mtd: cfi: fix deadloop in cfi_cmdset_0002.c do_write_buffer
authorLiu Jian <liujian56@huawei.com>
Sun, 3 Mar 2019 07:04:18 +0000 (15:04 +0800)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Sat, 2 May 2020 15:24:39 +0000 (17:24 +0200)
commit d9b8a67b3b95a5c5aae6422b8113adc1c2485f2b upstream.

In function do_write_buffer(), in the for loop, there is a case
chip_ready() returns 1 while chip_good() returns 0, so it never
break the loop.
To fix this, chip_good() is enough and it should timeout if it stay
bad for a while.

Fixes: dfeae1073583("mtd: cfi_cmdset_0002: Change write buffer to check correct value")
Signed-off-by: Yi Huaijie <yihuaijie@huawei.com>
Signed-off-by: Liu Jian <liujian56@huawei.com>
Reviewed-by: Tokunori Ikegami <ikegami_to@yahoo.co.jp>
Signed-off-by: Richard Weinberger <richard@nod.at>
Cc: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drivers/mtd/chips/cfi_cmdset_0002.c

index e773dc6fdd3c67106ecaea95c0b7969ed9a66f1e..1f0d83086cb092ed1b050a4ea989f4f755cbd30e 100644 (file)
@@ -1883,7 +1883,11 @@ static int __xipram do_write_buffer(struct map_info *map, struct flchip *chip,
                        continue;
                }
 
-               if (time_after(jiffies, timeo) && !chip_ready(map, adr))
+               /*
+                * We check "time_after" and "!chip_good" before checking "chip_good" to avoid
+                * the failure due to scheduling.
+                */
+               if (time_after(jiffies, timeo) && !chip_good(map, adr, datum))
                        break;
 
                if (chip_good(map, adr, datum)) {