[PATCH] tools: Fix math.h breakage

[PATCH] tools: Fix math.h breakage

[Matthew Wilcox (Oracle)]

Commit 98e1385ef24b broke the radix tree test suite in two different ways;
first by including math.h which didn't exist in the tools directory, and
second by removing an implicit include of spinlock.h before lockdep.h.
Fix both issues.

Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
---
 tools/include/linux/kernel.h             | 22 +--------------------
 tools/include/linux/math.h               | 25 ++++++++++++++++++++++++
 tools/testing/radix-tree/linux/lockdep.h |  3 +++
 3 files changed, 29 insertions(+), 21 deletions(-)
 create mode 100644 tools/include/linux/math.h

diff --git a/tools/include/linux/kernel.h b/tools/include/linux/kernel.h
index a7e54a08fb54..3e8df500cfbd 100644
--- a/tools/include/linux/kernel.h
+++ b/tools/include/linux/kernel.h
@@ -7,6 +7,7 @@
 #include <assert.h>
 #include <linux/build_bug.h>
 #include <linux/compiler.h>
+#include <linux/math.h>
 #include <endian.h>
 #include <byteswap.h>
 
@@ -14,8 +15,6 @@
 #define UINT_MAX	(~0U)
 #endif
 
-#define DIV_ROUND_UP(n,d) (((n) + (d) - 1) / (d))
-
 #define PERF_ALIGN(x, a)	__PERF_ALIGN_MASK(x, (typeof(x))(a)-1)
 #define __PERF_ALIGN_MASK(x, mask)	(((x)+(mask))&~(mask))
 
@@ -52,15 +51,6 @@
 	_min1 < _min2 ? _min1 : _min2; })
 #endif
 
-#ifndef roundup
-#define roundup(x, y) (                                \
-{                                                      \
-	const typeof(y) __y = y;		       \
-	(((x) + (__y - 1)) / __y) * __y;	       \
-}                                                      \
-)
-#endif
-
 #ifndef BUG_ON
 #ifdef NDEBUG
 #define BUG_ON(cond) do { if (cond) {} } while (0)
@@ -104,16 +94,6 @@ int scnprintf_pad(char * buf, size_t size, const char * fmt, ...);
 
 #define ARRAY_SIZE(arr) (sizeof(arr) / sizeof((arr)[0]) + __must_be_array(arr))
 
-/*
- * This looks more complex than it should be. But we need to
- * get the type for the ~ right in round_down (it needs to be
- * as wide as the result!), and we want to evaluate the macro
- * arguments just once each.
- */
-#define __round_mask(x, y) ((__typeof__(x))((y)-1))
-#define round_up(x, y) ((((x)-1) | __round_mask(x, y))+1)
-#define round_down(x, y) ((x) & ~__round_mask(x, y))
-
 #define current_gfp_context(k) 0
 #define synchronize_rcu()
 
diff --git a/tools/include/linux/math.h b/tools/include/linux/math.h
new file mode 100644
index 000000000000..4e7af99ec9eb
--- /dev/null
+++ b/tools/include/linux/math.h
@@ -0,0 +1,25 @@
+#ifndef _TOOLS_MATH_H
+#define _TOOLS_MATH_H
+
+/*
+ * This looks more complex than it should be. But we need to
+ * get the type for the ~ right in round_down (it needs to be
+ * as wide as the result!), and we want to evaluate the macro
+ * arguments just once each.
+ */
+#define __round_mask(x, y) ((__typeof__(x))((y)-1))
+#define round_up(x, y) ((((x)-1) | __round_mask(x, y))+1)
+#define round_down(x, y) ((x) & ~__round_mask(x, y))
+
+#define DIV_ROUND_UP(n,d) (((n) + (d) - 1) / (d))
+
+#ifndef roundup
+#define roundup(x, y) (                                \
+{                                                      \
+	const typeof(y) __y = y;		       \
+	(((x) + (__y - 1)) / __y) * __y;	       \
+}                                                      \
+)
+#endif
+
+#endif
diff --git a/tools/testing/radix-tree/linux/lockdep.h b/tools/testing/radix-tree/linux/lockdep.h
index 565fccdfe6e9..016cff473cfc 100644
--- a/tools/testing/radix-tree/linux/lockdep.h
+++ b/tools/testing/radix-tree/linux/lockdep.h
@@ -1,5 +1,8 @@
 #ifndef _LINUX_LOCKDEP_H
 #define _LINUX_LOCKDEP_H
+
+#include <linux/spinlock.h>
+
 struct lock_class_key {
 	unsigned int a;
 };
-- 
2.33.0

Re: [PATCH] tools: Fix math.h breakage

[Andy Shevchenko]

On Tue, Nov 30, 2021 at 02:13:16PM +0000, Matthew Wilcox (Oracle) wrote:
> Commit 98e1385ef24b broke the radix tree test suite in two different ways;
> first by including math.h which didn't exist in the tools directory, and
> second by removing an implicit include of spinlock.h before lockdep.h.
> Fix both issues.

