Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1591395 > unrolled thread

Re: [PATCH V2] arm64: hwpoison: add VM_FAULT_HWPOISON[_LARGE] handling

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: [PATCH V2] arm64: hwpoison: add VM_FAULT_HWPOISON[_LARGE] handling Punit Agrawal <punit.agrawal@arm.com> - 2017-03-02 19:50 +0100
    Re: [PATCH V2] arm64: hwpoison: add VM_FAULT_HWPOISON[_LARGE]  handling "Baicar, Tyler" <tbaicar@codeaurora.org> - 2017-03-07 21:40 +0100
      Re: [PATCH V2] arm64: hwpoison: add VM_FAULT_HWPOISON[_LARGE] handling Punit Agrawal <punit.agrawal@arm.com> - 2017-03-09 19:00 +0100
        Re: [PATCH V2] arm64: hwpoison: add VM_FAULT_HWPOISON[_LARGE]  handling "Baicar, Tyler" <tbaicar@codeaurora.org> - 2017-03-10 19:10 +0100
    Re: [PATCH V2] arm64: hwpoison: add VM_FAULT_HWPOISON[_LARGE] handling Punit Agrawal <punit.agrawal@arm.com> - 2017-03-07 22:00 +0100

#1591395 — Re: [PATCH V2] arm64: hwpoison: add VM_FAULT_HWPOISON[_LARGE] handling

Hi Tyler,

Tyler Baicar <tbaicar@codeaurora.org> writes:

> From: "Jonathan (Zhixiong) Zhang" <zjzhang@codeaurora.org>
>
> Add VM_FAULT_HWPOISON[_LARGE] handling to the arm64 page fault
> handler. Handling of VM_FAULT_HWPOISON[_LARGE] is very similar
> to VM_FAULT_OOM, the only difference is that a different si_code
> (BUS_MCEERR_AR) is passed to user space and si_addr_lsb field is
> initialized.
>
> Signed-off-by: Jonathan (Zhixiong) Zhang <zjzhang@codeaurora.org>
> Signed-off-by: Tyler Baicar <tbaicar@codeaurora.org>
> ---
>  arch/arm64/mm/fault.c | 22 +++++++++++++++++++---
>  1 file changed, 19 insertions(+), 3 deletions(-)
>
> diff --git a/arch/arm64/mm/fault.c b/arch/arm64/mm/fault.c
> index 156169c..ceaa82f 100644
> --- a/arch/arm64/mm/fault.c
> +++ b/arch/arm64/mm/fault.c
> @@ -30,6 +30,7 @@
>  #include <linux/highmem.h>
>  #include <linux/perf_event.h>
>  #include <linux/preempt.h>
> +#include <linux/hugetlb.h>
>  
>  #include <asm/bug.h>
>  #include <asm/cpufeature.h>
> @@ -193,9 +194,10 @@ static void __do_kernel_fault(struct mm_struct *mm, unsigned long addr,
>   */
>  static void __do_user_fault(struct task_struct *tsk, unsigned long addr,
>  			    unsigned int esr, unsigned int sig, int code,
> -			    struct pt_regs *regs)
> +			    struct pt_regs *regs, int fault)
>  {
>  	struct siginfo si;
> +	unsigned int lsb = 0;
>  
>  	if (unhandled_signal(tsk, sig) && show_unhandled_signals_ratelimited()) {
>  		pr_info("%s[%d]: unhandled %s (%d) at 0x%08lx, esr 0x%03x\n",
> @@ -211,6 +213,17 @@ static void __do_user_fault(struct task_struct *tsk, unsigned long addr,
>  	si.si_errno = 0;
>  	si.si_code = code;
>  	si.si_addr = (void __user *)addr;
> +	/*
> +	 * Either small page or large page may be poisoned.
> +	 * In other words, VM_FAULT_HWPOISON_LARGE and
> +	 * VM_FAULT_HWPOISON are mutually exclusive.
> +	 */
> +	if (fault & VM_FAULT_HWPOISON_LARGE)
> +		lsb = hstate_index_to_shift(VM_FAULT_GET_HINDEX(fault));
> +	else if (fault & VM_FAULT_HWPOISON)
> +		lsb = PAGE_SHIFT;
> +	si.si_addr_lsb = lsb;
> +
>  	force_sig_info(sig, &si, tsk);
>  }
>  
> @@ -224,7 +237,7 @@ static void do_bad_area(unsigned long addr, unsigned int esr, struct pt_regs *re
>  	 * handle this fault with.
>  	 */
>  	if (user_mode(regs))
> -		__do_user_fault(tsk, addr, esr, SIGSEGV, SEGV_MAPERR, regs);
> +		__do_user_fault(tsk, addr, esr, SIGSEGV, SEGV_MAPERR, regs, 0);
>  	else
>  		__do_kernel_fault(mm, addr, esr, regs);
>  }
> @@ -426,6 +439,9 @@ static int __kprobes do_page_fault(unsigned long addr, unsigned int esr,
>  		 */
>  		sig = SIGBUS;
>  		code = BUS_ADRERR;
> +	} else if (fault & (VM_FAULT_HWPOISON | VM_FAULT_HWPOISON_LARGE)) {
> +		sig = SIGBUS;
> +		code = BUS_MCEERR_AR;
>  	} else {
>  		/*
>  		 * Something tried to access memory that isn't in our memory
> @@ -436,7 +452,7 @@ static int __kprobes do_page_fault(unsigned long addr, unsigned int esr,
>  			SEGV_ACCERR : SEGV_MAPERR;
>  	}
>  
> -	__do_user_fault(tsk, addr, esr, sig, code, regs);
> +	__do_user_fault(tsk, addr, esr, sig, code, regs, fault);
>  	return 0;
>  
>  no_context:

