* [PATCH] bcache: fix wrong BITMASK offset value for BDEV_CACHE_MODE
@ 2021-07-20 10:32 Lin Feng
2022-05-13 15:20 ` Coly Li
0 siblings, 1 reply; 3+ messages in thread
From: Lin Feng @ 2021-07-20 10:32 UTC (permalink / raw)
To: colyli; +Cc: axboe, linf, linux-kernel, linux-bcache
Original codes:
BITMASK(CACHE_SYNC, struct cache_sb, flags, 0, 1);
BITMASK(CACHE_DISCARD, struct cache_sb, flags, 1, 1);
BITMASK(CACHE_REPLACEMENT, struct cache_sb, flags, 2, 3);
...
BITMASK(BDEV_CACHE_MODE, struct cache_sb, flags, 0, 4);
Should BDEV_CACHE_MODE bits start from bit-nr 5(2+3) else it overlaps
with previous defined bit chunks, since we have 4 types of cache modes,
BDEV_CACHE_MODE will overwrite CACHE_SYNC and CACHE_DISCARD bits.
This bug stays there since first upstream version of bcache, don't know
why it lives so long, or am i missing something, please point me out
if I'm wrong, thanks!
Signed-off-by: Lin Feng <linf@wangsu.com>
---
include/uapi/linux/bcache.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/include/uapi/linux/bcache.h b/include/uapi/linux/bcache.h
index cf7399f03b71..dccd89756451 100644
--- a/include/uapi/linux/bcache.h
+++ b/include/uapi/linux/bcache.h
@@ -288,7 +288,7 @@ BITMASK(CACHE_REPLACEMENT, struct cache_sb, flags, 2, 3);
#define CACHE_REPLACEMENT_FIFO 1U
#define CACHE_REPLACEMENT_RANDOM 2U
-BITMASK(BDEV_CACHE_MODE, struct cache_sb, flags, 0, 4);
+BITMASK(BDEV_CACHE_MODE, struct cache_sb, flags, 5, 4);
#define CACHE_MODE_WRITETHROUGH 0U
#define CACHE_MODE_WRITEBACK 1U
#define CACHE_MODE_WRITEAROUND 2U
--
2.31.1
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH] bcache: fix wrong BITMASK offset value for BDEV_CACHE_MODE
2021-07-20 10:32 [PATCH] bcache: fix wrong BITMASK offset value for BDEV_CACHE_MODE Lin Feng
@ 2022-05-13 15:20 ` Coly Li
2022-05-16 1:50 ` Lin Feng
0 siblings, 1 reply; 3+ messages in thread
From: Coly Li @ 2022-05-13 15:20 UTC (permalink / raw)
To: Lin Feng; +Cc: axboe, linux-kernel, linux-bcache
On 7/20/21 6:32 PM, Lin Feng wrote:
> Original codes:
> BITMASK(CACHE_SYNC, struct cache_sb, flags, 0, 1);
> BITMASK(CACHE_DISCARD, struct cache_sb, flags, 1, 1);
> BITMASK(CACHE_REPLACEMENT, struct cache_sb, flags, 2, 3);
> ...
> BITMASK(BDEV_CACHE_MODE, struct cache_sb, flags, 0, 4);
>
> Should BDEV_CACHE_MODE bits start from bit-nr 5(2+3) else it overlaps
> with previous defined bit chunks, since we have 4 types of cache modes,
> BDEV_CACHE_MODE will overwrite CACHE_SYNC and CACHE_DISCARD bits.
The overlap won't happen, previous lines are for cache device, and what
you modified is for backing device.
And your patch changes the on-disk format, which is unacceptable anyway.
Coly Li
>
> This bug stays there since first upstream version of bcache, don't know
> why it lives so long, or am i missing something, please point me out
> if I'm wrong, thanks!
