Le 10/06/2022 à 17:55, Hari Bathini a écrit :
> This adds two atomic opcodes BPF_XCHG and BPF_CMPXCHG on ppc32, both
> of which include the BPF_FETCH flag. The kernel's atomic_cmpxchg
> operation fundamentally has 3 operands, but we only have two register
> fields. Therefore the operand we
Le 10/06/2022 à 17:55, Hari Bathini a écrit :
> Adding instructions for ppc32 for
>
> atomic_and
> atomic_or
> atomic_xor
> atomic_fetch_add
> atomic_fetch_and
> atomic_fetch_or
> atomic_fetch_xor
>
> Signed-off-by: Hari Bathini
> ---
>
> Changes in v2:
> * Used an additional register
This is the end of the work started with commit 76222808fc25 ("powerpc:
Move C prototypes out of asm-prototypes.h")
Now that asm/machdep.h doesn't include asm/setup.h anymore, there are
no conflicts anymore with the function prom_init() defined in
asm/machdep.h doesn't need asm/setup.h
Remove it.
Add it directly in files that needs it.
Signed-off-by: Christophe Leroy
---
arch/powerpc/include/asm/machdep.h | 2 --
arch/powerpc/kernel/pci-common.c | 1 +
arch/powerpc/kernel/prom.c | 2 +-
powerpc's asm/prom.h brings some headers that it doesn't need itself.
Once those headers are removed from asm/prom.h, the following
errors occur:
CC [M] drivers/scsi/cxlflash/ocxl_hw.o
drivers/scsi/cxlflash/ocxl_hw.c: In function 'afu_map_irq':
drivers/scsi/cxlflash/ocxl_hw.c:195:16: error:
A lot of drivers were getting platform and of headers
indirectly via headers like asm/pci.h or asm/prom.h
Most of them were fixed during 5.19 cycle but a newissue was
introduced by commit 52b1b46c39ae ("of: Create platform devices
for OF framebuffers")
Include missing platform_device.h to allow
Remove all headers included from asm/prom.h which are not used
by asm/prom.h itself.
Declare struct device_node and struct property locally to
avoid including of.h
Signed-off-by: Christophe Leroy
---
arch/powerpc/include/asm/prom.h | 12 +++-
1 file changed, 3 insertions(+), 9
asm/pci.h and asm/mpc52xx.h don't need asm/prom.h
Declare struct device_node locally to avoid including of.h
Signed-off-by: Christophe Leroy
---
arch/powerpc/include/asm/mpc52xx.h | 3 ++-
arch/powerpc/include/asm/pci.h | 1 -
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git
Le 11/06/2022 à 17:10, Jason A. Donenfeld a écrit :
> The platform's RNG must be available before random_init() in order to be
> useful for initial seeding, which in turn means that it needs to be
> called from setup_arch(), rather than from an init call. Fortunately,
> each platform already has
Le 11/06/2022 à 17:10, Jason A. Donenfeld a écrit :
> The platform's RNG must be available before random_init() in order to be
> useful for initial seeding, which in turn means that it needs to be
> called from setup_arch(), rather than from an init call. Fortunately,
> each platform already has
Le 11/06/2022 à 17:10, Jason A. Donenfeld a écrit :
> The platform's RNG must be available before random_init() in order to be
> useful for initial seeding, which in turn means that it needs to be
> called from setup_arch(), rather than from an init call. Fortunately,
> each platform already has
The platform's RNG must be available before random_init() in order to be
useful for initial seeding, which in turn means that it needs to be
called from setup_arch(), rather than from an init call. Fortunately,
each platform already has a setup_arch function pointer, which means
it's easy to wire
The platform's RNG must be available before random_init() in order to be
useful for initial seeding, which in turn means that it needs to be
called from setup_arch(), rather than from an init call. Fortunately,
each platform already has a setup_arch function pointer, which means
it's easy to wire
The platform's RNG must be available before random_init() in order to be
useful for initial seeding, which in turn means that it needs to be
called from setup_arch(), rather than from an init call. Fortunately,
each platform already has a setup_arch function pointer, which means
it's easy to wire
The platform's RNG must be available before random_init() in order to be
useful for initial seeding, which in turn means that it needs to be
called from setup_arch(), rather than from an init call. This series
wires that up properly on the three platforms that currently initialize
the RNG from the
Hi Christophe,
On Sat, Jun 11, 2022 at 4:49 PM Christophe Leroy
wrote:
> Then you have:
>
> arch/powerpc/platforms/powernv/powernv.h
> arch/powerpc/platforms/pseries/pseries.h
>
> and you can add
>
> arch/powerpc/platforms/microwatt/microwatt.h
Oh, terrific, thanks. I'll do that.
