Commit Graph

415 Commits

Author SHA1 Message Date
Uwe Kleine-König
26dd48780b parisc: led: Convert to platform remove callback returning void
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.

To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().

Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Signed-off-by: Helge Deller <deller@gmx.de>
2024-03-08 10:00:07 +01:00
Helge Deller
6472036581 parisc/power: Fix power soft-off button emulation on qemu
Make sure to start the kthread to check the power button on qemu as
well if the power button address was provided.
This fixes the qemu built-in system_powerdown runtime command.

Fixes: d0c2194729 ("parisc/power: Add power soft-off when running on qemu")
Signed-off-by: Helge Deller <deller@gmx.de>
Cc: stable@vger.kernel.org # v6.0+
2024-01-07 22:59:16 +01:00
Helge Deller
6ad6e15a9c parisc/power: Fix power soft-off when running on qemu
Firmware returns the physical address of the power switch,
so need to use gsc_writel() instead of direct memory access.

Fixes: d0c2194729 ("parisc/power: Add power soft-off when running on qemu")
Signed-off-by: Helge Deller <deller@gmx.de>
Cc: stable@vger.kernel.org # v6.0+
2023-11-18 18:59:30 +01:00
Helge Deller
fe0a9b8b22 parisc/power: Trivial whitespace cleanups and license update
Signed-off-by: Helge Deller <deller@gmx.de>
2023-10-30 14:54:40 +01:00
Helge Deller
d0c2194729 parisc/power: Add power soft-off when running on qemu
Signed-off-by: Helge Deller <deller@gmx.de>
Cc: stable@vger.kernel.org # v6.0+
2023-10-30 14:54:40 +01:00
Helge Deller
58ad89e866 parisc: sba-iommu: Fix comment when calculating IOC number
Signed-off-by: Helge Deller <deller@gmx.de>
2023-10-30 14:54:40 +01:00
Helge Deller
927c6c8aa2 parisc: iosapic.c: Fix sparse warnings
Signed-off-by: Helge Deller <deller@gmx.de>
2023-08-31 21:42:42 +02:00
Helge Deller
9a47a710cf parisc: ccio-dma: Fix sparse warnings
Signed-off-by: Helge Deller <deller@gmx.de>
2023-08-31 21:42:42 +02:00
Helge Deller
c1ebb94071 parisc: sba-iommu: Fix sparse warnigs
Fix sparse warnings, as pdir is __le64 *.

Signed-off-by: Helge Deller <deller@gmx.de>
2023-08-31 21:42:42 +02:00
Helge Deller
6428bc7bd3 parisc: sba_iommu: Fix build warning if procfs if disabled
Clean up the code, e.g. make proc_mckinley_root static, drop the now
empty mckinley header file and remove some unneeded ifdefs around procfs
functions.

Signed-off-by: Helge Deller <deller@gmx.de>
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202308300800.Jod4sHzM-lkp@intel.com/
Fixes: 77e0ddf097 ("parisc: ccio-dma: Create private runway procfs root entry")
2023-08-31 21:42:42 +02:00
Helge Deller
77e0ddf097 parisc: ccio-dma: Create private runway procfs root entry
Create an own procfs "runway" root entry for the CCIO driver.
No need to share it with the sba_iommu driver, as only one
of those busses can be active in one machine anyway.

Signed-off-by: Helge Deller <deller@gmx.de>
Reported-by: kernel test robot <lkp@intel.com>
Fixes: 547259580d ("parisc: Move proc_mckinley_root and proc_runway_root to sba_iommu")
Cc: <stable@vger.kernel.org> # v6.5
2023-08-28 18:00:27 +02:00
Helge Deller
789e527adf parisc: led: Rewrite LED/LCD driver to utilizize Linux LED subsystem
Rewrite the whole driver and drop the own code to calculate load
average, disk and LAN load. Switch instead to use the in-kernel LED
subsystem, which gives us quite some advantages, e.g.
- existing triggers for heartbeat and disk/lan activity can be used
- users can configre the LEDs at will to any existing trigger via
  /sys/class/leds
- less overhead since we don't need to run own timers
- fully integrated in Linux and as such cleaner code.

