Skip to contentSkip to conversation

dead(?) MIPS config stuff

DISCUSSION Ralf Baechle 7 participants 13 messages Jul 5, 2007 – Jul 12, 2007 MIPS

lore
Share & tools
lore
#0 lore raw

----- Forwarded message from "Robert P. J. Day" <[email protected]> -----

From: "Robert P. J. Day" <[email protected]>
Date: Thu, 5 Jul 2007 06:38:14 -0400 (EDT)
To: Ralf Baechle <[email protected]>
Subject: dead(?) MIPS config stuff
Content-Type: TEXT/PLAIN; charset=US-ASCII

a brute force run of my latest dead CONFIG variable script, you can
decide if any of it is of interest.

========== AU1000_SRC_CLK ==========
arch/mips/au1000/common/time.c:206:#ifdef CONFIG_AU1000_SRC_CLK
arch/mips/au1000/common/time.c:207:#define AU1000_SRC_CLK CONFIG_AU1000_SRC_CLK
arch/mips/au1000/common/time.c:207:#define AU1000_SRC_CLK CONFIG_AU1000_SRC_CLK
arch/mips/au1000/common/time.c:209:#define AU1000_SRC_CLK 12000000
arch/mips/au1000/common/time.c:275: AU1000_SRC_CLK;
arch/mips/au1000/common/time.c:283: cpu_speed = (au_readl(SYS_CPUPLL) & 0x0000003f) * AU1000_SRC_CLK;
========== AU1000_USE32K ==========
arch/mips/au1000/common/time.c:250:#if defined(CONFIG_AU1000_USE32K)
========== AU1XXX_PSC_SPI ==========
arch/mips/au1000/pb1200/board_setup.c:134:#if defined(CONFIG_AU1XXX_PSC_SPI) && defined(CONFIG_I2C_AU1550) arch/mips/au1000/pb1200/board_setup.c:137:#elif defined( CONFIG_AU1XXX_PSC_SPI )
========== CPU_R4000 ==========
arch/mips/kernel/Makefile:20:obj-$(CONFIG_CPU_R4000) += r4k_fpu.o r4k_switch.o
arch/mips/kernel/cpu-bugs64.c:159:#ifndef CONFIG_CPU_R4000
arch/mips/kernel/cpu-bugs64.c:237:#if !defined(CONFIG_CPU_R4000) && !defined(CONFIG_CPU_R4400) arch/mips/kernel/cpu-bugs64.c:307:#if !defined(CONFIG_CPU_R4000) && !defined(CONFIG_CPU_R4400)
========== CPU_R4400 ==========
arch/mips/kernel/cpu-bugs64.c:237:#if !defined(CONFIG_CPU_R4000) && !defined(CONFIG_CPU_R4400) arch/mips/kernel/cpu-bugs64.c:307:#if !defined(CONFIG_CPU_R4000) && !defined(CONFIG_CPU_R4400)
========== CPU_SR71000 ==========
arch/mips/momentum/ocelot_c/prom.c:36:#ifdef CONFIG_CPU_SR71000 arch/mips/momentum/ocelot_c/setup.c:215:#ifdef CONFIG_CPU_SR71000 arch/mips/momentum/ocelot_c/setup.c:271:#ifdef CONFIG_CPU_SR71000
arch/mips/kernel/cpu-probe.c:702: c->cputype = CPU_SR71000;
arch/mips/kernel/proc.c:85: [CPU_SR71000] = "Sandcraft SR71000",
include/asm-mips/cpu.h:191:#define CPU_SR71000 53
========== EXCITE_FCAP_GPI ==========
arch/mips/basler/excite/excite_device.c:164:#if defined(CONFIG_EXCITE_FCAP_GPI) || defined(CONFIG_EXCITE_FCAP_GPI_MODULE) arch/mips/basler/excite/excite_device.c:227:#endif /* defined(CONFIG_EXCITE_FCAP_GPI) || defined(CONFIG_EXCITE_FCAP_GPI_MODULE) */ arch/mips/basler/excite/excite_device.c:164:#if defined(CONFIG_EXCITE_FCAP_GPI) || defined(CONFIG_EXCITE_FCAP_GPI_MODULE) arch/mips/basler/excite/excite_device.c:227:#endif /* defined(CONFIG_EXCITE_FCAP_GPI) || defined(CONFIG_EXCITE_FCAP_GPI_MODULE) */ ========== FB_XPERT98 ==========
arch/mips/au1000/common/setup.c:109:#ifdef CONFIG_FB_XPERT98
========== HT_LEVEL_TRIGGER ==========
arch/mips/pmc-sierra/yosemite/irq.c:77:#ifdef CONFIG_HT_LEVEL_TRIGGER arch/mips/pmc-sierra/yosemite/irq.c:106:#endif /* CONFIG_HT_LEVEL_TRIGGER */
========== MIPS_HYDROGEN3 ==========
arch/mips/au1000/common/setup.c:103:#ifdef CONFIG_MIPS_HYDROGEN3
