When we added the alignment tests recently we failed to check they were
actually passing - oops.

They weren't passing, because the bitmap was full. We should also be a
bit more careful when checking the return code, a negative error return
could by divisible by our alignment value.

Fixes: b0345bbc ("powerpc/msi: Improve IRQ bitmap allocator")
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

Currently msi_bitmap_alloc_hwirqs() will round up any IRQ allocation requests
to the nearest power of 2. eg. ask for 5 IRQs and you'll get 8. This wastes a
lot of IRQs which can be a scarce resource.

For cxl we may require multiple IRQs for every context that is attached to the
accelerator. There may be 1000s of contexts attached, hence we can easily run
out of IRQs, especially if we are needlessly wasting them.

This changes the msi_bitmap_alloc_hwirqs() to allocate only the required number
of IRQs, hence avoiding this wastage. It keeps the natural alignment
requirement though.
Signed-off-by: NIan Munsie <imunsie@au1.ibm.com>
Signed-off-by: NMichael Neuling <mikey@neuling.org>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

Signed-off-by: NAnton Blanchard <anton@samba.org>
Signed-off-by: NMichael Ellerman <mpe@ellerman.id.au>

This reverts commit c822e737.

This commit conflicted with a bitmap allocator change that partially
accomplishes the same thing, but which does so more correctly.  Revert
this one until it can be respun on top of the correct change.
Signed-off-by: NScott Wood <scottwood@freescale.com>

Allocate msis such that each time a new interrupt is requested,
the SRS (MSIR register select) to be used is allocated in a
round-robin fashion.
The end result is that the msi interrupts will be spread across
distinct MSIRs with the main benefit that now users can set
affinity to each msi int through the mpic irq backing up the
MSIR register.
This is achieved with the help of a newly introduced msi bitmap
api that allows specifying the starting point when searching
for a free msi interrupt.
Signed-off-by: NLaurentiu Tudor <Laurentiu.Tudor@freescale.com>
Cc: Scott Wood <scottwood@freescale.com>
Cc: Mihai Caraman <mihai.caraman@freescale.com>
Signed-off-by: NScott Wood <scottwood@freescale.com>

This patch uses of_node_init() to initialize the kobject in the fake
node used in test_of_node(), to avoid following kobject warning.

[    0.897654] kobject: '(null)' (c0000007ca183a08): is not initialized, yet kobject_put() is being called.
[    0.897682] ------------[ cut here ]------------
[    0.897688] WARNING: at lib/kobject.c:670
[    0.897692] Modules linked in:
[    0.897701] CPU: 4 PID: 1 Comm: swapper/0 Not tainted 3.14.0+ #1
[    0.897708] task: c0000007ca100000 ti: c0000007ca180000 task.ti: c0000007ca180000
[    0.897715] NIP: c00000000046a1f0 LR: c00000000046a1ec CTR: 0000000001704660
[    0.897721] REGS: c0000007ca1835c0 TRAP: 0700   Not tainted  (3.14.0+)
[    0.897727] MSR: 8000000000029032 <SF,EE,ME,IR,DR,RI>  CR: 28000024  XER: 0000000d
[    0.897749] CFAR: c0000000008ef4ec SOFTE: 1
GPR00: c00000000046a1ec c0000007ca183840 c0000000014c59b8 000000000000005c
GPR04: 0000000000000001 c000000000129770 0000000000000000 0000000000000001
GPR08: 0000000000000000 0000000000000000 0000000000000000 0000000000003fef
GPR12: 0000000000000000 c00000000f221200 c00000000000c350 0000000000000000
GPR16: 0000000000000000 0000000000000000 0000000000000000 0000000000000000
GPR20: 0000000000000000 0000000000000000 0000000000000000 0000000000000000
GPR24: 0000000000000000 c00000000144e808 c000000000c56f20 00000000000000d8
GPR28: c000000000cd5058 0000000000000000 c000000001454ca8 c0000007ca183a08
[    0.897856] NIP [c00000000046a1f0] .kobject_put+0xa0/0xb0
[    0.897863] LR [c00000000046a1ec] .kobject_put+0x9c/0xb0
[    0.897868] Call Trace:
[    0.897874] [c0000007ca183840] [c00000000046a1ec] .kobject_put+0x9c/0xb0 (unreliable)
[    0.897885] [c0000007ca1838c0] [c000000000743f9c] .of_node_put+0x2c/0x50
[    0.897894] [c0000007ca183940] [c000000000c83954] .test_of_node+0x1dc/0x208
[    0.897902] [c0000007ca183b80] [c000000000c839a4] .msi_bitmap_selftest+0x24/0x38
[    0.897913] [c0000007ca183bf0] [c00000000000bb34] .do_one_initcall+0x144/0x200
[    0.897922] [c0000007ca183ce0] [c000000000c748e4] .kernel_init_freeable+0x2b4/0x394
[    0.897931] [c0000007ca183db0] [c00000000000c374] .kernel_init+0x24/0x130
[    0.897940] [c0000007ca183e30] [c00000000000a2f4] .ret_from_kernel_thread+0x5c/0x68
[    0.897947] Instruction dump:
[    0.897952] 7fe3fb78 38210080 e8010010 ebe1fff8 7c0803a6 4800014c e89f0000 3c62ff6e
[    0.897971] 7fe5fb78 3863a950 48485279 60000000 <0fe00000> 39000000 393f0038 4bffff80
[    0.897992] ---[ end trace 1eeffdb9f825a556 ]---
Signed-off-by: NLi Zhong <zhong@linux.vnet.ibm.com>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

