Bug#929155: An Odd Error Message on GNU assembler
Package: binutils-arm-none-eabi
Version: 2.28-5+9+b3
Executing GNU assembler, "arm-none-eabi-as", in this package,
-
Error: register list must contain at least 1 and at most 16 registers --
`vpush {s0-s24}'
Error: register list must contain at least 1 and at most 16 registers --
`vpop {s0-s24}'
-
About "vpush" and "vpop" instructions; "s", single precision floating point
registers, don't have limitation at most 16.
This limitation is for "d", double precision floating point registers.
The older version 2.27-9+9 doesn't have this odd error message and going
well.
Thanks reading this report.
Bug#929155: An Odd Error Message on GNU assembler
Hi!
I'm looking on this and I find my self needing more information to
understand.
On Sat, 18 May 2019 18:50:32 +0900 Kenta Ishii wrote:
> Package: binutils-arm-none-eabi
> Version: 2.28-5+9+b3
>
> Executing GNU assembler, "arm-none-eabi-as", in this package,
>
>
-
>
> Error: register list must contain at least 1 and at most 16 registers --
> `vpush {s0-s24}'
> Error: register list must contain at least 1 and at most 16 registers --
> `vpop {s0-s24}'
>
>
-
>
Whats the target cpu? are you using any specific flag?
> About "vpush" and "vpop" instructions; "s", single precision floating
point
> registers, don't have limitation at most 16.
> This limitation is for "d", double precision floating point registers.
>
> The older version 2.27-9+9 doesn't have this odd error message and going
> well.
>
> Thanks reading this report.
Thanks!
Joa
Bug#929155: An Odd Error Message on GNU assembler
Hi Joa!
Options to execute "arm-none-eabi-as" in this issue are as follows.
About CPU: "-mcpu=arm1176jzf-s", "-mcpu=cortex-a7", or "-mcpu=cortex-a53"
About FPU: "-mfpu=vfp" on all CPU types
Note that these are 32-bit ARM.
"vstmdb sp!, {s0-s24}", the synonym of "vpush {s0-s24}" can pass without
this error.
"vldm sp!, {s0-s24}", the synonym of "vpop {s0-s24}" can pass without this
error too.
I think this issue occurs on the syntax checker of GNU assembler.
Thanks!
Bug#929155: An Odd Error Message on GNU assembler
Hi Kenta,
On 5/23/19 10:10 AM, Kenta Ishii wrote:
> Hi Joa!
>
> Options to execute "arm-none-eabi-as" in this issue are as follows.
>
> About CPU: "-mcpu=arm1176jzf-s", "-mcpu=cortex-a7", or "-mcpu=cortex-a53"
You provided 1 error report, and here you list 3 CPUs.
Which CPU the error belong to? All of them?
It makes sens you get this on the arm1176jzf-s, because the ARMv6 arch
doesn't provide the VLDM/VSTMDB instructions.
Per the datasheet "ARM DDI 0301H ID012310":
ARM1176JZF-S processors support: PKHBT, PKHTB, QADD16, QADD8, QADDSUBX,
QSUB16, QSUB8, QSUBADDX, SADD16, SADD8, SADDSUBX, SEL, SHADD16, SHADD8,
SHADDSUBX, SHSUB16, SHSUB8, SHSUBADDX, SSAT, SSAT16, SSUB16, SSUB8,
SSUBADDX, SXTAB16, SXTB16, UADD16, UADD8, UADDSUBX, UHADD16, UHADD8,
UHADDSUBX, UHSUB16, UHSUB8, UHSUBADDX, UQADD16, UQADD8, UQADDSUBX,
UQSUB16, UQSUB8, UQSUBADDX, USAD8, USADA8, USAT, USAT16, USUB16, USUB8,
USUBADDX, UXTAB16, UXTB16
Regards,
Phil.
> About FPU: "-mfpu=vfp" on all CPU types
>
> Note that these are 32-bit ARM.
>
> "vstmdb sp!, {s0-s24}", the synonym of "vpush {s0-s24}" can pass without
> this error.
> "vldm sp!, {s0-s24}", the synonym of "vpop {s0-s24}" can pass without
> this error too.
>
> I think this issue occurs on the syntax checker of GNU assembler.
>
> Thanks!
>
Bug#929155: An Odd Error Message on GNU assembler
Hi Phil! All CPU makes the error message on "vpush/vpop" (full descending stack), but the synonym "stmdb/ldm" doesn't make the error message. Later Releases of ARMv6 supports VFP, so VFP instructions are executable. CPUs are BCM2835 (arm1176jzf-s), BCM2836 (cortex-a7), and BCM2837 (cortex-a53). Thanks! > Which CPU the error belong to? All of them? > It makes sens you get this on the arm1176jzf-s, because the ARMv6 arch > doesn't provide the VLDM/VSTMDB instructions. > > Per the datasheet "ARM DDI 0301H ID012310": > > ARM1176JZF-S processors support: PKHBT, PKHTB, QADD16, QADD8, QADDSUBX, > QSUB16, QSUB8, QSUBADDX, SADD16, SADD8, SADDSUBX, SEL, SHADD16, SHADD8, > SHADDSUBX, SHSUB16, SHSUB8, SHSUBADDX, SSAT, SSAT16, SSUB16, SSUB8, > SSUBADDX, SXTAB16, SXTB16, UADD16, UADD8, UADDSUBX, UHADD16, UHADD8, > UHADDSUBX, UHSUB16, UHSUB8, UHSUBADDX, UQADD16, UQADD8, UQADDSUBX, > UQSUB16, UQSUB8, UQSUBADDX, USAD8, USADA8, USAT, USAT16, USUB16, USUB8, > USUBADDX, UXTAB16, UXTB16
