* Re: [PATCH v5 06/27] irq_domain/powerpc: eliminate irq_map; use irq_alloc_desc() instead
From: Grant Likely @ 2012-04-11 20:57 UTC (permalink / raw)
To: Benjamin Herrenschmidt, Andreas Schwab
Cc: Russell King - ARM Linux, devicetree-discuss, linux-kernel,
Milton Miller, Rob Herring, Thomas Gleixner, linuxppc-dev,
linux-arm-kernel
In-Reply-To: <1334122182.2984.33.camel@pasglop>
On Wed, 11 Apr 2012 15:29:42 +1000, Benjamin Herrenschmidt <benh@kernel.crashing.org> wrote:
> On Wed, 2012-04-11 at 11:33 +1000, Benjamin Herrenschmidt wrote:
> > On Wed, 2012-04-11 at 11:13 +1000, Benjamin Herrenschmidt wrote:
> > > On Sat, 2012-04-07 at 14:27 +0200, Andreas Schwab wrote:
> > > > Benjamin Herrenschmidt <benh@kernel.crashing.org> writes:
> > > >
> > > > > It's arguable that this irq_set_irq_type(,NONE) shouln't be there but
> > > > > still ... it's been around for ever and things worked :-) So something
> > > > > -else- is causing the problem and I'd like to understand what exactly.
> > > >
> > > > AFAICS before a09b659cd68c10ec6a30cb91ebd2c327fcd5bfe5
> > > > irq_set_irq_type(,NONE) was actually a no-op.
> > >
> > > So I'm still a bit baffled... ie, I understand some of what's happening
> > > but not why it breaks things, I haven't yet managed to reproduce but I
> > > haven't tried too hard just yet (was away from the HW) :
> >
> > Allright, I have a repro-case, I'll dig.
>
> Ok, so it's Grant's fault :-)
I pretty much expected it would be. :-p
> So basically, it's quite subtle and I'm only 99% sure of the details but
> I believe what happens is:
>
> - The audio interrupts get virq 64 and 65 (so above NR_IRQS)
>
> - The reverse map isn't pre-filled at map time (we should probably do
> it nowadays), we do it lazily so ...
Hmmm... I though I had merged a patch that does that.
/me goes to look again...
Okay, I did write that patch, but I never merged it because it didn't
get much review and I was already nervous about the other irq_domain
changes. I'll post it again and ask for feedback.
> - The whole business with irq_virq_count needs fixing. Basically the
> default value shouldn't be NR_IRQ. I suggest making it 0 and have all
> the use sites do something like:
>
> max = irq_virq_count ? irq_virq_count : nr_irqs;
>
> (Grant, can you take care of that ?)
Yeah, I've got a different way to fix it though. There is exactly one
user of irq_virq_count in-tree right now: PS3. Also, irq_virq_count
is only useful for the NOMAP mapping. So, instead of having a single
global irq_virq_count values, I've dropped it entirely and added a
max_irq argument to irq_domain_add_nomap(). That makes it a property
of an individual nomap irq domain instead of a global system settting.
Hopefully I'll have a draft patch ready today.
g.
^ permalink raw reply
* [PATCH] powerpc/e6500: add CPU_FTR_EMB_HV to CPU table
From: Scott Wood @ 2012-04-11 20:27 UTC (permalink / raw)
To: Marcelo Tosatti; +Cc: linuxppc-dev, kvm-ppc, kvm
e6500 support (commit 10241842fbe900276634fee8d37ec48a7d8a762f,
"powerpc: Add initial e6500 cpu support" and the introduction of
CPU_FTR_EMB_HV (commit 73196cd364a2d972d73fa08da9d81ca3215bed68,
"KVM: PPC: e500mc support") collided during merge, leaving e6500's CPU
table entry missing CPU_FTR_EMB_HV.
Signed-off-by: Scott Wood <scottwood@freescale.com>
---
Fixup patch for the KVM merge as requested by Marcelo.
arch/powerpc/include/asm/cputable.h | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/arch/powerpc/include/asm/cputable.h b/arch/powerpc/include/asm/cputable.h
index 67c34af..50d82c8 100644
--- a/arch/powerpc/include/asm/cputable.h
+++ b/arch/powerpc/include/asm/cputable.h
@@ -395,7 +395,7 @@ extern const char *powerpc_base_platform;
#define CPU_FTRS_E6500 (CPU_FTR_USE_TB | CPU_FTR_NODSISRALIGN | \
CPU_FTR_L2CSR | CPU_FTR_LWSYNC | CPU_FTR_NOEXECUTE | \
CPU_FTR_DBELL | CPU_FTR_POPCNTB | CPU_FTR_POPCNTD | \
- CPU_FTR_DEBUG_LVL_EXC)
+ CPU_FTR_DEBUG_LVL_EXC | CPU_FTR_EMB_HV)
#define CPU_FTRS_GENERIC_32 (CPU_FTR_COMMON | CPU_FTR_NODSISRALIGN)
/* 64-bit CPUs */
--
1.7.7.rc3.4.g8d714
^ permalink raw reply related
* Re: Network problem with 3.3
From: Gary Thomas @ 2012-04-11 17:06 UTC (permalink / raw)
To: Joakim Tjernlund; +Cc: linuxppc-dev
In-Reply-To: <OFBD338563.A7871AC2-ONC12579DD.005CD50B-C12579DD.005D2BA2@transmode.se>
On 2012-04-11 10:57, Joakim Tjernlund wrote:
>>
>> On 2012-04-11 10:14, Gary Thomas wrote:
>>>
>>> n.b. sorry if you receive this twice - ThunderBird seems very crashy lately!
>>>
>>> I'm trying to update my MPC8379 board from 3.0 to 3.3. I have
>>> this setup in my device tree:
>>>
>>> mdio@24520 {
>>> #address-cells =<1>;
>>> #size-cells =<0>;
>>> compatible = "fsl,gianfar-mdio";
>>> reg =<0x24520 0x20>;
>>>
>>> phy0: ethernet-phy@0 {
>>> interrupt-parent =<&ipic>;
>>> interrupts =<17 0x8>;
>>> reg =<0x01>;
>>> device_type = "ethernet-phy";
>>> };
>>> phy1: ethernet-phy@1 {
>>> interrupt-parent =<&ipic>;
>>> interrupts =<18 0x8>;
>>> reg =<0x00>;
>>> device_type = "ethernet-phy";
>>> };
>>> };
>>>
>>> enet0: ethernet@24000 {
>>> cell-index =<0>;
>>> device_type = "network";
>>> model = "TSEC";
>>> compatible = "gianfar";
>>> reg =<0x24000 0x1000>;
>>> local-mac-address = [ 00 08 e5 11 32 33 ];
>>> interrupts =<32 0x8 33 0x8 34 0x8>;
>>> interrupt-parent =<&ipic>;
>>> phy-handle =<&phy0>;
>>> linux,network-index =<0>;
>>> };
>>>
>>> enet1: ethernet@25000 {
>>> cell-index =<1>;
>>> device_type = "network";
>>> model = "TSEC";
>>> compatible = "gianfar";
>>> reg =<0x25000 0x1000>;
>>> local-mac-address = [ 00 08 e5 11 32 34 ];
>>> interrupts =<35 0x8 36 0x8 37 0x8>;
>>> interrupt-parent =<&ipic>;
>>> phy-handle =<&phy1>;
>>> linux,network-index =<1>;
>>> };
>>>
>>> On 3.0, this works just fine. When I boot the same setup on 3.3,
>>> I get this error:
>>> fsl-pq_mdio: probe of ff024520.mdio failed with error -16
>>>
>>> Is there something I need to change in my DTS file to make this work?
>>> Something else I might be missing?
>>
>> I think I found it - I need to add a TBI node to my PHY setup.
>> It's working now.
>
> Do you have a TBI PHY? There is a bug in 3.3 where every non-TBI PHY would fail, see
> http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=464b57da56910c8737ede75ad820b9a7afc46b3e
> for a fix.
Thanks. After applying this patch, networking is working fine on
my board, with no TBI node in the PHY.
--
------------------------------------------------------------
Gary Thomas | Consulting for the
MLB Associates | Embedded world
------------------------------------------------------------
^ permalink raw reply
* Re: Network problem with 3.3
From: Joakim Tjernlund @ 2012-04-11 16:57 UTC (permalink / raw)
To: Gary Thomas; +Cc: linuxppc-dev
In-Reply-To: <4F85AFB0.9020909@mlbassoc.com>
>
> On 2012-04-11 10:14, Gary Thomas wrote:
> >
> > n.b. sorry if you receive this twice - ThunderBird seems very crashy lately!
> >
> > I'm trying to update my MPC8379 board from 3.0 to 3.3. I have
> > this setup in my device tree:
> >
> > mdio@24520 {
> > #address-cells = <1>;
> > #size-cells = <0>;
> > compatible = "fsl,gianfar-mdio";
> > reg = <0x24520 0x20>;
> >
> > phy0: ethernet-phy@0 {
> > interrupt-parent = <&ipic>;
> > interrupts = <17 0x8>;
> > reg = <0x01>;
> > device_type = "ethernet-phy";
> > };
> > phy1: ethernet-phy@1 {
> > interrupt-parent = <&ipic>;
> > interrupts = <18 0x8>;
> > reg = <0x00>;
> > device_type = "ethernet-phy";
> > };
> > };
> >
> > enet0: ethernet@24000 {
> > cell-index = <0>;
> > device_type = "network";
> > model = "TSEC";
> > compatible = "gianfar";
> > reg = <0x24000 0x1000>;
> > local-mac-address = [ 00 08 e5 11 32 33 ];
> > interrupts = <32 0x8 33 0x8 34 0x8>;
> > interrupt-parent = <&ipic>;
> > phy-handle = <&phy0>;
> > linux,network-index = <0>;
> > };
> >
> > enet1: ethernet@25000 {
> > cell-index = <1>;
> > device_type = "network";
> > model = "TSEC";
> > compatible = "gianfar";
> > reg = <0x25000 0x1000>;
> > local-mac-address = [ 00 08 e5 11 32 34 ];
> > interrupts = <35 0x8 36 0x8 37 0x8>;
> > interrupt-parent = <&ipic>;
> > phy-handle = <&phy1>;
> > linux,network-index = <1>;
> > };
> >
> > On 3.0, this works just fine. When I boot the same setup on 3.3,
> > I get this error:
> > fsl-pq_mdio: probe of ff024520.mdio failed with error -16
> >
> > Is there something I need to change in my DTS file to make this work?
> > Something else I might be missing?
>
> I think I found it - I need to add a TBI node to my PHY setup.
> It's working now.
Do you have a TBI PHY? There is a bug in 3.3 where every non-TBI PHY would fail, see
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=464b57da56910c8737ede75ad820b9a7afc46b3e
for a fix.
Jocke
^ permalink raw reply
* Re: Network problem with 3.3
From: Gary Thomas @ 2012-04-11 16:22 UTC (permalink / raw)
To: linuxppc-dev
In-Reply-To: <4F85ADE7.5020800@mlbassoc.com>
On 2012-04-11 10:14, Gary Thomas wrote:
>
> n.b. sorry if you receive this twice - ThunderBird seems very crashy lately!
