The step 2 tests to make sure that the triggers are unique is not
needed for single source triggers. These were already trivially
validated in step 1.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Cc: Frank Mori Hess <fmhess@users.sourceforge.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The step 2 tests of the start_src, scan_begin_src, and scan_end_src
triggers to make sure they are unique are not needed.  These triggers
all only have one source and it was trivially validated in step 1.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Cc: Frank Mori Hess <fmhess@users.sourceforge.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The step 2 tests of the start_src, scan_begin_src, and scan_end_src
triggers to make sure they are unique are not needed.  These triggers
all only have one source and it was trivially validated in step 1.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Cc: Frank Mori Hess <fmhess@users.sourceforge.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

These printk's are #ifdef'ed out with an undefined symbol anyway.
Just remove them.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Cc: Frank Mori Hess <fmhess@users.sourceforge.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Currently the amount of pools each client can use is limited to 16, this is
and arbitrary limit which isn't really required by current implementation.

This places and arbitrary limit on the number of mounted filesystems that
can use cleancache.

This patch removes that limit and uses IDR to do sparse mapping of pools
in each client.
Signed-off-by: NSasha Levin <levinsasha928@gmail.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Changed setting the status bits to a byte-oriented approach. Number of
written status bits is now based on VHCI_NPORT, instead of casting
status buffer to a long. This fixes a stack corruption bug on 64-bit
based architectures, and potential byte order / endianness related
issues.

In addition updated function comments, and log statements.
Signed-off-by: NBart Westgeest <bart@elbrys.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch fixes an uninitialized variable warning in
alloc_zspage().  It also fixes the secondary issue of
prev_page leaving scope on each loop iteration.  The only
reason this ever worked was because prev_page was occupying
the same space on the stack on each iteration.
Signed-off-by: NSeth Jennings <sjenning@linux.vnet.ibm.com>
Reported-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Add whitespace to the _8255_* defines to improve readability.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Cc: Frank Mori Hess <fmhess@users.sourceforge.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The name 'do_config' is pretty generic, and this function is the
only one in this driver without namespace. Add namespace to it
just to avoid any issues.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Cc: Frank Mori Hess <fmhess@users.sourceforge.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The subdev_8255_interrupt function indirectly calls subdev_8255_io.
For aesthetic reasons, move the subdev_8255_io function.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Cc: Frank Mori Hess <fmhess@users.sourceforge.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Refactor the subdev_8255_insn function to follow what the user
space library is doing.

1) This function requires 'data' to contain 2 parameters. Add a
sanity check for this (insn->n != 2).

2) The 'data' parameters are actually a 'mask' of the valid bits and
the actual 'bits' to modify. Create local variables of these names
to help with maintainability.

3) The value returned in data[1] reflects the actual state of the
8255 io pins after the update of the masked bits. Fetch this data
in a local variable and then set data[1].

4) The user space library only checks for a < 0 return to indicate
an error. It appears that the 'return 2' in this function is supposed
to indicate the number of data parameters used to perform the command.
Return the insn->n value for this, the open coded '2' looks wrong.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Iam Abbott <abbotti@mev.co.uk>
Cc: Frank Mori Hess <fmhess@users.sourceforge.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

To improve readability, use a locale variable to hold the 8255's
private iobase value.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Cc: Frank Mori Hess <fmhess@users.sourceforge.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Cc: Frank Mori Hess <fmhess@users.sourceforge.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The initialization of the 8255 subdevice can fail. Make sure to
check for it.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Cc: Frank Mori Hess <fmhess@users.sourceforge.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The have_irq variable in the 8255 private data is not used by the driver,
remove it.  This removes the need to access the private data in the
subdev_8255_init_irq function.

Also, add a bit of whitespace to improve the readability.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Cc: Frank Mori Hess <fmhess@users.sourceforge.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Refactor the initialization of the 8255 so that the private data
allocation is done before the subdevice is setup. This makes sure
that the setup is complete.

Also, add a bit of whitespace to improve the readability.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Cc: Frank Mori Hess <fmhess@users.sourceforge.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Push out the rename of the private variables to improve the readability
of the driver.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Cc: Frank Mori Hess <fmhess@users.sourceforge.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The 8255's private data variables cb_arg and cb_func are actually
the iobase for the 8255 device and the pointer to the io function
to access the device. Rename them as such to improve the readability
of the driver.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Cc: Frank Mori Hess <fmhess@users.sourceforge.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The macros CALLBACK_ARG, CALLBACK_FUNC, and subdevpriv all rely
on a a local variable having a specific name. Replace the macros
with local variables wherever they occur.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Cc: Frank Mori Hess <fmhess@users.sourceforge.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Rename the 8255's private data structure from subdev_8255_struct
to subdev_8255_private.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Cc: Frank Mori Hess <fmhess@users.sourceforge.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Instead of accessing the dev->subdevices directly as an array,
use a pointer. This method is more common in the comedi subsystem.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Cc: Frank Mori Hess <fmhess@users.sourceforge.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Replace the printk calls with dev_info, dev_warn, etc.
Signed-off-by: NH Hartley Sweeten <hsweeten@visionengravers.com>
Cc: Ian Abbott <abbotti@mev.co.uk>
Cc: Frank Mori Hess <fmhess@users.sourceforge.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch cleans up the vb_device_info struct and the related
functions.
The cleanup decreases the size of the compiled module by
about 10kB.

== Remove fields in vb_device_info that are never read: ==
pOutputSelect
pRGBSenseData
pRGBSenseData2
pVideoSenseData
pVideoSenseData2
pYCSenseData
pYCSenseData2
CR49
pXGINew_I2CDefinition
pCR2E
pCR2F
pCR46
pCR47
pCRD0
pCRDE
pSR40
pSR41
pCR47

=== Remove the corresponding 'constants' ===
XGI330_RGBSenseData
XGI330_RGBSenseData2
XGI330_VideoSenseData
XGI330_VideoSenseData2
XGI330_YCSenseData
XGI330_YCSenseData2
XGI330_CR49
XG40_I2CDefinition
XG21_CR2E
XG21_CR2F
XG21_CR46
XG21_CR47
XG27_CRD0
XG27_CRDE
XGI330_OutputSelect

== Remove 'constant fields' and replace constant value with #define ==
pSR07 = XGI330_SR07 -> 0x18
pSR1F = XGI330_SR1F -> 0
pSR23 = XGI330_SR23 -> 0xf6
pSR24 = XGI330_SR24 -> 0x0d
pSR33 = XGI330_SR33 ->0
pCRT2Data_1_2 = XGI330_CRT2Data_1_2 -> 0
pCRT2Data_4_D = XGI330_CRT2Data_4_D -> 0
pCRT2Data_4_E = XGI330_CRT2Data_4_E -> 0
pCRT2Data_4_10 = XGI330_CRT2Data_4_10 -> 0x80
pSR36 = XG27_SR36 -> 0x30
pCR8F = &XG27_CR8F -> 0x0C
pSR40 = XG27_SR40 -> 0x04
pSR41 = XG27_SR41 ->0x00
pSR31 = XGI330_SR31 -> 0xc0
pSR32 = XGI330_SR32 -> 0xc0
SR25 = XGI330_sr25 -> 0 (we only use XGI330_sr25[0])

== Constant fields with 'dead' code: ==
pSoftSetting is set to XGI330_SoftSetting = 0x30
-> if (*pVBInfo->pSoftSetting & SoftDRAMType) is never true since
SoftDRAMType = 0x80
-> if (*pVBInfo->pSoftSetting & ModeSoftSetting) is never true since
ModeSoftSetting = 0x04
--> remove the code, remove pSoftSetting, remove XGI330_SoftSetting

pDVOSetting is set to XG21_DVOSetting = 0
-> if (((*pVBInfo->pDVOSetting) & 0xC0) == 0xC0) is never true
--> remove the code, remove pDVOSetting, remove XG21_DVOSetting

pXGINew_DRAMTypeDefinition is set to &XG40_DRAMTypeDefinition 0xFF
-> if (*pVBInfo->pXGINew_DRAMTypeDefinition != 0x0C) is always true
--> remove the if and remove pXGINew_DRAMTypeDefinition
remove XG40_DRAMTypeDefinition

== Replace pointer to unsigned char with unsigned char variable
and assign value of referenced pointer: ==
pSR21 -> SR21, remove XGI330_SR21
pSR22 -> SR22, remove XGI330_SR22
pXGINew_CR97 -> XGINew_CR97, remove XG20_CR97, XG27_CR97 and Z11m_CR97
Signed-off-by: NPeter Huewe <peterhuewe@gmx.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch replaces some of the XGI internal structs by their
counterparts in the SiS driver.
XGI330_LVDSDataStruct -> SiS_LVDSData
XGI330_LCDDataStruct -> SiS_LCDData
XGI330_CHTVDataStruct -> SiS_LVDSData
Signed-off-by: NPeter Huewe <peterhuewe@gmx.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

For TV Devices:
The values taken from XGI_TVDelayList are always overwritten with the
values from XGI_TVDelayList2 since the if condition for using the values
from XGI_TVDelayList2 is identical with the check to enter this scope and
thus always true.
The delay values in XGI_TVDelayList2 is always 0x22 so we
can simply replace it with this constant value.

