diff mbox

[v7,06/10] ppc64 ftrace: disable profiling for some functions

Message ID 20160204162107.65ECE692DF@newverein.lst.de (mailing list archive)
State Superseded, archived
Headers show

Commit Message

Torsten Duwe Jan. 25, 2016, 3:31 p.m. UTC 
At least POWER7/8 have MMUs that don't completely autoload;
a normal, recoverable memory fault might pass through these functions.
If a dynamic tracer function causes such a fault, any of these functions
being traced with -mprofile-kernel may cause an endless recursion.

Signed-off-by: Torsten Duwe <duwe@suse.de>
---
 arch/powerpc/kernel/process.c        |  2 +-
 arch/powerpc/mm/fault.c              |  2 +-
 arch/powerpc/mm/hash_utils_64.c      | 18 +++++++++---------
 arch/powerpc/mm/hugetlbpage-hash64.c |  2 +-
 arch/powerpc/mm/hugetlbpage.c        |  4 ++--
 arch/powerpc/mm/mem.c                |  2 +-
 arch/powerpc/mm/pgtable_64.c         |  2 +-
 arch/powerpc/mm/slb.c                |  6 +++---
 arch/powerpc/mm/slice.c              |  8 ++++----
 9 files changed, 23 insertions(+), 23 deletions(-)

Comments

Michael Ellerman Feb. 10, 2016, 1:50 a.m. UTC | #1 
On Mon, 2016-01-25 at 16:31 +0100, Torsten Duwe wrote:

> At least POWER7/8 have MMUs that don't completely autoload;
> a normal, recoverable memory fault might pass through these functions.
> If a dynamic tracer function causes such a fault, any of these functions
> being traced with -mprofile-kernel may cause an endless recursion.

I'm not really happy with this one, still :)

At the moment I can trace these without any problems, with either ftrace or
kprobes, but obviously it was causing you some trouble. So I'd like to
understand why you were having issues when regular tracing doesn't.

If it's the case that tracing can work for these functions, but live patching
doesn't (for some reason), then maybe these should be blocked by the live
patching infrastructure rather than at the ftrace/kprobes level.

cheers
Torsten Duwe Feb. 10, 2016, 6:01 p.m. UTC | #2 
On Wed, Feb 10, 2016 at 12:50:38PM +1100, Michael Ellerman wrote:
> On Mon, 2016-01-25 at 16:31 +0100, Torsten Duwe wrote:
> 
> > At least POWER7/8 have MMUs that don't completely autoload;
> > a normal, recoverable memory fault might pass through these functions.
> > If a dynamic tracer function causes such a fault, any of these functions
> > being traced with -mprofile-kernel may cause an endless recursion.
> 
> I'm not really happy with this one, still :)

I understand :)

> At the moment I can trace these without any problems, with either ftrace or
> kprobes, but obviously it was causing you some trouble. So I'd like to
> understand why you were having issues when regular tracing doesn't.

It was causing huge trouble during development. Make the smallest mistake
and the machine appears to freeze, unless you happen to have a gdb
beneath the VM, which tells you that it's "only" locking rock-hard into
an endless recursion. Hint: printk() is part of the recurse, usually :-/

> If it's the case that tracing can work for these functions, but live patching
> doesn't (for some reason), then maybe these should be blocked by the live
> patching infrastructure rather than at the ftrace/kprobes level.

Now with a matured patch set, the machine survives up to the point where the
NOPs are placed and all is fine. (until someone places probes there :)

But I still have a bad feeling about this. On other architectures, this
functionality is done in hardware; the functions cannot be traced either,
and the don't break (usually ;) I would rather not instrument them.

But if you don't like this patch, and it strictly isn't neccessary --
here you go: v8.

	Torsten
diff mbox

Patch

diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c
index dccc87e..5e72e8b 100644
--- a/arch/powerpc/kernel/process.c
+++ b/arch/powerpc/kernel/process.c
@@ -822,7 +822,7 @@  static inline void __switch_to_tm(struct task_struct *prev)
  * don't know which of the checkpointed state and the transactional
  * state to use.
  */
-void restore_tm_state(struct pt_regs *regs)
+notrace void restore_tm_state(struct pt_regs *regs)
 {
 	unsigned long msr_diff;
 
diff --git a/arch/powerpc/mm/fault.c b/arch/powerpc/mm/fault.c
index a67c6d7..125be37 100644
--- a/arch/powerpc/mm/fault.c
+++ b/arch/powerpc/mm/fault.c
@@ -205,7 +205,7 @@  static int mm_fault_error(struct pt_regs *regs, unsigned long addr, int fault)
  * The return value is 0 if the fault was handled, or the signal
  * number if this is a kernel fault that can't be handled here.
  */
-int __kprobes do_page_fault(struct pt_regs *regs, unsigned long address,
+notrace int __kprobes do_page_fault(struct pt_regs *regs, unsigned long address,
 			    unsigned long error_code)
 {
 	enum ctx_state prev_state = exception_enter();
diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c
index ba59d59..01d3dee 100644
--- a/arch/powerpc/mm/hash_utils_64.c
+++ b/arch/powerpc/mm/hash_utils_64.c
@@ -825,7 +825,7 @@  void early_init_mmu_secondary(void)
 /*
  * Called by asm hashtable.S for doing lazy icache flush
  */
-unsigned int hash_page_do_lazy_icache(unsigned int pp, pte_t pte, int trap)
+notrace unsigned int hash_page_do_lazy_icache(unsigned int pp, pte_t pte, int trap)
 {
 	struct page *page;
 