>
> I'm trying to update my MPC8379 board from 3.0 to 3.3. I have
> this setup in my device tree:
>
> mdio@24520 {
> #address-cells = <1>;
> #size-cells = <0>;
> compatible = "fsl,gianfar-mdio";
> reg = <0x24520 0x20>;
>
> phy0: ethernet-phy@0 {
> interrupt-parent = <&ipic>;
> interrupts = <17 0x8>;
> reg = <0x01>;
> device_type = "ethernet-phy";
> };
> phy1: ethernet-phy@1 {
> interrupt-parent = <&ipic>;
> interrupts = <18 0x8>;
> reg = <0x00>;
> device_type = "ethernet-phy";
> };
> };
>
> enet0: ethernet@24000 {
> cell-index = <0>;
> device_type = "network";
> model = "TSEC";
> compatible = "gianfar";
> reg = <0x24000 0x1000>;
> local-mac-address = [ 00 08 e5 11 32 33 ];
> interrupts = <32 0x8 33 0x8 34 0x8>;
> interrupt-parent = <&ipic>;
> phy-handle = <&phy0>;
> linux,network-index = <0>;
> };
>
> enet1: ethernet@25000 {
> cell-index = <1>;
> device_type = "network";
> model = "TSEC";
> compatible = "gianfar";
> reg = <0x25000 0x1000>;
> local-mac-address = [ 00 08 e5 11 32 34 ];
> interrupts = <35 0x8 36 0x8 37 0x8>;
> interrupt-parent = <&ipic>;
> phy-handle = <&phy1>;
> linux,network-index = <1>;
> };
>
> On 3.0, this works just fine. When I boot the same setup on 3.3,
> I get this error:
> fsl-pq_mdio: probe of ff024520.mdio failed with error -16
>
> Is there something I need to change in my DTS file to make this work?
> Something else I might be missing?
I think I found it - I need to add a TBI node to my PHY setup.
It's working now.
--
------------------------------------------------------------
Gary Thomas | Consulting for the
MLB Associates | Embedded world
------------------------------------------------------------
^ permalink raw reply
* Network problem with 3.3
From: Gary Thomas @ 2012-04-11 16:09 UTC (permalink / raw)
To: linuxppc-dev
I'm trying to port my MPC8379 board from 3.0 to 3.3. In 3.0, I have
this setup in my device tree:
mdio@24520 {
#address-cells = <1>;
#size-cells = <0>;
compatible = "fsl,gianfar-mdio";
reg = <0x24520 0x20>;
phy0: ethernet-phy@0 {
interrupt-parent = <&ipic>;
interrupts = <17 0x8>;
reg = <0x01>;
device_type = "ethernet-phy";
};
phy1: ethernet-phy@1 {
interrupt-parent = <&ipic>;
interrupts = <18 0x8>;
reg = <0x00>;
device_type = "ethernet-phy";
};
};
enet0: ethernet@24000 {
cell-index = <0>;
device_type = "network";
model = "TSEC";
compatible = "gianfar";
reg = <0x24000 0x1000>;
local-mac-address = [ 00 08 e5 11 32 33 ];
interrupts = <32 0x8 33 0x8 34 0x8>;
interrupt-parent = <&ipic>;
phy-handle = <&phy0>;
linux,network-index = <0>;
};
enet1: ethernet@25000 {
cell-index = <1>;
device_type = "network";
model = "TSEC";
compatible = "gianfar";
reg = <0x25000 0x1000>;
local-mac-address = [ 00 08 e5 11 32 34 ];
interrupts = <35 0x8 36 0x8 37 0x8>;
interrupt-parent = <&ipic>;
phy-handle = <&phy1>;
linux,network-index = <1>;
};
On 3.0, this works just fine. When I boot the same setup on 3.3,
I get this error:
fsl-pq_mdio: probe of ff024520.mdio failed with error -16
Is there something I need to change in my DTS file to make this work?
Something else I might be missing?
Thanks
--
------------------------------------------------------------
Gary Thomas | Consulting for the
MLB Associates | Embedded world
------------------------------------------------------------
^ permalink raw reply
* Network problem with 3.3
From: Gary Thomas @ 2012-04-11 16:14 UTC (permalink / raw)
To: linuxppc-dev
n.b. sorry if you receive this twice - ThunderBird seems very crashy lately!
I'm trying to update my MPC8379 board from 3.0 to 3.3. I have
this setup in my device tree:
mdio@24520 {
#address-cells = <1>;
#size-cells = <0>;
compatible = "fsl,gianfar-mdio";
reg = <0x24520 0x20>;
phy0: ethernet-phy@0 {
interrupt-parent = <&ipic>;
interrupts = <17 0x8>;
reg = <0x01>;
device_type = "ethernet-phy";
};
phy1: ethernet-phy@1 {
interrupt-parent = <&ipic>;
interrupts = <18 0x8>;
reg = <0x00>;
device_type = "ethernet-phy";
};
};
enet0: ethernet@24000 {
cell-index = <0>;
device_type = "network";
model = "TSEC";
compatible = "gianfar";
reg = <0x24000 0x1000>;
local-mac-address = [ 00 08 e5 11 32 33 ];
interrupts = <32 0x8 33 0x8 34 0x8>;
interrupt-parent = <&ipic>;
phy-handle = <&phy0>;
linux,network-index = <0>;
};
enet1: ethernet@25000 {
cell-index = <1>;
device_type = "network";
model = "TSEC";
compatible = "gianfar";
reg = <0x25000 0x1000>;
local-mac-address = [ 00 08 e5 11 32 34 ];
interrupts = <35 0x8 36 0x8 37 0x8>;
interrupt-parent = <&ipic>;
phy-handle = <&phy1>;
linux,network-index = <1>;
};
On 3.0, this works just fine. When I boot the same setup on 3.3,
I get this error:
fsl-pq_mdio: probe of ff024520.mdio failed with error -16
Is there something I need to change in my DTS file to make this work?
Something else I might be missing?
Thanks
--
------------------------------------------------------------
Gary Thomas | Consulting for the
MLB Associates | Embedded world
------------------------------------------------------------
^ permalink raw reply
* Re: [PATCH 03/12] powerpc: Rework runlatch code
From: Andreas Schwab @ 2012-04-11 9:09 UTC (permalink / raw)
To: Benjamin Herrenschmidt; +Cc: linuxppc-dev
In-Reply-To: <1334105211.2984.5.camel@pasglop>
Benjamin Herrenschmidt <benh@kernel.crashing.org> writes:
> Subject: [PATCH] powerpc: Fix typo in runlatch code
Thanks, this works for me. :-)
Andreas.
--
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4ED5
"And now for something completely different."
^ permalink raw reply
* Re: [PATCH] powerpc: Option FB_FSL_DIU is not really optional for mpc512x
From: Anatolij Gustschin @ 2012-04-11 9:08 UTC (permalink / raw)
To: Paul Gortmaker; +Cc: linuxppc-dev
In-Reply-To: <1333509300-27066-1-git-send-email-paul.gortmaker@windriver.com>
On Tue, 3 Apr 2012 23:15:00 -0400
Paul Gortmaker <paul.gortmaker@windriver.com> wrote:
> In powerpc randconfig builds, this keeps showing up:
>
> CC arch/powerpc/platforms/512x/mpc512x_shared.o
> arch/powerpc/platforms/512x/mpc512x_shared.c:70:9: warning: 'enum fsl_diu_monitor_port' declared inside parameter list
> arch/powerpc/platforms/512x/mpc512x_shared.c:70:9: warning: its scope is only this definition or declaration, which is probably not what you want
> arch/powerpc/platforms/512x/mpc512x_shared.c:69:56: error: parameter 1 ('port') has incomplete type
> arch/powerpc/platforms/512x/mpc512x_shared.c:69:5: warning: function declaration isn't a prototype
> arch/powerpc/platforms/512x/mpc512x_shared.c:84:9: warning: 'enum fsl_diu_monitor_port' declared inside parameter list
> arch/powerpc/platforms/512x/mpc512x_shared.c:83:56: error: parameter 1 ('port') has incomplete type
> arch/powerpc/platforms/512x/mpc512x_shared.c:83:6: warning: function declaration isn't a prototype
> arch/powerpc/platforms/512x/mpc512x_shared.c:88:36: warning: 'enum fsl_diu_monitor_port' declared inside parameter list
> arch/powerpc/platforms/512x/mpc512x_shared.c:88:57: error: parameter 1 ('port') has incomplete type
> arch/powerpc/platforms/512x/mpc512x_shared.c:88:6: warning: function declaration isn't a prototype
> arch/powerpc/platforms/512x/mpc512x_shared.c:187:54: error: parameter 1 ('port') has incomplete type
> arch/powerpc/platforms/512x/mpc512x_shared.c:187:1: error: return type is an incomplete type
> arch/powerpc/platforms/512x/mpc512x_shared.c:187:1: warning: function declaration isn't a prototype
> arch/powerpc/platforms/512x/mpc512x_shared.c: In function 'mpc512x_valid_monitor_port':
> arch/powerpc/platforms/512x/mpc512x_shared.c:189:9: error: 'FSL_DIU_PORT_DVI' undeclared (first use in this function)
> arch/powerpc/platforms/512x/mpc512x_shared.c:189:9: note: each undeclared identifier is reported only once for each function it appears in
> arch/powerpc/platforms/512x/mpc512x_shared.c:189:2: warning: 'return' with a value, in function returning void
> make[2]: *** [arch/powerpc/platforms/512x/mpc512x_shared.o] Error 1
>
> The reason is that mpc512x_shared.c has a couple token #ifdef
> on FB_FSL_DIU/FB_FSL_DIU_MODULE, but they don't come close to
> masking all the DIU dependencies, as the above fail shows.
>
> Rather than sprinkle more pointless #ifdef in this file, just
> remove the existing two, and make FB_FSL_DIU part of the
> dependency. The mpc512x_defconfig already has the line
> "CONFIG_FB_FSL_DIU=y" so this change should be zero impact
> on real world configs.
>
> Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Applied, thanks.
Anatolij
^ permalink raw reply
* Re: [PATCH] powerpc: 512x: Fix mpc5121_clk_get()
From: Anatolij Gustschin @ 2012-04-11 9:06 UTC (permalink / raw)
To: Richard Weinberger; +Cc: paulus, linuxppc-dev, linux-kernel
In-Reply-To: <1332788489-27435-1-git-send-email-richard@nod.at>
On Mon, 26 Mar 2012 21:01:29 +0200
Richard Weinberger <richard@nod.at> wrote:
> If try_module_get() fails, mpc5121_clk_get() might return
> a wrong clock.
>
> Signed-off-by: Richard Weinberger <richard@nod.at>
> ---
> arch/powerpc/platforms/512x/clock.c | 6 ++++--
> 1 files changed, 4 insertions(+), 2 deletions(-)
Applied, thanks.
