[PATCH] net: dsa: Fix OF fallback lookup for ports

The variable 'node' was already invalid, so using it for further lookup will not work. Signed-off-by: Marcus Comstedt --- net/dsa-uclass.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/net/dsa-uclass.c b/net/dsa-uclass.c index dd78e5744d..f64c68e340 100644 ---

Re: [PATCH] libfdt: Remove superfluous NUL character from overlay symbols

Hi Simon, Simon Glass writes: > You can send your patch to the mailing list on cc. > > Also see README.fdt-control fo the the git repo and other details. Thanks. Sent. // Marcus

Re: [PATCH] libfdt: Remove superfluous NUL character from overlay symbols

Hi Simon, Simon Glass writes: > Could you please send this patch upstream? Sure. But where is that, exactly? // Marcus

[PATCH] libfdt: Remove superfluous NUL character from overlay symbols

Signed-off-by: Marcus Comstedt --- scripts/dtc/libfdt/fdt_overlay.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/scripts/dtc/libfdt/fdt_overlay.c b/scripts/dtc/libfdt/fdt_overlay.c index c090e6991e..bd75e3dd78 100644 --- a/scripts/dtc/libfdt/fdt_overlay.c +++ b/scripts/

Re: [U-Boot] [PATCH] clk: sifive: Fix ethernet regression on HiFive Unleashed

Hi Bin, Bin Meng writes: > So 4.14 definitely was an out-of-tree kernel Everything before 5.2 was out-of-tree. > No one can guarantee an out-of-tree implementation will be > keeping compatible after it's accepted in-tree. Reviewers/maintainers can guarantee compatibility with existing hardw

Re: [U-Boot] [PATCH] clk: sifive: Fix ethernet regression on HiFive Unleashed

Hi Bin, Bin Meng writes: >> Yes, but in which Linux? This whole thing started because U-Boot will > > The latest Linux kernel v5.3. Thanks. I'll try it later. As if to prove my point, the one from 5.2.11 (which is the Linux version I'm currently running) did not work (I got MMC but no ether

Re: [U-Boot] [PATCH] clk: sifive: Fix ethernet regression on HiFive Unleashed

Hi Bin, Bin Meng writes: >> And which "Linux DTB" should that be? Since RISC-V seems to be going > > The one in arch/riscv/boot/dts/sifive. Yes, but in which Linux? This whole thing started because U-Boot will no longer accept the DTB used with Linux 4.14. If each Linux version is going to

Re: [U-Boot] [PATCH] clk: sifive: Fix ethernet regression on HiFive Unleashed

Bin Meng writes: > OpenSBI v0.4 works fine if you supply the the Linux DTB via > FW_PAYLOAD_FDT_PATH when building OpenSBI. And which "Linux DTB" should that be? Since RISC-V seems to be going the same annoying way as ARM with a requirement that you switch DTB every time you switch kernel vers

Re: [U-Boot] [PATCH] clk: sifive: Fix ethernet regression on HiFive Unleashed

Hi Bin, Bin Meng writes: > I believe the officially supported booting workflow is to use the > latest DT from Linux and compile it into OpenSBI, then boot to U-Boot. > The original DT shipped by FSBL is not used by OpenSBI or U-Boot. Ok, if U-Boot uses a DT compiled into OpenSBI then the incom

[U-Boot] [PATCH] clk: sifive: Fix ethernet regression on HiFive Unleashed

names "rtcclk" as its parent. Signed-off-by: Marcus Comstedt Cc: Anup Patel --- drivers/clk/sifive/fu540-prci.c | 11 ++- 1 file changed, 2 insertions(+), 9 deletions(-) diff --git a/drivers/clk/sifive/fu540-prci.c b/drivers/clk/sifive/fu540-prci.c index ce0769f2d1..78a9fe3770 10

[U-Boot] [PATCH] riscv: update fix_rela_dyn

The addend is now added for RELOC_TYPE relocs. Also, changed the loop structure so that all the R_RISCV_RELATIVE relocs are not required to be at the beginning of the list. Signed-off-by: Marcus Comstedt Cc: Rick Chen --- arch/riscv/cpu/start.S | 10 +- 1 file changed, 5 insertions

[U-Boot] [PATCH] riscv: tools: Handle addend to absolute reloc in prelink-riscv

80250970 R_RISCV_64systab+0x0060 In these cases the addend needs to be added to the symbol value to get the correct value for the reloc. Signed-off-by: Marcus Comstedt Cc: Rick Chen --- Note: This patch applies on top of the previous endianness patches. While the bug is

Re: [U-Boot] [PATCH v2 2/2] riscv: tools: Add big endian target support to prelink-riscv

Hi Rick, Rick Chen writes: > Thanks for your efforts about the support for big endian ELFs. > But I don't have big endian toolchain to verify. I don't think anyone has a big endian toolchain for riscv at this juncture. :-) (No specification for big/bi endian has been nailed down yet, AFAIK,

[U-Boot] [PATCH v2 2/2] riscv: tools: Add big endian target support to prelink-riscv

Signed-off-by: Marcus Comstedt Cc: Rick Chen --- Changes for v2: - Added tools/prelink-riscv.c | 34 ++ tools/prelink-riscv.inc | 62 ++--- 2 files changed, 63 insertions(+), 33 deletions(-) diff --git a/tools/prelink-riscv.c b

[U-Boot] [PATCH v2 1/2] riscv: tools: Fix prelink-riscv to work on big endian hosts

All ELF fields whose values are inspected by the code are converted to CPU byteorder first. Values which are copied verbatim (relocation fixups) are not swapped to CPU byteorder and back as it is not needed. Signed-off-by: Marcus Comstedt Cc: Rick Chen --- Changes for v2: - Now #undef:s

Re: [U-Boot] [PATCH] riscv: tools: Fix prelink-riscv to work on big endian hosts

Hi Rick, Rick Chen writes: > Nitpick: #undef lenn_to_cpu at the end of prelink-riscv.inc. > > Other than that, > Reviewed-by: Rick Chen Thanks for reviewing! Should I post a new version of the patch which adds the missing #undef? In the meantime I have also implemented support for big endi

[U-Boot] [PATCH] riscv: tools: Fix prelink-riscv to work on big endian hosts

big endian RISC-V targets as well, the calls to le*_to_cpu need to be replaced with macros target*_to_cpu defined by prelink-riscv.c, and prelink-riscv.inc included four times (32le, 64le, 32be, 32be) instead of two. Signed-off-by: Marcus Comstedt Cc: Rick Chen --- tools/prelink-riscv.c | 5