On 9/5/19 3:37 pm, Nicholas Piggin wrote:
Andrew Donnellan's on May 9, 2019 3:11 pm:
SCOM_DEBUGFS is really not needed for anything other than low-level
hardware debugging.
opal-prd uses its own interface (/dev/prd) for SCOM access, so it doesn't
need SCOM_DEBUGFS.
At some point in the future
On Thu, May 9, 2019 at 3:38 PM Nicholas Piggin wrote:
>
> Andrew Donnellan's on May 9, 2019 3:11 pm:
> > SCOM_DEBUGFS is really not needed for anything other than low-level
> > hardware debugging.
> >
> > opal-prd uses its own interface (/dev/prd) for SCOM access, so it doesn't
> > need
Andrew Donnellan's on May 9, 2019 3:11 pm:
> SCOM_DEBUGFS is really not needed for anything other than low-level
> hardware debugging.
>
> opal-prd uses its own interface (/dev/prd) for SCOM access, so it doesn't
> need SCOM_DEBUGFS.
>
> At some point in the future we'll introduce a debug config
On Thu, May 02, 2019 at 11:08:55AM +, Horia Geanta wrote:
>
> >> +
> >> static inline void wr_reg32(void __iomem *reg, u32 data)
> >> {
> >>if (caam_little_end)
> >
> > This crashes on my p5020ds. Did you test on powerpc?
> >
> > # first bad commit:
Once upon a time, the SCOM access code was used by the WSP platform as well
as powernv. Thus it made sense to have a generic SCOM access interface to
abstract between different platforms.
Now that it's just powernv, with no other platforms currently on the
horizon, let's rip out scom_controller
SCOM_DEBUGFS is really not needed for anything other than low-level
hardware debugging.
opal-prd uses its own interface (/dev/prd) for SCOM access, so it doesn't
need SCOM_DEBUGFS.
At some point in the future we'll introduce a debug config fragment where
this can go instead.
Signed-off-by:
Signed-off-by: Andrew Donnellan
---
v1->v2:
- new patch
---
arch/powerpc/platforms/powernv/opal-xscom.c | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/arch/powerpc/platforms/powernv/opal-xscom.c
b/arch/powerpc/platforms/powernv/opal-xscom.c
index
The powernv platform is the only one that directly accesses SCOMs. Move the
support code to platforms/powernv, and get rid of the PPC_SCOM Kconfig
option, as SCOM support is always selected when compiling for powernv.
This also means that the Kconfig item for CONFIG_SCOM_DEBUGFS will actually
Nothing is using scom_map_device() or scom_find_parent(). Remove them.
Also don't export scom_controller, there are no other users of it.
Signed-off-by: Andrew Donnellan
---
v1->v2:
- new patch
---
arch/powerpc/platforms/powernv/scom.c | 67 ---
On 07/05/19 10:43 PM, Mahesh J Salgaonkar wrote:
On 2019-04-16 16:06:13 Tue, Hari Bathini wrote:
OPAL loads kernel & initrd at 512MB offset (256MB size), also exported
as ibm,opal/dump/fw-load-area. So, if boot memory size of FADump is
less than 768MB, kernel memory to be exported as
On 08/05/19 5:55 AM, Nicholas Piggin wrote:
Hari Bathini's on April 16, 2019 8:35 pm:
From: Hari Bathini
Export /proc/opalcore file to analyze opal crashes. Since opalcore can
be generated independent of CONFIG_FA_DUMP support in kernel, add this
support under a new kernel config option
On 07/05/19 9:30 PM, Segher Boessenkool wrote:
On Tue, May 07, 2019 at 07:43:56PM +0530, Mahesh J Salgaonkar wrote:
Can we use SPRN_* #defines which are already present in asm/reg.h instead of
hard coding numbers for switch cases ? You may want to add new #defines
for NIP, MSR and CCR.
But
On 2019/5/1 17:29, Rasmus Villemoes wrote:
> -Original Message-
> From: Rasmus Villemoes
> Sent: 2019年5月1日 17:29
> To: devicet...@vger.kernel.org; Qiang Zhao ; Leo Li
>
> Cc: linuxppc-dev@lists.ozlabs.org; linux-arm-ker...@lists.infradead.org;
> linux-ker...@vger.kernel.org; Rob
On 2019/5/1 17:29, Rasmus Villemoes wrote:
> -Original Message-
> From: Rasmus Villemoes
> Sent: 2019年5月1日 17:29
> To: devicet...@vger.kernel.org; Qiang Zhao ; Leo Li
>
> Cc: linuxppc-dev@lists.ozlabs.org; linux-arm-ker...@lists.infradead.org;
> linux-ker...@vger.kernel.org; Rob
On 2019/5/1 17:29, Rasmus Villemoes wrote:
> -Original Message-
> From: Rasmus Villemoes
> Sent: 2019年5月1日 17:29
> To: devicet...@vger.kernel.org; Qiang Zhao ; Leo Li
>
> Cc: linuxppc-dev@lists.ozlabs.org; linux-arm-ker...@lists.infradead.org;
> linux-ker...@vger.kernel.org; Rob
On 2019/5/1 17:29, Rasmus Villemoes wrote:
> -Original Message-
> From: Rasmus Villemoes
> Sent: 2019年5月1日 17:29
> To: devicet...@vger.kernel.org; Qiang Zhao ; Leo Li
>
> Cc: linuxppc-dev@lists.ozlabs.org; linux-arm-ker...@lists.infradead.org;
> linux-ker...@vger.kernel.org; Rob
On 2019/5/1 17:29, Rasmus Villemoes wrote:
> -Original Message-
> From: Rasmus Villemoes
> Sent: 2019年5月1日 17:29
> To: devicet...@vger.kernel.org; Qiang Zhao ; Leo Li
>
> Cc: linuxppc-dev@lists.ozlabs.org; linux-arm-ker...@lists.infradead.org;
> linux-ker...@vger.kernel.org; Rob Herring
On 8/5/19 10:27 pm, Michael Ellerman wrote:
Andrew Donnellan writes:
The powernv platform is the only one that directly accesses SCOMs. Move the
support code to platforms/powernv, and get rid of the PPC_SCOM Kconfig
option, as SCOM support is always selected when compiling for powernv.
This
On Wed, 2019-05-08 at 09:39 +0200, David Hildenbrand wrote:
> However I haven't really thought it through yet, smalles like that
> could
> as well just be handled by the caller of
> arch_add_memory()/arch_remove_memory() eventually, passing it via
> something like the altmap parameter.
>
> Let's
On Wed, May 08, 2019 at 03:06:30PM +0530, Sachin Sant wrote:
> While running LTP tests(specifically ptrace01) following WARNING is observed
> on POWER8 LPAR running next-20190507 built using 4K page size.
>
> [ 3969.979492] msgrcv04 (433) used greatest stack depth: 9328 bytes left
> [
https://bugzilla.kernel.org/show_bug.cgi?id=203125
--- Comment #4 from Erhard F. (erhar...@mailbox.org) ---
Created attachment 282679
--> https://bugzilla.kernel.org/attachment.cgi?id=282679=edit
bisect.log
Finally, the results of the bisect...
f7354ccac844da7b1af8cc4f09da330fa3e960e4 is the
On Wed, May 8, 2019, at 10:31 PM, Michael Ellerman wrote:
> When implementing the KUAP support on Radix we fixed one case where
> mmu_has_feature() was being called too early in boot via
> __put_user_size().
>
> However since then some new code in linux-next has created a new path
> via which we
I see multiple instances of:
arch/powerpc/kernel/exceptions-64s.S:839: Error:
attempt to move .org backwards
in v4.4.y.queue (v4.4.179-143-gc4db218e9451).
This is due to commit 9b2d4e06d7f1 ("powerpc/64s: Add support for a store
forwarding barrier at kernel entry/exit"), which is part
Hi Pablo,
Le 07/05/2019 à 23:50, Rodriguez Quesada, Pablo a écrit :
Hi, Christophe and everyone from the LinuxPPC community,
Hope you are doing well!
I was working with Daniel, and we were reviewing the ipic source code
and came to these questions:
1.The first one came up while reviewing
On Wed, May 08, 2019 at 07:36:54AM -0700, Richard Cochran wrote:
> No the alarm functionality has been removed. It will not be coming
> back, unless there are really strong arguments to support it.
