mm,hwpoison: fix sending SIGBUS for Action Required MCE
Subsystems (5)
Naoya Horiguchi
reviewerv3 cover mm,hwpoison: fix sending SIGBUS for Action Required MCE
Hi,
I updated the series based on previous feedbacks/discussions.
Changelog v2 -> v3:
- 1/3 aligned returning code with "goto out" approach,
- 3/3 added one Tested-by tag,
- 3/3 fixed comment on kill_accessing_process().
Changelog v1 -> v2:
- 3/3 fixed on initializing local variables and calculation logic of error virtual address,
v1: https://lore.kernel.org/linux-mm/[email protected]/
v2 (only 3/3 is posted): https://lore.kernel.org/linux-mm/20210419023658.GA1962954@u2004/
Thanks,
Naoya Horiguchi
--- quote from cover letter of v1 ---
I wrote this patchset to materialize what I think is the current allowable solution mentioned by the previous discussion [1]. I simply borrowed Tony's mutex patch and Aili's return code patch, then I queued another one to find error virtual address in the best effort manner. I know that this is not a perfect solution, but should work for some typical case.
[1]: https://lore.kernel.org/linux-mm/20210331192540.2141052f@alex-virtual-machine/
---
Summary:
Aili Yao (1):
mm,hwpoison: return -EHWPOISON when page already
Naoya Horiguchi (1):
mm,hwpoison: add kill_accessing_process() to find error virtual address
Tony Luck (1):
mm/memory-failure: Use a mutex to avoid memory_failure() races
arch/x86/kernel/cpu/mce/core.c | 13 ++- include/linux/swapops.h | 5 ++ mm/memory-failure.c | 181 +++++++++++++++++++++++++++++++++++++---- 3 files changed, 183 insertions(+), 16 deletions(-)
Naoya Horiguchi
reviewerPATCH v3 1/3 mm/memory-failure: Use a mutex to avoid memory_failure() races
From: Tony Luck <[email protected]>
There can be races when multiple CPUs consume poison from the same page. The first into memory_failure() atomically sets the HWPoison page flag and begins hunting for tasks that map this page. Eventually it invalidates those mappings and may send a SIGBUS to the affected tasks.
But while all that work is going on, other CPUs see a "success" return code from memory_failure() and so they believe the error has been handled and continue executing.
Fix by wrapping most of the internal parts of memory_failure() in a mutex.
--- mm/memory-failure.c | 37 ++++++++++++++++++++++++------------- 1 file changed, 24 insertions(+), 13 deletions(-)
mm/memory-failure.c +24 -13
@@ -1381,6 +1381,8 @@ static int memory_failure_dev_pagemap(unsigned long pfn, int flags,return rc; } +static DEFINE_MUTEX (mf_mutex); +/** * memory_failure - Handle memory failure of a page. * @pfn: Page Number of the corrupted page
@@ -1404,7 +1406,7 @@ int memory_failure(unsigned long pfn, int flags)struct page *hpage;struct page *orig_head;struct dev_pagemap *pgmap; -int res; +int res =0 ;unsigned long page_flags;bool retry = true;
@@ -1424,13 +1426,18 @@ int memory_failure(unsigned long pfn, int flags)return -ENXIO ; } +mutex_lock (&mf_mutex); + try_again: -if (PageHuge (p)) -return memory_failure_hugetlb (pfn, flags); +if (PageHuge (p)) { + res =memory_failure_hugetlb (pfn, flags); + goto unlock_mutex; + } +if (TestSetPageHWPoison (p)) {pr_err ("Memory failure: %#lx: already hardware poisoned\n" , pfn); -return 0 ; + goto unlock_mutex; } orig_head = hpage =compound_head (p);
