a/mm/page_alloc.c
> > > > > +++ b/mm/page_alloc.c
> > > > > @@ -116,8 +116,10 @@ EXPORT_SYMBOL(latent_entropy);
> > > > > */
> > > > > nodemask_t node_states[NR_NODE_STATES] __read_mostly = {
> > > > > [N_POSSIBLE] = NODE_MASK_ALL,
> > > > > +#ifdef CONFIG_NUMA
> > > > > + [N_ONLINE] = NODE_MASK_NONE,
> > > > > +#else
> > > > > [N_ONLINE] = { { [0] = 1UL } },
> > > > > -#ifndef CONFIG_NUMA
> > > > > [N_NORMAL_MEMORY] = { { [0] = 1UL } },
> > > > > #ifdef CONFIG_HIGHMEM
> > > > > [N_HIGH_MEMORY] = { { [0] = 1UL } },
> > > >
> > > > So on all other NUMA machines, when does node 0 get marked online?
> > > >
> > > > This change means that for some time during boot, such machines will
> > > > now be running with node 0 marked as offline. What are the
> > > > implications of this? Will something break?
> > >
> > > Till the nodes are detected, marking Node 0 as online tends to be redundant.
> > > Because the system doesn't know if its a NUMA or a non-NUMA system.
> > > Once we detect the nodes, we online them immediately. Hence I don't see any
> > > side-effects or negative implications of this change.
> > >
> > > However if I am missing anything, please do let me know.
> > >
> > > >From my part, I have tested this on
> > > 1. Non-NUMA Single node but CPUs and memory coming from zero node.
> > > 2. Non-NUMA Single node but CPUs and memory coming from non-zero node.
> > > 3. NUMA Multi node but with CPUs and memory from node 0.
> > > 4. NUMA Multi node but with no CPUs and memory from node 0.
> >
> > Have you tested on something else than ppc? Each arch does the NUMA
> > setup separately and this is a big mess. E.g. x86 marks even memory less
> > nodes (see init_memory_less_node) as online.
> >
>
> while I have predominantly tested on ppc, I did test on X86 with CONFIG_NUMA
> enabled/disabled on both single node and multi node machines.
> However, I dont have a cpuless/memoryless x86 system.
This should be able to emulate inside kvm, I believe.
> > Honestly I have hard time to evaluate the effect of this patch. It makes
> > some sense to assume all nodes offline before they get online but this
> > is a land mine territory.
> >
> > I am also not sure what kind of problem this is going to address. You
> > have mentioned numa balancing without many details.
>
> 1. On a machine with just one node with node number not being 0,
> the current setup will end up showing 2 online nodes. And when there are
> more than one online nodes, numa_balancing gets enabled.
>
> Without patch
> $ grep numa /proc/vmstat
> numa_hit 95179
> numa_miss 0
> numa_foreign 0
> numa_interleave 3764
> numa_local 95179
> numa_other 0
> numa_pte_updates 1206973 <----------
> numa_huge_pte_updates 4654 <----------
> numa_hint_faults 19560 <----------
> numa_hint_faults_local 19560 <----------
> numa_pages_migrated 0
>
>
> With patch
> $ grep numa /proc/vmstat
> numa_hit 322338756
> numa_miss 0
> numa_foreign 0
> numa_interleave 3790
> numa_local 322338756
> numa_other 0
> numa_pte_updates 0 <----------
> numa_huge_pte_updates 0 <----------
> numa_hint_faults 0 <----------
> numa_hint_faults_local 0 <----------
> numa_pages_migrated 0
>
> So we have a redundant page hinting numa faults which we can avoid.
interesting. Does this lead to any observable differences? Btw. it would
be really great to describe how the online state influences the numa
balancing.
> 2. Few people have complained about existence of this dummy node when
> parsing lscpu and numactl o/p. They somehow start to think that the tools
> are reporting incorrectly or the kernel is not able to recognize resources
> connected to the node.
Please be more specific.
--
Michal Hocko
SUSE Labs
^ permalink raw reply
* Re: [PATCH 2/7] crypto: powerpc/sha1 - remove unused temporary workspace
From: Michael Ellerman @ 2020-05-04 10:27 UTC (permalink / raw)
To: Eric Biggers, linux-crypto
Cc: Jason A . Donenfeld, Theodore Ts'o, linux-kernel,
Paul Mackerras, linuxppc-dev
In-Reply-To: <20200502182427.104383-3-ebiggers@kernel.org>
Eric Biggers <ebiggers@kernel.org> writes:
> From: Eric Biggers <ebiggers@google.com>
>
> The PowerPC implementation of SHA-1 doesn't actually use the 16-word
> temporary array that's passed to the assembly code. This was probably
> meant to correspond to the 'W' array that lib/sha1.c uses. However, in
> sha1-powerpc-asm.S these values are actually stored in GPRs 16-31.
>
> Referencing SHA_WORKSPACE_WORDS from this code also isn't appropriate,
> since it's an implementation detail of lib/sha1.c.
>
> Therefore, just remove this unneeded array.
>
> Tested with:
>
> export ARCH=powerpc CROSS_COMPILE=powerpc-linux-gnu-
> make mpc85xx_defconfig
> cat >> .config << EOF
> # CONFIG_MODULES is not set
> # CONFIG_CRYPTO_MANAGER_DISABLE_TESTS is not set
> CONFIG_DEBUG_KERNEL=y
> CONFIG_CRYPTO_MANAGER_EXTRA_TESTS=y
> CONFIG_CRYPTO_SHA1_PPC=y
> EOF
> make olddefconfig
> make -j32
> qemu-system-ppc -M mpc8544ds -cpu e500 -nographic \
> -kernel arch/powerpc/boot/zImage \
> -append "cryptomgr.fuzz_iterations=1000 cryptomgr.panic_on_fail=1"
Thanks for testing.
I gave it a quick spin on a Power9 and it showed no issues.
Acked-by: Michael Ellerman <mpe@ellerman.id.au> (powerpc)
cheers
^ permalink raw reply
* Re: [PATCH] powerpc/64s: Fix unrecoverable SLB crashes due to preemption check
From: Michael Ellerman @ 2020-05-04 10:53 UTC (permalink / raw)
To: Hugh Dickins; +Cc: linuxppc-dev, hughd, npiggin
In-Reply-To: <alpine.LSU.2.11.2005030008400.1557@eggly.anvils>
Hugh Dickins <hughd@google.com> writes:
> On Sun, 3 May 2020, Michael Ellerman wrote:
>
>> Hugh reported that his trusty G5 crashed after a few hours under load
>> with an "Unrecoverable exception 380".
>>
>> The crash is in interrupt_return() where we check lazy_irq_pending(),
>> which calls get_paca() and with CONFIG_DEBUG_PREEMPT=y that goes to
>> check_preemption_disabled() via debug_smp_processor_id().
>>
>> As Nick explained on the list:
>>
>> Problem is MSR[RI] is cleared here, ready to do the last few things
>> for interrupt return where we're not allowed to take any other
>> interrupts.
>>
>> SLB interrupts can happen just about anywhere aside from kernel
>> text, global variables, and stack. When that hits, it appears to be
>> unrecoverable due to RI=0.
>>
>> The problematic access is in preempt_count() which is:
>>
>> return READ_ONCE(current_thread_info()->preempt_count);
>>
>> Because of THREAD_INFO_IN_TASK, current_thread_info() just points to
>> current, so the access is to somewhere in kernel memory, but not on
>> the stack or in .data, which means it can cause an SLB miss. If we
>> take an SLB miss with RI=0 it is fatal.
>>
>> The easiest solution is to add a version of lazy_irq_pending() that
>> doesn't do the preemption check and call it from the interrupt return
>> path.
>>
>> Fixes: 68b34588e202 ("powerpc/64/sycall: Implement syscall entry/exit logic in C")
>> Reported-by: Hugh Dickins <hughd@google.com>
>> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
>
> Thank you, Michael and Nick: this has been running fine all day for me.
Thanks Hugh.
cheers
^ permalink raw reply
* Re: [PATCH] powerpc/5200: update contact email
From: Michael Ellerman @ 2020-05-04 10:55 UTC (permalink / raw)
To: Wolfram Sang, linux-kernel
Cc: devicetree, Wolfram Sang, Rob Herring, Paul Mackerras, kernel,
linuxppc-dev
In-Reply-To: <20200502142642.18979-1-wsa@kernel.org>
Wolfram Sang <wsa@kernel.org> writes:
> My 'pengutronix' address is defunct for years. Merge the entries and use
> the proper contact address.
Is there any point adding the new address? It's just likely to bit-rot
one day too.
I figure the git history is a better source for more up-to-date emails.
cheers
> diff --git a/arch/powerpc/boot/dts/pcm032.dts b/arch/powerpc/boot/dts/pcm032.dts
> index c259c6b3ac5a..780e13d99e7b 100644
> --- a/arch/powerpc/boot/dts/pcm032.dts
> +++ b/arch/powerpc/boot/dts/pcm032.dts
> @@ -3,9 +3,7 @@
> * phyCORE-MPC5200B-IO (pcm032) board Device Tree Source
> *
> * Copyright (C) 2006-2009 Pengutronix
> - * Sascha Hauer <s.hauer@pengutronix.de>
> - * Juergen Beisert <j.beisert@pengutronix.de>
> - * Wolfram Sang <w.sang@pengutronix.de>
> + * Sascha Hauer, Juergen Beisert, Wolfram Sang <kernel@pengutronix.de>
> */
>
> /include/ "mpc5200b.dtsi"
> --
> 2.20.1
^ permalink raw reply
* [PATCH v2 1/5] refcount, kref: add dec-and-test wrappers for rw_semaphores
From: Emanuele Giuseppe Esposito @ 2020-05-04 11:03 UTC (permalink / raw)
To: kvm
Cc: Emanuele Giuseppe Esposito, linux-s390, David Hildenbrand,
Cornelia Huck, Emanuele Giuseppe Esposito, linux-kernel, kvm-ppc,
linux-mips, Christian Borntraeger, Alexander Viro, linux-fsdevel,
Paolo Bonzini, Vitaly Kuznetsov, linuxppc-dev, Jim Mattson
In-Reply-To: <20200504110344.17560-1-eesposit@redhat.com>
Similar to the existing functions that take a mutex or spinlock if and
only if a reference count is decremented to zero, these new function
take an rwsem for writing just before the refcount reaches 0 (and
call a user-provided function in the case of kref_put_rwsem).
These will be used for stats_fs_source data structures, which are
protected by an rw_semaphore to allow concurrent sysfs reads.
Signed-off-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
---
include/linux/kref.h | 11 +++++++++++
include/linux/refcount.h | 2 ++
lib/refcount.c | 32 ++++++++++++++++++++++++++++++++
3 files changed, 45 insertions(+)
diff --git a/include/linux/kref.h b/include/linux/kref.h
index d32e21a2538c..2dc935445f45 100644
--- a/include/linux/kref.h
+++ b/include/linux/kref.h
@@ -79,6 +79,17 @@ static inline int kref_put_mutex(struct kref *kref,
return 0;
}
+static inline int kref_put_rwsem(struct kref *kref,
+ void (*release)(struct kref *kref),
+ struct rw_semaphore *rwsem)
+{
+ if (refcount_dec_and_down_write(&kref->refcount, rwsem)) {
+ release(kref);
+ return 1;
+ }
+ return 0;
+}
+
static inline int kref_put_lock(struct kref *kref,
void (*release)(struct kref *kref),
spinlock_t *lock)
diff --git a/include/linux/refcount.h b/include/linux/refcount.h
index 0e3ee25eb156..a9d5038aec9a 100644
--- a/include/linux/refcount.h
+++ b/include/linux/refcount.h
@@ -99,6 +99,7 @@
#include <linux/spinlock_types.h>
struct mutex;
+struct rw_semaphore;
/**
* struct refcount_t - variant of atomic_t specialized for reference counts
@@ -313,6 +314,7 @@ static inline void refcount_dec(refcount_t *r)
extern __must_check bool refcount_dec_if_one(refcount_t *r);
extern __must_check bool refcount_dec_not_one(refcount_t *r);
extern __must_check bool refcount_dec_and_mutex_lock(refcount_t *r, struct mutex *lock);
+extern __must_check bool refcount_dec_and_down_write(refcount_t *r, struct rw_semaphore *rwsem);
extern __must_check bool refcount_dec_and_lock(refcount_t *r, spinlock_t *lock);
extern __must_check bool refcount_dec_and_lock_irqsave(refcount_t *r,
spinlock_t *lock,
diff --git a/lib/refcount.c b/lib/refcount.c
index ebac8b7d15a7..03e113e1b43a 100644
--- a/lib/refcount.c
+++ b/lib/refcount.c
@@ -4,6 +4,7 @@
*/
#include <linux/mutex.h>
+#include <linux/rwsem.h>
#include <linux/refcount.h>
#include <linux/spinlock.h>
#include <linux/bug.h>
@@ -94,6 +95,37 @@ bool refcount_dec_not_one(refcount_t *r)
}
EXPORT_SYMBOL(refcount_dec_not_one);
+/**
+ * refcount_dec_and_down_write - return holding rwsem for writing if able to decrement
+ * refcount to 0
+ * @r: the refcount
+ * @lock: the mutex to be locked
+ *
+ * Similar to atomic_dec_and_mutex_lock(), it will WARN on underflow and fail
+ * to decrement when saturated at REFCOUNT_SATURATED.
+ *
+ * Provides release memory ordering, such that prior loads and stores are done
+ * before, and provides a control dependency such that free() must come after.
+ * See the comment on top.
+ *
+ * Return: true and hold rwsem for writing if able to decrement refcount to 0, false
+ * otherwise
+ */
+bool refcount_dec_and_down_write(refcount_t *r, struct rw_semaphore *lock)
+{
+ if (refcount_dec_not_one(r))
+ return false;
+
+ down_write(lock);
+ if (!refcount_dec_and_test(r)) {
+ up_write(lock);
+ return false;
+ }
+
+ return true;
+}
+EXPORT_SYMBOL(refcount_dec_and_down_write);
+
/**
* refcount_dec_and_mutex_lock - return holding mutex if able to decrement
* refcount to 0
--
2.25.2
^ permalink raw reply related
* [PATCH v2 2/5] stats_fs API: create, add and remove stats_fs sources and values
From: Emanuele Giuseppe Esposito @ 2020-05-04 11:03 UTC (permalink / raw)
To: kvm
Cc: Emanuele Giuseppe Esposito, linux-s390, David Hildenbrand,
Cornelia Huck, Emanuele Giuseppe Esposito, linux-kernel, kvm-ppc,
linux-mips, Christian Borntraeger, Alexander Viro, linux-fsdevel,
Paolo Bonzini, Vitaly Kuznetsov, linuxppc-dev, Jim Mattson
In-Reply-To: <20200504110344.17560-1-eesposit@redhat.com>
Introduction to the stats_fs API, that allows to easily create, add
and remove stats_fs sources and values. The API allows to easily building
the statistics directory tree to automatically gather them for the linux
kernel. The main functionalities are: create a source, add child
sources/values/aggregates, register it to the root source (that on
the virtual fs would be /sys/kernel/statsfs), ad perform a search for
a value/aggregate.
This allows creating any kind of source tree, making it more flexible
also to future readjustments.
The API representation is only logical and will be backed up
by a virtual file system in patch 4.
Its usage will be shared between the stats_fs file system
and the end-users like kvm, the former calling it when it needs to
display and clear statistics, the latter to add values and sources.
Signed-off-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
---
MAINTAINERS | 7 +
fs/Kconfig | 6 +
fs/Makefile | 1 +
fs/stats_fs/Makefile | 4 +
fs/stats_fs/internal.h | 20 ++
fs/stats_fs/stats_fs.c | 610 +++++++++++++++++++++++++++++++++++++++
include/linux/stats_fs.h | 289 +++++++++++++++++++
7 files changed, 937 insertions(+)
create mode 100644 fs/stats_fs/Makefile
create mode 100644 fs/stats_fs/internal.h
create mode 100644 fs/stats_fs/stats_fs.c
create mode 100644 include/linux/stats_fs.h
diff --git a/MAINTAINERS b/MAINTAINERS
index b816a453b10e..a8403d07cee5 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -5229,6 +5229,13 @@ F: include/linux/debugfs.h
F: include/linux/kobj*
F: lib/kobj*
+STATS_FS
+M: Paolo Bonzini <pbonzini@redhat.com>
+R: Emanuele Giuseppe Esposito <e.emanuelegiuseppe@gmail.com>
+S: Supported
+F: include/linux/stats_fs.h
+F: fs/stats_fs
+
DRIVERS FOR ADAPTIVE VOLTAGE SCALING (AVS)
M: Kevin Hilman <khilman@kernel.org>
M: Nishanth Menon <nm@ti.com>
diff --git a/fs/Kconfig b/fs/Kconfig
index f08fbbfafd9a..1b0de0f19e96 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -328,4 +328,10 @@ source "fs/unicode/Kconfig"
config IO_WQ
bool
+config STATS_FS
+ bool "Statistics Filesystem"
+ help
+ stats_fs is a virtual file system that provides counters and
+ other statistics about the running kernel.
+
endmenu
diff --git a/fs/Makefile b/fs/Makefile
index 2ce5112b02c8..91558eca0cf7 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -125,6 +125,7 @@ obj-$(CONFIG_BEFS_FS) += befs/
obj-$(CONFIG_HOSTFS) += hostfs/
obj-$(CONFIG_CACHEFILES) += cachefiles/
obj-$(CONFIG_DEBUG_FS) += debugfs/
+obj-$(CONFIG_STATS_FS) += stats_fs/
obj-$(CONFIG_TRACING) += tracefs/
obj-$(CONFIG_OCFS2_FS) += ocfs2/
obj-$(CONFIG_BTRFS_FS) += btrfs/
diff --git a/fs/stats_fs/Makefile b/fs/stats_fs/Makefile
new file mode 100644
index 000000000000..94fe52d590d5
--- /dev/null
+++ b/fs/stats_fs/Makefile
@@ -0,0 +1,4 @@
+# SPDX-License-Identifier: GPL-2.0-only
+stats_fs-objs := stats_fs.o
+
+obj-$(CONFIG_STATS_FS) += stats_fs.o
diff --git a/fs/stats_fs/internal.h b/fs/stats_fs/internal.h
new file mode 100644
index 000000000000..ddf262a60736
--- /dev/null
+++ b/fs/stats_fs/internal.h
@@ -0,0 +1,20 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _STATS_FS_INTERNAL_H_
+#define _STATS_FS_INTERNAL_H_
+
+#include <linux/list.h>
+#include <linux/kref.h>
+#include <linux/rwsem.h>
+#include <linux/stats_fs.h>
+
+/* values, grouped by base */
+struct stats_fs_value_source {
+ void *base_addr;
+ bool files_created;
+ struct stats_fs_value *values;
+ struct list_head list_element;
+};
+
+int stats_fs_val_get_mode(struct stats_fs_value *val);
+
+#endif /* _STATS_FS_INTERNAL_H_ */
diff --git a/fs/stats_fs/stats_fs.c b/fs/stats_fs/stats_fs.c
new file mode 100644
index 000000000000..b63de12769e2
--- /dev/null
+++ b/fs/stats_fs/stats_fs.c
@@ -0,0 +1,610 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/module.h>
+#include <linux/errno.h>
+#include <linux/file.h>
+#include <linux/fs.h>
+#include <linux/slab.h>
+#include <linux/rwsem.h>
+#include <linux/list.h>
+#include <linux/kref.h>
+#include <linux/limits.h>
+#include <linux/stats_fs.h>
+
+#include "internal.h"
+
+struct stats_fs_aggregate_value {
+ uint64_t sum, min, max;
+ uint32_t count, count_zero;
+};
+
+static int is_val_signed(struct stats_fs_value *val)
+{
+ return val->type & STATS_FS_SIGN;
+}
+
+int stats_fs_val_get_mode(struct stats_fs_value *val)
+{
+ return val->mode ? val->mode : 0644;
+}
+
+static struct stats_fs_value *find_value(struct stats_fs_value_source *src,
+ struct stats_fs_value *val)
+{
+ struct stats_fs_value *entry;
+
+ for (entry = src->values; entry->name; entry++) {
+ if (entry == val)
+ return entry;
+ }
+ return NULL;
+}
+
+static struct stats_fs_value *
+search_value_in_source(struct stats_fs_source *src, struct stats_fs_value *arg,
+ struct stats_fs_value_source **val_src)
+{
+ struct stats_fs_value *entry;
+ struct stats_fs_value_source *src_entry;
+
+ list_for_each_entry (src_entry, &src->values_head, list_element) {
+ entry = find_value(src_entry, arg);
+ if (entry) {
+ *val_src = src_entry;
+ return entry;
+ }
+ }
+
+ return NULL;
+}
+
+/* Called with rwsem held for writing */
+static struct stats_fs_value_source *create_value_source(void *base)
+{
+ struct stats_fs_value_source *val_src;
+
+ val_src = kzalloc(sizeof(struct stats_fs_value_source), GFP_KERNEL);
+ if (!val_src)
+ return ERR_PTR(-ENOMEM);
+
+ val_src->base_addr = base;
+ INIT_LIST_HEAD(&val_src->list_element);
+
+ return val_src;
+}
+
+int stats_fs_source_add_values(struct stats_fs_source *source,
+ struct stats_fs_value *stat, void *ptr)
+{
+ struct stats_fs_value_source *val_src;
+ struct stats_fs_value_source *entry;
+
+ down_write(&source->rwsem);
+
+ list_for_each_entry (entry, &source->values_head, list_element) {
+ if (entry->base_addr == ptr && entry->values == stat) {
+ up_write(&source->rwsem);
+ return -EEXIST;
+ }
+ }
+
+ val_src = create_value_source(ptr);
+ val_src->values = (struct stats_fs_value *)stat;
+
+ /* add the val_src to the source list */
+ list_add(&val_src->list_element, &source->values_head);
+
+ up_write(&source->rwsem);
+
+ return 0;
+}
+EXPORT_SYMBOL_GPL(stats_fs_source_add_values);
+
+void stats_fs_source_add_subordinate(struct stats_fs_source *source,
+ struct stats_fs_source *sub)
+{
+ down_write(&source->rwsem);
+
+ stats_fs_source_get(sub);
+ list_add(&sub->list_element, &source->subordinates_head);
+
+ up_write(&source->rwsem);
+}
+EXPORT_SYMBOL_GPL(stats_fs_source_add_subordinate);
+
+/* Called with rwsem held for writing */
+static void
+stats_fs_source_remove_subordinate_locked(struct stats_fs_source *source,
+ struct stats_fs_source *sub)
+{
+ struct stats_fs_source *src_entry;
+
+ list_for_each_entry (src_entry, &source->subordinates_head,
+ list_element) {
+ if (src_entry == sub) {
+ list_del_init(&src_entry->list_element);
+ stats_fs_source_put(src_entry);
+ return;
+ }
+ }
+}
+
+void stats_fs_source_remove_subordinate(struct stats_fs_source *source,
+ struct stats_fs_source *sub)
+{
+ down_write(&source->rwsem);
+ stats_fs_source_remove_subordinate_locked(source, sub);
+ up_write(&source->rwsem);
+}
+EXPORT_SYMBOL_GPL(stats_fs_source_remove_subordinate);
+
+/* Called with rwsem held for reading */
+static uint64_t get_simple_value(struct stats_fs_value_source *src,
+ struct stats_fs_value *val)
+{
+ uint64_t value_found;
+ void *address;
+
+ address = src->base_addr + val->offset;
+
+ switch (val->type) {
+ case STATS_FS_U8:
+ value_found = *((uint8_t *)address);
+ break;
+ case STATS_FS_U8 | STATS_FS_SIGN:
+ value_found = *((int8_t *)address);
+ break;
+ case STATS_FS_U16:
+ value_found = *((uint16_t *)address);
+ break;
+ case STATS_FS_U16 | STATS_FS_SIGN:
+ value_found = *((int16_t *)address);
+ break;
+ case STATS_FS_U32:
+ value_found = *((uint32_t *)address);
+ break;
+ case STATS_FS_U32 | STATS_FS_SIGN:
+ value_found = *((int32_t *)address);
+ break;
+ case STATS_FS_U64:
+ value_found = *((uint64_t *)address);
+ break;
+ case STATS_FS_U64 | STATS_FS_SIGN:
+ value_found = *((int64_t *)address);
+ break;
+ case STATS_FS_BOOL:
+ value_found = *((uint8_t *)address);
+ break;
+ default:
+ value_found = 0;
+ break;
+ }
+
+ return value_found;
+}
+
+/* Called with rwsem held for reading */
+static void clear_simple_value(struct stats_fs_value_source *src,
+ struct stats_fs_value *val)
+{
+ void *address;
+
+ address = src->base_addr + val->offset;
+
+ switch (val->type) {
+ case STATS_FS_U8:
+ *((uint8_t *)address) = 0;
+ break;
+ case STATS_FS_U8 | STATS_FS_SIGN:
+ *((int8_t *)address) = 0;
+ break;
+ case STATS_FS_U16:
+ *((uint16_t *)address) = 0;
+ break;
+ case STATS_FS_U16 | STATS_FS_SIGN:
+ *((int16_t *)address) = 0;
+ break;
+ case STATS_FS_U32:
+ *((uint32_t *)address) = 0;
+ break;
+ case STATS_FS_U32 | STATS_FS_SIGN:
+ *((int32_t *)address) = 0;
+ break;
+ case STATS_FS_U64:
+ *((uint64_t *)address) = 0;
+ break;
+ case STATS_FS_U64 | STATS_FS_SIGN:
+ *((int64_t *)address) = 0;
+ break;
+ case STATS_FS_BOOL:
+ *((uint8_t *)address) = 0;
+ break;
+ default:
+ break;
+ }
+}
+
+/* Called with rwsem held for reading */
+static void
+search_all_simple_values(struct stats_fs_source *src,
+ struct stats_fs_value_source *ref_src_entry,
+ struct stats_fs_value *val,
+ struct stats_fs_aggregate_value *agg)
+{
+ struct stats_fs_value_source *src_entry;
+ uint64_t value_found;
+
+ list_for_each_entry (src_entry, &src->values_head, list_element) {
+ /* skip aggregates */
+ if (src_entry->base_addr == NULL)
+ continue;
+
+ /* useless to search here */
+ if (src_entry->values != ref_src_entry->values)
+ continue;
+
+ /* must be here */
+ value_found = get_simple_value(src_entry, val);
+ agg->sum += value_found;
+ agg->count++;
+ agg->count_zero += (value_found == 0);
+
+ if (is_val_signed(val)) {
+ agg->max = (((int64_t)value_found) >=
+ ((int64_t)agg->max)) ?
