[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] Re: [PATCH] Use hex for bitstream_size as expected by u-boot
On 10/11/2023 10:08, Julien Grall wrote: On 10/11/2023 10:01, Michal Orzel wrote:Hi Julien, On 10/11/2023 10:27, Julien Grall wrote:Hi Stefano, On 10/11/2023 01:44, Stefano Stabellini wrote:Reported-by: Brian Woods <brian@xxxxxxxxx> Signed-off-by: Stefano Stabellini <stefano.stabellini@xxxxxxx> diff --git a/scripts/uboot-script-gen b/scripts/uboot-script-gen index b284887..6e52da5 100755 --- a/scripts/uboot-script-gen +++ b/scripts/uboot-script-gen @@ -790,7 +790,7 @@ bitstream_load_and_config() check_file_type "$BITSTREAM" "BIT data" bitstream_addr=$memaddr load_file $BITSTREAM "fpga_bitstream" - bitstream_size=$filesize + bitstream_size=`printf "0x%X\n" $filesize`Looking at [1], there is no indication that the size parameter for "fpga load" should be hexadecimal. At the contrary, all the example I have found seems to use $filesize.U-boot expects size to be passed in hex format. You can see it here: https://github.com/u-boot/u-boot/blob/master/cmd/fpga.c#L60C20-L60C27Also, AFAICT $filesize var that gets updated after image load (e.g. tftpb) is in hex format.Hmmm, now that you are saying this... I vaguely recall we had some issues in the past with the size. The format was not consistent across U-boot.Maybe this is the same problem here? I had a chat with Michal on Matrix. 'filesize' is a variable internal to the image-builder rather than U-boot. Hence the confusion. The variable will be set in decimal. So it needs to be converted to hexadecimal. So the code seems to be correct. That said, I think some clarifications is needed in the commit message to help understanding the code and at least giving a hint whether this code was ever tested (or this was introduced by a follow-up change). Cheers, -- Julien Grall
|
Lists.xenproject.org is hosted with RackSpace, monitoring our |