@@ -1463,17 +1470,19 @@ int memory_failure(unsigned long pfn, int flags) res =MF_FAILED ; }action_result (pfn,MF_MSG_BUDDY , res); -return res ==MF_RECOVERED ?0 : -EBUSY ; + res = res ==MF_RECOVERED ?0 : -EBUSY ; }else {action_result (pfn,MF_MSG_KERNEL_HIGH_ORDER ,MF_IGNORED ); -return -EBUSY ; + res = -EBUSY ; } + goto unlock_mutex; }if (PageTransHuge (hpage)) {if (try_to_split_thp_page (p,"Memory Failure" ) <0 ) {action_result (pfn,MF_MSG_UNSPLIT_THP ,MF_IGNORED ); -return -EBUSY ; + res = -EBUSY ; + goto unlock_mutex; }VM_BUG_ON_PAGE (!page_count (p), p); }
@@ -1497,7 +1506,7 @@ int memory_failure(unsigned long pfn, int flags)if (PageCompound (p) &&compound_head (p) != orig_head) {action_result (pfn,MF_MSG_DIFFERENT_COMPOUND ,MF_IGNORED ); res = -EBUSY ; - goto out; + goto unlock_page; }/*
@@ -1517,14 +1526,14 @@ int memory_failure(unsigned long pfn, int flags)num_poisoned_pages_dec ();unlock_page (p);put_page (p); -return 0 ; + goto unlock_mutex; }if (hwpoison_filter (p)) {if (TestClearPageHWPoison (p))num_poisoned_pages_dec ();unlock_page (p);put_page (p); -return 0 ; + goto unlock_mutex; }if (!PageTransTail (p) && !PageLRU (p))
@@ -1543,7 +1552,7 @@ int memory_failure(unsigned long pfn, int flags)if (!hwpoison_user_mappings (p, pfn, flags, &p)) {action_result (pfn,MF_MSG_UNMAP_FAILED ,MF_IGNORED ); res = -EBUSY ; - goto out; + goto unlock_page; }/*
@@ -1552,13 +1561,15 @@ int memory_failure(unsigned long pfn, int flags)if (PageLRU (p) && !PageSwapCache (p) && p->mapping ==NULL ) {action_result (pfn,MF_MSG_TRUNCATED_LRU ,MF_IGNORED ); res = -EBUSY ; - goto out; + goto unlock_page; } identify_page_state: res =identify_page_state (pfn, p, page_flags); -out: +unlock_page:unlock_page (p); +unlock_mutex: +mutex_unlock (&mf_mutex);return res; }EXPORT_SYMBOL_GPL (memory_failure);
Signature · 1 line
2.25.1
Re: [PATCH v3 1/3] mm/memory-failure: Use a mutex to avoid memory_failure() races
Quoted 12 lines · Naoya Horiguchi
From: Tony Luck <[email protected]>
There can be races when multiple CPUs consume poison from the same page. The first into
memory_failure()atomically sets the HWPoison page flag and begins hunting for tasks that map this page. Eventually it invalidates those mappings and may send a SIGBUS to the affected tasks.But while all that work is going on, other CPUs see a "success" return code from
memory_failure()and so they believe the error has been handled and continue executing.
Quoted 8 lines
Fix by wrapping most of the internal parts of
memory_failure()in a mutex.Signed-off-by Tony Luck <[email protected]>Signed-off-by Naoya Horiguchi <[email protected]>--- mm/memory-failure.c | 37 ++++++++++++++++++++++++------------- 1 file changed, 24 insertions(+), 13 deletions(-)
Signature · 4 lines
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquetteNaoya Horiguchi
reviewerPATCH v3 2/3 mm,hwpoison: return -EHWPOISON when page already
In reply to Naoya HoriguchiFrom: Aili Yao <[email protected]>
When the page is already poisoned, another memory_failure() call in the same page now returns 0, meaning OK. For nested memory mce handling, this behavior may lead to one mce looping, Example:
1. When LCME is enabled, and there are two processes A && B running on different core X && Y separately, which will access one same page, then the page corrupted when process A access it, a MCE will be rasied to core X and the error process is just underway.