For LCD Devices:
The LCD_DelayCompensation field is always set to 0x12 so we can simply
replace this field with a constant value.

This saves about 500 bytes in compiled size.
Signed-off-by: NPeter Huewe <peterhuewe@gmx.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Since XGI_CloseCRTC does not perform anything useful we can simply
remove it.
Signed-off-by: NPeter Huewe <peterhuewe@gmx.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch simplyfies the XGI_GetVCLK2Ptr a bit by moving the +=25 to
a define and removing statements without effect.
Signed-off-by: NPeter Huewe <peterhuewe@gmx.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch removes the arrays in XGI_GetVCLK2Ptr which each contain only
one value four times and replaces them with their constant value.
Signed-off-by: NPeter Huewe <peterhuewe@gmx.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Since the first three entries in XGINew_DDRDRAM_TYPE{340,20} are never
used, we can simply remove them;
and instead of passing XGINew_DDRDRAM_TYPE with an index we can simply
pass the value directly to XGINew_SetDRAMSize20Reg.
Signed-off-by: NPeter Huewe <peterhuewe@gmx.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Since the function XGINew_SetDRAMSizingType is only called from one
location and consist only of 2 valuable lines we can simply inline it
here.
Signed-off-by: NPeter Huewe <peterhuewe@gmx.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Since XGINew_SetDRAMSize20Reg now handles both cases we can remove
the code duplication in XGINew_DDRSizing340.
Signed-off-by: NPeter Huewe <peterhuewe@gmx.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch consolidates the almost identical functions
XGINew_SetDRAMSizeReg and XGINew_SetDRAMSize20Reg as they are
implemented identically except one division factor.

The changed factor is now reflected in the input data.
Signed-off-by: NPeter Huewe <peterhuewe@gmx.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch removes all unnecessary, redundant and superfluous header
includes from xgifb.

Tested on hp t5325 (XGI Z11)
Signed-off-by: NPeter Huewe <peterhuewe@gmx.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch adds a simple #include guard to vb_table.h
Signed-off-by: NPeter Huewe <peterhuewe@gmx.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This adds a mutex for fb_mmap around smem_start and smem_len
so the mutex inside the fb_mmap() is actually used.  Changing of
these fields before calling the framebuffer_register() are not mutexed.

We check whether framebuffer_register has been called by reading
fbinfo->count.

See 537a1bf0 - "fbdev: add mutex for fb_mmap locking" by Krzysztof Helt
for details.
Signed-off-by: NPeter Huewe <peterhuewe@gmx.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

This patch removes assignments to the fb_fix_screeninfo struct which are
overwritten by the memset in XGIfb_get_fix() a few lines later.
Since the name/id might be useful this was moved to XGIfb_get_fix().
Signed-off-by: NPeter Huewe <peterhuewe@gmx.de>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

All the functionality is now supported by pstore and pstore_ram drivers.
Signed-off-by: NAnton Vorontsov <anton.vorontsov@linaro.org>
Acked-by: NKees Cook <keescook@chromium.org>
Acked-by: NColin Cross <ccross@android.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Having automatic updates seems pointless for production system, and
even dangerous and thus counter-productive:

1. If we can mount pstore, or read files, we can as well read
   /proc/kmsg. So, there's little point in duplicating the
   functionality and present the same information but via another
   userland ABI;

2. Expecting the kernel to behave sanely after oops/panic is naive.
   It might work, but you'd rather not try it. Screwed up kernel
   can do rather bad things, like recursive faults[1]; and pstore
   rather provoking bad things to happen. It uses:

   1. Timers (assumes sane interrupts state);
   2. Workqueues and mutexes (assumes scheduler in a sane state);
   3. kzalloc (a working slab allocator);

   That's too much for a dead kernel, so the debugging facility
   itself might just make debugging harder, which is not what
   we want.

Maybe for non-oops message types it would make sense to re-enable
automatic updates, but so far I don't see any use case for this.
Even for tracing, it has its own run-time/normal ABI, so we're
only interested in pstore upon next boot, to retrieve what has
gone wrong with HW or SW.

So, let's disable the updates by default.

