linux-toradex.git/lib/bitmap.c, branch v4.6-rc3

lib/bitmap.c: conversion routines to/from u32 array

2016-02-20T03:54:09+00:00

Aimed at transferring bitmaps to/from user-space in a 32/64-bit agnostic
way.

Tested:
  unit tests (next patch) on qemu i386, x86_64, ppc, ppc64 BE and LE,
  ARM.

Signed-off-by: David Decotigny 
Reviewed-by: Ben Hutchings 
Signed-off-by: David S. Miller

lib/bitmap.c: bitmap_parselist can accept string with whitespaces on head or tail

2015-09-10T20:29:01+00:00

In __bitmap_parselist we can accept whitespaces on head or tail during
every parsing procedure.  If input has valid ranges, there is no reason to
reject the user.

For example, bitmap_parselist(" 1-3, 5, ", &mask, nmaskbits).  After
separating the string, we get " 1-3", " 5", and " ".  It's possible and
reasonable to accept such string as long as the parsing result is correct.

Signed-off-by: Pan Xinhui 
Cc: Yury Norov 
Cc: Chris Metcalf 
Cc: Rasmus Villemoes 
Cc: Sudeep Holla 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

lib/bitmap.c: fix a special string handling bug in __bitmap_parselist

2015-09-10T20:29:01+00:00

If string end with '-', for exapmle, bitmap_parselist("1,0-",&mask,
nmaskbits), It is not in a valid pattern, so add a check after loop.
Return -EINVAL on such condition.

Signed-off-by: Pan Xinhui 
Cc: Yury Norov 
Cc: Chris Metcalf 
Cc: Rasmus Villemoes 
Cc: Sudeep Holla 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

lib/bitmap.c: correct a code style and do some, optimization

2015-09-10T20:29:01+00:00

We can avoid in-loop incrementation of ndigits.  Save current totaldigits
to ndigits before loop, and check ndigits against totaldigits after the
loop.

Signed-off-by: Pan Xinhui 
Cc: Yury Norov 
Cc: Chris Metcalf 
Cc: Rasmus Villemoes 
Cc: Sudeep Holla 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

bitmap: remove explicit newline handling using scnprintf format string

2015-06-26T00:00:40+00:00

bitmap_print_to_pagebuf uses scnprintf to copy the cpumask/list to page
buffer.  It handles the newline and trailing null character explicitly.

It's unnecessary and also partially duplicated as scnprintf already adds
trailing null character.  The newline can be passed through format
string to scnprintf.  This patch does that simplification.

However theoretically there's one behavior difference: when the buffer
is too small, the original code would still output '\n' at the end while
the new code(with this patch) would just continue to print the formatted
string.  Since this function is dealing with only page buffers, it's
highly unlikely to hit that corner case.

This patch will help in auditing the users of bitmap_print_to_pagebuf to
verify that the buffer passed is large enough and get rid of it
completely by replacing them with direct scnprintf()

[akpm@linux-foundation.org: tweak comment]
Signed-off-by: Sudeep Holla 
Suggested-by: Pawel Moll 
Cc: Tejun Heo 
Cc: "Peter Zijlstra (Intel)" 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

__bitmap_parselist: fix bug in empty string handling

2015-06-26T00:00:40+00:00

bitmap_parselist("", &mask, nmaskbits) will erroneously set bit zero in
the mask.  The same bug is visible in cpumask_parselist() since it is
layered on top of the bitmask code, e.g.  if you boot with "isolcpus=",
you will actually end up with cpu zero isolated.

The bug was introduced in commit 4b060420a596 ("bitmap, irq: add
smp_affinity_list interface to /proc/irq") when bitmap_parselist() was
generalized to support userspace as well as kernelspace.

Fixes: 4b060420a596 ("bitmap, irq: add smp_affinity_list interface to /proc/irq")
Signed-off-by: Chris Metcalf 
Cc: Rasmus Villemoes 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

lib/bitmap.c: bitmap_[empty,full]: remove code duplication

2015-04-17T13:03:56+00:00

bitmap_empty() has its own implementation.  But it's clearly as simple as:

	find_first_bit(src, nbits) == nbits

The same is true for 'bitmap_full'.