+ value_found :
+ agg->max;
+ agg->min = (((int64_t)value_found) <=
+ ((int64_t)agg->min)) ?
+ value_found :
+ agg->min;
+ } else {
+ agg->max = (value_found >= agg->max) ? value_found :
+ agg->max;
+ agg->min = (value_found <= agg->min) ? value_found :
+ agg->min;
+ }
+ }
+}
+
+/* Called with rwsem held for reading */
+static void
+do_recursive_aggregation(struct stats_fs_source *root,
+ struct stats_fs_value_source *ref_src_entry,
+ struct stats_fs_value *val,
+ struct stats_fs_aggregate_value *agg)
+{
+ struct stats_fs_source *subordinate;
+
+ /* search all simple values in this folder */
+ search_all_simple_values(root, ref_src_entry, val, agg);
+
+ /* recursively search in all subfolders */
+ list_for_each_entry (subordinate, &root->subordinates_head,
+ list_element) {
+ down_read(&subordinate->rwsem);
+ do_recursive_aggregation(subordinate, ref_src_entry, val, agg);
+ up_read(&subordinate->rwsem);
+ }
+}
+
+/* Called with rwsem held for reading */
+static void init_aggregate_value(struct stats_fs_aggregate_value *agg,
+ struct stats_fs_value *val)
+{
+ agg->count = agg->count_zero = agg->sum = 0;
+ if (is_val_signed(val)) {
+ agg->max = S64_MIN;
+ agg->min = S64_MAX;
+ } else {
+ agg->max = 0;
+ agg->min = U64_MAX;
+ }
+}
+
+/* Called with rwsem held for reading */
+static void store_final_value(struct stats_fs_aggregate_value *agg,
+ struct stats_fs_value *val, uint64_t *ret)
+{
+ int operation;
+
+ operation = val->aggr_kind | is_val_signed(val);
+
+ switch (operation) {
+ case STATS_FS_AVG:
+ *ret = agg->count ? agg->sum / agg->count : 0;
+ break;
+ case STATS_FS_AVG | STATS_FS_SIGN:
+ *ret = agg->count ? ((int64_t)agg->sum) / agg->count : 0;
+ break;
+ case STATS_FS_SUM:
+ case STATS_FS_SUM | STATS_FS_SIGN:
+ *ret = agg->sum;
+ break;
+ case STATS_FS_MIN:
+ case STATS_FS_MIN | STATS_FS_SIGN:
+ *ret = agg->min;
+ break;
+ case STATS_FS_MAX:
+ case STATS_FS_MAX | STATS_FS_SIGN:
+ *ret = agg->max;
+ break;
+ case STATS_FS_COUNT_ZERO:
+ case STATS_FS_COUNT_ZERO | STATS_FS_SIGN:
+ *ret = agg->count_zero;
+ break;
+ default:
+ break;
+ }
+}
+
+/* Called with rwsem held for reading */
+static int stats_fs_source_get_value_locked(struct stats_fs_source *source,
+ struct stats_fs_value *arg,
+ uint64_t *ret)
+{
+ struct stats_fs_value_source *src_entry;
+ struct stats_fs_value *found;
+ struct stats_fs_aggregate_value aggr;
+
+ *ret = 0;
+
+ if (!arg)
+ return -ENOENT;
+
+ /* look in simple values */
+ found = search_value_in_source(source, arg, &src_entry);
+
+ if (!found) {
+ printk(KERN_ERR "Stats_fs: Value in source \"%s\" not found!\n",
+ source->name);
+ return -ENOENT;
+ }
+
+ if (src_entry->base_addr != NULL) {
+ *ret = get_simple_value(src_entry, found);
+ return 0;
+ }
+
+ /* look in aggregates */
+ init_aggregate_value(&aggr, found);
+ do_recursive_aggregation(source, src_entry, found, &aggr);
+ store_final_value(&aggr, found, ret);
+
+ return 0;
+}
+
+int stats_fs_source_get_value(struct stats_fs_source *source,
+ struct stats_fs_value *arg, uint64_t *ret)
+{
+ int retval;
+
+ down_read(&source->rwsem);
+ retval = stats_fs_source_get_value_locked(source, arg, ret);
+ up_read(&source->rwsem);
+
+ return retval;
+}
+EXPORT_SYMBOL_GPL(stats_fs_source_get_value);
+
+/* Called with rwsem held for reading */
+static void set_all_simple_values(struct stats_fs_source *src,
+ struct stats_fs_value_source *ref_src_entry,
+ struct stats_fs_value *val)
+{
+ struct stats_fs_value_source *src_entry;
+
+ list_for_each_entry (src_entry, &src->values_head, list_element) {
+ /* skip aggregates */
+ if (src_entry->base_addr == NULL)
+ continue;
+
+ /* wrong to search here */
+ if (src_entry->values != ref_src_entry->values)
+ continue;
+
+ if (src_entry->base_addr &&
+ src_entry->values == ref_src_entry->values)
+ clear_simple_value(src_entry, val);
+ }
+}
+
+/* Called with rwsem held for reading */
+static void do_recursive_clean(struct stats_fs_source *root,
+ struct stats_fs_value_source *ref_src_entry,
+ struct stats_fs_value *val)
+{
+ struct stats_fs_source *subordinate;
+
+ /* search all simple values in this folder */
+ set_all_simple_values(root, ref_src_entry, val);
+
+ /* recursively search in all subfolders */
+ list_for_each_entry (subordinate, &root->subordinates_head,
+ list_element) {
+ down_read(&subordinate->rwsem);
+ do_recursive_clean(subordinate, ref_src_entry, val);
+ up_read(&subordinate->rwsem);
+ }
+}
+
+/* Called with rwsem held for reading */
+static int stats_fs_source_clear_locked(struct stats_fs_source *source,
+ struct stats_fs_value *val)
+{
+ struct stats_fs_value_source *src_entry;
+ struct stats_fs_value *found;
+
+ if (!val)
+ return -ENOENT;
+
+ /* look in simple values */
+ found = search_value_in_source(source, val, &src_entry);
+
+ if (!found) {
+ printk(KERN_ERR "Stats_fs: Value in source \"%s\" not found!\n",
+ source->name);
+ return -ENOENT;
+ }
+
+ if (src_entry->base_addr != NULL) {
+ clear_simple_value(src_entry, found);
+ return 0;
+ }
+
+ /* look in aggregates */
+ do_recursive_clean(source, src_entry, found);
+
+ return 0;
+}
+
+int stats_fs_source_clear(struct stats_fs_source *source,
+ struct stats_fs_value *val)
+{
+ int retval;
+
+ down_read(&source->rwsem);
+ retval = stats_fs_source_clear_locked(source, val);
+ up_read(&source->rwsem);
+
+ return retval;
+}
+
+/* Called with rwsem held for reading */
+static struct stats_fs_value *
+find_value_by_name(struct stats_fs_value_source *src, char *val)
+{
+ struct stats_fs_value *entry;
+
+ for (entry = src->values; entry->name; entry++)
+ if (!strcmp(entry->name, val))
+ return entry;
+
+ return NULL;
+}
+
+/* Called with rwsem held for reading */
+static struct stats_fs_value *
+search_in_source_by_name(struct stats_fs_source *src, char *name)
+{
+ struct stats_fs_value *entry;
+ struct stats_fs_value_source *src_entry;
+
+ list_for_each_entry (src_entry, &src->values_head, list_element) {
+ entry = find_value_by_name(src_entry, name);
+ if (entry)
+ return entry;
+ }
+
+ return NULL;
+}
+
+int stats_fs_source_get_value_by_name(struct stats_fs_source *source,
+ char *name, uint64_t *ret)
+{
+ struct stats_fs_value *val;
+ int retval;
+
+ down_read(&source->rwsem);
+ val = search_in_source_by_name(source, name);
+
+ if (!val) {
+ *ret = 0;
+ up_read(&source->rwsem);
+ return -ENOENT;
+ }
+
+ retval = stats_fs_source_get_value_locked(source, val, ret);
+ up_read(&source->rwsem);
+
+ return retval;
+}
+EXPORT_SYMBOL_GPL(stats_fs_source_get_value_by_name);
+
+void stats_fs_source_get(struct stats_fs_source *source)
+{
+ kref_get(&source->refcount);
+}
+EXPORT_SYMBOL_GPL(stats_fs_source_get);
+
+void stats_fs_source_revoke(struct stats_fs_source *source)
+{
+ struct stats_fs_value_source *val_src_entry;
+
+ down_write(&source->rwsem);
+
+ list_for_each_entry (val_src_entry, &source->values_head, list_element)
+ val_src_entry->base_addr = NULL;
+
+ up_write(&source->rwsem);
+}
+EXPORT_SYMBOL_GPL(stats_fs_source_revoke);
+
+/* Called with rwsem held for writing
+ *
+ * The refcount is 0 and the lock was taken before refcount
+ * went from 1 to 0
+ */
+static void stats_fs_source_destroy(struct kref *kref_source)
+{
+ struct stats_fs_value_source *val_src_entry;
+ struct list_head *it, *safe;
+ struct stats_fs_source *child, *source;
+
+ source = container_of(kref_source, struct stats_fs_source, refcount);
+
+ /* iterate through the values and delete them */
+ list_for_each_safe (it, safe, &source->values_head) {
+ val_src_entry = list_entry(it, struct stats_fs_value_source,
+ list_element);
+ kfree(val_src_entry);
+ }
+
+ /* iterate through the subordinates and delete them */
+ list_for_each_safe (it, safe, &source->subordinates_head) {
+ child = list_entry(it, struct stats_fs_source, list_element);
+ stats_fs_source_remove_subordinate_locked(source, child);
+ }
+
+ up_write(&source->rwsem);
+ kfree(source->name);
+ kfree(source);
+}
+
+void stats_fs_source_put(struct stats_fs_source *source)
+{
+ kref_put_rwsem(&source->refcount, stats_fs_source_destroy,
+ &source->rwsem);
+}
+EXPORT_SYMBOL_GPL(stats_fs_source_put);
+
+struct stats_fs_source *stats_fs_source_create(const char *fmt, ...)
+{
+ va_list ap;
+ char buf[100];
+ struct stats_fs_source *ret;
+ int char_needed;
+
+ va_start(ap, fmt);
+ char_needed = vsnprintf(buf, 100, fmt, ap);
+ va_end(ap);
+
+ ret = kzalloc(sizeof(struct stats_fs_source), GFP_KERNEL);
+ if (!ret)
+ return ERR_PTR(-ENOMEM);
+
+ ret->name = kstrdup(buf, GFP_KERNEL);
+ if (!ret->name) {
+ kfree(ret);
+ return ERR_PTR(-ENOMEM);
+ }
+
+ kref_init(&ret->refcount);
+ init_rwsem(&ret->rwsem);
+
+ INIT_LIST_HEAD(&ret->values_head);
+ INIT_LIST_HEAD(&ret->subordinates_head);
+ INIT_LIST_HEAD(&ret->list_element);
+
+ return ret;
+}
+EXPORT_SYMBOL_GPL(stats_fs_source_create);
diff --git a/include/linux/stats_fs.h b/include/linux/stats_fs.h
new file mode 100644
index 000000000000..dc2d2e11f5ea
--- /dev/null
+++ b/include/linux/stats_fs.h
@@ -0,0 +1,289 @@
+/* SPDX-License-Identifier: GPL-2.0
+ *
+ * stats_fs.h - a tiny little statistics file system
+ *
+ * Copyright (C) 2020 Emanuele Giuseppe Esposito
+ * Copyright (C) 2020 Redhat.
+ *
+ */
+
+#ifndef _STATS_FS_H_
+#define _STATS_FS_H_
+
+#include <linux/list.h>
+
+/* Used to distinguish signed types */
+#define STATS_FS_SIGN 0x8000
+
+enum stat_type {
+ STATS_FS_U8 = 0,
+ STATS_FS_U16 = 1,
+ STATS_FS_U32 = 2,
+ STATS_FS_U64 = 3,
+ STATS_FS_BOOL = 4,
+ STATS_FS_S8 = STATS_FS_U8 | STATS_FS_SIGN,
+ STATS_FS_S16 = STATS_FS_U16 | STATS_FS_SIGN,
+ STATS_FS_S32 = STATS_FS_U32 | STATS_FS_SIGN,
+ STATS_FS_S64 = STATS_FS_U64 | STATS_FS_SIGN,
+};
+
+enum stat_aggr {
+ STATS_FS_NONE = 0,
+ STATS_FS_SUM,
+ STATS_FS_MIN,
+ STATS_FS_MAX,
+ STATS_FS_COUNT_ZERO,
+ STATS_FS_AVG,
+};
+
+struct stats_fs_value {
+ /* Name of the stat */
+ char *name;
+
+ /* Offset from base address to field containing the value */
+ int offset;
+
+ /* Type of the stat BOOL,U64,... */
+ enum stat_type type;
+
+ /* Aggregate type: MIN, MAX, SUM,... */
+ enum stat_aggr aggr_kind;
+
+ /* File mode */
+ uint16_t mode;
+};
+
+struct stats_fs_source {
+ struct kref refcount;
+
+ char *name;
+
+ /* list of source stats_fs_value_source*/
+ struct list_head values_head;
+
+ /* list of struct stats_fs_source for subordinate sources */
+ struct list_head subordinates_head;
+
+ struct list_head list_element;
+
+ struct rw_semaphore rwsem;
+
+ struct dentry *source_dentry;
+};
+
+#if defined(CONFIG_STATS_FS)
+
+/**
+ * stats_fs_source_create - create a stats_fs_source
+ * Creates a stats_fs_source with the given name. This
+ * does not mean it will be backed by the filesystem yet, it will only
+ * be visible to the user once one of its parents (or itself) are
+ * registered in stats_fs.
+ *
+ * Returns a pointer to a stats_fs_source if it succeeds.
+ * This or one of the parents' pointer must be passed to the stats_fs_put()
+ * function when the file is to be removed. If an error occurs,
+ * ERR_PTR(-ERROR) will be returned.
+ */
+struct stats_fs_source *stats_fs_source_create(const char *fmt, ...);
+
+/**
+ * stats_fs_source_add_values - adds values to the given source
+ * @source: a pointer to the source that will receive the values
+ * @val: a pointer to the NULL terminated stats_fs_value array to add
+ * @base_ptr: a pointer to the base pointer used by these values
+ *
+ * In addition to adding values to the source, also create the
+ * files in the filesystem if the source already is backed up by a directory.
+ *
+ * Returns 0 it succeeds. If the value are already in the
+ * source and have the same base_ptr, -EEXIST is returned.
+ */
+int stats_fs_source_add_values(struct stats_fs_source *source,
+ struct stats_fs_value *val, void *base_ptr);
+
+/**
+ * stats_fs_source_add_subordinate - adds a child to the given source
+ * @parent: a pointer to the parent source
+ * @child: a pointer to child source to add
+ *
+ * Recursively create all files in the stats_fs filesystem
+ * only if the parent has already a dentry (created with
+ * stats_fs_source_register).
+ * This avoids the case where this function is called before register.
+ */
+void stats_fs_source_add_subordinate(struct stats_fs_source *parent,
+ struct stats_fs_source *child);
+
+/**
+ * stats_fs_source_remove_subordinate - removes a child from the given source
+ * @parent: a pointer to the parent source
+ * @child: a pointer to child source to remove
+ *
+ * Look if there is such child in the parent. If so,
+ * it will remove all its files and call stats_fs_put on the child.
+ */
+void stats_fs_source_remove_subordinate(struct stats_fs_source *parent,
+ struct stats_fs_source *child);
+
+/**
+ * stats_fs_source_get_value - search a value in the source (and
+ * subordinates)
+ * @source: a pointer to the source that will be searched
+ * @val: a pointer to the stats_fs_value to search
+ * @ret: a pointer to the uint64_t that will hold the found value
+ *
+ * Look up in the source if a value with same value pointer
+ * exists.
+ * If not, it will return -ENOENT. If it exists and it's a simple value
+ * (not an aggregate), the value that it points to will be returned.
+ * If it exists and it's an aggregate (aggr_type != STATS_FS_NONE), all
+ * subordinates will be recursively searched and every simple value match
+ * will be used to aggregate the final result. For example if it's a sum,
+ * all suboordinates having the same value will be sum together.
+ *
+ * This function will return 0 it succeeds.
+ */
+int stats_fs_source_get_value(struct stats_fs_source *source,
+ struct stats_fs_value *val, uint64_t *ret);
+
+/**
+ * stats_fs_source_get_value_by_name - search a value in the source (and
+ * subordinates)
+ * @source: a pointer to the source that will be searched
+ * @name: a pointer to the string representing the value to search
+ * (for example "exits")
+ * @ret: a pointer to the uint64_t that will hold the found value
+ *
+ * Same as stats_fs_source_get_value, but initially the name is used
+ * to search in the given source if there is a value with a matching
+ * name. If so, stats_fs_source_get_value will be called with the found
+ * value, otherwise -ENOENT will be returned.
+ */
+int stats_fs_source_get_value_by_name(struct stats_fs_source *source,
+ char *name, uint64_t *ret);
+
+/**
+ * stats_fs_source_clear - search and clears a value in the source (and
+ * subordinates)
+ * @source: a pointer to the source that will be searched
+ * @val: a pointer to the stats_fs_value to search
+ *
+ * Look up in the source if a value with same value pointer
+ * exists.
+ * If not, it will return -ENOENT. If it exists and it's a simple value
+ * (not an aggregate), the value that it points to will be set to 0.
+ * If it exists and it's an aggregate (aggr_type != STATS_FS_NONE), all
+ * subordinates will be recursively searched and every simple value match
+ * will be set to 0.
+ *
+ * This function will return 0 it succeeds.
+ */
+int stats_fs_source_clear(struct stats_fs_source *source,
+ struct stats_fs_value *val);
+
+/**
+ * stats_fs_source_revoke - disconnect the source from its backing data
+ * @source: a pointer to the source that will be revoked
+ *
+ * Ensure that stats_fs will not access the data that were passed to
+ * stats_fs_source_add_value for this source.
+ *
+ * Because open files increase the reference count for a stats_fs_source,
+ * the source can end up living longer than the data that provides the
+ * values for the source. Calling stats_fs_source_revoke just before the
+ * backing data is freed avoids accesses to freed data structures. The
+ * sources will return 0.
+ */
+void stats_fs_source_revoke(struct stats_fs_source *source);
+
+/**
+ * stats_fs_source_get - increases refcount of source
+ * @source: a pointer to the source whose refcount will be increased
+ */
+void stats_fs_source_get(struct stats_fs_source *source);
+
+/**
+ * stats_fs_source_put - decreases refcount of source and deletes if needed
+ * @source: a pointer to the source whose refcount will be decreased
+ *
+ * If refcount arrives to zero, take care of deleting
+ * and free the source resources and files, by firstly recursively calling
+ * stats_fs_source_remove_subordinate to the child and then deleting
+ * its own files and allocations.
+ */
+void stats_fs_source_put(struct stats_fs_source *source);
+
+/**
+ * stats_fs_initialized - returns true if stats_fs fs has been registered
+ */
+bool stats_fs_initialized(void);
+
+#else
+
+#include <linux/err.h>
+
+/*
+ * We do not return NULL from these functions if CONFIG_STATS_FS is not enabled
+ * so users have a chance to detect if there was a real error or not. We don't
+ * want to duplicate the design decision mistakes of procfs and devfs again.
+ */
+
+static inline struct stats_fs_source *stats_fs_source_create(const char *fmt,
+ ...)
+{
+ return ERR_PTR(-ENODEV);
+}
+
+static inline int stats_fs_source_add_values(struct stats_fs_source *source,
+ struct stats_fs_value *val,
+ void *base_ptr)
+{
+ return -ENODEV;
+}
+
+static inline void
+stats_fs_source_add_subordinate(struct stats_fs_source *parent,
+ struct stats_fs_source *child)
+{ }
+
+static inline void
+stats_fs_source_remove_subordinate(struct stats_fs_source *parent,
+ struct stats_fs_source *child)
+{ }
+
+static inline int stats_fs_source_get_value(struct stats_fs_source *source,
+ struct stats_fs_value *val,
+ uint64_t *ret)
+{
+ return -ENODEV;
+}
+
+static inline int
+stats_fs_source_get_value_by_name(struct stats_fs_source *source, char *name,
+ uint64_t *ret)
+{
+ return -ENODEV;
+}
+
+static inline int stats_fs_source_clear(struct stats_fs_source *source,
+ struct stats_fs_value *val)
+{
+ return -ENODEV;
+}
+
+static inline void stats_fs_source_revoke(struct stats_fs_source *source)
+{ }
+
+static inline void stats_fs_source_get(struct stats_fs_source *source)
+{ }
+
+static inline void stats_fs_source_put(struct stats_fs_source *source)
+{ }
+
+static inline bool stats_fs_initialized(void)
+{ }
+
+#endif
+
+#endif
--
2.25.2
^ permalink raw reply related
* [PATCH v2 0/5] Statsfs: a new ram-based file sytem for Linux kernel statistics
From: Emanuele Giuseppe Esposito @ 2020-05-04 11:03 UTC (permalink / raw)
To: kvm
Cc: Emanuele Giuseppe Esposito, linux-s390, David Hildenbrand,
Cornelia Huck, Emanuele Giuseppe Esposito, linux-kernel, kvm-ppc,
linux-mips, Christian Borntraeger, Alexander Viro, linux-fsdevel,
Paolo Bonzini, Vitaly Kuznetsov, linuxppc-dev, Jim Mattson
There is currently no common way for Linux kernel subsystems to expose
statistics to userspace shared throughout the Linux kernel; subsystems
have to take care of gathering and displaying statistics by themselves,
for example in the form of files in debugfs. For example KVM has its own
code section that takes care of this in virt/kvm/kvm_main.c, where it sets
up debugfs handlers for displaying values and aggregating them from
various subfolders to obtain information about the system state (i.e.
displaying the total number of exits, calculated by summing all exits of
all cpus of all running virtual machines).
Allowing each section of the kernel to do so has two disadvantages. First,
it will introduce redundant code. Second, debugfs is anyway not the right
place for statistics (for example it is affected by lockdown)
In this patch series I introduce statsfs, a synthetic ram-based virtual
filesystem that takes care of gathering and displaying statistics for the
Linux kernel subsystems.
The file system is mounted on /sys/kernel/stats and would be already used
by kvm. Statsfs was initially introduced by Paolo Bonzini [1].
Statsfs offers a generic and stable API, allowing any kind of
directory/file organization and supporting multiple kind of aggregations
(not only sum, but also average, max, min and count_zero) and data types
(all unsigned and signed types plus boolean). The implementation, which is
a generalization of KVM’s debugfs statistics code, takes care of gathering
and displaying information at run time; users only need to specify the
values to be included in each source.
Statsfs would also be a different mountpoint from debugfs, and would not
suffer from limited access due to the security lock down patches. Its main
function is to display each statistics as a file in the desired folder
hierarchy defined through the API. Statsfs files can be read, and possibly
cleared if their file mode allows it.
Statsfs has two main components: the public API defined by
include/linux/statsfs.h, and the virtual file system which should end up
in /sys/kernel/stats.
The API has two main elements, values and sources. Kernel subsystems like
KVM can use the API to create a source, add child
sources/values/aggregates and register it to the root source (that on the
virtual fs would be /sys/kernel/statsfs).
Sources are created via statsfs_source_create(), and each source becomes a
directory in the file system. Sources form a parent-child relationship;
root sources are added to the file system via statsfs_source_register().
Every other source is added to or removed from a parent through the
statsfs_source_add_subordinate and statsfs_source_remote_subordinate APIs.
Once a source is created and added to the tree (via add_subordinate), it
will be used to compute aggregate values in the parent source.
Values represent quantites that are gathered by the statsfs user. Examples
of values include the number of vm exits of a given kind, the amount of
memory used by some data structure, the length of the longest hash table
chain, or anything like that. Values are defined with the
statsfs_source_add_values function. Each value is defined by a struct
statsfs_value; the same statsfs_value can be added to many different
sources. A value can be considered "simple" if it fetches data from a
user-provided location, or "aggregate" if it groups all values in the
subordinates sources that include the same statsfs_value.
For more information, please consult the kerneldoc documentation in patch
2 and the sample uses in the kunit tests and in KVM.