Device tree nodes are already treated as objects, and we already want to
expose them to userspace which is done using the /proc filesystem today.
Right now the kernel has to do a lot of work to keep the /proc view in
sync with the in-kernel representation. If device_nodes are switched to
be kobjects then the device tree code can be a whole lot simpler. It
also turns out that switching to using /sysfs from /proc results in
smaller code and data size, and the userspace ABI won't change if
/proc/device-tree symlinks to /sys/firmware/devicetree/base.

v7: Add missing sysfs_bin_attr_init()
v6: Add __of_add_property() early init fixes from Pantelis
v5: Rename firmware/ofw to firmware/devicetree
    Fix updating property values in sysfs
v4: Fixed build error on Powerpc
    Fixed handling of dynamic nodes on powerpc
v3: Fixed handling of duplicate attribute and child node names
v2: switch to using sysfs bin_attributes which solve the problem of
    reporting incorrect property size.
Signed-off-by: NGrant Likely <grant.likely@secretlab.ca>
Tested-by: NSascha Hauer <s.hauer@pengutronix.de>
Cc: Rob Herring <rob.herring@calxeda.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: David S. Miller <davem@davemloft.net>
Cc: Nathan Fontenot <nfont@linux.vnet.ibm.com>
Cc: Pantelis Antoniou <panto@antoniou-consulting.com>

Disintegrate asm/system.h for PowerPC.
Signed-off-by: NDavid Howells <dhowells@redhat.com>
Acked-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
cc: linuxppc-dev@lists.ozlabs.org

include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h

percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files.  percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.

percpu.h -> slab.h dependency is about to be removed.  Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability.  As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.

  http://userweb.kernel.org/~tj/misc/slabh-sweep.py

The script does the followings.

* Scan files for gfp and slab usages and update includes such that
  only the necessary includes are there.  ie. if only gfp is used,
  gfp.h, if slab is used, slab.h.

* When the script inserts a new include, it looks at the include
  blocks and try to put the new include such that its order conforms
  to its surrounding.  It's put in the include block which contains
  core kernel includes, in the same order that the rest are ordered -
  alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
  doesn't seem to be any matching order.

* If the script can't find a place to put a new include (mostly
  because the file doesn't have fitting include block), it prints out
  an error message indicating which .h file needs to be added to the
  file.

The conversion was done in the following steps.

1. The initial automatic conversion of all .c files updated slightly
   over 4000 files, deleting around 700 includes and adding ~480 gfp.h
   and ~3000 slab.h inclusions.  The script emitted errors for ~400
   files.

2. Each error was manually checked.  Some didn't need the inclusion,
   some needed manual addition while adding it to implementation .h or
   embedding .c file was more appropriate for others.  This step added
   inclusions to around 150 files.

3. The script was run again and the output was compared to the edits
   from #2 to make sure no file was left behind.

4. Several build tests were done and a couple of problems were fixed.
   e.g. lib/decompress_*.c used malloc/free() wrappers around slab
   APIs requiring slab.h to be added manually.

5. The script was run on all .h files but without automatically
   editing them as sprinkling gfp.h and slab.h inclusions around .h
   files could easily lead to inclusion dependency hell.  Most gfp.h
   inclusion directives were ignored as stuff from gfp.h was usually
   wildly available and often used in preprocessor macros.  Each
   slab.h inclusion directive was examined and added manually as
   necessary.

6. percpu.h was updated not to include slab.h.

7. Build test were done on the following configurations and failures
   were fixed.  CONFIG_GCOV_KERNEL was turned off for all tests (as my
   distributed build env didn't work with gcov compiles) and a few
   more options had to be turned off depending on archs to make things
   build (like ipr on powerpc/64 which failed due to missing writeq).

   * x86 and x86_64 UP and SMP allmodconfig and a custom test config.
   * powerpc and powerpc64 SMP allmodconfig
   * sparc and sparc64 SMP allmodconfig
   * ia64 SMP allmodconfig
   * s390 SMP allmodconfig
   * alpha SMP allmodconfig
   * um on x86_64 SMP allmodconfig

8. percpu.h modifications were reverted so that it could be applied as
   a separate patch and serve as bisection point.

Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: NTejun Heo <tj@kernel.org>
Guess-its-ok-by: NChristoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>

Signed-off-by: NMichael Ellerman <michael@ellerman.id.au>
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>

There are now two almost identical implementations of an MSI bitmap
allocator, one in mpic_msi.c and the other in fsl_msi.c.

Merge them together and put the result in msi_bitmap.c.  Some of the
MPIC bits will remain to provide a nicer interface for the MPIC users.

In the process we fix two buglets.  The first is that the allocation
routines, now msi_bitmap_alloc_hwirqs(), returned an unsigned result,
even though they use -1 to indicate allocation failure.  Although all
the callers were checking correctly, it is much better for the routine
to just return an int.  At least until someone wants > ~2 billion MSIs.

The second buglet is that the device tree reservation logic only
allowed power-of-two reservations.  AFAICT that didn't effect any
existing code but it's nicer if we can reserve arbitrary irqs from MSI
use.

We also add some selftests, which exposed the two buglets and now test
for them, as well as some basic sanity tests.  The tests are only built
when CONFIG_DEBUG_KERNEL=y.
Signed-off-by: NMichael Ellerman <michael@ellerman.id.au>
Signed-off-by: NPaul Mackerras <paulus@samba.org>