From: Shuai Xue <xueshuai@linux.alibaba.com>
To: James Morse <james.morse@arm.com>,
rafael@kernel.org, wangkefeng.wang@huawei.com,
tanxiaofei@huawei.com, mawupeng1@huawei.com, tony.luck@intel.com,
linmiaohe@huawei.com, naoya.horiguchi@nec.com,
gregkh@linuxfoundation.org, will@kernel.org, jarkko@kernel.org
Cc: linux-acpi@vger.kernel.org, linux-mm@kvack.org,
linux-kernel@vger.kernel.org, akpm@linux-foundation.org,
linux-edac@vger.kernel.org,
acpica-devel@lists.linuxfoundation.org, stable@vger.kernel.org,
x86@kernel.org, justin.he@arm.com, ardb@kernel.org,
ying.huang@intel.com, ashish.kalra@amd.com,
baolin.wang@linux.alibaba.com, bp@alien8.de, tglx@linutronix.de,
mingo@redhat.com, dave.hansen@linux.intel.com, lenb@kernel.org,
hpa@zytor.com, robert.moore@intel.com, lvying6@huawei.com,
xiexiuqi@huawei.com, zhuo.song@linux.alibaba.com
Subject: Re: [PATCH v9 2/2] ACPI: APEI: handle synchronous exceptions in task work
Date: Fri, 1 Dec 2023 15:03:28 +0800 [thread overview]
Message-ID: <b56fbae2-0d9b-4c42-94bf-7fd58b3fd738@linux.alibaba.com> (raw)
In-Reply-To: <874f0170-a829-47db-8882-52b9ed8e869d@arm.com>
On 2023/12/1 01:39, James Morse wrote:
> Hi Shuai,
>
> On 07/10/2023 08:28, Shuai Xue wrote:
>> Hardware errors could be signaled by synchronous interrupt,
>
> I'm struggling with 'synchronous interrupt'. Do you mean arm64's 'precise' (all
> instructions before the exception were executed, and none after).
> Otherwise, surely any interrupt from a background scrubber is inherently asynchronous!
>
I am sorry, this is typo. I mean asynchronous interrupt.
>
>> e.g. when an
>> error is detected by a background scrubber, or signaled by synchronous
>> exception, e.g. when an uncorrected error is consumed. Both synchronous and
>> asynchronous error are queued and handled by a dedicated kthread in
>> workqueue.
>>
>> commit 7f17b4a121d0 ("ACPI: APEI: Kick the memory_failure() queue for
>> synchronous errors") keep track of whether memory_failure() work was
>> queued, and make task_work pending to flush out the workqueue so that the
>> work for synchronous error is processed before returning to user-space.
>
> It does it regardless, if user-space was interrupted by APEI any work queued as a result
> of that should be completed before we go back to user-space. Otherwise we can bounce
> between user-space and firmware, with the kernel only running the APEI code, and never
> making progress.
>
Agreed.
>
>> The trick ensures that the corrupted page is unmapped and poisoned. And
>> after returning to user-space, the task starts at current instruction which
>> triggering a page fault in which kernel will send SIGBUS to current process
>> due to VM_FAULT_HWPOISON.
>>
>> However, the memory failure recovery for hwpoison-aware mechanisms does not
>> work as expected. For example, hwpoison-aware user-space processes like
>> QEMU register their customized SIGBUS handler and enable early kill mode by
>> seting PF_MCE_EARLY at initialization. Then the kernel will directly notify
>
> (setting, directly)
Thank you. Will fix it.
>
>> the process by sending a SIGBUS signal in memory failure with wrong
>
>> si_code: the actual user-space process accessing the corrupt memory
>> location, but its memory failure work is handled in a kthread context, so
>> it will send SIGBUS with BUS_MCEERR_AO si_code to the actual user-space
>> process instead of BUS_MCEERR_AR in kill_proc().
