Re: [PATCH] raid5: use memalloc_noio_save()/restore in resize_chunks()

[kbuild test robot <lkp@intel.com>]

[-- Attachment #1: Type: text/plain, Size: 3509 bytes --]

Hi Coly,

I love your patch! Yet something to improve:

[auto build test ERROR on linus/master]
[also build test ERROR on v5.6 next-20200403]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system. BTW, we also suggest to use '--base' option to specify the
base tree in git format-patch, please see https://stackoverflow.com/a/37406982]

url:    https://github.com/0day-ci/linux/commits/Coly-Li/raid5-use-memalloc_noio_save-restore-in-resize_chunks/20200402-215014
base:   https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 919dce24701f7b34681a6a1d3ef95c9f6c4fb1cc
config: x86_64-rhel (attached as .config)
compiler: gcc-7 (Ubuntu 7.5.0-6ubuntu2) 7.5.0
reproduce:
        # save the attached .config to linux build tree
        make ARCH=x86_64 

If you fix the issue, kindly add following tag
Reported-by: kbuild test robot <lkp@intel.com>

All errors (new ones prefixed by >>):

   drivers/md/raid5.c: In function 'scribble_alloc':
   drivers/md/raid5.c:2237:15: warning: unused variable 'noio_flag' [-Wunused-variable]
     unsigned int noio_flag;
                  ^~~~~~~~~
   drivers/md/raid5.c: In function 'resize_chunks':
>> drivers/md/raid5.c:2275:14: error: implicit declaration of function 'memalloc_noio_save'; did you mean 'vmalloc_to_page'? [-Werror=implicit-function-declaration]
     noio_flag = memalloc_noio_save();
                 ^~~~~~~~~~~~~~~~~~
                 vmalloc_to_page
>> drivers/md/raid5.c:2285:2: error: implicit declaration of function 'memalloc_noio_restore'; did you mean '__vmalloc_node_range'? [-Werror=implicit-function-declaration]
     memalloc_noio_restore(noio_flag);
     ^~~~~~~~~~~~~~~~~~~~~
     __vmalloc_node_range
   cc1: some warnings being treated as errors

vim +2275 drivers/md/raid5.c

  2249	
  2250	static int resize_chunks(struct r5conf *conf, int new_disks, int new_sectors)
  2251	{
  2252		unsigned long cpu;
  2253		int err = 0;
  2254		unsigned int noio_flag;
  2255	
  2256		/*
  2257		 * Never shrink. And mddev_suspend() could deadlock if this is called
  2258		 * from raid5d. In that case, scribble_disks and scribble_sectors
  2259		 * should equal to new_disks and new_sectors
  2260		 */
  2261		if (conf->scribble_disks >= new_disks &&
  2262		    conf->scribble_sectors >= new_sectors)
  2263			return 0;
  2264		mddev_suspend(conf->mddev);
  2265		get_online_cpus();
  2266	
  2267		/*
  2268		 * scribble_alloc() allocates memory by kvmalloc_array(), if
  2269		 * the memory allocation triggers memory reclaim I/Os onto
  2270		 * this raid array, there might be potential deadlock if this
  2271		 * raid array happens to be suspended during memory allocation.
  2272		 * Here the scope APIs are used to disable such recursive memory
  2273		 * reclaim I/Os.
  2274		 */
> 2275		noio_flag = memalloc_noio_save();
  2276		for_each_present_cpu(cpu) {
  2277			struct raid5_percpu *percpu;
  2278	
  2279			percpu = per_cpu_ptr(conf->percpu, cpu);
  2280			err = scribble_alloc(percpu, new_disks,
  2281					     new_sectors / STRIPE_SECTORS);
  2282			if (err)
  2283				break;
  2284		}
> 2285		memalloc_noio_restore(noio_flag);
  2286	
  2287		put_online_cpus();
  2288		mddev_resume(conf->mddev);
  2289		if (!err) {
  2290			conf->scribble_disks = new_disks;
  2291			conf->scribble_sectors = new_sectors;
  2292		}
  2293		return err;
  2294	}
  2295	

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org

[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 44475 bytes --]