On Sat, Oct 29, 2016 at 07:59:09PM +0200, Marek Vasut wrote:
> On 10/28/2016 07:58 AM, Heiner Kallweit wrote:
> > - This patch was part of a series and the other patches of the
> > series went into 4.9. It's needed to correctly support
> > reads > 64k on fsl-espi as first user of spi_max_message_size.
> > See commit 02a595d5d6e4 ("spi: fsl-espi: eliminate spi nor
> > flash read loop"). CC it to stable.
> Does this imply that we have to fix every single driver this way ?
Ideally. Any SPI driver that has an upper limit really ought to export
it, and any driver that might bump into one of those limits should
check. A lot of drivers are either unlikely to be deployed in a system
that has relevant limits or tend to only do short enough transfers
though - if systems are currently working it's just neatening things up
really.
Attachment:
signature.asc
Description: PGP signature