>
> This is hard to parse, "the user-space process is accessing"? (dropping 'actual' and
> adding 'is')
Will fix it.
>
>
> Wasn't this behaviour fixed by the previous patch?
>
> What problem are you fixing here?
Nope. The memory_failure() runs in a kthread context, but not the
user-space process which consuming poison data.
// kill_proc() in memory-failure.c
if ((flags & MF_ACTION_REQUIRED) && (t == current))
ret = force_sig_mceerr(BUS_MCEERR_AR,
(void __user *)tk->addr, addr_lsb);
else
ret = send_sig_mceerr(BUS_MCEERR_AO, (void __user *)tk->addr,
addr_lsb, t);
So, even we queue memory_failure() with MF_ACTION_REQUIRED flags in
previous patch, it will still send a sigbus with BUS_MCEERR_AO in the else
branch of kill_proc().
>
>
>> To this end, separate synchronous and asynchronous error handling into
>> different paths like X86 platform does:
>>
>> - valid synchronous errors: queue a task_work to synchronously send SIGBUS
>> before ret_to_user.
>
>> - valid asynchronous errors: queue a work into workqueue to asynchronously
>> handle memory failure.
>
> Why? The signal issue was fixed by the previous patch. Why delay the handling of a
> poisoned memory location further?
The signal issue is not fixed completely. See my reply above.
>
>
>> - abnormal branches such as invalid PA, unexpected severity, no memory
>> failure config support, invalid GUID section, OOM, etc.
>
> ... do what?
If no memory failure work is queued for abnormal errors, do a force kill.
Will also add this comment to commit log.
>
>
>> Then for valid synchronous errors, the current context in memory failure is
>> exactly belongs to the task consuming poison data and it will send SIBBUS
>> with proper si_code.
>
>
>> diff --git a/arch/x86/kernel/cpu/mce/core.c b/arch/x86/kernel/cpu/mce/core.c
>> index 6f35f724cc14..1675ff77033d 100644
>> --- a/arch/x86/kernel/cpu/mce/core.c
>> +++ b/arch/x86/kernel/cpu/mce/core.c
>> @@ -1334,17 +1334,10 @@ static void kill_me_maybe(struct callback_head *cb)
>> return;
>> }
>>
>> - /*
>> - * -EHWPOISON from memory_failure() means that it already sent SIGBUS
>> - * to the current process with the proper error info,
>> - * -EOPNOTSUPP means hwpoison_filter() filtered the error event,
>> - *
>> - * In both cases, no further processing is required.
>> - */
>> if (ret == -EHWPOISON || ret == -EOPNOTSUPP)
>> return;
>>
>> - pr_err("Memory error not recovered");
>> + pr_err("Sending SIGBUS to current task due to memory error not recovered");
>> kill_me_now(cb);
>> }
>>
>
> I'm not sure how this hunk is relevant to the commit message.
I handle memory_failure() error code in its arm64 call site
memory_failure_cb() with some comments, similar to x86 call site
kill_me_maybe(). I moved these two part comments to function declaration,
followed by review comments from Kefeng.
I should split this into a separate patch. Will do it in next version.
>
>
>> diff --git a/drivers/acpi/apei/ghes.c b/drivers/acpi/apei/ghes.c
>> index 88178aa6222d..014401a65ed5 100644
>> --- a/drivers/acpi/apei/ghes.c
>> +++ b/drivers/acpi/apei/ghes.c
>> @@ -484,6 +497,18 @@ static bool ghes_do_memory_failure(u64 physical_addr, int flags)
>> return false;
>> }
>>
>> + if (flags == MF_ACTION_REQUIRED && current->mm) {
>> + twcb = kmalloc(sizeof(*twcb), GFP_ATOMIC);
>> + if (!twcb)
>> + return false;
>
> Yuck - New failure modes! This is why the existing code always has this memory allocated
> in struct ghes_estatus_node.
Are you suggesting to move fields of struct sync_task_work to struct
ghes_estatus_node, and use ghes_estatus_node here? Or we can just alloc
struct sync_task_work with gen_pool_alloc from ghes_estatus_pool.
>
>
>> + twcb->pfn = pfn;
>> + twcb->flags = flags;
>> + init_task_work(&twcb->twork, memory_failure_cb);
>> + task_work_add(current, &twcb->twork, TWA_RESUME);
>> + return true;
>> + }
>> +
>> memory_failure_queue(pfn, flags);
>> return true;
>> }
>
> [..]
>
>> @@ -696,7 +721,14 @@ static bool ghes_do_proc(struct ghes *ghes,
>> }
>> }
>>
>> - return queued;
>> + /*
>> + * If no memory failure work is queued for abnormal synchronous
>> + * errors, do a force kill.
>> + */
>> + if (sync && !queued) {
>> + pr_err("Sending SIGBUS to current task due to memory error not recovered");
>> + force_sig(SIGBUS);
>> + }
>> }
>
> I think this is a lot of churn, and this hunk is the the only meaningful change in
> behaviour. Can you explain how this happens?
For example:
- invalid GUID section in ghes_do_proc()
- CPER_MEM_VALID_PA is not set, unexpected severity in
ghes_handle_memory_failure().
- CONFIG_ACPI_APEI_MEMORY_FAILURE is not enabled, !pfn_vaild(pfn) in
ghes_do_memory_failure()
>
>
> Wouldn't it be simpler to split ghes_kick_task_work() to have a sync/async version.
> The synchronous version can unconditionally force_sig_mceerr(BUS_MCEERR_AR, ...) after
> memory_failure_queue_kick() - but that still means memory_failure() is unable to disappear
> errors that it fixed - see MF_RECOVERED.
Sorry, I don't think so. Unconditionally send a sigbus is not a good
choice. For example, if a sync memory error detected in instruction memory
error, the kernel should transparently fix and no signal should be send.
./einj_mem_uc instr
[168522.751671] Memory failure: 0x89dedd: corrupted page was clean: dropped without side effects
[168522.751679] Memory failure: 0x89dedd: recovery action for clean LRU page: Recovered
With this patch set, the instr case behaves consistently on both the arm64 and x86 platforms.
The complex page error_states are handled in memory_failure(). IMHO, we
should left this part to it.
>
>> diff --git a/mm/memory-failure.c b/mm/memory-failure.c
>> index 4d6e43c88489..0d02f8a0b556 100644
>> --- a/mm/memory-failure.c
>> +++ b/mm/memory-failure.c
>> @@ -2161,9 +2161,12 @@ static int memory_failure_dev_pagemap(unsigned long pfn, int flags,
>> * Must run in process context (e.g. a work queue) with interrupts
>> * enabled and no spinlocks held.
>> *
>> - * Return: 0 for successfully handled the memory error,
>> - * -EOPNOTSUPP for hwpoison_filter() filtered the error event,
>> - * < 0(except -EOPNOTSUPP) on failure.
>> + * Return values:
>> + * 0 - success
>> + * -EOPNOTSUPP - hwpoison_filter() filtered the error event.
>> + * -EHWPOISON - sent SIGBUS to the current process with the proper
>> + * error info by kill_accessing_process().
>> + * other negative values - failure
>> */
>> int memory_failure(unsigned long pfn, int flags)
>> {
>
> I'm not sure how this hunk is relevant to the commit message.
As mentioned, I will split this into a separate patch.
>
>
> Thanks,
>
> James
Thank you for valuable comments.
Best Regards,
Shuai
next prev parent reply other threads:[~2023-12-01 7:03 UTC|newest]
Thread overview: 121+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-10-27 4:24 [PATCH] ACPI: APEI: set memory failure flags as MF_ACTION_REQUIRED on action required events Shuai Xue
2022-10-28 17:08 ` Rafael J. Wysocki
2022-10-28 17:25 ` Luck, Tony
2022-11-02 11:53 ` Shuai Xue
2022-11-22 11:40 ` Shuai Xue
2022-11-02 7:07 ` Shuai Xue
2022-12-06 15:33 ` [RFC PATCH 0/2] ACPI: APEI: handle synchronous exceptions in task work Shuai Xue
2022-12-07 9:54 ` reply for " Lv Ying
2022-12-07 12:34 ` Bixuan Cui
2022-12-07 12:56 ` Shuai Xue
2022-12-07 14:04 ` Shuai Xue
2022-12-08 2:27 ` Lv Ying
2022-12-06 15:33 ` [RFC PATCH 1/2] ACPI: APEI: set memory failure flags as MF_ACTION_REQUIRED on synchronous events Shuai Xue
2022-12-06 15:33 ` [RFC PATCH 2/2] ACPI: APEI: separate synchronous error handling into task work Shuai Xue
2023-02-27 5:03 ` [PATCH v2 0/2] ACPI: APEI: handle synchronous exceptions with proper si_code Shuai Xue
2023-03-06 0:45 ` Shuai Xue
2023-02-27 5:03 ` [PATCH v2 1/2] ACPI: APEI: set memory failure flags as MF_ACTION_REQUIRED on synchronous events Shuai Xue
2023-03-16 7:21 ` HORIGUCHI NAOYA(堀口 直也)
2023-03-16 9:57 ` Shuai Xue
2023-03-16 16:45 ` Luck, Tony
2023-03-17 1:12 ` Shuai Xue
2023-02-27 5:03 ` [PATCH v2 2/2] ACPI: APEI: handle synchronous exceptions in task work Shuai Xue
2023-03-16 7:21 ` HORIGUCHI NAOYA(堀口 直也)
2023-03-16 11:10 ` Shuai Xue
2023-03-17 0:29 ` HORIGUCHI NAOYA(堀口 直也)
2023-03-17 1:24 ` Shuai Xue
2023-03-17 7:24 ` [PATCH v3 0/2] ACPI: APEI: handle synchronous exceptions with proper si_code Shuai Xue
2023-03-20 18:03 ` Rafael J. Wysocki
2023-03-30 6:11 ` Shuai Xue
2023-03-30 9:52 ` Rafael J. Wysocki
2023-03-21 7:17 ` mawupeng
2023-03-22 1:27 ` Shuai Xue
2023-03-17 7:24 ` [PATCH v3 1/2] ACPI: APEI: set memory failure flags as MF_ACTION_REQUIRED on synchronous events Shuai Xue
2023-03-17 7:24 ` [PATCH v3 2/2] ACPI: APEI: handle synchronous exceptions in task work Shuai Xue
2023-04-06 12:39 ` Xiaofei Tan
2023-04-07 2:21 ` Shuai Xue
2023-04-08 9:13 ` [PATCH v4 0/2] ACPI: APEI: handle synchronous exceptions with proper si_code Shuai Xue
2023-04-08 9:13 ` [PATCH v4 1/2] ACPI: APEI: set memory failure flags as MF_ACTION_REQUIRED on synchronous events Shuai Xue
2023-04-08 9:13 ` [PATCH v4 2/2] ACPI: APEI: handle synchronous exceptions in task work Shuai Xue
2023-04-11 1:44 ` Xiaofei Tan
2023-04-11 3:16 ` Shuai Xue
2023-04-11 9:02 ` Xiaofei Tan
2023-04-11 9:48 ` Shuai Xue
2023-04-11 10:48 ` [PATCH v5 0/2] ACPI: APEI: handle synchronous exceptions with proper si_code Shuai Xue
2023-04-11 10:48 ` [PATCH v5 1/2] ACPI: APEI: set memory failure flags as MF_ACTION_REQUIRED on synchronous events Shuai Xue
2023-04-11 14:17 ` Kefeng Wang
2023-04-12 2:54 ` Shuai Xue
2023-04-12 3:55 ` Xiaofei Tan
2023-04-13 1:42 ` Shuai Xue
2023-04-11 10:48 ` [PATCH v5 2/2] ACPI: APEI: handle synchronous exceptions in task work Shuai Xue
2023-04-11 14:28 ` Kefeng Wang
2023-04-12 2:58 ` Shuai Xue
2023-04-12 4:05 ` Xiaofei Tan
2023-04-13 1:49 ` Shuai Xue
2023-04-12 11:27 ` [PATCH v6 0/2] ACPI: APEI: handle synchronous exceptions with proper si_code Shuai Xue
2023-04-12 11:28 ` [PATCH v6 1/2] ACPI: APEI: set memory failure flags as MF_ACTION_REQUIRED on synchronous events Shuai Xue
2023-04-12 11:28 ` [PATCH v6 2/2] ACPI: APEI: handle synchronous exceptions in task work Shuai Xue
2023-04-17 1:14 ` [PATCH v7 0/2] ACPI: APEI: handle synchronous exceptions with proper si_code Shuai Xue
2023-04-24 6:24 ` Shuai Xue
2023-05-08 1:55 ` Shuai Xue
2023-04-17 1:14 ` [PATCH v7 1/2] ACPI: APEI: set memory failure flags as MF_ACTION_REQUIRED on synchronous events Shuai Xue
2023-04-17 1:14 ` [PATCH v7 2/2] ACPI: APEI: handle synchronous exceptions in task work Shuai Xue
2023-09-19 2:21 ` [RESEND PATCH v8 0/2] ACPI: APEI: handle synchronous errors in task work with proper si_code Shuai Xue
2023-09-19 2:21 ` [RESEND PATCH v8 1/2] ACPI: APEI: set memory failure flags as MF_ACTION_REQUIRED on synchronous events Shuai Xue
2023-09-25 14:43 ` Jarkko Sakkinen
2023-09-26 6:23 ` Shuai Xue
2023-09-19 2:21 ` [RESEND PATCH v8 2/2] ACPI: APEI: handle synchronous exceptions in task work Shuai Xue
2023-09-25 15:00 ` Jarkko Sakkinen
2023-09-26 6:38 ` Shuai Xue
2023-10-03 8:28 ` Naoya Horiguchi
2023-10-07 2:01 ` Shuai Xue
2023-10-07 7:28 ` [PATCH v9 0/2] ACPI: APEI: handle synchronous errors in task work with proper si_code Shuai Xue
2023-11-21 1:48 ` Shuai Xue
2023-11-23 15:07 ` Borislav Petkov
2023-11-25 6:44 ` Shuai Xue
2023-11-25 12:10 ` Borislav Petkov
2023-11-26 12:25 ` Shuai Xue
2023-11-29 18:54 ` Borislav Petkov
2023-11-30 2:58 ` Shuai Xue
2023-11-30 14:40 ` Borislav Petkov
2023-11-30 17:43 ` James Morse
2023-12-01 2:58 ` Shuai Xue
2023-11-30 17:39 ` James Morse
2023-12-01 3:37 ` Shuai Xue
2023-10-07 7:28 ` [PATCH v9 1/2] ACPI: APEI: set memory failure flags as MF_ACTION_REQUIRED on synchronous events Shuai Xue
2023-11-30 17:39 ` James Morse
2023-12-01 5:22 ` Shuai Xue
2023-10-07 7:28 ` [PATCH v9 2/2] ACPI: APEI: handle synchronous exceptions in task work Shuai Xue
2023-11-30 17:39 ` James Morse
2023-12-01 7:03 ` Shuai Xue [this message]
2023-12-18 6:45 ` [PATCH v10 0/4] ACPI: APEI: handle synchronous errors in task work with proper si_code Shuai Xue
2023-12-18 6:45 ` [PATCH v10 1/4] ACPI: APEI: set memory failure flags as MF_ACTION_REQUIRED on synchronous events Shuai Xue
2023-12-18 6:53 ` Greg KH
2023-12-21 13:55 ` Rafael J. Wysocki
2023-12-22 1:07 ` Shuai Xue
2023-12-18 6:45 ` [PATCH v10 2/4] ACPI: APEI: send SIGBUS to current task if synchronous memory error not recovered Shuai Xue
2023-12-18 6:54 ` Greg KH
2023-12-18 6:45 ` [PATCH v10 3/4] mm: memory-failure: move memory_failure() return value documentation to function declaration Shuai Xue
2023-12-18 6:54 ` Greg KH
2023-12-18 6:45 ` [PATCH v10 4/4] ACPI: APEI: handle synchronous exceptions in task work Shuai Xue
2023-12-18 6:54 ` Greg KH
2024-02-04 8:01 ` [PATCH v11 0/3] ACPI: APEI: handle synchronous exceptions in task work to send correct SIGBUS si_code Shuai Xue
2024-02-19 1:46 ` Shuai Xue
2024-02-04 8:01 ` [PATCH v11 1/3] ACPI: APEI: send SIGBUS to current task if synchronous memory error not recovered Shuai Xue
2024-02-19 9:25 ` Borislav Petkov
2024-02-22 2:07 ` Shuai Xue
2024-02-23 5:26 ` Dan Williams
2024-02-23 12:08 ` Jonathan Cameron
2024-02-23 12:17 ` Jonathan Cameron
2024-02-24 6:08 ` Shuai Xue
2024-02-26 10:29 ` Borislav Petkov
2024-02-27 1:23 ` Shuai Xue
2024-02-24 19:42 ` Dan Williams
2024-02-24 19:40 ` Dan Williams
2024-02-04 8:01 ` [PATCH v11 2/3] mm: memory-failure: move return value documentation to function declaration Shuai Xue
2024-02-26 10:46 ` Borislav Petkov
2024-02-27 1:27 ` Shuai Xue
2024-02-04 8:01 ` [PATCH v11 3/3] ACPI: APEI: handle synchronous exceptions in task work to send correct SIGBUS si_code Shuai Xue
2024-02-29 7:05 ` Shuai Xue
2024-03-08 10:18 ` Borislav Petkov
2024-03-12 6:05 ` Shuai Xue
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=b56fbae2-0d9b-4c42-94bf-7fd58b3fd738@linux.alibaba.com \
--to=xueshuai@linux.alibaba.com \
--cc=acpica-devel@lists.linuxfoundation.org \
--cc=akpm@linux-foundation.org \
--cc=ardb@kernel.org \
--cc=ashish.kalra@amd.com \
--cc=baolin.wang@linux.alibaba.com \
--cc=bp@alien8.de \
--cc=dave.hansen@linux.intel.com \
--cc=gregkh@linuxfoundation.org \
--cc=hpa@zytor.com \
--cc=james.morse@arm.com \
--cc=jarkko@kernel.org \
--cc=justin.he@arm.com \
--cc=lenb@kernel.org \
--cc=linmiaohe@huawei.com \
--cc=linux-acpi@vger.kernel.org \
--cc=linux-edac@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lvying6@huawei.com \
--cc=mawupeng1@huawei.com \
--cc=mingo@redhat.com \
--cc=naoya.horiguchi@nec.com \
--cc=rafael@kernel.org \
--cc=robert.moore@intel.com \
--cc=stable@vger.kernel.org \
--cc=tanxiaofei@huawei.com \
--cc=tglx@linutronix.de \
--cc=tony.luck@intel.com \
--cc=wangkefeng.wang@huawei.com \
--cc=will@kernel.org \
--cc=x86@kernel.org \
--cc=xiexiuqi@huawei.com \
--cc=ying.huang@intel.com \
--cc=zhuo.song@linux.alibaba.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).