This series of patches is based on my previous series "libfs: group and
simplify linux fs code" and the single patch sent to kvm "kvm_host: unify
VM_STAT and VCPU_STAT definitions in a single place". The former
simplifies code duplicated in debugfs and tracefs (from which statsfs is
based on), the latter groups all macros definition for statistics in kvm
in a single common file shared by all architectures.
Patch 1 adds a new refcount and kref destructor wrappers that take a
semaphore, as those are used later by statsfs. Patch 2 introduces the
statsfs API, patch 3 provides extensive tests that can also be used as
example on how to use the API and patch 4 adds the file system support.
Finally, patch 5 provides a real-life example of statsfs usage in KVM.
[1] https://lore.kernel.org/kvm/5d6cdcb1-d8ad-7ae6-7351-3544e2fa366d@redhat.com/?fbclid=IwAR18LHJ0PBcXcDaLzILFhHsl3qpT3z2vlG60RnqgbpGYhDv7L43n0ZXJY8M
Signed-off-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
v1->v2 remove unnecessary list_foreach_safe loops, fix wrong indentation,
change statsfs in stats_fs
Emanuele Giuseppe Esposito (5):
refcount, kref: add dec-and-test wrappers for rw_semaphores
stats_fs API: create, add and remove stats_fs sources and values
kunit: tests for stats_fs API
stats_fs fs: virtual fs to show stats to the end-user
kvm_main: replace debugfs with stats_fs
MAINTAINERS | 7 +
arch/arm64/kvm/Kconfig | 1 +
arch/arm64/kvm/guest.c | 2 +-
arch/mips/kvm/Kconfig | 1 +
arch/mips/kvm/mips.c | 2 +-
arch/powerpc/kvm/Kconfig | 1 +
arch/powerpc/kvm/book3s.c | 6 +-
arch/powerpc/kvm/booke.c | 8 +-
arch/s390/kvm/Kconfig | 1 +
arch/s390/kvm/kvm-s390.c | 16 +-
arch/x86/include/asm/kvm_host.h | 2 +-
arch/x86/kvm/Kconfig | 1 +
arch/x86/kvm/Makefile | 2 +-
arch/x86/kvm/debugfs.c | 64 --
arch/x86/kvm/stats_fs.c | 56 ++
arch/x86/kvm/x86.c | 6 +-
fs/Kconfig | 12 +
fs/Makefile | 1 +
fs/stats_fs/Makefile | 6 +
fs/stats_fs/inode.c | 337 ++++++++++
fs/stats_fs/internal.h | 35 +
fs/stats_fs/stats_fs-tests.c | 1088 +++++++++++++++++++++++++++++++
fs/stats_fs/stats_fs.c | 773 ++++++++++++++++++++++
include/linux/kref.h | 11 +
include/linux/kvm_host.h | 39 +-
include/linux/refcount.h | 2 +
include/linux/stats_fs.h | 304 +++++++++
include/uapi/linux/magic.h | 1 +
lib/refcount.c | 32 +
tools/lib/api/fs/fs.c | 21 +
virt/kvm/arm/arm.c | 2 +-
virt/kvm/kvm_main.c | 314 ++-------
32 files changed, 2772 insertions(+), 382 deletions(-)
delete mode 100644 arch/x86/kvm/debugfs.c
create mode 100644 arch/x86/kvm/stats_fs.c
create mode 100644 fs/stats_fs/Makefile
create mode 100644 fs/stats_fs/inode.c
create mode 100644 fs/stats_fs/internal.h
create mode 100644 fs/stats_fs/stats_fs-tests.c
create mode 100644 fs/stats_fs/stats_fs.c
create mode 100644 include/linux/stats_fs.h
--
2.25.2
^ permalink raw reply
* [PATCH v2 3/5] kunit: tests for stats_fs API
From: Emanuele Giuseppe Esposito @ 2020-05-04 11:03 UTC (permalink / raw)
To: kvm
Cc: Emanuele Giuseppe Esposito, linux-s390, David Hildenbrand,
Cornelia Huck, Emanuele Giuseppe Esposito, linux-kernel, kvm-ppc,
linux-mips, Christian Borntraeger, Alexander Viro, linux-fsdevel,
Paolo Bonzini, Vitaly Kuznetsov, linuxppc-dev, Jim Mattson
In-Reply-To: <20200504110344.17560-1-eesposit@redhat.com>
Add kunit tests to extensively test the stats_fs API functionality.
In order to run them, the kernel .config must set CONFIG_KUNIT=y
and a new .kunitconfig file must be created with CONFIG_STATS_FS=y
and CONFIG_STATS_FS_TEST=y
Tests can be then started by running the following command from the root
directory of the linux kernel source tree:
./tools/testing/kunit/kunit.py run --timeout=30 --jobs=`nproc --all`
Signed-off-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
---
fs/Kconfig | 6 +
fs/stats_fs/Makefile | 2 +
fs/stats_fs/stats_fs-tests.c | 1088 ++++++++++++++++++++++++++++++++++
3 files changed, 1096 insertions(+)
create mode 100644 fs/stats_fs/stats_fs-tests.c
diff --git a/fs/Kconfig b/fs/Kconfig
index 1b0de0f19e96..0844e8defd22 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -334,4 +334,10 @@ config STATS_FS
stats_fs is a virtual file system that provides counters and
other statistics about the running kernel.
+config STATS_FS_TEST
+ bool "Tests for stats_fs"
+ depends on STATS_FS && KUNIT
+ help
+ tests for the stats_fs API.
+
endmenu
diff --git a/fs/stats_fs/Makefile b/fs/stats_fs/Makefile
index 94fe52d590d5..9db130fac6b6 100644
--- a/fs/stats_fs/Makefile
+++ b/fs/stats_fs/Makefile
@@ -1,4 +1,6 @@
# SPDX-License-Identifier: GPL-2.0-only
stats_fs-objs := stats_fs.o
+stats_fs-tests-objs := stats_fs-tests.o
obj-$(CONFIG_STATS_FS) += stats_fs.o
+obj-$(CONFIG_STATS_FS_TEST) += stats_fs-tests.o
diff --git a/fs/stats_fs/stats_fs-tests.c b/fs/stats_fs/stats_fs-tests.c
new file mode 100644
index 000000000000..46c3fb510ee9
--- /dev/null
+++ b/fs/stats_fs/stats_fs-tests.c
@@ -0,0 +1,1088 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/module.h>
+#include <linux/errno.h>
+#include <linux/file.h>
+#include <linux/fs.h>
+#include <linux/slab.h>
+#include <linux/anon_inodes.h>
+#include <linux/spinlock.h>
+#include <linux/uaccess.h>
+#include <linux/rwsem.h>
+#include <linux/list.h>
+#include <linux/kref.h>
+
+#include <linux/limits.h>
+#include <linux/stats_fs.h>
+#include <kunit/test.h>
+#include "internal.h"
+
+#define STATS_FS_STAT(el, x, ...) \
+ { \
+ .name = #x, .offset = offsetof(struct container, el.x), \
+ ##__VA_ARGS__ \
+ }
+
+#define ARR_SIZE(el) ((int)(sizeof(el) / sizeof(struct stats_fs_value) - 1))
+
+struct test_values_struct {
+ uint64_t u64;
+ int32_t s32;
+ bool bo;
+ uint8_t u8;
+ int16_t s16;
+};
+
+struct container {
+ struct test_values_struct vals;
+};
+
+struct stats_fs_value test_values[6] = {
+ STATS_FS_STAT(vals, u64, .type = STATS_FS_U64,
+ .aggr_kind = STATS_FS_NONE, .mode = 0),
+ STATS_FS_STAT(vals, s32, .type = STATS_FS_S32,
+ .aggr_kind = STATS_FS_NONE, .mode = 0),
+ STATS_FS_STAT(vals, bo, .type = STATS_FS_BOOL,
+ .aggr_kind = STATS_FS_NONE, .mode = 0),
+ STATS_FS_STAT(vals, u8, .type = STATS_FS_U8, .aggr_kind = STATS_FS_NONE,
+ .mode = 0),
+ STATS_FS_STAT(vals, s16, .type = STATS_FS_S16,
+ .aggr_kind = STATS_FS_NONE, .mode = 0),
+ { NULL },
+};
+
+struct stats_fs_value test_aggr[4] = {
+ STATS_FS_STAT(vals, s32, .type = STATS_FS_S32,
+ .aggr_kind = STATS_FS_MIN, .mode = 0),
+ STATS_FS_STAT(vals, bo, .type = STATS_FS_BOOL,
+ .aggr_kind = STATS_FS_MAX, .mode = 0),
+ STATS_FS_STAT(vals, u64, .type = STATS_FS_U64,
+ .aggr_kind = STATS_FS_SUM, .mode = 0),
+ { NULL },
+};
+
+struct stats_fs_value test_same_name[3] = {
+ STATS_FS_STAT(vals, s32, .type = STATS_FS_S32,
+ .aggr_kind = STATS_FS_NONE, .mode = 0),
+ STATS_FS_STAT(vals, s32, .type = STATS_FS_S32,
+ .aggr_kind = STATS_FS_MIN, .mode = 0),
+ { NULL },
+};
+
+struct stats_fs_value test_all_aggr[6] = {
+ STATS_FS_STAT(vals, s32, .type = STATS_FS_S32,
+ .aggr_kind = STATS_FS_MIN, .mode = 0),
+ STATS_FS_STAT(vals, bo, .type = STATS_FS_BOOL,
+ .aggr_kind = STATS_FS_COUNT_ZERO, .mode = 0),
+ STATS_FS_STAT(vals, u64, .type = STATS_FS_U64,
+ .aggr_kind = STATS_FS_SUM, .mode = 0),
+ STATS_FS_STAT(vals, u8, .type = STATS_FS_U8, .aggr_kind = STATS_FS_AVG,
+ .mode = 0),
+ STATS_FS_STAT(vals, s16, .type = STATS_FS_S16,
+ .aggr_kind = STATS_FS_MAX, .mode = 0),
+ { NULL },
+};
+
+#define def_u64 ((uint64_t)64)
+
+#define def_val_s32 ((int32_t)S32_MIN)
+#define def_val_bool ((bool)true)
+#define def_val_u8 ((uint8_t)127)
+#define def_val_s16 ((int16_t)10000)
+
+#define def_val2_s32 ((int32_t)S16_MAX)
+#define def_val2_bool ((bool)false)
+#define def_val2_u8 ((uint8_t)255)
+#define def_val2_s16 ((int16_t)-20000)
+
+struct container cont = {
+ .vals = {
+ .u64 = def_u64,
+ .s32 = def_val_s32,
+ .bo = def_val_bool,
+ .u8 = def_val_u8,
+ .s16 = def_val_s16,
+ },
+};
+
+struct container cont2 = {
+ .vals = {
+ .u64 = def_u64,
+ .s32 = def_val2_s32,
+ .bo = def_val2_bool,
+ .u8 = def_val2_u8,
+ .s16 = def_val2_s16,
+ },
+};
+
+static void get_stats_at_addr(struct stats_fs_source *src, void *addr,
+ int *aggr, int *val, int use_addr)
+{
+ struct stats_fs_value *entry;
+ struct stats_fs_value_source *src_entry;
+ int counter_val = 0, counter_aggr = 0;
+
+ list_for_each_entry (src_entry, &src->values_head, list_element) {
+ if (use_addr && src_entry->base_addr != addr)
+ continue;
+
+ for (entry = src_entry->values; entry->name; entry++) {
+ if (entry->aggr_kind == STATS_FS_NONE)
+ counter_val++;
+ else
+ counter_aggr++;
+ }
+ }
+
+ if (aggr)
+ *aggr = counter_aggr;
+
+ if (val)
+ *val = counter_val;
+}
+
+int source_has_subsource(struct stats_fs_source *src,
+ struct stats_fs_source *sub)
+{
+ struct stats_fs_source *entry;
+
+ list_for_each_entry (entry, &src->subordinates_head, list_element) {
+ if (entry == sub)
+ return 1;
+ }
+ return 0;
+}
+
+int get_number_subsources(struct stats_fs_source *src)
+{
+ struct stats_fs_source *entry;
+ int counter = 0;
+
+ list_for_each_entry (entry, &src->subordinates_head, list_element) {
+ counter++;
+ }
+ return counter;
+}
+
+int get_number_values(struct stats_fs_source *src)
+{
+ int counter = 0;
+
+ get_stats_at_addr(src, NULL, NULL, &counter, 0);
+ return counter;
+}
+
+int get_total_number_values(struct stats_fs_source *src)
+{
+ struct stats_fs_source *sub_entry;
+ int counter = 0;
+
+ get_stats_at_addr(src, NULL, NULL, &counter, 0);
+
+ list_for_each_entry (sub_entry, &src->subordinates_head, list_element) {
+ counter += get_total_number_values(sub_entry);
+ }
+
+ return counter;
+}
+
+int get_number_aggregates(struct stats_fs_source *src)
+{
+ int counter = 0;
+
+ get_stats_at_addr(src, NULL, &counter, NULL, 1);
+ return counter;
+}
+
+int get_number_values_with_base(struct stats_fs_source *src, void *addr)
+{
+ int counter = 0;
+
+ get_stats_at_addr(src, addr, NULL, &counter, 1);
+ return counter;
+}
+
+int get_number_aggr_with_base(struct stats_fs_source *src, void *addr)
+{
+ int counter = 0;
+
+ get_stats_at_addr(src, addr, &counter, NULL, 1);
+ return counter;
+}
+
+static void test_empty_folder(struct kunit *test)
+{
+ struct stats_fs_source *src;
+
+ src = stats_fs_source_create("kvm_%d", 123);
+ KUNIT_EXPECT_EQ(test, strcmp(src->name, "kvm_123"), 0);
+ KUNIT_EXPECT_EQ(test, get_number_subsources(src), 0);
+ KUNIT_EXPECT_EQ(test, get_number_values(src), 0);
+ KUNIT_EXPECT_EQ(test, get_number_aggregates(src), 0);
+ stats_fs_source_put(src);
+}
+
+static void test_add_subfolder(struct kunit *test)
+{
+ struct stats_fs_source *src, *sub;
+
+ src = stats_fs_source_create("parent");
+ sub = stats_fs_source_create("child");
+ stats_fs_source_add_subordinate(src, sub);
+ KUNIT_EXPECT_EQ(test, source_has_subsource(src, sub), true);
+ KUNIT_EXPECT_EQ(test, get_number_subsources(src), 1);
+ KUNIT_EXPECT_EQ(test, get_number_values(src), 0);
+ KUNIT_EXPECT_EQ(test, get_number_aggregates(src), 0);
+ KUNIT_EXPECT_EQ(test, get_number_values(sub), 0);
+ KUNIT_EXPECT_EQ(test, get_number_aggregates(sub), 0);
+ KUNIT_EXPECT_EQ(test, get_total_number_values(src), 0);
+
+ stats_fs_source_put(sub);
+ sub = stats_fs_source_create("not a child");
+ KUNIT_EXPECT_EQ(test, source_has_subsource(src, sub), false);
+ KUNIT_EXPECT_EQ(test, get_number_subsources(src), 1);
+
+ stats_fs_source_put(sub);
+ stats_fs_source_put(src);
+}
+
+static void test_add_value(struct kunit *test)
+{
+ struct stats_fs_source *src;
+ int n;
+
+ src = stats_fs_source_create("parent");
+
+ // add values
+ n = stats_fs_source_add_values(src, test_values, &cont);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_values_with_base(src, &cont);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_values));
+
+ // add same values, nothing happens
+ n = stats_fs_source_add_values(src, test_values, &cont);
+ KUNIT_EXPECT_EQ(test, n, -EEXIST);
+ n = get_number_values_with_base(src, &cont);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_values));
+
+ // size is invaried
+ KUNIT_EXPECT_EQ(test, get_number_values(src), ARR_SIZE(test_values));
+
+ // no aggregates
+ n = get_number_aggr_with_base(src, &cont);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, get_number_values(src), ARR_SIZE(test_values));
+ KUNIT_EXPECT_EQ(test, get_number_aggregates(src), 0);
+
+ stats_fs_source_put(src);
+}
+
+static void test_add_value_in_subfolder(struct kunit *test)
+{
+ struct stats_fs_source *src, *sub, *sub_not;
+ int n;
+
+ src = stats_fs_source_create("parent");
+ sub = stats_fs_source_create("child");
+
+ // src -> sub
+ stats_fs_source_add_subordinate(src, sub);
+
+ // add values
+ n = stats_fs_source_add_values(sub, test_values, &cont);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_values_with_base(sub, &cont);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_values));
+ KUNIT_EXPECT_EQ(test, get_number_values(src), 0);
+ KUNIT_EXPECT_EQ(test, get_number_aggregates(src), 0);
+ KUNIT_EXPECT_EQ(test, get_total_number_values(src),
+ ARR_SIZE(test_values));
+
+ KUNIT_EXPECT_EQ(test, get_number_values(sub), ARR_SIZE(test_values));
+ // no values in sub
+ KUNIT_EXPECT_EQ(test, get_number_aggregates(sub), 0);
+
+ // different folder
+ sub_not = stats_fs_source_create("not a child");
+
+ // add values
+ n = stats_fs_source_add_values(sub_not, test_values, &cont);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_values_with_base(sub_not, &cont);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_values));
+ KUNIT_EXPECT_EQ(test, get_number_values(src), 0);
+ KUNIT_EXPECT_EQ(test, get_number_aggregates(src), 0);
+ KUNIT_EXPECT_EQ(test, get_total_number_values(src),
+ ARR_SIZE(test_values));
+
+ // remove sub, check values is 0
+ stats_fs_source_remove_subordinate(src, sub);
+ KUNIT_EXPECT_EQ(test, get_total_number_values(src), 0);
+
+ // re-add sub, check value are added
+ stats_fs_source_add_subordinate(src, sub);
+ KUNIT_EXPECT_EQ(test, get_total_number_values(src),
+ ARR_SIZE(test_values));
+
+ // add sub_not, check value are twice as many
+ stats_fs_source_add_subordinate(src, sub_not);
+ KUNIT_EXPECT_EQ(test, get_total_number_values(src),
+ ARR_SIZE(test_values) * 2);
+
+ KUNIT_EXPECT_EQ(test, get_number_values(sub_not),
+ ARR_SIZE(test_values));
+ KUNIT_EXPECT_EQ(test, get_number_aggregates(sub_not), 0);
+
+ stats_fs_source_put(sub);
+ stats_fs_source_put(sub_not);
+ stats_fs_source_put(src);
+}
+
+static void test_search_value(struct kunit *test)
+{
+ struct stats_fs_source *src;
+ uint64_t ret;
+ int n;
+
+ src = stats_fs_source_create("parent");
+
+ // add values
+ n = stats_fs_source_add_values(src, test_values, &cont);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_values_with_base(src, &cont);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_values));
+
+ // get u64
+ n = stats_fs_source_get_value_by_name(src, "u64", &ret);
+ KUNIT_EXPECT_EQ(test, ret, def_u64);
+ KUNIT_EXPECT_EQ(test, n, 0);
+
+ n = stats_fs_source_get_value_by_name(src, "s32", &ret);
+ KUNIT_EXPECT_EQ(test, ((int32_t)ret), def_val_s32);
+ KUNIT_EXPECT_EQ(test, n, 0);
+
+ n = stats_fs_source_get_value_by_name(src, "bo", &ret);
+ KUNIT_EXPECT_EQ(test, ((bool)ret), def_val_bool);
+ KUNIT_EXPECT_EQ(test, n, 0);
+
+ // get a non-added value
+ n = stats_fs_source_get_value_by_name(src, "does not exist", &ret);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ KUNIT_EXPECT_EQ(test, n, -ENOENT);
+
+ stats_fs_source_put(src);
+}
+
+static void test_search_value_in_subfolder(struct kunit *test)
+{
+ struct stats_fs_source *src, *sub;
+ uint64_t ret;
+ int n;
+
+ src = stats_fs_source_create("parent");
+ sub = stats_fs_source_create("child");
+
+ // src -> sub
+ stats_fs_source_add_subordinate(src, sub);
+
+ // add values to sub
+ n = stats_fs_source_add_values(sub, test_values, &cont);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_values_with_base(sub, &cont);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_values));
+
+ n = stats_fs_source_get_value_by_name(sub, "u64", &ret);
+ KUNIT_EXPECT_EQ(test, ret, def_u64);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = stats_fs_source_get_value_by_name(src, "u64", &ret);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ KUNIT_EXPECT_EQ(test, n, -ENOENT);
+
+ n = stats_fs_source_get_value_by_name(sub, "s32", &ret);
+ KUNIT_EXPECT_EQ(test, ((int32_t)ret), def_val_s32);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = stats_fs_source_get_value_by_name(src, "s32", &ret);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ KUNIT_EXPECT_EQ(test, n, -ENOENT);
+
+ n = stats_fs_source_get_value_by_name(sub, "bo", &ret);
+ KUNIT_EXPECT_EQ(test, ((bool)ret), def_val_bool);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = stats_fs_source_get_value_by_name(src, "bo", &ret);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ KUNIT_EXPECT_EQ(test, n, -ENOENT);
+
+ n = stats_fs_source_get_value_by_name(sub, "does not exist", &ret);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ KUNIT_EXPECT_EQ(test, n, -ENOENT);
+ n = stats_fs_source_get_value_by_name(src, "does not exist", &ret);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ KUNIT_EXPECT_EQ(test, n, -ENOENT);
+
+ stats_fs_source_put(sub);
+ stats_fs_source_put(src);
+}
+
+static void test_search_value_in_empty_folder(struct kunit *test)
+{
+ struct stats_fs_source *src;
+ uint64_t ret;
+ int n;
+
+ src = stats_fs_source_create("empty folder");
+ KUNIT_EXPECT_EQ(test, get_number_aggregates(src), 0);
+ KUNIT_EXPECT_EQ(test, get_number_subsources(src), 0);
+ KUNIT_EXPECT_EQ(test, get_number_values(src), 0);
+
+ n = stats_fs_source_get_value_by_name(src, "u64", &ret);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ KUNIT_EXPECT_EQ(test, n, -ENOENT);
+
+ n = stats_fs_source_get_value_by_name(src, "s32", &ret);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ KUNIT_EXPECT_EQ(test, n, -ENOENT);
+
+ n = stats_fs_source_get_value_by_name(src, "bo", &ret);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ KUNIT_EXPECT_EQ(test, n, -ENOENT);
+
+ n = stats_fs_source_get_value_by_name(src, "does not exist", &ret);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ KUNIT_EXPECT_EQ(test, n, -ENOENT);
+
+ stats_fs_source_put(src);
+}
+
+static void test_add_aggregate(struct kunit *test)
+{
+ struct stats_fs_source *src;
+ int n;
+
+ src = stats_fs_source_create("parent");
+
+ // add aggr to src, no values
+ n = stats_fs_source_add_values(src, test_aggr, NULL);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_values_with_base(src, NULL);
+ KUNIT_EXPECT_EQ(test, n, 0);
+
+ // count values
+ n = get_number_aggr_with_base(src, NULL);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_aggr));
+
+ // add same array again, should not be added
+ n = stats_fs_source_add_values(src, test_aggr, NULL);
+ KUNIT_EXPECT_EQ(test, n, -EEXIST);
+ n = get_number_aggr_with_base(src, NULL);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_aggr));
+
+ KUNIT_EXPECT_EQ(test, get_number_values(src), 0);
+ KUNIT_EXPECT_EQ(test, get_number_aggregates(src), ARR_SIZE(test_aggr));
+
+ stats_fs_source_put(src);
+}
+
+static void test_add_aggregate_in_subfolder(struct kunit *test)
+{
+ struct stats_fs_source *src, *sub, *sub_not;
+ int n;
+
+ src = stats_fs_source_create("parent");
+ sub = stats_fs_source_create("child");
+ // src->sub
+ stats_fs_source_add_subordinate(src, sub);
+
+ // add aggr to sub
+ n = stats_fs_source_add_values(sub, test_aggr, NULL);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_aggr_with_base(sub, NULL);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_aggr));
+ KUNIT_EXPECT_EQ(test, get_number_values(src), 0);
+ KUNIT_EXPECT_EQ(test, get_number_aggregates(src), 0);
+ KUNIT_EXPECT_EQ(test, get_total_number_values(src), 0);
+
+ KUNIT_EXPECT_EQ(test, get_number_values(sub), 0);
+ KUNIT_EXPECT_EQ(test, get_number_aggregates(sub), ARR_SIZE(test_aggr));
+
+ // not a child
+ sub_not = stats_fs_source_create("not a child");
+
+ // add aggr to "not a child"
+ n = stats_fs_source_add_values(sub_not, test_aggr, NULL);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_aggr_with_base(sub_not, NULL);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_aggr));
+ KUNIT_EXPECT_EQ(test, get_number_values(src), 0);
+ KUNIT_EXPECT_EQ(test, get_number_aggregates(src), 0);
+ KUNIT_EXPECT_EQ(test, get_total_number_values(src), 0);
+
+ // remove sub
+ stats_fs_source_remove_subordinate(src, sub);
+ KUNIT_EXPECT_EQ(test, get_total_number_values(src), 0);
+
+ // re-add both
+ stats_fs_source_add_subordinate(src, sub);
+ KUNIT_EXPECT_EQ(test, get_total_number_values(src), 0);
+ stats_fs_source_add_subordinate(src, sub_not);
+ KUNIT_EXPECT_EQ(test, get_total_number_values(src), 0);
+
+ KUNIT_EXPECT_EQ(test, get_number_values(sub_not), 0);
+ KUNIT_EXPECT_EQ(test, get_number_aggregates(sub_not),
+ ARR_SIZE(test_aggr));
+
+ stats_fs_source_put(sub);
+ stats_fs_source_put(sub_not);
+ stats_fs_source_put(src);
+}
+
+static void test_search_aggregate(struct kunit *test)
+{
+ struct stats_fs_source *src;
+ uint64_t ret;
+ int n;
+
+ src = stats_fs_source_create("parent");
+ n = stats_fs_source_add_values(src, test_aggr, NULL);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_aggr_with_base(src, NULL);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_aggr));
+ n = get_number_aggr_with_base(src, &cont);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = stats_fs_source_get_value_by_name(src, "u64", &ret);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ KUNIT_EXPECT_EQ(test, n, 0);
+
+ n = stats_fs_source_get_value_by_name(src, "s32", &ret);
+ KUNIT_EXPECT_EQ(test, (int64_t)ret, S64_MAX);
+ KUNIT_EXPECT_EQ(test, n, 0);
+
+ n = stats_fs_source_get_value_by_name(src, "bo", &ret);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ KUNIT_EXPECT_EQ(test, n, 0);
+
+ n = stats_fs_source_get_value_by_name(src, "does not exist", &ret);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ KUNIT_EXPECT_EQ(test, n, -ENOENT);
+ stats_fs_source_put(src);
+}
+
+static void test_search_aggregate_in_subfolder(struct kunit *test)
+{
+ struct stats_fs_source *src, *sub;
+ uint64_t ret;
+ int n;
+
+ src = stats_fs_source_create("parent");
+ sub = stats_fs_source_create("child");
+
+ stats_fs_source_add_subordinate(src, sub);
+
+ n = stats_fs_source_add_values(sub, test_aggr, NULL);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_aggr_with_base(sub, NULL);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_aggr));
+ n = get_number_aggr_with_base(sub, &cont);
+ KUNIT_EXPECT_EQ(test, n, 0);
+
+ // no u64 in test_aggr
+ n = stats_fs_source_get_value_by_name(sub, "u64", &ret);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = stats_fs_source_get_value_by_name(src, "u64", &ret);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ KUNIT_EXPECT_EQ(test, n, -ENOENT);
+
+ n = stats_fs_source_get_value_by_name(sub, "s32", &ret);
+ KUNIT_EXPECT_EQ(test, (int64_t)ret, S64_MAX);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = stats_fs_source_get_value_by_name(src, "s32", &ret);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ KUNIT_EXPECT_EQ(test, n, -ENOENT);
+
+ n = stats_fs_source_get_value_by_name(sub, "bo", &ret);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = stats_fs_source_get_value_by_name(src, "bo", &ret);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ KUNIT_EXPECT_EQ(test, n, -ENOENT);
+
+ n = stats_fs_source_get_value_by_name(sub, "does not exist", &ret);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ KUNIT_EXPECT_EQ(test, n, -ENOENT);