2. Then B access the page and trigger another MCE to core Y, it will also do error process, it will see TestSetPageHWPoison be true, and 0 is returned.
3. The kill_me_maybe will check the return:
1244 static void kill_me_maybe(struct callback_head *cb)
1245 {
...
1254 if (!memory_failure(p->mce_addr >> PAGE_SHIFT, flags) &&
1255 !(p->mce_kflags & MCE_IN_KERNEL_COPYIN)) {
1256 set_mce_nospec(p->mce_addr >> PAGE_SHIFT, p->mce_whole_page);
1257 sync_core();
1258 return;
1259 }
...
1267 }4. The error process for B will end, and may nothing happened if kill-early is not set, The process B will re-excute instruction and get into mce again and then loop happens. And also the set_mce_nospec() here is not proper, may refer to commit fd0e786d9d09 ("x86/mm, mm/hwpoison: Don't unconditionally unmap kernel 1:1 pages").
For other cases which care the return value of memory_failure() should check why they want to process a memory error which have already been processed. This behavior seems reasonable.
--- mm/memory-failure.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
mm/memory-failure.c +2 -1
@@ -1228,7 +1228,7 @@ static int memory_failure_hugetlb(unsigned long pfn, int flags)if (TestSetPageHWPoison (head)) {pr_err ("Memory failure: %#lx: already hardware poisoned\n" , pfn); -return 0 ; +return -EHWPOISON ; }num_poisoned_pages_inc ();
@@ -1437,6 +1437,7 @@ int memory_failure(unsigned long pfn, int flags)if (TestSetPageHWPoison (p)) {pr_err ("Memory failure: %#lx: already hardware poisoned\n" , pfn); + res = -EHWPOISON ; goto unlock_mutex; }
Signature · 1 line
2.25.1
Borislav Petkov
maintainerRe: [PATCH v3 2/3] mm,hwpoison: return -EHWPOISON when page already
Quoted 3 lines · Naoya Horiguchi
From: Aili Yao <[email protected]>
Subject: Re: [PATCH v3 2/3] mm,hwpoison: return
-EHWPOISONwhen page already
... Return -EHWPOISON to denote that the page has already been poisoned"
Quoted 3 lines
When the page is already poisoned, another
memory_failure()call in the same page now returns 0, meaning OK. For nested memory mce handling, this behavior may lead to one mce looping,
s/mce/MCE/g
Quoted 1 line
Example:
For example:
Quoted 2 lines
1. When LCME is enabled, and there are two processes A && B running on different core X && Y separately, which will access one same page, then
which access the same page...
s/&&/and/g
Quoted 2 lines
the page corrupted when process A access it, a MCE will be rasied to core X and the error process is just underway.
... and you lost me here. I don't understand what that is trying to say. Is that trying to say that when process A encounters the error, the MCE will be raised on CPU X?
Quoted 3 lines
2. Then B access the page and trigger another MCE to core Y, it will also do error process, it will see TestSetPageHWPoison be true, and 0 is returned.
That sentence needs massaging.
Quoted 13 lines
3. The
kill_me_maybewill check the return:1244 static void kill_me_maybe(struct callback_head *cb) 1245 { ... 1254 if (!memory_failure(p->mce_addr >> PAGE_SHIFT, flags) && 1255 !(p->mce_kflags & MCE_IN_KERNEL_COPYIN)) { 1256 set_mce_nospec(p->mce_addr >> PAGE_SHIFT, p->mce_whole_page); 1257 sync_core(); 1258 return; 1259 } ... 1267 }
No need for the line numbers.
Quoted 5 lines
4. The error process for B will end, and may nothing happened if kill-early is not set, The process B will re-excute instruction and get into mce again and then loop happens. And also the
set_mce_nospec()here is not proper, may refer to commitfd0e786d9d09("x86/mm, mm/hwpoison: Don't unconditionally unmap kernel 1:1 pages").