[1]
BUG: unable to handle kernel paging request at fffffffffffffff8
IP: [<ffffffff8104801b>] kthread_data+0xb/0x20
[...]
Process kworker/0:1 (pid: 14, threadinfo ffff8800072c0000, task ffff88000725b100)
[...
Call Trace:
 [<ffffffff81043710>] wq_worker_sleeping+0x10/0xa0
 [<ffffffff813687a8>] __schedule+0x568/0x7d0
 [<ffffffff8106c24d>] ? trace_hardirqs_on+0xd/0x10
 [<ffffffff81087e22>] ? call_rcu_sched+0x12/0x20
 [<ffffffff8102b596>] ? release_task+0x156/0x2d0
 [<ffffffff8102b45e>] ? release_task+0x1e/0x2d0
 [<ffffffff8106c24d>] ? trace_hardirqs_on+0xd/0x10
 [<ffffffff81368ac4>] schedule+0x24/0x70
 [<ffffffff8102cba8>] do_exit+0x1f8/0x370
 [<ffffffff810051e7>] oops_end+0x77/0xb0
 [<ffffffff8135c301>] no_context+0x1a6/0x1b5
 [<ffffffff8135c4de>] __bad_area_nosemaphore+0x1ce/0x1ed
 [<ffffffff81053156>] ? ttwu_queue+0xc6/0xe0
 [<ffffffff8135c50b>] bad_area_nosemaphore+0xe/0x10
 [<ffffffff8101fa47>] do_page_fault+0x2c7/0x450
 [<ffffffff8106e34b>] ? __lock_release+0x6b/0xe0
 [<ffffffff8106bf21>] ? mark_held_locks+0x61/0x140
 [<ffffffff810502fe>] ? __wake_up+0x4e/0x70
 [<ffffffff81185f7d>] ? trace_hardirqs_off_thunk+0x3a/0x3c
 [<ffffffff81158970>] ? pstore_register+0x120/0x120
 [<ffffffff8136a37f>] page_fault+0x1f/0x30
 [<ffffffff81158970>] ? pstore_register+0x120/0x120
 [<ffffffff81185ab8>] ? memcpy+0x68/0x110
 [<ffffffff8115875a>] ? pstore_get_records+0x3a/0x130
 [<ffffffff811590f4>] ? persistent_ram_copy_old+0x64/0x90
 [<ffffffff81158bf4>] ramoops_pstore_read+0x84/0x130
 [<ffffffff81158799>] pstore_get_records+0x79/0x130
 [<ffffffff81042536>] ? process_one_work+0x116/0x450
 [<ffffffff81158970>] ? pstore_register+0x120/0x120
 [<ffffffff8115897e>] pstore_dowork+0xe/0x10
 [<ffffffff81042594>] process_one_work+0x174/0x450
 [<ffffffff81042536>] ? process_one_work+0x116/0x450
 [<ffffffff81042e13>] worker_thread+0x123/0x2d0
 [<ffffffff81042cf0>] ? manage_workers.isra.28+0x120/0x120
 [<ffffffff81047d8e>] kthread+0x8e/0xa0
 [<ffffffff8136ba74>] kernel_thread_helper+0x4/0x10
 [<ffffffff8136a199>] ? retint_restore_args+0xe/0xe
 [<ffffffff81047d00>] ? __init_kthread_worker+0x70/0x70
 [<ffffffff8136ba70>] ? gs_change+0xb/0xb
Code: be e2 00 00 00 48 c7 c7 d1 2a 4e 81 e8 bf fb fd ff 48 8b 5d f0 4c 8b 65 f8 c9 c3 0f 1f 44 00 00 48 8b 87 08 02 00 00 55 48 89 e5 <48> 8b 40 f8 5d c3 66 66 66 66 66 66 2e 0f 1f 84 00 00 00 00 00
RIP  [<ffffffff8104801b>] kthread_data+0xb/0x20
 RSP <ffff8800072c1888>
CR2: fffffffffffffff8
---[ end trace 996a332dc399111d ]---
Fixing recursive fault but reboot is needed!
Signed-off-by: NAnton Vorontsov <anton.vorontsov@linaro.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

There is no behavioural change, the default value is still 60 seconds.
Signed-off-by: NAnton Vorontsov <anton.vorontsov@linaro.org>
Acked-by: NKees Cook <keescook@chromium.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

The code tried to maintain the global list of persistent ram zones,
which isn't a great idea overall, plus since Android's ram_console
is no longer there, we can remove some unused functions.
Signed-off-by: NAnton Vorontsov <anton.vorontsov@linaro.org>
Acked-by: NKees Cook <keescook@chromium.org>
Acked-by: NColin Cross <ccross@android.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>