Skip to contentSkip to conversation

asm-{alpha,h8300,um,v850,xtensa}/param.h: unbreak HZ for userspace

PATCH Mike Frysinger 3 participants 8 messages May 10, 2008 – May 14, 2008

lore
Share & tools
lore
#0 lore raw

I noticed this because alpha was broken due to the recent bdc807871d58285737. Most arches do something like this in their asm/param.h:
#ifdef __KERNEL__

# define HZ CONFIG_HZ
#else
# define HZ 100
#endif
A few arches though (namely alpha/h8300/um/v850/xtensa) either do no set HZ at
all for !__KERNEL__, or they set it wrongly.  This should bring all arches in
line by setting up HZ for userspace.

---

include/asm-alpha/param.h +4 -0
index 0982f1d..d300950 100644
@@ -5,8 +5,12 @@
    hardware ignores reprogramming.  We also need userland buy-in to the 
    change in HZ, since this is visible in the wait4 resources etc.  */
 
+#ifdef __KERNEL__
 #define HZ		CONFIG_HZ
 #define USER_HZ		HZ
+#else
+#define HZ		100
+#endif
 
 #define EXEC_PAGESIZE	8192
 
include/asm-h8300/param.h +3 -5
index 04f64f1..1c72fb8 100644
@@ -1,14 +1,12 @@
 #ifndef _H8300_PARAM_H
 #define _H8300_PARAM_H
 
-
-#ifndef HZ
-#define HZ CONFIG_HZ
-#endif
-
 #ifdef __KERNEL__
+#define HZ		CONFIG_HZ
 #define	USER_HZ		HZ
 #define	CLOCKS_PER_SEC	(USER_HZ)
+#else
+#define HZ		100
 #endif
 
 #define EXEC_PAGESIZE	4096
include/asm-um/param.h +2 -0
index 4cd4a22..e44f4e6 100644
@@ -13,6 +13,8 @@
 #define HZ CONFIG_HZ
 #define USER_HZ	100	   /* .. some user interfaces are in "ticks" */
 #define CLOCKS_PER_SEC (USER_HZ)  /* frequency at which times() counts */
+#else
+#define HZ 100
 #endif
 
 #endif
include/asm-v850/param.h +2 -0
index 2818326..4391f5f 100644
@@ -26,6 +26,8 @@
 # define HZ		CONFIG_HZ
 # define USER_HZ	100
 # define CLOCKS_PER_SEC	USER_HZ
+#else
+# define HZ		100
 #endif
 
 #endif /* __V850_PARAM_H__ */
include/asm-xtensa/param.h +2 -0
index 82ad34d..ba03d5a 100644
@@ -15,6 +15,8 @@
 # define HZ		CONFIG_HZ	/* internal timer frequency */
 # define USER_HZ	100		/* for user interfaces in "ticks" */
 # define CLOCKS_PER_SEC (USER_HZ)	/* frequnzy at which times() counts */
+#else
+# define HZ		100
 #endif
 
 #define EXEC_PAGESIZE	4096
#1 lore raw

could we please get an ack or nack on that patch.
checked linux-next and it doesn't appear there.
without it currently perl 5.10 doesn't build on alpha:
perl.c: In function 'perl_construct':
perl.c:388: error: 'CONFIG_HZ' undeclared (first use in this function)
-> http://buildd.debian.org/fetch.cgi?pkg=perl;ver=5.10.0-10;arch=alpha;stamp=1210252894

Signature · 15 lines
maks

I noticed this because alpha was broken due to the recent bdc807871d58285737. 
Most arches do something like this in their asm/param.h:
#ifdef __KERNEL__
# define HZ CONFIG_HZ
#else
# define HZ 100
#endif
A few arches though (namely alpha/h8300/um/v850/xtensa) either do no set HZ at
all for !__KERNEL__, or they set it wrongly.  This should bring all arches in
line by setting up HZ for userspace.

Signed-off-by: Mike Frysinger <[email protected]>
---
include/asm-alpha/param.h +4 -0
index 0982f1d..d300950 100644
@@ -5,8 +5,12 @@
    hardware ignores reprogramming.  We also need userland buy-in to the 
    change in HZ, since this is visible in the wait4 resources etc.  */
 
+#ifdef __KERNEL__
 #define HZ		CONFIG_HZ
 #define USER_HZ		HZ
+#else
+#define HZ		100
+#endif
 
 #define EXEC_PAGESIZE	8192
 
include/asm-h8300/param.h +3 -5
index 04f64f1..1c72fb8 100644
@@ -1,14 +1,12 @@
 #ifndef _H8300_PARAM_H
 #define _H8300_PARAM_H
 
-
-#ifndef HZ
-#define HZ CONFIG_HZ
-#endif
-
 #ifdef __KERNEL__
+#define HZ		CONFIG_HZ
 #define	USER_HZ		HZ
 #define	CLOCKS_PER_SEC	(USER_HZ)
+#else
+#define HZ		100
 #endif
 
 #define EXEC_PAGESIZE	4096
include/asm-um/param.h +2 -0
index 4cd4a22..e44f4e6 100644
@@ -13,6 +13,8 @@
 #define HZ CONFIG_HZ
 #define USER_HZ	100	   /* .. some user interfaces are in "ticks" */
 #define CLOCKS_PER_SEC (USER_HZ)  /* frequency at which times() counts */
+#else
+#define HZ 100
 #endif
 