That needs massaging too.
Quoted 3 lines
For other cases which care the return value of
memory_failure()should check why they want to process a memory error which have already been processed. This behavior seems reasonable.
This whole commit message needs sanitizing.
Also, looking at the next patch, you can merge this one into the next because the next one is acting on -EHWPOISON so it all belongs together in a single patch.
Thx.
Signature · 4 lines
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquetteRe: [PATCH v3 2/3] mm,hwpoison: return -EHWPOISON when page already
Quoted 3 lines · Borislav Petkov
Quoted 1 line · Naoya Horiguchi
From: Aili Yao <[email protected]>
...
Quoted 5 lines
This whole commit message needs sanitizing.
Also, looking at the next patch, you can merge this one into the next because the next one is acting on
-EHWPOISONso it all belongs together in a single patch.
OK, I'll fold this into 3/3.
Thanks,
Naoya Horiguchi
Naoya Horiguchi
reviewerPATCH v3 3/3 mm,hwpoison: add kill_accessing_process() to find error virtual address
In reply to Naoya HoriguchiFrom: Naoya Horiguchi <[email protected]>
The previous patch solves the infinite MCE loop issue when multiple MCE events races. The remaining issue is to make sure that all threads processing Action Required MCEs send to the current processes the SIGBUS with the proper virtual address and the error size.
This patch suggests to do page table walk to find the error virtual address. If we find multiple virtual addresses in walking, we now can't determine which one is correct, so we fall back to sending SIGBUS in kill_me_maybe() without error info as we do now. This corner case needs to be solved in the future.
---
change log v1 -> v2:
- initialize local variables in check_hwpoisoned_entry() and hwpoison_pte_range()
- fix and improve logic to calculate error address offset.
--- arch/x86/kernel/cpu/mce/core.c | 13 ++- include/linux/swapops.h | 5 ++ mm/memory-failure.c | 143 ++++++++++++++++++++++++++++++++- 3 files changed, 158 insertions(+), 3 deletions(-)
arch/x86/kernel/cpu/mce/core.c +11 -2
@@ -1257,19 +1257,28 @@ static void kill_me_maybe(struct callback_head *cb) { struct task_struct *p = container_of(cb, struct task_struct, mce_kill_me);int flags =MF_ACTION_REQUIRED ; +int ret;pr_err ("Uncorrected hardware memory error in user-access at %llx" , p->mce_addr);if (!p->mce_ripv) flags |=MF_MUST_KILL ; -if (!memory_failure (p->mce_addr >>PAGE_SHIFT , flags) && - !(p->mce_kflags &MCE_IN_KERNEL_COPYIN )) { + ret =memory_failure (p->mce_addr >>PAGE_SHIFT , flags); +if (!ret && !(p->mce_kflags &MCE_IN_KERNEL_COPYIN )) {set_mce_nospec (p->mce_addr >>PAGE_SHIFT , p->mce_whole_page);sync_core ();return ; } +/* +* -EHWPOISON from memory_failure() means that it already sent SIGBUS +* to the current process with the proper error info, so no need to +* send it here again. +*/ +if (ret == -EHWPOISON ) +return ; +if (p->mce_vaddr != (void __user *)-1l ) {force_sig_mceerr (BUS_MCEERR_AR , p->mce_vaddr,PAGE_SHIFT ); }else {
include/linux/swapops.h +5 -0
@@ -323,6 +323,11 @@ static inline int is_hwpoison_entry(swp_entry_t entry)return swp_type (entry) ==SWP_HWPOISON ; } +static inline unsigned long hwpoison_entry_to_pfn (swp_entry_t entry) +{ +return swp_offset (entry); +} +static inline void num_poisoned_pages_inc (void ) {atomic_long_inc (&num_poisoned_pages);
mm/memory-failure.c +142 -1
167 lines not on this page · raw message
Signature · 1 line
2.25.1
Borislav Petkov
maintainerRe: [PATCH v3 3/3] mm,hwpoison: add kill_accessing_process() to find error virtual address
Quoted 3 lines · Naoya Horiguchi
From: Naoya Horiguchi <[email protected]>
The previous patch solves the infinite MCE loop issue when multiple
"previous patch" has no meaning when it is in git.