========== PCMCIA_XXS1500 ==========
arch/mips/au1000/xxs1500/board_setup.c:66:#ifdef CONFIG_PCMCIA_XXS1500
========== RM9K_GE ==========
arch/mips/basler/excite/excite_device.c:314:#if defined(CONFIG_RM9K_GE) || defined(CONFIG_RM9K_GE_MODULE) arch/mips/basler/excite/excite_device.c:367:#endif /* defined(CONFIG_RM9K_GE) || defined(CONFIG_RM9K_GE_MODULE) */ arch/mips/basler/excite/excite_device.c:314:#if defined(CONFIG_RM9K_GE) || defined(CONFIG_RM9K_GE_MODULE) arch/mips/basler/excite/excite_device.c:367:#endif /* defined(CONFIG_RM9K_GE) || defined(CONFIG_RM9K_GE_MODULE) */ ========== SIBYTE_BCM1480_PROF ==========
arch/mips/sibyte/bcm1480/irq.c:460:#ifdef CONFIG_SIBYTE_BCM1480_PROF arch/mips/sibyte/bcm1480/irq.c:467:#ifdef CONFIG_SIBYTE_BCM1480_PROF
========== SIBYTE_SB1250_DUART ========== arch/mips/configs/sb1250-swarm_defconfig:665:CONFIG_SIBYTE_SB1250_DUART=y arch/mips/configs/bigsur_defconfig:673:CONFIG_SIBYTE_SB1250_DUART=y
arch/mips/sibyte/bcm1480/irq.c:79:#ifdef CONFIG_SIBYTE_SB1250_DUART
arch/mips/sibyte/bcm1480/irq.c:407:#ifdef CONFIG_SIBYTE_SB1250_DUART
arch/mips/sibyte/sb1250/irq.c:64:#ifdef CONFIG_SIBYTE_SB1250_DUART
arch/mips/sibyte/sb1250/irq.c:362:#ifdef CONFIG_SIBYTE_SB1250_DUART
arch/mips/sibyte/cfe/console.c:49:#ifdef CONFIG_SIBYTE_SB1250_DUART
========== SMTC_IDLE_HOOK_DEBUG ==========
arch/mips/kernel/smtc.c:146:#ifdef CONFIG_SMTC_IDLE_HOOK_DEBUG
arch/mips/kernel/smtc.c:181:#endif /* CONFIG_SMTC_IDLE_HOOK_DEBUG */
arch/mips/kernel/smtc.c:399:#ifdef CONFIG_SMTC_IDLE_HOOK_DEBUG
arch/mips/kernel/smtc.c:402:#endif /* CONFIG_SMTC_IDLE_HOOK_DEBUG */
arch/mips/kernel/smtc.c:614:#ifdef CONFIG_SMTC_IDLE_HOOK_DEBUG
arch/mips/kernel/smtc.c:828:#ifdef CONFIG_SMTC_IDLE_HOOK_DEBUG
arch/mips/kernel/smtc.c:830:#endif /* CONFIG_SMTC_IDLE_HOOK_DEBUG */
arch/mips/kernel/smtc.c:1055:#ifdef CONFIG_SMTC_IDLE_HOOK_DEBUG
arch/mips/kernel/smtc.c:1148:#endif /* CONFIG_SMTC_IDLE_HOOK_DEBUG */
arch/mips/kernel/process.c:54:#ifdef CONFIG_SMTC_IDLE_HOOK_DEBUG
arch/mips/Kconfig.debug:40:config CONFIG_SMTC_IDLE_HOOK_DEBUG
========== SOUND_AU1X00 ==========
arch/mips/au1000/common/setup.c:116:#if defined(CONFIG_SOUND_AU1X00) && !defined(CONFIG_SOC_AU1000) ========== TX4927BUG_WORKAROUND ========== arch/mips/tx4927/toshiba_rbtx4927/toshiba_rbtx4927_setup.c:141:#define CONFIG_TX4927BUG_WORKAROUND arch/mips/tx4927/toshiba_rbtx4927/toshiba_rbtx4927_setup.c:672:#ifdef CONFIG_TX4927BUG_WORKAROUND arch/mips/tx4927/toshiba_rbtx4927/toshiba_rbtx4927_setup.c:914:#ifdef CONFIG_TX4927BUG_WORKAROUND
========== USB_OHCI ==========
arch/mips/au1000/pb1000/board_setup.c:57:#ifdef CONFIG_USB_OHCI arch/mips/au1000/pb1000/board_setup.c:105:#ifdef CONFIG_USB_OHCI arch/mips/au1000/pb1000/board_setup.c:119:#endif // defined (CONFIG_USB_OHCI) arch/mips/au1000/pb1500/board_setup.c:59:#ifdef CONFIG_USB_OHCI arch/mips/au1000/pb1500/board_setup.c:88:#ifdef CONFIG_USB_OHCI arch/mips/au1000/pb1500/board_setup.c:98:#endif // defined (CONFIG_USB_OHCI) arch/mips/au1000/pb1100/board_setup.c:57:#ifdef CONFIG_USB_OHCI arch/mips/au1000/pb1100/board_setup.c:101:#endif // defined (CONFIG_USB_OHCI) arch/mips/au1000/mtx-1/board_setup.c:57:#ifdef CONFIG_USB_OHCI
arch/mips/au1000/mtx-1/board_setup.c:61:#endif // defined (CONFIG_USB_OHCI)