Sorry for that and thank you for the fix.
Acked-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>

I'm wondering if there is a way of not copying kernel headers manually,
otherwise we always will have such breakages.

> Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
> ---
>  tools/include/linux/kernel.h             | 22 +--------------------
>  tools/include/linux/math.h               | 25 ++++++++++++++++++++++++
>  tools/testing/radix-tree/linux/lockdep.h |  3 +++
>  3 files changed, 29 insertions(+), 21 deletions(-)
>  create mode 100644 tools/include/linux/math.h
> 
> diff --git a/tools/include/linux/kernel.h b/tools/include/linux/kernel.h
> index a7e54a08fb54..3e8df500cfbd 100644
> --- a/tools/include/linux/kernel.h
> +++ b/tools/include/linux/kernel.h
> @@ -7,6 +7,7 @@
>  #include <assert.h>
>  #include <linux/build_bug.h>
>  #include <linux/compiler.h>
> +#include <linux/math.h>
>  #include <endian.h>
>  #include <byteswap.h>
>  
> @@ -14,8 +15,6 @@
>  #define UINT_MAX	(~0U)
>  #endif
>  
> -#define DIV_ROUND_UP(n,d) (((n) + (d) - 1) / (d))
> -
>  #define PERF_ALIGN(x, a)	__PERF_ALIGN_MASK(x, (typeof(x))(a)-1)
>  #define __PERF_ALIGN_MASK(x, mask)	(((x)+(mask))&~(mask))
>  
> @@ -52,15 +51,6 @@
>  	_min1 < _min2 ? _min1 : _min2; })
>  #endif
>  
> -#ifndef roundup
> -#define roundup(x, y) (                                \
> -{                                                      \
> -	const typeof(y) __y = y;		       \
> -	(((x) + (__y - 1)) / __y) * __y;	       \
> -}                                                      \
> -)
> -#endif
> -
>  #ifndef BUG_ON
>  #ifdef NDEBUG
>  #define BUG_ON(cond) do { if (cond) {} } while (0)
> @@ -104,16 +94,6 @@ int scnprintf_pad(char * buf, size_t size, const char * fmt, ...);
>  
>  #define ARRAY_SIZE(arr) (sizeof(arr) / sizeof((arr)[0]) + __must_be_array(arr))
>  
> -/*
> - * This looks more complex than it should be. But we need to
> - * get the type for the ~ right in round_down (it needs to be
> - * as wide as the result!), and we want to evaluate the macro
> - * arguments just once each.
> - */
> -#define __round_mask(x, y) ((__typeof__(x))((y)-1))
> -#define round_up(x, y) ((((x)-1) | __round_mask(x, y))+1)
> -#define round_down(x, y) ((x) & ~__round_mask(x, y))
> -
>  #define current_gfp_context(k) 0
>  #define synchronize_rcu()
>  
> diff --git a/tools/include/linux/math.h b/tools/include/linux/math.h
> new file mode 100644
> index 000000000000..4e7af99ec9eb
> --- /dev/null
> +++ b/tools/include/linux/math.h
> @@ -0,0 +1,25 @@
> +#ifndef _TOOLS_MATH_H
> +#define _TOOLS_MATH_H
> +
> +/*
> + * This looks more complex than it should be. But we need to
> + * get the type for the ~ right in round_down (it needs to be
> + * as wide as the result!), and we want to evaluate the macro
> + * arguments just once each.
> + */
> +#define __round_mask(x, y) ((__typeof__(x))((y)-1))
> +#define round_up(x, y) ((((x)-1) | __round_mask(x, y))+1)
> +#define round_down(x, y) ((x) & ~__round_mask(x, y))
> +
> +#define DIV_ROUND_UP(n,d) (((n) + (d) - 1) / (d))
> +
> +#ifndef roundup
> +#define roundup(x, y) (                                \
> +{                                                      \
> +	const typeof(y) __y = y;		       \
> +	(((x) + (__y - 1)) / __y) * __y;	       \
> +}                                                      \
> +)
> +#endif
> +
> +#endif
> diff --git a/tools/testing/radix-tree/linux/lockdep.h b/tools/testing/radix-tree/linux/lockdep.h
> index 565fccdfe6e9..016cff473cfc 100644
> --- a/tools/testing/radix-tree/linux/lockdep.h
> +++ b/tools/testing/radix-tree/linux/lockdep.h
> @@ -1,5 +1,8 @@
>  #ifndef _LINUX_LOCKDEP_H
>  #define _LINUX_LOCKDEP_H
> +
> +#include <linux/spinlock.h>
> +
>  struct lock_class_key {
>  	unsigned int a;
>  };
> -- 
> 2.33.0
> 