The code looks good but I ran into some failures while running the
hugepages hwpoison tests from mce-tests suite[0]. I get a bad pmd error
in dmesg -

[  344.165544] mm/pgtable-generic.c:33: bad pmd 000000083af00074.

I suspect that this is due to the huge pte accessors not correctly
dealing with poisoned entries (which are represented as swap entries).

I am investigating the failure but could you try running the tests at
your end as well.

To run the tests, I cloned the repository[0]. It test needs a simple fix
at the end of this mail to run correctly. With that applied and running
as root -

# cd mce-test/cases/function/hwpoison
# ./run_hugepage.sh


[0] https://git.kernel.org/cgit/utils/cpu/mce/mce-test.git/

--------->8--------------
commit cb5c61f18dd86baf01b90404d4ecf51dd3d176c7
Author: Punit Agrawal <punit.agrawal@arm.com>
Date:   Thu Mar 2 18:24:40 2017 +0000

    Use correct return type for getopt_long

    getopt_long returns an int. Fix the return type to avoid issues when
    checking for negative error codes on architectures with unsigned char,
    e.g., arm.

    Signed-off-by: Punit Agrawal <punit.agrawal@arm.com>

diff --git a/cases/function/hwpoison/thugetlb.c b/cases/function/hwpoison/thugetlb.c
index 92dc7d2..fbcf426 100644
--- a/cases/function/hwpoison/thugetlb.c
+++ b/cases/function/hwpoison/thugetlb.c
@@ -125,7 +125,7 @@ int main(int argc, char *argv[])
        int forkflag = 0;
        int privateflag = 0;
        int cowflag = 0;
-   char c;
+ int c;
        pid_t pid = 0;
        void *expected_addr = NULL;
        struct sembuf sembuffer;

[toc] | [next] | [standalone]

#1594621 — Re: [PATCH V2] arm64: hwpoison: add VM_FAULT_HWPOISON[_LARGE] handling

On 3/7/2017 12:56 PM, Punit Agrawal wrote:
> Punit Agrawal <punit.agrawal@arm.com> writes:
>
> [...]
>
>> The code looks good but I ran into some failures while running the
>> hugepages hwpoison tests from mce-tests suite[0]. I get a bad pmd error
>> in dmesg -
>>
>> [  344.165544] mm/pgtable-generic.c:33: bad pmd 000000083af00074.
>>
>> I suspect that this is due to the huge pte accessors not correctly
>> dealing with poisoned entries (which are represented as swap entries).
> I think I've got to the bottom of the issue - the problem is due to
> huge_pte_at() returning NULL for poisoned pmd entries (which in turn is
> due to pmd_present() not handling poisoned pmd entries correctly)
>
> The following is the call chain for the failure case.
>
> do_munmap
>    unmap_region
>      unmap_vmas
>        unmap_single_vma
>          __unmap_hugepage_range_final    # The test case uses hugepages
>            __unmap_hugepage_range
>              huge_pte_offset             # Returns NULL for a poisoned pmd
>
> Reverting 5bb1cc0ff9a6 ("arm64: Ensure pmd_present() returns false after
> pmd_mknotpresent()") fixes the problem for me but I don't think that is
> the right fix.
>
> While I work on a proper fix, it would be great if you can confirm that
> reverting 5bb1cc0ff9a6 makes the problem go away at your end.
Thanks Punit! I haven't got a chance to do this yet, but I will let you 
know once I get it tested :)
>
>> I am investigating the failure but could you try running the tests at
>> your end as well.
>>
>> To run the tests, I cloned the repository[0]. It test needs a simple fix
>> at the end of this mail to run correctly. With that applied and running
>> as root -
>>
>> # cd mce-test/cases/function/hwpoison
>> # ./run_hugepage.sh
>>
>>
>> [0] https://git.kernel.org/cgit/utils/cpu/mce/mce-test.git/
>>
>> --------->8--------------
>> commit cb5c61f18dd86baf01b90404d4ecf51dd3d176c7
>> Author: Punit Agrawal <punit.agrawal@arm.com>
>> Date:   Thu Mar 2 18:24:40 2017 +0000
>>
>>      Use correct return type for getopt_long
>>
>>      getopt_long returns an int. Fix the return type to avoid issues when
>>      checking for negative error codes on architectures with unsigned char,
>>      e.g., arm.
>>
>>      Signed-off-by: Punit Agrawal <punit.agrawal@arm.com>
>>
>> diff --git a/cases/function/hwpoison/thugetlb.c b/cases/function/hwpoison/thugetlb.c
>> index 92dc7d2..fbcf426 100644
>> --- a/cases/function/hwpoison/thugetlb.c
>> +++ b/cases/function/hwpoison/thugetlb.c
>> @@ -125,7 +125,7 @@ int main(int argc, char *argv[])
>>          int forkflag = 0;
>>          int privateflag = 0;
>>          int cowflag = 0;
>> -   char c;
>> + int c;
>>          pid_t pid = 0;
>>          void *expected_addr = NULL;
>>          struct sembuf sembuffer;

