From: Qi Yong <qiyong@fc-cn.com>
To: Dave Hansen <haveblue@us.ibm.com>
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org,
linux-ia64@vger.kernel.org
Subject: Re: [RFC][PATCH 3/9] actual generic PAGE_SIZE infrastructure
Date: Tue, 17 Oct 2006 07:29:34 +0000 [thread overview]
Message-ID: <4534865E.7020504@fc-cn.com> (raw)
In-Reply-To: <20060830221606.40937644@localhost.localdomain>
Dave Hansen wrote:
>* Add _ALIGN_UP() which we'll use now and _ALIGN_DOWN(), just for
> parity.
>* Define ASM_CONST() macro to help using constants in both assembly
> and C code. Several architectures have some form of this, and
> they will be consolidated around this one.
>* Actually create PAGE_SHIFT and PAGE_SIZE macros
>* For now, require that architectures enable GENERIC_PAGE_SIZE in
> order to get this new code. This option will be removed by the
> last patch in the series, and makes the series bisect-safe.
>* Note that this moves the compiler.h define outside of the
> #ifdef __KERNEL__, but that's OK because it has its own.
>
>Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
>---
>
> threadalloc-dave/include/asm-generic/page.h | 31 ++++++++++++++++++--
> threadalloc-dave/mm/Kconfig | 43 ++++++++++++++++++++++++++++
> 2 files changed, 71 insertions(+), 3 deletions(-)
>
>diff -puN include/asm-generic/page.h~generic-PAGE_SIZE-infrastructure include/asm-generic/page.h
>--- threadalloc/include/asm-generic/page.h~generic-PAGE_SIZE-infrastructure 2006-08-30 15:15:00.000000000 -0700
>+++ threadalloc-dave/include/asm-generic/page.h 2006-08-30 15:15:01.000000000 -0700
>@@ -1,11 +1,36 @@
> #ifndef _ASM_GENERIC_PAGE_H
> #define _ASM_GENERIC_PAGE_H
>
>+#include <linux/compiler.h>
>+#include <linux/align.h>
>+
> #ifdef __KERNEL__
>-#ifndef __ASSEMBLY__
>
>-#include <linux/compiler.h>
>+#ifdef __ASSEMBLY__
>+#define ASM_CONST(x) x
>+#else
>+#define __ASM_CONST(x) x##UL
>+#define ASM_CONST(x) __ASM_CONST(x)
>+#endif
>+
>+#ifdef CONFIG_ARCH_GENERIC_PAGE_SIZE
>+
>+#define PAGE_SHIFT CONFIG_PAGE_SHIFT
>+#define PAGE_SIZE (ASM_CONST(1) << PAGE_SHIFT)
>
>
Your generic page.h hides PAGE_SIZE under "#ifdef __KERNEL__".
That would cause severe userland compile failures.
Most archs have PAGE_SIZE visible to userland.
Please keep PAGE_SIZE and its friends outside "#ifdef __KERNEL__".
-- qiyong
>+
>+/*
>+ * Subtle: (1 << PAGE_SHIFT) is an int, not an unsigned long. So if we
>+ * assign PAGE_MASK to a larger type it gets extended the way we want
>+ * (i.e. with 1s in the high bits)
>+ */
>+#define PAGE_MASK (~((1 << PAGE_SHIFT) - 1))
>
>+/* to align the pointer to the (next) page boundary */
>+#define PAGE_ALIGN(addr) ALIGN(addr, PAGE_SIZE)
>+
>+#endif /* CONFIG_ARCH_GENERIC_PAGE_SIZE */
>+
>+#ifndef __ASSEMBLY__
> #ifndef CONFIG_ARCH_HAVE_GET_ORDER
> /* Pure 2^n version of get_order */
> static __inline__ __attribute_const__ int get_order(unsigned long size)
>@@ -22,7 +47,7 @@ static __inline__ __attribute_const__ in
> }
>
> #endif /* CONFIG_ARCH_HAVE_GET_ORDER */
>-#endif /* __ASSEMBLY__ */
>+#endif /* __ASSEMBLY__ */
> #endif /* __KERNEL__ */
>
> #endif /* _ASM_GENERIC_PAGE_H */
>
>
--
Qi Yong
WARNING: multiple messages have this Message-ID (diff)
From: Qi Yong <qiyong@fc-cn.com>
To: Dave Hansen <haveblue@us.ibm.com>
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org,
linux-ia64@vger.kernel.org
Subject: Re: [RFC][PATCH 3/9] actual generic PAGE_SIZE infrastructure
Date: Tue, 17 Oct 2006 15:29:34 +0800 [thread overview]
Message-ID: <4534865E.7020504@fc-cn.com> (raw)
In-Reply-To: <20060830221606.40937644@localhost.localdomain>
Dave Hansen wrote:
>* Add _ALIGN_UP() which we'll use now and _ALIGN_DOWN(), just for
> parity.