 #endif
include/asm-v850/param.h +2 -0
index 2818326..4391f5f 100644
@@ -26,6 +26,8 @@
 # define HZ		CONFIG_HZ
 # define USER_HZ	100
 # define CLOCKS_PER_SEC	USER_HZ
+#else
+# define HZ		100
 #endif
 
 #endif /* __V850_PARAM_H__ */
include/asm-xtensa/param.h +2 -0
index 82ad34d..ba03d5a 100644
@@ -15,6 +15,8 @@
 # define HZ		CONFIG_HZ	/* internal timer frequency */
 # define USER_HZ	100		/* for user interfaces in "ticks" */
 # define CLOCKS_PER_SEC (USER_HZ)	/* frequnzy at which times() counts */
+#else
+# define HZ		100
 #endif
 
 #define EXEC_PAGESIZE	4096

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

#2 lore raw
Quoted 6 lines · maximilian attems

could we please get an ack or nack on that patch.
checked linux-next and it doesn't appear there.
without it currently perl 5.10 doesn't build on alpha:
perl.c: In function 'perl_construct':
perl.c:388: error: 'CONFIG_HZ' undeclared (first use in this function)
-> http://buildd.debian.org/fetch.cgi?pkg=perl;ver=5.10.0-10;arch=alpha;stamp=1210252894

hmm it seems more userspace expects CONFIG_HZ, so it's not only a perl build breakage..

#3 lore raw
Quoted 10 lines · maximilian attems
Quoted 8 lines · maximilian attems

could we please get an ack or nack on that patch.
checked linux-next and it doesn't appear there.
without it currently perl 5.10 doesn't build on alpha:
perl.c: In function 'perl_construct':
perl.c:388: error: 'CONFIG_HZ' undeclared (first use in this function) ->
http://buildd.debian.org/fetch.cgi?pkg=perl;ver=5.10.0-10;arch=alpha;stam
p=1210252894

Quoted 2 lines

hmm it seems more userspace expects CONFIG_HZ, so it's not only a perl build breakage..

anything using CONFIG_HZ is absolutely broken. go fix the package.

anything using HZ should be converted to sysconf(), but i'm not about to go on that crusade to fix things up
-mike

#4 lore raw
Quoted 11 lines · Mike Frysinger
Quoted 10 lines · maximilian attems
Quoted 8 lines · maximilian attems

could we please get an ack or nack on that patch.
checked linux-next and it doesn't appear there.
without it currently perl 5.10 doesn't build on alpha:
perl.c: In function 'perl_construct':
perl.c:388: error: 'CONFIG_HZ' undeclared (first use in this function) ->
http://buildd.debian.org/fetch.cgi?pkg=perl;ver=5.10.0-10;arch=alpha;stam
p=1210252894

Quoted 7 lines
Quoted 2 lines

hmm it seems more userspace expects CONFIG_HZ, so it's not only a perl build breakage..

anything using CONFIG_HZ is absolutely broken. go fix the package.

anything using HZ should be converted to sysconf(), but i'm not about to go on that crusade to fix things up

news at 11

okay great that the patch went into mm.

Signature · 1 line
maks
#5 lore raw
Quoted 15 lines · maximilian attems
Quoted 14 lines · Mike Frysinger
Quoted 12 lines · maximilian attems
Quoted 8 lines · maximilian attems

could we please get an ack or nack on that patch. checked linux-next and it doesn't appear there. without it currently perl 5.10 doesn't build on alpha:
perl.c: In function 'perl_construct':
perl.c:388: error: 'CONFIG_HZ' undeclared (first use in this function) ->
http://buildd.debian.org/fetch.cgi?pkg=perl;ver=5.10.0-10;arch=alpha;
stam p=1210252894

hmm it seems more userspace expects CONFIG_HZ, so it's not only a perl build breakage..

Quoted 8 lines
Quoted 4 lines

anything using CONFIG_HZ is absolutely broken. go fix the package.

anything using HZ should be converted to sysconf(), but i'm not about to go on that crusade to fix things up

news at 11

okay great that the patch went into mm.

i'm guessing you're thinking packages are using CONFIG_HZ when in reality they arent. the asm-alpha/param.h in 2.6.25 does:
#define HZ CONFIG_HZ
so anything using HZ would end up with an error message about CONFIG_HZ being undefined
-mike

#7 lore raw
In reply to maximilian attems
Quoted 4 lines · On Wed, 14 May 2008 13:05:31 +0200 maximilian attems <[email protected]> wrote:

without it currently perl 5.10 doesn't build on alpha:
perl.c: In function 'perl_construct':
perl.c:388: error: 'CONFIG_HZ' undeclared (first use in this function)
-> http://buildd.debian.org/fetch.cgi?pkg=perl;ver=5.10.0-10;arch=alpha;stamp=1210252894

um, that was a fairly important piece of information which was cruelly hidden from us - thanks.

I happened to have already planned on sending this fix in for 2.6.26 but it wasn't an obvious call.