* [PATCH] ARM: hyp-stub: Fix Thumb-2 compilation
@ 2017-04-20 10:16 Marc Zyngier
2017-04-20 11:23 ` Christoffer Dall
0 siblings, 1 reply; 3+ messages in thread
From: Marc Zyngier @ 2017-04-20 10:16 UTC (permalink / raw)
To: linux-arm-kernel
The assembler defaults to emiting the short form of ADR, leading
to an out-of-range immediate. Using the wide version solves this
issue.
Fixes: bc845e4fbbbb ("ARM: KVM: Implement HVC_RESET_VECTORS stub hypercall in the init code")
Reported-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
---
arch/arm/kernel/hyp-stub.S | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/arm/kernel/hyp-stub.S b/arch/arm/kernel/hyp-stub.S
index d8523cc47a54..ec7e7377d423 100644
--- a/arch/arm/kernel/hyp-stub.S
+++ b/arch/arm/kernel/hyp-stub.S
@@ -125,7 +125,7 @@ ENTRY(__hyp_stub_install_secondary)
* (see safe_svcmode_maskall).
*/
@ Now install the hypervisor stub:
- adr r7, __hyp_stub_vectors
+ W(adr) r7, __hyp_stub_vectors
mcr p15, 4, r7, c12, c0, 0 @ set hypervisor vector base (HVBAR)
@ Disable all traps, so we don't get any nasty surprise
--
2.11.0
^ permalink raw reply related [flat|nested] 3+ messages in thread* [PATCH] ARM: hyp-stub: Fix Thumb-2 compilation
2017-04-20 10:16 [PATCH] ARM: hyp-stub: Fix Thumb-2 compilation Marc Zyngier
@ 2017-04-20 11:23 ` Christoffer Dall
2017-04-20 14:56 ` Marc Zyngier
0 siblings, 1 reply; 3+ messages in thread
From: Christoffer Dall @ 2017-04-20 11:23 UTC (permalink / raw)
To: linux-arm-kernel
On Thu, Apr 20, 2017 at 11:16:20AM +0100, Marc Zyngier wrote:
> The assembler defaults to emiting the short form of ADR, leading
> to an out-of-range immediate. Using the wide version solves this
> issue.
>
> Fixes: bc845e4fbbbb ("ARM: KVM: Implement HVC_RESET_VECTORS stub hypercall in the init code")
> Reported-by: Arnd Bergmann <arnd@arndb.de>
> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
> ---
> arch/arm/kernel/hyp-stub.S | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/arch/arm/kernel/hyp-stub.S b/arch/arm/kernel/hyp-stub.S
> index d8523cc47a54..ec7e7377d423 100644
> --- a/arch/arm/kernel/hyp-stub.S
> +++ b/arch/arm/kernel/hyp-stub.S
> @@ -125,7 +125,7 @@ ENTRY(__hyp_stub_install_secondary)
> * (see safe_svcmode_maskall).
> */
> @ Now install the hypervisor stub:
> - adr r7, __hyp_stub_vectors
> + W(adr) r7, __hyp_stub_vectors
> mcr p15, 4, r7, c12, c0, 0 @ set hypervisor vector base (HVBAR)
>
> @ Disable all traps, so we don't get any nasty surprise
> --
> 2.11.0
>
This fixes the compile, but I get this when booting it on TC2:
kvm [1]: 8-bit VMID
kvm [1]: IDMAP page: 80201000
kvm [1]: HYP VA range: 80000000:ffffffff
kvm [1]: Hyp mode initialized successfully
kvm [1]: vgic-v2 at 2c004000
kvm [1]: vgic interrupt IRQ16
kvm [1]: virtual timer IRQ19
Kernel panic - not syncing:
HYP panic: UNDEF PC:802010ac CPSR:800001da
CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.11.0-rc3+ #140
Hardware name: ARM-Versatile Express
[<8021c5f1>] (unwind_backtrace) from [<802199eb>] (show_stack+0xb/0xc)
[<802199eb>] (show_stack) from [<803ccc15>] (dump_stack+0x71/0x80)
[<803ccc15>] (dump_stack) from [<80289467>] (panic+0xbb/0x1d4)
[<80289467>] (panic) from [<8020b1f5>]
(_kvm_arch_hardware_disable+0x11/0x20)
[<8020b1f5>] (_kvm_arch_hardware_disable) from [<8026c497>]
(flush_smp_call_function_queue+0x77/0x134)
[<8026c497>] (flush_smp_call_function_queue) from [<8021b8f9>]
(handle_IPI+0xd9/0x150)
[<8021b8f9>] (handle_IPI) from [<80201425>] (gic_handle_irq+0x65/0x6c)
[<80201425>] (gic_handle_irq) from [<8021a213>] (__irq_svc+0x53/0x7c)
Exception stack(0x80a01f18 to 0x80a01f60)
1f00: 00000001
00000000
1f20: 80a01f70 802231c1 00000000 80a00000 00000000 80a03c80 80a03c34
808445d8
1f40: 80a01f70 80a03c88 805b31d1 80a01f68 80217ce1 80217ce2 40000033
ffffffff
[<8021a213>] (__irq_svc) from [<80217ce2>] (arch_cpu_idle+0x22/0x24)
[<80217ce2>] (arch_cpu_idle) from [<8024ca89>] (do_idle+0xf9/0x158)
[<8024ca89>] (do_idle) from [<8024ccaf>] (cpu_startup_entry+0x13/0x14)
[<8024ccaf>] (cpu_startup_entry) from [<80800a4d>]
(start_kernel+0x309/0x314)
[<80800a4d>] (start_kernel) from [<80008095>] (0x80008095)
CPU1: stopping
CPU: 1 PID: 1 Comm: swapper/0 Not tainted 4.11.0-rc3+ #140
Hardware name: ARM-Versatile Express
[<8021c5f1>] (unwind_backtrace) from [<802199eb>] (show_stack+0xb/0xc)
[<802199eb>] (show_stack) from [<803ccc15>] (dump_stack+0x71/0x80)
[<803ccc15>] (dump_stack) from [<8021b95d>] (handle_IPI+0x13d/0x150)
[<8021b95d>] (handle_IPI) from [<80201425>] (gic_handle_irq+0x65/0x6c)
[<80201425>] (gic_handle_irq) from [<8021a213>] (__irq_svc+0x53/0x7c)
Exception stack(0xeec47e20 to 0xeec47e68)
7e20: 00000000 ef1a96dc 80000000 80000000 80a03de4 80a03cf8 ef1b5a84
ef1b5a80
7e40: 00000001 00000008 80a03de4 80845a00 00000000 eec47e70 8026c9ed
8026ca02
7e60: a0000033 ffffffff
[<8021a213>] (__irq_svc) from [<8026ca02>]
(smp_call_function_many+0x1aa/0x1f0)
[<8026ca02>] (smp_call_function_many) from [<8026ca89>]
(on_each_cpu+0x1d/0x30)
[<8026ca89>] (on_each_cpu) from [<8020c369>] (kvm_arch_init+0x339/0x3fc)
[<8020c369>] (kvm_arch_init) from [<80207149>] (kvm_init+0x11/0x214)
[<80207149>] (kvm_init) from [<8020177d>] (do_one_initcall+0x2d/0x104)
[<8020177d>] (do_one_initcall) from [<80800bed>]
(kernel_init_freeable+0x195/0x208)
[<80800bed>] (kernel_init_freeable) from [<805b2383>]
(kernel_init+0x7/0xd4)
[<805b2383>] (kernel_init) from [<80217361>] (ret_from_fork+0x11/0x30)
SMP: failed to stop secondary CPUs
---[ end Kernel panic - not syncing:
HYP panic: UNDEF PC:802010ac CPSR:800001da
Thanks,
-Christoffer
^ permalink raw reply [flat|nested] 3+ messages in thread* [PATCH] ARM: hyp-stub: Fix Thumb-2 compilation
2017-04-20 11:23 ` Christoffer Dall
@ 2017-04-20 14:56 ` Marc Zyngier
0 siblings, 0 replies; 3+ messages in thread
From: Marc Zyngier @ 2017-04-20 14:56 UTC (permalink / raw)
To: linux-arm-kernel
On 20/04/17 12:23, Christoffer Dall wrote:
> On Thu, Apr 20, 2017 at 11:16:20AM +0100, Marc Zyngier wrote:
>> The assembler defaults to emiting the short form of ADR, leading
>> to an out-of-range immediate. Using the wide version solves this
>> issue.
>>
>> Fixes: bc845e4fbbbb ("ARM: KVM: Implement HVC_RESET_VECTORS stub hypercall in the init code")
>> Reported-by: Arnd Bergmann <arnd@arndb.de>
>> Signed-off-by: Marc Zyngier <marc.zyngier@arm.com>
>> ---
>> arch/arm/kernel/hyp-stub.S | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/arch/arm/kernel/hyp-stub.S b/arch/arm/kernel/hyp-stub.S
>> index d8523cc47a54..ec7e7377d423 100644
>> --- a/arch/arm/kernel/hyp-stub.S
>> +++ b/arch/arm/kernel/hyp-stub.S
>> @@ -125,7 +125,7 @@ ENTRY(__hyp_stub_install_secondary)
>> * (see safe_svcmode_maskall).