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH] tools: Fix math.h breakage

[Matthew Wilcox]

On Tue, Nov 30, 2021 at 04:39:00PM +0200, Andy Shevchenko wrote:
> On Tue, Nov 30, 2021 at 02:13:16PM +0000, Matthew Wilcox (Oracle) wrote:
> > Commit 98e1385ef24b broke the radix tree test suite in two different ways;
> > first by including math.h which didn't exist in the tools directory, and
> > second by removing an implicit include of spinlock.h before lockdep.h.
> > Fix both issues.
> 
> Sorry for that and thank you for the fix.
> Acked-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> 
> I'm wondering if there is a way of not copying kernel headers manually,
> otherwise we always will have such breakages.

It's not necessarily that these are _copies_ of kernel headers, so much as
they're ways of mocking kernel interfaces when building userspace code.
We could separate out pieces and include them from each direction, but
that has its own problems, and doesn't necessarily solve these kinds of
problems either.

I think the only way to prevent these kinds of breakages is to make sure
the build bots are also building things.

Re: [PATCH] tools: Fix math.h breakage

[Andy Shevchenko]

On Tue, Nov 30, 2021 at 03:07:12PM +0000, Matthew Wilcox wrote:
> On Tue, Nov 30, 2021 at 04:39:00PM +0200, Andy Shevchenko wrote:
> > On Tue, Nov 30, 2021 at 02:13:16PM +0000, Matthew Wilcox (Oracle) wrote:
> > > Commit 98e1385ef24b broke the radix tree test suite in two different ways;
> > > first by including math.h which didn't exist in the tools directory, and
> > > second by removing an implicit include of spinlock.h before lockdep.h.
> > > Fix both issues.
> > 
> > Sorry for that and thank you for the fix.
> > Acked-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> > 
> > I'm wondering if there is a way of not copying kernel headers manually,
> > otherwise we always will have such breakages.
> 
> It's not necessarily that these are _copies_ of kernel headers, so much as
> they're ways of mocking kernel interfaces when building userspace code.
> We could separate out pieces and include them from each direction, but
> that has its own problems, and doesn't necessarily solve these kinds of
> problems either.
> 
> I think the only way to prevent these kinds of breakages is to make sure
> the build bots are also building things.

I don't know how to achieve this locally because I'm using `make O=...` and
it's broken for many tools/ folders. At some point I simply gave up.

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH] tools: Fix math.h breakage

[Andy Shevchenko]

On Tue, Nov 30, 2021 at 05:22:30PM +0200, Andy Shevchenko wrote:
> On Tue, Nov 30, 2021 at 03:07:12PM +0000, Matthew Wilcox wrote:
> > On Tue, Nov 30, 2021 at 04:39:00PM +0200, Andy Shevchenko wrote:
> > > On Tue, Nov 30, 2021 at 02:13:16PM +0000, Matthew Wilcox (Oracle) wrote:
> > > > Commit 98e1385ef24b broke the radix tree test suite in two different ways;
> > > > first by including math.h which didn't exist in the tools directory, and
> > > > second by removing an implicit include of spinlock.h before lockdep.h.
> > > > Fix both issues.
> > > 
> > > Sorry for that and thank you for the fix.
> > > Acked-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> > > 
> > > I'm wondering if there is a way of not copying kernel headers manually,
> > > otherwise we always will have such breakages.
> > 
> > It's not necessarily that these are _copies_ of kernel headers, so much as
> > they're ways of mocking kernel interfaces when building userspace code.
> > We could separate out pieces and include them from each direction, but
> > that has its own problems, and doesn't necessarily solve these kinds of
> > problems either.
> > 
> > I think the only way to prevent these kinds of breakages is to make sure
> > the build bots are also building things.
> 
> I don't know how to achieve this locally because I'm using `make O=...` and
> it's broken for many tools/ folders. At some point I simply gave up.

To be clear, it's an argument to support your idea that CI can do it for us.

-- 
With Best Regards,
Andy Shevchenko