-- 
Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project.

[toc] | [prev] | [next] | [standalone]

#1596270

[ +steve for arm64 mm and hugepages chops ]

"Baicar, Tyler" <tbaicar@codeaurora.org> writes:

> On 3/7/2017 12:56 PM, Punit Agrawal wrote:
>> Punit Agrawal <punit.agrawal@arm.com> writes:
>>
>> [...]
>>
>>> The code looks good but I ran into some failures while running the
>>> hugepages hwpoison tests from mce-tests suite[0]. I get a bad pmd error
>>> in dmesg -
>>>
>>> [  344.165544] mm/pgtable-generic.c:33: bad pmd 000000083af00074.
>>>
>>> I suspect that this is due to the huge pte accessors not correctly
>>> dealing with poisoned entries (which are represented as swap entries).
>> I think I've got to the bottom of the issue - the problem is due to
>> huge_pte_at() returning NULL for poisoned pmd entries (which in turn is
>> due to pmd_present() not handling poisoned pmd entries correctly)
>>
>> The following is the call chain for the failure case.
>>
>> do_munmap
>>    unmap_region
>>      unmap_vmas
>>        unmap_single_vma
>>          __unmap_hugepage_range_final    # The test case uses hugepages
>>            __unmap_hugepage_range
>>              huge_pte_offset             # Returns NULL for a poisoned pmd
>>
>> Reverting 5bb1cc0ff9a6 ("arm64: Ensure pmd_present() returns false after
>> pmd_mknotpresent()") fixes the problem for me but I don't think that is
>> the right fix.
>>
>> While I work on a proper fix, it would be great if you can confirm that
>> reverting 5bb1cc0ff9a6 makes the problem go away at your end.
> Thanks Punit! I haven't got a chance to do this yet, but I will let
> you know once I get it tested :)

This time with a patch. Please test this instead.

After a lot of head scratching, I've bit the bullet and added a check to
return the poisoned entry from huge_pte_offset(). What with having to
deal with contiguous hugepages et al., there just doesn't seem to be any
leeway in how we handle the situation here.

Let's see if there are any other ideas. Patch follows.

Thanks,
Punit

----------->8-------------
From d5ad3f428e629c80b0f93f2bbdf99b4cae28c9bc Mon Sep 17 00:00:00 2001
From: Punit Agrawal <punit.agrawal@arm.com>
Date: Thu, 9 Mar 2017 16:16:29 +0000
Subject: [PATCH] arm64: hugetlb: Fix huge_pte_offset to return poisoned pmd

When memory failure is enabled, a poisoned hugepage PMD is marked as a
swap entry. As pmd_present() only checks for VALID and PROT_NONE
bits (turned off for swap entries), it causues huge_pte_offset() to
return NULL for poisoned PMDs.

This behaviour of huge_pte_offset() leads to the error such as below
when munmap is called on poisoned hugepages.

[  344.165544] mm/pgtable-generic.c:33: bad pmd 000000083af00074.

Fix huge_pte_offset() to return the poisoned PMD which is then
appropriately handled by the generic layer code.