+ n = stats_fs_source_get_value_by_name(src, "does not exist", &ret);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ KUNIT_EXPECT_EQ(test, n, -ENOENT);
+
+ stats_fs_source_put(sub);
+ stats_fs_source_put(src);
+}
+
+void test_search_same(struct kunit *test)
+{
+ struct stats_fs_source *src;
+ uint64_t ret;
+ int n;
+
+ src = stats_fs_source_create("parent");
+ n = stats_fs_source_add_values(src, test_same_name, &cont);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_values_with_base(src, &cont);
+ KUNIT_EXPECT_EQ(test, n, 1);
+ n = get_number_aggr_with_base(src, &cont);
+ KUNIT_EXPECT_EQ(test, n, 1);
+
+ n = stats_fs_source_add_values(src, test_same_name, &cont);
+ KUNIT_EXPECT_EQ(test, n, -EEXIST);
+ n = get_number_values_with_base(src, &cont);
+ KUNIT_EXPECT_EQ(test, n, 1);
+ n = get_number_aggr_with_base(src, &cont);
+ KUNIT_EXPECT_EQ(test, n, 1);
+
+ // returns first the value
+ n = stats_fs_source_get_value_by_name(src, "s32", &ret);
+ KUNIT_EXPECT_EQ(test, ((int32_t)ret), def_val_s32);
+ KUNIT_EXPECT_EQ(test, n, 0);
+
+ stats_fs_source_put(src);
+}
+
+static void test_add_mixed(struct kunit *test)
+{
+ struct stats_fs_source *src;
+ int n;
+
+ src = stats_fs_source_create("parent");
+
+ n = stats_fs_source_add_values(src, test_aggr, NULL);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_values_with_base(src, NULL);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = stats_fs_source_add_values(src, test_values, &cont);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_aggr_with_base(src, NULL);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_aggr));
+
+ n = stats_fs_source_add_values(src, test_values, &cont);
+ KUNIT_EXPECT_EQ(test, n, -EEXIST);
+ n = get_number_values_with_base(src, &cont);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_values));
+ n = stats_fs_source_add_values(src, test_aggr, NULL);
+ KUNIT_EXPECT_EQ(test, n, -EEXIST);
+ n = get_number_aggr_with_base(src, NULL);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_aggr));
+
+ KUNIT_EXPECT_EQ(test, get_number_values(src), ARR_SIZE(test_values));
+ KUNIT_EXPECT_EQ(test, get_number_aggregates(src), ARR_SIZE(test_aggr));
+ stats_fs_source_put(src);
+}
+
+static void test_search_mixed(struct kunit *test)
+{
+ struct stats_fs_source *src, *sub;
+ uint64_t ret;
+ int n;
+
+ src = stats_fs_source_create("parent");
+ sub = stats_fs_source_create("child");
+ stats_fs_source_add_subordinate(src, sub);
+
+ // src has the aggregates, sub the values. Just search
+ n = stats_fs_source_add_values(sub, test_values, &cont);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_values_with_base(sub, &cont);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_values));
+ n = stats_fs_source_add_values(src, test_aggr, &cont);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_aggr_with_base(src, &cont);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_aggr));
+
+ // u64 is sum so again same value
+ n = stats_fs_source_get_value_by_name(sub, "u64", &ret);
+ KUNIT_EXPECT_EQ(test, ret, def_u64);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = stats_fs_source_get_value_by_name(src, "u64", &ret);
+ KUNIT_EXPECT_EQ(test, ret, def_u64);
+ KUNIT_EXPECT_EQ(test, n, 0);
+
+ // s32 is min so return the value also in the aggregate
+ n = stats_fs_source_get_value_by_name(sub, "s32", &ret);
+ KUNIT_EXPECT_EQ(test, ((int32_t)ret), def_val_s32);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = stats_fs_source_get_value_by_name(src, "s32", &ret);
+ KUNIT_EXPECT_EQ(test, ((int32_t)ret), def_val_s32);
+ KUNIT_EXPECT_EQ(test, n, 0);
+
+ // bo is max
+ n = stats_fs_source_get_value_by_name(sub, "bo", &ret);
+ KUNIT_EXPECT_EQ(test, (bool)ret, def_val_bool);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = stats_fs_source_get_value_by_name(src, "bo", &ret);
+ KUNIT_EXPECT_EQ(test, (bool)ret, def_val_bool);
+ KUNIT_EXPECT_EQ(test, n, 0);
+
+ n = stats_fs_source_get_value_by_name(sub, "does not exist", &ret);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ KUNIT_EXPECT_EQ(test, n, -ENOENT);
+ n = stats_fs_source_get_value_by_name(src, "does not exist", &ret);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ KUNIT_EXPECT_EQ(test, n, -ENOENT);
+
+ stats_fs_source_put(sub);
+ stats_fs_source_put(src);
+}
+
+static void test_all_aggregations_agg_val_val(struct kunit *test)
+{
+ struct stats_fs_source *src, *sub1, *sub2;
+ uint64_t ret;
+ int n;
+
+ src = stats_fs_source_create("parent");
+ sub1 = stats_fs_source_create("child1");
+ sub2 = stats_fs_source_create("child2");
+ stats_fs_source_add_subordinate(src, sub1);
+ stats_fs_source_add_subordinate(src, sub2);
+
+ n = stats_fs_source_add_values(sub1, test_all_aggr, &cont);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_aggr_with_base(sub1, &cont);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_all_aggr));
+ n = stats_fs_source_add_values(sub2, test_all_aggr, &cont2);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_aggr_with_base(sub2, &cont2);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_all_aggr));
+
+ n = stats_fs_source_add_values(src, test_all_aggr, NULL);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_aggr_with_base(src, NULL);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_all_aggr));
+
+ // sum
+ n = stats_fs_source_get_value_by_name(src, "u64", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, ret, def_u64 * 2);
+
+ // min
+ n = stats_fs_source_get_value_by_name(src, "s32", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, ((int32_t)ret), def_val_s32);
+
+ // count_0
+ n = stats_fs_source_get_value_by_name(src, "bo", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, ret, 1ull);
+
+ // avg
+ n = stats_fs_source_get_value_by_name(src, "u8", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, ret, 191ull);
+
+ // max
+ n = stats_fs_source_get_value_by_name(src, "s16", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, (int16_t)ret, def_val_s16);
+
+ stats_fs_source_put(sub1);
+ stats_fs_source_put(sub2);
+ stats_fs_source_put(src);
+}
+
+static void test_all_aggregations_val_agg_val(struct kunit *test)
+{
+ struct stats_fs_source *src, *sub1, *sub2;
+ uint64_t ret;
+ int n;
+
+ src = stats_fs_source_create("parent");
+ sub1 = stats_fs_source_create("child1");
+ sub2 = stats_fs_source_create("child2");
+ stats_fs_source_add_subordinate(src, sub1);
+ stats_fs_source_add_subordinate(src, sub2);
+
+ n = stats_fs_source_add_values(src, test_all_aggr, &cont);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_aggr_with_base(src, &cont);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_all_aggr));
+ n = stats_fs_source_add_values(sub2, test_all_aggr, &cont2);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_aggr_with_base(sub2, &cont2);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_all_aggr));
+
+ n = stats_fs_source_add_values(sub1, test_all_aggr, NULL);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_aggr_with_base(sub1, NULL);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_all_aggr));
+
+ n = stats_fs_source_get_value_by_name(src, "u64", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, ret, def_u64);
+ n = stats_fs_source_get_value_by_name(sub1, "u64", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ n = stats_fs_source_get_value_by_name(sub2, "u64", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, ret, def_u64);
+
+ n = stats_fs_source_get_value_by_name(src, "s32", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, ((int32_t)ret), def_val_s32);
+ n = stats_fs_source_get_value_by_name(sub1, "s32", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, (int64_t)ret, S64_MAX); // MIN
+ n = stats_fs_source_get_value_by_name(sub2, "s32", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, ((int32_t)ret), def_val2_s32);
+
+ n = stats_fs_source_get_value_by_name(src, "bo", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, (bool)ret, def_val_bool);
+ n = stats_fs_source_get_value_by_name(sub1, "bo", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ n = stats_fs_source_get_value_by_name(sub2, "bo", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, (bool)ret, def_val2_bool);
+
+ n = stats_fs_source_get_value_by_name(src, "u8", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, (uint8_t)ret, def_val_u8);
+ n = stats_fs_source_get_value_by_name(sub1, "u8", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, ret, 0ull);
+ n = stats_fs_source_get_value_by_name(sub2, "u8", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, (uint8_t)ret, def_val2_u8);
+
+ n = stats_fs_source_get_value_by_name(src, "s16", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, (int16_t)ret, def_val_s16);
+ n = stats_fs_source_get_value_by_name(sub1, "s16", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, (int64_t)ret, S64_MIN); // MAX
+ n = stats_fs_source_get_value_by_name(sub2, "s16", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, (int16_t)ret, def_val2_s16);
+
+ stats_fs_source_put(sub1);
+ stats_fs_source_put(sub2);
+ stats_fs_source_put(src);
+}
+
+static void test_all_aggregations_agg_val_val_sub(struct kunit *test)
+{
+ struct stats_fs_source *src, *sub1, *sub11;
+ uint64_t ret;
+ int n;
+
+ src = stats_fs_source_create("parent");
+ sub1 = stats_fs_source_create("child1");
+ sub11 = stats_fs_source_create("child11");
+ stats_fs_source_add_subordinate(src, sub1);
+ stats_fs_source_add_subordinate(sub1, sub11); // changes here!
+
+ n = stats_fs_source_add_values(sub1, test_values, &cont);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_values_with_base(sub1, &cont);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_values));
+ n = stats_fs_source_add_values(sub11, test_values, &cont2);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_values_with_base(sub11, &cont2);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_values));
+
+ KUNIT_EXPECT_EQ(test, get_total_number_values(src),
+ ARR_SIZE(test_values) * 2);
+
+ n = stats_fs_source_add_values(sub1, test_all_aggr, &cont);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_aggr_with_base(sub1, &cont);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_all_aggr));
+ n = stats_fs_source_add_values(sub11, test_all_aggr, &cont2);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_aggr_with_base(sub11, &cont2);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_all_aggr));
+
+ n = stats_fs_source_add_values(src, test_all_aggr, NULL);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_aggr_with_base(src, NULL);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_all_aggr));
+
+ // sum
+ n = stats_fs_source_get_value_by_name(src, "u64", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, ret, def_u64 * 2);
+
+ // min
+ n = stats_fs_source_get_value_by_name(src, "s32", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, ((int32_t)ret), def_val_s32);
+
+ // count_0
+ n = stats_fs_source_get_value_by_name(src, "bo", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, ret, 1ull);
+
+ // avg
+ n = stats_fs_source_get_value_by_name(src, "u8", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, ret, 191ull);
+
+ // max
+ n = stats_fs_source_get_value_by_name(src, "s16", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, (int16_t)ret, def_val_s16);
+
+ stats_fs_source_put(sub1);
+ stats_fs_source_put(sub11);
+ stats_fs_source_put(src);
+}
+
+static void test_all_aggregations_agg_no_val_sub(struct kunit *test)
+{
+ struct stats_fs_source *src, *sub1, *sub11;
+ uint64_t ret;
+ int n;
+
+ src = stats_fs_source_create("parent");
+ sub1 = stats_fs_source_create("child1");
+ sub11 = stats_fs_source_create("child11");
+ stats_fs_source_add_subordinate(src, sub1);
+ stats_fs_source_add_subordinate(sub1, sub11);
+
+ n = stats_fs_source_add_values(sub11, test_all_aggr, &cont2);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_aggr_with_base(sub11, &cont2);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_all_aggr));
+
+ KUNIT_EXPECT_EQ(test, get_total_number_values(src), 0);
+
+ n = stats_fs_source_add_values(src, test_all_aggr, NULL);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_aggr_with_base(src, NULL);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_all_aggr));
+
+ // sum
+ n = stats_fs_source_get_value_by_name(src, "u64", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, ret, def_u64);
+
+ // min
+ n = stats_fs_source_get_value_by_name(src, "s32", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, ((int32_t)ret), def_val2_s32);
+
+ // count_0
+ n = stats_fs_source_get_value_by_name(src, "bo", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, ret, 1ull);
+
+ // avg
+ n = stats_fs_source_get_value_by_name(src, "u8", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, (uint8_t)ret, def_val2_u8);
+
+ // max
+ n = stats_fs_source_get_value_by_name(src, "s16", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, (int16_t)ret, def_val2_s16);
+
+ stats_fs_source_put(sub1);
+ stats_fs_source_put(sub11);
+ stats_fs_source_put(src);
+}
+
+static void test_all_aggregations_agg_agg_val_sub(struct kunit *test)
+{
+ struct stats_fs_source *src, *sub1, *sub11, *sub12;
+ uint64_t ret;
+ int n;
+
+ src = stats_fs_source_create("parent");
+ sub1 = stats_fs_source_create("child1");
+ sub11 = stats_fs_source_create("child11");
+ sub12 = stats_fs_source_create("child12");
+ stats_fs_source_add_subordinate(src, sub1);
+ stats_fs_source_add_subordinate(sub1, sub11);
+ stats_fs_source_add_subordinate(sub1, sub12);
+
+ n = stats_fs_source_add_values(sub11, test_all_aggr, &cont2);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_aggr_with_base(sub11, &cont2);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_all_aggr));
+
+ n = stats_fs_source_add_values(sub12, test_all_aggr, &cont);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_aggr_with_base(sub12, &cont);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_all_aggr));
+
+ KUNIT_EXPECT_EQ(test, get_total_number_values(src), 0);
+
+ n = stats_fs_source_add_values(src, test_all_aggr, NULL);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_aggr_with_base(src, NULL);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_all_aggr));
+
+ n = stats_fs_source_add_values(sub1, test_all_aggr, NULL);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ n = get_number_aggr_with_base(sub1, NULL);
+ KUNIT_EXPECT_EQ(test, n, ARR_SIZE(test_all_aggr));
+
+ // sum
+ n = stats_fs_source_get_value_by_name(src, "u64", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, ret, def_u64 * 2);
+
+ // min
+ n = stats_fs_source_get_value_by_name(src, "s32", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, ((int32_t)ret), def_val_s32);
+
+ // count_0
+ n = stats_fs_source_get_value_by_name(src, "bo", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, ret, 1ull);
+
+ // avg
+ n = stats_fs_source_get_value_by_name(src, "u8", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, (uint8_t)ret,
+ (uint8_t)((def_val2_u8 + def_val_u8) / 2));
+
+ // max
+ n = stats_fs_source_get_value_by_name(src, "s16", &ret);
+ KUNIT_EXPECT_EQ(test, n, 0);
+ KUNIT_EXPECT_EQ(test, (int16_t)ret, def_val_s16);
+
+ stats_fs_source_put(sub1);
+ stats_fs_source_put(sub11);
+ stats_fs_source_put(sub12);
+ stats_fs_source_put(src);
+}
+
+static struct kunit_case stats_fs_test_cases[] = {
+ KUNIT_CASE(test_empty_folder),
+ KUNIT_CASE(test_add_subfolder),
+ KUNIT_CASE(test_add_value),
+ KUNIT_CASE(test_add_value_in_subfolder),
+ KUNIT_CASE(test_search_value),
+ KUNIT_CASE(test_search_value_in_subfolder),
+ KUNIT_CASE(test_search_value_in_empty_folder),
+ KUNIT_CASE(test_add_aggregate),
+ KUNIT_CASE(test_add_aggregate_in_subfolder),
+ KUNIT_CASE(test_search_aggregate),
+ KUNIT_CASE(test_search_aggregate_in_subfolder),
+ KUNIT_CASE(test_search_same),
+ KUNIT_CASE(test_add_mixed),
+ KUNIT_CASE(test_search_mixed),
+ KUNIT_CASE(test_all_aggregations_agg_val_val),
+ KUNIT_CASE(test_all_aggregations_val_agg_val),
+ KUNIT_CASE(test_all_aggregations_agg_val_val_sub),
+ KUNIT_CASE(test_all_aggregations_agg_no_val_sub),
+ KUNIT_CASE(test_all_aggregations_agg_agg_val_sub),
+ {}
+};
+
+static struct kunit_suite stats_fs_test_suite = {
+ .name = "stats_fs",
+ .test_cases = stats_fs_test_cases,
+};
+
+kunit_test_suite(stats_fs_test_suite);
--
2.25.2
^ permalink raw reply related
* [PATCH v2 4/5] stats_fs fs: virtual fs to show stats to the end-user
From: Emanuele Giuseppe Esposito @ 2020-05-04 11:03 UTC (permalink / raw)
To: kvm
Cc: Emanuele Giuseppe Esposito, linux-s390, David Hildenbrand,
Cornelia Huck, Emanuele Giuseppe Esposito, linux-kernel, kvm-ppc,
linux-mips, Christian Borntraeger, Alexander Viro, linux-fsdevel,
Paolo Bonzini, Vitaly Kuznetsov, linuxppc-dev, Jim Mattson
In-Reply-To: <20200504110344.17560-1-eesposit@redhat.com>
Add virtual fs that maps stats_fs sources with directories, and values
(simple or aggregates) to files.
Every time a file is read/cleared, the fs internally invokes the stats_fs
API to get/set the requested value.
fs/stats_fs/inode.c is pretty much similar to what is done in
fs/debugfs/inode.c, with the exception that the API is only
composed by stats_fs_create_file, stats_fs_create_dir and stats_fs_remove.
Signed-off-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
---
fs/stats_fs/Makefile | 2 +-
fs/stats_fs/inode.c | 337 +++++++++++++++++++++++++++++++++++++
fs/stats_fs/internal.h | 15 ++
fs/stats_fs/stats_fs.c | 163 ++++++++++++++++++
include/linux/stats_fs.h | 15 ++
include/uapi/linux/magic.h | 1 +
tools/lib/api/fs/fs.c | 21 +++
7 files changed, 553 insertions(+), 1 deletion(-)
create mode 100644 fs/stats_fs/inode.c
diff --git a/fs/stats_fs/Makefile b/fs/stats_fs/Makefile
index 9db130fac6b6..ac12c27545f6 100644
--- a/fs/stats_fs/Makefile
+++ b/fs/stats_fs/Makefile
@@ -1,5 +1,5 @@
# SPDX-License-Identifier: GPL-2.0-only
-stats_fs-objs := stats_fs.o
+stats_fs-objs := inode.o stats_fs.o
stats_fs-tests-objs := stats_fs-tests.o
obj-$(CONFIG_STATS_FS) += stats_fs.o
diff --git a/fs/stats_fs/inode.c b/fs/stats_fs/inode.c
new file mode 100644
index 000000000000..865ee91656ba
--- /dev/null
+++ b/fs/stats_fs/inode.c
@@ -0,0 +1,337 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * inode.c - part of stats_fs, a tiny little stats_fs file system
+ *
+ * Copyright (C) 2020 Emanuele Giuseppe Esposito <eesposit@redhat.com>
+ * Copyright (C) 2020 Redhat
+ */
+#define pr_fmt(fmt) "stats_fs: " fmt
+
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/mount.h>
+#include <linux/init.h>
+#include <linux/stats_fs.h>
+#include <linux/string.h>
+#include <linux/seq_file.h>
+#include <linux/parser.h>
+#include <linux/magic.h>
+#include <linux/slab.h>
+
+#include "internal.h"
+
+#define STATS_FS_DEFAULT_MODE 0700
+
+static struct simple_fs stats_fs;
+static bool stats_fs_registered;
+
+struct stats_fs_mount_opts {
+ kuid_t uid;
+ kgid_t gid;
+ umode_t mode;
+};
+
+enum {
+ Opt_uid,
+ Opt_gid,
+ Opt_mode,
+ Opt_err
+};
+
+static const match_table_t tokens = {
+ {Opt_uid, "uid=%u"},
+ {Opt_gid, "gid=%u"},
+ {Opt_mode, "mode=%o"},
+ {Opt_err, NULL}
+};
+
+struct stats_fs_fs_info {
+ struct stats_fs_mount_opts mount_opts;
+};
+
+static int stats_fs_parse_options(char *data, struct stats_fs_mount_opts *opts)
+{
+ substring_t args[MAX_OPT_ARGS];
+ int option;
+ int token;
+ kuid_t uid;
+ kgid_t gid;
+ char *p;
+
+ opts->mode = STATS_FS_DEFAULT_MODE;
+
+ while ((p = strsep(&data, ",")) != NULL) {
+ if (!*p)
+ continue;
+
+ token = match_token(p, tokens, args);
+ switch (token) {
+ case Opt_uid:
+ if (match_int(&args[0], &option))
+ return -EINVAL;
+ uid = make_kuid(current_user_ns(), option);
+ if (!uid_valid(uid))
+ return -EINVAL;
+ opts->uid = uid;
+ break;
+ case Opt_gid:
+ if (match_int(&args[0], &option))
+ return -EINVAL;
+ gid = make_kgid(current_user_ns(), option);
+ if (!gid_valid(gid))
+ return -EINVAL;
+ opts->gid = gid;
+ break;
+ case Opt_mode:
+ if (match_octal(&args[0], &option))
+ return -EINVAL;
+ opts->mode = option & S_IALLUGO;
+ break;
+ /*
+ * We might like to report bad mount options here;
+ * but traditionally stats_fs has ignored all mount options
+ */
+ }
+ }
+
+ return 0;
+}
+
+static int stats_fs_apply_options(struct super_block *sb)
+{
+ struct stats_fs_fs_info *fsi = sb->s_fs_info;
+ struct inode *inode = d_inode(sb->s_root);
+ struct stats_fs_mount_opts *opts = &fsi->mount_opts;
+
+ inode->i_mode &= ~S_IALLUGO;
+ inode->i_mode |= opts->mode;
+
+ inode->i_uid = opts->uid;
+ inode->i_gid = opts->gid;
+
+ return 0;
+}
+
+static int stats_fs_remount(struct super_block *sb, int *flags, char *data)
+{
+ int err;
+ struct stats_fs_fs_info *fsi = sb->s_fs_info;
+
+ sync_filesystem(sb);
+ err = stats_fs_parse_options(data, &fsi->mount_opts);
+ if (err)
+ goto fail;
+
+ stats_fs_apply_options(sb);
+
+fail:
+ return err;
+}
+
+static int stats_fs_show_options(struct seq_file *m, struct dentry *root)
+{
+ struct stats_fs_fs_info *fsi = root->d_sb->s_fs_info;
+ struct stats_fs_mount_opts *opts = &fsi->mount_opts;
+
+ if (!uid_eq(opts->uid, GLOBAL_ROOT_UID))
+ seq_printf(m, ",uid=%u",
+ from_kuid_munged(&init_user_ns, opts->uid));
+ if (!gid_eq(opts->gid, GLOBAL_ROOT_GID))
+ seq_printf(m, ",gid=%u",
+ from_kgid_munged(&init_user_ns, opts->gid));
+ if (opts->mode != STATS_FS_DEFAULT_MODE)
+ seq_printf(m, ",mode=%o", opts->mode);
+
+ return 0;
+}
+
+
+static void stats_fs_free_inode(struct inode *inode)
+{
+ kfree(inode->i_private);
+ free_inode_nonrcu(inode);
+}
+
+static const struct super_operations stats_fs_super_operations = {
+ .statfs = simple_statfs,
+ .remount_fs = stats_fs_remount,
+ .show_options = stats_fs_show_options,
+ .free_inode = stats_fs_free_inode,
+};
+
+static int stats_fs_fill_super(struct super_block *sb, void *data, int silent)
+{
+ static const struct tree_descr stats_fs_files[] = {{""}};
+ struct stats_fs_fs_info *fsi;
+ int err;
+
+ fsi = kzalloc(sizeof(struct stats_fs_fs_info), GFP_KERNEL);
+ sb->s_fs_info = fsi;
+ if (!fsi) {
+ err = -ENOMEM;
+ goto fail;
+ }
+
+ err = stats_fs_parse_options(data, &fsi->mount_opts);
+ if (err)
+ goto fail;
+
+ err = simple_fill_super(sb, STATSFS_MAGIC, stats_fs_files);
+ if (err)
+ goto fail;
+
+ sb->s_op = &stats_fs_super_operations;
+
+ stats_fs_apply_options(sb);
+
+ return 0;
+
+fail:
+ kfree(fsi);
+ sb->s_fs_info = NULL;
+ return err;
+}
+
+static struct dentry *stats_fs_mount(struct file_system_type *fs_type,
+ int flags, const char *dev_name,
+ void *data)
+{
+ return mount_single(fs_type, flags, data, stats_fs_fill_super);
+}
+
+static struct file_system_type stats_fs_fs_type = {
+ .owner = THIS_MODULE,
+ .name = "statsfs",
+ .mount = stats_fs_mount,
+ .kill_sb = kill_litter_super,
+};
+MODULE_ALIAS_FS("statsfs");
+
+
+/**
+ * stats_fs_create_file - create a file in the stats_fs filesystem
+ * @val: a pointer to a stats_fs_value containing all the infos of
+ * the file to create (name, permission)
+ * @src: a pointer to a stats_fs_source containing the dentry of where
+ * to add this file
+ *
+ * This function will return a pointer to a dentry if it succeeds. This
+ * pointer must be passed to the stats_fs_remove() function when the file is
+ * to be removed (no automatic cleanup happens if your module is unloaded,
+ * you are responsible here.) If an error occurs, ERR_PTR(-ERROR) will be
+ * returned.
