Re: [PATCH] xen/blkfront: Only check REQ_FUA for writes

2023-05-03 Thread Roger Pau Monné
ksandr_tyshche...@epam.com > > ; ax...@kernel.dk > > Subject: Re: [PATCH] xen/blkfront: Only check REQ_FUA for writes > >   > > On Wed, Apr 26, 2023 at 05:40:05PM +0100, Ross Lagerwall wrote: > > > The existing code silently converts read operations with the > > >

Re: [PATCH] xen/blkfront: Only check REQ_FUA for writes

2023-05-02 Thread Ross Lagerwall
> From: Roger Pau Monne > Sent: Tuesday, May 2, 2023 4:57 PM > To: Ross Lagerwall > Cc: xen-devel@lists.xenproject.org ; > jgr...@suse.com ; sstabell...@kernel.org > ; oleksandr_tyshche...@epam.com > ; ax...@kernel.dk > Subject: Re: [PATCH] xen/blkfront: Only

Re: [PATCH] xen/blkfront: Only check REQ_FUA for writes

2023-05-02 Thread Roger Pau Monné
On Wed, Apr 26, 2023 at 05:40:05PM +0100, Ross Lagerwall wrote: > The existing code silently converts read operations with the > REQ_FUA bit set into write-barrier operations. This results in data > loss as the backend scribbles zeroes over the data instead of returning > it. > > While the

Re: [PATCH] xen/blkfront: Only check REQ_FUA for writes

2023-05-02 Thread Juergen Gross
On 26.04.23 18:40, Ross Lagerwall wrote: The existing code silently converts read operations with the REQ_FUA bit set into write-barrier operations. This results in data loss as the backend scribbles zeroes over the data instead of returning it. While the REQ_FUA bit doesn't make sense on a

[PATCH] xen/blkfront: Only check REQ_FUA for writes

2023-04-26 Thread Ross Lagerwall
The existing code silently converts read operations with the REQ_FUA bit set into write-barrier operations. This results in data loss as the backend scribbles zeroes over the data instead of returning it. While the REQ_FUA bit doesn't make sense on a read operation, at least one well-known