Note that the driver now depends on CONFIG_LEDS_CLASS which has to
be built-in and not as module.

Signed-off-by: Helge Deller <deller@gmx.de>
2023-08-28 17:58:14 +02:00
Helge Deller
d75ef5994d parisc: lasi: Initialize LASI driver via arch_initcall()
Move initialization code for LASI out of the GSC driver.
Since ASP and WAX have been moved in previous commits,
the GSC driver is now just a driver which provides library
functions for LASI, ASP and WAX and as such doesn't need
an own initialization function any longer.

Signed-off-by: Helge Deller <deller@gmx.de>
2023-08-27 11:09:10 +02:00
Helge Deller
ff0e833e98 parisc: asp: Initialize asp driver via arch_initcall()
Signed-off-by: Helge Deller <deller@gmx.de>
2023-08-27 11:09:10 +02:00
Helge Deller
ac65d9c90e parisc: wax: Initialize wax driver via arch_initcall()
Signed-off-by: Helge Deller <deller@gmx.de>
2023-08-27 11:09:10 +02:00
Helge Deller
9c2ca106c9 parisc: iosapic: Convert I/O Sapic driver to use arch_initcall()
Signed-off-by: Helge Deller <deller@gmx.de>
2023-08-27 10:54:10 +02:00
Helge Deller
53861a915a parisc: sba_iommu: Convert SBA IOMMU driver to use arch_initcall()
Signed-off-by: Helge Deller <deller@gmx.de>
2023-08-27 09:38:32 +02:00
Helge Deller
3b425dd2ae parisc: led: Move register_led_regions() to late_initcall()
Signed-off-by: Helge Deller <deller@gmx.de>
2023-08-27 09:36:23 +02:00
Helge Deller
df3f93596c parisc: lba: Convert LBA PCI bus driver to use arch_initcall()
Signed-off-by: Helge Deller <deller@gmx.de>
2023-08-27 09:25:48 +02:00
Helge Deller
ba8723b1ed parisc: gsc: Convert GSC bus driver to use arch_initcall()
This conversion includes LASI, ASP and WAX drivers for now.

Signed-off-by: Helge Deller <deller@gmx.de>
2023-08-27 09:17:37 +02:00
Helge Deller
63c1ce56ab parisc: ccio: Convert CCIO driver to use arch_initcall()
Signed-off-by: Helge Deller <deller@gmx.de>
2023-08-27 09:17:37 +02:00
Helge Deller
5f4f870a44 parisc: eisa: Convert HP EISA bus driver to use arch_initcall()
Signed-off-by: Helge Deller <deller@gmx.de>
2023-08-27 09:17:37 +02:00
Helge Deller
49663185d0 parisc: hppb: Convert HP PB bus driver to use arch_initcall()
Signed-off-by: Helge Deller <deller@gmx.de>
2023-08-27 09:17:37 +02:00
Helge Deller
07c34e9fdc parisc: dino: Convert dino PCI bus driver to use arch_initcall()
Signed-off-by: Helge Deller <deller@gmx.de>
2023-08-27 09:17:37 +02:00
Helge Deller
59bf860a97 parisc: Makefile: Adjust order in which drivers should be loaded
This is the order in which the drivers are initialized in setup.c.
The order is important when the drivers are convertet to use the
initcall_*() startup framework.

Signed-off-by: Helge Deller <deller@gmx.de>
2023-08-27 09:17:37 +02:00
Helge Deller
358ad816e5 parisc: led: Reduce CPU overhead for disk & lan LED computation
Older PA-RISC machines have LEDs which show the disk- and LAN-activity.
The computation is done in software and takes quite some time, e.g. on a
J6500 this may take up to 60% time of one CPU if the machine is loaded
via network traffic.

Since most people don't care about the LEDs, start with LEDs disabled and
just show a CPU heartbeat LED. The disk and LAN LEDs can be turned on
manually via /proc/pdc/led.

