Linux-fbdev Archive mirror
 help / color / mirror / Atom feed
From: Andreas Larsson <andreas@gaisler.com>
To: sam@ravnborg.org, "David S. Miller" <davem@davemloft.net>,
	Arnd Bergmann <arnd@kernel.org>
Cc: Helge Deller <deller@gmx.de>,
	Alexander Viro <viro@zeniv.linux.org.uk>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Alan Stern <stern@rowland.harvard.edu>,
	Jaroslav Kysela <perex@perex.cz>, Takashi Iwai <tiwai@suse.com>,
	sparclinux@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-usb@vger.kernel.org, linux-fbdev@vger.kernel.org,
	dri-devel@lists.freedesktop.org, linux-sound@vger.kernel.org
Subject: Re: [PATCH 17/27] sparc32: Drop run-time patching of ASI instructions
Date: Fri, 16 Feb 2024 18:12:08 +0100	[thread overview]
Message-ID: <d1a616f1-0fd2-4af0-8b89-e0d0a8842a6d@gaisler.com> (raw)
In-Reply-To: <20231219-sam-sparc32-sunset-v3-v1-17-64bb44b598c5@ravnborg.org>

On 2023-12-19 23:03, Sam Ravnborg via B4 Relay wrote:
> From: Sam Ravnborg <sam@ravnborg.org>
> 
> With only LEON supported there is no need to run-time patch
> the instructions to match ASI.
> 
> Move a few functions back to C with inline asm, now that
> run-time patching is not needed.
> 
> Deleted a few functions that turns out not to be used rather
> than re-implement them in C.
...
> diff --git a/arch/sparc/include/asm/sections.h b/arch/sparc/include/asm/sections.h
> index 08f833453ab3..e9d28148850b 100644
> --- a/arch/sparc/include/asm/sections.h
> +++ b/arch/sparc/include/asm/sections.h
> @@ -8,7 +8,4 @@
>  /* sparc entry point */
>  extern char _start[];
>  
> -extern char __leon_1insn_patch[];
> -extern char __leon_1insn_patch_end[];
> -
>  #endif
> diff --git a/arch/sparc/include/asm/winmacro.h b/arch/sparc/include/asm/winmacro.h
> index b6e911f5d93c..c496b04cdfaf 100644
> --- a/arch/sparc/include/asm/winmacro.h
> +++ b/arch/sparc/include/asm/winmacro.h
> @@ -108,18 +108,11 @@
>  661:	rd	%tbr, %idreg;				\
>  	srl	%idreg, 10, %idreg;			\
>  	and	%idreg, 0xc, %idreg;			\

These three lines, including the label, should also be removed as they
are not for LEON. Additionally, I think it would be best to split out
removing the cpuid instruction fixups to one patch and the MMU ASI
instruction fixups to another patch.

> -	.section	.cpuid_patch, "ax";		\
> -	/* Instruction location. */			\
> -	.word		661b;				\
> -	/* SUN4D implementation. */			\
> -	lda	 [%g0] ASI_M_VIKING_TMP1, %idreg;	\
> -	sll	 %idreg, 2, %idreg;			\
> -	nop;						\
> -	/* LEON implementation. */			\
> +							\
>  	rd 	%asr17, %idreg;				\
>  	srl	%idreg, 0x1c, %idreg;			\
>  	sll	%idreg, 0x02, %idreg;			\
> -	.previous;					\
> +							\
>  	sethi    %hi(current_set), %dest_reg; 		\
>  	or       %dest_reg, %lo(current_set), %dest_reg;\
>  	ld       [%idreg + %dest_reg], %dest_reg;
> diff --git a/arch/sparc/kernel/entry.S b/arch/sparc/kernel/entry.S
> index 0f2417ee3f95..9cf8f87e8c42 100644
> --- a/arch/sparc/kernel/entry.S
> +++ b/arch/sparc/kernel/entry.S

The hard_smp_processor_id function also needs to be reduced to just the
LEON code. With the patching removed, SMP otherwise breaks with CPUs
other than CPU 0 getting stuck.