Quoted 1 line
MCE events races. The remaining issue is to make sure that all threads
"race."
Quoted 1 line
processing Action Required MCEs send to the current processes the
s/the //
Quoted 3 lines
SIGBUS with the proper virtual address and the error size.
This patch suggests to do page table walk to find the error virtual
Avoid having "This patch" or "This commit" in the commit message. It is tautologically useless.
Also, do
$ git grep 'This patch' Documentation/process
for more details.
Quoted 1 line
address. If we find multiple virtual addresses in walking, we now can't
Who's "we"? during the pagetable walk
Quoted 3 lines
determine which one is correct, so we fall back to sending SIGBUS in
kill_me_maybe()without error info as we do now. This corner case needs to be solved in the future.
Solved how? If you can't map which error comes from which process, you can't do anything here. You could send SIGBUS to all but you might injure some innocent bystanders this way.
Just code structuring suggestions below - mm stuff is for someone else to review properly.
mm/memory-failure.c static void collect_procs(struct page *page, struct list_head *tokill, · line 603
Quoted 10 lines
mm/memory-failure.c +10
-0+static int hwpoison_pte_range (pmd_t *pmdp,unsigned long addr, +unsigned long end,struct mm_walk *walk) +{ +struct hwp_walk *hwp = (struct hwp_walk *)walk->private; +int ret =0 ; +pte_t *ptep; +spinlock_t *ptl; + + ptl =pmd_trans_huge_lock (pmdp, walk->vma); +if (ptl) {
Save yourself an indentation level:
if (!ptl) goto unlock;
mm/memory-failure.c static void collect_procs(struct page *page, struct list_head *tokill, · line 606
Quoted 3 lines
mm/memory-failure.c +3
-0+pmd_t pmd = *pmdp; + +if (pmd_present (pmd)) {
... ditto...
mm/memory-failure.c static void collect_procs(struct page *page, struct list_head *tokill, · line 611
Quoted 5 lines
mm/memory-failure.c +5
-0+unsigned long pfn =pmd_pfn (pmd); + +if (pfn <= hwp->pfn && hwp->pfn < pfn +HPAGE_PMD_NR ) { +unsigned long hwpoison_vaddr = addr + + ((hwp->pfn - pfn) <<PAGE_SHIFT );
... which will allow you to not break those.
Quoted 24 lines
mm/memory-failure.c +24
-0+ + ret =set_to_kill (&hwp->tk, hwpoison_vaddr, +PAGE_SHIFT ); + } + } +spin_unlock (ptl); + goto out; + } + +if (pmd_trans_unstable (pmdp)) + goto out; + + ptep =pte_offset_map_lock (walk->vma->vm_mm, pmdp, addr, &ptl); +for (; addr != end; ptep++, addr +=PAGE_SIZE ) { + ret =check_hwpoisoned_entry (*ptep, addr,PAGE_SHIFT , + hwp->pfn, &hwp->tk); +if (ret ==1 ) +break ; + } +pte_unmap_unlock (ptep -1 , ptl); +out: +cond_resched (); +return ret; +}
Signature · 4 lines
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquetteRe: [PATCH v3 3/3] mm,hwpoison: add kill_accessing_process() to find error virtual address
Quoted 7 lines · Borislav Petkov
Quoted 3 lines · Naoya Horiguchi
From: Naoya Horiguchi <[email protected]>
The previous patch solves the infinite MCE loop issue when multiple
"previous patch" has no meaning when it is in git.
Quoted 7 lines
Quoted 1 line
MCE events races. The remaining issue is to make sure that all threads
"race."Quoted 1 line
processing Action Required MCEs send to the current processes the
s/the //
I'll fix these grammar errors.