Signed-off-by: Helge Deller <deller@gmx.de>
Cc: <stable@vger.kernel.org>
2023-08-26 10:14:42 +02:00
Helge Deller
8f01caf0c5 parisc: Avoid ioremap() for same addresss in iosapic_register()
The LBA has already called ioremap() to get it's virtual address,
which can be used for the IOSAPIC as well.
Avoid calling ioremap() again and just reuse the correct
iomem address for the IOSAPIC.

Signed-off-by: Helge Deller <deller@gmx.de>
2023-08-22 10:24:47 +02:00
Helge Deller
e71d47dc2a parisc: lasi: Register LASI power-off feature as sys_off_handler
Prefer the Linux kernel sys_off_handler functionality over a
home-grown implementation.

Signed-off-by: Helge Deller <deller@gmx.de>
2023-08-20 20:23:46 +02:00
Helge Deller
547259580d parisc: Move proc_mckinley_root and proc_runway_root to sba_iommu
Clean up the procfs root entries for gsc, runway, and mckinley busses.

Signed-off-by: Helge Deller <deller@gmx.de>
2023-08-10 22:22:03 +02:00
Helge Deller
f310f8dd14 parisc: Move init function declarations into header file
Clean up the code to not have external function declarations
inside the C source files. Reduces warnings when compiled with W=1.

Signed-off-by: Helge Deller <deller@gmx.de>
2023-06-30 17:14:15 +02:00
Helge Deller
c8080024e0 parisc: dino: Make dino_init() returning void
Signed-off-by: Helge Deller <deller@gmx.de>
2023-06-30 17:14:15 +02:00
Helge Deller
1d72e83d63 parisc: lba_pci: Mark two variables __maybe_unused
Signed-off-by: Helge Deller <deller@gmx.de>
2023-06-30 17:14:15 +02:00
Helge Deller
8829428c16 parisc: ccio-dma: Fix kdoc and compiler warnings
Signed-off-by: Helge Deller <deller@gmx.de>
2023-06-30 17:14:14 +02:00
Helge Deller
f28a98779d parisc: pdc_stable: Fix kdoc and compiler warnings
Signed-off-by: Helge Deller <deller@gmx.de>
2023-06-30 17:14:14 +02:00
Helge Deller
226b8ab875 parisc: sba_iommu: Fix kdoc warnings
Signed-off-by: Helge Deller <deller@gmx.de>
2023-06-30 17:14:14 +02:00
Guilherme G. Piccoli
829632dae8 parisc: Replace regular spinlock with spin_trylock on panic path
The panic notifiers' callbacks execute in an atomic context, with
interrupts/preemption disabled, and all CPUs not running the panic
function are off, so it's very dangerous to wait on a regular
spinlock, there's a risk of deadlock.

Refactor the panic notifier of parisc/power driver to make use
of spin_trylock - for that, we've added a second version of the
soft-power function. Also, some comments were reorganized and
trailing white spaces, useless header inclusion and blank lines
were removed.

Cc: "James E.J. Bottomley" <James.Bottomley@HansenPartnership.com>
Cc: Jeroen Roovers <jer@xs4all.nl>
Acked-by: Helge Deller <deller@gmx.de> # parisc
Signed-off-by: Guilherme G. Piccoli <gpiccoli@igalia.com>
Signed-off-by: Helge Deller <deller@gmx.de>
2023-05-03 17:43:26 +02:00
Niklas Schnelle
fcbfe8121a
Kconfig: introduce HAS_IOPORT option and select it as necessary
We introduce a new HAS_IOPORT Kconfig option to indicate support for I/O
Port access. In a future patch HAS_IOPORT=n will disable compilation of
the I/O accessor functions inb()/outb() and friends on architectures
which can not meaningfully support legacy I/O spaces such as s390.

The following architectures do not select HAS_IOPORT:

* ARC
* C-SKY
* Hexagon
* Nios II
* OpenRISC
* s390
* User-Mode Linux
* Xtensa

All other architectures select HAS_IOPORT at least conditionally.