Thanks,
Andreas

  reply	other threads:[~2024-02-16 17:12 UTC|newest]

Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-12-19 22:03 [PATCH 00/27] sparc32: sunset sun4m and sun4d Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 01/27] sparc32: Update defconfig to LEON SMP Sam Ravnborg via B4 Relay
2023-12-19 22:23   ` Arnd Bergmann
2023-12-20  6:43     ` Sam Ravnborg
2023-12-20  8:37       ` Arnd Bergmann
2023-12-19 22:03 ` [PATCH 02/27] sparc32: Drop sun4m/sun4d support from head_32.S Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 03/27] sparc32: Drop floppy support Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 04/27] sparc32: Drop sun4m specific led driver Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 05/27] sparc32: Drop sun specific power management drivers Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 06/27] sparc32: Drop auxio support Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 07/27] sparc32: Drop run-time patching of ipi trap Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 08/27] sparc32: Drop patching of interrupt vector Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 09/27] sparc32: Drop sun4m/sun4d specific irq handling Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 10/27] sparc32: Drop sun4d/sun4m smp support Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 11/27] sparc32: Drop pcic support Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 12/27] sparc32: Drop mbus support Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 13/27] sparc32: Drop unused function __get_{phys,iospace} Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 14/27] sparc32: Drop unused mmu models Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 15/27] sparc32: Drop check for sparc_model Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 16/27] sparc32: Drop use of sparc_config Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 17/27] sparc32: Drop run-time patching of ASI instructions Sam Ravnborg via B4 Relay
2024-02-16 17:12   ` Andreas Larsson [this message]
2024-02-17 11:59     ` Sam Ravnborg
2023-12-19 22:03 ` [PATCH 18/27] sparc32: Drop support for 7 register windows Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 19/27] sparc32: Drop additional sun4d bits Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 20/27] sparc32: Drop unused prom ranges support Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 21/27] sparc32: Drop unused iommu support Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 22/27] sparc32: Drop sun4m irq support Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 23/27] sparc32: Drop unused trampoline code Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 24/27] sparc32: Drop config SPARC_LEON Sam Ravnborg via B4 Relay
2023-12-20  5:52   ` Greg Kroah-Hartman
2023-12-19 22:03 ` [PATCH 25/27] sparc32: Drop sbus support Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 26/27] sbus: char: Drop now unused uctrl driver Sam Ravnborg via B4 Relay
2023-12-19 22:03 ` [PATCH 27/27] fbdev/p9100: Drop now unused driver p9100 Sam Ravnborg via B4 Relay
2024-01-02  8:26   ` Thomas Zimmermann
2024-01-02 16:31     ` Sam Ravnborg
2023-12-20  8:36 ` [PATCH 00/27] sparc32: sunset sun4m and sun4d Arnd Bergmann
2023-12-20  9:28   ` John Paul Adrian Glaubitz
2023-12-20  9:34   ` David Laight
2023-12-20  9:51     ` Arnd Bergmann
2023-12-20  9:54   ` John Paul Adrian Glaubitz
2023-12-20 10:47     ` Arnd Bergmann
2023-12-20 11:30       ` Mark Cave-Ayland
2023-12-20 15:22         ` Sam Ravnborg
2023-12-20 17:25           ` John Paul Adrian Glaubitz
2023-12-22 15:27             ` Andreas Larsson
2023-12-20 14:42 ` Kjetil Oftedal
2024-02-04 19:21 ` Sam Ravnborg
2024-02-05  5:12   ` John Paul Adrian Glaubitz

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=d1a616f1-0fd2-4af0-8b89-e0d0a8842a6d@gaisler.com \
    --to=andreas@gaisler.com \
    --cc=arnd@kernel.org \
    --cc=davem@davemloft.net \
    --cc=deller@gmx.de \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=gregkh@linuxfoundation.org \
    --cc=linux-fbdev@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-sound@vger.kernel.org \
    --cc=linux-usb@vger.kernel.org \
    --cc=perex@perex.cz \
    --cc=sam@ravnborg.org \
    --cc=sparclinux@vger.kernel.org \
    --cc=stern@rowland.harvard.edu \
    --cc=tiwai@suse.com \
    --cc=viro@zeniv.linux.org.uk \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).