Signed-off-by: Yury Norov 
Cc: George Spelvin 
Cc: Alexey Klimov 
Cc: Rasmus Villemoes 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

bitmap, cpumask, nodemask: remove dedicated formatting functions

2015-02-14T05:21:39+00:00

Now that all bitmap formatting usages have been converted to
'%*pb[l]', the separate formatting functions are unnecessary.  The
following functions are removed.

* bitmap_scn[list]printf()
* cpumask_scnprintf(), cpulist_scnprintf()
* [__]nodemask_scnprintf(), [__]nodelist_scnprintf()
* seq_bitmap[_list](), seq_cpumask[_list](), seq_nodemask[_list]()
* seq_buf_bitmask()

Signed-off-by: Tejun Heo 
Cc: Rusty Russell 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

bitmap: use %*pb[l] to print bitmaps including cpumasks and nodemasks

2015-02-14T05:21:36+00:00

printk and friends can now format bitmaps using '%*pb[l]'.  cpumask
and nodemask also provide cpumask_pr_args() and nodemask_pr_args()
respectively which can be used to generate the two printf arguments
necessary to format the specified cpu/nodemask.

Signed-off-by: Tejun Heo 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

lib/vsprintf: implement bitmap printing through '%*pb[l]'

2015-02-14T05:21:36+00:00

bitmap and its derivatives such as cpumask and nodemask currently only
provide formatting functions which put the output string into the
provided buffer; however, how long this buffer should be isn't defined
anywhere and given that some of these bitmaps can be too large to be
formatted into an on-stack buffer it users sometimes are unnecessarily
forced to come up with creative solutions and compromises for the
buffer just to printk these bitmaps.

There have been a couple different attempts at making this easier.

1. Way back, PeterZ tried printk '%pb' extension with the precision
   for bit width - '%.*pb'.  This was intuitive and made sense but
   unfortunately triggered a compile warning about using precision
   for a pointer.

   http://lkml.kernel.org/g/1336577562.2527.58.camel@twins

2. I implemented bitmap_pr_cont[_list]() and its wrappers for cpumask
   and nodemask.  This works but PeterZ pointed out that pr_cont's
   tendency to produce broken lines when multiple CPUs are printing is
   bothering considering the usages.

   http://lkml.kernel.org/g/1418226774-30215-3-git-send-email-tj@kernel.org

So, this patch is another attempt at teaching printk and friends how
to print bitmaps.  It's almost identical to what PeterZ tried with
precision but it uses the field width for the number of bits instead
of precision.  The format used is '%*pb[l]', with the optional
trailing 'l' specifying list format instead of hex masks.

This is a valid format string and doesn't trigger compiler warnings;
however, it does make it impossible to specify output field width when
printing bitmaps.  I think this is an acceptable trade-off given how
much easier it makes printing bitmaps and that we don't have any
in-kernel user which is using the field width specification.  If any
future user wants to use field width with a bitmap, it'd have to
format the bitmap into a string buffer and then print that buffer with
width spec, which isn't different from how it should be done now.

This patch implements bitmap[_list]_string() which are called from the
vsprintf pointer() formatting function.  The implementation is mostly
identical to bitmap_scn[list]printf() except that the output is
performed in the vsprintf way.  These functions handle formatting into
too small buffers and sprintf() family of functions report the correct
overrun output length.

bitmap_scn[list]printf() are now thin wrappers around scnprintf().

Signed-off-by: Tejun Heo 
Acked-by: Peter Zijlstra (Intel) 
Cc: "David S. Miller" 
Cc: "James E.J. Bottomley" 
Cc: "John W. Linville" 
Cc: "Paul E. McKenney" 
Cc: Benjamin Herrenschmidt 
Cc: Chris Metcalf 
Cc: Chris Zankel 
Cc: Christoph Lameter 
Cc: Dmitry Torokhov 
Cc: Fenghua Yu 
Cc: Greg Kroah-Hartman 
Cc: Ingo Molnar 
Cc: Li Zefan 
Cc: Max Filippov 
Cc: Mike Travis 
Cc: Pekka Enberg 
Cc: Russell King 
Cc: Rusty Russell 
Cc: Steffen Klassert 
Cc: Steven Rostedt 
Cc: Thomas Gleixner 
Cc: Tony Luck 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds