Hi Adrian,
On Mon, May 9, 2022 at 1:30 PM Adrian Fiergolski
<adrian.fiergol...@fastree3d.com> wrote:
>
> Hi Oleksandr,
>
> On 04.05.2022 20:26, Oleksandr Suvorov wrote:
> > Hi Adrian,
> >
> > On Wed, May 4, 2022 at 5:28 PM Adrian Fiergolski
> > <adrian.fiergol...@fastree3d.com> wrote:
> >> Hi Oleksandr,
> >>
> >> On 03.05.2022 09:42, Michal Simek wrote:
> >>>
> >>> On 4/11/22 20:00, Adrian Fiergolski wrote:
> >>>> From: Oleksandr Suvorov <oleksandr.suvo...@foundries.io>
> >>>>
> >>>> Introduce a function which passes an fpga compatible string from
> >>>> FIT images to FPGA drivers. This lets the different implementations
> >>>> decide how to handle it.
> >>>>
> >>>> Some code of Jorge Ramirez-Ortiz <jo...@foundries.io> is reused.
> >>>>
> >>>> Signed-off-by: Oleksandr Suvorov <oleksandr.suvo...@foundries.io>
> >>>> Tested-by: Ricardo Salveti <rica...@foundries.io>
> >>>> Co-developed-by: Adrian Fiergolski <adrian.fiergol...@fastree3d.com>
> >>>> Signed-off-by: Adrian Fiergolski <adrian.fiergol...@fastree3d.com>
> >>>> ---
> >>>> common/spl/spl_fit.c | 6 ++----
> >>>> drivers/fpga/fpga.c | 23 ++++++++++++++++++++++-
> >>>> include/fpga.h | 4 ++++
> >>>> 3 files changed, 28 insertions(+), 5 deletions(-)
> >>>>
> >>>> diff --git a/common/spl/spl_fit.c b/common/spl/spl_fit.c
> >>>> index 1bbf824684..0e3c2a94b6 100644
> >>>> --- a/common/spl/spl_fit.c
> >>>> +++ b/common/spl/spl_fit.c
> >>>> @@ -588,11 +588,9 @@ static int spl_fit_upload_fpga(struct
> >>>> spl_fit_info *ctx, int node,
> >>>> compatible = fdt_getprop(ctx->fit, node, "compatible", NULL);
> >>>> if (!compatible)
> >>>> warn_deprecated("'fpga' image without 'compatible' property");
> >>>> - else if (strcmp(compatible, "u-boot,fpga-legacy"))
> >>>> - printf("Ignoring compatible = %s property\n", compatible);
> >>>> - ret = fpga_load(0, (void *)fpga_image->load_addr,
> >>>> fpga_image->size,
> >>>> - BIT_FULL);
> >>>> + ret = fit_fpga_load(0, (void *)fpga_image->load_addr,
> >>>> fpga_image->size,
> >>>> + BIT_FULL, compatible);
> >>>> if (ret) {
> >>>> printf("%s: Cannot load the image to the FPGA\n", __func__);
> >>>> return ret;
> >>>> diff --git a/drivers/fpga/fpga.c b/drivers/fpga/fpga.c
> >>>> index 3b0a44b242..a306dd81f9 100644
> >>>> --- a/drivers/fpga/fpga.c
> >>>> +++ b/drivers/fpga/fpga.c
> >>>> @@ -249,8 +249,23 @@ int fpga_loads(int devnum, const void *buf,
> >>>> size_t size,
> >>>> }
> >>>> #endif
> >>>> +int fit_fpga_load(int devnum, const void *buf, size_t bsize,
> >>>> + bitstream_type bstype, const char *compatible)
> >>>> +{
> >>>> + fpga_desc *desc = fpga_get_desc(devnum);
> >>> this generates warning which you should fix.
> >>>
> >>> + fpga_desc *desc = fpga_get_desc(devnum);
> >>> + ^~~~~~~~~~~~~
> >>> w+../drivers/fpga/fpga.c: In function ‘fit_fpga_load’:
> >>> w+../drivers/fpga/fpga.c:255:20: warning: initialization discards
> >>> ‘const’ qualifier from pointer target type [-Wdiscarded-qualifiers]
> >> As it's your patch, could you address it?
> >>
> >> The 'compatible' field can't be set here, as fpga_get_desc returns
> >> 'const fpga_desc * const'. The descriptors table is populated (fpga_add)
> >> in board_init. At that stage, I am afraid, we don't have access to the
> >> fit image information yet to set a proper 'compatible' (would need to
> >> check the code more carefully). Moreover, IMHO it's not the cleanest way
> >> to change the 'compatible' field after the driver's initialisation.
> > Thanks for catching this. I'm addressing the issue this Friday.
>
> Perfect. I understand you will address all Michal's comments, including
> those with my SOB missing, and you will release the new version of the
> patches, right?
Yes, I've already addressed Michal's comments and going to release the new
patchset version in 1-2 hours after finishing tests.
>
> Thanks,
> Adrian
>
--
Best regards
Oleksandr
Oleksandr Suvorov
cryo...@gmail.com