>
> Signed-off-by: Lin Feng <linf@wangsu.com>
> ---
> include/uapi/linux/bcache.h | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/include/uapi/linux/bcache.h b/include/uapi/linux/bcache.h
> index cf7399f03b71..dccd89756451 100644
> --- a/include/uapi/linux/bcache.h
> +++ b/include/uapi/linux/bcache.h
> @@ -288,7 +288,7 @@ BITMASK(CACHE_REPLACEMENT, struct cache_sb, flags, 2, 3);
> #define CACHE_REPLACEMENT_FIFO 1U
> #define CACHE_REPLACEMENT_RANDOM 2U
>
> -BITMASK(BDEV_CACHE_MODE, struct cache_sb, flags, 0, 4);
> +BITMASK(BDEV_CACHE_MODE, struct cache_sb, flags, 5, 4);
> #define CACHE_MODE_WRITETHROUGH 0U
> #define CACHE_MODE_WRITEBACK 1U
> #define CACHE_MODE_WRITEAROUND 2U
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] bcache: fix wrong BITMASK offset value for BDEV_CACHE_MODE
2022-05-13 15:20 ` Coly Li
@ 2022-05-16 1:50 ` Lin Feng
0 siblings, 0 replies; 3+ messages in thread
From: Lin Feng @ 2022-05-16 1:50 UTC (permalink / raw)
To: Coly Li; +Cc: axboe, linux-kernel, linux-bcache
On 5/13/22 23:20, Coly Li wrote:
> On 7/20/21 6:32 PM, Lin Feng wrote:
>> Original codes:
>> BITMASK(CACHE_SYNC, struct cache_sb, flags, 0, 1);
>> BITMASK(CACHE_DISCARD, struct cache_sb, flags, 1, 1);
>> BITMASK(CACHE_REPLACEMENT, struct cache_sb, flags, 2, 3);
>> ...
>> BITMASK(BDEV_CACHE_MODE, struct cache_sb, flags, 0, 4);
>>
>> Should BDEV_CACHE_MODE bits start from bit-nr 5(2+3) else it overlaps
>> with previous defined bit chunks, since we have 4 types of cache modes,
>> BDEV_CACHE_MODE will overwrite CACHE_SYNC and CACHE_DISCARD bits.
>
> The overlap won't happen, previous lines are for cache device, and what
> you modified is for backing device.
>
> And your patch changes the on-disk format, which is unacceptable anyway.
>
Yes, you are right, this patch happened as I started reading bcache codes,
I did not get a whole picture of bcache at that time, I'm sorry for making
noise.
Thank you, Coly!
linfeng
>
> Coly Li
>
>>
>> This bug stays there since first upstream version of bcache, don't know
>> why it lives so long, or am i missing something, please point me out
>> if I'm wrong, thanks!
>>
>> Signed-off-by: Lin Feng <linf@wangsu.com>
>> ---
>> include/uapi/linux/bcache.h | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/include/uapi/linux/bcache.h b/include/uapi/linux/bcache.h
>> index cf7399f03b71..dccd89756451 100644
>> --- a/include/uapi/linux/bcache.h
>> +++ b/include/uapi/linux/bcache.h
>> @@ -288,7 +288,7 @@ BITMASK(CACHE_REPLACEMENT, struct cache_sb, flags, 2, 3);
>> #define CACHE_REPLACEMENT_FIFO 1U
>> #define CACHE_REPLACEMENT_RANDOM 2U
>>
>> -BITMASK(BDEV_CACHE_MODE, struct cache_sb, flags, 0, 4);
>> +BITMASK(BDEV_CACHE_MODE, struct cache_sb, flags, 5, 4);
>> #define CACHE_MODE_WRITETHROUGH 0U
>> #define CACHE_MODE_WRITEBACK 1U
>> #define CACHE_MODE_WRITEAROUND 2U
>
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2022-05-16 1:50 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-07-20 10:32 [PATCH] bcache: fix wrong BITMASK offset value for BDEV_CACHE_MODE Lin Feng
2022-05-13 15:20 ` Coly Li
2022-05-16 1:50 ` Lin Feng
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).