>* Define ASM_CONST() macro to help using constants in both assembly
> and C code. Several architectures have some form of this, and
> they will be consolidated around this one.
>* Actually create PAGE_SHIFT and PAGE_SIZE macros
>* For now, require that architectures enable GENERIC_PAGE_SIZE in
> order to get this new code. This option will be removed by the
> last patch in the series, and makes the series bisect-safe.
>* Note that this moves the compiler.h define outside of the
> #ifdef __KERNEL__, but that's OK because it has its own.
>
>Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
>---
>
> threadalloc-dave/include/asm-generic/page.h | 31 ++++++++++++++++++--
> threadalloc-dave/mm/Kconfig | 43 ++++++++++++++++++++++++++++
> 2 files changed, 71 insertions(+), 3 deletions(-)
>
>diff -puN include/asm-generic/page.h~generic-PAGE_SIZE-infrastructure include/asm-generic/page.h
>--- threadalloc/include/asm-generic/page.h~generic-PAGE_SIZE-infrastructure 2006-08-30 15:15:00.000000000 -0700
>+++ threadalloc-dave/include/asm-generic/page.h 2006-08-30 15:15:01.000000000 -0700
>@@ -1,11 +1,36 @@
> #ifndef _ASM_GENERIC_PAGE_H
> #define _ASM_GENERIC_PAGE_H
>
>+#include <linux/compiler.h>
>+#include <linux/align.h>
>+
> #ifdef __KERNEL__
>-#ifndef __ASSEMBLY__
>
>-#include <linux/compiler.h>
>+#ifdef __ASSEMBLY__
>+#define ASM_CONST(x) x
>+#else
>+#define __ASM_CONST(x) x##UL
>+#define ASM_CONST(x) __ASM_CONST(x)
>+#endif
>+
>+#ifdef CONFIG_ARCH_GENERIC_PAGE_SIZE
>+
>+#define PAGE_SHIFT CONFIG_PAGE_SHIFT
>+#define PAGE_SIZE (ASM_CONST(1) << PAGE_SHIFT)
>
>
Your generic page.h hides PAGE_SIZE under "#ifdef __KERNEL__".
That would cause severe userland compile failures.
Most archs have PAGE_SIZE visible to userland.
Please keep PAGE_SIZE and its friends outside "#ifdef __KERNEL__".
-- qiyong
>+
>+/*
>+ * Subtle: (1 << PAGE_SHIFT) is an int, not an unsigned long. So if we
>+ * assign PAGE_MASK to a larger type it gets extended the way we want
>+ * (i.e. with 1s in the high bits)
>+ */
>+#define PAGE_MASK (~((1 << PAGE_SHIFT) - 1))
>
>+/* to align the pointer to the (next) page boundary */
>+#define PAGE_ALIGN(addr) ALIGN(addr, PAGE_SIZE)
>+
>+#endif /* CONFIG_ARCH_GENERIC_PAGE_SIZE */
>+
>+#ifndef __ASSEMBLY__
> #ifndef CONFIG_ARCH_HAVE_GET_ORDER
> /* Pure 2^n version of get_order */
> static __inline__ __attribute_const__ int get_order(unsigned long size)
>@@ -22,7 +47,7 @@ static __inline__ __attribute_const__ in
> }
>
> #endif /* CONFIG_ARCH_HAVE_GET_ORDER */
>-#endif /* __ASSEMBLY__ */
>+#endif /* __ASSEMBLY__ */
> #endif /* __KERNEL__ */
>
> #endif /* _ASM_GENERIC_PAGE_H */
>
>
--
Qi Yong
WARNING: multiple messages have this Message-ID (diff)
From: Qi Yong <qiyong@fc-cn.com>
To: Dave Hansen <haveblue@us.ibm.com>
Cc: linux-mm@kvack.org, linux-kernel@vger.kernel.org,
linux-ia64@vger.kernel.org
Subject: Re: [RFC][PATCH 3/9] actual generic PAGE_SIZE infrastructure
Date: Tue, 17 Oct 2006 15:29:34 +0800 [thread overview]
Message-ID: <4534865E.7020504@fc-cn.com> (raw)
In-Reply-To: <20060830221606.40937644@localhost.localdomain>
Dave Hansen wrote:
>* Add _ALIGN_UP() which we'll use now and _ALIGN_DOWN(), just for
> parity.