Signature · 4 lines
Some of the stuff does exist intentionally such as the R4000 bits.  Much
of the remainder however seems to be waiting to be chainsawed.

  Ralf
#1 lore raw

... all CONFIG output snipped ...

in any event, i'll leave all that output in your capable hands, and you can deal with it any way you wish. ciao.

rday

Signature · 7 lines
========================================================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry
Waterloo, Ontario, CANADA

http://fsdev.net/wiki/index.php?title=Main_Page
========================================================================
#2 lore raw
Quoted 4 lines · Robert P. J. Day

... all CONFIG output snipped ...

in any event, i'll leave all that output in your capable hands, and you can deal with it any way you wish. ciao.

Are you planning to contribute your script for the kernel's scripts/ directory?

Ralf

#4 lore raw
In reply to Ralf Baechle
Quoted 8 lines · Ralf Baechle

========== CPU_R4000 ==========
arch/mips/kernel/Makefile:20:obj-$(CONFIG_CPU_R4000) += r4k_fpu.o r4k_switch.o
arch/mips/kernel/cpu-bugs64.c:159:#ifndef CONFIG_CPU_R4000 arch/mips/kernel/cpu-bugs64.c:237:#if !defined(CONFIG_CPU_R4000) && !defined(CONFIG_CPU_R4400) arch/mips/kernel/cpu-bugs64.c:307:#if !defined(CONFIG_CPU_R4000) && !defined(CONFIG_CPU_R4400) ========== CPU_R4400 ==========
arch/mips/kernel/cpu-bugs64.c:237:#if !defined(CONFIG_CPU_R4000) && !defined(CONFIG_CPU_R4400) arch/mips/kernel/cpu-bugs64.c:307:#if !defined(CONFIG_CPU_R4000) && !defined(CONFIG_CPU_R4400)

I should at last port that patch with a workaround for R4000/R4400 errata
-- it has been long overdue now. Let's make it the item #3 on my to-do list. ;-) I'm not sure that'll keep the names of the options though.

Thanks for the reminder.

Maciej

#3 lore raw
In reply to Ralf Baechle
Quoted 9 lines · Ralf Baechle
Quoted 4 lines · Robert P. J. Day

... all CONFIG output snipped ...

in any event, i'll leave all that output in your capable hands, and you can deal with it any way you wish. ciao.

Are you planning to contribute your script for the kernel's scripts/ directory?

ah, i'd never thought of that. after i clean it up, sure.

rday

Signature · 7 lines
========================================================================
Robert P. J. Day
Linux Consulting, Training and Annoying Kernel Pedantry
Waterloo, Ontario, CANADA

http://fsdev.net/wiki/index.php?title=Main_Page
========================================================================
#5 lore raw
In reply to Ralf Baechle

Hello.

Quoted 12 lines · Ralf Baechle

a brute force run of my latest dead CONFIG variable script, you can
decide if any of it is of interest.

========== AU1000_SRC_CLK ========== arch/mips/au1000/common/time.c:206:#ifdef CONFIG_AU1000_SRC_CLK
arch/mips/au1000/common/time.c:207:#define AU1000_SRC_CLK CONFIG_AU1000_SRC_CLK
arch/mips/au1000/common/time.c:207:#define AU1000_SRC_CLK CONFIG_AU1000_SRC_CLK
arch/mips/au1000/common/time.c:209:#define AU1000_SRC_CLK 12000000
arch/mips/au1000/common/time.c:275: AU1000_SRC_CLK;
arch/mips/au1000/common/time.c:283: cpu_speed = (au_readl(SYS_CPUPLL) & 0x0000003f) * AU1000_SRC_CLK;
========== AU1000_USE32K ========== arch/mips/au1000/common/time.c:250:#if defined(CONFIG_AU1000_USE32K)

Erm, maybe it's worth to declare these options instead?

Quoted 3 lines

========== AU1XXX_PSC_SPI ==========
arch/mips/au1000/pb1200/board_setup.c:134:#if defined(CONFIG_AU1XXX_PSC_SPI) && defined(CONFIG_I2C_AU1550) arch/mips/au1000/pb1200/board_setup.c:137:#elif defined( CONFIG_AU1XXX_PSC_SPI )

I think that CONFIG_AU1XXX_PSC_SPI needs to be changed to CONFIG_SPI_AU1550...

Quoted 2 lines

========== MIPS_HYDROGEN3 ========== arch/mips/au1000/common/setup.c:103:#ifdef CONFIG_MIPS_HYDROGEN3

Hm, hasn't Hydrogen3 support been removed?

Quoted 2 lines

========== PCMCIA_XXS1500 ========== arch/mips/au1000/xxs1500/board_setup.c:66:#ifdef CONFIG_PCMCIA_XXS1500

Actually, that PCMCIA driver is controlled by CONFIG_MIPS_XXS1500 itself,
so #ifdef/#endif should just go, leaving what's between them always compiled.

Quoted 3 lines

========== SIBYTE_BCM1480_PROF ========== arch/mips/sibyte/bcm1480/irq.c:460:#ifdef CONFIG_SIBYTE_BCM1480_PROF arch/mips/sibyte/bcm1480/irq.c:467:#ifdef CONFIG_SIBYTE_BCM1480_PROF

Hm, that could probably be killed. The referenced function isn't defined
anywhere. Same about CONFIG_SIBYTE_BCM1250_PROF which is here but the kernel won't build w/it anyway...

Quoted 8 lines

========== SIBYTE_SB1250_DUART ========== arch/mips/configs/sb1250-swarm_defconfig:665:CONFIG_SIBYTE_SB1250_DUART=y arch/mips/configs/bigsur_defconfig:673:CONFIG_SIBYTE_SB1250_DUART=y arch/mips/sibyte/bcm1480/irq.c:79:#ifdef CONFIG_SIBYTE_SB1250_DUART arch/mips/sibyte/bcm1480/irq.c:407:#ifdef CONFIG_SIBYTE_SB1250_DUART arch/mips/sibyte/sb1250/irq.c:64:#ifdef CONFIG_SIBYTE_SB1250_DUART arch/mips/sibyte/sb1250/irq.c:362:#ifdef CONFIG_SIBYTE_SB1250_DUART arch/mips/sibyte/cfe/console.c:49:#ifdef CONFIG_SIBYTE_SB1250_DUART

This belongs to drivers/char/sb1350_duart.c which is only maintained in
the Linux/MIPS tree...