Here is some more background:
commit 3a06c7ac24f9f24ec059cd77c2dbdf7fbfd0aaaf
Author:
On Wed, 2019-05-08 at 15:18 +0200, Greg KH wrote:
>
>
> On Wed, May 08, 2019 at 04:11:28PM +0300, Andy Shevchenko wrote:
> > On Wed, May 08, 2019 at 02:28:29PM +0300, Alexandru Ardelean wrote:
> > > This change re-introduces `match_string()` as a macro that uses
> > > ARRAY_SIZE() to compute the
On Wed, 2019-05-08 at 15:20 +0300, Dan Carpenter wrote:
>
>
> On Wed, May 08, 2019 at 02:28:35PM +0300, Alexandru Ardelean wrote:
> > -static const char * const phy_types[] = {
> > - "emmc 5.0 phy",
> > - "emmc 5.1 phy"
> > -};
> > -
> > enum xenon_phy_type_enum {
> >
On Wed, 2019-05-08 at 14:28 +0300, Alexandru Ardelean wrote:
> The documentation the `_match_string()` helper mentions that `n`
> should be:
> * @n: number of strings in the array or -1 for NULL terminated arrays
>
> The behavior of the function is different, in the sense that it exits on
> the
The `lcd_types` array is a static array of strings.
Using match_string() (which computes the array size via ARRAY_SIZE())
is possible.
This reduces the array by 1 element, since the NULL (at the end of the
array) is no longer needed.
Signed-off-by: Alexandru Ardelean
---
__match_string() is called on 2 static array of strings in this file. For
this reason, the conversion to the new match_string() macro/helper, was
done in this separate commit.
Using the new match_string() helper is mostly a cosmetic change (at this
point in time). The sizes of the arrays will be
The `sched_feat_names` array is a static array of strings.
Using match_string() (which computes the array size via ARRAY_SIZE())
is possible.
The change is mostly cosmetic.
No functionality change.
Signed-off-by: Alexandru Ardelean
---
kernel/sched/debug.c | 2 +-
1 file changed, 1
The `DRIVER_STRING` array is a static array of strings.
Using match_string() (which computes the array size via ARRAY_SIZE())
is possible.
The change is mostly cosmetic.
No functionality change.
Signed-off-by: Alexandru Ardelean
---
drivers/staging/gdm724x/gdm_tty.c | 3 +--
1 file changed, 1
The `generic_edid_name` is a static array of strings.
Using match_string() (which computes the array size via ARRAY_SIZE())
is possible.
The change is mostly cosmetic.
No functionality change.
Signed-off-by: Alexandru Ardelean
---
drivers/gpu/drm/drm_edid_load.c | 2 +-
1 file changed, 1
The `rdmacg_resource_names` array is a static array of strings.
Using match_string() (which computes the array size via ARRAY_SIZE())
is possible.
The change is mostly cosmetic.
No functionality change.
Signed-off-by: Alexandru Ardelean
---
kernel/cgroup/rdma.c | 2 +-
1 file changed, 1
The change is mostly cosmetic.
The `armada_37xx_pin_group` struct is defined as.
struct armada_37xx_pin_group {
const char *name;
unsigned intstart_pin;
unsigned intnpins;
u32 reg_mask;
u32 val[NB_FUNCS];
The change is also cosmetic, but it also does a tighter coupling between
the enums & the string values. This way, the ARRAY_SIZE(phy_types) that is
implicitly done in the match_string() macro is also a bit safer.
Signed-off-by: Alexandru Ardelean
---
drivers/mmc/host/sdhci-xenon-phy.c | 12
The change is mostly cosmetic.
The `energy_perf_strings` array is static, so match_string() can be used
(which will implicitly do a ARRAY_SIZE(energy_perf_strings)).
The only small benefit here, is the reduction of the array size by 1
element.
Signed-off-by: Alexandru Ardelean
---
The `device_connection` struct is defined as:
struct device_connection {
struct fwnode_handle*fwnode;
const char *endpoint[2];
const char *id;
struct list_headlist;
};
The `endpoint` member is a static array of strings (on the
This change is a bit more than cosmetic.
It replaces 2 values in mtrr_strings with NULL. Previously, they were
defined as "?", which is not great because you could technically pass "?",
and you would get value 2.
It's not sure whether that was intended (likely it wasn't), but this fixes
that.
The change is purely cosmetic at this point in time, but it does highlight
the change done in lib/string.c for match_string().
Particularly for this change, a control mode can be removed/added at a
different index/enum-value, and the match_string() helper will continue
until the end of the array
The change is purely cosmetic at this point in time, but it does highlight
the change done in lib/string.c for match_string().
Particularly for this change, if a regname is removed (replaced with NULL)
in the list, the match_string() helper will continue until the end of the
array and ignore the
This change re-introduces `match_string()` as a macro that uses
ARRAY_SIZE() to compute the size of the array.
The macro is added in all the places that do
`match_string(_a, ARRAY_SIZE(_a), s)`, since the change is pretty
straightforward.
Signed-off-by: Alexandru Ardelean
---
This change does a rename of match_string() -> __match_string().
There are a few parts to the intention here (with this change):
1. Align with sysfs_match_string()/__sysfs_match_string()
2. This helps to group users of `match_string()` into simple users:
a. those that use ARRAY_SIZE(_a) to
The intent of this patch series is to make a case for fixing the
match_string() string helper.
The doc-string of the `__sysfs_match_string()` helper mentions that `n`
(the size of the given array) should be:
* @n: number of strings in the array or -1 for NULL terminated arrays
However, this is
The documentation the `_match_string()` helper mentions that `n`
(size of the given array) should be:
* @n: number of strings in the array or -1 for NULL terminated arrays
The behavior of the function is different, in the sense that it exits on
the first NULL element in the array, regardless of
The documentation the `_match_string()` helper mentions that `n`
should be:
* @n: number of strings in the array or -1 for NULL terminated arrays
The behavior of the function is different, in the sense that it exits on
the first NULL element in the array, regardless of whether `n` is -1 or a
On Tue, May 07, 2019 at 06:53:30PM +0200, Christophe Leroy wrote:
> Le 07/05/2019 à 17:10, Segher Boessenkool a écrit :
> >On Tue, May 07, 2019 at 01:31:39PM +, Christophe Leroy wrote:
> >>Cache instructions (dcbz, dcbi, dcbf and dcbst) take two registers
> >>that are summed to obtain the
On Wed, May 08, 2019 at 03:30:01AM +, Po Liu wrote:
> > Sorry, NAK, since we decided some time ago not to support timer_* operations
> > on dynamic clocks. You get much better application level timer performance
> > by synchronizing CLOCK_REALTIME to your PHC and using clock_nanosleep()
> >
On Wed, May 8, 2019 at 12:22 AM David Hildenbrand wrote:
>
>
> >> drivers/base/node.c | 18 +-
> >> include/linux/node.h | 5 ++---
> >> 2 files changed, 7 insertions(+), 16 deletions(-)
> >>
> >> diff --git a/drivers/base/node.c b/drivers/base/node.c
> >> index
On Wed, May 08, 2019 at 04:11:28PM +0300, Andy Shevchenko wrote:
> On Wed, May 08, 2019 at 02:28:29PM +0300, Alexandru Ardelean wrote:
> > This change re-introduces `match_string()` as a macro that uses
> > ARRAY_SIZE() to compute the size of the array.
> > The macro is added in all the places
On Wed, May 08, 2019 at 02:28:29PM +0300, Alexandru Ardelean wrote:
> This change re-introduces `match_string()` as a macro that uses
> ARRAY_SIZE() to compute the size of the array.
> The macro is added in all the places that do
> `match_string(_a, ARRAY_SIZE(_a), s)`, since the change is pretty
When implementing the KUAP support on Radix we fixed one case where
mmu_has_feature() was being called too early in boot via
__put_user_size().
However since then some new code in linux-next has created a new path
via which we can end up calling mmu_has_feature() too early.
On P9 this leads to
Andrew Donnellan writes:
> The powernv platform is the only one that directly accesses SCOMs. Move the
> support code to platforms/powernv, and get rid of the PPC_SCOM Kconfig
> option, as SCOM support is always selected when compiling for powernv.