Signed-off-by: Punit Agrawal <punit.agrawal@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Steve Capper <steve.capper@arm.com>
---
 arch/arm64/mm/hugetlbpage.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

diff --git a/arch/arm64/mm/hugetlbpage.c b/arch/arm64/mm/hugetlbpage.c
index e25584d72396..9263f206353c 100644
--- a/arch/arm64/mm/hugetlbpage.c
+++ b/arch/arm64/mm/hugetlbpage.c
@@ -150,8 +150,17 @@ pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
        if (pud_huge(*pud))
                return (pte_t *)pud;
        pmd = pmd_offset(pud, addr);
+
+       /*
+        * In case of HW Poisoning, a hugepage pmd can contain
+        * poisoned entries. Poisoned entries are marked as swap
+        * entries.
+        *
+        * For pmds that are not present, check to see if it could be
+        * a swap entry (!present and !none) before giving up.
+        */
        if (!pmd_present(*pmd))
-               return NULL;
+               return !pmd_none(*pmd) ? (pte_t *)pmd : NULL;

        if (pte_cont(pmd_pte(*pmd))) {
                pmd = pmd_offset(
--
2.11.0

[toc] | [prev] | [next] | [standalone]

#1597997 — Re: [PATCH V2] arm64: hwpoison: add VM_FAULT_HWPOISON[_LARGE] handling

Hello Punit,

I ran the test with and without the kernel patch you're suggesting 
below. I do not see the "bad pmd ..." print that you are seeing in 
either case. All the tests are not passing for me though, it runs 42 
test cases and 14 show up as failed for some reason.

Thanks,

Tyler


On 3/9/2017 10:46 AM, Punit Agrawal wrote:
> [ +steve for arm64 mm and hugepages chops ]
>
> "Baicar, Tyler" <tbaicar@codeaurora.org> writes:
>
>> On 3/7/2017 12:56 PM, Punit Agrawal wrote:
>>> Punit Agrawal <punit.agrawal@arm.com> writes:
>>>
>>> [...]
>>>
>>>> The code looks good but I ran into some failures while running the
>>>> hugepages hwpoison tests from mce-tests suite[0]. I get a bad pmd error
>>>> in dmesg -
>>>>
>>>> [  344.165544] mm/pgtable-generic.c:33: bad pmd 000000083af00074.
>>>>
>>>> I suspect that this is due to the huge pte accessors not correctly
>>>> dealing with poisoned entries (which are represented as swap entries).
>>> I think I've got to the bottom of the issue - the problem is due to
>>> huge_pte_at() returning NULL for poisoned pmd entries (which in turn is
>>> due to pmd_present() not handling poisoned pmd entries correctly)
>>>
>>> The following is the call chain for the failure case.
>>>
>>> do_munmap
>>>     unmap_region
>>>       unmap_vmas
>>>         unmap_single_vma
>>>           __unmap_hugepage_range_final    # The test case uses hugepages
>>>             __unmap_hugepage_range
>>>               huge_pte_offset             # Returns NULL for a poisoned pmd
>>>
>>> Reverting 5bb1cc0ff9a6 ("arm64: Ensure pmd_present() returns false after
>>> pmd_mknotpresent()") fixes the problem for me but I don't think that is
>>> the right fix.
>>>
>>> While I work on a proper fix, it would be great if you can confirm that
>>> reverting 5bb1cc0ff9a6 makes the problem go away at your end.
>> Thanks Punit! I haven't got a chance to do this yet, but I will let
>> you know once I get it tested :)
> This time with a patch. Please test this instead.
>
> After a lot of head scratching, I've bit the bullet and added a check to
> return the poisoned entry from huge_pte_offset(). What with having to
> deal with contiguous hugepages et al., there just doesn't seem to be any
> leeway in how we handle the situation here.
>
> Let's see if there are any other ideas. Patch follows.
>
> Thanks,
> Punit
>
> ----------->8-------------
>  From d5ad3f428e629c80b0f93f2bbdf99b4cae28c9bc Mon Sep 17 00:00:00 2001
> From: Punit Agrawal <punit.agrawal@arm.com>
> Date: Thu, 9 Mar 2017 16:16:29 +0000
> Subject: [PATCH] arm64: hugetlb: Fix huge_pte_offset to return poisoned pmd
>
> When memory failure is enabled, a poisoned hugepage PMD is marked as a
> swap entry. As pmd_present() only checks for VALID and PROT_NONE
> bits (turned off for swap entries), it causues huge_pte_offset() to
> return NULL for poisoned PMDs.
>
> This behaviour of huge_pte_offset() leads to the error such as below
> when munmap is called on poisoned hugepages.
>
> [  344.165544] mm/pgtable-generic.c:33: bad pmd 000000083af00074.
>
> Fix huge_pte_offset() to return the poisoned PMD which is then
> appropriately handled by the generic layer code.
>
> Signed-off-by: Punit Agrawal <punit.agrawal@arm.com>
> Cc: Catalin Marinas <catalin.marinas@arm.com>
> Cc: Steve Capper <steve.capper@arm.com>
> ---
>   arch/arm64/mm/hugetlbpage.c | 11 ++++++++++-
>   1 file changed, 10 insertions(+), 1 deletion(-)
>
> diff --git a/arch/arm64/mm/hugetlbpage.c b/arch/arm64/mm/hugetlbpage.c
> index e25584d72396..9263f206353c 100644
> --- a/arch/arm64/mm/hugetlbpage.c
> +++ b/arch/arm64/mm/hugetlbpage.c
> @@ -150,8 +150,17 @@ pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
>          if (pud_huge(*pud))
>                  return (pte_t *)pud;
>          pmd = pmd_offset(pud, addr);
> +
> +       /*
> +        * In case of HW Poisoning, a hugepage pmd can contain
> +        * poisoned entries. Poisoned entries are marked as swap
> +        * entries.
> +        *
> +        * For pmds that are not present, check to see if it could be
> +        * a swap entry (!present and !none) before giving up.
> +        */
>          if (!pmd_present(*pmd))
> -               return NULL;
> +               return !pmd_none(*pmd) ? (pte_t *)pmd : NULL;
>
>          if (pte_cont(pmd_pte(*pmd))) {
>                  pmd = pmd_offset(
> --
> 2.11.0