>* Define ASM_CONST() macro to help using constants in both assembly
> and C code. Several architectures have some form of this, and
> they will be consolidated around this one.
>* Actually create PAGE_SHIFT and PAGE_SIZE macros
>* For now, require that architectures enable GENERIC_PAGE_SIZE in
> order to get this new code. This option will be removed by the
> last patch in the series, and makes the series bisect-safe.
>* Note that this moves the compiler.h define outside of the
> #ifdef __KERNEL__, but that's OK because it has its own.
>
>Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
>---
>
> threadalloc-dave/include/asm-generic/page.h | 31 ++++++++++++++++++--
> threadalloc-dave/mm/Kconfig | 43 ++++++++++++++++++++++++++++
> 2 files changed, 71 insertions(+), 3 deletions(-)
>
>diff -puN include/asm-generic/page.h~generic-PAGE_SIZE-infrastructure include/asm-generic/page.h
>--- threadalloc/include/asm-generic/page.h~generic-PAGE_SIZE-infrastructure 2006-08-30 15:15:00.000000000 -0700
>+++ threadalloc-dave/include/asm-generic/page.h 2006-08-30 15:15:01.000000000 -0700
>@@ -1,11 +1,36 @@
> #ifndef _ASM_GENERIC_PAGE_H
> #define _ASM_GENERIC_PAGE_H
>
>+#include <linux/compiler.h>
>+#include <linux/align.h>
>+
> #ifdef __KERNEL__
>-#ifndef __ASSEMBLY__
>
>-#include <linux/compiler.h>
>+#ifdef __ASSEMBLY__
>+#define ASM_CONST(x) x
>+#else
>+#define __ASM_CONST(x) x##UL
>+#define ASM_CONST(x) __ASM_CONST(x)
>+#endif
>+
>+#ifdef CONFIG_ARCH_GENERIC_PAGE_SIZE
>+
>+#define PAGE_SHIFT CONFIG_PAGE_SHIFT
>+#define PAGE_SIZE (ASM_CONST(1) << PAGE_SHIFT)
>
>
Your generic page.h hides PAGE_SIZE under "#ifdef __KERNEL__".
That would cause severe userland compile failures.
Most archs have PAGE_SIZE visible to userland.
Please keep PAGE_SIZE and its friends outside "#ifdef __KERNEL__".
-- qiyong
>+
>+/*
>+ * Subtle: (1 << PAGE_SHIFT) is an int, not an unsigned long. So if we
>+ * assign PAGE_MASK to a larger type it gets extended the way we want
>+ * (i.e. with 1s in the high bits)
>+ */
>+#define PAGE_MASK (~((1 << PAGE_SHIFT) - 1))
>
>+/* to align the pointer to the (next) page boundary */
>+#define PAGE_ALIGN(addr) ALIGN(addr, PAGE_SIZE)
>+
>+#endif /* CONFIG_ARCH_GENERIC_PAGE_SIZE */
>+
>+#ifndef __ASSEMBLY__
> #ifndef CONFIG_ARCH_HAVE_GET_ORDER
> /* Pure 2^n version of get_order */
> static __inline__ __attribute_const__ int get_order(unsigned long size)
>@@ -22,7 +47,7 @@ static __inline__ __attribute_const__ in
> }
>
> #endif /* CONFIG_ARCH_HAVE_GET_ORDER */
>-#endif /* __ASSEMBLY__ */
>+#endif /* __ASSEMBLY__ */
> #endif /* __KERNEL__ */
>
> #endif /* _ASM_GENERIC_PAGE_H */
>
>
--
Qi Yong
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2006-10-17 7:29 UTC|newest]
Thread overview: 80+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-08-30 22:16 [RFC][PATCH 0/9] generic PAGE_SIZE infrastructure (v4) Dave Hansen
2006-08-30 22:16 ` Dave Hansen
2006-08-30 22:16 ` Dave Hansen
2006-08-30 22:16 ` [RFC][PATCH 2/9] conditionally define generic get_order() (ARCH_HAS_GET_ORDER) Dave Hansen
2006-08-30 22:16 ` Dave Hansen
2006-08-30 22:16 ` Dave Hansen