@@ -846,7 +846,7 @@  unsigned int hash_page_do_lazy_icache(unsigned int pp, pte_t pte, int trap)
 }
 
 #ifdef CONFIG_PPC_MM_SLICES
-static unsigned int get_paca_psize(unsigned long addr)
+static notrace unsigned int get_paca_psize(unsigned long addr)
 {
 	u64 lpsizes;
 	unsigned char *hpsizes;
@@ -875,7 +875,7 @@  unsigned int get_paca_psize(unsigned long addr)
  * For now this makes the whole process use 4k pages.
  */
 #ifdef CONFIG_PPC_64K_PAGES
-void demote_segment_4k(struct mm_struct *mm, unsigned long addr)
+notrace void demote_segment_4k(struct mm_struct *mm, unsigned long addr)
 {
 	if (get_slice_psize(mm, addr) == MMU_PAGE_4K)
 		return;
@@ -897,7 +897,7 @@  void demote_segment_4k(struct mm_struct *mm, unsigned long addr)
  * Result is 0: full permissions, _PAGE_RW: read-only,
  * _PAGE_USER or _PAGE_USER|_PAGE_RW: no access.
  */
-static int subpage_protection(struct mm_struct *mm, unsigned long ea)
+static notrace int subpage_protection(struct mm_struct *mm, unsigned long ea)
 {
 	struct subpage_prot_table *spt = &mm->context.spt;
 	u32 spp = 0;
@@ -945,7 +945,7 @@  void hash_failure_debug(unsigned long ea, unsigned long access,
 		trap, vsid, ssize, psize, lpsize, pte);
 }
 
-static void check_paca_psize(unsigned long ea, struct mm_struct *mm,
+static notrace void check_paca_psize(unsigned long ea, struct mm_struct *mm,
 			     int psize, bool user_region)
 {
 	if (user_region) {
@@ -967,7 +967,7 @@  static void check_paca_psize(unsigned long ea, struct mm_struct *mm,
  * -1 - critical hash insertion error
  * -2 - access not permitted by subpage protection mechanism
  */
-int hash_page_mm(struct mm_struct *mm, unsigned long ea,
+notrace int hash_page_mm(struct mm_struct *mm, unsigned long ea,
 		 unsigned long access, unsigned long trap,
 		 unsigned long flags)
 {
@@ -1165,7 +1165,7 @@  bail:
 }
 EXPORT_SYMBOL_GPL(hash_page_mm);
 
-int hash_page(unsigned long ea, unsigned long access, unsigned long trap,
+notrace int hash_page(unsigned long ea, unsigned long access, unsigned long trap,
 	      unsigned long dsisr)
 {
 	unsigned long flags = 0;
@@ -1296,7 +1296,7 @@  out_exit:
 /* WARNING: This is called from hash_low_64.S, if you change this prototype,
  *          do not forget to update the assembly call site !
  */
-void flush_hash_page(unsigned long vpn, real_pte_t pte, int psize, int ssize,
+notrace void flush_hash_page(unsigned long vpn, real_pte_t pte, int psize, int ssize,
 		     unsigned long flags)
 {
 	unsigned long hash, index, shift, hidx, slot;
@@ -1444,7 +1444,7 @@  void low_hash_fault(struct pt_regs *regs, unsigned long address, int rc)
 	exception_exit(prev_state);
 }
 
-long hpte_insert_repeating(unsigned long hash, unsigned long vpn,
+notrace long hpte_insert_repeating(unsigned long hash, unsigned long vpn,
 			   unsigned long pa, unsigned long rflags,
 			   unsigned long vflags, int psize, int ssize)
 {
diff --git a/arch/powerpc/mm/hugetlbpage-hash64.c b/arch/powerpc/mm/hugetlbpage-hash64.c
index e2138c7..17fc139 100644
--- a/arch/powerpc/mm/hugetlbpage-hash64.c
+++ b/arch/powerpc/mm/hugetlbpage-hash64.c
@@ -18,7 +18,7 @@  extern long hpte_insert_repeating(unsigned long hash, unsigned long vpn,
 				  unsigned long pa, unsigned long rlags,
 				  unsigned long vflags, int psize, int ssize);
 
-int __hash_page_huge(unsigned long ea, unsigned long access, unsigned long vsid,
+notrace int __hash_page_huge(unsigned long ea, unsigned long access, unsigned long vsid,
 		     pte_t *ptep, unsigned long trap, unsigned long flags,
 		     int ssize, unsigned int shift, unsigned int mmu_psize)
 {
diff --git a/arch/powerpc/mm/hugetlbpage.c b/arch/powerpc/mm/hugetlbpage.c
index 744e24b..70dda66 100644
--- a/arch/powerpc/mm/hugetlbpage.c
+++ b/arch/powerpc/mm/hugetlbpage.c
@@ -870,7 +870,7 @@  static int __init hugetlbpage_init(void)
 #endif
 arch_initcall(hugetlbpage_init);
 
