[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20191101220939.GK25745@shell.armlinux.org.uk>
Date: Fri, 1 Nov 2019 22:09:39 +0000
From: Russell King - ARM Linux admin <linux@...linux.org.uk>
To: Stefan Agner <stefan@...er.ch>
Cc: ndesaulniers@...gle.com, nico@...xnic.net, rfranz@...vell.com,
linus.walleij@...aro.org, ard.biesheuvel@...aro.org,
linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
clang-built-linux@...glegroups.com
Subject: Re: [PATCH] ARM: use APSR_nzcv instead of r15 as mrc operand
On Fri, Nov 01, 2019 at 10:47:58PM +0100, Stefan Agner wrote:
> LLVM's integrated assembler does not accept r15 as mrc operand.
> arch/arm/boot/compressed/head.S:1267:16: error: operand must be a register in range [r0, r14] or apsr_nzcv
> 1: mrc p15, 0, r15, c7, c14, 3 @ test,clean,invalidate D cache
> ^
>
> Use APSR_nzcv instead of r15. The GNU assembler supports this
> syntax since binutils 2.21 [0].
>
> [0] https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;a=commit;h=db472d6ff0f438a21b357249a9b48e4b74498076
>
> Signed-off-by: Stefan Agner <stefan@...er.ch>
Looks fine, please put it in the patch system; however, please note
that I've been tweaking the patch system over the last week (mainly
with the database, which has impacted almost everything) so there
may be issues that I've not yet found...
Thanks.
> ---
> arch/arm/boot/compressed/head.S | 2 +-
> arch/arm/mm/proc-arm1026.S | 4 ++--
> arch/arm/mm/proc-arm926.S | 4 ++--
> 3 files changed, 5 insertions(+), 5 deletions(-)
>
> diff --git a/arch/arm/boot/compressed/head.S b/arch/arm/boot/compressed/head.S
> index 15ecad944847..ead21e5f2b80 100644
> --- a/arch/arm/boot/compressed/head.S
> +++ b/arch/arm/boot/compressed/head.S
> @@ -1273,7 +1273,7 @@ iflush:
> __armv5tej_mmu_cache_flush:
> tst r4, #1
> movne pc, lr
> -1: mrc p15, 0, r15, c7, c14, 3 @ test,clean,invalidate D cache
> +1: mrc p15, 0, APSR_nzcv, c7, c14, 3 @ test,clean,invalidate D cache
> bne 1b
> mcr p15, 0, r0, c7, c5, 0 @ flush I cache
> mcr p15, 0, r0, c7, c10, 4 @ drain WB
> diff --git a/arch/arm/mm/proc-arm1026.S b/arch/arm/mm/proc-arm1026.S
> index 10e21012380b..0bdf25a95b10 100644
> --- a/arch/arm/mm/proc-arm1026.S
> +++ b/arch/arm/mm/proc-arm1026.S
> @@ -138,7 +138,7 @@ ENTRY(arm1026_flush_kern_cache_all)
> mov ip, #0
> __flush_whole_cache:
> #ifndef CONFIG_CPU_DCACHE_DISABLE
> -1: mrc p15, 0, r15, c7, c14, 3 @ test, clean, invalidate
> +1: mrc p15, 0, APSR_nzcv, c7, c14, 3 @ test, clean, invalidate
> bne 1b
> #endif
> tst r2, #VM_EXEC
> @@ -363,7 +363,7 @@ ENTRY(cpu_arm1026_switch_mm)
> #ifdef CONFIG_MMU
> mov r1, #0
> #ifndef CONFIG_CPU_DCACHE_DISABLE
> -1: mrc p15, 0, r15, c7, c14, 3 @ test, clean, invalidate
> +1: mrc p15, 0, APSR_nzcv, c7, c14, 3 @ test, clean, invalidate
> bne 1b
> #endif
> #ifndef CONFIG_CPU_ICACHE_DISABLE
> diff --git a/arch/arm/mm/proc-arm926.S b/arch/arm/mm/proc-arm926.S
> index 3188ab2bac61..1ba253c2bce1 100644
> --- a/arch/arm/mm/proc-arm926.S
> +++ b/arch/arm/mm/proc-arm926.S
> @@ -131,7 +131,7 @@ __flush_whole_cache:
> #ifdef CONFIG_CPU_DCACHE_WRITETHROUGH
> mcr p15, 0, ip, c7, c6, 0 @ invalidate D cache
> #else
> -1: mrc p15, 0, r15, c7, c14, 3 @ test,clean,invalidate
> +1: mrc p15, 0, APSR_nzcv, c7, c14, 3 @ test,clean,invalidate
> bne 1b
> #endif
> tst r2, #VM_EXEC
> @@ -358,7 +358,7 @@ ENTRY(cpu_arm926_switch_mm)
> mcr p15, 0, ip, c7, c6, 0 @ invalidate D cache
> #else
> @ && 'Clean & Invalidate whole DCache'
> -1: mrc p15, 0, r15, c7, c14, 3 @ test,clean,invalidate
> +1: mrc p15, 0, APSR_nzcv, c7, c14, 3 @ test,clean,invalidate
> bne 1b
> #endif
> mcr p15, 0, ip, c7, c5, 0 @ invalidate I cache
> --
> 2.23.0
>
>
--
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTC broadband for 0.8mile line in suburbia: sync at 12.1Mbps down 622kbps up
According to speedtest.net: 11.9Mbps down 500kbps up
Powered by blists - more mailing lists