Anatolij
^ permalink raw reply
* linux-next: boot failures with next-20120411
From: Stephen Rothwell @ 2012-04-11 6:58 UTC (permalink / raw)
To: LKML; +Cc: ppc-dev
[-- Attachment #1: Type: text/plain, Size: 2453 bytes --]
Hi all,
Some (not all) of my PowerPC boot tests have failed like this after
getting into user mode (this one was just after udev started, but others
are after other processes getting going):
Unable to handle kernel paging request for data at address 0xc0000003f9d550
Faulting instruction address: 0xc0000000001b7f40
Oops: Kernel access of bad area, sig: 11 [#1]
SMP NR_CPUS=32 NUMA pSeries
Modules linked in: ehea
NIP: c0000000001b7f40 LR: c0000000001b7f14 CTR: c0000000000e04f0
REGS: c0000003f68bf6b0 TRAP: 0300 Not tainted (3.4.0-rc2-autokern1)
MSR: 800000000280b032 <SF,VEC,VSX,EE,FP,ME,IR,DR,RI> CR: 24422424 XER: 20000001
SOFTE: 1
CFAR: 000000000000562c
DAR: 00c0000003f9d550, DSISR: 40000000
TASK = c0000003f8818000[3192] 'kdump' THREAD: c0000003f68bc000 CPU: 5
GPR00: 0000000000000000 c0000003f68bf930 c000000000ce1d40 c0000003fe00ec00
GPR04: 00000000000002d0 0000000000000038 c0000003f8f935e8 c000000000e55280
GPR08: 0000000000000011 c000000000bcb280 c000000000bcb1e8 000000000028a000
GPR12: 0000000024422424 c00000000f33bc80 00000fffdd90a770 0000000000081000
GPR16: c0000003f846c000 000000000de4f7a0 f00000000de4f7a0 0000000000000000
GPR20: c0000003f8365408 c0000003f8365480 c0000003f8e5d110 0000000000000000
GPR24: 0000000000000100 c0000003f8365400 c0000000001e5424 00000000000002d0
GPR28: 0000000000000800 00c0000003f9d550 c000000000c5b718 c0000003fe00ec00
NIP [c0000000001b7f40] .__kmalloc+0x70/0x230
LR [c0000000001b7f14] .__kmalloc+0x44/0x230
Call Trace:
[c0000003f68bf930] [c0000003f68bf9b0] 0xc0000003f68bf9b0 (unreliable)
[c0000003f68bf9e0] [c0000000001e5424] .alloc_fdmem+0x24/0x70
[c0000003f68bfa60] [c0000000001e54f8] .alloc_fdtable+0x88/0x130
[c0000003f68bfaf0] [c0000000001e5924] .dup_fd+0x384/0x450
[c0000003f68bfbd0] [c00000000009a310] .copy_process+0x880/0x11d0
[c0000003f68bfcd0] [c00000000009aee0] .do_fork+0x70/0x400
[c0000003f68bfdc0] [c0000000000141c4] .sys_clone+0x54/0x70
[c0000003f68bfe30] [c000000000009aa0] .ppc_clone+0x8/0xc
Instruction dump:
4bff9281 2ba30010 7c7f1b78 40dd00f4 e96d0040 e93f0000 7ce95a14 e9070008
7fa9582a 2fbd0000 41de0054 e81f0022 <7f3d002a> 38000000 886d01f2 980d01f2
---[ end trace 366fe6c7ced3bfb0 ]---
This did not happen yesterday. Just wondering if anyone can think of
anything obvious. Full console log at
http://ozlabs.org/~sfr/next-20120411.log.bz2
--
Cheers,
Stephen Rothwell sfr@canb.auug.org.au
[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]
^ permalink raw reply
* Re: [PATCH v5 06/27] irq_domain/powerpc: eliminate irq_map; use irq_alloc_desc() instead
From: Benjamin Herrenschmidt @ 2012-04-11 5:29 UTC (permalink / raw)
To: Andreas Schwab
Cc: Russell King - ARM Linux, devicetree-discuss, linux-kernel,
Rob Herring, Milton Miller, Thomas Gleixner, linuxppc-dev,
linux-arm-kernel
In-Reply-To: <1334107996.2984.20.camel@pasglop>
On Wed, 2012-04-11 at 11:33 +1000, Benjamin Herrenschmidt wrote:
> On Wed, 2012-04-11 at 11:13 +1000, Benjamin Herrenschmidt wrote:
> > On Sat, 2012-04-07 at 14:27 +0200, Andreas Schwab wrote:
> > > Benjamin Herrenschmidt <benh@kernel.crashing.org> writes:
> > >
> > > > It's arguable that this irq_set_irq_type(,NONE) shouln't be there but
> > > > still ... it's been around for ever and things worked :-) So something
> > > > -else- is causing the problem and I'd like to understand what exactly.
> > >
> > > AFAICS before a09b659cd68c10ec6a30cb91ebd2c327fcd5bfe5
> > > irq_set_irq_type(,NONE) was actually a no-op.
> >
> > So I'm still a bit baffled... ie, I understand some of what's happening
> > but not why it breaks things, I haven't yet managed to reproduce but I
> > haven't tried too hard just yet (was away from the HW) :
>
> Allright, I have a repro-case, I'll dig.
Ok, so it's Grant's fault :-)
So basically, it's quite subtle and I'm only 99% sure of the details but
I believe what happens is:
- The audio interrupts get virq 64 and 65 (so above NR_IRQS)
- The reverse map isn't pre-filled at map time (we should probably do
it nowadays), we do it lazily so ...
- The audio interrupt happens, it tries to pre-fill the reverse map
and ... fails (see below)
- This cause irq_linear_revmap() to returns 0
- This hits another bug in mpic where when that happens it doesn't EOI
the interrupt, which means the priority remains stuck high and we don't
get any other interrupt.
There's thus two bugs here:
- We don't properly establish the reverse mapping for 64 and 65 (well,
not always, again see below)
- We don't EOI an interrupt we can't reverse map (we should warn & EOI,
I'll send a patch for that).
The second problem is a bug we shouldn't hit if the first one didn't
happen, the first one comes from way irq_find_mapping() works. Basically
when the revmap entry is 0, we call it to find the mapping & populate
the revmap... and that fails.
That fails because it will only search up to irq_virq_count which is
statically initialized to NR_IRQS, which in our case is 64 so it won't
find our interrupts 64 and 65...
The reason it works if you don't do the set_type(NONE) is a fluke, it
will crash as soon as you actually do audio:
The set_type(NONE) has the effect in mpic to configure the interrupt to
level low (default). This is also the idle state of the audio interrupt
(which is positive edge), and the MPIC appears to be latching it (yeah
odd, it does seem to latch level interrupts).
So as soon as the audio driver enables it, it fires, triggering the bug.
Without the set_type(NONE) the occurence of the bug is delayed to the
fist time you get a real audio interrupt.
Now what is the solution ? Well, there's several things I think we need
to do:
- MPIC should properly EOI interrupts it can't revmap (I'll fix that)
- MPIC should probably not do the set_type(NONE) on map, it's not
useful
- However, we do have a risk of discrepency between the default trigger
type in the irq_desc vs. the default HW value at startup/map time, so we
do need to do something to reconcile them (that was the intend of NONE I
think)... without that, we risk having irq_create_of_mapping() not
setting the trigger because it thinks it thinks it's right... What do
you think is best here ? We can probably read the HW config and sync the
irq desc appropriately ...
- The whole business with irq_virq_count needs fixing. Basically the
default value shouldn't be NR_IRQ. I suggest making it 0 and have all
the use sites do something like:
max = irq_virq_count ? irq_virq_count : nr_irqs;
(Grant, can you take care of that ?)
- We still need to clear up all the NR_IRQS occurences in arch/powerpc
Cheers,
Ben.
^ permalink raw reply
* [PATCH 3/3] powerpc: Reformat lppaca.h
From: Anton Blanchard @ 2012-04-11 2:22 UTC (permalink / raw)
To: benh, paulus, sfr; +Cc: linuxppc-dev
In-Reply-To: <20120411122054.6297f2fb@kryten>
Reformat lppaca.h to match Linux coding standards.
Signed-off-by: Anton Blanchard <anton@samba.org>
---
Index: linux-build/arch/powerpc/include/asm/lppaca.h
===================================================================
--- linux-build.orig/arch/powerpc/include/asm/lppaca.h 2012-04-11 12:07:53.000000000 +1000
+++ linux-build/arch/powerpc/include/asm/lppaca.h 2012-04-11 12:08:25.794532160 +1000
@@ -20,18 +20,16 @@
#define _ASM_POWERPC_LPPACA_H
#ifdef __KERNEL__
-/* These definitions relate to hypervisors that only exist when using
+/*
+ * These definitions relate to hypervisors that only exist when using
* a server type processor
*/
#ifdef CONFIG_PPC_BOOK3S
-//=============================================================================
-//
-// This control block contains the data that is shared between the
-// hypervisor and the OS.
-//
-//
-//----------------------------------------------------------------------------
+/*
+ * This control block contains the data that is shared between the
+ * hypervisor and the OS.
+ */
#include <linux/cache.h>
#include <linux/threads.h>
#include <asm/types.h>
@@ -43,67 +41,65 @@
*/
#define NR_LPPACAS 1
-
-/* The Hypervisor barfs if the lppaca crosses a page boundary. A 1k
- * alignment is sufficient to prevent this */
+/*
+ * The Hypervisor barfs if the lppaca crosses a page boundary. A 1k
+ * alignment is sufficient to prevent this
+ */
struct lppaca {
-//=============================================================================
-// CACHE_LINE_1 0x0000 - 0x007F Contains read-only data
-//=============================================================================
- u32 desc; // Eye catcher 0xD397D781 x00-x03
- u16 size; // Size of this struct x04-x05
- u16 reserved1; // Reserved x06-x07
- u16 reserved2:14; // Reserved x08-x09
- u8 shared_proc:1; // Shared processor indicator ...
- u8 secondary_thread:1; // Secondary thread indicator ...
- u8 reserved3[14]; // x0A-x17
- volatile u32 dyn_hw_node_id; // Dynamic Hardware Node id x18-x1B
- volatile u32 dyn_hw_proc_id; // Dynamic Hardware Proc Id x1C-x1F
- u8 reserved4[56]; // Reserved x20-x57
- volatile u8 vphn_assoc_counts[8]; // Virtual processor home node
- // associativity change counters x58-x5F
- u8 reserved5[32]; // Reserved x60-x7F
-
-//=============================================================================
-// CACHE_LINE_2 0x0080 - 0x00FF Contains local read-write data
-//=============================================================================
-
- u8 reserved6[48]; // x00-x2f
- u8 cede_latency_hint; /* x30 */
- u8 reserved7[7]; /* x31-x37 */
- u8 dtl_enable_mask; // Dispatch Trace Log mask x38-x38
- u8 donate_dedicated_cpu; // Donate dedicated CPU cycles x39-x39
- u8 fpregs_in_use; // FP regs in use x3A-x3A
- u8 pmcregs_in_use; // PMC regs in use x3B-x3B
- u8 reserved8[28]; // x3C-x57
- u64 wait_state_cycles; // Wait cycles for this proc x58-x5F
- u8 reserved9[28]; // x60-x7B
- u16 slb_count; // # of SLBs to maintain x7C-x7D
- u8 idle; // Indicate OS is idle x7E
- u8 vmxregs_in_use; // VMX registers in use x7F
-
-//=============================================================================
-// CACHE_LINE_3 0x0100 - 0x017F: This line is shared with other processors
-//=============================================================================
- // This is the yield_count. An "odd" value (low bit on) means that
- // the processor is yielded (either because of an OS yield or a PLIC
- // preempt). An even value implies that the processor is currently
- // executing.
- // NOTE: This value will ALWAYS be zero for dedicated processors and
- // will NEVER be zero for shared processors (ie, initialized to a 1).
- volatile u32 yield_count; // PLIC increments each dispatchx00-x03
- volatile u32 dispersion_count; // dispatch changed phys cpu x04-x07
- volatile u64 cmo_faults; // CMO page fault count x08-x0F
- volatile u64 cmo_fault_time; // CMO page fault time x10-x17
- u8 reserved10[104]; // Reserved x18-x7F
-
-//=============================================================================
-// CACHE_LINE_4-5 0x0180 - 0x027F Contains PMC interrupt data
-//=============================================================================
- u32 page_ins; // CMO Hint - # page ins by OS x00-x03
- u8 reserved11[148]; // Reserved x04-x97
- volatile u64 dtl_idx; // Dispatch Trace Log head idx x98-x9F
- u8 reserved12[96]; // Reserved xA0-xFF
+ /* cacheline 1 contains read-only data */
+
+ u32 desc; /* Eye catcher 0xD397D781 */
+ u16 size; /* Size of this struct */
+ u16 reserved1;
+ u16 reserved2:14;
+ u8 shared_proc:1; /* Shared processor indicator */
+ u8 secondary_thread:1; /* Secondary thread indicator */
+ u8 reserved3[14];
+ volatile u32 dyn_hw_node_id; /* Dynamic hardware node id */
+ volatile u32 dyn_hw_proc_id; /* Dynamic hardware proc id */
+ u8 reserved4[56];
+ volatile u8 vphn_assoc_counts[8]; /* Virtual processor home node */
+ /* associativity change counters */
+ u8 reserved5[32];
+
+ /* cacheline 2 contains local read-write data */
+
+ u8 reserved6[48];
+ u8 cede_latency_hint;
+ u8 reserved7[7];
+ u8 dtl_enable_mask; /* Dispatch Trace Log mask */
+ u8 donate_dedicated_cpu; /* Donate dedicated CPU cycles */
+ u8 fpregs_in_use;
+ u8 pmcregs_in_use;
+ u8 reserved8[28];
+ u64 wait_state_cycles; /* Wait cycles for this proc */
+ u8 reserved9[28];
+ u16 slb_count; /* # of SLBs to maintain */
+ u8 idle; /* Indicate OS is idle */
+ u8 vmxregs_in_use;
+
+ /* cacheline 3 is shared with other processors */
+
+ /*
+ * This is the yield_count. An "odd" value (low bit on) means that
+ * the processor is yielded (either because of an OS yield or a
+ * hypervisor preempt). An even value implies that the processor is
+ * currently executing.