This code was used on WSP / A2, back in the
On Wed, May 08, 2019 at 02:28:35PM +0300, Alexandru Ardelean wrote:
> -static const char * const phy_types[] = {
> - "emmc 5.0 phy",
> - "emmc 5.1 phy"
> -};
> -
> enum xenon_phy_type_enum {
> EMMC_5_0_PHY,
> EMMC_5_1_PHY,
> NR_PHY_TYPES
There is no need for
While running LTP tests (specifically futex_wake04) against next-20199597
build with 4K page size on a POWER8 LPAR following crash is observed.
[ 4233.214876] BUG: Kernel NULL pointer dereference at 0x001c
[ 4233.214898] Faulting instruction address: 0xc1d1e58c
[ 4233.214905] Oops:
On Mon, May 06, 2019 at 08:50:45AM +0200, Johannes Thumshirn wrote:
> Acked-by: Johannes Thumshirn
Queued, thanks.
--
Regards/Gruss,
Boris.
Good mailing practices for 400: avoid top-posting and trim the reply.
On 18/04/2019 14:57, Zhen Lei wrote:
First, add build option IOMMU_DEFAULT_{LAZY|STRICT}, so that we have the
opportunity to set {lazy|strict} mode as default at build time. Then put
the three config options in an choice, make people can only choose one of
the three at a time.
The default IOMMU
While running LTP tests(specifically ptrace01) following WARNING is observed
on POWER8 LPAR running next-20190507 built using 4K page size.
[ 3969.979492] msgrcv04 (433) used greatest stack depth: 9328 bytes left
[ 3981.452911] madvise06 (515): drop_caches: 3
[ 4004.575752] WARNING: CPU: 5 PID:
On 07.05.19 20:38, David Hildenbrand wrote:
> Only memory to be added to the buddy and to be onlined/offlined by
> user space using memory block devices needs (and should have!) memory
> block devices.
>
> Factor out creation of memory block devices Create all devices after
> arch_add_memory()
On 07.05.19 23:25, Dan Williams wrote:
> On Tue, May 7, 2019 at 2:24 PM David Hildenbrand wrote:
>>
>> On 07.05.19 23:19, Dan Williams wrote:
>>> On Tue, May 7, 2019 at 11:38 AM David Hildenbrand wrote:
No longer needed, the callers of arch_add_memory() can handle this
manually.
>> drivers/base/node.c | 18 +-
>> include/linux/node.h | 5 ++---
>> 2 files changed, 7 insertions(+), 16 deletions(-)
>>
>> diff --git a/drivers/base/node.c b/drivers/base/node.c
>> index 04fdfa99b8bc..9be88fd05147 100644
>> --- a/drivers/base/node.c
>> +++
(Resent to include relevant mailing lists - sorry about that!)
Hello all,
This patch set implements support for user-specified PCI resource
alignment on the pseries platform for hotplugged PCI devices.
Currently on pseries, PCI resource alignments specified with the
pci=resource_alignment
Enable the pcibios_after_init hook on all powerpc platforms.
This hook is executed at the end of pcibios_init and was previously
only available on CONFIG_PPC32.
Since it is useful and not inherently limited to 32-bit mode,
remove the limitation and allow it on all powerpc platforms.
On Wed, May 01, 2019 at 10:59:00PM -0500, Gustavo A. R. Silva wrote:
> Mark,
>
> I wonder if you are going to take this patch.
Please don't send content free pings and please allow a reasonable time
for review. People get busy, go on holiday, attend conferences and so
on so unless there is
The powernv platform is the only one that directly accesses SCOMs. Move the
support code to platforms/powernv, and get rid of the PPC_SCOM Kconfig
option, as SCOM support is always selected when compiling for powernv.
This also means that the Kconfig item for CONFIG_SCOM_DEBUGFS will actually
Replace __get_free_page() and alloc_pages() calls with the generic
__pte_alloc_one_kernel() and __pte_alloc_one().
There is no functional change for the kernel PTE allocation.
The difference for the user PTEs, is that the clear_pte_table() is now
called after pgtable_page_ctor() and the addition
um allocates PTE pages with __get_free_page() and uses
GFP_KERNEL | __GFP_ZERO for the allocations.
