Re: [Devel] [PATCH] dm-qcow2: parse bitmap extension

2025-02-20 Thread Alexander Atanasov
Hello Andrey, On 19.02.25 18:17, Andrey Zhadchenko wrote: If qcow2 file has only inactive bitmaps, we should not reset QCOW2_AUTOCLEAR_BITMAPS bit. Bitmaps are already allocated and we are not expected to change them. Still reset this bit if we encounter enabled bitmap. In what situation this

Re: [Devel] [PATCH] dm-qcow2: parse bitmap extension

2025-02-20 Thread Andrey Zhadchenko
On 2/20/25 11:17, Alexander Atanasov wrote: Hello Andrey, On 19.02.25 18:17, Andrey Zhadchenko wrote: If qcow2 file has only inactive bitmaps, we should not reset QCOW2_AUTOCLEAR_BITMAPS bit. Bitmaps are already allocated and we are not expected to change them. Still reset this bit if we enco

Re: [Devel] [PATCH] dm-qcow2: parse bitmap extension

2025-02-20 Thread Andrey Zhadchenko
On 2/20/25 11:17, Alexander Atanasov wrote: Hello Andrey, On 19.02.25 18:17, Andrey Zhadchenko wrote: If qcow2 file has only inactive bitmaps, we should not reset QCOW2_AUTOCLEAR_BITMAPS bit. Bitmaps are already allocated and we are not expected to change them. Still reset this bit if we enco

[Devel] [PATCH VZ9] fs/ext4: bug in swapext ioctl

2025-02-20 Thread Alexey Kuznetsov
Stupid and very dangerous mistake. Plain luck it did not result in corruptions, was exposed as failure to convert csd journal from old to new format. Affects: #VSTOR-100586 Signed-off-by: Alexey Kuznetsov --- fs/ext4/move_extent.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --gi

[Devel] [PATCH VZ9 v2] fs/ext4: bug in swapext ioctl

2025-02-20 Thread Alexey Kuznetsov
Stupid and very dangerous mistake. Plain luck it did not result in corruptions, was exposed as failure to convert csd journal from old to new format. Affects: #VSTOR-100610 Signed-off-by: Alexey Kuznetsov --- fs/ext4/move_extent.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --gi

Re: [Devel] [PATCH v2] dm-qcow2: parse bitmap extension

2025-02-20 Thread Andrey Zhadchenko
On 2/20/25 13:08, Alexander Atanasov wrote: Hi, you missed my inline comments. Oops, my bad. Will fix it asap On 20.02.25 14:07, Andrey Zhadchenko wrote: Programs, which work with qcow2 files as writables, are expected to reset autoclear_features bits which are unknown to them. If qcow2 f

[Devel] [PATCH v3] dm-qcow2: parse bitmap extension

2025-02-20 Thread Andrey Zhadchenko
Programs, which work with qcow2 files as writables, are expected to reset autoclear_features bits which are unknown to them. If qcow2 file has only inactive bitmaps, we may not reset QCOW2_AUTOCLEAR_BITMAPS bit. Bitmaps are already allocated and we are not expected to change them. Still reset this

[Devel] [PATCH v2] dm-qcow2: parse bitmap extension

2025-02-20 Thread Andrey Zhadchenko
Programs, which work with qcow2 files as writables, are expected to reset autoclear_features bits which are unknown to them. If qcow2 file has only inactive bitmaps, we may not reset QCOW2_AUTOCLEAR_BITMAPS bit. Bitmaps are already allocated and we are not expected to change them. Still reset this

Re: [Devel] [PATCH v2] dm-qcow2: parse bitmap extension

2025-02-20 Thread Alexander Atanasov
Hi, you missed my inline comments. On 20.02.25 14:07, Andrey Zhadchenko wrote: Programs, which work with qcow2 files as writables, are expected to reset autoclear_features bits which are unknown to them. If qcow2 file has only inactive bitmaps, we may not reset QCOW2_AUTOCLEAR_BITMAPS bit. Bitma