Quoted 5 lines
Quoted 4 lines
SIGBUS with the proper virtual address and the error size.
This patch suggests to do page table walk to find the error virtual
Quoted 8 lines
Avoid having "This patch" or "This commit" in the commit message. It is tautologically useless.
Also, do
$ git grep 'This patch' Documentation/processfor more details.
I didn't know the following rule:
Describe your changes in imperative mood, e.g. "make xyzzy do frotz" instead of "[This patch] makes xyzzy do frotz" or "[I] changed xyzzy to do frotz", as if you are giving orders to the codebase to change its behaviour.
I'll follow this in my future post.
Quoted 4 lines
Quoted 2 lines
address. If we find multiple virtual addresses in walking, we now can't
Who's "we"? during the pagetable walk
I wrongly abused rhetorical "we". I'll change this sentence in passive form.
Quoted 6 lines
Quoted 4 lines
determine which one is correct, so we fall back to sending SIGBUS in
kill_me_maybe()without error info as we do now. This corner case needs to be solved in the future.Solved how?
I don't know exactly. MCE subsystem seems to have code extracting linear address, so I wonder that that could be used as a hint to memory_failure() to find the proper virtual address.
Quoted 3 lines
If you can't map which error comes from which process, you can't do anything here. You could send SIGBUS to all but you might injure some innocent bystanders this way.
The situation in question is caused by action required MCE, so we know which process we should send SIGBUS to. So if we choose to send SIGBUS to all, no innocent bystanders would be affected. But when the process have multiple virtual addresses associated with the error physical address, the process receives multiple SIGBUSs and all but one have wrong value in si_addr in siginfo_t, so that's confusing.
Quoted 3 lines
Just code structuring suggestions below - mm stuff is for someone else to review properly.
Thank you, I'll update with them.
- Naoya Horiguchi
Quoted 63 lines
mm/memory-failure.c static void collect_procs(struct page *page, struct list_head *tokill, · line 603
Quoted 11 lines
mm/memory-failure.c +10
-0+static int hwpoison_pte_range (pmd_t *pmdp,unsigned long addr, +unsigned long end,struct mm_walk *walk) +{ +struct hwp_walk *hwp = (struct hwp_walk *)walk->private; +int ret =0 ; +pte_t *ptep; +spinlock_t *ptl; + + ptl =pmd_trans_huge_lock (pmdp, walk->vma); +if (ptl) {Save yourself an indentation level:
if (!ptl) goto unlock;mm/memory-failure.c static void collect_procs(struct page *page, struct list_head *tokill, · line 606
Quoted 3 lines
mm/memory-failure.c +3
-0+pmd_t pmd = *pmdp; + +if (pmd_present (pmd)) {... ditto...
mm/memory-failure.c static void collect_procs(struct page *page, struct list_head *tokill, · line 611
Quoted 5 lines
mm/memory-failure.c +5
-0+unsigned long pfn =pmd_pfn (pmd); + +if (pfn <= hwp->pfn && hwp->pfn < pfn +HPAGE_PMD_NR ) { +unsigned long hwpoison_vaddr = addr + + ((hwp->pfn - pfn) <<PAGE_SHIFT );... which will allow you to not break those.
Quoted 24 lines
mm/memory-failure.c +24
-0+ + ret =set_to_kill (&hwp->tk, hwpoison_vaddr, +PAGE_SHIFT ); + } + } +spin_unlock (ptl); + goto out; + } + +if (pmd_trans_unstable (pmdp)) + goto out; + + ptep =pte_offset_map_lock (walk->vma->vm_mm, pmdp, addr, &ptl); +for (; addr != end; ptep++, addr +=PAGE_SIZE ) { + ret =check_hwpoisoned_entry (*ptep, addr,PAGE_SHIFT , + hwp->pfn, &hwp->tk); +if (ret ==1 ) +break ; + } +pte_unmap_unlock (ptep -1 , ptl); +out: +cond_resched (); +return ret; +}Signature · 4 lines
Regards/Gruss, Boris. https://people.kernel.org/tglx/notes-about-netiquette
Borislav Petkov
maintainerRe: [PATCH v3 3/3] mm,hwpoison: add kill_accessing_process() to find error virtual address
Quoted 3 lines · HORIGUCHI NAOYA(堀口 直也)
I don't know exactly. MCE subsystem seems to have code extracting linear address, so I wonder that that could be used as a hint to
memory_failure()to find the proper virtual address.