Quoted 2 lines

========== SOUND_AU1X00 ==========
arch/mips/au1000/common/setup.c:116:#if defined(CONFIG_SOUND_AU1X00) && !defined(CONFIG_SOC_AU1000)

The OSS driver has been obsoleted. Kill this piece, it doesn't apply to
the ALSA driver.

Quoted 4 lines

========== TX4927BUG_WORKAROUND ========== arch/mips/tx4927/toshiba_rbtx4927/toshiba_rbtx4927_setup.c:141:#define CONFIG_TX4927BUG_WORKAROUND arch/mips/tx4927/toshiba_rbtx4927/toshiba_rbtx4927_setup.c:672:#ifdef CONFIG_TX4927BUG_WORKAROUND arch/mips/tx4927/toshiba_rbtx4927/toshiba_rbtx4927_setup.c:914:#ifdef CONFIG_TX4927BUG_WORKAROUND

This one shouldn't have been called CONFIG_* and might be just killed as
well....

Quoted 11 lines

========== USB_OHCI ========== arch/mips/au1000/pb1000/board_setup.c:57:#ifdef CONFIG_USB_OHCI arch/mips/au1000/pb1000/board_setup.c:105:#ifdef CONFIG_USB_OHCI arch/mips/au1000/pb1000/board_setup.c:119:#endif // defined (CONFIG_USB_OHCI) arch/mips/au1000/pb1500/board_setup.c:59:#ifdef CONFIG_USB_OHCI arch/mips/au1000/pb1500/board_setup.c:88:#ifdef CONFIG_USB_OHCI arch/mips/au1000/pb1500/board_setup.c:98:#endif // defined (CONFIG_USB_OHCI) arch/mips/au1000/pb1100/board_setup.c:57:#ifdef CONFIG_USB_OHCI arch/mips/au1000/pb1100/board_setup.c:101:#endif // defined (CONFIG_USB_OHCI) arch/mips/au1000/mtx-1/board_setup.c:57:#ifdef CONFIG_USB_OHCI arch/mips/au1000/mtx-1/board_setup.c:61:#endif // defined (CONFIG_USB_OHCI)

Should be changed to CONFIG_USB_OHCI_HCD.

WBR, Sergei

#7 lore raw
In reply to Ralf Baechle
Quoted 4 lines

========== SMTC_IDLE_HOOK_DEBUG ========== arch/mips/kernel/smtc.c:146:#ifdef CONFIG_SMTC_IDLE_HOOK_DEBUG arch/mips/kernel/smtc.c:181:#endif /* CONFIG_SMTC_IDLE_HOOK_DEBUG */ arch/mips/kernel/smtc.c:399:#ifdef CONFIG_SMTC_IDLE_HOOK_DEBUG

Quoted 8 lines

arch/mips/kernel/smtc.c:402:#endif /* CONFIG_SMTC_IDLE_HOOK_DEBUG */ arch/mips/kernel/smtc.c:614:#ifdef CONFIG_SMTC_IDLE_HOOK_DEBUG arch/mips/kernel/smtc.c:828:#ifdef CONFIG_SMTC_IDLE_HOOK_DEBUG arch/mips/kernel/smtc.c:830:#endif /* CONFIG_SMTC_IDLE_HOOK_DEBUG */ arch/mips/kernel/smtc.c:1055:#ifdef CONFIG_SMTC_IDLE_HOOK_DEBUG arch/mips/kernel/smtc.c:1148:#endif /* CONFIG_SMTC_IDLE_HOOK_DEBUG */ arch/mips/kernel/process.c:54:#ifdef CONFIG_SMTC_IDLE_HOOK_DEBUG arch/mips/Kconfig.debug:40:config CONFIG_SMTC_IDLE_HOOK_DEBUG

It's cruft, but it's cruft that does have some utility when porting SMTC to a new platform, as it helps identify interrupt management bugs. It needs to be eliminated or replaced with a revision of the interface between architecture and platform support for SMTC, toward which I've made some small progress recently, but I'd prefer not to see it disappear before reinforcements arrive.

Regards,

Kevin K.

#8 lore raw

if it makes it any easier, there's no need to continue to CC me on
these messages. there's nothing i'm going to do beyond just reporting what i found.

rday

#11 lore raw
In reply to Kevin D. Kissell
Quoted 5 lines · Kevin D. Kissell

It's cruft, but it's cruft that does have some utility when porting SMTC to a new platform, as it helps identify interrupt management bugs. It needs to be eliminated or replaced with a revision of the interface between architecture and platform support for SMTC, toward which I've made some small progress recently, but I'd prefer not to see it disappear before reinforcements arrive.

All this email was about were kernel configuration options that can't be enabled for some reason. Of course these automated checkers such as Robert's make the assumption that the CONFIG_* preprocessor namespace is rserved for the Kconfig system. In this particular case the reason is quite simple:

Quoted 2 lines

arch/mips/kernel/process.c:54:#ifdef CONFIG_SMTC_IDLE_HOOK_DEBUG arch/mips/Kconfig.debug:40:config CONFIG_SMTC_IDLE_HOOK_DEBUG

The kconfig system itself does prefix names with CONFIG_ so kconfig ends up defining CONFIG_CONFIG_SMTC_IDLE_HOOK_DEBUG (are we stuttuttering?) So this one is fixed trivially.

Ralf

#9 lore raw
In reply to Robert P. J. Day
Quoted 3 lines · Robert P. J. Day

if it makes it any easier, there's no need to continue to CC me on
these messages. there's nothing i'm going to do beyond just reporting what i found.

Bah, we want you to enjoy the email storm you started ;-)

Ralf

#12 lore raw
In reply to Ralf Baechle
Quoted 11 lines · Ralf Baechle

All this email was about were kernel configuration options that can't be enabled for some reason. Of course these automated checkers such as Robert's make the assumption that the CONFIG_* preprocessor namespace is rserved for the Kconfig system. In this particular case the reason is quite simple:

Quoted 2 lines

arch/mips/kernel/process.c:54:#ifdef CONFIG_SMTC_IDLE_HOOK_DEBUG arch/mips/Kconfig.debug:40:config CONFIG_SMTC_IDLE_HOOK_DEBUG

The kconfig system itself does prefix names with CONFIG_ so kconfig ends up defining CONFIG_CONFIG_SMTC_IDLE_HOOK_DEBUG (are we stuttuttering?)

From: Ralf Baechle <[email protected]>
Date: Thu, 5 Jul 2007 17:39:48 +0100
Subject: [MIPS] SMTC: Fix cut'n'paste bug in Kconfig.debug

This effectivly turned the SMTC_IDLE_HOOK_DEBUG debug option into a no-op.

arch/mips/Kconfig.debug +1 -1
index 72d5c19..3efe117 100644
@@ -37,7 +37,7 @@ config DEBUG_STACK_USAGE
 
 	  This option will slow down process creation somewhat.
 
-config CONFIG_SMTC_IDLE_HOOK_DEBUG
+config SMTC_IDLE_HOOK_DEBUG
 	bool "Enable additional debug checks before going into CPU idle loop"
 	depends on DEBUG_KERNEL && MIPS_MT_SMTC
 	help
#10 lore raw
In reply to Ralf Baechle
Quoted 6 lines · Ralf Baechle
Quoted 3 lines · Robert P. J. Day

if it makes it any easier, there's no need to continue to CC me on
these messages. there's nothing i'm going to do beyond just reporting what i found.

Bah, we want you to enjoy the email storm you started ;-)

Indeed, and what happened to `Show me the code^Wpatch?' ;-)

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds

#6 lore raw
In reply to Sergei Shtylyov
Quoted 7 lines · Sergei Shtylyov
Quoted 4 lines

========== TX4927BUG_WORKAROUND ========== arch/mips/tx4927/toshiba_rbtx4927/toshiba_rbtx4927_setup.c:141:#define CONFIG_TX4927BUG_WORKAROUND arch/mips/tx4927/toshiba_rbtx4927/toshiba_rbtx4927_setup.c:672:#ifdef CONFIG_TX4927BUG_WORKAROUND arch/mips/tx4927/toshiba_rbtx4927/toshiba_rbtx4927_setup.c:914:#ifdef CONFIG_TX4927BUG_WORKAROUND

This one shouldn't have been called CONFIG_* and might be just killed as
well....

That was workarounds for very early chip. I believe we can kill them all.

---
Atsushi Nemoto