The "depends on" relations on HAS_IOPORT in drivers as well as ifdefs
for HAS_IOPORT specific sections will be added in subsequent patches on
a per subsystem basis.

Co-developed-by: Arnd Bergmann <arnd@kernel.org>
Signed-off-by: Arnd Bergmann <arnd@kernel.org>
Acked-by: Johannes Berg <johannes@sipsolutions.net> # for ARCH=um
Acked-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Niklas Schnelle <schnelle@linux.ibm.com>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
2023-04-05 22:15:19 +02:00
Xu Panda
f2193bb2ee parisc: pdc_stable: use strscpy() to instead of strncpy()
The implementation of strscpy() is more robust and safer.
That's now the recommended way to copy NUL-terminated strings.

Signed-off-by: Xu Panda <xu.panda@zte.com.cn>
Signed-off-by: Yang Yang <yang.yang29@zte.com>
Signed-off-by: Helge Deller <deller@gmx.de>
2023-01-09 09:51:56 +01:00
Shang XiaoJing
41f563ab3c parisc: led: Fix potential null-ptr-deref in start_task()
start_task() calls create_singlethread_workqueue() and not checked the
ret value, which may return NULL. And a null-ptr-deref may happen:

start_task()
    create_singlethread_workqueue() # failed, led_wq is NULL
    queue_delayed_work()
        queue_delayed_work_on()
            __queue_delayed_work()  # warning here, but continue
                __queue_work()      # access wq->flags, null-ptr-deref

Check the ret value and return -ENOMEM if it is NULL.

Fixes: 3499495205 ("[PARISC] Use work queue in LED/LCD driver instead of tasklet.")
Signed-off-by: Shang XiaoJing <shangxiaojing@huawei.com>
Signed-off-by: Helge Deller <deller@gmx.de>
Cc: <stable@vger.kernel.org>
2022-12-17 23:19:38 +01:00
Helge Deller
a0c9f1f2e5 parisc: Export iosapic_serial_irq() symbol for serial port driver
The parisc serial port driver needs this symbol when it's compiled
as module.

Signed-off-by: Helge Deller <deller@gmx.de>
Reported-by: kernel test robot <lkp@intel.com>
Cc: <stable@vger.kernel.org>
2022-10-27 09:12:05 +02:00
Helge Deller
50f19697dd parisc: Use signed char for hardware path in pdc.h
Clean up the struct for hardware_path and drop the struct device_path
with a proper assignment of bc[] and mod members as signed chars.

This patch prepares for the kbuild change from Jason A. Donenfeld to
treat char as always unsigned.

Signed-off-by: Helge Deller <deller@gmx.de>
Cc: Jason A. Donenfeld <Jason@zx2c4.com>
2022-10-21 09:15:20 +02:00
Linus Torvalds
f2e44139f3 parisc architecture fixes and updates for kernel v6.1-rc1:
* Convert the PDC console to an early console
 * Unbreak mmap() of graphics card memory due to PAGE_SPECIAL pgtable flag
 * Reduce the size of the alternative tables
 * Align stifb graphics card memory size to 4MB
 * Spelling fixes
 -----BEGIN PGP SIGNATURE-----
 
 iHUEABYKAB0WIQS86RI+GtKfB8BJu973ErUQojoPXwUCY0mnYAAKCRD3ErUQojoP
 X1z3AQC204Pq725WM3rgEomIqFjk8QaU2UFcNQ74a52V3AUiiQD8C0DFmWDbssUv
 ZE5uC+ySlNd1hLkL3QYxrbbwuL+y+Q0=
 =YvFZ
 -----END PGP SIGNATURE-----

Merge tag 'parisc-for-6.1-1' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux

Pull parisc updates from Helge Deller:
 "Fixes:

   - When we added basic vDSO support in kernel 5.18 we introduced a bug
     which prevented a mmap() of graphic card memory. This is because we
     used the DMB (data memory break trap bit) page flag as special-bit,
     but missed to clear that bit when loading the TLB.

   - Graphics card memory size was not correctly aligned

   - Spelling fixes (from Colin Ian King)

  Enhancements:

   - PDC console (which uses firmware calls) now rewritten as early
     console

   - Reduced size of alternative tables"

* tag 'parisc-for-6.1-1' of git://git.kernel.org/pub/scm/linux/kernel/git/deller/parisc-linux:
  parisc: Fix spelling mistake "mis-match" -> "mismatch" in eisa driver
  parisc: Fix userspace graphics card breakage due to pgtable special bit
  parisc: fbdev/stifb: Align graphics memory size to 4MB
  parisc: Convert PDC console to an early console
  parisc: Reduce kernel size by packing alternative tables
2022-10-14 12:10:01 -07:00
Colin Ian King
34314cd615 parisc: Fix spelling mistake "mis-match" -> "mismatch" in eisa driver
There are several spelling mistakes in kernel error messages. Fix them.

Signed-off-by: Colin Ian King <colin.i.king@gmail.com>
Signed-off-by: Helge Deller <deller@gmx.de>
2022-10-14 20:06:30 +02:00
Linus Torvalds
93ed07a23f asm-generic updates for v6.1
This contains a series from Linus Walleij to unify the linux/io.h
 interface by making the ia64, alpha, parisc and sparc include
 asm-generic/io.h. All functions provided by the generic header are
 now available to all drivers, but the architectures can still override
 this. For the moment, mips and sh still don't include asm-generic/io.h
 but provide a full set of functions themselves.
 
 There are also a few minor cleanups unrelated to this.
 -----BEGIN PGP SIGNATURE-----
 
 iQIzBAABCgAdFiEEo6/YBQwIrVS28WGKmmx57+YAGNkFAmM8L6sACgkQmmx57+YA
 GNnmKg/9EgTf+0+KoQ193Xhvav2niOud/AQP2H5wvKlt4csIq5CRb8cHMYYSpmn5
 MzifFA4/meeVkbh0LPX4Hue6XqegIe14CRZtIYep5yc1DGjLzEGHTWxkxDWZ5LK9
 8mUVPsm7ddCDt99XmI2mZNgy4Qktn3ePo1N+4tgmeadYdYXQXXsL1H1HlfQeVNRk
 IGvsGx/75QkhOrA20xNL+uwQ9n6ekdwOgj2Z2Q3SHNJPNpyPJLcspBbZ6HEMUG6/
 NnEnfqVOz9i1Nonx7Gf6gKXvgRzyU4AIaUW0xADBwtY5/Yq3bC0LQj4HTANkmWo8
 lBHAoSm3GicrzMMFb5eeKHkePX5iGSiTChB+TBWN/3Ofxzfma668lDuxdFYGOW+F
 gK5UsYZXpMuKuVTdWpp2k+swFkNtlBOQoYE6kE2NOLkR40FjaxS67LPAbDMtY9mb
 WIw10waN5dX0Dxv+ecJoaiQ7TeoIT5zI1plXeh1eJqARHmBX+o45tsVDSZAEvwNG
 EoVRjtG/aJDXwPDI2OiohEwckadIUnDbnXQI4DOgACy8MPkNITrth6E/Xnv3sAtx
 QLAojsMp6A3BNlen7LoyWpCWeb177OdnQZJmVVEar8cqAcWX+6HtMtlDIGi1dvEq
 Wl8JLysHk5jXdqNGYiKYtT8ocGZux0x2fYT/x/OnetBkn5D3DrE=
 =4dNP
 -----END PGP SIGNATURE-----

Merge tag 'asm-generic-6.1' of git://git.kernel.org/pub/scm/linux/kernel/git/arnd/asm-generic

Pull asm-generic updates from Arnd Bergmann:
 "This contains a series from Linus Walleij to unify the linux/io.h
  interface by making the ia64, alpha, parisc and sparc include
  asm-generic/io.h.

  All functions provided by the generic header are now available to all
  drivers, but the architectures can still override this.

  For the moment, mips and sh still don't include asm-generic/io.h but
  provide a full set of functions themselves.