Switch it to the generic version that does exactly the same thing for the
kernel page tables and adds __GFP_ACCOUNT for the user PTEs.
The pte_free() and pte_free_kernel() versions are identical
The only difference between the generic and RISC-V implementation of PTE
allocation is the usage of __GFP_RETRY_MAYFAIL for both kernel and user
PTEs and the absence of __GFP_ACCOUNT for the user PTEs.
The conversion to the generic version removes the __GFP_RETRY_MAYFAIL and
ensures that
parisc allocates PTE pages with __get_free_page() and uses
GFP_KERNEL | __GFP_ZERO for the allocations.
Switch it to the generic version that does exactly the same thing for the
kernel page tables and adds __GFP_ACCOUNT for the user PTEs.
The pte_free_kernel() and pte_free() versions on are
nios2 allocates kernel PTE pages with
__get_free_pages(GFP_KERNEL | __GFP_ZERO, PTE_ORDER);
and user page tables with
pte = alloc_pages(GFP_KERNEL, PTE_ORDER);
if (pte)
clear_highpage();
The PTE_ORDER is hardwired to zero, which makes nios2
The nds32 implementation of pte_alloc_one_kernel() differs from the generic
in the use of __GFP_RETRY_MAYFAIL flag, which is removed after the
conversion.
The nds32 version of pte_alloc_one() missed the call to pgtable_page_ctor()
and also used __GFP_RETRY_MAYFAIL. Switching it to use generic
MIPS allocates kernel PTE pages with
__get_free_pages(GFP_KERNEL | __GFP_ZERO, PTE_ORDER)
and user PTE pages with
pte = alloc_pages(GFP_KERNEL, PTE_ORDER)
and then uses clear_highpage(pte) to zero out the allocated page for the
user page tables.
The PTE_ORDER is hardwired to
The sun3 MMU variant of m68k uses GFP_KERNEL to allocate a PTE page and
then memset(0) or clear_highpage() to clear it.
This is equivalent to allocating the page with GFP_KERNEL | __GFP_ZERO,
which allows replacing sun3 implementation of pte_alloc_one() and
pte_alloc_one_kernel() with the generic
The hexagon implementation pte_alloc_one(), pte_alloc_one_kernel(),
pte_free_kernel() and pte_free() is identical to the generic except of
lack of __GFP_ACCOUNT for the user PTEs allocation.
Switch hexagon to use generic version of these functions.
Signed-off-by: Mike Rapoport
---
The csky implementation pte_alloc_one(), pte_free_kernel() and pte_free()
is identical to the generic except of lack of __GFP_ACCOUNT for the user
PTEs allocation.
Switch csky to use generic version of these functions.
The csky implementation of pte_alloc_one_kernel() is not replaced because
it
The PTE allocations in arm64 are identical to the generic ones modulo the
GFP flags.
Using the generic pte_alloc_one() functions ensures that the user page
tables are allocated with __GFP_ACCOUNT set.
The arm64 definition of PGALLOC_GFP is removed and replaced with
GFP_PGTABLE_USER for
Replace __get_free_page() and alloc_pages() calls with the generic
__pte_alloc_one_kernel() and __pte_alloc_one().
There is no functional change for the kernel PTE allocation.
The difference for the user PTEs, is that the clear_pte_table() is now
called after pgtable_page_ctor() and the addition
alpha allocates PTE pages with __get_free_page() and uses
GFP_KERNEL | __GFP_ZERO for the allocations.
Switch it to the generic version that does exactly the same thing for the
kernel page tables and adds __GFP_ACCOUNT for the user PTEs.
The alpha pte_free() and pte_free_kernel() versions are
Most architectures have identical or very similar implementation of
pte_alloc_one_kernel(), pte_alloc_one(), pte_free_kernel() and pte_free().
Add a generic implementation that can be reused across architectures and
enable its use on x86.
The generic implementation uses
GFP_KERNEL |
Hi,
Many architectures have similar, if not identical implementation of
pte_alloc_one_kernel(), pte_alloc_one(), pte_free_kernel() and pte_free().
A while ago Anshuman suggested to introduce a common definition of
GFP_PGTABLE and during the discussion it was suggested to rather
consolidate the
80 matches
Mail list logo
