[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [RFC 0/2] Split padded I/O vectors exceeding IOV_MAX
|
From: |
Stefan Hajnoczi |
|
Subject: |
Re: [RFC 0/2] Split padded I/O vectors exceeding IOV_MAX |
|
Date: |
Wed, 15 Mar 2023 11:29:41 -0400 |
On Wed, Mar 15, 2023 at 01:13:28PM +0100, Hanna Czenczek wrote:
> Hi,
>
> We accept I/O vectors with up to 1024 (IOV_MAX) elements from guests.
> When a guest request does not conform to the underlying storage's
> alignment requirements, we pad it with head and/or tail buffers as
> necessary, which are simply appended to the I/O vector.
>
> As of 4c002cef, we (sensibly) check that such-padded vectors do not then
> exceed IOV_MAX. However, there seems to be no sensible error handling;
> instead, the guest simply receives an I/O error.
>
> That???s a problem, because it submitted a perfectly sensible I/O request
Looks like there is an encoding issue. I get 3 question marks instead of
an apostrophe. lore.kernel.org also renders mojibake:
20230315121330.29679-1-hreitz@redhat.com/">https://lore.kernel.org/qemu-devel/20230315121330.29679-1-hreitz@redhat.com/
Stefan
signature.asc
Description: PGP signature