+ * NOTE: This value will ALWAYS be zero for dedicated processors and
+ * will NEVER be zero for shared processors (ie, initialized to a 1).
+ */
+ volatile u32 yield_count;
+ volatile u32 dispersion_count; /* dispatch changed physical cpu */
+ volatile u64 cmo_faults; /* CMO page fault count */
+ volatile u64 cmo_fault_time; /* CMO page fault time */
+ u8 reserved10[104];
+
+ /* cacheline 4-5 */
+
+ u32 page_ins; /* CMO Hint - # page ins by OS */
+ u8 reserved11[148];
+ volatile u64 dtl_idx; /* Dispatch Trace Log head index */
+ u8 reserved12[96];
} __attribute__((__aligned__(0x400)));
extern struct lppaca lppaca[];
@@ -116,13 +112,13 @@ extern struct lppaca lppaca[];
* ESID is stored in the lower 64bits, then the VSID.
*/
struct slb_shadow {
- u32 persistent; // Number of persistent SLBs x00-x03
- u32 buffer_length; // Total shadow buffer length x04-x07
- u64 reserved; // Alignment x08-x0f
+ u32 persistent; /* Number of persistent SLBs */
+ u32 buffer_length; /* Total shadow buffer length */
+ u64 reserved;
struct {
u64 esid;
u64 vsid;
- } save_area[SLB_NUM_BOLTED]; // x10-x40
+ } save_area[SLB_NUM_BOLTED];
} ____cacheline_aligned;
extern struct slb_shadow slb_shadow[];
^ permalink raw reply
* [PATCH 2/3] powerpc: Remove iseries specific fields in lppaca
From: Anton Blanchard @ 2012-04-11 2:22 UTC (permalink / raw)
To: benh, paulus, sfr; +Cc: linuxppc-dev
In-Reply-To: <20120411122054.6297f2fb@kryten>
Remove all the iseries specific fields in the lppaca.
Signed-off-by: Anton Blanchard <anton@samba.org>
---
Index: linux-build/arch/powerpc/include/asm/lppaca.h
===================================================================
--- linux-build.orig/arch/powerpc/include/asm/lppaca.h 2012-04-11 12:05:54.000000000 +1000
+++ linux-build/arch/powerpc/include/asm/lppaca.h 2012-04-11 12:07:53.537907718 +1000
@@ -28,7 +28,7 @@
//=============================================================================
//
// This control block contains the data that is shared between the
-// hypervisor (PLIC) and the OS.
+// hypervisor and the OS.
//
//
//----------------------------------------------------------------------------
@@ -49,9 +49,6 @@
struct lppaca {
//=============================================================================
// CACHE_LINE_1 0x0000 - 0x007F Contains read-only data
-// NOTE: The xDynXyz fields are fields that will be dynamically changed by
-// PLIC when preparing to bring a processor online or when dispatching a
-// virtual processor!
//=============================================================================
u32 desc; // Eye catcher 0xD397D781 x00-x03
u16 size; // Size of this struct x04-x05
@@ -59,75 +56,32 @@ struct lppaca {
u16 reserved2:14; // Reserved x08-x09
u8 shared_proc:1; // Shared processor indicator ...
u8 secondary_thread:1; // Secondary thread indicator ...
- volatile u8 dyn_proc_status:8; // Dynamic Status of this proc x0A-x0A
- u8 secondary_thread_count; // Secondary thread count x0B-x0B
- volatile u16 dyn_hv_phys_proc_index;// Dynamic HV Physical Proc Index0C-x0D
- volatile u16 dyn_hv_log_proc_index;// Dynamic HV Logical Proc Indexx0E-x0F
- u32 decr_val; // Value for Decr programming x10-x13
- u32 pmc_val; // Value for PMC regs x14-x17
+ u8 reserved3[14]; // x0A-x17
volatile u32 dyn_hw_node_id; // Dynamic Hardware Node id x18-x1B
volatile u32 dyn_hw_proc_id; // Dynamic Hardware Proc Id x1C-x1F
- volatile u32 dyn_pir; // Dynamic ProcIdReg value x20-x23
- u32 dsei_data; // DSEI data x24-x27
- u64 sprg3; // SPRG3 value x28-x2F
- u8 reserved3[40]; // Reserved x30-x57
+ u8 reserved4[56]; // Reserved x20-x57
volatile u8 vphn_assoc_counts[8]; // Virtual processor home node
// associativity change counters x58-x5F
- u8 reserved4[32]; // Reserved x60-x7F
+ u8 reserved5[32]; // Reserved x60-x7F
//=============================================================================
// CACHE_LINE_2 0x0080 - 0x00FF Contains local read-write data
//=============================================================================
- // This Dword contains a byte for each type of interrupt that can occur.
- // The IPI is a count while the others are just a binary 1 or 0.
- union {
- u64 any_int;
- struct {
- u16 reserved; // Reserved - cleared by #mpasmbl
- u8 xirr_int; // Indicates xXirrValue is valid or Immed IO
- u8 ipi_cnt; // IPI Count
- u8 decr_int; // DECR interrupt occurred
- u8 pdc_int; // PDC interrupt occurred
- u8 quantum_int; // Interrupt quantum reached
- u8 old_plic_deferred_ext_int; // Old PLIC has a deferred XIRR pending
- } fields;
- } int_dword;
-
- // Whenever any fields in this Dword are set then PLIC will defer the
- // processing of external interrupts. Note that PLIC will store the
- // XIRR directly into the xXirrValue field so that another XIRR will
- // not be presented until this one clears. The layout of the low
- // 4-bytes of this Dword is up to SLIC - PLIC just checks whether the
- // entire Dword is zero or not. A non-zero value in the low order
- // 2-bytes will result in SLIC being granted the highest thread
- // priority upon return. A 0 will return to SLIC as medium priority.
- u64 plic_defer_ints_area; // Entire Dword
-
- // Used to pass the real SRR0/1 from PLIC to SLIC as well as to
- // pass the target SRR0/1 from SLIC to PLIC on a SetAsrAndRfid.
- u64 saved_srr0; // Saved SRR0 x10-x17
- u64 saved_srr1; // Saved SRR1 x18-x1F
- u64 reserved5[2]; /* x20-x2F */
+
+ u8 reserved6[48]; // x00-x2f
u8 cede_latency_hint; /* x30 */
- u8 reserved[7]; /* x31-x37 */
+ u8 reserved7[7]; /* x31-x37 */
u8 dtl_enable_mask; // Dispatch Trace Log mask x38-x38
u8 donate_dedicated_cpu; // Donate dedicated CPU cycles x39-x39
u8 fpregs_in_use; // FP regs in use x3A-x3A
u8 pmcregs_in_use; // PMC regs in use x3B-x3B
- volatile u32 saved_decr; // Saved Decr Value x3C-x3F
- volatile u64 emulated_time_base;// Emulated TB for this thread x40-x47
- volatile u64 cur_plic_latency; // Unaccounted PLIC latency x48-x4F
- u64 tot_plic_latency; // Accumulated PLIC latency x50-x57
+ u8 reserved8[28]; // x3C-x57
u64 wait_state_cycles; // Wait cycles for this proc x58-x5F
- u64 end_of_quantum; // TB at end of quantum x60-x67
- u64 pdc_saved_sprg1; // Saved SPRG1 for PMC int x68-x6F
- u64 pdc_saved_srr0; // Saved SRR0 for PMC int x70-x77
- volatile u32 virtual_decr; // Virtual DECR for shared procsx78-x7B
+ u8 reserved9[28]; // x60-x7B
u16 slb_count; // # of SLBs to maintain x7C-x7D
u8 idle; // Indicate OS is idle x7E
u8 vmxregs_in_use; // VMX registers in use x7F
-
//=============================================================================
// CACHE_LINE_3 0x0100 - 0x017F: This line is shared with other processors
//=============================================================================
@@ -141,15 +95,15 @@ struct lppaca {
volatile u32 dispersion_count; // dispatch changed phys cpu x04-x07
volatile u64 cmo_faults; // CMO page fault count x08-x0F
volatile u64 cmo_fault_time; // CMO page fault time x10-x17
- u8 reserved7[104]; // Reserved x18-x7F
+ u8 reserved10[104]; // Reserved x18-x7F
//=============================================================================
// CACHE_LINE_4-5 0x0180 - 0x027F Contains PMC interrupt data
//=============================================================================
u32 page_ins; // CMO Hint - # page ins by OS x00-x03
- u8 reserved8[148]; // Reserved x04-x97
+ u8 reserved11[148]; // Reserved x04-x97
volatile u64 dtl_idx; // Dispatch Trace Log head idx x98-x9F
- u8 reserved9[96]; // Reserved xA0-xFF
+ u8 reserved12[96]; // Reserved xA0-xFF
} __attribute__((__aligned__(0x400)));
extern struct lppaca lppaca[];
Index: linux-build/arch/powerpc/kernel/paca.c
===================================================================
--- linux-build.orig/arch/powerpc/kernel/paca.c 2012-04-11 12:05:15.000000000 +1000
+++ linux-build/arch/powerpc/kernel/paca.c 2012-04-11 12:06:43.012541685 +1000
@@ -36,10 +36,7 @@ struct lppaca lppaca[] = {
[0 ... (NR_LPPACAS-1)] = {
.desc = 0xd397d781, /* "LpPa" */
.size = sizeof(struct lppaca),
- .dyn_proc_status = 2,
- .decr_val = 0x00ff0000,
.fpregs_in_use = 1,
- .end_of_quantum = 0xfffffffffffffffful,
.slb_count = 64,
.vmxregs_in_use = 0,
.page_ins = 0,
Index: linux-build/arch/powerpc/kernel/asm-offsets.c
===================================================================
--- linux-build.orig/arch/powerpc/kernel/asm-offsets.c 2012-04-11 12:05:15.000000000 +1000
+++ linux-build/arch/powerpc/kernel/asm-offsets.c 2012-04-11 12:06:43.012541685 +1000
@@ -188,10 +188,6 @@ int main(void)
DEFINE(SLBSHADOW_STACKESID,
offsetof(struct slb_shadow, save_area[SLB_NUM_BOLTED - 1].esid));
DEFINE(SLBSHADOW_SAVEAREA, offsetof(struct slb_shadow, save_area));
- DEFINE(LPPACASRR0, offsetof(struct lppaca, saved_srr0));
- DEFINE(LPPACASRR1, offsetof(struct lppaca, saved_srr1));
- DEFINE(LPPACAANYINT, offsetof(struct lppaca, int_dword.any_int));
- DEFINE(LPPACADECRINT, offsetof(struct lppaca, int_dword.fields.decr_int));
DEFINE(LPPACA_PMCINUSE, offsetof(struct lppaca, pmcregs_in_use));
DEFINE(LPPACA_DTLIDX, offsetof(struct lppaca, dtl_idx));
DEFINE(LPPACA_YIELDCOUNT, offsetof(struct lppaca, yield_count));
^ permalink raw reply
* [PATCH 1/3] powerpc: Clean up lppaca->cede_latency_hint
From: Anton Blanchard @ 2012-04-11 2:20 UTC (permalink / raw)
To: benh, paulus, sfr; +Cc: linuxppc-dev
We have a union containing fields from the old iseries hypervisor
that has been reused for the cede latency hint. Since we no
longer support iseries, remove the union completely.