See "Table 15-3. Address Mode in IA32_MCi_MISC[8:6]" in the SDM - apparently it can report all kinds of address types, depending on the hw incarnation or MCA bank type or whatnot. Tony knows :)
Quoted 7 lines
The situation in question is caused by action required MCE, so we know which process we should send SIGBUS to. So if we choose to send SIGBUS to all, no innocent bystanders would be affected. But when the process have multiple virtual addresses associated with the error physical address, the process receives multiple SIGBUSs and all but one have wrong value in
si_addrinsiginfo_t, so that's confusing.
Is that scenario real or hypothetical?
Because I'd expect that if we send it a SIGBUS and we poison that page, then all the VAs mapping it will have to handle the situation that that page has been poisoned and pulled from under them.
So from a hw perspective, there won't be any more accesses to the faulty physical page.
In a perfect world, that is...
Signature · 4 lines
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquetteRe: [PATCH v3 3/3] mm,hwpoison: add kill_accessing_process() to find error virtual address
Quoted 8 lines · Borislav Petkov
Quoted 3 lines · HORIGUCHI NAOYA(堀口 直也)
I don't know exactly. MCE subsystem seems to have code extracting linear address, so I wonder that that could be used as a hint to
memory_failure()to find the proper virtual address.See "Table 15-3. Address Mode in IA32_MCi_MISC[8:6]" in the SDM - apparently it can report all kinds of address types, depending on the hw incarnation or MCA bank type or whatnot. Tony knows :)
"15.9.3.2 Architecturally Defined SRAR Errors" says that the register is supposed to have physical address.
For both the data load and instruction fetch errors, the ADDRV and MISCV flags in the IA32_MCi_STATUS register are set to indicate that the offending physical address information is available from the IA32_MCi_MISC and the IA32_MCi_ADDR registers.
Quoted 8 lines
Quoted 7 lines
The situation in question is caused by action required MCE, so we know which process we should send SIGBUS to. So if we choose to send SIGBUS to all, no innocent bystanders would be affected. But when the process have multiple virtual addresses associated with the error physical address, the process receives multiple SIGBUSs and all but one have wrong value in
si_addrinsiginfo_t, so that's confusing.
Quoted 5 lines
Is that scenario real or hypothetical?
Because I'd expect that if we send it a SIGBUS and we poison that page, then all the VAs mapping it will have to handle the situation that that page has been poisoned and pulled from under them.
IIUC, the above should be done by the first MCE handling. In "already hwpoisoned" case, the page has already been poisoned and all mapping for it should be already unmapped, then what we need additionally is to send SIGBUS to report to the application that it should take some action or abort immediately.
Thanks,
Naoya Horiguchi
Quoted 1 line
I updated the series based on previous feedbacks/discussions.
-Tony
Luck, Tony
maintainerQuoted 3 lines
Quoted 1 line
I updated the series based on previous feedbacks/discussions.
Tested-by Tony Luck <[email protected]>
Maybe this series should get a "Cc: stable" tag too?
I don't have a specific commit to provide for Fixes:, but it deals with an issue where sometimes processes return from the machine check handler and hit the same machine check again.
It's hard[1] to hit in v5.8, but somehow gets progressively easier in newer versions.