+ *
+ * Val and src will be also inglobated in a ststsfs_data_inode struct
+ * that will be internally stored as inode->i_private and used in the
+ * get/set attribute functions (see stats_fs_ops in stats_fs.c).
+ */
+struct dentry *stats_fs_create_file(struct stats_fs_value *val, struct stats_fs_source *src)
+{
+ struct dentry *dentry;
+ struct inode *inode;
+ struct stats_fs_data_inode *val_inode;
+
+ val_inode = kzalloc(sizeof(struct stats_fs_data_inode), GFP_KERNEL);
+ if (!val_inode) {
+ printk(KERN_ERR
+ "Kzalloc failure in stats_fs_create_files (ENOMEM)\n");
+ return ERR_PTR(-ENOMEM);
+ }
+
+ val_inode->src = src;
+ val_inode->val = val;
+
+
+ dentry = simplefs_create_file(&stats_fs, &stats_fs_fs_type,
+ val->name, stats_fs_val_get_mode(val),
+ src->source_dentry, val_inode, &inode);
+ if (IS_ERR(dentry))
+ return dentry;
+
+ inode->i_fop = &stats_fs_ops;
+
+ return simplefs_finish_dentry(dentry, inode);
+}
+/**
+ * stats_fs_create_dir - create a directory in the stats_fs filesystem
+ * @name: a pointer to a string containing the name of the directory to
+ * create.
+ * @parent: a pointer to the parent dentry for this file. This should be a
+ * directory dentry if set. If this parameter is NULL, then the
+ * directory will be created in the root of the stats_fs filesystem.
+ *
+ * This function creates a directory in stats_fs with the given name.
+ *
+ * This function will return a pointer to a dentry if it succeeds. This
+ * pointer must be passed to the stats_fs_remove() function when the file is
+ * to be removed (no automatic cleanup happens if your module is unloaded,
+ * you are responsible here.) If an error occurs, ERR_PTR(-ERROR) will be
+ * returned.
+ */
+struct dentry *stats_fs_create_dir(const char *name, struct dentry *parent)
+{
+ struct dentry *dentry;
+ struct inode *inode;
+
+ dentry = simplefs_create_dir(&stats_fs, &stats_fs_fs_type,
+ name, 0755, parent, &inode);
+ if (IS_ERR(dentry))
+ return dentry;
+
+ inode->i_op = &simple_dir_inode_operations;
+ return simplefs_finish_dentry(dentry, inode);
+}
+
+static void remove_one(struct dentry *victim)
+{
+ simple_release_fs(&stats_fs);
+}
+
+/**
+ * stats_fs_remove - recursively removes a directory
+ * @dentry: a pointer to a the dentry of the directory to be removed. If this
+ * parameter is NULL or an error value, nothing will be done.
+ *
+ * This function recursively removes a directory tree in stats_fs that
+ * was previously created with a call to another stats_fs function
+ * (like stats_fs_create_file() or variants thereof.)
+ *
+ * This function is required to be called in order for the file to be
+ * removed, no automatic cleanup of files will happen when a module is
+ * removed, you are responsible here.
+ */
+void stats_fs_remove(struct dentry *dentry)
+{
+ if (IS_ERR_OR_NULL(dentry))
+ return;
+
+ simple_pin_fs(&stats_fs, &stats_fs_fs_type);
+ simple_recursive_removal(dentry, remove_one);
+ simple_release_fs(&stats_fs);
+}
+/**
+ * stats_fs_initialized - Tells whether stats_fs has been registered
+ */
+bool stats_fs_initialized(void)
+{
+ return stats_fs_registered;
+}
+EXPORT_SYMBOL_GPL(stats_fs_initialized);
+
+static int __init stats_fs_init(void)
+{
+ int retval;
+
+ retval = sysfs_create_mount_point(kernel_kobj, "statsfs");
+ if (retval)
+ return retval;
+
+ retval = register_filesystem(&stats_fs_fs_type);
+ if (retval)
+ sysfs_remove_mount_point(kernel_kobj, "statsfs");
+ else
+ stats_fs_registered = true;
+
+ return retval;
+}
+core_initcall(stats_fs_init);
diff --git a/fs/stats_fs/internal.h b/fs/stats_fs/internal.h
index ddf262a60736..1f7bb1da6c3c 100644
--- a/fs/stats_fs/internal.h
+++ b/fs/stats_fs/internal.h
@@ -15,6 +15,21 @@ struct stats_fs_value_source {
struct list_head list_element;
};
+struct stats_fs_data_inode {
+ struct stats_fs_source *src;
+ struct stats_fs_value *val;
+};
+
+extern const struct file_operations stats_fs_ops;
+
+struct dentry *stats_fs_create_file(struct stats_fs_value *val,
+ struct stats_fs_source *src);
+
+struct dentry *stats_fs_create_dir(const char *name, struct dentry *parent);
+
+void stats_fs_remove(struct dentry *dentry);
+#define stats_fs_remove_recursive stats_fs_remove
+
int stats_fs_val_get_mode(struct stats_fs_value *val);
#endif /* _STATS_FS_INTERNAL_H_ */
diff --git a/fs/stats_fs/stats_fs.c b/fs/stats_fs/stats_fs.c
index b63de12769e2..4ac6fe1ec62e 100644
--- a/fs/stats_fs/stats_fs.c
+++ b/fs/stats_fs/stats_fs.c
@@ -17,16 +17,114 @@ struct stats_fs_aggregate_value {
uint32_t count, count_zero;
};
+static void stats_fs_source_remove_files(struct stats_fs_source *src);
+
static int is_val_signed(struct stats_fs_value *val)
{
return val->type & STATS_FS_SIGN;
}
+static int stats_fs_attr_get(void *data, u64 *val)
+{
+ int r = -EFAULT;
+ struct stats_fs_data_inode *val_inode =
+ (struct stats_fs_data_inode *)data;
+
+ r = stats_fs_source_get_value(val_inode->src, val_inode->val, val);
+ return r;
+}
+
+static int stats_fs_attr_clear(void *data, u64 val)
+{
+ int r = -EFAULT;
+ struct stats_fs_data_inode *val_inode =
+ (struct stats_fs_data_inode *)data;
+
+ if (val)
+ return -EINVAL;
+
+ r = stats_fs_source_clear(val_inode->src, val_inode->val);
+ return r;
+}
+
int stats_fs_val_get_mode(struct stats_fs_value *val)
{
return val->mode ? val->mode : 0644;
}
+static int stats_fs_attr_data_open(struct inode *inode, struct file *file)
+{
+ struct stats_fs_data_inode *val_inode;
+ char *fmt;
+
+ val_inode = (struct stats_fs_data_inode *)inode->i_private;
+
+ /* Inodes hold a pointer to the source which is not included in the
+ * refcount, so they files be opened while destroy is running, but
+ * values are removed (base_addr = NULL) before the source is destroyed.
+ */
+ if (!kref_get_unless_zero(&val_inode->src->refcount))
+ return -ENOENT;
+
+ if (is_val_signed(val_inode->val))
+ fmt = "%lld\n";
+ else
+ fmt = "%llu\n";
+
+ if (simple_attr_open(inode, file, stats_fs_attr_get,
+ stats_fs_val_get_mode(val_inode->val) & 0222 ?
+ stats_fs_attr_clear :
+ NULL,
+ fmt)) {
+ stats_fs_source_put(val_inode->src);
+ return -ENOMEM;
+ }
+ return 0;
+}
+
+static int stats_fs_attr_release(struct inode *inode, struct file *file)
+{
+ struct stats_fs_data_inode *val_inode;
+
+ val_inode = (struct stats_fs_data_inode *)inode->i_private;
+
+ simple_attr_release(inode, file);
+ stats_fs_source_put(val_inode->src);
+
+ return 0;
+}
+
+const struct file_operations stats_fs_ops = {
+ .owner = THIS_MODULE,
+ .open = stats_fs_attr_data_open,
+ .release = stats_fs_attr_release,
+ .read = simple_attr_read,
+ .write = simple_attr_write,
+ .llseek = no_llseek,
+};
+
+/* Called with rwsem held for writing */
+static void stats_fs_source_remove_files_locked(struct stats_fs_source *src)
+{
+ struct stats_fs_source *child;
+
+ if (src->source_dentry == NULL)
+ return;
+
+ list_for_each_entry (child, &src->subordinates_head, list_element)
+ stats_fs_source_remove_files(child);
+
+ stats_fs_remove_recursive(src->source_dentry);
+ src->source_dentry = NULL;
+}
+
+static void stats_fs_source_remove_files(struct stats_fs_source *src)
+{
+ down_write(&src->rwsem);
+ stats_fs_source_remove_files_locked(src);
+ up_write(&src->rwsem);
+}
+
static struct stats_fs_value *find_value(struct stats_fs_value_source *src,
struct stats_fs_value *val)
{
@@ -57,6 +155,62 @@ search_value_in_source(struct stats_fs_source *src, struct stats_fs_value *arg,
return NULL;
}
+/* Called with rwsem held for writing */
+static void stats_fs_create_files_locked(struct stats_fs_source *source)
+{
+ struct stats_fs_value_source *val_src;
+ struct stats_fs_value *val;
+
+ if (!source->source_dentry)
+ return;
+
+ list_for_each_entry (val_src, &source->values_head, list_element) {
+ if (val_src->files_created)
+ continue;
+
+ for (val = val_src->values; val->name; val++)
+ stats_fs_create_file(val, source);
+
+ val_src->files_created = true;
+ }
+}
+
+/* Called with rwsem held for writing */
+static void
+stats_fs_create_files_recursive_locked(struct stats_fs_source *source,
+ struct dentry *parent_dentry)
+{
+ struct stats_fs_source *child;
+
+ /* first check values in this folder, since it might be new */
+ if (!source->source_dentry) {
+ source->source_dentry =
+ stats_fs_create_dir(source->name, parent_dentry);
+ }
+
+ stats_fs_create_files_locked(source);
+
+ list_for_each_entry (child, &source->subordinates_head, list_element) {
+ if (child->source_dentry == NULL) {
+ /* assume that if child has a folder,
+ * also the sub-child have that.
+ */
+ down_write(&child->rwsem);
+ stats_fs_create_files_recursive_locked(
+ child, source->source_dentry);
+ up_write(&child->rwsem);
+ }
+ }
+}
+
+void stats_fs_source_register(struct stats_fs_source *source)
+{
+ down_write(&source->rwsem);
+ stats_fs_create_files_recursive_locked(source, NULL);
+ up_write(&source->rwsem);
+}
+EXPORT_SYMBOL_GPL(stats_fs_source_register);
+
/* Called with rwsem held for writing */
static struct stats_fs_value_source *create_value_source(void *base)
{
@@ -93,6 +247,9 @@ int stats_fs_source_add_values(struct stats_fs_source *source,
/* add the val_src to the source list */
list_add(&val_src->list_element, &source->values_head);
+ /* create child if it's the case */
+ stats_fs_create_files_locked(source);
+
up_write(&source->rwsem);
return 0;
@@ -106,6 +263,9 @@ void stats_fs_source_add_subordinate(struct stats_fs_source *source,
stats_fs_source_get(sub);
list_add(&sub->list_element, &source->subordinates_head);
+ if (source->source_dentry)
+ stats_fs_create_files_recursive_locked(sub,
+ source->source_dentry);
up_write(&source->rwsem);
}
@@ -122,6 +282,7 @@ stats_fs_source_remove_subordinate_locked(struct stats_fs_source *source,
list_element) {
if (src_entry == sub) {
list_del_init(&src_entry->list_element);
+ stats_fs_source_remove_files(src_entry);
stats_fs_source_put(src_entry);
return;
}
@@ -565,6 +726,8 @@ static void stats_fs_source_destroy(struct kref *kref_source)
stats_fs_source_remove_subordinate_locked(source, child);
}
+ stats_fs_source_remove_files_locked(source);
+
up_write(&source->rwsem);
kfree(source->name);
kfree(source);
diff --git a/include/linux/stats_fs.h b/include/linux/stats_fs.h
index dc2d2e11f5ea..b04c42d827cf 100644
--- a/include/linux/stats_fs.h
+++ b/include/linux/stats_fs.h
@@ -87,6 +87,18 @@ struct stats_fs_source {
*/
struct stats_fs_source *stats_fs_source_create(const char *fmt, ...);
+/**
+ * stats_fs_source_register - register a source in the stats_fs filesystem
+ * @source: a pointer to the source that will be registered
+ *
+ * Add the given folder as direct child of /sys/kernel/statsfs.
+ * It also starts to recursively search its own child and create all folders
+ * and files if they weren't already. All subsequent add_subordinate calls
+ * on the same source that is used in this function will create corresponding
+ * files and directories.
+ */
+void stats_fs_source_register(struct stats_fs_source *source);
+
/**
* stats_fs_source_add_values - adds values to the given source
* @source: a pointer to the source that will receive the values
@@ -235,6 +247,9 @@ static inline struct stats_fs_source *stats_fs_source_create(const char *fmt,
return ERR_PTR(-ENODEV);
}
+static inline void stats_fs_source_register(struct stats_fs_source *source)
+{ }
+
static inline int stats_fs_source_add_values(struct stats_fs_source *source,
struct stats_fs_value *val,
void *base_ptr)
diff --git a/include/uapi/linux/magic.h b/include/uapi/linux/magic.h
index d78064007b17..46c66ea3fc9e 100644
--- a/include/uapi/linux/magic.h
+++ b/include/uapi/linux/magic.h
@@ -10,6 +10,7 @@
#define CRAMFS_MAGIC 0x28cd3d45 /* some random number */
#define CRAMFS_MAGIC_WEND 0x453dcd28 /* magic number with the wrong endianess */
#define DEBUGFS_MAGIC 0x64626720
+#define STATSFS_MAGIC 0x73746174
#define SECURITYFS_MAGIC 0x73636673
#define SELINUX_MAGIC 0xf97cff8c
#define SMACK_MAGIC 0x43415d53 /* "SMAC" */
diff --git a/tools/lib/api/fs/fs.c b/tools/lib/api/fs/fs.c
index 027b18f7ed8c..6fe306206dfb 100644
--- a/tools/lib/api/fs/fs.c
+++ b/tools/lib/api/fs/fs.c
@@ -35,6 +35,10 @@
#define TRACEFS_MAGIC 0x74726163
#endif
+#ifndef STATSFS_MAGIC
+#define STATSFS_MAGIC 0x73746174
+#endif
+
#ifndef HUGETLBFS_MAGIC
#define HUGETLBFS_MAGIC 0x958458f6
#endif
@@ -76,6 +80,16 @@ static const char * const tracefs__known_mountpoints[] = {
0,
};
+#ifndef STATSFS_DEFAULT_PATH
+#define STATSFS_DEFAULT_PATH "/sys/kernel/statsfs"
+#endif
+
+static const char * const statsfs__known_mountpoints[] = {
+ STATSFS_DEFAULT_PATH,
+ "/statsfs",
+ 0,
+};
+
static const char * const hugetlbfs__known_mountpoints[] = {
0,
};
@@ -100,6 +114,7 @@ enum {
FS__TRACEFS = 3,
FS__HUGETLBFS = 4,
FS__BPF_FS = 5,
+ FS__STATSFS = 6,
};
#ifndef TRACEFS_MAGIC
@@ -127,6 +142,11 @@ static struct fs fs__entries[] = {
.mounts = tracefs__known_mountpoints,
.magic = TRACEFS_MAGIC,
},
+ [FS__STATSFS] = {
+ .name = "statsfs",
+ .mounts = statsfs__known_mountpoints,
+ .magic = STATSFS_MAGIC,
+ },
[FS__HUGETLBFS] = {
.name = "hugetlbfs",
.mounts = hugetlbfs__known_mountpoints,
@@ -297,6 +317,7 @@ FS(sysfs, FS__SYSFS);
FS(procfs, FS__PROCFS);
FS(debugfs, FS__DEBUGFS);
FS(tracefs, FS__TRACEFS);
+FS(statsfs, FS__STATSFS);
FS(hugetlbfs, FS__HUGETLBFS);
FS(bpf_fs, FS__BPF_FS);
--
2.25.2
^ permalink raw reply related
* [PATCH v2 5/5] kvm_main: replace debugfs with stats_fs
From: Emanuele Giuseppe Esposito @ 2020-05-04 11:03 UTC (permalink / raw)
To: kvm
Cc: Emanuele Giuseppe Esposito, linux-s390, David Hildenbrand,
Cornelia Huck, Emanuele Giuseppe Esposito, linux-kernel, kvm-ppc,
linux-mips, Christian Borntraeger, Alexander Viro, linux-fsdevel,
Paolo Bonzini, Vitaly Kuznetsov, linuxppc-dev, Jim Mattson
In-Reply-To: <20200504110344.17560-1-eesposit@redhat.com>
Use stats_fs API instead of debugfs to create sources and add values.
This also requires to change all architecture files to replace the old
debugfs_entries with stats_fs_vcpu_entries and statsfs_vm_entries.
The files/folders name and organization is kept unchanged, and a symlink
in sys/kernel/debugfs/kvm is left for backward compatibility.
