lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date:   Mon, 10 Feb 2020 19:19:33 -0800
From:   Greg KH <gregkh@...uxfoundation.org>
To:     Jing Xia <jing.xia.mail@...il.com>
Cc:     akpm@...ux-foundation.org, npiggin@...il.com, rppt@...ux.ibm.com,
        allison@...utok.net, tglx@...utronix.de, linux-mm@...ck.org,
        linux-kernel@...r.kernel.org, yuming.han@...soc.com,
        chunyang.zhang@...soc.com, Jing Xia <jing.xia@...soc.com>
Subject: Re: [RFC] lib/show_mem.c: extend show_mem() to support showing
 drivers' memory consumption

On Fri, Jan 17, 2020 at 04:24:05PM +0800, Jing Xia wrote:
> From: Jing Xia <jing.xia@...soc.com>
> 
> In low memory situations, sometimes we need to check how much memory
> a driver using.This patch add a notifer chain to show_mem.So a driver
> can show their memory usage when show_mem_extend() is called.
> 
> Co-developed-by: Yuming Han <yuming.han@...soc.com>
> Signed-off-by: Yuming Han <yuming.han@...soc.com>
> Signed-off-by: Jing Xia <jing.xia@...soc.com>
> ---
>  include/linux/mm.h |  9 +++++++++
>  lib/show_mem.c     | 36 ++++++++++++++++++++++++++++++++++++
>  2 files changed, 45 insertions(+)
> 
> diff --git a/include/linux/mm.h b/include/linux/mm.h
> index cfaa8fe..a37274a 100644
> --- a/include/linux/mm.h
> +++ b/include/linux/mm.h
> @@ -2201,6 +2201,15 @@ extern void memmap_init_zone(unsigned long, int, unsigned long, unsigned long,
>  #ifdef __HAVE_ARCH_RESERVED_KERNEL_PAGES
>  extern unsigned long arch_reserved_kernel_pages(void);
>  #endif
> +enum show_mem_extend_type {
> +	SHOW_MEM_EXTEND_BASIC,
> +	SHOW_MEM_EXTEND_CLASSIC,
> +	SHOW_MEM_EXTEND_ALL
> +};
> +extern int register_show_mem_notifier(struct notifier_block *nb);
> +extern int unregister_show_mem_notifier(struct notifier_block *nb);
> +extern void show_mem_extend(unsigned int flags, nodemask_t *nodemask,
> +			    enum show_mem_extend_type type);
>  
>  extern __printf(3, 4)
>  void warn_alloc(gfp_t gfp_mask, nodemask_t *nodemask, const char *fmt, ...);
> diff --git a/lib/show_mem.c b/lib/show_mem.c
> index 1c26c14..6b013cb 100644
> --- a/lib/show_mem.c
> +++ b/lib/show_mem.c
> @@ -7,6 +7,8 @@
>  
>  #include <linux/mm.h>
>  #include <linux/cma.h>
> +#include <linux/notifier.h>
> +#include <linux/swap.h>
>  
>  void show_mem(unsigned int filter, nodemask_t *nodemask)
>  {
> @@ -42,3 +44,37 @@ void show_mem(unsigned int filter, nodemask_t *nodemask)
>  	printk("%lu pages hwpoisoned\n", atomic_long_read(&num_poisoned_pages));
>  #endif
>  }
> +
> +static BLOCKING_NOTIFIER_HEAD(show_mem_notify_list);
> +
> +int register_show_mem_notifier(struct notifier_block *nb)
> +{
> +	return blocking_notifier_chain_register(&show_mem_notify_list, nb);
> +}
> +EXPORT_SYMBOL_GPL(register_show_mem_notifier);
> +
> +int unregister_show_mem_notifier(struct notifier_block *nb)
> +{
> +	return blocking_notifier_chain_unregister(&show_mem_notify_list, nb);
> +}
> +EXPORT_SYMBOL_GPL(unregister_show_mem_notifier);

You are exporting functions that are never used at all.  We can't do
that.


> +
> +void show_mem_extend(unsigned int filter, nodemask_t *nodemask,
> +		     enum show_mem_extend_type type)
> +{
> +	unsigned long used = 0;
> +	struct sysinfo si;
> +
> +	pr_info("Mem-Info-Extend:\n");
> +	show_mem(filter, NULL);
> +	si_meminfo(&si);
> +	pr_info("MemTotal:	%8lu KB\n"
> +		"Buffers:	%8lu KB\n"
> +		"SwapCached:	%8lu KB\n",
> +		(si.totalram) << (PAGE_SHIFT - 10),
> +		(si.bufferram) << (PAGE_SHIFT - 10),
> +		total_swapcache_pages() << (PAGE_SHIFT - 10));
> +
> +	blocking_notifier_call_chain(&show_mem_notify_list,
> +				     (unsigned long)type, &used);
> +}

Who calls this function?

As a stand-alone patch, this makes no sense as it doesn't seem to do
anything, so there's no way it can be accepted.

Please submit this as a patch series, with actual users of these
functions, if you wish to have it properly considered.

thanks,

greg k-h

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