  There are also a few minor cleanups unrelated to this"

* tag 'asm-generic-6.1' of git://git.kernel.org/pub/scm/linux/kernel/git/arnd/asm-generic:
  alpha: add full ioread64/iowrite64 implementation
  parisc: Drop homebrewn io[read|write]64_[lo_hi|hi_lo]
  parisc: hide ioread64 declaration on 32-bit
  ia64: export memory_add_physaddr_to_nid to fix cxl build error
  asm-generic: Remove empty #ifdef SA_RESTORER
  parisc: Use the generic IO helpers
  parisc: Remove 64bit access on 32bit machines
  sparc: Fix the generic IO helpers
  alpha: Use generic <asm-generic/io.h>
2022-10-06 12:10:37 -07:00
Rolf Eike Beer
e359b70cc1 parisc: remove obsolete manual allocation aligning in iosapic
kmalloc() returns memory with __assume_kmalloc_alignment, which is
__alignof__(unsigned long long) for parisc.

Signed-off-by: Rolf Eike Beer <eike-kernel@sf-tec.de>
Signed-off-by: Helge Deller <deller@gmx.de>
2022-09-15 07:33:26 +02:00
Yang Yingliang
38238be4e8 parisc: ccio-dma: Add missing iounmap in error path in ccio_probe()
Add missing iounmap() before return from ccio_probe(), if ccio_init_resources()
fails.

Fixes: d46c742f82 ("parisc: ccio-dma: Handle kmalloc failure in ccio_init_resources()")
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Signed-off-by: Helge Deller <deller@gmx.de>
2022-09-13 08:50:31 +02:00
Linus Walleij
77bfc8bdb5
parisc: Remove 64bit access on 32bit machines
The parisc was using some readq/writeq accessors without special
considerations as to what will happen on 32bit CPUs if you do
this. Maybe we have been lucky that it "just worked" on 32bit
due to the compiler behaviour, or the code paths were never
executed.

Fix the two offending code sites like this:

arch/parisc/lib/iomap.c:

- Put ifdefs around the 64bit accessors and make sure
  that ioread64, ioread64be, iowrite64 and iowrite64be
  are not available on 32bit builds.

- Also fold in a bug fix where 64bit access was by
  mistake using 32bit writel() accessors rather
  than 64bit writeq().

drivers/parisc/sba_iommu.c:

- Access any 64bit registers using _lo_hi-semantics by way
  of the readq and writeq operations provided by
  <linux/io-64-nonatomic-lo-hi.h>

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Cc: James E.J. Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Helge Deller <deller@gmx.de>
Cc: linux-parisc@vger.kernel.org
Cc: linux-arch@vger.kernel.org
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: John David Anglin <dave.anglin@bell.net>
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
2022-09-10 09:55:55 +02:00
Li Qiong
d46c742f82 parisc: ccio-dma: Handle kmalloc failure in ccio_init_resources()
As the possible failure of the kmalloc(), it should be better
to fix this error path, check and return '-ENOMEM' error code.

Signed-off-by: Li Qiong <liqiong@nfschina.com>
Signed-off-by: Helge Deller <deller@gmx.de>
2022-08-22 11:09:17 +02:00
Wolfram Sang
4cb2643667 parisc: led: Move from strlcpy with unused retval to strscpy
Follow the advice of the below link and prefer 'strscpy' in this
subsystem. Conversion is 1:1 because the return value is not used.
Generated by a coccinelle script.

Link: https://lore.kernel.org/r/CAHk-=wgfRnXz0W3D37d01q3JFkr_i_uTL=V6A6G1oUZcprmknw@mail.gmail.com/
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Helge Deller <deller@gmx.de>
2022-08-22 11:09:17 +02:00
Jason Wang
db4538ad4d parisc: ccio-dma: Fix typo in comment
The double `was' is duplicated in the comment, remove one.

Signed-off-by: Jason Wang <wangborong@cdjrlc.com>
Signed-off-by: Helge Deller <deller@gmx.de>
2022-08-22 11:09:17 +02:00