Re: [Qemu-devel] [RFC PATCH 03/12] block/qcow2-bitmap: avoid adjusting bm->flags for RO bitmaps

[John Snow <jsnow@redhat.com>]

On 05/14/2018 08:44 AM, Vladimir Sementsov-Ogievskiy wrote:
> 12.05.2018 04:25, John Snow wrote:
>> Instead of always setting IN_USE, handle whether or not the bitmap
>> is read-only instead of a two-loop pass. This will allow us to show
>> the flags exactly as they appear for bitmaps in `qemu-img info`.
>>
>> Signed-off-by: John Snow <jsnow@redhat.com>
>> ---
>>   block/qcow2-bitmap.c | 48
>> ++++++++++++++++++++----------------------------
>>   1 file changed, 20 insertions(+), 28 deletions(-)
>>
>> diff --git a/block/qcow2-bitmap.c b/block/qcow2-bitmap.c
>> index d89758f235..fc8d52fc3e 100644
>> --- a/block/qcow2-bitmap.c
>> +++ b/block/qcow2-bitmap.c
>> @@ -942,13 +942,6 @@ fail:
>>       return ret;
>>   }
>>   -/* for g_slist_foreach for GSList of BdrvDirtyBitmap* elements */
>> -static void release_dirty_bitmap_helper(gpointer bitmap,
>> -                                        gpointer bs)
>> -{
>> -    bdrv_release_dirty_bitmap(bs, bitmap);
>> -}
>> -
>>   /* for g_slist_foreach for GSList of BdrvDirtyBitmap* elements */
>>   static void set_readonly_helper(gpointer bitmap, gpointer value)
>>   {
>> @@ -967,8 +960,8 @@ bool qcow2_load_dirty_bitmaps(BlockDriverState
>> *bs, Error **errp)
>>       BDRVQcow2State *s = bs->opaque;
>>       Qcow2BitmapList *bm_list;
>>       Qcow2Bitmap *bm;
>> -    GSList *created_dirty_bitmaps = NULL;
>>       bool header_updated = false;
>> +    bool needs_update = false;
>>         if (s->nb_bitmaps == 0) {
>>           /* No bitmaps - nothing to do */
>> @@ -992,33 +985,32 @@ bool qcow2_load_dirty_bitmaps(BlockDriverState
>> *bs, Error **errp)
>>                   bdrv_disable_dirty_bitmap(bitmap);
>>               }
>>               bdrv_dirty_bitmap_set_persistance(bitmap, true);
>> -            bm->flags |= BME_FLAG_IN_USE;
>> -            created_dirty_bitmaps =
>> -                    g_slist_append(created_dirty_bitmaps, bitmap);
>> -        }
>> -    }
>> -
>> -    if (created_dirty_bitmaps != NULL) {
>> -        if (can_write(bs)) {
>> -            /* in_use flags must be updated */
>> -            int ret = update_ext_header_and_dir_in_place(bs, bm_list);
>> -            if (ret < 0) {
>> -                error_setg_errno(errp, -ret, "Can't update bitmap
>> directory");
>> -                goto fail;
>> +            if (can_write(bs)) {
>> +                bm->flags |= BME_FLAG_IN_USE;
>> +                needs_update = true;
>> +            } else {
>> +                bdrv_dirty_bitmap_set_readonly(bitmap, true);
> 
> oops. This shows, that patch 01 was incorrect: before this (03) patch,
> in this case we'll have IN_USE set in cache but not in the image.
> 

Ah, I can try to order this patch first so that the cache stays
semantically correct.

>>               }
>> -            header_updated = true;
>> -        } else {
>> -            g_slist_foreach(created_dirty_bitmaps, set_readonly_helper,
>> -                            (gpointer)true);
>>           }
>>       }
>>   -    g_slist_free(created_dirty_bitmaps);
>> +    /* in_use flags must be updated */
>> +    if (needs_update) {
>> +        int ret = update_ext_header_and_dir_in_place(bs, bm_list);
>> +        if (ret < 0) {
>> +            error_setg_errno(errp, -ret, "Can't update bitmap
>> directory");
>> +            goto fail;
>> +        }
>> +        header_updated = true;
>> +    }
>>       return header_updated;
> 
> may be safely changed to return needs_update, as if we failed to update,
> we'll go to fail.
> 

Ugh, yes.

>>     fail:
>> -    g_slist_foreach(created_dirty_bitmaps,
>> release_dirty_bitmap_helper, bs);
>> -    g_slist_free(created_dirty_bitmaps);
>> +    QSIMPLEQ_FOREACH(bm, bm_list, entry) {
>> +        if (bm->dirty_bitmap) {
>> +            bdrv_release_dirty_bitmap(bs, bm->dirty_bitmap);
>> +        }
>> +    }
>>       del_bitmap_list(bs);
>>         return false;
> 
> overall looks good, may be worth move it before 01.
>