-void flush_dcache_icache_hugepage(struct page *page)
+notrace void flush_dcache_icache_hugepage(struct page *page)
 {
 	int i;
 	void *start;
@@ -903,7 +903,7 @@  void flush_dcache_icache_hugepage(struct page *page)
  * when we have MSR[EE] = 0 but the paca->soft_enabled = 1
  */
 
-pte_t *__find_linux_pte_or_hugepte(pgd_t *pgdir, unsigned long ea,
+notrace pte_t *__find_linux_pte_or_hugepte(pgd_t *pgdir, unsigned long ea,
 				   bool *is_thp, unsigned *shift)
 {
 	pgd_t pgd, *pgdp;
diff --git a/arch/powerpc/mm/mem.c b/arch/powerpc/mm/mem.c
index 22d94c3..f690e8a 100644
--- a/arch/powerpc/mm/mem.c
+++ b/arch/powerpc/mm/mem.c
@@ -406,7 +406,7 @@  void flush_dcache_page(struct page *page)
 }
 EXPORT_SYMBOL(flush_dcache_page);
 
-void flush_dcache_icache_page(struct page *page)
+notrace void flush_dcache_icache_page(struct page *page)
 {
 #ifdef CONFIG_HUGETLB_PAGE
 	if (PageCompound(page)) {
diff --git a/arch/powerpc/mm/pgtable_64.c b/arch/powerpc/mm/pgtable_64.c
index 3124a20..bb9041b 100644
--- a/arch/powerpc/mm/pgtable_64.c
+++ b/arch/powerpc/mm/pgtable_64.c
@@ -442,7 +442,7 @@  static void page_table_free_rcu(void *table)
 	}
 }
 
-void pgtable_free_tlb(struct mmu_gather *tlb, void *table, int shift)
+notrace void pgtable_free_tlb(struct mmu_gather *tlb, void *table, int shift)
 {
 	unsigned long pgf = (unsigned long)table;
 
diff --git a/arch/powerpc/mm/slb.c b/arch/powerpc/mm/slb.c
index 825b687..852bd54 100644
--- a/arch/powerpc/mm/slb.c
+++ b/arch/powerpc/mm/slb.c
@@ -96,7 +96,7 @@  static inline void create_shadowed_slbe(unsigned long ea, int ssize,
 		     : "memory" );
 }
 
-static void __slb_flush_and_rebolt(void)
+static notrace void __slb_flush_and_rebolt(void)
 {
 	/* If you change this make sure you change SLB_NUM_BOLTED
 	 * and PR KVM appropriately too. */
@@ -136,7 +136,7 @@  static void __slb_flush_and_rebolt(void)
 		     : "memory");
 }
 
-void slb_flush_and_rebolt(void)
+notrace void slb_flush_and_rebolt(void)
 {
 
 	WARN_ON(!irqs_disabled());
@@ -151,7 +151,7 @@  void slb_flush_and_rebolt(void)
 	get_paca()->slb_cache_ptr = 0;
 }
 
-void slb_vmalloc_update(void)
+notrace void slb_vmalloc_update(void)
 {
 	unsigned long vflags;
 
diff --git a/arch/powerpc/mm/slice.c b/arch/powerpc/mm/slice.c
index 42954f0..5fb0e5b 100644
--- a/arch/powerpc/mm/slice.c
+++ b/arch/powerpc/mm/slice.c
@@ -76,8 +76,8 @@  static void slice_print_mask(const char *label, struct slice_mask mask) {}
 
 #endif
 
-static struct slice_mask slice_range_to_mask(unsigned long start,
-					     unsigned long len)
+static notrace struct slice_mask slice_range_to_mask(unsigned long start,
+						     unsigned long len)
 {
 	unsigned long end = start + len - 1;
 	struct slice_mask ret = { 0, 0 };
@@ -563,7 +563,7 @@  unsigned long arch_get_unmapped_area_topdown(struct file *filp,
 				       current->mm->context.user_psize, 1);
 }
 
-unsigned int get_slice_psize(struct mm_struct *mm, unsigned long addr)
+notrace unsigned int get_slice_psize(struct mm_struct *mm, unsigned long addr)
 {
 	unsigned char *hpsizes;
 	int index, mask_index;
@@ -644,7 +644,7 @@  void slice_set_user_psize(struct mm_struct *mm, unsigned int psize)
 	spin_unlock_irqrestore(&slice_convert_lock, flags);
 }
 
-void slice_set_range_psize(struct mm_struct *mm, unsigned long start,
+notrace void slice_set_range_psize(struct mm_struct *mm, unsigned long start,
 			   unsigned long len, unsigned int psize)
 {
 	struct slice_mask mask = slice_range_to_mask(start, len);