-- 
Qualcomm Datacenter Technologies, Inc. as an affiliate of Qualcomm Technologies, Inc.
Qualcomm Technologies, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project.

[toc] | [prev] | [next] | [standalone]

#1594641

Punit Agrawal <punit.agrawal@arm.com> writes:

[...]

>
> The code looks good but I ran into some failures while running the
> hugepages hwpoison tests from mce-tests suite[0]. I get a bad pmd error
> in dmesg -
>
> [  344.165544] mm/pgtable-generic.c:33: bad pmd 000000083af00074.
>
> I suspect that this is due to the huge pte accessors not correctly
> dealing with poisoned entries (which are represented as swap entries).

I think I've got to the bottom of the issue - the problem is due to
huge_pte_at() returning NULL for poisoned pmd entries (which in turn is
due to pmd_present() not handling poisoned pmd entries correctly)

The following is the call chain for the failure case.

do_munmap
  unmap_region
    unmap_vmas
      unmap_single_vma
        __unmap_hugepage_range_final    # The test case uses hugepages
          __unmap_hugepage_range
            huge_pte_offset             # Returns NULL for a poisoned pmd

Reverting 5bb1cc0ff9a6 ("arm64: Ensure pmd_present() returns false after
pmd_mknotpresent()") fixes the problem for me but I don't think that is
the right fix.

While I work on a proper fix, it would be great if you can confirm that
reverting 5bb1cc0ff9a6 makes the problem go away at your end.

>
> I am investigating the failure but could you try running the tests at
> your end as well.
>
> To run the tests, I cloned the repository[0]. It test needs a simple fix
> at the end of this mail to run correctly. With that applied and running
> as root -
>
> # cd mce-test/cases/function/hwpoison
> # ./run_hugepage.sh
>
>
> [0] https://git.kernel.org/cgit/utils/cpu/mce/mce-test.git/
>
> --------->8--------------
> commit cb5c61f18dd86baf01b90404d4ecf51dd3d176c7
> Author: Punit Agrawal <punit.agrawal@arm.com>
> Date:   Thu Mar 2 18:24:40 2017 +0000
>
>     Use correct return type for getopt_long
>
>     getopt_long returns an int. Fix the return type to avoid issues when
>     checking for negative error codes on architectures with unsigned char,
>     e.g., arm.
>
>     Signed-off-by: Punit Agrawal <punit.agrawal@arm.com>
>
> diff --git a/cases/function/hwpoison/thugetlb.c b/cases/function/hwpoison/thugetlb.c
> index 92dc7d2..fbcf426 100644
> --- a/cases/function/hwpoison/thugetlb.c
> +++ b/cases/function/hwpoison/thugetlb.c
> @@ -125,7 +125,7 @@ int main(int argc, char *argv[])
>         int forkflag = 0;
>         int privateflag = 0;
>         int cowflag = 0;
> -   char c;
> + int c;
>         pid_t pid = 0;
>         void *expected_addr = NULL;
>         struct sembuf sembuffer;

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web