Signed-off-by: Anton Blanchard <anton@samba.org>
---
Index: linux-build/arch/powerpc/include/asm/lppaca.h
===================================================================
--- linux-build.orig/arch/powerpc/include/asm/lppaca.h 2012-04-11 11:59:57.632828804 +1000
+++ linux-build/arch/powerpc/include/asm/lppaca.h 2012-04-11 12:05:54.767606593 +1000
@@ -107,19 +107,9 @@ struct lppaca {
// pass the target SRR0/1 from SLIC to PLIC on a SetAsrAndRfid.
u64 saved_srr0; // Saved SRR0 x10-x17
u64 saved_srr1; // Saved SRR1 x18-x1F
-
- // Used to pass parms from the OS to PLIC for SetAsrAndRfid
- u64 saved_gpr3; // Saved GPR3 x20-x27
- u64 saved_gpr4; // Saved GPR4 x28-x2F
- union {
- u64 saved_gpr5; /* Saved GPR5 x30-x37 */
- struct {
- u8 cede_latency_hint; /* x30 */
- u8 reserved[7]; /* x31-x36 */
- } fields;
- } gpr5_dword;
-
-
+ u64 reserved5[2]; /* x20-x2F */
+ u8 cede_latency_hint; /* x30 */
+ u8 reserved[7]; /* x31-x37 */
u8 dtl_enable_mask; // Dispatch Trace Log mask x38-x38
u8 donate_dedicated_cpu; // Donate dedicated CPU cycles x39-x39
u8 fpregs_in_use; // FP regs in use x3A-x3A
Index: linux-build/arch/powerpc/platforms/pseries/plpar_wrappers.h
===================================================================
--- linux-build.orig/arch/powerpc/platforms/pseries/plpar_wrappers.h 2012-04-11 11:59:57.652829173 +1000
+++ linux-build/arch/powerpc/platforms/pseries/plpar_wrappers.h 2012-04-11 12:00:13.473120748 +1000
@@ -22,12 +22,12 @@ static inline long poll_pending(void)
static inline u8 get_cede_latency_hint(void)
{
- return get_lppaca()->gpr5_dword.fields.cede_latency_hint;
+ return get_lppaca()->cede_latency_hint;
}
static inline void set_cede_latency_hint(u8 latency_hint)
{
- get_lppaca()->gpr5_dword.fields.cede_latency_hint = latency_hint;
+ get_lppaca()->cede_latency_hint = latency_hint;
}
static inline long cede_processor(void)
^ permalink raw reply
* Re: [PATCH v5 06/27] irq_domain/powerpc: eliminate irq_map; use irq_alloc_desc() instead
From: Benjamin Herrenschmidt @ 2012-04-11 1:33 UTC (permalink / raw)
To: Andreas Schwab
Cc: Russell King - ARM Linux, devicetree-discuss, linux-kernel,
Rob Herring, Milton Miller, Thomas Gleixner, linuxppc-dev,
linux-arm-kernel
In-Reply-To: <1334106834.2984.17.camel@pasglop>
On Wed, 2012-04-11 at 11:13 +1000, Benjamin Herrenschmidt wrote:
> On Sat, 2012-04-07 at 14:27 +0200, Andreas Schwab wrote:
> > Benjamin Herrenschmidt <benh@kernel.crashing.org> writes:
> >
> > > It's arguable that this irq_set_irq_type(,NONE) shouln't be there but
> > > still ... it's been around for ever and things worked :-) So something
> > > -else- is causing the problem and I'd like to understand what exactly.
> >
> > AFAICS before a09b659cd68c10ec6a30cb91ebd2c327fcd5bfe5
> > irq_set_irq_type(,NONE) was actually a no-op.
>
> So I'm still a bit baffled... ie, I understand some of what's happening
> but not why it breaks things, I haven't yet managed to reproduce but I
> haven't tried too hard just yet (was away from the HW) :
Allright, I have a repro-case, I'll dig.
Cheers,
Ben.
^ permalink raw reply
* Re: [PATCH v5 06/27] irq_domain/powerpc: eliminate irq_map; use irq_alloc_desc() instead
From: Benjamin Herrenschmidt @ 2012-04-11 1:13 UTC (permalink / raw)
To: Andreas Schwab
Cc: Russell King - ARM Linux, devicetree-discuss, linux-kernel,
Rob Herring, Milton Miller, Thomas Gleixner, linuxppc-dev,
linux-arm-kernel
In-Reply-To: <m2aa2n7mmw.fsf@igel.home>
On Sat, 2012-04-07 at 14:27 +0200, Andreas Schwab wrote:
> Benjamin Herrenschmidt <benh@kernel.crashing.org> writes:
>
> > It's arguable that this irq_set_irq_type(,NONE) shouln't be there but
> > still ... it's been around for ever and things worked :-) So something
> > -else- is causing the problem and I'd like to understand what exactly.
>
> AFAICS before a09b659cd68c10ec6a30cb91ebd2c327fcd5bfe5
> irq_set_irq_type(,NONE) was actually a no-op.
So I'm still a bit baffled... ie, I understand some of what's happening
but not why it breaks things, I haven't yet managed to reproduce but I
haven't tried too hard just yet (was away from the HW) :
Basically, we end up setting the mpic to IRQ_TYPE_LEVEL_LOW as a result
of a request to set it to IRQ_TYPE_NONE, ie, we apply a "default"
setting instead of just setting it to "none" (in part because we don't
have a way in HW to disable the trigger completely. We then write that
into the irq_desc with irqd_set_trigger_type().
Then we return IRQ_SET_MASK_OK_NOCOPY, which means that the generic
__irq_set_trigger() code will read back what we've set in the desc and
adjust things accordingly, rather than writing the original value in,
ie, our descriptor ends up being effectively set to IRQ_TYPE_LEVEL_LOW
rather than IRQ_TYPE_NONE.
Later, when we actually try to set it to IRQ_TYPE_LEVEL_LOW as a result
of parsing the device-tree, we then hit the code
irq_create_of_mapping(), at the bottom, which will skip the call to
irq_set_irq_type() if we are trying to set it to the same type that it's
already set to....
But I don't see why that breaks anything, ie we -have- set things to
LEVEL_LOW as a result of IRQ_TYPE_NONE, which is the right setting, so
things should work despite the fact that we skip the second call.
Out of curiosity, can you try removing the check in irqdomain and always
apply the setting regardless of the previous value (ie, remove the check
of type against irqd_get_trigger at the end of irq_create_of_mapping),
see if that makes a difference ?
Cheers,
Ben.
^ permalink raw reply
* Re: [PATCH 03/12] powerpc: Rework runlatch code
From: Benjamin Herrenschmidt @ 2012-04-11 0:46 UTC (permalink / raw)
To: Andreas Schwab; +Cc: linuxppc-dev
In-Reply-To: <m2vcldamh2.fsf@igel.home>
On Thu, 2012-04-05 at 23:38 +0200, Andreas Schwab wrote:
> Benjamin Herrenschmidt <benh@kernel.crashing.org> writes:
>
> > This moves the inlines into system.h and changes the runlatch
> > code to use the thread local flags (non-atomic) rather than
> > the TIF flags (atomic) to keep track of the latch state.
> >
> > The code to turn it back on in an asynchronous interrupt is
> > now simplified and partially inlined.
>
> This breaks the X server in that it no longer receives SIGIO signals.
I think I found my st00pid bug, can you try this patch ?
>From fae2e0fb24c61ca68c98d854a34732549ebc1854 Mon Sep 17 00:00:00 2001
From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Date: Wed, 11 Apr 2012 10:42:15 +1000
Subject: [PATCH] powerpc: Fix typo in runlatch code
Commit fe1952fc0afb9a2e4c79f103c08aef5d13db1873
"powerpc: Rework runlatch code" has a nasty typo
where it uses "TLF_RUNLATCH" instead of "_TLF_RUNLATCH"
(bit number instead of bit mask), causing some flags to
be potentially lost such as _TLF_RESTORE_SIGMASK
(Brown paper bag for me ! We should be able to make
that break at compile time with a bit of magic, any
volunteer ?)
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
---
arch/powerpc/kernel/process.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c
index f88698c..4937c96 100644
--- a/arch/powerpc/kernel/process.c
+++ b/arch/powerpc/kernel/process.c
@@ -1235,7 +1235,7 @@ void __ppc64_runlatch_on(void)
ctrl |= CTRL_RUNLATCH;
mtspr(SPRN_CTRLT, ctrl);
- ti->local_flags |= TLF_RUNLATCH;
+ ti->local_flags |= _TLF_RUNLATCH;
}
/* Called with hard IRQs off */
@@ -1244,7 +1244,7 @@ void __ppc64_runlatch_off(void)
struct thread_info *ti = current_thread_info();
unsigned long ctrl;
- ti->local_flags &= ~TLF_RUNLATCH;
+ ti->local_flags &= ~_TLF_RUNLATCH;
ctrl = mfspr(SPRN_CTRLF);
ctrl &= ~CTRL_RUNLATCH;
--
1.7.9.5
^ permalink raw reply related
* Re: [PATCH v2 14/17] powerpc: crypto: nx driver code supporting nx encryption
From: Kent Yoder @ 2012-04-10 18:53 UTC (permalink / raw)
To: David Miller; +Cc: rcj, linuxppc-dev, linux-kernel, linux-crypto
In-Reply-To: <20120410.144136.1777978593334560368.davem@davemloft.net>
Hi Dave,
On Tue, 2012-04-10 at 14:41 -0400, David Miller wrote:
> From: Kent Yoder <key@linux.vnet.ibm.com>
> Date: Tue, 10 Apr 2012 10:10:57 -0500
>
> > These routines add the base device driver code supporting the Power7+
> > in-Nest encryption accelerator (nx) device.
> >
> > Signed-off-by: Kent Yoder <key@linux.vnet.ibm.com>
>
> This patch should be earlier in the series, so it can
> appear before the drivers which use the helper routines
> in here.
I wasn't sure which order to post these in since the structures I
register with the crypto api in this patch live in the earlier .c files,
so there's a circular dependency. I can reorder if you want though.
Kent
>
^ permalink raw reply
* Re: [PATCH v2 14/17] powerpc: crypto: nx driver code supporting nx encryption
From: David Miller @ 2012-04-10 18:41 UTC (permalink / raw)
To: key; +Cc: rcj, linuxppc-dev, linux-kernel, linux-crypto
In-Reply-To: <1334070657.16827.29.camel@key-ThinkPad-W510>
From: Kent Yoder <key@linux.vnet.ibm.com>
Date: Tue, 10 Apr 2012 10:10:57 -0500
> These routines add the base device driver code supporting the Power7+
> in-Nest encryption accelerator (nx) device.
>
> Signed-off-by: Kent Yoder <key@linux.vnet.ibm.com>
This patch should be earlier in the series, so it can
appear before the drivers which use the helper routines
in here.
^ permalink raw reply
* Re: 3.4.0-rc1: No init found
From: Christian Kujau @ 2012-04-10 16:56 UTC (permalink / raw)
To: Benjamin Herrenschmidt; +Cc: linuxppc-dev, Suzuki K. Poulose, LKML
In-Reply-To: <1334042632.3040.54.camel@pasglop>
On Tue, 10 Apr 2012 at 17:23, Benjamin Herrenschmidt wrote:
> can you test the patch below ?
When applied to 3.4-rc2, Linux boots again, yay! :-)
Tested-by: Christian Kujau <lists@nerdbynature.de>
Thanks!
Christian.
> From 08f1ec8a594c60bf3856e3c45b6d15fd691d90bb Mon Sep 17 00:00:00 2001
> From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
> Date: Tue, 10 Apr 2012 17:21:35 +1000
> Subject: [PATCH] powerpc: Fix page fault with lockdep regression
>
> commit a546498f3bf9aac311c66f965186373aee2ca0b0
> introduced a regression on 32-bit when irq tracing
> is enabled by exposing an old bug in our irq tracing
> code for exception entry.
>
> The code would save and restore some GPRs around the
> calls to the C lockdep code, however, it tries to be
> too smart for its own good and restores some of the
> GPRs from the exception frame (as saved there on
> exception entry).
>
> However, for page faults, we do replace those GPRs with
> arguments to do_page_fault before we call transfer_to_handler
> and so restoring from the exception frame is plain wrong in
> this case.
>
> This was fine as long as we didn't touch the interrupt state
> when taking page fault, but when I started doing it, it would
> trigger the lockdep calls and the bug.
>
> This fixes it by cleaning up that code a bit. It did create
> a small stack frame for the sake of backtraces, so let's
> make it a bit bigger and use it to save and restore the
> stuff we care about.