Jason
Hi Christophe,
On Sat, Jun 11, 2022 at 4:45 PM Christophe Leroy
wrote:
>
> There must be a empty line between declarations and code.
Ack.
> Prototype has to go in a header file
Already voiced disagreement about this in the other thread.
> and should be pSeries maybe
> allthough camelCase in
Le 11/06/2022 à 16:46, Jason A. Donenfeld a écrit :
> Hi again,
>
> On Sat, Jun 11, 2022 at 04:41:58PM +0200, Jason A. Donenfeld wrote:
>> Hi Christophe,
>>
>> On Sat, Jun 11, 2022 at 4:40 PM Christophe Leroy
>> wrote:
+__init void microwatt_rng_init(void);
>>>
>>> This prototype
Hi Christophe,
On Sat, Jun 11, 2022 at 4:42 PM Christophe Leroy
wrote:
> Same here, the prototype should go in a header file., should be 'void
> __init' (and indeed the __init is pointless in the prototype, only
> matters in the function definition).
I'll change the order, but I don't see a
Hi again,
On Sat, Jun 11, 2022 at 04:41:58PM +0200, Jason A. Donenfeld wrote:
> Hi Christophe,
>
> On Sat, Jun 11, 2022 at 4:40 PM Christophe Leroy
> wrote:
> > >
> > > +__init void microwatt_rng_init(void);
> >
> > This prototype should be declared in a header file, for instance asm/setup.h
>
Le 11/06/2022 à 12:04, Jason A. Donenfeld a écrit :
> The platform's RNG must be available before random_init() in order to be
> useful for initial seeding, which in turn means that it needs to be
> called from setup_arch(), rather than from an init call. Fortunately,
> each platform already has
Le 11/06/2022 à 12:04, Jason A. Donenfeld a écrit :
> The platform's RNG must be available before random_init() in order to be
> useful for initial seeding, which in turn means that it needs to be
> called from setup_arch(), rather than from an init call. Fortunately,
> each platform already has
Hi Christophe,
On Sat, Jun 11, 2022 at 4:40 PM Christophe Leroy
wrote:
> >
> > +__init void microwatt_rng_init(void);
>
> This prototype should be declared in a header file, for instance asm/setup.h
Alright.
> And I think the __init keyword usually goes after the type, so should be
> 'void
Le 11/06/2022 à 12:04, Jason A. Donenfeld a écrit :
> The platform's RNG must be available before random_init() in order to be
> useful for initial seeding, which in turn means that it needs to be
> called from setup_arch(), rather than from an init call. Fortunately,
> each platform already has
commit 8e127846fc97778a5e5c99bca1ce0bbc5ec9 upstream.
The ptrace PEEKUSR/POKEUSR (aka PEEKUSER/POKEUSER) API allows a process
to read/write registers of another process.
To get/set a register, the API takes an index into an imaginary address
space called the "USER area", where the registers
Le 11/06/2022 à 11:58, Jason A. Donenfeld a écrit :
> Hi Christophe,
>
> On Sat, Jun 11, 2022 at 09:27:43AM +, Christophe Leroy wrote:
>> Le 11/06/2022 à 11:22, Jason A. Donenfeld a écrit :
>>> Hi Christophe,
>>>
>>> On Sat, Jun 11, 2022 at 11:17:23AM +0200, Christophe Leroy wrote:
Hey again,
On Sat, Jun 11, 2022 at 11:58:23AM +0200, Jason A. Donenfeld wrote:
> Anyway, sure, I'll do that and send a v2 series.
This is now done here:
https://lore.kernel.org/lkml/20220611100447.5066-1-ja...@zx2c4.com/
Jason
The platform's RNG must be available before random_init() in order to be
useful for initial seeding, which in turn means that it needs to be
called from setup_arch(), rather than from an init call. Fortunately,
each platform already has a setup_arch function pointer, which means
it's easy to wire
The platform's RNG must be available before random_init() in order to be
useful for initial seeding, which in turn means that it needs to be
called from setup_arch(), rather than from an init call. Fortunately,
each platform already has a setup_arch function pointer, which means
it's easy to wire
The platform's RNG must be available before random_init() in order to be
useful for initial seeding, which in turn means that it needs to be
called from setup_arch(), rather than from an init call. Fortunately,
each platform already has a setup_arch function pointer, which means
it's easy to wire
The platform's RNG must be available before random_init() in order to be
useful for initial seeding, which in turn means that it needs to be
called from setup_arch(), rather than from an init call. This series
wires that up properly on the three platforms that currently initialize
the RNG from the
Hi Christophe,
On Sat, Jun 11, 2022 at 09:27:43AM +, Christophe Leroy wrote:
> Le 11/06/2022 à 11:22, Jason A. Donenfeld a écrit :
> > Hi Christophe,
> >
> > On Sat, Jun 11, 2022 at 11:17:23AM +0200, Christophe Leroy wrote:
> >> Also, you copied stable. Should you add a Fixes: tag so that we
Le 11/06/2022 à 11:22, Jason A. Donenfeld a écrit :
> Hi Christophe,
>
> On Sat, Jun 11, 2022 at 11:17:23AM +0200, Christophe Leroy wrote:
>> Also, you copied stable. Should you add a Fixes: tag so that we know
>> what it fixes ?
>
> I suppose the fixes tag would be whatever introduced those
Le 11/06/2022 à 11:20, Jason A. Donenfeld a écrit :
> Hi Christophe,
>
> On Sat, Jun 11, 2022 at 09:16:24AM +, Christophe Leroy wrote:
>> Le 11/06/2022 à 10:11, Jason A. Donenfeld a écrit :
>>> The platform's RNG must be available before random_init() in order to be
>>> useful for initial
Hi Christophe,
On Sat, Jun 11, 2022 at 11:17:23AM +0200, Christophe Leroy wrote:
> Also, you copied stable. Should you add a Fixes: tag so that we know
> what it fixes ?
I suppose the fixes tag would be whatever introduced those files in the
first place, so not all together useful. But if you
Hi Christophe,
On Sat, Jun 11, 2022 at 09:16:24AM +, Christophe Leroy wrote:
> Le 11/06/2022 à 10:11, Jason A. Donenfeld a écrit :
> > The platform's RNG must be available before random_init() in order to be
> > useful for initial seeding, which in turn means that it needs to be
> > called
Le 11/06/2022 à 11:16, Christophe Leroy a écrit :
Le 11/06/2022 à 10:11, Jason A. Donenfeld a écrit :
The platform's RNG must be available before random_init() in order to be
useful for initial seeding, which in turn means that it needs to be
called from setup_arch(), rather than from an
Le 11/06/2022 à 10:11, Jason A. Donenfeld a écrit :
> The platform's RNG must be available before random_init() in order to be
> useful for initial seeding, which in turn means that it needs to be
> called from setup_arch(), rather than from an init call. Fortunately,
> each platform already has
Le 09/06/2022 à 11:52, Ariel Miculas a écrit :
> This fixes the gdbserver issue on PPC32 described here:
> Link:
> https://linuxppc-dev.ozlabs.narkive.com/C46DRek4/debug-problems-on-ppc-83xx-target-due-to-changed-struct-task-struct
>
> On PPC32, the user space code considers the floating point
Le 10/06/2022 à 05:32, Rohan McLure a écrit :
>> On 2 Jun 2022, at 2:00 am, Segher Boessenkool
>> wrote:
>>
>> Hi!
>>
>> On Wed, Jun 01, 2022 at 03:48:45PM +1000, Rohan McLure wrote:
>>> +.macro BINOP_REGS op, rhs, start, end
>>> + .Lreg=\start
>>> + .rept (\end - \start + 1)
>>> + \op
The platform's RNG must be available before random_init() in order to be
useful for initial seeding, which in turn means that it needs to be
called from setup_arch(), rather than from an init call. Fortunately,
each platform already has a setup_arch function pointer, which means
it's easy to wire
Commit f9b3cd245784 ("Kconfig.debug: make DEBUG_INFO selectable from a
choice") broke the selection of CONFIG_DEBUG_INFO by powerpc defconfigs.
It is now necessary to select one of the three DEBUG_INFO_DWARF*
options to get DEBUG_INFO enabled.
Replace DEBUG_INFO=y by
42 matches
Mail list logo