Signed-off-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
---
arch/arm64/kvm/Kconfig | 1 +
arch/arm64/kvm/guest.c | 2 +-
arch/mips/kvm/Kconfig | 1 +
arch/mips/kvm/mips.c | 2 +-
arch/powerpc/kvm/Kconfig | 1 +
arch/powerpc/kvm/book3s.c | 6 +-
arch/powerpc/kvm/booke.c | 8 +-
arch/s390/kvm/Kconfig | 1 +
arch/s390/kvm/kvm-s390.c | 16 +-
arch/x86/include/asm/kvm_host.h | 2 +-
arch/x86/kvm/Kconfig | 1 +
arch/x86/kvm/Makefile | 2 +-
arch/x86/kvm/debugfs.c | 64 -------
arch/x86/kvm/stats_fs.c | 56 ++++++
arch/x86/kvm/x86.c | 6 +-
include/linux/kvm_host.h | 39 +---
virt/kvm/arm/arm.c | 2 +-
virt/kvm/kvm_main.c | 314 ++++----------------------------
18 files changed, 142 insertions(+), 382 deletions(-)
delete mode 100644 arch/x86/kvm/debugfs.c
create mode 100644 arch/x86/kvm/stats_fs.c
diff --git a/arch/arm64/kvm/Kconfig b/arch/arm64/kvm/Kconfig
index 449386d76441..8c125387b673 100644
--- a/arch/arm64/kvm/Kconfig
+++ b/arch/arm64/kvm/Kconfig
@@ -23,6 +23,7 @@ config KVM
depends on OF
# for TASKSTATS/TASK_DELAY_ACCT:
depends on NET && MULTIUSER
+ imply STATS_FS
select MMU_NOTIFIER
select PREEMPT_NOTIFIERS
select HAVE_KVM_CPU_RELAX_INTERCEPT
diff --git a/arch/arm64/kvm/guest.c b/arch/arm64/kvm/guest.c
index 8417b200bec9..235ed44e4353 100644
--- a/arch/arm64/kvm/guest.c
+++ b/arch/arm64/kvm/guest.c
@@ -29,7 +29,7 @@
#include "trace.h"
-struct kvm_stats_debugfs_item debugfs_entries[] = {
+struct stats_fs_value stats_fs_vcpu_entries[] = {
VCPU_STAT("halt_successful_poll", halt_successful_poll),
VCPU_STAT("halt_attempted_poll", halt_attempted_poll),
VCPU_STAT("halt_poll_invalid", halt_poll_invalid),
diff --git a/arch/mips/kvm/Kconfig b/arch/mips/kvm/Kconfig
index b91d145aa2d5..19d14e979e5f 100644
--- a/arch/mips/kvm/Kconfig
+++ b/arch/mips/kvm/Kconfig
@@ -19,6 +19,7 @@ config KVM
tristate "Kernel-based Virtual Machine (KVM) support"
depends on HAVE_KVM
depends on MIPS_FP_SUPPORT
+ imply STATS_FS
select EXPORT_UASM
select PREEMPT_NOTIFIERS
select KVM_GENERIC_DIRTYLOG_READ_PROTECT
diff --git a/arch/mips/kvm/mips.c b/arch/mips/kvm/mips.c
index fdf1c14d9205..a47d21f35444 100644
--- a/arch/mips/kvm/mips.c
+++ b/arch/mips/kvm/mips.c
@@ -39,7 +39,7 @@
#define VECTORSPACING 0x100 /* for EI/VI mode */
#endif
-struct kvm_stats_debugfs_item debugfs_entries[] = {
+struct stats_fs_value stats_fs_vcpu_entries[] = {
VCPU_STAT("wait", wait_exits),
VCPU_STAT("cache", cache_exits),
VCPU_STAT("signal", signal_exits),
diff --git a/arch/powerpc/kvm/Kconfig b/arch/powerpc/kvm/Kconfig
index 12885eda324e..feb5e110ebb0 100644
--- a/arch/powerpc/kvm/Kconfig
+++ b/arch/powerpc/kvm/Kconfig
@@ -19,6 +19,7 @@ if VIRTUALIZATION
config KVM
bool
+ imply STATS_FS
select PREEMPT_NOTIFIERS
select HAVE_KVM_EVENTFD
select HAVE_KVM_VCPU_ASYNC_IOCTL
diff --git a/arch/powerpc/kvm/book3s.c b/arch/powerpc/kvm/book3s.c
index 37508a356f28..76222ab148da 100644
--- a/arch/powerpc/kvm/book3s.c
+++ b/arch/powerpc/kvm/book3s.c
@@ -38,7 +38,7 @@
/* #define EXIT_DEBUG */
-struct kvm_stats_debugfs_item debugfs_entries[] = {
+struct stats_fs_value stats_fs_vcpu_entries[] = {
VCPU_STAT("exits", sum_exits),
VCPU_STAT("mmio", mmio_exits),
VCPU_STAT("sig", signal_exits),
@@ -66,6 +66,10 @@ struct kvm_stats_debugfs_item debugfs_entries[] = {
VCPU_STAT("pthru_all", pthru_all),
VCPU_STAT("pthru_host", pthru_host),
VCPU_STAT("pthru_bad_aff", pthru_bad_aff),
+ { NULL }
+};
+
+struct stats_fs_value stats_fs_vm_entries[] = {
VM_STAT("largepages_2M", num_2M_pages, .mode = 0444),
VM_STAT("largepages_1G", num_1G_pages, .mode = 0444),
{ NULL }
diff --git a/arch/powerpc/kvm/booke.c b/arch/powerpc/kvm/booke.c
index c2984cb6dfa7..b14c07786cc8 100644
--- a/arch/powerpc/kvm/booke.c
+++ b/arch/powerpc/kvm/booke.c
@@ -35,7 +35,12 @@
unsigned long kvmppc_booke_handlers;
-struct kvm_stats_debugfs_item debugfs_entries[] = {
+struct stats_fs_value stats_fs_vm_entries[] = {
+ VM_STAT("remote_tlb_flush", remote_tlb_flush),
+ { NULL }
+};
+
+struct stats_fs_value stats_fs_vcpu_entries[] = {
VCPU_STAT("mmio", mmio_exits),
VCPU_STAT("sig", signal_exits),
VCPU_STAT("itlb_r", itlb_real_miss_exits),
@@ -54,7 +59,6 @@ struct kvm_stats_debugfs_item debugfs_entries[] = {
VCPU_STAT("halt_wakeup", halt_wakeup),
VCPU_STAT("doorbell", dbell_exits),
VCPU_STAT("guest doorbell", gdbell_exits),
- VM_STAT("remote_tlb_flush", remote_tlb_flush),
{ NULL }
};
diff --git a/arch/s390/kvm/Kconfig b/arch/s390/kvm/Kconfig
index def3b60f1fe8..4e912ffcde78 100644
--- a/arch/s390/kvm/Kconfig
+++ b/arch/s390/kvm/Kconfig
@@ -20,6 +20,7 @@ config KVM
def_tristate y
prompt "Kernel-based Virtual Machine (KVM) support"
depends on HAVE_KVM
+ imply STATS_FS
select PREEMPT_NOTIFIERS
select HAVE_KVM_CPU_RELAX_INTERCEPT
select HAVE_KVM_VCPU_ASYNC_IOCTL
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index dbeb7da07f18..f2f090b78529 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -57,7 +57,16 @@
#define VCPU_IRQS_MAX_BUF (sizeof(struct kvm_s390_irq) * \
(KVM_MAX_VCPUS + LOCAL_IRQS))
-struct kvm_stats_debugfs_item debugfs_entries[] = {
+struct stats_fs_value stats_fs_vm_entries[] = {
+ VM_STAT("inject_float_mchk", inject_float_mchk),
+ VM_STAT("inject_io", inject_io),
+ VM_STAT("inject_pfault_done", inject_pfault_done),
+ VM_STAT("inject_service_signal", inject_service_signal),
+ VM_STAT("inject_virtio", inject_virtio),
+ { NULL }
+};
+
+struct stats_fs_value stats_fs_vcpu_entries[] = {
VCPU_STAT("userspace_handled", exit_userspace),
VCPU_STAT("exit_null", exit_null),
VCPU_STAT("exit_validity", exit_validity),
@@ -95,18 +104,13 @@ struct kvm_stats_debugfs_item debugfs_entries[] = {
VCPU_STAT("inject_ckc", inject_ckc),
VCPU_STAT("inject_cputm", inject_cputm),
VCPU_STAT("inject_external_call", inject_external_call),
- VM_STAT("inject_float_mchk", inject_float_mchk),
VCPU_STAT("inject_emergency_signal", inject_emergency_signal),
- VM_STAT("inject_io", inject_io),
VCPU_STAT("inject_mchk", inject_mchk),
- VM_STAT("inject_pfault_done", inject_pfault_done),
VCPU_STAT("inject_program", inject_program),
VCPU_STAT("inject_restart", inject_restart),
- VM_STAT("inject_service_signal", inject_service_signal),
VCPU_STAT("inject_set_prefix", inject_set_prefix),
VCPU_STAT("inject_stop_signal", inject_stop_signal),
VCPU_STAT("inject_pfault_init", inject_pfault_init),
- VM_STAT("inject_virtio", inject_virtio),
VCPU_STAT("instruction_epsw", instruction_epsw),
VCPU_STAT("instruction_gs", instruction_gs),
VCPU_STAT("instruction_io_other", instruction_io_other),
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 42a2d0d3984a..6a04f590963f 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -35,7 +35,7 @@
#include <asm/kvm_vcpu_regs.h>
#include <asm/hyperv-tlfs.h>
-#define __KVM_HAVE_ARCH_VCPU_DEBUGFS
+#define __KVM_HAVE_ARCH_VCPU_STATS_FS
#define KVM_MAX_VCPUS 288
#define KVM_SOFT_MAX_VCPUS 240
diff --git a/arch/x86/kvm/Kconfig b/arch/x86/kvm/Kconfig
index d8154e0684b6..834b8f4790a7 100644
--- a/arch/x86/kvm/Kconfig
+++ b/arch/x86/kvm/Kconfig
@@ -25,6 +25,7 @@ config KVM
# for TASKSTATS/TASK_DELAY_ACCT:
depends on NET && MULTIUSER
depends on X86_LOCAL_APIC
+ imply STATS_FS
select PREEMPT_NOTIFIERS
select MMU_NOTIFIER
select HAVE_KVM_IRQCHIP
diff --git a/arch/x86/kvm/Makefile b/arch/x86/kvm/Makefile
index a789759b7261..18285a382eba 100644
--- a/arch/x86/kvm/Makefile
+++ b/arch/x86/kvm/Makefile
@@ -11,7 +11,7 @@ kvm-$(CONFIG_KVM_ASYNC_PF) += $(KVM)/async_pf.o
kvm-y += x86.o emulate.o i8259.o irq.o lapic.o \
i8254.o ioapic.o irq_comm.o cpuid.o pmu.o mtrr.o \
- hyperv.o debugfs.o mmu/mmu.o mmu/page_track.o
+ hyperv.o stats_fs.o mmu/mmu.o mmu/page_track.o
kvm-intel-y += vmx/vmx.o vmx/vmenter.o vmx/pmu_intel.o vmx/vmcs12.o vmx/evmcs.o vmx/nested.o
kvm-amd-y += svm/svm.o svm/vmenter.o svm/pmu.o svm/nested.o svm/avic.o svm/sev.o
diff --git a/arch/x86/kvm/debugfs.c b/arch/x86/kvm/debugfs.c
deleted file mode 100644
index 018aebce33ff..000000000000
--- a/arch/x86/kvm/debugfs.c
+++ /dev/null
@@ -1,64 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0-only
-/*
- * Kernel-based Virtual Machine driver for Linux
- *
- * Copyright 2016 Red Hat, Inc. and/or its affiliates.
- */
-#include <linux/kvm_host.h>
-#include <linux/debugfs.h>
-#include "lapic.h"
-
-static int vcpu_get_timer_advance_ns(void *data, u64 *val)
-{
- struct kvm_vcpu *vcpu = (struct kvm_vcpu *) data;
- *val = vcpu->arch.apic->lapic_timer.timer_advance_ns;
- return 0;
-}
-
-DEFINE_SIMPLE_ATTRIBUTE(vcpu_timer_advance_ns_fops, vcpu_get_timer_advance_ns, NULL, "%llu\n");
-
-static int vcpu_get_tsc_offset(void *data, u64 *val)
-{
- struct kvm_vcpu *vcpu = (struct kvm_vcpu *) data;
- *val = vcpu->arch.tsc_offset;
- return 0;
-}
-
-DEFINE_SIMPLE_ATTRIBUTE(vcpu_tsc_offset_fops, vcpu_get_tsc_offset, NULL, "%lld\n");
-
-static int vcpu_get_tsc_scaling_ratio(void *data, u64 *val)
-{
- struct kvm_vcpu *vcpu = (struct kvm_vcpu *) data;
- *val = vcpu->arch.tsc_scaling_ratio;
- return 0;
-}
-
-DEFINE_SIMPLE_ATTRIBUTE(vcpu_tsc_scaling_fops, vcpu_get_tsc_scaling_ratio, NULL, "%llu\n");
-
-static int vcpu_get_tsc_scaling_frac_bits(void *data, u64 *val)
-{
- *val = kvm_tsc_scaling_ratio_frac_bits;
- return 0;
-}
-
-DEFINE_SIMPLE_ATTRIBUTE(vcpu_tsc_scaling_frac_fops, vcpu_get_tsc_scaling_frac_bits, NULL, "%llu\n");
-
-void kvm_arch_create_vcpu_debugfs(struct kvm_vcpu *vcpu)
-{
- debugfs_create_file("tsc-offset", 0444, vcpu->debugfs_dentry, vcpu,
- &vcpu_tsc_offset_fops);
-
- if (lapic_in_kernel(vcpu))
- debugfs_create_file("lapic_timer_advance_ns", 0444,
- vcpu->debugfs_dentry, vcpu,
- &vcpu_timer_advance_ns_fops);
-
- if (kvm_has_tsc_control) {
- debugfs_create_file("tsc-scaling-ratio", 0444,
- vcpu->debugfs_dentry, vcpu,
- &vcpu_tsc_scaling_fops);
- debugfs_create_file("tsc-scaling-ratio-frac-bits", 0444,
- vcpu->debugfs_dentry, vcpu,
- &vcpu_tsc_scaling_frac_fops);
- }
-}
diff --git a/arch/x86/kvm/stats_fs.c b/arch/x86/kvm/stats_fs.c
new file mode 100644
index 000000000000..f2ac6ed8b01b
--- /dev/null
+++ b/arch/x86/kvm/stats_fs.c
@@ -0,0 +1,56 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Kernel-based Virtual Machine driver for Linux
+ *
+ * Copyright 2016 Red Hat, Inc. and/or its affiliates.
+ */
+#include <linux/kvm_host.h>
+#include <linux/stats_fs.h>
+#include "lapic.h"
+
+#define VCPU_ARCH_STATS_FS(n, s, x, ...) \
+ { n, offsetof(struct s, x), .aggr_kind = STATS_FS_SUM, \
+ ##__VA_ARGS__ }
+
+struct stats_fs_value stats_fs_vcpu_tsc_offset[] = {
+ VCPU_ARCH_STATS_FS("tsc-offset", kvm_vcpu_arch, tsc_offset,
+ .type = STATS_FS_S64, .mode = 0444),
+ { NULL }
+};
+
+struct stats_fs_value stats_fs_vcpu_arch_lapic_timer[] = {
+ VCPU_ARCH_STATS_FS("lapic_timer_advance_ns", kvm_timer, timer_advance_ns,
+ .type = STATS_FS_U64, .mode = 0444),
+ { NULL }
+};
+
+struct stats_fs_value stats_fs_vcpu_arch_tsc_ratio[] = {
+ VCPU_ARCH_STATS_FS("tsc-scaling-ratio", kvm_vcpu_arch, tsc_scaling_ratio,
+ .type = STATS_FS_U64, .mode = 0444),
+ { NULL }
+};
+
+struct stats_fs_value stats_fs_vcpu_arch_tsc_frac[] = {
+ { "tsc-scaling-ratio-frac-bits", 0, .type = STATS_FS_U64, .mode = 0444 },
+ { NULL } /* base is &kvm_tsc_scaling_ratio_frac_bits */
+};
+
+void kvm_arch_create_vcpu_stats_fs(struct kvm_vcpu *vcpu)
+{
+ stats_fs_source_add_values(vcpu->stats_fs_src, stats_fs_vcpu_tsc_offset,
+ &vcpu->arch);
+
+ if (lapic_in_kernel(vcpu))
+ stats_fs_source_add_values(vcpu->stats_fs_src,
+ stats_fs_vcpu_arch_lapic_timer,
+ &vcpu->arch.apic->lapic_timer);
+
+ if (kvm_has_tsc_control) {
+ stats_fs_source_add_values(vcpu->stats_fs_src,
+ stats_fs_vcpu_arch_tsc_ratio,
+ &vcpu->arch);
+ stats_fs_source_add_values(vcpu->stats_fs_src,
+ stats_fs_vcpu_arch_tsc_frac,
+ &kvm_tsc_scaling_ratio_frac_bits);
+ }
+}
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 35723dafedeb..01fed7ac2e49 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -190,7 +190,7 @@ static u64 __read_mostly host_xss;
u64 __read_mostly supported_xss;
EXPORT_SYMBOL_GPL(supported_xss);
-struct kvm_stats_debugfs_item debugfs_entries[] = {
+struct stats_fs_value stats_fs_vcpu_entries[] = {
VCPU_STAT("pf_fixed", pf_fixed),
VCPU_STAT("pf_guest", pf_guest),
VCPU_STAT("tlb_flush", tlb_flush),
@@ -217,6 +217,10 @@ struct kvm_stats_debugfs_item debugfs_entries[] = {
VCPU_STAT("nmi_injections", nmi_injections),
VCPU_STAT("req_event", req_event),
VCPU_STAT("l1d_flush", l1d_flush),
+ { NULL }
+};
+
+struct stats_fs_value stats_fs_vm_entries[] = {
VM_STAT("mmu_shadow_zapped", mmu_shadow_zapped),
VM_STAT("mmu_pte_write", mmu_pte_write),
VM_STAT("mmu_pte_updated", mmu_pte_updated),
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 3845f857ef7b..1282799aa46e 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -27,6 +27,7 @@
#include <linux/refcount.h>
#include <linux/nospec.h>
#include <asm/signal.h>
+#include <linux/stats_fs.h>
#include <linux/kvm.h>
#include <linux/kvm_para.h>
@@ -318,7 +319,7 @@ struct kvm_vcpu {
bool preempted;
bool ready;
struct kvm_vcpu_arch arch;
- struct dentry *debugfs_dentry;
+ struct stats_fs_source *stats_fs_src;
};
static inline int kvm_vcpu_exiting_guest_mode(struct kvm_vcpu *vcpu)
@@ -498,8 +499,7 @@ struct kvm {
long tlbs_dirty;
struct list_head devices;
u64 manual_dirty_log_protect;
- struct dentry *debugfs_dentry;
- struct kvm_stat_data **debugfs_stat_data;
+ struct stats_fs_source *stats_fs_src;
struct srcu_struct srcu;
struct srcu_struct irq_srcu;
pid_t userspace_pid;
@@ -880,8 +880,8 @@ int kvm_arch_vcpu_create(struct kvm_vcpu *vcpu);
void kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu);
void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu);
-#ifdef __KVM_HAVE_ARCH_VCPU_DEBUGFS
-void kvm_arch_create_vcpu_debugfs(struct kvm_vcpu *vcpu);
+#ifdef __KVM_HAVE_ARCH_VCPU_STATS_FS
+void kvm_arch_create_vcpu_stats_fs(struct kvm_vcpu *vcpu);
#endif
int kvm_arch_hardware_enable(void);
@@ -1110,33 +1110,14 @@ static inline bool kvm_is_error_gpa(struct kvm *kvm, gpa_t gpa)
return kvm_is_error_hva(hva);
}
-enum kvm_stat_kind {
- KVM_STAT_VM,
- KVM_STAT_VCPU,
-};
-
-struct kvm_stat_data {
- struct kvm *kvm;
- struct kvm_stats_debugfs_item *dbgfs_item;
-};
-
-struct kvm_stats_debugfs_item {
- const char *name;
- int offset;
- enum kvm_stat_kind kind;
- int mode;
-};
-
-#define KVM_DBGFS_GET_MODE(dbgfs_item) \
- ((dbgfs_item)->mode ? (dbgfs_item)->mode : 0644)
-
#define VM_STAT(n, x, ...) \
- { n, offsetof(struct kvm, stat.x), KVM_STAT_VM, ## __VA_ARGS__ }
+ { n, offsetof(struct kvm, stat.x), STATS_FS_U64, STATS_FS_SUM, ## __VA_ARGS__ }
#define VCPU_STAT(n, x, ...) \
- { n, offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU, ## __VA_ARGS__ }
+ { n, offsetof(struct kvm_vcpu, stat.x), STATS_FS_U64, STATS_FS_SUM, ## __VA_ARGS__ }
-extern struct kvm_stats_debugfs_item debugfs_entries[];
-extern struct dentry *kvm_debugfs_dir;
+extern struct stats_fs_value stats_fs_vcpu_entries[];
+extern struct stats_fs_value stats_fs_vm_entries[];
+extern struct stats_fs_source *kvm_stats_fs_dir;
#if defined(CONFIG_MMU_NOTIFIER) && defined(KVM_ARCH_WANT_MMU_NOTIFIER)
static inline int mmu_notifier_retry(struct kvm *kvm, unsigned long mmu_seq)
diff --git a/virt/kvm/arm/arm.c b/virt/kvm/arm/arm.c
index 48d0ec44ad77..4171f92fa473 100644
--- a/virt/kvm/arm/arm.c
+++ b/virt/kvm/arm/arm.c
@@ -140,7 +140,7 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
return ret;
}
-int kvm_arch_create_vcpu_debugfs(struct kvm_vcpu *vcpu)
+int kvm_arch_create_vcpu_stats_fs(struct kvm_vcpu *vcpu)
{
return 0;
}
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 74bdb7bf3295..093150125bc2 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -25,6 +25,7 @@
#include <linux/vmalloc.h>
#include <linux/reboot.h>
#include <linux/debugfs.h>
+#include <linux/stats_fs.h>
#include <linux/highmem.h>
#include <linux/file.h>
#include <linux/syscore_ops.h>
@@ -109,11 +110,8 @@ static struct kmem_cache *kvm_vcpu_cache;
static __read_mostly struct preempt_ops kvm_preempt_ops;
static DEFINE_PER_CPU(struct kvm_vcpu *, kvm_running_vcpu);
-struct dentry *kvm_debugfs_dir;
-EXPORT_SYMBOL_GPL(kvm_debugfs_dir);
-
-static int kvm_debugfs_num_entries;
-static const struct file_operations stat_fops_per_vm;
+struct stats_fs_source *kvm_stats_fs_dir;
+EXPORT_SYMBOL_GPL(kvm_stats_fs_dir);
static long kvm_vcpu_ioctl(struct file *file, unsigned int ioctl,
unsigned long arg);
@@ -356,6 +354,8 @@ static void kvm_vcpu_init(struct kvm_vcpu *vcpu, struct kvm *kvm, unsigned id)
void kvm_vcpu_destroy(struct kvm_vcpu *vcpu)
{
+ stats_fs_source_revoke(vcpu->stats_fs_src);
+ stats_fs_source_put(vcpu->stats_fs_src);
kvm_arch_vcpu_destroy(vcpu);
/*
@@ -601,52 +601,27 @@ static void kvm_free_memslots(struct kvm *kvm, struct kvm_memslots *slots)
kvfree(slots);
}
-static void kvm_destroy_vm_debugfs(struct kvm *kvm)
+static void kvm_destroy_vm_stats_fs(struct kvm *kvm)
{
- int i;
-
- if (!kvm->debugfs_dentry)
- return;
-
- debugfs_remove_recursive(kvm->debugfs_dentry);
-
- if (kvm->debugfs_stat_data) {
- for (i = 0; i < kvm_debugfs_num_entries; i++)
- kfree(kvm->debugfs_stat_data[i]);
- kfree(kvm->debugfs_stat_data);
- }
+ stats_fs_source_remove_subordinate(kvm_stats_fs_dir, kvm->stats_fs_src);
+ stats_fs_source_revoke(kvm->stats_fs_src);
+ stats_fs_source_put(kvm->stats_fs_src);
}
-static int kvm_create_vm_debugfs(struct kvm *kvm, int fd)
+static int kvm_create_vm_stats_fs(struct kvm *kvm, int fd)
{
char dir_name[ITOA_MAX_LEN * 2];
- struct kvm_stat_data *stat_data;
- struct kvm_stats_debugfs_item *p;
- if (!debugfs_initialized())
+ if (!stats_fs_initialized())
return 0;
snprintf(dir_name, sizeof(dir_name), "%d-%d", task_pid_nr(current), fd);
- kvm->debugfs_dentry = debugfs_create_dir(dir_name, kvm_debugfs_dir);
+ kvm->stats_fs_src = stats_fs_source_create(dir_name);
+ stats_fs_source_add_subordinate(kvm_stats_fs_dir, kvm->stats_fs_src);
- kvm->debugfs_stat_data = kcalloc(kvm_debugfs_num_entries,
- sizeof(*kvm->debugfs_stat_data),
- GFP_KERNEL_ACCOUNT);
- if (!kvm->debugfs_stat_data)
- return -ENOMEM;
+ stats_fs_source_add_values(kvm->stats_fs_src, stats_fs_vm_entries, kvm);
- for (p = debugfs_entries; p->name; p++) {
- stat_data = kzalloc(sizeof(*stat_data), GFP_KERNEL_ACCOUNT);
- if (!stat_data)
- return -ENOMEM;
-
- stat_data->kvm = kvm;
- stat_data->dbgfs_item = p;
- kvm->debugfs_stat_data[p - debugfs_entries] = stat_data;
- debugfs_create_file(p->name, KVM_DBGFS_GET_MODE(p),
- kvm->debugfs_dentry, stat_data,
- &stat_fops_per_vm);
- }
+ stats_fs_source_add_values(kvm->stats_fs_src, stats_fs_vcpu_entries, NULL);
return 0;
}
@@ -783,7 +758,7 @@ static void kvm_destroy_vm(struct kvm *kvm)
struct mm_struct *mm = kvm->mm;
kvm_uevent_notify_change(KVM_EVENT_DESTROY_VM, kvm);
- kvm_destroy_vm_debugfs(kvm);
+ kvm_destroy_vm_stats_fs(kvm);
kvm_arch_sync_events(kvm);
mutex_lock(&kvm_lock);
list_del(&kvm->vm_list);
@@ -2946,7 +2921,6 @@ static int kvm_vcpu_release(struct inode *inode, struct file *filp)
{
struct kvm_vcpu *vcpu = filp->private_data;
- debugfs_remove_recursive(vcpu->debugfs_dentry);
kvm_put_kvm(vcpu->kvm);
return 0;
}
@@ -2970,19 +2944,22 @@ static int create_vcpu_fd(struct kvm_vcpu *vcpu)
return anon_inode_getfd(name, &kvm_vcpu_fops, vcpu, O_RDWR | O_CLOEXEC);
}
-static void kvm_create_vcpu_debugfs(struct kvm_vcpu *vcpu)
+static void kvm_create_vcpu_stats_fs(struct kvm_vcpu *vcpu)
{
-#ifdef __KVM_HAVE_ARCH_VCPU_DEBUGFS
char dir_name[ITOA_MAX_LEN * 2];
- if (!debugfs_initialized())
+ if (!stats_fs_initialized())
return;
snprintf(dir_name, sizeof(dir_name), "vcpu%d", vcpu->vcpu_id);
- vcpu->debugfs_dentry = debugfs_create_dir(dir_name,
- vcpu->kvm->debugfs_dentry);
- kvm_arch_create_vcpu_debugfs(vcpu);
+ vcpu->stats_fs_src = stats_fs_source_create(dir_name);
+ stats_fs_source_add_subordinate(vcpu->kvm->stats_fs_src, vcpu->stats_fs_src);
+
+ stats_fs_source_add_values(vcpu->stats_fs_src, stats_fs_vcpu_entries, vcpu);
+
+#ifdef __KVM_HAVE_ARCH_VCPU_STATS_FS
+ kvm_arch_create_vcpu_stats_fs(vcpu);
#endif
}
@@ -3031,8 +3008,6 @@ static int kvm_vm_ioctl_create_vcpu(struct kvm *kvm, u32 id)
if (r)
goto vcpu_free_run_page;
- kvm_create_vcpu_debugfs(vcpu);
-
mutex_lock(&kvm->lock);
if (kvm_get_vcpu_by_id(kvm, id)) {
r = -EEXIST;
@@ -3061,11 +3036,11 @@ static int kvm_vm_ioctl_create_vcpu(struct kvm *kvm, u32 id)
mutex_unlock(&kvm->lock);
kvm_arch_vcpu_postcreate(vcpu);
+ kvm_create_vcpu_stats_fs(vcpu);
return r;
unlock_vcpu_destroy:
mutex_unlock(&kvm->lock);
- debugfs_remove_recursive(vcpu->debugfs_dentry);
kvm_arch_vcpu_destroy(vcpu);
vcpu_free_run_page:
free_page((unsigned long)vcpu->run);
@@ -3839,7 +3814,7 @@ static int kvm_dev_ioctl_create_vm(unsigned long type)
* cases it will be called by the final fput(file) and will take
* care of doing kvm_put_kvm(kvm).
*/
- if (kvm_create_vm_debugfs(kvm, r) < 0) {
+ if (kvm_create_vm_stats_fs(kvm, r) < 0) {
put_unused_fd(r);
fput(file);
return -ENOMEM;
@@ -4295,214 +4270,6 @@ struct kvm_io_device *kvm_io_bus_get_dev(struct kvm *kvm, enum kvm_bus bus_idx,
}
EXPORT_SYMBOL_GPL(kvm_io_bus_get_dev);
-static int kvm_debugfs_open(struct inode *inode, struct file *file,
- int (*get)(void *, u64 *), int (*set)(void *, u64),
- const char *fmt)
-{
- struct kvm_stat_data *stat_data = (struct kvm_stat_data *)
- inode->i_private;
-
- /* The debugfs files are a reference to the kvm struct which
- * is still valid when kvm_destroy_vm is called.
- * To avoid the race between open and the removal of the debugfs
- * directory we test against the users count.
- */
- if (!refcount_inc_not_zero(&stat_data->kvm->users_count))
- return -ENOENT;
-
- if (simple_attr_open(inode, file, get,
- KVM_DBGFS_GET_MODE(stat_data->dbgfs_item) & 0222
- ? set : NULL,
- fmt)) {
- kvm_put_kvm(stat_data->kvm);
- return -ENOMEM;
- }
-
- return 0;
-}
-
-static int kvm_debugfs_release(struct inode *inode, struct file *file)
-{
- struct kvm_stat_data *stat_data = (struct kvm_stat_data *)
- inode->i_private;
-
- simple_attr_release(inode, file);
- kvm_put_kvm(stat_data->kvm);
-
- return 0;
-}
-
-static int kvm_get_stat_per_vm(struct kvm *kvm, size_t offset, u64 *val)
-{
- *val = *(ulong *)((void *)kvm + offset);
-
- return 0;
-}
-
-static int kvm_clear_stat_per_vm(struct kvm *kvm, size_t offset)
-{
- *(ulong *)((void *)kvm + offset) = 0;
-
- return 0;
-}
-
-static int kvm_get_stat_per_vcpu(struct kvm *kvm, size_t offset, u64 *val)
-{
- int i;
- struct kvm_vcpu *vcpu;
-
- *val = 0;
-
- kvm_for_each_vcpu(i, vcpu, kvm)
- *val += *(u64 *)((void *)vcpu + offset);
-
- return 0;
-}
-
-static int kvm_clear_stat_per_vcpu(struct kvm *kvm, size_t offset)
-{
- int i;
- struct kvm_vcpu *vcpu;
-
- kvm_for_each_vcpu(i, vcpu, kvm)
- *(u64 *)((void *)vcpu + offset) = 0;
-
- return 0;
-}
-
-static int kvm_stat_data_get(void *data, u64 *val)
-{
- int r = -EFAULT;
- struct kvm_stat_data *stat_data = (struct kvm_stat_data *)data;
-
- switch (stat_data->dbgfs_item->kind) {
- case KVM_STAT_VM:
- r = kvm_get_stat_per_vm(stat_data->kvm,
- stat_data->dbgfs_item->offset, val);
- break;
- case KVM_STAT_VCPU:
- r = kvm_get_stat_per_vcpu(stat_data->kvm,
- stat_data->dbgfs_item->offset, val);
- break;
- }
-
- return r;
-}
-
-static int kvm_stat_data_clear(void *data, u64 val)
-{
- int r = -EFAULT;
- struct kvm_stat_data *stat_data = (struct kvm_stat_data *)data;
-
- if (val)
- return -EINVAL;
-
- switch (stat_data->dbgfs_item->kind) {
- case KVM_STAT_VM:
- r = kvm_clear_stat_per_vm(stat_data->kvm,
- stat_data->dbgfs_item->offset);
- break;
- case KVM_STAT_VCPU:
- r = kvm_clear_stat_per_vcpu(stat_data->kvm,
- stat_data->dbgfs_item->offset);
- break;
- }
-
- return r;
-}
-
-static int kvm_stat_data_open(struct inode *inode, struct file *file)
-{
- __simple_attr_check_format("%llu\n", 0ull);
- return kvm_debugfs_open(inode, file, kvm_stat_data_get,
- kvm_stat_data_clear, "%llu\n");
-}
-
-static const struct file_operations stat_fops_per_vm = {
- .owner = THIS_MODULE,
- .open = kvm_stat_data_open,
- .release = kvm_debugfs_release,
- .read = simple_attr_read,
- .write = simple_attr_write,
- .llseek = no_llseek,
-};
-
-static int vm_stat_get(void *_offset, u64 *val)
-{
- unsigned offset = (long)_offset;
- struct kvm *kvm;
- u64 tmp_val;
-
- *val = 0;
- mutex_lock(&kvm_lock);
- list_for_each_entry(kvm, &vm_list, vm_list) {
- kvm_get_stat_per_vm(kvm, offset, &tmp_val);
- *val += tmp_val;
- }
- mutex_unlock(&kvm_lock);
- return 0;
-}
-
-static int vm_stat_clear(void *_offset, u64 val)
-{
- unsigned offset = (long)_offset;
- struct kvm *kvm;
-
- if (val)
- return -EINVAL;
-
- mutex_lock(&kvm_lock);
- list_for_each_entry(kvm, &vm_list, vm_list) {
- kvm_clear_stat_per_vm(kvm, offset);
- }
- mutex_unlock(&kvm_lock);
-
- return 0;
-}
-
-DEFINE_SIMPLE_ATTRIBUTE(vm_stat_fops, vm_stat_get, vm_stat_clear, "%llu\n");
-
-static int vcpu_stat_get(void *_offset, u64 *val)
-{
- unsigned offset = (long)_offset;
- struct kvm *kvm;
- u64 tmp_val;
-
- *val = 0;
- mutex_lock(&kvm_lock);
- list_for_each_entry(kvm, &vm_list, vm_list) {
- kvm_get_stat_per_vcpu(kvm, offset, &tmp_val);
- *val += tmp_val;
- }
- mutex_unlock(&kvm_lock);
- return 0;
-}
-
-static int vcpu_stat_clear(void *_offset, u64 val)
-{
- unsigned offset = (long)_offset;
- struct kvm *kvm;
-
- if (val)
- return -EINVAL;
-
- mutex_lock(&kvm_lock);
- list_for_each_entry(kvm, &vm_list, vm_list) {
- kvm_clear_stat_per_vcpu(kvm, offset);
- }
- mutex_unlock(&kvm_lock);
-
- return 0;
-}
-
-DEFINE_SIMPLE_ATTRIBUTE(vcpu_stat_fops, vcpu_stat_get, vcpu_stat_clear,
- "%llu\n");
-
-static const struct file_operations *stat_fops[] = {
- [KVM_STAT_VCPU] = &vcpu_stat_fops,
- [KVM_STAT_VM] = &vm_stat_fops,
-};
-
static void kvm_uevent_notify_change(unsigned int type, struct kvm *kvm)
{
struct kobj_uevent_env *env;
@@ -4537,34 +4304,32 @@ static void kvm_uevent_notify_change(unsigned int type, struct kvm *kvm)
}
add_uevent_var(env, "PID=%d", kvm->userspace_pid);
- if (!IS_ERR_OR_NULL(kvm->debugfs_dentry)) {
+ if (!IS_ERR_OR_NULL(kvm->stats_fs_src->source_dentry)) {
char *tmp, *p = kmalloc(PATH_MAX, GFP_KERNEL_ACCOUNT);
if (p) {
- tmp = dentry_path_raw(kvm->debugfs_dentry, p, PATH_MAX);
+ tmp = dentry_path_raw(kvm->stats_fs_src->source_dentry, p, PATH_MAX);
if (!IS_ERR(tmp))
add_uevent_var(env, "STATS_PATH=%s", tmp);
kfree(p);
}
}
+
/* no need for checks, since we are adding at most only 5 keys */
env->envp[env->envp_idx++] = NULL;
kobject_uevent_env(&kvm_dev.this_device->kobj, KOBJ_CHANGE, env->envp);
kfree(env);
}
-static void kvm_init_debug(void)
+static void kvm_init_stats_fs(void)
{
- struct kvm_stats_debugfs_item *p;
+ kvm_stats_fs_dir = stats_fs_source_create("kvm");
+ /* symlink to debugfs */
+ debugfs_create_symlink("kvm", NULL, "/sys/kernel/stats_fs/kvm");
+ stats_fs_source_register(kvm_stats_fs_dir);
- kvm_debugfs_dir = debugfs_create_dir("kvm", NULL);
-
- kvm_debugfs_num_entries = 0;
- for (p = debugfs_entries; p->name; ++p, kvm_debugfs_num_entries++) {
- debugfs_create_file(p->name, KVM_DBGFS_GET_MODE(p),
- kvm_debugfs_dir, (void *)(long)p->offset,
- stat_fops[p->kind]);
- }
+ stats_fs_source_add_values(kvm_stats_fs_dir, stats_fs_vcpu_entries, NULL);
+ stats_fs_source_add_values(kvm_stats_fs_dir, stats_fs_vm_entries, NULL);
}
static int kvm_suspend(void)
@@ -4738,7 +4503,7 @@ int kvm_init(void *opaque, unsigned vcpu_size, unsigned vcpu_align,
kvm_preempt_ops.sched_in = kvm_sched_in;
kvm_preempt_ops.sched_out = kvm_sched_out;
- kvm_init_debug();
+ kvm_init_stats_fs();
r = kvm_vfio_ops_init();
WARN_ON(r);
@@ -4767,7 +4532,8 @@ EXPORT_SYMBOL_GPL(kvm_init);
void kvm_exit(void)
{
- debugfs_remove_recursive(kvm_debugfs_dir);
+ stats_fs_source_revoke(kvm_stats_fs_dir);
+ stats_fs_source_put(kvm_stats_fs_dir);
misc_deregister(&kvm_dev);
kmem_cache_destroy(kvm_vcpu_cache);
kvm_async_pf_deinit();
--
2.25.2
^ permalink raw reply related
* Re: [PATCH V2 11/11] drm: Remove drm specific kmap_atomic code
From: Daniel Vetter @ 2020-05-04 11:18 UTC (permalink / raw)
To: Ira Weiny
Cc: Peter Zijlstra, Dave Hansen, dri-devel, linux-mips,
James E.J. Bottomley, Max Filippov, Huang Rui, Paul Mackerras,
H. Peter Anvin, sparclinux, Christoph Hellwig, Helge Deller,
X86 ML, linux-csky, Ingo Molnar, arcml, linux-xtensa,
Dan Williams, Borislav Petkov, Andy Lutomirski, Thomas Gleixner,
Linux ARM, Chris Zankel, Thomas Bogendoerfer, linux-parisc,
Linux Kernel Mailing List, David S. Miller, Andrew Morton,
linuxppc-dev, Christian Koenig
In-Reply-To: <20200504010912.982044-12-ira.weiny@intel.com>
On Mon, May 4, 2020 at 3:09 AM <ira.weiny@intel.com> wrote:
>
> From: Ira Weiny <ira.weiny@intel.com>
>
> kmap_atomic_prot() is now exported by all architectures. Use this
> function rather than open coding a driver specific kmap_atomic.
>
> Reviewed-by: Christian König <christian.koenig@amd.com>
> Reviewed-by: Christoph Hellwig <hch@lst.de>
> Signed-off-by: Ira Weiny <ira.weiny@intel.com>
I'm assuming this lands through some other tree or a topic branch or whatever.
Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Cheers, Daniel
> ---
> drivers/gpu/drm/ttm/ttm_bo_util.c | 56 ++--------------------------
> drivers/gpu/drm/vmwgfx/vmwgfx_blit.c | 16 ++++----
> include/drm/ttm/ttm_bo_api.h | 4 --
> 3 files changed, 12 insertions(+), 64 deletions(-)
>
> diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c
> index 52d2b71f1588..f09b096ba4fd 100644
> --- a/drivers/gpu/drm/ttm/ttm_bo_util.c
> +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c
> @@ -257,54 +257,6 @@ static int ttm_copy_io_page(void *dst, void *src, unsigned long page)
> return 0;
> }
>
> -#ifdef CONFIG_X86
> -#define __ttm_kmap_atomic_prot(__page, __prot) kmap_atomic_prot(__page, __prot)
> -#define __ttm_kunmap_atomic(__addr) kunmap_atomic(__addr)
> -#else
> -#define __ttm_kmap_atomic_prot(__page, __prot) vmap(&__page, 1, 0, __prot)
> -#define __ttm_kunmap_atomic(__addr) vunmap(__addr)
> -#endif
> -
> -
> -/**
> - * ttm_kmap_atomic_prot - Efficient kernel map of a single page with
> - * specified page protection.
> - *
> - * @page: The page to map.
> - * @prot: The page protection.
> - *
> - * This function maps a TTM page using the kmap_atomic api if available,
> - * otherwise falls back to vmap. The user must make sure that the
> - * specified page does not have an aliased mapping with a different caching
> - * policy unless the architecture explicitly allows it. Also mapping and
> - * unmapping using this api must be correctly nested. Unmapping should
> - * occur in the reverse order of mapping.
> - */
> -void *ttm_kmap_atomic_prot(struct page *page, pgprot_t prot)
> -{
> - if (pgprot_val(prot) == pgprot_val(PAGE_KERNEL))
> - return kmap_atomic(page);
> - else
> - return __ttm_kmap_atomic_prot(page, prot);
> -}
> -EXPORT_SYMBOL(ttm_kmap_atomic_prot);
> -
> -/**
> - * ttm_kunmap_atomic_prot - Unmap a page that was mapped using
> - * ttm_kmap_atomic_prot.
> - *
> - * @addr: The virtual address from the map.
> - * @prot: The page protection.
> - */
> -void ttm_kunmap_atomic_prot(void *addr, pgprot_t prot)
> -{
> - if (pgprot_val(prot) == pgprot_val(PAGE_KERNEL))
> - kunmap_atomic(addr);
> - else
> - __ttm_kunmap_atomic(addr);
> -}
> -EXPORT_SYMBOL(ttm_kunmap_atomic_prot);
> -
> static int ttm_copy_io_ttm_page(struct ttm_tt *ttm, void *src,
> unsigned long page,
> pgprot_t prot)
> @@ -316,13 +268,13 @@ static int ttm_copy_io_ttm_page(struct ttm_tt *ttm, void *src,
> return -ENOMEM;
>
> src = (void *)((unsigned long)src + (page << PAGE_SHIFT));
> - dst = ttm_kmap_atomic_prot(d, prot);
> + dst = kmap_atomic_prot(d, prot);
> if (!dst)
> return -ENOMEM;
>
> memcpy_fromio(dst, src, PAGE_SIZE);
>
> - ttm_kunmap_atomic_prot(dst, prot);
> + kunmap_atomic(dst);
>
> return 0;
> }
> @@ -338,13 +290,13 @@ static int ttm_copy_ttm_io_page(struct ttm_tt *ttm, void *dst,
> return -ENOMEM;
>
> dst = (void *)((unsigned long)dst + (page << PAGE_SHIFT));
> - src = ttm_kmap_atomic_prot(s, prot);
> + src = kmap_atomic_prot(s, prot);
> if (!src)
> return -ENOMEM;
>
> memcpy_toio(dst, src, PAGE_SIZE);
>
> - ttm_kunmap_atomic_prot(src, prot);
> + kunmap_atomic(src);
>
> return 0;
> }
> diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_blit.c b/drivers/gpu/drm/vmwgfx/vmwgfx_blit.c
> index bb46ca0c458f..94d456a1d1a9 100644
> --- a/drivers/gpu/drm/vmwgfx/vmwgfx_blit.c
> +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_blit.c
> @@ -374,12 +374,12 @@ static int vmw_bo_cpu_blit_line(struct vmw_bo_blit_line_data *d,
> copy_size = min_t(u32, copy_size, PAGE_SIZE - src_page_offset);
>
> if (unmap_src) {
> - ttm_kunmap_atomic_prot(d->src_addr, d->src_prot);
> + kunmap_atomic(d->src_addr);
> d->src_addr = NULL;
> }
>
> if (unmap_dst) {
> - ttm_kunmap_atomic_prot(d->dst_addr, d->dst_prot);
> + kunmap_atomic(d->dst_addr);
> d->dst_addr = NULL;
> }
>
> @@ -388,8 +388,8 @@ static int vmw_bo_cpu_blit_line(struct vmw_bo_blit_line_data *d,
> return -EINVAL;
>
> d->dst_addr =
> - ttm_kmap_atomic_prot(d->dst_pages[dst_page],
> - d->dst_prot);
> + kmap_atomic_prot(d->dst_pages[dst_page],
> + d->dst_prot);
> if (!d->dst_addr)
> return -ENOMEM;
>
> @@ -401,8 +401,8 @@ static int vmw_bo_cpu_blit_line(struct vmw_bo_blit_line_data *d,
> return -EINVAL;
>
> d->src_addr =
> - ttm_kmap_atomic_prot(d->src_pages[src_page],
> - d->src_prot);
> + kmap_atomic_prot(d->src_pages[src_page],
> + d->src_prot);
> if (!d->src_addr)
> return -ENOMEM;
>
> @@ -499,9 +499,9 @@ int vmw_bo_cpu_blit(struct ttm_buffer_object *dst,
> }
> out:
> if (d.src_addr)
> - ttm_kunmap_atomic_prot(d.src_addr, d.src_prot);
> + kunmap_atomic(d.src_addr);
> if (d.dst_addr)
> - ttm_kunmap_atomic_prot(d.dst_addr, d.dst_prot);
> + kunmap_atomic(d.dst_addr);
>
> return ret;
> }
> diff --git a/include/drm/ttm/ttm_bo_api.h b/include/drm/ttm/ttm_bo_api.h
> index 0a9d042e075a..de1ccdcd5703 100644
> --- a/include/drm/ttm/ttm_bo_api.h
> +++ b/include/drm/ttm/ttm_bo_api.h
> @@ -668,10 +668,6 @@ int ttm_bo_mmap_obj(struct vm_area_struct *vma, struct ttm_buffer_object *bo);
> int ttm_bo_mmap(struct file *filp, struct vm_area_struct *vma,
> struct ttm_bo_device *bdev);
>
> -void *ttm_kmap_atomic_prot(struct page *page, pgprot_t prot);
> -
> -void ttm_kunmap_atomic_prot(void *addr, pgprot_t prot);
> -
> /**
> * ttm_bo_io
> *
> --
> 2.25.1
>
> _______________________________________________
> dri-devel mailing list
> dri-devel@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/dri-devel
--
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch
^ permalink raw reply
* [PATCH] powerpc/64s/radix: Don't prefetch DAR in update_mmu_cache
From: Nicholas Piggin @ 2020-05-04 12:29 UTC (permalink / raw)
To: linuxppc-dev; +Cc: Nicholas Piggin
The idea behind this prefetch was to kick off a page table walk before
returning from the fault, getting some pipelining advantage.
But this never showed up any noticable performance advantage, and in
fact with KUAP the prefetches are actually blocked and cause some
kind of micro-architectural fault. Removing this improves page fault
microbenchmark performance by about 9%.
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/book3s/64/pgtable.h | 7 +++++--
arch/powerpc/mm/book3s64/hash_utils.c | 5 -----
arch/powerpc/mm/book3s64/pgtable.c | 13 -------------
3 files changed, 5 insertions(+), 20 deletions(-)
diff --git a/arch/powerpc/include/asm/book3s/64/pgtable.h b/arch/powerpc/include/asm/book3s/64/pgtable.h
index 368b136517e0..59ed15e43e89 100644
--- a/arch/powerpc/include/asm/book3s/64/pgtable.h
+++ b/arch/powerpc/include/asm/book3s/64/pgtable.h
@@ -1139,8 +1139,11 @@ extern pmd_t mk_pmd(struct page *page, pgprot_t pgprot);
extern pmd_t pmd_modify(pmd_t pmd, pgprot_t newprot);
extern void set_pmd_at(struct mm_struct *mm, unsigned long addr,
pmd_t *pmdp, pmd_t pmd);
-extern void update_mmu_cache_pmd(struct vm_area_struct *vma, unsigned long addr,
- pmd_t *pmd);
+static inline void update_mmu_cache_pmd(struct vm_area_struct *vma,
+ unsigned long addr, pmd_t *pmd)
+{
+}
+
extern int hash__has_transparent_hugepage(void);
static inline int has_transparent_hugepage(void)
{
diff --git a/arch/powerpc/mm/book3s64/hash_utils.c b/arch/powerpc/mm/book3s64/hash_utils.c
index 845da1e8ca4f..2458615805ee 100644
--- a/arch/powerpc/mm/book3s64/hash_utils.c
+++ b/arch/powerpc/mm/book3s64/hash_utils.c
@@ -1672,11 +1672,6 @@ void update_mmu_cache(struct vm_area_struct *vma, unsigned long address,
unsigned long trap;
bool is_exec;
- if (radix_enabled()) {
- prefetch((void *)address);
- return;
- }
-
/* We only want HPTEs for linux PTEs that have _PAGE_ACCESSED set */
if (!pte_young(*ptep) || address >= TASK_SIZE)
return;
diff --git a/arch/powerpc/mm/book3s64/pgtable.c b/arch/powerpc/mm/book3s64/pgtable.c
index e0bb69c616e4..821b483a5ac3 100644
--- a/arch/powerpc/mm/book3s64/pgtable.c
+++ b/arch/powerpc/mm/book3s64/pgtable.c
@@ -146,19 +146,6 @@ pmd_t pmd_modify(pmd_t pmd, pgprot_t newprot)
pmdv &= _HPAGE_CHG_MASK;
return pmd_set_protbits(__pmd(pmdv), newprot);
}
-
-/*
- * This is called at the end of handling a user page fault, when the
- * fault has been handled by updating a HUGE PMD entry in the linux page tables.
- * We use it to preload an HPTE into the hash table corresponding to
- * the updated linux HUGE PMD entry.
- */
-void update_mmu_cache_pmd(struct vm_area_struct *vma, unsigned long addr,
- pmd_t *pmd)
-{
- if (radix_enabled())
- prefetch((void *)addr);
-}
#endif /* CONFIG_TRANSPARENT_HUGEPAGE */
/* For use by kexec */
--
2.23.0
^ permalink raw reply related
* Re: [PATCH 1/2] powerpc/spufs: fix copy_to_user while atomic
From: Christoph Hellwig @ 2020-05-04 13:04 UTC (permalink / raw)
To: Jeremy Kerr; +Cc: linuxppc-dev, Christoph Hellwig, Arnd Bergmann
In-Reply-To: <20200429070303.17599-1-jk@ozlabs.org>
powerpc mantainers,
are you going to pick this up for the next -rc1? I'm waiting for it to
hit upstream before resending the coredump series.
On Wed, Apr 29, 2020 at 03:03:02PM +0800, Jeremy Kerr wrote:
> Currently, we may perform a copy_to_user (through
> simple_read_from_buffer()) while holding a context's register_lock,
> while accessing the context save area.
>
> This change uses a temporary buffer for the context save area data,
> which we then pass to simple_read_from_buffer.
>
> Includes changes from Christoph Hellwig <hch@lst.de>.
>
> Fixes: bf1ab978be23 ("[POWERPC] coredump: Add SPU elf notes to coredump.")
> Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
> Reviewed-by: Arnd Bergmann <arnd@arndb.de>
> Reviewed-by: Christoph Hellwig <hch@lst.de>
> ---
> arch/powerpc/platforms/cell/spufs/file.c | 113 +++++++++++++++--------
> 1 file changed, 75 insertions(+), 38 deletions(-)
>
> diff --git a/arch/powerpc/platforms/cell/spufs/file.c b/arch/powerpc/platforms/cell/spufs/file.c
> index c0f950a3f4e1..b4e1ef650b40 100644
> --- a/arch/powerpc/platforms/cell/spufs/file.c
> +++ b/arch/powerpc/platforms/cell/spufs/file.c
> @@ -1978,8 +1978,9 @@ static ssize_t __spufs_mbox_info_read(struct spu_context *ctx,
> static ssize_t spufs_mbox_info_read(struct file *file, char __user *buf,
> size_t len, loff_t *pos)
> {
> - int ret;
> struct spu_context *ctx = file->private_data;
> + u32 stat, data;
> + int ret;
>
> if (!access_ok(buf, len))
> return -EFAULT;
> @@ -1988,11 +1989,16 @@ static ssize_t spufs_mbox_info_read(struct file *file, char __user *buf,
> if (ret)
> return ret;
> spin_lock(&ctx->csa.register_lock);
> - ret = __spufs_mbox_info_read(ctx, buf, len, pos);
> + stat = ctx->csa.prob.mb_stat_R;
> + data = ctx->csa.prob.pu_mb_R;
> spin_unlock(&ctx->csa.register_lock);
> spu_release_saved(ctx);
>
> - return ret;
> + /* EOF if there's no entry in the mbox */
> + if (!(stat & 0x0000ff))
> + return 0;
> +
> + return simple_read_from_buffer(buf, len, pos, &data, sizeof(data));
> }
>
> static const struct file_operations spufs_mbox_info_fops = {
> @@ -2019,6 +2025,7 @@ static ssize_t spufs_ibox_info_read(struct file *file, char __user *buf,
> size_t len, loff_t *pos)
> {
> struct spu_context *ctx = file->private_data;
> + u32 stat, data;
> int ret;
>
> if (!access_ok(buf, len))
> @@ -2028,11 +2035,16 @@ static ssize_t spufs_ibox_info_read(struct file *file, char __user *buf,
> if (ret)
> return ret;
> spin_lock(&ctx->csa.register_lock);
> - ret = __spufs_ibox_info_read(ctx, buf, len, pos);
> + stat = ctx->csa.prob.mb_stat_R;
> + data = ctx->csa.priv2.puint_mb_R;
> spin_unlock(&ctx->csa.register_lock);
> spu_release_saved(ctx);
>
> - return ret;
> + /* EOF if there's no entry in the ibox */
> + if (!(stat & 0xff0000))
> + return 0;
> +
> + return simple_read_from_buffer(buf, len, pos, &data, sizeof(data));
> }
>
> static const struct file_operations spufs_ibox_info_fops = {
> @@ -2041,6 +2053,11 @@ static const struct file_operations spufs_ibox_info_fops = {
> .llseek = generic_file_llseek,
> };
>
> +static size_t spufs_wbox_info_cnt(struct spu_context *ctx)
> +{
> + return (4 - ((ctx->csa.prob.mb_stat_R & 0x00ff00) >> 8)) * sizeof(u32);
> +}
> +
> static ssize_t __spufs_wbox_info_read(struct spu_context *ctx,
> char __user *buf, size_t len, loff_t *pos)
> {
> @@ -2049,7 +2066,7 @@ static ssize_t __spufs_wbox_info_read(struct spu_context *ctx,
> u32 wbox_stat;
>
> wbox_stat = ctx->csa.prob.mb_stat_R;
> - cnt = 4 - ((wbox_stat & 0x00ff00) >> 8);
> + cnt = spufs_wbox_info_cnt(ctx);
> for (i = 0; i < cnt; i++) {
> data[i] = ctx->csa.spu_mailbox_data[i];
> }
> @@ -2062,7 +2079,8 @@ static ssize_t spufs_wbox_info_read(struct file *file, char __user *buf,
> size_t len, loff_t *pos)
> {
> struct spu_context *ctx = file->private_data;
> - int ret;
> + u32 data[ARRAY_SIZE(ctx->csa.spu_mailbox_data)];
> + int ret, count;
>
> if (!access_ok(buf, len))
> return -EFAULT;
> @@ -2071,11 +2089,13 @@ static ssize_t spufs_wbox_info_read(struct file *file, char __user *buf,
> if (ret)
> return ret;
> spin_lock(&ctx->csa.register_lock);
> - ret = __spufs_wbox_info_read(ctx, buf, len, pos);
> + count = spufs_wbox_info_cnt(ctx);
> + memcpy(&data, &ctx->csa.spu_mailbox_data, sizeof(data));
> spin_unlock(&ctx->csa.register_lock);
> spu_release_saved(ctx);
>
> - return ret;
> + return simple_read_from_buffer(buf, len, pos, &data,
> + count * sizeof(u32));
> }
>
> static const struct file_operations spufs_wbox_info_fops = {
> @@ -2084,27 +2104,33 @@ static const struct file_operations spufs_wbox_info_fops = {
> .llseek = generic_file_llseek,
> };
>
> -static ssize_t __spufs_dma_info_read(struct spu_context *ctx,
> - char __user *buf, size_t len, loff_t *pos)
> +static void ___spufs_dma_info_read(struct spu_context *ctx,
> + struct spu_dma_info *info)
> {
> - struct spu_dma_info info;
> - struct mfc_cq_sr *qp, *spuqp;
> int i;
>
> - info.dma_info_type = ctx->csa.priv2.spu_tag_status_query_RW;
> - info.dma_info_mask = ctx->csa.lscsa->tag_mask.slot[0];
> - info.dma_info_status = ctx->csa.spu_chnldata_RW[24];
> - info.dma_info_stall_and_notify = ctx->csa.spu_chnldata_RW[25];
> - info.dma_info_atomic_command_status = ctx->csa.spu_chnldata_RW[27];
> + info->dma_info_type = ctx->csa.priv2.spu_tag_status_query_RW;
> + info->dma_info_mask = ctx->csa.lscsa->tag_mask.slot[0];
> + info->dma_info_status = ctx->csa.spu_chnldata_RW[24];
> + info->dma_info_stall_and_notify = ctx->csa.spu_chnldata_RW[25];
> + info->dma_info_atomic_command_status = ctx->csa.spu_chnldata_RW[27];
> for (i = 0; i < 16; i++) {
> - qp = &info.dma_info_command_data[i];
> - spuqp = &ctx->csa.priv2.spuq[i];
> + struct mfc_cq_sr *qp = &info->dma_info_command_data[i];
> + struct mfc_cq_sr *spuqp = &ctx->csa.priv2.spuq[i];
>
> qp->mfc_cq_data0_RW = spuqp->mfc_cq_data0_RW;
> qp->mfc_cq_data1_RW = spuqp->mfc_cq_data1_RW;
> qp->mfc_cq_data2_RW = spuqp->mfc_cq_data2_RW;
> qp->mfc_cq_data3_RW = spuqp->mfc_cq_data3_RW;
> }
> +}
> +
> +static ssize_t __spufs_dma_info_read(struct spu_context *ctx,
> + char __user *buf, size_t len, loff_t *pos)
> +{
> + struct spu_dma_info info;
> +
> + ___spufs_dma_info_read(ctx, &info);
>
> return simple_read_from_buffer(buf, len, pos, &info,
> sizeof info);
> @@ -2114,6 +2140,7 @@ static ssize_t spufs_dma_info_read(struct file *file, char __user *buf,
> size_t len, loff_t *pos)
> {
> struct spu_context *ctx = file->private_data;
> + struct spu_dma_info info;
> int ret;
>
> if (!access_ok(buf, len))
> @@ -2123,11 +2150,12 @@ static ssize_t spufs_dma_info_read(struct file *file, char __user *buf,
> if (ret)
> return ret;
> spin_lock(&ctx->csa.register_lock);
> - ret = __spufs_dma_info_read(ctx, buf, len, pos);
> + ___spufs_dma_info_read(ctx, &info);
> spin_unlock(&ctx->csa.register_lock);
> spu_release_saved(ctx);
>
> - return ret;
> + return simple_read_from_buffer(buf, len, pos, &info,
> + sizeof(info));
> }
>
> static const struct file_operations spufs_dma_info_fops = {
> @@ -2136,13 +2164,31 @@ static const struct file_operations spufs_dma_info_fops = {
> .llseek = no_llseek,
> };
>
> +static void ___spufs_proxydma_info_read(struct spu_context *ctx,
> + struct spu_proxydma_info *info)
> +{
> + int i;
> +
> + info->proxydma_info_type = ctx->csa.prob.dma_querytype_RW;
> + info->proxydma_info_mask = ctx->csa.prob.dma_querymask_RW;
> + info->proxydma_info_status = ctx->csa.prob.dma_tagstatus_R;
> +
> + for (i = 0; i < 8; i++) {
> + struct mfc_cq_sr *qp = &info->proxydma_info_command_data[i];
> + struct mfc_cq_sr *puqp = &ctx->csa.priv2.puq[i];
> +
> + qp->mfc_cq_data0_RW = puqp->mfc_cq_data0_RW;
> + qp->mfc_cq_data1_RW = puqp->mfc_cq_data1_RW;
> + qp->mfc_cq_data2_RW = puqp->mfc_cq_data2_RW;
> + qp->mfc_cq_data3_RW = puqp->mfc_cq_data3_RW;
> + }
> +}
> +
> static ssize_t __spufs_proxydma_info_read(struct spu_context *ctx,
> char __user *buf, size_t len, loff_t *pos)
> {
> struct spu_proxydma_info info;
> - struct mfc_cq_sr *qp, *puqp;
> int ret = sizeof info;
> - int i;
>
> if (len < ret)
> return -EINVAL;
> @@ -2150,18 +2196,7 @@ static ssize_t __spufs_proxydma_info_read(struct spu_context *ctx,
> if (!access_ok(buf, len))
> return -EFAULT;
>
> - info.proxydma_info_type = ctx->csa.prob.dma_querytype_RW;
> - info.proxydma_info_mask = ctx->csa.prob.dma_querymask_RW;
> - info.proxydma_info_status = ctx->csa.prob.dma_tagstatus_R;
> - for (i = 0; i < 8; i++) {
> - qp = &info.proxydma_info_command_data[i];
> - puqp = &ctx->csa.priv2.puq[i];
> -
> - qp->mfc_cq_data0_RW = puqp->mfc_cq_data0_RW;
> - qp->mfc_cq_data1_RW = puqp->mfc_cq_data1_RW;
> - qp->mfc_cq_data2_RW = puqp->mfc_cq_data2_RW;
> - qp->mfc_cq_data3_RW = puqp->mfc_cq_data3_RW;
> - }
> + ___spufs_proxydma_info_read(ctx, &info);
>
> return simple_read_from_buffer(buf, len, pos, &info,
> sizeof info);
> @@ -2171,17 +2206,19 @@ static ssize_t spufs_proxydma_info_read(struct file *file, char __user *buf,
> size_t len, loff_t *pos)
> {
> struct spu_context *ctx = file->private_data;
> + struct spu_proxydma_info info;
> int ret;
>
> ret = spu_acquire_saved(ctx);
> if (ret)
> return ret;
> spin_lock(&ctx->csa.register_lock);
> - ret = __spufs_proxydma_info_read(ctx, buf, len, pos);
> + ___spufs_proxydma_info_read(ctx, &info);
> spin_unlock(&ctx->csa.register_lock);
> spu_release_saved(ctx);
>
> - return ret;
> + return simple_read_from_buffer(buf, len, pos, &info,
> + sizeof(info));
> }
>
> static const struct file_operations spufs_proxydma_info_fops = {
> --
> 2.17.1
---end quoted text---
^ permalink raw reply
* [PATCH] powerpc/64/signal: balance return predictor stack in signal trampoline
From: Nicholas Piggin @ 2020-05-04 14:20 UTC (permalink / raw)
To: linuxppc-dev; +Cc: Nicholas Piggin, Alan Modra
Returning from an interrupt or syscall to a signal handler currently
begins execution directly at the handler's entry point, with LR set to
the address of the sigreturn trampoline. When the signal handler
function returns, it runs the trampoline. It looks like this:
# interrupt at user address xyz
# kernel stuff... signal is raised
rfid
# void handler(int sig)
addis 2,12,.TOC.-.LCF0@ha
addi 2,2,.TOC.-.LCF0@l
mflr 0
std 0,16(1)
stdu 1,-96(1)
# handler stuff
ld 0,16(1)
mtlr 0
blr
# __kernel_sigtramp_rt64
addi r1,r1,__SIGNAL_FRAMESIZE
li r0,__NR_rt_sigreturn
sc
# kernel executes rt_sigreturn
rfid
# back to user address xyz
Note the blr with no matching bl. This can corrupt the return predictor.
Solve this by instead resuming execution at the signal trampoline which
then calls the signal handler. qtrace-tools link_stack checker confirms
the entire user/kernel/vdso cycle is balanced after this patch, whereas
it's not upstream.
Alan confirms the dwarf unwind info still looks good. gdb still
recognises the signal frame and can step into parent frames if it break
inside a signal handler.
Performance is pretty noisy, not a very significant change on a POWER9
here, but branch misses are consistently a lot lower on a microbenchmark:
Performance counter stats for './signal':
13,085.72 msec task-clock # 1.000 CPUs utilized
45,024,760,101 cycles # 3.441 GHz
65,102,895,542 instructions # 1.45 insn per cycle
11,271,673,787 branches # 861.372 M/sec
59,468,979 branch-misses # 0.53% of all branches
12,989.09 msec task-clock # 1.000 CPUs utilized
44,692,719,559 cycles # 3.441 GHz
65,109,984,964 instructions # 1.46 insn per cycle
11,282,136,057 branches # 868.585 M/sec
39,786,942 branch-misses # 0.35% of all branches
Cc: Alan Modra <amodra@gmail.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
---
arch/powerpc/include/asm/ppc-opcode.h | 1 +
arch/powerpc/kernel/signal_64.c | 20 +++++++++++---------
arch/powerpc/kernel/vdso64/sigtramp.S | 13 +++++--------
3 files changed, 17 insertions(+), 17 deletions(-)
diff --git a/arch/powerpc/include/asm/ppc-opcode.h b/arch/powerpc/include/asm/ppc-opcode.h
index c1df75edde44..747b37f1ce09 100644
--- a/arch/powerpc/include/asm/ppc-opcode.h
+++ b/arch/powerpc/include/asm/ppc-opcode.h
@@ -329,6 +329,7 @@
#define PPC_INST_BLR 0x4e800020
#define PPC_INST_BLRL 0x4e800021
#define PPC_INST_BCTR 0x4e800420
+#define PPC_INST_BCTRL 0x4e800421
#define PPC_INST_MULLD 0x7c0001d2
#define PPC_INST_MULLW 0x7c0001d6
#define PPC_INST_MULHWU 0x7c000016
diff --git a/arch/powerpc/kernel/signal_64.c b/arch/powerpc/kernel/signal_64.c
index adfde59cf4ba..6c17e2456ccc 100644
--- a/arch/powerpc/kernel/signal_64.c
+++ b/arch/powerpc/kernel/signal_64.c
@@ -41,7 +41,7 @@
#define FP_REGS_SIZE sizeof(elf_fpregset_t)
#define TRAMP_TRACEBACK 3
-#define TRAMP_SIZE 6
+#define TRAMP_SIZE 7
/*
* When we have signals to deliver, we set up on the user stack,
@@ -603,13 +603,15 @@ static long setup_trampoline(unsigned int syscall, unsigned int __user *tramp)
int i;
long err = 0;
+ /* bctrl # call the handler */
+ err |= __put_user(PPC_INST_BCTRL, &tramp[0]);
/* addi r1, r1, __SIGNAL_FRAMESIZE # Pop the dummy stackframe */
err |= __put_user(PPC_INST_ADDI | __PPC_RT(R1) | __PPC_RA(R1) |
- (__SIGNAL_FRAMESIZE & 0xffff), &tramp[0]);
+ (__SIGNAL_FRAMESIZE & 0xffff), &tramp[1]);
/* li r0, __NR_[rt_]sigreturn| */
- err |= __put_user(PPC_INST_ADDI | (syscall & 0xffff), &tramp[1]);
+ err |= __put_user(PPC_INST_ADDI | (syscall & 0xffff), &tramp[2]);
/* sc */
- err |= __put_user(PPC_INST_SC, &tramp[2]);
+ err |= __put_user(PPC_INST_SC, &tramp[3]);
/* Minimal traceback info */
for (i=TRAMP_TRACEBACK; i < TRAMP_SIZE ;i++)
@@ -867,12 +869,12 @@ int handle_rt_signal64(struct ksignal *ksig, sigset_t *set,
/* Set up to return from userspace. */
if (vdso64_rt_sigtramp && tsk->mm->context.vdso_base) {
- regs->link = tsk->mm->context.vdso_base + vdso64_rt_sigtramp;
+ regs->nip = tsk->mm->context.vdso_base + vdso64_rt_sigtramp;
} else {
err |= setup_trampoline(__NR_rt_sigreturn, &frame->tramp[0]);
if (err)
goto badframe;
- regs->link = (unsigned long) &frame->tramp[0];
+ regs->nip = (unsigned long) &frame->tramp[0];
}
/* Allocate a dummy caller frame for the signal handler. */
@@ -881,8 +883,8 @@ int handle_rt_signal64(struct ksignal *ksig, sigset_t *set,
/* Set up "regs" so we "return" to the signal handler. */
if (is_elf2_task()) {
- regs->nip = (unsigned long) ksig->ka.sa.sa_handler;
- regs->gpr[12] = regs->nip;
+ regs->ctr = (unsigned long) ksig->ka.sa.sa_handler;
+ regs->gpr[12] = regs->ctr;
} else {
/* Handler is *really* a pointer to the function descriptor for
* the signal routine. The first entry in the function
@@ -892,7 +894,7 @@ int handle_rt_signal64(struct ksignal *ksig, sigset_t *set,
func_descr_t __user *funct_desc_ptr =
(func_descr_t __user *) ksig->ka.sa.sa_handler;
- err |= get_user(regs->nip, &funct_desc_ptr->entry);
+ err |= get_user(regs->ctr, &funct_desc_ptr->entry);
err |= get_user(regs->gpr[2], &funct_desc_ptr->toc);
}
diff --git a/arch/powerpc/kernel/vdso64/sigtramp.S b/arch/powerpc/kernel/vdso64/sigtramp.S
index a8cc0409d7d2..bbf68cd01088 100644
--- a/arch/powerpc/kernel/vdso64/sigtramp.S
+++ b/arch/powerpc/kernel/vdso64/sigtramp.S
@@ -6,6 +6,7 @@
* Copyright (C) 2004 Benjamin Herrenschmuidt (benh@kernel.crashing.org), IBM Corp.
* Copyright (C) 2004 Alan Modra (amodra@au.ibm.com)), IBM Corp.
*/
+#include <asm/cache.h> /* IFETCH_ALIGN_BYTES */
#include <asm/processor.h>
#include <asm/ppc_asm.h>
#include <asm/unistd.h>
@@ -14,21 +15,17 @@
.text
-/* The nop here is a hack. The dwarf2 unwind routines subtract 1 from
- the return address to get an address in the middle of the presumed
- call instruction. Since we don't have a call here, we artificially
- extend the range covered by the unwind info by padding before the
- real start. */
- nop
.balign 8
+ .balign IFETCH_ALIGN_BYTES
V_FUNCTION_BEGIN(__kernel_sigtramp_rt64)
-.Lsigrt_start = . - 4
+.Lsigrt_start:
+ bctrl /* call the handler */
addi r1, r1, __SIGNAL_FRAMESIZE
li r0,__NR_rt_sigreturn
sc
.Lsigrt_end:
V_FUNCTION_END(__kernel_sigtramp_rt64)
-/* The ".balign 8" above and the following zeros mimic the old stack
+/* The .balign 8 above and the following zeros mimic the old stack
trampoline layout. The last magic value is the ucontext pointer,
chosen in such a way that older libgcc unwind code returns a zero
for a sigcontext pointer. */
--
2.23.0
^ permalink raw reply related
* [powerpc:merge] BUILD SUCCESS 1bc92fe3175eb26ff37e580c0383d7a9abe06835
From: kbuild test robot @ 2020-05-04 14:39 UTC (permalink / raw)
To: Michael Ellerman; +Cc: linuxppc-dev
tree/branch: https://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux.git merge
branch HEAD: 1bc92fe3175eb26ff37e580c0383d7a9abe06835 Automatic merge of branches 'master', 'next' and 'fixes' into merge
elapsed time: 5894m
configs tested: 248
configs skipped: 0
The following configs have been built successfully.
More configs may be tested in the coming days.
arm efm32_defconfig
arm at91_dt_defconfig
arm shmobile_defconfig
arm64 defconfig
arm multi_v5_defconfig
arm sunxi_defconfig
arm multi_v7_defconfig
arm exynos_defconfig
arm64 allyesconfig
arm allyesconfig
arm64 allmodconfig
arm allmodconfig
arm64 allnoconfig
arm allnoconfig
sparc allyesconfig
powerpc defconfig
ia64 defconfig
powerpc chrp32_defconfig
c6x evmc6678_defconfig
xtensa common_defconfig
ia64 alldefconfig
nds32 allnoconfig
mips capcella_defconfig
riscv nommu_virt_defconfig
mips allmodconfig
s390 zfcpdump_defconfig
powerpc mpc512x_defconfig
sh sh7785lcr_32bit_defconfig
xtensa iss_defconfig
um defconfig
sh rsk7269_defconfig
mips allnoconfig
csky defconfig
i386 allnoconfig
i386 allyesconfig
i386 alldefconfig
i386 defconfig
i386 debian-10.3
ia64 allmodconfig
ia64 allnoconfig
ia64 generic_defconfig
ia64 tiger_defconfig
ia64 bigsur_defconfig
ia64 allyesconfig
m68k m5475evb_defconfig
m68k allmodconfig
m68k bvme6000_defconfig
m68k sun3_defconfig
m68k multi_defconfig
nios2 3c120_defconfig
nios2 10m50_defconfig
c6x allyesconfig
openrisc simple_smp_defconfig
openrisc or1ksim_defconfig
nds32 defconfig
alpha defconfig
h8300 h8s-sim_defconfig
h8300 edosk2674_defconfig
h8300 h8300h-sim_defconfig
arc defconfig
arc allyesconfig
microblaze mmu_defconfig
microblaze nommu_defconfig
mips fuloong2e_defconfig
mips malta_kvm_defconfig
mips ar7_defconfig
mips allyesconfig
mips 64r6el_defconfig
mips 32r2_defconfig
mips malta_kvm_guest_defconfig
mips tb0287_defconfig
mips ip32_defconfig
mips decstation_64_defconfig
mips loongson3_defconfig
mips ath79_defconfig
mips bcm63xx_defconfig
parisc allnoconfig
parisc generic-64bit_defconfig
parisc generic-32bit_defconfig
parisc allyesconfig
parisc allmodconfig
powerpc holly_defconfig
powerpc ppc64_defconfig
powerpc rhel-kconfig
powerpc allnoconfig
powerpc mpc866_ads_defconfig
powerpc amigaone_defconfig
powerpc adder875_defconfig
powerpc ep8248e_defconfig
powerpc g5_defconfig
parisc randconfig-a001-20200430
mips randconfig-a001-20200430
m68k randconfig-a001-20200430
riscv randconfig-a001-20200430
alpha randconfig-a001-20200430
nds32 randconfig-a001-20200430
m68k randconfig-a001-20200502
mips randconfig-a001-20200502
nds32 randconfig-a001-20200502
alpha randconfig-a001-20200502
parisc randconfig-a001-20200502
riscv randconfig-a001-20200502
microblaze randconfig-a001-20200430
nios2 randconfig-a001-20200430
h8300 randconfig-a001-20200430
c6x randconfig-a001-20200430
sparc64 randconfig-a001-20200430
h8300 randconfig-a001-20200502
nios2 randconfig-a001-20200502
microblaze randconfig-a001-20200502
c6x randconfig-a001-20200502
sparc64 randconfig-a001-20200502
s390 randconfig-a001-20200430
xtensa randconfig-a001-20200430
csky randconfig-a001-20200430
openrisc randconfig-a001-20200430
sh randconfig-a001-20200430
s390 randconfig-a001-20200502
xtensa randconfig-a001-20200502
sh randconfig-a001-20200502
openrisc randconfig-a001-20200502
csky randconfig-a001-20200502
s390 randconfig-a001-20200501
xtensa randconfig-a001-20200501
sh randconfig-a001-20200501
openrisc randconfig-a001-20200501
csky randconfig-a001-20200501
x86_64 randconfig-a003-20200501
x86_64 randconfig-a001-20200501
i386 randconfig-a003-20200501
i386 randconfig-a002-20200501
i386 randconfig-a001-20200501
i386 randconfig-b003-20200501
x86_64 randconfig-b002-20200501
i386 randconfig-b001-20200501
x86_64 randconfig-b003-20200501
x86_64 randconfig-b001-20200501
i386 randconfig-b002-20200501
i386 randconfig-b001-20200430
i386 randconfig-b002-20200430
x86_64 randconfig-b001-20200430
i386 randconfig-b003-20200430
x86_64 randconfig-b002-20200430
x86_64 randconfig-b003-20200430
i386 randconfig-b003-20200502
i386 randconfig-b001-20200502
x86_64 randconfig-b003-20200502
x86_64 randconfig-b001-20200502
i386 randconfig-b002-20200502
x86_64 randconfig-d001-20200501
i386 randconfig-d003-20200501
x86_64 randconfig-d003-20200501
i386 randconfig-d001-20200501
x86_64 randconfig-d002-20200501
i386 randconfig-d002-20200501
x86_64 randconfig-e003-20200503
x86_64 randconfig-e002-20200503
i386 randconfig-e003-20200503
x86_64 randconfig-e001-20200503
i386 randconfig-e002-20200503
i386 randconfig-e001-20200503
x86_64 randconfig-e002-20200430
i386 randconfig-e003-20200430
x86_64 randconfig-e003-20200430
i386 randconfig-e002-20200430
x86_64 randconfig-e001-20200430
i386 randconfig-e001-20200430
i386 randconfig-f003-20200501
x86_64 randconfig-f001-20200501
x86_64 randconfig-f003-20200501
i386 randconfig-f001-20200501
i386 randconfig-f002-20200501
x86_64 randconfig-a003-20200502
x86_64 randconfig-a001-20200502
x86_64 randconfig-a002-20200502
i386 randconfig-a002-20200502
i386 randconfig-a003-20200502
i386 randconfig-a001-20200502
x86_64 randconfig-g003-20200502
i386 randconfig-g003-20200502
i386 randconfig-g002-20200502
x86_64 randconfig-g001-20200502
x86_64 randconfig-g002-20200502
i386 randconfig-g001-20200502
i386 randconfig-g003-20200501
i386 randconfig-g002-20200501
x86_64 randconfig-g002-20200501
i386 randconfig-g001-20200501
i386 randconfig-h002-20200430
i386 randconfig-h003-20200430
x86_64 randconfig-h001-20200430
x86_64 randconfig-h003-20200430
i386 randconfig-h001-20200430
i386 randconfig-h001-20200501
i386 randconfig-h002-20200501
i386 randconfig-h003-20200501
x86_64 randconfig-h001-20200501
x86_64 randconfig-h003-20200501
i386 randconfig-h001-20200502
i386 randconfig-h002-20200502
i386 randconfig-h003-20200502
x86_64 randconfig-h002-20200502
x86_64 randconfig-h001-20200502
x86_64 randconfig-h003-20200502
ia64 randconfig-a001-20200502
arm64 randconfig-a001-20200502
arc randconfig-a001-20200502
powerpc randconfig-a001-20200502
arm randconfig-a001-20200502
sparc randconfig-a001-20200502
ia64 randconfig-a001-20200501
arc randconfig-a001-20200501
powerpc randconfig-a001-20200501
arm randconfig-a001-20200501
sparc randconfig-a001-20200501
sparc randconfig-a001-20200430
arc randconfig-a001-20200430
ia64 randconfig-a001-20200430
powerpc randconfig-a001-20200430
arm randconfig-a001-20200430
riscv allyesconfig
riscv allnoconfig
riscv defconfig
riscv rv32_defconfig
riscv allmodconfig
s390 debug_defconfig
s390 allyesconfig
s390 allnoconfig
s390 allmodconfig
s390 alldefconfig
s390 defconfig
sh allmodconfig
sh titan_defconfig
sh allnoconfig
sparc defconfig
sparc64 defconfig
sparc64 allnoconfig
sparc64 allyesconfig
sparc64 allmodconfig
um x86_64_defconfig
um i386_defconfig
x86_64 rhel
x86_64 rhel-7.6
x86_64 rhel-7.6-kselftests
x86_64 rhel-7.2-clear
x86_64 lkp
x86_64 fedora-25
x86_64 kexec
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox