public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Baoquan He <bhe@redhat.com>
To: Zhen Lei <thunder.leizhen@huawei.com>
Cc: Eric Biederman <ebiederm@xmission.com>,
	kexec@lists.infradead.org, linux-kernel@vger.kernel.org,
	Michael Holzheu <holzheu@linux.vnet.ibm.com>,
	Andrew Morton <akpm@linux-foundation.org>
Subject: Re: [PATCH 6/6] kexec: enable kexec_crash_size to support two crash kernel regions
Date: Wed, 31 May 2023 17:53:34 +0800	[thread overview]
Message-ID: <ZHcY/jsExa8t7hJW@MiWiFi-R3L-srv> (raw)
In-Reply-To: <20230527123439.772-7-thunder.leizhen@huawei.com>

On 05/27/23 at 08:34pm, Zhen Lei wrote:
> The crashk_low_res should be considered by /sys/kernel/kexec_crash_size
> to support two crash kernel regions. Since crashk_res manages the memory
> with high address and crashk_low_res manages the memory with low address,
> crashk_low_res is shrunken only when all crashk_res is shrunken. And
> because when there is only one crash kernel region, crashk_res is always
> used. Therefore, if all crashk_res is shrunken and crashk_low_res still
> exists, swap them.

This looks good, otherwise someone else won't stop attempting to add
support of crashk_low_res shrinking. Not sure if this will bring corner
case issue in testing, let's see. For the patch log, I tried to
rephrase, feel free to refer to.

=====
The crashk_low_res should be considered by /sys/kernel/kexec_crash_size
to support two crash kernel regions shrinking if existing.

While doing it, crashk_low_res will only be shrunk when the entire
crashk_res is empty; and if the crashk_res is empty and crahk_low_res
is not, change crashk_low_res to be crashk_res.
=====

With the log updated, you can add:

Acked-by: Baoquan He <bhe@redhat.com>

> 
> Signed-off-by: Zhen Lei <thunder.leizhen@huawei.com>
> ---
>  kernel/kexec_core.c | 43 ++++++++++++++++++++++++++++++++++++++-----
>  1 file changed, 38 insertions(+), 5 deletions(-)
> 
> diff --git a/kernel/kexec_core.c b/kernel/kexec_core.c
> index e82bc6d6634136a..c1d50f6566300d9 100644
> --- a/kernel/kexec_core.c
> +++ b/kernel/kexec_core.c
> @@ -1091,6 +1091,11 @@ __bpf_kfunc void crash_kexec(struct pt_regs *regs)
>  	}
>  }
>  
> +static inline resource_size_t crash_resource_size(const struct resource *res)
> +{
> +	return !res->end ? 0 : resource_size(res);
> +}
> +
>  ssize_t crash_get_memory_size(void)
>  {
>  	ssize_t size = 0;
> @@ -1098,8 +1103,8 @@ ssize_t crash_get_memory_size(void)
>  	if (!kexec_trylock())
>  		return -EBUSY;
>  
> -	if (crashk_res.end != crashk_res.start)
> -		size = resource_size(&crashk_res);
> +	size += crash_resource_size(&crashk_res);
> +	size += crash_resource_size(&crashk_low_res);
>  
>  	kexec_unlock();
>  	return size;
> @@ -1135,7 +1140,7 @@ int __crash_shrink_memory(struct resource *old_res, unsigned long new_size)
>  int crash_shrink_memory(unsigned long new_size)
>  {
>  	int ret = 0;
> -	unsigned long old_size;
> +	unsigned long old_size, low_size;
>  
>  	if (!kexec_trylock())
>  		return -EBUSY;
> @@ -1144,14 +1149,42 @@ int crash_shrink_memory(unsigned long new_size)
>  		ret = -ENOENT;
>  		goto unlock;
>  	}
> -	old_size = !crashk_res.end ? 0 : resource_size(&crashk_res);
> +
> +	low_size = crash_resource_size(&crashk_low_res);
> +	old_size = crash_resource_size(&crashk_res) + low_size;
>  	new_size = roundup(new_size, KEXEC_CRASH_MEM_ALIGN);
>  	if (new_size >= old_size) {
>  		ret = (new_size == old_size) ? 0 : -EINVAL;
>  		goto unlock;
>  	}
>  
> -	ret = __crash_shrink_memory(&crashk_res, new_size);
> +	/*
> +	 * (low_size > new_size) implies that low_size is greater than zero.
> +	 * This also means that if low_size is zero, the else branch is taken.
> +	 *
> +	 * If low_size is greater than 0, (low_size > new_size) indicates that
> +	 * crashk_low_res also needs to be shrunken. Otherwise, only crashk_res
> +	 * needs to be shrunken.
> +	 */
> +	if (low_size > new_size) {
> +		ret = __crash_shrink_memory(&crashk_res, 0);
> +		if (ret)
> +			goto unlock;
> +
> +		ret = __crash_shrink_memory(&crashk_low_res, new_size);
> +	} else {
> +		ret = __crash_shrink_memory(&crashk_res, new_size - low_size);
> +	}
> +
> +	/* Swap crashk_res and crashk_low_res if needed */
> +	if (!crashk_res.end && crashk_low_res.end) {
> +		crashk_res.start = crashk_low_res.start;
> +		crashk_res.end   = crashk_low_res.end;
> +		release_resource(&crashk_low_res);
> +		crashk_low_res.start = 0;
> +		crashk_low_res.end   = 0;
> +		insert_resource(&iomem_resource, &crashk_res);
> +	}
>  
>  unlock:
>  	kexec_unlock();
> -- 
> 2.25.1
> 


  reply	other threads:[~2023-05-31  9:54 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-05-27 12:34 [PATCH 0/6] kexec: enable kexec_crash_size to support two crash kernel regions Zhen Lei
2023-05-27 12:34 ` [PATCH 1/6] kexec: fix a memory leak in crash_shrink_memory() Zhen Lei
2023-05-31  0:13   ` Baoquan He
2023-05-31  1:16     ` Leizhen (ThunderTown)
2023-05-31  7:31       ` Baoquan He
2023-05-27 12:34 ` [PATCH 2/6] kexec: delete a useless check " Zhen Lei
2023-05-31  0:17   ` Baoquan He
2023-05-31  2:19     ` Leizhen (ThunderTown)
2023-05-31  7:41       ` Baoquan He
2023-05-31  8:26         ` Leizhen (ThunderTown)
2023-05-27 12:34 ` [PATCH 3/6] kexec: clear crashk_res if all its memory has been released Zhen Lei
2023-05-31  0:33   ` Baoquan He
2023-05-27 12:34 ` [PATCH 4/6] kexec: improve the readability of crash_shrink_memory() Zhen Lei
2023-05-31  7:48   ` Baoquan He
2023-05-27 12:34 ` [PATCH 5/6] kexec: add helper __crash_shrink_memory() Zhen Lei
2023-05-28  0:08   ` kernel test robot
2023-05-29  0:37     ` Leizhen (ThunderTown)
2023-05-28  1:44   ` kernel test robot
2023-05-28  6:26   ` kernel test robot
2023-05-31  7:50   ` Baoquan He
2023-05-27 12:34 ` [PATCH 6/6] kexec: enable kexec_crash_size to support two crash kernel regions Zhen Lei
2023-05-31  9:53   ` Baoquan He [this message]
2023-05-31 14:25     ` Leizhen (ThunderTown)

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=ZHcY/jsExa8t7hJW@MiWiFi-R3L-srv \
    --to=bhe@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=ebiederm@xmission.com \
    --cc=holzheu@linux.vnet.ibm.com \
    --cc=kexec@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=thunder.leizhen@huawei.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