2006-08-31 18:41 ` Haavard Skinnemoen
2006-08-31 18:41 ` Haavard Skinnemoen
2006-08-31 18:41 ` Haavard Skinnemoen
2006-08-31 19:51 ` [RFC][PATCH 2/9] conditionally define generic get_order() Dave Hansen
2006-08-31 19:51 ` [RFC][PATCH 2/9] conditionally define generic get_order() (ARCH_HAS_GET_ORDER) Dave Hansen
2006-08-31 19:51 ` Dave Hansen
2006-08-30 22:16 ` [RFC][PATCH 1/9] put alignment macros in align.h Dave Hansen
2006-08-30 22:16 ` Dave Hansen
2006-08-30 22:16 ` Dave Hansen
2006-08-30 22:16 ` [RFC][PATCH 3/9] actual generic PAGE_SIZE infrastructure Dave Hansen
2006-08-30 22:16 ` Dave Hansen
2006-08-30 22:16 ` Dave Hansen
2006-08-31 0:08 ` Christoph Lameter
2006-08-31 0:08 ` Christoph Lameter
2006-08-31 0:08 ` Christoph Lameter
2006-08-31 17:57 ` Dave Hansen
2006-08-31 17:57 ` Dave Hansen
2006-08-31 17:57 ` Dave Hansen
2006-08-31 18:06 ` Christoph Lameter
2006-08-31 18:06 ` Christoph Lameter
2006-08-31 18:06 ` Christoph Lameter
2006-08-31 20:50 ` Dave Hansen
2006-08-31 20:50 ` Dave Hansen
2006-08-31 20:50 ` Dave Hansen
2006-09-05 11:20 ` Martin Waitz
2006-09-05 11:20 ` Martin Waitz
2006-09-05 16:47 ` Dave Hansen
2006-09-05 16:47 ` Dave Hansen
2006-09-05 16:47 ` Dave Hansen
2006-10-17 7:29 ` Qi Yong [this message]
2006-10-17 7:29 ` Qi Yong
2006-10-17 7:29 ` Qi Yong
2006-08-30 22:16 ` [RFC][PATCH 4/9] ia64 generic PAGE_SIZE Dave Hansen
2006-08-30 22:16 ` Dave Hansen
2006-08-30 22:16 ` Dave Hansen
2006-08-30 23:57 ` Christoph Lameter
2006-08-30 23:57 ` Christoph Lameter
2006-08-30 23:57 ` Christoph Lameter
2006-08-31 17:38 ` Dave Hansen
2006-08-31 17:38 ` Dave Hansen
2006-08-31 17:38 ` Dave Hansen
2006-08-31 17:39 ` Christoph Lameter
2006-08-31 17:39 ` Christoph Lameter
2006-08-31 17:39 ` Christoph Lameter
2006-08-30 22:16 ` [RFC][PATCH 5/9] sparc64 " Dave Hansen
2006-08-30 22:16 ` Dave Hansen
2006-08-30 22:16 ` Dave Hansen
2006-08-30 22:27 ` David Miller
2006-08-30 22:27 ` David Miller, Dave Hansen
2006-08-30 22:27 ` David Miller
2006-08-30 22:16 ` [RFC][PATCH 6/9] mips " Dave Hansen
2006-08-30 22:16 ` Dave Hansen
2006-08-30 22:16 ` Dave Hansen
2006-08-30 22:16 ` [RFC][PATCH 7/9] parisc " Dave Hansen
2006-08-30 22:16 ` Dave Hansen
2006-08-30 22:16 ` Dave Hansen
2006-08-30 22:40 ` Kyle McMartin
2006-08-30 22:40 ` Kyle McMartin
2006-08-30 22:40 ` Kyle McMartin
2006-08-30 22:48 ` Dave Hansen
2006-08-30 22:48 ` Dave Hansen
2006-08-30 22:48 ` Dave Hansen
2006-08-30 22:16 ` [RFC][PATCH 8/9] powerpc " Dave Hansen
2006-08-30 22:16 ` Dave Hansen
2006-08-30 22:16 ` Dave Hansen
2006-08-30 22:16 ` [RFC][PATCH 9/9] convert the "easy" architectures to " Dave Hansen
2006-08-30 22:16 ` Dave Hansen
2006-08-30 22:16 ` Dave Hansen
2006-08-31 0:33 ` [RFC][PATCH 0/9] generic PAGE_SIZE infrastructure (v4) Paul Mackerras
2006-08-31 0:33 ` Paul Mackerras
2006-08-31 0:33 ` Paul Mackerras
2006-08-31 21:03 ` Dave Hansen
2006-08-31 21:03 ` Dave Hansen
2006-08-31 21:03 ` Dave Hansen
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=4534865E.7020504@fc-cn.com \
--to=qiyong@fc-cn.com \
--cc=haveblue@us.ibm.com \
--cc=linux-ia64@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.