>
> Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
> ---
> arch/powerpc/kernel/entry_32.S | 39 +++++++++++++++++++++------------------
> 1 file changed, 21 insertions(+), 18 deletions(-)
>
> diff --git a/arch/powerpc/kernel/entry_32.S b/arch/powerpc/kernel/entry_32.S
> index 3e57a00..ba3aeb4 100644
> --- a/arch/powerpc/kernel/entry_32.S
> +++ b/arch/powerpc/kernel/entry_32.S
> @@ -206,40 +206,43 @@ reenable_mmu: /* re-enable mmu so we can */
> andi. r10,r10,MSR_EE /* Did EE change? */
> beq 1f
>
> - /* Save handler and return address into the 2 unused words
> - * of the STACK_FRAME_OVERHEAD (sneak sneak sneak). Everything
> - * else can be recovered from the pt_regs except r3 which for
> - * normal interrupts has been set to pt_regs and for syscalls
> - * is an argument, so we temporarily use ORIG_GPR3 to save it
> - */
> - stw r9,8(r1)
> - stw r11,12(r1)
> - stw r3,ORIG_GPR3(r1)
> /*
> * The trace_hardirqs_off will use CALLER_ADDR0 and CALLER_ADDR1.
> * If from user mode there is only one stack frame on the stack, and
> * accessing CALLER_ADDR1 will cause oops. So we need create a dummy
> * stack frame to make trace_hardirqs_off happy.
> + *
> + * This is handy because we also need to save a bunch of GPRs,
> + * r3 can be different from GPR3(r1) at this point, r9 and r11
> + * contains the old MSR and handler address respectively,
> + * r4 & r5 can contain page fault arguments that need to be passed
> + * along as well. r12, CCR, CTR, XER etc... are left clobbered as
> + * they aren't useful past this point (aren't syscall arguments),
> + * the rest is restored from the exception frame.
> */
> + stwu r1,-32(r1)
> + stw r9,8(r1)
> + stw r11,12(r1)
> + stw r3,16(r1)
> + stw r4,20(r1)
> + stw r5,24(r1)
> andi. r12,r12,MSR_PR
> - beq 11f
> - stwu r1,-16(r1)
> + b 11f
> bl trace_hardirqs_off
> - addi r1,r1,16
> b 12f
> -
> 11:
> bl trace_hardirqs_off
> 12:
> + lwz r5,24(r1)
> + lwz r4,20(r1)
> + lwz r3,16(r1)
> + lwz r11,12(r1)
> + lwz r9,8(r1)
> + addi r1,r1,32
> lwz r0,GPR0(r1)
> - lwz r3,ORIG_GPR3(r1)
> - lwz r4,GPR4(r1)
> - lwz r5,GPR5(r1)
> lwz r6,GPR6(r1)
> lwz r7,GPR7(r1)
> lwz r8,GPR8(r1)
> - lwz r9,8(r1)
> - lwz r11,12(r1)
> 1: mtctr r11
> mtlr r9
> bctr /* jump to handler */
> --
> 1.7.9.5
>
--
BOFH excuse #292:
We ran out of dial tone and we're and waiting for the phone company to deliver another bottle.
^ permalink raw reply
* [PATCH v2 01/17] powerpc: Add new hvcall constants to support PFO
From: Kent Yoder @ 2012-04-10 15:08 UTC (permalink / raw)
To: linux-kernel; +Cc: key, rcj, linuxppc-dev, linux-crypto
In-Reply-To: <1334070249.16827.1.camel@key-ThinkPad-W510>
The Platform Facilities Option (PFO) adds several new h_calls and
more return codes.
Signed-off-by: Robert Jennings <rcj@linux.vnet.ibm.com>
Signed-off-by: Kent Yoder <key@linux.vnet.ibm.com>
---
arch/powerpc/include/asm/hvcall.h | 25 +++++++++++++++++++++++--
1 files changed, 23 insertions(+), 2 deletions(-)
diff --git a/arch/powerpc/include/asm/hvcall.h b/arch/powerpc/include/asm/hvcall.h
index 1c324ff..6122523 100644
--- a/arch/powerpc/include/asm/hvcall.h
+++ b/arch/powerpc/include/asm/hvcall.h
@@ -77,8 +77,27 @@
#define H_MR_CONDITION -43
#define H_NOT_ENOUGH_RESOURCES -44
#define H_R_STATE -45
-#define H_RESCINDEND -46
-#define H_MULTI_THREADS_ACTIVE -9005
+#define H_RESCINDED -46
+#define H_P2 -55
+#define H_P3 -56
+#define H_P4 -57
+#define H_P5 -58
+#define H_P6 -59
+#define H_P7 -60
+#define H_P8 -61
+#define H_P9 -62
+#define H_TOO_BIG -64
+#define H_OVERLAP -68
+#define H_INTERRUPT -69
+#define H_BAD_DATA -70
+#define H_NOT_ACTIVE -71
+#define H_SG_LIST -72
+#define H_OP_MODE -73
+#define H_COP_HW -74
+#define H_UNSUPPORTED_FLAG_START -256
+#define H_UNSUPPORTED_FLAG_END -511
+#define H_MULTI_THREADS_ACTIVE -9005
+#define H_OUTSTANDING_COP_OPS -9006
/* Long Busy is a condition that can be returned by the firmware
@@ -240,6 +259,8 @@
#define H_GET_MPP 0x2D4
#define H_HOME_NODE_ASSOCIATIVITY 0x2EC
#define H_BEST_ENERGY 0x2F4
+#define H_RANDOM 0x300
+#define H_COP 0x304
#define H_GET_MPP_X 0x314
#define MAX_HCALL_OPCODE H_GET_MPP_X
--
1.7.1
^ permalink raw reply related
* [PATCH v2 03/17] powerpc: Add PFO support to the VIO bus
From: Kent Yoder @ 2012-04-10 15:08 UTC (permalink / raw)
To: linux-kernel; +Cc: key, rcj, linuxppc-dev, linux-crypto
In-Reply-To: <1334070249.16827.1.camel@key-ThinkPad-W510>
Add support for the Platform Facilities Option (PFO) to the VIO bus.
These devices have a separate root node in OpenFirmware which
requires additional parsing to map into the existing VIO device
structure fields. This adds the interface for PFO device drivers to
make synchronous hypervisor calls.
Signed-off-by: Robert Jennings <rcj@linux.vnet.ibm.com>
Signed-off-by: Kent Yoder <key@linux.vnet.ibm.com>
---
arch/powerpc/include/asm/vio.h | 46 +++++++
arch/powerpc/kernel/vio.c | 273 ++++++++++++++++++++++++++++++++++------
2 files changed, 280 insertions(+), 39 deletions(-)
diff --git a/arch/powerpc/include/asm/vio.h b/arch/powerpc/include/asm/vio.h
index 6bfd5ff..b19adf7 100644
--- a/arch/powerpc/include/asm/vio.h
+++ b/arch/powerpc/include/asm/vio.h
@@ -46,6 +46,48 @@
struct iommu_table;
+/*
+ * Platform Facilities Option (PFO)-specific data
+ */
+
+/* Starting unit address for PFO devices on the VIO BUS */
+#define VIO_BASE_PFO_UA 0x50000000
+
+/**
+ * vio_pfo_op - PFO operation parameters
+ *
+ * @flags: h_call subfunctions and modifiers
+ * @in: Input data block logical real address
+ * @inlen: If non-negative, the length of the input data block. If negative,
+ * the length of the input data descriptor list in bytes.
+ * @out: Output data block logical real address
+ * @outlen: If non-negative, the length of the input data block. If negative,
+ * the length of the input data descriptor list in bytes.
+ * @csbcpb: Logical real address of the 4k naturally-aligned storage block
+ * containing the CSB & optional FC field specific CPB
+ * @timeout: # of milliseconds to retry h_call, 0 for no timeout.
+ * @hcall_err: pointer to return the h_call return value, else NULL
+ */
+struct vio_pfo_op {
+ u64 flags;
+ s64 in;
+ s64 inlen;
+ s64 out;
+ s64 outlen;
+ u64 csbcpb;
+ void *done;
+ unsigned long handle;
+ unsigned int timeout;
+ long hcall_err;
+};
+
+/* End PFO specific data */
+
+enum vio_dev_family {
+ VDEVICE, /* The OF node is a child of /vdevice */
+ PFO, /* The OF node is a child of /ibm,platform-facilities */
+};
+
/**
* vio_dev - This structure is used to describe virtual I/O devices.
*
@@ -58,6 +100,7 @@ struct vio_dev {
const char *name;
const char *type;
uint32_t unit_address;
+ uint32_t resource_id;
unsigned int irq;
struct {
size_t desired;
@@ -65,6 +108,7 @@ struct vio_dev {
size_t allocated;
atomic_t allocs_failed;
} cmo;
+ enum vio_dev_family family;
struct device dev;
};
@@ -95,6 +139,8 @@ extern void vio_cmo_set_dev_desired(struct vio_dev *viodev, size_t desired);
extern void __devinit vio_unregister_device(struct vio_dev *dev);
+extern int vio_h_cop_sync(struct vio_dev *vdev, struct vio_pfo_op *op);
+
struct device_node;
extern struct vio_dev *vio_register_device_node(
diff --git a/arch/powerpc/kernel/vio.c b/arch/powerpc/kernel/vio.c
index a3a9990..cb87301 100644
--- a/arch/powerpc/kernel/vio.c
+++ b/arch/powerpc/kernel/vio.c
@@ -14,7 +14,9 @@
* 2 of the License, or (at your option) any later version.
*/
+#include <linux/cpu.h>
#include <linux/types.h>
+#include <linux/delay.h>
#include <linux/stat.h>
#include <linux/device.h>
#include <linux/init.h>
@@ -709,13 +711,26 @@ static int vio_cmo_bus_probe(struct vio_dev *viodev)
struct vio_driver *viodrv = to_vio_driver(dev->driver);
unsigned long flags;
size_t size;
+ bool dma_capable = false;
+
+ /* A device requires entitlement if it has a DMA window property */
+ switch (viodev->family) {
+ case VDEVICE:
+ if (of_get_property(viodev->dev.of_node,
+ "ibm,my-dma-window", NULL))
+ dma_capable = true;
+ break;
+ case PFO:
+ dma_capable = false;
+ break;
+ default:
+ dev_warn(dev, "unknown device family: %d\n", viodev->family);
+ BUG();
+ break;
+ }
- /*
- * Check to see that device has a DMA window and configure
- * entitlement for the device.
- */
- if (of_get_property(viodev->dev.of_node,
- "ibm,my-dma-window", NULL)) {
+ /* Configure entitlement for the device. */
+ if (dma_capable) {
/* Check that the driver is CMO enabled and get desired DMA */
if (!viodrv->get_desired_dma) {
dev_err(dev, "%s: device driver does not support CMO\n",
@@ -1050,6 +1065,94 @@ static void vio_cmo_sysfs_init(void) { }
EXPORT_SYMBOL(vio_cmo_entitlement_update);
EXPORT_SYMBOL(vio_cmo_set_dev_desired);
+
+/*
+ * Platform Facilities Option (PFO) support
+ */
+
+/**
+ * vio_h_cop_sync - Perform a synchronous PFO co-processor operation
+ *
+ * @vdev - Pointer to a struct vio_dev for device
+ * @op - Pointer to a struct vio_pfo_op for the operation parameters
+ *
+ * Calls the hypervisor to synchronously perform the PFO operation
+ * described in @op. In the case of a busy response from the hypervisor,
+ * the operation will be re-submitted indefinitely unless a non-zero timeout
+ * is specified or an error occurs. The timeout places a limit on when to
+ * stop re-submitting a operation, the total time can be exceeded if an
+ * operation is in progress.
+ *
+ * If op->hcall_ret is not NULL, this will be set to the return from the
+ * last h_cop_op call or it will be 0 if an error not involving the h_call
+ * was encountered.