-Tony
[1] Initially I thought it didn't happen at all in v5.8. Then wasted a bunch of time trying to bisect. But my "git bisect good" responses turned out to be randomly assigned when I just hadn't tried enough times to make the problem show up.
Quoted 5 lines · Tony
Quoted 3 lines
Quoted 1 line
I updated the series based on previous feedbacks/discussions.
Tested-by Tony Luck <[email protected]>Maybe this series should get a "Cc: stable" tag too?
The second patch (I squashed 2/3 and 3/3 into one in v4) is a little too large (more than 100 lines of change), which doesn't meet stable rule. But the first patch looks suitable to me for stable.
Quoted 4 lines
I don't have a specific commit to provide for Fixes:, but it deals with an issue where sometimes processes return from the machine check handler and hit the same machine check again.
If we consider that this issue is visible after LMCE is supported, the following commit could be the target of Fixes tag.
commit 243d657eaf540db882f73497060da5a4f7d86a90
Author: Ashok Raj <[email protected]>
Date: Thu Jun 4 18:55:24 2015 +0200
x86/mce: Handle Local MCE events
but according to your test result, any other factor might affect when the issue actually got reproducible.
Quoted 10 lines
It's hard[1] to hit in v5.8, but somehow gets progressively easier in newer versions.
-Tony
[1] Initially I thought it didn't happen at all in v5.8. Then wasted a bunch of time trying to bisect. But my "git bisect good" responses turned out to be randomly assigned when I just hadn't tried enough times to make the problem show up.
Thanks for time and effort for testing.
- Naoya
Quoted 10 lines · On Fri, 7 May 2021 05:24:22 +0000 HORIGUCHI NAOYA(堀口 直也)…
Quoted 5 lines · Tony
Quoted 3 lines
Quoted 1 line
I updated the series based on previous feedbacks/discussions.
Tested-by Tony Luck <[email protected]>Maybe this series should get a "Cc: stable" tag too?
The second patch (I squashed 2/3 and 3/3 into one in v4) is a little too large (more than 100 lines of change), which doesn't meet stable rule. But the first patch looks suitable to me for stable.
Before cc:stable, would you please do one stress test first?
It failed in my server, but I didn't dig into it, maybe the fail is meaningless. Just a small suggestion.
Thanks!
Aili Yao
Luck, Tony
maintainerQuoted 3 lines
Before cc:stable, would you please do one stress test first?
It failed in my server, but I didn't dig into it, maybe the fail is meaningless. Just a small suggestion.
Upstream plus these three patches passed an overnight 8-hour stress test on four machines here (running a test that's been failing with hung kernels and kernel crashes w/o these patches).
What were the symptoms of your failed server?
-Tony
Quoted 10 lines · Tony
Quoted 3 lines
Before cc:stable, would you please do one stress test first?
It failed in my server, but I didn't dig into it, maybe the fail is meaningless. Just a small suggestion.Upstream plus these three patches passed an overnight 8-hour stress test on four machines here (running a test that's been failing with hung kernels and kernel crashes w/o these patches).
What were the symptoms of your failed server?
Sorry, I am not sure if the stress test in my server is same with yours. Usually, I will do one RAS stress test from mce-test/cases/stress before post something. Maybe this test is not general useful or you have tested it and passed or you may think the test is not proper though or you tested and you don't think it's one real issue, then just ignore it please.
Thanks!
Aili Yao
Luck, Tony
maintainerQuoted 1 line
Sorry, I am not sure if the stress test in my server is same with yours.
It's better if you test different things ... that's how we get better coverage.
Quoted 4 lines
Usually, I will do one RAS stress test from mce-test/cases/stress before post something. Maybe this test is not general useful or you have tested it and passed or you may think the test is not proper though or you tested and you don't think it's one real issue, then just ignore it please.
I haven't re-run the mce-test cases for a while. The test that was run on these patches involves a multi-threaded application running on all logical CPUs. When poison is injected many of the CPUs can hit the poison close together. So it's a good stress test for the new series.
-Tony