>> */
>> @ Now install the hypervisor stub:
>> - adr r7, __hyp_stub_vectors
>> + W(adr) r7, __hyp_stub_vectors
>> mcr p15, 4, r7, c12, c0, 0 @ set hypervisor vector base (HVBAR)
>>
>> @ Disable all traps, so we don't get any nasty surprise
>> --
>> 2.11.0
>>
>
> This fixes the compile, but I get this when booting it on TC2:
>
> kvm [1]: 8-bit VMID
> kvm [1]: IDMAP page: 80201000
> kvm [1]: HYP VA range: 80000000:ffffffff
> kvm [1]: Hyp mode initialized successfully
> kvm [1]: vgic-v2 at 2c004000
> kvm [1]: vgic interrupt IRQ16
> kvm [1]: virtual timer IRQ19
> Kernel panic - not syncing:
> HYP panic: UNDEF PC:802010ac CPSR:800001da
> CPU: 0 PID: 0 Comm: swapper/0 Not tainted 4.11.0-rc3+ #140
> Hardware name: ARM-Versatile Express
> [<8021c5f1>] (unwind_backtrace) from [<802199eb>] (show_stack+0xb/0xc)
> [<802199eb>] (show_stack) from [<803ccc15>] (dump_stack+0x71/0x80)
> [<803ccc15>] (dump_stack) from [<80289467>] (panic+0xbb/0x1d4)
> [<80289467>] (panic) from [<8020b1f5>]
> (_kvm_arch_hardware_disable+0x11/0x20)
> [<8020b1f5>] (_kvm_arch_hardware_disable) from [<8026c497>]
> (flush_smp_call_function_queue+0x77/0x134)
> [<8026c497>] (flush_smp_call_function_queue) from [<8021b8f9>]
> (handle_IPI+0xd9/0x150)
> [<8021b8f9>] (handle_IPI) from [<80201425>] (gic_handle_irq+0x65/0x6c)
> [<80201425>] (gic_handle_irq) from [<8021a213>] (__irq_svc+0x53/0x7c)
> Exception stack(0x80a01f18 to 0x80a01f60)
> 1f00: 00000001
> 00000000
> 1f20: 80a01f70 802231c1 00000000 80a00000 00000000 80a03c80 80a03c34
> 808445d8
> 1f40: 80a01f70 80a03c88 805b31d1 80a01f68 80217ce1 80217ce2 40000033
> ffffffff
> [<8021a213>] (__irq_svc) from [<80217ce2>] (arch_cpu_idle+0x22/0x24)
> [<80217ce2>] (arch_cpu_idle) from [<8024ca89>] (do_idle+0xf9/0x158)
> [<8024ca89>] (do_idle) from [<8024ccaf>] (cpu_startup_entry+0x13/0x14)
> [<8024ccaf>] (cpu_startup_entry) from [<80800a4d>]
> (start_kernel+0x309/0x314)
> [<80800a4d>] (start_kernel) from [<80008095>] (0x80008095)
> CPU1: stopping
> CPU: 1 PID: 1 Comm: swapper/0 Not tainted 4.11.0-rc3+ #140
> Hardware name: ARM-Versatile Express
> [<8021c5f1>] (unwind_backtrace) from [<802199eb>] (show_stack+0xb/0xc)
> [<802199eb>] (show_stack) from [<803ccc15>] (dump_stack+0x71/0x80)
> [<803ccc15>] (dump_stack) from [<8021b95d>] (handle_IPI+0x13d/0x150)
> [<8021b95d>] (handle_IPI) from [<80201425>] (gic_handle_irq+0x65/0x6c)
> [<80201425>] (gic_handle_irq) from [<8021a213>] (__irq_svc+0x53/0x7c)
> Exception stack(0xeec47e20 to 0xeec47e68)
> 7e20: 00000000 ef1a96dc 80000000 80000000 80a03de4 80a03cf8 ef1b5a84
> ef1b5a80
> 7e40: 00000001 00000008 80a03de4 80845a00 00000000 eec47e70 8026c9ed
> 8026ca02
> 7e60: a0000033 ffffffff
> [<8021a213>] (__irq_svc) from [<8026ca02>]
> (smp_call_function_many+0x1aa/0x1f0)
> [<8026ca02>] (smp_call_function_many) from [<8026ca89>]
> (on_each_cpu+0x1d/0x30)
> [<8026ca89>] (on_each_cpu) from [<8020c369>] (kvm_arch_init+0x339/0x3fc)
> [<8020c369>] (kvm_arch_init) from [<80207149>] (kvm_init+0x11/0x214)
> [<80207149>] (kvm_init) from [<8020177d>] (do_one_initcall+0x2d/0x104)
> [<8020177d>] (do_one_initcall) from [<80800bed>]
> (kernel_init_freeable+0x195/0x208)
> [<80800bed>] (kernel_init_freeable) from [<805b2383>]
> (kernel_init+0x7/0xd4)
> [<805b2383>] (kernel_init) from [<80217361>] (ret_from_fork+0x11/0x30)
> SMP: failed to stop secondary CPUs
> ---[ end Kernel panic - not syncing:
> HYP panic: UNDEF PC:802010ac CPSR:800001da
Grmbl... We inadvertently switched to ARM by applying a pretty silly
correction to the target PC when branching to the stub, while the
assembler has already provided us with a valid Thumb address.
/me hates Thumb, patch incoming.
Thanks,
M.
--
Jazz is not dead. It just smells funny...
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2017-04-20 14:56 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2017-04-20 10:16 [PATCH] ARM: hyp-stub: Fix Thumb-2 compilation Marc Zyngier
2017-04-20 11:23 ` Christoffer Dall
2017-04-20 14:56 ` Marc Zyngier
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).