+ *
+ * Returns:
+ * 0 on success,
+ * -EINVAL if the h_call fails due to an invalid parameter,
+ * -E2BIG if the h_call can not be performed synchronously,
+ * -EBUSY if a timeout is specified and has elapsed,
+ * -EACCES if the memory area for data/status has been rescinded, or
+ * -EPERM if a hardware fault has been indicated
+ */
+int vio_h_cop_sync(struct vio_dev *vdev, struct vio_pfo_op *op)
+{
+ struct device *dev = &vdev->dev;
+ unsigned long deadline = 0;
+ long hret = 0;
+ int ret = 0;
+
+ if (op->timeout)
+ deadline = jiffies + msecs_to_jiffies(op->timeout);
+
+ while (true) {
+ hret = plpar_hcall_norets(H_COP, op->flags,
+ vdev->resource_id,
+ op->in, op->inlen, op->out,
+ op->outlen, op->csbcpb);
+
+ if (hret == H_SUCCESS ||
+ (hret != H_NOT_ENOUGH_RESOURCES &&
+ hret != H_BUSY && hret != H_RESOURCE) ||
+ (op->timeout && time_after(deadline, jiffies)))
+ break;
+
+ dev_dbg(dev, "%s: hcall ret(%ld), retrying.\n", __func__, hret);
+ }
+
+ switch (hret) {
+ case H_SUCCESS:
+ ret = 0;
+ break;
+ case H_OP_MODE:
+ case H_TOO_BIG:
+ ret = -E2BIG;
+ break;
+ case H_RESCINDED:
+ ret = -EACCES;
+ break;
+ case H_HARDWARE:
+ ret = -EPERM;
+ break;
+ case H_NOT_ENOUGH_RESOURCES:
+ case H_RESOURCE:
+ case H_BUSY:
+ ret = -EBUSY;
+ break;
+ default:
+ ret = -EINVAL;
+ break;
+ }
+
+ if (ret)
+ dev_dbg(dev, "%s: Sync h_cop_op failure (ret:%d) (hret:%ld)\n",
+ __func__, ret, hret);
+
+ op->hcall_err = hret;
+ return ret;
+}
+EXPORT_SYMBOL(vio_h_cop_sync);
+
static struct iommu_table *vio_build_iommu_table(struct vio_dev *dev)
{
const unsigned char *dma_window;
@@ -1211,35 +1314,87 @@ static void __devinit vio_dev_release(struct device *dev)
struct vio_dev *vio_register_device_node(struct device_node *of_node)
{
struct vio_dev *viodev;
+ struct device_node *parent_node;
const unsigned int *unit_address;
+ const unsigned int *pfo_resid = NULL;
+ enum vio_dev_family family;
+ const char *of_node_name = of_node->name ? of_node->name : "<unknown>";
- /* we need the 'device_type' property, in order to match with drivers */
- if (of_node->type == NULL) {
- printk(KERN_WARNING "%s: node %s missing 'device_type'\n",
- __func__,
- of_node->name ? of_node->name : "<unknown>");
+ /*
+ * Determine if this node is a under the /vdevice node or under the
+ * /ibm,platform-facilities node. This decides the device's family.
+ */
+ parent_node = of_get_parent(of_node);
+ if (parent_node) {
+ if (!strcmp(parent_node->full_name, "/ibm,platform-facilities"))
+ family = PFO;
+ else if (!strcmp(parent_node->full_name, "/vdevice"))
+ family = VDEVICE;
+ else {
+ pr_warn("%s: parent(%s) of %s not recognized.\n",
+ __func__,
+ parent_node->full_name,
+ of_node_name);
+ of_node_put(parent_node);
+ return NULL;
+ }
+ of_node_put(parent_node);
+ } else {
+ pr_warn("%s: could not determine the parent of node %s.\n",
+ __func__, of_node_name);
return NULL;
}
- unit_address = of_get_property(of_node, "reg", NULL);
- if (unit_address == NULL) {
- printk(KERN_WARNING "%s: node %s missing 'reg'\n",
- __func__,
- of_node->name ? of_node->name : "<unknown>");
- return NULL;
+ if (family == PFO) {
+ if (of_get_property(of_node, "interrupt-controller", NULL)) {
+ pr_debug("%s: Skipping the interrupt controller %s.\n",
+ __func__, of_node_name);
+ return NULL;
+ }
}
/* allocate a vio_dev for this node */
viodev = kzalloc(sizeof(struct vio_dev), GFP_KERNEL);
- if (viodev == NULL)
+ if (viodev == NULL) {
+ pr_warn("%s: allocation failure for VIO device.\n", __func__);
return NULL;
+ }
- viodev->irq = irq_of_parse_and_map(of_node, 0);
+ /* we need the 'device_type' property, in order to match with drivers */
+ viodev->family = family;
+ if (viodev->family == VDEVICE) {
+ if (of_node->type != NULL)
+ viodev->type = of_node->type;
+ else {
+ pr_warn("%s: node %s is missing the 'device_type' "
+ "property.\n", __func__, of_node_name);
+ goto out;
+ }
+
+ unit_address = of_get_property(of_node, "reg", NULL);
+ if (unit_address == NULL) {
+ pr_warn("%s: node %s missing 'reg'\n",
+ __func__, of_node_name);
+ goto out;
+ }
+ dev_set_name(&viodev->dev, "%x", *unit_address);
+ viodev->irq = irq_of_parse_and_map(of_node, 0);
+ viodev->unit_address = *unit_address;
+ } else {
+ /* PFO devices need their resource_id for submitting COP_OPs
+ * This is an optional field for devices, but is required when
+ * performing synchronous ops */
+ pfo_resid = of_get_property(of_node, "ibm,resource-id", NULL);
+ if (pfo_resid != NULL)
+ viodev->resource_id = *pfo_resid;
+
+ unit_address = NULL;
+ dev_set_name(&viodev->dev, "%s", of_node_name);
+ viodev->type = of_node_name;
+ viodev->irq = 0;
+ }
- dev_set_name(&viodev->dev, "%x", *unit_address);
viodev->name = of_node->name;
- viodev->type = of_node->type;
- viodev->unit_address = *unit_address;
viodev->dev.of_node = of_node_get(of_node);
if (firmware_has_feature(FW_FEATURE_CMO))
@@ -1267,16 +1422,51 @@ struct vio_dev *vio_register_device_node(struct device_node *of_node)
}
return viodev;
+
+out: /* Use this exit point for any return prior to device_register */
+ kfree(viodev);
+
+ return NULL;
}
EXPORT_SYMBOL(vio_register_device_node);
+/*
+ * vio_bus_scan_for_devices - Scan OF and register each child device
+ * @root_name - OF node name for the root of the subtree to search.
+ * This must be non-NULL
+ *
+ * Starting from the root node provide, register the device node for
+ * each child beneath the root.
+ */
+static void vio_bus_scan_register_devices(char *root_name)
+{
+ struct device_node *node_root, *node_child;
+
+ if (!root_name)
+ return;
+
+ node_root = of_find_node_by_name(NULL, root_name);
+ if (node_root) {
+
+ /*
+ * Create struct vio_devices for each virtual device in
+ * the device tree. Drivers will associate with them later.
+ */
+ node_child = of_get_next_child(node_root, NULL);
+ while (node_child) {
+ vio_register_device_node(node_child);
+ node_child = of_get_next_child(node_root, node_child);
+ }
+ of_node_put(node_root);
+ }
+}
+
/**
* vio_bus_init: - Initialize the virtual IO bus
*/
static int __init vio_bus_init(void)
{
int err;
- struct device_node *node_vroot;
if (firmware_has_feature(FW_FEATURE_CMO))
vio_cmo_sysfs_init();
@@ -1301,19 +1491,8 @@ static int __init vio_bus_init(void)
if (firmware_has_feature(FW_FEATURE_CMO))
vio_cmo_bus_init();
- node_vroot = of_find_node_by_name(NULL, "vdevice");
- if (node_vroot) {
- struct device_node *of_node;
-
- /*
- * Create struct vio_devices for each virtual device in
- * the device tree. Drivers will associate with them later.
- */
- for (of_node = node_vroot->child; of_node != NULL;
- of_node = of_node->sibling)
- vio_register_device_node(of_node);
- of_node_put(node_vroot);
- }
+ vio_bus_scan_register_devices("vdevice");
+ vio_bus_scan_register_devices("ibm,platform-facilities");
return 0;
}
@@ -1436,12 +1615,28 @@ struct vio_dev *vio_find_node(struct device_node *vnode)
{
const uint32_t *unit_address;
char kobj_name[20];
+ struct device_node *vnode_parent;
+ const char *dev_type;
+
+ vnode_parent = of_get_parent(vnode);
+ if (!vnode_parent)
+ return NULL;
+
+ dev_type = of_get_property(vnode_parent, "device_type", NULL);
+ of_node_put(vnode_parent);
+ if (!dev_type)
+ return NULL;
/* construct the kobject name from the device node */
- unit_address = of_get_property(vnode, "reg", NULL);
- if (!unit_address)
+ if (!strcmp(dev_type, "vdevice")) {
+ unit_address = of_get_property(vnode, "reg", NULL);
+ if (!unit_address)
+ return NULL;
+ snprintf(kobj_name, sizeof(kobj_name), "%x", *unit_address);
+ } else if (!strcmp(dev_type, "ibm,platform-facilities"))
+ snprintf(kobj_name, sizeof(kobj_name), "%s", vnode->name);
+ else
return NULL;
- snprintf(kobj_name, sizeof(kobj_name), "%x", *unit_address);
return vio_find_name(kobj_name);
}
--
1.7.1
^ permalink raw reply related
* [PATCH v2 12/17] powerpc: crypto: SHA256 hash routines for nx encryption
From: Kent Yoder @ 2012-04-10 15:10 UTC (permalink / raw)
To: linux-kernel; +Cc: key, rcj, linuxppc-dev, linux-crypto
In-Reply-To: <1334070249.16827.1.camel@key-ThinkPad-W510>
These routines add support for SHA-256 hashing on the Power7+ CPU's
in-Nest accelerator driver.
Signed-off-by: Kent Yoder <key@linux.vnet.ibm.com>
---
drivers/crypto/nx/nx-sha256.c | 244 +++++++++++++++++++++++++++++++++++++++++
1 files changed, 244 insertions(+), 0 deletions(-)
create mode 100644 drivers/crypto/nx/nx-sha256.c
diff --git a/drivers/crypto/nx/nx-sha256.c b/drivers/crypto/nx/nx-sha256.c
new file mode 100644
index 0000000..239cc3b
--- /dev/null
+++ b/drivers/crypto/nx/nx-sha256.c
@@ -0,0 +1,244 @@
+/**
+ * SHA-256 routines supporting the Power 7+ Nest Accelerators driver
+ *
+ * Copyright (C) 2011-2012 International Business Machines Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 only.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ *
+ * Author: Kent Yoder <yoder1@us.ibm.com>
+ */
+
+#include <crypto/internal/hash.h>
+#include <crypto/sha.h>
+#include <linux/module.h>
+#include <asm/vio.h>
+
+#include "nx_csbcpb.h"
+#include "nx.h"
+
+
+static int nx_sha256_init(struct shash_desc *desc)
+{
+ struct sha256_state *sctx = shash_desc_ctx(desc);
+ struct nx_crypto_ctx *nx_ctx = crypto_tfm_ctx(&desc->tfm->base);
+ struct nx_sg *out_sg;
+
+ nx_ctx_init(nx_ctx, HCOP_FC_SHA);
+
+ memset(sctx, 0, sizeof *sctx);
+
+ nx_ctx->ap = &nx_ctx->props[NX_PROPS_SHA256];
+
+ NX_CPB_SET_DIGEST_SIZE(nx_ctx->csbcpb, NX_DS_SHA256);
+ out_sg = nx_build_sg_list(nx_ctx->out_sg, (u8 *)sctx->state,
+ SHA256_DIGEST_SIZE, nx_ctx->ap->sglen);
+ nx_ctx->op.outlen = (nx_ctx->out_sg - out_sg) * sizeof(struct nx_sg);
+
+ return 0;
+}
+
+static int nx_sha256_update(struct shash_desc *desc, const u8 *data,
+ unsigned int len)
+{
+ struct sha256_state *sctx = shash_desc_ctx(desc);
+ struct nx_crypto_ctx *nx_ctx = crypto_tfm_ctx(&desc->tfm->base);
+ struct nx_csbcpb *csbcpb = (struct nx_csbcpb *)nx_ctx->csbcpb;
+ struct nx_sg *in_sg;
+ u64 to_process, leftover;
+ int rc = 0;
+
+ if (NX_CPB_FDM(csbcpb) & NX_FDM_CONTINUATION) {
+ /* we've hit the nx chip previously and we're updating again,
+ * so copy over the partial digest */
+ memcpy(csbcpb->cpb.sha256.input_partial_digest,
+ csbcpb->cpb.sha256.message_digest, SHA256_DIGEST_SIZE);
+ }
+
+ /* 2 cases for total data len:
+ * 1: <= SHA256_BLOCK_SIZE: copy into state, return 0
+ * 2: > SHA256_BLOCK_SIZE: process X blocks, copy in leftover
+ */
+ if (len + sctx->count <= SHA256_BLOCK_SIZE) {
+ memcpy(sctx->buf + sctx->count, data, len);
+ sctx->count += len;
+ goto out;
+ }
+
+ /* to_process: the SHA256_BLOCK_SIZE data chunk to process in this
+ * update */
+ to_process = (sctx->count + len) & ~(SHA256_BLOCK_SIZE - 1);
+ leftover = (sctx->count + len) & (SHA256_BLOCK_SIZE - 1);
+
+ if (sctx->count) {
+ in_sg = nx_build_sg_list(nx_ctx->in_sg, (u8 *)sctx->buf,
+ sctx->count, nx_ctx->ap->sglen);
+ in_sg = nx_build_sg_list(in_sg, (u8 *)data,
+ to_process - sctx->count,
+ nx_ctx->ap->sglen);
+ nx_ctx->op.inlen = (nx_ctx->in_sg - in_sg) *
+ sizeof(struct nx_sg);
+ } else {
+ in_sg = nx_build_sg_list(nx_ctx->in_sg, (u8 *)data,
+ to_process, nx_ctx->ap->sglen);
+ nx_ctx->op.inlen = (nx_ctx->in_sg - in_sg) *
+ sizeof(struct nx_sg);
+ }
+
+ NX_CPB_FDM(csbcpb) |= NX_FDM_INTERMEDIATE;
+
+ if (!nx_ctx->op.inlen || !nx_ctx->op.outlen) {
+ rc = -EINVAL;
+ goto out;
+ }
+
+ rc = nx_hcall_sync(nx_ctx, &nx_ctx->op);
+ if (rc)
+ goto out;
+
+ atomic_inc(&(nx_ctx->stats->sha256_ops));
+
+ /* copy the leftover back into the state struct */
+ memcpy(sctx->buf, data + len - leftover, leftover);
+ sctx->count = leftover;
+
+ csbcpb->cpb.sha256.message_bit_length += (u64)
+ (csbcpb->cpb.sha256.spbc * 8);
+
+ /* everything after the first update is continuation */
+ NX_CPB_FDM(csbcpb) |= NX_FDM_CONTINUATION;
+out:
+ return rc;
+}
+
+static int nx_sha256_final(struct shash_desc *desc, u8 *out)
+{
+ struct sha256_state *sctx = shash_desc_ctx(desc);
+ struct nx_crypto_ctx *nx_ctx = crypto_tfm_ctx(&desc->tfm->base);
+ struct nx_csbcpb *csbcpb = (struct nx_csbcpb *)nx_ctx->csbcpb;
+ struct nx_sg *in_sg, *out_sg;
+ int rc;
+
+ if (NX_CPB_FDM(csbcpb) & NX_FDM_CONTINUATION) {
+ /* we've hit the nx chip previously, now we're finalizing,
+ * so copy over the partial digest */
+ memcpy(csbcpb->cpb.sha256.input_partial_digest,
+ csbcpb->cpb.sha256.message_digest, SHA256_DIGEST_SIZE);
+ }
+
+ /* final is represented by continuing the operation and indicating that
+ * this is not an intermediate operation */
+ NX_CPB_FDM(csbcpb) &= ~NX_FDM_INTERMEDIATE;
+
+ csbcpb->cpb.sha256.message_bit_length += (u64)(sctx->count * 8);
+
+ in_sg = nx_build_sg_list(nx_ctx->in_sg, (u8 *)sctx->buf,
+ sctx->count, nx_ctx->ap->sglen);
+ out_sg = nx_build_sg_list(nx_ctx->out_sg, out, SHA256_DIGEST_SIZE,
+ nx_ctx->ap->sglen);
+ nx_ctx->op.inlen = (nx_ctx->in_sg - in_sg) * sizeof(struct nx_sg);
+ nx_ctx->op.outlen = (nx_ctx->out_sg - out_sg) * sizeof(struct nx_sg);
+
+ if (!nx_ctx->op.outlen) {
+ rc = -EINVAL;
+ goto out;
+ }
+
+ rc = nx_hcall_sync(nx_ctx, &nx_ctx->op);
+ if (rc)
+ goto out;
+
+ atomic_inc(&(nx_ctx->stats->sha256_ops));
+
+ atomic64_add(csbcpb->cpb.sha256.message_bit_length,
+ &(nx_ctx->stats->sha256_bytes));
+ memcpy(out, csbcpb->cpb.sha256.message_digest, SHA256_DIGEST_SIZE);
+out:
+ return rc;
+}
+
+static int nx_sha256_export(struct shash_desc *desc, void *out)
+{
+ struct sha256_state *sctx = shash_desc_ctx(desc);
+ struct nx_crypto_ctx *nx_ctx = crypto_tfm_ctx(&desc->tfm->base);
+ struct nx_csbcpb *csbcpb = (struct nx_csbcpb *)nx_ctx->csbcpb;
+ struct sha256_state *octx = out;
+
+ octx->count = sctx->count +
+ (csbcpb->cpb.sha256.message_bit_length / 8);
+ memcpy(octx->buf, sctx->buf, sizeof(octx->buf));
+
+ /* if no data has been processed yet, we need to export SHA256's
+ * initial data, in case this context gets imported into a software
+ * context */
+ if (csbcpb->cpb.sha256.message_bit_length)
+ memcpy(octx->state, csbcpb->cpb.sha256.message_digest,
+ SHA256_DIGEST_SIZE);
+ else {
+ octx->state[0] = SHA256_H0;
+ octx->state[1] = SHA256_H1;
+ octx->state[2] = SHA256_H2;
+ octx->state[3] = SHA256_H3;
+ octx->state[4] = SHA256_H4;
+ octx->state[5] = SHA256_H5;
+ octx->state[6] = SHA256_H6;
+ octx->state[7] = SHA256_H7;
+ }
+
+ return 0;
+}
+
+static int nx_sha256_import(struct shash_desc *desc, const void *in)
+{
+ struct sha256_state *sctx = shash_desc_ctx(desc);
+ struct nx_crypto_ctx *nx_ctx = crypto_tfm_ctx(&desc->tfm->base);
+ struct nx_csbcpb *csbcpb = (struct nx_csbcpb *)nx_ctx->csbcpb;
+ const struct sha256_state *ictx = in;
+
+ memcpy(sctx->buf, ictx->buf, sizeof(ictx->buf));
+
+ sctx->count = ictx->count & 0x3f;
+ csbcpb->cpb.sha256.message_bit_length = (ictx->count & ~0x3f) * 8;
+
+ if (csbcpb->cpb.sha256.message_bit_length) {
+ memcpy(csbcpb->cpb.sha256.message_digest, ictx->state,
+ SHA256_DIGEST_SIZE);
+
+ NX_CPB_FDM(csbcpb) |= NX_FDM_CONTINUATION;
+ NX_CPB_FDM(csbcpb) |= NX_FDM_INTERMEDIATE;
+ }
+
+ return 0;
+}
+
+struct shash_alg nx_shash_sha256_alg = {
+ .digestsize = SHA256_DIGEST_SIZE,
+ .init = nx_sha256_init,
+ .update = nx_sha256_update,
+ .final = nx_sha256_final,
+ .export = nx_sha256_export,
+ .import = nx_sha256_import,
+ .descsize = sizeof(struct sha256_state),
+ .statesize = sizeof(struct sha256_state),
+ .base = {
+ .cra_name = "sha256",
+ .cra_driver_name = "sha256-nx",
+ .cra_priority = 300,
+ .cra_flags = CRYPTO_ALG_TYPE_SHASH,
+ .cra_blocksize = SHA256_BLOCK_SIZE,
+ .cra_module = THIS_MODULE,
+ .cra_ctxsize = sizeof(struct nx_crypto_ctx),
+ .cra_init = nx_crypto_ctx_sha_init,
+ .cra_exit = nx_crypto_ctx_exit,
+ }
+};
--
1.7.1
^ permalink raw reply related
* [PATCH v2 16/17] powerpc: crypto: Build files for the nx device driver
From: Kent Yoder @ 2012-04-10 15:11 UTC (permalink / raw)
To: linux-kernel; +Cc: key, rcj, linuxppc-dev, linux-crypto
In-Reply-To: <1334070249.16827.1.camel@key-ThinkPad-W510>
These files support configuring and building the nx device driver.
Signed-off-by: Kent Yoder <key@linux.vnet.ibm.com>
---
arch/powerpc/Makefile | 1 +
drivers/crypto/Kconfig | 17 +++++++++++++++++
drivers/crypto/nx/Makefile | 11 +++++++++++
3 files changed, 29 insertions(+), 0 deletions(-)
create mode 100644 drivers/crypto/nx/Makefile
diff --git a/arch/powerpc/Makefile b/arch/powerpc/Makefile
index 6524c6e..16e1bd7 100644
--- a/arch/powerpc/Makefile
+++ b/arch/powerpc/Makefile
@@ -160,6 +160,7 @@ core-$(CONFIG_KVM) += arch/powerpc/kvm/
core-$(CONFIG_PERF_EVENTS) += arch/powerpc/perf/
drivers-$(CONFIG_OPROFILE) += arch/powerpc/oprofile/
+drivers-$(CONFIG_CRYPTO_DEV_NX) += drivers/crypto/nx/
# Default to zImage, override when needed
all: zImage
diff --git a/drivers/crypto/Kconfig b/drivers/crypto/Kconfig
index ab9abb4..4319248 100644
--- a/drivers/crypto/Kconfig
+++ b/drivers/crypto/Kconfig
@@ -295,4 +295,21 @@ config CRYPTO_DEV_TEGRA_AES
To compile this driver as a module, choose M here: the module
will be called tegra-aes.
+config CRYPTO_DEV_NX
+ tristate "Support for Power7+ in-Nest cryptographic accleration"
+ depends on PPC64 && IBMVIO
+ select CRYPTO_AES
+ select CRYPTO_CBC
+ select CRYPTO_ECB
+ select CRYPTO_CCM
+ select CRYPTO_GCM
+ select CRYPTO_AUTHENC
+ select CRYPTO_XCBC
+ select CRYPTO_SHA256
+ select CRYPTO_SHA512
+ help
+ Support for Power7+ in-Nest cryptographic acceleration. This
+ module supports acceleration for AES and SHA2 algorithms. If you
+ choose 'M' here, this module will be called nx_crypto.
+
endif # CRYPTO_HW
diff --git a/drivers/crypto/nx/Makefile b/drivers/crypto/nx/Makefile
new file mode 100644
index 0000000..411ce59
--- /dev/null
+++ b/drivers/crypto/nx/Makefile
@@ -0,0 +1,11 @@
+obj-$(CONFIG_CRYPTO_DEV_NX) += nx-crypto.o
+nx-crypto-objs := nx.o \
+ nx_debugfs.o \
+ nx-aes-cbc.o \
+ nx-aes-ecb.o \
+ nx-aes-gcm.o \
+ nx-aes-ccm.o \
+ nx-aes-ctr.o \
+ nx-aes-xcbc.o \
+ nx-sha256.o \
+ nx-sha512.o
--
1.7.1
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox